From: Michael Tremer Date: Sun, 7 Feb 2010 14:54:18 +0000 (+0100) Subject: kernel: Copy patch files into repository. X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=58c5fc13210353a151f8b2d3c62d19da140b10ab;p=ipfire-3.x.git kernel: Copy patch files into repository. --- diff --git a/pkgs/core/kernel/kernel.nm b/pkgs/core/kernel/kernel.nm index e9244a183..32e7f5bec 100644 --- a/pkgs/core/kernel/kernel.nm +++ b/pkgs/core/kernel/kernel.nm @@ -41,40 +41,14 @@ define PKG_DESCRIPTION input and output, etc. endef -PKG_BUILD_DEPS+= -PKG_DEPS += - CFLAGS = CXXFLAGS = -LOCALVERSION = -ipfire1 -FULLVER= $(PKG_VER)$(LOCALVERSION) - -# Another Union FS -AUFS_PATCH = aufs2-$(PKG_VER)-1.patch -PKG_PATCHES += $(AUFS_PATCH) - -# Grsecurity -GRSEC_PATCH = grsecurity-2.1.14-$(PKG_VER)-200910012153.patch -PKG_PATCHES += $(GRSEC_PATCH) - -# OCF -#OCF_PATCH = ocf-linux-26-20080917.patch -PKG_PATCHES += $(OCF_PATCH) - -# Reiser4 Filesystem -REISER4_PATCH = reiser4-for-$(PKG_VER).patch -PKG_PATCHES += $(REISER4_PATCH) - -# Routing improvements taken from http://www.ssi.bg/~ja/#routes -ROUTES_PATCH = routes-$(PKG_VER)-16.diff -PKG_PATCHES += $(ROUTES_PATCH) +LOCALVERSION = -ipfire1 +FULLVER = $(PKG_VER)$(LOCALVERSION) PKG_TARBALL = $(THISAPP).tar.bz2 -#PKG_PATCHES += $(THISAPP)-disable-compat_vdso-1.patch -PKG_PATCHES += linux-$(PKG_VER)-scsi.h-fix-1.patch - ############################################################################### # Installation Details ############################################################################### diff --git a/pkgs/core/kernel/patches/aufs2-2.6.31.1-1.patch b/pkgs/core/kernel/patches/aufs2-2.6.31.1-1.patch new file mode 100644 index 000000000..1f6f6126e --- /dev/null +++ b/pkgs/core/kernel/patches/aufs2-2.6.31.1-1.patch @@ -0,0 +1,25456 @@ +diff -Nur linux-2.6.31-vanilla/Documentation/ABI/testing/debugfs-aufs linux-2.6.31/Documentation/ABI/testing/debugfs-aufs +--- linux-2.6.31-vanilla/Documentation/ABI/testing/debugfs-aufs 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/Documentation/ABI/testing/debugfs-aufs 2009-09-16 13:55:29.000000000 +0200 +@@ -0,0 +1,40 @@ ++What: /debug/aufs/si_/ ++Date: March 2009 ++Contact: J. R. Okajima ++Description: ++ Under /debug/aufs, a directory named si_ is created ++ per aufs mount, where is a unique id generated ++ internally. ++ ++What: /debug/aufs/si_/xib ++Date: March 2009 ++Contact: J. R. Okajima ++Description: ++ It shows the consumed blocks by xib (External Inode Number ++ Bitmap), its block size and file size. ++ When the aufs mount option 'noxino' is specified, it ++ will be empty. About XINO files, see ++ Documentation/filesystems/aufs/aufs.5 in detail. ++ ++What: /debug/aufs/si_/xino0, xino1 ... xinoN ++Date: March 2009 ++Contact: J. R. Okajima ++Description: ++ It shows the consumed blocks by xino (External Inode Number ++ Translation Table), its link count, block size and file ++ size. ++ When the aufs mount option 'noxino' is specified, it ++ will be empty. About XINO files, see ++ Documentation/filesystems/aufs/aufs.5 in detail. ++ ++What: /debug/aufs/si_/xigen ++Date: March 2009 ++Contact: J. R. Okajima ++Description: ++ It shows the consumed blocks by xigen (External Inode ++ Generation Table), its block size and file size. ++ If CONFIG_AUFS_EXPORT is disabled, this entry will not ++ be created. ++ When the aufs mount option 'noxino' is specified, it ++ will be empty. About XINO files, see ++ Documentation/filesystems/aufs/aufs.5 in detail. +diff -Nur linux-2.6.31-vanilla/Documentation/ABI/testing/sysfs-aufs linux-2.6.31/Documentation/ABI/testing/sysfs-aufs +--- linux-2.6.31-vanilla/Documentation/ABI/testing/sysfs-aufs 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/Documentation/ABI/testing/sysfs-aufs 2009-09-16 13:55:29.000000000 +0200 +@@ -0,0 +1,25 @@ ++What: /sys/fs/aufs/si_/ ++Date: March 2009 ++Contact: J. R. Okajima ++Description: ++ Under /sys/fs/aufs, a directory named si_ is created ++ per aufs mount, where is a unique id generated ++ internally. ++ ++What: /sys/fs/aufs/si_/br0, br1 ... brN ++Date: March 2009 ++Contact: J. R. Okajima ++Description: ++ It shows the abolute path of a member directory (which ++ is called branch) in aufs, and its permission. ++ ++What: /sys/fs/aufs/si_/xi_path ++Date: March 2009 ++Contact: J. R. Okajima ++Description: ++ It shows the abolute path of XINO (External Inode Number ++ Bitmap, Translation Table and Generation Table) file ++ even if it is the default path. ++ When the aufs mount option 'noxino' is specified, it ++ will be empty. About XINO files, see ++ Documentation/filesystems/aufs/aufs.5 in detail. +diff -Nur linux-2.6.31-vanilla/fs/aufs/aufs.h linux-2.6.31/fs/aufs/aufs.h +--- linux-2.6.31-vanilla/fs/aufs/aufs.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/aufs.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,51 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * all header files ++ */ ++ ++#ifndef __AUFS_H__ ++#define __AUFS_H__ ++ ++#ifdef __KERNEL__ ++ ++#include "debug.h" ++ ++#include "branch.h" ++#include "cpup.h" ++#include "dcsub.h" ++#include "dbgaufs.h" ++#include "dentry.h" ++#include "dir.h" ++#include "file.h" ++#include "fstype.h" ++#include "inode.h" ++#include "loop.h" ++#include "module.h" ++#include "opts.h" ++#include "rwsem.h" ++#include "spl.h" ++#include "super.h" ++#include "sysaufs.h" ++#include "vfsub.h" ++#include "whout.h" ++#include "wkq.h" ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/branch.c linux-2.6.31/fs/aufs/branch.c +--- linux-2.6.31-vanilla/fs/aufs/branch.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/branch.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,969 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * branch management ++ */ ++ ++#include ++#include "aufs.h" ++ ++/* ++ * free a single branch ++ */ ++static void au_br_do_free(struct au_branch *br) ++{ ++ int i; ++ struct au_wbr *wbr; ++ ++ if (br->br_xino.xi_file) ++ fput(br->br_xino.xi_file); ++ mutex_destroy(&br->br_xino.xi_nondir_mtx); ++ ++ AuDebugOn(atomic_read(&br->br_count)); ++ ++ wbr = br->br_wbr; ++ if (wbr) { ++ for (i = 0; i < AuBrWh_Last; i++) ++ dput(wbr->wbr_wh[i]); ++ AuDebugOn(atomic_read(&wbr->wbr_wh_running)); ++ AuRwDestroy(&wbr->wbr_wh_rwsem); ++ } ++ ++ /* some filesystems acquire extra lock */ ++ lockdep_off(); ++ mntput(br->br_mnt); ++ lockdep_on(); ++ ++ kfree(wbr); ++ kfree(br); ++} ++ ++/* ++ * frees all branches ++ */ ++void au_br_free(struct au_sbinfo *sbinfo) ++{ ++ aufs_bindex_t bmax; ++ struct au_branch **br; ++ ++ AuRwMustWriteLock(&sbinfo->si_rwsem); ++ ++ bmax = sbinfo->si_bend + 1; ++ br = sbinfo->si_branch; ++ while (bmax--) ++ au_br_do_free(*br++); ++} ++ ++/* ++ * find the index of a branch which is specified by @br_id. ++ */ ++int au_br_index(struct super_block *sb, aufs_bindex_t br_id) ++{ ++ aufs_bindex_t bindex, bend; ++ ++ bend = au_sbend(sb); ++ for (bindex = 0; bindex <= bend; bindex++) ++ if (au_sbr_id(sb, bindex) == br_id) ++ return bindex; ++ return -1; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * add a branch ++ */ ++ ++static int test_overlap(struct super_block *sb, struct dentry *h_d1, ++ struct dentry *h_d2) ++{ ++ if (unlikely(h_d1 == h_d2)) ++ return 1; ++ return !!au_test_subdir(h_d1, h_d2) ++ || !!au_test_subdir(h_d2, h_d1) ++ || au_test_loopback_overlap(sb, h_d1, h_d2) ++ || au_test_loopback_overlap(sb, h_d2, h_d1); ++} ++ ++/* ++ * returns a newly allocated branch. @new_nbranch is a number of branches ++ * after adding a branch. ++ */ ++static struct au_branch *au_br_alloc(struct super_block *sb, int new_nbranch, ++ int perm) ++{ ++ struct au_branch *add_branch; ++ struct dentry *root; ++ ++ root = sb->s_root; ++ add_branch = kmalloc(sizeof(*add_branch), GFP_NOFS); ++ if (unlikely(!add_branch)) ++ goto out; ++ ++ add_branch->br_wbr = NULL; ++ if (au_br_writable(perm)) { ++ /* may be freed separately at changing the branch permission */ ++ add_branch->br_wbr = kmalloc(sizeof(*add_branch->br_wbr), ++ GFP_NOFS); ++ if (unlikely(!add_branch->br_wbr)) ++ goto out_br; ++ } ++ ++ if (unlikely(au_sbr_realloc(au_sbi(sb), new_nbranch) ++ || au_di_realloc(au_di(root), new_nbranch) ++ || au_ii_realloc(au_ii(root->d_inode), new_nbranch))) ++ goto out_wbr; ++ return add_branch; /* success */ ++ ++ out_wbr: ++ kfree(add_branch->br_wbr); ++ out_br: ++ kfree(add_branch); ++ out: ++ return ERR_PTR(-ENOMEM); ++} ++ ++/* ++ * test if the branch permission is legal or not. ++ */ ++static int test_br(struct inode *inode, int brperm, char *path) ++{ ++ int err; ++ ++ err = 0; ++ if (unlikely(au_br_writable(brperm) && IS_RDONLY(inode))) { ++ AuErr("write permission for readonly mount or inode, %s\n", ++ path); ++ err = -EINVAL; ++ } ++ ++ return err; ++} ++ ++/* ++ * returns: ++ * 0: success, the caller will add it ++ * plus: success, it is already unified, the caller should ignore it ++ * minus: error ++ */ ++static int test_add(struct super_block *sb, struct au_opt_add *add, int remount) ++{ ++ int err; ++ aufs_bindex_t bend, bindex; ++ struct dentry *root; ++ struct inode *inode, *h_inode; ++ ++ root = sb->s_root; ++ bend = au_sbend(sb); ++ if (unlikely(bend >= 0 ++ && au_find_dbindex(root, add->path.dentry) >= 0)) { ++ err = 1; ++ if (!remount) { ++ err = -EINVAL; ++ AuErr("%s duplicated\n", add->pathname); ++ } ++ goto out; ++ } ++ ++ err = -ENOSPC; /* -E2BIG; */ ++ if (unlikely(AUFS_BRANCH_MAX <= add->bindex ++ || AUFS_BRANCH_MAX - 1 <= bend)) { ++ AuErr("number of branches exceeded %s\n", add->pathname); ++ goto out; ++ } ++ ++ err = -EDOM; ++ if (unlikely(add->bindex < 0 || bend + 1 < add->bindex)) { ++ AuErr("bad index %d\n", add->bindex); ++ goto out; ++ } ++ ++ inode = add->path.dentry->d_inode; ++ err = -ENOENT; ++ if (unlikely(!inode->i_nlink)) { ++ AuErr("no existence %s\n", add->pathname); ++ goto out; ++ } ++ ++ err = -EINVAL; ++ if (unlikely(inode->i_sb == sb)) { ++ AuErr("%s must be outside\n", add->pathname); ++ goto out; ++ } ++ ++ if (unlikely(au_test_fs_unsuppoted(inode->i_sb))) { ++ AuErr("unsupported filesystem, %s (%s)\n", ++ add->pathname, au_sbtype(inode->i_sb)); ++ goto out; ++ } ++ ++ err = test_br(add->path.dentry->d_inode, add->perm, add->pathname); ++ if (unlikely(err)) ++ goto out; ++ ++ if (bend < 0) ++ return 0; /* success */ ++ ++ err = -EINVAL; ++ for (bindex = 0; bindex <= bend; bindex++) ++ if (unlikely(test_overlap(sb, add->path.dentry, ++ au_h_dptr(root, bindex)))) { ++ AuErr("%s is overlapped\n", add->pathname); ++ goto out; ++ } ++ ++ err = 0; ++ if (au_opt_test(au_mntflags(sb), WARN_PERM)) { ++ h_inode = au_h_dptr(root, 0)->d_inode; ++ if ((h_inode->i_mode & S_IALLUGO) != (inode->i_mode & S_IALLUGO) ++ || h_inode->i_uid != inode->i_uid ++ || h_inode->i_gid != inode->i_gid) ++ AuWarn("uid/gid/perm %s %u/%u/0%o, %u/%u/0%o\n", ++ add->pathname, ++ inode->i_uid, inode->i_gid, ++ (inode->i_mode & S_IALLUGO), ++ h_inode->i_uid, h_inode->i_gid, ++ (h_inode->i_mode & S_IALLUGO)); ++ } ++ ++ out: ++ return err; ++} ++ ++/* ++ * initialize or clean the whiteouts for an adding branch ++ */ ++static int au_br_init_wh(struct super_block *sb, struct au_branch *br, ++ int new_perm, struct dentry *h_root) ++{ ++ int err, old_perm; ++ aufs_bindex_t bindex; ++ struct mutex *h_mtx; ++ struct au_wbr *wbr; ++ struct au_hinode *hdir; ++ ++ wbr = br->br_wbr; ++ old_perm = br->br_perm; ++ br->br_perm = new_perm; ++ hdir = NULL; ++ h_mtx = NULL; ++ bindex = au_br_index(sb, br->br_id); ++ if (0 <= bindex) { ++ hdir = au_hi(sb->s_root->d_inode, bindex); ++ au_hin_imtx_lock_nested(hdir, AuLsc_I_PARENT); ++ } else { ++ h_mtx = &h_root->d_inode->i_mutex; ++ mutex_lock_nested(h_mtx, AuLsc_I_PARENT); ++ } ++ if (!wbr) ++ err = au_wh_init(h_root, br, sb); ++ else { ++ wbr_wh_write_lock(wbr); ++ err = au_wh_init(h_root, br, sb); ++ wbr_wh_write_unlock(wbr); ++ } ++ if (hdir) ++ au_hin_imtx_unlock(hdir); ++ else ++ mutex_unlock(h_mtx); ++ br->br_perm = old_perm; ++ ++ if (!err && wbr && !au_br_writable(new_perm)) { ++ kfree(wbr); ++ br->br_wbr = NULL; ++ } ++ ++ return err; ++} ++ ++static int au_wbr_init(struct au_branch *br, struct super_block *sb, ++ int perm, struct path *path) ++{ ++ int err; ++ struct au_wbr *wbr; ++ ++ wbr = br->br_wbr; ++ au_rw_init(&wbr->wbr_wh_rwsem); ++ memset(wbr->wbr_wh, 0, sizeof(wbr->wbr_wh)); ++ atomic_set(&wbr->wbr_wh_running, 0); ++ wbr->wbr_bytes = 0; ++ ++ err = au_br_init_wh(sb, br, perm, path->dentry); ++ ++ return err; ++} ++ ++/* intialize a new branch */ ++static int au_br_init(struct au_branch *br, struct super_block *sb, ++ struct au_opt_add *add) ++{ ++ int err; ++ ++ err = 0; ++ memset(&br->br_xino, 0, sizeof(br->br_xino)); ++ mutex_init(&br->br_xino.xi_nondir_mtx); ++ br->br_perm = add->perm; ++ br->br_mnt = add->path.mnt; /* set first, mntget() later */ ++ atomic_set(&br->br_count, 0); ++ br->br_xino_upper = AUFS_XINO_TRUNC_INIT; ++ atomic_set(&br->br_xino_running, 0); ++ br->br_id = au_new_br_id(sb); ++ ++ if (au_br_writable(add->perm)) { ++ err = au_wbr_init(br, sb, add->perm, &add->path); ++ if (unlikely(err)) ++ goto out; ++ } ++ ++ if (au_opt_test(au_mntflags(sb), XINO)) { ++ err = au_xino_br(sb, br, add->path.dentry->d_inode->i_ino, ++ au_sbr(sb, 0)->br_xino.xi_file, /*do_test*/1); ++ if (unlikely(err)) { ++ AuDebugOn(br->br_xino.xi_file); ++ goto out; ++ } ++ } ++ ++ sysaufs_br_init(br); ++ mntget(add->path.mnt); ++ ++ out: ++ return err; ++} ++ ++static void au_br_do_add_brp(struct au_sbinfo *sbinfo, aufs_bindex_t bindex, ++ struct au_branch *br, aufs_bindex_t bend, ++ aufs_bindex_t amount) ++{ ++ struct au_branch **brp; ++ ++ AuRwMustWriteLock(&sbinfo->si_rwsem); ++ ++ brp = sbinfo->si_branch + bindex; ++ memmove(brp + 1, brp, sizeof(*brp) * amount); ++ *brp = br; ++ sbinfo->si_bend++; ++ if (unlikely(bend < 0)) ++ sbinfo->si_bend = 0; ++} ++ ++static void au_br_do_add_hdp(struct au_dinfo *dinfo, aufs_bindex_t bindex, ++ aufs_bindex_t bend, aufs_bindex_t amount) ++{ ++ struct au_hdentry *hdp; ++ ++ AuRwMustWriteLock(&dinfo->di_rwsem); ++ ++ hdp = dinfo->di_hdentry + bindex; ++ memmove(hdp + 1, hdp, sizeof(*hdp) * amount); ++ au_h_dentry_init(hdp); ++ dinfo->di_bend++; ++ if (unlikely(bend < 0)) ++ dinfo->di_bstart = 0; ++} ++ ++static void au_br_do_add_hip(struct au_iinfo *iinfo, aufs_bindex_t bindex, ++ aufs_bindex_t bend, aufs_bindex_t amount) ++{ ++ struct au_hinode *hip; ++ ++ AuRwMustWriteLock(&iinfo->ii_rwsem); ++ ++ hip = iinfo->ii_hinode + bindex; ++ memmove(hip + 1, hip, sizeof(*hip) * amount); ++ hip->hi_inode = NULL; ++ au_hin_init(hip, NULL); ++ iinfo->ii_bend++; ++ if (unlikely(bend < 0)) ++ iinfo->ii_bstart = 0; ++} ++ ++static void au_br_do_add(struct super_block *sb, struct dentry *h_dentry, ++ struct au_branch *br, aufs_bindex_t bindex) ++{ ++ struct dentry *root; ++ struct inode *root_inode; ++ aufs_bindex_t bend, amount; ++ ++ root = sb->s_root; ++ root_inode = root->d_inode; ++ au_plink_block_maintain(sb); ++ bend = au_sbend(sb); ++ amount = bend + 1 - bindex; ++ au_br_do_add_brp(au_sbi(sb), bindex, br, bend, amount); ++ au_br_do_add_hdp(au_di(root), bindex, bend, amount); ++ au_br_do_add_hip(au_ii(root_inode), bindex, bend, amount); ++ au_set_h_dptr(root, bindex, dget(h_dentry)); ++ au_set_h_iptr(root_inode, bindex, au_igrab(h_dentry->d_inode), ++ /*flags*/0); ++} ++ ++int au_br_add(struct super_block *sb, struct au_opt_add *add, int remount) ++{ ++ int err; ++ aufs_bindex_t bend, add_bindex; ++ struct dentry *root, *h_dentry; ++ struct inode *root_inode; ++ struct au_branch *add_branch; ++ ++ root = sb->s_root; ++ root_inode = root->d_inode; ++ IMustLock(root_inode); ++ err = test_add(sb, add, remount); ++ if (unlikely(err < 0)) ++ goto out; ++ if (err) { ++ err = 0; ++ goto out; /* success */ ++ } ++ ++ bend = au_sbend(sb); ++ add_branch = au_br_alloc(sb, bend + 2, add->perm); ++ err = PTR_ERR(add_branch); ++ if (IS_ERR(add_branch)) ++ goto out; ++ ++ err = au_br_init(add_branch, sb, add); ++ if (unlikely(err)) { ++ au_br_do_free(add_branch); ++ goto out; ++ } ++ ++ add_bindex = add->bindex; ++ h_dentry = add->path.dentry; ++ if (!remount) ++ au_br_do_add(sb, h_dentry, add_branch, add_bindex); ++ else { ++ sysaufs_brs_del(sb, add_bindex); ++ au_br_do_add(sb, h_dentry, add_branch, add_bindex); ++ sysaufs_brs_add(sb, add_bindex); ++ } ++ ++ if (!add_bindex) { ++ au_cpup_attr_all(root_inode, /*force*/1); ++ sb->s_maxbytes = h_dentry->d_sb->s_maxbytes; ++ } else ++ au_add_nlink(root_inode, h_dentry->d_inode); ++ ++ /* ++ * this test/set prevents aufs from handling unnecesary inotify events ++ * of xino files, in a case of re-adding a writable branch which was ++ * once detached from aufs. ++ */ ++ if (au_xino_brid(sb) < 0 ++ && au_br_writable(add_branch->br_perm) ++ && !au_test_fs_bad_xino(h_dentry->d_sb) ++ && add_branch->br_xino.xi_file ++ && add_branch->br_xino.xi_file->f_dentry->d_parent == h_dentry) ++ au_xino_brid_set(sb, add_branch->br_id); ++ ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * delete a branch ++ */ ++ ++/* to show the line number, do not make it inlined function */ ++#define AuVerbose(do_info, fmt, args...) do { \ ++ if (do_info) \ ++ AuInfo(fmt, ##args); \ ++} while (0) ++ ++/* ++ * test if the branch is deletable or not. ++ */ ++static int test_dentry_busy(struct dentry *root, aufs_bindex_t bindex, ++ unsigned int sigen) ++{ ++ int err, i, j, ndentry; ++ aufs_bindex_t bstart, bend; ++ unsigned char verbose; ++ struct au_dcsub_pages dpages; ++ struct au_dpage *dpage; ++ struct dentry *d; ++ struct inode *inode; ++ ++ err = au_dpages_init(&dpages, GFP_NOFS); ++ if (unlikely(err)) ++ goto out; ++ err = au_dcsub_pages(&dpages, root, NULL, NULL); ++ if (unlikely(err)) ++ goto out_dpages; ++ ++ verbose = !!au_opt_test(au_mntflags(root->d_sb), VERBOSE); ++ for (i = 0; !err && i < dpages.ndpage; i++) { ++ dpage = dpages.dpages + i; ++ ndentry = dpage->ndentry; ++ for (j = 0; !err && j < ndentry; j++) { ++ d = dpage->dentries[j]; ++ AuDebugOn(!atomic_read(&d->d_count)); ++ inode = d->d_inode; ++ if (au_digen(d) == sigen && au_iigen(inode) == sigen) ++ di_read_lock_child(d, AuLock_IR); ++ else { ++ di_write_lock_child(d); ++ err = au_reval_dpath(d, sigen); ++ if (!err) ++ di_downgrade_lock(d, AuLock_IR); ++ else { ++ di_write_unlock(d); ++ break; ++ } ++ } ++ ++ bstart = au_dbstart(d); ++ bend = au_dbend(d); ++ if (bstart <= bindex ++ && bindex <= bend ++ && au_h_dptr(d, bindex) ++ && (!S_ISDIR(inode->i_mode) || bstart == bend)) { ++ err = -EBUSY; ++ AuVerbose(verbose, "busy %.*s\n", AuDLNPair(d)); ++ } ++ di_read_unlock(d, AuLock_IR); ++ } ++ } ++ ++ out_dpages: ++ au_dpages_free(&dpages); ++ out: ++ return err; ++} ++ ++static int test_inode_busy(struct super_block *sb, aufs_bindex_t bindex, ++ unsigned int sigen) ++{ ++ int err; ++ struct inode *i; ++ aufs_bindex_t bstart, bend; ++ unsigned char verbose; ++ ++ err = 0; ++ verbose = !!au_opt_test(au_mntflags(sb), VERBOSE); ++ list_for_each_entry(i, &sb->s_inodes, i_sb_list) { ++ AuDebugOn(!atomic_read(&i->i_count)); ++ if (!list_empty(&i->i_dentry)) ++ continue; ++ ++ if (au_iigen(i) == sigen) ++ ii_read_lock_child(i); ++ else { ++ ii_write_lock_child(i); ++ err = au_refresh_hinode_self(i, /*do_attr*/1); ++ if (!err) ++ ii_downgrade_lock(i); ++ else { ++ ii_write_unlock(i); ++ break; ++ } ++ } ++ ++ bstart = au_ibstart(i); ++ bend = au_ibend(i); ++ if (bstart <= bindex ++ && bindex <= bend ++ && au_h_iptr(i, bindex) ++ && (!S_ISDIR(i->i_mode) || bstart == bend)) { ++ err = -EBUSY; ++ AuVerbose(verbose, "busy i%lu\n", i->i_ino); ++ ii_read_unlock(i); ++ break; ++ } ++ ii_read_unlock(i); ++ } ++ ++ return err; ++} ++ ++static int test_children_busy(struct dentry *root, aufs_bindex_t bindex) ++{ ++ int err; ++ unsigned int sigen; ++ ++ sigen = au_sigen(root->d_sb); ++ DiMustNoWaiters(root); ++ IiMustNoWaiters(root->d_inode); ++ di_write_unlock(root); ++ err = test_dentry_busy(root, bindex, sigen); ++ if (!err) ++ err = test_inode_busy(root->d_sb, bindex, sigen); ++ di_write_lock_child(root); /* aufs_write_lock() calls ..._child() */ ++ ++ return err; ++} ++ ++static void au_br_do_del_brp(struct au_sbinfo *sbinfo, ++ const aufs_bindex_t bindex, ++ const aufs_bindex_t bend) ++{ ++ struct au_branch **brp, **p; ++ ++ AuRwMustWriteLock(&sbinfo->si_rwsem); ++ ++ brp = sbinfo->si_branch + bindex; ++ if (bindex < bend) ++ memmove(brp, brp + 1, sizeof(*brp) * (bend - bindex)); ++ sbinfo->si_branch[0 + bend] = NULL; ++ sbinfo->si_bend--; ++ ++ p = krealloc(sbinfo->si_branch, sizeof(*p) * bend, GFP_NOFS); ++ if (p) ++ sbinfo->si_branch = p; ++} ++ ++static void au_br_do_del_hdp(struct au_dinfo *dinfo, const aufs_bindex_t bindex, ++ const aufs_bindex_t bend) ++{ ++ struct au_hdentry *hdp, *p; ++ ++ AuRwMustWriteLock(&dinfo->di_rwsem); ++ ++ hdp = dinfo->di_hdentry + bindex; ++ if (bindex < bend) ++ memmove(hdp, hdp + 1, sizeof(*hdp) * (bend - bindex)); ++ dinfo->di_hdentry[0 + bend].hd_dentry = NULL; ++ dinfo->di_bend--; ++ ++ p = krealloc(dinfo->di_hdentry, sizeof(*p) * bend, GFP_NOFS); ++ if (p) ++ dinfo->di_hdentry = p; ++} ++ ++static void au_br_do_del_hip(struct au_iinfo *iinfo, const aufs_bindex_t bindex, ++ const aufs_bindex_t bend) ++{ ++ struct au_hinode *hip, *p; ++ ++ AuRwMustWriteLock(&iinfo->ii_rwsem); ++ ++ hip = iinfo->ii_hinode + bindex; ++ if (bindex < bend) ++ memmove(hip, hip + 1, sizeof(*hip) * (bend - bindex)); ++ iinfo->ii_hinode[0 + bend].hi_inode = NULL; ++ au_hin_init(iinfo->ii_hinode + bend, NULL); ++ iinfo->ii_bend--; ++ ++ p = krealloc(iinfo->ii_hinode, sizeof(*p) * bend, GFP_NOFS); ++ if (p) ++ iinfo->ii_hinode = p; ++} ++ ++static void au_br_do_del(struct super_block *sb, aufs_bindex_t bindex, ++ struct au_branch *br) ++{ ++ aufs_bindex_t bend; ++ struct au_sbinfo *sbinfo; ++ struct dentry *root; ++ struct inode *inode; ++ ++ SiMustWriteLock(sb); ++ ++ root = sb->s_root; ++ inode = root->d_inode; ++ au_plink_block_maintain(sb); ++ sbinfo = au_sbi(sb); ++ bend = sbinfo->si_bend; ++ ++ dput(au_h_dptr(root, bindex)); ++ au_hiput(au_hi(inode, bindex)); ++ au_br_do_free(br); ++ ++ au_br_do_del_brp(sbinfo, bindex, bend); ++ au_br_do_del_hdp(au_di(root), bindex, bend); ++ au_br_do_del_hip(au_ii(inode), bindex, bend); ++} ++ ++int au_br_del(struct super_block *sb, struct au_opt_del *del, int remount) ++{ ++ int err, rerr, i; ++ unsigned int mnt_flags; ++ aufs_bindex_t bindex, bend, br_id; ++ unsigned char do_wh, verbose; ++ struct au_branch *br; ++ struct au_wbr *wbr; ++ ++ err = 0; ++ bindex = au_find_dbindex(sb->s_root, del->h_path.dentry); ++ if (bindex < 0) { ++ if (remount) ++ goto out; /* success */ ++ err = -ENOENT; ++ AuErr("%s no such branch\n", del->pathname); ++ goto out; ++ } ++ AuDbg("bindex b%d\n", bindex); ++ ++ err = -EBUSY; ++ mnt_flags = au_mntflags(sb); ++ verbose = !!au_opt_test(mnt_flags, VERBOSE); ++ bend = au_sbend(sb); ++ if (unlikely(!bend)) { ++ AuVerbose(verbose, "no more branches left\n"); ++ goto out; ++ } ++ br = au_sbr(sb, bindex); ++ i = atomic_read(&br->br_count); ++ if (unlikely(i)) { ++ AuVerbose(verbose, "%d file(s) opened\n", i); ++ goto out; ++ } ++ ++ wbr = br->br_wbr; ++ do_wh = wbr && (wbr->wbr_whbase || wbr->wbr_plink || wbr->wbr_orph); ++ if (do_wh) { ++ /* instead of WbrWhMustWriteLock(wbr) */ ++ SiMustWriteLock(sb); ++ for (i = 0; i < AuBrWh_Last; i++) { ++ dput(wbr->wbr_wh[i]); ++ wbr->wbr_wh[i] = NULL; ++ } ++ } ++ ++ err = test_children_busy(sb->s_root, bindex); ++ if (unlikely(err)) { ++ if (do_wh) ++ goto out_wh; ++ goto out; ++ } ++ ++ err = 0; ++ br_id = br->br_id; ++ if (!remount) ++ au_br_do_del(sb, bindex, br); ++ else { ++ sysaufs_brs_del(sb, bindex); ++ au_br_do_del(sb, bindex, br); ++ sysaufs_brs_add(sb, bindex); ++ } ++ ++ if (!bindex) { ++ au_cpup_attr_all(sb->s_root->d_inode, /*force*/1); ++ sb->s_maxbytes = au_sbr_sb(sb, 0)->s_maxbytes; ++ } else ++ au_sub_nlink(sb->s_root->d_inode, del->h_path.dentry->d_inode); ++ if (au_opt_test(mnt_flags, PLINK)) ++ au_plink_half_refresh(sb, br_id); ++ ++ if (au_xino_brid(sb) == br->br_id) ++ au_xino_brid_set(sb, -1); ++ goto out; /* success */ ++ ++ out_wh: ++ /* revert */ ++ rerr = au_br_init_wh(sb, br, br->br_perm, del->h_path.dentry); ++ if (rerr) ++ AuWarn("failed re-creating base whiteout, %s. (%d)\n", ++ del->pathname, rerr); ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * change a branch permission ++ */ ++ ++static void au_warn_ima(void) ++{ ++#ifdef CONFIG_IMA ++ AuWarn("RW -> RO makes IMA to produce wrong message"); ++#endif ++} ++ ++static int do_need_sigen_inc(int a, int b) ++{ ++ return au_br_whable(a) && !au_br_whable(b); ++} ++ ++static int need_sigen_inc(int old, int new) ++{ ++ return do_need_sigen_inc(old, new) ++ || do_need_sigen_inc(new, old); ++} ++ ++static int au_br_mod_files_ro(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ int err; ++ unsigned long n, ul, bytes, files; ++ aufs_bindex_t bstart; ++ struct file *file, *hf, **a; ++ const int step_bytes = 1024, /* memory allocation unit */ ++ step_files = step_bytes / sizeof(*a); ++ ++ err = -ENOMEM; ++ n = 0; ++ bytes = step_bytes; ++ files = step_files; ++ a = kmalloc(bytes, GFP_NOFS); ++ if (unlikely(!a)) ++ goto out; ++ ++ /* no need file_list_lock() since sbinfo is locked? defered? */ ++ list_for_each_entry(file, &sb->s_files, f_u.fu_list) { ++ if (special_file(file->f_dentry->d_inode->i_mode)) ++ continue; ++ ++ AuDbg("%.*s\n", AuDLNPair(file->f_dentry)); ++ fi_read_lock(file); ++ if (unlikely(au_test_mmapped(file))) { ++ err = -EBUSY; ++ FiMustNoWaiters(file); ++ fi_read_unlock(file); ++ goto out_free; ++ } ++ ++ bstart = au_fbstart(file); ++ if (!S_ISREG(file->f_dentry->d_inode->i_mode) ++ || !(file->f_mode & FMODE_WRITE) ++ || bstart != bindex) { ++ FiMustNoWaiters(file); ++ fi_read_unlock(file); ++ continue; ++ } ++ ++ hf = au_h_fptr(file, bstart); ++ FiMustNoWaiters(file); ++ fi_read_unlock(file); ++ ++ if (n < files) ++ a[n++] = hf; ++ else { ++ void *p; ++ ++ err = -ENOMEM; ++ bytes += step_bytes; ++ files += step_files; ++ p = krealloc(a, bytes, GFP_NOFS); ++ if (p) { ++ a = p; ++ a[n++] = hf; ++ } else ++ goto out_free; ++ } ++ } ++ ++ err = 0; ++ if (n) ++ au_warn_ima(); ++ for (ul = 0; ul < n; ul++) { ++ /* todo: already flushed? */ ++ /* cf. fs/super.c:mark_files_ro() */ ++ hf = a[ul]; ++ hf->f_mode &= ~FMODE_WRITE; ++ if (!file_check_writeable(hf)) { ++ file_release_write(hf); ++ mnt_drop_write(hf->f_vfsmnt); ++ } ++ } ++ ++ out_free: ++ kfree(a); ++ out: ++ return err; ++} ++ ++int au_br_mod(struct super_block *sb, struct au_opt_mod *mod, int remount, ++ int *do_update) ++{ ++ int err, rerr; ++ aufs_bindex_t bindex; ++ struct path path; ++ struct dentry *root; ++ struct au_branch *br; ++ ++ root = sb->s_root; ++ au_plink_block_maintain(sb); ++ bindex = au_find_dbindex(root, mod->h_root); ++ if (bindex < 0) { ++ if (remount) ++ return 0; /* success */ ++ err = -ENOENT; ++ AuErr("%s no such branch\n", mod->path); ++ goto out; ++ } ++ AuDbg("bindex b%d\n", bindex); ++ ++ err = test_br(mod->h_root->d_inode, mod->perm, mod->path); ++ if (unlikely(err)) ++ goto out; ++ ++ br = au_sbr(sb, bindex); ++ if (br->br_perm == mod->perm) ++ return 0; /* success */ ++ ++ if (au_br_writable(br->br_perm)) { ++ /* remove whiteout base */ ++ err = au_br_init_wh(sb, br, mod->perm, mod->h_root); ++ if (unlikely(err)) ++ goto out; ++ ++ if (!au_br_writable(mod->perm)) { ++ /* rw --> ro, file might be mmapped */ ++ DiMustNoWaiters(root); ++ IiMustNoWaiters(root->d_inode); ++ di_write_unlock(root); ++ err = au_br_mod_files_ro(sb, bindex); ++ /* aufs_write_lock() calls ..._child() */ ++ di_write_lock_child(root); ++ ++ if (unlikely(err)) { ++ rerr = -ENOMEM; ++ br->br_wbr = kmalloc(sizeof(*br->br_wbr), ++ GFP_NOFS); ++ if (br->br_wbr) { ++ path.mnt = br->br_mnt; ++ path.dentry = mod->h_root; ++ rerr = au_wbr_init(br, sb, br->br_perm, ++ &path); ++ } ++ if (unlikely(rerr)) { ++ AuIOErr("nested error %d (%d)\n", ++ rerr, err); ++ br->br_perm = mod->perm; ++ } ++ } ++ } ++ } else if (au_br_writable(mod->perm)) { ++ /* ro --> rw */ ++ err = -ENOMEM; ++ br->br_wbr = kmalloc(sizeof(*br->br_wbr), GFP_NOFS); ++ if (br->br_wbr) { ++ path.mnt = br->br_mnt; ++ path.dentry = mod->h_root; ++ err = au_wbr_init(br, sb, mod->perm, &path); ++ if (unlikely(err)) { ++ kfree(br->br_wbr); ++ br->br_wbr = NULL; ++ } ++ } ++ } ++ ++ if (!err) { ++ *do_update |= need_sigen_inc(br->br_perm, mod->perm); ++ br->br_perm = mod->perm; ++ } ++ ++ out: ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/branch.h linux-2.6.31/fs/aufs/branch.h +--- linux-2.6.31-vanilla/fs/aufs/branch.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/branch.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,219 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * branch filesystems and xino for them ++ */ ++ ++#ifndef __AUFS_BRANCH_H__ ++#define __AUFS_BRANCH_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++#include ++#include "rwsem.h" ++#include "super.h" ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* a xino file */ ++struct au_xino_file { ++ struct file *xi_file; ++ struct mutex xi_nondir_mtx; ++ ++ /* todo: make xino files an array to support huge inode number */ ++ ++#ifdef CONFIG_DEBUG_FS ++ struct dentry *xi_dbgaufs; ++#endif ++}; ++ ++/* members for writable branch only */ ++enum {AuBrWh_BASE, AuBrWh_PLINK, AuBrWh_ORPH, AuBrWh_Last}; ++struct au_wbr { ++ struct au_rwsem wbr_wh_rwsem; ++ struct dentry *wbr_wh[AuBrWh_Last]; ++ atomic_t wbr_wh_running; ++#define wbr_whbase wbr_wh[AuBrWh_BASE] /* whiteout base */ ++#define wbr_plink wbr_wh[AuBrWh_PLINK] /* pseudo-link dir */ ++#define wbr_orph wbr_wh[AuBrWh_ORPH] /* dir for orphans */ ++ ++ /* mfs mode */ ++ unsigned long long wbr_bytes; ++}; ++ ++/* protected by superblock rwsem */ ++struct au_branch { ++ struct au_xino_file br_xino; ++ ++ aufs_bindex_t br_id; ++ ++ int br_perm; ++ struct vfsmount *br_mnt; ++ atomic_t br_count; ++ ++ struct au_wbr *br_wbr; ++ ++ /* xino truncation */ ++ blkcnt_t br_xino_upper; /* watermark in blocks */ ++ atomic_t br_xino_running; ++ ++#ifdef CONFIG_SYSFS ++ /* an entry under sysfs per mount-point */ ++ char br_name[8]; ++ struct attribute br_attr; ++#endif ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* branch permission and attribute */ ++enum { ++ AuBrPerm_RW, /* writable, linkable wh */ ++ AuBrPerm_RO, /* readonly, no wh */ ++ AuBrPerm_RR, /* natively readonly, no wh */ ++ ++ AuBrPerm_RWNoLinkWH, /* un-linkable whiteouts */ ++ ++ AuBrPerm_ROWH, /* whiteout-able */ ++ AuBrPerm_RRWH, /* whiteout-able */ ++ ++ AuBrPerm_Last ++}; ++ ++static inline int au_br_writable(int brperm) ++{ ++ return brperm == AuBrPerm_RW || brperm == AuBrPerm_RWNoLinkWH; ++} ++ ++static inline int au_br_whable(int brperm) ++{ ++ return brperm == AuBrPerm_RW ++ || brperm == AuBrPerm_ROWH ++ || brperm == AuBrPerm_RRWH; ++} ++ ++static inline int au_br_rdonly(struct au_branch *br) ++{ ++ return ((br->br_mnt->mnt_sb->s_flags & MS_RDONLY) ++ || !au_br_writable(br->br_perm)) ++ ? -EROFS : 0; ++} ++ ++static inline int au_br_hinotifyable(int brperm __maybe_unused) ++{ ++#ifdef CONFIG_AUFS_HINOTIFY ++ return brperm != AuBrPerm_RR && brperm != AuBrPerm_RRWH; ++#else ++ return 0; ++#endif ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* branch.c */ ++struct au_sbinfo; ++void au_br_free(struct au_sbinfo *sinfo); ++int au_br_index(struct super_block *sb, aufs_bindex_t br_id); ++struct au_opt_add; ++int au_br_add(struct super_block *sb, struct au_opt_add *add, int remount); ++struct au_opt_del; ++int au_br_del(struct super_block *sb, struct au_opt_del *del, int remount); ++struct au_opt_mod; ++int au_br_mod(struct super_block *sb, struct au_opt_mod *mod, int remount, ++ int *do_update); ++ ++/* xino.c */ ++static const loff_t au_loff_max = LLONG_MAX; ++ ++int au_xib_trunc(struct super_block *sb); ++ssize_t xino_fread(au_readf_t func, struct file *file, void *buf, size_t size, ++ loff_t *pos); ++ssize_t xino_fwrite(au_writef_t func, struct file *file, void *buf, size_t size, ++ loff_t *pos); ++struct file *au_xino_create2(struct file *base_file, struct file *copy_src); ++struct file *au_xino_create(struct super_block *sb, char *fname, int silent); ++ino_t au_xino_new_ino(struct super_block *sb); ++int au_xino_write0(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, ++ ino_t ino); ++int au_xino_write(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, ++ ino_t ino); ++int au_xino_read(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, ++ ino_t *ino); ++int au_xino_br(struct super_block *sb, struct au_branch *br, ino_t hino, ++ struct file *base_file, int do_test); ++int au_xino_trunc(struct super_block *sb, aufs_bindex_t bindex); ++ ++struct au_opt_xino; ++int au_xino_set(struct super_block *sb, struct au_opt_xino *xino, int remount); ++void au_xino_clr(struct super_block *sb); ++struct file *au_xino_def(struct super_block *sb); ++int au_xino_path(struct seq_file *seq, struct file *file); ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* Superblock to branch */ ++static inline ++aufs_bindex_t au_sbr_id(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ return au_sbr(sb, bindex)->br_id; ++} ++ ++static inline ++struct vfsmount *au_sbr_mnt(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ return au_sbr(sb, bindex)->br_mnt; ++} ++ ++static inline ++struct super_block *au_sbr_sb(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ return au_sbr_mnt(sb, bindex)->mnt_sb; ++} ++ ++static inline void au_sbr_put(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ atomic_dec_return(&au_sbr(sb, bindex)->br_count); ++} ++ ++static inline int au_sbr_perm(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ return au_sbr(sb, bindex)->br_perm; ++} ++ ++static inline int au_sbr_whable(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ return au_br_whable(au_sbr_perm(sb, bindex)); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * wbr_wh_read_lock, wbr_wh_write_lock ++ * wbr_wh_read_unlock, wbr_wh_write_unlock, wbr_wh_downgrade_lock ++ */ ++AuSimpleRwsemFuncs(wbr_wh, struct au_wbr *wbr, &wbr->wbr_wh_rwsem); ++ ++#define WbrWhMustNoWaiters(wbr) AuRwMustNoWaiters(&wbr->wbr_wh_rwsem) ++#define WbrWhMustAnyLock(wbr) AuRwMustAnyLock(&wbr->wbr_wh_rwsem) ++#define WbrWhMustWriteLock(wbr) AuRwMustWriteLock(&wbr->wbr_wh_rwsem) ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_BRANCH_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/cpup.c linux-2.6.31/fs/aufs/cpup.c +--- linux-2.6.31-vanilla/fs/aufs/cpup.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/cpup.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,1048 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * copy-up functions, see wbr_policy.c for copy-down ++ */ ++ ++#include ++#include ++#include ++#include ++#include "aufs.h" ++ ++void au_cpup_attr_flags(struct inode *dst, struct inode *src) ++{ ++ const unsigned int mask = S_DEAD | S_SWAPFILE | S_PRIVATE ++ | S_NOATIME | S_NOCMTIME; ++ ++ dst->i_flags |= src->i_flags & ~mask; ++ if (au_test_fs_notime(dst->i_sb)) ++ dst->i_flags |= S_NOATIME | S_NOCMTIME; ++} ++ ++void au_cpup_attr_timesizes(struct inode *inode) ++{ ++ struct inode *h_inode; ++ ++ h_inode = au_h_iptr(inode, au_ibstart(inode)); ++ fsstack_copy_attr_times(inode, h_inode); ++ vfsub_copy_inode_size(inode, h_inode); ++} ++ ++void au_cpup_attr_nlink(struct inode *inode, int force) ++{ ++ struct inode *h_inode; ++ struct super_block *sb; ++ aufs_bindex_t bindex, bend; ++ ++ sb = inode->i_sb; ++ bindex = au_ibstart(inode); ++ h_inode = au_h_iptr(inode, bindex); ++ if (!force ++ && !S_ISDIR(h_inode->i_mode) ++ && au_opt_test(au_mntflags(sb), PLINK) ++ && au_plink_test(inode)) ++ return; ++ ++ inode->i_nlink = h_inode->i_nlink; ++ ++ /* ++ * fewer nlink makes find(1) noisy, but larger nlink doesn't. ++ * it may includes whplink directory. ++ */ ++ if (S_ISDIR(h_inode->i_mode)) { ++ bend = au_ibend(inode); ++ for (bindex++; bindex <= bend; bindex++) { ++ h_inode = au_h_iptr(inode, bindex); ++ if (h_inode) ++ au_add_nlink(inode, h_inode); ++ } ++ } ++} ++ ++void au_cpup_attr_changeable(struct inode *inode) ++{ ++ struct inode *h_inode; ++ ++ h_inode = au_h_iptr(inode, au_ibstart(inode)); ++ inode->i_mode = h_inode->i_mode; ++ inode->i_uid = h_inode->i_uid; ++ inode->i_gid = h_inode->i_gid; ++ au_cpup_attr_timesizes(inode); ++ au_cpup_attr_flags(inode, h_inode); ++} ++ ++void au_cpup_igen(struct inode *inode, struct inode *h_inode) ++{ ++ struct au_iinfo *iinfo = au_ii(inode); ++ ++ IiMustWriteLock(inode); ++ ++ iinfo->ii_higen = h_inode->i_generation; ++ iinfo->ii_hsb1 = h_inode->i_sb; ++} ++ ++void au_cpup_attr_all(struct inode *inode, int force) ++{ ++ struct inode *h_inode; ++ ++ h_inode = au_h_iptr(inode, au_ibstart(inode)); ++ au_cpup_attr_changeable(inode); ++ if (inode->i_nlink > 0) ++ au_cpup_attr_nlink(inode, force); ++ inode->i_rdev = h_inode->i_rdev; ++ inode->i_blkbits = h_inode->i_blkbits; ++ au_cpup_igen(inode, h_inode); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* Note: dt_dentry and dt_h_dentry are not dget/dput-ed */ ++ ++/* keep the timestamps of the parent dir when cpup */ ++void au_dtime_store(struct au_dtime *dt, struct dentry *dentry, ++ struct path *h_path) ++{ ++ struct inode *h_inode; ++ ++ dt->dt_dentry = dentry; ++ dt->dt_h_path = *h_path; ++ h_inode = h_path->dentry->d_inode; ++ dt->dt_atime = h_inode->i_atime; ++ dt->dt_mtime = h_inode->i_mtime; ++ /* smp_mb(); */ ++} ++ ++void au_dtime_revert(struct au_dtime *dt) ++{ ++ struct iattr attr; ++ int err; ++ ++ attr.ia_atime = dt->dt_atime; ++ attr.ia_mtime = dt->dt_mtime; ++ attr.ia_valid = ATTR_FORCE | ATTR_MTIME | ATTR_MTIME_SET ++ | ATTR_ATIME | ATTR_ATIME_SET; ++ ++ err = vfsub_notify_change(&dt->dt_h_path, &attr); ++ if (unlikely(err)) ++ AuWarn("restoring timestamps failed(%d). ignored\n", err); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static noinline_for_stack ++int cpup_iattr(struct dentry *dst, aufs_bindex_t bindex, struct dentry *h_src) ++{ ++ int err, sbits; ++ struct iattr ia; ++ struct path h_path; ++ struct inode *h_isrc, *h_idst; ++ ++ h_path.dentry = au_h_dptr(dst, bindex); ++ h_idst = h_path.dentry->d_inode; ++ h_path.mnt = au_sbr_mnt(dst->d_sb, bindex); ++ h_isrc = h_src->d_inode; ++ ia.ia_valid = ATTR_FORCE | ATTR_UID | ATTR_GID ++ | ATTR_ATIME | ATTR_MTIME ++ | ATTR_ATIME_SET | ATTR_MTIME_SET; ++ ia.ia_uid = h_isrc->i_uid; ++ ia.ia_gid = h_isrc->i_gid; ++ ia.ia_atime = h_isrc->i_atime; ++ ia.ia_mtime = h_isrc->i_mtime; ++ if (h_idst->i_mode != h_isrc->i_mode ++ && !S_ISLNK(h_idst->i_mode)) { ++ ia.ia_valid |= ATTR_MODE; ++ ia.ia_mode = h_isrc->i_mode; ++ } ++ sbits = !!(h_isrc->i_mode & (S_ISUID | S_ISGID)); ++ au_cpup_attr_flags(h_idst, h_isrc); ++ err = vfsub_notify_change(&h_path, &ia); ++ ++ /* is this nfs only? */ ++ if (!err && sbits && au_test_nfs(h_path.dentry->d_sb)) { ++ ia.ia_valid = ATTR_FORCE | ATTR_MODE; ++ ia.ia_mode = h_isrc->i_mode; ++ err = vfsub_notify_change(&h_path, &ia); ++ } ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int au_do_copy_file(struct file *dst, struct file *src, loff_t len, ++ char *buf, unsigned long blksize) ++{ ++ int err; ++ size_t sz, rbytes, wbytes; ++ unsigned char all_zero; ++ char *p, *zp; ++ struct mutex *h_mtx; ++ /* reduce stack usage */ ++ struct iattr *ia; ++ ++ zp = page_address(ZERO_PAGE(0)); ++ if (unlikely(!zp)) ++ return -ENOMEM; /* possible? */ ++ ++ err = 0; ++ all_zero = 0; ++ while (len) { ++ AuDbg("len %lld\n", len); ++ sz = blksize; ++ if (len < blksize) ++ sz = len; ++ ++ rbytes = 0; ++ /* todo: signal_pending? */ ++ while (!rbytes || err == -EAGAIN || err == -EINTR) { ++ rbytes = vfsub_read_k(src, buf, sz, &src->f_pos); ++ err = rbytes; ++ } ++ if (unlikely(err < 0)) ++ break; ++ ++ all_zero = 0; ++ if (len >= rbytes && rbytes == blksize) ++ all_zero = !memcmp(buf, zp, rbytes); ++ if (!all_zero) { ++ wbytes = rbytes; ++ p = buf; ++ while (wbytes) { ++ size_t b; ++ ++ b = vfsub_write_k(dst, p, wbytes, &dst->f_pos); ++ err = b; ++ /* todo: signal_pending? */ ++ if (unlikely(err == -EAGAIN || err == -EINTR)) ++ continue; ++ if (unlikely(err < 0)) ++ break; ++ wbytes -= b; ++ p += b; ++ } ++ } else { ++ loff_t res; ++ ++ AuLabel(hole); ++ res = vfsub_llseek(dst, rbytes, SEEK_CUR); ++ err = res; ++ if (unlikely(res < 0)) ++ break; ++ } ++ len -= rbytes; ++ err = 0; ++ } ++ ++ /* the last block may be a hole */ ++ if (!err && all_zero) { ++ AuLabel(last hole); ++ ++ err = 1; ++ if (au_test_nfs(dst->f_dentry->d_sb)) { ++ /* nfs requires this step to make last hole */ ++ /* is this only nfs? */ ++ do { ++ /* todo: signal_pending? */ ++ err = vfsub_write_k(dst, "\0", 1, &dst->f_pos); ++ } while (err == -EAGAIN || err == -EINTR); ++ if (err == 1) ++ dst->f_pos--; ++ } ++ ++ if (err == 1) { ++ ia = (void *)buf; ++ ia->ia_size = dst->f_pos; ++ ia->ia_valid = ATTR_SIZE | ATTR_FILE; ++ ia->ia_file = dst; ++ h_mtx = &dst->f_dentry->d_inode->i_mutex; ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD2); ++ err = vfsub_notify_change(&dst->f_path, ia); ++ mutex_unlock(h_mtx); ++ } ++ } ++ ++ return err; ++} ++ ++int au_copy_file(struct file *dst, struct file *src, loff_t len) ++{ ++ int err; ++ unsigned long blksize; ++ unsigned char do_kfree; ++ char *buf; ++ ++ err = -ENOMEM; ++ blksize = dst->f_dentry->d_sb->s_blocksize; ++ if (!blksize || PAGE_SIZE < blksize) ++ blksize = PAGE_SIZE; ++ AuDbg("blksize %lu\n", blksize); ++ do_kfree = (blksize != PAGE_SIZE && blksize >= sizeof(struct iattr *)); ++ if (do_kfree) ++ buf = kmalloc(blksize, GFP_NOFS); ++ else ++ buf = (void *)__get_free_page(GFP_NOFS); ++ if (unlikely(!buf)) ++ goto out; ++ ++ if (len > (1 << 22)) ++ AuDbg("copying a large file %lld\n", (long long)len); ++ ++ src->f_pos = 0; ++ dst->f_pos = 0; ++ err = au_do_copy_file(dst, src, len, buf, blksize); ++ if (do_kfree) ++ kfree(buf); ++ else ++ free_page((unsigned long)buf); ++ ++ out: ++ return err; ++} ++ ++/* ++ * to support a sparse file which is opened with O_APPEND, ++ * we need to close the file. ++ */ ++static int au_cp_regular(struct dentry *dentry, aufs_bindex_t bdst, ++ aufs_bindex_t bsrc, loff_t len) ++{ ++ int err, i; ++ enum { SRC, DST }; ++ struct { ++ aufs_bindex_t bindex; ++ unsigned int flags; ++ struct dentry *dentry; ++ struct file *file; ++ void *label, *label_file; ++ } *f, file[] = { ++ { ++ .bindex = bsrc, ++ .flags = O_RDONLY | O_NOATIME | O_LARGEFILE, ++ .file = NULL, ++ .label = &&out, ++ .label_file = &&out_src ++ }, ++ { ++ .bindex = bdst, ++ .flags = O_WRONLY | O_NOATIME | O_LARGEFILE, ++ .file = NULL, ++ .label = &&out_src, ++ .label_file = &&out_dst ++ } ++ }; ++ struct super_block *sb; ++ ++ /* bsrc branch can be ro/rw. */ ++ sb = dentry->d_sb; ++ f = file; ++ for (i = 0; i < 2; i++, f++) { ++ f->dentry = au_h_dptr(dentry, f->bindex); ++ f->file = au_h_open(dentry, f->bindex, f->flags, /*file*/NULL); ++ err = PTR_ERR(f->file); ++ if (IS_ERR(f->file)) ++ goto *f->label; ++ err = -EINVAL; ++ if (unlikely(!f->file->f_op)) ++ goto *f->label_file; ++ } ++ ++ /* try stopping to update while we copyup */ ++ IMustLock(file[SRC].dentry->d_inode); ++ err = au_copy_file(file[DST].file, file[SRC].file, len); ++ ++ out_dst: ++ fput(file[DST].file); ++ au_sbr_put(sb, file[DST].bindex); ++ out_src: ++ fput(file[SRC].file); ++ au_sbr_put(sb, file[SRC].bindex); ++ out: ++ return err; ++} ++ ++static int au_do_cpup_regular(struct dentry *dentry, aufs_bindex_t bdst, ++ aufs_bindex_t bsrc, loff_t len, ++ struct inode *h_dir, struct path *h_path) ++{ ++ int err, rerr; ++ loff_t l; ++ ++ err = 0; ++ l = i_size_read(au_h_iptr(dentry->d_inode, bsrc)); ++ if (len == -1 || l < len) ++ len = l; ++ if (len) ++ err = au_cp_regular(dentry, bdst, bsrc, len); ++ if (!err) ++ goto out; /* success */ ++ ++ rerr = vfsub_unlink(h_dir, h_path, /*force*/0); ++ if (rerr) { ++ AuIOErr("failed unlinking cpup-ed %.*s(%d, %d)\n", ++ AuDLNPair(h_path->dentry), err, rerr); ++ err = -EIO; ++ } ++ ++ out: ++ return err; ++} ++ ++static int au_do_cpup_symlink(struct path *h_path, struct dentry *h_src, ++ struct inode *h_dir) ++{ ++ int err, symlen; ++ mm_segment_t old_fs; ++ char *sym; ++ ++ err = -ENOSYS; ++ if (unlikely(!h_src->d_inode->i_op->readlink)) ++ goto out; ++ ++ err = -ENOMEM; ++ sym = __getname(); ++ if (unlikely(!sym)) ++ goto out; ++ ++ old_fs = get_fs(); ++ set_fs(KERNEL_DS); ++ symlen = h_src->d_inode->i_op->readlink(h_src, (char __user *)sym, ++ PATH_MAX); ++ err = symlen; ++ set_fs(old_fs); ++ ++ if (symlen > 0) { ++ sym[symlen] = 0; ++ err = vfsub_symlink(h_dir, h_path, sym); ++ } ++ __putname(sym); ++ ++ out: ++ return err; ++} ++ ++/* return with the lower dst inode is locked */ ++static noinline_for_stack ++int cpup_entry(struct dentry *dentry, aufs_bindex_t bdst, ++ aufs_bindex_t bsrc, loff_t len, unsigned int flags, ++ struct dentry *dst_parent) ++{ ++ int err; ++ umode_t mode; ++ unsigned int mnt_flags; ++ unsigned char isdir; ++ const unsigned char do_dt = !!au_ftest_cpup(flags, DTIME); ++ struct au_dtime dt; ++ struct path h_path; ++ struct dentry *h_src, *h_dst, *h_parent; ++ struct inode *h_inode, *h_dir; ++ struct super_block *sb; ++ ++ /* bsrc branch can be ro/rw. */ ++ h_src = au_h_dptr(dentry, bsrc); ++ h_inode = h_src->d_inode; ++ AuDebugOn(h_inode != au_h_iptr(dentry->d_inode, bsrc)); ++ ++ /* try stopping to be referenced while we are creating */ ++ h_dst = au_h_dptr(dentry, bdst); ++ h_parent = h_dst->d_parent; /* dir inode is locked */ ++ h_dir = h_parent->d_inode; ++ IMustLock(h_dir); ++ AuDebugOn(h_parent != h_dst->d_parent); ++ ++ sb = dentry->d_sb; ++ h_path.mnt = au_sbr_mnt(sb, bdst); ++ if (do_dt) { ++ h_path.dentry = h_parent; ++ au_dtime_store(&dt, dst_parent, &h_path); ++ } ++ h_path.dentry = h_dst; ++ ++ isdir = 0; ++ mode = h_inode->i_mode; ++ switch (mode & S_IFMT) { ++ case S_IFREG: ++ /* try stopping to update while we are referencing */ ++ IMustLock(h_inode); ++ err = vfsub_create(h_dir, &h_path, mode | S_IWUSR); ++ if (!err) ++ err = au_do_cpup_regular ++ (dentry, bdst, bsrc, len, ++ au_h_iptr(dst_parent->d_inode, bdst), &h_path); ++ break; ++ case S_IFDIR: ++ isdir = 1; ++ err = vfsub_mkdir(h_dir, &h_path, mode); ++ if (!err) { ++ /* ++ * strange behaviour from the users view, ++ * particularry setattr case ++ */ ++ if (au_ibstart(dst_parent->d_inode) == bdst) ++ au_cpup_attr_nlink(dst_parent->d_inode, ++ /*force*/1); ++ au_cpup_attr_nlink(dentry->d_inode, /*force*/1); ++ } ++ break; ++ case S_IFLNK: ++ err = au_do_cpup_symlink(&h_path, h_src, h_dir); ++ break; ++ case S_IFCHR: ++ case S_IFBLK: ++ AuDebugOn(!capable(CAP_MKNOD)); ++ /*FALLTHROUGH*/ ++ case S_IFIFO: ++ case S_IFSOCK: ++ err = vfsub_mknod(h_dir, &h_path, mode, h_inode->i_rdev); ++ break; ++ default: ++ AuIOErr("Unknown inode type 0%o\n", mode); ++ err = -EIO; ++ } ++ ++ mnt_flags = au_mntflags(sb); ++ if (!au_opt_test(mnt_flags, UDBA_NONE) ++ && !isdir ++ && au_opt_test(mnt_flags, XINO) ++ && h_inode->i_nlink == 1 ++ /* todo: unnecessary? */ ++ /* && dentry->d_inode->i_nlink == 1 */ ++ && bdst < bsrc ++ && !au_ftest_cpup(flags, KEEPLINO)) ++ au_xino_write(sb, bsrc, h_inode->i_ino, /*ino*/0); ++ /* ignore this error */ ++ ++ if (do_dt) ++ au_dtime_revert(&dt); ++ return err; ++} ++ ++/* ++ * copyup the @dentry from @bsrc to @bdst. ++ * the caller must set the both of lower dentries. ++ * @len is for truncating when it is -1 copyup the entire file. ++ * in link/rename cases, @dst_parent may be different from the real one. ++ */ ++static int au_cpup_single(struct dentry *dentry, aufs_bindex_t bdst, ++ aufs_bindex_t bsrc, loff_t len, unsigned int flags, ++ struct dentry *dst_parent) ++{ ++ int err, rerr; ++ aufs_bindex_t old_ibstart; ++ unsigned char isdir, plink; ++ struct au_dtime dt; ++ struct path h_path; ++ struct dentry *h_src, *h_dst, *h_parent; ++ struct inode *dst_inode, *h_dir, *inode; ++ struct super_block *sb; ++ ++ AuDebugOn(bsrc <= bdst); ++ ++ sb = dentry->d_sb; ++ h_path.mnt = au_sbr_mnt(sb, bdst); ++ h_dst = au_h_dptr(dentry, bdst); ++ h_parent = h_dst->d_parent; /* dir inode is locked */ ++ h_dir = h_parent->d_inode; ++ IMustLock(h_dir); ++ ++ h_src = au_h_dptr(dentry, bsrc); ++ inode = dentry->d_inode; ++ ++ if (!dst_parent) ++ dst_parent = dget_parent(dentry); ++ else ++ dget(dst_parent); ++ ++ plink = !!au_opt_test(au_mntflags(sb), PLINK); ++ dst_inode = au_h_iptr(inode, bdst); ++ if (dst_inode) { ++ if (unlikely(!plink)) { ++ err = -EIO; ++ AuIOErr("i%lu exists on a upper branch " ++ "but plink is disabled\n", inode->i_ino); ++ goto out; ++ } ++ ++ if (dst_inode->i_nlink) { ++ const int do_dt = au_ftest_cpup(flags, DTIME); ++ ++ h_src = au_plink_lkup(inode, bdst); ++ err = PTR_ERR(h_src); ++ if (IS_ERR(h_src)) ++ goto out; ++ if (unlikely(!h_src->d_inode)) { ++ err = -EIO; ++ AuIOErr("i%lu exists on a upper branch " ++ "but plink is broken\n", inode->i_ino); ++ dput(h_src); ++ goto out; ++ } ++ ++ if (do_dt) { ++ h_path.dentry = h_parent; ++ au_dtime_store(&dt, dst_parent, &h_path); ++ } ++ h_path.dentry = h_dst; ++ err = vfsub_link(h_src, h_dir, &h_path); ++ if (do_dt) ++ au_dtime_revert(&dt); ++ dput(h_src); ++ goto out; ++ } else ++ /* todo: cpup_wh_file? */ ++ /* udba work */ ++ au_update_brange(inode, 1); ++ } ++ ++ old_ibstart = au_ibstart(inode); ++ err = cpup_entry(dentry, bdst, bsrc, len, flags, dst_parent); ++ if (unlikely(err)) ++ goto out; ++ dst_inode = h_dst->d_inode; ++ mutex_lock_nested(&dst_inode->i_mutex, AuLsc_I_CHILD2); ++ ++ err = cpup_iattr(dentry, bdst, h_src); ++ isdir = S_ISDIR(dst_inode->i_mode); ++ if (!err) { ++ if (bdst < old_ibstart) ++ au_set_ibstart(inode, bdst); ++ au_set_h_iptr(inode, bdst, au_igrab(dst_inode), ++ au_hi_flags(inode, isdir)); ++ mutex_unlock(&dst_inode->i_mutex); ++ if (!isdir ++ && h_src->d_inode->i_nlink > 1 ++ && plink) ++ au_plink_append(inode, bdst, h_dst); ++ goto out; /* success */ ++ } ++ ++ /* revert */ ++ h_path.dentry = h_parent; ++ mutex_unlock(&dst_inode->i_mutex); ++ au_dtime_store(&dt, dst_parent, &h_path); ++ h_path.dentry = h_dst; ++ if (!isdir) ++ rerr = vfsub_unlink(h_dir, &h_path, /*force*/0); ++ else ++ rerr = vfsub_rmdir(h_dir, &h_path); ++ au_dtime_revert(&dt); ++ if (rerr) { ++ AuIOErr("failed removing broken entry(%d, %d)\n", err, rerr); ++ err = -EIO; ++ } ++ ++ out: ++ dput(dst_parent); ++ return err; ++} ++ ++struct au_cpup_single_args { ++ int *errp; ++ struct dentry *dentry; ++ aufs_bindex_t bdst, bsrc; ++ loff_t len; ++ unsigned int flags; ++ struct dentry *dst_parent; ++}; ++ ++static void au_call_cpup_single(void *args) ++{ ++ struct au_cpup_single_args *a = args; ++ *a->errp = au_cpup_single(a->dentry, a->bdst, a->bsrc, a->len, ++ a->flags, a->dst_parent); ++} ++ ++int au_sio_cpup_single(struct dentry *dentry, aufs_bindex_t bdst, ++ aufs_bindex_t bsrc, loff_t len, unsigned int flags, ++ struct dentry *dst_parent) ++{ ++ int err, wkq_err; ++ umode_t mode; ++ struct dentry *h_dentry; ++ ++ h_dentry = au_h_dptr(dentry, bsrc); ++ mode = h_dentry->d_inode->i_mode & S_IFMT; ++ if ((mode != S_IFCHR && mode != S_IFBLK) ++ || capable(CAP_MKNOD)) ++ err = au_cpup_single(dentry, bdst, bsrc, len, flags, ++ dst_parent); ++ else { ++ struct au_cpup_single_args args = { ++ .errp = &err, ++ .dentry = dentry, ++ .bdst = bdst, ++ .bsrc = bsrc, ++ .len = len, ++ .flags = flags, ++ .dst_parent = dst_parent ++ }; ++ wkq_err = au_wkq_wait(au_call_cpup_single, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ } ++ ++ return err; ++} ++ ++/* ++ * copyup the @dentry from the first active lower branch to @bdst, ++ * using au_cpup_single(). ++ */ ++static int au_cpup_simple(struct dentry *dentry, aufs_bindex_t bdst, loff_t len, ++ unsigned int flags) ++{ ++ int err; ++ aufs_bindex_t bsrc, bend; ++ ++ bend = au_dbend(dentry); ++ for (bsrc = bdst + 1; bsrc <= bend; bsrc++) ++ if (au_h_dptr(dentry, bsrc)) ++ break; ++ ++ err = au_lkup_neg(dentry, bdst); ++ if (!err) { ++ err = au_cpup_single(dentry, bdst, bsrc, len, flags, NULL); ++ if (!err) ++ return 0; /* success */ ++ ++ /* revert */ ++ au_set_h_dptr(dentry, bdst, NULL); ++ au_set_dbstart(dentry, bsrc); ++ } ++ ++ return err; ++} ++ ++struct au_cpup_simple_args { ++ int *errp; ++ struct dentry *dentry; ++ aufs_bindex_t bdst; ++ loff_t len; ++ unsigned int flags; ++}; ++ ++static void au_call_cpup_simple(void *args) ++{ ++ struct au_cpup_simple_args *a = args; ++ *a->errp = au_cpup_simple(a->dentry, a->bdst, a->len, a->flags); ++} ++ ++int au_sio_cpup_simple(struct dentry *dentry, aufs_bindex_t bdst, loff_t len, ++ unsigned int flags) ++{ ++ int err, wkq_err; ++ unsigned char do_sio; ++ struct dentry *parent; ++ struct inode *h_dir; ++ ++ parent = dget_parent(dentry); ++ h_dir = au_h_iptr(parent->d_inode, bdst); ++ do_sio = !!au_test_h_perm_sio(h_dir, MAY_EXEC | MAY_WRITE); ++ if (!do_sio) { ++ /* ++ * testing CAP_MKNOD is for generic fs, ++ * but CAP_FSETID is for xfs only, currently. ++ */ ++ umode_t mode = dentry->d_inode->i_mode; ++ do_sio = (((mode & (S_IFCHR | S_IFBLK)) ++ && !capable(CAP_MKNOD)) ++ || ((mode & (S_ISUID | S_ISGID)) ++ && !capable(CAP_FSETID))); ++ } ++ if (!do_sio) ++ err = au_cpup_simple(dentry, bdst, len, flags); ++ else { ++ struct au_cpup_simple_args args = { ++ .errp = &err, ++ .dentry = dentry, ++ .bdst = bdst, ++ .len = len, ++ .flags = flags ++ }; ++ wkq_err = au_wkq_wait(au_call_cpup_simple, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ } ++ ++ dput(parent); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * copyup the deleted file for writing. ++ */ ++static int au_do_cpup_wh(struct dentry *dentry, aufs_bindex_t bdst, ++ struct dentry *wh_dentry, struct file *file, ++ loff_t len) ++{ ++ int err; ++ aufs_bindex_t bstart; ++ struct au_dinfo *dinfo; ++ struct dentry *h_d_dst, *h_d_start; ++ ++ dinfo = au_di(dentry); ++ AuRwMustWriteLock(&dinfo->di_rwsem); ++ ++ bstart = dinfo->di_bstart; ++ h_d_dst = dinfo->di_hdentry[0 + bdst].hd_dentry; ++ dinfo->di_bstart = bdst; ++ dinfo->di_hdentry[0 + bdst].hd_dentry = wh_dentry; ++ h_d_start = dinfo->di_hdentry[0 + bstart].hd_dentry; ++ if (file) ++ dinfo->di_hdentry[0 + bstart].hd_dentry ++ = au_h_fptr(file, au_fbstart(file))->f_dentry; ++ err = au_cpup_single(dentry, bdst, bstart, len, !AuCpup_DTIME, ++ /*h_parent*/NULL); ++ if (!err && file) { ++ err = au_reopen_nondir(file); ++ dinfo->di_hdentry[0 + bstart].hd_dentry = h_d_start; ++ } ++ dinfo->di_hdentry[0 + bdst].hd_dentry = h_d_dst; ++ dinfo->di_bstart = bstart; ++ ++ return err; ++} ++ ++static int au_cpup_wh(struct dentry *dentry, aufs_bindex_t bdst, loff_t len, ++ struct file *file) ++{ ++ int err; ++ struct au_dtime dt; ++ struct dentry *parent, *h_parent, *wh_dentry; ++ struct au_branch *br; ++ struct path h_path; ++ ++ br = au_sbr(dentry->d_sb, bdst); ++ parent = dget_parent(dentry); ++ h_parent = au_h_dptr(parent, bdst); ++ wh_dentry = au_whtmp_lkup(h_parent, br, &dentry->d_name); ++ err = PTR_ERR(wh_dentry); ++ if (IS_ERR(wh_dentry)) ++ goto out; ++ ++ h_path.dentry = h_parent; ++ h_path.mnt = br->br_mnt; ++ au_dtime_store(&dt, parent, &h_path); ++ err = au_do_cpup_wh(dentry, bdst, wh_dentry, file, len); ++ if (unlikely(err)) ++ goto out_wh; ++ ++ dget(wh_dentry); ++ h_path.dentry = wh_dentry; ++ err = vfsub_unlink(h_parent->d_inode, &h_path, /*force*/0); ++ if (unlikely(err)) { ++ AuIOErr("failed remove copied-up tmp file %.*s(%d)\n", ++ AuDLNPair(wh_dentry), err); ++ err = -EIO; ++ } ++ au_dtime_revert(&dt); ++ au_set_hi_wh(dentry->d_inode, bdst, wh_dentry); ++ ++ out_wh: ++ dput(wh_dentry); ++ out: ++ dput(parent); ++ return err; ++} ++ ++struct au_cpup_wh_args { ++ int *errp; ++ struct dentry *dentry; ++ aufs_bindex_t bdst; ++ loff_t len; ++ struct file *file; ++}; ++ ++static void au_call_cpup_wh(void *args) ++{ ++ struct au_cpup_wh_args *a = args; ++ *a->errp = au_cpup_wh(a->dentry, a->bdst, a->len, a->file); ++} ++ ++int au_sio_cpup_wh(struct dentry *dentry, aufs_bindex_t bdst, loff_t len, ++ struct file *file) ++{ ++ int err, wkq_err; ++ struct dentry *parent, *h_orph, *h_parent, *h_dentry; ++ struct inode *dir, *h_dir, *h_tmpdir, *h_inode; ++ struct au_wbr *wbr; ++ ++ parent = dget_parent(dentry); ++ dir = parent->d_inode; ++ h_orph = NULL; ++ h_parent = NULL; ++ h_dir = au_igrab(au_h_iptr(dir, bdst)); ++ h_tmpdir = h_dir; ++ if (!h_dir->i_nlink) { ++ wbr = au_sbr(dentry->d_sb, bdst)->br_wbr; ++ h_orph = wbr->wbr_orph; ++ ++ h_parent = dget(au_h_dptr(parent, bdst)); ++ au_set_h_dptr(parent, bdst, NULL); ++ au_set_h_dptr(parent, bdst, dget(h_orph)); ++ h_tmpdir = h_orph->d_inode; ++ au_set_h_iptr(dir, bdst, NULL, 0); ++ au_set_h_iptr(dir, bdst, au_igrab(h_tmpdir), /*flags*/0); ++ ++ /* this temporary unlock is safe */ ++ if (file) ++ h_dentry = au_h_fptr(file, au_fbstart(file))->f_dentry; ++ else ++ h_dentry = au_h_dptr(dentry, au_dbstart(dentry)); ++ h_inode = h_dentry->d_inode; ++ IMustLock(h_inode); ++ mutex_unlock(&h_inode->i_mutex); ++ mutex_lock_nested(&h_tmpdir->i_mutex, AuLsc_I_PARENT3); ++ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); ++ } ++ ++ if (!au_test_h_perm_sio(h_tmpdir, MAY_EXEC | MAY_WRITE)) ++ err = au_cpup_wh(dentry, bdst, len, file); ++ else { ++ struct au_cpup_wh_args args = { ++ .errp = &err, ++ .dentry = dentry, ++ .bdst = bdst, ++ .len = len, ++ .file = file ++ }; ++ wkq_err = au_wkq_wait(au_call_cpup_wh, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ } ++ ++ if (h_orph) { ++ mutex_unlock(&h_tmpdir->i_mutex); ++ au_set_h_iptr(dir, bdst, NULL, 0); ++ au_set_h_iptr(dir, bdst, au_igrab(h_dir), /*flags*/0); ++ au_set_h_dptr(parent, bdst, NULL); ++ au_set_h_dptr(parent, bdst, h_parent); ++ } ++ iput(h_dir); ++ dput(parent); ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * generic routine for both of copy-up and copy-down. ++ */ ++/* cf. revalidate function in file.c */ ++int au_cp_dirs(struct dentry *dentry, aufs_bindex_t bdst, ++ int (*cp)(struct dentry *dentry, aufs_bindex_t bdst, ++ struct dentry *h_parent, void *arg), ++ void *arg) ++{ ++ int err; ++ struct au_pin pin; ++ struct dentry *d, *parent, *h_parent, *real_parent; ++ ++ err = 0; ++ parent = dget_parent(dentry); ++ if (IS_ROOT(parent)) ++ goto out; ++ ++ au_pin_init(&pin, dentry, bdst, AuLsc_DI_PARENT2, AuLsc_I_PARENT2, ++ au_opt_udba(dentry->d_sb), AuPin_MNT_WRITE); ++ ++ /* do not use au_dpage */ ++ real_parent = parent; ++ while (1) { ++ dput(parent); ++ parent = dget_parent(dentry); ++ h_parent = au_h_dptr(parent, bdst); ++ if (h_parent) ++ goto out; /* success */ ++ ++ /* find top dir which is necessary to cpup */ ++ do { ++ d = parent; ++ dput(parent); ++ parent = dget_parent(d); ++ di_read_lock_parent3(parent, !AuLock_IR); ++ h_parent = au_h_dptr(parent, bdst); ++ di_read_unlock(parent, !AuLock_IR); ++ } while (!h_parent); ++ ++ if (d != real_parent) ++ di_write_lock_child3(d); ++ ++ /* somebody else might create while we were sleeping */ ++ if (!au_h_dptr(d, bdst) || !au_h_dptr(d, bdst)->d_inode) { ++ if (au_h_dptr(d, bdst)) ++ au_update_dbstart(d); ++ ++ au_pin_set_dentry(&pin, d); ++ err = au_do_pin(&pin); ++ if (!err) { ++ err = cp(d, bdst, h_parent, arg); ++ au_unpin(&pin); ++ } ++ } ++ ++ if (d != real_parent) ++ di_write_unlock(d); ++ if (unlikely(err)) ++ break; ++ } ++ ++ out: ++ dput(parent); ++ return err; ++} ++ ++static int au_cpup_dir(struct dentry *dentry, aufs_bindex_t bdst, ++ struct dentry *h_parent __maybe_unused , ++ void *arg __maybe_unused) ++{ ++ return au_sio_cpup_simple(dentry, bdst, -1, AuCpup_DTIME); ++} ++ ++int au_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst) ++{ ++ return au_cp_dirs(dentry, bdst, au_cpup_dir, NULL); ++} ++ ++int au_test_and_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst) ++{ ++ int err; ++ struct dentry *parent; ++ struct inode *dir; ++ ++ parent = dget_parent(dentry); ++ dir = parent->d_inode; ++ err = 0; ++ if (au_h_iptr(dir, bdst)) ++ goto out; ++ ++ di_read_unlock(parent, AuLock_IR); ++ di_write_lock_parent(parent); ++ /* someone else might change our inode while we were sleeping */ ++ if (!au_h_iptr(dir, bdst)) ++ err = au_cpup_dirs(dentry, bdst); ++ di_downgrade_lock(parent, AuLock_IR); ++ ++ out: ++ dput(parent); ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/cpup.h linux-2.6.31/fs/aufs/cpup.h +--- linux-2.6.31-vanilla/fs/aufs/cpup.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/cpup.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,81 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * copy-up/down functions ++ */ ++ ++#ifndef __AUFS_CPUP_H__ ++#define __AUFS_CPUP_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++#include ++ ++struct inode; ++struct file; ++ ++void au_cpup_attr_flags(struct inode *dst, struct inode *src); ++void au_cpup_attr_timesizes(struct inode *inode); ++void au_cpup_attr_nlink(struct inode *inode, int force); ++void au_cpup_attr_changeable(struct inode *inode); ++void au_cpup_igen(struct inode *inode, struct inode *h_inode); ++void au_cpup_attr_all(struct inode *inode, int force); ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* cpup flags */ ++#define AuCpup_DTIME 1 /* do dtime_store/revert */ ++#define AuCpup_KEEPLINO (1 << 1) /* do not clear the lower xino, ++ for link(2) */ ++#define au_ftest_cpup(flags, name) ((flags) & AuCpup_##name) ++#define au_fset_cpup(flags, name) { (flags) |= AuCpup_##name; } ++#define au_fclr_cpup(flags, name) { (flags) &= ~AuCpup_##name; } ++ ++int au_copy_file(struct file *dst, struct file *src, loff_t len); ++int au_sio_cpup_single(struct dentry *dentry, aufs_bindex_t bdst, ++ aufs_bindex_t bsrc, loff_t len, unsigned int flags, ++ struct dentry *dst_parent); ++int au_sio_cpup_simple(struct dentry *dentry, aufs_bindex_t bdst, loff_t len, ++ unsigned int flags); ++int au_sio_cpup_wh(struct dentry *dentry, aufs_bindex_t bdst, loff_t len, ++ struct file *file); ++ ++int au_cp_dirs(struct dentry *dentry, aufs_bindex_t bdst, ++ int (*cp)(struct dentry *dentry, aufs_bindex_t bdst, ++ struct dentry *h_parent, void *arg), ++ void *arg); ++int au_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst); ++int au_test_and_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst); ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* keep timestamps when copyup */ ++struct au_dtime { ++ struct dentry *dt_dentry; ++ struct path dt_h_path; ++ struct timespec dt_atime, dt_mtime; ++}; ++void au_dtime_store(struct au_dtime *dt, struct dentry *dentry, ++ struct path *h_path); ++void au_dtime_revert(struct au_dtime *dt); ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_CPUP_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/dbgaufs.c linux-2.6.31/fs/aufs/dbgaufs.c +--- linux-2.6.31-vanilla/fs/aufs/dbgaufs.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/dbgaufs.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,331 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * debugfs interface ++ */ ++ ++#include ++#include "aufs.h" ++ ++#ifndef CONFIG_SYSFS ++#error DEBUG_FS depends upon SYSFS ++#endif ++ ++static struct dentry *dbgaufs; ++static const mode_t dbgaufs_mode = S_IRUSR | S_IRGRP | S_IROTH; ++ ++/* 20 is max digits length of ulong 64 */ ++struct dbgaufs_arg { ++ int n; ++ char a[20 * 4]; ++}; ++ ++/* ++ * common function for all XINO files ++ */ ++static int dbgaufs_xi_release(struct inode *inode __maybe_unused, ++ struct file *file) ++{ ++ kfree(file->private_data); ++ return 0; ++} ++ ++static int dbgaufs_xi_open(struct file *xf, struct file *file, int do_fcnt) ++{ ++ int err; ++ struct kstat st; ++ struct dbgaufs_arg *p; ++ ++ err = -ENOMEM; ++ p = kmalloc(sizeof(*p), GFP_NOFS); ++ if (unlikely(!p)) ++ goto out; ++ ++ err = 0; ++ p->n = 0; ++ file->private_data = p; ++ if (!xf) ++ goto out; ++ ++ err = vfs_getattr(xf->f_vfsmnt, xf->f_dentry, &st); ++ if (!err) { ++ if (do_fcnt) ++ p->n = snprintf ++ (p->a, sizeof(p->a), "%ld, %llux%lu %lld\n", ++ (long)file_count(xf), st.blocks, st.blksize, ++ (long long)st.size); ++ else ++ p->n = snprintf(p->a, sizeof(p->a), "%llux%lu %lld\n", ++ st.blocks, st.blksize, ++ (long long)st.size); ++ AuDebugOn(p->n >= sizeof(p->a)); ++ } else { ++ p->n = snprintf(p->a, sizeof(p->a), "err %d\n", err); ++ err = 0; ++ } ++ ++ out: ++ return err; ++ ++} ++ ++static ssize_t dbgaufs_xi_read(struct file *file, char __user *buf, ++ size_t count, loff_t *ppos) ++{ ++ struct dbgaufs_arg *p; ++ ++ p = file->private_data; ++ return simple_read_from_buffer(buf, count, ppos, p->a, p->n); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int dbgaufs_xib_open(struct inode *inode, struct file *file) ++{ ++ int err; ++ struct au_sbinfo *sbinfo; ++ struct super_block *sb; ++ ++ sbinfo = inode->i_private; ++ sb = sbinfo->si_sb; ++ si_noflush_read_lock(sb); ++ err = dbgaufs_xi_open(sbinfo->si_xib, file, /*do_fcnt*/0); ++ si_read_unlock(sb); ++ return err; ++} ++ ++static const struct file_operations dbgaufs_xib_fop = { ++ .open = dbgaufs_xib_open, ++ .release = dbgaufs_xi_release, ++ .read = dbgaufs_xi_read ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++#define DbgaufsXi_PREFIX "xi" ++ ++static int dbgaufs_xino_open(struct inode *inode, struct file *file) ++{ ++ int err; ++ long l; ++ struct au_sbinfo *sbinfo; ++ struct super_block *sb; ++ struct file *xf; ++ struct qstr *name; ++ ++ err = -ENOENT; ++ xf = NULL; ++ name = &file->f_dentry->d_name; ++ if (unlikely(name->len < sizeof(DbgaufsXi_PREFIX) ++ || memcmp(name->name, DbgaufsXi_PREFIX, ++ sizeof(DbgaufsXi_PREFIX) - 1))) ++ goto out; ++ err = strict_strtol(name->name + sizeof(DbgaufsXi_PREFIX) - 1, 10, &l); ++ if (unlikely(err)) ++ goto out; ++ ++ sbinfo = inode->i_private; ++ sb = sbinfo->si_sb; ++ si_noflush_read_lock(sb); ++ if (l <= au_sbend(sb)) { ++ xf = au_sbr(sb, (aufs_bindex_t)l)->br_xino.xi_file; ++ err = dbgaufs_xi_open(xf, file, /*do_fcnt*/1); ++ } else ++ err = -ENOENT; ++ si_read_unlock(sb); ++ ++ out: ++ return err; ++} ++ ++static const struct file_operations dbgaufs_xino_fop = { ++ .open = dbgaufs_xino_open, ++ .release = dbgaufs_xi_release, ++ .read = dbgaufs_xi_read ++}; ++ ++void dbgaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ aufs_bindex_t bend; ++ struct au_branch *br; ++ struct au_xino_file *xi; ++ ++ if (!au_sbi(sb)->si_dbgaufs) ++ return; ++ ++ bend = au_sbend(sb); ++ for (; bindex <= bend; bindex++) { ++ br = au_sbr(sb, bindex); ++ xi = &br->br_xino; ++ if (xi->xi_dbgaufs) { ++ debugfs_remove(xi->xi_dbgaufs); ++ xi->xi_dbgaufs = NULL; ++ } ++ } ++} ++ ++void dbgaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ struct au_sbinfo *sbinfo; ++ struct dentry *parent; ++ struct au_branch *br; ++ struct au_xino_file *xi; ++ aufs_bindex_t bend; ++ char name[sizeof(DbgaufsXi_PREFIX) + 5]; /* "xi" bindex NULL */ ++ ++ sbinfo = au_sbi(sb); ++ parent = sbinfo->si_dbgaufs; ++ if (!parent) ++ return; ++ ++ bend = au_sbend(sb); ++ for (; bindex <= bend; bindex++) { ++ snprintf(name, sizeof(name), DbgaufsXi_PREFIX "%d", bindex); ++ br = au_sbr(sb, bindex); ++ xi = &br->br_xino; ++ AuDebugOn(xi->xi_dbgaufs); ++ xi->xi_dbgaufs = debugfs_create_file(name, dbgaufs_mode, parent, ++ sbinfo, &dbgaufs_xino_fop); ++ /* ignore an error */ ++ if (unlikely(!xi->xi_dbgaufs)) ++ AuWarn1("failed %s under debugfs\n", name); ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++#ifdef CONFIG_AUFS_EXPORT ++static int dbgaufs_xigen_open(struct inode *inode, struct file *file) ++{ ++ int err; ++ struct au_sbinfo *sbinfo; ++ struct super_block *sb; ++ ++ sbinfo = inode->i_private; ++ sb = sbinfo->si_sb; ++ si_noflush_read_lock(sb); ++ err = dbgaufs_xi_open(sbinfo->si_xigen, file, /*do_fcnt*/0); ++ si_read_unlock(sb); ++ return err; ++} ++ ++static const struct file_operations dbgaufs_xigen_fop = { ++ .open = dbgaufs_xigen_open, ++ .release = dbgaufs_xi_release, ++ .read = dbgaufs_xi_read ++}; ++ ++static int dbgaufs_xigen_init(struct au_sbinfo *sbinfo) ++{ ++ int err; ++ ++ /* ++ * This function is a dynamic '__init' fucntion actually, ++ * so the tiny check for si_rwsem is unnecessary. ++ */ ++ /* AuRwMustWriteLock(&sbinfo->si_rwsem); */ ++ ++ err = -EIO; ++ sbinfo->si_dbgaufs_xigen = debugfs_create_file ++ ("xigen", dbgaufs_mode, sbinfo->si_dbgaufs, sbinfo, ++ &dbgaufs_xigen_fop); ++ if (sbinfo->si_dbgaufs_xigen) ++ err = 0; ++ ++ return err; ++} ++#else ++static int dbgaufs_xigen_init(struct au_sbinfo *sbinfo) ++{ ++ return 0; ++} ++#endif /* CONFIG_AUFS_EXPORT */ ++ ++/* ---------------------------------------------------------------------- */ ++ ++void dbgaufs_si_fin(struct au_sbinfo *sbinfo) ++{ ++ /* ++ * This function is a dynamic '__init' fucntion actually, ++ * so the tiny check for si_rwsem is unnecessary. ++ */ ++ /* AuRwMustWriteLock(&sbinfo->si_rwsem); */ ++ ++ debugfs_remove_recursive(sbinfo->si_dbgaufs); ++ sbinfo->si_dbgaufs = NULL; ++ kobject_put(&sbinfo->si_kobj); ++} ++ ++int dbgaufs_si_init(struct au_sbinfo *sbinfo) ++{ ++ int err; ++ char name[SysaufsSiNameLen]; ++ ++ /* ++ * This function is a dynamic '__init' fucntion actually, ++ * so the tiny check for si_rwsem is unnecessary. ++ */ ++ /* AuRwMustWriteLock(&sbinfo->si_rwsem); */ ++ ++ err = -ENOENT; ++ if (!dbgaufs) { ++ AuErr1("/debug/aufs is uninitialized\n"); ++ goto out; ++ } ++ ++ err = -EIO; ++ sysaufs_name(sbinfo, name); ++ sbinfo->si_dbgaufs = debugfs_create_dir(name, dbgaufs); ++ if (unlikely(!sbinfo->si_dbgaufs)) ++ goto out; ++ kobject_get(&sbinfo->si_kobj); ++ ++ sbinfo->si_dbgaufs_xib = debugfs_create_file ++ ("xib", dbgaufs_mode, sbinfo->si_dbgaufs, sbinfo, ++ &dbgaufs_xib_fop); ++ if (unlikely(!sbinfo->si_dbgaufs_xib)) ++ goto out_dir; ++ ++ err = dbgaufs_xigen_init(sbinfo); ++ if (!err) ++ goto out; /* success */ ++ ++ out_dir: ++ dbgaufs_si_fin(sbinfo); ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void dbgaufs_fin(void) ++{ ++ debugfs_remove(dbgaufs); ++} ++ ++int __init dbgaufs_init(void) ++{ ++ int err; ++ ++ err = -EIO; ++ dbgaufs = debugfs_create_dir(AUFS_NAME, NULL); ++ if (dbgaufs) ++ err = 0; ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/dbgaufs.h linux-2.6.31/fs/aufs/dbgaufs.h +--- linux-2.6.31-vanilla/fs/aufs/dbgaufs.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/dbgaufs.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,79 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * debugfs interface ++ */ ++ ++#ifndef __DBGAUFS_H__ ++#define __DBGAUFS_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++ ++struct super_block; ++struct au_sbinfo; ++ ++#ifdef CONFIG_DEBUG_FS ++/* dbgaufs.c */ ++void dbgaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex); ++void dbgaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex); ++void dbgaufs_si_fin(struct au_sbinfo *sbinfo); ++int dbgaufs_si_init(struct au_sbinfo *sbinfo); ++void dbgaufs_fin(void); ++int __init dbgaufs_init(void); ++ ++#else ++ ++static inline ++void dbgaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ /* empty */ ++} ++ ++static inline ++void dbgaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ /* empty */ ++} ++ ++static inline ++void dbgaufs_si_fin(struct au_sbinfo *sbinfo) ++{ ++ /* empty */ ++} ++ ++static inline ++int dbgaufs_si_init(struct au_sbinfo *sbinfo) ++{ ++ return 0; ++} ++ ++#define dbgaufs_fin() do {} while (0) ++ ++static inline ++int __init dbgaufs_init(void) ++{ ++ return 0; ++} ++#endif /* CONFIG_DEBUG_FS */ ++ ++#endif /* __KERNEL__ */ ++#endif /* __DBGAUFS_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/dcsub.c linux-2.6.31/fs/aufs/dcsub.c +--- linux-2.6.31-vanilla/fs/aufs/dcsub.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/dcsub.c 2009-09-16 13:55:29.000000000 +0200 +@@ -0,0 +1,223 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * sub-routines for dentry cache ++ */ ++ ++#include "aufs.h" ++ ++static void au_dpage_free(struct au_dpage *dpage) ++{ ++ int i; ++ struct dentry **p; ++ ++ p = dpage->dentries; ++ for (i = 0; i < dpage->ndentry; i++) ++ dput(*p++); ++ free_page((unsigned long)dpage->dentries); ++} ++ ++int au_dpages_init(struct au_dcsub_pages *dpages, gfp_t gfp) ++{ ++ int err; ++ void *p; ++ ++ err = -ENOMEM; ++ dpages->dpages = kmalloc(sizeof(*dpages->dpages), gfp); ++ if (unlikely(!dpages->dpages)) ++ goto out; ++ ++ p = (void *)__get_free_page(gfp); ++ if (unlikely(!p)) ++ goto out_dpages; ++ ++ dpages->dpages[0].ndentry = 0; ++ dpages->dpages[0].dentries = p; ++ dpages->ndpage = 1; ++ return 0; /* success */ ++ ++ out_dpages: ++ kfree(dpages->dpages); ++ out: ++ return err; ++} ++ ++void au_dpages_free(struct au_dcsub_pages *dpages) ++{ ++ int i; ++ struct au_dpage *p; ++ ++ p = dpages->dpages; ++ for (i = 0; i < dpages->ndpage; i++) ++ au_dpage_free(p++); ++ kfree(dpages->dpages); ++} ++ ++static int au_dpages_append(struct au_dcsub_pages *dpages, ++ struct dentry *dentry, gfp_t gfp) ++{ ++ int err, sz; ++ struct au_dpage *dpage; ++ void *p; ++ ++ dpage = dpages->dpages + dpages->ndpage - 1; ++ sz = PAGE_SIZE / sizeof(dentry); ++ if (unlikely(dpage->ndentry >= sz)) { ++ AuLabel(new dpage); ++ err = -ENOMEM; ++ sz = dpages->ndpage * sizeof(*dpages->dpages); ++ p = au_kzrealloc(dpages->dpages, sz, ++ sz + sizeof(*dpages->dpages), gfp); ++ if (unlikely(!p)) ++ goto out; ++ ++ dpages->dpages = p; ++ dpage = dpages->dpages + dpages->ndpage; ++ p = (void *)__get_free_page(gfp); ++ if (unlikely(!p)) ++ goto out; ++ ++ dpage->ndentry = 0; ++ dpage->dentries = p; ++ dpages->ndpage++; ++ } ++ ++ dpage->dentries[dpage->ndentry++] = dget(dentry); ++ return 0; /* success */ ++ ++ out: ++ return err; ++} ++ ++int au_dcsub_pages(struct au_dcsub_pages *dpages, struct dentry *root, ++ au_dpages_test test, void *arg) ++{ ++ int err; ++ struct dentry *this_parent = root; ++ struct list_head *next; ++ struct super_block *sb = root->d_sb; ++ ++ err = 0; ++ spin_lock(&dcache_lock); ++ repeat: ++ next = this_parent->d_subdirs.next; ++ resume: ++ if (this_parent->d_sb == sb ++ && !IS_ROOT(this_parent) ++ && atomic_read(&this_parent->d_count) ++ && this_parent->d_inode ++ && (!test || test(this_parent, arg))) { ++ err = au_dpages_append(dpages, this_parent, GFP_ATOMIC); ++ if (unlikely(err)) ++ goto out; ++ } ++ ++ while (next != &this_parent->d_subdirs) { ++ struct list_head *tmp = next; ++ struct dentry *dentry = list_entry(tmp, struct dentry, ++ d_u.d_child); ++ next = tmp->next; ++ if (/*d_unhashed(dentry) || */!dentry->d_inode) ++ continue; ++ if (!list_empty(&dentry->d_subdirs)) { ++ this_parent = dentry; ++ goto repeat; ++ } ++ if (dentry->d_sb == sb ++ && atomic_read(&dentry->d_count) ++ && (!test || test(dentry, arg))) { ++ err = au_dpages_append(dpages, dentry, GFP_ATOMIC); ++ if (unlikely(err)) ++ goto out; ++ } ++ } ++ ++ if (this_parent != root) { ++ next = this_parent->d_u.d_child.next; ++ this_parent = this_parent->d_parent; /* dcache_lock is locked */ ++ goto resume; ++ } ++ out: ++ spin_unlock(&dcache_lock); ++ return err; ++} ++ ++int au_dcsub_pages_rev(struct au_dcsub_pages *dpages, struct dentry *dentry, ++ int do_include, au_dpages_test test, void *arg) ++{ ++ int err; ++ ++ err = 0; ++ spin_lock(&dcache_lock); ++ if (do_include && (!test || test(dentry, arg))) { ++ err = au_dpages_append(dpages, dentry, GFP_ATOMIC); ++ if (unlikely(err)) ++ goto out; ++ } ++ while (!IS_ROOT(dentry)) { ++ dentry = dentry->d_parent; /* dcache_lock is locked */ ++ if (!test || test(dentry, arg)) { ++ err = au_dpages_append(dpages, dentry, GFP_ATOMIC); ++ if (unlikely(err)) ++ break; ++ } ++ } ++ ++ out: ++ spin_unlock(&dcache_lock); ++ ++ return err; ++} ++ ++struct dentry *au_test_subdir(struct dentry *d1, struct dentry *d2) ++{ ++ struct dentry *trap, **dentries; ++ int err, i, j; ++ struct au_dcsub_pages dpages; ++ struct au_dpage *dpage; ++ ++ trap = ERR_PTR(-ENOMEM); ++ err = au_dpages_init(&dpages, GFP_NOFS); ++ if (unlikely(err)) ++ goto out; ++ err = au_dcsub_pages_rev(&dpages, d1, /*do_include*/1, NULL, NULL); ++ if (unlikely(err)) ++ goto out_dpages; ++ ++ trap = d1; ++ for (i = 0; !err && i < dpages.ndpage; i++) { ++ dpage = dpages.dpages + i; ++ dentries = dpage->dentries; ++ for (j = 0; !err && j < dpage->ndentry; j++) { ++ struct dentry *d; ++ ++ d = dentries[j]; ++ err = (d == d2); ++ if (!err) ++ trap = d; ++ } ++ } ++ if (!err) ++ trap = NULL; ++ ++ out_dpages: ++ au_dpages_free(&dpages); ++ out: ++ return trap; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/dcsub.h linux-2.6.31/fs/aufs/dcsub.h +--- linux-2.6.31-vanilla/fs/aufs/dcsub.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/dcsub.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,54 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * sub-routines for dentry cache ++ */ ++ ++#ifndef __AUFS_DCSUB_H__ ++#define __AUFS_DCSUB_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++ ++struct dentry; ++ ++struct au_dpage { ++ int ndentry; ++ struct dentry **dentries; ++}; ++ ++struct au_dcsub_pages { ++ int ndpage; ++ struct au_dpage *dpages; ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++int au_dpages_init(struct au_dcsub_pages *dpages, gfp_t gfp); ++void au_dpages_free(struct au_dcsub_pages *dpages); ++typedef int (*au_dpages_test)(struct dentry *dentry, void *arg); ++int au_dcsub_pages(struct au_dcsub_pages *dpages, struct dentry *root, ++ au_dpages_test test, void *arg); ++int au_dcsub_pages_rev(struct au_dcsub_pages *dpages, struct dentry *dentry, ++ int do_include, au_dpages_test test, void *arg); ++struct dentry *au_test_subdir(struct dentry *d1, struct dentry *d2); ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_DCSUB_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/debug.c linux-2.6.31/fs/aufs/debug.c +--- linux-2.6.31-vanilla/fs/aufs/debug.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/debug.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,431 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * debug print functions ++ */ ++ ++#include ++#include ++#include "aufs.h" ++ ++int aufs_debug; ++MODULE_PARM_DESC(debug, "debug print"); ++module_param_named(debug, aufs_debug, int, S_IRUGO | S_IWUSR | S_IWGRP); ++ ++char *au_plevel = KERN_DEBUG; ++#define dpri(fmt, arg...) do { \ ++ if (au_debug_test()) \ ++ printk("%s" fmt, au_plevel, ##arg); \ ++} while (0) ++ ++/* ---------------------------------------------------------------------- */ ++ ++void au_dpri_whlist(struct au_nhash *whlist) ++{ ++ unsigned long ul, n; ++ struct hlist_head *head; ++ struct au_vdir_wh *tpos; ++ struct hlist_node *pos; ++ ++ n = whlist->nh_num; ++ head = whlist->nh_head; ++ for (ul = 0; ul < n; ul++) { ++ hlist_for_each_entry(tpos, pos, head, wh_hash) ++ dpri("b%d, %.*s, %d\n", ++ tpos->wh_bindex, ++ tpos->wh_str.len, tpos->wh_str.name, ++ tpos->wh_str.len); ++ head++; ++ } ++} ++ ++void au_dpri_vdir(struct au_vdir *vdir) ++{ ++ unsigned long ul; ++ union au_vdir_deblk_p p; ++ unsigned char *o; ++ ++ if (!vdir || IS_ERR(vdir)) { ++ dpri("err %ld\n", PTR_ERR(vdir)); ++ return; ++ } ++ ++ dpri("deblk %u, nblk %lu, deblk %p, last{%lu, %p}, ver %lu\n", ++ vdir->vd_deblk_sz, vdir->vd_nblk, vdir->vd_deblk, ++ vdir->vd_last.ul, vdir->vd_last.p.deblk, vdir->vd_version); ++ for (ul = 0; ul < vdir->vd_nblk; ul++) { ++ p.deblk = vdir->vd_deblk[ul]; ++ o = p.deblk; ++ dpri("[%lu]: %p\n", ul, o); ++ } ++} ++ ++static int do_pri_inode(aufs_bindex_t bindex, struct inode *inode, ++ struct dentry *wh) ++{ ++ char *n = NULL; ++ int l = 0; ++ ++ if (!inode || IS_ERR(inode)) { ++ dpri("i%d: err %ld\n", bindex, PTR_ERR(inode)); ++ return -1; ++ } ++ ++ /* the type of i_blocks depends upon CONFIG_LSF */ ++ BUILD_BUG_ON(sizeof(inode->i_blocks) != sizeof(unsigned long) ++ && sizeof(inode->i_blocks) != sizeof(u64)); ++ if (wh) { ++ n = (void *)wh->d_name.name; ++ l = wh->d_name.len; ++ } ++ ++ dpri("i%d: i%lu, %s, cnt %d, nl %u, 0%o, sz %llu, blk %llu," ++ " ct %lld, np %lu, st 0x%lx, f 0x%x, g %x%s%.*s\n", ++ bindex, ++ inode->i_ino, inode->i_sb ? au_sbtype(inode->i_sb) : "??", ++ atomic_read(&inode->i_count), inode->i_nlink, inode->i_mode, ++ i_size_read(inode), (unsigned long long)inode->i_blocks, ++ (long long)timespec_to_ns(&inode->i_ctime) & 0x0ffff, ++ inode->i_mapping ? inode->i_mapping->nrpages : 0, ++ inode->i_state, inode->i_flags, inode->i_generation, ++ l ? ", wh " : "", l, n); ++ return 0; ++} ++ ++void au_dpri_inode(struct inode *inode) ++{ ++ struct au_iinfo *iinfo; ++ aufs_bindex_t bindex; ++ int err; ++ ++ err = do_pri_inode(-1, inode, NULL); ++ if (err || !au_test_aufs(inode->i_sb)) ++ return; ++ ++ iinfo = au_ii(inode); ++ if (!iinfo) ++ return; ++ dpri("i-1: bstart %d, bend %d, gen %d\n", ++ iinfo->ii_bstart, iinfo->ii_bend, au_iigen(inode)); ++ if (iinfo->ii_bstart < 0) ++ return; ++ for (bindex = iinfo->ii_bstart; bindex <= iinfo->ii_bend; bindex++) ++ do_pri_inode(bindex, iinfo->ii_hinode[0 + bindex].hi_inode, ++ iinfo->ii_hinode[0 + bindex].hi_whdentry); ++} ++ ++static int do_pri_dentry(aufs_bindex_t bindex, struct dentry *dentry) ++{ ++ struct dentry *wh = NULL; ++ ++ if (!dentry || IS_ERR(dentry)) { ++ dpri("d%d: err %ld\n", bindex, PTR_ERR(dentry)); ++ return -1; ++ } ++ /* do not call dget_parent() here */ ++ dpri("d%d: %.*s?/%.*s, %s, cnt %d, flags 0x%x\n", ++ bindex, ++ AuDLNPair(dentry->d_parent), AuDLNPair(dentry), ++ dentry->d_sb ? au_sbtype(dentry->d_sb) : "??", ++ atomic_read(&dentry->d_count), dentry->d_flags); ++ if (bindex >= 0 && dentry->d_inode && au_test_aufs(dentry->d_sb)) { ++ struct au_iinfo *iinfo = au_ii(dentry->d_inode); ++ if (iinfo) ++ wh = iinfo->ii_hinode[0 + bindex].hi_whdentry; ++ } ++ do_pri_inode(bindex, dentry->d_inode, wh); ++ return 0; ++} ++ ++void au_dpri_dentry(struct dentry *dentry) ++{ ++ struct au_dinfo *dinfo; ++ aufs_bindex_t bindex; ++ int err; ++ ++ err = do_pri_dentry(-1, dentry); ++ if (err || !au_test_aufs(dentry->d_sb)) ++ return; ++ ++ dinfo = au_di(dentry); ++ if (!dinfo) ++ return; ++ dpri("d-1: bstart %d, bend %d, bwh %d, bdiropq %d, gen %d\n", ++ dinfo->di_bstart, dinfo->di_bend, ++ dinfo->di_bwh, dinfo->di_bdiropq, au_digen(dentry)); ++ if (dinfo->di_bstart < 0) ++ return; ++ for (bindex = dinfo->di_bstart; bindex <= dinfo->di_bend; bindex++) ++ do_pri_dentry(bindex, dinfo->di_hdentry[0 + bindex].hd_dentry); ++} ++ ++static int do_pri_file(aufs_bindex_t bindex, struct file *file) ++{ ++ char a[32]; ++ ++ if (!file || IS_ERR(file)) { ++ dpri("f%d: err %ld\n", bindex, PTR_ERR(file)); ++ return -1; ++ } ++ a[0] = 0; ++ if (bindex < 0 ++ && file->f_dentry ++ && au_test_aufs(file->f_dentry->d_sb) ++ && au_fi(file)) ++ snprintf(a, sizeof(a), ", mmapped %d", au_test_mmapped(file)); ++ dpri("f%d: mode 0x%x, flags 0%o, cnt %ld, pos %llu%s\n", ++ bindex, file->f_mode, file->f_flags, (long)file_count(file), ++ file->f_pos, a); ++ if (file->f_dentry) ++ do_pri_dentry(bindex, file->f_dentry); ++ return 0; ++} ++ ++void au_dpri_file(struct file *file) ++{ ++ struct au_finfo *finfo; ++ aufs_bindex_t bindex; ++ int err; ++ ++ err = do_pri_file(-1, file); ++ if (err || !file->f_dentry || !au_test_aufs(file->f_dentry->d_sb)) ++ return; ++ ++ finfo = au_fi(file); ++ if (!finfo) ++ return; ++ if (finfo->fi_bstart < 0) ++ return; ++ for (bindex = finfo->fi_bstart; bindex <= finfo->fi_bend; bindex++) { ++ struct au_hfile *hf; ++ ++ hf = finfo->fi_hfile + bindex; ++ do_pri_file(bindex, hf ? hf->hf_file : NULL); ++ } ++} ++ ++static int do_pri_br(aufs_bindex_t bindex, struct au_branch *br) ++{ ++ struct vfsmount *mnt; ++ struct super_block *sb; ++ ++ if (!br || IS_ERR(br)) ++ goto out; ++ mnt = br->br_mnt; ++ if (!mnt || IS_ERR(mnt)) ++ goto out; ++ sb = mnt->mnt_sb; ++ if (!sb || IS_ERR(sb)) ++ goto out; ++ ++ dpri("s%d: {perm 0x%x, cnt %d, wbr %p}, " ++ "%s, dev 0x%02x%02x, flags 0x%lx, cnt(BIAS) %d, active %d, " ++ "xino %d\n", ++ bindex, br->br_perm, atomic_read(&br->br_count), br->br_wbr, ++ au_sbtype(sb), MAJOR(sb->s_dev), MINOR(sb->s_dev), ++ sb->s_flags, sb->s_count - S_BIAS, ++ atomic_read(&sb->s_active), !!br->br_xino.xi_file); ++ return 0; ++ ++ out: ++ dpri("s%d: err %ld\n", bindex, PTR_ERR(br)); ++ return -1; ++} ++ ++void au_dpri_sb(struct super_block *sb) ++{ ++ struct au_sbinfo *sbinfo; ++ aufs_bindex_t bindex; ++ int err; ++ /* to reuduce stack size */ ++ struct { ++ struct vfsmount mnt; ++ struct au_branch fake; ++ } *a; ++ ++ /* this function can be called from magic sysrq */ ++ a = kzalloc(sizeof(*a), GFP_ATOMIC); ++ if (unlikely(!a)) { ++ dpri("no memory\n"); ++ return; ++ } ++ ++ a->mnt.mnt_sb = sb; ++ a->fake.br_perm = 0; ++ a->fake.br_mnt = &a->mnt; ++ a->fake.br_xino.xi_file = NULL; ++ atomic_set(&a->fake.br_count, 0); ++ smp_mb(); /* atomic_set */ ++ err = do_pri_br(-1, &a->fake); ++ kfree(a); ++ dpri("dev 0x%x\n", sb->s_dev); ++ if (err || !au_test_aufs(sb)) ++ return; ++ ++ sbinfo = au_sbi(sb); ++ if (!sbinfo) ++ return; ++ dpri("nw %d, gen %u, kobj %d\n", ++ atomic_read(&sbinfo->si_nowait.nw_len), sbinfo->si_generation, ++ atomic_read(&sbinfo->si_kobj.kref.refcount)); ++ for (bindex = 0; bindex <= sbinfo->si_bend; bindex++) ++ do_pri_br(bindex, sbinfo->si_branch[0 + bindex]); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void au_dbg_sleep_jiffy(int jiffy) ++{ ++ while (jiffy) ++ jiffy = schedule_timeout_uninterruptible(jiffy); ++} ++ ++void au_dbg_iattr(struct iattr *ia) ++{ ++#define AuBit(name) if (ia->ia_valid & ATTR_ ## name) \ ++ dpri(#name "\n") ++ AuBit(MODE); ++ AuBit(UID); ++ AuBit(GID); ++ AuBit(SIZE); ++ AuBit(ATIME); ++ AuBit(MTIME); ++ AuBit(CTIME); ++ AuBit(ATIME_SET); ++ AuBit(MTIME_SET); ++ AuBit(FORCE); ++ AuBit(ATTR_FLAG); ++ AuBit(KILL_SUID); ++ AuBit(KILL_SGID); ++ AuBit(FILE); ++ AuBit(KILL_PRIV); ++ AuBit(OPEN); ++ AuBit(TIMES_SET); ++#undef AuBit ++ dpri("ia_file %p\n", ia->ia_file); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void au_dbg_verify_dir_parent(struct dentry *dentry, unsigned int sigen) ++{ ++ struct dentry *parent; ++ ++ parent = dget_parent(dentry); ++ AuDebugOn(!S_ISDIR(dentry->d_inode->i_mode) ++ || IS_ROOT(dentry) ++ || au_digen(parent) != sigen); ++ dput(parent); ++} ++ ++void au_dbg_verify_nondir_parent(struct dentry *dentry, unsigned int sigen) ++{ ++ struct dentry *parent; ++ ++ parent = dget_parent(dentry); ++ AuDebugOn(S_ISDIR(dentry->d_inode->i_mode) ++ || au_digen(parent) != sigen); ++ dput(parent); ++} ++ ++void au_dbg_verify_gen(struct dentry *parent, unsigned int sigen) ++{ ++ int err, i, j; ++ struct au_dcsub_pages dpages; ++ struct au_dpage *dpage; ++ struct dentry **dentries; ++ ++ err = au_dpages_init(&dpages, GFP_NOFS); ++ AuDebugOn(err); ++ err = au_dcsub_pages_rev(&dpages, parent, /*do_include*/1, NULL, NULL); ++ AuDebugOn(err); ++ for (i = dpages.ndpage - 1; !err && i >= 0; i--) { ++ dpage = dpages.dpages + i; ++ dentries = dpage->dentries; ++ for (j = dpage->ndentry - 1; !err && j >= 0; j--) ++ AuDebugOn(au_digen(dentries[j]) != sigen); ++ } ++ au_dpages_free(&dpages); ++} ++ ++void au_dbg_verify_hf(struct au_finfo *finfo) ++{ ++ struct au_hfile *hf; ++ aufs_bindex_t bend, bindex; ++ ++ if (finfo->fi_bstart >= 0) { ++ bend = finfo->fi_bend; ++ for (bindex = finfo->fi_bstart; bindex <= bend; bindex++) { ++ hf = finfo->fi_hfile + bindex; ++ AuDebugOn(hf->hf_file || hf->hf_br); ++ } ++ } ++} ++ ++void au_dbg_verify_kthread(void) ++{ ++ if (au_test_wkq(current)) { ++ au_dbg_blocked(); ++ BUG(); ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void au_debug_sbinfo_init(struct au_sbinfo *sbinfo __maybe_unused) ++{ ++#ifdef AuForceNoPlink ++ au_opt_clr(sbinfo->si_mntflags, PLINK); ++#endif ++#ifdef AuForceNoXino ++ au_opt_clr(sbinfo->si_mntflags, XINO); ++#endif ++#ifdef AuForceNoRefrof ++ au_opt_clr(sbinfo->si_mntflags, REFROF); ++#endif ++#ifdef AuForceHinotify ++ au_opt_set_udba(sbinfo->si_mntflags, UDBA_HINOTIFY); ++#endif ++#ifdef AuForceRd0 ++ sbinfo->si_rdblk = 0; ++ sbinfo->si_rdhash = 0; ++#endif ++} ++ ++int __init au_debug_init(void) ++{ ++ aufs_bindex_t bindex; ++ struct au_vdir_destr destr; ++ ++ bindex = -1; ++ AuDebugOn(bindex >= 0); ++ ++ destr.len = -1; ++ AuDebugOn(destr.len < NAME_MAX); ++ ++#ifdef CONFIG_4KSTACKS ++ AuWarn("CONFIG_4KSTACKS is defined.\n"); ++#endif ++ ++#ifdef AuForceNoBrs ++ sysaufs_brs = 0; ++#endif ++ ++ return 0; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/debug.h linux-2.6.31/fs/aufs/debug.h +--- linux-2.6.31-vanilla/fs/aufs/debug.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/debug.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,263 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * debug print functions ++ */ ++ ++#ifndef __AUFS_DEBUG_H__ ++#define __AUFS_DEBUG_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++/* #include */ ++#include ++/* #include */ ++#include ++/* #include */ ++/* #include */ ++#include ++#include ++ ++#include ++ ++#ifdef CONFIG_AUFS_DEBUG ++#define AuDebugOn(a) BUG_ON(a) ++ ++/* module parameter */ ++extern int aufs_debug; ++static inline void au_debug(int n) ++{ ++ aufs_debug = n; ++ smp_mb(); ++} ++ ++static inline int au_debug_test(void) ++{ ++ return aufs_debug; ++} ++#else ++#define AuDebugOn(a) do {} while (0) ++#define au_debug() do {} while (0) ++static inline int au_debug_test(void) ++{ ++ return 0; ++} ++#endif /* CONFIG_AUFS_DEBUG */ ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* debug print */ ++ ++#define AuDpri(lvl, fmt, arg...) \ ++ printk(lvl AUFS_NAME " %s:%d:%s[%d]: " fmt, \ ++ __func__, __LINE__, current->comm, current->pid, ##arg) ++#define AuDbg(fmt, arg...) do { \ ++ if (au_debug_test()) \ ++ AuDpri(KERN_DEBUG, "DEBUG: " fmt, ##arg); \ ++} while (0) ++#define AuLabel(l) AuDbg(#l "\n") ++#define AuInfo(fmt, arg...) AuDpri(KERN_INFO, fmt, ##arg) ++#define AuWarn(fmt, arg...) AuDpri(KERN_WARNING, fmt, ##arg) ++#define AuErr(fmt, arg...) AuDpri(KERN_ERR, fmt, ##arg) ++#define AuIOErr(fmt, arg...) AuErr("I/O Error, " fmt, ##arg) ++#define AuWarn1(fmt, arg...) do { \ ++ static unsigned char _c; \ ++ if (!_c++) \ ++ AuWarn(fmt, ##arg); \ ++} while (0) ++ ++#define AuErr1(fmt, arg...) do { \ ++ static unsigned char _c; \ ++ if (!_c++) \ ++ AuErr(fmt, ##arg); \ ++} while (0) ++ ++#define AuIOErr1(fmt, arg...) do { \ ++ static unsigned char _c; \ ++ if (!_c++) \ ++ AuIOErr(fmt, ##arg); \ ++} while (0) ++ ++#define AuUnsupportMsg "This operation is not supported." \ ++ " Please report this application to aufs-users ML." ++#define AuUnsupport(fmt, args...) do { \ ++ AuErr(AuUnsupportMsg "\n" fmt, ##args); \ ++ dump_stack(); \ ++} while (0) ++ ++#define AuTraceErr(e) do { \ ++ if (unlikely((e) < 0)) \ ++ AuDbg("err %d\n", (int)(e)); \ ++} while (0) ++ ++#define AuTraceErrPtr(p) do { \ ++ if (IS_ERR(p)) \ ++ AuDbg("err %ld\n", PTR_ERR(p)); \ ++} while (0) ++ ++/* dirty macros for debug print, use with "%.*s" and caution */ ++#define AuLNPair(qstr) (qstr)->len, (qstr)->name ++#define AuDLNPair(d) AuLNPair(&(d)->d_name) ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct au_sbinfo; ++struct au_finfo; ++struct dentry; ++#ifdef CONFIG_AUFS_DEBUG ++extern char *au_plevel; ++struct au_nhash; ++void au_dpri_whlist(struct au_nhash *whlist); ++struct au_vdir; ++void au_dpri_vdir(struct au_vdir *vdir); ++struct inode; ++void au_dpri_inode(struct inode *inode); ++void au_dpri_dentry(struct dentry *dentry); ++struct file; ++void au_dpri_file(struct file *filp); ++struct super_block; ++void au_dpri_sb(struct super_block *sb); ++ ++void au_dbg_sleep_jiffy(int jiffy); ++struct iattr; ++void au_dbg_iattr(struct iattr *ia); ++ ++void au_dbg_verify_dir_parent(struct dentry *dentry, unsigned int sigen); ++void au_dbg_verify_nondir_parent(struct dentry *dentry, unsigned int sigen); ++void au_dbg_verify_gen(struct dentry *parent, unsigned int sigen); ++void au_dbg_verify_hf(struct au_finfo *finfo); ++void au_dbg_verify_kthread(void); ++ ++int __init au_debug_init(void); ++void au_debug_sbinfo_init(struct au_sbinfo *sbinfo); ++#define AuDbgWhlist(w) do { \ ++ AuDbg(#w "\n"); \ ++ au_dpri_whlist(w); \ ++} while (0) ++ ++#define AuDbgVdir(v) do { \ ++ AuDbg(#v "\n"); \ ++ au_dpri_vdir(v); \ ++} while (0) ++ ++#define AuDbgInode(i) do { \ ++ AuDbg(#i "\n"); \ ++ au_dpri_inode(i); \ ++} while (0) ++ ++#define AuDbgDentry(d) do { \ ++ AuDbg(#d "\n"); \ ++ au_dpri_dentry(d); \ ++} while (0) ++ ++#define AuDbgFile(f) do { \ ++ AuDbg(#f "\n"); \ ++ au_dpri_file(f); \ ++} while (0) ++ ++#define AuDbgSb(sb) do { \ ++ AuDbg(#sb "\n"); \ ++ au_dpri_sb(sb); \ ++} while (0) ++ ++#define AuDbgSleep(sec) do { \ ++ AuDbg("sleep %d sec\n", sec); \ ++ ssleep(sec); \ ++} while (0) ++ ++#define AuDbgSleepJiffy(jiffy) do { \ ++ AuDbg("sleep %d jiffies\n", jiffy); \ ++ au_dbg_sleep_jiffy(jiffy); \ ++} while (0) ++ ++#define AuDbgIAttr(ia) do { \ ++ AuDbg("ia_valid 0x%x\n", (ia)->ia_valid); \ ++ au_dbg_iattr(ia); \ ++} while (0) ++#else ++static inline void au_dbg_verify_dir_parent(struct dentry *dentry, ++ unsigned int sigen) ++{ ++ /* empty */ ++} ++static inline void au_dbg_verify_nondir_parent(struct dentry *dentry, ++ unsigned int sigen) ++{ ++ /* empty */ ++} ++static inline void au_dbg_verify_gen(struct dentry *parent, unsigned int sigen) ++{ ++ /* empty */ ++} ++static inline void au_dbg_verify_hf(struct au_finfo *finfo) ++{ ++ /* empty */ ++} ++static inline void au_dbg_verify_kthread(void) ++{ ++ /* empty */ ++} ++ ++static inline int au_debug_init(void) ++{ ++ return 0; ++} ++static inline void au_debug_sbinfo_init(struct au_sbinfo *sbinfo) ++{ ++ /* empty */ ++} ++#define AuDbgWhlist(w) do {} while (0) ++#define AuDbgVdir(v) do {} while (0) ++#define AuDbgInode(i) do {} while (0) ++#define AuDbgDentry(d) do {} while (0) ++#define AuDbgFile(f) do {} while (0) ++#define AuDbgSb(sb) do {} while (0) ++#define AuDbgSleep(sec) do {} while (0) ++#define AuDbgSleepJiffy(jiffy) do {} while (0) ++#define AuDbgIAttr(ia) do {} while (0) ++#endif /* CONFIG_AUFS_DEBUG */ ++ ++/* ---------------------------------------------------------------------- */ ++ ++#ifdef CONFIG_AUFS_MAGIC_SYSRQ ++int __init au_sysrq_init(void); ++void au_sysrq_fin(void); ++ ++#ifdef CONFIG_HW_CONSOLE ++#define au_dbg_blocked() do { \ ++ WARN_ON(1); \ ++ handle_sysrq('w', vc_cons[fg_console].d->vc_tty); \ ++} while (0) ++#else ++#define au_dbg_blocked() do {} while (0) ++#endif ++ ++#else ++static inline int au_sysrq_init(void) ++{ ++ return 0; ++} ++#define au_sysrq_fin() do {} while (0) ++#define au_dbg_blocked() do {} while (0) ++#endif /* CONFIG_AUFS_MAGIC_SYSRQ */ ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_DEBUG_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/dentry.c linux-2.6.31/fs/aufs/dentry.c +--- linux-2.6.31-vanilla/fs/aufs/dentry.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/dentry.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,879 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * lookup and dentry operations ++ */ ++ ++#include ++#include "aufs.h" ++ ++static void au_h_nd(struct nameidata *h_nd, struct nameidata *nd) ++{ ++ if (nd) { ++ *h_nd = *nd; ++ ++ /* ++ * gave up supporting LOOKUP_CREATE/OPEN for lower fs, ++ * due to whiteout and branch permission. ++ */ ++ h_nd->flags &= ~(/*LOOKUP_PARENT |*/ LOOKUP_OPEN | LOOKUP_CREATE ++ | LOOKUP_FOLLOW); ++ /* unnecessary? */ ++ h_nd->intent.open.file = NULL; ++ } else ++ memset(h_nd, 0, sizeof(*h_nd)); ++} ++ ++struct au_lkup_one_args { ++ struct dentry **errp; ++ struct qstr *name; ++ struct dentry *h_parent; ++ struct au_branch *br; ++ struct nameidata *nd; ++}; ++ ++struct dentry *au_lkup_one(struct qstr *name, struct dentry *h_parent, ++ struct au_branch *br, struct nameidata *nd) ++{ ++ struct dentry *h_dentry; ++ int err; ++ struct nameidata h_nd; ++ ++ if (au_test_fs_null_nd(h_parent->d_sb)) ++ return vfsub_lookup_one_len(name->name, h_parent, name->len); ++ ++ au_h_nd(&h_nd, nd); ++ h_nd.path.dentry = h_parent; ++ h_nd.path.mnt = br->br_mnt; ++ ++ err = __lookup_one_len(name->name, &h_nd.last, NULL, name->len); ++ h_dentry = ERR_PTR(err); ++ if (!err) { ++ path_get(&h_nd.path); ++ h_dentry = vfsub_lookup_hash(&h_nd); ++ path_put(&h_nd.path); ++ } ++ ++ return h_dentry; ++} ++ ++static void au_call_lkup_one(void *args) ++{ ++ struct au_lkup_one_args *a = args; ++ *a->errp = au_lkup_one(a->name, a->h_parent, a->br, a->nd); ++} ++ ++#define AuLkup_ALLOW_NEG 1 ++#define au_ftest_lkup(flags, name) ((flags) & AuLkup_##name) ++#define au_fset_lkup(flags, name) { (flags) |= AuLkup_##name; } ++#define au_fclr_lkup(flags, name) { (flags) &= ~AuLkup_##name; } ++ ++struct au_do_lookup_args { ++ unsigned int flags; ++ mode_t type; ++ struct nameidata *nd; ++}; ++ ++/* ++ * returns positive/negative dentry, NULL or an error. ++ * NULL means whiteout-ed or not-found. ++ */ ++static struct dentry* ++au_do_lookup(struct dentry *h_parent, struct dentry *dentry, ++ aufs_bindex_t bindex, struct qstr *wh_name, ++ struct au_do_lookup_args *args) ++{ ++ struct dentry *h_dentry; ++ struct inode *h_inode, *inode; ++ struct qstr *name; ++ struct au_branch *br; ++ int wh_found, opq; ++ unsigned char wh_able; ++ const unsigned char allow_neg = !!au_ftest_lkup(args->flags, ALLOW_NEG); ++ ++ name = &dentry->d_name; ++ wh_found = 0; ++ br = au_sbr(dentry->d_sb, bindex); ++ wh_able = !!au_br_whable(br->br_perm); ++ if (wh_able) ++ wh_found = au_wh_test(h_parent, wh_name, br, /*try_sio*/0); ++ h_dentry = ERR_PTR(wh_found); ++ if (!wh_found) ++ goto real_lookup; ++ if (unlikely(wh_found < 0)) ++ goto out; ++ ++ /* We found a whiteout */ ++ /* au_set_dbend(dentry, bindex); */ ++ au_set_dbwh(dentry, bindex); ++ if (!allow_neg) ++ return NULL; /* success */ ++ ++ real_lookup: ++ h_dentry = au_lkup_one(name, h_parent, br, args->nd); ++ if (IS_ERR(h_dentry)) ++ goto out; ++ ++ h_inode = h_dentry->d_inode; ++ if (!h_inode) { ++ if (!allow_neg) ++ goto out_neg; ++ } else if (wh_found ++ || (args->type && args->type != (h_inode->i_mode & S_IFMT))) ++ goto out_neg; ++ ++ if (au_dbend(dentry) <= bindex) ++ au_set_dbend(dentry, bindex); ++ if (au_dbstart(dentry) < 0 || bindex < au_dbstart(dentry)) ++ au_set_dbstart(dentry, bindex); ++ au_set_h_dptr(dentry, bindex, h_dentry); ++ ++ inode = dentry->d_inode; ++ if (!h_inode || !S_ISDIR(h_inode->i_mode) || !wh_able ++ || (inode && !S_ISDIR(inode->i_mode))) ++ goto out; /* success */ ++ ++ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); ++ opq = au_diropq_test(h_dentry, br); ++ mutex_unlock(&h_inode->i_mutex); ++ if (opq > 0) ++ au_set_dbdiropq(dentry, bindex); ++ else if (unlikely(opq < 0)) { ++ au_set_h_dptr(dentry, bindex, NULL); ++ h_dentry = ERR_PTR(opq); ++ } ++ goto out; ++ ++ out_neg: ++ dput(h_dentry); ++ h_dentry = NULL; ++ out: ++ return h_dentry; ++} ++ ++static int au_test_shwh(struct super_block *sb, const struct qstr *name) ++{ ++ if (unlikely(!au_opt_test(au_mntflags(sb), SHWH) ++ && !strncmp(name->name, AUFS_WH_PFX, AUFS_WH_PFX_LEN))) ++ return -EPERM; ++ return 0; ++} ++ ++/* ++ * returns the number of lower positive dentries, ++ * otherwise an error. ++ * can be called at unlinking with @type is zero. ++ */ ++int au_lkup_dentry(struct dentry *dentry, aufs_bindex_t bstart, mode_t type, ++ struct nameidata *nd) ++{ ++ int npositive, err; ++ aufs_bindex_t bindex, btail, bdiropq; ++ unsigned char isdir; ++ struct qstr whname; ++ struct au_do_lookup_args args = { ++ .flags = 0, ++ .type = type, ++ .nd = nd ++ }; ++ const struct qstr *name = &dentry->d_name; ++ struct dentry *parent; ++ struct inode *inode; ++ ++ parent = dget_parent(dentry); ++ err = au_test_shwh(dentry->d_sb, name); ++ if (unlikely(err)) ++ goto out; ++ ++ err = au_wh_name_alloc(&whname, name); ++ if (unlikely(err)) ++ goto out; ++ ++ inode = dentry->d_inode; ++ isdir = !!(inode && S_ISDIR(inode->i_mode)); ++ if (!type) ++ au_fset_lkup(args.flags, ALLOW_NEG); ++ ++ npositive = 0; ++ btail = au_dbtaildir(parent); ++ for (bindex = bstart; bindex <= btail; bindex++) { ++ struct dentry *h_parent, *h_dentry; ++ struct inode *h_inode, *h_dir; ++ ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (h_dentry) { ++ if (h_dentry->d_inode) ++ npositive++; ++ if (type != S_IFDIR) ++ break; ++ continue; ++ } ++ h_parent = au_h_dptr(parent, bindex); ++ if (!h_parent) ++ continue; ++ h_dir = h_parent->d_inode; ++ if (!h_dir || !S_ISDIR(h_dir->i_mode)) ++ continue; ++ ++ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT); ++ h_dentry = au_do_lookup(h_parent, dentry, bindex, &whname, ++ &args); ++ mutex_unlock(&h_dir->i_mutex); ++ err = PTR_ERR(h_dentry); ++ if (IS_ERR(h_dentry)) ++ goto out_wh; ++ au_fclr_lkup(args.flags, ALLOW_NEG); ++ ++ if (au_dbwh(dentry) >= 0) ++ break; ++ if (!h_dentry) ++ continue; ++ h_inode = h_dentry->d_inode; ++ if (!h_inode) ++ continue; ++ npositive++; ++ if (!args.type) ++ args.type = h_inode->i_mode & S_IFMT; ++ if (args.type != S_IFDIR) ++ break; ++ else if (isdir) { ++ /* the type of lower may be different */ ++ bdiropq = au_dbdiropq(dentry); ++ if (bdiropq >= 0 && bdiropq <= bindex) ++ break; ++ } ++ } ++ ++ if (npositive) { ++ AuLabel(positive); ++ au_update_dbstart(dentry); ++ } ++ err = npositive; ++ if (unlikely(!au_opt_test(au_mntflags(dentry->d_sb), UDBA_NONE) ++ && au_dbstart(dentry) < 0)) ++ /* both of real entry and whiteout found */ ++ err = -EIO; ++ ++ out_wh: ++ kfree(whname.name); ++ out: ++ dput(parent); ++ return err; ++} ++ ++struct dentry *au_sio_lkup_one(struct qstr *name, struct dentry *parent, ++ struct au_branch *br) ++{ ++ struct dentry *dentry; ++ int wkq_err; ++ ++ if (!au_test_h_perm_sio(parent->d_inode, MAY_EXEC)) ++ dentry = au_lkup_one(name, parent, br, /*nd*/NULL); ++ else { ++ struct au_lkup_one_args args = { ++ .errp = &dentry, ++ .name = name, ++ .h_parent = parent, ++ .br = br, ++ .nd = NULL ++ }; ++ ++ wkq_err = au_wkq_wait(au_call_lkup_one, &args); ++ if (unlikely(wkq_err)) ++ dentry = ERR_PTR(wkq_err); ++ } ++ ++ return dentry; ++} ++ ++/* ++ * lookup @dentry on @bindex which should be negative. ++ */ ++int au_lkup_neg(struct dentry *dentry, aufs_bindex_t bindex) ++{ ++ int err; ++ struct dentry *parent, *h_parent, *h_dentry; ++ struct qstr *name; ++ ++ name = &dentry->d_name; ++ parent = dget_parent(dentry); ++ h_parent = au_h_dptr(parent, bindex); ++ h_dentry = au_sio_lkup_one(name, h_parent, ++ au_sbr(dentry->d_sb, bindex)); ++ err = PTR_ERR(h_dentry); ++ if (IS_ERR(h_dentry)) ++ goto out; ++ if (unlikely(h_dentry->d_inode)) { ++ err = -EIO; ++ AuIOErr("b%d %.*s should be negative.\n", ++ bindex, AuDLNPair(h_dentry)); ++ dput(h_dentry); ++ goto out; ++ } ++ ++ if (bindex < au_dbstart(dentry)) ++ au_set_dbstart(dentry, bindex); ++ if (au_dbend(dentry) < bindex) ++ au_set_dbend(dentry, bindex); ++ au_set_h_dptr(dentry, bindex, h_dentry); ++ err = 0; ++ ++ out: ++ dput(parent); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* subset of struct inode */ ++struct au_iattr { ++ unsigned long i_ino; ++ /* unsigned int i_nlink; */ ++ uid_t i_uid; ++ gid_t i_gid; ++ u64 i_version; ++/* ++ loff_t i_size; ++ blkcnt_t i_blocks; ++*/ ++ umode_t i_mode; ++}; ++ ++static void au_iattr_save(struct au_iattr *ia, struct inode *h_inode) ++{ ++ ia->i_ino = h_inode->i_ino; ++ /* ia->i_nlink = h_inode->i_nlink; */ ++ ia->i_uid = h_inode->i_uid; ++ ia->i_gid = h_inode->i_gid; ++ ia->i_version = h_inode->i_version; ++/* ++ ia->i_size = h_inode->i_size; ++ ia->i_blocks = h_inode->i_blocks; ++*/ ++ ia->i_mode = (h_inode->i_mode & S_IFMT); ++} ++ ++static int au_iattr_test(struct au_iattr *ia, struct inode *h_inode) ++{ ++ return ia->i_ino != h_inode->i_ino ++ /* || ia->i_nlink != h_inode->i_nlink */ ++ || ia->i_uid != h_inode->i_uid ++ || ia->i_gid != h_inode->i_gid ++ || ia->i_version != h_inode->i_version ++/* ++ || ia->i_size != h_inode->i_size ++ || ia->i_blocks != h_inode->i_blocks ++*/ ++ || ia->i_mode != (h_inode->i_mode & S_IFMT); ++} ++ ++static int au_h_verify_dentry(struct dentry *h_dentry, struct dentry *h_parent, ++ struct au_branch *br) ++{ ++ int err; ++ struct au_iattr ia; ++ struct inode *h_inode; ++ struct dentry *h_d; ++ struct super_block *h_sb; ++ ++ err = 0; ++ memset(&ia, -1, sizeof(ia)); ++ h_sb = h_dentry->d_sb; ++ h_inode = h_dentry->d_inode; ++ if (h_inode) ++ au_iattr_save(&ia, h_inode); ++ else if (au_test_nfs(h_sb) || au_test_fuse(h_sb)) ++ /* nfs d_revalidate may return 0 for negative dentry */ ++ /* fuse d_revalidate always return 0 for negative dentry */ ++ goto out; ++ ++ /* main purpose is namei.c:cached_lookup() and d_revalidate */ ++ h_d = au_lkup_one(&h_dentry->d_name, h_parent, br, /*nd*/NULL); ++ err = PTR_ERR(h_d); ++ if (IS_ERR(h_d)) ++ goto out; ++ ++ err = 0; ++ if (unlikely(h_d != h_dentry ++ || h_d->d_inode != h_inode ++ || (h_inode && au_iattr_test(&ia, h_inode)))) ++ err = au_busy_or_stale(); ++ dput(h_d); ++ ++ out: ++ AuTraceErr(err); ++ return err; ++} ++ ++int au_h_verify(struct dentry *h_dentry, unsigned int udba, struct inode *h_dir, ++ struct dentry *h_parent, struct au_branch *br) ++{ ++ int err; ++ ++ err = 0; ++ if (udba == AuOpt_UDBA_REVAL) { ++ IMustLock(h_dir); ++ err = (h_dentry->d_parent->d_inode != h_dir); ++ } else if (udba == AuOpt_UDBA_HINOTIFY) ++ err = au_h_verify_dentry(h_dentry, h_parent, br); ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static void au_do_refresh_hdentry(struct au_hdentry *p, struct au_dinfo *dinfo, ++ struct dentry *parent) ++{ ++ struct dentry *h_d, *h_dp; ++ struct au_hdentry tmp, *q; ++ struct super_block *sb; ++ aufs_bindex_t new_bindex, bindex, bend, bwh, bdiropq; ++ ++ AuRwMustWriteLock(&dinfo->di_rwsem); ++ ++ bend = dinfo->di_bend; ++ bwh = dinfo->di_bwh; ++ bdiropq = dinfo->di_bdiropq; ++ for (bindex = dinfo->di_bstart; bindex <= bend; bindex++, p++) { ++ h_d = p->hd_dentry; ++ if (!h_d) ++ continue; ++ ++ h_dp = dget_parent(h_d); ++ if (h_dp == au_h_dptr(parent, bindex)) { ++ dput(h_dp); ++ continue; ++ } ++ ++ new_bindex = au_find_dbindex(parent, h_dp); ++ dput(h_dp); ++ if (dinfo->di_bwh == bindex) ++ bwh = new_bindex; ++ if (dinfo->di_bdiropq == bindex) ++ bdiropq = new_bindex; ++ if (new_bindex < 0) { ++ au_hdput(p); ++ p->hd_dentry = NULL; ++ continue; ++ } ++ ++ /* swap two lower dentries, and loop again */ ++ q = dinfo->di_hdentry + new_bindex; ++ tmp = *q; ++ *q = *p; ++ *p = tmp; ++ if (tmp.hd_dentry) { ++ bindex--; ++ p--; ++ } ++ } ++ ++ sb = parent->d_sb; ++ dinfo->di_bwh = -1; ++ if (bwh >= 0 && bwh <= au_sbend(sb) && au_sbr_whable(sb, bwh)) ++ dinfo->di_bwh = bwh; ++ ++ dinfo->di_bdiropq = -1; ++ if (bdiropq >= 0 ++ && bdiropq <= au_sbend(sb) ++ && au_sbr_whable(sb, bdiropq)) ++ dinfo->di_bdiropq = bdiropq; ++ ++ bend = au_dbend(parent); ++ p = dinfo->di_hdentry; ++ for (bindex = 0; bindex <= bend; bindex++, p++) ++ if (p->hd_dentry) { ++ dinfo->di_bstart = bindex; ++ break; ++ } ++ ++ p = dinfo->di_hdentry + bend; ++ for (bindex = bend; bindex >= 0; bindex--, p--) ++ if (p->hd_dentry) { ++ dinfo->di_bend = bindex; ++ break; ++ } ++} ++ ++/* ++ * returns the number of found lower positive dentries, ++ * otherwise an error. ++ */ ++int au_refresh_hdentry(struct dentry *dentry, mode_t type) ++{ ++ int npositive, err; ++ unsigned int sigen; ++ aufs_bindex_t bstart; ++ struct au_dinfo *dinfo; ++ struct super_block *sb; ++ struct dentry *parent; ++ ++ DiMustWriteLock(dentry); ++ ++ sb = dentry->d_sb; ++ AuDebugOn(IS_ROOT(dentry)); ++ sigen = au_sigen(sb); ++ parent = dget_parent(dentry); ++ AuDebugOn(au_digen(parent) != sigen ++ || au_iigen(parent->d_inode) != sigen); ++ ++ dinfo = au_di(dentry); ++ err = au_di_realloc(dinfo, au_sbend(sb) + 1); ++ npositive = err; ++ if (unlikely(err)) ++ goto out; ++ au_do_refresh_hdentry(dinfo->di_hdentry + dinfo->di_bstart, dinfo, ++ parent); ++ ++ npositive = 0; ++ bstart = au_dbstart(parent); ++ if (type != S_IFDIR && dinfo->di_bstart == bstart) ++ goto out_dgen; /* success */ ++ ++ npositive = au_lkup_dentry(dentry, bstart, type, /*nd*/NULL); ++ if (npositive < 0) ++ goto out; ++ if (dinfo->di_bwh >= 0 && dinfo->di_bwh <= dinfo->di_bstart) ++ d_drop(dentry); ++ ++ out_dgen: ++ au_update_digen(dentry); ++ out: ++ dput(parent); ++ AuTraceErr(npositive); ++ return npositive; ++} ++ ++static noinline_for_stack ++int au_do_h_d_reval(struct dentry *h_dentry, struct nameidata *nd, ++ struct dentry *dentry, aufs_bindex_t bindex) ++{ ++ int err, valid; ++ int (*reval)(struct dentry *, struct nameidata *); ++ ++ err = 0; ++ reval = NULL; ++ if (h_dentry->d_op) ++ reval = h_dentry->d_op->d_revalidate; ++ if (!reval) ++ goto out; ++ ++ AuDbg("b%d\n", bindex); ++ if (au_test_fs_null_nd(h_dentry->d_sb)) ++ /* it may return tri-state */ ++ valid = reval(h_dentry, NULL); ++ else { ++ struct nameidata h_nd; ++ int locked; ++ struct dentry *parent; ++ ++ au_h_nd(&h_nd, nd); ++ parent = nd->path.dentry; ++ locked = (nd && nd->path.dentry != dentry); ++ if (locked) ++ di_read_lock_parent(parent, AuLock_IR); ++ BUG_ON(bindex > au_dbend(parent)); ++ h_nd.path.dentry = au_h_dptr(parent, bindex); ++ BUG_ON(!h_nd.path.dentry); ++ h_nd.path.mnt = au_sbr(parent->d_sb, bindex)->br_mnt; ++ path_get(&h_nd.path); ++ valid = reval(h_dentry, &h_nd); ++ path_put(&h_nd.path); ++ if (locked) ++ di_read_unlock(parent, AuLock_IR); ++ } ++ ++ if (unlikely(valid < 0)) ++ err = valid; ++ else if (!valid) ++ err = -EINVAL; ++ ++ out: ++ AuTraceErr(err); ++ return err; ++} ++ ++/* todo: remove this */ ++static int h_d_revalidate(struct dentry *dentry, struct inode *inode, ++ struct nameidata *nd, int do_udba) ++{ ++ int err; ++ umode_t mode, h_mode; ++ aufs_bindex_t bindex, btail, bstart, ibs, ibe; ++ unsigned char plus, unhashed, is_root, h_plus; ++ struct inode *first, *h_inode, *h_cached_inode; ++ struct dentry *h_dentry; ++ struct qstr *name, *h_name; ++ ++ err = 0; ++ plus = 0; ++ mode = 0; ++ first = NULL; ++ ibs = -1; ++ ibe = -1; ++ unhashed = !!d_unhashed(dentry); ++ is_root = !!IS_ROOT(dentry); ++ name = &dentry->d_name; ++ ++ /* ++ * Theoretically, REVAL test should be unnecessary in case of INOTIFY. ++ * But inotify doesn't fire some necessary events, ++ * IN_ATTRIB for atime/nlink/pageio ++ * IN_DELETE for NFS dentry ++ * Let's do REVAL test too. ++ */ ++ if (do_udba && inode) { ++ mode = (inode->i_mode & S_IFMT); ++ plus = (inode->i_nlink > 0); ++ first = au_h_iptr(inode, au_ibstart(inode)); ++ ibs = au_ibstart(inode); ++ ibe = au_ibend(inode); ++ } ++ ++ bstart = au_dbstart(dentry); ++ btail = bstart; ++ if (inode && S_ISDIR(inode->i_mode)) ++ btail = au_dbtaildir(dentry); ++ for (bindex = bstart; bindex <= btail; bindex++) { ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (!h_dentry) ++ continue; ++ ++ AuDbg("b%d, %.*s\n", bindex, AuDLNPair(h_dentry)); ++ h_name = &h_dentry->d_name; ++ if (unlikely(do_udba ++ && !is_root ++ && (unhashed != !!d_unhashed(h_dentry) ++ || name->len != h_name->len ++ || memcmp(name->name, h_name->name, name->len)) ++ )) { ++ AuDbg("unhash 0x%x 0x%x, %.*s %.*s\n", ++ unhashed, d_unhashed(h_dentry), ++ AuDLNPair(dentry), AuDLNPair(h_dentry)); ++ goto err; ++ } ++ ++ err = au_do_h_d_reval(h_dentry, nd, dentry, bindex); ++ if (unlikely(err)) ++ /* do not goto err, to keep the errno */ ++ break; ++ ++ /* todo: plink too? */ ++ if (!do_udba) ++ continue; ++ ++ /* UDBA tests */ ++ h_inode = h_dentry->d_inode; ++ if (unlikely(!!inode != !!h_inode)) ++ goto err; ++ ++ h_plus = plus; ++ h_mode = mode; ++ h_cached_inode = h_inode; ++ if (h_inode) { ++ h_mode = (h_inode->i_mode & S_IFMT); ++ h_plus = (h_inode->i_nlink > 0); ++ } ++ if (inode && ibs <= bindex && bindex <= ibe) ++ h_cached_inode = au_h_iptr(inode, bindex); ++ ++ if (unlikely(plus != h_plus ++ || mode != h_mode ++ || h_cached_inode != h_inode)) ++ goto err; ++ continue; ++ ++ err: ++ err = -EINVAL; ++ break; ++ } ++ ++ return err; ++} ++ ++static int simple_reval_dpath(struct dentry *dentry, unsigned int sigen) ++{ ++ int err; ++ struct dentry *parent; ++ struct inode *inode; ++ ++ inode = dentry->d_inode; ++ if (au_digen(dentry) == sigen && au_iigen(inode) == sigen) ++ return 0; ++ ++ parent = dget_parent(dentry); ++ di_read_lock_parent(parent, AuLock_IR); ++ AuDebugOn(au_digen(parent) != sigen ++ || au_iigen(parent->d_inode) != sigen); ++ au_dbg_verify_gen(parent, sigen); ++ ++ /* returns a number of positive dentries */ ++ err = au_refresh_hdentry(dentry, inode->i_mode & S_IFMT); ++ if (err >= 0) ++ err = au_refresh_hinode(inode, dentry); ++ ++ di_read_unlock(parent, AuLock_IR); ++ dput(parent); ++ return err; ++} ++ ++int au_reval_dpath(struct dentry *dentry, unsigned int sigen) ++{ ++ int err; ++ struct dentry *d, *parent; ++ struct inode *inode; ++ ++ if (!au_ftest_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIRS)) ++ return simple_reval_dpath(dentry, sigen); ++ ++ /* slow loop, keep it simple and stupid */ ++ /* cf: au_cpup_dirs() */ ++ err = 0; ++ parent = NULL; ++ while (au_digen(dentry) != sigen ++ || au_iigen(dentry->d_inode) != sigen) { ++ d = dentry; ++ while (1) { ++ dput(parent); ++ parent = dget_parent(d); ++ if (au_digen(parent) == sigen ++ && au_iigen(parent->d_inode) == sigen) ++ break; ++ d = parent; ++ } ++ ++ inode = d->d_inode; ++ if (d != dentry) ++ di_write_lock_child(d); ++ ++ /* someone might update our dentry while we were sleeping */ ++ if (au_digen(d) != sigen || au_iigen(d->d_inode) != sigen) { ++ di_read_lock_parent(parent, AuLock_IR); ++ /* returns a number of positive dentries */ ++ err = au_refresh_hdentry(d, inode->i_mode & S_IFMT); ++ if (err >= 0) ++ err = au_refresh_hinode(inode, d); ++ di_read_unlock(parent, AuLock_IR); ++ } ++ ++ if (d != dentry) ++ di_write_unlock(d); ++ dput(parent); ++ if (unlikely(err)) ++ break; ++ } ++ ++ return err; ++} ++ ++/* ++ * if valid returns 1, otherwise 0. ++ */ ++static int aufs_d_revalidate(struct dentry *dentry, struct nameidata *nd) ++{ ++ int valid, err; ++ unsigned int sigen; ++ unsigned char do_udba; ++ struct super_block *sb; ++ struct inode *inode; ++ ++ err = -EINVAL; ++ sb = dentry->d_sb; ++ inode = dentry->d_inode; ++ aufs_read_lock(dentry, AuLock_FLUSH | AuLock_DW); ++ sigen = au_sigen(sb); ++ if (au_digen(dentry) != sigen) { ++ AuDebugOn(IS_ROOT(dentry)); ++ if (inode) ++ err = au_reval_dpath(dentry, sigen); ++ if (unlikely(err)) ++ goto out_dgrade; ++ AuDebugOn(au_digen(dentry) != sigen); ++ } ++ if (inode && au_iigen(inode) != sigen) { ++ AuDebugOn(IS_ROOT(dentry)); ++ err = au_refresh_hinode(inode, dentry); ++ if (unlikely(err)) ++ goto out_dgrade; ++ AuDebugOn(au_iigen(inode) != sigen); ++ } ++ di_downgrade_lock(dentry, AuLock_IR); ++ ++ AuDebugOn(au_digen(dentry) != sigen); ++ AuDebugOn(inode && au_iigen(inode) != sigen); ++ err = -EINVAL; ++ do_udba = !au_opt_test(au_mntflags(sb), UDBA_NONE); ++ if (do_udba && inode) { ++ aufs_bindex_t bstart = au_ibstart(inode); ++ ++ if (bstart >= 0 ++ && au_test_higen(inode, au_h_iptr(inode, bstart))) ++ goto out; ++ } ++ ++ err = h_d_revalidate(dentry, inode, nd, do_udba); ++ if (unlikely(!err && do_udba && au_dbstart(dentry) < 0)) ++ /* both of real entry and whiteout found */ ++ err = -EIO; ++ goto out; ++ ++ out_dgrade: ++ di_downgrade_lock(dentry, AuLock_IR); ++ out: ++ aufs_read_unlock(dentry, AuLock_IR); ++ AuTraceErr(err); ++ valid = !err; ++ if (!valid) ++ AuDbg("%.*s invalid\n", AuDLNPair(dentry)); ++ return valid; ++} ++ ++static void aufs_d_release(struct dentry *dentry) ++{ ++ struct au_dinfo *dinfo; ++ aufs_bindex_t bend, bindex; ++ ++ dinfo = dentry->d_fsdata; ++ if (!dinfo) ++ return; ++ ++ /* dentry may not be revalidated */ ++ bindex = dinfo->di_bstart; ++ if (bindex >= 0) { ++ struct au_hdentry *p; ++ ++ bend = dinfo->di_bend; ++ p = dinfo->di_hdentry + bindex; ++ while (bindex++ <= bend) { ++ if (p->hd_dentry) ++ au_hdput(p); ++ p++; ++ } ++ } ++ kfree(dinfo->di_hdentry); ++ AuRwDestroy(&dinfo->di_rwsem); ++ au_cache_free_dinfo(dinfo); ++ au_hin_di_reinit(dentry); ++} ++ ++struct dentry_operations aufs_dop = { ++ .d_revalidate = aufs_d_revalidate, ++ .d_release = aufs_d_release ++}; +diff -Nur linux-2.6.31-vanilla/fs/aufs/dentry.h linux-2.6.31/fs/aufs/dentry.h +--- linux-2.6.31-vanilla/fs/aufs/dentry.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/dentry.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,231 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * lookup and dentry operations ++ */ ++ ++#ifndef __AUFS_DENTRY_H__ ++#define __AUFS_DENTRY_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++#include "rwsem.h" ++ ++/* make a single member structure for future use */ ++/* todo: remove this structure */ ++struct au_hdentry { ++ struct dentry *hd_dentry; ++}; ++ ++struct au_dinfo { ++ atomic_t di_generation; ++ ++ struct au_rwsem di_rwsem; ++ aufs_bindex_t di_bstart, di_bend, di_bwh, di_bdiropq; ++ struct au_hdentry *di_hdentry; ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* dentry.c */ ++extern struct dentry_operations aufs_dop; ++struct au_branch; ++struct dentry *au_lkup_one(struct qstr *name, struct dentry *h_parent, ++ struct au_branch *br, struct nameidata *nd); ++struct dentry *au_sio_lkup_one(struct qstr *name, struct dentry *parent, ++ struct au_branch *br); ++int au_h_verify(struct dentry *h_dentry, unsigned int udba, struct inode *h_dir, ++ struct dentry *h_parent, struct au_branch *br); ++ ++int au_lkup_dentry(struct dentry *dentry, aufs_bindex_t bstart, mode_t type, ++ struct nameidata *nd); ++int au_lkup_neg(struct dentry *dentry, aufs_bindex_t bindex); ++int au_refresh_hdentry(struct dentry *dentry, mode_t type); ++int au_reval_dpath(struct dentry *dentry, unsigned int sigen); ++ ++/* dinfo.c */ ++int au_alloc_dinfo(struct dentry *dentry); ++int au_di_realloc(struct au_dinfo *dinfo, int nbr); ++ ++void di_read_lock(struct dentry *d, int flags, unsigned int lsc); ++void di_read_unlock(struct dentry *d, int flags); ++void di_downgrade_lock(struct dentry *d, int flags); ++void di_write_lock(struct dentry *d, unsigned int lsc); ++void di_write_unlock(struct dentry *d); ++void di_write_lock2_child(struct dentry *d1, struct dentry *d2, int isdir); ++void di_write_lock2_parent(struct dentry *d1, struct dentry *d2, int isdir); ++void di_write_unlock2(struct dentry *d1, struct dentry *d2); ++ ++struct dentry *au_h_dptr(struct dentry *dentry, aufs_bindex_t bindex); ++aufs_bindex_t au_dbtail(struct dentry *dentry); ++aufs_bindex_t au_dbtaildir(struct dentry *dentry); ++ ++void au_set_h_dptr(struct dentry *dentry, aufs_bindex_t bindex, ++ struct dentry *h_dentry); ++void au_update_digen(struct dentry *dentry); ++void au_update_dbrange(struct dentry *dentry, int do_put_zero); ++void au_update_dbstart(struct dentry *dentry); ++void au_update_dbend(struct dentry *dentry); ++int au_find_dbindex(struct dentry *dentry, struct dentry *h_dentry); ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline struct au_dinfo *au_di(struct dentry *dentry) ++{ ++ return dentry->d_fsdata; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* lock subclass for dinfo */ ++enum { ++ AuLsc_DI_CHILD, /* child first */ ++ AuLsc_DI_CHILD2, /* rename(2), link(2), and cpup at hinotify */ ++ AuLsc_DI_CHILD3, /* copyup dirs */ ++ AuLsc_DI_PARENT, ++ AuLsc_DI_PARENT2, ++ AuLsc_DI_PARENT3 ++}; ++ ++/* ++ * di_read_lock_child, di_write_lock_child, ++ * di_read_lock_child2, di_write_lock_child2, ++ * di_read_lock_child3, di_write_lock_child3, ++ * di_read_lock_parent, di_write_lock_parent, ++ * di_read_lock_parent2, di_write_lock_parent2, ++ * di_read_lock_parent3, di_write_lock_parent3, ++ */ ++#define AuReadLockFunc(name, lsc) \ ++static inline void di_read_lock_##name(struct dentry *d, int flags) \ ++{ di_read_lock(d, flags, AuLsc_DI_##lsc); } ++ ++#define AuWriteLockFunc(name, lsc) \ ++static inline void di_write_lock_##name(struct dentry *d) \ ++{ di_write_lock(d, AuLsc_DI_##lsc); } ++ ++#define AuRWLockFuncs(name, lsc) \ ++ AuReadLockFunc(name, lsc) \ ++ AuWriteLockFunc(name, lsc) ++ ++AuRWLockFuncs(child, CHILD); ++AuRWLockFuncs(child2, CHILD2); ++AuRWLockFuncs(child3, CHILD3); ++AuRWLockFuncs(parent, PARENT); ++AuRWLockFuncs(parent2, PARENT2); ++AuRWLockFuncs(parent3, PARENT3); ++ ++#undef AuReadLockFunc ++#undef AuWriteLockFunc ++#undef AuRWLockFuncs ++ ++#define DiMustNoWaiters(d) AuRwMustNoWaiters(&au_di(d)->di_rwsem) ++#define DiMustAnyLock(d) AuRwMustAnyLock(&au_di(d)->di_rwsem) ++#define DiMustWriteLock(d) AuRwMustWriteLock(&au_di(d)->di_rwsem) ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* todo: memory barrier? */ ++static inline unsigned int au_digen(struct dentry *d) ++{ ++ return atomic_read(&au_di(d)->di_generation); ++} ++ ++static inline void au_h_dentry_init(struct au_hdentry *hdentry) ++{ ++ hdentry->hd_dentry = NULL; ++} ++ ++static inline void au_hdput(struct au_hdentry *hd) ++{ ++ dput(hd->hd_dentry); ++} ++ ++static inline aufs_bindex_t au_dbstart(struct dentry *dentry) ++{ ++ DiMustAnyLock(dentry); ++ return au_di(dentry)->di_bstart; ++} ++ ++static inline aufs_bindex_t au_dbend(struct dentry *dentry) ++{ ++ DiMustAnyLock(dentry); ++ return au_di(dentry)->di_bend; ++} ++ ++static inline aufs_bindex_t au_dbwh(struct dentry *dentry) ++{ ++ DiMustAnyLock(dentry); ++ return au_di(dentry)->di_bwh; ++} ++ ++static inline aufs_bindex_t au_dbdiropq(struct dentry *dentry) ++{ ++ DiMustAnyLock(dentry); ++ return au_di(dentry)->di_bdiropq; ++} ++ ++/* todo: hard/soft set? */ ++static inline void au_set_dbstart(struct dentry *dentry, aufs_bindex_t bindex) ++{ ++ DiMustWriteLock(dentry); ++ au_di(dentry)->di_bstart = bindex; ++} ++ ++static inline void au_set_dbend(struct dentry *dentry, aufs_bindex_t bindex) ++{ ++ DiMustWriteLock(dentry); ++ au_di(dentry)->di_bend = bindex; ++} ++ ++static inline void au_set_dbwh(struct dentry *dentry, aufs_bindex_t bindex) ++{ ++ DiMustWriteLock(dentry); ++ /* dbwh can be outside of bstart - bend range */ ++ au_di(dentry)->di_bwh = bindex; ++} ++ ++static inline void au_set_dbdiropq(struct dentry *dentry, aufs_bindex_t bindex) ++{ ++ DiMustWriteLock(dentry); ++ au_di(dentry)->di_bdiropq = bindex; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++#ifdef CONFIG_AUFS_HINOTIFY ++static inline void au_digen_dec(struct dentry *d) ++{ ++ atomic_dec_return(&au_di(d)->di_generation); ++} ++ ++static inline void au_hin_di_reinit(struct dentry *dentry) ++{ ++ dentry->d_fsdata = NULL; ++} ++#else ++static inline void au_hin_di_reinit(struct dentry *dentry __maybe_unused) ++{ ++ /* empty */ ++} ++#endif /* CONFIG_AUFS_HINOTIFY */ ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_DENTRY_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/dinfo.c linux-2.6.31/fs/aufs/dinfo.c +--- linux-2.6.31-vanilla/fs/aufs/dinfo.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/dinfo.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,367 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * dentry private data ++ */ ++ ++#include "aufs.h" ++ ++int au_alloc_dinfo(struct dentry *dentry) ++{ ++ struct au_dinfo *dinfo; ++ struct super_block *sb; ++ int nbr; ++ ++ dinfo = au_cache_alloc_dinfo(); ++ if (unlikely(!dinfo)) ++ goto out; ++ ++ sb = dentry->d_sb; ++ nbr = au_sbend(sb) + 1; ++ if (nbr <= 0) ++ nbr = 1; ++ dinfo->di_hdentry = kcalloc(nbr, sizeof(*dinfo->di_hdentry), GFP_NOFS); ++ if (unlikely(!dinfo->di_hdentry)) ++ goto out_dinfo; ++ ++ atomic_set(&dinfo->di_generation, au_sigen(sb)); ++ /* smp_mb(); */ /* atomic_set */ ++ au_rw_init_wlock_nested(&dinfo->di_rwsem, AuLsc_DI_CHILD); ++ dinfo->di_bstart = -1; ++ dinfo->di_bend = -1; ++ dinfo->di_bwh = -1; ++ dinfo->di_bdiropq = -1; ++ ++ dentry->d_fsdata = dinfo; ++ dentry->d_op = &aufs_dop; ++ return 0; /* success */ ++ ++ out_dinfo: ++ au_cache_free_dinfo(dinfo); ++ out: ++ return -ENOMEM; ++} ++ ++int au_di_realloc(struct au_dinfo *dinfo, int nbr) ++{ ++ int err, sz; ++ struct au_hdentry *hdp; ++ ++ AuRwMustWriteLock(&dinfo->di_rwsem); ++ ++ err = -ENOMEM; ++ sz = sizeof(*hdp) * (dinfo->di_bend + 1); ++ if (!sz) ++ sz = sizeof(*hdp); ++ hdp = au_kzrealloc(dinfo->di_hdentry, sz, sizeof(*hdp) * nbr, GFP_NOFS); ++ if (hdp) { ++ dinfo->di_hdentry = hdp; ++ err = 0; ++ } ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static void do_ii_write_lock(struct inode *inode, unsigned int lsc) ++{ ++ switch (lsc) { ++ case AuLsc_DI_CHILD: ++ ii_write_lock_child(inode); ++ break; ++ case AuLsc_DI_CHILD2: ++ ii_write_lock_child2(inode); ++ break; ++ case AuLsc_DI_CHILD3: ++ ii_write_lock_child3(inode); ++ break; ++ case AuLsc_DI_PARENT: ++ ii_write_lock_parent(inode); ++ break; ++ case AuLsc_DI_PARENT2: ++ ii_write_lock_parent2(inode); ++ break; ++ case AuLsc_DI_PARENT3: ++ ii_write_lock_parent3(inode); ++ break; ++ default: ++ BUG(); ++ } ++} ++ ++static void do_ii_read_lock(struct inode *inode, unsigned int lsc) ++{ ++ switch (lsc) { ++ case AuLsc_DI_CHILD: ++ ii_read_lock_child(inode); ++ break; ++ case AuLsc_DI_CHILD2: ++ ii_read_lock_child2(inode); ++ break; ++ case AuLsc_DI_CHILD3: ++ ii_read_lock_child3(inode); ++ break; ++ case AuLsc_DI_PARENT: ++ ii_read_lock_parent(inode); ++ break; ++ case AuLsc_DI_PARENT2: ++ ii_read_lock_parent2(inode); ++ break; ++ case AuLsc_DI_PARENT3: ++ ii_read_lock_parent3(inode); ++ break; ++ default: ++ BUG(); ++ } ++} ++ ++void di_read_lock(struct dentry *d, int flags, unsigned int lsc) ++{ ++ au_rw_read_lock_nested(&au_di(d)->di_rwsem, lsc); ++ if (d->d_inode) { ++ if (au_ftest_lock(flags, IW)) ++ do_ii_write_lock(d->d_inode, lsc); ++ else if (au_ftest_lock(flags, IR)) ++ do_ii_read_lock(d->d_inode, lsc); ++ } ++} ++ ++void di_read_unlock(struct dentry *d, int flags) ++{ ++ if (d->d_inode) { ++ if (au_ftest_lock(flags, IW)) ++ ii_write_unlock(d->d_inode); ++ else if (au_ftest_lock(flags, IR)) ++ ii_read_unlock(d->d_inode); ++ } ++ au_rw_read_unlock(&au_di(d)->di_rwsem); ++} ++ ++void di_downgrade_lock(struct dentry *d, int flags) ++{ ++ if (d->d_inode && au_ftest_lock(flags, IR)) ++ ii_downgrade_lock(d->d_inode); ++ au_rw_dgrade_lock(&au_di(d)->di_rwsem); ++} ++ ++void di_write_lock(struct dentry *d, unsigned int lsc) ++{ ++ au_rw_write_lock_nested(&au_di(d)->di_rwsem, lsc); ++ if (d->d_inode) ++ do_ii_write_lock(d->d_inode, lsc); ++} ++ ++void di_write_unlock(struct dentry *d) ++{ ++ if (d->d_inode) ++ ii_write_unlock(d->d_inode); ++ au_rw_write_unlock(&au_di(d)->di_rwsem); ++} ++ ++void di_write_lock2_child(struct dentry *d1, struct dentry *d2, int isdir) ++{ ++ AuDebugOn(d1 == d2 ++ || d1->d_inode == d2->d_inode ++ || d1->d_sb != d2->d_sb); ++ ++ if (isdir && au_test_subdir(d1, d2)) { ++ di_write_lock_child(d1); ++ di_write_lock_child2(d2); ++ } else { ++ /* there should be no races */ ++ di_write_lock_child(d2); ++ di_write_lock_child2(d1); ++ } ++} ++ ++void di_write_lock2_parent(struct dentry *d1, struct dentry *d2, int isdir) ++{ ++ AuDebugOn(d1 == d2 ++ || d1->d_inode == d2->d_inode ++ || d1->d_sb != d2->d_sb); ++ ++ if (isdir && au_test_subdir(d1, d2)) { ++ di_write_lock_parent(d1); ++ di_write_lock_parent2(d2); ++ } else { ++ /* there should be no races */ ++ di_write_lock_parent(d2); ++ di_write_lock_parent2(d1); ++ } ++} ++ ++void di_write_unlock2(struct dentry *d1, struct dentry *d2) ++{ ++ di_write_unlock(d1); ++ if (d1->d_inode == d2->d_inode) ++ au_rw_write_unlock(&au_di(d2)->di_rwsem); ++ else ++ di_write_unlock(d2); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct dentry *au_h_dptr(struct dentry *dentry, aufs_bindex_t bindex) ++{ ++ struct dentry *d; ++ ++ DiMustAnyLock(dentry); ++ ++ if (au_dbstart(dentry) < 0 || bindex < au_dbstart(dentry)) ++ return NULL; ++ AuDebugOn(bindex < 0); ++ d = au_di(dentry)->di_hdentry[0 + bindex].hd_dentry; ++ AuDebugOn(d && (atomic_read(&d->d_count) <= 0)); ++ return d; ++} ++ ++aufs_bindex_t au_dbtail(struct dentry *dentry) ++{ ++ aufs_bindex_t bend, bwh; ++ ++ bend = au_dbend(dentry); ++ if (0 <= bend) { ++ bwh = au_dbwh(dentry); ++ if (!bwh) ++ return bwh; ++ if (0 < bwh && bwh < bend) ++ return bwh - 1; ++ } ++ return bend; ++} ++ ++aufs_bindex_t au_dbtaildir(struct dentry *dentry) ++{ ++ aufs_bindex_t bend, bopq; ++ ++ bend = au_dbtail(dentry); ++ if (0 <= bend) { ++ bopq = au_dbdiropq(dentry); ++ if (0 <= bopq && bopq < bend) ++ bend = bopq; ++ } ++ return bend; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void au_set_h_dptr(struct dentry *dentry, aufs_bindex_t bindex, ++ struct dentry *h_dentry) ++{ ++ struct au_hdentry *hd = au_di(dentry)->di_hdentry + bindex; ++ ++ DiMustWriteLock(dentry); ++ ++ if (hd->hd_dentry) ++ au_hdput(hd); ++ hd->hd_dentry = h_dentry; ++} ++ ++void au_update_digen(struct dentry *dentry) ++{ ++ atomic_set(&au_di(dentry)->di_generation, au_sigen(dentry->d_sb)); ++ /* smp_mb(); */ /* atomic_set */ ++} ++ ++void au_update_dbrange(struct dentry *dentry, int do_put_zero) ++{ ++ struct au_dinfo *dinfo; ++ struct dentry *h_d; ++ ++ DiMustWriteLock(dentry); ++ ++ dinfo = au_di(dentry); ++ if (!dinfo || dinfo->di_bstart < 0) ++ return; ++ ++ if (do_put_zero) { ++ aufs_bindex_t bindex, bend; ++ ++ bend = dinfo->di_bend; ++ for (bindex = dinfo->di_bstart; bindex <= bend; bindex++) { ++ h_d = dinfo->di_hdentry[0 + bindex].hd_dentry; ++ if (h_d && !h_d->d_inode) ++ au_set_h_dptr(dentry, bindex, NULL); ++ } ++ } ++ ++ dinfo->di_bstart = -1; ++ while (++dinfo->di_bstart <= dinfo->di_bend) ++ if (dinfo->di_hdentry[0 + dinfo->di_bstart].hd_dentry) ++ break; ++ if (dinfo->di_bstart > dinfo->di_bend) { ++ dinfo->di_bstart = -1; ++ dinfo->di_bend = -1; ++ return; ++ } ++ ++ dinfo->di_bend++; ++ while (0 <= --dinfo->di_bend) ++ if (dinfo->di_hdentry[0 + dinfo->di_bend].hd_dentry) ++ break; ++ AuDebugOn(dinfo->di_bstart > dinfo->di_bend || dinfo->di_bend < 0); ++} ++ ++void au_update_dbstart(struct dentry *dentry) ++{ ++ aufs_bindex_t bindex, bend; ++ struct dentry *h_dentry; ++ ++ bend = au_dbend(dentry); ++ for (bindex = au_dbstart(dentry); bindex <= bend; bindex++) { ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (!h_dentry) ++ continue; ++ if (h_dentry->d_inode) { ++ au_set_dbstart(dentry, bindex); ++ return; ++ } ++ au_set_h_dptr(dentry, bindex, NULL); ++ } ++} ++ ++void au_update_dbend(struct dentry *dentry) ++{ ++ aufs_bindex_t bindex, bstart; ++ struct dentry *h_dentry; ++ ++ bstart = au_dbstart(dentry); ++ for (bindex = au_dbend(dentry); bindex <= bstart; bindex--) { ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (!h_dentry) ++ continue; ++ if (h_dentry->d_inode) { ++ au_set_dbend(dentry, bindex); ++ return; ++ } ++ au_set_h_dptr(dentry, bindex, NULL); ++ } ++} ++ ++int au_find_dbindex(struct dentry *dentry, struct dentry *h_dentry) ++{ ++ aufs_bindex_t bindex, bend; ++ ++ bend = au_dbend(dentry); ++ for (bindex = au_dbstart(dentry); bindex <= bend; bindex++) ++ if (au_h_dptr(dentry, bindex) == h_dentry) ++ return bindex; ++ return -1; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/dir.c linux-2.6.31/fs/aufs/dir.c +--- linux-2.6.31-vanilla/fs/aufs/dir.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/dir.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,593 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * directory operations ++ */ ++ ++#include ++#include ++#include "aufs.h" ++ ++void au_add_nlink(struct inode *dir, struct inode *h_dir) ++{ ++ AuDebugOn(!S_ISDIR(dir->i_mode) || !S_ISDIR(h_dir->i_mode)); ++ ++ dir->i_nlink += h_dir->i_nlink - 2; ++ if (h_dir->i_nlink < 2) ++ dir->i_nlink += 2; ++} ++ ++void au_sub_nlink(struct inode *dir, struct inode *h_dir) ++{ ++ AuDebugOn(!S_ISDIR(dir->i_mode) || !S_ISDIR(h_dir->i_mode)); ++ ++ dir->i_nlink -= h_dir->i_nlink - 2; ++ if (h_dir->i_nlink < 2) ++ dir->i_nlink -= 2; ++} ++ ++loff_t au_dir_size(struct file *file, struct dentry *dentry) ++{ ++ loff_t sz; ++ aufs_bindex_t bindex, bend; ++ struct file *h_file; ++ struct dentry *h_dentry; ++ ++ sz = 0; ++ if (file) { ++ AuDebugOn(!file->f_dentry); ++ AuDebugOn(!file->f_dentry->d_inode); ++ AuDebugOn(!S_ISDIR(file->f_dentry->d_inode->i_mode)); ++ ++ bend = au_fbend(file); ++ for (bindex = au_fbstart(file); ++ bindex <= bend && sz < KMALLOC_MAX_SIZE; ++ bindex++) { ++ h_file = au_h_fptr(file, bindex); ++ if (h_file ++ && h_file->f_dentry ++ && h_file->f_dentry->d_inode) ++ sz += i_size_read(h_file->f_dentry->d_inode); ++ } ++ } else { ++ AuDebugOn(!dentry); ++ AuDebugOn(!dentry->d_inode); ++ AuDebugOn(!S_ISDIR(dentry->d_inode->i_mode)); ++ ++ bend = au_dbtaildir(dentry); ++ for (bindex = au_dbstart(dentry); ++ bindex <= bend && sz < KMALLOC_MAX_SIZE; ++ bindex++) { ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (h_dentry && h_dentry->d_inode) ++ sz += i_size_read(h_dentry->d_inode); ++ } ++ } ++ if (sz < KMALLOC_MAX_SIZE) ++ sz = roundup_pow_of_two(sz); ++ if (sz > KMALLOC_MAX_SIZE) ++ sz = KMALLOC_MAX_SIZE; ++ else if (sz < NAME_MAX) { ++ BUILD_BUG_ON(AUFS_RDBLK_DEF < NAME_MAX); ++ sz = AUFS_RDBLK_DEF; ++ } ++ return sz; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int reopen_dir(struct file *file) ++{ ++ int err; ++ unsigned int flags; ++ aufs_bindex_t bindex, btail, bstart; ++ struct dentry *dentry, *h_dentry; ++ struct file *h_file; ++ ++ /* open all lower dirs */ ++ dentry = file->f_dentry; ++ bstart = au_dbstart(dentry); ++ for (bindex = au_fbstart(file); bindex < bstart; bindex++) ++ au_set_h_fptr(file, bindex, NULL); ++ au_set_fbstart(file, bstart); ++ ++ btail = au_dbtaildir(dentry); ++ for (bindex = au_fbend(file); btail < bindex; bindex--) ++ au_set_h_fptr(file, bindex, NULL); ++ au_set_fbend(file, btail); ++ ++ spin_lock(&file->f_lock); ++ flags = file->f_flags; ++ spin_unlock(&file->f_lock); ++ for (bindex = bstart; bindex <= btail; bindex++) { ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (!h_dentry) ++ continue; ++ h_file = au_h_fptr(file, bindex); ++ if (h_file) ++ continue; ++ ++ h_file = au_h_open(dentry, bindex, flags, file); ++ err = PTR_ERR(h_file); ++ if (IS_ERR(h_file)) ++ goto out; /* close all? */ ++ au_set_h_fptr(file, bindex, h_file); ++ } ++ au_update_figen(file); ++ /* todo: necessary? */ ++ /* file->f_ra = h_file->f_ra; */ ++ err = 0; ++ ++ out: ++ return err; ++} ++ ++static int do_open_dir(struct file *file, int flags) ++{ ++ int err; ++ aufs_bindex_t bindex, btail; ++ struct dentry *dentry, *h_dentry; ++ struct file *h_file; ++ ++ FiMustWriteLock(file); ++ ++ err = 0; ++ dentry = file->f_dentry; ++ au_set_fvdir_cache(file, NULL); ++ au_fi(file)->fi_maintain_plink = 0; ++ file->f_version = dentry->d_inode->i_version; ++ bindex = au_dbstart(dentry); ++ au_set_fbstart(file, bindex); ++ btail = au_dbtaildir(dentry); ++ au_set_fbend(file, btail); ++ for (; !err && bindex <= btail; bindex++) { ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (!h_dentry) ++ continue; ++ ++ h_file = au_h_open(dentry, bindex, flags, file); ++ if (IS_ERR(h_file)) { ++ err = PTR_ERR(h_file); ++ break; ++ } ++ au_set_h_fptr(file, bindex, h_file); ++ } ++ au_update_figen(file); ++ /* todo: necessary? */ ++ /* file->f_ra = h_file->f_ra; */ ++ if (!err) ++ return 0; /* success */ ++ ++ /* close all */ ++ for (bindex = au_fbstart(file); bindex <= btail; bindex++) ++ au_set_h_fptr(file, bindex, NULL); ++ au_set_fbstart(file, -1); ++ au_set_fbend(file, -1); ++ return err; ++} ++ ++static int aufs_open_dir(struct inode *inode __maybe_unused, ++ struct file *file) ++{ ++ return au_do_open(file, do_open_dir); ++} ++ ++static int aufs_release_dir(struct inode *inode __maybe_unused, ++ struct file *file) ++{ ++ struct au_vdir *vdir_cache; ++ struct super_block *sb; ++ struct au_sbinfo *sbinfo; ++ ++ sb = file->f_dentry->d_sb; ++ si_noflush_read_lock(sb); ++ fi_write_lock(file); ++ vdir_cache = au_fvdir_cache(file); ++ if (vdir_cache) ++ au_vdir_free(vdir_cache); ++ if (au_fi(file)->fi_maintain_plink) { ++ sbinfo = au_sbi(sb); ++ /* clear the flag without write-lock */ ++ sbinfo->au_si_status &= ~AuSi_MAINTAIN_PLINK; ++ smp_mb(); ++ wake_up_all(&sbinfo->si_plink_wq); ++ } ++ fi_write_unlock(file); ++ au_finfo_fin(file); ++ si_read_unlock(sb); ++ return 0; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int au_do_fsync_dir_no_file(struct dentry *dentry, int datasync) ++{ ++ int err; ++ aufs_bindex_t bend, bindex; ++ struct inode *inode; ++ struct super_block *sb; ++ ++ err = 0; ++ sb = dentry->d_sb; ++ inode = dentry->d_inode; ++ IMustLock(inode); ++ bend = au_dbend(dentry); ++ for (bindex = au_dbstart(dentry); !err && bindex <= bend; bindex++) { ++ struct path h_path; ++ struct inode *h_inode; ++ ++ if (au_test_ro(sb, bindex, inode)) ++ continue; ++ h_path.dentry = au_h_dptr(dentry, bindex); ++ if (!h_path.dentry) ++ continue; ++ h_inode = h_path.dentry->d_inode; ++ if (!h_inode) ++ continue; ++ ++ /* no mnt_want_write() */ ++ /* cf. fs/nsfd/vfs.c and fs/nfsd/nfs4recover.c */ ++ /* todo: inotiry fired? */ ++ h_path.mnt = au_sbr_mnt(sb, bindex); ++ mutex_lock(&h_inode->i_mutex); ++ err = filemap_fdatawrite(h_inode->i_mapping); ++ AuDebugOn(!h_inode->i_fop); ++ if (!err && h_inode->i_fop->fsync) ++ err = h_inode->i_fop->fsync(NULL, h_path.dentry, ++ datasync); ++ if (!err) ++ err = filemap_fdatawrite(h_inode->i_mapping); ++ if (!err) ++ vfsub_update_h_iattr(&h_path, /*did*/NULL); /*ignore*/ ++ mutex_unlock(&h_inode->i_mutex); ++ } ++ ++ return err; ++} ++ ++static int au_do_fsync_dir(struct file *file, int datasync) ++{ ++ int err; ++ aufs_bindex_t bend, bindex; ++ struct file *h_file; ++ struct super_block *sb; ++ struct inode *inode; ++ struct mutex *h_mtx; ++ ++ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1); ++ if (unlikely(err)) ++ goto out; ++ ++ sb = file->f_dentry->d_sb; ++ inode = file->f_dentry->d_inode; ++ bend = au_fbend(file); ++ for (bindex = au_fbstart(file); !err && bindex <= bend; bindex++) { ++ h_file = au_h_fptr(file, bindex); ++ if (!h_file || au_test_ro(sb, bindex, inode)) ++ continue; ++ ++ err = vfs_fsync(h_file, h_file->f_dentry, datasync); ++ if (!err) { ++ h_mtx = &h_file->f_dentry->d_inode->i_mutex; ++ mutex_lock(h_mtx); ++ vfsub_update_h_iattr(&h_file->f_path, /*did*/NULL); ++ /*ignore*/ ++ mutex_unlock(h_mtx); ++ } ++ } ++ ++ out: ++ return err; ++} ++ ++/* ++ * @file may be NULL ++ */ ++static int aufs_fsync_dir(struct file *file, struct dentry *dentry, ++ int datasync) ++{ ++ int err; ++ struct super_block *sb; ++ ++ IMustLock(dentry->d_inode); ++ ++ err = 0; ++ sb = dentry->d_sb; ++ si_noflush_read_lock(sb); ++ if (file) ++ err = au_do_fsync_dir(file, datasync); ++ else { ++ di_write_lock_child(dentry); ++ err = au_do_fsync_dir_no_file(dentry, datasync); ++ } ++ au_cpup_attr_timesizes(dentry->d_inode); ++ di_write_unlock(dentry); ++ if (file) ++ fi_write_unlock(file); ++ ++ si_read_unlock(sb); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int aufs_readdir(struct file *file, void *dirent, filldir_t filldir) ++{ ++ int err; ++ struct dentry *dentry; ++ struct inode *inode; ++ struct super_block *sb; ++ ++ dentry = file->f_dentry; ++ inode = dentry->d_inode; ++ IMustLock(inode); ++ ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1); ++ if (unlikely(err)) ++ goto out; ++ err = au_vdir_init(file); ++ di_downgrade_lock(dentry, AuLock_IR); ++ if (unlikely(err)) ++ goto out_unlock; ++ ++ if (!au_test_nfsd(current)) { ++ err = au_vdir_fill_de(file, dirent, filldir); ++ fsstack_copy_attr_atime(inode, ++ au_h_iptr(inode, au_ibstart(inode))); ++ } else { ++ /* ++ * nfsd filldir may call lookup_one_len(), vfs_getattr(), ++ * encode_fh() and others. ++ */ ++ struct inode *h_inode = au_h_iptr(inode, au_ibstart(inode)); ++ ++ di_read_unlock(dentry, AuLock_IR); ++ si_read_unlock(sb); ++ lockdep_off(); ++ err = au_vdir_fill_de(file, dirent, filldir); ++ lockdep_on(); ++ fsstack_copy_attr_atime(inode, h_inode); ++ fi_write_unlock(file); ++ ++ AuTraceErr(err); ++ return err; ++ } ++ ++ out_unlock: ++ di_read_unlock(dentry, AuLock_IR); ++ fi_write_unlock(file); ++ out: ++ si_read_unlock(sb); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++#define AuTestEmpty_WHONLY 1 ++#define AuTestEmpty_CALLED (1 << 1) ++#define AuTestEmpty_SHWH (1 << 2) ++#define au_ftest_testempty(flags, name) ((flags) & AuTestEmpty_##name) ++#define au_fset_testempty(flags, name) { (flags) |= AuTestEmpty_##name; } ++#define au_fclr_testempty(flags, name) { (flags) &= ~AuTestEmpty_##name; } ++ ++#ifndef CONFIG_AUFS_SHWH ++#undef AuTestEmpty_SHWH ++#define AuTestEmpty_SHWH 0 ++#endif ++ ++struct test_empty_arg { ++ struct au_nhash *whlist; ++ unsigned int flags; ++ int err; ++ aufs_bindex_t bindex; ++}; ++ ++static int test_empty_cb(void *__arg, const char *__name, int namelen, ++ loff_t offset __maybe_unused, u64 ino, ++ unsigned int d_type) ++{ ++ struct test_empty_arg *arg = __arg; ++ char *name = (void *)__name; ++ ++ arg->err = 0; ++ au_fset_testempty(arg->flags, CALLED); ++ /* smp_mb(); */ ++ if (name[0] == '.' ++ && (namelen == 1 || (name[1] == '.' && namelen == 2))) ++ goto out; /* success */ ++ ++ if (namelen <= AUFS_WH_PFX_LEN ++ || memcmp(name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) { ++ if (au_ftest_testempty(arg->flags, WHONLY) ++ && !au_nhash_test_known_wh(arg->whlist, name, namelen)) ++ arg->err = -ENOTEMPTY; ++ goto out; ++ } ++ ++ name += AUFS_WH_PFX_LEN; ++ namelen -= AUFS_WH_PFX_LEN; ++ if (!au_nhash_test_known_wh(arg->whlist, name, namelen)) ++ arg->err = au_nhash_append_wh ++ (arg->whlist, name, namelen, ino, d_type, arg->bindex, ++ au_ftest_testempty(arg->flags, SHWH)); ++ ++ out: ++ /* smp_mb(); */ ++ AuTraceErr(arg->err); ++ return arg->err; ++} ++ ++static int do_test_empty(struct dentry *dentry, struct test_empty_arg *arg) ++{ ++ int err; ++ struct file *h_file; ++ ++ h_file = au_h_open(dentry, arg->bindex, ++ O_RDONLY | O_NONBLOCK | O_DIRECTORY | O_LARGEFILE, ++ /*file*/NULL); ++ err = PTR_ERR(h_file); ++ if (IS_ERR(h_file)) ++ goto out; ++ ++ err = 0; ++ if (!au_opt_test(au_mntflags(dentry->d_sb), UDBA_NONE) ++ && !h_file->f_dentry->d_inode->i_nlink) ++ goto out_put; ++ ++ do { ++ arg->err = 0; ++ au_fclr_testempty(arg->flags, CALLED); ++ /* smp_mb(); */ ++ err = vfsub_readdir(h_file, test_empty_cb, arg); ++ if (err >= 0) ++ err = arg->err; ++ } while (!err && au_ftest_testempty(arg->flags, CALLED)); ++ ++ out_put: ++ fput(h_file); ++ au_sbr_put(dentry->d_sb, arg->bindex); ++ out: ++ return err; ++} ++ ++struct do_test_empty_args { ++ int *errp; ++ struct dentry *dentry; ++ struct test_empty_arg *arg; ++}; ++ ++static void call_do_test_empty(void *args) ++{ ++ struct do_test_empty_args *a = args; ++ *a->errp = do_test_empty(a->dentry, a->arg); ++} ++ ++static int sio_test_empty(struct dentry *dentry, struct test_empty_arg *arg) ++{ ++ int err, wkq_err; ++ struct dentry *h_dentry; ++ struct inode *h_inode; ++ ++ h_dentry = au_h_dptr(dentry, arg->bindex); ++ h_inode = h_dentry->d_inode; ++ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); ++ err = au_test_h_perm_sio(h_inode, MAY_EXEC | MAY_READ); ++ mutex_unlock(&h_inode->i_mutex); ++ if (!err) ++ err = do_test_empty(dentry, arg); ++ else { ++ struct do_test_empty_args args = { ++ .errp = &err, ++ .dentry = dentry, ++ .arg = arg ++ }; ++ unsigned int flags = arg->flags; ++ ++ wkq_err = au_wkq_wait(call_do_test_empty, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ arg->flags = flags; ++ } ++ ++ return err; ++} ++ ++int au_test_empty_lower(struct dentry *dentry) ++{ ++ int err; ++ unsigned int rdhash; ++ aufs_bindex_t bindex, bstart, btail; ++ struct au_nhash whlist; ++ struct test_empty_arg arg; ++ ++ SiMustAnyLock(dentry->d_sb); ++ ++ rdhash = au_sbi(dentry->d_sb)->si_rdhash; ++ if (!rdhash) ++ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, dentry)); ++ err = au_nhash_alloc(&whlist, rdhash, GFP_NOFS); ++ if (unlikely(err)) ++ goto out; ++ ++ arg.flags = 0; ++ arg.whlist = &whlist; ++ bstart = au_dbstart(dentry); ++ if (au_opt_test(au_mntflags(dentry->d_sb), SHWH)) ++ au_fset_testempty(arg.flags, SHWH); ++ arg.bindex = bstart; ++ err = do_test_empty(dentry, &arg); ++ if (unlikely(err)) ++ goto out_whlist; ++ ++ au_fset_testempty(arg.flags, WHONLY); ++ btail = au_dbtaildir(dentry); ++ for (bindex = bstart + 1; !err && bindex <= btail; bindex++) { ++ struct dentry *h_dentry; ++ ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (h_dentry && h_dentry->d_inode) { ++ arg.bindex = bindex; ++ err = do_test_empty(dentry, &arg); ++ } ++ } ++ ++ out_whlist: ++ au_nhash_wh_free(&whlist); ++ out: ++ return err; ++} ++ ++int au_test_empty(struct dentry *dentry, struct au_nhash *whlist) ++{ ++ int err; ++ struct test_empty_arg arg; ++ aufs_bindex_t bindex, btail; ++ ++ err = 0; ++ arg.whlist = whlist; ++ arg.flags = AuTestEmpty_WHONLY; ++ if (au_opt_test(au_mntflags(dentry->d_sb), SHWH)) ++ au_fset_testempty(arg.flags, SHWH); ++ btail = au_dbtaildir(dentry); ++ for (bindex = au_dbstart(dentry); !err && bindex <= btail; bindex++) { ++ struct dentry *h_dentry; ++ ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (h_dentry && h_dentry->d_inode) { ++ arg.bindex = bindex; ++ err = sio_test_empty(dentry, &arg); ++ } ++ } ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++const struct file_operations aufs_dir_fop = { ++ .read = generic_read_dir, ++ .readdir = aufs_readdir, ++ .unlocked_ioctl = aufs_ioctl_dir, ++ .open = aufs_open_dir, ++ .release = aufs_release_dir, ++ .flush = aufs_flush, ++ .fsync = aufs_fsync_dir ++}; +diff -Nur linux-2.6.31-vanilla/fs/aufs/dir.h linux-2.6.31/fs/aufs/dir.h +--- linux-2.6.31-vanilla/fs/aufs/dir.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/dir.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,127 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * directory operations ++ */ ++ ++#ifndef __AUFS_DIR_H__ ++#define __AUFS_DIR_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* need to be faster and smaller */ ++ ++struct au_nhash { ++ unsigned int nh_num; ++ struct hlist_head *nh_head; ++}; ++ ++struct au_vdir_destr { ++ unsigned char len; ++ unsigned char name[0]; ++} __packed; ++ ++struct au_vdir_dehstr { ++ struct hlist_node hash; ++ struct au_vdir_destr *str; ++}; ++ ++struct au_vdir_de { ++ ino_t de_ino; ++ unsigned char de_type; ++ /* caution: packed */ ++ struct au_vdir_destr de_str; ++} __packed; ++ ++struct au_vdir_wh { ++ struct hlist_node wh_hash; ++#ifdef CONFIG_AUFS_SHWH ++ ino_t wh_ino; ++ aufs_bindex_t wh_bindex; ++ unsigned char wh_type; ++#else ++ aufs_bindex_t wh_bindex; ++#endif ++ /* caution: packed */ ++ struct au_vdir_destr wh_str; ++} __packed; ++ ++union au_vdir_deblk_p { ++ unsigned char *deblk; ++ struct au_vdir_de *de; ++}; ++ ++struct au_vdir { ++ unsigned char **vd_deblk; ++ unsigned long vd_nblk; ++ struct { ++ unsigned long ul; ++ union au_vdir_deblk_p p; ++ } vd_last; ++ ++ unsigned long vd_version; ++ unsigned int vd_deblk_sz; ++ unsigned long vd_jiffy; ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* dir.c */ ++extern const struct file_operations aufs_dir_fop; ++void au_add_nlink(struct inode *dir, struct inode *h_dir); ++void au_sub_nlink(struct inode *dir, struct inode *h_dir); ++loff_t au_dir_size(struct file *file, struct dentry *dentry); ++int au_test_empty_lower(struct dentry *dentry); ++int au_test_empty(struct dentry *dentry, struct au_nhash *whlist); ++ ++/* vdir.c */ ++unsigned int au_rdhash_est(loff_t sz); ++int au_nhash_alloc(struct au_nhash *nhash, unsigned int num_hash, gfp_t gfp); ++void au_nhash_wh_free(struct au_nhash *whlist); ++int au_nhash_test_longer_wh(struct au_nhash *whlist, aufs_bindex_t btgt, ++ int limit); ++int au_nhash_test_known_wh(struct au_nhash *whlist, char *name, int nlen); ++int au_nhash_append_wh(struct au_nhash *whlist, char *name, int nlen, ino_t ino, ++ unsigned int d_type, aufs_bindex_t bindex, ++ unsigned char shwh); ++void au_vdir_free(struct au_vdir *vdir); ++int au_vdir_init(struct file *file); ++int au_vdir_fill_de(struct file *file, void *dirent, filldir_t filldir); ++ ++/* ioctl.c */ ++long aufs_ioctl_dir(struct file *file, unsigned int cmd, unsigned long arg); ++ ++#ifdef CONFIG_AUFS_RDU ++/* rdu.c */ ++long au_rdu_ioctl(struct file *file, unsigned int cmd, unsigned long arg); ++#else ++static inline long au_rdu_ioctl(struct file *file, unsigned int cmd, ++ unsigned long arg) ++{ ++ return -EINVAL; ++} ++#endif ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_DIR_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/export.c linux-2.6.31/fs/aufs/export.c +--- linux-2.6.31-vanilla/fs/aufs/export.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/export.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,746 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * export via nfs ++ */ ++ ++#include ++#include ++#include ++#include ++#include ++#include ++#include "aufs.h" ++ ++union conv { ++#ifdef CONFIG_AUFS_INO_T_64 ++ __u32 a[2]; ++#else ++ __u32 a[1]; ++#endif ++ ino_t ino; ++}; ++ ++static ino_t decode_ino(__u32 *a) ++{ ++ union conv u; ++ ++ BUILD_BUG_ON(sizeof(u.ino) != sizeof(u.a)); ++ u.a[0] = a[0]; ++#ifdef CONFIG_AUFS_INO_T_64 ++ u.a[1] = a[1]; ++#endif ++ return u.ino; ++} ++ ++static void encode_ino(__u32 *a, ino_t ino) ++{ ++ union conv u; ++ ++ u.ino = ino; ++ a[0] = u.a[0]; ++#ifdef CONFIG_AUFS_INO_T_64 ++ a[1] = u.a[1]; ++#endif ++} ++ ++/* NFS file handle */ ++enum { ++ Fh_br_id, ++ Fh_sigen, ++#ifdef CONFIG_AUFS_INO_T_64 ++ /* support 64bit inode number */ ++ Fh_ino1, ++ Fh_ino2, ++ Fh_dir_ino1, ++ Fh_dir_ino2, ++#else ++ Fh_ino1, ++ Fh_dir_ino1, ++#endif ++ Fh_igen, ++ Fh_h_type, ++ Fh_tail, ++ ++ Fh_ino = Fh_ino1, ++ Fh_dir_ino = Fh_dir_ino1 ++}; ++ ++static int au_test_anon(struct dentry *dentry) ++{ ++ return !!(dentry->d_flags & DCACHE_DISCONNECTED); ++} ++ ++/* ---------------------------------------------------------------------- */ ++/* inode generation external table */ ++ ++int au_xigen_inc(struct inode *inode) ++{ ++ int err; ++ loff_t pos; ++ ssize_t sz; ++ __u32 igen; ++ struct super_block *sb; ++ struct au_sbinfo *sbinfo; ++ ++ err = 0; ++ sb = inode->i_sb; ++ sbinfo = au_sbi(sb); ++ /* ++ * temporary workaround for escaping from SiMustAnyLock() in ++ * au_mntflags(), since this function is called from au_iinfo_fin(). ++ */ ++ if (unlikely(!au_opt_test(sbinfo->si_mntflags, XINO))) ++ goto out; ++ ++ pos = inode->i_ino; ++ pos *= sizeof(igen); ++ igen = inode->i_generation + 1; ++ sz = xino_fwrite(sbinfo->si_xwrite, sbinfo->si_xigen, &igen, ++ sizeof(igen), &pos); ++ if (sz == sizeof(igen)) ++ goto out; /* success */ ++ ++ err = sz; ++ if (unlikely(sz >= 0)) { ++ err = -EIO; ++ AuIOErr("xigen error (%zd)\n", sz); ++ } ++ ++ out: ++ return err; ++} ++ ++int au_xigen_new(struct inode *inode) ++{ ++ int err; ++ loff_t pos; ++ ssize_t sz; ++ struct super_block *sb; ++ struct au_sbinfo *sbinfo; ++ struct file *file; ++ ++ err = 0; ++ /* todo: dirty, at mount time */ ++ if (inode->i_ino == AUFS_ROOT_INO) ++ goto out; ++ sb = inode->i_sb; ++ SiMustAnyLock(sb); ++ if (unlikely(!au_opt_test(au_mntflags(sb), XINO))) ++ goto out; ++ ++ err = -EFBIG; ++ pos = inode->i_ino; ++ if (unlikely(au_loff_max / sizeof(inode->i_generation) - 1 < pos)) { ++ AuIOErr1("too large i%lld\n", pos); ++ goto out; ++ } ++ pos *= sizeof(inode->i_generation); ++ ++ err = 0; ++ sbinfo = au_sbi(sb); ++ file = sbinfo->si_xigen; ++ BUG_ON(!file); ++ ++ if (i_size_read(file->f_dentry->d_inode) ++ < pos + sizeof(inode->i_generation)) { ++ inode->i_generation = atomic_inc_return(&sbinfo->si_xigen_next); ++ sz = xino_fwrite(sbinfo->si_xwrite, file, &inode->i_generation, ++ sizeof(inode->i_generation), &pos); ++ } else ++ sz = xino_fread(sbinfo->si_xread, file, &inode->i_generation, ++ sizeof(inode->i_generation), &pos); ++ if (sz == sizeof(inode->i_generation)) ++ goto out; /* success */ ++ ++ err = sz; ++ if (unlikely(sz >= 0)) { ++ err = -EIO; ++ AuIOErr("xigen error (%zd)\n", sz); ++ } ++ ++ out: ++ return err; ++} ++ ++int au_xigen_set(struct super_block *sb, struct file *base) ++{ ++ int err; ++ struct au_sbinfo *sbinfo; ++ struct file *file; ++ ++ SiMustWriteLock(sb); ++ ++ sbinfo = au_sbi(sb); ++ file = au_xino_create2(base, sbinfo->si_xigen); ++ err = PTR_ERR(file); ++ if (IS_ERR(file)) ++ goto out; ++ err = 0; ++ if (sbinfo->si_xigen) ++ fput(sbinfo->si_xigen); ++ sbinfo->si_xigen = file; ++ ++ out: ++ return err; ++} ++ ++void au_xigen_clr(struct super_block *sb) ++{ ++ struct au_sbinfo *sbinfo; ++ ++ SiMustWriteLock(sb); ++ ++ sbinfo = au_sbi(sb); ++ if (sbinfo->si_xigen) { ++ fput(sbinfo->si_xigen); ++ sbinfo->si_xigen = NULL; ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static struct dentry *decode_by_ino(struct super_block *sb, ino_t ino, ++ ino_t dir_ino) ++{ ++ struct dentry *dentry, *d; ++ struct inode *inode; ++ unsigned int sigen; ++ ++ dentry = NULL; ++ inode = ilookup(sb, ino); ++ if (!inode) ++ goto out; ++ ++ dentry = ERR_PTR(-ESTALE); ++ sigen = au_sigen(sb); ++ if (unlikely(is_bad_inode(inode) ++ || IS_DEADDIR(inode) ++ || sigen != au_iigen(inode))) ++ goto out_iput; ++ ++ dentry = NULL; ++ if (!dir_ino || S_ISDIR(inode->i_mode)) ++ dentry = d_find_alias(inode); ++ else { ++ spin_lock(&dcache_lock); ++ list_for_each_entry(d, &inode->i_dentry, d_alias) ++ if (!au_test_anon(d) ++ && d->d_parent->d_inode->i_ino == dir_ino) { ++ dentry = dget_locked(d); ++ break; ++ } ++ spin_unlock(&dcache_lock); ++ } ++ if (unlikely(dentry && sigen != au_digen(dentry))) { ++ dput(dentry); ++ dentry = ERR_PTR(-ESTALE); ++ } ++ ++ out_iput: ++ iput(inode); ++ out: ++ return dentry; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* todo: dirty? */ ++/* if exportfs_decode_fh() passed vfsmount*, we could be happy */ ++static struct vfsmount *au_mnt_get(struct super_block *sb) ++{ ++ struct mnt_namespace *ns; ++ struct vfsmount *pos, *mnt; ++ ++ spin_lock(&vfsmount_lock); ++ /* no get/put ?? */ ++ AuDebugOn(!current->nsproxy); ++ ns = current->nsproxy->mnt_ns; ++ AuDebugOn(!ns); ++ mnt = NULL; ++ /* the order (reverse) will not be a problem */ ++ list_for_each_entry(pos, &ns->list, mnt_list) ++ if (pos->mnt_sb == sb) { ++ mnt = mntget(pos); ++ break; ++ } ++ spin_unlock(&vfsmount_lock); ++ AuDebugOn(!mnt); ++ ++ return mnt; ++} ++ ++struct au_nfsd_si_lock { ++ const unsigned int sigen; ++ const aufs_bindex_t br_id; ++ unsigned char force_lock; ++}; ++ ++static aufs_bindex_t si_nfsd_read_lock(struct super_block *sb, ++ struct au_nfsd_si_lock *nsi_lock) ++{ ++ aufs_bindex_t bindex; ++ ++ si_read_lock(sb, AuLock_FLUSH); ++ ++ /* branch id may be wrapped around */ ++ bindex = au_br_index(sb, nsi_lock->br_id); ++ if (bindex >= 0 && nsi_lock->sigen + AUFS_BRANCH_MAX > au_sigen(sb)) ++ goto out; /* success */ ++ ++ if (!nsi_lock->force_lock) ++ si_read_unlock(sb); ++ bindex = -1; ++ ++ out: ++ return bindex; ++} ++ ++struct find_name_by_ino { ++ int called, found; ++ ino_t ino; ++ char *name; ++ int namelen; ++}; ++ ++static int ++find_name_by_ino(void *arg, const char *name, int namelen, loff_t offset, ++ u64 ino, unsigned int d_type) ++{ ++ struct find_name_by_ino *a = arg; ++ ++ a->called++; ++ if (a->ino != ino) ++ return 0; ++ ++ memcpy(a->name, name, namelen); ++ a->namelen = namelen; ++ a->found = 1; ++ return 1; ++} ++ ++static struct dentry *au_lkup_by_ino(struct path *path, ino_t ino, ++ struct au_nfsd_si_lock *nsi_lock) ++{ ++ struct dentry *dentry, *parent; ++ struct file *file; ++ struct inode *dir; ++ struct find_name_by_ino arg; ++ int err; ++ ++ parent = path->dentry; ++ if (nsi_lock) ++ si_read_unlock(parent->d_sb); ++ path_get(path); ++ file = vfsub_dentry_open(path, au_dir_roflags, current_cred()); ++ dentry = (void *)file; ++ if (IS_ERR(file)) ++ goto out; ++ ++ dentry = ERR_PTR(-ENOMEM); ++ arg.name = __getname(); ++ if (unlikely(!arg.name)) ++ goto out_file; ++ arg.ino = ino; ++ arg.found = 0; ++ do { ++ arg.called = 0; ++ /* smp_mb(); */ ++ err = vfsub_readdir(file, find_name_by_ino, &arg); ++ } while (!err && !arg.found && arg.called); ++ dentry = ERR_PTR(err); ++ if (unlikely(err)) ++ goto out_name; ++ dentry = ERR_PTR(-ENOENT); ++ if (!arg.found) ++ goto out_name; ++ ++ /* do not call au_lkup_one() */ ++ dir = parent->d_inode; ++ mutex_lock(&dir->i_mutex); ++ dentry = vfsub_lookup_one_len(arg.name, parent, arg.namelen); ++ mutex_unlock(&dir->i_mutex); ++ AuTraceErrPtr(dentry); ++ if (IS_ERR(dentry)) ++ goto out_name; ++ AuDebugOn(au_test_anon(dentry)); ++ if (unlikely(!dentry->d_inode)) { ++ dput(dentry); ++ dentry = ERR_PTR(-ENOENT); ++ } ++ ++ out_name: ++ __putname(arg.name); ++ out_file: ++ fput(file); ++ out: ++ if (unlikely(nsi_lock ++ && si_nfsd_read_lock(parent->d_sb, nsi_lock) < 0)) ++ if (!IS_ERR(dentry)) { ++ dput(dentry); ++ dentry = ERR_PTR(-ESTALE); ++ } ++ AuTraceErrPtr(dentry); ++ return dentry; ++} ++ ++static struct dentry *decode_by_dir_ino(struct super_block *sb, ino_t ino, ++ ino_t dir_ino, ++ struct au_nfsd_si_lock *nsi_lock) ++{ ++ struct dentry *dentry; ++ struct path path; ++ ++ if (dir_ino != AUFS_ROOT_INO) { ++ path.dentry = decode_by_ino(sb, dir_ino, 0); ++ dentry = path.dentry; ++ if (!path.dentry || IS_ERR(path.dentry)) ++ goto out; ++ AuDebugOn(au_test_anon(path.dentry)); ++ } else ++ path.dentry = dget(sb->s_root); ++ ++ path.mnt = au_mnt_get(sb); ++ dentry = au_lkup_by_ino(&path, ino, nsi_lock); ++ path_put(&path); ++ ++ out: ++ AuTraceErrPtr(dentry); ++ return dentry; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int h_acceptable(void *expv, struct dentry *dentry) ++{ ++ return 1; ++} ++ ++static char *au_build_path(struct dentry *h_parent, struct path *h_rootpath, ++ char *buf, int len, struct super_block *sb) ++{ ++ char *p; ++ int n; ++ struct path path; ++ ++ p = d_path(h_rootpath, buf, len); ++ if (IS_ERR(p)) ++ goto out; ++ n = strlen(p); ++ ++ path.mnt = h_rootpath->mnt; ++ path.dentry = h_parent; ++ p = d_path(&path, buf, len); ++ if (IS_ERR(p)) ++ goto out; ++ if (n != 1) ++ p += n; ++ ++ path.mnt = au_mnt_get(sb); ++ path.dentry = sb->s_root; ++ p = d_path(&path, buf, len - strlen(p)); ++ mntput(path.mnt); ++ if (IS_ERR(p)) ++ goto out; ++ if (n != 1) ++ p[strlen(p)] = '/'; ++ ++ out: ++ AuTraceErrPtr(p); ++ return p; ++} ++ ++static ++struct dentry *decode_by_path(struct super_block *sb, aufs_bindex_t bindex, ++ ino_t ino, __u32 *fh, int fh_len, ++ struct au_nfsd_si_lock *nsi_lock) ++{ ++ struct dentry *dentry, *h_parent, *root; ++ struct super_block *h_sb; ++ char *pathname, *p; ++ struct vfsmount *h_mnt; ++ struct au_branch *br; ++ int err; ++ struct path path; ++ ++ br = au_sbr(sb, bindex); ++ /* au_br_get(br); */ ++ h_mnt = br->br_mnt; ++ h_sb = h_mnt->mnt_sb; ++ /* todo: call lower fh_to_dentry()? fh_to_parent()? */ ++ h_parent = exportfs_decode_fh(h_mnt, (void *)(fh + Fh_tail), ++ fh_len - Fh_tail, fh[Fh_h_type], ++ h_acceptable, /*context*/NULL); ++ dentry = h_parent; ++ if (unlikely(!h_parent || IS_ERR(h_parent))) { ++ AuWarn1("%s decode_fh failed, %ld\n", ++ au_sbtype(h_sb), PTR_ERR(h_parent)); ++ goto out; ++ } ++ dentry = NULL; ++ if (unlikely(au_test_anon(h_parent))) { ++ AuWarn1("%s decode_fh returned a disconnected dentry\n", ++ au_sbtype(h_sb)); ++ goto out_h_parent; ++ } ++ ++ dentry = ERR_PTR(-ENOMEM); ++ pathname = (void *)__get_free_page(GFP_NOFS); ++ if (unlikely(!pathname)) ++ goto out_h_parent; ++ ++ root = sb->s_root; ++ path.mnt = h_mnt; ++ di_read_lock_parent(root, !AuLock_IR); ++ path.dentry = au_h_dptr(root, bindex); ++ di_read_unlock(root, !AuLock_IR); ++ p = au_build_path(h_parent, &path, pathname, PAGE_SIZE, sb); ++ dentry = (void *)p; ++ if (IS_ERR(p)) ++ goto out_pathname; ++ ++ si_read_unlock(sb); ++ err = vfsub_kern_path(p, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &path); ++ dentry = ERR_PTR(err); ++ if (unlikely(err)) ++ goto out_relock; ++ ++ dentry = ERR_PTR(-ENOENT); ++ AuDebugOn(au_test_anon(path.dentry)); ++ if (unlikely(!path.dentry->d_inode)) ++ goto out_path; ++ ++ if (ino != path.dentry->d_inode->i_ino) ++ dentry = au_lkup_by_ino(&path, ino, /*nsi_lock*/NULL); ++ else ++ dentry = dget(path.dentry); ++ ++ out_path: ++ path_put(&path); ++ out_relock: ++ if (unlikely(si_nfsd_read_lock(sb, nsi_lock) < 0)) ++ if (!IS_ERR(dentry)) { ++ dput(dentry); ++ dentry = ERR_PTR(-ESTALE); ++ } ++ out_pathname: ++ free_page((unsigned long)pathname); ++ out_h_parent: ++ dput(h_parent); ++ out: ++ /* au_br_put(br); */ ++ AuTraceErrPtr(dentry); ++ return dentry; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static struct dentry * ++aufs_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len, ++ int fh_type) ++{ ++ struct dentry *dentry; ++ __u32 *fh = fid->raw; ++ ino_t ino, dir_ino; ++ aufs_bindex_t bindex; ++ struct au_nfsd_si_lock nsi_lock = { ++ .sigen = fh[Fh_sigen], ++ .br_id = fh[Fh_br_id], ++ .force_lock = 0 ++ }; ++ ++ AuDebugOn(fh_len < Fh_tail); ++ ++ dentry = ERR_PTR(-ESTALE); ++ /* branch id may be wrapped around */ ++ bindex = si_nfsd_read_lock(sb, &nsi_lock); ++ if (unlikely(bindex < 0)) ++ goto out; ++ nsi_lock.force_lock = 1; ++ ++ /* is this inode still cached? */ ++ ino = decode_ino(fh + Fh_ino); ++ AuDebugOn(ino == AUFS_ROOT_INO); ++ dir_ino = decode_ino(fh + Fh_dir_ino); ++ dentry = decode_by_ino(sb, ino, dir_ino); ++ if (IS_ERR(dentry)) ++ goto out_unlock; ++ if (dentry) ++ goto accept; ++ ++ /* is the parent dir cached? */ ++ dentry = decode_by_dir_ino(sb, ino, dir_ino, &nsi_lock); ++ if (IS_ERR(dentry)) ++ goto out_unlock; ++ if (dentry) ++ goto accept; ++ ++ /* lookup path */ ++ dentry = decode_by_path(sb, bindex, ino, fh, fh_len, &nsi_lock); ++ if (IS_ERR(dentry)) ++ goto out_unlock; ++ if (unlikely(!dentry)) ++ /* todo?: make it ESTALE */ ++ goto out_unlock; ++ ++ accept: ++ if (dentry->d_inode->i_generation == fh[Fh_igen]) ++ goto out_unlock; /* success */ ++ ++ dput(dentry); ++ dentry = ERR_PTR(-ESTALE); ++ out_unlock: ++ si_read_unlock(sb); ++ out: ++ AuTraceErrPtr(dentry); ++ return dentry; ++} ++ ++#if 0 /* reserved for future use */ ++/* support subtreecheck option */ ++static struct dentry *aufs_fh_to_parent(struct super_block *sb, struct fid *fid, ++ int fh_len, int fh_type) ++{ ++ struct dentry *parent; ++ __u32 *fh = fid->raw; ++ ino_t dir_ino; ++ ++ dir_ino = decode_ino(fh + Fh_dir_ino); ++ parent = decode_by_ino(sb, dir_ino, 0); ++ if (IS_ERR(parent)) ++ goto out; ++ if (!parent) ++ parent = decode_by_path(sb, au_br_index(sb, fh[Fh_br_id]), ++ dir_ino, fh, fh_len); ++ ++ out: ++ AuTraceErrPtr(parent); ++ return parent; ++} ++#endif ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int aufs_encode_fh(struct dentry *dentry, __u32 *fh, int *max_len, ++ int connectable) ++{ ++ int err; ++ aufs_bindex_t bindex, bend; ++ struct super_block *sb, *h_sb; ++ struct inode *inode; ++ struct dentry *parent, *h_parent; ++ struct au_branch *br; ++ ++ AuDebugOn(au_test_anon(dentry)); ++ ++ parent = NULL; ++ err = -ENOSPC; ++ if (unlikely(*max_len <= Fh_tail)) { ++ AuWarn1("NFSv2 client (max_len %d)?\n", *max_len); ++ goto out; ++ } ++ ++ err = FILEID_ROOT; ++ if (IS_ROOT(dentry)) { ++ AuDebugOn(dentry->d_inode->i_ino != AUFS_ROOT_INO); ++ goto out; ++ } ++ ++ err = -EIO; ++ h_parent = NULL; ++ sb = dentry->d_sb; ++ aufs_read_lock(dentry, AuLock_FLUSH | AuLock_IR); ++ parent = dget_parent(dentry); ++ di_read_lock_parent(parent, !AuLock_IR); ++ inode = dentry->d_inode; ++ AuDebugOn(!inode); ++#ifdef CONFIG_AUFS_DEBUG ++ if (unlikely(!au_opt_test(au_mntflags(sb), XINO))) ++ AuWarn1("NFS-exporting requires xino\n"); ++#endif ++ ++ bend = au_dbtaildir(parent); ++ for (bindex = au_dbstart(parent); bindex <= bend; bindex++) { ++ h_parent = au_h_dptr(parent, bindex); ++ if (h_parent) { ++ dget(h_parent); ++ break; ++ } ++ } ++ if (unlikely(!h_parent)) ++ goto out_unlock; ++ ++ err = -EPERM; ++ br = au_sbr(sb, bindex); ++ h_sb = br->br_mnt->mnt_sb; ++ if (unlikely(!h_sb->s_export_op)) { ++ AuErr1("%s branch is not exportable\n", au_sbtype(h_sb)); ++ goto out_dput; ++ } ++ ++ fh[Fh_br_id] = br->br_id; ++ fh[Fh_sigen] = au_sigen(sb); ++ encode_ino(fh + Fh_ino, inode->i_ino); ++ encode_ino(fh + Fh_dir_ino, parent->d_inode->i_ino); ++ fh[Fh_igen] = inode->i_generation; ++ ++ *max_len -= Fh_tail; ++ fh[Fh_h_type] = exportfs_encode_fh(h_parent, (void *)(fh + Fh_tail), ++ max_len, ++ /*connectable or subtreecheck*/0); ++ err = fh[Fh_h_type]; ++ *max_len += Fh_tail; ++ /* todo: macros? */ ++ if (err != 255) ++ err = 99; ++ else ++ AuWarn1("%s encode_fh failed\n", au_sbtype(h_sb)); ++ ++ out_dput: ++ dput(h_parent); ++ out_unlock: ++ di_read_unlock(parent, !AuLock_IR); ++ dput(parent); ++ aufs_read_unlock(dentry, AuLock_IR); ++ out: ++ if (unlikely(err < 0)) ++ err = 255; ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static struct export_operations aufs_export_op = { ++ .fh_to_dentry = aufs_fh_to_dentry, ++ /* .fh_to_parent = aufs_fh_to_parent, */ ++ .encode_fh = aufs_encode_fh ++}; ++ ++void au_export_init(struct super_block *sb) ++{ ++ struct au_sbinfo *sbinfo; ++ __u32 u; ++ ++ sb->s_export_op = &aufs_export_op; ++ sbinfo = au_sbi(sb); ++ sbinfo->si_xigen = NULL; ++ get_random_bytes(&u, sizeof(u)); ++ BUILD_BUG_ON(sizeof(u) != sizeof(int)); ++ atomic_set(&sbinfo->si_xigen_next, u); ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/file.c linux-2.6.31/fs/aufs/file.c +--- linux-2.6.31-vanilla/fs/aufs/file.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/file.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,568 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * handling file/dir, and address_space operation ++ */ ++ ++#include ++#include ++#include ++#include ++#include "aufs.h" ++ ++/* drop flags for writing */ ++unsigned int au_file_roflags(unsigned int flags) ++{ ++ flags &= ~(O_WRONLY | O_RDWR | O_APPEND | O_CREAT | O_TRUNC); ++ flags |= O_RDONLY | O_NOATIME; ++ return flags; ++} ++ ++/* common functions to regular file and dir */ ++struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags, ++ struct file *file) ++{ ++ struct file *h_file; ++ struct dentry *h_dentry; ++ struct inode *h_inode; ++ struct super_block *sb; ++ struct au_branch *br; ++ int err, exec_flag; ++ struct path h_path; ++ ++ /* a race condition can happen between open and unlink/rmdir */ ++ h_file = ERR_PTR(-ENOENT); ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (au_test_nfsd(current) && !h_dentry) ++ goto out; ++ h_inode = h_dentry->d_inode; ++ if (au_test_nfsd(current) && !h_inode) ++ goto out; ++ if (unlikely((!d_unhashed(dentry) && d_unhashed(h_dentry)) ++ || !h_inode)) ++ goto out; ++ ++ sb = dentry->d_sb; ++ br = au_sbr(sb, bindex); ++ h_file = ERR_PTR(-EACCES); ++ exec_flag = flags & vfsub_fmode_to_uint(FMODE_EXEC); ++ if (exec_flag && (br->br_mnt->mnt_flags & MNT_NOEXEC)) ++ goto out; ++ ++ /* drop flags for writing */ ++ if (au_test_ro(sb, bindex, dentry->d_inode)) ++ flags = au_file_roflags(flags); ++ flags &= ~O_CREAT; ++ atomic_inc(&br->br_count); ++ h_path.dentry = h_dentry; ++ h_path.mnt = br->br_mnt; ++ path_get(&h_path); ++ h_file = vfsub_dentry_open(&h_path, flags, current_cred()); ++ if (IS_ERR(h_file)) ++ goto out_br; ++ ++ if (exec_flag) { ++ err = deny_write_access(h_file); ++ if (unlikely(err)) { ++ fput(h_file); ++ h_file = ERR_PTR(err); ++ goto out_br; ++ } ++ } ++ fsnotify_open(h_dentry); ++ goto out; /* success */ ++ ++ out_br: ++ atomic_dec(&br->br_count); ++ out: ++ return h_file; ++} ++ ++int au_do_open(struct file *file, int (*open)(struct file *file, int flags)) ++{ ++ int err; ++ unsigned int flags; ++ struct dentry *dentry; ++ struct super_block *sb; ++ ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ err = au_finfo_init(file); ++ if (unlikely(err)) ++ goto out; ++ ++ di_read_lock_child(dentry, AuLock_IR); ++ spin_lock(&file->f_lock); ++ flags = file->f_flags; ++ spin_unlock(&file->f_lock); ++ err = open(file, flags); ++ di_read_unlock(dentry, AuLock_IR); ++ ++ fi_write_unlock(file); ++ if (unlikely(err)) ++ au_finfo_fin(file); ++ out: ++ si_read_unlock(sb); ++ return err; ++} ++ ++int au_reopen_nondir(struct file *file) ++{ ++ int err; ++ unsigned int flags; ++ aufs_bindex_t bstart, bindex, bend; ++ struct dentry *dentry; ++ struct file *h_file, *h_file_tmp; ++ ++ dentry = file->f_dentry; ++ bstart = au_dbstart(dentry); ++ h_file_tmp = NULL; ++ if (au_fbstart(file) == bstart) { ++ h_file = au_h_fptr(file, bstart); ++ if (file->f_mode == h_file->f_mode) ++ return 0; /* success */ ++ h_file_tmp = h_file; ++ get_file(h_file_tmp); ++ au_set_h_fptr(file, bstart, NULL); ++ } ++ AuDebugOn(au_fbstart(file) < bstart ++ || au_fi(file)->fi_hfile[0 + bstart].hf_file); ++ ++ spin_lock(&file->f_lock); ++ flags = file->f_flags & ~O_TRUNC; ++ spin_unlock(&file->f_lock); ++ h_file = au_h_open(dentry, bstart, flags, file); ++ err = PTR_ERR(h_file); ++ if (IS_ERR(h_file)) ++ goto out; /* todo: close all? */ ++ ++ err = 0; ++ au_set_fbstart(file, bstart); ++ au_set_h_fptr(file, bstart, h_file); ++ au_update_figen(file); ++ /* todo: necessary? */ ++ /* file->f_ra = h_file->f_ra; */ ++ ++ /* close lower files */ ++ bend = au_fbend(file); ++ for (bindex = bstart + 1; bindex <= bend; bindex++) ++ au_set_h_fptr(file, bindex, NULL); ++ au_set_fbend(file, bstart); ++ ++ out: ++ if (h_file_tmp) ++ fput(h_file_tmp); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int au_reopen_wh(struct file *file, aufs_bindex_t btgt, ++ struct dentry *hi_wh) ++{ ++ int err; ++ aufs_bindex_t bstart; ++ struct au_dinfo *dinfo; ++ struct dentry *h_dentry; ++ ++ dinfo = au_di(file->f_dentry); ++ AuRwMustWriteLock(&dinfo->di_rwsem); ++ ++ bstart = dinfo->di_bstart; ++ dinfo->di_bstart = btgt; ++ h_dentry = dinfo->di_hdentry[0 + btgt].hd_dentry; ++ dinfo->di_hdentry[0 + btgt].hd_dentry = hi_wh; ++ err = au_reopen_nondir(file); ++ dinfo->di_hdentry[0 + btgt].hd_dentry = h_dentry; ++ dinfo->di_bstart = bstart; ++ ++ return err; ++} ++ ++static int au_ready_to_write_wh(struct file *file, loff_t len, ++ aufs_bindex_t bcpup) ++{ ++ int err; ++ struct inode *inode; ++ struct dentry *dentry, *hi_wh; ++ struct super_block *sb; ++ ++ dentry = file->f_dentry; ++ inode = dentry->d_inode; ++ hi_wh = au_hi_wh(inode, bcpup); ++ if (!hi_wh) ++ err = au_sio_cpup_wh(dentry, bcpup, len, file); ++ else ++ /* already copied-up after unlink */ ++ err = au_reopen_wh(file, bcpup, hi_wh); ++ ++ sb = dentry->d_sb; ++ if (!err && inode->i_nlink > 1 && au_opt_test(au_mntflags(sb), PLINK)) ++ au_plink_append(inode, bcpup, au_h_dptr(dentry, bcpup)); ++ ++ return err; ++} ++ ++/* ++ * prepare the @file for writing. ++ */ ++int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin) ++{ ++ int err; ++ aufs_bindex_t bstart, bcpup; ++ struct dentry *dentry, *parent, *h_dentry; ++ struct inode *h_inode, *inode; ++ struct super_block *sb; ++ ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ bstart = au_fbstart(file); ++ inode = dentry->d_inode; ++ err = au_test_ro(sb, bstart, inode); ++ if (!err && (au_h_fptr(file, bstart)->f_mode & FMODE_WRITE)) { ++ err = au_pin(pin, dentry, bstart, AuOpt_UDBA_NONE, /*flags*/0); ++ goto out; ++ } ++ ++ /* need to cpup */ ++ parent = dget_parent(dentry); ++ di_write_lock_parent(parent); ++ err = AuWbrCopyup(au_sbi(sb), dentry); ++ bcpup = err; ++ if (unlikely(err < 0)) ++ goto out_dgrade; ++ err = 0; ++ ++ if (!au_h_dptr(parent, bcpup)) { ++ err = au_cpup_dirs(dentry, bcpup); ++ if (unlikely(err)) ++ goto out_dgrade; ++ } ++ ++ err = au_pin(pin, dentry, bcpup, AuOpt_UDBA_NONE, ++ AuPin_DI_LOCKED | AuPin_MNT_WRITE); ++ if (unlikely(err)) ++ goto out_dgrade; ++ ++ h_dentry = au_h_fptr(file, bstart)->f_dentry; ++ h_inode = h_dentry->d_inode; ++ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); ++ if (d_unhashed(dentry) /* || d_unhashed(h_dentry) */ ++ /* || !h_inode->i_nlink */) { ++ err = au_ready_to_write_wh(file, len, bcpup); ++ di_downgrade_lock(parent, AuLock_IR); ++ } else { ++ di_downgrade_lock(parent, AuLock_IR); ++ if (!au_h_dptr(dentry, bcpup)) ++ err = au_sio_cpup_simple(dentry, bcpup, len, ++ AuCpup_DTIME); ++ if (!err) ++ err = au_reopen_nondir(file); ++ } ++ mutex_unlock(&h_inode->i_mutex); ++ ++ if (!err) { ++ au_pin_set_parent_lflag(pin, /*lflag*/0); ++ goto out_dput; /* success */ ++ } ++ au_unpin(pin); ++ goto out_unlock; ++ ++ out_dgrade: ++ di_downgrade_lock(parent, AuLock_IR); ++ out_unlock: ++ di_read_unlock(parent, AuLock_IR); ++ out_dput: ++ dput(parent); ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int au_file_refresh_by_inode(struct file *file, int *need_reopen) ++{ ++ int err; ++ aufs_bindex_t bstart; ++ struct au_pin pin; ++ struct au_finfo *finfo; ++ struct dentry *dentry, *parent, *hi_wh; ++ struct inode *inode; ++ struct super_block *sb; ++ ++ FiMustWriteLock(file); ++ ++ err = 0; ++ finfo = au_fi(file); ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ inode = dentry->d_inode; ++ bstart = au_ibstart(inode); ++ if (bstart == finfo->fi_bstart) ++ goto out; ++ ++ parent = dget_parent(dentry); ++ if (au_test_ro(sb, bstart, inode)) { ++ di_read_lock_parent(parent, !AuLock_IR); ++ err = AuWbrCopyup(au_sbi(sb), dentry); ++ bstart = err; ++ di_read_unlock(parent, !AuLock_IR); ++ if (unlikely(err < 0)) ++ goto out_parent; ++ err = 0; ++ } ++ ++ di_read_lock_parent(parent, AuLock_IR); ++ hi_wh = au_hi_wh(inode, bstart); ++ if (au_opt_test(au_mntflags(sb), PLINK) ++ && au_plink_test(inode) ++ && !d_unhashed(dentry)) { ++ err = au_test_and_cpup_dirs(dentry, bstart); ++ if (unlikely(err)) ++ goto out_unlock; ++ ++ /* always superio. */ ++ err = au_pin(&pin, dentry, bstart, AuOpt_UDBA_NONE, ++ AuPin_DI_LOCKED | AuPin_MNT_WRITE); ++ if (!err) ++ err = au_sio_cpup_simple(dentry, bstart, -1, ++ AuCpup_DTIME); ++ au_unpin(&pin); ++ } else if (hi_wh) { ++ /* already copied-up after unlink */ ++ err = au_reopen_wh(file, bstart, hi_wh); ++ *need_reopen = 0; ++ } ++ ++ out_unlock: ++ di_read_unlock(parent, AuLock_IR); ++ out_parent: ++ dput(parent); ++ out: ++ return err; ++} ++ ++static void au_do_refresh_file(struct file *file) ++{ ++ aufs_bindex_t bindex, bend, new_bindex, brid; ++ struct au_hfile *p, tmp, *q; ++ struct au_finfo *finfo; ++ struct super_block *sb; ++ ++ FiMustWriteLock(file); ++ ++ sb = file->f_dentry->d_sb; ++ finfo = au_fi(file); ++ p = finfo->fi_hfile + finfo->fi_bstart; ++ brid = p->hf_br->br_id; ++ bend = finfo->fi_bend; ++ for (bindex = finfo->fi_bstart; bindex <= bend; bindex++, p++) { ++ if (!p->hf_file) ++ continue; ++ ++ new_bindex = au_br_index(sb, p->hf_br->br_id); ++ if (new_bindex == bindex) ++ continue; ++ if (new_bindex < 0) { ++ au_set_h_fptr(file, bindex, NULL); ++ continue; ++ } ++ ++ /* swap two lower inode, and loop again */ ++ q = finfo->fi_hfile + new_bindex; ++ tmp = *q; ++ *q = *p; ++ *p = tmp; ++ if (tmp.hf_file) { ++ bindex--; ++ p--; ++ } ++ } ++ ++ p = finfo->fi_hfile; ++ if (!au_test_mmapped(file) && !d_unhashed(file->f_dentry)) { ++ bend = au_sbend(sb); ++ for (finfo->fi_bstart = 0; finfo->fi_bstart <= bend; ++ finfo->fi_bstart++, p++) ++ if (p->hf_file) { ++ if (p->hf_file->f_dentry ++ && p->hf_file->f_dentry->d_inode) ++ break; ++ else ++ au_hfput(p, file); ++ } ++ } else { ++ bend = au_br_index(sb, brid); ++ for (finfo->fi_bstart = 0; finfo->fi_bstart < bend; ++ finfo->fi_bstart++, p++) ++ if (p->hf_file) ++ au_hfput(p, file); ++ bend = au_sbend(sb); ++ } ++ ++ p = finfo->fi_hfile + bend; ++ for (finfo->fi_bend = bend; finfo->fi_bend >= finfo->fi_bstart; ++ finfo->fi_bend--, p--) ++ if (p->hf_file) { ++ if (p->hf_file->f_dentry ++ && p->hf_file->f_dentry->d_inode) ++ break; ++ else ++ au_hfput(p, file); ++ } ++ AuDebugOn(finfo->fi_bend < finfo->fi_bstart); ++} ++ ++/* ++ * after branch manipulating, refresh the file. ++ */ ++static int refresh_file(struct file *file, int (*reopen)(struct file *file)) ++{ ++ int err, need_reopen; ++ struct dentry *dentry; ++ aufs_bindex_t bend, bindex; ++ ++ dentry = file->f_dentry; ++ err = au_fi_realloc(au_fi(file), au_sbend(dentry->d_sb) + 1); ++ if (unlikely(err)) ++ goto out; ++ au_do_refresh_file(file); ++ ++ err = 0; ++ need_reopen = 1; ++ if (!au_test_mmapped(file)) ++ err = au_file_refresh_by_inode(file, &need_reopen); ++ if (!err && need_reopen && !d_unhashed(dentry)) ++ err = reopen(file); ++ if (!err) { ++ au_update_figen(file); ++ return 0; /* success */ ++ } ++ ++ /* error, close all lower files */ ++ bend = au_fbend(file); ++ for (bindex = au_fbstart(file); bindex <= bend; bindex++) ++ au_set_h_fptr(file, bindex, NULL); ++ ++ out: ++ return err; ++} ++ ++/* common function to regular file and dir */ ++int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file), ++ int wlock) ++{ ++ int err; ++ unsigned int sigen, figen; ++ aufs_bindex_t bstart; ++ unsigned char pseudo_link; ++ struct dentry *dentry; ++ ++ err = 0; ++ dentry = file->f_dentry; ++ sigen = au_sigen(dentry->d_sb); ++ fi_write_lock(file); ++ figen = au_figen(file); ++ di_write_lock_child(dentry); ++ bstart = au_dbstart(dentry); ++ pseudo_link = (bstart != au_ibstart(dentry->d_inode)); ++ if (sigen == figen && !pseudo_link && au_fbstart(file) == bstart) { ++ if (!wlock) { ++ di_downgrade_lock(dentry, AuLock_IR); ++ fi_downgrade_lock(file); ++ } ++ goto out; /* success */ ++ } ++ ++ AuDbg("sigen %d, figen %d\n", sigen, figen); ++ if (sigen != au_digen(dentry) ++ || sigen != au_iigen(dentry->d_inode)) { ++ err = au_reval_dpath(dentry, sigen); ++ if (unlikely(err < 0)) ++ goto out; ++ AuDebugOn(au_digen(dentry) != sigen ++ || au_iigen(dentry->d_inode) != sigen); ++ } ++ ++ err = refresh_file(file, reopen); ++ if (!err) { ++ if (!wlock) { ++ di_downgrade_lock(dentry, AuLock_IR); ++ fi_downgrade_lock(file); ++ } ++ } else { ++ di_write_unlock(dentry); ++ fi_write_unlock(file); ++ } ++ ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* cf. aufs_nopage() */ ++/* for madvise(2) */ ++static int aufs_readpage(struct file *file __maybe_unused, struct page *page) ++{ ++ unlock_page(page); ++ return 0; ++} ++ ++/* they will never be called. */ ++#ifdef CONFIG_AUFS_DEBUG ++static int aufs_write_begin(struct file *file, struct address_space *mapping, ++ loff_t pos, unsigned len, unsigned flags, ++ struct page **pagep, void **fsdata) ++{ AuUnsupport(); return 0; } ++static int aufs_write_end(struct file *file, struct address_space *mapping, ++ loff_t pos, unsigned len, unsigned copied, ++ struct page *page, void *fsdata) ++{ AuUnsupport(); return 0; } ++static int aufs_writepage(struct page *page, struct writeback_control *wbc) ++{ AuUnsupport(); return 0; } ++static void aufs_sync_page(struct page *page) ++{ AuUnsupport(); } ++ ++static int aufs_set_page_dirty(struct page *page) ++{ AuUnsupport(); return 0; } ++static void aufs_invalidatepage(struct page *page, unsigned long offset) ++{ AuUnsupport(); } ++static int aufs_releasepage(struct page *page, gfp_t gfp) ++{ AuUnsupport(); return 0; } ++static ssize_t aufs_direct_IO(int rw, struct kiocb *iocb, ++ const struct iovec *iov, loff_t offset, ++ unsigned long nr_segs) ++{ AuUnsupport(); return 0; } ++#endif /* CONFIG_AUFS_DEBUG */ ++ ++struct address_space_operations aufs_aop = { ++ .readpage = aufs_readpage, ++#ifdef CONFIG_AUFS_DEBUG ++ .writepage = aufs_writepage, ++ .sync_page = aufs_sync_page, ++ .set_page_dirty = aufs_set_page_dirty, ++ .write_begin = aufs_write_begin, ++ .write_end = aufs_write_end, ++ .invalidatepage = aufs_invalidatepage, ++ .releasepage = aufs_releasepage, ++ .direct_IO = aufs_direct_IO, ++#endif /* CONFIG_AUFS_DEBUG */ ++}; +diff -Nur linux-2.6.31-vanilla/fs/aufs/file.h linux-2.6.31/fs/aufs/file.h +--- linux-2.6.31-vanilla/fs/aufs/file.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/file.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,174 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * file operations ++ */ ++ ++#ifndef __AUFS_FILE_H__ ++#define __AUFS_FILE_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++#include ++#include "rwsem.h" ++ ++struct au_branch; ++struct au_hfile { ++ struct file *hf_file; ++ struct au_branch *hf_br; ++}; ++ ++struct au_vdir; ++struct au_finfo { ++ atomic_t fi_generation; ++ ++ struct au_rwsem fi_rwsem; ++ struct au_hfile *fi_hfile; ++ aufs_bindex_t fi_bstart, fi_bend; ++ ++ union { ++ /* non-dir only */ ++ struct { ++ struct vm_operations_struct *fi_h_vm_ops; ++ struct vm_operations_struct *fi_vm_ops; ++ }; ++ ++ /* dir only */ ++ struct { ++ struct au_vdir *fi_vdir_cache; ++ int fi_maintain_plink; ++ }; ++ }; ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* file.c */ ++extern struct address_space_operations aufs_aop; ++unsigned int au_file_roflags(unsigned int flags); ++struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags, ++ struct file *file); ++int au_do_open(struct file *file, int (*open)(struct file *file, int flags)); ++int au_reopen_nondir(struct file *file); ++struct au_pin; ++int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin); ++int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file), ++ int wlock); ++ ++/* poll.c */ ++#ifdef CONFIG_AUFS_POLL ++unsigned int aufs_poll(struct file *file, poll_table *wait); ++#endif ++ ++/* f_op.c */ ++extern const struct file_operations aufs_file_fop; ++int aufs_flush(struct file *file, fl_owner_t id); ++ ++/* finfo.c */ ++void au_hfput(struct au_hfile *hf, struct file *file); ++void au_set_h_fptr(struct file *file, aufs_bindex_t bindex, ++ struct file *h_file); ++ ++void au_update_figen(struct file *file); ++ ++void au_finfo_fin(struct file *file); ++int au_finfo_init(struct file *file); ++int au_fi_realloc(struct au_finfo *finfo, int nbr); ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline struct au_finfo *au_fi(struct file *file) ++{ ++ return file->private_data; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * fi_read_lock, fi_write_lock, ++ * fi_read_unlock, fi_write_unlock, fi_downgrade_lock ++ */ ++AuSimpleRwsemFuncs(fi, struct file *f, &au_fi(f)->fi_rwsem); ++ ++#define FiMustNoWaiters(f) AuRwMustNoWaiters(&au_fi(f)->fi_rwsem) ++#define FiMustAnyLock(f) AuRwMustAnyLock(&au_fi(f)->fi_rwsem) ++#define FiMustWriteLock(f) AuRwMustWriteLock(&au_fi(f)->fi_rwsem) ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* todo: hard/soft set? */ ++static inline aufs_bindex_t au_fbstart(struct file *file) ++{ ++ FiMustAnyLock(file); ++ return au_fi(file)->fi_bstart; ++} ++ ++static inline aufs_bindex_t au_fbend(struct file *file) ++{ ++ FiMustAnyLock(file); ++ return au_fi(file)->fi_bend; ++} ++ ++static inline struct au_vdir *au_fvdir_cache(struct file *file) ++{ ++ FiMustAnyLock(file); ++ return au_fi(file)->fi_vdir_cache; ++} ++ ++static inline void au_set_fbstart(struct file *file, aufs_bindex_t bindex) ++{ ++ FiMustWriteLock(file); ++ au_fi(file)->fi_bstart = bindex; ++} ++ ++static inline void au_set_fbend(struct file *file, aufs_bindex_t bindex) ++{ ++ FiMustWriteLock(file); ++ au_fi(file)->fi_bend = bindex; ++} ++ ++static inline void au_set_fvdir_cache(struct file *file, ++ struct au_vdir *vdir_cache) ++{ ++ FiMustWriteLock(file); ++ au_fi(file)->fi_vdir_cache = vdir_cache; ++} ++ ++static inline struct file *au_h_fptr(struct file *file, aufs_bindex_t bindex) ++{ ++ FiMustAnyLock(file); ++ return au_fi(file)->fi_hfile[0 + bindex].hf_file; ++} ++ ++/* todo: memory barrier? */ ++static inline unsigned int au_figen(struct file *f) ++{ ++ return atomic_read(&au_fi(f)->fi_generation); ++} ++ ++static inline int au_test_mmapped(struct file *f) ++{ ++ /* FiMustAnyLock(f); */ ++ return !!(au_fi(f)->fi_h_vm_ops); ++} ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_FILE_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/finfo.c linux-2.6.31/fs/aufs/finfo.c +--- linux-2.6.31-vanilla/fs/aufs/finfo.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/finfo.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,128 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * file private data ++ */ ++ ++#include ++#include "aufs.h" ++ ++void au_hfput(struct au_hfile *hf, struct file *file) ++{ ++ if (file->f_flags & vfsub_fmode_to_uint(FMODE_EXEC)) ++ allow_write_access(hf->hf_file); ++ fput(hf->hf_file); ++ hf->hf_file = NULL; ++ atomic_dec_return(&hf->hf_br->br_count); ++ hf->hf_br = NULL; ++} ++ ++void au_set_h_fptr(struct file *file, aufs_bindex_t bindex, struct file *val) ++{ ++ struct au_finfo *finfo = au_fi(file); ++ struct au_hfile *hf; ++ ++ hf = finfo->fi_hfile + bindex; ++ if (hf->hf_file) ++ au_hfput(hf, file); ++ if (val) { ++ hf->hf_file = val; ++ hf->hf_br = au_sbr(file->f_dentry->d_sb, bindex); ++ } ++} ++ ++void au_update_figen(struct file *file) ++{ ++ atomic_set(&au_fi(file)->fi_generation, au_digen(file->f_dentry)); ++ /* smp_mb(); */ /* atomic_set */ ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void au_finfo_fin(struct file *file) ++{ ++ struct au_finfo *finfo; ++ aufs_bindex_t bindex, bend; ++ ++ fi_write_lock(file); ++ bend = au_fbend(file); ++ bindex = au_fbstart(file); ++ if (bindex >= 0) ++ /* ++ * calls fput() instead of filp_close(), ++ * since no dnotify or lock for the lower file. ++ */ ++ for (; bindex <= bend; bindex++) ++ au_set_h_fptr(file, bindex, NULL); ++ ++ finfo = au_fi(file); ++ au_dbg_verify_hf(finfo); ++ kfree(finfo->fi_hfile); ++ fi_write_unlock(file); ++ AuRwDestroy(&finfo->fi_rwsem); ++ au_cache_free_finfo(finfo); ++} ++ ++int au_finfo_init(struct file *file) ++{ ++ struct au_finfo *finfo; ++ struct dentry *dentry; ++ ++ dentry = file->f_dentry; ++ finfo = au_cache_alloc_finfo(); ++ if (unlikely(!finfo)) ++ goto out; ++ ++ finfo->fi_hfile = kcalloc(au_sbend(dentry->d_sb) + 1, ++ sizeof(*finfo->fi_hfile), GFP_NOFS); ++ if (unlikely(!finfo->fi_hfile)) ++ goto out_finfo; ++ ++ au_rw_init_wlock(&finfo->fi_rwsem); ++ finfo->fi_bstart = -1; ++ finfo->fi_bend = -1; ++ atomic_set(&finfo->fi_generation, au_digen(dentry)); ++ /* smp_mb(); */ /* atomic_set */ ++ ++ file->private_data = finfo; ++ return 0; /* success */ ++ ++ out_finfo: ++ au_cache_free_finfo(finfo); ++ out: ++ return -ENOMEM; ++} ++ ++int au_fi_realloc(struct au_finfo *finfo, int nbr) ++{ ++ int err, sz; ++ struct au_hfile *hfp; ++ ++ err = -ENOMEM; ++ sz = sizeof(*hfp) * (finfo->fi_bend + 1); ++ if (!sz) ++ sz = sizeof(*hfp); ++ hfp = au_kzrealloc(finfo->fi_hfile, sz, sizeof(*hfp) * nbr, GFP_NOFS); ++ if (hfp) { ++ finfo->fi_hfile = hfp; ++ err = 0; ++ } ++ ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/f_op.c linux-2.6.31/fs/aufs/f_op.c +--- linux-2.6.31-vanilla/fs/aufs/f_op.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/f_op.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,823 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * file and vm operations ++ */ ++ ++#include ++#include ++#include ++#include ++#include ++#include ++#include "aufs.h" ++ ++/* common function to regular file and dir */ ++int aufs_flush(struct file *file, fl_owner_t id) ++{ ++ int err; ++ aufs_bindex_t bindex, bend; ++ struct dentry *dentry; ++ struct file *h_file; ++ ++ dentry = file->f_dentry; ++ si_noflush_read_lock(dentry->d_sb); ++ fi_read_lock(file); ++ di_read_lock_child(dentry, AuLock_IW); ++ ++ err = 0; ++ bend = au_fbend(file); ++ for (bindex = au_fbstart(file); !err && bindex <= bend; bindex++) { ++ h_file = au_h_fptr(file, bindex); ++ if (!h_file || !h_file->f_op || !h_file->f_op->flush) ++ continue; ++ ++ err = h_file->f_op->flush(h_file, id); ++ if (!err) ++ vfsub_update_h_iattr(&h_file->f_path, /*did*/NULL); ++ /*ignore*/ ++ } ++ au_cpup_attr_timesizes(dentry->d_inode); ++ ++ di_read_unlock(dentry, AuLock_IW); ++ fi_read_unlock(file); ++ si_read_unlock(dentry->d_sb); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int do_open_nondir(struct file *file, int flags) ++{ ++ int err; ++ aufs_bindex_t bindex; ++ struct file *h_file; ++ struct dentry *dentry; ++ struct au_finfo *finfo; ++ ++ FiMustWriteLock(file); ++ ++ err = 0; ++ dentry = file->f_dentry; ++ finfo = au_fi(file); ++ finfo->fi_h_vm_ops = NULL; ++ finfo->fi_vm_ops = NULL; ++ bindex = au_dbstart(dentry); ++ /* O_TRUNC is processed already */ ++ BUG_ON(au_test_ro(dentry->d_sb, bindex, dentry->d_inode) ++ && (flags & O_TRUNC)); ++ ++ h_file = au_h_open(dentry, bindex, flags, file); ++ if (IS_ERR(h_file)) ++ err = PTR_ERR(h_file); ++ else { ++ au_set_fbstart(file, bindex); ++ au_set_fbend(file, bindex); ++ au_set_h_fptr(file, bindex, h_file); ++ au_update_figen(file); ++ /* todo: necessary? */ ++ /* file->f_ra = h_file->f_ra; */ ++ } ++ return err; ++} ++ ++static int aufs_open_nondir(struct inode *inode __maybe_unused, ++ struct file *file) ++{ ++ return au_do_open(file, do_open_nondir); ++} ++ ++static int aufs_release_nondir(struct inode *inode __maybe_unused, ++ struct file *file) ++{ ++ struct super_block *sb = file->f_dentry->d_sb; ++ ++ si_noflush_read_lock(sb); ++ kfree(au_fi(file)->fi_vm_ops); ++ au_finfo_fin(file); ++ si_read_unlock(sb); ++ return 0; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static ssize_t aufs_read(struct file *file, char __user *buf, size_t count, ++ loff_t *ppos) ++{ ++ ssize_t err; ++ struct dentry *dentry; ++ struct file *h_file; ++ struct super_block *sb; ++ ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); ++ if (unlikely(err)) ++ goto out; ++ ++ h_file = au_h_fptr(file, au_fbstart(file)); ++ err = vfsub_read_u(h_file, buf, count, ppos); ++ /* todo: necessary? */ ++ /* file->f_ra = h_file->f_ra; */ ++ fsstack_copy_attr_atime(dentry->d_inode, h_file->f_dentry->d_inode); ++ ++ di_read_unlock(dentry, AuLock_IR); ++ fi_read_unlock(file); ++ out: ++ si_read_unlock(sb); ++ return err; ++} ++ ++static ssize_t aufs_write(struct file *file, const char __user *ubuf, ++ size_t count, loff_t *ppos) ++{ ++ ssize_t err; ++ aufs_bindex_t bstart; ++ struct au_pin pin; ++ struct dentry *dentry; ++ struct inode *inode; ++ struct super_block *sb; ++ struct file *h_file; ++ char __user *buf = (char __user *)ubuf; ++ ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ inode = dentry->d_inode; ++ mutex_lock(&inode->i_mutex); ++ si_read_lock(sb, AuLock_FLUSH); ++ ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); ++ if (unlikely(err)) ++ goto out; ++ ++ err = au_ready_to_write(file, -1, &pin); ++ di_downgrade_lock(dentry, AuLock_IR); ++ if (unlikely(err)) ++ goto out_unlock; ++ ++ bstart = au_fbstart(file); ++ h_file = au_h_fptr(file, bstart); ++ au_unpin(&pin); ++ err = vfsub_write_u(h_file, buf, count, ppos); ++ au_cpup_attr_timesizes(inode); ++ inode->i_mode = h_file->f_dentry->d_inode->i_mode; ++ ++ out_unlock: ++ di_read_unlock(dentry, AuLock_IR); ++ fi_write_unlock(file); ++ out: ++ si_read_unlock(sb); ++ mutex_unlock(&inode->i_mutex); ++ return err; ++} ++ ++static ssize_t aufs_aio_read(struct kiocb *kio, const struct iovec *iov, ++ unsigned long nv, loff_t pos) ++{ ++ ssize_t err; ++ struct file *file, *h_file; ++ struct dentry *dentry; ++ struct super_block *sb; ++ ++ file = kio->ki_filp; ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); ++ if (unlikely(err)) ++ goto out; ++ ++ err = -ENOSYS; ++ h_file = au_h_fptr(file, au_fbstart(file)); ++ if (h_file->f_op && h_file->f_op->aio_read) { ++ err = security_file_permission(h_file, MAY_READ); ++ if (unlikely(err)) ++ goto out_unlock; ++ if (!is_sync_kiocb(kio)) { ++ get_file(h_file); ++ fput(file); ++ } ++ kio->ki_filp = h_file; ++ err = h_file->f_op->aio_read(kio, iov, nv, pos); ++ /* todo: necessary? */ ++ /* file->f_ra = h_file->f_ra; */ ++ fsstack_copy_attr_atime(dentry->d_inode, ++ h_file->f_dentry->d_inode); ++ } else ++ /* currently there is no such fs */ ++ WARN_ON_ONCE(h_file->f_op && h_file->f_op->read); ++ ++ out_unlock: ++ di_read_unlock(dentry, AuLock_IR); ++ fi_read_unlock(file); ++ out: ++ si_read_unlock(sb); ++ return err; ++} ++ ++static ssize_t aufs_aio_write(struct kiocb *kio, const struct iovec *iov, ++ unsigned long nv, loff_t pos) ++{ ++ ssize_t err; ++ aufs_bindex_t bstart; ++ struct au_pin pin; ++ struct dentry *dentry; ++ struct inode *inode; ++ struct super_block *sb; ++ struct file *file, *h_file; ++ ++ file = kio->ki_filp; ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ inode = dentry->d_inode; ++ mutex_lock(&inode->i_mutex); ++ si_read_lock(sb, AuLock_FLUSH); ++ ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); ++ if (unlikely(err)) ++ goto out; ++ ++ err = au_ready_to_write(file, -1, &pin); ++ di_downgrade_lock(dentry, AuLock_IR); ++ if (unlikely(err)) ++ goto out_unlock; ++ ++ err = -ENOSYS; ++ bstart = au_fbstart(file); ++ h_file = au_h_fptr(file, bstart); ++ au_unpin(&pin); ++ if (h_file->f_op && h_file->f_op->aio_write) { ++ err = security_file_permission(h_file, MAY_WRITE); ++ if (unlikely(err)) ++ goto out_unlock; ++ if (!is_sync_kiocb(kio)) { ++ get_file(h_file); ++ fput(file); ++ } ++ kio->ki_filp = h_file; ++ err = h_file->f_op->aio_write(kio, iov, nv, pos); ++ au_cpup_attr_timesizes(inode); ++ inode->i_mode = h_file->f_dentry->d_inode->i_mode; ++ } else ++ /* currently there is no such fs */ ++ WARN_ON_ONCE(h_file->f_op && h_file->f_op->write); ++ ++ out_unlock: ++ di_read_unlock(dentry, AuLock_IR); ++ fi_write_unlock(file); ++ out: ++ si_read_unlock(sb); ++ mutex_unlock(&inode->i_mutex); ++ return err; ++} ++ ++static ssize_t aufs_splice_read(struct file *file, loff_t *ppos, ++ struct pipe_inode_info *pipe, size_t len, ++ unsigned int flags) ++{ ++ ssize_t err; ++ struct file *h_file; ++ struct dentry *dentry; ++ struct super_block *sb; ++ ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); ++ if (unlikely(err)) ++ goto out; ++ ++ err = -EINVAL; ++ h_file = au_h_fptr(file, au_fbstart(file)); ++ if (au_test_loopback_kthread()) { ++ file->f_mapping = h_file->f_mapping; ++ smp_mb(); /* unnecessary? */ ++ } ++ err = vfsub_splice_to(h_file, ppos, pipe, len, flags); ++ /* todo: necessasry? */ ++ /* file->f_ra = h_file->f_ra; */ ++ fsstack_copy_attr_atime(dentry->d_inode, h_file->f_dentry->d_inode); ++ ++ di_read_unlock(dentry, AuLock_IR); ++ fi_read_unlock(file); ++ ++ out: ++ si_read_unlock(sb); ++ return err; ++} ++ ++static ssize_t ++aufs_splice_write(struct pipe_inode_info *pipe, struct file *file, loff_t *ppos, ++ size_t len, unsigned int flags) ++{ ++ ssize_t err; ++ struct au_pin pin; ++ struct dentry *dentry; ++ struct inode *inode; ++ struct super_block *sb; ++ struct file *h_file; ++ ++ dentry = file->f_dentry; ++ inode = dentry->d_inode; ++ mutex_lock(&inode->i_mutex); ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); ++ if (unlikely(err)) ++ goto out; ++ ++ err = au_ready_to_write(file, -1, &pin); ++ di_downgrade_lock(dentry, AuLock_IR); ++ if (unlikely(err)) ++ goto out_unlock; ++ ++ h_file = au_h_fptr(file, au_fbstart(file)); ++ au_unpin(&pin); ++ err = vfsub_splice_from(pipe, h_file, ppos, len, flags); ++ au_cpup_attr_timesizes(inode); ++ inode->i_mode = h_file->f_dentry->d_inode->i_mode; ++ ++ out_unlock: ++ di_read_unlock(dentry, AuLock_IR); ++ fi_write_unlock(file); ++ out: ++ si_read_unlock(sb); ++ mutex_unlock(&inode->i_mutex); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static struct file *au_safe_file(struct vm_area_struct *vma) ++{ ++ struct file *file; ++ ++ file = vma->vm_file; ++ if (file->private_data && au_test_aufs(file->f_dentry->d_sb)) ++ return file; ++ return NULL; ++} ++ ++static void au_reset_file(struct vm_area_struct *vma, struct file *file) ++{ ++ vma->vm_file = file; ++ /* smp_mb(); */ /* flush vm_file */ ++} ++ ++static int aufs_fault(struct vm_area_struct *vma, struct vm_fault *vmf) ++{ ++ int err; ++ static DECLARE_WAIT_QUEUE_HEAD(wq); ++ struct file *file, *h_file; ++ struct au_finfo *finfo; ++ ++ /* todo: non-robr mode, user vm_file as it is? */ ++ wait_event(wq, (file = au_safe_file(vma))); ++ ++ /* do not revalidate, no si lock */ ++ finfo = au_fi(file); ++ h_file = finfo->fi_hfile[0 + finfo->fi_bstart].hf_file; ++ AuDebugOn(!h_file || !finfo->fi_h_vm_ops); ++ ++ fi_write_lock(file); ++ vma->vm_file = h_file; ++ err = finfo->fi_h_vm_ops->fault(vma, vmf); ++ /* todo: necessary? */ ++ /* file->f_ra = h_file->f_ra; */ ++ au_reset_file(vma, file); ++ fi_write_unlock(file); ++#if 0 /* def CONFIG_SMP */ ++ /* wake_up_nr(&wq, online_cpu - 1); */ ++ wake_up_all(&wq); ++#else ++ wake_up(&wq); ++#endif ++ ++ return err; ++} ++ ++static int aufs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) ++{ ++ int err; ++ static DECLARE_WAIT_QUEUE_HEAD(wq); ++ struct file *file, *h_file; ++ struct au_finfo *finfo; ++ ++ wait_event(wq, (file = au_safe_file(vma))); ++ ++ finfo = au_fi(file); ++ h_file = finfo->fi_hfile[0 + finfo->fi_bstart].hf_file; ++ AuDebugOn(!h_file || !finfo->fi_h_vm_ops); ++ ++ fi_write_lock(file); ++ vma->vm_file = h_file; ++ err = finfo->fi_h_vm_ops->page_mkwrite(vma, vmf); ++ au_reset_file(vma, file); ++ fi_write_unlock(file); ++ wake_up(&wq); ++ ++ return err; ++} ++ ++static void aufs_vm_close(struct vm_area_struct *vma) ++{ ++ static DECLARE_WAIT_QUEUE_HEAD(wq); ++ struct file *file, *h_file; ++ struct au_finfo *finfo; ++ ++ wait_event(wq, (file = au_safe_file(vma))); ++ ++ finfo = au_fi(file); ++ h_file = finfo->fi_hfile[0 + finfo->fi_bstart].hf_file; ++ AuDebugOn(!h_file || !finfo->fi_h_vm_ops); ++ ++ fi_write_lock(file); ++ vma->vm_file = h_file; ++ finfo->fi_h_vm_ops->close(vma); ++ au_reset_file(vma, file); ++ fi_write_unlock(file); ++ wake_up(&wq); ++} ++ ++static struct vm_operations_struct aufs_vm_ops = { ++ /* .close and .page_mkwrite are not set by default */ ++ .fault = aufs_fault, ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++static unsigned long au_prot_conv(unsigned long flags) ++{ ++ unsigned long prot; ++ ++ prot = 0; ++ if (flags & VM_READ) ++ prot |= PROT_READ; ++ if (flags & VM_WRITE) ++ prot |= PROT_WRITE; ++ if (flags & VM_EXEC) ++ prot |= PROT_EXEC; ++ return prot; ++} ++ ++static struct vm_operations_struct *au_vm_ops(struct file *h_file, ++ struct vm_area_struct *vma) ++{ ++ struct vm_operations_struct *vm_ops; ++ int err; ++ ++ vm_ops = ERR_PTR(-ENODEV); ++ if (!h_file->f_op || !h_file->f_op->mmap) ++ goto out; ++ ++ err = ima_file_mmap(h_file, au_prot_conv(vma->vm_flags)); ++ vm_ops = ERR_PTR(err); ++ if (err) ++ goto out; ++ ++ err = h_file->f_op->mmap(h_file, vma); ++ vm_ops = ERR_PTR(err); ++ if (unlikely(err)) ++ goto out; ++ ++ vm_ops = vma->vm_ops; ++ err = do_munmap(current->mm, vma->vm_start, ++ vma->vm_end - vma->vm_start); ++ if (unlikely(err)) { ++ AuIOErr("failed internal unmapping %.*s, %d\n", ++ AuDLNPair(h_file->f_dentry), err); ++ vm_ops = ERR_PTR(-EIO); ++ } ++ ++ out: ++ return vm_ops; ++} ++ ++static int au_custom_vm_ops(struct au_finfo *finfo, struct vm_area_struct *vma) ++{ ++ int err; ++ struct vm_operations_struct *h_ops; ++ ++ AuRwMustAnyLock(&finfo->fi_rwsem); ++ ++ err = 0; ++ h_ops = finfo->fi_h_vm_ops; ++ AuDebugOn(!h_ops); ++ if ((!h_ops->page_mkwrite && !h_ops->close) ++ || finfo->fi_vm_ops) ++ goto out; ++ ++ err = -ENOMEM; ++ finfo->fi_vm_ops = kmemdup(&aufs_vm_ops, sizeof(aufs_vm_ops), GFP_NOFS); ++ if (unlikely(!finfo->fi_vm_ops)) ++ goto out; ++ ++ err = 0; ++ if (h_ops->page_mkwrite) ++ finfo->fi_vm_ops->page_mkwrite = aufs_page_mkwrite; ++ if (h_ops->close) ++ finfo->fi_vm_ops->close = aufs_vm_close; ++ ++ vma->vm_ops = finfo->fi_vm_ops; ++ ++ out: ++ return err; ++} ++ ++static int aufs_mmap(struct file *file, struct vm_area_struct *vma) ++{ ++ int err; ++ unsigned char wlock, mmapped; ++ struct dentry *dentry; ++ struct super_block *sb; ++ struct file *h_file; ++ struct vm_operations_struct *vm_ops; ++ ++ dentry = file->f_dentry; ++ wlock = !!(file->f_mode & FMODE_WRITE) && (vma->vm_flags & VM_SHARED); ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); ++ if (unlikely(err)) ++ goto out; ++ ++ mmapped = !!au_test_mmapped(file); ++ if (wlock) { ++ struct au_pin pin; ++ ++ err = au_ready_to_write(file, -1, &pin); ++ di_downgrade_lock(dentry, AuLock_IR); ++ if (unlikely(err)) ++ goto out_unlock; ++ au_unpin(&pin); ++ } else ++ di_downgrade_lock(dentry, AuLock_IR); ++ ++ h_file = au_h_fptr(file, au_fbstart(file)); ++ if (!mmapped && au_test_fs_bad_mapping(h_file->f_dentry->d_sb)) { ++ /* ++ * by this assignment, f_mapping will differs from aufs inode ++ * i_mapping. ++ * if someone else mixes the use of f_dentry->d_inode and ++ * f_mapping->host, then a problem may arise. ++ */ ++ file->f_mapping = h_file->f_mapping; ++ } ++ ++ vm_ops = NULL; ++ if (!mmapped) { ++ vm_ops = au_vm_ops(h_file, vma); ++ err = PTR_ERR(vm_ops); ++ if (IS_ERR(vm_ops)) ++ goto out_unlock; ++ } ++ ++ /* ++ * unnecessary to handle MAP_DENYWRITE and deny_write_access()? ++ * currently MAP_DENYWRITE from userspace is ignored, but elf loader ++ * sets it. when FMODE_EXEC is set (by open_exec() or sys_uselib()), ++ * both of the aufs file and the lower file is deny_write_access()-ed. ++ * finally I hope we can skip handlling MAP_DENYWRITE here. ++ */ ++ err = generic_file_mmap(file, vma); ++ if (unlikely(err)) ++ goto out_unlock; ++ ++ vma->vm_ops = &aufs_vm_ops; ++ /* test again */ ++ if (!au_test_mmapped(file)) ++ au_fi(file)->fi_h_vm_ops = vm_ops; ++ ++ err = au_custom_vm_ops(au_fi(file), vma); ++ if (unlikely(err)) ++ goto out_unlock; ++ ++ vfsub_file_accessed(h_file); ++ fsstack_copy_attr_atime(dentry->d_inode, h_file->f_dentry->d_inode); ++ ++ out_unlock: ++ di_read_unlock(dentry, AuLock_IR); ++ fi_write_unlock(file); ++ out: ++ si_read_unlock(sb); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int aufs_fsync_nondir(struct file *file, struct dentry *dentry, ++ int datasync) ++{ ++ int err; ++ struct au_pin pin; ++ struct inode *inode; ++ struct file *h_file; ++ struct super_block *sb; ++ ++ inode = dentry->d_inode; ++ IMustLock(file->f_mapping->host); ++ if (inode != file->f_mapping->host) { ++ mutex_unlock(&file->f_mapping->host->i_mutex); ++ mutex_lock(&inode->i_mutex); ++ } ++ IMustLock(inode); ++ ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ ++ err = 0; /* -EBADF; */ /* posix? */ ++ if (unlikely(!(file->f_mode & FMODE_WRITE))) ++ goto out; ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); ++ if (unlikely(err)) ++ goto out; ++ ++ err = au_ready_to_write(file, -1, &pin); ++ di_downgrade_lock(dentry, AuLock_IR); ++ if (unlikely(err)) ++ goto out_unlock; ++ au_unpin(&pin); ++ ++ err = -EINVAL; ++ h_file = au_h_fptr(file, au_fbstart(file)); ++ if (h_file->f_op && h_file->f_op->fsync) { ++ struct dentry *h_d; ++ struct mutex *h_mtx; ++ ++ /* ++ * no filemap_fdatawrite() since aufs file has no its own ++ * mapping, but dir. ++ */ ++ h_d = h_file->f_dentry; ++ h_mtx = &h_d->d_inode->i_mutex; ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD); ++ err = h_file->f_op->fsync(h_file, h_d, datasync); ++ if (!err) ++ vfsub_update_h_iattr(&h_file->f_path, /*did*/NULL); ++ /*ignore*/ ++ au_cpup_attr_timesizes(inode); ++ mutex_unlock(h_mtx); ++ } ++ ++ out_unlock: ++ di_read_unlock(dentry, AuLock_IR); ++ fi_write_unlock(file); ++ out: ++ si_read_unlock(sb); ++ if (inode != file->f_mapping->host) { ++ mutex_unlock(&inode->i_mutex); ++ mutex_lock(&file->f_mapping->host->i_mutex); ++ } ++ return err; ++} ++ ++/* no one supports this operation, currently */ ++#if 0 ++static int aufs_aio_fsync_nondir(struct kiocb *kio, int datasync) ++{ ++ int err; ++ struct au_pin pin; ++ struct dentry *dentry; ++ struct inode *inode; ++ struct file *file, *h_file; ++ struct super_block *sb; ++ ++ file = kio->ki_filp; ++ dentry = file->f_dentry; ++ inode = dentry->d_inode; ++ mutex_lock(&inode->i_mutex); ++ ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ ++ err = 0; /* -EBADF; */ /* posix? */ ++ if (unlikely(!(file->f_mode & FMODE_WRITE))) ++ goto out; ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1); ++ if (unlikely(err)) ++ goto out; ++ ++ err = au_ready_to_write(file, -1, &pin); ++ di_downgrade_lock(dentry, AuLock_IR); ++ if (unlikely(err)) ++ goto out_unlock; ++ au_unpin(&pin); ++ ++ err = -ENOSYS; ++ h_file = au_h_fptr(file, au_fbstart(file)); ++ if (h_file->f_op && h_file->f_op->aio_fsync) { ++ struct dentry *h_d; ++ struct mutex *h_mtx; ++ ++ h_d = h_file->f_dentry; ++ h_mtx = &h_d->d_inode->i_mutex; ++ if (!is_sync_kiocb(kio)) { ++ get_file(h_file); ++ fput(file); ++ } ++ kio->ki_filp = h_file; ++ err = h_file->f_op->aio_fsync(kio, datasync); ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD); ++ if (!err) ++ vfsub_update_h_iattr(&h_file->f_path, /*did*/NULL); ++ /*ignore*/ ++ au_cpup_attr_timesizes(inode); ++ mutex_unlock(h_mtx); ++ } ++ ++ out_unlock: ++ di_read_unlock(dentry, AuLock_IR); ++ fi_write_unlock(file); ++ out: ++ si_read_unlock(sb); ++ mutex_unlock(&inode->i_mutex); ++ return err; ++} ++#endif ++ ++static int aufs_fasync(int fd, struct file *file, int flag) ++{ ++ int err; ++ struct file *h_file; ++ struct dentry *dentry; ++ struct super_block *sb; ++ ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); ++ if (unlikely(err)) ++ goto out; ++ ++ h_file = au_h_fptr(file, au_fbstart(file)); ++ if (h_file->f_op && h_file->f_op->fasync) ++ err = h_file->f_op->fasync(fd, h_file, flag); ++ ++ di_read_unlock(dentry, AuLock_IR); ++ fi_read_unlock(file); ++ ++ out: ++ si_read_unlock(sb); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* no one supports this operation, currently */ ++#if 0 ++static ssize_t aufs_sendpage(struct file *file, struct page *page, int offset, ++ size_t len, loff_t *pos , int more) ++{ ++} ++#endif ++ ++/* ---------------------------------------------------------------------- */ ++ ++const struct file_operations aufs_file_fop = { ++ /* ++ * while generic_file_llseek/_unlocked() don't use BKL, ++ * don't use it since it operates file->f_mapping->host. ++ * in aufs, it may be a real file and may confuse users by UDBA. ++ */ ++ /* .llseek = generic_file_llseek, */ ++ ++ .read = aufs_read, ++ .write = aufs_write, ++ .aio_read = aufs_aio_read, ++ .aio_write = aufs_aio_write, ++#ifdef CONFIG_AUFS_POLL ++ .poll = aufs_poll, ++#endif ++ .mmap = aufs_mmap, ++ .open = aufs_open_nondir, ++ .flush = aufs_flush, ++ .release = aufs_release_nondir, ++ .fsync = aufs_fsync_nondir, ++ /* .aio_fsync = aufs_aio_fsync_nondir, */ ++ .fasync = aufs_fasync, ++ /* .sendpage = aufs_sendpage, */ ++ .splice_write = aufs_splice_write, ++ .splice_read = aufs_splice_read, ++#if 0 ++ .aio_splice_write = aufs_aio_splice_write, ++ .aio_splice_read = aufs_aio_splice_read ++#endif ++}; +diff -Nur linux-2.6.31-vanilla/fs/aufs/fstype.h linux-2.6.31/fs/aufs/fstype.h +--- linux-2.6.31-vanilla/fs/aufs/fstype.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/fstype.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,485 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * judging filesystem type ++ */ ++ ++#ifndef __AUFS_FSTYPE_H__ ++#define __AUFS_FSTYPE_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++#include ++#include ++#include ++ ++static inline int au_test_aufs(struct super_block *sb) ++{ ++ return sb->s_magic == AUFS_SUPER_MAGIC; ++} ++ ++static inline const char *au_sbtype(struct super_block *sb) ++{ ++ return sb->s_type->name; ++} ++ ++static inline int au_test_iso9660(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_ROMFS_FS) || defined(CONFIG_ROMFS_FS_MODULE) ++ return sb->s_magic == ROMFS_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_romfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_ISO9660_FS) || defined(CONFIG_ISO9660_FS_MODULE) ++ return sb->s_magic == ISOFS_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_cramfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_CRAMFS) || defined(CONFIG_CRAMFS_MODULE) ++ return sb->s_magic == CRAMFS_MAGIC; ++#endif ++ return 0; ++} ++ ++static inline int au_test_nfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_NFS_FS) || defined(CONFIG_NFS_FS_MODULE) ++ return sb->s_magic == NFS_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_fuse(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_FUSE_FS) || defined(CONFIG_FUSE_FS_MODULE) ++ return sb->s_magic == FUSE_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_xfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_XFS_FS) || defined(CONFIG_XFS_FS_MODULE) ++ return sb->s_magic == XFS_SB_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_tmpfs(struct super_block *sb __maybe_unused) ++{ ++#ifdef CONFIG_TMPFS ++ return sb->s_magic == TMPFS_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_ecryptfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_ECRYPT_FS) || defined(CONFIG_ECRYPT_FS_MODULE) ++ return !strcmp(au_sbtype(sb), "ecryptfs"); ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_smbfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_SMB_FS) || defined(CONFIG_SMB_FS_MODULE) ++ return sb->s_magic == SMB_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_ocfs2(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_OCFS2_FS) || defined(CONFIG_OCFS2_FS_MODULE) ++ return sb->s_magic == OCFS2_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_ocfs2_dlmfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_OCFS2_FS_O2CB) || defined(CONFIG_OCFS2_FS_O2CB_MODULE) ++ return sb->s_magic == DLMFS_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_coda(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_CODA_FS) || defined(CONFIG_CODA_FS_MODULE) ++ return sb->s_magic == CODA_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_v9fs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_9P_FS) || defined(CONFIG_9P_FS_MODULE) ++ return sb->s_magic == V9FS_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_ext4(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_EXT4DEV_FS) || defined(CONFIG_EXT4DEV_FS_MODULE) ++ return sb->s_magic == EXT4_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_sysv(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_SYSV_FS) || defined(CONFIG_SYSV_FS_MODULE) ++ return !strcmp(au_sbtype(sb), "sysv"); ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_ramfs(struct super_block *sb) ++{ ++ return sb->s_magic == RAMFS_MAGIC; ++} ++ ++static inline int au_test_ubifs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_UBIFS_FS) || defined(CONFIG_UBIFS_FS_MODULE) ++ return sb->s_magic == UBIFS_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_procfs(struct super_block *sb __maybe_unused) ++{ ++#ifdef CONFIG_PROC_FS ++ return sb->s_magic == PROC_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_sysfs(struct super_block *sb __maybe_unused) ++{ ++#ifdef CONFIG_SYSFS ++ return sb->s_magic == SYSFS_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_configfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_CONFIGFS_FS) || defined(CONFIG_CONFIGFS_FS_MODULE) ++ return sb->s_magic == CONFIGFS_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_minix(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_MINIX_FS) || defined(CONFIG_MINIX_FS_MODULE) ++ return sb->s_magic == MINIX3_SUPER_MAGIC ++ || sb->s_magic == MINIX2_SUPER_MAGIC ++ || sb->s_magic == MINIX2_SUPER_MAGIC2 ++ || sb->s_magic == MINIX_SUPER_MAGIC ++ || sb->s_magic == MINIX_SUPER_MAGIC2; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_cifs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_CIFS_FS) || defined(CONFIGCIFS_FS_MODULE) ++ return sb->s_magic == CIFS_MAGIC_NUMBER; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_fat(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_FAT_FS) || defined(CONFIG_FAT_FS_MODULE) ++ return sb->s_magic == MSDOS_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_msdos(struct super_block *sb) ++{ ++ return au_test_fat(sb); ++} ++ ++static inline int au_test_vfat(struct super_block *sb) ++{ ++ return au_test_fat(sb); ++} ++ ++static inline int au_test_securityfs(struct super_block *sb __maybe_unused) ++{ ++#ifdef CONFIG_SECURITYFS ++ return sb->s_magic == SECURITYFS_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_squashfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_SQUASHFS) || defined(CONFIG_SQUASHFS_MODULE) ++ return sb->s_magic == SQUASHFS_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_btrfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_BTRFS_FS) || defined(CONFIG_BTRFS_FS_MODULE) ++ return sb->s_magic == BTRFS_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_xenfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_XENFS) || defined(CONFIG_XENFS_MODULE) ++ return sb->s_magic == XENFS_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_debugfs(struct super_block *sb __maybe_unused) ++{ ++#ifdef CONFIG_DEBUG_FS ++ return sb->s_magic == DEBUGFS_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++static inline int au_test_nilfs(struct super_block *sb __maybe_unused) ++{ ++#if defined(CONFIG_NILFS) || defined(CONFIG_NILFS_MODULE) ++ return sb->s_magic == NILFS_SUPER_MAGIC; ++#else ++ return 0; ++#endif ++} ++ ++/* ---------------------------------------------------------------------- */ ++/* ++ * they can't be an aufs branch. ++ */ ++static inline int au_test_fs_unsuppoted(struct super_block *sb) ++{ ++ return ++#ifndef CONFIG_AUFS_BR_RAMFS ++ au_test_ramfs(sb) || ++#endif ++ au_test_procfs(sb) ++ || au_test_sysfs(sb) ++ || au_test_configfs(sb) ++ || au_test_debugfs(sb) ++ || au_test_securityfs(sb) ++ || au_test_xenfs(sb) ++ || au_test_ecryptfs(sb) ++ /* || !strcmp(au_sbtype(sb), "unionfs") */ ++ || au_test_aufs(sb); /* will be supported in next version */ ++} ++ ++/* ++ * If the filesystem supports NFS-export, then it has to support NULL as ++ * a nameidata parameter for ->create(), ->lookup() and ->d_revalidate(). ++ * We can apply this principle when we handle a lower filesystem. ++ */ ++static inline int au_test_fs_null_nd(struct super_block *sb) ++{ ++ return !!sb->s_export_op; ++} ++ ++static inline int au_test_fs_remote(struct super_block *sb) ++{ ++ return !au_test_tmpfs(sb) ++#ifdef CONFIG_AUFS_BR_RAMFS ++ && !au_test_ramfs(sb) ++#endif ++ && !(sb->s_type->fs_flags & FS_REQUIRES_DEV); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * Note: these functions (below) are created after reading ->getattr() in all ++ * filesystems under linux/fs. it means we have to do so in every update... ++ */ ++ ++/* ++ * some filesystems require getattr to refresh the inode attributes before ++ * referencing. ++ * in most cases, we can rely on the inode attribute in NFS (or every remote fs) ++ * and leave the work for d_revalidate() ++ */ ++static inline int au_test_fs_refresh_iattr(struct super_block *sb) ++{ ++ return au_test_nfs(sb) ++ || au_test_fuse(sb) ++ /* || au_test_smbfs(sb) */ /* untested */ ++ /* || au_test_ocfs2(sb) */ /* untested */ ++ /* || au_test_btrfs(sb) */ /* untested */ ++ /* || au_test_coda(sb) */ /* untested */ ++ /* || au_test_v9fs(sb) */ /* untested */ ++ ; ++} ++ ++/* ++ * filesystems which don't maintain i_size or i_blocks. ++ */ ++static inline int au_test_fs_bad_iattr_size(struct super_block *sb) ++{ ++ return au_test_xfs(sb) ++ /* || au_test_ext4(sb) */ /* untested */ ++ /* || au_test_ocfs2(sb) */ /* untested */ ++ /* || au_test_ocfs2_dlmfs(sb) */ /* untested */ ++ /* || au_test_sysv(sb) */ /* untested */ ++ /* || au_test_ubifs(sb) */ /* untested */ ++ /* || au_test_minix(sb) */ /* untested */ ++ ; ++} ++ ++/* ++ * filesystems which don't store the correct value in some of their inode ++ * attributes. ++ */ ++static inline int au_test_fs_bad_iattr(struct super_block *sb) ++{ ++ return au_test_fs_bad_iattr_size(sb) ++ /* || au_test_cifs(sb) */ /* untested */ ++ || au_test_fat(sb) ++ || au_test_msdos(sb) ++ || au_test_vfat(sb); ++} ++ ++/* they don't check i_nlink in link(2) */ ++static inline int au_test_fs_no_limit_nlink(struct super_block *sb) ++{ ++ return au_test_tmpfs(sb) ++#ifdef CONFIG_AUFS_BR_RAMFS ++ || au_test_ramfs(sb) ++#endif ++ || au_test_ubifs(sb); ++} ++ ++/* ++ * filesystems which sets S_NOATIME and S_NOCMTIME. ++ */ ++static inline int au_test_fs_notime(struct super_block *sb) ++{ ++ return au_test_nfs(sb) ++ || au_test_fuse(sb) ++ || au_test_ubifs(sb) ++ /* || au_test_cifs(sb) */ /* untested */ ++ ; ++} ++ ++/* ++ * filesystems which requires replacing i_mapping. ++ */ ++static inline int au_test_fs_bad_mapping(struct super_block *sb) ++{ ++ return au_test_fuse(sb) ++ || au_test_ubifs(sb); ++} ++ ++/* temporary support for i#1 in cramfs */ ++static inline int au_test_fs_unique_ino(struct inode *inode) ++{ ++ if (au_test_cramfs(inode->i_sb)) ++ return inode->i_ino != 1; ++ return 1; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * the filesystem where the xino files placed must support i/o after unlink and ++ * maintain i_size and i_blocks. ++ */ ++static inline int au_test_fs_bad_xino(struct super_block *sb) ++{ ++ return au_test_fs_remote(sb) ++ || au_test_fs_bad_iattr_size(sb) ++#ifdef CONFIG_AUFS_BR_RAMFS ++ || !(au_test_ramfs(sb) || au_test_fs_null_nd(sb)) ++#else ++ || !au_test_fs_null_nd(sb) /* to keep xino code simple */ ++#endif ++ /* don't want unnecessary work for xino */ ++ || au_test_aufs(sb) ++ || au_test_ecryptfs(sb) ++ || au_test_nilfs(sb); ++} ++ ++static inline int au_test_fs_trunc_xino(struct super_block *sb) ++{ ++ return au_test_tmpfs(sb) ++ || au_test_ramfs(sb); ++} ++ ++/* ++ * test if the @sb is real-readonly. ++ */ ++static inline int au_test_fs_rr(struct super_block *sb) ++{ ++ return au_test_squashfs(sb) ++ || au_test_iso9660(sb) ++ || au_test_cramfs(sb) ++ || au_test_romfs(sb); ++} ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_FSTYPE_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/hinotify.c linux-2.6.31/fs/aufs/hinotify.c +--- linux-2.6.31-vanilla/fs/aufs/hinotify.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/hinotify.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,755 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * inotify for the lower directories ++ */ ++ ++#include "aufs.h" ++ ++static const __u32 AuHinMask = (IN_MOVE | IN_DELETE | IN_CREATE); ++static struct inotify_handle *au_hin_handle; ++ ++AuCacheFuncs(hinotify, HINOTIFY); ++ ++int au_hin_alloc(struct au_hinode *hinode, struct inode *inode, ++ struct inode *h_inode) ++{ ++ int err; ++ struct au_hinotify *hin; ++ s32 wd; ++ ++ err = -ENOMEM; ++ hin = au_cache_alloc_hinotify(); ++ if (hin) { ++ AuDebugOn(hinode->hi_notify); ++ hinode->hi_notify = hin; ++ hin->hin_aufs_inode = inode; ++ ++ inotify_init_watch(&hin->hin_watch); ++ wd = inotify_add_watch(au_hin_handle, &hin->hin_watch, h_inode, ++ AuHinMask); ++ if (wd >= 0) ++ return 0; /* success */ ++ ++ err = wd; ++ put_inotify_watch(&hin->hin_watch); ++ au_cache_free_hinotify(hin); ++ hinode->hi_notify = NULL; ++ } ++ ++ return err; ++} ++ ++void au_hin_free(struct au_hinode *hinode) ++{ ++ int err; ++ struct au_hinotify *hin; ++ ++ hin = hinode->hi_notify; ++ if (hin) { ++ err = 0; ++ if (atomic_read(&hin->hin_watch.count)) ++ err = inotify_rm_watch(au_hin_handle, &hin->hin_watch); ++ if (unlikely(err)) ++ /* it means the watch is already removed */ ++ AuWarn("failed inotify_rm_watch() %d\n", err); ++ au_cache_free_hinotify(hin); ++ hinode->hi_notify = NULL; ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void au_hin_ctl(struct au_hinode *hinode, int do_set) ++{ ++ struct inode *h_inode; ++ struct inotify_watch *watch; ++ ++ if (!hinode->hi_notify) ++ return; ++ ++ h_inode = hinode->hi_inode; ++ IMustLock(h_inode); ++ ++ /* todo: try inotify_find_update_watch()? */ ++ watch = &hinode->hi_notify->hin_watch; ++ mutex_lock(&h_inode->inotify_mutex); ++ /* mutex_lock(&watch->ih->mutex); */ ++ if (do_set) { ++ AuDebugOn(watch->mask & AuHinMask); ++ watch->mask |= AuHinMask; ++ } else { ++ AuDebugOn(!(watch->mask & AuHinMask)); ++ watch->mask &= ~AuHinMask; ++ } ++ /* mutex_unlock(&watch->ih->mutex); */ ++ mutex_unlock(&h_inode->inotify_mutex); ++} ++ ++void au_reset_hinotify(struct inode *inode, unsigned int flags) ++{ ++ aufs_bindex_t bindex, bend; ++ struct inode *hi; ++ struct dentry *iwhdentry; ++ ++ bend = au_ibend(inode); ++ for (bindex = au_ibstart(inode); bindex <= bend; bindex++) { ++ hi = au_h_iptr(inode, bindex); ++ if (!hi) ++ continue; ++ ++ /* mutex_lock_nested(&hi->i_mutex, AuLsc_I_CHILD); */ ++ iwhdentry = au_hi_wh(inode, bindex); ++ if (iwhdentry) ++ dget(iwhdentry); ++ au_igrab(hi); ++ au_set_h_iptr(inode, bindex, NULL, 0); ++ au_set_h_iptr(inode, bindex, au_igrab(hi), ++ flags & ~AuHi_XINO); ++ iput(hi); ++ dput(iwhdentry); ++ /* mutex_unlock(&hi->i_mutex); */ ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int hin_xino(struct inode *inode, struct inode *h_inode) ++{ ++ int err; ++ aufs_bindex_t bindex, bend, bfound, bstart; ++ struct inode *h_i; ++ ++ err = 0; ++ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) { ++ AuWarn("branch root dir was changed\n"); ++ goto out; ++ } ++ ++ bfound = -1; ++ bend = au_ibend(inode); ++ bstart = au_ibstart(inode); ++#if 0 /* reserved for future use */ ++ if (bindex == bend) { ++ /* keep this ino in rename case */ ++ goto out; ++ } ++#endif ++ for (bindex = bstart; bindex <= bend; bindex++) { ++ if (au_h_iptr(inode, bindex) == h_inode) { ++ bfound = bindex; ++ break; ++ } ++ } ++ if (bfound < 0) ++ goto out; ++ ++ for (bindex = bstart; bindex <= bend; bindex++) { ++ h_i = au_h_iptr(inode, bindex); ++ if (!h_i) ++ continue; ++ ++ err = au_xino_write(inode->i_sb, bindex, h_i->i_ino, /*ino*/0); ++ /* ignore this error */ ++ /* bad action? */ ++ } ++ ++ /* children inode number will be broken */ ++ ++ out: ++ AuTraceErr(err); ++ return err; ++} ++ ++static int hin_gen_tree(struct dentry *dentry) ++{ ++ int err, i, j, ndentry; ++ struct au_dcsub_pages dpages; ++ struct au_dpage *dpage; ++ struct dentry **dentries; ++ ++ err = au_dpages_init(&dpages, GFP_NOFS); ++ if (unlikely(err)) ++ goto out; ++ err = au_dcsub_pages(&dpages, dentry, NULL, NULL); ++ if (unlikely(err)) ++ goto out_dpages; ++ ++ for (i = 0; i < dpages.ndpage; i++) { ++ dpage = dpages.dpages + i; ++ dentries = dpage->dentries; ++ ndentry = dpage->ndentry; ++ for (j = 0; j < ndentry; j++) { ++ struct dentry *d; ++ ++ d = dentries[j]; ++ if (IS_ROOT(d)) ++ continue; ++ ++ d_drop(d); ++ au_digen_dec(d); ++ if (d->d_inode) ++ /* todo: reset children xino? ++ cached children only? */ ++ au_iigen_dec(d->d_inode); ++ } ++ } ++ ++ out_dpages: ++ au_dpages_free(&dpages); ++ ++ /* discard children */ ++ dentry_unhash(dentry); ++ dput(dentry); ++ out: ++ return err; ++} ++ ++/* ++ * return 0 if processed. ++ */ ++static int hin_gen_by_inode(char *name, unsigned int nlen, struct inode *inode, ++ const unsigned int isdir) ++{ ++ int err; ++ struct dentry *d; ++ struct qstr *dname; ++ ++ err = 1; ++ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) { ++ AuWarn("branch root dir was changed\n"); ++ err = 0; ++ goto out; ++ } ++ ++ if (!isdir) { ++ AuDebugOn(!name); ++ au_iigen_dec(inode); ++ spin_lock(&dcache_lock); ++ list_for_each_entry(d, &inode->i_dentry, d_alias) { ++ dname = &d->d_name; ++ if (dname->len != nlen ++ && memcmp(dname->name, name, nlen)) ++ continue; ++ err = 0; ++ spin_lock(&d->d_lock); ++ __d_drop(d); ++ au_digen_dec(d); ++ spin_unlock(&d->d_lock); ++ break; ++ } ++ spin_unlock(&dcache_lock); ++ } else { ++ au_fset_si(au_sbi(inode->i_sb), FAILED_REFRESH_DIRS); ++ d = d_find_alias(inode); ++ if (!d) { ++ au_iigen_dec(inode); ++ goto out; ++ } ++ ++ dname = &d->d_name; ++ if (dname->len == nlen && !memcmp(dname->name, name, nlen)) ++ err = hin_gen_tree(d); ++ dput(d); ++ } ++ ++ out: ++ AuTraceErr(err); ++ return err; ++} ++ ++static int hin_gen_by_name(struct dentry *dentry, const unsigned int isdir) ++{ ++ int err; ++ struct inode *inode; ++ ++ inode = dentry->d_inode; ++ if (IS_ROOT(dentry) ++ /* || (inode && inode->i_ino == AUFS_ROOT_INO) */ ++ ) { ++ AuWarn("branch root dir was changed\n"); ++ return 0; ++ } ++ ++ err = 0; ++ if (!isdir) { ++ d_drop(dentry); ++ au_digen_dec(dentry); ++ if (inode) ++ au_iigen_dec(inode); ++ } else { ++ au_fset_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIRS); ++ if (inode) ++ err = hin_gen_tree(dentry); ++ } ++ ++ AuTraceErr(err); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* hinotify job flags */ ++#define AuHinJob_XINO0 1 ++#define AuHinJob_GEN (1 << 1) ++#define AuHinJob_DIRENT (1 << 2) ++#define AuHinJob_ISDIR (1 << 3) ++#define AuHinJob_TRYXINO0 (1 << 4) ++#define AuHinJob_MNTPNT (1 << 5) ++#define au_ftest_hinjob(flags, name) ((flags) & AuHinJob_##name) ++#define au_fset_hinjob(flags, name) { (flags) |= AuHinJob_##name; } ++#define au_fclr_hinjob(flags, name) { (flags) &= ~AuHinJob_##name; } ++ ++struct hin_job_args { ++ unsigned int flags; ++ struct inode *inode, *h_inode, *dir, *h_dir; ++ struct dentry *dentry; ++ char *h_name; ++ int h_nlen; ++}; ++ ++static int hin_job(struct hin_job_args *a) ++{ ++ const unsigned int isdir = au_ftest_hinjob(a->flags, ISDIR); ++ ++ /* reset xino */ ++ if (au_ftest_hinjob(a->flags, XINO0) && a->inode) ++ hin_xino(a->inode, a->h_inode); /* ignore this error */ ++ ++ if (au_ftest_hinjob(a->flags, TRYXINO0) ++ && a->inode ++ && a->h_inode) { ++ mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD); ++ if (!a->h_inode->i_nlink) ++ hin_xino(a->inode, a->h_inode); /* ignore this error */ ++ mutex_unlock(&a->h_inode->i_mutex); ++ } ++ ++ /* make the generation obsolete */ ++ if (au_ftest_hinjob(a->flags, GEN)) { ++ int err = -1; ++ if (a->inode) ++ err = hin_gen_by_inode(a->h_name, a->h_nlen, a->inode, ++ isdir); ++ if (err && a->dentry) ++ hin_gen_by_name(a->dentry, isdir); ++ /* ignore this error */ ++ } ++ ++ /* make dir entries obsolete */ ++ if (au_ftest_hinjob(a->flags, DIRENT) && a->inode) { ++ struct au_vdir *vdir; ++ ++ vdir = au_ivdir(a->inode); ++ if (vdir) ++ vdir->vd_jiffy = 0; ++ /* IMustLock(a->inode); */ ++ /* a->inode->i_version++; */ ++ } ++ ++ /* can do nothing but warn */ ++ if (au_ftest_hinjob(a->flags, MNTPNT) ++ && a->dentry ++ && d_mountpoint(a->dentry)) ++ AuWarn("mount-point %.*s is removed or renamed\n", ++ AuDLNPair(a->dentry)); ++ ++ return 0; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static char *in_name(u32 mask) ++{ ++#ifdef CONFIG_AUFS_DEBUG ++#define test_ret(flag) if (mask & flag) \ ++ return #flag; ++ test_ret(IN_ACCESS); ++ test_ret(IN_MODIFY); ++ test_ret(IN_ATTRIB); ++ test_ret(IN_CLOSE_WRITE); ++ test_ret(IN_CLOSE_NOWRITE); ++ test_ret(IN_OPEN); ++ test_ret(IN_MOVED_FROM); ++ test_ret(IN_MOVED_TO); ++ test_ret(IN_CREATE); ++ test_ret(IN_DELETE); ++ test_ret(IN_DELETE_SELF); ++ test_ret(IN_MOVE_SELF); ++ test_ret(IN_UNMOUNT); ++ test_ret(IN_Q_OVERFLOW); ++ test_ret(IN_IGNORED); ++ return ""; ++#undef test_ret ++#else ++ return "??"; ++#endif ++} ++ ++static struct dentry *lookup_wlock_by_name(char *name, unsigned int nlen, ++ struct inode *dir) ++{ ++ struct dentry *dentry, *d, *parent; ++ struct qstr *dname; ++ ++ parent = d_find_alias(dir); ++ if (!parent) ++ return NULL; ++ ++ dentry = NULL; ++ spin_lock(&dcache_lock); ++ list_for_each_entry(d, &parent->d_subdirs, d_u.d_child) { ++ /* AuDbg("%.*s\n", AuDLNPair(d)); */ ++ dname = &d->d_name; ++ if (dname->len != nlen || memcmp(dname->name, name, nlen)) ++ continue; ++ if (!atomic_read(&d->d_count) || !d->d_fsdata) { ++ spin_lock(&d->d_lock); ++ __d_drop(d); ++ spin_unlock(&d->d_lock); ++ continue; ++ } ++ ++ dentry = dget(d); ++ break; ++ } ++ spin_unlock(&dcache_lock); ++ dput(parent); ++ ++ if (dentry) ++ di_write_lock_child(dentry); ++ ++ return dentry; ++} ++ ++static struct inode *lookup_wlock_by_ino(struct super_block *sb, ++ aufs_bindex_t bindex, ino_t h_ino) ++{ ++ struct inode *inode; ++ ino_t ino; ++ int err; ++ ++ inode = NULL; ++ err = au_xino_read(sb, bindex, h_ino, &ino); ++ if (!err && ino) ++ inode = ilookup(sb, ino); ++ if (!inode) ++ goto out; ++ ++ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) { ++ AuWarn("wrong root branch\n"); ++ iput(inode); ++ inode = NULL; ++ goto out; ++ } ++ ++ ii_write_lock_child(inode); ++ ++ out: ++ return inode; ++} ++ ++enum { CHILD, PARENT }; ++struct postproc_args { ++ struct inode *h_dir, *dir, *h_child_inode; ++ u32 mask; ++ unsigned int flags[2]; ++ unsigned int h_child_nlen; ++ char h_child_name[]; ++}; ++ ++static void postproc(void *_args) ++{ ++ struct postproc_args *a = _args; ++ struct super_block *sb; ++ aufs_bindex_t bindex, bend, bfound; ++ unsigned char xino, try_iput; ++ int err; ++ struct inode *inode; ++ ino_t h_ino; ++ struct hin_job_args args; ++ struct dentry *dentry; ++ struct au_sbinfo *sbinfo; ++ ++ AuDebugOn(!_args); ++ AuDebugOn(!a->h_dir); ++ AuDebugOn(!a->dir); ++ AuDebugOn(!a->mask); ++ AuDbg("mask 0x%x %s, i%lu, hi%lu, hci%lu\n", ++ a->mask, in_name(a->mask), a->dir->i_ino, a->h_dir->i_ino, ++ a->h_child_inode ? a->h_child_inode->i_ino : 0); ++ ++ inode = NULL; ++ dentry = NULL; ++ /* ++ * do not lock a->dir->i_mutex here ++ * because of d_revalidate() may cause a deadlock. ++ */ ++ sb = a->dir->i_sb; ++ AuDebugOn(!sb); ++ sbinfo = au_sbi(sb); ++ AuDebugOn(!sbinfo); ++ /* big aufs lock */ ++ si_noflush_write_lock(sb); ++ ++ ii_read_lock_parent(a->dir); ++ bfound = -1; ++ bend = au_ibend(a->dir); ++ for (bindex = au_ibstart(a->dir); bindex <= bend; bindex++) ++ if (au_h_iptr(a->dir, bindex) == a->h_dir) { ++ bfound = bindex; ++ break; ++ } ++ ii_read_unlock(a->dir); ++ if (unlikely(bfound < 0)) ++ goto out; ++ ++ xino = !!au_opt_test(au_mntflags(sb), XINO); ++ h_ino = 0; ++ if (a->h_child_inode) ++ h_ino = a->h_child_inode->i_ino; ++ ++ if (a->h_child_nlen ++ && (au_ftest_hinjob(a->flags[CHILD], GEN) ++ || au_ftest_hinjob(a->flags[CHILD], MNTPNT))) ++ dentry = lookup_wlock_by_name(a->h_child_name, a->h_child_nlen, ++ a->dir); ++ try_iput = 0; ++ if (dentry) ++ inode = dentry->d_inode; ++ if (xino && !inode && h_ino ++ && (au_ftest_hinjob(a->flags[CHILD], XINO0) ++ || au_ftest_hinjob(a->flags[CHILD], TRYXINO0) ++ || au_ftest_hinjob(a->flags[CHILD], GEN))) { ++ inode = lookup_wlock_by_ino(sb, bfound, h_ino); ++ try_iput = 1; ++ } ++ ++ args.flags = a->flags[CHILD]; ++ args.dentry = dentry; ++ args.inode = inode; ++ args.h_inode = a->h_child_inode; ++ args.dir = a->dir; ++ args.h_dir = a->h_dir; ++ args.h_name = a->h_child_name; ++ args.h_nlen = a->h_child_nlen; ++ err = hin_job(&args); ++ if (dentry) { ++ if (dentry->d_fsdata) ++ di_write_unlock(dentry); ++ dput(dentry); ++ } ++ if (inode && try_iput) { ++ ii_write_unlock(inode); ++ iput(inode); ++ } ++ ++ ii_write_lock_parent(a->dir); ++ args.flags = a->flags[PARENT]; ++ args.dentry = NULL; ++ args.inode = a->dir; ++ args.h_inode = a->h_dir; ++ args.dir = NULL; ++ args.h_dir = NULL; ++ args.h_name = NULL; ++ args.h_nlen = 0; ++ err = hin_job(&args); ++ ii_write_unlock(a->dir); ++ ++ out: ++ au_nwt_done(&sbinfo->si_nowait); ++ si_write_unlock(sb); ++ ++ iput(a->h_child_inode); ++ iput(a->h_dir); ++ iput(a->dir); ++ kfree(a); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static void aufs_inotify(struct inotify_watch *watch, u32 wd __maybe_unused, ++ u32 mask, u32 cookie __maybe_unused, ++ const char *h_child_name, struct inode *h_child_inode) ++{ ++ struct au_hinotify *hinotify; ++ struct postproc_args *args; ++ int len, wkq_err; ++ unsigned char isdir, isroot, wh; ++ char *p; ++ struct inode *dir; ++ unsigned int flags[2]; ++ ++ /* if IN_UNMOUNT happens, there must be another bug */ ++ AuDebugOn(mask & IN_UNMOUNT); ++ if (mask & (IN_IGNORED | IN_UNMOUNT)) { ++ put_inotify_watch(watch); ++ return; ++ } ++#ifdef AuDbgHinotify ++ au_debug(1); ++ if (1 || !h_child_name || strcmp(h_child_name, AUFS_XINO_FNAME)) { ++ AuDbg("i%lu, wd %d, mask 0x%x %s, cookie 0x%x, hcname %s," ++ " hi%lu\n", ++ watch->inode->i_ino, wd, mask, in_name(mask), cookie, ++ h_child_name ? h_child_name : "", ++ h_child_inode ? h_child_inode->i_ino : 0); ++ WARN_ON(1); ++ } ++ au_debug(0); ++#endif ++ ++ hinotify = container_of(watch, struct au_hinotify, hin_watch); ++ AuDebugOn(!hinotify || !hinotify->hin_aufs_inode); ++ dir = igrab(hinotify->hin_aufs_inode); ++ if (!dir) ++ return; ++ ++ isroot = (dir->i_ino == AUFS_ROOT_INO); ++ len = 0; ++ wh = 0; ++ if (h_child_name) { ++ len = strlen(h_child_name); ++ if (!memcmp(h_child_name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) { ++ h_child_name += AUFS_WH_PFX_LEN; ++ len -= AUFS_WH_PFX_LEN; ++ wh = 1; ++ } ++ } ++ ++ isdir = 0; ++ if (h_child_inode) ++ isdir = !!S_ISDIR(h_child_inode->i_mode); ++ flags[PARENT] = AuHinJob_ISDIR; ++ flags[CHILD] = 0; ++ if (isdir) ++ flags[CHILD] = AuHinJob_ISDIR; ++ switch (mask & IN_ALL_EVENTS) { ++ case IN_MOVED_FROM: ++ case IN_MOVED_TO: ++ AuDebugOn(!h_child_name || !h_child_inode); ++ au_fset_hinjob(flags[CHILD], GEN); ++ au_fset_hinjob(flags[CHILD], XINO0); ++ au_fset_hinjob(flags[CHILD], MNTPNT); ++ au_fset_hinjob(flags[PARENT], DIRENT); ++ break; ++ ++ case IN_CREATE: ++ AuDebugOn(!h_child_name || !h_child_inode); ++ au_fset_hinjob(flags[PARENT], DIRENT); ++ au_fset_hinjob(flags[CHILD], GEN); ++ break; ++ ++ case IN_DELETE: ++ /* ++ * aufs never be able to get this child inode. ++ * revalidation should be in d_revalidate() ++ * by checking i_nlink, i_generation or d_unhashed(). ++ */ ++ AuDebugOn(!h_child_name); ++ au_fset_hinjob(flags[PARENT], DIRENT); ++ au_fset_hinjob(flags[CHILD], GEN); ++ au_fset_hinjob(flags[CHILD], TRYXINO0); ++ au_fset_hinjob(flags[CHILD], MNTPNT); ++ break; ++ ++ default: ++ AuDebugOn(1); ++ } ++ ++ if (wh) ++ h_child_inode = NULL; ++ ++ /* iput() and kfree() will be called in postproc() */ ++ /* ++ * inotify_mutex is already acquired and kmalloc/prune_icache may lock ++ * iprune_mutex. strange. ++ */ ++ lockdep_off(); ++ args = kmalloc(sizeof(*args) + len + 1, GFP_NOFS); ++ lockdep_on(); ++ if (unlikely(!args)) { ++ AuErr1("no memory\n"); ++ iput(dir); ++ return; ++ } ++ args->flags[PARENT] = flags[PARENT]; ++ args->flags[CHILD] = flags[CHILD]; ++ args->mask = mask; ++ args->dir = dir; ++ args->h_dir = igrab(watch->inode); ++ if (h_child_inode) ++ h_child_inode = igrab(h_child_inode); /* can be NULL */ ++ args->h_child_inode = h_child_inode; ++ args->h_child_nlen = len; ++ if (len) { ++ p = (void *)args; ++ p += sizeof(*args); ++ memcpy(p, h_child_name, len + 1); ++ } ++ ++ lockdep_off(); ++ wkq_err = au_wkq_nowait(postproc, args, dir->i_sb); ++ lockdep_on(); ++ if (unlikely(wkq_err)) ++ AuErr("wkq %d\n", wkq_err); ++} ++ ++static void aufs_inotify_destroy(struct inotify_watch *watch __maybe_unused) ++{ ++ return; ++} ++ ++static struct inotify_operations aufs_inotify_ops = { ++ .handle_event = aufs_inotify, ++ .destroy_watch = aufs_inotify_destroy ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++static void au_hin_destroy_cache(void) ++{ ++ kmem_cache_destroy(au_cachep[AuCache_HINOTIFY]); ++ au_cachep[AuCache_HINOTIFY] = NULL; ++} ++ ++int __init au_hinotify_init(void) ++{ ++ int err; ++ ++ err = -ENOMEM; ++ au_cachep[AuCache_HINOTIFY] = AuCache(au_hinotify); ++ if (au_cachep[AuCache_HINOTIFY]) { ++ err = 0; ++ au_hin_handle = inotify_init(&aufs_inotify_ops); ++ if (IS_ERR(au_hin_handle)) { ++ err = PTR_ERR(au_hin_handle); ++ au_hin_destroy_cache(); ++ } ++ } ++ AuTraceErr(err); ++ return err; ++} ++ ++void au_hinotify_fin(void) ++{ ++ inotify_destroy(au_hin_handle); ++ if (au_cachep[AuCache_HINOTIFY]) ++ au_hin_destroy_cache(); ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/iinfo.c linux-2.6.31/fs/aufs/iinfo.c +--- linux-2.6.31-vanilla/fs/aufs/iinfo.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/iinfo.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,283 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * inode private data ++ */ ++ ++#include "aufs.h" ++ ++struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex) ++{ ++ struct inode *h_inode; ++ ++ IiMustAnyLock(inode); ++ ++ h_inode = au_ii(inode)->ii_hinode[0 + bindex].hi_inode; ++ AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0); ++ return h_inode; ++} ++ ++/* todo: hard/soft set? */ ++void au_set_ibstart(struct inode *inode, aufs_bindex_t bindex) ++{ ++ struct au_iinfo *iinfo = au_ii(inode); ++ struct inode *h_inode; ++ ++ IiMustWriteLock(inode); ++ ++ iinfo->ii_bstart = bindex; ++ h_inode = iinfo->ii_hinode[bindex + 0].hi_inode; ++ if (h_inode) ++ au_cpup_igen(inode, h_inode); ++} ++ ++void au_hiput(struct au_hinode *hinode) ++{ ++ au_hin_free(hinode); ++ dput(hinode->hi_whdentry); ++ iput(hinode->hi_inode); ++} ++ ++unsigned int au_hi_flags(struct inode *inode, int isdir) ++{ ++ unsigned int flags; ++ const unsigned int mnt_flags = au_mntflags(inode->i_sb); ++ ++ flags = 0; ++ if (au_opt_test(mnt_flags, XINO)) ++ au_fset_hi(flags, XINO); ++ if (isdir && au_opt_test(mnt_flags, UDBA_HINOTIFY)) ++ au_fset_hi(flags, HINOTIFY); ++ return flags; ++} ++ ++void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex, ++ struct inode *h_inode, unsigned int flags) ++{ ++ struct au_hinode *hinode; ++ struct inode *hi; ++ struct au_iinfo *iinfo = au_ii(inode); ++ ++ IiMustWriteLock(inode); ++ ++ hinode = iinfo->ii_hinode + bindex; ++ hi = hinode->hi_inode; ++ AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0); ++ AuDebugOn(h_inode && hi); ++ ++ if (hi) ++ au_hiput(hinode); ++ hinode->hi_inode = h_inode; ++ if (h_inode) { ++ int err; ++ struct super_block *sb = inode->i_sb; ++ struct au_branch *br; ++ ++ if (bindex == iinfo->ii_bstart) ++ au_cpup_igen(inode, h_inode); ++ br = au_sbr(sb, bindex); ++ hinode->hi_id = br->br_id; ++ if (au_ftest_hi(flags, XINO)) { ++ err = au_xino_write(sb, bindex, h_inode->i_ino, ++ inode->i_ino); ++ if (unlikely(err)) ++ AuIOErr1("failed au_xino_write() %d\n", err); ++ } ++ ++ if (au_ftest_hi(flags, HINOTIFY) ++ && au_br_hinotifyable(br->br_perm)) { ++ err = au_hin_alloc(hinode, inode, h_inode); ++ if (unlikely(err)) ++ AuIOErr1("au_hin_alloc() %d\n", err); ++ } ++ } ++} ++ ++void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex, ++ struct dentry *h_wh) ++{ ++ struct au_hinode *hinode; ++ ++ IiMustWriteLock(inode); ++ ++ hinode = au_ii(inode)->ii_hinode + bindex; ++ AuDebugOn(hinode->hi_whdentry); ++ hinode->hi_whdentry = h_wh; ++} ++ ++void au_update_iigen(struct inode *inode) ++{ ++ atomic_set(&au_ii(inode)->ii_generation, au_sigen(inode->i_sb)); ++ /* smp_mb(); */ /* atomic_set */ ++} ++ ++/* it may be called at remount time, too */ ++void au_update_brange(struct inode *inode, int do_put_zero) ++{ ++ struct au_iinfo *iinfo; ++ ++ iinfo = au_ii(inode); ++ if (!iinfo || iinfo->ii_bstart < 0) ++ return; ++ ++ IiMustWriteLock(inode); ++ ++ if (do_put_zero) { ++ aufs_bindex_t bindex; ++ ++ for (bindex = iinfo->ii_bstart; bindex <= iinfo->ii_bend; ++ bindex++) { ++ struct inode *h_i; ++ ++ h_i = iinfo->ii_hinode[0 + bindex].hi_inode; ++ if (h_i && !h_i->i_nlink) ++ au_set_h_iptr(inode, bindex, NULL, 0); ++ } ++ } ++ ++ iinfo->ii_bstart = -1; ++ while (++iinfo->ii_bstart <= iinfo->ii_bend) ++ if (iinfo->ii_hinode[0 + iinfo->ii_bstart].hi_inode) ++ break; ++ if (iinfo->ii_bstart > iinfo->ii_bend) { ++ iinfo->ii_bstart = -1; ++ iinfo->ii_bend = -1; ++ return; ++ } ++ ++ iinfo->ii_bend++; ++ while (0 <= --iinfo->ii_bend) ++ if (iinfo->ii_hinode[0 + iinfo->ii_bend].hi_inode) ++ break; ++ AuDebugOn(iinfo->ii_bstart > iinfo->ii_bend || iinfo->ii_bend < 0); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int au_iinfo_init(struct inode *inode) ++{ ++ struct au_iinfo *iinfo; ++ struct super_block *sb; ++ int nbr, i; ++ ++ sb = inode->i_sb; ++ iinfo = &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo); ++ nbr = au_sbend(sb) + 1; ++ if (unlikely(nbr <= 0)) ++ nbr = 1; ++ iinfo->ii_hinode = kcalloc(nbr, sizeof(*iinfo->ii_hinode), GFP_NOFS); ++ if (iinfo->ii_hinode) { ++ for (i = 0; i < nbr; i++) ++ iinfo->ii_hinode[i].hi_id = -1; ++ ++ atomic_set(&iinfo->ii_generation, au_sigen(sb)); ++ /* smp_mb(); */ /* atomic_set */ ++ au_rw_init(&iinfo->ii_rwsem); ++ iinfo->ii_bstart = -1; ++ iinfo->ii_bend = -1; ++ iinfo->ii_vdir = NULL; ++ return 0; ++ } ++ return -ENOMEM; ++} ++ ++int au_ii_realloc(struct au_iinfo *iinfo, int nbr) ++{ ++ int err, sz; ++ struct au_hinode *hip; ++ ++ AuRwMustWriteLock(&iinfo->ii_rwsem); ++ ++ err = -ENOMEM; ++ sz = sizeof(*hip) * (iinfo->ii_bend + 1); ++ if (!sz) ++ sz = sizeof(*hip); ++ hip = au_kzrealloc(iinfo->ii_hinode, sz, sizeof(*hip) * nbr, GFP_NOFS); ++ if (hip) { ++ iinfo->ii_hinode = hip; ++ err = 0; ++ } ++ ++ return err; ++} ++ ++static int au_iinfo_write0(struct super_block *sb, struct au_hinode *hinode, ++ ino_t ino) ++{ ++ int err; ++ aufs_bindex_t bindex; ++ unsigned char locked; ++ ++ err = 0; ++ locked = !!si_noflush_read_trylock(sb); ++ bindex = au_br_index(sb, hinode->hi_id); ++ if (bindex >= 0) ++ err = au_xino_write0(sb, bindex, hinode->hi_inode->i_ino, ino); ++ /* error action? */ ++ if (locked) ++ si_read_unlock(sb); ++ return err; ++} ++ ++void au_iinfo_fin(struct inode *inode) ++{ ++ ino_t ino; ++ aufs_bindex_t bend; ++ unsigned char unlinked = !inode->i_nlink; ++ struct au_iinfo *iinfo; ++ struct au_hinode *hi; ++ struct super_block *sb; ++ ++ if (unlinked) { ++ int err = au_xigen_inc(inode); ++ if (unlikely(err)) ++ AuWarn1("failed resetting i_generation, %d\n", err); ++ } ++ ++ iinfo = au_ii(inode); ++ /* bad_inode case */ ++ if (!iinfo) ++ return; ++ ++ if (iinfo->ii_vdir) ++ au_vdir_free(iinfo->ii_vdir); ++ ++ if (iinfo->ii_bstart >= 0) { ++ sb = inode->i_sb; ++ ino = 0; ++ if (unlinked) ++ ino = inode->i_ino; ++ hi = iinfo->ii_hinode + iinfo->ii_bstart; ++ bend = iinfo->ii_bend; ++ while (iinfo->ii_bstart++ <= bend) { ++ if (hi->hi_inode) { ++ if (unlinked || !hi->hi_inode->i_nlink) { ++ au_iinfo_write0(sb, hi, ino); ++ /* ignore this error */ ++ ino = 0; ++ } ++ au_hiput(hi); ++ } ++ hi++; ++ } ++ } ++ ++ kfree(iinfo->ii_hinode); ++ AuRwDestroy(&iinfo->ii_rwsem); ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/inode.c linux-2.6.31/fs/aufs/inode.c +--- linux-2.6.31-vanilla/fs/aufs/inode.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/inode.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,413 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * inode functions ++ */ ++ ++#include "aufs.h" ++ ++struct inode *au_igrab(struct inode *inode) ++{ ++ if (inode) { ++ AuDebugOn(!atomic_read(&inode->i_count)); ++ atomic_inc_return(&inode->i_count); ++ } ++ return inode; ++} ++ ++static void au_refresh_hinode_attr(struct inode *inode, int do_version) ++{ ++ au_cpup_attr_all(inode, /*force*/0); ++ au_update_iigen(inode); ++ if (do_version) ++ inode->i_version++; ++} ++ ++int au_refresh_hinode_self(struct inode *inode, int do_attr) ++{ ++ int err; ++ aufs_bindex_t bindex, new_bindex; ++ unsigned char update; ++ struct inode *first; ++ struct au_hinode *p, *q, tmp; ++ struct super_block *sb; ++ struct au_iinfo *iinfo; ++ ++ IiMustWriteLock(inode); ++ ++ update = 0; ++ sb = inode->i_sb; ++ iinfo = au_ii(inode); ++ err = au_ii_realloc(iinfo, au_sbend(sb) + 1); ++ if (unlikely(err)) ++ goto out; ++ ++ p = iinfo->ii_hinode + iinfo->ii_bstart; ++ first = p->hi_inode; ++ err = 0; ++ for (bindex = iinfo->ii_bstart; bindex <= iinfo->ii_bend; ++ bindex++, p++) { ++ if (!p->hi_inode) ++ continue; ++ ++ new_bindex = au_br_index(sb, p->hi_id); ++ if (new_bindex == bindex) ++ continue; ++ ++ if (new_bindex < 0) { ++ update++; ++ au_hiput(p); ++ p->hi_inode = NULL; ++ continue; ++ } ++ ++ if (new_bindex < iinfo->ii_bstart) ++ iinfo->ii_bstart = new_bindex; ++ if (iinfo->ii_bend < new_bindex) ++ iinfo->ii_bend = new_bindex; ++ /* swap two lower inode, and loop again */ ++ q = iinfo->ii_hinode + new_bindex; ++ tmp = *q; ++ *q = *p; ++ *p = tmp; ++ if (tmp.hi_inode) { ++ bindex--; ++ p--; ++ } ++ } ++ au_update_brange(inode, /*do_put_zero*/0); ++ if (do_attr) ++ au_refresh_hinode_attr(inode, update && S_ISDIR(inode->i_mode)); ++ ++ out: ++ return err; ++} ++ ++int au_refresh_hinode(struct inode *inode, struct dentry *dentry) ++{ ++ int err, update; ++ unsigned int flags; ++ aufs_bindex_t bindex, bend; ++ unsigned char isdir; ++ struct inode *first; ++ struct au_hinode *p; ++ struct au_iinfo *iinfo; ++ ++ err = au_refresh_hinode_self(inode, /*do_attr*/0); ++ if (unlikely(err)) ++ goto out; ++ ++ update = 0; ++ iinfo = au_ii(inode); ++ p = iinfo->ii_hinode + iinfo->ii_bstart; ++ first = p->hi_inode; ++ isdir = S_ISDIR(inode->i_mode); ++ flags = au_hi_flags(inode, isdir); ++ bend = au_dbend(dentry); ++ for (bindex = au_dbstart(dentry); bindex <= bend; bindex++) { ++ struct inode *h_i; ++ struct dentry *h_d; ++ ++ h_d = au_h_dptr(dentry, bindex); ++ if (!h_d || !h_d->d_inode) ++ continue; ++ ++ if (iinfo->ii_bstart <= bindex && bindex <= iinfo->ii_bend) { ++ h_i = au_h_iptr(inode, bindex); ++ if (h_i) { ++ if (h_i == h_d->d_inode) ++ continue; ++ err = -EIO; ++ break; ++ } ++ } ++ if (bindex < iinfo->ii_bstart) ++ iinfo->ii_bstart = bindex; ++ if (iinfo->ii_bend < bindex) ++ iinfo->ii_bend = bindex; ++ au_set_h_iptr(inode, bindex, au_igrab(h_d->d_inode), flags); ++ update = 1; ++ } ++ au_update_brange(inode, /*do_put_zero*/0); ++ ++ if (unlikely(err)) ++ goto out; ++ ++ au_refresh_hinode_attr(inode, update && isdir); ++ ++ out: ++ return err; ++} ++ ++static int set_inode(struct inode *inode, struct dentry *dentry) ++{ ++ int err; ++ unsigned int flags; ++ umode_t mode; ++ aufs_bindex_t bindex, bstart, btail; ++ unsigned char isdir; ++ struct dentry *h_dentry; ++ struct inode *h_inode; ++ struct au_iinfo *iinfo; ++ ++ IiMustWriteLock(inode); ++ ++ err = 0; ++ isdir = 0; ++ bstart = au_dbstart(dentry); ++ h_inode = au_h_dptr(dentry, bstart)->d_inode; ++ mode = h_inode->i_mode; ++ switch (mode & S_IFMT) { ++ case S_IFREG: ++ btail = au_dbtail(dentry); ++ inode->i_op = &aufs_iop; ++ inode->i_fop = &aufs_file_fop; ++ inode->i_mapping->a_ops = &aufs_aop; ++ break; ++ case S_IFDIR: ++ isdir = 1; ++ btail = au_dbtaildir(dentry); ++ inode->i_op = &aufs_dir_iop; ++ inode->i_fop = &aufs_dir_fop; ++ break; ++ case S_IFLNK: ++ btail = au_dbtail(dentry); ++ inode->i_op = &aufs_symlink_iop; ++ break; ++ case S_IFBLK: ++ case S_IFCHR: ++ case S_IFIFO: ++ case S_IFSOCK: ++ btail = au_dbtail(dentry); ++ inode->i_op = &aufs_iop; ++ init_special_inode(inode, mode, h_inode->i_rdev); ++ break; ++ default: ++ AuIOErr("Unknown file type 0%o\n", mode); ++ err = -EIO; ++ goto out; ++ } ++ ++ /* do not set inotify for whiteouted dirs (SHWH mode) */ ++ flags = au_hi_flags(inode, isdir); ++ if (au_opt_test(au_mntflags(dentry->d_sb), SHWH) ++ && au_ftest_hi(flags, HINOTIFY) ++ && dentry->d_name.len > AUFS_WH_PFX_LEN ++ && !memcmp(dentry->d_name.name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) ++ au_fclr_hi(flags, HINOTIFY); ++ iinfo = au_ii(inode); ++ iinfo->ii_bstart = bstart; ++ iinfo->ii_bend = btail; ++ for (bindex = bstart; bindex <= btail; bindex++) { ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (h_dentry) ++ au_set_h_iptr(inode, bindex, ++ au_igrab(h_dentry->d_inode), flags); ++ } ++ au_cpup_attr_all(inode, /*force*/1); ++ ++ out: ++ return err; ++} ++ ++/* successful returns with iinfo write_locked */ ++static int reval_inode(struct inode *inode, struct dentry *dentry, int *matched) ++{ ++ int err; ++ aufs_bindex_t bindex, bend; ++ struct inode *h_inode, *h_dinode; ++ ++ *matched = 0; ++ ++ /* ++ * before this function, if aufs got any iinfo lock, it must be only ++ * one, the parent dir. ++ * it can happen by UDBA and the obsoleted inode number. ++ */ ++ err = -EIO; ++ if (unlikely(inode->i_ino == parent_ino(dentry))) ++ goto out; ++ ++ err = 0; ++ ii_write_lock_new_child(inode); ++ h_dinode = au_h_dptr(dentry, au_dbstart(dentry))->d_inode; ++ bend = au_ibend(inode); ++ for (bindex = au_ibstart(inode); bindex <= bend; bindex++) { ++ h_inode = au_h_iptr(inode, bindex); ++ if (h_inode && h_inode == h_dinode) { ++ *matched = 1; ++ err = 0; ++ if (au_iigen(inode) != au_digen(dentry)) ++ err = au_refresh_hinode(inode, dentry); ++ break; ++ } ++ } ++ ++ if (unlikely(err)) ++ ii_write_unlock(inode); ++ out: ++ return err; ++} ++ ++int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, ++ unsigned int d_type, ino_t *ino) ++{ ++ int err; ++ struct mutex *mtx; ++ const int isdir = (d_type == DT_DIR); ++ ++ /* prevent hardlinks from race condition */ ++ mtx = NULL; ++ if (!isdir) { ++ mtx = &au_sbr(sb, bindex)->br_xino.xi_nondir_mtx; ++ mutex_lock(mtx); ++ } ++ err = au_xino_read(sb, bindex, h_ino, ino); ++ if (unlikely(err)) ++ goto out; ++ ++ if (!*ino) { ++ err = -EIO; ++ *ino = au_xino_new_ino(sb); ++ if (unlikely(!*ino)) ++ goto out; ++ err = au_xino_write(sb, bindex, h_ino, *ino); ++ if (unlikely(err)) ++ goto out; ++ } ++ ++ out: ++ if (!isdir) ++ mutex_unlock(mtx); ++ return err; ++} ++ ++/* successful returns with iinfo write_locked */ ++/* todo: return with unlocked? */ ++struct inode *au_new_inode(struct dentry *dentry, int must_new) ++{ ++ struct inode *inode; ++ struct dentry *h_dentry; ++ struct super_block *sb; ++ ino_t h_ino, ino; ++ int err, match; ++ aufs_bindex_t bstart; ++ ++ sb = dentry->d_sb; ++ bstart = au_dbstart(dentry); ++ h_dentry = au_h_dptr(dentry, bstart); ++ h_ino = h_dentry->d_inode->i_ino; ++ err = au_xino_read(sb, bstart, h_ino, &ino); ++ inode = ERR_PTR(err); ++ if (unlikely(err)) ++ goto out; ++ new_ino: ++ if (!ino) { ++ ino = au_xino_new_ino(sb); ++ if (unlikely(!ino)) { ++ inode = ERR_PTR(-EIO); ++ goto out; ++ } ++ } ++ ++ AuDbg("i%lu\n", (unsigned long)ino); ++ inode = au_iget_locked(sb, ino); ++ err = PTR_ERR(inode); ++ if (IS_ERR(inode)) ++ goto out; ++ ++ AuDbg("%lx, new %d\n", inode->i_state, !!(inode->i_state & I_NEW)); ++ if (inode->i_state & I_NEW) { ++ ii_write_lock_new_child(inode); ++ err = set_inode(inode, dentry); ++ unlock_new_inode(inode); ++ if (!err) ++ goto out; /* success */ ++ ++ iget_failed(inode); ++ ii_write_unlock(inode); ++ goto out_iput; ++ } else if (!must_new) { ++ err = reval_inode(inode, dentry, &match); ++ if (!err) ++ goto out; /* success */ ++ else if (match) ++ goto out_iput; ++ } ++ ++ if (unlikely(au_test_fs_unique_ino(h_dentry->d_inode))) ++ AuWarn1("Warning: Un-notified UDBA or repeatedly renamed dir," ++ " b%d, %s, %.*s, hi%lu, i%lu.\n", ++ bstart, au_sbtype(h_dentry->d_sb), AuDLNPair(dentry), ++ (unsigned long)h_ino, (unsigned long)ino); ++ ino = 0; ++ err = au_xino_write(sb, bstart, h_ino, /*ino*/0); ++ if (!err) { ++ iput(inode); ++ goto new_ino; ++ } ++ ++ out_iput: ++ iput(inode); ++ inode = ERR_PTR(err); ++ out: ++ return inode; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int au_test_ro(struct super_block *sb, aufs_bindex_t bindex, ++ struct inode *inode) ++{ ++ int err; ++ ++ err = au_br_rdonly(au_sbr(sb, bindex)); ++ ++ /* pseudo-link after flushed may happen out of bounds */ ++ if (!err ++ && inode ++ && au_ibstart(inode) <= bindex ++ && bindex <= au_ibend(inode)) { ++ /* ++ * permission check is unnecessary since vfsub routine ++ * will be called later ++ */ ++ struct inode *hi = au_h_iptr(inode, bindex); ++ if (hi) ++ err = IS_IMMUTABLE(hi) ? -EROFS : 0; ++ } ++ ++ return err; ++} ++ ++int au_test_h_perm(struct inode *h_inode, int mask) ++{ ++ if (!current_fsuid()) ++ return 0; ++ return inode_permission(h_inode, mask); ++} ++ ++int au_test_h_perm_sio(struct inode *h_inode, int mask) ++{ ++ if (au_test_nfs(h_inode->i_sb) ++ && (mask & MAY_WRITE) ++ && S_ISDIR(h_inode->i_mode)) ++ mask |= MAY_READ; /* force permission check */ ++ return au_test_h_perm(h_inode, mask); ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/inode.h linux-2.6.31/fs/aufs/inode.h +--- linux-2.6.31-vanilla/fs/aufs/inode.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/inode.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,497 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * inode operations ++ */ ++ ++#ifndef __AUFS_INODE_H__ ++#define __AUFS_INODE_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++#include ++#include "rwsem.h" ++ ++struct vfsmount; ++ ++struct au_hinotify { ++#ifdef CONFIG_AUFS_HINOTIFY ++ struct inotify_watch hin_watch; ++ struct inode *hin_aufs_inode; /* no get/put */ ++#endif ++}; ++ ++struct au_hinode { ++ struct inode *hi_inode; ++ aufs_bindex_t hi_id; ++#ifdef CONFIG_AUFS_HINOTIFY ++ struct au_hinotify *hi_notify; ++#endif ++ ++ /* reference to the copied-up whiteout with get/put */ ++ struct dentry *hi_whdentry; ++}; ++ ++struct au_vdir; ++struct au_iinfo { ++ atomic_t ii_generation; ++ struct super_block *ii_hsb1; /* no get/put */ ++ ++ struct au_rwsem ii_rwsem; ++ aufs_bindex_t ii_bstart, ii_bend; ++ __u32 ii_higen; ++ struct au_hinode *ii_hinode; ++ struct au_vdir *ii_vdir; ++}; ++ ++struct au_icntnr { ++ struct au_iinfo iinfo; ++ struct inode vfs_inode; ++}; ++ ++/* au_pin flags */ ++#define AuPin_DI_LOCKED 1 ++#define AuPin_MNT_WRITE (1 << 1) ++#define au_ftest_pin(flags, name) ((flags) & AuPin_##name) ++#define au_fset_pin(flags, name) { (flags) |= AuPin_##name; } ++#define au_fclr_pin(flags, name) { (flags) &= ~AuPin_##name; } ++ ++struct au_pin { ++ /* input */ ++ struct dentry *dentry; ++ unsigned int udba; ++ unsigned char lsc_di, lsc_hi, flags; ++ aufs_bindex_t bindex; ++ ++ /* output */ ++ struct dentry *parent; ++ struct au_hinode *hdir; ++ struct vfsmount *h_mnt; ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline struct au_iinfo *au_ii(struct inode *inode) ++{ ++ struct au_iinfo *iinfo; ++ ++ iinfo = &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo); ++ if (iinfo->ii_hinode) ++ return iinfo; ++ return NULL; /* debugging bad_inode case */ ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* inode.c */ ++struct inode *au_igrab(struct inode *inode); ++int au_refresh_hinode_self(struct inode *inode, int do_attr); ++int au_refresh_hinode(struct inode *inode, struct dentry *dentry); ++int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, ++ unsigned int d_type, ino_t *ino); ++struct inode *au_new_inode(struct dentry *dentry, int must_new); ++int au_test_ro(struct super_block *sb, aufs_bindex_t bindex, ++ struct inode *inode); ++int au_test_h_perm(struct inode *h_inode, int mask); ++int au_test_h_perm_sio(struct inode *h_inode, int mask); ++ ++static inline int au_wh_ino(struct super_block *sb, aufs_bindex_t bindex, ++ ino_t h_ino, unsigned int d_type, ino_t *ino) ++{ ++#ifdef CONFIG_AUFS_SHWH ++ return au_ino(sb, bindex, h_ino, d_type, ino); ++#else ++ return 0; ++#endif ++} ++ ++/* i_op.c */ ++extern struct inode_operations aufs_iop, aufs_symlink_iop, aufs_dir_iop; ++ ++/* au_wr_dir flags */ ++#define AuWrDir_ADD_ENTRY 1 ++#define AuWrDir_ISDIR (1 << 1) ++#define au_ftest_wrdir(flags, name) ((flags) & AuWrDir_##name) ++#define au_fset_wrdir(flags, name) { (flags) |= AuWrDir_##name; } ++#define au_fclr_wrdir(flags, name) { (flags) &= ~AuWrDir_##name; } ++ ++struct au_wr_dir_args { ++ aufs_bindex_t force_btgt; ++ unsigned char flags; ++}; ++int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry, ++ struct au_wr_dir_args *args); ++ ++struct dentry *au_pinned_h_parent(struct au_pin *pin); ++void au_pin_init(struct au_pin *pin, struct dentry *dentry, ++ aufs_bindex_t bindex, int lsc_di, int lsc_hi, ++ unsigned int udba, unsigned char flags); ++int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex, ++ unsigned int udba, unsigned char flags) __must_check; ++int au_do_pin(struct au_pin *pin) __must_check; ++void au_unpin(struct au_pin *pin); ++ ++/* i_op_add.c */ ++int au_may_add(struct dentry *dentry, aufs_bindex_t bindex, ++ struct dentry *h_parent, int isdir); ++int aufs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev); ++int aufs_symlink(struct inode *dir, struct dentry *dentry, const char *symname); ++int aufs_create(struct inode *dir, struct dentry *dentry, int mode, ++ struct nameidata *nd); ++int aufs_link(struct dentry *src_dentry, struct inode *dir, ++ struct dentry *dentry); ++int aufs_mkdir(struct inode *dir, struct dentry *dentry, int mode); ++ ++/* i_op_del.c */ ++int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup); ++int au_may_del(struct dentry *dentry, aufs_bindex_t bindex, ++ struct dentry *h_parent, int isdir); ++int aufs_unlink(struct inode *dir, struct dentry *dentry); ++int aufs_rmdir(struct inode *dir, struct dentry *dentry); ++ ++/* i_op_ren.c */ ++int au_wbr(struct dentry *dentry, aufs_bindex_t btgt); ++int aufs_rename(struct inode *src_dir, struct dentry *src_dentry, ++ struct inode *dir, struct dentry *dentry); ++ ++/* iinfo.c */ ++struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex); ++void au_hiput(struct au_hinode *hinode); ++void au_set_ibstart(struct inode *inode, aufs_bindex_t bindex); ++void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex, ++ struct dentry *h_wh); ++unsigned int au_hi_flags(struct inode *inode, int isdir); ++ ++/* hinode flags */ ++#define AuHi_XINO 1 ++#define AuHi_HINOTIFY (1 << 1) ++#define au_ftest_hi(flags, name) ((flags) & AuHi_##name) ++#define au_fset_hi(flags, name) { (flags) |= AuHi_##name; } ++#define au_fclr_hi(flags, name) { (flags) &= ~AuHi_##name; } ++ ++#ifndef CONFIG_AUFS_HINOTIFY ++#undef AuHi_HINOTIFY ++#define AuHi_HINOTIFY 0 ++#endif ++ ++void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex, ++ struct inode *h_inode, unsigned int flags); ++ ++void au_update_iigen(struct inode *inode); ++void au_update_brange(struct inode *inode, int do_put_zero); ++ ++int au_iinfo_init(struct inode *inode); ++void au_iinfo_fin(struct inode *inode); ++int au_ii_realloc(struct au_iinfo *iinfo, int nbr); ++ ++/* plink.c */ ++void au_plink_block_maintain(struct super_block *sb); ++#ifdef CONFIG_AUFS_DEBUG ++void au_plink_list(struct super_block *sb); ++#else ++static inline void au_plink_list(struct super_block *sb) ++{ ++ /* nothing */ ++} ++#endif ++int au_plink_test(struct inode *inode); ++struct dentry *au_plink_lkup(struct inode *inode, aufs_bindex_t bindex); ++void au_plink_append(struct inode *inode, aufs_bindex_t bindex, ++ struct dentry *h_dentry); ++void au_plink_put(struct super_block *sb); ++void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id); ++long au_plink_ioctl(struct file *file, unsigned int cmd); ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* lock subclass for iinfo */ ++enum { ++ AuLsc_II_CHILD, /* child first */ ++ AuLsc_II_CHILD2, /* rename(2), link(2), and cpup at hinotify */ ++ AuLsc_II_CHILD3, /* copyup dirs */ ++ AuLsc_II_PARENT, /* see AuLsc_I_PARENT in vfsub.h */ ++ AuLsc_II_PARENT2, ++ AuLsc_II_PARENT3, /* copyup dirs */ ++ AuLsc_II_NEW_CHILD ++}; ++ ++/* ++ * ii_read_lock_child, ii_write_lock_child, ++ * ii_read_lock_child2, ii_write_lock_child2, ++ * ii_read_lock_child3, ii_write_lock_child3, ++ * ii_read_lock_parent, ii_write_lock_parent, ++ * ii_read_lock_parent2, ii_write_lock_parent2, ++ * ii_read_lock_parent3, ii_write_lock_parent3, ++ * ii_read_lock_new_child, ii_write_lock_new_child, ++ */ ++#define AuReadLockFunc(name, lsc) \ ++static inline void ii_read_lock_##name(struct inode *i) \ ++{ \ ++ au_rw_read_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \ ++} ++ ++#define AuWriteLockFunc(name, lsc) \ ++static inline void ii_write_lock_##name(struct inode *i) \ ++{ \ ++ au_rw_write_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \ ++} ++ ++#define AuRWLockFuncs(name, lsc) \ ++ AuReadLockFunc(name, lsc) \ ++ AuWriteLockFunc(name, lsc) ++ ++AuRWLockFuncs(child, CHILD); ++AuRWLockFuncs(child2, CHILD2); ++AuRWLockFuncs(child3, CHILD3); ++AuRWLockFuncs(parent, PARENT); ++AuRWLockFuncs(parent2, PARENT2); ++AuRWLockFuncs(parent3, PARENT3); ++AuRWLockFuncs(new_child, NEW_CHILD); ++ ++#undef AuReadLockFunc ++#undef AuWriteLockFunc ++#undef AuRWLockFuncs ++ ++/* ++ * ii_read_unlock, ii_write_unlock, ii_downgrade_lock ++ */ ++AuSimpleUnlockRwsemFuncs(ii, struct inode *i, &au_ii(i)->ii_rwsem); ++ ++#define IiMustNoWaiters(i) AuRwMustNoWaiters(&au_ii(i)->ii_rwsem) ++#define IiMustAnyLock(i) AuRwMustAnyLock(&au_ii(i)->ii_rwsem) ++#define IiMustWriteLock(i) AuRwMustWriteLock(&au_ii(i)->ii_rwsem) ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline unsigned int au_iigen(struct inode *inode) ++{ ++ return atomic_read(&au_ii(inode)->ii_generation); ++} ++ ++/* tiny test for inode number */ ++/* tmpfs generation is too rough */ ++static inline int au_test_higen(struct inode *inode, struct inode *h_inode) ++{ ++ struct au_iinfo *iinfo; ++ ++ iinfo = au_ii(inode); ++ AuRwMustAnyLock(&iinfo->ii_rwsem); ++ return !(iinfo->ii_hsb1 == h_inode->i_sb ++ && iinfo->ii_higen == h_inode->i_generation); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline aufs_bindex_t au_ii_br_id(struct inode *inode, ++ aufs_bindex_t bindex) ++{ ++ IiMustAnyLock(inode); ++ return au_ii(inode)->ii_hinode[0 + bindex].hi_id; ++} ++ ++static inline aufs_bindex_t au_ibstart(struct inode *inode) ++{ ++ IiMustAnyLock(inode); ++ return au_ii(inode)->ii_bstart; ++} ++ ++static inline aufs_bindex_t au_ibend(struct inode *inode) ++{ ++ IiMustAnyLock(inode); ++ return au_ii(inode)->ii_bend; ++} ++ ++static inline struct au_vdir *au_ivdir(struct inode *inode) ++{ ++ IiMustAnyLock(inode); ++ return au_ii(inode)->ii_vdir; ++} ++ ++static inline struct dentry *au_hi_wh(struct inode *inode, aufs_bindex_t bindex) ++{ ++ IiMustAnyLock(inode); ++ return au_ii(inode)->ii_hinode[0 + bindex].hi_whdentry; ++} ++ ++static inline void au_set_ibend(struct inode *inode, aufs_bindex_t bindex) ++{ ++ IiMustWriteLock(inode); ++ au_ii(inode)->ii_bend = bindex; ++} ++ ++static inline void au_set_ivdir(struct inode *inode, struct au_vdir *vdir) ++{ ++ IiMustWriteLock(inode); ++ au_ii(inode)->ii_vdir = vdir; ++} ++ ++static inline struct au_hinode *au_hi(struct inode *inode, aufs_bindex_t bindex) ++{ ++ IiMustAnyLock(inode); ++ return au_ii(inode)->ii_hinode + bindex; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline struct dentry *au_pinned_parent(struct au_pin *pin) ++{ ++ if (pin) ++ return pin->parent; ++ return NULL; ++} ++ ++static inline struct inode *au_pinned_h_dir(struct au_pin *pin) ++{ ++ if (pin && pin->hdir) ++ return pin->hdir->hi_inode; ++ return NULL; ++} ++ ++static inline struct au_hinode *au_pinned_hdir(struct au_pin *pin) ++{ ++ if (pin) ++ return pin->hdir; ++ return NULL; ++} ++ ++static inline void au_pin_set_dentry(struct au_pin *pin, struct dentry *dentry) ++{ ++ if (pin) ++ pin->dentry = dentry; ++} ++ ++static inline void au_pin_set_parent_lflag(struct au_pin *pin, ++ unsigned char lflag) ++{ ++ if (pin) { ++ /* dirty macros require brackets */ ++ if (lflag) { ++ au_fset_pin(pin->flags, DI_LOCKED); ++ } else { ++ au_fclr_pin(pin->flags, DI_LOCKED); ++ } ++ } ++} ++ ++static inline void au_pin_set_parent(struct au_pin *pin, struct dentry *parent) ++{ ++ if (pin) { ++ dput(pin->parent); ++ pin->parent = dget(parent); ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++#ifdef CONFIG_AUFS_HINOTIFY ++/* hinotify.c */ ++int au_hin_alloc(struct au_hinode *hinode, struct inode *inode, ++ struct inode *h_inode); ++void au_hin_free(struct au_hinode *hinode); ++void au_hin_ctl(struct au_hinode *hinode, int do_set); ++void au_reset_hinotify(struct inode *inode, unsigned int flags); ++ ++int __init au_hinotify_init(void); ++void au_hinotify_fin(void); ++ ++static inline ++void au_hin_init(struct au_hinode *hinode, struct au_hinotify *val) ++{ ++ hinode->hi_notify = val; ++} ++ ++static inline void au_iigen_dec(struct inode *inode) ++{ ++ atomic_dec_return(&au_ii(inode)->ii_generation); ++} ++ ++#else ++static inline ++int au_hin_alloc(struct au_hinode *hinode __maybe_unused, ++ struct inode *inode __maybe_unused, ++ struct inode *h_inode __maybe_unused) ++{ ++ return -EOPNOTSUPP; ++} ++ ++static inline void au_hin_free(struct au_hinode *hinode __maybe_unused) ++{ ++ /* nothing */ ++} ++ ++static inline void au_hin_ctl(struct au_hinode *hinode __maybe_unused, ++ int do_set __maybe_unused) ++{ ++ /* nothing */ ++} ++ ++static inline void au_reset_hinotify(struct inode *inode __maybe_unused, ++ unsigned int flags __maybe_unused) ++{ ++ /* nothing */ ++} ++ ++static inline int au_hinotify_init(void) ++{ ++ return 0; ++} ++ ++#define au_hinotify_fin() do {} while (0) ++ ++static inline ++void au_hin_init(struct au_hinode *hinode __maybe_unused, ++ struct au_hinotify *val __maybe_unused) ++{ ++ /* empty */ ++} ++#endif /* CONFIG_AUFS_HINOTIFY */ ++ ++static inline void au_hin_suspend(struct au_hinode *hdir) ++{ ++ au_hin_ctl(hdir, /*do_set*/0); ++} ++ ++static inline void au_hin_resume(struct au_hinode *hdir) ++{ ++ au_hin_ctl(hdir, /*do_set*/1); ++} ++ ++static inline void au_hin_imtx_lock(struct au_hinode *hdir) ++{ ++ mutex_lock(&hdir->hi_inode->i_mutex); ++ au_hin_suspend(hdir); ++} ++ ++static inline void au_hin_imtx_lock_nested(struct au_hinode *hdir, ++ unsigned int sc __maybe_unused) ++{ ++ mutex_lock_nested(&hdir->hi_inode->i_mutex, sc); ++ au_hin_suspend(hdir); ++} ++ ++static inline void au_hin_imtx_unlock(struct au_hinode *hdir) ++{ ++ au_hin_resume(hdir); ++ mutex_unlock(&hdir->hi_inode->i_mutex); ++} ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_INODE_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/ioctl.c linux-2.6.31/fs/aufs/ioctl.c +--- linux-2.6.31-vanilla/fs/aufs/ioctl.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/ioctl.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,47 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * ioctl ++ * plink-management and readdir in userspace. ++ */ ++ ++#include "aufs.h" ++ ++long aufs_ioctl_dir(struct file *file, unsigned int cmd, unsigned long arg) ++{ ++ long err; ++ ++ switch (cmd) { ++ case AUFS_CTL_PLINK_MAINT: ++ case AUFS_CTL_PLINK_CLEAN: ++ err = au_plink_ioctl(file, cmd); ++ break; ++ ++ case AUFS_CTL_RDU: ++ case AUFS_CTL_RDU_INO: ++ err = au_rdu_ioctl(file, cmd, arg); ++ break; ++ ++ default: ++ err = -EINVAL; ++ } ++ ++ AuTraceErr(err); ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/i_op_add.c linux-2.6.31/fs/aufs/i_op_add.c +--- linux-2.6.31-vanilla/fs/aufs/i_op_add.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/i_op_add.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,649 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * inode operations (add entry) ++ */ ++ ++#include "aufs.h" ++ ++/* ++ * final procedure of adding a new entry, except link(2). ++ * remove whiteout, instantiate, copyup the parent dir's times and size ++ * and update version. ++ * if it failed, re-create the removed whiteout. ++ */ ++static int epilog(struct inode *dir, aufs_bindex_t bindex, ++ struct dentry *wh_dentry, struct dentry *dentry) ++{ ++ int err, rerr; ++ aufs_bindex_t bwh; ++ struct path h_path; ++ struct inode *inode, *h_dir; ++ struct dentry *wh; ++ ++ bwh = -1; ++ if (wh_dentry) { ++ h_dir = wh_dentry->d_parent->d_inode; /* dir inode is locked */ ++ IMustLock(h_dir); ++ AuDebugOn(au_h_iptr(dir, bindex) != h_dir); ++ bwh = au_dbwh(dentry); ++ h_path.dentry = wh_dentry; ++ h_path.mnt = au_sbr_mnt(dir->i_sb, bindex); ++ err = au_wh_unlink_dentry(au_h_iptr(dir, bindex), &h_path, ++ dentry); ++ if (unlikely(err)) ++ goto out; ++ } ++ ++ inode = au_new_inode(dentry, /*must_new*/1); ++ if (!IS_ERR(inode)) { ++ d_instantiate(dentry, inode); ++ dir = dentry->d_parent->d_inode; /* dir inode is locked */ ++ IMustLock(dir); ++ if (au_ibstart(dir) == au_dbstart(dentry)) ++ au_cpup_attr_timesizes(dir); ++ dir->i_version++; ++ return 0; /* success */ ++ } ++ ++ err = PTR_ERR(inode); ++ if (!wh_dentry) ++ goto out; ++ ++ /* revert */ ++ /* dir inode is locked */ ++ wh = au_wh_create(dentry, bwh, wh_dentry->d_parent); ++ rerr = PTR_ERR(wh); ++ if (IS_ERR(wh)) { ++ AuIOErr("%.*s reverting whiteout failed(%d, %d)\n", ++ AuDLNPair(dentry), err, rerr); ++ err = -EIO; ++ } else ++ dput(wh); ++ ++ out: ++ return err; ++} ++ ++/* ++ * simple tests for the adding inode operations. ++ * following the checks in vfs, plus the parent-child relationship. ++ */ ++int au_may_add(struct dentry *dentry, aufs_bindex_t bindex, ++ struct dentry *h_parent, int isdir) ++{ ++ int err; ++ umode_t h_mode; ++ struct dentry *h_dentry; ++ struct inode *h_inode; ++ ++ h_dentry = au_h_dptr(dentry, bindex); ++ h_inode = h_dentry->d_inode; ++ if (!dentry->d_inode) { ++ err = -EEXIST; ++ if (unlikely(h_inode)) ++ goto out; ++ } else { ++ /* rename(2) case */ ++ err = -EIO; ++ if (unlikely(!h_inode || !h_inode->i_nlink)) ++ goto out; ++ ++ h_mode = h_inode->i_mode; ++ if (!isdir) { ++ err = -EISDIR; ++ if (unlikely(S_ISDIR(h_mode))) ++ goto out; ++ } else if (unlikely(!S_ISDIR(h_mode))) { ++ err = -ENOTDIR; ++ goto out; ++ } ++ } ++ ++ err = -EIO; ++ /* expected parent dir is locked */ ++ if (unlikely(h_parent != h_dentry->d_parent)) ++ goto out; ++ err = 0; ++ ++ out: ++ return err; ++} ++ ++/* ++ * initial procedure of adding a new entry. ++ * prepare writable branch and the parent dir, lock it, ++ * and lookup whiteout for the new entry. ++ */ ++static struct dentry* ++lock_hdir_lkup_wh(struct dentry *dentry, struct au_dtime *dt, ++ struct dentry *src_dentry, struct au_pin *pin, ++ struct au_wr_dir_args *wr_dir_args) ++{ ++ struct dentry *wh_dentry, *h_parent; ++ struct super_block *sb; ++ struct au_branch *br; ++ int err; ++ unsigned int udba; ++ aufs_bindex_t bcpup; ++ ++ err = au_wr_dir(dentry, src_dentry, wr_dir_args); ++ bcpup = err; ++ wh_dentry = ERR_PTR(err); ++ if (unlikely(err < 0)) ++ goto out; ++ ++ sb = dentry->d_sb; ++ udba = au_opt_udba(sb); ++ err = au_pin(pin, dentry, bcpup, udba, ++ AuPin_DI_LOCKED | AuPin_MNT_WRITE); ++ wh_dentry = ERR_PTR(err); ++ if (unlikely(err)) ++ goto out; ++ ++ h_parent = au_pinned_h_parent(pin); ++ if (udba != AuOpt_UDBA_NONE ++ && au_dbstart(dentry) == bcpup) { ++ err = au_may_add(dentry, bcpup, h_parent, ++ au_ftest_wrdir(wr_dir_args->flags, ISDIR)); ++ wh_dentry = ERR_PTR(err); ++ if (unlikely(err)) ++ goto out_unpin; ++ } ++ ++ br = au_sbr(sb, bcpup); ++ if (dt) { ++ struct path tmp = { ++ .dentry = h_parent, ++ .mnt = br->br_mnt ++ }; ++ au_dtime_store(dt, au_pinned_parent(pin), &tmp); ++ } ++ ++ wh_dentry = NULL; ++ if (bcpup != au_dbwh(dentry)) ++ goto out; /* success */ ++ ++ wh_dentry = au_wh_lkup(h_parent, &dentry->d_name, br); ++ ++ out_unpin: ++ if (IS_ERR(wh_dentry)) ++ au_unpin(pin); ++ out: ++ return wh_dentry; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++enum { Mknod, Symlink, Creat }; ++struct simple_arg { ++ int type; ++ union { ++ struct { ++ int mode; ++ struct nameidata *nd; ++ } c; ++ struct { ++ const char *symname; ++ } s; ++ struct { ++ int mode; ++ dev_t dev; ++ } m; ++ } u; ++}; ++ ++static int add_simple(struct inode *dir, struct dentry *dentry, ++ struct simple_arg *arg) ++{ ++ int err; ++ aufs_bindex_t bstart; ++ unsigned char created; ++ struct au_dtime dt; ++ struct au_pin pin; ++ struct path h_path; ++ struct dentry *wh_dentry, *parent; ++ struct inode *h_dir; ++ struct au_wr_dir_args wr_dir_args = { ++ .force_btgt = -1, ++ .flags = AuWrDir_ADD_ENTRY ++ }; ++ ++ IMustLock(dir); ++ ++ parent = dentry->d_parent; /* dir inode is locked */ ++ aufs_read_lock(dentry, AuLock_DW); ++ di_write_lock_parent(parent); ++ wh_dentry = lock_hdir_lkup_wh(dentry, &dt, /*src_dentry*/NULL, &pin, ++ &wr_dir_args); ++ err = PTR_ERR(wh_dentry); ++ if (IS_ERR(wh_dentry)) ++ goto out; ++ ++ bstart = au_dbstart(dentry); ++ h_path.dentry = au_h_dptr(dentry, bstart); ++ h_path.mnt = au_sbr_mnt(dentry->d_sb, bstart); ++ h_dir = au_pinned_h_dir(&pin); ++ switch (arg->type) { ++ case Creat: ++ err = vfsub_create(h_dir, &h_path, arg->u.c.mode); ++ break; ++ case Symlink: ++ err = vfsub_symlink(h_dir, &h_path, arg->u.s.symname); ++ break; ++ case Mknod: ++ err = vfsub_mknod(h_dir, &h_path, arg->u.m.mode, arg->u.m.dev); ++ break; ++ default: ++ BUG(); ++ } ++ created = !err; ++ if (!err) ++ err = epilog(dir, bstart, wh_dentry, dentry); ++ ++ /* revert */ ++ if (unlikely(created && err && h_path.dentry->d_inode)) { ++ int rerr; ++ rerr = vfsub_unlink(h_dir, &h_path, /*force*/0); ++ if (rerr) { ++ AuIOErr("%.*s revert failure(%d, %d)\n", ++ AuDLNPair(dentry), err, rerr); ++ err = -EIO; ++ } ++ au_dtime_revert(&dt); ++ d_drop(dentry); ++ } ++ ++ au_unpin(&pin); ++ dput(wh_dentry); ++ ++ out: ++ if (unlikely(err)) { ++ au_update_dbstart(dentry); ++ d_drop(dentry); ++ } ++ di_write_unlock(parent); ++ aufs_read_unlock(dentry, AuLock_DW); ++ return err; ++} ++ ++int aufs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) ++{ ++ struct simple_arg arg = { ++ .type = Mknod, ++ .u.m = { ++ .mode = mode, ++ .dev = dev ++ } ++ }; ++ return add_simple(dir, dentry, &arg); ++} ++ ++int aufs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) ++{ ++ struct simple_arg arg = { ++ .type = Symlink, ++ .u.s.symname = symname ++ }; ++ return add_simple(dir, dentry, &arg); ++} ++ ++int aufs_create(struct inode *dir, struct dentry *dentry, int mode, ++ struct nameidata *nd) ++{ ++ struct simple_arg arg = { ++ .type = Creat, ++ .u.c = { ++ .mode = mode, ++ .nd = nd ++ } ++ }; ++ return add_simple(dir, dentry, &arg); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct au_link_args { ++ aufs_bindex_t bdst, bsrc; ++ struct au_pin pin; ++ struct path h_path; ++ struct dentry *src_parent, *parent; ++}; ++ ++static int au_cpup_before_link(struct dentry *src_dentry, ++ struct au_link_args *a) ++{ ++ int err; ++ struct dentry *h_src_dentry; ++ struct mutex *h_mtx; ++ ++ di_read_lock_parent(a->src_parent, AuLock_IR); ++ err = au_test_and_cpup_dirs(src_dentry, a->bdst); ++ if (unlikely(err)) ++ goto out; ++ ++ h_src_dentry = au_h_dptr(src_dentry, a->bsrc); ++ h_mtx = &h_src_dentry->d_inode->i_mutex; ++ err = au_pin(&a->pin, src_dentry, a->bdst, ++ au_opt_udba(src_dentry->d_sb), ++ AuPin_DI_LOCKED | AuPin_MNT_WRITE); ++ if (unlikely(err)) ++ goto out; ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD); ++ err = au_sio_cpup_simple(src_dentry, a->bdst, -1, ++ AuCpup_DTIME /* | AuCpup_KEEPLINO */); ++ mutex_unlock(h_mtx); ++ au_unpin(&a->pin); ++ ++ out: ++ di_read_unlock(a->src_parent, AuLock_IR); ++ return err; ++} ++ ++static int au_cpup_or_link(struct dentry *src_dentry, struct au_link_args *a) ++{ ++ int err; ++ unsigned char plink; ++ struct inode *h_inode, *inode; ++ struct dentry *h_src_dentry; ++ struct super_block *sb; ++ ++ plink = 0; ++ h_inode = NULL; ++ sb = src_dentry->d_sb; ++ inode = src_dentry->d_inode; ++ if (au_ibstart(inode) <= a->bdst) ++ h_inode = au_h_iptr(inode, a->bdst); ++ if (!h_inode || !h_inode->i_nlink) { ++ /* copyup src_dentry as the name of dentry. */ ++ au_set_dbstart(src_dentry, a->bdst); ++ au_set_h_dptr(src_dentry, a->bdst, dget(a->h_path.dentry)); ++ h_inode = au_h_dptr(src_dentry, a->bsrc)->d_inode; ++ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); ++ err = au_sio_cpup_single(src_dentry, a->bdst, a->bsrc, -1, ++ AuCpup_KEEPLINO, a->parent); ++ mutex_unlock(&h_inode->i_mutex); ++ au_set_h_dptr(src_dentry, a->bdst, NULL); ++ au_set_dbstart(src_dentry, a->bsrc); ++ } else { ++ /* the inode of src_dentry already exists on a.bdst branch */ ++ h_src_dentry = d_find_alias(h_inode); ++ if (!h_src_dentry && au_plink_test(inode)) { ++ plink = 1; ++ h_src_dentry = au_plink_lkup(inode, a->bdst); ++ err = PTR_ERR(h_src_dentry); ++ if (IS_ERR(h_src_dentry)) ++ goto out; ++ ++ if (unlikely(!h_src_dentry->d_inode)) { ++ dput(h_src_dentry); ++ h_src_dentry = NULL; ++ } ++ ++ } ++ if (h_src_dentry) { ++ err = vfsub_link(h_src_dentry, au_pinned_h_dir(&a->pin), ++ &a->h_path); ++ dput(h_src_dentry); ++ } else { ++ AuIOErr("no dentry found for hi%lu on b%d\n", ++ h_inode->i_ino, a->bdst); ++ err = -EIO; ++ } ++ } ++ ++ if (!err && !plink) ++ au_plink_append(inode, a->bdst, a->h_path.dentry); ++ ++out: ++ return err; ++} ++ ++int aufs_link(struct dentry *src_dentry, struct inode *dir, ++ struct dentry *dentry) ++{ ++ int err, rerr; ++ struct au_dtime dt; ++ struct au_link_args *a; ++ struct dentry *wh_dentry, *h_src_dentry; ++ struct inode *inode; ++ struct super_block *sb; ++ struct au_wr_dir_args wr_dir_args = { ++ /* .force_btgt = -1, */ ++ .flags = AuWrDir_ADD_ENTRY ++ }; ++ ++ IMustLock(dir); ++ inode = src_dentry->d_inode; ++ IMustLock(inode); ++ ++ err = -ENOENT; ++ if (unlikely(!inode->i_nlink)) ++ goto out; ++ ++ err = -ENOMEM; ++ a = kzalloc(sizeof(*a), GFP_NOFS); ++ if (unlikely(!a)) ++ goto out; ++ ++ a->parent = dentry->d_parent; /* dir inode is locked */ ++ aufs_read_and_write_lock2(dentry, src_dentry, /*AuLock_FLUSH*/0); ++ a->src_parent = dget_parent(src_dentry); ++ wr_dir_args.force_btgt = au_dbstart(src_dentry); ++ ++ di_write_lock_parent(a->parent); ++ wr_dir_args.force_btgt = au_wbr(dentry, wr_dir_args.force_btgt); ++ wh_dentry = lock_hdir_lkup_wh(dentry, &dt, src_dentry, &a->pin, ++ &wr_dir_args); ++ err = PTR_ERR(wh_dentry); ++ if (IS_ERR(wh_dentry)) ++ goto out_unlock; ++ ++ err = 0; ++ sb = dentry->d_sb; ++ a->bdst = au_dbstart(dentry); ++ a->h_path.dentry = au_h_dptr(dentry, a->bdst); ++ a->h_path.mnt = au_sbr_mnt(sb, a->bdst); ++ a->bsrc = au_dbstart(src_dentry); ++ if (au_opt_test(au_mntflags(sb), PLINK)) { ++ if (a->bdst < a->bsrc ++ /* && h_src_dentry->d_sb != a->h_path.dentry->d_sb */) ++ err = au_cpup_or_link(src_dentry, a); ++ else { ++ h_src_dentry = au_h_dptr(src_dentry, a->bdst); ++ err = vfsub_link(h_src_dentry, au_pinned_h_dir(&a->pin), ++ &a->h_path); ++ } ++ } else { ++ /* ++ * copyup src_dentry to the branch we process, ++ * and then link(2) to it. ++ */ ++ if (a->bdst < a->bsrc ++ /* && h_src_dentry->d_sb != a->h_path.dentry->d_sb */) { ++ au_unpin(&a->pin); ++ di_write_unlock(a->parent); ++ err = au_cpup_before_link(src_dentry, a); ++ di_write_lock_parent(a->parent); ++ if (!err) ++ err = au_pin(&a->pin, dentry, a->bdst, ++ au_opt_udba(sb), ++ AuPin_DI_LOCKED | AuPin_MNT_WRITE); ++ if (unlikely(err)) ++ goto out_wh; ++ } ++ if (!err) { ++ h_src_dentry = au_h_dptr(src_dentry, a->bdst); ++ err = -ENOENT; ++ if (h_src_dentry && h_src_dentry->d_inode) ++ err = vfsub_link(h_src_dentry, ++ au_pinned_h_dir(&a->pin), ++ &a->h_path); ++ } ++ } ++ if (unlikely(err)) ++ goto out_unpin; ++ ++ if (wh_dentry) { ++ a->h_path.dentry = wh_dentry; ++ err = au_wh_unlink_dentry(au_pinned_h_dir(&a->pin), &a->h_path, ++ dentry); ++ if (unlikely(err)) ++ goto out_revert; ++ } ++ ++ dir->i_version++; ++ if (au_ibstart(dir) == au_dbstart(dentry)) ++ au_cpup_attr_timesizes(dir); ++ inc_nlink(inode); ++ inode->i_ctime = dir->i_ctime; ++ if (!d_unhashed(a->h_path.dentry)) ++ d_instantiate(dentry, au_igrab(inode)); ++ else ++ /* some filesystem calls d_drop() */ ++ d_drop(dentry); ++ goto out_unpin; /* success */ ++ ++ out_revert: ++ rerr = vfsub_unlink(au_pinned_h_dir(&a->pin), &a->h_path, /*force*/0); ++ if (!rerr) ++ goto out_dt; ++ AuIOErr("%.*s reverting failed(%d, %d)\n", ++ AuDLNPair(dentry), err, rerr); ++ err = -EIO; ++ out_dt: ++ d_drop(dentry); ++ au_dtime_revert(&dt); ++ out_unpin: ++ au_unpin(&a->pin); ++ out_wh: ++ dput(wh_dentry); ++ out_unlock: ++ if (unlikely(err)) { ++ au_update_dbstart(dentry); ++ d_drop(dentry); ++ } ++ di_write_unlock(a->parent); ++ dput(a->src_parent); ++ aufs_read_and_write_unlock2(dentry, src_dentry); ++ kfree(a); ++ out: ++ return err; ++} ++ ++int aufs_mkdir(struct inode *dir, struct dentry *dentry, int mode) ++{ ++ int err, rerr; ++ aufs_bindex_t bindex; ++ unsigned char diropq; ++ struct path h_path; ++ struct dentry *wh_dentry, *parent, *opq_dentry; ++ struct mutex *h_mtx; ++ struct super_block *sb; ++ struct { ++ struct au_pin pin; ++ struct au_dtime dt; ++ } *a; /* reduce the stack usage */ ++ struct au_wr_dir_args wr_dir_args = { ++ .force_btgt = -1, ++ .flags = AuWrDir_ADD_ENTRY | AuWrDir_ISDIR ++ }; ++ ++ IMustLock(dir); ++ ++ err = -ENOMEM; ++ a = kmalloc(sizeof(*a), GFP_NOFS); ++ if (unlikely(!a)) ++ goto out; ++ ++ aufs_read_lock(dentry, AuLock_DW); ++ parent = dentry->d_parent; /* dir inode is locked */ ++ di_write_lock_parent(parent); ++ wh_dentry = lock_hdir_lkup_wh(dentry, &a->dt, /*src_dentry*/NULL, ++ &a->pin, &wr_dir_args); ++ err = PTR_ERR(wh_dentry); ++ if (IS_ERR(wh_dentry)) ++ goto out_free; ++ ++ sb = dentry->d_sb; ++ bindex = au_dbstart(dentry); ++ h_path.dentry = au_h_dptr(dentry, bindex); ++ h_path.mnt = au_sbr_mnt(sb, bindex); ++ err = vfsub_mkdir(au_pinned_h_dir(&a->pin), &h_path, mode); ++ if (unlikely(err)) ++ goto out_unlock; ++ ++ /* make the dir opaque */ ++ diropq = 0; ++ h_mtx = &h_path.dentry->d_inode->i_mutex; ++ if (wh_dentry ++ || au_opt_test(au_mntflags(sb), ALWAYS_DIROPQ)) { ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD); ++ opq_dentry = au_diropq_create(dentry, bindex); ++ mutex_unlock(h_mtx); ++ err = PTR_ERR(opq_dentry); ++ if (IS_ERR(opq_dentry)) ++ goto out_dir; ++ dput(opq_dentry); ++ diropq = 1; ++ } ++ ++ err = epilog(dir, bindex, wh_dentry, dentry); ++ if (!err) { ++ inc_nlink(dir); ++ goto out_unlock; /* success */ ++ } ++ ++ /* revert */ ++ if (diropq) { ++ AuLabel(revert opq); ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD); ++ rerr = au_diropq_remove(dentry, bindex); ++ mutex_unlock(h_mtx); ++ if (rerr) { ++ AuIOErr("%.*s reverting diropq failed(%d, %d)\n", ++ AuDLNPair(dentry), err, rerr); ++ err = -EIO; ++ } ++ } ++ ++ out_dir: ++ AuLabel(revert dir); ++ rerr = vfsub_rmdir(au_pinned_h_dir(&a->pin), &h_path); ++ if (rerr) { ++ AuIOErr("%.*s reverting dir failed(%d, %d)\n", ++ AuDLNPair(dentry), err, rerr); ++ err = -EIO; ++ } ++ d_drop(dentry); ++ au_dtime_revert(&a->dt); ++ out_unlock: ++ au_unpin(&a->pin); ++ dput(wh_dentry); ++ out_free: ++ if (unlikely(err)) { ++ au_update_dbstart(dentry); ++ d_drop(dentry); ++ } ++ di_write_unlock(parent); ++ aufs_read_unlock(dentry, AuLock_DW); ++ kfree(a); ++ out: ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/i_op.c linux-2.6.31/fs/aufs/i_op.c +--- linux-2.6.31-vanilla/fs/aufs/i_op.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/i_op.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,891 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * inode operations (except add/del/rename) ++ */ ++ ++#include ++#include ++#include ++#include ++#include ++#include ++#include "aufs.h" ++ ++static int h_permission(struct inode *h_inode, int mask, ++ struct vfsmount *h_mnt, int brperm) ++{ ++ int err; ++ const unsigned char write_mask = !!(mask & (MAY_WRITE | MAY_APPEND)); ++ ++ err = -EACCES; ++ if ((write_mask && IS_IMMUTABLE(h_inode)) ++ || ((mask & MAY_EXEC) ++ && S_ISREG(h_inode->i_mode) ++ && ((h_mnt->mnt_flags & MNT_NOEXEC) ++ || !(h_inode->i_mode & S_IXUGO)))) ++ goto out; ++ ++ /* ++ * - skip the lower fs test in the case of write to ro branch. ++ * - nfs dir permission write check is optimized, but a policy for ++ * link/rename requires a real check. ++ */ ++ if ((write_mask && !au_br_writable(brperm)) ++ || (au_test_nfs(h_inode->i_sb) && S_ISDIR(h_inode->i_mode) ++ && write_mask && !(mask & MAY_READ)) ++ || !h_inode->i_op->permission) { ++ /* AuLabel(generic_permission); */ ++ err = generic_permission(h_inode, mask, NULL); ++ } else { ++ /* AuLabel(h_inode->permission); */ ++ err = h_inode->i_op->permission(h_inode, mask); ++ AuTraceErr(err); ++ } ++ ++ if (!err) ++ err = devcgroup_inode_permission(h_inode, mask); ++ if (!err) ++ err = security_inode_permission ++ (h_inode, mask & (MAY_READ | MAY_WRITE | MAY_EXEC ++ | MAY_APPEND)); ++ ++#if 0 ++ if (!err) { ++ /* todo: do we need to call ima_path_check()? */ ++ struct path h_path = { ++ .dentry = ++ .mnt = h_mnt ++ }; ++ err = ima_path_check(&h_path, ++ mask & (MAY_READ | MAY_WRITE | MAY_EXEC), ++ IMA_COUNT_LEAVE); ++ } ++#endif ++ ++ out: ++ return err; ++} ++ ++static int aufs_permission(struct inode *inode, int mask) ++{ ++ int err; ++ aufs_bindex_t bindex, bend; ++ const unsigned char isdir = !!S_ISDIR(inode->i_mode); ++ const unsigned char write_mask = !!(mask & (MAY_WRITE | MAY_APPEND)); ++ struct inode *h_inode; ++ struct super_block *sb; ++ struct au_branch *br; ++ ++ sb = inode->i_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ ii_read_lock_child(inode); ++ ++ if (!isdir || write_mask) { ++ err = au_busy_or_stale(); ++ h_inode = au_h_iptr(inode, au_ibstart(inode)); ++ if (unlikely(!h_inode ++ || (h_inode->i_mode & S_IFMT) ++ != (inode->i_mode & S_IFMT))) ++ goto out; ++ ++ err = 0; ++ bindex = au_ibstart(inode); ++ br = au_sbr(sb, bindex); ++ err = h_permission(h_inode, mask, br->br_mnt, br->br_perm); ++ if (write_mask && !err) { ++ /* test whether the upper writable branch exists */ ++ err = -EROFS; ++ for (; bindex >= 0; bindex--) ++ if (!au_br_rdonly(au_sbr(sb, bindex))) { ++ err = 0; ++ break; ++ } ++ } ++ goto out; ++ } ++ ++ /* non-write to dir */ ++ err = 0; ++ bend = au_ibend(inode); ++ for (bindex = au_ibstart(inode); !err && bindex <= bend; bindex++) { ++ h_inode = au_h_iptr(inode, bindex); ++ if (h_inode) { ++ err = au_busy_or_stale(); ++ if (unlikely(!S_ISDIR(h_inode->i_mode))) ++ break; ++ ++ br = au_sbr(sb, bindex); ++ err = h_permission(h_inode, mask, br->br_mnt, ++ br->br_perm); ++ } ++ } ++ ++ out: ++ ii_read_unlock(inode); ++ si_read_unlock(sb); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static struct dentry *aufs_lookup(struct inode *dir, struct dentry *dentry, ++ struct nameidata *nd) ++{ ++ struct dentry *ret, *parent; ++ struct inode *inode, *h_inode; ++ struct mutex *mtx; ++ struct super_block *sb; ++ int err, npositive; ++ aufs_bindex_t bstart; ++ ++ IMustLock(dir); ++ ++ sb = dir->i_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ err = au_alloc_dinfo(dentry); ++ ret = ERR_PTR(err); ++ if (unlikely(err)) ++ goto out; ++ ++ parent = dentry->d_parent; /* dir inode is locked */ ++ di_read_lock_parent(parent, AuLock_IR); ++ npositive = au_lkup_dentry(dentry, au_dbstart(parent), /*type*/0, nd); ++ di_read_unlock(parent, AuLock_IR); ++ err = npositive; ++ ret = ERR_PTR(err); ++ if (unlikely(err < 0)) ++ goto out_unlock; ++ ++ inode = NULL; ++ if (npositive) { ++ bstart = au_dbstart(dentry); ++ h_inode = au_h_dptr(dentry, bstart)->d_inode; ++ if (!S_ISDIR(h_inode->i_mode)) { ++ /* ++ * stop 'race'-ing between hardlinks under different ++ * parents. ++ */ ++ mtx = &au_sbr(sb, bstart)->br_xino.xi_nondir_mtx; ++ mutex_lock(mtx); ++ inode = au_new_inode(dentry, /*must_new*/0); ++ mutex_unlock(mtx); ++ } else ++ inode = au_new_inode(dentry, /*must_new*/0); ++ ret = (void *)inode; ++ } ++ if (IS_ERR(inode)) ++ goto out_unlock; ++ ++ ret = d_splice_alias(inode, dentry); ++ if (unlikely(IS_ERR(ret) && inode)) ++ ii_write_unlock(inode); ++ ++ out_unlock: ++ di_write_unlock(dentry); ++ out: ++ si_read_unlock(sb); ++ return ret; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int au_wr_dir_cpup(struct dentry *dentry, struct dentry *parent, ++ const unsigned char add_entry, aufs_bindex_t bcpup, ++ aufs_bindex_t bstart) ++{ ++ int err; ++ struct dentry *h_parent; ++ struct inode *h_dir; ++ ++ if (add_entry) { ++ au_update_dbstart(dentry); ++ IMustLock(parent->d_inode); ++ } else ++ di_write_lock_parent(parent); ++ ++ err = 0; ++ if (!au_h_dptr(parent, bcpup)) { ++ if (bstart < bcpup) ++ err = au_cpdown_dirs(dentry, bcpup); ++ else ++ err = au_cpup_dirs(dentry, bcpup); ++ } ++ if (!err && add_entry) { ++ h_parent = au_h_dptr(parent, bcpup); ++ h_dir = h_parent->d_inode; ++ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT); ++ err = au_lkup_neg(dentry, bcpup); ++ /* todo: no unlock here */ ++ mutex_unlock(&h_dir->i_mutex); ++ if (bstart < bcpup && au_dbstart(dentry) < 0) { ++ au_set_dbstart(dentry, 0); ++ au_update_dbrange(dentry, /*do_put_zero*/0); ++ } ++ } ++ ++ if (!add_entry) ++ di_write_unlock(parent); ++ if (!err) ++ err = bcpup; /* success */ ++ ++ return err; ++} ++ ++/* ++ * decide the branch and the parent dir where we will create a new entry. ++ * returns new bindex or an error. ++ * copyup the parent dir if needed. ++ */ ++int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry, ++ struct au_wr_dir_args *args) ++{ ++ int err; ++ aufs_bindex_t bcpup, bstart, src_bstart; ++ const unsigned char add_entry = !!au_ftest_wrdir(args->flags, ++ ADD_ENTRY); ++ struct super_block *sb; ++ struct dentry *parent; ++ struct au_sbinfo *sbinfo; ++ ++ sb = dentry->d_sb; ++ sbinfo = au_sbi(sb); ++ parent = dget_parent(dentry); ++ bstart = au_dbstart(dentry); ++ bcpup = bstart; ++ if (args->force_btgt < 0) { ++ if (src_dentry) { ++ src_bstart = au_dbstart(src_dentry); ++ if (src_bstart < bstart) ++ bcpup = src_bstart; ++ } else if (add_entry) { ++ err = AuWbrCreate(sbinfo, dentry, ++ au_ftest_wrdir(args->flags, ISDIR)); ++ bcpup = err; ++ } ++ ++ if (bcpup < 0 || au_test_ro(sb, bcpup, dentry->d_inode)) { ++ if (add_entry) ++ err = AuWbrCopyup(sbinfo, dentry); ++ else { ++ if (!IS_ROOT(dentry)) { ++ di_read_lock_parent(parent, !AuLock_IR); ++ err = AuWbrCopyup(sbinfo, dentry); ++ di_read_unlock(parent, !AuLock_IR); ++ } else ++ err = AuWbrCopyup(sbinfo, dentry); ++ } ++ bcpup = err; ++ if (unlikely(err < 0)) ++ goto out; ++ } ++ } else { ++ bcpup = args->force_btgt; ++ AuDebugOn(au_test_ro(sb, bcpup, dentry->d_inode)); ++ } ++ AuDbg("bstart %d, bcpup %d\n", bstart, bcpup); ++ if (bstart < bcpup) ++ au_update_dbrange(dentry, /*do_put_zero*/1); ++ ++ err = bcpup; ++ if (bcpup == bstart) ++ goto out; /* success */ ++ ++ /* copyup the new parent into the branch we process */ ++ err = au_wr_dir_cpup(dentry, parent, add_entry, bcpup, bstart); ++ ++ out: ++ dput(parent); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct dentry *au_pinned_h_parent(struct au_pin *pin) ++{ ++ if (pin && pin->parent) ++ return au_h_dptr(pin->parent, pin->bindex); ++ return NULL; ++} ++ ++void au_unpin(struct au_pin *p) ++{ ++ if (au_ftest_pin(p->flags, MNT_WRITE)) ++ mnt_drop_write(p->h_mnt); ++ if (!p->hdir) ++ return; ++ ++ au_hin_imtx_unlock(p->hdir); ++ if (!au_ftest_pin(p->flags, DI_LOCKED)) ++ di_read_unlock(p->parent, AuLock_IR); ++ iput(p->hdir->hi_inode); ++ dput(p->parent); ++ p->parent = NULL; ++ p->hdir = NULL; ++ p->h_mnt = NULL; ++} ++ ++int au_do_pin(struct au_pin *p) ++{ ++ int err; ++ struct super_block *sb; ++ struct dentry *h_dentry, *h_parent; ++ struct au_branch *br; ++ struct inode *h_dir; ++ ++ err = 0; ++ sb = p->dentry->d_sb; ++ br = au_sbr(sb, p->bindex); ++ if (IS_ROOT(p->dentry)) { ++ if (au_ftest_pin(p->flags, MNT_WRITE)) { ++ p->h_mnt = br->br_mnt; ++ err = mnt_want_write(p->h_mnt); ++ if (unlikely(err)) { ++ au_fclr_pin(p->flags, MNT_WRITE); ++ goto out_err; ++ } ++ } ++ goto out; ++ } ++ ++ h_dentry = NULL; ++ if (p->bindex <= au_dbend(p->dentry)) ++ h_dentry = au_h_dptr(p->dentry, p->bindex); ++ ++ p->parent = dget_parent(p->dentry); ++ if (!au_ftest_pin(p->flags, DI_LOCKED)) ++ di_read_lock(p->parent, AuLock_IR, p->lsc_di); ++ ++ h_dir = NULL; ++ h_parent = au_h_dptr(p->parent, p->bindex); ++ p->hdir = au_hi(p->parent->d_inode, p->bindex); ++ if (p->hdir) ++ h_dir = p->hdir->hi_inode; ++ ++ /* udba case */ ++ if (unlikely(!p->hdir || !h_dir)) { ++ if (!au_ftest_pin(p->flags, DI_LOCKED)) ++ di_read_unlock(p->parent, AuLock_IR); ++ dput(p->parent); ++ p->parent = NULL; ++ goto out_err; ++ } ++ ++ au_igrab(h_dir); ++ au_hin_imtx_lock_nested(p->hdir, p->lsc_hi); ++ ++ if (unlikely(p->hdir->hi_inode != h_parent->d_inode)) { ++ err = -EBUSY; ++ goto out_unpin; ++ } ++ if (h_dentry) { ++ err = au_h_verify(h_dentry, p->udba, h_dir, h_parent, br); ++ if (unlikely(err)) { ++ au_fclr_pin(p->flags, MNT_WRITE); ++ goto out_unpin; ++ } ++ } ++ ++ if (au_ftest_pin(p->flags, MNT_WRITE)) { ++ p->h_mnt = br->br_mnt; ++ err = mnt_want_write(p->h_mnt); ++ if (unlikely(err)) { ++ au_fclr_pin(p->flags, MNT_WRITE); ++ goto out_unpin; ++ } ++ } ++ goto out; /* success */ ++ ++ out_unpin: ++ au_unpin(p); ++ out_err: ++ AuErr("err %d\n", err); ++ err = au_busy_or_stale(); ++ out: ++ return err; ++} ++ ++void au_pin_init(struct au_pin *p, struct dentry *dentry, ++ aufs_bindex_t bindex, int lsc_di, int lsc_hi, ++ unsigned int udba, unsigned char flags) ++{ ++ p->dentry = dentry; ++ p->udba = udba; ++ p->lsc_di = lsc_di; ++ p->lsc_hi = lsc_hi; ++ p->flags = flags; ++ p->bindex = bindex; ++ ++ p->parent = NULL; ++ p->hdir = NULL; ++ p->h_mnt = NULL; ++} ++ ++int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex, ++ unsigned int udba, unsigned char flags) ++{ ++ au_pin_init(pin, dentry, bindex, AuLsc_DI_PARENT, AuLsc_I_PARENT2, ++ udba, flags); ++ return au_do_pin(pin); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++#define AuIcpup_DID_CPUP 1 ++#define au_ftest_icpup(flags, name) ((flags) & AuIcpup_##name) ++#define au_fset_icpup(flags, name) { (flags) |= AuIcpup_##name; } ++#define au_fclr_icpup(flags, name) { (flags) &= ~AuIcpup_##name; } ++ ++struct au_icpup_args { ++ unsigned char flags; ++ unsigned char pin_flags; ++ aufs_bindex_t btgt; ++ struct au_pin pin; ++ struct path h_path; ++ struct inode *h_inode; ++}; ++ ++static int au_lock_and_icpup(struct dentry *dentry, struct iattr *ia, ++ struct au_icpup_args *a) ++{ ++ int err; ++ unsigned int udba; ++ loff_t sz; ++ aufs_bindex_t bstart; ++ struct dentry *hi_wh, *parent; ++ struct inode *inode; ++ struct au_wr_dir_args wr_dir_args = { ++ .force_btgt = -1, ++ .flags = 0 ++ }; ++ ++ di_write_lock_child(dentry); ++ bstart = au_dbstart(dentry); ++ inode = dentry->d_inode; ++ if (S_ISDIR(inode->i_mode)) ++ au_fset_wrdir(wr_dir_args.flags, ISDIR); ++ /* plink or hi_wh() case */ ++ if (bstart != au_ibstart(inode)) ++ wr_dir_args.force_btgt = au_ibstart(inode); ++ err = au_wr_dir(dentry, /*src_dentry*/NULL, &wr_dir_args); ++ if (unlikely(err < 0)) ++ goto out_dentry; ++ a->btgt = err; ++ if (err != bstart) ++ au_fset_icpup(a->flags, DID_CPUP); ++ ++ err = 0; ++ a->pin_flags = AuPin_MNT_WRITE; ++ parent = NULL; ++ if (!IS_ROOT(dentry)) { ++ au_fset_pin(a->pin_flags, DI_LOCKED); ++ parent = dget_parent(dentry); ++ di_write_lock_parent(parent); ++ } ++ ++ udba = au_opt_udba(dentry->d_sb); ++ if (d_unhashed(dentry) || (ia->ia_valid & ATTR_FILE)) ++ udba = AuOpt_UDBA_NONE; ++ err = au_pin(&a->pin, dentry, a->btgt, udba, a->pin_flags); ++ if (unlikely(err)) { ++ if (parent) { ++ di_write_unlock(parent); ++ dput(parent); ++ } ++ goto out_dentry; ++ } ++ a->h_path.dentry = au_h_dptr(dentry, bstart); ++ a->h_inode = a->h_path.dentry->d_inode; ++ mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD); ++ sz = -1; ++ if ((ia->ia_valid & ATTR_SIZE) && ia->ia_size < i_size_read(a->h_inode)) ++ sz = ia->ia_size; ++ ++ hi_wh = NULL; ++ if (au_ftest_icpup(a->flags, DID_CPUP) && d_unhashed(dentry)) { ++ hi_wh = au_hi_wh(inode, a->btgt); ++ if (!hi_wh) { ++ err = au_sio_cpup_wh(dentry, a->btgt, sz, /*file*/NULL); ++ if (unlikely(err)) ++ goto out_unlock; ++ hi_wh = au_hi_wh(inode, a->btgt); ++ /* todo: revalidate hi_wh? */ ++ } ++ } ++ ++ if (parent) { ++ au_pin_set_parent_lflag(&a->pin, /*lflag*/0); ++ di_downgrade_lock(parent, AuLock_IR); ++ dput(parent); ++ } ++ if (!au_ftest_icpup(a->flags, DID_CPUP)) ++ goto out; /* success */ ++ ++ if (!d_unhashed(dentry)) { ++ err = au_sio_cpup_simple(dentry, a->btgt, sz, AuCpup_DTIME); ++ if (!err) ++ a->h_path.dentry = au_h_dptr(dentry, a->btgt); ++ } else if (!hi_wh) ++ a->h_path.dentry = au_h_dptr(dentry, a->btgt); ++ else ++ a->h_path.dentry = hi_wh; /* do not dget here */ ++ ++ out_unlock: ++ mutex_unlock(&a->h_inode->i_mutex); ++ a->h_inode = a->h_path.dentry->d_inode; ++ if (!err) { ++ mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD); ++ goto out; /* success */ ++ } ++ ++ au_unpin(&a->pin); ++ ++ out_dentry: ++ di_write_unlock(dentry); ++ out: ++ return err; ++} ++ ++static int aufs_setattr(struct dentry *dentry, struct iattr *ia) ++{ ++ int err; ++ struct inode *inode; ++ struct super_block *sb; ++ struct file *file; ++ struct au_icpup_args *a; ++ ++ err = -ENOMEM; ++ a = kzalloc(sizeof(*a), GFP_NOFS); ++ if (unlikely(!a)) ++ goto out; ++ ++ inode = dentry->d_inode; ++ IMustLock(inode); ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ ++ file = NULL; ++ if (ia->ia_valid & ATTR_FILE) { ++ /* currently ftruncate(2) only */ ++ file = ia->ia_file; ++ fi_write_lock(file); ++ ia->ia_file = au_h_fptr(file, au_fbstart(file)); ++ } ++ ++ if (ia->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID)) ++ ia->ia_valid &= ~ATTR_MODE; ++ ++ err = au_lock_and_icpup(dentry, ia, a); ++ if (unlikely(err < 0)) ++ goto out_si; ++ if (au_ftest_icpup(a->flags, DID_CPUP)) { ++ ia->ia_file = NULL; ++ ia->ia_valid &= ~ATTR_FILE; ++ } ++ ++ a->h_path.mnt = au_sbr_mnt(sb, a->btgt); ++ if (ia->ia_valid & ATTR_SIZE) { ++ struct file *f; ++ ++ if (ia->ia_size < i_size_read(inode)) { ++ /* unmap only */ ++ err = vmtruncate(inode, ia->ia_size); ++ if (unlikely(err)) ++ goto out_unlock; ++ } ++ ++ f = NULL; ++ if (ia->ia_valid & ATTR_FILE) ++ f = ia->ia_file; ++ mutex_unlock(&a->h_inode->i_mutex); ++ err = vfsub_trunc(&a->h_path, ia->ia_size, ia->ia_valid, f); ++ mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD); ++ } else ++ err = vfsub_notify_change(&a->h_path, ia); ++ if (!err) ++ au_cpup_attr_changeable(inode); ++ ++ out_unlock: ++ mutex_unlock(&a->h_inode->i_mutex); ++ au_unpin(&a->pin); ++ di_write_unlock(dentry); ++ out_si: ++ if (file) { ++ fi_write_unlock(file); ++ ia->ia_file = file; ++ ia->ia_valid |= ATTR_FILE; ++ } ++ si_read_unlock(sb); ++ kfree(a); ++ out: ++ return err; ++} ++ ++static int au_getattr_lock_reval(struct dentry *dentry, unsigned int sigen) ++{ ++ int err; ++ struct inode *inode; ++ struct dentry *parent; ++ ++ err = 0; ++ inode = dentry->d_inode; ++ di_write_lock_child(dentry); ++ if (au_digen(dentry) != sigen || au_iigen(inode) != sigen) { ++ parent = dget_parent(dentry); ++ di_read_lock_parent(parent, AuLock_IR); ++ /* returns a number of positive dentries */ ++ err = au_refresh_hdentry(dentry, inode->i_mode & S_IFMT); ++ if (err > 0) ++ err = au_refresh_hinode(inode, dentry); ++ di_read_unlock(parent, AuLock_IR); ++ dput(parent); ++ if (unlikely(!err)) ++ err = -EIO; ++ } ++ di_downgrade_lock(dentry, AuLock_IR); ++ if (unlikely(err)) ++ di_read_unlock(dentry, AuLock_IR); ++ ++ return err; ++} ++ ++static void au_refresh_iattr(struct inode *inode, struct kstat *st, ++ unsigned int nlink) ++{ ++ inode->i_mode = st->mode; ++ inode->i_uid = st->uid; ++ inode->i_gid = st->gid; ++ inode->i_atime = st->atime; ++ inode->i_mtime = st->mtime; ++ inode->i_ctime = st->ctime; ++ ++ au_cpup_attr_nlink(inode, /*force*/0); ++ if (S_ISDIR(inode->i_mode)) { ++ inode->i_nlink -= nlink; ++ inode->i_nlink += st->nlink; ++ } ++ ++ spin_lock(&inode->i_lock); ++ inode->i_blocks = st->blocks; ++ i_size_write(inode, st->size); ++ spin_unlock(&inode->i_lock); ++} ++ ++static int aufs_getattr(struct vfsmount *mnt __maybe_unused, ++ struct dentry *dentry, struct kstat *st) ++{ ++ int err; ++ unsigned int mnt_flags; ++ aufs_bindex_t bindex; ++ unsigned char udba_none, positive; ++ struct super_block *sb, *h_sb; ++ struct inode *inode; ++ struct vfsmount *h_mnt; ++ struct dentry *h_dentry; ++ ++ err = 0; ++ sb = dentry->d_sb; ++ inode = dentry->d_inode; ++ si_read_lock(sb, AuLock_FLUSH); ++ mnt_flags = au_mntflags(sb); ++ udba_none = !!au_opt_test(mnt_flags, UDBA_NONE); ++ ++ /* support fstat(2) */ ++ if (!d_unhashed(dentry) && !udba_none) { ++ unsigned int sigen = au_sigen(sb); ++ if (au_digen(dentry) == sigen && au_iigen(inode) == sigen) ++ di_read_lock_child(dentry, AuLock_IR); ++ else { ++ /* NFSD may skip the revalidation */ ++ if (!au_test_nfsd(current)) ++ AuDebugOn(!IS_ROOT(dentry)); ++ else { ++ err = au_busy_or_stale(); ++ if (unlikely(!IS_ROOT(dentry))) ++ goto out; ++ } ++ err = au_getattr_lock_reval(dentry, sigen); ++ if (unlikely(err)) ++ goto out; ++ } ++ } else ++ di_read_lock_child(dentry, AuLock_IR); ++ ++ bindex = au_ibstart(inode); ++ h_mnt = au_sbr_mnt(sb, bindex); ++ h_sb = h_mnt->mnt_sb; ++ if (!au_test_fs_bad_iattr(h_sb) && udba_none) ++ goto out_fill; /* success */ ++ ++ h_dentry = NULL; ++ if (au_dbstart(dentry) == bindex) ++ h_dentry = dget(au_h_dptr(dentry, bindex)); ++ else if (au_opt_test(mnt_flags, PLINK) && au_plink_test(inode)) { ++ h_dentry = au_plink_lkup(inode, bindex); ++ if (IS_ERR(h_dentry)) ++ goto out_fill; /* pretending success */ ++ } ++ /* illegally overlapped or something */ ++ if (unlikely(!h_dentry)) ++ goto out_fill; /* pretending success */ ++ ++ positive = !!h_dentry->d_inode; ++ if (positive) ++ err = vfs_getattr(h_mnt, h_dentry, st); ++ dput(h_dentry); ++ if (!err) { ++ if (positive) ++ au_refresh_iattr(inode, st, h_dentry->d_inode->i_nlink); ++ goto out_fill; /* success */ ++ } ++ goto out_unlock; ++ ++ out_fill: ++ generic_fillattr(inode, st); ++ out_unlock: ++ di_read_unlock(dentry, AuLock_IR); ++ out: ++ si_read_unlock(sb); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int h_readlink(struct dentry *dentry, int bindex, char __user *buf, ++ int bufsiz) ++{ ++ int err; ++ struct super_block *sb; ++ struct dentry *h_dentry; ++ ++ err = -EINVAL; ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (unlikely(/* !h_dentry ++ || !h_dentry->d_inode ++ || !h_dentry->d_inode->i_op ++ || */ !h_dentry->d_inode->i_op->readlink)) ++ goto out; ++ ++ err = security_inode_readlink(h_dentry); ++ if (unlikely(err)) ++ goto out; ++ ++ sb = dentry->d_sb; ++ if (!au_test_ro(sb, bindex, dentry->d_inode)) { ++ vfsub_touch_atime(au_sbr_mnt(sb, bindex), h_dentry); ++ fsstack_copy_attr_atime(dentry->d_inode, h_dentry->d_inode); ++ } ++ err = h_dentry->d_inode->i_op->readlink(h_dentry, buf, bufsiz); ++ ++ out: ++ return err; ++} ++ ++static int aufs_readlink(struct dentry *dentry, char __user *buf, int bufsiz) ++{ ++ int err; ++ ++ aufs_read_lock(dentry, AuLock_IR); ++ err = h_readlink(dentry, au_dbstart(dentry), buf, bufsiz); ++ aufs_read_unlock(dentry, AuLock_IR); ++ ++ return err; ++} ++ ++static void *aufs_follow_link(struct dentry *dentry, struct nameidata *nd) ++{ ++ int err; ++ char *buf; ++ mm_segment_t old_fs; ++ ++ err = -ENOMEM; ++ buf = __getname(); ++ if (unlikely(!buf)) ++ goto out; ++ ++ aufs_read_lock(dentry, AuLock_IR); ++ old_fs = get_fs(); ++ set_fs(KERNEL_DS); ++ err = h_readlink(dentry, au_dbstart(dentry), (char __user *)buf, ++ PATH_MAX); ++ set_fs(old_fs); ++ aufs_read_unlock(dentry, AuLock_IR); ++ ++ if (err >= 0) { ++ buf[err] = 0; ++ /* will be freed by put_link */ ++ nd_set_link(nd, buf); ++ return NULL; /* success */ ++ } ++ __putname(buf); ++ ++ out: ++ path_put(&nd->path); ++ AuTraceErr(err); ++ return ERR_PTR(err); ++} ++ ++static void aufs_put_link(struct dentry *dentry __maybe_unused, ++ struct nameidata *nd, void *cookie __maybe_unused) ++{ ++ __putname(nd_get_link(nd)); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static void aufs_truncate_range(struct inode *inode __maybe_unused, ++ loff_t start __maybe_unused, ++ loff_t end __maybe_unused) ++{ ++ AuUnsupport(); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct inode_operations aufs_symlink_iop = { ++ .permission = aufs_permission, ++ .setattr = aufs_setattr, ++ .getattr = aufs_getattr, ++ .readlink = aufs_readlink, ++ .follow_link = aufs_follow_link, ++ .put_link = aufs_put_link ++}; ++ ++struct inode_operations aufs_dir_iop = { ++ .create = aufs_create, ++ .lookup = aufs_lookup, ++ .link = aufs_link, ++ .unlink = aufs_unlink, ++ .symlink = aufs_symlink, ++ .mkdir = aufs_mkdir, ++ .rmdir = aufs_rmdir, ++ .mknod = aufs_mknod, ++ .rename = aufs_rename, ++ ++ .permission = aufs_permission, ++ .setattr = aufs_setattr, ++ .getattr = aufs_getattr ++}; ++ ++struct inode_operations aufs_iop = { ++ .permission = aufs_permission, ++ .setattr = aufs_setattr, ++ .getattr = aufs_getattr, ++ .truncate_range = aufs_truncate_range ++}; +diff -Nur linux-2.6.31-vanilla/fs/aufs/i_op_del.c linux-2.6.31/fs/aufs/i_op_del.c +--- linux-2.6.31-vanilla/fs/aufs/i_op_del.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/i_op_del.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,468 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * inode operations (del entry) ++ */ ++ ++#include "aufs.h" ++ ++/* ++ * decide if a new whiteout for @dentry is necessary or not. ++ * when it is necessary, prepare the parent dir for the upper branch whose ++ * branch index is @bcpup for creation. the actual creation of the whiteout will ++ * be done by caller. ++ * return value: ++ * 0: wh is unnecessary ++ * plus: wh is necessary ++ * minus: error ++ */ ++int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup) ++{ ++ int need_wh, err; ++ aufs_bindex_t bstart; ++ struct super_block *sb; ++ ++ sb = dentry->d_sb; ++ bstart = au_dbstart(dentry); ++ if (*bcpup < 0) { ++ *bcpup = bstart; ++ if (au_test_ro(sb, bstart, dentry->d_inode)) { ++ err = AuWbrCopyup(au_sbi(sb), dentry); ++ *bcpup = err; ++ if (unlikely(err < 0)) ++ goto out; ++ } ++ } else ++ AuDebugOn(bstart < *bcpup ++ || au_test_ro(sb, *bcpup, dentry->d_inode)); ++ AuDbg("bcpup %d, bstart %d\n", *bcpup, bstart); ++ ++ if (*bcpup != bstart) { ++ err = au_cpup_dirs(dentry, *bcpup); ++ if (unlikely(err)) ++ goto out; ++ need_wh = 1; ++ } else { ++ aufs_bindex_t old_bend, new_bend, bdiropq = -1; ++ ++ old_bend = au_dbend(dentry); ++ if (isdir) { ++ bdiropq = au_dbdiropq(dentry); ++ au_set_dbdiropq(dentry, -1); ++ } ++ need_wh = au_lkup_dentry(dentry, bstart + 1, /*type*/0, ++ /*nd*/NULL); ++ err = need_wh; ++ if (isdir) ++ au_set_dbdiropq(dentry, bdiropq); ++ if (unlikely(err < 0)) ++ goto out; ++ new_bend = au_dbend(dentry); ++ if (!need_wh && old_bend != new_bend) { ++ au_set_h_dptr(dentry, new_bend, NULL); ++ au_set_dbend(dentry, old_bend); ++ } ++ } ++ AuDbg("need_wh %d\n", need_wh); ++ err = need_wh; ++ ++ out: ++ return err; ++} ++ ++/* ++ * simple tests for the del-entry operations. ++ * following the checks in vfs, plus the parent-child relationship. ++ */ ++int au_may_del(struct dentry *dentry, aufs_bindex_t bindex, ++ struct dentry *h_parent, int isdir) ++{ ++ int err; ++ umode_t h_mode; ++ struct dentry *h_dentry, *h_latest; ++ struct inode *h_inode; ++ ++ h_dentry = au_h_dptr(dentry, bindex); ++ h_inode = h_dentry->d_inode; ++ if (dentry->d_inode) { ++ err = -ENOENT; ++ if (unlikely(!h_inode || !h_inode->i_nlink)) ++ goto out; ++ ++ h_mode = h_inode->i_mode; ++ if (!isdir) { ++ err = -EISDIR; ++ if (unlikely(S_ISDIR(h_mode))) ++ goto out; ++ } else if (unlikely(!S_ISDIR(h_mode))) { ++ err = -ENOTDIR; ++ goto out; ++ } ++ } else { ++ /* rename(2) case */ ++ err = -EIO; ++ if (unlikely(h_inode)) ++ goto out; ++ } ++ ++ err = -ENOENT; ++ /* expected parent dir is locked */ ++ if (unlikely(h_parent != h_dentry->d_parent)) ++ goto out; ++ err = 0; ++ ++ /* ++ * rmdir a dir may break the consistency on some filesystem. ++ * let's try heavy test. ++ */ ++ err = -EACCES; ++ if (unlikely(au_test_h_perm(h_parent->d_inode, MAY_EXEC | MAY_WRITE))) ++ goto out; ++ ++ h_latest = au_sio_lkup_one(&dentry->d_name, h_parent, ++ au_sbr(dentry->d_sb, bindex)); ++ err = -EIO; ++ if (IS_ERR(h_latest)) ++ goto out; ++ if (h_latest == h_dentry) ++ err = 0; ++ dput(h_latest); ++ ++ out: ++ return err; ++} ++ ++/* ++ * decide the branch where we operate for @dentry. the branch index will be set ++ * @rbcpup. after diciding it, 'pin' it and store the timestamps of the parent ++ * dir for reverting. ++ * when a new whiteout is necessary, create it. ++ */ ++static struct dentry* ++lock_hdir_create_wh(struct dentry *dentry, int isdir, aufs_bindex_t *rbcpup, ++ struct au_dtime *dt, struct au_pin *pin) ++{ ++ struct dentry *wh_dentry; ++ struct super_block *sb; ++ struct path h_path; ++ int err, need_wh; ++ unsigned int udba; ++ aufs_bindex_t bcpup; ++ ++ need_wh = au_wr_dir_need_wh(dentry, isdir, rbcpup); ++ wh_dentry = ERR_PTR(need_wh); ++ if (unlikely(need_wh < 0)) ++ goto out; ++ ++ sb = dentry->d_sb; ++ udba = au_opt_udba(sb); ++ bcpup = *rbcpup; ++ err = au_pin(pin, dentry, bcpup, udba, ++ AuPin_DI_LOCKED | AuPin_MNT_WRITE); ++ wh_dentry = ERR_PTR(err); ++ if (unlikely(err)) ++ goto out; ++ ++ h_path.dentry = au_pinned_h_parent(pin); ++ if (udba != AuOpt_UDBA_NONE ++ && au_dbstart(dentry) == bcpup) { ++ err = au_may_del(dentry, bcpup, h_path.dentry, isdir); ++ wh_dentry = ERR_PTR(err); ++ if (unlikely(err)) ++ goto out_unpin; ++ } ++ ++ h_path.mnt = au_sbr_mnt(sb, bcpup); ++ au_dtime_store(dt, au_pinned_parent(pin), &h_path); ++ wh_dentry = NULL; ++ if (!need_wh) ++ goto out; /* success, no need to create whiteout */ ++ ++ wh_dentry = au_wh_create(dentry, bcpup, h_path.dentry); ++ if (!IS_ERR(wh_dentry)) ++ goto out; /* success */ ++ /* returns with the parent is locked and wh_dentry is dget-ed */ ++ ++ out_unpin: ++ au_unpin(pin); ++ out: ++ return wh_dentry; ++} ++ ++/* ++ * when removing a dir, rename it to a unique temporary whiteout-ed name first ++ * in order to be revertible and save time for removing many child whiteouts ++ * under the dir. ++ * returns 1 when there are too many child whiteout and caller should remove ++ * them asynchronously. returns 0 when the number of children is enough small to ++ * remove now or the branch fs is a remote fs. ++ * otherwise return an error. ++ */ ++static int renwh_and_rmdir(struct dentry *dentry, aufs_bindex_t bindex, ++ struct au_nhash *whlist, struct inode *dir) ++{ ++ int rmdir_later, err, dirwh; ++ struct dentry *h_dentry; ++ struct super_block *sb; ++ ++ sb = dentry->d_sb; ++ SiMustAnyLock(sb); ++ h_dentry = au_h_dptr(dentry, bindex); ++ err = au_whtmp_ren(h_dentry, au_sbr(sb, bindex)); ++ if (unlikely(err)) ++ goto out; ++ ++ /* stop monitoring */ ++ au_hin_free(au_hi(dentry->d_inode, bindex)); ++ ++ if (!au_test_fs_remote(h_dentry->d_sb)) { ++ dirwh = au_sbi(sb)->si_dirwh; ++ rmdir_later = (dirwh <= 1); ++ if (!rmdir_later) ++ rmdir_later = au_nhash_test_longer_wh(whlist, bindex, ++ dirwh); ++ if (rmdir_later) ++ return rmdir_later; ++ } ++ ++ err = au_whtmp_rmdir(dir, bindex, h_dentry, whlist); ++ if (unlikely(err)) { ++ AuIOErr("rmdir %.*s, b%d failed, %d. ignored\n", ++ AuDLNPair(h_dentry), bindex, err); ++ err = 0; ++ } ++ ++ out: ++ return err; ++} ++ ++/* ++ * final procedure for deleting a entry. ++ * maintain dentry and iattr. ++ */ ++static void epilog(struct inode *dir, struct dentry *dentry, ++ aufs_bindex_t bindex) ++{ ++ struct inode *inode; ++ ++ inode = dentry->d_inode; ++ d_drop(dentry); ++ inode->i_ctime = dir->i_ctime; ++ ++ if (atomic_read(&dentry->d_count) == 1) { ++ au_set_h_dptr(dentry, au_dbstart(dentry), NULL); ++ au_update_dbstart(dentry); ++ } ++ if (au_ibstart(dir) == bindex) ++ au_cpup_attr_timesizes(dir); ++ dir->i_version++; ++} ++ ++/* ++ * when an error happened, remove the created whiteout and revert everything. ++ */ ++static int do_revert(int err, struct inode *dir, aufs_bindex_t bwh, ++ struct dentry *wh_dentry, struct dentry *dentry, ++ struct au_dtime *dt) ++{ ++ int rerr; ++ struct path h_path = { ++ .dentry = wh_dentry, ++ .mnt = au_sbr_mnt(dir->i_sb, bwh) ++ }; ++ ++ rerr = au_wh_unlink_dentry(au_h_iptr(dir, bwh), &h_path, dentry); ++ if (!rerr) { ++ au_set_dbwh(dentry, bwh); ++ au_dtime_revert(dt); ++ return 0; ++ } ++ ++ AuIOErr("%.*s reverting whiteout failed(%d, %d)\n", ++ AuDLNPair(dentry), err, rerr); ++ return -EIO; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int aufs_unlink(struct inode *dir, struct dentry *dentry) ++{ ++ int err; ++ aufs_bindex_t bwh, bindex, bstart; ++ struct au_dtime dt; ++ struct au_pin pin; ++ struct path h_path; ++ struct inode *inode, *h_dir; ++ struct dentry *parent, *wh_dentry; ++ ++ IMustLock(dir); ++ inode = dentry->d_inode; ++ if (unlikely(!inode)) ++ return -ENOENT; /* possible? */ ++ IMustLock(inode); ++ ++ aufs_read_lock(dentry, AuLock_DW); ++ parent = dentry->d_parent; /* dir inode is locked */ ++ di_write_lock_parent(parent); ++ ++ bstart = au_dbstart(dentry); ++ bwh = au_dbwh(dentry); ++ bindex = -1; ++ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/0, &bindex, &dt, &pin); ++ err = PTR_ERR(wh_dentry); ++ if (IS_ERR(wh_dentry)) ++ goto out; ++ ++ h_path.mnt = au_sbr_mnt(dentry->d_sb, bstart); ++ h_path.dentry = au_h_dptr(dentry, bstart); ++ dget(h_path.dentry); ++ if (bindex == bstart) { ++ h_dir = au_pinned_h_dir(&pin); ++ err = vfsub_unlink(h_dir, &h_path, /*force*/0); ++ } else { ++ /* dir inode is locked */ ++ h_dir = wh_dentry->d_parent->d_inode; ++ IMustLock(h_dir); ++ err = 0; ++ } ++ ++ if (!err) { ++ drop_nlink(inode); ++ epilog(dir, dentry, bindex); ++ ++ /* update target timestamps */ ++ if (bindex == bstart) { ++ vfsub_update_h_iattr(&h_path, /*did*/NULL); /*ignore*/ ++ inode->i_ctime = h_path.dentry->d_inode->i_ctime; ++ } else ++ /* todo: this timestamp may be reverted later */ ++ inode->i_ctime = h_dir->i_ctime; ++ goto out_unlock; /* success */ ++ } ++ ++ /* revert */ ++ if (wh_dentry) { ++ int rerr; ++ ++ rerr = do_revert(err, dir, bwh, wh_dentry, dentry, &dt); ++ if (rerr) ++ err = rerr; ++ } ++ ++ out_unlock: ++ au_unpin(&pin); ++ dput(wh_dentry); ++ dput(h_path.dentry); ++ out: ++ di_write_unlock(parent); ++ aufs_read_unlock(dentry, AuLock_DW); ++ return err; ++} ++ ++int aufs_rmdir(struct inode *dir, struct dentry *dentry) ++{ ++ int err, rmdir_later; ++ aufs_bindex_t bwh, bindex, bstart; ++ struct au_dtime dt; ++ struct au_pin pin; ++ struct inode *inode; ++ struct dentry *parent, *wh_dentry, *h_dentry; ++ struct au_whtmp_rmdir *args; ++ ++ IMustLock(dir); ++ inode = dentry->d_inode; ++ err = -ENOENT; /* possible? */ ++ if (unlikely(!inode)) ++ goto out; ++ IMustLock(inode); ++ ++ aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH); ++ err = -ENOMEM; ++ args = au_whtmp_rmdir_alloc(dir->i_sb, GFP_NOFS); ++ if (unlikely(!args)) ++ goto out_unlock; ++ ++ parent = dentry->d_parent; /* dir inode is locked */ ++ di_write_lock_parent(parent); ++ err = au_test_empty(dentry, &args->whlist); ++ if (unlikely(err)) ++ goto out_args; ++ ++ bstart = au_dbstart(dentry); ++ bwh = au_dbwh(dentry); ++ bindex = -1; ++ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/1, &bindex, &dt, &pin); ++ err = PTR_ERR(wh_dentry); ++ if (IS_ERR(wh_dentry)) ++ goto out_args; ++ ++ h_dentry = au_h_dptr(dentry, bstart); ++ dget(h_dentry); ++ rmdir_later = 0; ++ if (bindex == bstart) { ++ err = renwh_and_rmdir(dentry, bstart, &args->whlist, dir); ++ if (err > 0) { ++ rmdir_later = err; ++ err = 0; ++ } ++ } else { ++ /* stop monitoring */ ++ au_hin_free(au_hi(inode, bstart)); ++ ++ /* dir inode is locked */ ++ IMustLock(wh_dentry->d_parent->d_inode); ++ err = 0; ++ } ++ ++ if (!err) { ++ clear_nlink(inode); ++ au_set_dbdiropq(dentry, -1); ++ epilog(dir, dentry, bindex); ++ ++ if (rmdir_later) { ++ au_whtmp_kick_rmdir(dir, bstart, h_dentry, args); ++ args = NULL; ++ } ++ ++ goto out_unpin; /* success */ ++ } ++ ++ /* revert */ ++ AuLabel(revert); ++ if (wh_dentry) { ++ int rerr; ++ ++ rerr = do_revert(err, dir, bwh, wh_dentry, dentry, &dt); ++ if (rerr) ++ err = rerr; ++ } ++ ++ out_unpin: ++ au_unpin(&pin); ++ dput(wh_dentry); ++ dput(h_dentry); ++ out_args: ++ di_write_unlock(parent); ++ if (args) ++ au_whtmp_rmdir_free(args); ++ out_unlock: ++ aufs_read_unlock(dentry, AuLock_DW); ++ out: ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/i_op_ren.c linux-2.6.31/fs/aufs/i_op_ren.c +--- linux-2.6.31-vanilla/fs/aufs/i_op_ren.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/i_op_ren.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,957 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * inode operation (rename entry) ++ * todo: this is crazy monster ++ */ ++ ++#include "aufs.h" ++ ++enum { AuSRC, AuDST, AuSrcDst }; ++enum { AuPARENT, AuCHILD, AuParentChild }; ++ ++#define AuRen_ISDIR 1 ++#define AuRen_ISSAMEDIR (1 << 1) ++#define AuRen_WHSRC (1 << 2) ++#define AuRen_WHDST (1 << 3) ++#define AuRen_MNT_WRITE (1 << 4) ++#define AuRen_DT_DSTDIR (1 << 5) ++#define AuRen_DIROPQ (1 << 6) ++#define AuRen_CPUP (1 << 7) ++#define au_ftest_ren(flags, name) ((flags) & AuRen_##name) ++#define au_fset_ren(flags, name) { (flags) |= AuRen_##name; } ++#define au_fclr_ren(flags, name) { (flags) &= ~AuRen_##name; } ++ ++struct au_ren_args { ++ struct { ++ struct dentry *dentry, *h_dentry, *parent, *h_parent, ++ *wh_dentry; ++ struct inode *dir, *inode; ++ struct au_hinode *hdir; ++ struct au_dtime dt[AuParentChild]; ++ aufs_bindex_t bstart; ++ } sd[AuSrcDst]; ++ ++#define src_dentry sd[AuSRC].dentry ++#define src_dir sd[AuSRC].dir ++#define src_inode sd[AuSRC].inode ++#define src_h_dentry sd[AuSRC].h_dentry ++#define src_parent sd[AuSRC].parent ++#define src_h_parent sd[AuSRC].h_parent ++#define src_wh_dentry sd[AuSRC].wh_dentry ++#define src_hdir sd[AuSRC].hdir ++#define src_h_dir sd[AuSRC].hdir->hi_inode ++#define src_dt sd[AuSRC].dt ++#define src_bstart sd[AuSRC].bstart ++ ++#define dst_dentry sd[AuDST].dentry ++#define dst_dir sd[AuDST].dir ++#define dst_inode sd[AuDST].inode ++#define dst_h_dentry sd[AuDST].h_dentry ++#define dst_parent sd[AuDST].parent ++#define dst_h_parent sd[AuDST].h_parent ++#define dst_wh_dentry sd[AuDST].wh_dentry ++#define dst_hdir sd[AuDST].hdir ++#define dst_h_dir sd[AuDST].hdir->hi_inode ++#define dst_dt sd[AuDST].dt ++#define dst_bstart sd[AuDST].bstart ++ ++ struct dentry *h_trap; ++ struct au_branch *br; ++ struct au_hinode *src_hinode; ++ struct path h_path; ++ struct au_nhash whlist; ++ aufs_bindex_t btgt; ++ ++ unsigned int flags; ++ ++ struct au_whtmp_rmdir *thargs; ++ struct dentry *h_dst; ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * functions for reverting. ++ * when an error happened in a single rename systemcall, we should revert ++ * everything as if nothing happend. ++ * we don't need to revert the copied-up/down the parent dir since they are ++ * harmless. ++ */ ++ ++#define RevertFailure(fmt, args...) do { \ ++ AuIOErr("revert failure: " fmt " (%d, %d)\n", \ ++ ##args, err, rerr); \ ++ err = -EIO; \ ++} while (0) ++ ++static void au_ren_rev_diropq(int err, struct au_ren_args *a) ++{ ++ int rerr; ++ ++ au_hin_imtx_lock_nested(a->src_hinode, AuLsc_I_CHILD); ++ rerr = au_diropq_remove(a->src_dentry, a->btgt); ++ au_hin_imtx_unlock(a->src_hinode); ++ if (rerr) ++ RevertFailure("remove diropq %.*s", AuDLNPair(a->src_dentry)); ++} ++ ++ ++static void au_ren_rev_rename(int err, struct au_ren_args *a) ++{ ++ int rerr; ++ ++ a->h_path.dentry = au_lkup_one(&a->src_dentry->d_name, a->src_h_parent, ++ a->br, /*nd*/NULL); ++ rerr = PTR_ERR(a->h_path.dentry); ++ if (IS_ERR(a->h_path.dentry)) { ++ RevertFailure("au_lkup_one %.*s", AuDLNPair(a->src_dentry)); ++ return; ++ } ++ ++ rerr = vfsub_rename(a->dst_h_dir, ++ au_h_dptr(a->src_dentry, a->btgt), ++ a->src_h_dir, &a->h_path); ++ d_drop(a->h_path.dentry); ++ dput(a->h_path.dentry); ++ /* au_set_h_dptr(a->src_dentry, a->btgt, NULL); */ ++ if (rerr) ++ RevertFailure("rename %.*s", AuDLNPair(a->src_dentry)); ++} ++ ++static void au_ren_rev_cpup(int err, struct au_ren_args *a) ++{ ++ int rerr; ++ ++ a->h_path.dentry = a->dst_h_dentry; ++ rerr = vfsub_unlink(a->dst_h_dir, &a->h_path, /*force*/0); ++ au_set_h_dptr(a->src_dentry, a->btgt, NULL); ++ au_set_dbstart(a->src_dentry, a->src_bstart); ++ if (rerr) ++ RevertFailure("unlink %.*s", AuDLNPair(a->dst_h_dentry)); ++} ++ ++ ++static void au_ren_rev_whtmp(int err, struct au_ren_args *a) ++{ ++ int rerr; ++ ++ a->h_path.dentry = au_lkup_one(&a->dst_dentry->d_name, a->dst_h_parent, ++ a->br, /*nd*/NULL); ++ rerr = PTR_ERR(a->h_path.dentry); ++ if (IS_ERR(a->h_path.dentry)) { ++ RevertFailure("lookup %.*s", AuDLNPair(a->dst_dentry)); ++ return; ++ } ++ if (a->h_path.dentry->d_inode) { ++ d_drop(a->h_path.dentry); ++ dput(a->h_path.dentry); ++ return; ++ } ++ ++ rerr = vfsub_rename(a->dst_h_dir, a->h_dst, a->dst_h_dir, &a->h_path); ++ d_drop(a->h_path.dentry); ++ dput(a->h_path.dentry); ++ if (!rerr) { ++ au_set_h_dptr(a->dst_dentry, a->btgt, NULL); ++ au_set_h_dptr(a->dst_dentry, a->btgt, dget(a->h_dst)); ++ } else ++ RevertFailure("rename %.*s", AuDLNPair(a->h_dst)); ++} ++ ++static void au_ren_rev_whsrc(int err, struct au_ren_args *a) ++{ ++ int rerr; ++ ++ a->h_path.dentry = a->src_wh_dentry; ++ rerr = au_wh_unlink_dentry(a->src_h_dir, &a->h_path, a->src_dentry); ++ if (rerr) ++ RevertFailure("unlink %.*s", AuDLNPair(a->src_wh_dentry)); ++} ++ ++static void au_ren_rev_drop(struct au_ren_args *a) ++{ ++ struct dentry *d, *h_d; ++ int i; ++ aufs_bindex_t bend, bindex; ++ ++ for (i = 0; i < AuSrcDst; i++) { ++ d = a->sd[i].dentry; ++ d_drop(d); ++ bend = au_dbend(d); ++ for (bindex = au_dbstart(d); bindex <= bend; bindex++) { ++ h_d = au_h_dptr(d, bindex); ++ if (h_d) ++ d_drop(h_d); ++ } ++ } ++ ++ au_update_dbstart(a->dst_dentry); ++ if (a->thargs) ++ d_drop(a->h_dst); ++} ++#undef RevertFailure ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * when we have to copyup the renaming entry, do it with the rename-target name ++ * in order to minimize the cost (the later actual rename is unnecessary). ++ * otherwise rename it on the target branch. ++ */ ++static int au_ren_or_cpup(struct au_ren_args *a) ++{ ++ int err; ++ struct dentry *d; ++ ++ d = a->src_dentry; ++ if (au_dbstart(d) == a->btgt) { ++ a->h_path.dentry = a->dst_h_dentry; ++ if (au_ftest_ren(a->flags, DIROPQ) ++ && au_dbdiropq(d) == a->btgt) ++ au_fclr_ren(a->flags, DIROPQ); ++ AuDebugOn(au_dbstart(d) != a->btgt); ++ err = vfsub_rename(a->src_h_dir, au_h_dptr(d, a->btgt), ++ a->dst_h_dir, &a->h_path); ++ } else { ++ struct mutex *h_mtx = &a->src_h_dentry->d_inode->i_mutex; ++ ++ au_fset_ren(a->flags, CPUP); ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD); ++ au_set_dbstart(d, a->btgt); ++ au_set_h_dptr(d, a->btgt, dget(a->dst_h_dentry)); ++ err = au_sio_cpup_single(d, a->btgt, a->src_bstart, -1, ++ !AuCpup_DTIME, a->dst_parent); ++ if (unlikely(err)) { ++ au_set_h_dptr(d, a->btgt, NULL); ++ au_set_dbstart(d, a->src_bstart); ++ } ++ mutex_unlock(h_mtx); ++ } ++ ++ return err; ++} ++ ++/* cf. aufs_rmdir() */ ++static int au_ren_del_whtmp(struct au_ren_args *a) ++{ ++ int err; ++ struct inode *dir; ++ ++ dir = a->dst_dir; ++ SiMustAnyLock(dir->i_sb); ++ if (!au_nhash_test_longer_wh(&a->whlist, a->btgt, ++ au_sbi(dir->i_sb)->si_dirwh) ++ || au_test_fs_remote(a->h_dst->d_sb)) { ++ err = au_whtmp_rmdir(dir, a->btgt, a->h_dst, &a->whlist); ++ if (unlikely(err)) ++ AuWarn("failed removing whtmp dir %.*s (%d), " ++ "ignored.\n", AuDLNPair(a->h_dst), err); ++ } else { ++ au_nhash_wh_free(&a->thargs->whlist); ++ a->thargs->whlist = a->whlist; ++ a->whlist.nh_num = 0; ++ au_whtmp_kick_rmdir(dir, a->btgt, a->h_dst, a->thargs); ++ dput(a->h_dst); ++ a->thargs = NULL; ++ } ++ ++ return 0; ++} ++ ++/* make it 'opaque' dir. */ ++static int au_ren_diropq(struct au_ren_args *a) ++{ ++ int err; ++ struct dentry *diropq; ++ ++ err = 0; ++ a->src_hinode = au_hi(a->src_inode, a->btgt); ++ au_hin_imtx_lock_nested(a->src_hinode, AuLsc_I_CHILD); ++ diropq = au_diropq_create(a->src_dentry, a->btgt); ++ au_hin_imtx_unlock(a->src_hinode); ++ if (IS_ERR(diropq)) ++ err = PTR_ERR(diropq); ++ dput(diropq); ++ ++ return err; ++} ++ ++static int do_rename(struct au_ren_args *a) ++{ ++ int err; ++ struct dentry *d, *h_d; ++ ++ /* prepare workqueue args for asynchronous rmdir */ ++ h_d = a->dst_h_dentry; ++ if (au_ftest_ren(a->flags, ISDIR) && h_d->d_inode) { ++ err = -ENOMEM; ++ a->thargs = au_whtmp_rmdir_alloc(a->src_dentry->d_sb, GFP_NOFS); ++ if (unlikely(!a->thargs)) ++ goto out; ++ a->h_dst = dget(h_d); ++ } ++ ++ /* create whiteout for src_dentry */ ++ if (au_ftest_ren(a->flags, WHSRC)) { ++ a->src_wh_dentry ++ = au_wh_create(a->src_dentry, a->btgt, a->src_h_parent); ++ err = PTR_ERR(a->src_wh_dentry); ++ if (IS_ERR(a->src_wh_dentry)) ++ goto out_thargs; ++ } ++ ++ /* lookup whiteout for dentry */ ++ if (au_ftest_ren(a->flags, WHDST)) { ++ h_d = au_wh_lkup(a->dst_h_parent, &a->dst_dentry->d_name, ++ a->br); ++ err = PTR_ERR(h_d); ++ if (IS_ERR(h_d)) ++ goto out_whsrc; ++ if (!h_d->d_inode) ++ dput(h_d); ++ else ++ a->dst_wh_dentry = h_d; ++ } ++ ++ /* rename dentry to tmpwh */ ++ if (a->thargs) { ++ err = au_whtmp_ren(a->dst_h_dentry, a->br); ++ if (unlikely(err)) ++ goto out_whdst; ++ ++ d = a->dst_dentry; ++ au_set_h_dptr(d, a->btgt, NULL); ++ err = au_lkup_neg(d, a->btgt); ++ if (unlikely(err)) ++ goto out_whtmp; ++ a->dst_h_dentry = au_h_dptr(d, a->btgt); ++ } ++ ++ /* cpup src */ ++ if (a->dst_h_dentry->d_inode && a->src_bstart != a->btgt) { ++ struct mutex *h_mtx = &a->src_h_dentry->d_inode->i_mutex; ++ ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD); ++ err = au_sio_cpup_simple(a->src_dentry, a->btgt, -1, ++ !AuCpup_DTIME); ++ mutex_unlock(h_mtx); ++ if (unlikely(err)) ++ goto out_whtmp; ++ } ++ ++ /* rename by vfs_rename or cpup */ ++ d = a->dst_dentry; ++ if (au_ftest_ren(a->flags, ISDIR) ++ && (a->dst_wh_dentry ++ || au_dbdiropq(d) == a->btgt ++ /* hide the lower to keep xino */ ++ || a->btgt < au_dbend(d) ++ || au_opt_test(au_mntflags(d->d_sb), ALWAYS_DIROPQ))) ++ au_fset_ren(a->flags, DIROPQ); ++ err = au_ren_or_cpup(a); ++ if (unlikely(err)) ++ /* leave the copied-up one */ ++ goto out_whtmp; ++ ++ /* make dir opaque */ ++ if (au_ftest_ren(a->flags, DIROPQ)) { ++ err = au_ren_diropq(a); ++ if (unlikely(err)) ++ goto out_rename; ++ } ++ ++ /* update target timestamps */ ++ AuDebugOn(au_dbstart(a->src_dentry) != a->btgt); ++ a->h_path.dentry = au_h_dptr(a->src_dentry, a->btgt); ++ vfsub_update_h_iattr(&a->h_path, /*did*/NULL); /*ignore*/ ++ a->src_inode->i_ctime = a->h_path.dentry->d_inode->i_ctime; ++ ++ /* remove whiteout for dentry */ ++ if (a->dst_wh_dentry) { ++ a->h_path.dentry = a->dst_wh_dentry; ++ err = au_wh_unlink_dentry(a->dst_h_dir, &a->h_path, ++ a->dst_dentry); ++ if (unlikely(err)) ++ goto out_diropq; ++ } ++ ++ /* remove whtmp */ ++ if (a->thargs) ++ au_ren_del_whtmp(a); /* ignore this error */ ++ ++ err = 0; ++ goto out_success; ++ ++ out_diropq: ++ if (au_ftest_ren(a->flags, DIROPQ)) ++ au_ren_rev_diropq(err, a); ++ out_rename: ++ if (!au_ftest_ren(a->flags, CPUP)) ++ au_ren_rev_rename(err, a); ++ else ++ au_ren_rev_cpup(err, a); ++ out_whtmp: ++ if (a->thargs) ++ au_ren_rev_whtmp(err, a); ++ out_whdst: ++ dput(a->dst_wh_dentry); ++ a->dst_wh_dentry = NULL; ++ out_whsrc: ++ if (a->src_wh_dentry) ++ au_ren_rev_whsrc(err, a); ++ au_ren_rev_drop(a); ++ out_success: ++ dput(a->src_wh_dentry); ++ dput(a->dst_wh_dentry); ++ out_thargs: ++ if (a->thargs) { ++ dput(a->h_dst); ++ au_whtmp_rmdir_free(a->thargs); ++ a->thargs = NULL; ++ } ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * test if @dentry dir can be rename destination or not. ++ * success means, it is a logically empty dir. ++ */ ++static int may_rename_dstdir(struct dentry *dentry, struct au_nhash *whlist) ++{ ++ return au_test_empty(dentry, whlist); ++} ++ ++/* ++ * test if @dentry dir can be rename source or not. ++ * if it can, return 0 and @children is filled. ++ * success means, ++ * - it is a logically empty dir. ++ * - or, it exists on writable branch and has no children including whiteouts ++ * on the lower branch. ++ */ ++static int may_rename_srcdir(struct dentry *dentry, aufs_bindex_t btgt) ++{ ++ int err; ++ unsigned int rdhash; ++ aufs_bindex_t bstart; ++ ++ bstart = au_dbstart(dentry); ++ if (bstart != btgt) { ++ struct au_nhash whlist; ++ ++ SiMustAnyLock(dentry->d_sb); ++ rdhash = au_sbi(dentry->d_sb)->si_rdhash; ++ if (!rdhash) ++ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, ++ dentry)); ++ err = au_nhash_alloc(&whlist, rdhash, GFP_NOFS); ++ if (unlikely(err)) ++ goto out; ++ err = au_test_empty(dentry, &whlist); ++ au_nhash_wh_free(&whlist); ++ goto out; ++ } ++ ++ if (bstart == au_dbtaildir(dentry)) ++ return 0; /* success */ ++ ++ err = au_test_empty_lower(dentry); ++ ++ out: ++ if (err == -ENOTEMPTY) { ++ AuWarn1("renaming dir who has child(ren) on multiple branches," ++ " is not supported\n"); ++ err = -EXDEV; ++ } ++ return err; ++} ++ ++/* side effect: sets whlist and h_dentry */ ++static int au_ren_may_dir(struct au_ren_args *a) ++{ ++ int err; ++ unsigned int rdhash; ++ struct dentry *d; ++ ++ d = a->dst_dentry; ++ SiMustAnyLock(d->d_sb); ++ ++ err = 0; ++ if (au_ftest_ren(a->flags, ISDIR) && a->dst_inode) { ++ rdhash = au_sbi(d->d_sb)->si_rdhash; ++ if (!rdhash) ++ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, d)); ++ err = au_nhash_alloc(&a->whlist, rdhash, GFP_NOFS); ++ if (unlikely(err)) ++ goto out; ++ ++ au_set_dbstart(d, a->dst_bstart); ++ err = may_rename_dstdir(d, &a->whlist); ++ au_set_dbstart(d, a->btgt); ++ } ++ a->dst_h_dentry = au_h_dptr(d, au_dbstart(d)); ++ if (unlikely(err)) ++ goto out; ++ ++ d = a->src_dentry; ++ a->src_h_dentry = au_h_dptr(d, au_dbstart(d)); ++ if (au_ftest_ren(a->flags, ISDIR)) { ++ err = may_rename_srcdir(d, a->btgt); ++ if (unlikely(err)) { ++ au_nhash_wh_free(&a->whlist); ++ a->whlist.nh_num = 0; ++ } ++ } ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * simple tests for rename. ++ * following the checks in vfs, plus the parent-child relationship. ++ */ ++static int au_may_ren(struct au_ren_args *a) ++{ ++ int err, isdir; ++ struct inode *h_inode; ++ ++ if (a->src_bstart == a->btgt) { ++ err = au_may_del(a->src_dentry, a->btgt, a->src_h_parent, ++ au_ftest_ren(a->flags, ISDIR)); ++ if (unlikely(err)) ++ goto out; ++ err = -EINVAL; ++ if (unlikely(a->src_h_dentry == a->h_trap)) ++ goto out; ++ } ++ ++ err = 0; ++ if (a->dst_bstart != a->btgt) ++ goto out; ++ ++ err = -EIO; ++ h_inode = a->dst_h_dentry->d_inode; ++ isdir = !!au_ftest_ren(a->flags, ISDIR); ++ if (!a->dst_dentry->d_inode) { ++ if (unlikely(h_inode)) ++ goto out; ++ err = au_may_add(a->dst_dentry, a->btgt, a->dst_h_parent, ++ isdir); ++ } else { ++ if (unlikely(!h_inode || !h_inode->i_nlink)) ++ goto out; ++ err = au_may_del(a->dst_dentry, a->btgt, a->dst_h_parent, ++ isdir); ++ if (unlikely(err)) ++ goto out; ++ err = -ENOTEMPTY; ++ if (unlikely(a->dst_h_dentry == a->h_trap)) ++ goto out; ++ err = 0; ++ } ++ ++ out: ++ if (unlikely(err == -ENOENT || err == -EEXIST)) ++ err = -EIO; ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * locking order ++ * (VFS) ++ * - src_dir and dir by lock_rename() ++ * - inode if exitsts ++ * (aufs) ++ * - lock all ++ * + src_dentry and dentry by aufs_read_and_write_lock2() which calls, ++ * + si_read_lock ++ * + di_write_lock2_child() ++ * + di_write_lock_child() ++ * + ii_write_lock_child() ++ * + di_write_lock_child2() ++ * + ii_write_lock_child2() ++ * + src_parent and parent ++ * + di_write_lock_parent() ++ * + ii_write_lock_parent() ++ * + di_write_lock_parent2() ++ * + ii_write_lock_parent2() ++ * + lower src_dir and dir by vfsub_lock_rename() ++ * + verify the every relationships between child and parent. if any ++ * of them failed, unlock all and return -EBUSY. ++ */ ++static void au_ren_unlock(struct au_ren_args *a) ++{ ++ struct super_block *sb; ++ ++ sb = a->dst_dentry->d_sb; ++ if (au_ftest_ren(a->flags, MNT_WRITE)) ++ mnt_drop_write(a->br->br_mnt); ++ vfsub_unlock_rename(a->src_h_parent, a->src_hdir, ++ a->dst_h_parent, a->dst_hdir); ++} ++ ++static int au_ren_lock(struct au_ren_args *a) ++{ ++ int err; ++ unsigned int udba; ++ ++ err = 0; ++ a->src_h_parent = au_h_dptr(a->src_parent, a->btgt); ++ a->src_hdir = au_hi(a->src_dir, a->btgt); ++ a->dst_h_parent = au_h_dptr(a->dst_parent, a->btgt); ++ a->dst_hdir = au_hi(a->dst_dir, a->btgt); ++ a->h_trap = vfsub_lock_rename(a->src_h_parent, a->src_hdir, ++ a->dst_h_parent, a->dst_hdir); ++ udba = au_opt_udba(a->src_dentry->d_sb); ++ if (unlikely(a->src_hdir->hi_inode != a->src_h_parent->d_inode ++ || a->dst_hdir->hi_inode != a->dst_h_parent->d_inode)) ++ err = au_busy_or_stale(); ++ if (!err && au_dbstart(a->src_dentry) == a->btgt) ++ err = au_h_verify(a->src_h_dentry, udba, ++ a->src_h_parent->d_inode, a->src_h_parent, ++ a->br); ++ if (!err && au_dbstart(a->dst_dentry) == a->btgt) ++ err = au_h_verify(a->dst_h_dentry, udba, ++ a->dst_h_parent->d_inode, a->dst_h_parent, ++ a->br); ++ if (!err) { ++ err = mnt_want_write(a->br->br_mnt); ++ if (unlikely(err)) ++ goto out_unlock; ++ au_fset_ren(a->flags, MNT_WRITE); ++ goto out; /* success */ ++ } ++ ++ err = au_busy_or_stale(); ++ ++ out_unlock: ++ au_ren_unlock(a); ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static void au_ren_refresh_dir(struct au_ren_args *a) ++{ ++ struct inode *dir; ++ ++ dir = a->dst_dir; ++ dir->i_version++; ++ if (au_ftest_ren(a->flags, ISDIR)) { ++ /* is this updating defined in POSIX? */ ++ au_cpup_attr_timesizes(a->src_inode); ++ au_cpup_attr_nlink(dir, /*force*/1); ++ if (a->dst_inode) { ++ clear_nlink(a->dst_inode); ++ au_cpup_attr_timesizes(a->dst_inode); ++ } ++ } ++ if (au_ibstart(dir) == a->btgt) ++ au_cpup_attr_timesizes(dir); ++ ++ if (au_ftest_ren(a->flags, ISSAMEDIR)) ++ return; ++ ++ dir = a->src_dir; ++ dir->i_version++; ++ if (au_ftest_ren(a->flags, ISDIR)) ++ au_cpup_attr_nlink(dir, /*force*/1); ++ if (au_ibstart(dir) == a->btgt) ++ au_cpup_attr_timesizes(dir); ++} ++ ++static void au_ren_refresh(struct au_ren_args *a) ++{ ++ aufs_bindex_t bend, bindex; ++ struct dentry *d, *h_d; ++ struct inode *i, *h_i; ++ struct super_block *sb; ++ ++ d = a->src_dentry; ++ au_set_dbwh(d, -1); ++ bend = au_dbend(d); ++ for (bindex = a->btgt + 1; bindex <= bend; bindex++) { ++ h_d = au_h_dptr(d, bindex); ++ if (h_d) ++ au_set_h_dptr(d, bindex, NULL); ++ } ++ au_set_dbend(d, a->btgt); ++ ++ sb = d->d_sb; ++ i = a->src_inode; ++ if (au_opt_test(au_mntflags(sb), PLINK) && au_plink_test(i)) ++ return; /* success */ ++ ++ bend = au_ibend(i); ++ for (bindex = a->btgt + 1; bindex <= bend; bindex++) { ++ h_i = au_h_iptr(i, bindex); ++ if (h_i) { ++ au_xino_write(sb, bindex, h_i->i_ino, /*ino*/0); ++ /* ignore this error */ ++ au_set_h_iptr(i, bindex, NULL, 0); ++ } ++ } ++ au_set_ibend(i, a->btgt); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* mainly for link(2) and rename(2) */ ++int au_wbr(struct dentry *dentry, aufs_bindex_t btgt) ++{ ++ aufs_bindex_t bdiropq, bwh; ++ struct dentry *parent; ++ struct au_branch *br; ++ ++ parent = dentry->d_parent; ++ IMustLock(parent->d_inode); /* dir is locked */ ++ ++ bdiropq = au_dbdiropq(parent); ++ bwh = au_dbwh(dentry); ++ br = au_sbr(dentry->d_sb, btgt); ++ if (au_br_rdonly(br) ++ || (0 <= bdiropq && bdiropq < btgt) ++ || (0 <= bwh && bwh < btgt)) ++ btgt = -1; ++ ++ AuDbg("btgt %d\n", btgt); ++ return btgt; ++} ++ ++/* sets src_bstart, dst_bstart and btgt */ ++static int au_ren_wbr(struct au_ren_args *a) ++{ ++ int err; ++ struct au_wr_dir_args wr_dir_args = { ++ /* .force_btgt = -1, */ ++ .flags = AuWrDir_ADD_ENTRY ++ }; ++ ++ a->src_bstart = au_dbstart(a->src_dentry); ++ a->dst_bstart = au_dbstart(a->dst_dentry); ++ if (au_ftest_ren(a->flags, ISDIR)) ++ au_fset_wrdir(wr_dir_args.flags, ISDIR); ++ wr_dir_args.force_btgt = a->src_bstart; ++ if (a->dst_inode && a->dst_bstart < a->src_bstart) ++ wr_dir_args.force_btgt = a->dst_bstart; ++ wr_dir_args.force_btgt = au_wbr(a->dst_dentry, wr_dir_args.force_btgt); ++ err = au_wr_dir(a->dst_dentry, a->src_dentry, &wr_dir_args); ++ a->btgt = err; ++ ++ return err; ++} ++ ++static void au_ren_dt(struct au_ren_args *a) ++{ ++ a->h_path.dentry = a->src_h_parent; ++ au_dtime_store(a->src_dt + AuPARENT, a->src_parent, &a->h_path); ++ if (!au_ftest_ren(a->flags, ISSAMEDIR)) { ++ a->h_path.dentry = a->dst_h_parent; ++ au_dtime_store(a->dst_dt + AuPARENT, a->dst_parent, &a->h_path); ++ } ++ ++ au_fclr_ren(a->flags, DT_DSTDIR); ++ if (!au_ftest_ren(a->flags, ISDIR)) ++ return; ++ ++ a->h_path.dentry = a->src_h_dentry; ++ au_dtime_store(a->src_dt + AuCHILD, a->src_dentry, &a->h_path); ++ if (a->dst_h_dentry->d_inode) { ++ au_fset_ren(a->flags, DT_DSTDIR); ++ a->h_path.dentry = a->dst_h_dentry; ++ au_dtime_store(a->dst_dt + AuCHILD, a->dst_dentry, &a->h_path); ++ } ++} ++ ++static void au_ren_rev_dt(int err, struct au_ren_args *a) ++{ ++ struct dentry *h_d; ++ struct mutex *h_mtx; ++ ++ au_dtime_revert(a->src_dt + AuPARENT); ++ if (!au_ftest_ren(a->flags, ISSAMEDIR)) ++ au_dtime_revert(a->dst_dt + AuPARENT); ++ ++ if (au_ftest_ren(a->flags, ISDIR) && err != -EIO) { ++ h_d = a->src_dt[AuCHILD].dt_h_path.dentry; ++ h_mtx = &h_d->d_inode->i_mutex; ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD); ++ au_dtime_revert(a->src_dt + AuCHILD); ++ mutex_unlock(h_mtx); ++ ++ if (au_ftest_ren(a->flags, DT_DSTDIR)) { ++ h_d = a->dst_dt[AuCHILD].dt_h_path.dentry; ++ h_mtx = &h_d->d_inode->i_mutex; ++ mutex_lock_nested(h_mtx, AuLsc_I_CHILD); ++ au_dtime_revert(a->dst_dt + AuCHILD); ++ mutex_unlock(h_mtx); ++ } ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int aufs_rename(struct inode *_src_dir, struct dentry *_src_dentry, ++ struct inode *_dst_dir, struct dentry *_dst_dentry) ++{ ++ int err; ++ /* reduce stack space */ ++ struct au_ren_args *a; ++ ++ IMustLock(_src_dir); ++ IMustLock(_dst_dir); ++ ++ err = -ENOMEM; ++ BUILD_BUG_ON(sizeof(*a) > PAGE_SIZE); ++ a = kzalloc(sizeof(*a), GFP_NOFS); ++ if (unlikely(!a)) ++ goto out; ++ ++ a->src_dir = _src_dir; ++ a->src_dentry = _src_dentry; ++ a->src_inode = a->src_dentry->d_inode; ++ a->src_parent = a->src_dentry->d_parent; /* dir inode is locked */ ++ a->dst_dir = _dst_dir; ++ a->dst_dentry = _dst_dentry; ++ a->dst_inode = a->dst_dentry->d_inode; ++ a->dst_parent = a->dst_dentry->d_parent; /* dir inode is locked */ ++ if (a->dst_inode) { ++ IMustLock(a->dst_inode); ++ au_igrab(a->dst_inode); ++ } ++ ++ err = -ENOTDIR; ++ if (S_ISDIR(a->src_inode->i_mode)) { ++ au_fset_ren(a->flags, ISDIR); ++ if (unlikely(a->dst_inode && !S_ISDIR(a->dst_inode->i_mode))) ++ goto out_free; ++ aufs_read_and_write_lock2(a->dst_dentry, a->src_dentry, ++ AuLock_DIR | AuLock_FLUSH); ++ } else ++ aufs_read_and_write_lock2(a->dst_dentry, a->src_dentry, ++ AuLock_FLUSH); ++ ++ au_fset_ren(a->flags, ISSAMEDIR); /* temporary */ ++ di_write_lock_parent(a->dst_parent); ++ ++ /* which branch we process */ ++ err = au_ren_wbr(a); ++ if (unlikely(err < 0)) ++ goto out_unlock; ++ a->br = au_sbr(a->dst_dentry->d_sb, a->btgt); ++ a->h_path.mnt = a->br->br_mnt; ++ ++ /* are they available to be renamed */ ++ err = au_ren_may_dir(a); ++ if (unlikely(err)) ++ goto out_children; ++ ++ /* prepare the writable parent dir on the same branch */ ++ if (a->dst_bstart == a->btgt) { ++ au_fset_ren(a->flags, WHDST); ++ } else { ++ err = au_cpup_dirs(a->dst_dentry, a->btgt); ++ if (unlikely(err)) ++ goto out_children; ++ } ++ ++ if (a->src_dir != a->dst_dir) { ++ /* ++ * this temporary unlock is safe, ++ * because both dir->i_mutex are locked. ++ */ ++ di_write_unlock(a->dst_parent); ++ di_write_lock_parent(a->src_parent); ++ err = au_wr_dir_need_wh(a->src_dentry, ++ au_ftest_ren(a->flags, ISDIR), ++ &a->btgt); ++ di_write_unlock(a->src_parent); ++ di_write_lock2_parent(a->src_parent, a->dst_parent, /*isdir*/1); ++ au_fclr_ren(a->flags, ISSAMEDIR); ++ } else ++ err = au_wr_dir_need_wh(a->src_dentry, ++ au_ftest_ren(a->flags, ISDIR), ++ &a->btgt); ++ if (unlikely(err < 0)) ++ goto out_children; ++ if (err) ++ au_fset_ren(a->flags, WHSRC); ++ ++ /* lock them all */ ++ err = au_ren_lock(a); ++ if (unlikely(err)) ++ goto out_children; ++ ++ if (!au_opt_test(au_mntflags(a->dst_dir->i_sb), UDBA_NONE)) { ++ err = au_may_ren(a); ++ if (unlikely(err)) ++ goto out_hdir; ++ } ++ ++ /* store timestamps to be revertible */ ++ au_ren_dt(a); ++ ++ /* here we go */ ++ err = do_rename(a); ++ if (unlikely(err)) ++ goto out_dt; ++ ++ /* update dir attributes */ ++ au_ren_refresh_dir(a); ++ ++ /* dput/iput all lower dentries */ ++ au_ren_refresh(a); ++ ++ goto out_hdir; /* success */ ++ ++ out_dt: ++ au_ren_rev_dt(err, a); ++ out_hdir: ++ au_ren_unlock(a); ++ out_children: ++ au_nhash_wh_free(&a->whlist); ++ out_unlock: ++ if (unlikely(err && au_ftest_ren(a->flags, ISDIR))) { ++ au_update_dbstart(a->dst_dentry); ++ d_drop(a->dst_dentry); ++ } ++ if (!err) ++ d_move(a->src_dentry, a->dst_dentry); ++ if (au_ftest_ren(a->flags, ISSAMEDIR)) ++ di_write_unlock(a->dst_parent); ++ else ++ di_write_unlock2(a->src_parent, a->dst_parent); ++ aufs_read_and_write_unlock2(a->dst_dentry, a->src_dentry); ++ out_free: ++ iput(a->dst_inode); ++ if (a->thargs) ++ au_whtmp_rmdir_free(a->thargs); ++ kfree(a); ++ out: ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/Kconfig linux-2.6.31/fs/aufs/Kconfig +--- linux-2.6.31-vanilla/fs/aufs/Kconfig 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/Kconfig 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,140 @@ ++config AUFS_FS ++ tristate "Aufs (Advanced multi layered unification filesystem) support" ++ depends on EXPERIMENTAL ++ help ++ Aufs is a stackable unification filesystem such as Unionfs, ++ which unifies several directories and provides a merged single ++ directory. ++ In the early days, aufs was entirely re-designed and ++ re-implemented Unionfs Version 1.x series. Introducing many ++ original ideas, approaches and improvements, it becomes totally ++ different from Unionfs while keeping the basic features. ++ ++if AUFS_FS ++choice ++ prompt "Maximum number of branches" ++ default AUFS_BRANCH_MAX_127 ++ help ++ Specifies the maximum number of branches (or member directories) ++ in a single aufs. The larger value consumes more system ++ resources and has a minor impact to performance. ++config AUFS_BRANCH_MAX_127 ++ bool "127" ++ help ++ Specifies the maximum number of branches (or member directories) ++ in a single aufs. The larger value consumes more system ++ resources and has a minor impact to performance. ++config AUFS_BRANCH_MAX_511 ++ bool "511" ++ help ++ Specifies the maximum number of branches (or member directories) ++ in a single aufs. The larger value consumes more system ++ resources and has a minor impact to performance. ++config AUFS_BRANCH_MAX_1023 ++ bool "1023" ++ help ++ Specifies the maximum number of branches (or member directories) ++ in a single aufs. The larger value consumes more system ++ resources and has a minor impact to performance. ++config AUFS_BRANCH_MAX_32767 ++ bool "32767" ++ help ++ Specifies the maximum number of branches (or member directories) ++ in a single aufs. The larger value consumes more system ++ resources and has a minor impact to performance. ++endchoice ++ ++config AUFS_HINOTIFY ++ bool "Use inotify to detect actions on a branch" ++ depends on INOTIFY ++ help ++ If you want to modify files on branches directly, eg. bypassing aufs, ++ and want aufs to detect the changes of them fully, then enable this ++ option and use 'udba=inotify' mount option. ++ It will have a negative impact to the performance. ++ See detail in aufs.5. ++ ++config AUFS_EXPORT ++ bool "NFS-exportable aufs" ++ depends on (AUFS_FS = y && EXPORTFS = y) || (AUFS_FS = m && EXPORTFS) ++ help ++ If you want to export your mounted aufs via NFS, then enable this ++ option. There are several requirements for this configuration. ++ See detail in aufs.5. ++ ++config AUFS_RDU ++ bool "Readdir in userspace" ++ help ++ If you have millions of files under a single aufs directory, and ++ meet the out of memory, then enable this option and set ++ environment variables for your readdir(3). ++ See detail in aufs.5. ++ ++config AUFS_SHWH ++ bool "Show whiteouts" ++ help ++ If you want to make the whiteouts in aufs visible, then enable ++ this option and specify 'shwh' mount option. Although it may ++ sounds like philosophy or something, but in technically it ++ simply shows the name of whiteout with keeping its behaviour. ++ ++config AUFS_BR_RAMFS ++ bool "Ramfs (initramfs/rootfs) as an aufs branch" ++ help ++ If you want to use ramfs as an aufs branch fs, then enable this ++ option. Generally tmpfs is recommended. ++ Aufs prohibited them to be a branch fs by default, because ++ initramfs becomes unusable after switch_root or something ++ generally. If you sets initramfs as an aufs branch and boot your ++ system by switch_root, you will meet a problem easily since the ++ files in initramfs may be inaccessible. ++ Unless you are going to use ramfs as an aufs branch fs without ++ switch_root or something, leave it N. ++ ++config AUFS_BR_FUSE ++ bool "Fuse fs as an aufs branch" ++ depends on FUSE_FS ++ select AUFS_POLL ++ help ++ If you want to use fuse-based userspace filesystem as an aufs ++ branch fs, then enable this option. ++ It implements the internal poll(2) operation which is ++ implemented by fuse only (curretnly). ++ ++config AUFS_DEBUG ++ bool "Debug aufs" ++ help ++ Enable this to compile aufs internal debug code. ++ It will have a negative impact to the performance. ++ ++config AUFS_MAGIC_SYSRQ ++ bool ++ depends on AUFS_DEBUG && MAGIC_SYSRQ ++ default y ++ help ++ Automatic configuration for internal use. ++ When aufs supports Magic SysRq, enabled automatically. ++ ++config AUFS_BDEV_LOOP ++ bool ++ depends on BLK_DEV_LOOP ++ default y ++ help ++ Automatic configuration for internal use. ++ Convert =[ym] into =y. ++ ++config AUFS_INO_T_64 ++ bool ++ depends on AUFS_EXPORT ++ depends on 64BIT && !(ALPHA || S390) ++ default y ++ help ++ Automatic configuration for internal use. ++ /* typedef unsigned long/int __kernel_ino_t */ ++ /* alpha and s390x are int */ ++ ++config AUFS_POLL ++ bool ++ help ++ Automatic configuration for internal use. ++endif +diff -Nur linux-2.6.31-vanilla/fs/aufs/loop.c linux-2.6.31/fs/aufs/loop.c +--- linux-2.6.31-vanilla/fs/aufs/loop.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/loop.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,55 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * support for loopback block device as a branch ++ */ ++ ++#include ++#include "aufs.h" ++ ++/* ++ * test if two lower dentries have overlapping branches. ++ */ ++int au_test_loopback_overlap(struct super_block *sb, struct dentry *h_d1, ++ struct dentry *h_d2) ++{ ++ struct inode *h_inode; ++ struct loop_device *l; ++ ++ h_inode = h_d1->d_inode; ++ if (MAJOR(h_inode->i_sb->s_dev) != LOOP_MAJOR) ++ return 0; ++ ++ l = h_inode->i_sb->s_bdev->bd_disk->private_data; ++ h_d1 = l->lo_backing_file->f_dentry; ++ /* h_d1 can be local NFS. in this case aufs cannot detect the loop */ ++ if (unlikely(h_d1->d_sb == sb)) ++ return 1; ++ return !!au_test_subdir(h_d1, h_d2); ++} ++ ++/* true if a kernel thread named 'loop[0-9].*' accesses a file */ ++int au_test_loopback_kthread(void) ++{ ++ const char c = current->comm[4]; ++ ++ return current->mm == NULL ++ && '0' <= c && c <= '9' ++ && strncmp(current->comm, "loop", 4) == 0; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/loop.h linux-2.6.31/fs/aufs/loop.h +--- linux-2.6.31-vanilla/fs/aufs/loop.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/loop.h 2009-09-16 13:55:29.000000000 +0200 +@@ -0,0 +1,51 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * support for loopback mount as a branch ++ */ ++ ++#ifndef __AUFS_LOOP_H__ ++#define __AUFS_LOOP_H__ ++ ++#ifdef __KERNEL__ ++ ++struct dentry; ++struct super_block; ++ ++#ifdef CONFIG_AUFS_BDEV_LOOP ++/* loop.c */ ++int au_test_loopback_overlap(struct super_block *sb, struct dentry *h_d1, ++ struct dentry *h_d2); ++int au_test_loopback_kthread(void); ++#else ++static inline ++int au_test_loopback_overlap(struct super_block *sb, struct dentry *h_d1, ++ struct dentry *h_d2) ++{ ++ return 0; ++} ++ ++static inline int au_test_loopback_kthread(void) ++{ ++ return 0; ++} ++#endif /* BLK_DEV_LOOP */ ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_LOOP_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/magic.mk linux-2.6.31/fs/aufs/magic.mk +--- linux-2.6.31-vanilla/fs/aufs/magic.mk 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/magic.mk 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,52 @@ ++ ++# defined in ${srctree}/fs/fuse/inode.c ++# tristate ++ifdef CONFIG_FUSE_FS ++ccflags-y += -DFUSE_SUPER_MAGIC=0x65735546 ++endif ++ ++# defined in ${srctree}/fs/ocfs2/ocfs2_fs.h ++# tristate ++ifdef CONFIG_OCFS2_FS ++ccflags-y += -DOCFS2_SUPER_MAGIC=0x7461636f ++endif ++ ++# defined in ${srctree}/fs/ocfs2/dlm/userdlm.h ++# tristate ++ifdef CONFIG_OCFS2_FS_O2CB ++ccflags-y += -DDLMFS_MAGIC=0x76a9f425 ++endif ++ ++# defined in ${srctree}/fs/ramfs/inode.c ++# always true ++ccflags-y += -DRAMFS_MAGIC=0x858458f6 ++ ++# defined in ${srctree}/fs/cifs/cifsfs.c ++# tristate ++ifdef CONFIG_CIFS_FS ++ccflags-y += -DCIFS_MAGIC_NUMBER=0xFF534D42 ++endif ++ ++# defined in ${srctree}/fs/xfs/xfs_sb.h ++# tristate ++ifdef CONFIG_XFS_FS ++ccflags-y += -DXFS_SB_MAGIC=0x58465342 ++endif ++ ++# defined in ${srctree}/fs/configfs/mount.c ++# tristate ++ifdef CONFIG_CONFIGFS_FS ++ccflags-y += -DCONFIGFS_MAGIC=0x62656570 ++endif ++ ++# defined in ${srctree}/fs/9p/v9fs.h ++# tristate ++ifdef CONFIG_9P_FS ++ccflags-y += -DV9FS_MAGIC=0x01021997 ++endif ++ ++# defined in ${srctree}/fs/ubifs/ubifs.h ++# tristate ++ifdef CONFIG_UBIFS_FS ++ccflags-y += -DUBIFS_SUPER_MAGIC=0x24051905 ++endif +diff -Nur linux-2.6.31-vanilla/fs/aufs/Makefile linux-2.6.31/fs/aufs/Makefile +--- linux-2.6.31-vanilla/fs/aufs/Makefile 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/Makefile 2009-09-16 13:55:29.000000000 +0200 +@@ -0,0 +1,24 @@ ++ ++include ${src}/magic.mk ++-include ${src}/priv_def.mk ++ ++obj-$(CONFIG_AUFS_FS) += aufs.o ++aufs-y := module.o sbinfo.o super.o branch.o xino.o sysaufs.o opts.o \ ++ wkq.o vfsub.o dcsub.o \ ++ cpup.o whout.o plink.o wbr_policy.o \ ++ dinfo.o dentry.o \ ++ finfo.o file.o f_op.o \ ++ dir.o vdir.o \ ++ iinfo.o inode.o i_op.o i_op_add.o i_op_del.o i_op_ren.o \ ++ ioctl.o ++ ++# all are boolean ++aufs-$(CONFIG_SYSFS) += sysfs.o ++aufs-$(CONFIG_DEBUG_FS) += dbgaufs.o ++aufs-$(CONFIG_AUFS_BDEV_LOOP) += loop.o ++aufs-$(CONFIG_AUFS_HINOTIFY) += hinotify.o ++aufs-$(CONFIG_AUFS_EXPORT) += export.o ++aufs-$(CONFIG_AUFS_POLL) += poll.o ++aufs-$(CONFIG_AUFS_RDU) += rdu.o ++aufs-$(CONFIG_AUFS_DEBUG) += debug.o ++aufs-$(CONFIG_AUFS_MAGIC_SYSRQ) += sysrq.o +diff -Nur linux-2.6.31-vanilla/fs/aufs/module.c linux-2.6.31/fs/aufs/module.c +--- linux-2.6.31-vanilla/fs/aufs/module.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/module.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,173 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * module global variables and operations ++ */ ++ ++#include ++#include ++#include "aufs.h" ++ ++void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp) ++{ ++ if (new_sz <= nused) ++ return p; ++ ++ p = krealloc(p, new_sz, gfp); ++ if (p) ++ memset(p + nused, 0, new_sz - nused); ++ return p; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * aufs caches ++ */ ++struct kmem_cache *au_cachep[AuCache_Last]; ++static int __init au_cache_init(void) ++{ ++ au_cachep[AuCache_DINFO] = AuCache(au_dinfo); ++ if (au_cachep[AuCache_DINFO]) ++ au_cachep[AuCache_ICNTNR] = AuCache(au_icntnr); ++ if (au_cachep[AuCache_ICNTNR]) ++ au_cachep[AuCache_FINFO] = AuCache(au_finfo); ++ if (au_cachep[AuCache_FINFO]) ++ au_cachep[AuCache_VDIR] = AuCache(au_vdir); ++ if (au_cachep[AuCache_VDIR]) ++ au_cachep[AuCache_DEHSTR] = AuCache(au_vdir_dehstr); ++ if (au_cachep[AuCache_DEHSTR]) ++ return 0; ++ ++ return -ENOMEM; ++} ++ ++static void au_cache_fin(void) ++{ ++ int i; ++ for (i = 0; i < AuCache_Last; i++) ++ if (au_cachep[i]) { ++ kmem_cache_destroy(au_cachep[i]); ++ au_cachep[i] = NULL; ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int au_dir_roflags; ++ ++/* ++ * functions for module interface. ++ */ ++MODULE_LICENSE("GPL"); ++/* MODULE_LICENSE("GPL v2"); */ ++MODULE_AUTHOR("Junjiro R. Okajima "); ++MODULE_DESCRIPTION(AUFS_NAME ++ " -- Advanced multi layered unification filesystem"); ++MODULE_VERSION(AUFS_VERSION); ++ ++/* it should be 'byte', but param_set_byte() prints it by "%c" */ ++short aufs_nwkq = AUFS_NWKQ_DEF; ++MODULE_PARM_DESC(nwkq, "the number of workqueue thread, " AUFS_WKQ_NAME); ++module_param_named(nwkq, aufs_nwkq, short, S_IRUGO); ++ ++/* this module parameter has no meaning when SYSFS is disabled */ ++int sysaufs_brs = 1; ++MODULE_PARM_DESC(brs, "use /fs/aufs/si_*/brN"); ++module_param_named(brs, sysaufs_brs, int, S_IRUGO); ++ ++/* ---------------------------------------------------------------------- */ ++ ++static char au_esc_chars[0x20 + 3]; /* 0x01-0x20, backslash, del, and NULL */ ++ ++int au_seq_path(struct seq_file *seq, struct path *path) ++{ ++ return seq_path(seq, path, au_esc_chars); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int __init aufs_init(void) ++{ ++ int err, i; ++ char *p; ++ ++ p = au_esc_chars; ++ for (i = 1; i <= ' '; i++) ++ *p++ = i; ++ *p++ = '\\'; ++ *p++ = '\x7f'; ++ *p = 0; ++ ++ au_dir_roflags = au_file_roflags(O_DIRECTORY | O_LARGEFILE); ++ ++ sysaufs_brs_init(); ++ au_debug_init(); ++ ++ err = -EINVAL; ++ if (unlikely(aufs_nwkq <= 0)) ++ goto out; ++ ++ err = sysaufs_init(); ++ if (unlikely(err)) ++ goto out; ++ err = au_wkq_init(); ++ if (unlikely(err)) ++ goto out_sysaufs; ++ err = au_hinotify_init(); ++ if (unlikely(err)) ++ goto out_wkq; ++ err = au_sysrq_init(); ++ if (unlikely(err)) ++ goto out_hin; ++ err = au_cache_init(); ++ if (unlikely(err)) ++ goto out_sysrq; ++ err = register_filesystem(&aufs_fs_type); ++ if (unlikely(err)) ++ goto out_cache; ++ pr_info(AUFS_NAME " " AUFS_VERSION "\n"); ++ goto out; /* success */ ++ ++ out_cache: ++ au_cache_fin(); ++ out_sysrq: ++ au_sysrq_fin(); ++ out_hin: ++ au_hinotify_fin(); ++ out_wkq: ++ au_wkq_fin(); ++ out_sysaufs: ++ sysaufs_fin(); ++ out: ++ return err; ++} ++ ++static void __exit aufs_exit(void) ++{ ++ unregister_filesystem(&aufs_fs_type); ++ au_cache_fin(); ++ au_sysrq_fin(); ++ au_hinotify_fin(); ++ au_wkq_fin(); ++ sysaufs_fin(); ++} ++ ++module_init(aufs_init); ++module_exit(aufs_exit); +diff -Nur linux-2.6.31-vanilla/fs/aufs/module.h linux-2.6.31/fs/aufs/module.h +--- linux-2.6.31-vanilla/fs/aufs/module.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/module.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,78 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * module initialization and module-global ++ */ ++ ++#ifndef __AUFS_MODULE_H__ ++#define __AUFS_MODULE_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++ ++struct path; ++struct seq_file; ++ ++/* module parameters */ ++extern short aufs_nwkq; ++extern int sysaufs_brs; ++ ++/* ---------------------------------------------------------------------- */ ++ ++extern int au_dir_roflags; ++ ++void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp); ++int au_seq_path(struct seq_file *seq, struct path *path); ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* kmem cache */ ++enum { ++ AuCache_DINFO, ++ AuCache_ICNTNR, ++ AuCache_FINFO, ++ AuCache_VDIR, ++ AuCache_DEHSTR, ++#ifdef CONFIG_AUFS_HINOTIFY ++ AuCache_HINOTIFY, ++#endif ++ AuCache_Last ++}; ++ ++#define AuCache(type) KMEM_CACHE(type, SLAB_RECLAIM_ACCOUNT) ++ ++extern struct kmem_cache *au_cachep[]; ++ ++#define AuCacheFuncs(name, index) \ ++static inline void *au_cache_alloc_##name(void) \ ++{ return kmem_cache_alloc(au_cachep[AuCache_##index], GFP_NOFS); } \ ++static inline void au_cache_free_##name(void *p) \ ++{ kmem_cache_free(au_cachep[AuCache_##index], p); } ++ ++AuCacheFuncs(dinfo, DINFO); ++AuCacheFuncs(icntnr, ICNTNR); ++AuCacheFuncs(finfo, FINFO); ++AuCacheFuncs(vdir, VDIR); ++AuCacheFuncs(dehstr, DEHSTR); ++ ++/* ---------------------------------------------------------------------- */ ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_MODULE_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/opts.c linux-2.6.31/fs/aufs/opts.c +--- linux-2.6.31-vanilla/fs/aufs/opts.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/opts.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,1546 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * mount options/flags ++ */ ++ ++#include ++#include ++#include /* a distribution requires */ ++#include ++#include "aufs.h" ++ ++/* ---------------------------------------------------------------------- */ ++ ++enum { ++ Opt_br, ++ Opt_add, Opt_del, Opt_mod, Opt_reorder, Opt_append, Opt_prepend, ++ Opt_idel, Opt_imod, Opt_ireorder, ++ Opt_dirwh, Opt_rdcache, Opt_rdblk, Opt_rdhash, Opt_rendir, ++ Opt_rdblk_def, Opt_rdhash_def, ++ Opt_xino, Opt_zxino, Opt_noxino, ++ Opt_trunc_xino, Opt_trunc_xino_v, Opt_notrunc_xino, ++ Opt_trunc_xino_path, Opt_itrunc_xino, ++ Opt_trunc_xib, Opt_notrunc_xib, ++ Opt_shwh, Opt_noshwh, ++ Opt_plink, Opt_noplink, Opt_list_plink, ++ Opt_udba, ++ /* Opt_lock, Opt_unlock, */ ++ Opt_cmd, Opt_cmd_args, ++ Opt_diropq_a, Opt_diropq_w, ++ Opt_warn_perm, Opt_nowarn_perm, ++ Opt_wbr_copyup, Opt_wbr_create, ++ Opt_refrof, Opt_norefrof, ++ Opt_verbose, Opt_noverbose, ++ Opt_sum, Opt_nosum, Opt_wsum, ++ Opt_tail, Opt_ignore, Opt_ignore_silent, Opt_err ++}; ++ ++static match_table_t options = { ++ {Opt_br, "br=%s"}, ++ {Opt_br, "br:%s"}, ++ ++ {Opt_add, "add=%d:%s"}, ++ {Opt_add, "add:%d:%s"}, ++ {Opt_add, "ins=%d:%s"}, ++ {Opt_add, "ins:%d:%s"}, ++ {Opt_append, "append=%s"}, ++ {Opt_append, "append:%s"}, ++ {Opt_prepend, "prepend=%s"}, ++ {Opt_prepend, "prepend:%s"}, ++ ++ {Opt_del, "del=%s"}, ++ {Opt_del, "del:%s"}, ++ /* {Opt_idel, "idel:%d"}, */ ++ {Opt_mod, "mod=%s"}, ++ {Opt_mod, "mod:%s"}, ++ /* {Opt_imod, "imod:%d:%s"}, */ ++ ++ {Opt_dirwh, "dirwh=%d"}, ++ ++ {Opt_xino, "xino=%s"}, ++ {Opt_noxino, "noxino"}, ++ {Opt_trunc_xino, "trunc_xino"}, ++ {Opt_trunc_xino_v, "trunc_xino_v=%d:%d"}, ++ {Opt_notrunc_xino, "notrunc_xino"}, ++ {Opt_trunc_xino_path, "trunc_xino=%s"}, ++ {Opt_itrunc_xino, "itrunc_xino=%d"}, ++ /* {Opt_zxino, "zxino=%s"}, */ ++ {Opt_trunc_xib, "trunc_xib"}, ++ {Opt_notrunc_xib, "notrunc_xib"}, ++ ++ {Opt_plink, "plink"}, ++ {Opt_noplink, "noplink"}, ++#ifdef CONFIG_AUFS_DEBUG ++ {Opt_list_plink, "list_plink"}, ++#endif ++ ++ {Opt_udba, "udba=%s"}, ++ ++ {Opt_diropq_a, "diropq=always"}, ++ {Opt_diropq_a, "diropq=a"}, ++ {Opt_diropq_w, "diropq=whiteouted"}, ++ {Opt_diropq_w, "diropq=w"}, ++ ++ {Opt_warn_perm, "warn_perm"}, ++ {Opt_nowarn_perm, "nowarn_perm"}, ++ ++ /* keep them temporary */ ++ {Opt_ignore_silent, "coo=%s"}, ++ {Opt_ignore_silent, "nodlgt"}, ++ {Opt_ignore_silent, "nodirperm1"}, ++ {Opt_ignore_silent, "clean_plink"}, ++ ++#ifdef CONFIG_AUFS_SHWH ++ {Opt_shwh, "shwh"}, ++#endif ++ {Opt_noshwh, "noshwh"}, ++ ++ {Opt_rendir, "rendir=%d"}, ++ ++ {Opt_refrof, "refrof"}, ++ {Opt_norefrof, "norefrof"}, ++ ++ {Opt_verbose, "verbose"}, ++ {Opt_verbose, "v"}, ++ {Opt_noverbose, "noverbose"}, ++ {Opt_noverbose, "quiet"}, ++ {Opt_noverbose, "q"}, ++ {Opt_noverbose, "silent"}, ++ ++ {Opt_sum, "sum"}, ++ {Opt_nosum, "nosum"}, ++ {Opt_wsum, "wsum"}, ++ ++ {Opt_rdcache, "rdcache=%d"}, ++ {Opt_rdblk, "rdblk=%d"}, ++ {Opt_rdblk_def, "rdblk=def"}, ++ {Opt_rdhash, "rdhash=%d"}, ++ {Opt_rdhash_def, "rdhash=def"}, ++ ++ {Opt_wbr_create, "create=%s"}, ++ {Opt_wbr_create, "create_policy=%s"}, ++ {Opt_wbr_copyup, "cpup=%s"}, ++ {Opt_wbr_copyup, "copyup=%s"}, ++ {Opt_wbr_copyup, "copyup_policy=%s"}, ++ ++ /* internal use for the scripts */ ++ {Opt_ignore_silent, "si=%s"}, ++ ++ {Opt_br, "dirs=%s"}, ++ {Opt_ignore, "debug=%d"}, ++ {Opt_ignore, "delete=whiteout"}, ++ {Opt_ignore, "delete=all"}, ++ {Opt_ignore, "imap=%s"}, ++ ++ /* temporary workaround, due to old mount(8)? */ ++ {Opt_ignore_silent, "relatime"}, ++ ++ {Opt_err, NULL} ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++static const char *au_parser_pattern(int val, struct match_token *token) ++{ ++ while (token->pattern) { ++ if (token->token == val) ++ return token->pattern; ++ token++; ++ } ++ BUG(); ++ return "??"; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static match_table_t brperms = { ++ {AuBrPerm_RO, AUFS_BRPERM_RO}, ++ {AuBrPerm_RR, AUFS_BRPERM_RR}, ++ {AuBrPerm_RW, AUFS_BRPERM_RW}, ++ ++ {AuBrPerm_ROWH, AUFS_BRPERM_ROWH}, ++ {AuBrPerm_RRWH, AUFS_BRPERM_RRWH}, ++ {AuBrPerm_RWNoLinkWH, AUFS_BRPERM_RWNLWH}, ++ ++ {AuBrPerm_ROWH, "nfsro"}, ++ {AuBrPerm_RO, NULL} ++}; ++ ++static int br_perm_val(char *perm) ++{ ++ int val; ++ substring_t args[MAX_OPT_ARGS]; ++ ++ val = match_token(perm, brperms, args); ++ return val; ++} ++ ++const char *au_optstr_br_perm(int brperm) ++{ ++ return au_parser_pattern(brperm, (void *)brperms); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static match_table_t udbalevel = { ++ {AuOpt_UDBA_REVAL, "reval"}, ++ {AuOpt_UDBA_NONE, "none"}, ++#ifdef CONFIG_AUFS_HINOTIFY ++ {AuOpt_UDBA_HINOTIFY, "inotify"}, ++#endif ++ {-1, NULL} ++}; ++ ++static int udba_val(char *str) ++{ ++ substring_t args[MAX_OPT_ARGS]; ++ ++ return match_token(str, udbalevel, args); ++} ++ ++const char *au_optstr_udba(int udba) ++{ ++ return au_parser_pattern(udba, (void *)udbalevel); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static match_table_t au_wbr_create_policy = { ++ {AuWbrCreate_TDP, "tdp"}, ++ {AuWbrCreate_TDP, "top-down-parent"}, ++ {AuWbrCreate_RR, "rr"}, ++ {AuWbrCreate_RR, "round-robin"}, ++ {AuWbrCreate_MFS, "mfs"}, ++ {AuWbrCreate_MFS, "most-free-space"}, ++ {AuWbrCreate_MFSV, "mfs:%d"}, ++ {AuWbrCreate_MFSV, "most-free-space:%d"}, ++ ++ {AuWbrCreate_MFSRR, "mfsrr:%d"}, ++ {AuWbrCreate_MFSRRV, "mfsrr:%d:%d"}, ++ {AuWbrCreate_PMFS, "pmfs"}, ++ {AuWbrCreate_PMFSV, "pmfs:%d"}, ++ ++ {-1, NULL} ++}; ++ ++/* ++ * cf. linux/lib/parser.c and cmdline.c ++ * gave up calling memparse() since it uses simple_strtoull() instead of ++ * strict_...(). ++ */ ++static int au_match_ull(substring_t *s, unsigned long long *result) ++{ ++ int err; ++ unsigned int len; ++ char a[32]; ++ ++ err = -ERANGE; ++ len = s->to - s->from; ++ if (len + 1 <= sizeof(a)) { ++ memcpy(a, s->from, len); ++ a[len] = '\0'; ++ err = strict_strtoull(a, 0, result); ++ } ++ return err; ++} ++ ++static int au_wbr_mfs_wmark(substring_t *arg, char *str, ++ struct au_opt_wbr_create *create) ++{ ++ int err; ++ unsigned long long ull; ++ ++ err = 0; ++ if (!au_match_ull(arg, &ull)) ++ create->mfsrr_watermark = ull; ++ else { ++ AuErr("bad integer in %s\n", str); ++ err = -EINVAL; ++ } ++ ++ return err; ++} ++ ++static int au_wbr_mfs_sec(substring_t *arg, char *str, ++ struct au_opt_wbr_create *create) ++{ ++ int n, err; ++ ++ err = 0; ++ if (!match_int(arg, &n) && 0 <= n) ++ create->mfs_second = n; ++ else { ++ AuErr("bad integer in %s\n", str); ++ err = -EINVAL; ++ } ++ ++ return err; ++} ++ ++static int au_wbr_create_val(char *str, struct au_opt_wbr_create *create) ++{ ++ int err, e; ++ substring_t args[MAX_OPT_ARGS]; ++ ++ err = match_token(str, au_wbr_create_policy, args); ++ create->wbr_create = err; ++ switch (err) { ++ case AuWbrCreate_MFSRRV: ++ e = au_wbr_mfs_wmark(&args[0], str, create); ++ if (!e) ++ e = au_wbr_mfs_sec(&args[1], str, create); ++ if (unlikely(e)) ++ err = e; ++ break; ++ case AuWbrCreate_MFSRR: ++ e = au_wbr_mfs_wmark(&args[0], str, create); ++ if (unlikely(e)) { ++ err = e; ++ break; ++ } ++ /*FALLTHROUGH*/ ++ case AuWbrCreate_MFS: ++ case AuWbrCreate_PMFS: ++ create->mfs_second = AUFS_MFS_SECOND_DEF; ++ break; ++ case AuWbrCreate_MFSV: ++ case AuWbrCreate_PMFSV: ++ e = au_wbr_mfs_sec(&args[0], str, create); ++ if (unlikely(e)) ++ err = e; ++ break; ++ } ++ ++ return err; ++} ++ ++const char *au_optstr_wbr_create(int wbr_create) ++{ ++ return au_parser_pattern(wbr_create, (void *)au_wbr_create_policy); ++} ++ ++static match_table_t au_wbr_copyup_policy = { ++ {AuWbrCopyup_TDP, "tdp"}, ++ {AuWbrCopyup_TDP, "top-down-parent"}, ++ {AuWbrCopyup_BUP, "bup"}, ++ {AuWbrCopyup_BUP, "bottom-up-parent"}, ++ {AuWbrCopyup_BU, "bu"}, ++ {AuWbrCopyup_BU, "bottom-up"}, ++ {-1, NULL} ++}; ++ ++static int au_wbr_copyup_val(char *str) ++{ ++ substring_t args[MAX_OPT_ARGS]; ++ ++ return match_token(str, au_wbr_copyup_policy, args); ++} ++ ++const char *au_optstr_wbr_copyup(int wbr_copyup) ++{ ++ return au_parser_pattern(wbr_copyup, (void *)au_wbr_copyup_policy); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static const int lkup_dirflags = LOOKUP_FOLLOW | LOOKUP_DIRECTORY; ++ ++static void dump_opts(struct au_opts *opts) ++{ ++#ifdef CONFIG_AUFS_DEBUG ++ /* reduce stack space */ ++ union { ++ struct au_opt_add *add; ++ struct au_opt_del *del; ++ struct au_opt_mod *mod; ++ struct au_opt_xino *xino; ++ struct au_opt_xino_itrunc *xino_itrunc; ++ struct au_opt_wbr_create *create; ++ } u; ++ struct au_opt *opt; ++ ++ opt = opts->opt; ++ while (opt->type != Opt_tail) { ++ switch (opt->type) { ++ case Opt_add: ++ u.add = &opt->add; ++ AuDbg("add {b%d, %s, 0x%x, %p}\n", ++ u.add->bindex, u.add->pathname, u.add->perm, ++ u.add->path.dentry); ++ break; ++ case Opt_del: ++ case Opt_idel: ++ u.del = &opt->del; ++ AuDbg("del {%s, %p}\n", ++ u.del->pathname, u.del->h_path.dentry); ++ break; ++ case Opt_mod: ++ case Opt_imod: ++ u.mod = &opt->mod; ++ AuDbg("mod {%s, 0x%x, %p}\n", ++ u.mod->path, u.mod->perm, u.mod->h_root); ++ break; ++ case Opt_append: ++ u.add = &opt->add; ++ AuDbg("append {b%d, %s, 0x%x, %p}\n", ++ u.add->bindex, u.add->pathname, u.add->perm, ++ u.add->path.dentry); ++ break; ++ case Opt_prepend: ++ u.add = &opt->add; ++ AuDbg("prepend {b%d, %s, 0x%x, %p}\n", ++ u.add->bindex, u.add->pathname, u.add->perm, ++ u.add->path.dentry); ++ break; ++ case Opt_dirwh: ++ AuDbg("dirwh %d\n", opt->dirwh); ++ break; ++ case Opt_rdcache: ++ AuDbg("rdcache %d\n", opt->rdcache); ++ break; ++ case Opt_rdblk: ++ AuDbg("rdblk %u\n", opt->rdblk); ++ break; ++ case Opt_rdblk_def: ++ AuDbg("rdblk_def\n"); ++ break; ++ case Opt_rdhash: ++ AuDbg("rdhash %u\n", opt->rdhash); ++ break; ++ case Opt_rdhash_def: ++ AuDbg("rdhash_def\n"); ++ break; ++ case Opt_xino: ++ u.xino = &opt->xino; ++ AuDbg("xino {%s %.*s}\n", ++ u.xino->path, ++ AuDLNPair(u.xino->file->f_dentry)); ++ break; ++ case Opt_trunc_xino: ++ AuLabel(trunc_xino); ++ break; ++ case Opt_notrunc_xino: ++ AuLabel(notrunc_xino); ++ break; ++ case Opt_trunc_xino_path: ++ case Opt_itrunc_xino: ++ u.xino_itrunc = &opt->xino_itrunc; ++ AuDbg("trunc_xino %d\n", u.xino_itrunc->bindex); ++ break; ++ ++ case Opt_noxino: ++ AuLabel(noxino); ++ break; ++ case Opt_trunc_xib: ++ AuLabel(trunc_xib); ++ break; ++ case Opt_notrunc_xib: ++ AuLabel(notrunc_xib); ++ break; ++ case Opt_shwh: ++ AuLabel(shwh); ++ break; ++ case Opt_noshwh: ++ AuLabel(noshwh); ++ break; ++ case Opt_plink: ++ AuLabel(plink); ++ break; ++ case Opt_noplink: ++ AuLabel(noplink); ++ break; ++ case Opt_list_plink: ++ AuLabel(list_plink); ++ break; ++ case Opt_udba: ++ AuDbg("udba %d, %s\n", ++ opt->udba, au_optstr_udba(opt->udba)); ++ break; ++ case Opt_diropq_a: ++ AuLabel(diropq_a); ++ break; ++ case Opt_diropq_w: ++ AuLabel(diropq_w); ++ break; ++ case Opt_warn_perm: ++ AuLabel(warn_perm); ++ break; ++ case Opt_nowarn_perm: ++ AuLabel(nowarn_perm); ++ break; ++ case Opt_refrof: ++ AuLabel(refrof); ++ break; ++ case Opt_norefrof: ++ AuLabel(norefrof); ++ break; ++ case Opt_verbose: ++ AuLabel(verbose); ++ break; ++ case Opt_noverbose: ++ AuLabel(noverbose); ++ break; ++ case Opt_sum: ++ AuLabel(sum); ++ break; ++ case Opt_nosum: ++ AuLabel(nosum); ++ break; ++ case Opt_wsum: ++ AuLabel(wsum); ++ break; ++ case Opt_wbr_create: ++ u.create = &opt->wbr_create; ++ AuDbg("create %d, %s\n", u.create->wbr_create, ++ au_optstr_wbr_create(u.create->wbr_create)); ++ switch (u.create->wbr_create) { ++ case AuWbrCreate_MFSV: ++ case AuWbrCreate_PMFSV: ++ AuDbg("%d sec\n", u.create->mfs_second); ++ break; ++ case AuWbrCreate_MFSRR: ++ AuDbg("%llu watermark\n", ++ u.create->mfsrr_watermark); ++ break; ++ case AuWbrCreate_MFSRRV: ++ AuDbg("%llu watermark, %d sec\n", ++ u.create->mfsrr_watermark, ++ u.create->mfs_second); ++ break; ++ } ++ break; ++ case Opt_wbr_copyup: ++ AuDbg("copyup %d, %s\n", opt->wbr_copyup, ++ au_optstr_wbr_copyup(opt->wbr_copyup)); ++ break; ++ default: ++ BUG(); ++ } ++ opt++; ++ } ++#endif ++} ++ ++void au_opts_free(struct au_opts *opts) ++{ ++ struct au_opt *opt; ++ ++ opt = opts->opt; ++ while (opt->type != Opt_tail) { ++ switch (opt->type) { ++ case Opt_add: ++ case Opt_append: ++ case Opt_prepend: ++ path_put(&opt->add.path); ++ break; ++ case Opt_del: ++ case Opt_idel: ++ path_put(&opt->del.h_path); ++ break; ++ case Opt_mod: ++ case Opt_imod: ++ dput(opt->mod.h_root); ++ break; ++ case Opt_xino: ++ fput(opt->xino.file); ++ break; ++ } ++ opt++; ++ } ++} ++ ++static int opt_add(struct au_opt *opt, char *opt_str, unsigned long sb_flags, ++ aufs_bindex_t bindex) ++{ ++ int err; ++ struct au_opt_add *add = &opt->add; ++ char *p; ++ ++ add->bindex = bindex; ++ add->perm = AuBrPerm_Last; ++ add->pathname = opt_str; ++ p = strchr(opt_str, '='); ++ if (p) { ++ *p++ = 0; ++ if (*p) ++ add->perm = br_perm_val(p); ++ } ++ ++ err = vfsub_kern_path(add->pathname, lkup_dirflags, &add->path); ++ if (!err) { ++ if (!p) { ++ add->perm = AuBrPerm_RO; ++ if (au_test_fs_rr(add->path.dentry->d_sb)) ++ add->perm = AuBrPerm_RR; ++ else if (!bindex && !(sb_flags & MS_RDONLY)) ++ add->perm = AuBrPerm_RW; ++ } ++ opt->type = Opt_add; ++ goto out; ++ } ++ AuErr("lookup failed %s (%d)\n", add->pathname, err); ++ err = -EINVAL; ++ ++ out: ++ return err; ++} ++ ++static int au_opts_parse_del(struct au_opt_del *del, substring_t args[]) ++{ ++ int err; ++ ++ del->pathname = args[0].from; ++ AuDbg("del path %s\n", del->pathname); ++ ++ err = vfsub_kern_path(del->pathname, lkup_dirflags, &del->h_path); ++ if (unlikely(err)) ++ AuErr("lookup failed %s (%d)\n", del->pathname, err); ++ ++ return err; ++} ++ ++#if 0 /* reserved for future use */ ++static int au_opts_parse_idel(struct super_block *sb, aufs_bindex_t bindex, ++ struct au_opt_del *del, substring_t args[]) ++{ ++ int err; ++ struct dentry *root; ++ ++ err = -EINVAL; ++ root = sb->s_root; ++ aufs_read_lock(root, AuLock_FLUSH); ++ if (bindex < 0 || au_sbend(sb) < bindex) { ++ AuErr("out of bounds, %d\n", bindex); ++ goto out; ++ } ++ ++ err = 0; ++ del->h_path.dentry = dget(au_h_dptr(root, bindex)); ++ del->h_path.mnt = mntget(au_sbr_mnt(sb, bindex)); ++ ++ out: ++ aufs_read_unlock(root, !AuLock_IR); ++ return err; ++} ++#endif ++ ++static int au_opts_parse_mod(struct au_opt_mod *mod, substring_t args[]) ++{ ++ int err; ++ struct path path; ++ char *p; ++ ++ err = -EINVAL; ++ mod->path = args[0].from; ++ p = strchr(mod->path, '='); ++ if (unlikely(!p)) { ++ AuErr("no permssion %s\n", args[0].from); ++ goto out; ++ } ++ ++ *p++ = 0; ++ err = vfsub_kern_path(mod->path, lkup_dirflags, &path); ++ if (unlikely(err)) { ++ AuErr("lookup failed %s (%d)\n", mod->path, err); ++ goto out; ++ } ++ ++ mod->perm = br_perm_val(p); ++ AuDbg("mod path %s, perm 0x%x, %s\n", mod->path, mod->perm, p); ++ mod->h_root = dget(path.dentry); ++ path_put(&path); ++ ++ out: ++ return err; ++} ++ ++#if 0 /* reserved for future use */ ++static int au_opts_parse_imod(struct super_block *sb, aufs_bindex_t bindex, ++ struct au_opt_mod *mod, substring_t args[]) ++{ ++ int err; ++ struct dentry *root; ++ ++ err = -EINVAL; ++ root = sb->s_root; ++ aufs_read_lock(root, AuLock_FLUSH); ++ if (bindex < 0 || au_sbend(sb) < bindex) { ++ AuErr("out of bounds, %d\n", bindex); ++ goto out; ++ } ++ ++ err = 0; ++ mod->perm = br_perm_val(args[1].from); ++ AuDbg("mod path %s, perm 0x%x, %s\n", ++ mod->path, mod->perm, args[1].from); ++ mod->h_root = dget(au_h_dptr(root, bindex)); ++ ++ out: ++ aufs_read_unlock(root, !AuLock_IR); ++ return err; ++} ++#endif ++ ++static int au_opts_parse_xino(struct super_block *sb, struct au_opt_xino *xino, ++ substring_t args[]) ++{ ++ int err; ++ struct file *file; ++ ++ file = au_xino_create(sb, args[0].from, /*silent*/0); ++ err = PTR_ERR(file); ++ if (IS_ERR(file)) ++ goto out; ++ ++ err = -EINVAL; ++ if (unlikely(file->f_dentry->d_sb == sb)) { ++ fput(file); ++ AuErr("%s must be outside\n", args[0].from); ++ goto out; ++ } ++ ++ err = 0; ++ xino->file = file; ++ xino->path = args[0].from; ++ ++ out: ++ return err; ++} ++ ++static ++int au_opts_parse_xino_itrunc_path(struct super_block *sb, ++ struct au_opt_xino_itrunc *xino_itrunc, ++ substring_t args[]) ++{ ++ int err; ++ aufs_bindex_t bend, bindex; ++ struct path path; ++ struct dentry *root; ++ ++ err = vfsub_kern_path(args[0].from, lkup_dirflags, &path); ++ if (unlikely(err)) { ++ AuErr("lookup failed %s (%d)\n", args[0].from, err); ++ goto out; ++ } ++ ++ xino_itrunc->bindex = -1; ++ root = sb->s_root; ++ aufs_read_lock(root, AuLock_FLUSH); ++ bend = au_sbend(sb); ++ for (bindex = 0; bindex <= bend; bindex++) { ++ if (au_h_dptr(root, bindex) == path.dentry) { ++ xino_itrunc->bindex = bindex; ++ break; ++ } ++ } ++ aufs_read_unlock(root, !AuLock_IR); ++ path_put(&path); ++ ++ if (unlikely(xino_itrunc->bindex < 0)) { ++ AuErr("no such branch %s\n", args[0].from); ++ err = -EINVAL; ++ } ++ ++ out: ++ return err; ++} ++ ++/* called without aufs lock */ ++int au_opts_parse(struct super_block *sb, char *str, struct au_opts *opts) ++{ ++ int err, n, token; ++ aufs_bindex_t bindex; ++ unsigned char skipped; ++ struct dentry *root; ++ struct au_opt *opt, *opt_tail; ++ char *opt_str; ++ /* reduce the stack space */ ++ union { ++ struct au_opt_xino_itrunc *xino_itrunc; ++ struct au_opt_wbr_create *create; ++ } u; ++ struct { ++ substring_t args[MAX_OPT_ARGS]; ++ } *a; ++ ++ err = -ENOMEM; ++ a = kmalloc(sizeof(*a), GFP_NOFS); ++ if (unlikely(!a)) ++ goto out; ++ ++ root = sb->s_root; ++ err = 0; ++ bindex = 0; ++ opt = opts->opt; ++ opt_tail = opt + opts->max_opt - 1; ++ opt->type = Opt_tail; ++ while (!err && (opt_str = strsep(&str, ",")) && *opt_str) { ++ err = -EINVAL; ++ skipped = 0; ++ token = match_token(opt_str, options, a->args); ++ switch (token) { ++ case Opt_br: ++ err = 0; ++ while (!err && (opt_str = strsep(&a->args[0].from, ":")) ++ && *opt_str) { ++ err = opt_add(opt, opt_str, opts->sb_flags, ++ bindex++); ++ if (unlikely(!err && ++opt > opt_tail)) { ++ err = -E2BIG; ++ break; ++ } ++ opt->type = Opt_tail; ++ skipped = 1; ++ } ++ break; ++ case Opt_add: ++ if (unlikely(match_int(&a->args[0], &n))) { ++ AuErr("bad integer in %s\n", opt_str); ++ break; ++ } ++ bindex = n; ++ err = opt_add(opt, a->args[1].from, opts->sb_flags, ++ bindex); ++ if (!err) ++ opt->type = token; ++ break; ++ case Opt_append: ++ err = opt_add(opt, a->args[0].from, opts->sb_flags, ++ /*dummy bindex*/1); ++ if (!err) ++ opt->type = token; ++ break; ++ case Opt_prepend: ++ err = opt_add(opt, a->args[0].from, opts->sb_flags, ++ /*bindex*/0); ++ if (!err) ++ opt->type = token; ++ break; ++ case Opt_del: ++ err = au_opts_parse_del(&opt->del, a->args); ++ if (!err) ++ opt->type = token; ++ break; ++#if 0 /* reserved for future use */ ++ case Opt_idel: ++ del->pathname = "(indexed)"; ++ if (unlikely(match_int(&args[0], &n))) { ++ AuErr("bad integer in %s\n", opt_str); ++ break; ++ } ++ err = au_opts_parse_idel(sb, n, &opt->del, a->args); ++ if (!err) ++ opt->type = token; ++ break; ++#endif ++ case Opt_mod: ++ err = au_opts_parse_mod(&opt->mod, a->args); ++ if (!err) ++ opt->type = token; ++ break; ++#ifdef IMOD /* reserved for future use */ ++ case Opt_imod: ++ u.mod->path = "(indexed)"; ++ if (unlikely(match_int(&a->args[0], &n))) { ++ AuErr("bad integer in %s\n", opt_str); ++ break; ++ } ++ err = au_opts_parse_imod(sb, n, &opt->mod, a->args); ++ if (!err) ++ opt->type = token; ++ break; ++#endif ++ case Opt_xino: ++ err = au_opts_parse_xino(sb, &opt->xino, a->args); ++ if (!err) ++ opt->type = token; ++ break; ++ ++ case Opt_trunc_xino_path: ++ err = au_opts_parse_xino_itrunc_path ++ (sb, &opt->xino_itrunc, a->args); ++ if (!err) ++ opt->type = token; ++ break; ++ ++ case Opt_itrunc_xino: ++ u.xino_itrunc = &opt->xino_itrunc; ++ if (unlikely(match_int(&a->args[0], &n))) { ++ AuErr("bad integer in %s\n", opt_str); ++ break; ++ } ++ u.xino_itrunc->bindex = n; ++ aufs_read_lock(root, AuLock_FLUSH); ++ if (n < 0 || au_sbend(sb) < n) { ++ AuErr("out of bounds, %d\n", n); ++ aufs_read_unlock(root, !AuLock_IR); ++ break; ++ } ++ aufs_read_unlock(root, !AuLock_IR); ++ err = 0; ++ opt->type = token; ++ break; ++ ++ case Opt_dirwh: ++ if (unlikely(match_int(&a->args[0], &opt->dirwh))) ++ break; ++ err = 0; ++ opt->type = token; ++ break; ++ ++ case Opt_rdcache: ++ if (unlikely(match_int(&a->args[0], &opt->rdcache))) ++ break; ++ err = 0; ++ opt->type = token; ++ break; ++ case Opt_rdblk: ++ if (unlikely(match_int(&a->args[0], &n) ++ || n < 0 ++ || n > KMALLOC_MAX_SIZE)) { ++ AuErr("bad integer in %s\n", opt_str); ++ break; ++ } ++ if (unlikely(n && n < NAME_MAX)) { ++ AuErr("rdblk must be larger than %d\n", ++ NAME_MAX); ++ break; ++ } ++ opt->rdblk = n; ++ err = 0; ++ opt->type = token; ++ break; ++ case Opt_rdhash: ++ if (unlikely(match_int(&a->args[0], &n) ++ || n < 0 ++ || n * sizeof(struct hlist_head) ++ > KMALLOC_MAX_SIZE)) { ++ AuErr("bad integer in %s\n", opt_str); ++ break; ++ } ++ opt->rdhash = n; ++ err = 0; ++ opt->type = token; ++ break; ++ ++ case Opt_trunc_xino: ++ case Opt_notrunc_xino: ++ case Opt_noxino: ++ case Opt_trunc_xib: ++ case Opt_notrunc_xib: ++ case Opt_shwh: ++ case Opt_noshwh: ++ case Opt_plink: ++ case Opt_noplink: ++ case Opt_list_plink: ++ case Opt_diropq_a: ++ case Opt_diropq_w: ++ case Opt_warn_perm: ++ case Opt_nowarn_perm: ++ case Opt_refrof: ++ case Opt_norefrof: ++ case Opt_verbose: ++ case Opt_noverbose: ++ case Opt_sum: ++ case Opt_nosum: ++ case Opt_wsum: ++ case Opt_rdblk_def: ++ case Opt_rdhash_def: ++ err = 0; ++ opt->type = token; ++ break; ++ ++ case Opt_udba: ++ opt->udba = udba_val(a->args[0].from); ++ if (opt->udba >= 0) { ++ err = 0; ++ opt->type = token; ++ } else ++ AuErr("wrong value, %s\n", opt_str); ++ break; ++ ++ case Opt_wbr_create: ++ u.create = &opt->wbr_create; ++ u.create->wbr_create ++ = au_wbr_create_val(a->args[0].from, u.create); ++ if (u.create->wbr_create >= 0) { ++ err = 0; ++ opt->type = token; ++ } else ++ AuErr("wrong value, %s\n", opt_str); ++ break; ++ case Opt_wbr_copyup: ++ opt->wbr_copyup = au_wbr_copyup_val(a->args[0].from); ++ if (opt->wbr_copyup >= 0) { ++ err = 0; ++ opt->type = token; ++ } else ++ AuErr("wrong value, %s\n", opt_str); ++ break; ++ ++ case Opt_ignore: ++ AuWarn("ignored %s\n", opt_str); ++ /*FALLTHROUGH*/ ++ case Opt_ignore_silent: ++ skipped = 1; ++ err = 0; ++ break; ++ case Opt_err: ++ AuErr("unknown option %s\n", opt_str); ++ break; ++ } ++ ++ if (!err && !skipped) { ++ if (unlikely(++opt > opt_tail)) { ++ err = -E2BIG; ++ opt--; ++ opt->type = Opt_tail; ++ break; ++ } ++ opt->type = Opt_tail; ++ } ++ } ++ ++ kfree(a); ++ dump_opts(opts); ++ if (unlikely(err)) ++ au_opts_free(opts); ++ ++ out: ++ return err; ++} ++ ++static int au_opt_wbr_create(struct super_block *sb, ++ struct au_opt_wbr_create *create) ++{ ++ int err; ++ struct au_sbinfo *sbinfo; ++ ++ SiMustWriteLock(sb); ++ ++ err = 1; /* handled */ ++ sbinfo = au_sbi(sb); ++ if (sbinfo->si_wbr_create_ops->fin) { ++ err = sbinfo->si_wbr_create_ops->fin(sb); ++ if (!err) ++ err = 1; ++ } ++ ++ sbinfo->si_wbr_create = create->wbr_create; ++ sbinfo->si_wbr_create_ops = au_wbr_create_ops + create->wbr_create; ++ switch (create->wbr_create) { ++ case AuWbrCreate_MFSRRV: ++ case AuWbrCreate_MFSRR: ++ sbinfo->si_wbr_mfs.mfsrr_watermark = create->mfsrr_watermark; ++ /*FALLTHROUGH*/ ++ case AuWbrCreate_MFS: ++ case AuWbrCreate_MFSV: ++ case AuWbrCreate_PMFS: ++ case AuWbrCreate_PMFSV: ++ sbinfo->si_wbr_mfs.mfs_expire = create->mfs_second * HZ; ++ break; ++ } ++ ++ if (sbinfo->si_wbr_create_ops->init) ++ sbinfo->si_wbr_create_ops->init(sb); /* ignore */ ++ ++ return err; ++} ++ ++/* ++ * returns, ++ * plus: processed without an error ++ * zero: unprocessed ++ */ ++static int au_opt_simple(struct super_block *sb, struct au_opt *opt, ++ struct au_opts *opts) ++{ ++ int err; ++ struct au_sbinfo *sbinfo; ++ ++ SiMustWriteLock(sb); ++ ++ err = 1; /* handled */ ++ sbinfo = au_sbi(sb); ++ switch (opt->type) { ++ case Opt_udba: ++ sbinfo->si_mntflags &= ~AuOptMask_UDBA; ++ sbinfo->si_mntflags |= opt->udba; ++ opts->given_udba |= opt->udba; ++ break; ++ ++ case Opt_plink: ++ au_opt_set(sbinfo->si_mntflags, PLINK); ++ break; ++ case Opt_noplink: ++ if (au_opt_test(sbinfo->si_mntflags, PLINK)) ++ au_plink_put(sb); ++ au_opt_clr(sbinfo->si_mntflags, PLINK); ++ break; ++ case Opt_list_plink: ++ if (au_opt_test(sbinfo->si_mntflags, PLINK)) ++ au_plink_list(sb); ++ break; ++ ++ case Opt_diropq_a: ++ au_opt_set(sbinfo->si_mntflags, ALWAYS_DIROPQ); ++ break; ++ case Opt_diropq_w: ++ au_opt_clr(sbinfo->si_mntflags, ALWAYS_DIROPQ); ++ break; ++ ++ case Opt_warn_perm: ++ au_opt_set(sbinfo->si_mntflags, WARN_PERM); ++ break; ++ case Opt_nowarn_perm: ++ au_opt_clr(sbinfo->si_mntflags, WARN_PERM); ++ break; ++ ++ case Opt_refrof: ++ au_opt_set(sbinfo->si_mntflags, REFROF); ++ break; ++ case Opt_norefrof: ++ au_opt_clr(sbinfo->si_mntflags, REFROF); ++ break; ++ ++ case Opt_verbose: ++ au_opt_set(sbinfo->si_mntflags, VERBOSE); ++ break; ++ case Opt_noverbose: ++ au_opt_clr(sbinfo->si_mntflags, VERBOSE); ++ break; ++ ++ case Opt_sum: ++ au_opt_set(sbinfo->si_mntflags, SUM); ++ break; ++ case Opt_wsum: ++ au_opt_clr(sbinfo->si_mntflags, SUM); ++ au_opt_set(sbinfo->si_mntflags, SUM_W); ++ case Opt_nosum: ++ au_opt_clr(sbinfo->si_mntflags, SUM); ++ au_opt_clr(sbinfo->si_mntflags, SUM_W); ++ break; ++ ++ case Opt_wbr_create: ++ err = au_opt_wbr_create(sb, &opt->wbr_create); ++ break; ++ case Opt_wbr_copyup: ++ sbinfo->si_wbr_copyup = opt->wbr_copyup; ++ sbinfo->si_wbr_copyup_ops = au_wbr_copyup_ops + opt->wbr_copyup; ++ break; ++ ++ case Opt_dirwh: ++ sbinfo->si_dirwh = opt->dirwh; ++ break; ++ ++ case Opt_rdcache: ++ sbinfo->si_rdcache = opt->rdcache * HZ; ++ break; ++ case Opt_rdblk: ++ sbinfo->si_rdblk = opt->rdblk; ++ break; ++ case Opt_rdblk_def: ++ sbinfo->si_rdblk = AUFS_RDBLK_DEF; ++ break; ++ case Opt_rdhash: ++ sbinfo->si_rdhash = opt->rdhash; ++ break; ++ case Opt_rdhash_def: ++ sbinfo->si_rdhash = AUFS_RDHASH_DEF; ++ break; ++ ++ case Opt_shwh: ++ au_opt_set(sbinfo->si_mntflags, SHWH); ++ break; ++ case Opt_noshwh: ++ au_opt_clr(sbinfo->si_mntflags, SHWH); ++ break; ++ ++ case Opt_trunc_xino: ++ au_opt_set(sbinfo->si_mntflags, TRUNC_XINO); ++ break; ++ case Opt_notrunc_xino: ++ au_opt_clr(sbinfo->si_mntflags, TRUNC_XINO); ++ break; ++ ++ case Opt_trunc_xino_path: ++ case Opt_itrunc_xino: ++ err = au_xino_trunc(sb, opt->xino_itrunc.bindex); ++ if (!err) ++ err = 1; ++ break; ++ ++ case Opt_trunc_xib: ++ au_fset_opts(opts->flags, TRUNC_XIB); ++ break; ++ case Opt_notrunc_xib: ++ au_fclr_opts(opts->flags, TRUNC_XIB); ++ break; ++ ++ default: ++ err = 0; ++ break; ++ } ++ ++ return err; ++} ++ ++/* ++ * returns tri-state. ++ * plus: processed without an error ++ * zero: unprocessed ++ * minus: error ++ */ ++static int au_opt_br(struct super_block *sb, struct au_opt *opt, ++ struct au_opts *opts) ++{ ++ int err, do_refresh; ++ ++ err = 0; ++ switch (opt->type) { ++ case Opt_append: ++ opt->add.bindex = au_sbend(sb) + 1; ++ if (opt->add.bindex < 0) ++ opt->add.bindex = 0; ++ goto add; ++ case Opt_prepend: ++ opt->add.bindex = 0; ++ add: ++ case Opt_add: ++ err = au_br_add(sb, &opt->add, ++ au_ftest_opts(opts->flags, REMOUNT)); ++ if (!err) { ++ err = 1; ++ au_fset_opts(opts->flags, REFRESH_DIR); ++ if (au_br_whable(opt->add.perm)) ++ au_fset_opts(opts->flags, REFRESH_NONDIR); ++ } ++ break; ++ ++ case Opt_del: ++ case Opt_idel: ++ err = au_br_del(sb, &opt->del, ++ au_ftest_opts(opts->flags, REMOUNT)); ++ if (!err) { ++ err = 1; ++ au_fset_opts(opts->flags, TRUNC_XIB); ++ au_fset_opts(opts->flags, REFRESH_DIR); ++ au_fset_opts(opts->flags, REFRESH_NONDIR); ++ } ++ break; ++ ++ case Opt_mod: ++ case Opt_imod: ++ err = au_br_mod(sb, &opt->mod, ++ au_ftest_opts(opts->flags, REMOUNT), ++ &do_refresh); ++ if (!err) { ++ err = 1; ++ if (do_refresh) { ++ au_fset_opts(opts->flags, REFRESH_DIR); ++ au_fset_opts(opts->flags, REFRESH_NONDIR); ++ } ++ } ++ break; ++ } ++ ++ return err; ++} ++ ++static int au_opt_xino(struct super_block *sb, struct au_opt *opt, ++ struct au_opt_xino **opt_xino, ++ struct au_opts *opts) ++{ ++ int err; ++ aufs_bindex_t bend, bindex; ++ struct dentry *root, *parent, *h_root; ++ ++ err = 0; ++ switch (opt->type) { ++ case Opt_xino: ++ err = au_xino_set(sb, &opt->xino, ++ !!au_ftest_opts(opts->flags, REMOUNT)); ++ if (unlikely(err)) ++ break; ++ ++ *opt_xino = &opt->xino; ++ au_xino_brid_set(sb, -1); ++ ++ /* safe d_parent access */ ++ parent = opt->xino.file->f_dentry->d_parent; ++ root = sb->s_root; ++ bend = au_sbend(sb); ++ for (bindex = 0; bindex <= bend; bindex++) { ++ h_root = au_h_dptr(root, bindex); ++ if (h_root == parent) { ++ au_xino_brid_set(sb, au_sbr_id(sb, bindex)); ++ break; ++ } ++ } ++ break; ++ ++ case Opt_noxino: ++ au_xino_clr(sb); ++ au_xino_brid_set(sb, -1); ++ *opt_xino = (void *)-1; ++ break; ++ } ++ ++ return err; ++} ++ ++int au_opts_verify(struct super_block *sb, unsigned long sb_flags, ++ unsigned int pending) ++{ ++ int err; ++ aufs_bindex_t bindex, bend; ++ unsigned char do_plink, skip, do_free; ++ struct au_branch *br; ++ struct au_wbr *wbr; ++ struct dentry *root; ++ struct inode *dir, *h_dir; ++ struct au_sbinfo *sbinfo; ++ struct au_hinode *hdir; ++ ++ SiMustAnyLock(sb); ++ ++ sbinfo = au_sbi(sb); ++ AuDebugOn(!(sbinfo->si_mntflags & AuOptMask_UDBA)); ++ ++ if (!(sb_flags & MS_RDONLY)) { ++ if (unlikely(!au_br_writable(au_sbr_perm(sb, 0)))) ++ AuWarn("first branch should be rw\n"); ++ if (unlikely(au_opt_test(sbinfo->si_mntflags, SHWH))) ++ AuWarn("shwh should be used with ro\n"); ++ } ++ ++ if (au_opt_test((sbinfo->si_mntflags | pending), UDBA_HINOTIFY) ++ && !au_opt_test(sbinfo->si_mntflags, XINO)) ++ AuWarn("udba=inotify requires xino\n"); ++ ++ err = 0; ++ root = sb->s_root; ++ dir = sb->s_root->d_inode; ++ do_plink = !!au_opt_test(sbinfo->si_mntflags, PLINK); ++ bend = au_sbend(sb); ++ for (bindex = 0; !err && bindex <= bend; bindex++) { ++ skip = 0; ++ h_dir = au_h_iptr(dir, bindex); ++ br = au_sbr(sb, bindex); ++ do_free = 0; ++ ++ wbr = br->br_wbr; ++ if (wbr) ++ wbr_wh_read_lock(wbr); ++ ++ switch (br->br_perm) { ++ case AuBrPerm_RO: ++ case AuBrPerm_ROWH: ++ case AuBrPerm_RR: ++ case AuBrPerm_RRWH: ++ do_free = !!wbr; ++ skip = (!wbr ++ || (!wbr->wbr_whbase ++ && !wbr->wbr_plink ++ && !wbr->wbr_orph)); ++ break; ++ ++ case AuBrPerm_RWNoLinkWH: ++ /* skip = (!br->br_whbase && !br->br_orph); */ ++ skip = (!wbr || !wbr->wbr_whbase); ++ if (skip && wbr) { ++ if (do_plink) ++ skip = !!wbr->wbr_plink; ++ else ++ skip = !wbr->wbr_plink; ++ } ++ break; ++ ++ case AuBrPerm_RW: ++ /* skip = (br->br_whbase && br->br_ohph); */ ++ skip = (wbr && wbr->wbr_whbase); ++ if (skip) { ++ if (do_plink) ++ skip = !!wbr->wbr_plink; ++ else ++ skip = !wbr->wbr_plink; ++ } ++ break; ++ ++ default: ++ BUG(); ++ } ++ if (wbr) ++ wbr_wh_read_unlock(wbr); ++ ++ if (skip) ++ continue; ++ ++ hdir = au_hi(dir, bindex); ++ au_hin_imtx_lock_nested(hdir, AuLsc_I_PARENT); ++ if (wbr) ++ wbr_wh_write_lock(wbr); ++ err = au_wh_init(au_h_dptr(root, bindex), br, sb); ++ if (wbr) ++ wbr_wh_write_unlock(wbr); ++ au_hin_imtx_unlock(hdir); ++ ++ if (!err && do_free) { ++ kfree(wbr); ++ br->br_wbr = NULL; ++ } ++ } ++ ++ return err; ++} ++ ++int au_opts_mount(struct super_block *sb, struct au_opts *opts) ++{ ++ int err; ++ unsigned int tmp; ++ aufs_bindex_t bend; ++ struct au_opt *opt; ++ struct au_opt_xino *opt_xino, xino; ++ struct au_sbinfo *sbinfo; ++ ++ SiMustWriteLock(sb); ++ ++ err = 0; ++ opt_xino = NULL; ++ opt = opts->opt; ++ while (err >= 0 && opt->type != Opt_tail) ++ err = au_opt_simple(sb, opt++, opts); ++ if (err > 0) ++ err = 0; ++ else if (unlikely(err < 0)) ++ goto out; ++ ++ /* disable xino and udba temporary */ ++ sbinfo = au_sbi(sb); ++ tmp = sbinfo->si_mntflags; ++ au_opt_clr(sbinfo->si_mntflags, XINO); ++ au_opt_set_udba(sbinfo->si_mntflags, UDBA_REVAL); ++ ++ opt = opts->opt; ++ while (err >= 0 && opt->type != Opt_tail) ++ err = au_opt_br(sb, opt++, opts); ++ if (err > 0) ++ err = 0; ++ else if (unlikely(err < 0)) ++ goto out; ++ ++ bend = au_sbend(sb); ++ if (unlikely(bend < 0)) { ++ err = -EINVAL; ++ AuErr("no branches\n"); ++ goto out; ++ } ++ ++ if (au_opt_test(tmp, XINO)) ++ au_opt_set(sbinfo->si_mntflags, XINO); ++ opt = opts->opt; ++ while (!err && opt->type != Opt_tail) ++ err = au_opt_xino(sb, opt++, &opt_xino, opts); ++ if (unlikely(err)) ++ goto out; ++ ++ err = au_opts_verify(sb, sb->s_flags, tmp); ++ if (unlikely(err)) ++ goto out; ++ ++ /* restore xino */ ++ if (au_opt_test(tmp, XINO) && !opt_xino) { ++ xino.file = au_xino_def(sb); ++ err = PTR_ERR(xino.file); ++ if (IS_ERR(xino.file)) ++ goto out; ++ ++ err = au_xino_set(sb, &xino, /*remount*/0); ++ fput(xino.file); ++ if (unlikely(err)) ++ goto out; ++ } ++ ++ /* restore udba */ ++ sbinfo->si_mntflags &= ~AuOptMask_UDBA; ++ sbinfo->si_mntflags |= (tmp & AuOptMask_UDBA); ++ if (au_opt_test(tmp, UDBA_HINOTIFY)) { ++ struct inode *dir = sb->s_root->d_inode; ++ au_reset_hinotify(dir, ++ au_hi_flags(dir, /*isdir*/1) & ~AuHi_XINO); ++ } ++ ++ out: ++ return err; ++} ++ ++int au_opts_remount(struct super_block *sb, struct au_opts *opts) ++{ ++ int err, rerr; ++ struct inode *dir; ++ struct au_opt_xino *opt_xino; ++ struct au_opt *opt; ++ struct au_sbinfo *sbinfo; ++ ++ SiMustWriteLock(sb); ++ ++ dir = sb->s_root->d_inode; ++ sbinfo = au_sbi(sb); ++ err = 0; ++ opt_xino = NULL; ++ opt = opts->opt; ++ while (err >= 0 && opt->type != Opt_tail) { ++ err = au_opt_simple(sb, opt, opts); ++ if (!err) ++ err = au_opt_br(sb, opt, opts); ++ if (!err) ++ err = au_opt_xino(sb, opt, &opt_xino, opts); ++ opt++; ++ } ++ if (err > 0) ++ err = 0; ++ AuTraceErr(err); ++ /* go on even err */ ++ ++ rerr = au_opts_verify(sb, opts->sb_flags, /*pending*/0); ++ if (unlikely(rerr && !err)) ++ err = rerr; ++ ++ if (au_ftest_opts(opts->flags, TRUNC_XIB)) { ++ rerr = au_xib_trunc(sb); ++ if (unlikely(rerr && !err)) ++ err = rerr; ++ } ++ ++ /* will be handled by the caller */ ++ if (!au_ftest_opts(opts->flags, REFRESH_DIR) ++ && (opts->given_udba || au_opt_test(sbinfo->si_mntflags, XINO))) ++ au_fset_opts(opts->flags, REFRESH_DIR); ++ ++ AuDbg("status 0x%x\n", opts->flags); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++unsigned int au_opt_udba(struct super_block *sb) ++{ ++ return au_mntflags(sb) & AuOptMask_UDBA; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/opts.h linux-2.6.31/fs/aufs/opts.h +--- linux-2.6.31-vanilla/fs/aufs/opts.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/opts.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,196 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * mount options/flags ++ */ ++ ++#ifndef __AUFS_OPTS_H__ ++#define __AUFS_OPTS_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++ ++struct file; ++struct super_block; ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* mount flags */ ++#define AuOpt_XINO 1 /* external inode number bitmap ++ and translation table */ ++#define AuOpt_TRUNC_XINO (1 << 1) /* truncate xino files */ ++#define AuOpt_UDBA_NONE (1 << 2) /* users direct branch access */ ++#define AuOpt_UDBA_REVAL (1 << 3) ++#define AuOpt_UDBA_HINOTIFY (1 << 4) ++#define AuOpt_SHWH (1 << 5) /* show whiteout */ ++#define AuOpt_PLINK (1 << 6) /* pseudo-link */ ++#define AuOpt_DIRPERM1 (1 << 7) /* unimplemented */ ++#define AuOpt_REFROF (1 << 8) /* unimplemented */ ++#define AuOpt_ALWAYS_DIROPQ (1 << 9) /* policy to creating diropq */ ++#define AuOpt_SUM (1 << 10) /* summation for statfs(2) */ ++#define AuOpt_SUM_W (1 << 11) /* unimplemented */ ++#define AuOpt_WARN_PERM (1 << 12) /* warn when add-branch */ ++#define AuOpt_VERBOSE (1 << 13) /* busy inode when del-branch */ ++ ++#ifndef CONFIG_AUFS_HINOTIFY ++#undef AuOpt_UDBA_HINOTIFY ++#define AuOpt_UDBA_HINOTIFY 0 ++#endif ++#ifndef CONFIG_AUFS_SHWH ++#undef AuOpt_SHWH ++#define AuOpt_SHWH 0 ++#endif ++ ++#define AuOpt_Def (AuOpt_XINO \ ++ | AuOpt_UDBA_REVAL \ ++ | AuOpt_PLINK \ ++ /* | AuOpt_DIRPERM1 */ \ ++ | AuOpt_WARN_PERM) ++#define AuOptMask_UDBA (AuOpt_UDBA_NONE \ ++ | AuOpt_UDBA_REVAL \ ++ | AuOpt_UDBA_HINOTIFY) ++ ++#define au_opt_test(flags, name) (flags & AuOpt_##name) ++#define au_opt_set(flags, name) do { \ ++ BUILD_BUG_ON(AuOpt_##name & AuOptMask_UDBA); \ ++ ((flags) |= AuOpt_##name); \ ++} while (0) ++#define au_opt_set_udba(flags, name) do { \ ++ (flags) &= ~AuOptMask_UDBA; \ ++ ((flags) |= AuOpt_##name); \ ++} while (0) ++#define au_opt_clr(flags, name) { ((flags) &= ~AuOpt_##name); } ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* policies to select one among multiple writable branches */ ++enum { ++ AuWbrCreate_TDP, /* top down parent */ ++ AuWbrCreate_RR, /* round robin */ ++ AuWbrCreate_MFS, /* most free space */ ++ AuWbrCreate_MFSV, /* mfs with seconds */ ++ AuWbrCreate_MFSRR, /* mfs then rr */ ++ AuWbrCreate_MFSRRV, /* mfs then rr with seconds */ ++ AuWbrCreate_PMFS, /* parent and mfs */ ++ AuWbrCreate_PMFSV, /* parent and mfs with seconds */ ++ ++ AuWbrCreate_Def = AuWbrCreate_TDP ++}; ++ ++enum { ++ AuWbrCopyup_TDP, /* top down parent */ ++ AuWbrCopyup_BUP, /* bottom up parent */ ++ AuWbrCopyup_BU, /* bottom up */ ++ ++ AuWbrCopyup_Def = AuWbrCopyup_TDP ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct au_opt_add { ++ aufs_bindex_t bindex; ++ char *pathname; ++ int perm; ++ struct path path; ++}; ++ ++struct au_opt_del { ++ char *pathname; ++ struct path h_path; ++}; ++ ++struct au_opt_mod { ++ char *path; ++ int perm; ++ struct dentry *h_root; ++}; ++ ++struct au_opt_xino { ++ char *path; ++ struct file *file; ++}; ++ ++struct au_opt_xino_itrunc { ++ aufs_bindex_t bindex; ++}; ++ ++struct au_opt_wbr_create { ++ int wbr_create; ++ int mfs_second; ++ unsigned long long mfsrr_watermark; ++}; ++ ++struct au_opt { ++ int type; ++ union { ++ struct au_opt_xino xino; ++ struct au_opt_xino_itrunc xino_itrunc; ++ struct au_opt_add add; ++ struct au_opt_del del; ++ struct au_opt_mod mod; ++ int dirwh; ++ int rdcache; ++ unsigned int rdblk; ++ unsigned int rdhash; ++ int udba; ++ struct au_opt_wbr_create wbr_create; ++ int wbr_copyup; ++ }; ++}; ++ ++/* opts flags */ ++#define AuOpts_REMOUNT 1 ++#define AuOpts_REFRESH_DIR (1 << 1) ++#define AuOpts_REFRESH_NONDIR (1 << 2) ++#define AuOpts_TRUNC_XIB (1 << 3) ++#define au_ftest_opts(flags, name) ((flags) & AuOpts_##name) ++#define au_fset_opts(flags, name) { (flags) |= AuOpts_##name; } ++#define au_fclr_opts(flags, name) { (flags) &= ~AuOpts_##name; } ++ ++struct au_opts { ++ struct au_opt *opt; ++ int max_opt; ++ ++ unsigned int given_udba; ++ unsigned int flags; ++ unsigned long sb_flags; ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++const char *au_optstr_br_perm(int brperm); ++const char *au_optstr_udba(int udba); ++const char *au_optstr_wbr_copyup(int wbr_copyup); ++const char *au_optstr_wbr_create(int wbr_create); ++ ++void au_opts_free(struct au_opts *opts); ++int au_opts_parse(struct super_block *sb, char *str, struct au_opts *opts); ++int au_opts_verify(struct super_block *sb, unsigned long sb_flags, ++ unsigned int pending); ++int au_opts_mount(struct super_block *sb, struct au_opts *opts); ++int au_opts_remount(struct super_block *sb, struct au_opts *opts); ++ ++unsigned int au_opt_udba(struct super_block *sb); ++ ++/* ---------------------------------------------------------------------- */ ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_OPTS_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/plink.c linux-2.6.31/fs/aufs/plink.c +--- linux-2.6.31-vanilla/fs/aufs/plink.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/plink.c 2009-09-16 13:55:29.000000000 +0200 +@@ -0,0 +1,396 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * pseudo-link ++ */ ++ ++#include "aufs.h" ++ ++/* ++ * during a user process maintains the pseudo-links, ++ * prohibit adding a new plink and branch manipulation. ++ */ ++void au_plink_block_maintain(struct super_block *sb) ++{ ++ struct au_sbinfo *sbi = au_sbi(sb); ++ ++ SiMustAnyLock(sb); ++ ++ /* gave up wake_up_bit() */ ++ wait_event(sbi->si_plink_wq, !au_ftest_si(sbi, MAINTAIN_PLINK)); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct pseudo_link { ++ struct list_head list; ++ struct inode *inode; ++}; ++ ++#ifdef CONFIG_AUFS_DEBUG ++void au_plink_list(struct super_block *sb) ++{ ++ struct au_sbinfo *sbinfo; ++ struct list_head *plink_list; ++ struct pseudo_link *plink; ++ ++ SiMustAnyLock(sb); ++ ++ sbinfo = au_sbi(sb); ++ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK)); ++ ++ plink_list = &sbinfo->si_plink.head; ++ spin_lock(&sbinfo->si_plink.spin); ++ list_for_each_entry(plink, plink_list, list) ++ AuDbg("%lu\n", plink->inode->i_ino); ++ spin_unlock(&sbinfo->si_plink.spin); ++} ++#endif ++ ++/* is the inode pseudo-linked? */ ++int au_plink_test(struct inode *inode) ++{ ++ int found; ++ struct au_sbinfo *sbinfo; ++ struct list_head *plink_list; ++ struct pseudo_link *plink; ++ ++ sbinfo = au_sbi(inode->i_sb); ++ AuRwMustAnyLock(&sbinfo->si_rwsem); ++ AuDebugOn(!au_opt_test(au_mntflags(inode->i_sb), PLINK)); ++ ++ found = 0; ++ plink_list = &sbinfo->si_plink.head; ++ spin_lock(&sbinfo->si_plink.spin); ++ list_for_each_entry(plink, plink_list, list) ++ if (plink->inode == inode) { ++ found = 1; ++ break; ++ } ++ spin_unlock(&sbinfo->si_plink.spin); ++ return found; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * generate a name for plink. ++ * the file will be stored under AUFS_WH_PLINKDIR. ++ */ ++/* 20 is max digits length of ulong 64 */ ++#define PLINK_NAME_LEN ((20 + 1) * 2) ++ ++static int plink_name(char *name, int len, struct inode *inode, ++ aufs_bindex_t bindex) ++{ ++ int rlen; ++ struct inode *h_inode; ++ ++ h_inode = au_h_iptr(inode, bindex); ++ rlen = snprintf(name, len, "%lu.%lu", inode->i_ino, h_inode->i_ino); ++ return rlen; ++} ++ ++/* lookup the plink-ed @inode under the branch at @bindex */ ++struct dentry *au_plink_lkup(struct inode *inode, aufs_bindex_t bindex) ++{ ++ struct dentry *h_dentry, *h_parent; ++ struct au_branch *br; ++ struct inode *h_dir; ++ char a[PLINK_NAME_LEN]; ++ struct qstr tgtname = { ++ .name = a ++ }; ++ ++ br = au_sbr(inode->i_sb, bindex); ++ h_parent = br->br_wbr->wbr_plink; ++ h_dir = h_parent->d_inode; ++ tgtname.len = plink_name(a, sizeof(a), inode, bindex); ++ ++ /* always superio. */ ++ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_CHILD2); ++ h_dentry = au_sio_lkup_one(&tgtname, h_parent, br); ++ mutex_unlock(&h_dir->i_mutex); ++ return h_dentry; ++} ++ ++/* create a pseudo-link */ ++static int do_whplink(struct qstr *tgt, struct dentry *h_parent, ++ struct dentry *h_dentry, struct au_branch *br) ++{ ++ int err; ++ struct path h_path = { ++ .mnt = br->br_mnt ++ }; ++ struct inode *h_dir; ++ ++ h_dir = h_parent->d_inode; ++ again: ++ h_path.dentry = au_lkup_one(tgt, h_parent, br, /*nd*/NULL); ++ err = PTR_ERR(h_path.dentry); ++ if (IS_ERR(h_path.dentry)) ++ goto out; ++ ++ err = 0; ++ /* wh.plink dir is not monitored */ ++ if (h_path.dentry->d_inode ++ && h_path.dentry->d_inode != h_dentry->d_inode) { ++ err = vfsub_unlink(h_dir, &h_path, /*force*/0); ++ dput(h_path.dentry); ++ h_path.dentry = NULL; ++ if (!err) ++ goto again; ++ } ++ if (!err && !h_path.dentry->d_inode) ++ err = vfsub_link(h_dentry, h_dir, &h_path); ++ dput(h_path.dentry); ++ ++ out: ++ return err; ++} ++ ++struct do_whplink_args { ++ int *errp; ++ struct qstr *tgt; ++ struct dentry *h_parent; ++ struct dentry *h_dentry; ++ struct au_branch *br; ++}; ++ ++static void call_do_whplink(void *args) ++{ ++ struct do_whplink_args *a = args; ++ *a->errp = do_whplink(a->tgt, a->h_parent, a->h_dentry, a->br); ++} ++ ++static int whplink(struct dentry *h_dentry, struct inode *inode, ++ aufs_bindex_t bindex, struct au_branch *br) ++{ ++ int err, wkq_err; ++ struct au_wbr *wbr; ++ struct dentry *h_parent; ++ struct inode *h_dir; ++ char a[PLINK_NAME_LEN]; ++ struct qstr tgtname = { ++ .name = a ++ }; ++ ++ wbr = au_sbr(inode->i_sb, bindex)->br_wbr; ++ h_parent = wbr->wbr_plink; ++ h_dir = h_parent->d_inode; ++ tgtname.len = plink_name(a, sizeof(a), inode, bindex); ++ ++ /* always superio. */ ++ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_CHILD2); ++ if (!au_test_wkq(current)) { ++ struct do_whplink_args args = { ++ .errp = &err, ++ .tgt = &tgtname, ++ .h_parent = h_parent, ++ .h_dentry = h_dentry, ++ .br = br ++ }; ++ wkq_err = au_wkq_wait(call_do_whplink, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ } else ++ err = do_whplink(&tgtname, h_parent, h_dentry, br); ++ mutex_unlock(&h_dir->i_mutex); ++ ++ return err; ++} ++ ++/* free a single plink */ ++static void do_put_plink(struct pseudo_link *plink, int do_del) ++{ ++ iput(plink->inode); ++ if (do_del) ++ list_del(&plink->list); ++ kfree(plink); ++} ++ ++/* ++ * create a new pseudo-link for @h_dentry on @bindex. ++ * the linked inode is held in aufs @inode. ++ */ ++void au_plink_append(struct inode *inode, aufs_bindex_t bindex, ++ struct dentry *h_dentry) ++{ ++ struct super_block *sb; ++ struct au_sbinfo *sbinfo; ++ struct list_head *plink_list; ++ struct pseudo_link *plink; ++ int found, err, cnt; ++ ++ sb = inode->i_sb; ++ sbinfo = au_sbi(sb); ++ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK)); ++ ++ err = 0; ++ cnt = 0; ++ found = 0; ++ plink_list = &sbinfo->si_plink.head; ++ spin_lock(&sbinfo->si_plink.spin); ++ list_for_each_entry(plink, plink_list, list) { ++ cnt++; ++ if (plink->inode == inode) { ++ found = 1; ++ break; ++ } ++ } ++ if (found) { ++ spin_unlock(&sbinfo->si_plink.spin); ++ return; ++ } ++ ++ plink = NULL; ++ if (!found) { ++ plink = kmalloc(sizeof(*plink), GFP_ATOMIC); ++ if (plink) { ++ plink->inode = au_igrab(inode); ++ list_add(&plink->list, plink_list); ++ cnt++; ++ } else ++ err = -ENOMEM; ++ } ++ spin_unlock(&sbinfo->si_plink.spin); ++ ++ if (!err) { ++ au_plink_block_maintain(sb); ++ err = whplink(h_dentry, inode, bindex, au_sbr(sb, bindex)); ++ } ++ ++ if (unlikely(cnt > AUFS_PLINK_WARN)) ++ AuWarn1("unexpectedly many pseudo links, %d\n", cnt); ++ if (unlikely(err)) { ++ AuWarn("err %d, damaged pseudo link.\n", err); ++ if (!found && plink) ++ do_put_plink(plink, /*do_del*/1); ++ } ++} ++ ++/* free all plinks */ ++void au_plink_put(struct super_block *sb) ++{ ++ struct au_sbinfo *sbinfo; ++ struct list_head *plink_list; ++ struct pseudo_link *plink, *tmp; ++ ++ SiMustWriteLock(sb); ++ ++ sbinfo = au_sbi(sb); ++ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK)); ++ ++ plink_list = &sbinfo->si_plink.head; ++ /* no spin_lock since sbinfo is write-locked */ ++ list_for_each_entry_safe(plink, tmp, plink_list, list) ++ do_put_plink(plink, 0); ++ INIT_LIST_HEAD(plink_list); ++} ++ ++/* free the plinks on a branch specified by @br_id */ ++void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id) ++{ ++ struct au_sbinfo *sbinfo; ++ struct list_head *plink_list; ++ struct pseudo_link *plink, *tmp; ++ struct inode *inode; ++ aufs_bindex_t bstart, bend, bindex; ++ unsigned char do_put; ++ ++ SiMustWriteLock(sb); ++ ++ sbinfo = au_sbi(sb); ++ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK)); ++ ++ plink_list = &sbinfo->si_plink.head; ++ /* no spin_lock since sbinfo is write-locked */ ++ list_for_each_entry_safe(plink, tmp, plink_list, list) { ++ do_put = 0; ++ inode = au_igrab(plink->inode); ++ ii_write_lock_child(inode); ++ bstart = au_ibstart(inode); ++ bend = au_ibend(inode); ++ if (bstart >= 0) { ++ for (bindex = bstart; bindex <= bend; bindex++) { ++ if (!au_h_iptr(inode, bindex) ++ || au_ii_br_id(inode, bindex) != br_id) ++ continue; ++ au_set_h_iptr(inode, bindex, NULL, 0); ++ do_put = 1; ++ break; ++ } ++ } else ++ do_put_plink(plink, 1); ++ ++ if (do_put) { ++ for (bindex = bstart; bindex <= bend; bindex++) ++ if (au_h_iptr(inode, bindex)) { ++ do_put = 0; ++ break; ++ } ++ if (do_put) ++ do_put_plink(plink, 1); ++ } ++ ii_write_unlock(inode); ++ iput(inode); ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++long au_plink_ioctl(struct file *file, unsigned int cmd) ++{ ++ long err; ++ struct super_block *sb; ++ struct au_sbinfo *sbinfo; ++ ++ err = -EACCES; ++ if (!capable(CAP_SYS_ADMIN)) ++ goto out; ++ ++ err = 0; ++ sb = file->f_dentry->d_sb; ++ sbinfo = au_sbi(sb); ++ switch (cmd) { ++ case AUFS_CTL_PLINK_MAINT: ++ /* ++ * pseudo-link maintenance mode, ++ * cleared by aufs_release_dir() ++ */ ++ si_write_lock(sb); ++ if (!au_ftest_si(sbinfo, MAINTAIN_PLINK)) { ++ au_fset_si(sbinfo, MAINTAIN_PLINK); ++ au_fi(file)->fi_maintain_plink = 1; ++ } else ++ err = -EBUSY; ++ si_write_unlock(sb); ++ break; ++ case AUFS_CTL_PLINK_CLEAN: ++ aufs_write_lock(sb->s_root); ++ if (au_opt_test(sbinfo->si_mntflags, PLINK)) ++ au_plink_put(sb); ++ aufs_write_unlock(sb->s_root); ++ break; ++ default: ++ err = -EINVAL; ++ } ++ out: ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/poll.c linux-2.6.31/fs/aufs/poll.c +--- linux-2.6.31-vanilla/fs/aufs/poll.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/poll.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,56 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * poll operation ++ * There is only one filesystem which implements ->poll operation, currently. ++ */ ++ ++#include "aufs.h" ++ ++unsigned int aufs_poll(struct file *file, poll_table *wait) ++{ ++ unsigned int mask; ++ int err; ++ struct file *h_file; ++ struct dentry *dentry; ++ struct super_block *sb; ++ ++ /* We should pretend an error happened. */ ++ mask = POLLERR /* | POLLIN | POLLOUT */; ++ dentry = file->f_dentry; ++ sb = dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0); ++ if (unlikely(err)) ++ goto out; ++ ++ /* it is not an error if h_file has no operation */ ++ mask = DEFAULT_POLLMASK; ++ h_file = au_h_fptr(file, au_fbstart(file)); ++ if (h_file->f_op && h_file->f_op->poll) ++ mask = h_file->f_op->poll(h_file, wait); ++ ++ di_read_unlock(dentry, AuLock_IR); ++ fi_read_unlock(file); ++ ++ out: ++ si_read_unlock(sb); ++ AuTraceErr((int)mask); ++ return mask; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/rdu.c linux-2.6.31/fs/aufs/rdu.c +--- linux-2.6.31-vanilla/fs/aufs/rdu.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/rdu.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,331 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * readdir in userspace. ++ */ ++ ++#include ++#include ++#include ++#include "aufs.h" ++ ++/* bits for struct aufs_rdu.flags */ ++#define AuRdu_CALLED 1 ++#define AuRdu_CONT (1 << 1) ++#define AuRdu_FULL (1 << 2) ++#define au_ftest_rdu(flags, name) ((flags) & AuRdu_##name) ++#define au_fset_rdu(flags, name) { (flags) |= AuRdu_##name; } ++#define au_fclr_rdu(flags, name) { (flags) &= ~AuRdu_##name; } ++ ++struct au_rdu_arg { ++ struct aufs_rdu *rdu; ++ union au_rdu_ent_ul ent; ++ unsigned long end; ++ ++ struct super_block *sb; ++ int err; ++}; ++ ++static int au_rdu_fill(void *__arg, const char *name, int nlen, ++ loff_t offset, u64 h_ino, unsigned int d_type) ++{ ++ int err, len; ++ struct au_rdu_arg *arg = __arg; ++ struct aufs_rdu *rdu = arg->rdu; ++ struct au_rdu_ent ent; ++ ++ err = 0; ++ arg->err = 0; ++ au_fset_rdu(rdu->cookie.flags, CALLED); ++ len = au_rdu_len(nlen); ++ if (arg->ent.ul + len < arg->end) { ++ ent.ino = h_ino; ++ ent.bindex = rdu->cookie.bindex; ++ ent.type = d_type; ++ ent.nlen = nlen; ++ ++ err = -EFAULT; ++ if (copy_to_user(arg->ent.e, &ent, sizeof(ent))) ++ goto out; ++ if (copy_to_user(arg->ent.e->name, name, nlen)) ++ goto out; ++ /* the terminating NULL */ ++ if (__put_user(0, arg->ent.e->name + nlen)) ++ goto out; ++ err = 0; ++ /* AuDbg("%p, %.*s\n", arg->ent.p, nlen, name); */ ++ arg->ent.ul += len; ++ rdu->rent++; ++ } else { ++ err = -EFAULT; ++ au_fset_rdu(rdu->cookie.flags, FULL); ++ rdu->full = 1; ++ rdu->tail = arg->ent; ++ } ++ ++ out: ++ /* AuTraceErr(err); */ ++ return err; ++} ++ ++static int au_rdu_do(struct file *h_file, struct au_rdu_arg *arg) ++{ ++ int err; ++ loff_t offset; ++ struct au_rdu_cookie *cookie = &arg->rdu->cookie; ++ ++ offset = vfsub_llseek(h_file, cookie->h_pos, SEEK_SET); ++ err = offset; ++ if (unlikely(offset != cookie->h_pos)) ++ goto out; ++ ++ err = 0; ++ do { ++ arg->err = 0; ++ au_fclr_rdu(cookie->flags, CALLED); ++ /* smp_mb(); */ ++ err = vfsub_readdir(h_file, au_rdu_fill, arg); ++ if (err >= 0) ++ err = arg->err; ++ } while (!err ++ && au_ftest_rdu(cookie->flags, CALLED) ++ && !au_ftest_rdu(cookie->flags, FULL)); ++ cookie->h_pos = h_file->f_pos; ++ ++ out: ++ AuTraceErr(err); ++ return err; ++} ++ ++static int au_rdu(struct file *file, struct aufs_rdu *rdu) ++{ ++ int err; ++ aufs_bindex_t bend; ++ struct au_rdu_arg arg; ++ struct dentry *dentry; ++ struct inode *inode; ++ struct file *h_file; ++ struct au_rdu_cookie *cookie = &rdu->cookie; ++ ++ err = !access_ok(VERIFY_WRITE, rdu->ent.e, rdu->sz); ++ if (unlikely(err)) { ++ err = -EFAULT; ++ AuTraceErr(err); ++ goto out; ++ } ++ rdu->rent = 0; ++ rdu->tail = rdu->ent; ++ rdu->full = 0; ++ arg.rdu = rdu; ++ arg.ent = rdu->ent; ++ arg.end = arg.ent.ul; ++ arg.end += rdu->sz; ++ ++ err = -ENOTDIR; ++ if (unlikely(!file->f_op || !file->f_op->readdir)) ++ goto out; ++ ++ err = security_file_permission(file, MAY_READ); ++ AuTraceErr(err); ++ if (unlikely(err)) ++ goto out; ++ ++ dentry = file->f_dentry; ++ inode = dentry->d_inode; ++#if 1 ++ mutex_lock(&inode->i_mutex); ++#else ++ err = mutex_lock_killable(&inode->i_mutex); ++ AuTraceErr(err); ++ if (unlikely(err)) ++ goto out; ++#endif ++ err = -ENOENT; ++ if (unlikely(IS_DEADDIR(inode))) ++ goto out_mtx; ++ ++ arg.sb = inode->i_sb; ++ si_read_lock(arg.sb, AuLock_FLUSH); ++ fi_read_lock(file); ++ ++ err = -EAGAIN; ++ if (unlikely(au_ftest_rdu(cookie->flags, CONT) ++ && cookie->generation != au_figen(file))) ++ goto out_unlock; ++ ++ err = 0; ++ if (!rdu->blk) { ++ rdu->blk = au_sbi(arg.sb)->si_rdblk; ++ if (!rdu->blk) ++ rdu->blk = au_dir_size(file, /*dentry*/NULL); ++ } ++ bend = au_fbstart(file); ++ if (cookie->bindex < bend) ++ cookie->bindex = bend; ++ bend = au_fbend(file); ++ /* AuDbg("b%d, b%d\n", cookie->bindex, bend); */ ++ for (; !err && cookie->bindex <= bend; ++ cookie->bindex++, cookie->h_pos = 0) { ++ h_file = au_h_fptr(file, cookie->bindex); ++ if (!h_file) ++ continue; ++ ++ au_fclr_rdu(cookie->flags, FULL); ++ err = au_rdu_do(h_file, &arg); ++ AuTraceErr(err); ++ if (unlikely(au_ftest_rdu(cookie->flags, FULL) || err)) ++ break; ++ } ++ AuDbg("rent %llu\n", rdu->rent); ++ ++ if (!err && !au_ftest_rdu(cookie->flags, CONT)) { ++ rdu->shwh = !!au_opt_test(au_sbi(arg.sb)->si_mntflags, SHWH); ++ au_fset_rdu(cookie->flags, CONT); ++ cookie->generation = au_figen(file); ++ } ++ ++ ii_read_lock_child(inode); ++ fsstack_copy_attr_atime(inode, au_h_iptr(inode, au_ibstart(inode))); ++ ii_read_unlock(inode); ++ ++ out_unlock: ++ fi_read_unlock(file); ++ si_read_unlock(arg.sb); ++ out_mtx: ++ mutex_unlock(&inode->i_mutex); ++ out: ++ AuTraceErr(err); ++ return err; ++} ++ ++static int au_rdu_ino(struct file *file, struct aufs_rdu *rdu) ++{ ++ int err; ++ ino_t ino; ++ unsigned long long nent; ++ union au_rdu_ent_ul *u; ++ struct au_rdu_ent ent; ++ struct super_block *sb; ++ ++ err = 0; ++ nent = rdu->nent; ++ u = &rdu->ent; ++ sb = file->f_dentry->d_sb; ++ si_read_lock(sb, AuLock_FLUSH); ++ while (nent-- > 0) { ++ err = !access_ok(VERIFY_WRITE, u->e, sizeof(ent)); ++ if (unlikely(err)) { ++ err = -EFAULT; ++ AuTraceErr(err); ++ break; ++ } ++ ++ err = copy_from_user(&ent, u->e, sizeof(ent)); ++ if (unlikely(err)) { ++ err = -EFAULT; ++ AuTraceErr(err); ++ break; ++ } ++ ++ /* AuDbg("b%d, i%llu\n", ent.bindex, ent.ino); */ ++ if (!ent.wh) ++ err = au_ino(sb, ent.bindex, ent.ino, ent.type, &ino); ++ else ++ err = au_wh_ino(sb, ent.bindex, ent.ino, ent.type, ++ &ino); ++ if (unlikely(err)) { ++ AuTraceErr(err); ++ break; ++ } ++ ++ err = __put_user(ino, &u->e->ino); ++ if (unlikely(err)) { ++ err = -EFAULT; ++ AuTraceErr(err); ++ break; ++ } ++ u->ul += au_rdu_len(ent.nlen); ++ } ++ si_read_unlock(sb); ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int au_rdu_verify(struct aufs_rdu *rdu) ++{ ++ AuDbg("rdu{%llu, %p, (%u, %u) | %u | %llu, %u, %u | " ++ "%llu, b%d, 0x%x, g%u}\n", ++ rdu->sz, rdu->ent.e, rdu->verify[0], rdu->verify[1], ++ rdu->blk, ++ rdu->rent, rdu->shwh, rdu->full, ++ rdu->cookie.h_pos, rdu->cookie.bindex, rdu->cookie.flags, ++ rdu->cookie.generation); ++ ++ if (rdu->verify[AufsCtlRduV_SZ] == sizeof(*rdu) ++ && rdu->verify[AufsCtlRduV_SZ_PTR] == sizeof(rdu)) ++ return 0; ++ ++ AuDbg("%u:%u, %u:%u\n", ++ rdu->verify[AufsCtlRduV_SZ], (unsigned int)sizeof(*rdu), ++ rdu->verify[AufsCtlRduV_SZ_PTR], (unsigned int)sizeof(rdu)); ++ return -EINVAL; ++} ++ ++long au_rdu_ioctl(struct file *file, unsigned int cmd, unsigned long arg) ++{ ++ long err, e; ++ struct aufs_rdu rdu; ++ void __user *p = (void __user *)arg; ++ ++ err = copy_from_user(&rdu, p, sizeof(rdu)); ++ if (unlikely(err)) { ++ err = -EFAULT; ++ AuTraceErr(err); ++ goto out; ++ } ++ err = au_rdu_verify(&rdu); ++ if (unlikely(err)) ++ goto out; ++ ++ switch (cmd) { ++ case AUFS_CTL_RDU: ++ err = au_rdu(file, &rdu); ++ if (unlikely(err)) ++ break; ++ ++ e = copy_to_user(p, &rdu, sizeof(rdu)); ++ if (unlikely(e)) { ++ err = -EFAULT; ++ AuTraceErr(err); ++ } ++ break; ++ case AUFS_CTL_RDU_INO: ++ err = au_rdu_ino(file, &rdu); ++ break; ++ ++ default: ++ err = -EINVAL; ++ } ++ ++ out: ++ AuTraceErr(err); ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/rwsem.h linux-2.6.31/fs/aufs/rwsem.h +--- linux-2.6.31-vanilla/fs/aufs/rwsem.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/rwsem.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,186 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * simple read-write semaphore wrappers ++ */ ++ ++#ifndef __AUFS_RWSEM_H__ ++#define __AUFS_RWSEM_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++ ++struct au_rwsem { ++ struct rw_semaphore rwsem; ++#ifdef CONFIG_AUFS_DEBUG ++ /* just for debugging, not almighty counter */ ++ atomic_t rcnt, wcnt; ++#endif ++}; ++ ++#ifdef CONFIG_AUFS_DEBUG ++#define AuDbgCntInit(rw) do { \ ++ atomic_set(&(rw)->rcnt, 0); \ ++ atomic_set(&(rw)->wcnt, 0); \ ++ smp_mb(); /* atomic set */ \ ++} while (0) ++ ++#define AuDbgRcntInc(rw) atomic_inc_return(&(rw)->rcnt) ++#define AuDbgRcntDec(rw) WARN_ON(atomic_dec_return(&(rw)->rcnt) < 0) ++#define AuDbgWcntInc(rw) WARN_ON(atomic_inc_return(&(rw)->wcnt) > 1) ++#define AuDbgWcntDec(rw) WARN_ON(atomic_dec_return(&(rw)->wcnt) < 0) ++#else ++#define AuDbgCntInit(rw) do {} while (0) ++#define AuDbgRcntInc(rw) do {} while (0) ++#define AuDbgRcntDec(rw) do {} while (0) ++#define AuDbgWcntInc(rw) do {} while (0) ++#define AuDbgWcntDec(rw) do {} while (0) ++#endif /* CONFIG_AUFS_DEBUG */ ++ ++/* to debug easier, do not make them inlined functions */ ++#define AuRwMustNoWaiters(rw) AuDebugOn(!list_empty(&(rw)->rwsem.wait_list)) ++/* rwsem_is_locked() is unusable */ ++#define AuRwMustReadLock(rw) AuDebugOn(atomic_read(&(rw)->rcnt) <= 0) ++#define AuRwMustWriteLock(rw) AuDebugOn(atomic_read(&(rw)->wcnt) <= 0) ++#define AuRwMustAnyLock(rw) AuDebugOn(atomic_read(&(rw)->rcnt) <= 0 \ ++ && atomic_read(&(rw)->wcnt) <= 0) ++#define AuRwDestroy(rw) AuDebugOn(atomic_read(&(rw)->rcnt) \ ++ || atomic_read(&(rw)->wcnt)) ++ ++static inline void au_rw_init(struct au_rwsem *rw) ++{ ++ AuDbgCntInit(rw); ++ init_rwsem(&rw->rwsem); ++} ++ ++static inline void au_rw_init_wlock(struct au_rwsem *rw) ++{ ++ au_rw_init(rw); ++ down_write(&rw->rwsem); ++ AuDbgWcntInc(rw); ++} ++ ++static inline void au_rw_init_wlock_nested(struct au_rwsem *rw, ++ unsigned int lsc) ++{ ++ au_rw_init(rw); ++ down_write_nested(&rw->rwsem, lsc); ++ AuDbgWcntInc(rw); ++} ++ ++static inline void au_rw_read_lock(struct au_rwsem *rw) ++{ ++ down_read(&rw->rwsem); ++ AuDbgRcntInc(rw); ++} ++ ++static inline void au_rw_read_lock_nested(struct au_rwsem *rw, unsigned int lsc) ++{ ++ down_read_nested(&rw->rwsem, lsc); ++ AuDbgRcntInc(rw); ++} ++ ++static inline void au_rw_read_unlock(struct au_rwsem *rw) ++{ ++ AuRwMustReadLock(rw); ++ AuDbgRcntDec(rw); ++ up_read(&rw->rwsem); ++} ++ ++static inline void au_rw_dgrade_lock(struct au_rwsem *rw) ++{ ++ AuRwMustWriteLock(rw); ++ AuDbgRcntInc(rw); ++ AuDbgWcntDec(rw); ++ downgrade_write(&rw->rwsem); ++} ++ ++static inline void au_rw_write_lock(struct au_rwsem *rw) ++{ ++ down_write(&rw->rwsem); ++ AuDbgWcntInc(rw); ++} ++ ++static inline void au_rw_write_lock_nested(struct au_rwsem *rw, ++ unsigned int lsc) ++{ ++ down_write_nested(&rw->rwsem, lsc); ++ AuDbgWcntInc(rw); ++} ++ ++static inline void au_rw_write_unlock(struct au_rwsem *rw) ++{ ++ AuRwMustWriteLock(rw); ++ AuDbgWcntDec(rw); ++ up_write(&rw->rwsem); ++} ++ ++/* why is not _nested version defined */ ++static inline int au_rw_read_trylock(struct au_rwsem *rw) ++{ ++ int ret = down_read_trylock(&rw->rwsem); ++ if (ret) ++ AuDbgRcntInc(rw); ++ return ret; ++} ++ ++static inline int au_rw_write_trylock(struct au_rwsem *rw) ++{ ++ int ret = down_write_trylock(&rw->rwsem); ++ if (ret) ++ AuDbgWcntInc(rw); ++ return ret; ++} ++ ++#undef AuDbgCntInit ++#undef AuDbgRcntInc ++#undef AuDbgRcntDec ++#undef AuDbgWcntInc ++#undef AuDbgWcntDec ++ ++#define AuSimpleLockRwsemFuncs(prefix, param, rwsem) \ ++static inline void prefix##_read_lock(param) \ ++{ au_rw_read_lock(rwsem); } \ ++static inline void prefix##_write_lock(param) \ ++{ au_rw_write_lock(rwsem); } \ ++static inline int prefix##_read_trylock(param) \ ++{ return au_rw_read_trylock(rwsem); } \ ++static inline int prefix##_write_trylock(param) \ ++{ return au_rw_write_trylock(rwsem); } ++/* why is not _nested version defined */ ++/* static inline void prefix##_read_trylock_nested(param, lsc) ++{ au_rw_read_trylock_nested(rwsem, lsc)); } ++static inline void prefix##_write_trylock_nestd(param, lsc) ++{ au_rw_write_trylock_nested(rwsem, lsc); } */ ++ ++#define AuSimpleUnlockRwsemFuncs(prefix, param, rwsem) \ ++static inline void prefix##_read_unlock(param) \ ++{ au_rw_read_unlock(rwsem); } \ ++static inline void prefix##_write_unlock(param) \ ++{ au_rw_write_unlock(rwsem); } \ ++static inline void prefix##_downgrade_lock(param) \ ++{ au_rw_dgrade_lock(rwsem); } ++ ++#define AuSimpleRwsemFuncs(prefix, param, rwsem) \ ++ AuSimpleLockRwsemFuncs(prefix, param, rwsem) \ ++ AuSimpleUnlockRwsemFuncs(prefix, param, rwsem) ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_RWSEM_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/sbinfo.c linux-2.6.31/fs/aufs/sbinfo.c +--- linux-2.6.31-vanilla/fs/aufs/sbinfo.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/sbinfo.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,208 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * superblock private data ++ */ ++ ++#include "aufs.h" ++ ++/* ++ * they are necessary regardless sysfs is disabled. ++ */ ++void au_si_free(struct kobject *kobj) ++{ ++ struct au_sbinfo *sbinfo; ++ struct super_block *sb; ++ ++ sbinfo = container_of(kobj, struct au_sbinfo, si_kobj); ++ AuDebugOn(!list_empty(&sbinfo->si_plink.head)); ++ ++ sb = sbinfo->si_sb; ++ si_write_lock(sb); ++ au_xino_clr(sb); ++ au_br_free(sbinfo); ++ kfree(sbinfo->si_branch); ++ mutex_destroy(&sbinfo->si_xib_mtx); ++ si_write_unlock(sb); ++ AuRwDestroy(&sbinfo->si_rwsem); ++ ++ kfree(sbinfo); ++} ++ ++int au_si_alloc(struct super_block *sb) ++{ ++ int err; ++ struct au_sbinfo *sbinfo; ++ ++ err = -ENOMEM; ++ sbinfo = kmalloc(sizeof(*sbinfo), GFP_NOFS); ++ if (unlikely(!sbinfo)) ++ goto out; ++ ++ /* will be reallocated separately */ ++ sbinfo->si_branch = kzalloc(sizeof(*sbinfo->si_branch), GFP_NOFS); ++ if (unlikely(!sbinfo->si_branch)) ++ goto out_sbinfo; ++ ++ memset(&sbinfo->si_kobj, 0, sizeof(sbinfo->si_kobj)); ++ err = sysaufs_si_init(sbinfo); ++ if (unlikely(err)) ++ goto out_br; ++ ++ au_nwt_init(&sbinfo->si_nowait); ++ au_rw_init_wlock(&sbinfo->si_rwsem); ++ sbinfo->si_generation = 0; ++ sbinfo->au_si_status = 0; ++ sbinfo->si_bend = -1; ++ sbinfo->si_last_br_id = 0; ++ ++ sbinfo->si_wbr_copyup = AuWbrCopyup_Def; ++ sbinfo->si_wbr_create = AuWbrCreate_Def; ++ sbinfo->si_wbr_copyup_ops = au_wbr_copyup_ops + AuWbrCopyup_Def; ++ sbinfo->si_wbr_create_ops = au_wbr_create_ops + AuWbrCreate_Def; ++ ++ sbinfo->si_mntflags = AuOpt_Def; ++ ++ sbinfo->si_xread = NULL; ++ sbinfo->si_xwrite = NULL; ++ sbinfo->si_xib = NULL; ++ mutex_init(&sbinfo->si_xib_mtx); ++ sbinfo->si_xib_buf = NULL; ++ sbinfo->si_xino_brid = -1; ++ /* leave si_xib_last_pindex and si_xib_next_bit */ ++ ++ sbinfo->si_rdcache = AUFS_RDCACHE_DEF * HZ; ++ sbinfo->si_rdblk = AUFS_RDBLK_DEF; ++ sbinfo->si_rdhash = AUFS_RDHASH_DEF; ++ sbinfo->si_dirwh = AUFS_DIRWH_DEF; ++ ++ au_spl_init(&sbinfo->si_plink); ++ init_waitqueue_head(&sbinfo->si_plink_wq); ++ ++ /* leave other members for sysaufs and si_mnt. */ ++ sbinfo->si_sb = sb; ++ sb->s_fs_info = sbinfo; ++ au_debug_sbinfo_init(sbinfo); ++ return 0; /* success */ ++ ++ out_br: ++ kfree(sbinfo->si_branch); ++ out_sbinfo: ++ kfree(sbinfo); ++ out: ++ return err; ++} ++ ++int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr) ++{ ++ int err, sz; ++ struct au_branch **brp; ++ ++ AuRwMustWriteLock(&sbinfo->si_rwsem); ++ ++ err = -ENOMEM; ++ sz = sizeof(*brp) * (sbinfo->si_bend + 1); ++ if (unlikely(!sz)) ++ sz = sizeof(*brp); ++ brp = au_kzrealloc(sbinfo->si_branch, sz, sizeof(*brp) * nbr, GFP_NOFS); ++ if (brp) { ++ sbinfo->si_branch = brp; ++ err = 0; ++ } ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++unsigned int au_sigen_inc(struct super_block *sb) ++{ ++ unsigned int gen; ++ ++ SiMustWriteLock(sb); ++ ++ gen = ++au_sbi(sb)->si_generation; ++ au_update_digen(sb->s_root); ++ au_update_iigen(sb->s_root->d_inode); ++ sb->s_root->d_inode->i_version++; ++ return gen; ++} ++ ++aufs_bindex_t au_new_br_id(struct super_block *sb) ++{ ++ aufs_bindex_t br_id; ++ int i; ++ struct au_sbinfo *sbinfo; ++ ++ SiMustWriteLock(sb); ++ ++ sbinfo = au_sbi(sb); ++ for (i = 0; i <= AUFS_BRANCH_MAX; i++) { ++ br_id = ++sbinfo->si_last_br_id; ++ if (br_id && au_br_index(sb, br_id) < 0) ++ return br_id; ++ } ++ ++ return -1; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* dentry and super_block lock. call at entry point */ ++void aufs_read_lock(struct dentry *dentry, int flags) ++{ ++ si_read_lock(dentry->d_sb, flags); ++ if (au_ftest_lock(flags, DW)) ++ di_write_lock_child(dentry); ++ else ++ di_read_lock_child(dentry, flags); ++} ++ ++void aufs_read_unlock(struct dentry *dentry, int flags) ++{ ++ if (au_ftest_lock(flags, DW)) ++ di_write_unlock(dentry); ++ else ++ di_read_unlock(dentry, flags); ++ si_read_unlock(dentry->d_sb); ++} ++ ++void aufs_write_lock(struct dentry *dentry) ++{ ++ si_write_lock(dentry->d_sb); ++ di_write_lock_child(dentry); ++} ++ ++void aufs_write_unlock(struct dentry *dentry) ++{ ++ di_write_unlock(dentry); ++ si_write_unlock(dentry->d_sb); ++} ++ ++void aufs_read_and_write_lock2(struct dentry *d1, struct dentry *d2, int flags) ++{ ++ si_read_lock(d1->d_sb, flags); ++ di_write_lock2_child(d1, d2, au_ftest_lock(flags, DIR)); ++} ++ ++void aufs_read_and_write_unlock2(struct dentry *d1, struct dentry *d2) ++{ ++ di_write_unlock2(d1, d2); ++ si_read_unlock(d1->d_sb); ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/spl.h linux-2.6.31/fs/aufs/spl.h +--- linux-2.6.31-vanilla/fs/aufs/spl.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/spl.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,57 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * simple list protected by a spinlock ++ */ ++ ++#ifndef __AUFS_SPL_H__ ++#define __AUFS_SPL_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++ ++struct au_splhead { ++ spinlock_t spin; ++ struct list_head head; ++}; ++ ++static inline void au_spl_init(struct au_splhead *spl) ++{ ++ spin_lock_init(&spl->spin); ++ INIT_LIST_HEAD(&spl->head); ++} ++ ++static inline void au_spl_add(struct list_head *list, struct au_splhead *spl) ++{ ++ spin_lock(&spl->spin); ++ list_add(list, &spl->head); ++ spin_unlock(&spl->spin); ++} ++ ++static inline void au_spl_del(struct list_head *list, struct au_splhead *spl) ++{ ++ spin_lock(&spl->spin); ++ list_del(list); ++ spin_unlock(&spl->spin); ++} ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_SPL_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/super.c linux-2.6.31/fs/aufs/super.c +--- linux-2.6.31-vanilla/fs/aufs/super.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/super.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,874 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * mount and super_block operations ++ */ ++ ++#include ++#include ++#include ++#include ++#include "aufs.h" ++ ++/* ++ * super_operations ++ */ ++static struct inode *aufs_alloc_inode(struct super_block *sb __maybe_unused) ++{ ++ struct au_icntnr *c; ++ ++ c = au_cache_alloc_icntnr(); ++ if (c) { ++ inode_init_once(&c->vfs_inode); ++ c->vfs_inode.i_version = 1; /* sigen(sb); */ ++ c->iinfo.ii_hinode = NULL; ++ return &c->vfs_inode; ++ } ++ return NULL; ++} ++ ++static void aufs_destroy_inode(struct inode *inode) ++{ ++ au_iinfo_fin(inode); ++ au_cache_free_icntnr(container_of(inode, struct au_icntnr, vfs_inode)); ++} ++ ++struct inode *au_iget_locked(struct super_block *sb, ino_t ino) ++{ ++ struct inode *inode; ++ int err; ++ ++ inode = iget_locked(sb, ino); ++ if (unlikely(!inode)) { ++ inode = ERR_PTR(-ENOMEM); ++ goto out; ++ } ++ if (!(inode->i_state & I_NEW)) ++ goto out; ++ ++ err = au_xigen_new(inode); ++ if (!err) ++ err = au_iinfo_init(inode); ++ if (!err) ++ inode->i_version++; ++ else { ++ iget_failed(inode); ++ inode = ERR_PTR(err); ++ } ++ ++ out: ++ /* never return NULL */ ++ AuDebugOn(!inode); ++ AuTraceErrPtr(inode); ++ return inode; ++} ++ ++/* lock free root dinfo */ ++static int au_show_brs(struct seq_file *seq, struct super_block *sb) ++{ ++ int err; ++ aufs_bindex_t bindex, bend; ++ struct path path; ++ struct au_hdentry *hd; ++ struct au_branch *br; ++ ++ err = 0; ++ bend = au_sbend(sb); ++ hd = au_di(sb->s_root)->di_hdentry; ++ for (bindex = 0; !err && bindex <= bend; bindex++) { ++ br = au_sbr(sb, bindex); ++ path.mnt = br->br_mnt; ++ path.dentry = hd[bindex].hd_dentry; ++ err = au_seq_path(seq, &path); ++ if (err > 0) ++ err = seq_printf(seq, "=%s", ++ au_optstr_br_perm(br->br_perm)); ++ if (!err && bindex != bend) ++ err = seq_putc(seq, ':'); ++ } ++ ++ return err; ++} ++ ++static void au_show_wbr_create(struct seq_file *m, int v, ++ struct au_sbinfo *sbinfo) ++{ ++ const char *pat; ++ ++ AuRwMustAnyLock(&sbinfo->si_rwsem); ++ ++ seq_printf(m, ",create="); ++ pat = au_optstr_wbr_create(v); ++ switch (v) { ++ case AuWbrCreate_TDP: ++ case AuWbrCreate_RR: ++ case AuWbrCreate_MFS: ++ case AuWbrCreate_PMFS: ++ seq_printf(m, pat); ++ break; ++ case AuWbrCreate_MFSV: ++ seq_printf(m, /*pat*/"mfs:%lu", ++ sbinfo->si_wbr_mfs.mfs_expire / HZ); ++ break; ++ case AuWbrCreate_PMFSV: ++ seq_printf(m, /*pat*/"pmfs:%lu", ++ sbinfo->si_wbr_mfs.mfs_expire / HZ); ++ break; ++ case AuWbrCreate_MFSRR: ++ seq_printf(m, /*pat*/"mfsrr:%llu", ++ sbinfo->si_wbr_mfs.mfsrr_watermark); ++ break; ++ case AuWbrCreate_MFSRRV: ++ seq_printf(m, /*pat*/"mfsrr:%llu:%lu", ++ sbinfo->si_wbr_mfs.mfsrr_watermark, ++ sbinfo->si_wbr_mfs.mfs_expire / HZ); ++ break; ++ } ++} ++ ++static int au_show_xino(struct seq_file *seq, struct vfsmount *mnt) ++{ ++#ifdef CONFIG_SYSFS ++ return 0; ++#else ++ int err; ++ const int len = sizeof(AUFS_XINO_FNAME) - 1; ++ aufs_bindex_t bindex, brid; ++ struct super_block *sb; ++ struct qstr *name; ++ struct file *f; ++ struct dentry *d, *h_root; ++ ++ AuRwMustAnyLock(&sbinfo->si_rwsem); ++ ++ err = 0; ++ sb = mnt->mnt_sb; ++ f = au_sbi(sb)->si_xib; ++ if (!f) ++ goto out; ++ ++ /* stop printing the default xino path on the first writable branch */ ++ h_root = NULL; ++ brid = au_xino_brid(sb); ++ if (brid >= 0) { ++ bindex = au_br_index(sb, brid); ++ h_root = au_di(sb->s_root)->di_hdentry[0 + bindex].hd_dentry; ++ } ++ d = f->f_dentry; ++ name = &d->d_name; ++ /* safe ->d_parent because the file is unlinked */ ++ if (d->d_parent == h_root ++ && name->len == len ++ && !memcmp(name->name, AUFS_XINO_FNAME, len)) ++ goto out; ++ ++ seq_puts(seq, ",xino="); ++ err = au_xino_path(seq, f); ++ ++ out: ++ return err; ++#endif ++} ++ ++/* seq_file will re-call me in case of too long string */ ++static int aufs_show_options(struct seq_file *m, struct vfsmount *mnt) ++{ ++ int err, n; ++ unsigned int mnt_flags, v; ++ struct super_block *sb; ++ struct au_sbinfo *sbinfo; ++ ++#define AuBool(name, str) do { \ ++ v = au_opt_test(mnt_flags, name); \ ++ if (v != au_opt_test(AuOpt_Def, name)) \ ++ seq_printf(m, ",%s" #str, v ? "" : "no"); \ ++} while (0) ++ ++#define AuStr(name, str) do { \ ++ v = mnt_flags & AuOptMask_##name; \ ++ if (v != (AuOpt_Def & AuOptMask_##name)) \ ++ seq_printf(m, "," #str "=%s", au_optstr_##str(v)); \ ++} while (0) ++ ++#define AuUInt(name, str, val) do { \ ++ if (val != AUFS_##name##_DEF) \ ++ seq_printf(m, "," #str "=%u", val); \ ++} while (0) ++ ++ /* lock free root dinfo */ ++ sb = mnt->mnt_sb; ++ si_noflush_read_lock(sb); ++ sbinfo = au_sbi(sb); ++ seq_printf(m, ",si=%lx", sysaufs_si_id(sbinfo)); ++ ++ mnt_flags = au_mntflags(sb); ++ if (au_opt_test(mnt_flags, XINO)) { ++ err = au_show_xino(m, mnt); ++ if (unlikely(err)) ++ goto out; ++ } else ++ seq_puts(m, ",noxino"); ++ ++ AuBool(TRUNC_XINO, trunc_xino); ++ AuStr(UDBA, udba); ++ AuBool(SHWH, shwh); ++ AuBool(PLINK, plink); ++ /* AuBool(DIRPERM1, dirperm1); */ ++ /* AuBool(REFROF, refrof); */ ++ ++ v = sbinfo->si_wbr_create; ++ if (v != AuWbrCreate_Def) ++ au_show_wbr_create(m, v, sbinfo); ++ ++ v = sbinfo->si_wbr_copyup; ++ if (v != AuWbrCopyup_Def) ++ seq_printf(m, ",cpup=%s", au_optstr_wbr_copyup(v)); ++ ++ v = au_opt_test(mnt_flags, ALWAYS_DIROPQ); ++ if (v != au_opt_test(AuOpt_Def, ALWAYS_DIROPQ)) ++ seq_printf(m, ",diropq=%c", v ? 'a' : 'w'); ++ ++ AuUInt(DIRWH, dirwh, sbinfo->si_dirwh); ++ ++ n = sbinfo->si_rdcache / HZ; ++ AuUInt(RDCACHE, rdcache, n); ++ ++ AuUInt(RDBLK, rdblk, sbinfo->si_rdblk); ++ AuUInt(RDHASH, rdhash, sbinfo->si_rdhash); ++ ++ AuBool(SUM, sum); ++ /* AuBool(SUM_W, wsum); */ ++ AuBool(WARN_PERM, warn_perm); ++ AuBool(VERBOSE, verbose); ++ ++ out: ++ /* be sure to print "br:" last */ ++ if (!sysaufs_brs) { ++ seq_puts(m, ",br:"); ++ au_show_brs(m, sb); ++ } ++ si_read_unlock(sb); ++ return 0; ++ ++#undef Deleted ++#undef AuBool ++#undef AuStr ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* sum mode which returns the summation for statfs(2) */ ++ ++static u64 au_add_till_max(u64 a, u64 b) ++{ ++ u64 old; ++ ++ old = a; ++ a += b; ++ if (old < a) ++ return a; ++ return ULLONG_MAX; ++} ++ ++static int au_statfs_sum(struct super_block *sb, struct kstatfs *buf) ++{ ++ int err; ++ u64 blocks, bfree, bavail, files, ffree; ++ aufs_bindex_t bend, bindex, i; ++ unsigned char shared; ++ struct vfsmount *h_mnt; ++ struct super_block *h_sb; ++ ++ blocks = 0; ++ bfree = 0; ++ bavail = 0; ++ files = 0; ++ ffree = 0; ++ ++ err = 0; ++ bend = au_sbend(sb); ++ for (bindex = bend; bindex >= 0; bindex--) { ++ h_mnt = au_sbr_mnt(sb, bindex); ++ h_sb = h_mnt->mnt_sb; ++ shared = 0; ++ for (i = bindex + 1; !shared && i <= bend; i++) ++ shared = (au_sbr_sb(sb, i) == h_sb); ++ if (shared) ++ continue; ++ ++ /* sb->s_root for NFS is unreliable */ ++ err = vfs_statfs(h_mnt->mnt_root, buf); ++ if (unlikely(err)) ++ goto out; ++ ++ blocks = au_add_till_max(blocks, buf->f_blocks); ++ bfree = au_add_till_max(bfree, buf->f_bfree); ++ bavail = au_add_till_max(bavail, buf->f_bavail); ++ files = au_add_till_max(files, buf->f_files); ++ ffree = au_add_till_max(ffree, buf->f_ffree); ++ } ++ ++ buf->f_blocks = blocks; ++ buf->f_bfree = bfree; ++ buf->f_bavail = bavail; ++ buf->f_files = files; ++ buf->f_ffree = ffree; ++ ++ out: ++ return err; ++} ++ ++static int aufs_statfs(struct dentry *dentry, struct kstatfs *buf) ++{ ++ int err; ++ struct super_block *sb; ++ ++ /* lock free root dinfo */ ++ sb = dentry->d_sb; ++ si_noflush_read_lock(sb); ++ if (!au_opt_test(au_mntflags(sb), SUM)) ++ /* sb->s_root for NFS is unreliable */ ++ err = vfs_statfs(au_sbr_mnt(sb, 0)->mnt_root, buf); ++ else ++ err = au_statfs_sum(sb, buf); ++ si_read_unlock(sb); ++ ++ if (!err) { ++ buf->f_type = AUFS_SUPER_MAGIC; ++ buf->f_namelen -= AUFS_WH_PFX_LEN; ++ memset(&buf->f_fsid, 0, sizeof(buf->f_fsid)); ++ } ++ /* buf->f_bsize = buf->f_blocks = buf->f_bfree = buf->f_bavail = -1; */ ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* try flushing the lower fs at aufs remount/unmount time */ ++ ++static void au_fsync_br(struct super_block *sb) ++{ ++ aufs_bindex_t bend, bindex; ++ int brperm; ++ struct au_branch *br; ++ struct super_block *h_sb; ++ ++ bend = au_sbend(sb); ++ for (bindex = 0; bindex < bend; bindex++) { ++ br = au_sbr(sb, bindex); ++ brperm = br->br_perm; ++ if (brperm == AuBrPerm_RR || brperm == AuBrPerm_RRWH) ++ continue; ++ h_sb = br->br_mnt->mnt_sb; ++ if (bdev_read_only(h_sb->s_bdev)) ++ continue; ++ ++ lockdep_off(); ++ down_write(&h_sb->s_umount); ++ shrink_dcache_sb(h_sb); ++ sync_filesystem(h_sb); ++ up_write(&h_sb->s_umount); ++ lockdep_on(); ++ } ++} ++ ++/* ++ * this IS NOT for super_operations. ++ * I guess it will be reverted someday. ++ */ ++static void aufs_umount_begin(struct super_block *sb) ++{ ++ struct au_sbinfo *sbinfo; ++ ++ sbinfo = au_sbi(sb); ++ if (!sbinfo) ++ return; ++ ++ si_write_lock(sb); ++ au_fsync_br(sb); ++ if (au_opt_test(au_mntflags(sb), PLINK)) ++ au_plink_put(sb); ++ if (sbinfo->si_wbr_create_ops->fin) ++ sbinfo->si_wbr_create_ops->fin(sb); ++ si_write_unlock(sb); ++} ++ ++/* final actions when unmounting a file system */ ++static void aufs_put_super(struct super_block *sb) ++{ ++ struct au_sbinfo *sbinfo; ++ ++ sbinfo = au_sbi(sb); ++ if (!sbinfo) ++ return; ++ ++ aufs_umount_begin(sb); ++ dbgaufs_si_fin(sbinfo); ++ kobject_put(&sbinfo->si_kobj); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * refresh dentry and inode at remount time. ++ */ ++static int do_refresh(struct dentry *dentry, mode_t type, ++ unsigned int dir_flags) ++{ ++ int err; ++ struct dentry *parent; ++ ++ di_write_lock_child(dentry); ++ parent = dget_parent(dentry); ++ di_read_lock_parent(parent, AuLock_IR); ++ ++ /* returns the number of positive dentries */ ++ err = au_refresh_hdentry(dentry, type); ++ if (err >= 0) { ++ struct inode *inode = dentry->d_inode; ++ err = au_refresh_hinode(inode, dentry); ++ if (!err && type == S_IFDIR) ++ au_reset_hinotify(inode, dir_flags); ++ } ++ if (unlikely(err)) ++ AuErr("unrecoverable error %d, %.*s\n", err, AuDLNPair(dentry)); ++ ++ di_read_unlock(parent, AuLock_IR); ++ dput(parent); ++ di_write_unlock(dentry); ++ ++ return err; ++} ++ ++static int test_dir(struct dentry *dentry, void *arg __maybe_unused) ++{ ++ return S_ISDIR(dentry->d_inode->i_mode); ++} ++ ++/* gave up consolidating with refresh_nondir() */ ++static int refresh_dir(struct dentry *root, unsigned int sigen) ++{ ++ int err, i, j, ndentry, e; ++ struct au_dcsub_pages dpages; ++ struct au_dpage *dpage; ++ struct dentry **dentries; ++ struct inode *inode; ++ const unsigned int flags = au_hi_flags(root->d_inode, /*isdir*/1); ++ ++ err = 0; ++ list_for_each_entry(inode, &root->d_sb->s_inodes, i_sb_list) ++ if (S_ISDIR(inode->i_mode) && au_iigen(inode) != sigen) { ++ ii_write_lock_child(inode); ++ e = au_refresh_hinode_self(inode, /*do_attr*/1); ++ ii_write_unlock(inode); ++ if (unlikely(e)) { ++ AuDbg("e %d, i%lu\n", e, inode->i_ino); ++ if (!err) ++ err = e; ++ /* go on even if err */ ++ } ++ } ++ ++ e = au_dpages_init(&dpages, GFP_NOFS); ++ if (unlikely(e)) { ++ if (!err) ++ err = e; ++ goto out; ++ } ++ e = au_dcsub_pages(&dpages, root, test_dir, NULL); ++ if (unlikely(e)) { ++ if (!err) ++ err = e; ++ goto out_dpages; ++ } ++ ++ for (i = 0; !e && i < dpages.ndpage; i++) { ++ dpage = dpages.dpages + i; ++ dentries = dpage->dentries; ++ ndentry = dpage->ndentry; ++ for (j = 0; !e && j < ndentry; j++) { ++ struct dentry *d; ++ ++ d = dentries[j]; ++ au_dbg_verify_dir_parent(d, sigen); ++ if (au_digen(d) != sigen) { ++ e = do_refresh(d, S_IFDIR, flags); ++ if (unlikely(e && !err)) ++ err = e; ++ /* break on err */ ++ } ++ } ++ } ++ ++ out_dpages: ++ au_dpages_free(&dpages); ++ out: ++ return err; ++} ++ ++static int test_nondir(struct dentry *dentry, void *arg __maybe_unused) ++{ ++ return !S_ISDIR(dentry->d_inode->i_mode); ++} ++ ++static int refresh_nondir(struct dentry *root, unsigned int sigen, ++ int do_dentry) ++{ ++ int err, i, j, ndentry, e; ++ struct au_dcsub_pages dpages; ++ struct au_dpage *dpage; ++ struct dentry **dentries; ++ struct inode *inode; ++ ++ err = 0; ++ list_for_each_entry(inode, &root->d_sb->s_inodes, i_sb_list) ++ if (!S_ISDIR(inode->i_mode) && au_iigen(inode) != sigen) { ++ ii_write_lock_child(inode); ++ e = au_refresh_hinode_self(inode, /*do_attr*/1); ++ ii_write_unlock(inode); ++ if (unlikely(e)) { ++ AuDbg("e %d, i%lu\n", e, inode->i_ino); ++ if (!err) ++ err = e; ++ /* go on even if err */ ++ } ++ } ++ ++ if (!do_dentry) ++ goto out; ++ ++ e = au_dpages_init(&dpages, GFP_NOFS); ++ if (unlikely(e)) { ++ if (!err) ++ err = e; ++ goto out; ++ } ++ e = au_dcsub_pages(&dpages, root, test_nondir, NULL); ++ if (unlikely(e)) { ++ if (!err) ++ err = e; ++ goto out_dpages; ++ } ++ ++ for (i = 0; i < dpages.ndpage; i++) { ++ dpage = dpages.dpages + i; ++ dentries = dpage->dentries; ++ ndentry = dpage->ndentry; ++ for (j = 0; j < ndentry; j++) { ++ struct dentry *d; ++ ++ d = dentries[j]; ++ au_dbg_verify_nondir_parent(d, sigen); ++ inode = d->d_inode; ++ if (inode && au_digen(d) != sigen) { ++ e = do_refresh(d, inode->i_mode & S_IFMT, ++ /*dir_flags*/0); ++ if (unlikely(e && !err)) ++ err = e; ++ /* go on even err */ ++ } ++ } ++ } ++ ++ out_dpages: ++ au_dpages_free(&dpages); ++ out: ++ return err; ++} ++ ++static void au_remount_refresh(struct super_block *sb, unsigned int flags) ++{ ++ int err; ++ unsigned int sigen; ++ struct au_sbinfo *sbinfo; ++ struct dentry *root; ++ struct inode *inode; ++ ++ au_sigen_inc(sb); ++ sigen = au_sigen(sb); ++ sbinfo = au_sbi(sb); ++ au_fclr_si(sbinfo, FAILED_REFRESH_DIRS); ++ ++ root = sb->s_root; ++ DiMustNoWaiters(root); ++ inode = root->d_inode; ++ IiMustNoWaiters(inode); ++ au_reset_hinotify(inode, au_hi_flags(inode, /*isdir*/1)); ++ di_write_unlock(root); ++ ++ err = refresh_dir(root, sigen); ++ if (unlikely(err)) { ++ au_fset_si(sbinfo, FAILED_REFRESH_DIRS); ++ AuWarn("Refreshing directories failed, ignored (%d)\n", err); ++ } ++ ++ if (au_ftest_opts(flags, REFRESH_NONDIR)) { ++ err = refresh_nondir(root, sigen, !err); ++ if (unlikely(err)) ++ AuWarn("Refreshing non-directories failed, ignored" ++ "(%d)\n", err); ++ } ++ ++ /* aufs_write_lock() calls ..._child() */ ++ di_write_lock_child(root); ++ au_cpup_attr_all(root->d_inode, /*force*/1); ++} ++ ++/* stop extra interpretation of errno in mount(8), and strange error messages */ ++static int cvt_err(int err) ++{ ++ AuTraceErr(err); ++ ++ switch (err) { ++ case -ENOENT: ++ case -ENOTDIR: ++ case -EEXIST: ++ case -EIO: ++ err = -EINVAL; ++ } ++ return err; ++} ++ ++static int aufs_remount_fs(struct super_block *sb, int *flags, char *data) ++{ ++ int err; ++ struct au_opts opts; ++ struct dentry *root; ++ struct inode *inode; ++ struct au_sbinfo *sbinfo; ++ ++ err = 0; ++ root = sb->s_root; ++ if (!data || !*data) { ++ aufs_write_lock(root); ++ err = au_opts_verify(sb, *flags, /*pending*/0); ++ if (!err) ++ au_fsync_br(sb); ++ aufs_write_unlock(root); ++ goto out; ++ } ++ ++ err = -ENOMEM; ++ memset(&opts, 0, sizeof(opts)); ++ opts.opt = (void *)__get_free_page(GFP_NOFS); ++ if (unlikely(!opts.opt)) ++ goto out; ++ opts.max_opt = PAGE_SIZE / sizeof(*opts.opt); ++ opts.flags = AuOpts_REMOUNT; ++ opts.sb_flags = *flags; ++ ++ /* parse it before aufs lock */ ++ err = au_opts_parse(sb, data, &opts); ++ if (unlikely(err)) ++ goto out_opts; ++ ++ sbinfo = au_sbi(sb); ++ inode = root->d_inode; ++ mutex_lock(&inode->i_mutex); ++ aufs_write_lock(root); ++ au_fsync_br(sb); ++ ++ /* au_opts_remount() may return an error */ ++ err = au_opts_remount(sb, &opts); ++ au_opts_free(&opts); ++ ++ if (au_ftest_opts(opts.flags, REFRESH_DIR) ++ || au_ftest_opts(opts.flags, REFRESH_NONDIR)) ++ au_remount_refresh(sb, opts.flags); ++ ++ aufs_write_unlock(root); ++ mutex_unlock(&inode->i_mutex); ++ ++ out_opts: ++ free_page((unsigned long)opts.opt); ++ out: ++ err = cvt_err(err); ++ AuTraceErr(err); ++ return err; ++} ++ ++static struct super_operations aufs_sop = { ++ .alloc_inode = aufs_alloc_inode, ++ .destroy_inode = aufs_destroy_inode, ++ .drop_inode = generic_delete_inode, ++ .show_options = aufs_show_options, ++ .statfs = aufs_statfs, ++ .put_super = aufs_put_super, ++ .remount_fs = aufs_remount_fs ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int alloc_root(struct super_block *sb) ++{ ++ int err; ++ struct inode *inode; ++ struct dentry *root; ++ ++ err = -ENOMEM; ++ inode = au_iget_locked(sb, AUFS_ROOT_INO); ++ err = PTR_ERR(inode); ++ if (IS_ERR(inode)) ++ goto out; ++ ++ inode->i_op = &aufs_dir_iop; ++ inode->i_fop = &aufs_dir_fop; ++ inode->i_mode = S_IFDIR; ++ inode->i_nlink = 2; ++ unlock_new_inode(inode); ++ ++ root = d_alloc_root(inode); ++ if (unlikely(!root)) ++ goto out_iput; ++ err = PTR_ERR(root); ++ if (IS_ERR(root)) ++ goto out_iput; ++ ++ err = au_alloc_dinfo(root); ++ if (!err) { ++ sb->s_root = root; ++ return 0; /* success */ ++ } ++ dput(root); ++ goto out; /* do not iput */ ++ ++ out_iput: ++ iget_failed(inode); ++ iput(inode); ++ out: ++ return err; ++ ++} ++ ++static int aufs_fill_super(struct super_block *sb, void *raw_data, ++ int silent __maybe_unused) ++{ ++ int err; ++ struct au_opts opts; ++ struct dentry *root; ++ struct inode *inode; ++ char *arg = raw_data; ++ ++ if (unlikely(!arg || !*arg)) { ++ err = -EINVAL; ++ AuErr("no arg\n"); ++ goto out; ++ } ++ ++ err = -ENOMEM; ++ memset(&opts, 0, sizeof(opts)); ++ opts.opt = (void *)__get_free_page(GFP_NOFS); ++ if (unlikely(!opts.opt)) ++ goto out; ++ opts.max_opt = PAGE_SIZE / sizeof(*opts.opt); ++ opts.sb_flags = sb->s_flags; ++ ++ err = au_si_alloc(sb); ++ if (unlikely(err)) ++ goto out_opts; ++ ++ /* all timestamps always follow the ones on the branch */ ++ sb->s_flags |= MS_NOATIME | MS_NODIRATIME; ++ sb->s_op = &aufs_sop; ++ sb->s_magic = AUFS_SUPER_MAGIC; ++ sb->s_maxbytes = 0; ++ au_export_init(sb); ++ ++ err = alloc_root(sb); ++ if (unlikely(err)) { ++ si_write_unlock(sb); ++ goto out_info; ++ } ++ root = sb->s_root; ++ inode = root->d_inode; ++ ++ /* ++ * actually we can parse options regardless aufs lock here. ++ * but at remount time, parsing must be done before aufs lock. ++ * so we follow the same rule. ++ */ ++ ii_write_lock_parent(inode); ++ aufs_write_unlock(root); ++ err = au_opts_parse(sb, arg, &opts); ++ if (unlikely(err)) ++ goto out_root; ++ ++ /* lock vfs_inode first, then aufs. */ ++ mutex_lock(&inode->i_mutex); ++ inode->i_op = &aufs_dir_iop; ++ inode->i_fop = &aufs_dir_fop; ++ aufs_write_lock(root); ++ err = au_opts_mount(sb, &opts); ++ au_opts_free(&opts); ++ if (unlikely(err)) ++ goto out_unlock; ++ aufs_write_unlock(root); ++ mutex_unlock(&inode->i_mutex); ++ goto out_opts; /* success */ ++ ++ out_unlock: ++ aufs_write_unlock(root); ++ mutex_unlock(&inode->i_mutex); ++ out_root: ++ dput(root); ++ sb->s_root = NULL; ++ out_info: ++ kobject_put(&au_sbi(sb)->si_kobj); ++ sb->s_fs_info = NULL; ++ out_opts: ++ free_page((unsigned long)opts.opt); ++ out: ++ AuTraceErr(err); ++ err = cvt_err(err); ++ AuTraceErr(err); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int aufs_get_sb(struct file_system_type *fs_type, int flags, ++ const char *dev_name __maybe_unused, void *raw_data, ++ struct vfsmount *mnt) ++{ ++ int err; ++ struct super_block *sb; ++ ++ /* all timestamps always follow the ones on the branch */ ++ /* mnt->mnt_flags |= MNT_NOATIME | MNT_NODIRATIME; */ ++ err = get_sb_nodev(fs_type, flags, raw_data, aufs_fill_super, mnt); ++ if (!err) { ++ sb = mnt->mnt_sb; ++ si_write_lock(sb); ++ sysaufs_brs_add(sb, 0); ++ si_write_unlock(sb); ++ } ++ return err; ++} ++ ++struct file_system_type aufs_fs_type = { ++ .name = AUFS_FSTYPE, ++ .fs_flags = ++ FS_RENAME_DOES_D_MOVE /* a race between rename and others */ ++ | FS_REVAL_DOT, /* for NFS branch and udba */ ++ .get_sb = aufs_get_sb, ++ .kill_sb = generic_shutdown_super, ++ /* no need to __module_get() and module_put(). */ ++ .owner = THIS_MODULE, ++}; +diff -Nur linux-2.6.31-vanilla/fs/aufs/super.h linux-2.6.31/fs/aufs/super.h +--- linux-2.6.31-vanilla/fs/aufs/super.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/super.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,384 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * super_block operations ++ */ ++ ++#ifndef __AUFS_SUPER_H__ ++#define __AUFS_SUPER_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++#include "rwsem.h" ++#include "spl.h" ++#include "wkq.h" ++ ++typedef ssize_t (*au_readf_t)(struct file *, char __user *, size_t, loff_t *); ++typedef ssize_t (*au_writef_t)(struct file *, const char __user *, size_t, ++ loff_t *); ++ ++/* policies to select one among multiple writable branches */ ++struct au_wbr_copyup_operations { ++ int (*copyup)(struct dentry *dentry); ++}; ++ ++struct au_wbr_create_operations { ++ int (*create)(struct dentry *dentry, int isdir); ++ int (*init)(struct super_block *sb); ++ int (*fin)(struct super_block *sb); ++}; ++ ++struct au_wbr_mfs { ++ struct mutex mfs_lock; /* protect this structure */ ++ unsigned long mfs_jiffy; ++ unsigned long mfs_expire; ++ aufs_bindex_t mfs_bindex; ++ ++ unsigned long long mfsrr_bytes; ++ unsigned long long mfsrr_watermark; ++}; ++ ++struct au_branch; ++struct au_sbinfo { ++ /* nowait tasks in the system-wide workqueue */ ++ struct au_nowait_tasks si_nowait; ++ ++ struct au_rwsem si_rwsem; ++ ++ /* branch management */ ++ unsigned int si_generation; ++ ++ /* see above flags */ ++ unsigned char au_si_status; ++ ++ aufs_bindex_t si_bend; ++ aufs_bindex_t si_last_br_id; ++ struct au_branch **si_branch; ++ ++ /* policy to select a writable branch */ ++ unsigned char si_wbr_copyup; ++ unsigned char si_wbr_create; ++ struct au_wbr_copyup_operations *si_wbr_copyup_ops; ++ struct au_wbr_create_operations *si_wbr_create_ops; ++ ++ /* round robin */ ++ atomic_t si_wbr_rr_next; ++ ++ /* most free space */ ++ struct au_wbr_mfs si_wbr_mfs; ++ ++ /* mount flags */ ++ /* include/asm-ia64/siginfo.h defines a macro named si_flags */ ++ unsigned int si_mntflags; ++ ++ /* external inode number (bitmap and translation table) */ ++ au_readf_t si_xread; ++ au_writef_t si_xwrite; ++ struct file *si_xib; ++ struct mutex si_xib_mtx; /* protect xib members */ ++ unsigned long *si_xib_buf; ++ unsigned long si_xib_last_pindex; ++ int si_xib_next_bit; ++ aufs_bindex_t si_xino_brid; ++ /* reserved for future use */ ++ /* unsigned long long si_xib_limit; */ /* Max xib file size */ ++ ++#ifdef CONFIG_AUFS_EXPORT ++ /* i_generation */ ++ struct file *si_xigen; ++ atomic_t si_xigen_next; ++#endif ++ ++ /* vdir parameters */ ++ unsigned long si_rdcache; /* max cache time in HZ */ ++ unsigned int si_rdblk; /* deblk size */ ++ unsigned int si_rdhash; /* hash size */ ++ ++ /* ++ * If the number of whiteouts are larger than si_dirwh, leave all of ++ * them after au_whtmp_ren to reduce the cost of rmdir(2). ++ * future fsck.aufs or kernel thread will remove them later. ++ * Otherwise, remove all whiteouts and the dir in rmdir(2). ++ */ ++ unsigned int si_dirwh; ++ ++ /* ++ * rename(2) a directory with all children. ++ */ ++ /* reserved for future use */ ++ /* int si_rendir; */ ++ ++ /* pseudo_link list */ ++ struct au_splhead si_plink; ++ wait_queue_head_t si_plink_wq; ++ ++ /* ++ * sysfs and lifetime management. ++ * this is not a small structure and it may be a waste of memory in case ++ * of sysfs is disabled, particulary when many aufs-es are mounted. ++ * but using sysfs is majority. ++ */ ++ struct kobject si_kobj; ++#ifdef CONFIG_DEBUG_FS ++ struct dentry *si_dbgaufs, *si_dbgaufs_xib; ++#ifdef CONFIG_AUFS_EXPORT ++ struct dentry *si_dbgaufs_xigen; ++#endif ++#endif ++ ++ /* dirty, necessary for unmounting, sysfs and sysrq */ ++ struct super_block *si_sb; ++}; ++ ++/* sbinfo status flags */ ++/* ++ * set true when refresh_dirs() failed at remount time. ++ * then try refreshing dirs at access time again. ++ * if it is false, refreshing dirs at access time is unnecesary ++ */ ++#define AuSi_FAILED_REFRESH_DIRS 1 ++#define AuSi_MAINTAIN_PLINK (1 << 1) /* ioctl */ ++static inline unsigned char au_do_ftest_si(struct au_sbinfo *sbi, ++ unsigned int flag) ++{ ++ AuRwMustAnyLock(&sbi->si_rwsem); ++ return sbi->au_si_status & flag; ++} ++#define au_ftest_si(sbinfo, name) au_do_ftest_si(sbinfo, AuSi_##name) ++#define au_fset_si(sbinfo, name) do { \ ++ AuRwMustWriteLock(&(sbinfo)->si_rwsem); \ ++ (sbinfo)->au_si_status |= AuSi_##name; \ ++} while (0) ++#define au_fclr_si(sbinfo, name) do { \ ++ AuRwMustWriteLock(&(sbinfo)->si_rwsem); \ ++ (sbinfo)->au_si_status &= ~AuSi_##name; \ ++} while (0) ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* policy to select one among writable branches */ ++#define AuWbrCopyup(sbinfo, args...) \ ++ ((sbinfo)->si_wbr_copyup_ops->copyup(args)) ++#define AuWbrCreate(sbinfo, args...) \ ++ ((sbinfo)->si_wbr_create_ops->create(args)) ++ ++/* flags for si_read_lock()/aufs_read_lock()/di_read_lock() */ ++#define AuLock_DW 1 /* write-lock dentry */ ++#define AuLock_IR (1 << 1) /* read-lock inode */ ++#define AuLock_IW (1 << 2) /* write-lock inode */ ++#define AuLock_FLUSH (1 << 3) /* wait for 'nowait' tasks */ ++#define AuLock_DIR (1 << 4) /* target is a dir */ ++#define au_ftest_lock(flags, name) ((flags) & AuLock_##name) ++#define au_fset_lock(flags, name) { (flags) |= AuLock_##name; } ++#define au_fclr_lock(flags, name) { (flags) &= ~AuLock_##name; } ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* super.c */ ++extern struct file_system_type aufs_fs_type; ++struct inode *au_iget_locked(struct super_block *sb, ino_t ino); ++ ++/* sbinfo.c */ ++void au_si_free(struct kobject *kobj); ++int au_si_alloc(struct super_block *sb); ++int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr); ++ ++unsigned int au_sigen_inc(struct super_block *sb); ++aufs_bindex_t au_new_br_id(struct super_block *sb); ++ ++void aufs_read_lock(struct dentry *dentry, int flags); ++void aufs_read_unlock(struct dentry *dentry, int flags); ++void aufs_write_lock(struct dentry *dentry); ++void aufs_write_unlock(struct dentry *dentry); ++void aufs_read_and_write_lock2(struct dentry *d1, struct dentry *d2, int isdir); ++void aufs_read_and_write_unlock2(struct dentry *d1, struct dentry *d2); ++ ++/* wbr_policy.c */ ++extern struct au_wbr_copyup_operations au_wbr_copyup_ops[]; ++extern struct au_wbr_create_operations au_wbr_create_ops[]; ++int au_cpdown_dirs(struct dentry *dentry, aufs_bindex_t bdst); ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline struct au_sbinfo *au_sbi(struct super_block *sb) ++{ ++ return sb->s_fs_info; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++#ifdef CONFIG_AUFS_EXPORT ++void au_export_init(struct super_block *sb); ++ ++static inline int au_test_nfsd(struct task_struct *tsk) ++{ ++ return !tsk->mm && !strcmp(tsk->comm, "nfsd"); ++} ++ ++int au_xigen_inc(struct inode *inode); ++int au_xigen_new(struct inode *inode); ++int au_xigen_set(struct super_block *sb, struct file *base); ++void au_xigen_clr(struct super_block *sb); ++ ++static inline int au_busy_or_stale(void) ++{ ++ if (!au_test_nfsd(current)) ++ return -EBUSY; ++ return -ESTALE; ++} ++#else ++static inline void au_export_init(struct super_block *sb) ++{ ++ /* nothing */ ++} ++ ++static inline int au_test_nfsd(struct task_struct *tsk) ++{ ++ return 0; ++} ++ ++static inline int au_xigen_inc(struct inode *inode) ++{ ++ return 0; ++} ++ ++static inline int au_xigen_new(struct inode *inode) ++{ ++ return 0; ++} ++ ++static inline int au_xigen_set(struct super_block *sb, struct file *base) ++{ ++ return 0; ++} ++ ++static inline void au_xigen_clr(struct super_block *sb) ++{ ++ /* empty */ ++} ++ ++static inline int au_busy_or_stale(void) ++{ ++ return -EBUSY; ++} ++#endif /* CONFIG_AUFS_EXPORT */ ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline void dbgaufs_si_null(struct au_sbinfo *sbinfo) ++{ ++ /* ++ * This function is a dynamic '__init' fucntion actually, ++ * so the tiny check for si_rwsem is unnecessary. ++ */ ++ /* AuRwMustWriteLock(&sbinfo->si_rwsem); */ ++#ifdef CONFIG_DEBUG_FS ++ sbinfo->si_dbgaufs = NULL; ++ sbinfo->si_dbgaufs_xib = NULL; ++#ifdef CONFIG_AUFS_EXPORT ++ sbinfo->si_dbgaufs_xigen = NULL; ++#endif ++#endif ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* lock superblock. mainly for entry point functions */ ++/* ++ * si_noflush_read_lock, si_noflush_write_lock, ++ * si_read_unlock, si_write_unlock, si_downgrade_lock ++ */ ++AuSimpleLockRwsemFuncs(si_noflush, struct super_block *sb, ++ &au_sbi(sb)->si_rwsem); ++AuSimpleUnlockRwsemFuncs(si, struct super_block *sb, &au_sbi(sb)->si_rwsem); ++ ++#define SiMustNoWaiters(sb) AuRwMustNoWaiters(&au_sbi(sb)->si_rwsem) ++#define SiMustAnyLock(sb) AuRwMustAnyLock(&au_sbi(sb)->si_rwsem) ++#define SiMustWriteLock(sb) AuRwMustWriteLock(&au_sbi(sb)->si_rwsem) ++ ++static inline void si_read_lock(struct super_block *sb, int flags) ++{ ++ if (au_ftest_lock(flags, FLUSH)) ++ au_nwt_flush(&au_sbi(sb)->si_nowait); ++ si_noflush_read_lock(sb); ++} ++ ++static inline void si_write_lock(struct super_block *sb) ++{ ++ au_nwt_flush(&au_sbi(sb)->si_nowait); ++ si_noflush_write_lock(sb); ++} ++ ++static inline int si_read_trylock(struct super_block *sb, int flags) ++{ ++ if (au_ftest_lock(flags, FLUSH)) ++ au_nwt_flush(&au_sbi(sb)->si_nowait); ++ return si_noflush_read_trylock(sb); ++} ++ ++static inline int si_write_trylock(struct super_block *sb, int flags) ++{ ++ if (au_ftest_lock(flags, FLUSH)) ++ au_nwt_flush(&au_sbi(sb)->si_nowait); ++ return si_noflush_write_trylock(sb); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline aufs_bindex_t au_sbend(struct super_block *sb) ++{ ++ SiMustAnyLock(sb); ++ return au_sbi(sb)->si_bend; ++} ++ ++static inline unsigned int au_mntflags(struct super_block *sb) ++{ ++ SiMustAnyLock(sb); ++ return au_sbi(sb)->si_mntflags; ++} ++ ++static inline unsigned int au_sigen(struct super_block *sb) ++{ ++ SiMustAnyLock(sb); ++ return au_sbi(sb)->si_generation; ++} ++ ++static inline struct au_branch *au_sbr(struct super_block *sb, ++ aufs_bindex_t bindex) ++{ ++ SiMustAnyLock(sb); ++ return au_sbi(sb)->si_branch[0 + bindex]; ++} ++ ++static inline void au_xino_brid_set(struct super_block *sb, aufs_bindex_t brid) ++{ ++ SiMustWriteLock(sb); ++ au_sbi(sb)->si_xino_brid = brid; ++} ++ ++static inline aufs_bindex_t au_xino_brid(struct super_block *sb) ++{ ++ SiMustAnyLock(sb); ++ return au_sbi(sb)->si_xino_brid; ++} ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_SUPER_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/sysaufs.c linux-2.6.31/fs/aufs/sysaufs.c +--- linux-2.6.31-vanilla/fs/aufs/sysaufs.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/sysaufs.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,104 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * sysfs interface and lifetime management ++ * they are necessary regardless sysfs is disabled. ++ */ ++ ++#include ++#include ++#include ++#include "aufs.h" ++ ++unsigned long sysaufs_si_mask; ++struct kset *sysaufs_ket; ++ ++#define AuSiAttr(_name) { \ ++ .attr = { .name = __stringify(_name), .mode = 0444 }, \ ++ .show = sysaufs_si_##_name, \ ++} ++ ++static struct sysaufs_si_attr sysaufs_si_attr_xi_path = AuSiAttr(xi_path); ++struct attribute *sysaufs_si_attrs[] = { ++ &sysaufs_si_attr_xi_path.attr, ++ NULL, ++}; ++ ++static struct sysfs_ops au_sbi_ops = { ++ .show = sysaufs_si_show ++}; ++ ++static struct kobj_type au_sbi_ktype = { ++ .release = au_si_free, ++ .sysfs_ops = &au_sbi_ops, ++ .default_attrs = sysaufs_si_attrs ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++int sysaufs_si_init(struct au_sbinfo *sbinfo) ++{ ++ int err; ++ ++ sbinfo->si_kobj.kset = sysaufs_ket; ++ /* cf. sysaufs_name() */ ++ err = kobject_init_and_add ++ (&sbinfo->si_kobj, &au_sbi_ktype, /*&sysaufs_ket->kobj*/NULL, ++ SysaufsSiNamePrefix "%lx", sysaufs_si_id(sbinfo)); ++ ++ dbgaufs_si_null(sbinfo); ++ if (!err) { ++ err = dbgaufs_si_init(sbinfo); ++ if (unlikely(err)) ++ kobject_put(&sbinfo->si_kobj); ++ } ++ return err; ++} ++ ++void sysaufs_fin(void) ++{ ++ dbgaufs_fin(); ++ sysfs_remove_group(&sysaufs_ket->kobj, sysaufs_attr_group); ++ kset_unregister(sysaufs_ket); ++} ++ ++int __init sysaufs_init(void) ++{ ++ int err; ++ ++ do { ++ get_random_bytes(&sysaufs_si_mask, sizeof(sysaufs_si_mask)); ++ } while (!sysaufs_si_mask); ++ ++ sysaufs_ket = kset_create_and_add(AUFS_NAME, NULL, fs_kobj); ++ err = PTR_ERR(sysaufs_ket); ++ if (IS_ERR(sysaufs_ket)) ++ goto out; ++ err = sysfs_create_group(&sysaufs_ket->kobj, sysaufs_attr_group); ++ if (unlikely(err)) { ++ kset_unregister(sysaufs_ket); ++ goto out; ++ } ++ ++ err = dbgaufs_init(); ++ if (unlikely(err)) ++ sysaufs_fin(); ++ out: ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/sysaufs.h linux-2.6.31/fs/aufs/sysaufs.h +--- linux-2.6.31-vanilla/fs/aufs/sysaufs.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/sysaufs.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,120 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * sysfs interface and mount lifetime management ++ */ ++ ++#ifndef __SYSAUFS_H__ ++#define __SYSAUFS_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++#include "module.h" ++ ++struct super_block; ++struct au_sbinfo; ++ ++struct sysaufs_si_attr { ++ struct attribute attr; ++ int (*show)(struct seq_file *seq, struct super_block *sb); ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* sysaufs.c */ ++extern unsigned long sysaufs_si_mask; ++extern struct kset *sysaufs_ket; ++extern struct attribute *sysaufs_si_attrs[]; ++int sysaufs_si_init(struct au_sbinfo *sbinfo); ++int __init sysaufs_init(void); ++void sysaufs_fin(void); ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* some people doesn't like to show a pointer in kernel */ ++static inline unsigned long sysaufs_si_id(struct au_sbinfo *sbinfo) ++{ ++ return sysaufs_si_mask ^ (unsigned long)sbinfo; ++} ++ ++#define SysaufsSiNamePrefix "si_" ++#define SysaufsSiNameLen (sizeof(SysaufsSiNamePrefix) + 16) ++static inline void sysaufs_name(struct au_sbinfo *sbinfo, char *name) ++{ ++ snprintf(name, SysaufsSiNameLen, SysaufsSiNamePrefix "%lx", ++ sysaufs_si_id(sbinfo)); ++} ++ ++struct au_branch; ++#ifdef CONFIG_SYSFS ++/* sysfs.c */ ++extern struct attribute_group *sysaufs_attr_group; ++ ++int sysaufs_si_xi_path(struct seq_file *seq, struct super_block *sb); ++ssize_t sysaufs_si_show(struct kobject *kobj, struct attribute *attr, ++ char *buf); ++ ++void sysaufs_br_init(struct au_branch *br); ++void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex); ++void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex); ++ ++#define sysaufs_brs_init() do {} while (0) ++ ++#else ++#define sysaufs_attr_group NULL ++ ++static inline ++int sysaufs_si_xi_path(struct seq_file *seq, struct super_block *sb) ++{ ++ return 0; ++} ++ ++static inline ++ssize_t sysaufs_si_show(struct kobject *kobj, struct attribute *attr, ++ char *buf) ++{ ++ return 0; ++} ++ ++static inline void sysaufs_br_init(struct au_branch *br) ++{ ++ /* empty */ ++} ++ ++static inline void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ /* nothing */ ++} ++ ++static inline void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ /* nothing */ ++} ++ ++static inline void sysaufs_brs_init(void) ++{ ++ sysaufs_brs = 0; ++} ++ ++#endif /* CONFIG_SYSFS */ ++ ++#endif /* __KERNEL__ */ ++#endif /* __SYSAUFS_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/sysfs.c linux-2.6.31/fs/aufs/sysfs.c +--- linux-2.6.31-vanilla/fs/aufs/sysfs.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/sysfs.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,210 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * sysfs interface ++ */ ++ ++#include ++#include ++#include ++#include ++#include "aufs.h" ++ ++static struct attribute *au_attr[] = { ++ NULL, /* need to NULL terminate the list of attributes */ ++}; ++ ++static struct attribute_group sysaufs_attr_group_body = { ++ .attrs = au_attr ++}; ++ ++struct attribute_group *sysaufs_attr_group = &sysaufs_attr_group_body; ++ ++/* ---------------------------------------------------------------------- */ ++ ++int sysaufs_si_xi_path(struct seq_file *seq, struct super_block *sb) ++{ ++ int err; ++ ++ SiMustAnyLock(sb); ++ ++ err = 0; ++ if (au_opt_test(au_mntflags(sb), XINO)) { ++ err = au_xino_path(seq, au_sbi(sb)->si_xib); ++ seq_putc(seq, '\n'); ++ } ++ return err; ++} ++ ++/* ++ * the lifetime of branch is independent from the entry under sysfs. ++ * sysfs handles the lifetime of the entry, and never call ->show() after it is ++ * unlinked. ++ */ ++static int sysaufs_si_br(struct seq_file *seq, struct super_block *sb, ++ aufs_bindex_t bindex) ++{ ++ struct path path; ++ struct dentry *root; ++ struct au_branch *br; ++ ++ AuDbg("b%d\n", bindex); ++ ++ root = sb->s_root; ++ di_read_lock_parent(root, !AuLock_IR); ++ br = au_sbr(sb, bindex); ++ path.mnt = br->br_mnt; ++ path.dentry = au_h_dptr(root, bindex); ++ au_seq_path(seq, &path); ++ di_read_unlock(root, !AuLock_IR); ++ seq_printf(seq, "=%s\n", au_optstr_br_perm(br->br_perm)); ++ return 0; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static struct seq_file *au_seq(char *p, ssize_t len) ++{ ++ struct seq_file *seq; ++ ++ seq = kzalloc(sizeof(*seq), GFP_NOFS); ++ if (seq) { ++ /* mutex_init(&seq.lock); */ ++ seq->buf = p; ++ seq->size = len; ++ return seq; /* success */ ++ } ++ ++ seq = ERR_PTR(-ENOMEM); ++ return seq; ++} ++ ++#define SysaufsBr_PREFIX "br" ++ ++/* todo: file size may exceed PAGE_SIZE */ ++ssize_t sysaufs_si_show(struct kobject *kobj, struct attribute *attr, ++ char *buf) ++{ ++ ssize_t err; ++ long l; ++ aufs_bindex_t bend; ++ struct au_sbinfo *sbinfo; ++ struct super_block *sb; ++ struct seq_file *seq; ++ char *name; ++ struct attribute **cattr; ++ ++ sbinfo = container_of(kobj, struct au_sbinfo, si_kobj); ++ sb = sbinfo->si_sb; ++ si_noflush_read_lock(sb); ++ ++ seq = au_seq(buf, PAGE_SIZE); ++ err = PTR_ERR(seq); ++ if (IS_ERR(seq)) ++ goto out; ++ ++ name = (void *)attr->name; ++ cattr = sysaufs_si_attrs; ++ while (*cattr) { ++ if (!strcmp(name, (*cattr)->name)) { ++ err = container_of(*cattr, struct sysaufs_si_attr, attr) ++ ->show(seq, sb); ++ goto out_seq; ++ } ++ cattr++; ++ } ++ ++ bend = au_sbend(sb); ++ if (!strncmp(name, SysaufsBr_PREFIX, sizeof(SysaufsBr_PREFIX) - 1)) { ++ name += sizeof(SysaufsBr_PREFIX) - 1; ++ err = strict_strtol(name, 10, &l); ++ if (!err) { ++ if (l <= bend) ++ err = sysaufs_si_br(seq, sb, (aufs_bindex_t)l); ++ else ++ err = -ENOENT; ++ } ++ goto out_seq; ++ } ++ BUG(); ++ ++ out_seq: ++ if (!err) { ++ err = seq->count; ++ /* sysfs limit */ ++ if (unlikely(err == PAGE_SIZE)) ++ err = -EFBIG; ++ } ++ kfree(seq); ++ out: ++ si_read_unlock(sb); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void sysaufs_br_init(struct au_branch *br) ++{ ++ br->br_attr.name = br->br_name; ++ br->br_attr.mode = S_IRUGO; ++ br->br_attr.owner = THIS_MODULE; ++} ++ ++void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ struct au_branch *br; ++ struct kobject *kobj; ++ aufs_bindex_t bend; ++ ++ dbgaufs_brs_del(sb, bindex); ++ ++ if (!sysaufs_brs) ++ return; ++ ++ kobj = &au_sbi(sb)->si_kobj; ++ bend = au_sbend(sb); ++ for (; bindex <= bend; bindex++) { ++ br = au_sbr(sb, bindex); ++ sysfs_remove_file(kobj, &br->br_attr); ++ } ++} ++ ++void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ int err; ++ aufs_bindex_t bend; ++ struct kobject *kobj; ++ struct au_branch *br; ++ ++ dbgaufs_brs_add(sb, bindex); ++ ++ if (!sysaufs_brs) ++ return; ++ ++ kobj = &au_sbi(sb)->si_kobj; ++ bend = au_sbend(sb); ++ for (; bindex <= bend; bindex++) { ++ br = au_sbr(sb, bindex); ++ snprintf(br->br_name, sizeof(br->br_name), SysaufsBr_PREFIX ++ "%d", bindex); ++ err = sysfs_create_file(kobj, &br->br_attr); ++ if (unlikely(err)) ++ AuWarn("failed %s under sysfs(%d)\n", br->br_name, err); ++ } ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/sysrq.c linux-2.6.31/fs/aufs/sysrq.c +--- linux-2.6.31-vanilla/fs/aufs/sysrq.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/sysrq.c 2009-09-16 13:55:29.000000000 +0200 +@@ -0,0 +1,115 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * magic sysrq hanlder ++ */ ++ ++#include ++#include ++#include ++/* #include */ ++#include "aufs.h" ++ ++/* ---------------------------------------------------------------------- */ ++ ++static void sysrq_sb(struct super_block *sb) ++{ ++ char *plevel; ++ struct au_sbinfo *sbinfo; ++ struct file *file; ++ ++ plevel = au_plevel; ++ au_plevel = KERN_WARNING; ++ au_debug(1); ++ ++ sbinfo = au_sbi(sb); ++ pr_warning("si=%lx\n", sysaufs_si_id(sbinfo)); ++ pr_warning(AUFS_NAME ": superblock\n"); ++ au_dpri_sb(sb); ++ pr_warning(AUFS_NAME ": root dentry\n"); ++ au_dpri_dentry(sb->s_root); ++ pr_warning(AUFS_NAME ": root inode\n"); ++ au_dpri_inode(sb->s_root->d_inode); ++#if 0 ++ struct inode *i; ++ pr_warning(AUFS_NAME ": isolated inode\n"); ++ list_for_each_entry(i, &sb->s_inodes, i_sb_list) ++ if (list_empty(&i->i_dentry)) ++ au_dpri_inode(i); ++#endif ++ pr_warning(AUFS_NAME ": files\n"); ++ list_for_each_entry(file, &sb->s_files, f_u.fu_list) ++ if (!special_file(file->f_dentry->d_inode->i_mode)) ++ au_dpri_file(file); ++ ++ au_plevel = plevel; ++ au_debug(0); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* module parameter */ ++static char *aufs_sysrq_key = "a"; ++module_param_named(sysrq, aufs_sysrq_key, charp, S_IRUGO); ++MODULE_PARM_DESC(sysrq, "MagicSysRq key for " AUFS_NAME); ++ ++static void au_sysrq(int key __maybe_unused, ++ struct tty_struct *tty __maybe_unused) ++{ ++ struct kobject *kobj; ++ struct au_sbinfo *sbinfo; ++ ++ /* spin_lock(&sysaufs_ket->list_lock); */ ++ list_for_each_entry(kobj, &sysaufs_ket->list, entry) { ++ sbinfo = container_of(kobj, struct au_sbinfo, si_kobj); ++ sysrq_sb(sbinfo->si_sb); ++ } ++ /* spin_unlock(&sysaufs_ket->list_lock); */ ++} ++ ++static struct sysrq_key_op au_sysrq_op = { ++ .handler = au_sysrq, ++ .help_msg = "Aufs", ++ .action_msg = "Aufs", ++ .enable_mask = SYSRQ_ENABLE_DUMP ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++int __init au_sysrq_init(void) ++{ ++ int err; ++ char key; ++ ++ err = -1; ++ key = *aufs_sysrq_key; ++ if ('a' <= key && key <= 'z') ++ err = register_sysrq_key(key, &au_sysrq_op); ++ if (unlikely(err)) ++ AuErr("err %d, sysrq=%c\n", err, key); ++ return err; ++} ++ ++void au_sysrq_fin(void) ++{ ++ int err; ++ err = unregister_sysrq_key(*aufs_sysrq_key, &au_sysrq_op); ++ if (unlikely(err)) ++ AuErr("err %d (ignored)\n", err); ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/vdir.c linux-2.6.31/fs/aufs/vdir.c +--- linux-2.6.31-vanilla/fs/aufs/vdir.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/vdir.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,879 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * virtual or vertical directory ++ */ ++ ++#include ++#include "aufs.h" ++ ++static unsigned int calc_size(int nlen) ++{ ++ BUILD_BUG_ON(sizeof(ino_t) != sizeof(long)); ++ return ALIGN(sizeof(struct au_vdir_de) + nlen, sizeof(ino_t)); ++} ++ ++static int set_deblk_end(union au_vdir_deblk_p *p, ++ union au_vdir_deblk_p *deblk_end) ++{ ++ if (calc_size(0) <= deblk_end->deblk - p->deblk) { ++ p->de->de_str.len = 0; ++ /* smp_mb(); */ ++ return 0; ++ } ++ return -1; /* error */ ++} ++ ++/* returns true or false */ ++static int is_deblk_end(union au_vdir_deblk_p *p, ++ union au_vdir_deblk_p *deblk_end) ++{ ++ if (calc_size(0) <= deblk_end->deblk - p->deblk) ++ return !p->de->de_str.len; ++ return 1; ++} ++ ++static unsigned char *last_deblk(struct au_vdir *vdir) ++{ ++ return vdir->vd_deblk[vdir->vd_nblk - 1]; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* estimate the apropriate size for name hash table */ ++unsigned int au_rdhash_est(loff_t sz) ++{ ++ unsigned int n; ++ ++ n = UINT_MAX; ++ sz >>= 10; ++ if (sz < n) ++ n = sz; ++ if (sz < AUFS_RDHASH_DEF) ++ n = AUFS_RDHASH_DEF; ++ /* AuInfo("n %u\n", n); */ ++ return n; ++} ++ ++/* ++ * the allocated memory has to be freed by ++ * au_nhash_wh_free() or au_nhash_de_free(). ++ */ ++int au_nhash_alloc(struct au_nhash *nhash, unsigned int num_hash, gfp_t gfp) ++{ ++ struct hlist_head *head; ++ unsigned int u; ++ ++ head = kmalloc(sizeof(*nhash->nh_head) * num_hash, gfp); ++ if (head) { ++ nhash->nh_num = num_hash; ++ nhash->nh_head = head; ++ for (u = 0; u < num_hash; u++) ++ INIT_HLIST_HEAD(head++); ++ return 0; /* success */ ++ } ++ ++ return -ENOMEM; ++} ++ ++static void nhash_count(struct hlist_head *head) ++{ ++#if 0 ++ unsigned long n; ++ struct hlist_node *pos; ++ ++ n = 0; ++ hlist_for_each(pos, head) ++ n++; ++ AuInfo("%lu\n", n); ++#endif ++} ++ ++static void au_nhash_wh_do_free(struct hlist_head *head) ++{ ++ struct au_vdir_wh *tpos; ++ struct hlist_node *pos, *node; ++ ++ hlist_for_each_entry_safe(tpos, pos, node, head, wh_hash) { ++ /* hlist_del(pos); */ ++ kfree(tpos); ++ } ++} ++ ++static void au_nhash_de_do_free(struct hlist_head *head) ++{ ++ struct au_vdir_dehstr *tpos; ++ struct hlist_node *pos, *node; ++ ++ hlist_for_each_entry_safe(tpos, pos, node, head, hash) { ++ /* hlist_del(pos); */ ++ au_cache_free_dehstr(tpos); ++ } ++} ++ ++static void au_nhash_do_free(struct au_nhash *nhash, ++ void (*free)(struct hlist_head *head)) ++{ ++ unsigned int n; ++ struct hlist_head *head; ++ ++ n = nhash->nh_num; ++ if (!n) ++ return; ++ ++ head = nhash->nh_head; ++ while (n-- > 0) { ++ nhash_count(head); ++ free(head++); ++ } ++ kfree(nhash->nh_head); ++} ++ ++void au_nhash_wh_free(struct au_nhash *whlist) ++{ ++ au_nhash_do_free(whlist, au_nhash_wh_do_free); ++} ++ ++static void au_nhash_de_free(struct au_nhash *delist) ++{ ++ au_nhash_do_free(delist, au_nhash_de_do_free); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int au_nhash_test_longer_wh(struct au_nhash *whlist, aufs_bindex_t btgt, ++ int limit) ++{ ++ int num; ++ unsigned int u, n; ++ struct hlist_head *head; ++ struct au_vdir_wh *tpos; ++ struct hlist_node *pos; ++ ++ num = 0; ++ n = whlist->nh_num; ++ head = whlist->nh_head; ++ for (u = 0; u < n; u++, head++) ++ hlist_for_each_entry(tpos, pos, head, wh_hash) ++ if (tpos->wh_bindex == btgt && ++num > limit) ++ return 1; ++ return 0; ++} ++ ++static struct hlist_head *au_name_hash(struct au_nhash *nhash, ++ unsigned char *name, ++ unsigned int len) ++{ ++ unsigned int v; ++ /* const unsigned int magic_bit = 12; */ ++ ++ AuDebugOn(!nhash->nh_num || !nhash->nh_head); ++ ++ v = 0; ++ while (len--) ++ v += *name++; ++ /* v = hash_long(v, magic_bit); */ ++ v %= nhash->nh_num; ++ return nhash->nh_head + v; ++} ++ ++static int au_nhash_test_name(struct au_vdir_destr *str, const char *name, ++ int nlen) ++{ ++ return str->len == nlen && !memcmp(str->name, name, nlen); ++} ++ ++/* returns found or not */ ++int au_nhash_test_known_wh(struct au_nhash *whlist, char *name, int nlen) ++{ ++ struct hlist_head *head; ++ struct au_vdir_wh *tpos; ++ struct hlist_node *pos; ++ struct au_vdir_destr *str; ++ ++ head = au_name_hash(whlist, name, nlen); ++ hlist_for_each_entry(tpos, pos, head, wh_hash) { ++ str = &tpos->wh_str; ++ AuDbg("%.*s\n", str->len, str->name); ++ if (au_nhash_test_name(str, name, nlen)) ++ return 1; ++ } ++ return 0; ++} ++ ++/* returns found(true) or not */ ++static int test_known(struct au_nhash *delist, char *name, int nlen) ++{ ++ struct hlist_head *head; ++ struct au_vdir_dehstr *tpos; ++ struct hlist_node *pos; ++ struct au_vdir_destr *str; ++ ++ head = au_name_hash(delist, name, nlen); ++ hlist_for_each_entry(tpos, pos, head, hash) { ++ str = tpos->str; ++ AuDbg("%.*s\n", str->len, str->name); ++ if (au_nhash_test_name(str, name, nlen)) ++ return 1; ++ } ++ return 0; ++} ++ ++static void au_shwh_init_wh(struct au_vdir_wh *wh, ino_t ino, ++ unsigned char d_type) ++{ ++#ifdef CONFIG_AUFS_SHWH ++ wh->wh_ino = ino; ++ wh->wh_type = d_type; ++#endif ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int au_nhash_append_wh(struct au_nhash *whlist, char *name, int nlen, ino_t ino, ++ unsigned int d_type, aufs_bindex_t bindex, ++ unsigned char shwh) ++{ ++ int err; ++ struct au_vdir_destr *str; ++ struct au_vdir_wh *wh; ++ ++ AuDbg("%.*s\n", nlen, name); ++ AuDebugOn(!whlist->nh_num || !whlist->nh_head); ++ ++ err = -ENOMEM; ++ wh = kmalloc(sizeof(*wh) + nlen, GFP_NOFS); ++ if (unlikely(!wh)) ++ goto out; ++ ++ err = 0; ++ wh->wh_bindex = bindex; ++ if (shwh) ++ au_shwh_init_wh(wh, ino, d_type); ++ str = &wh->wh_str; ++ str->len = nlen; ++ memcpy(str->name, name, nlen); ++ hlist_add_head(&wh->wh_hash, au_name_hash(whlist, name, nlen)); ++ /* smp_mb(); */ ++ ++ out: ++ return err; ++} ++ ++static int append_deblk(struct au_vdir *vdir) ++{ ++ int err; ++ unsigned long ul; ++ const unsigned int deblk_sz = vdir->vd_deblk_sz; ++ union au_vdir_deblk_p p, deblk_end; ++ unsigned char **o; ++ ++ err = -ENOMEM; ++ o = krealloc(vdir->vd_deblk, sizeof(*o) * (vdir->vd_nblk + 1), ++ GFP_NOFS); ++ if (unlikely(!o)) ++ goto out; ++ ++ vdir->vd_deblk = o; ++ p.deblk = kmalloc(deblk_sz, GFP_NOFS); ++ if (p.deblk) { ++ ul = vdir->vd_nblk++; ++ vdir->vd_deblk[ul] = p.deblk; ++ vdir->vd_last.ul = ul; ++ vdir->vd_last.p.deblk = p.deblk; ++ deblk_end.deblk = p.deblk + deblk_sz; ++ err = set_deblk_end(&p, &deblk_end); ++ } ++ ++ out: ++ return err; ++} ++ ++static int append_de(struct au_vdir *vdir, char *name, int nlen, ino_t ino, ++ unsigned int d_type, struct au_nhash *delist) ++{ ++ int err; ++ unsigned int sz; ++ const unsigned int deblk_sz = vdir->vd_deblk_sz; ++ union au_vdir_deblk_p p, *room, deblk_end; ++ struct au_vdir_dehstr *dehstr; ++ ++ p.deblk = last_deblk(vdir); ++ deblk_end.deblk = p.deblk + deblk_sz; ++ room = &vdir->vd_last.p; ++ AuDebugOn(room->deblk < p.deblk || deblk_end.deblk <= room->deblk ++ || !is_deblk_end(room, &deblk_end)); ++ ++ sz = calc_size(nlen); ++ if (unlikely(sz > deblk_end.deblk - room->deblk)) { ++ err = append_deblk(vdir); ++ if (unlikely(err)) ++ goto out; ++ ++ p.deblk = last_deblk(vdir); ++ deblk_end.deblk = p.deblk + deblk_sz; ++ /* smp_mb(); */ ++ AuDebugOn(room->deblk != p.deblk); ++ } ++ ++ err = -ENOMEM; ++ dehstr = au_cache_alloc_dehstr(); ++ if (unlikely(!dehstr)) ++ goto out; ++ ++ dehstr->str = &room->de->de_str; ++ hlist_add_head(&dehstr->hash, au_name_hash(delist, name, nlen)); ++ room->de->de_ino = ino; ++ room->de->de_type = d_type; ++ room->de->de_str.len = nlen; ++ memcpy(room->de->de_str.name, name, nlen); ++ ++ err = 0; ++ room->deblk += sz; ++ if (unlikely(set_deblk_end(room, &deblk_end))) ++ err = append_deblk(vdir); ++ /* smp_mb(); */ ++ ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void au_vdir_free(struct au_vdir *vdir) ++{ ++ unsigned char **deblk; ++ ++ deblk = vdir->vd_deblk; ++ while (vdir->vd_nblk--) ++ kfree(*deblk++); ++ kfree(vdir->vd_deblk); ++ au_cache_free_vdir(vdir); ++} ++ ++static struct au_vdir *alloc_vdir(struct file *file) ++{ ++ struct au_vdir *vdir; ++ struct super_block *sb; ++ int err; ++ ++ sb = file->f_dentry->d_sb; ++ SiMustAnyLock(sb); ++ ++ err = -ENOMEM; ++ vdir = au_cache_alloc_vdir(); ++ if (unlikely(!vdir)) ++ goto out; ++ ++ vdir->vd_deblk = kzalloc(sizeof(*vdir->vd_deblk), GFP_NOFS); ++ if (unlikely(!vdir->vd_deblk)) ++ goto out_free; ++ ++ vdir->vd_deblk_sz = au_sbi(sb)->si_rdblk; ++ if (!vdir->vd_deblk_sz) { ++ /* estimate the apropriate size for deblk */ ++ vdir->vd_deblk_sz = au_dir_size(file, /*dentry*/NULL); ++ /* AuInfo("vd_deblk_sz %u\n", vdir->vd_deblk_sz); */ ++ } ++ vdir->vd_nblk = 0; ++ vdir->vd_version = 0; ++ vdir->vd_jiffy = 0; ++ err = append_deblk(vdir); ++ if (!err) ++ return vdir; /* success */ ++ ++ kfree(vdir->vd_deblk); ++ ++ out_free: ++ au_cache_free_vdir(vdir); ++ out: ++ vdir = ERR_PTR(err); ++ return vdir; ++} ++ ++static int reinit_vdir(struct au_vdir *vdir) ++{ ++ int err; ++ union au_vdir_deblk_p p, deblk_end; ++ ++ while (vdir->vd_nblk > 1) { ++ kfree(vdir->vd_deblk[vdir->vd_nblk - 1]); ++ /* vdir->vd_deblk[vdir->vd_nblk - 1] = NULL; */ ++ vdir->vd_nblk--; ++ } ++ p.deblk = vdir->vd_deblk[0]; ++ deblk_end.deblk = p.deblk + vdir->vd_deblk_sz; ++ err = set_deblk_end(&p, &deblk_end); ++ /* keep vd_dblk_sz */ ++ vdir->vd_last.ul = 0; ++ vdir->vd_last.p.deblk = vdir->vd_deblk[0]; ++ vdir->vd_version = 0; ++ vdir->vd_jiffy = 0; ++ /* smp_mb(); */ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++#define AuFillVdir_CALLED 1 ++#define AuFillVdir_WHABLE (1 << 1) ++#define AuFillVdir_SHWH (1 << 2) ++#define au_ftest_fillvdir(flags, name) ((flags) & AuFillVdir_##name) ++#define au_fset_fillvdir(flags, name) { (flags) |= AuFillVdir_##name; } ++#define au_fclr_fillvdir(flags, name) { (flags) &= ~AuFillVdir_##name; } ++ ++#ifndef CONFIG_AUFS_SHWH ++#undef AuFillVdir_SHWH ++#define AuFillVdir_SHWH 0 ++#endif ++ ++struct fillvdir_arg { ++ struct file *file; ++ struct au_vdir *vdir; ++ struct au_nhash delist; ++ struct au_nhash whlist; ++ aufs_bindex_t bindex; ++ unsigned int flags; ++ int err; ++}; ++ ++static int fillvdir(void *__arg, const char *__name, int nlen, ++ loff_t offset __maybe_unused, u64 h_ino, ++ unsigned int d_type) ++{ ++ struct fillvdir_arg *arg = __arg; ++ char *name = (void *)__name; ++ struct super_block *sb; ++ ino_t ino; ++ const unsigned char shwh = !!au_ftest_fillvdir(arg->flags, SHWH); ++ ++ arg->err = 0; ++ sb = arg->file->f_dentry->d_sb; ++ au_fset_fillvdir(arg->flags, CALLED); ++ /* smp_mb(); */ ++ if (nlen <= AUFS_WH_PFX_LEN ++ || memcmp(name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) { ++ if (test_known(&arg->delist, name, nlen) ++ || au_nhash_test_known_wh(&arg->whlist, name, nlen)) ++ goto out; /* already exists or whiteouted */ ++ ++ sb = arg->file->f_dentry->d_sb; ++ arg->err = au_ino(sb, arg->bindex, h_ino, d_type, &ino); ++ if (!arg->err) ++ arg->err = append_de(arg->vdir, name, nlen, ino, ++ d_type, &arg->delist); ++ } else if (au_ftest_fillvdir(arg->flags, WHABLE)) { ++ name += AUFS_WH_PFX_LEN; ++ nlen -= AUFS_WH_PFX_LEN; ++ if (au_nhash_test_known_wh(&arg->whlist, name, nlen)) ++ goto out; /* already whiteouted */ ++ ++ if (shwh) ++ arg->err = au_wh_ino(sb, arg->bindex, h_ino, d_type, ++ &ino); ++ if (!arg->err) ++ arg->err = au_nhash_append_wh ++ (&arg->whlist, name, nlen, ino, d_type, ++ arg->bindex, shwh); ++ } ++ ++ out: ++ if (!arg->err) ++ arg->vdir->vd_jiffy = jiffies; ++ /* smp_mb(); */ ++ AuTraceErr(arg->err); ++ return arg->err; ++} ++ ++static int au_handle_shwh(struct super_block *sb, struct au_vdir *vdir, ++ struct au_nhash *whlist, struct au_nhash *delist) ++{ ++#ifdef CONFIG_AUFS_SHWH ++ int err; ++ unsigned int nh, u; ++ struct hlist_head *head; ++ struct au_vdir_wh *tpos; ++ struct hlist_node *pos, *n; ++ char *p, *o; ++ struct au_vdir_destr *destr; ++ ++ AuDebugOn(!au_opt_test(au_mntflags(sb), SHWH)); ++ ++ err = -ENOMEM; ++ o = p = __getname(); ++ if (unlikely(!p)) ++ goto out; ++ ++ err = 0; ++ nh = whlist->nh_num; ++ memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN); ++ p += AUFS_WH_PFX_LEN; ++ for (u = 0; u < nh; u++) { ++ head = whlist->nh_head + u; ++ hlist_for_each_entry_safe(tpos, pos, n, head, wh_hash) { ++ destr = &tpos->wh_str; ++ memcpy(p, destr->name, destr->len); ++ err = append_de(vdir, o, destr->len + AUFS_WH_PFX_LEN, ++ tpos->wh_ino, tpos->wh_type, delist); ++ if (unlikely(err)) ++ break; ++ } ++ } ++ ++ __putname(o); ++ ++ out: ++ AuTraceErr(err); ++ return err; ++#else ++ return 0; ++#endif ++} ++ ++static int au_do_read_vdir(struct fillvdir_arg *arg) ++{ ++ int err; ++ unsigned int rdhash; ++ loff_t offset; ++ aufs_bindex_t bend, bindex, bstart; ++ unsigned char shwh; ++ struct file *hf, *file; ++ struct super_block *sb; ++ ++ file = arg->file; ++ sb = file->f_dentry->d_sb; ++ SiMustAnyLock(sb); ++ ++ rdhash = au_sbi(sb)->si_rdhash; ++ if (!rdhash) ++ rdhash = au_rdhash_est(au_dir_size(file, /*dentry*/NULL)); ++ err = au_nhash_alloc(&arg->delist, rdhash, GFP_NOFS); ++ if (unlikely(err)) ++ goto out; ++ err = au_nhash_alloc(&arg->whlist, rdhash, GFP_NOFS); ++ if (unlikely(err)) ++ goto out_delist; ++ ++ err = 0; ++ arg->flags = 0; ++ shwh = 0; ++ if (au_opt_test(au_mntflags(sb), SHWH)) { ++ shwh = 1; ++ au_fset_fillvdir(arg->flags, SHWH); ++ } ++ bstart = au_fbstart(file); ++ bend = au_fbend(file); ++ for (bindex = bstart; !err && bindex <= bend; bindex++) { ++ hf = au_h_fptr(file, bindex); ++ if (!hf) ++ continue; ++ ++ offset = vfsub_llseek(hf, 0, SEEK_SET); ++ err = offset; ++ if (unlikely(offset)) ++ break; ++ ++ arg->bindex = bindex; ++ au_fclr_fillvdir(arg->flags, WHABLE); ++ if (shwh ++ || (bindex != bend ++ && au_br_whable(au_sbr_perm(sb, bindex)))) ++ au_fset_fillvdir(arg->flags, WHABLE); ++ do { ++ arg->err = 0; ++ au_fclr_fillvdir(arg->flags, CALLED); ++ /* smp_mb(); */ ++ err = vfsub_readdir(hf, fillvdir, arg); ++ if (err >= 0) ++ err = arg->err; ++ } while (!err && au_ftest_fillvdir(arg->flags, CALLED)); ++ } ++ ++ if (!err && shwh) ++ err = au_handle_shwh(sb, arg->vdir, &arg->whlist, &arg->delist); ++ ++ au_nhash_wh_free(&arg->whlist); ++ ++ out_delist: ++ au_nhash_de_free(&arg->delist); ++ out: ++ return err; ++} ++ ++static int read_vdir(struct file *file, int may_read) ++{ ++ int err; ++ unsigned long expire; ++ unsigned char do_read; ++ struct fillvdir_arg arg; ++ struct inode *inode; ++ struct au_vdir *vdir, *allocated; ++ ++ err = 0; ++ inode = file->f_dentry->d_inode; ++ IMustLock(inode); ++ SiMustAnyLock(inode->i_sb); ++ ++ allocated = NULL; ++ do_read = 0; ++ expire = au_sbi(inode->i_sb)->si_rdcache; ++ vdir = au_ivdir(inode); ++ if (!vdir) { ++ do_read = 1; ++ vdir = alloc_vdir(file); ++ err = PTR_ERR(vdir); ++ if (IS_ERR(vdir)) ++ goto out; ++ err = 0; ++ allocated = vdir; ++ } else if (may_read ++ && (inode->i_version != vdir->vd_version ++ || time_after(jiffies, vdir->vd_jiffy + expire))) { ++ do_read = 1; ++ err = reinit_vdir(vdir); ++ if (unlikely(err)) ++ goto out; ++ } ++ ++ if (!do_read) ++ return 0; /* success */ ++ ++ arg.file = file; ++ arg.vdir = vdir; ++ err = au_do_read_vdir(&arg); ++ if (!err) { ++ /* file->f_pos = 0; */ ++ vdir->vd_version = inode->i_version; ++ vdir->vd_last.ul = 0; ++ vdir->vd_last.p.deblk = vdir->vd_deblk[0]; ++ if (allocated) ++ au_set_ivdir(inode, allocated); ++ } else if (allocated) ++ au_vdir_free(allocated); ++ ++ out: ++ return err; ++} ++ ++static int copy_vdir(struct au_vdir *tgt, struct au_vdir *src) ++{ ++ int err, rerr; ++ unsigned long ul, n; ++ const unsigned int deblk_sz = src->vd_deblk_sz; ++ ++ AuDebugOn(tgt->vd_nblk != 1); ++ ++ err = -ENOMEM; ++ if (tgt->vd_nblk < src->vd_nblk) { ++ unsigned char **p; ++ ++ p = krealloc(tgt->vd_deblk, sizeof(*p) * src->vd_nblk, ++ GFP_NOFS); ++ if (unlikely(!p)) ++ goto out; ++ tgt->vd_deblk = p; ++ } ++ ++ if (tgt->vd_deblk_sz != deblk_sz) { ++ unsigned char *p; ++ ++ tgt->vd_deblk_sz = deblk_sz; ++ p = krealloc(tgt->vd_deblk[0], deblk_sz, GFP_NOFS); ++ if (unlikely(!p)) ++ goto out; ++ tgt->vd_deblk[0] = p; ++ } ++ memcpy(tgt->vd_deblk[0], src->vd_deblk[0], deblk_sz); ++ tgt->vd_version = src->vd_version; ++ tgt->vd_jiffy = src->vd_jiffy; ++ ++ n = src->vd_nblk; ++ for (ul = 1; ul < n; ul++) { ++ tgt->vd_deblk[ul] = kmemdup(src->vd_deblk[ul], deblk_sz, ++ GFP_NOFS); ++ if (unlikely(!tgt->vd_deblk[ul])) ++ goto out; ++ tgt->vd_nblk++; ++ } ++ tgt->vd_nblk = n; ++ tgt->vd_last.ul = tgt->vd_last.ul; ++ tgt->vd_last.p.deblk = tgt->vd_deblk[tgt->vd_last.ul]; ++ tgt->vd_last.p.deblk += src->vd_last.p.deblk ++ - src->vd_deblk[src->vd_last.ul]; ++ /* smp_mb(); */ ++ return 0; /* success */ ++ ++ out: ++ rerr = reinit_vdir(tgt); ++ BUG_ON(rerr); ++ return err; ++} ++ ++int au_vdir_init(struct file *file) ++{ ++ int err; ++ struct inode *inode; ++ struct au_vdir *vdir_cache, *allocated; ++ ++ err = read_vdir(file, !file->f_pos); ++ if (unlikely(err)) ++ goto out; ++ ++ allocated = NULL; ++ vdir_cache = au_fvdir_cache(file); ++ if (!vdir_cache) { ++ vdir_cache = alloc_vdir(file); ++ err = PTR_ERR(vdir_cache); ++ if (IS_ERR(vdir_cache)) ++ goto out; ++ allocated = vdir_cache; ++ } else if (!file->f_pos && vdir_cache->vd_version != file->f_version) { ++ err = reinit_vdir(vdir_cache); ++ if (unlikely(err)) ++ goto out; ++ } else ++ return 0; /* success */ ++ ++ inode = file->f_dentry->d_inode; ++ err = copy_vdir(vdir_cache, au_ivdir(inode)); ++ if (!err) { ++ file->f_version = inode->i_version; ++ if (allocated) ++ au_set_fvdir_cache(file, allocated); ++ } else if (allocated) ++ au_vdir_free(allocated); ++ ++ out: ++ return err; ++} ++ ++static loff_t calc_offset(struct au_vdir *vdir) ++{ ++ loff_t offset; ++ union au_vdir_deblk_p p; ++ ++ p.deblk = vdir->vd_deblk[vdir->vd_last.ul]; ++ offset = vdir->vd_last.p.deblk - p.deblk; ++ offset += vdir->vd_deblk_sz * vdir->vd_last.ul; ++ return offset; ++} ++ ++/* returns true or false */ ++static int seek_vdir(struct file *file) ++{ ++ int valid; ++ unsigned int deblk_sz; ++ unsigned long ul, n; ++ loff_t offset; ++ union au_vdir_deblk_p p, deblk_end; ++ struct au_vdir *vdir_cache; ++ ++ valid = 1; ++ vdir_cache = au_fvdir_cache(file); ++ offset = calc_offset(vdir_cache); ++ AuDbg("offset %lld\n", offset); ++ if (file->f_pos == offset) ++ goto out; ++ ++ vdir_cache->vd_last.ul = 0; ++ vdir_cache->vd_last.p.deblk = vdir_cache->vd_deblk[0]; ++ if (!file->f_pos) ++ goto out; ++ ++ valid = 0; ++ deblk_sz = vdir_cache->vd_deblk_sz; ++ ul = div64_u64(file->f_pos, deblk_sz); ++ AuDbg("ul %lu\n", ul); ++ if (ul >= vdir_cache->vd_nblk) ++ goto out; ++ ++ n = vdir_cache->vd_nblk; ++ for (; ul < n; ul++) { ++ p.deblk = vdir_cache->vd_deblk[ul]; ++ deblk_end.deblk = p.deblk + deblk_sz; ++ offset = ul; ++ offset *= deblk_sz; ++ while (!is_deblk_end(&p, &deblk_end) && offset < file->f_pos) { ++ unsigned int l; ++ ++ l = calc_size(p.de->de_str.len); ++ offset += l; ++ p.deblk += l; ++ } ++ if (!is_deblk_end(&p, &deblk_end)) { ++ valid = 1; ++ vdir_cache->vd_last.ul = ul; ++ vdir_cache->vd_last.p = p; ++ break; ++ } ++ } ++ ++ out: ++ /* smp_mb(); */ ++ AuTraceErr(!valid); ++ return valid; ++} ++ ++int au_vdir_fill_de(struct file *file, void *dirent, filldir_t filldir) ++{ ++ int err; ++ unsigned int l, deblk_sz; ++ union au_vdir_deblk_p deblk_end; ++ struct au_vdir *vdir_cache; ++ struct au_vdir_de *de; ++ ++ vdir_cache = au_fvdir_cache(file); ++ if (!seek_vdir(file)) ++ return 0; ++ ++ deblk_sz = vdir_cache->vd_deblk_sz; ++ while (1) { ++ deblk_end.deblk = vdir_cache->vd_deblk[vdir_cache->vd_last.ul]; ++ deblk_end.deblk += deblk_sz; ++ while (!is_deblk_end(&vdir_cache->vd_last.p, &deblk_end)) { ++ de = vdir_cache->vd_last.p.de; ++ AuDbg("%.*s, off%lld, i%lu, dt%d\n", ++ de->de_str.len, de->de_str.name, file->f_pos, ++ (unsigned long)de->de_ino, de->de_type); ++ err = filldir(dirent, de->de_str.name, de->de_str.len, ++ file->f_pos, de->de_ino, de->de_type); ++ if (unlikely(err)) { ++ AuTraceErr(err); ++ /* todo: ignore the error caused by udba? */ ++ /* return err; */ ++ return 0; ++ } ++ ++ l = calc_size(de->de_str.len); ++ vdir_cache->vd_last.p.deblk += l; ++ file->f_pos += l; ++ } ++ if (vdir_cache->vd_last.ul < vdir_cache->vd_nblk - 1) { ++ vdir_cache->vd_last.ul++; ++ vdir_cache->vd_last.p.deblk ++ = vdir_cache->vd_deblk[vdir_cache->vd_last.ul]; ++ file->f_pos = deblk_sz * vdir_cache->vd_last.ul; ++ continue; ++ } ++ break; ++ } ++ ++ /* smp_mb(); */ ++ return 0; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/vfsub.c linux-2.6.31/fs/aufs/vfsub.c +--- linux-2.6.31-vanilla/fs/aufs/vfsub.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/vfsub.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,755 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * sub-routines for VFS ++ */ ++ ++#include ++#include ++#include ++#include ++#include ++#include "aufs.h" ++ ++int vfsub_update_h_iattr(struct path *h_path, int *did) ++{ ++ int err; ++ struct kstat st; ++ struct super_block *h_sb; ++ ++ /* for remote fs, leave work for its getattr or d_revalidate */ ++ /* for bad i_attr fs, handle them in aufs_getattr() */ ++ /* still some fs may acquire i_mutex. we need to skip them */ ++ err = 0; ++ if (!did) ++ did = &err; ++ h_sb = h_path->dentry->d_sb; ++ *did = (!au_test_fs_remote(h_sb) && au_test_fs_refresh_iattr(h_sb)); ++ if (*did) ++ err = vfs_getattr(h_path->mnt, h_path->dentry, &st); ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++#ifdef CONFIG_IMA ++#error IMA is not supported since it does not work well. Wait for their fixing. ++#endif ++ ++struct file *vfsub_dentry_open(struct path *path, int flags, ++ const struct cred *cred) ++{ ++ struct file *file; ++ ++ file = dentry_open(path->dentry, path->mnt, flags, cred); ++ if (IS_ERR(file)) ++ return file; ++ /* as NFSD does, just call ima_..._get() simply after dentry_open */ ++ ima_counts_get(file); ++ return file; ++} ++ ++struct file *vfsub_filp_open(const char *path, int oflags, int mode) ++{ ++ struct file *file; ++ ++ lockdep_off(); ++ file = filp_open(path, oflags, mode); ++ lockdep_on(); ++ if (IS_ERR(file)) ++ goto out; ++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/ ++ ++ out: ++ return file; ++} ++ ++int vfsub_kern_path(const char *name, unsigned int flags, struct path *path) ++{ ++ int err; ++ ++ /* lockdep_off(); */ ++ err = kern_path(name, flags, path); ++ /* lockdep_on(); */ ++ if (!err && path->dentry->d_inode) ++ vfsub_update_h_iattr(path, /*did*/NULL); /*ignore*/ ++ return err; ++} ++ ++struct dentry *vfsub_lookup_one_len(const char *name, struct dentry *parent, ++ int len) ++{ ++ struct path path = { ++ .mnt = NULL ++ }; ++ ++ /* VFS checks it too, but by WARN_ON_ONCE() */ ++ IMustLock(parent->d_inode); ++ ++ path.dentry = lookup_one_len(name, parent, len); ++ if (IS_ERR(path.dentry)) ++ goto out; ++ if (path.dentry->d_inode) ++ vfsub_update_h_iattr(&path, /*did*/NULL); /*ignore*/ ++ ++ out: ++ return path.dentry; ++} ++ ++struct dentry *vfsub_lookup_hash(struct nameidata *nd) ++{ ++ struct path path = { ++ .mnt = nd->path.mnt ++ }; ++ ++ IMustLock(nd->path.dentry->d_inode); ++ ++ path.dentry = lookup_hash(nd); ++ if (!IS_ERR(path.dentry) && path.dentry->d_inode) ++ vfsub_update_h_iattr(&path, /*did*/NULL); /*ignore*/ ++ ++ return path.dentry; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct dentry *vfsub_lock_rename(struct dentry *d1, struct au_hinode *hdir1, ++ struct dentry *d2, struct au_hinode *hdir2) ++{ ++ struct dentry *d; ++ ++ lockdep_off(); ++ d = lock_rename(d1, d2); ++ lockdep_on(); ++ au_hin_suspend(hdir1); ++ if (hdir1 != hdir2) ++ au_hin_suspend(hdir2); ++ ++ return d; ++} ++ ++void vfsub_unlock_rename(struct dentry *d1, struct au_hinode *hdir1, ++ struct dentry *d2, struct au_hinode *hdir2) ++{ ++ au_hin_resume(hdir1); ++ if (hdir1 != hdir2) ++ au_hin_resume(hdir2); ++ lockdep_off(); ++ unlock_rename(d1, d2); ++ lockdep_on(); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int vfsub_create(struct inode *dir, struct path *path, int mode) ++{ ++ int err; ++ struct dentry *d; ++ ++ IMustLock(dir); ++ ++ d = path->dentry; ++ path->dentry = d->d_parent; ++ err = security_path_mknod(path, path->dentry, mode, 0); ++ path->dentry = d; ++ if (unlikely(err)) ++ goto out; ++ ++ if (au_test_fs_null_nd(dir->i_sb)) ++ err = vfs_create(dir, path->dentry, mode, NULL); ++ else { ++ struct nameidata h_nd; ++ ++ memset(&h_nd, 0, sizeof(h_nd)); ++ h_nd.flags = LOOKUP_CREATE; ++ h_nd.intent.open.flags = O_CREAT ++ | vfsub_fmode_to_uint(FMODE_READ); ++ h_nd.intent.open.create_mode = mode; ++ h_nd.path.dentry = path->dentry->d_parent; ++ h_nd.path.mnt = path->mnt; ++ path_get(&h_nd.path); ++ err = vfs_create(dir, path->dentry, mode, &h_nd); ++ path_put(&h_nd.path); ++ } ++ ++ if (!err) { ++ struct path tmp = *path; ++ int did; ++ ++ vfsub_update_h_iattr(&tmp, &did); ++ if (did) { ++ tmp.dentry = path->dentry->d_parent; ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); ++ } ++ /*ignore*/ ++ } ++ ++ out: ++ return err; ++} ++ ++int vfsub_symlink(struct inode *dir, struct path *path, const char *symname) ++{ ++ int err; ++ struct dentry *d; ++ ++ IMustLock(dir); ++ ++ d = path->dentry; ++ path->dentry = d->d_parent; ++ err = security_path_symlink(path, path->dentry, symname); ++ path->dentry = d; ++ if (unlikely(err)) ++ goto out; ++ ++ err = vfs_symlink(dir, path->dentry, symname); ++ if (!err) { ++ struct path tmp = *path; ++ int did; ++ ++ vfsub_update_h_iattr(&tmp, &did); ++ if (did) { ++ tmp.dentry = path->dentry->d_parent; ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); ++ } ++ /*ignore*/ ++ } ++ ++ out: ++ return err; ++} ++ ++int vfsub_mknod(struct inode *dir, struct path *path, int mode, dev_t dev) ++{ ++ int err; ++ struct dentry *d; ++ ++ IMustLock(dir); ++ ++ d = path->dentry; ++ path->dentry = d->d_parent; ++ err = security_path_mknod(path, path->dentry, mode, dev); ++ path->dentry = d; ++ if (unlikely(err)) ++ goto out; ++ ++ err = vfs_mknod(dir, path->dentry, mode, dev); ++ if (!err) { ++ struct path tmp = *path; ++ int did; ++ ++ vfsub_update_h_iattr(&tmp, &did); ++ if (did) { ++ tmp.dentry = path->dentry->d_parent; ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); ++ } ++ /*ignore*/ ++ } ++ ++ out: ++ return err; ++} ++ ++static int au_test_nlink(struct inode *inode) ++{ ++ const unsigned int link_max = UINT_MAX >> 1; /* rough margin */ ++ ++ if (!au_test_fs_no_limit_nlink(inode->i_sb) ++ || inode->i_nlink < link_max) ++ return 0; ++ return -EMLINK; ++} ++ ++int vfsub_link(struct dentry *src_dentry, struct inode *dir, struct path *path) ++{ ++ int err; ++ struct dentry *d; ++ ++ IMustLock(dir); ++ ++ err = au_test_nlink(src_dentry->d_inode); ++ if (unlikely(err)) ++ return err; ++ ++ d = path->dentry; ++ path->dentry = d->d_parent; ++ err = security_path_link(src_dentry, path, path->dentry); ++ path->dentry = d; ++ if (unlikely(err)) ++ goto out; ++ ++ lockdep_off(); ++ err = vfs_link(src_dentry, dir, path->dentry); ++ lockdep_on(); ++ if (!err) { ++ struct path tmp = *path; ++ int did; ++ ++ /* fuse has different memory inode for the same inumber */ ++ vfsub_update_h_iattr(&tmp, &did); ++ if (did) { ++ tmp.dentry = path->dentry->d_parent; ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); ++ tmp.dentry = src_dentry; ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); ++ } ++ /*ignore*/ ++ } ++ ++ out: ++ return err; ++} ++ ++int vfsub_rename(struct inode *src_dir, struct dentry *src_dentry, ++ struct inode *dir, struct path *path) ++{ ++ int err; ++ struct path tmp = { ++ .mnt = path->mnt ++ }; ++ struct dentry *d; ++ ++ IMustLock(dir); ++ IMustLock(src_dir); ++ ++ d = path->dentry; ++ path->dentry = d->d_parent; ++ tmp.dentry = src_dentry->d_parent; ++ err = security_path_rename(&tmp, src_dentry, path, path->dentry); ++ path->dentry = d; ++ if (unlikely(err)) ++ goto out; ++ ++ lockdep_off(); ++ err = vfs_rename(src_dir, src_dentry, dir, path->dentry); ++ lockdep_on(); ++ if (!err) { ++ int did; ++ ++ tmp.dentry = d->d_parent; ++ vfsub_update_h_iattr(&tmp, &did); ++ if (did) { ++ tmp.dentry = src_dentry; ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); ++ tmp.dentry = src_dentry->d_parent; ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); ++ } ++ /*ignore*/ ++ } ++ ++ out: ++ return err; ++} ++ ++int vfsub_mkdir(struct inode *dir, struct path *path, int mode) ++{ ++ int err; ++ struct dentry *d; ++ ++ IMustLock(dir); ++ ++ d = path->dentry; ++ path->dentry = d->d_parent; ++ err = security_path_mkdir(path, path->dentry, mode); ++ path->dentry = d; ++ if (unlikely(err)) ++ goto out; ++ ++ err = vfs_mkdir(dir, path->dentry, mode); ++ if (!err) { ++ struct path tmp = *path; ++ int did; ++ ++ vfsub_update_h_iattr(&tmp, &did); ++ if (did) { ++ tmp.dentry = path->dentry->d_parent; ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); ++ } ++ /*ignore*/ ++ } ++ ++ out: ++ return err; ++} ++ ++int vfsub_rmdir(struct inode *dir, struct path *path) ++{ ++ int err; ++ struct dentry *d; ++ ++ IMustLock(dir); ++ ++ d = path->dentry; ++ path->dentry = d->d_parent; ++ err = security_path_rmdir(path, path->dentry); ++ path->dentry = d; ++ if (unlikely(err)) ++ goto out; ++ ++ lockdep_off(); ++ err = vfs_rmdir(dir, path->dentry); ++ lockdep_on(); ++ if (!err) { ++ struct path tmp = { ++ .dentry = path->dentry->d_parent, ++ .mnt = path->mnt ++ }; ++ ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); /*ignore*/ ++ } ++ ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++ssize_t vfsub_read_u(struct file *file, char __user *ubuf, size_t count, ++ loff_t *ppos) ++{ ++ ssize_t err; ++ ++ err = vfs_read(file, ubuf, count, ppos); ++ if (err >= 0) ++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/ ++ return err; ++} ++ ++/* todo: kernel_read()? */ ++ssize_t vfsub_read_k(struct file *file, void *kbuf, size_t count, ++ loff_t *ppos) ++{ ++ ssize_t err; ++ mm_segment_t oldfs; ++ ++ oldfs = get_fs(); ++ set_fs(KERNEL_DS); ++ err = vfsub_read_u(file, (char __user *)kbuf, count, ppos); ++ set_fs(oldfs); ++ return err; ++} ++ ++ssize_t vfsub_write_u(struct file *file, const char __user *ubuf, size_t count, ++ loff_t *ppos) ++{ ++ ssize_t err; ++ ++ lockdep_off(); ++ err = vfs_write(file, ubuf, count, ppos); ++ lockdep_on(); ++ if (err >= 0) ++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/ ++ return err; ++} ++ ++ssize_t vfsub_write_k(struct file *file, void *kbuf, size_t count, loff_t *ppos) ++{ ++ ssize_t err; ++ mm_segment_t oldfs; ++ ++ oldfs = get_fs(); ++ set_fs(KERNEL_DS); ++ err = vfsub_write_u(file, (const char __user *)kbuf, count, ppos); ++ set_fs(oldfs); ++ return err; ++} ++ ++int vfsub_readdir(struct file *file, filldir_t filldir, void *arg) ++{ ++ int err; ++ ++ lockdep_off(); ++ err = vfs_readdir(file, filldir, arg); ++ lockdep_on(); ++ if (err >= 0) ++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/ ++ return err; ++} ++ ++long vfsub_splice_to(struct file *in, loff_t *ppos, ++ struct pipe_inode_info *pipe, size_t len, ++ unsigned int flags) ++{ ++ long err; ++ ++ lockdep_off(); ++ err = do_splice_to(in, ppos, pipe, len, flags); ++ lockdep_on(); ++ if (err >= 0) ++ vfsub_update_h_iattr(&in->f_path, /*did*/NULL); /*ignore*/ ++ return err; ++} ++ ++long vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out, ++ loff_t *ppos, size_t len, unsigned int flags) ++{ ++ long err; ++ ++ lockdep_off(); ++ err = do_splice_from(pipe, out, ppos, len, flags); ++ lockdep_on(); ++ if (err >= 0) ++ vfsub_update_h_iattr(&out->f_path, /*did*/NULL); /*ignore*/ ++ return err; ++} ++ ++/* cf. open.c:do_sys_truncate() and do_sys_ftruncate() */ ++int vfsub_trunc(struct path *h_path, loff_t length, unsigned int attr, ++ struct file *h_file) ++{ ++ int err; ++ struct inode *h_inode; ++ ++ h_inode = h_path->dentry->d_inode; ++ if (!h_file) { ++ err = mnt_want_write(h_path->mnt); ++ if (err) ++ goto out; ++ err = inode_permission(h_inode, MAY_WRITE); ++ if (err) ++ goto out_mnt; ++ err = get_write_access(h_inode); ++ if (err) ++ goto out_mnt; ++ err = break_lease(h_inode, vfsub_fmode_to_uint(FMODE_WRITE)); ++ if (err) ++ goto out_inode; ++ } ++ ++ err = locks_verify_truncate(h_inode, h_file, length); ++ if (!err) ++ err = security_path_truncate(h_path, length, attr); ++ if (!err) { ++ lockdep_off(); ++ err = do_truncate(h_path->dentry, length, attr, h_file); ++ lockdep_on(); ++ } ++ ++ out_inode: ++ if (!h_file) ++ put_write_access(h_inode); ++ out_mnt: ++ if (!h_file) ++ mnt_drop_write(h_path->mnt); ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct au_vfsub_mkdir_args { ++ int *errp; ++ struct inode *dir; ++ struct path *path; ++ int mode; ++}; ++ ++static void au_call_vfsub_mkdir(void *args) ++{ ++ struct au_vfsub_mkdir_args *a = args; ++ *a->errp = vfsub_mkdir(a->dir, a->path, a->mode); ++} ++ ++int vfsub_sio_mkdir(struct inode *dir, struct path *path, int mode) ++{ ++ int err, do_sio, wkq_err; ++ ++ do_sio = au_test_h_perm_sio(dir, MAY_EXEC | MAY_WRITE); ++ if (!do_sio) ++ err = vfsub_mkdir(dir, path, mode); ++ else { ++ struct au_vfsub_mkdir_args args = { ++ .errp = &err, ++ .dir = dir, ++ .path = path, ++ .mode = mode ++ }; ++ wkq_err = au_wkq_wait(au_call_vfsub_mkdir, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ } ++ ++ return err; ++} ++ ++struct au_vfsub_rmdir_args { ++ int *errp; ++ struct inode *dir; ++ struct path *path; ++}; ++ ++static void au_call_vfsub_rmdir(void *args) ++{ ++ struct au_vfsub_rmdir_args *a = args; ++ *a->errp = vfsub_rmdir(a->dir, a->path); ++} ++ ++int vfsub_sio_rmdir(struct inode *dir, struct path *path) ++{ ++ int err, do_sio, wkq_err; ++ ++ do_sio = au_test_h_perm_sio(dir, MAY_EXEC | MAY_WRITE); ++ if (!do_sio) ++ err = vfsub_rmdir(dir, path); ++ else { ++ struct au_vfsub_rmdir_args args = { ++ .errp = &err, ++ .dir = dir, ++ .path = path ++ }; ++ wkq_err = au_wkq_wait(au_call_vfsub_rmdir, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ } ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct notify_change_args { ++ int *errp; ++ struct path *path; ++ struct iattr *ia; ++}; ++ ++static void call_notify_change(void *args) ++{ ++ struct notify_change_args *a = args; ++ struct inode *h_inode; ++ ++ h_inode = a->path->dentry->d_inode; ++ IMustLock(h_inode); ++ ++ *a->errp = -EPERM; ++ if (!IS_IMMUTABLE(h_inode) && !IS_APPEND(h_inode)) { ++ lockdep_off(); ++ *a->errp = notify_change(a->path->dentry, a->ia); ++ lockdep_on(); ++ if (!*a->errp) ++ vfsub_update_h_iattr(a->path, /*did*/NULL); /*ignore*/ ++ } ++ AuTraceErr(*a->errp); ++} ++ ++int vfsub_notify_change(struct path *path, struct iattr *ia) ++{ ++ int err; ++ struct notify_change_args args = { ++ .errp = &err, ++ .path = path, ++ .ia = ia ++ }; ++ ++ call_notify_change(&args); ++ ++ return err; ++} ++ ++int vfsub_sio_notify_change(struct path *path, struct iattr *ia) ++{ ++ int err, wkq_err; ++ struct notify_change_args args = { ++ .errp = &err, ++ .path = path, ++ .ia = ia ++ }; ++ ++ wkq_err = au_wkq_wait(call_notify_change, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct unlink_args { ++ int *errp; ++ struct inode *dir; ++ struct path *path; ++}; ++ ++static void call_unlink(void *args) ++{ ++ struct unlink_args *a = args; ++ struct dentry *d = a->path->dentry; ++ struct inode *h_inode; ++ const int stop_sillyrename = (au_test_nfs(d->d_sb) ++ && atomic_read(&d->d_count) == 1); ++ ++ IMustLock(a->dir); ++ ++ a->path->dentry = d->d_parent; ++ *a->errp = security_path_unlink(a->path, d); ++ a->path->dentry = d; ++ if (unlikely(*a->errp)) ++ return; ++ ++ if (!stop_sillyrename) ++ dget(d); ++ h_inode = d->d_inode; ++ if (h_inode) ++ atomic_inc(&h_inode->i_count); ++ ++ lockdep_off(); ++ *a->errp = vfs_unlink(a->dir, d); ++ lockdep_on(); ++ if (!*a->errp) { ++ struct path tmp = { ++ .dentry = d->d_parent, ++ .mnt = a->path->mnt ++ }; ++ vfsub_update_h_iattr(&tmp, /*did*/NULL); /*ignore*/ ++ } ++ ++ if (!stop_sillyrename) ++ dput(d); ++ if (h_inode) ++ iput(h_inode); ++ ++ AuTraceErr(*a->errp); ++} ++ ++/* ++ * @dir: must be locked. ++ * @dentry: target dentry. ++ */ ++int vfsub_unlink(struct inode *dir, struct path *path, int force) ++{ ++ int err; ++ struct unlink_args args = { ++ .errp = &err, ++ .dir = dir, ++ .path = path ++ }; ++ ++ if (!force) ++ call_unlink(&args); ++ else { ++ int wkq_err; ++ ++ wkq_err = au_wkq_wait(call_unlink, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ } ++ ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/vfsub.h linux-2.6.31/fs/aufs/vfsub.h +--- linux-2.6.31-vanilla/fs/aufs/vfsub.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/vfsub.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,172 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * sub-routines for VFS ++ */ ++ ++#ifndef __AUFS_VFSUB_H__ ++#define __AUFS_VFSUB_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* lock subclass for lower inode */ ++/* default MAX_LOCKDEP_SUBCLASSES(8) is not enough */ ++/* reduce? gave up. */ ++enum { ++ AuLsc_I_Begin = I_MUTEX_QUOTA, /* 4 */ ++ AuLsc_I_PARENT, /* lower inode, parent first */ ++ AuLsc_I_PARENT2, /* copyup dirs */ ++ AuLsc_I_PARENT3, /* copyup wh */ ++ AuLsc_I_CHILD, ++ AuLsc_I_CHILD2, ++ AuLsc_I_End ++}; ++ ++/* to debug easier, do not make them inlined functions */ ++#define MtxMustLock(mtx) AuDebugOn(!mutex_is_locked(mtx)) ++#define IMustLock(i) MtxMustLock(&(i)->i_mutex) ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline void vfsub_copy_inode_size(struct inode *inode, ++ struct inode *h_inode) ++{ ++ spin_lock(&inode->i_lock); ++ fsstack_copy_inode_size(inode, h_inode); ++ spin_unlock(&inode->i_lock); ++} ++ ++int vfsub_update_h_iattr(struct path *h_path, int *did); ++struct file *vfsub_filp_open(const char *path, int oflags, int mode); ++struct file *vfsub_dentry_open(struct path *path, int flags, ++ const struct cred *cred); ++int vfsub_kern_path(const char *name, unsigned int flags, struct path *path); ++struct dentry *vfsub_lookup_one_len(const char *name, struct dentry *parent, ++ int len); ++struct dentry *vfsub_lookup_hash(struct nameidata *nd); ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct au_hinode; ++struct dentry *vfsub_lock_rename(struct dentry *d1, struct au_hinode *hdir1, ++ struct dentry *d2, struct au_hinode *hdir2); ++void vfsub_unlock_rename(struct dentry *d1, struct au_hinode *hdir1, ++ struct dentry *d2, struct au_hinode *hdir2); ++ ++int vfsub_create(struct inode *dir, struct path *path, int mode); ++int vfsub_symlink(struct inode *dir, struct path *path, ++ const char *symname); ++int vfsub_mknod(struct inode *dir, struct path *path, int mode, dev_t dev); ++int vfsub_link(struct dentry *src_dentry, struct inode *dir, ++ struct path *path); ++int vfsub_rename(struct inode *src_hdir, struct dentry *src_dentry, ++ struct inode *hdir, struct path *path); ++int vfsub_mkdir(struct inode *dir, struct path *path, int mode); ++int vfsub_rmdir(struct inode *dir, struct path *path); ++ ++int vfsub_sio_mkdir(struct inode *dir, struct path *path, int mode); ++int vfsub_sio_rmdir(struct inode *dir, struct path *path); ++int vfsub_sio_notify_change(struct path *path, struct iattr *ia); ++int vfsub_notify_change(struct path *path, struct iattr *ia); ++int vfsub_unlink(struct inode *dir, struct path *path, int force); ++ ++/* ---------------------------------------------------------------------- */ ++ ++ssize_t vfsub_read_u(struct file *file, char __user *ubuf, size_t count, ++ loff_t *ppos); ++ssize_t vfsub_read_k(struct file *file, void *kbuf, size_t count, ++ loff_t *ppos); ++ssize_t vfsub_write_u(struct file *file, const char __user *ubuf, size_t count, ++ loff_t *ppos); ++ssize_t vfsub_write_k(struct file *file, void *kbuf, size_t count, ++ loff_t *ppos); ++int vfsub_readdir(struct file *file, filldir_t filldir, void *arg); ++ ++long vfsub_splice_to(struct file *in, loff_t *ppos, ++ struct pipe_inode_info *pipe, size_t len, ++ unsigned int flags); ++long vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out, ++ loff_t *ppos, size_t len, unsigned int flags); ++int vfsub_trunc(struct path *h_path, loff_t length, unsigned int attr, ++ struct file *h_file); ++ ++static inline void vfsub_file_accessed(struct file *h_file) ++{ ++ file_accessed(h_file); ++ vfsub_update_h_iattr(&h_file->f_path, /*did*/NULL); /*ignore*/ ++} ++ ++static inline void vfsub_touch_atime(struct vfsmount *h_mnt, ++ struct dentry *h_dentry) ++{ ++ struct path h_path = { ++ .dentry = h_dentry, ++ .mnt = h_mnt ++ }; ++ touch_atime(h_mnt, h_dentry); ++ vfsub_update_h_iattr(&h_path, /*did*/NULL); /*ignore*/ ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline loff_t vfsub_llseek(struct file *file, loff_t offset, int origin) ++{ ++ loff_t err; ++ ++ lockdep_off(); ++ err = vfs_llseek(file, offset, origin); ++ lockdep_on(); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* dirty workaround for strict type of fmode_t */ ++union vfsub_fmu { ++ fmode_t fm; ++ unsigned int ui; ++}; ++ ++static inline unsigned int vfsub_fmode_to_uint(fmode_t fm) ++{ ++ union vfsub_fmu u = { ++ .fm = fm ++ }; ++ ++ BUILD_BUG_ON(sizeof(u.fm) != sizeof(u.ui)); ++ ++ return u.ui; ++} ++ ++static inline fmode_t vfsub_uint_to_fmode(unsigned int ui) ++{ ++ union vfsub_fmu u = { ++ .ui = ui ++ }; ++ ++ return u.fm; ++} ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_VFSUB_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/wbr_policy.c linux-2.6.31/fs/aufs/wbr_policy.c +--- linux-2.6.31-vanilla/fs/aufs/wbr_policy.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/wbr_policy.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,641 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * policies for selecting one among multiple writable branches ++ */ ++ ++#include ++#include "aufs.h" ++ ++/* subset of cpup_attr() */ ++static noinline_for_stack ++int au_cpdown_attr(struct path *h_path, struct dentry *h_src) ++{ ++ int err, sbits; ++ struct iattr ia; ++ struct inode *h_isrc; ++ ++ h_isrc = h_src->d_inode; ++ ia.ia_valid = ATTR_FORCE | ATTR_MODE | ATTR_UID | ATTR_GID; ++ ia.ia_mode = h_isrc->i_mode; ++ ia.ia_uid = h_isrc->i_uid; ++ ia.ia_gid = h_isrc->i_gid; ++ sbits = !!(ia.ia_mode & (S_ISUID | S_ISGID)); ++ au_cpup_attr_flags(h_path->dentry->d_inode, h_isrc); ++ err = vfsub_sio_notify_change(h_path, &ia); ++ ++ /* is this nfs only? */ ++ if (!err && sbits && au_test_nfs(h_path->dentry->d_sb)) { ++ ia.ia_valid = ATTR_FORCE | ATTR_MODE; ++ ia.ia_mode = h_isrc->i_mode; ++ err = vfsub_sio_notify_change(h_path, &ia); ++ } ++ ++ return err; ++} ++ ++#define AuCpdown_PARENT_OPQ 1 ++#define AuCpdown_WHED (1 << 1) ++#define AuCpdown_MADE_DIR (1 << 2) ++#define AuCpdown_DIROPQ (1 << 3) ++#define au_ftest_cpdown(flags, name) ((flags) & AuCpdown_##name) ++#define au_fset_cpdown(flags, name) { (flags) |= AuCpdown_##name; } ++#define au_fclr_cpdown(flags, name) { (flags) &= ~AuCpdown_##name; } ++ ++struct au_cpdown_dir_args { ++ struct dentry *parent; ++ unsigned int flags; ++}; ++ ++static int au_cpdown_dir_opq(struct dentry *dentry, aufs_bindex_t bdst, ++ struct au_cpdown_dir_args *a) ++{ ++ int err; ++ struct dentry *opq_dentry; ++ ++ opq_dentry = au_diropq_create(dentry, bdst); ++ err = PTR_ERR(opq_dentry); ++ if (IS_ERR(opq_dentry)) ++ goto out; ++ dput(opq_dentry); ++ au_fset_cpdown(a->flags, DIROPQ); ++ ++ out: ++ return err; ++} ++ ++static int au_cpdown_dir_wh(struct dentry *dentry, struct dentry *h_parent, ++ struct inode *dir, aufs_bindex_t bdst) ++{ ++ int err; ++ struct path h_path; ++ struct au_branch *br; ++ ++ br = au_sbr(dentry->d_sb, bdst); ++ h_path.dentry = au_wh_lkup(h_parent, &dentry->d_name, br); ++ err = PTR_ERR(h_path.dentry); ++ if (IS_ERR(h_path.dentry)) ++ goto out; ++ ++ err = 0; ++ if (h_path.dentry->d_inode) { ++ h_path.mnt = br->br_mnt; ++ err = au_wh_unlink_dentry(au_h_iptr(dir, bdst), &h_path, ++ dentry); ++ } ++ dput(h_path.dentry); ++ ++ out: ++ return err; ++} ++ ++static int au_cpdown_dir(struct dentry *dentry, aufs_bindex_t bdst, ++ struct dentry *h_parent, void *arg) ++{ ++ int err, rerr; ++ aufs_bindex_t bend, bopq, bstart; ++ unsigned char parent_opq; ++ struct path h_path; ++ struct dentry *parent; ++ struct inode *h_dir, *h_inode, *inode, *dir; ++ struct au_cpdown_dir_args *args = arg; ++ ++ bstart = au_dbstart(dentry); ++ /* dentry is di-locked */ ++ parent = dget_parent(dentry); ++ dir = parent->d_inode; ++ h_dir = h_parent->d_inode; ++ AuDebugOn(h_dir != au_h_iptr(dir, bdst)); ++ IMustLock(h_dir); ++ ++ err = au_lkup_neg(dentry, bdst); ++ if (unlikely(err < 0)) ++ goto out; ++ h_path.dentry = au_h_dptr(dentry, bdst); ++ h_path.mnt = au_sbr_mnt(dentry->d_sb, bdst); ++ err = vfsub_sio_mkdir(au_h_iptr(dir, bdst), &h_path, ++ S_IRWXU | S_IRUGO | S_IXUGO); ++ if (unlikely(err)) ++ goto out_put; ++ au_fset_cpdown(args->flags, MADE_DIR); ++ ++ bend = au_dbend(dentry); ++ bopq = au_dbdiropq(dentry); ++ au_fclr_cpdown(args->flags, WHED); ++ au_fclr_cpdown(args->flags, DIROPQ); ++ if (au_dbwh(dentry) == bdst) ++ au_fset_cpdown(args->flags, WHED); ++ if (!au_ftest_cpdown(args->flags, PARENT_OPQ) && bopq <= bdst) ++ au_fset_cpdown(args->flags, PARENT_OPQ); ++ parent_opq = (au_ftest_cpdown(args->flags, PARENT_OPQ) ++ && args->parent == dentry); ++ h_inode = h_path.dentry->d_inode; ++ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); ++ if (au_ftest_cpdown(args->flags, WHED)) { ++ err = au_cpdown_dir_opq(dentry, bdst, args); ++ if (unlikely(err)) { ++ mutex_unlock(&h_inode->i_mutex); ++ goto out_dir; ++ } ++ } ++ ++ err = au_cpdown_attr(&h_path, au_h_dptr(dentry, bstart)); ++ mutex_unlock(&h_inode->i_mutex); ++ if (unlikely(err)) ++ goto out_opq; ++ ++ if (au_ftest_cpdown(args->flags, WHED)) { ++ err = au_cpdown_dir_wh(dentry, h_parent, dir, bdst); ++ if (unlikely(err)) ++ goto out_opq; ++ } ++ ++ inode = dentry->d_inode; ++ if (au_ibend(inode) < bdst) ++ au_set_ibend(inode, bdst); ++ au_set_h_iptr(inode, bdst, au_igrab(h_inode), ++ au_hi_flags(inode, /*isdir*/1)); ++ goto out; /* success */ ++ ++ /* revert */ ++ out_opq: ++ if (au_ftest_cpdown(args->flags, DIROPQ)) { ++ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD); ++ rerr = au_diropq_remove(dentry, bdst); ++ mutex_unlock(&h_inode->i_mutex); ++ if (unlikely(rerr)) { ++ AuIOErr("failed removing diropq for %.*s b%d (%d)\n", ++ AuDLNPair(dentry), bdst, rerr); ++ err = -EIO; ++ goto out; ++ } ++ } ++ out_dir: ++ if (au_ftest_cpdown(args->flags, MADE_DIR)) { ++ rerr = vfsub_sio_rmdir(au_h_iptr(dir, bdst), &h_path); ++ if (unlikely(rerr)) { ++ AuIOErr("failed removing %.*s b%d (%d)\n", ++ AuDLNPair(dentry), bdst, rerr); ++ err = -EIO; ++ } ++ } ++ out_put: ++ au_set_h_dptr(dentry, bdst, NULL); ++ if (au_dbend(dentry) == bdst) ++ au_update_dbend(dentry); ++ out: ++ dput(parent); ++ return err; ++} ++ ++int au_cpdown_dirs(struct dentry *dentry, aufs_bindex_t bdst) ++{ ++ int err; ++ struct au_cpdown_dir_args args = { ++ .parent = dget_parent(dentry), ++ .flags = 0 ++ }; ++ ++ err = au_cp_dirs(dentry, bdst, au_cpdown_dir, &args); ++ dput(args.parent); ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* policies for create */ ++ ++static int au_wbr_bu(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ for (; bindex >= 0; bindex--) ++ if (!au_br_rdonly(au_sbr(sb, bindex))) ++ return bindex; ++ return -EROFS; ++} ++ ++/* top down parent */ ++static int au_wbr_create_tdp(struct dentry *dentry, int isdir __maybe_unused) ++{ ++ int err; ++ aufs_bindex_t bstart, bindex; ++ struct super_block *sb; ++ struct dentry *parent, *h_parent; ++ ++ sb = dentry->d_sb; ++ bstart = au_dbstart(dentry); ++ err = bstart; ++ if (!au_br_rdonly(au_sbr(sb, bstart))) ++ goto out; ++ ++ err = -EROFS; ++ parent = dget_parent(dentry); ++ for (bindex = au_dbstart(parent); bindex < bstart; bindex++) { ++ h_parent = au_h_dptr(parent, bindex); ++ if (!h_parent || !h_parent->d_inode) ++ continue; ++ ++ if (!au_br_rdonly(au_sbr(sb, bindex))) { ++ err = bindex; ++ break; ++ } ++ } ++ dput(parent); ++ ++ /* bottom up here */ ++ if (unlikely(err < 0)) ++ err = au_wbr_bu(sb, bstart - 1); ++ ++ out: ++ AuDbg("b%d\n", err); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* an exception for the policy other than tdp */ ++static int au_wbr_create_exp(struct dentry *dentry) ++{ ++ int err; ++ aufs_bindex_t bwh, bdiropq; ++ struct dentry *parent; ++ ++ err = -1; ++ bwh = au_dbwh(dentry); ++ parent = dget_parent(dentry); ++ bdiropq = au_dbdiropq(parent); ++ if (bwh >= 0) { ++ if (bdiropq >= 0) ++ err = min(bdiropq, bwh); ++ else ++ err = bwh; ++ AuDbg("%d\n", err); ++ } else if (bdiropq >= 0) { ++ err = bdiropq; ++ AuDbg("%d\n", err); ++ } ++ dput(parent); ++ ++ if (err >= 0 && au_br_rdonly(au_sbr(dentry->d_sb, err))) ++ err = -1; ++ ++ AuDbg("%d\n", err); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* round robin */ ++static int au_wbr_create_init_rr(struct super_block *sb) ++{ ++ int err; ++ ++ err = au_wbr_bu(sb, au_sbend(sb)); ++ atomic_set(&au_sbi(sb)->si_wbr_rr_next, -err); /* less important */ ++ /* smp_mb(); */ ++ ++ AuDbg("b%d\n", err); ++ return err; ++} ++ ++static int au_wbr_create_rr(struct dentry *dentry, int isdir) ++{ ++ int err, nbr; ++ unsigned int u; ++ aufs_bindex_t bindex, bend; ++ struct super_block *sb; ++ atomic_t *next; ++ ++ err = au_wbr_create_exp(dentry); ++ if (err >= 0) ++ goto out; ++ ++ sb = dentry->d_sb; ++ next = &au_sbi(sb)->si_wbr_rr_next; ++ bend = au_sbend(sb); ++ nbr = bend + 1; ++ for (bindex = 0; bindex <= bend; bindex++) { ++ if (!isdir) { ++ err = atomic_dec_return(next) + 1; ++ /* modulo for 0 is meaningless */ ++ if (unlikely(!err)) ++ err = atomic_dec_return(next) + 1; ++ } else ++ err = atomic_read(next); ++ AuDbg("%d\n", err); ++ u = err; ++ err = u % nbr; ++ AuDbg("%d\n", err); ++ if (!au_br_rdonly(au_sbr(sb, err))) ++ break; ++ err = -EROFS; ++ } ++ ++ out: ++ AuDbg("%d\n", err); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* most free space */ ++static void au_mfs(struct dentry *dentry) ++{ ++ struct super_block *sb; ++ struct au_branch *br; ++ struct au_wbr_mfs *mfs; ++ aufs_bindex_t bindex, bend; ++ int err; ++ unsigned long long b, bavail; ++ /* reduce the stack usage */ ++ struct kstatfs *st; ++ ++ st = kmalloc(sizeof(*st), GFP_NOFS); ++ if (unlikely(!st)) { ++ AuWarn1("failed updating mfs(%d), ignored\n", -ENOMEM); ++ return; ++ } ++ ++ bavail = 0; ++ sb = dentry->d_sb; ++ mfs = &au_sbi(sb)->si_wbr_mfs; ++ MtxMustLock(&mfs->mfs_lock); ++ mfs->mfs_bindex = -EROFS; ++ mfs->mfsrr_bytes = 0; ++ bend = au_sbend(sb); ++ for (bindex = 0; bindex <= bend; bindex++) { ++ br = au_sbr(sb, bindex); ++ if (au_br_rdonly(br)) ++ continue; ++ ++ /* sb->s_root for NFS is unreliable */ ++ err = vfs_statfs(br->br_mnt->mnt_root, st); ++ if (unlikely(err)) { ++ AuWarn1("failed statfs, b%d, %d\n", bindex, err); ++ continue; ++ } ++ ++ /* when the available size is equal, select the lower one */ ++ BUILD_BUG_ON(sizeof(b) < sizeof(st->f_bavail) ++ || sizeof(b) < sizeof(st->f_bsize)); ++ b = st->f_bavail * st->f_bsize; ++ br->br_wbr->wbr_bytes = b; ++ if (b >= bavail) { ++ bavail = b; ++ mfs->mfs_bindex = bindex; ++ mfs->mfs_jiffy = jiffies; ++ } ++ } ++ ++ mfs->mfsrr_bytes = bavail; ++ AuDbg("b%d\n", mfs->mfs_bindex); ++ kfree(st); ++} ++ ++static int au_wbr_create_mfs(struct dentry *dentry, int isdir __maybe_unused) ++{ ++ int err; ++ struct super_block *sb; ++ struct au_wbr_mfs *mfs; ++ ++ err = au_wbr_create_exp(dentry); ++ if (err >= 0) ++ goto out; ++ ++ sb = dentry->d_sb; ++ mfs = &au_sbi(sb)->si_wbr_mfs; ++ mutex_lock(&mfs->mfs_lock); ++ if (time_after(jiffies, mfs->mfs_jiffy + mfs->mfs_expire) ++ || mfs->mfs_bindex < 0 ++ || au_br_rdonly(au_sbr(sb, mfs->mfs_bindex))) ++ au_mfs(dentry); ++ mutex_unlock(&mfs->mfs_lock); ++ err = mfs->mfs_bindex; ++ ++ out: ++ AuDbg("b%d\n", err); ++ return err; ++} ++ ++static int au_wbr_create_init_mfs(struct super_block *sb) ++{ ++ struct au_wbr_mfs *mfs; ++ ++ mfs = &au_sbi(sb)->si_wbr_mfs; ++ mutex_init(&mfs->mfs_lock); ++ mfs->mfs_jiffy = 0; ++ mfs->mfs_bindex = -EROFS; ++ ++ return 0; ++} ++ ++static int au_wbr_create_fin_mfs(struct super_block *sb __maybe_unused) ++{ ++ mutex_destroy(&au_sbi(sb)->si_wbr_mfs.mfs_lock); ++ return 0; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* most free space and then round robin */ ++static int au_wbr_create_mfsrr(struct dentry *dentry, int isdir) ++{ ++ int err; ++ struct au_wbr_mfs *mfs; ++ ++ err = au_wbr_create_mfs(dentry, isdir); ++ if (err >= 0) { ++ mfs = &au_sbi(dentry->d_sb)->si_wbr_mfs; ++ mutex_lock(&mfs->mfs_lock); ++ if (mfs->mfsrr_bytes < mfs->mfsrr_watermark) ++ err = au_wbr_create_rr(dentry, isdir); ++ mutex_unlock(&mfs->mfs_lock); ++ } ++ ++ AuDbg("b%d\n", err); ++ return err; ++} ++ ++static int au_wbr_create_init_mfsrr(struct super_block *sb) ++{ ++ int err; ++ ++ au_wbr_create_init_mfs(sb); /* ignore */ ++ err = au_wbr_create_init_rr(sb); ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* top down parent and most free space */ ++static int au_wbr_create_pmfs(struct dentry *dentry, int isdir) ++{ ++ int err, e2; ++ unsigned long long b; ++ aufs_bindex_t bindex, bstart, bend; ++ struct super_block *sb; ++ struct dentry *parent, *h_parent; ++ struct au_branch *br; ++ ++ err = au_wbr_create_tdp(dentry, isdir); ++ if (unlikely(err < 0)) ++ goto out; ++ parent = dget_parent(dentry); ++ bstart = au_dbstart(parent); ++ bend = au_dbtaildir(parent); ++ if (bstart == bend) ++ goto out_parent; /* success */ ++ ++ e2 = au_wbr_create_mfs(dentry, isdir); ++ if (e2 < 0) ++ goto out_parent; /* success */ ++ ++ /* when the available size is equal, select upper one */ ++ sb = dentry->d_sb; ++ br = au_sbr(sb, err); ++ b = br->br_wbr->wbr_bytes; ++ AuDbg("b%d, %llu\n", err, b); ++ ++ for (bindex = bstart; bindex <= bend; bindex++) { ++ h_parent = au_h_dptr(parent, bindex); ++ if (!h_parent || !h_parent->d_inode) ++ continue; ++ ++ br = au_sbr(sb, bindex); ++ if (!au_br_rdonly(br) && br->br_wbr->wbr_bytes > b) { ++ b = br->br_wbr->wbr_bytes; ++ err = bindex; ++ AuDbg("b%d, %llu\n", err, b); ++ } ++ } ++ ++ out_parent: ++ dput(parent); ++ out: ++ AuDbg("b%d\n", err); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* policies for copyup */ ++ ++/* top down parent */ ++static int au_wbr_copyup_tdp(struct dentry *dentry) ++{ ++ return au_wbr_create_tdp(dentry, /*isdir, anything is ok*/0); ++} ++ ++/* bottom up parent */ ++static int au_wbr_copyup_bup(struct dentry *dentry) ++{ ++ int err; ++ aufs_bindex_t bindex, bstart; ++ struct dentry *parent, *h_parent; ++ struct super_block *sb; ++ ++ err = -EROFS; ++ sb = dentry->d_sb; ++ parent = dget_parent(dentry); ++ bstart = au_dbstart(parent); ++ for (bindex = au_dbstart(dentry); bindex >= bstart; bindex--) { ++ h_parent = au_h_dptr(parent, bindex); ++ if (!h_parent || !h_parent->d_inode) ++ continue; ++ ++ if (!au_br_rdonly(au_sbr(sb, bindex))) { ++ err = bindex; ++ break; ++ } ++ } ++ dput(parent); ++ ++ /* bottom up here */ ++ if (unlikely(err < 0)) ++ err = au_wbr_bu(sb, bstart - 1); ++ ++ AuDbg("b%d\n", err); ++ return err; ++} ++ ++/* bottom up */ ++static int au_wbr_copyup_bu(struct dentry *dentry) ++{ ++ int err; ++ ++ err = au_wbr_bu(dentry->d_sb, au_dbstart(dentry)); ++ ++ AuDbg("b%d\n", err); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct au_wbr_copyup_operations au_wbr_copyup_ops[] = { ++ [AuWbrCopyup_TDP] = { ++ .copyup = au_wbr_copyup_tdp ++ }, ++ [AuWbrCopyup_BUP] = { ++ .copyup = au_wbr_copyup_bup ++ }, ++ [AuWbrCopyup_BU] = { ++ .copyup = au_wbr_copyup_bu ++ } ++}; ++ ++struct au_wbr_create_operations au_wbr_create_ops[] = { ++ [AuWbrCreate_TDP] = { ++ .create = au_wbr_create_tdp ++ }, ++ [AuWbrCreate_RR] = { ++ .create = au_wbr_create_rr, ++ .init = au_wbr_create_init_rr ++ }, ++ [AuWbrCreate_MFS] = { ++ .create = au_wbr_create_mfs, ++ .init = au_wbr_create_init_mfs, ++ .fin = au_wbr_create_fin_mfs ++ }, ++ [AuWbrCreate_MFSV] = { ++ .create = au_wbr_create_mfs, ++ .init = au_wbr_create_init_mfs, ++ .fin = au_wbr_create_fin_mfs ++ }, ++ [AuWbrCreate_MFSRR] = { ++ .create = au_wbr_create_mfsrr, ++ .init = au_wbr_create_init_mfsrr, ++ .fin = au_wbr_create_fin_mfs ++ }, ++ [AuWbrCreate_MFSRRV] = { ++ .create = au_wbr_create_mfsrr, ++ .init = au_wbr_create_init_mfsrr, ++ .fin = au_wbr_create_fin_mfs ++ }, ++ [AuWbrCreate_PMFS] = { ++ .create = au_wbr_create_pmfs, ++ .init = au_wbr_create_init_mfs, ++ .fin = au_wbr_create_fin_mfs ++ }, ++ [AuWbrCreate_PMFSV] = { ++ .create = au_wbr_create_pmfs, ++ .init = au_wbr_create_init_mfs, ++ .fin = au_wbr_create_fin_mfs ++ } ++}; +diff -Nur linux-2.6.31-vanilla/fs/aufs/whout.c linux-2.6.31/fs/aufs/whout.c +--- linux-2.6.31-vanilla/fs/aufs/whout.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/whout.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,1052 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * whiteout for logical deletion and opaque directory ++ */ ++ ++#include ++#include "aufs.h" ++ ++#define WH_MASK S_IRUGO ++ ++/* ++ * If a directory contains this file, then it is opaque. We start with the ++ * .wh. flag so that it is blocked by lookup. ++ */ ++static struct qstr diropq_name = { ++ .name = AUFS_WH_DIROPQ, ++ .len = sizeof(AUFS_WH_DIROPQ) - 1 ++}; ++ ++/* ++ * generate whiteout name, which is NOT terminated by NULL. ++ * @name: original d_name.name ++ * @len: original d_name.len ++ * @wh: whiteout qstr ++ * returns zero when succeeds, otherwise error. ++ * succeeded value as wh->name should be freed by kfree(). ++ */ ++int au_wh_name_alloc(struct qstr *wh, const struct qstr *name) ++{ ++ char *p; ++ ++ if (unlikely(name->len > PATH_MAX - AUFS_WH_PFX_LEN)) ++ return -ENAMETOOLONG; ++ ++ wh->len = name->len + AUFS_WH_PFX_LEN; ++ p = kmalloc(wh->len, GFP_NOFS); ++ wh->name = p; ++ if (p) { ++ memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN); ++ memcpy(p + AUFS_WH_PFX_LEN, name->name, name->len); ++ /* smp_mb(); */ ++ return 0; ++ } ++ return -ENOMEM; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * test if the @wh_name exists under @h_parent. ++ * @try_sio specifies the necessary of super-io. ++ */ ++int au_wh_test(struct dentry *h_parent, struct qstr *wh_name, ++ struct au_branch *br, int try_sio) ++{ ++ int err; ++ struct dentry *wh_dentry; ++ struct inode *h_dir; ++ ++ h_dir = h_parent->d_inode; ++ if (!try_sio) ++ wh_dentry = au_lkup_one(wh_name, h_parent, br, /*nd*/NULL); ++ else ++ wh_dentry = au_sio_lkup_one(wh_name, h_parent, br); ++ err = PTR_ERR(wh_dentry); ++ if (IS_ERR(wh_dentry)) ++ goto out; ++ ++ err = 0; ++ if (!wh_dentry->d_inode) ++ goto out_wh; /* success */ ++ ++ err = 1; ++ if (S_ISREG(wh_dentry->d_inode->i_mode)) ++ goto out_wh; /* success */ ++ ++ err = -EIO; ++ AuIOErr("%.*s Invalid whiteout entry type 0%o.\n", ++ AuDLNPair(wh_dentry), wh_dentry->d_inode->i_mode); ++ ++ out_wh: ++ dput(wh_dentry); ++ out: ++ return err; ++} ++ ++/* ++ * test if the @h_dentry sets opaque or not. ++ */ ++int au_diropq_test(struct dentry *h_dentry, struct au_branch *br) ++{ ++ int err; ++ struct inode *h_dir; ++ ++ h_dir = h_dentry->d_inode; ++ err = au_wh_test(h_dentry, &diropq_name, br, ++ au_test_h_perm_sio(h_dir, MAY_EXEC)); ++ return err; ++} ++ ++/* ++ * returns a negative dentry whose name is unique and temporary. ++ */ ++struct dentry *au_whtmp_lkup(struct dentry *h_parent, struct au_branch *br, ++ struct qstr *prefix) ++{ ++#define HEX_LEN 4 ++ struct dentry *dentry; ++ int i; ++ char defname[AUFS_WH_PFX_LEN * 2 + DNAME_INLINE_LEN_MIN + 1 ++ + HEX_LEN + 1], *name, *p; ++ static unsigned short cnt; ++ struct qstr qs; ++ ++ name = defname; ++ qs.len = sizeof(defname) - DNAME_INLINE_LEN_MIN + prefix->len - 1; ++ if (unlikely(prefix->len > DNAME_INLINE_LEN_MIN)) { ++ dentry = ERR_PTR(-ENAMETOOLONG); ++ if (unlikely(qs.len >= PATH_MAX)) ++ goto out; ++ dentry = ERR_PTR(-ENOMEM); ++ name = kmalloc(qs.len + 1, GFP_NOFS); ++ if (unlikely(!name)) ++ goto out; ++ } ++ ++ /* doubly whiteout-ed */ ++ memcpy(name, AUFS_WH_PFX AUFS_WH_PFX, AUFS_WH_PFX_LEN * 2); ++ p = name + AUFS_WH_PFX_LEN * 2; ++ memcpy(p, prefix->name, prefix->len); ++ p += prefix->len; ++ *p++ = '.'; ++ AuDebugOn(name + qs.len + 1 - p <= HEX_LEN); ++ ++ qs.name = name; ++ for (i = 0; i < 3; i++) { ++ sprintf(p, "%.*d", HEX_LEN, cnt++); ++ dentry = au_sio_lkup_one(&qs, h_parent, br); ++ if (IS_ERR(dentry) || !dentry->d_inode) ++ goto out_name; ++ dput(dentry); ++ } ++ /* AuWarn("could not get random name\n"); */ ++ dentry = ERR_PTR(-EEXIST); ++ AuDbg("%.*s\n", AuLNPair(&qs)); ++ BUG(); ++ ++ out_name: ++ if (name != defname) ++ kfree(name); ++ out: ++ return dentry; ++#undef HEX_LEN ++} ++ ++/* ++ * rename the @h_dentry on @br to the whiteouted temporary name. ++ */ ++int au_whtmp_ren(struct dentry *h_dentry, struct au_branch *br) ++{ ++ int err; ++ struct path h_path = { ++ .mnt = br->br_mnt ++ }; ++ struct inode *h_dir; ++ struct dentry *h_parent; ++ ++ h_parent = h_dentry->d_parent; /* dir inode is locked */ ++ h_dir = h_parent->d_inode; ++ IMustLock(h_dir); ++ ++ h_path.dentry = au_whtmp_lkup(h_parent, br, &h_dentry->d_name); ++ err = PTR_ERR(h_path.dentry); ++ if (IS_ERR(h_path.dentry)) ++ goto out; ++ ++ /* under the same dir, no need to lock_rename() */ ++ err = vfsub_rename(h_dir, h_dentry, h_dir, &h_path); ++ AuTraceErr(err); ++ dput(h_path.dentry); ++ ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++/* ++ * functions for removing a whiteout ++ */ ++ ++static int do_unlink_wh(struct inode *h_dir, struct path *h_path) ++{ ++ int force; ++ ++ /* ++ * forces superio when the dir has a sticky bit. ++ * this may be a violation of unix fs semantics. ++ */ ++ force = (h_dir->i_mode & S_ISVTX) ++ && h_path->dentry->d_inode->i_uid != current_fsuid(); ++ return vfsub_unlink(h_dir, h_path, force); ++} ++ ++int au_wh_unlink_dentry(struct inode *h_dir, struct path *h_path, ++ struct dentry *dentry) ++{ ++ int err; ++ ++ err = do_unlink_wh(h_dir, h_path); ++ if (!err && dentry) ++ au_set_dbwh(dentry, -1); ++ ++ return err; ++} ++ ++static int unlink_wh_name(struct dentry *h_parent, struct qstr *wh, ++ struct au_branch *br) ++{ ++ int err; ++ struct path h_path = { ++ .mnt = br->br_mnt ++ }; ++ ++ err = 0; ++ h_path.dentry = au_lkup_one(wh, h_parent, br, /*nd*/NULL); ++ if (IS_ERR(h_path.dentry)) ++ err = PTR_ERR(h_path.dentry); ++ else { ++ if (h_path.dentry->d_inode ++ && S_ISREG(h_path.dentry->d_inode->i_mode)) ++ err = do_unlink_wh(h_parent->d_inode, &h_path); ++ dput(h_path.dentry); ++ } ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++/* ++ * initialize/clean whiteout for a branch ++ */ ++ ++static void au_wh_clean(struct inode *h_dir, struct path *whpath, ++ const int isdir) ++{ ++ int err; ++ ++ if (!whpath->dentry->d_inode) ++ return; ++ ++ err = mnt_want_write(whpath->mnt); ++ if (!err) { ++ if (isdir) ++ err = vfsub_rmdir(h_dir, whpath); ++ else ++ err = vfsub_unlink(h_dir, whpath, /*force*/0); ++ mnt_drop_write(whpath->mnt); ++ } ++ if (unlikely(err)) ++ AuWarn("failed removing %.*s (%d), ignored.\n", ++ AuDLNPair(whpath->dentry), err); ++} ++ ++static int test_linkable(struct dentry *h_root) ++{ ++ struct inode *h_dir = h_root->d_inode; ++ ++ if (h_dir->i_op->link) ++ return 0; ++ ++ AuErr("%.*s (%s) doesn't support link(2), use noplink and rw+nolwh\n", ++ AuDLNPair(h_root), au_sbtype(h_root->d_sb)); ++ return -ENOSYS; ++} ++ ++/* todo: should this mkdir be done in /sbin/mount.aufs helper? */ ++static int au_whdir(struct inode *h_dir, struct path *path) ++{ ++ int err; ++ ++ err = -EEXIST; ++ if (!path->dentry->d_inode) { ++ int mode = S_IRWXU; ++ ++ if (au_test_nfs(path->dentry->d_sb)) ++ mode |= S_IXUGO; ++ err = mnt_want_write(path->mnt); ++ if (!err) { ++ err = vfsub_mkdir(h_dir, path, mode); ++ mnt_drop_write(path->mnt); ++ } ++ } else if (S_ISDIR(path->dentry->d_inode->i_mode)) ++ err = 0; ++ else ++ AuErr("unknown %.*s exists\n", AuDLNPair(path->dentry)); ++ ++ return err; ++} ++ ++struct au_wh_base { ++ const struct qstr *name; ++ struct dentry *dentry; ++}; ++ ++static void au_wh_init_ro(struct inode *h_dir, struct au_wh_base base[], ++ struct path *h_path) ++{ ++ h_path->dentry = base[AuBrWh_BASE].dentry; ++ au_wh_clean(h_dir, h_path, /*isdir*/0); ++ h_path->dentry = base[AuBrWh_PLINK].dentry; ++ au_wh_clean(h_dir, h_path, /*isdir*/1); ++ h_path->dentry = base[AuBrWh_ORPH].dentry; ++ au_wh_clean(h_dir, h_path, /*isdir*/1); ++} ++ ++/* ++ * returns tri-state, ++ * minus: error, caller should print the mesage ++ * zero: succuess ++ * plus: error, caller should NOT print the mesage ++ */ ++static int au_wh_init_rw_nolink(struct dentry *h_root, struct au_wbr *wbr, ++ int do_plink, struct au_wh_base base[], ++ struct path *h_path) ++{ ++ int err; ++ struct inode *h_dir; ++ ++ h_dir = h_root->d_inode; ++ h_path->dentry = base[AuBrWh_BASE].dentry; ++ au_wh_clean(h_dir, h_path, /*isdir*/0); ++ h_path->dentry = base[AuBrWh_PLINK].dentry; ++ if (do_plink) { ++ err = test_linkable(h_root); ++ if (unlikely(err)) { ++ err = 1; ++ goto out; ++ } ++ ++ err = au_whdir(h_dir, h_path); ++ if (unlikely(err)) ++ goto out; ++ wbr->wbr_plink = dget(base[AuBrWh_PLINK].dentry); ++ } else ++ au_wh_clean(h_dir, h_path, /*isdir*/1); ++ h_path->dentry = base[AuBrWh_ORPH].dentry; ++ err = au_whdir(h_dir, h_path); ++ if (unlikely(err)) ++ goto out; ++ wbr->wbr_orph = dget(base[AuBrWh_ORPH].dentry); ++ ++ out: ++ return err; ++} ++ ++/* ++ * for the moment, aufs supports the branch filesystem which does not support ++ * link(2). testing on FAT which does not support i_op->setattr() fully either, ++ * copyup failed. finally, such filesystem will not be used as the writable ++ * branch. ++ * ++ * returns tri-state, see above. ++ */ ++static int au_wh_init_rw(struct dentry *h_root, struct au_wbr *wbr, ++ int do_plink, struct au_wh_base base[], ++ struct path *h_path) ++{ ++ int err; ++ struct inode *h_dir; ++ ++ WbrWhMustWriteLock(wbr); ++ ++ err = test_linkable(h_root); ++ if (unlikely(err)) { ++ err = 1; ++ goto out; ++ } ++ ++ /* ++ * todo: should this create be done in /sbin/mount.aufs helper? ++ */ ++ err = -EEXIST; ++ h_dir = h_root->d_inode; ++ if (!base[AuBrWh_BASE].dentry->d_inode) { ++ err = mnt_want_write(h_path->mnt); ++ if (!err) { ++ h_path->dentry = base[AuBrWh_BASE].dentry; ++ err = vfsub_create(h_dir, h_path, WH_MASK); ++ mnt_drop_write(h_path->mnt); ++ } ++ } else if (S_ISREG(base[AuBrWh_BASE].dentry->d_inode->i_mode)) ++ err = 0; ++ else ++ AuErr("unknown %.*s/%.*s exists\n", ++ AuDLNPair(h_root), AuDLNPair(base[AuBrWh_BASE].dentry)); ++ if (unlikely(err)) ++ goto out; ++ ++ h_path->dentry = base[AuBrWh_PLINK].dentry; ++ if (do_plink) { ++ err = au_whdir(h_dir, h_path); ++ if (unlikely(err)) ++ goto out; ++ wbr->wbr_plink = dget(base[AuBrWh_PLINK].dentry); ++ } else ++ au_wh_clean(h_dir, h_path, /*isdir*/1); ++ wbr->wbr_whbase = dget(base[AuBrWh_BASE].dentry); ++ ++ h_path->dentry = base[AuBrWh_ORPH].dentry; ++ err = au_whdir(h_dir, h_path); ++ if (unlikely(err)) ++ goto out; ++ wbr->wbr_orph = dget(base[AuBrWh_ORPH].dentry); ++ ++ out: ++ return err; ++} ++ ++/* ++ * initialize the whiteout base file/dir for @br. ++ */ ++int au_wh_init(struct dentry *h_root, struct au_branch *br, ++ struct super_block *sb) ++{ ++ int err, i; ++ const unsigned char do_plink ++ = !!au_opt_test(au_mntflags(sb), PLINK); ++ struct path path = { ++ .mnt = br->br_mnt ++ }; ++ struct inode *h_dir; ++ struct au_wbr *wbr = br->br_wbr; ++ static const struct qstr base_name[] = { ++ [AuBrWh_BASE] = { ++ .name = AUFS_BASE_NAME, ++ .len = sizeof(AUFS_BASE_NAME) - 1 ++ }, ++ [AuBrWh_PLINK] = { ++ .name = AUFS_PLINKDIR_NAME, ++ .len = sizeof(AUFS_PLINKDIR_NAME) - 1 ++ }, ++ [AuBrWh_ORPH] = { ++ .name = AUFS_ORPHDIR_NAME, ++ .len = sizeof(AUFS_ORPHDIR_NAME) - 1 ++ } ++ }; ++ struct au_wh_base base[] = { ++ [AuBrWh_BASE] = { ++ .name = base_name + AuBrWh_BASE, ++ .dentry = NULL ++ }, ++ [AuBrWh_PLINK] = { ++ .name = base_name + AuBrWh_PLINK, ++ .dentry = NULL ++ }, ++ [AuBrWh_ORPH] = { ++ .name = base_name + AuBrWh_ORPH, ++ .dentry = NULL ++ } ++ }; ++ ++ if (wbr) ++ WbrWhMustWriteLock(wbr); ++ ++ h_dir = h_root->d_inode; ++ for (i = 0; i < AuBrWh_Last; i++) { ++ /* doubly whiteouted */ ++ struct dentry *d; ++ ++ d = au_wh_lkup(h_root, (void *)base[i].name, br); ++ err = PTR_ERR(d); ++ if (IS_ERR(d)) ++ goto out; ++ ++ base[i].dentry = d; ++ AuDebugOn(wbr ++ && wbr->wbr_wh[i] ++ && wbr->wbr_wh[i] != base[i].dentry); ++ } ++ ++ if (wbr) ++ for (i = 0; i < AuBrWh_Last; i++) { ++ dput(wbr->wbr_wh[i]); ++ wbr->wbr_wh[i] = NULL; ++ } ++ ++ err = 0; ++ ++ switch (br->br_perm) { ++ case AuBrPerm_RO: ++ case AuBrPerm_ROWH: ++ case AuBrPerm_RR: ++ case AuBrPerm_RRWH: ++ au_wh_init_ro(h_dir, base, &path); ++ break; ++ ++ case AuBrPerm_RWNoLinkWH: ++ err = au_wh_init_rw_nolink(h_root, wbr, do_plink, base, &path); ++ if (err > 0) ++ goto out; ++ else if (err) ++ goto out_err; ++ break; ++ ++ case AuBrPerm_RW: ++ err = au_wh_init_rw(h_root, wbr, do_plink, base, &path); ++ if (err > 0) ++ goto out; ++ else if (err) ++ goto out_err; ++ break; ++ ++ default: ++ BUG(); ++ } ++ goto out; /* success */ ++ ++ out_err: ++ AuErr("an error(%d) on the writable branch %.*s(%s)\n", ++ err, AuDLNPair(h_root), au_sbtype(h_root->d_sb)); ++ out: ++ for (i = 0; i < AuBrWh_Last; i++) ++ dput(base[i].dentry); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++/* ++ * whiteouts are all hard-linked usually. ++ * when its link count reaches a ceiling, we create a new whiteout base ++ * asynchronously. ++ */ ++ ++struct reinit_br_wh { ++ struct super_block *sb; ++ struct au_branch *br; ++}; ++ ++static void reinit_br_wh(void *arg) ++{ ++ int err; ++ aufs_bindex_t bindex; ++ struct path h_path; ++ struct reinit_br_wh *a = arg; ++ struct au_wbr *wbr; ++ struct inode *dir; ++ struct dentry *h_root; ++ struct au_hinode *hdir; ++ ++ err = 0; ++ wbr = a->br->br_wbr; ++ /* big aufs lock */ ++ si_noflush_write_lock(a->sb); ++ if (!au_br_writable(a->br->br_perm)) ++ goto out; ++ bindex = au_br_index(a->sb, a->br->br_id); ++ if (unlikely(bindex < 0)) ++ goto out; ++ ++ di_read_lock_parent(a->sb->s_root, AuLock_IR); ++ dir = a->sb->s_root->d_inode; ++ hdir = au_hi(dir, bindex); ++ h_root = au_h_dptr(a->sb->s_root, bindex); ++ ++ au_hin_imtx_lock_nested(hdir, AuLsc_I_PARENT); ++ wbr_wh_write_lock(wbr); ++ err = au_h_verify(wbr->wbr_whbase, au_opt_udba(a->sb), hdir->hi_inode, ++ h_root, a->br); ++ if (!err) { ++ err = mnt_want_write(a->br->br_mnt); ++ if (!err) { ++ h_path.dentry = wbr->wbr_whbase; ++ h_path.mnt = a->br->br_mnt; ++ err = vfsub_unlink(hdir->hi_inode, &h_path, /*force*/0); ++ mnt_drop_write(a->br->br_mnt); ++ } ++ } else { ++ AuWarn("%.*s is moved, ignored\n", AuDLNPair(wbr->wbr_whbase)); ++ err = 0; ++ } ++ dput(wbr->wbr_whbase); ++ wbr->wbr_whbase = NULL; ++ if (!err) ++ err = au_wh_init(h_root, a->br, a->sb); ++ wbr_wh_write_unlock(wbr); ++ au_hin_imtx_unlock(hdir); ++ di_read_unlock(a->sb->s_root, AuLock_IR); ++ ++ out: ++ if (wbr) ++ atomic_dec(&wbr->wbr_wh_running); ++ atomic_dec(&a->br->br_count); ++ au_nwt_done(&au_sbi(a->sb)->si_nowait); ++ si_write_unlock(a->sb); ++ kfree(arg); ++ if (unlikely(err)) ++ AuIOErr("err %d\n", err); ++} ++ ++static void kick_reinit_br_wh(struct super_block *sb, struct au_branch *br) ++{ ++ int do_dec, wkq_err; ++ struct reinit_br_wh *arg; ++ ++ do_dec = 1; ++ if (atomic_inc_return(&br->br_wbr->wbr_wh_running) != 1) ++ goto out; ++ ++ /* ignore ENOMEM */ ++ arg = kmalloc(sizeof(*arg), GFP_NOFS); ++ if (arg) { ++ /* ++ * dec(wh_running), kfree(arg) and dec(br_count) ++ * in reinit function ++ */ ++ arg->sb = sb; ++ arg->br = br; ++ atomic_inc(&br->br_count); ++ wkq_err = au_wkq_nowait(reinit_br_wh, arg, sb); ++ if (unlikely(wkq_err)) { ++ atomic_dec(&br->br_wbr->wbr_wh_running); ++ atomic_dec(&br->br_count); ++ kfree(arg); ++ } ++ do_dec = 0; ++ } ++ ++ out: ++ if (do_dec) ++ atomic_dec(&br->br_wbr->wbr_wh_running); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * create the whiteout @wh. ++ */ ++static int link_or_create_wh(struct super_block *sb, aufs_bindex_t bindex, ++ struct dentry *wh) ++{ ++ int err; ++ struct path h_path = { ++ .dentry = wh ++ }; ++ struct au_branch *br; ++ struct au_wbr *wbr; ++ struct dentry *h_parent; ++ struct inode *h_dir; ++ ++ h_parent = wh->d_parent; /* dir inode is locked */ ++ h_dir = h_parent->d_inode; ++ IMustLock(h_dir); ++ ++ br = au_sbr(sb, bindex); ++ h_path.mnt = br->br_mnt; ++ wbr = br->br_wbr; ++ wbr_wh_read_lock(wbr); ++ if (wbr->wbr_whbase) { ++ err = vfsub_link(wbr->wbr_whbase, h_dir, &h_path); ++ if (!err || err != -EMLINK) ++ goto out; ++ ++ /* link count full. re-initialize br_whbase. */ ++ kick_reinit_br_wh(sb, br); ++ } ++ ++ /* return this error in this context */ ++ err = vfsub_create(h_dir, &h_path, WH_MASK); ++ ++ out: ++ wbr_wh_read_unlock(wbr); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * create or remove the diropq. ++ */ ++static struct dentry *do_diropq(struct dentry *dentry, aufs_bindex_t bindex, ++ unsigned int flags) ++{ ++ struct dentry *opq_dentry, *h_dentry; ++ struct super_block *sb; ++ struct au_branch *br; ++ int err; ++ ++ sb = dentry->d_sb; ++ br = au_sbr(sb, bindex); ++ h_dentry = au_h_dptr(dentry, bindex); ++ opq_dentry = au_lkup_one(&diropq_name, h_dentry, br, /*nd*/NULL); ++ if (IS_ERR(opq_dentry)) ++ goto out; ++ ++ if (au_ftest_diropq(flags, CREATE)) { ++ err = link_or_create_wh(sb, bindex, opq_dentry); ++ if (!err) { ++ au_set_dbdiropq(dentry, bindex); ++ goto out; /* success */ ++ } ++ } else { ++ struct path tmp = { ++ .dentry = opq_dentry, ++ .mnt = br->br_mnt ++ }; ++ err = do_unlink_wh(au_h_iptr(dentry->d_inode, bindex), &tmp); ++ if (!err) ++ au_set_dbdiropq(dentry, -1); ++ } ++ dput(opq_dentry); ++ opq_dentry = ERR_PTR(err); ++ ++ out: ++ return opq_dentry; ++} ++ ++struct do_diropq_args { ++ struct dentry **errp; ++ struct dentry *dentry; ++ aufs_bindex_t bindex; ++ unsigned int flags; ++}; ++ ++static void call_do_diropq(void *args) ++{ ++ struct do_diropq_args *a = args; ++ *a->errp = do_diropq(a->dentry, a->bindex, a->flags); ++} ++ ++struct dentry *au_diropq_sio(struct dentry *dentry, aufs_bindex_t bindex, ++ unsigned int flags) ++{ ++ struct dentry *diropq, *h_dentry; ++ ++ h_dentry = au_h_dptr(dentry, bindex); ++ if (!au_test_h_perm_sio(h_dentry->d_inode, MAY_EXEC | MAY_WRITE)) ++ diropq = do_diropq(dentry, bindex, flags); ++ else { ++ int wkq_err; ++ struct do_diropq_args args = { ++ .errp = &diropq, ++ .dentry = dentry, ++ .bindex = bindex, ++ .flags = flags ++ }; ++ ++ wkq_err = au_wkq_wait(call_do_diropq, &args); ++ if (unlikely(wkq_err)) ++ diropq = ERR_PTR(wkq_err); ++ } ++ ++ return diropq; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * lookup whiteout dentry. ++ * @h_parent: lower parent dentry which must exist and be locked ++ * @base_name: name of dentry which will be whiteouted ++ * returns dentry for whiteout. ++ */ ++struct dentry *au_wh_lkup(struct dentry *h_parent, struct qstr *base_name, ++ struct au_branch *br) ++{ ++ int err; ++ struct qstr wh_name; ++ struct dentry *wh_dentry; ++ ++ err = au_wh_name_alloc(&wh_name, base_name); ++ wh_dentry = ERR_PTR(err); ++ if (!err) { ++ wh_dentry = au_lkup_one(&wh_name, h_parent, br, /*nd*/NULL); ++ kfree(wh_name.name); ++ } ++ return wh_dentry; ++} ++ ++/* ++ * link/create a whiteout for @dentry on @bindex. ++ */ ++struct dentry *au_wh_create(struct dentry *dentry, aufs_bindex_t bindex, ++ struct dentry *h_parent) ++{ ++ struct dentry *wh_dentry; ++ struct super_block *sb; ++ int err; ++ ++ sb = dentry->d_sb; ++ wh_dentry = au_wh_lkup(h_parent, &dentry->d_name, au_sbr(sb, bindex)); ++ if (!IS_ERR(wh_dentry) && !wh_dentry->d_inode) { ++ err = link_or_create_wh(sb, bindex, wh_dentry); ++ if (!err) ++ au_set_dbwh(dentry, bindex); ++ else { ++ dput(wh_dentry); ++ wh_dentry = ERR_PTR(err); ++ } ++ } ++ ++ return wh_dentry; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* Delete all whiteouts in this directory on branch bindex. */ ++static int del_wh_children(struct dentry *h_dentry, struct au_nhash *whlist, ++ aufs_bindex_t bindex, struct au_branch *br) ++{ ++ int err; ++ unsigned long ul, n; ++ struct qstr wh_name; ++ char *p; ++ struct hlist_head *head; ++ struct au_vdir_wh *tpos; ++ struct hlist_node *pos; ++ struct au_vdir_destr *str; ++ ++ err = -ENOMEM; ++ p = __getname(); ++ wh_name.name = p; ++ if (unlikely(!wh_name.name)) ++ goto out; ++ ++ err = 0; ++ memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN); ++ p += AUFS_WH_PFX_LEN; ++ n = whlist->nh_num; ++ head = whlist->nh_head; ++ for (ul = 0; !err && ul < n; ul++, head++) { ++ hlist_for_each_entry(tpos, pos, head, wh_hash) { ++ if (tpos->wh_bindex != bindex) ++ continue; ++ ++ str = &tpos->wh_str; ++ if (str->len + AUFS_WH_PFX_LEN <= PATH_MAX) { ++ memcpy(p, str->name, str->len); ++ wh_name.len = AUFS_WH_PFX_LEN + str->len; ++ err = unlink_wh_name(h_dentry, &wh_name, br); ++ if (!err) ++ continue; ++ break; ++ } ++ AuIOErr("whiteout name too long %.*s\n", ++ str->len, str->name); ++ err = -EIO; ++ break; ++ } ++ } ++ __putname(wh_name.name); ++ ++ out: ++ return err; ++} ++ ++struct del_wh_children_args { ++ int *errp; ++ struct dentry *h_dentry; ++ struct au_nhash *whlist; ++ aufs_bindex_t bindex; ++ struct au_branch *br; ++}; ++ ++static void call_del_wh_children(void *args) ++{ ++ struct del_wh_children_args *a = args; ++ *a->errp = del_wh_children(a->h_dentry, a->whlist, a->bindex, a->br); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++struct au_whtmp_rmdir *au_whtmp_rmdir_alloc(struct super_block *sb, gfp_t gfp) ++{ ++ struct au_whtmp_rmdir *whtmp; ++ int err; ++ unsigned int rdhash; ++ ++ SiMustAnyLock(sb); ++ ++ whtmp = kmalloc(sizeof(*whtmp), gfp); ++ if (unlikely(!whtmp)) { ++ whtmp = ERR_PTR(-ENOMEM); ++ goto out; ++ } ++ ++ whtmp->dir = NULL; ++ whtmp->wh_dentry = NULL; ++ /* no estimation for dir size */ ++ rdhash = au_sbi(sb)->si_rdhash; ++ if (!rdhash) ++ rdhash = AUFS_RDHASH_DEF; ++ err = au_nhash_alloc(&whtmp->whlist, rdhash, gfp); ++ if (unlikely(err)) { ++ kfree(whtmp); ++ whtmp = ERR_PTR(err); ++ } ++ ++ out: ++ return whtmp; ++} ++ ++void au_whtmp_rmdir_free(struct au_whtmp_rmdir *whtmp) ++{ ++ dput(whtmp->wh_dentry); ++ iput(whtmp->dir); ++ au_nhash_wh_free(&whtmp->whlist); ++ kfree(whtmp); ++} ++ ++/* ++ * rmdir the whiteouted temporary named dir @h_dentry. ++ * @whlist: whiteouted children. ++ */ ++int au_whtmp_rmdir(struct inode *dir, aufs_bindex_t bindex, ++ struct dentry *wh_dentry, struct au_nhash *whlist) ++{ ++ int err; ++ struct path h_tmp; ++ struct inode *wh_inode, *h_dir; ++ struct au_branch *br; ++ ++ h_dir = wh_dentry->d_parent->d_inode; /* dir inode is locked */ ++ IMustLock(h_dir); ++ ++ br = au_sbr(dir->i_sb, bindex); ++ wh_inode = wh_dentry->d_inode; ++ mutex_lock_nested(&wh_inode->i_mutex, AuLsc_I_CHILD); ++ ++ /* ++ * someone else might change some whiteouts while we were sleeping. ++ * it means this whlist may have an obsoleted entry. ++ */ ++ if (!au_test_h_perm_sio(wh_inode, MAY_EXEC | MAY_WRITE)) ++ err = del_wh_children(wh_dentry, whlist, bindex, br); ++ else { ++ int wkq_err; ++ struct del_wh_children_args args = { ++ .errp = &err, ++ .h_dentry = wh_dentry, ++ .whlist = whlist, ++ .bindex = bindex, ++ .br = br ++ }; ++ ++ wkq_err = au_wkq_wait(call_del_wh_children, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ } ++ mutex_unlock(&wh_inode->i_mutex); ++ ++ if (!err) { ++ h_tmp.dentry = wh_dentry; ++ h_tmp.mnt = br->br_mnt; ++ err = vfsub_rmdir(h_dir, &h_tmp); ++ /* d_drop(h_dentry); */ ++ } ++ ++ if (!err) { ++ if (au_ibstart(dir) == bindex) { ++ au_cpup_attr_timesizes(dir); ++ drop_nlink(dir); ++ } ++ return 0; /* success */ ++ } ++ ++ AuWarn("failed removing %.*s(%d), ignored\n", ++ AuDLNPair(wh_dentry), err); ++ return err; ++} ++ ++static void call_rmdir_whtmp(void *args) ++{ ++ int err; ++ struct au_whtmp_rmdir *a = args; ++ struct super_block *sb; ++ struct dentry *h_parent; ++ struct inode *h_dir; ++ struct au_branch *br; ++ struct au_hinode *hdir; ++ ++ /* rmdir by nfsd may cause deadlock with this i_mutex */ ++ /* mutex_lock(&a->dir->i_mutex); */ ++ sb = a->dir->i_sb; ++ si_noflush_read_lock(sb); ++ err = au_test_ro(sb, a->bindex, NULL); ++ if (unlikely(err)) ++ goto out; ++ ++ err = -EIO; ++ br = au_sbr(sb, a->bindex); ++ ii_write_lock_parent(a->dir); ++ h_parent = dget_parent(a->wh_dentry); ++ h_dir = h_parent->d_inode; ++ hdir = au_hi(a->dir, a->bindex); ++ au_hin_imtx_lock_nested(hdir, AuLsc_I_PARENT); ++ err = au_h_verify(a->wh_dentry, au_opt_udba(sb), h_dir, h_parent, br); ++ if (!err) { ++ err = mnt_want_write(br->br_mnt); ++ if (!err) { ++ err = au_whtmp_rmdir(a->dir, a->bindex, a->wh_dentry, ++ &a->whlist); ++ mnt_drop_write(br->br_mnt); ++ } ++ } ++ au_hin_imtx_unlock(hdir); ++ dput(h_parent); ++ ii_write_unlock(a->dir); ++ ++ out: ++ /* mutex_unlock(&a->dir->i_mutex); */ ++ au_nwt_done(&au_sbi(sb)->si_nowait); ++ si_read_unlock(sb); ++ au_whtmp_rmdir_free(a); ++ if (unlikely(err)) ++ AuIOErr("err %d\n", err); ++} ++ ++void au_whtmp_kick_rmdir(struct inode *dir, aufs_bindex_t bindex, ++ struct dentry *wh_dentry, struct au_whtmp_rmdir *args) ++{ ++ int wkq_err; ++ ++ IMustLock(dir); ++ ++ /* all post-process will be done in do_rmdir_whtmp(). */ ++ args->dir = au_igrab(dir); ++ args->bindex = bindex; ++ args->wh_dentry = dget(wh_dentry); ++ wkq_err = au_wkq_nowait(call_rmdir_whtmp, args, dir->i_sb); ++ if (unlikely(wkq_err)) { ++ AuWarn("rmdir error %.*s (%d), ignored\n", ++ AuDLNPair(wh_dentry), wkq_err); ++ au_whtmp_rmdir_free(args); ++ } ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/whout.h linux-2.6.31/fs/aufs/whout.h +--- linux-2.6.31-vanilla/fs/aufs/whout.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/whout.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,87 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * whiteout for logical deletion and opaque directory ++ */ ++ ++#ifndef __AUFS_WHOUT_H__ ++#define __AUFS_WHOUT_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include "dir.h" ++ ++/* whout.c */ ++int au_wh_name_alloc(struct qstr *wh, const struct qstr *name); ++struct au_branch; ++int au_wh_test(struct dentry *h_parent, struct qstr *wh_name, ++ struct au_branch *br, int try_sio); ++int au_diropq_test(struct dentry *h_dentry, struct au_branch *br); ++struct dentry *au_whtmp_lkup(struct dentry *h_parent, struct au_branch *br, ++ struct qstr *prefix); ++int au_whtmp_ren(struct dentry *h_dentry, struct au_branch *br); ++int au_wh_unlink_dentry(struct inode *h_dir, struct path *h_path, ++ struct dentry *dentry); ++int au_wh_init(struct dentry *h_parent, struct au_branch *br, ++ struct super_block *sb); ++ ++/* diropq flags */ ++#define AuDiropq_CREATE 1 ++#define au_ftest_diropq(flags, name) ((flags) & AuDiropq_##name) ++#define au_fset_diropq(flags, name) { (flags) |= AuDiropq_##name; } ++#define au_fclr_diropq(flags, name) { (flags) &= ~AuDiropq_##name; } ++ ++struct dentry *au_diropq_sio(struct dentry *dentry, aufs_bindex_t bindex, ++ unsigned int flags); ++struct dentry *au_wh_lkup(struct dentry *h_parent, struct qstr *base_name, ++ struct au_branch *br); ++struct dentry *au_wh_create(struct dentry *dentry, aufs_bindex_t bindex, ++ struct dentry *h_parent); ++ ++/* real rmdir for the whiteout-ed dir */ ++struct au_whtmp_rmdir { ++ struct inode *dir; ++ aufs_bindex_t bindex; ++ struct dentry *wh_dentry; ++ struct au_nhash whlist; ++}; ++ ++struct au_whtmp_rmdir *au_whtmp_rmdir_alloc(struct super_block *sb, gfp_t gfp); ++void au_whtmp_rmdir_free(struct au_whtmp_rmdir *whtmp); ++int au_whtmp_rmdir(struct inode *dir, aufs_bindex_t bindex, ++ struct dentry *wh_dentry, struct au_nhash *whlist); ++void au_whtmp_kick_rmdir(struct inode *dir, aufs_bindex_t bindex, ++ struct dentry *wh_dentry, struct au_whtmp_rmdir *args); ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline struct dentry *au_diropq_create(struct dentry *dentry, ++ aufs_bindex_t bindex) ++{ ++ return au_diropq_sio(dentry, bindex, AuDiropq_CREATE); ++} ++ ++static inline int au_diropq_remove(struct dentry *dentry, aufs_bindex_t bindex) ++{ ++ return PTR_ERR(au_diropq_sio(dentry, bindex, !AuDiropq_CREATE)); ++} ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_WHOUT_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/wkq.c linux-2.6.31/fs/aufs/wkq.c +--- linux-2.6.31-vanilla/fs/aufs/wkq.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/wkq.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,259 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * workqueue for asynchronous/super-io operations ++ * todo: try new dredential scheme ++ */ ++ ++#include ++#include "aufs.h" ++ ++/* internal workqueue named AUFS_WKQ_NAME */ ++static struct au_wkq { ++ struct workqueue_struct *q; ++ ++ /* balancing */ ++ atomic_t busy; ++} *au_wkq; ++ ++struct au_wkinfo { ++ struct work_struct wk; ++ struct super_block *sb; ++ ++ unsigned int flags; /* see wkq.h */ ++ ++ au_wkq_func_t func; ++ void *args; ++ ++ atomic_t *busyp; ++ struct completion *comp; ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int enqueue(struct au_wkq *wkq, struct au_wkinfo *wkinfo) ++{ ++ wkinfo->busyp = &wkq->busy; ++ if (au_ftest_wkq(wkinfo->flags, WAIT)) ++ return !queue_work(wkq->q, &wkinfo->wk); ++ else ++ return !schedule_work(&wkinfo->wk); ++} ++ ++static void do_wkq(struct au_wkinfo *wkinfo) ++{ ++ unsigned int idle, n; ++ int i, idle_idx; ++ ++ while (1) { ++ if (au_ftest_wkq(wkinfo->flags, WAIT)) { ++ idle_idx = 0; ++ idle = UINT_MAX; ++ for (i = 0; i < aufs_nwkq; i++) { ++ n = atomic_inc_return(&au_wkq[i].busy); ++ if (n == 1 && !enqueue(au_wkq + i, wkinfo)) ++ return; /* success */ ++ ++ if (n < idle) { ++ idle_idx = i; ++ idle = n; ++ } ++ atomic_dec(&au_wkq[i].busy); ++ } ++ } else ++ idle_idx = aufs_nwkq; ++ ++ atomic_inc(&au_wkq[idle_idx].busy); ++ if (!enqueue(au_wkq + idle_idx, wkinfo)) ++ return; /* success */ ++ ++ /* impossible? */ ++ AuWarn1("failed to queue_work()\n"); ++ yield(); ++ } ++} ++ ++static void wkq_func(struct work_struct *wk) ++{ ++ struct au_wkinfo *wkinfo = container_of(wk, struct au_wkinfo, wk); ++ ++ wkinfo->func(wkinfo->args); ++ atomic_dec_return(wkinfo->busyp); ++ if (au_ftest_wkq(wkinfo->flags, WAIT)) ++ complete(wkinfo->comp); ++ else { ++ kobject_put(&au_sbi(wkinfo->sb)->si_kobj); ++ module_put(THIS_MODULE); ++ kfree(wkinfo); ++ } ++} ++ ++/* ++ * Since struct completion is large, try allocating it dynamically. ++ */ ++#if defined(CONFIG_4KSTACKS) || defined(AuTest4KSTACKS) ++#define AuWkqCompDeclare(name) struct completion *comp = NULL ++ ++static int au_wkq_comp_alloc(struct au_wkinfo *wkinfo, struct completion **comp) ++{ ++ *comp = kmalloc(sizeof(**comp), GFP_NOFS); ++ if (*comp) { ++ init_completion(*comp); ++ wkinfo->comp = *comp; ++ return 0; ++ } ++ return -ENOMEM; ++} ++ ++static void au_wkq_comp_free(struct completion *comp) ++{ ++ kfree(comp); ++} ++ ++#else ++ ++/* no braces */ ++#define AuWkqCompDeclare(name) \ ++ DECLARE_COMPLETION_ONSTACK(_ ## name); \ ++ struct completion *comp = &_ ## name ++ ++static int au_wkq_comp_alloc(struct au_wkinfo *wkinfo, struct completion **comp) ++{ ++ wkinfo->comp = *comp; ++ return 0; ++} ++ ++static void au_wkq_comp_free(struct completion *comp __maybe_unused) ++{ ++ /* empty */ ++} ++#endif /* 4KSTACKS */ ++ ++static void au_wkq_run(struct au_wkinfo *wkinfo) ++{ ++ au_dbg_verify_kthread(); ++ INIT_WORK(&wkinfo->wk, wkq_func); ++ do_wkq(wkinfo); ++} ++ ++int au_wkq_wait(au_wkq_func_t func, void *args) ++{ ++ int err; ++ AuWkqCompDeclare(comp); ++ struct au_wkinfo wkinfo = { ++ .flags = AuWkq_WAIT, ++ .func = func, ++ .args = args ++ }; ++ ++ err = au_wkq_comp_alloc(&wkinfo, &comp); ++ if (!err) { ++ au_wkq_run(&wkinfo); ++ /* no timeout, no interrupt */ ++ wait_for_completion(wkinfo.comp); ++ au_wkq_comp_free(comp); ++ } ++ ++ return err; ++ ++} ++ ++int au_wkq_nowait(au_wkq_func_t func, void *args, struct super_block *sb) ++{ ++ int err; ++ struct au_wkinfo *wkinfo; ++ ++ atomic_inc(&au_sbi(sb)->si_nowait.nw_len); ++ ++ /* ++ * wkq_func() must free this wkinfo. ++ * it highly depends upon the implementation of workqueue. ++ */ ++ err = 0; ++ wkinfo = kmalloc(sizeof(*wkinfo), GFP_NOFS); ++ if (wkinfo) { ++ wkinfo->sb = sb; ++ wkinfo->flags = !AuWkq_WAIT; ++ wkinfo->func = func; ++ wkinfo->args = args; ++ wkinfo->comp = NULL; ++ kobject_get(&au_sbi(sb)->si_kobj); ++ __module_get(THIS_MODULE); ++ ++ au_wkq_run(wkinfo); ++ } else { ++ err = -ENOMEM; ++ atomic_dec(&au_sbi(sb)->si_nowait.nw_len); ++ } ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++void au_nwt_init(struct au_nowait_tasks *nwt) ++{ ++ atomic_set(&nwt->nw_len, 0); ++ /* smp_mb();*/ /* atomic_set */ ++ init_waitqueue_head(&nwt->nw_wq); ++} ++ ++void au_wkq_fin(void) ++{ ++ int i; ++ ++ for (i = 0; i < aufs_nwkq; i++) ++ if (au_wkq[i].q && !IS_ERR(au_wkq[i].q)) ++ destroy_workqueue(au_wkq[i].q); ++ kfree(au_wkq); ++} ++ ++int __init au_wkq_init(void) ++{ ++ int err, i; ++ struct au_wkq *nowaitq; ++ ++ /* '+1' is for accounting of nowait queue */ ++ err = -ENOMEM; ++ au_wkq = kcalloc(aufs_nwkq + 1, sizeof(*au_wkq), GFP_NOFS); ++ if (unlikely(!au_wkq)) ++ goto out; ++ ++ err = 0; ++ for (i = 0; i < aufs_nwkq; i++) { ++ au_wkq[i].q = create_singlethread_workqueue(AUFS_WKQ_NAME); ++ if (au_wkq[i].q && !IS_ERR(au_wkq[i].q)) { ++ atomic_set(&au_wkq[i].busy, 0); ++ continue; ++ } ++ ++ err = PTR_ERR(au_wkq[i].q); ++ au_wkq_fin(); ++ goto out; ++ } ++ ++ /* nowait accounting */ ++ nowaitq = au_wkq + aufs_nwkq; ++ atomic_set(&nowaitq->busy, 0); ++ nowaitq->q = NULL; ++ /* smp_mb(); */ /* atomic_set */ ++ ++ out: ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/aufs/wkq.h linux-2.6.31/fs/aufs/wkq.h +--- linux-2.6.31-vanilla/fs/aufs/wkq.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/wkq.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,82 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * workqueue for asynchronous/super-io operations ++ * todo: try new credentials management scheme ++ */ ++ ++#ifndef __AUFS_WKQ_H__ ++#define __AUFS_WKQ_H__ ++ ++#ifdef __KERNEL__ ++ ++#include ++#include ++#include ++ ++struct super_block; ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * in the next operation, wait for the 'nowait' tasks in system-wide workqueue ++ */ ++struct au_nowait_tasks { ++ atomic_t nw_len; ++ wait_queue_head_t nw_wq; ++}; ++ ++/* ---------------------------------------------------------------------- */ ++ ++typedef void (*au_wkq_func_t)(void *args); ++ ++/* wkq flags */ ++#define AuWkq_WAIT 1 ++#define au_ftest_wkq(flags, name) ((flags) & AuWkq_##name) ++#define au_fset_wkq(flags, name) { (flags) |= AuWkq_##name; } ++#define au_fclr_wkq(flags, name) { (flags) &= ~AuWkq_##name; } ++ ++/* wkq.c */ ++int au_wkq_wait(au_wkq_func_t func, void *args); ++int au_wkq_nowait(au_wkq_func_t func, void *args, struct super_block *sb); ++void au_nwt_init(struct au_nowait_tasks *nwt); ++int __init au_wkq_init(void); ++void au_wkq_fin(void); ++ ++/* ---------------------------------------------------------------------- */ ++ ++static inline int au_test_wkq(struct task_struct *tsk) ++{ ++ return !tsk->mm && !strcmp(tsk->comm, AUFS_WKQ_NAME); ++} ++ ++static inline void au_nwt_done(struct au_nowait_tasks *nwt) ++{ ++ if (!atomic_dec_return(&nwt->nw_len)) ++ wake_up_all(&nwt->nw_wq); ++} ++ ++static inline int au_nwt_flush(struct au_nowait_tasks *nwt) ++{ ++ wait_event(nwt->nw_wq, !atomic_read(&nwt->nw_len)); ++ return 0; ++} ++ ++#endif /* __KERNEL__ */ ++#endif /* __AUFS_WKQ_H__ */ +diff -Nur linux-2.6.31-vanilla/fs/aufs/xino.c linux-2.6.31/fs/aufs/xino.c +--- linux-2.6.31-vanilla/fs/aufs/xino.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/fs/aufs/xino.c 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,1203 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++/* ++ * external inode number translation table and bitmap ++ */ ++ ++#include ++#include ++#include ++#include "aufs.h" ++ ++ssize_t xino_fread(au_readf_t func, struct file *file, void *buf, size_t size, ++ loff_t *pos) ++{ ++ ssize_t err; ++ mm_segment_t oldfs; ++ ++ oldfs = get_fs(); ++ set_fs(KERNEL_DS); ++ do { ++ /* todo: signal_pending? */ ++ err = func(file, (char __user *)buf, size, pos); ++ } while (err == -EAGAIN || err == -EINTR); ++ set_fs(oldfs); ++ ++#if 0 /* reserved for future use */ ++ if (err > 0) ++ fsnotify_access(file->f_dentry); ++#endif ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static ssize_t do_xino_fwrite(au_writef_t func, struct file *file, void *buf, ++ size_t size, loff_t *pos) ++{ ++ ssize_t err; ++ mm_segment_t oldfs; ++ ++ oldfs = get_fs(); ++ set_fs(KERNEL_DS); ++ lockdep_off(); ++ do { ++ /* todo: signal_pending? */ ++ err = func(file, (const char __user *)buf, size, pos); ++ } while (err == -EAGAIN || err == -EINTR); ++ lockdep_on(); ++ set_fs(oldfs); ++ ++#if 0 /* reserved for future use */ ++ if (err > 0) ++ fsnotify_modify(file->f_dentry); ++#endif ++ ++ return err; ++} ++ ++struct do_xino_fwrite_args { ++ ssize_t *errp; ++ au_writef_t func; ++ struct file *file; ++ void *buf; ++ size_t size; ++ loff_t *pos; ++}; ++ ++static void call_do_xino_fwrite(void *args) ++{ ++ struct do_xino_fwrite_args *a = args; ++ *a->errp = do_xino_fwrite(a->func, a->file, a->buf, a->size, a->pos); ++} ++ ++ssize_t xino_fwrite(au_writef_t func, struct file *file, void *buf, size_t size, ++ loff_t *pos) ++{ ++ ssize_t err; ++ ++ /* todo: signal block and no wkq? */ ++ /* todo: new credential scheme */ ++ /* ++ * it breaks RLIMIT_FSIZE and normal user's limit, ++ * users should care about quota and real 'filesystem full.' ++ */ ++ if (!au_test_wkq(current)) { ++ int wkq_err; ++ struct do_xino_fwrite_args args = { ++ .errp = &err, ++ .func = func, ++ .file = file, ++ .buf = buf, ++ .size = size, ++ .pos = pos ++ }; ++ ++ wkq_err = au_wkq_wait(call_do_xino_fwrite, &args); ++ if (unlikely(wkq_err)) ++ err = wkq_err; ++ } else ++ err = do_xino_fwrite(func, file, buf, size, pos); ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * create a new xinofile at the same place/path as @base_file. ++ */ ++struct file *au_xino_create2(struct file *base_file, struct file *copy_src) ++{ ++ struct file *file; ++ struct dentry *base, *dentry, *parent; ++ struct inode *dir; ++ struct qstr *name; ++ int err; ++ struct path path; ++ ++ base = base_file->f_dentry; ++ parent = base->d_parent; /* dir inode is locked */ ++ dir = parent->d_inode; ++ IMustLock(dir); ++ ++ file = ERR_PTR(-EINVAL); ++ name = &base->d_name; ++ dentry = vfsub_lookup_one_len(name->name, parent, name->len); ++ if (IS_ERR(dentry)) { ++ file = (void *)dentry; ++ AuErr("%.*s lookup err %ld\n", AuLNPair(name), PTR_ERR(dentry)); ++ goto out; ++ } ++ ++ /* no need to mnt_want_write() since we call dentry_open() later */ ++ err = vfs_create(dir, dentry, S_IRUGO | S_IWUGO, NULL); ++ if (unlikely(err)) { ++ file = ERR_PTR(err); ++ AuErr("%.*s create err %d\n", AuLNPair(name), err); ++ goto out_dput; ++ } ++ ++ path.dentry = dentry; ++ path.mnt = base_file->f_vfsmnt; ++ path_get(&path); ++ file = vfsub_dentry_open(&path, O_RDWR | O_CREAT | O_EXCL | O_LARGEFILE, ++ current_cred()); ++ if (IS_ERR(file)) { ++ AuErr("%.*s open err %ld\n", AuLNPair(name), PTR_ERR(file)); ++ goto out_dput; ++ } ++ ++ err = vfsub_unlink(dir, &file->f_path, /*force*/0); ++ if (unlikely(err)) { ++ AuErr("%.*s unlink err %d\n", AuLNPair(name), err); ++ goto out_fput; ++ } ++ ++ if (copy_src) { ++ /* no one can touch copy_src xino */ ++ err = au_copy_file(file, copy_src, ++ i_size_read(copy_src->f_dentry->d_inode)); ++ if (unlikely(err)) { ++ AuErr("%.*s copy err %d\n", AuLNPair(name), err); ++ goto out_fput; ++ } ++ } ++ goto out_dput; /* success */ ++ ++ out_fput: ++ fput(file); ++ file = ERR_PTR(err); ++ out_dput: ++ dput(dentry); ++ out: ++ return file; ++} ++ ++struct au_xino_lock_dir { ++ struct au_hinode *hdir; ++ struct dentry *parent; ++ struct mutex *mtx; ++}; ++ ++static void au_xino_lock_dir(struct super_block *sb, struct file *xino, ++ struct au_xino_lock_dir *ldir) ++{ ++ aufs_bindex_t brid, bindex; ++ ++ ldir->hdir = NULL; ++ bindex = -1; ++ brid = au_xino_brid(sb); ++ if (brid >= 0) ++ bindex = au_br_index(sb, brid); ++ if (bindex >= 0) { ++ ldir->hdir = au_hi(sb->s_root->d_inode, bindex); ++ au_hin_imtx_lock_nested(ldir->hdir, AuLsc_I_PARENT); ++ } else { ++ ldir->parent = dget_parent(xino->f_dentry); ++ ldir->mtx = &ldir->parent->d_inode->i_mutex; ++ mutex_lock_nested(ldir->mtx, AuLsc_I_PARENT); ++ } ++} ++ ++static void au_xino_unlock_dir(struct au_xino_lock_dir *ldir) ++{ ++ if (ldir->hdir) ++ au_hin_imtx_unlock(ldir->hdir); ++ else { ++ mutex_unlock(ldir->mtx); ++ dput(ldir->parent); ++ } ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* trucate xino files asynchronously */ ++ ++int au_xino_trunc(struct super_block *sb, aufs_bindex_t bindex) ++{ ++ int err; ++ aufs_bindex_t bi, bend; ++ struct au_branch *br; ++ struct file *new_xino, *file; ++ struct super_block *h_sb; ++ struct au_xino_lock_dir ldir; ++ ++ err = -EINVAL; ++ bend = au_sbend(sb); ++ if (unlikely(bindex < 0 || bend < bindex)) ++ goto out; ++ br = au_sbr(sb, bindex); ++ file = br->br_xino.xi_file; ++ if (!file) ++ goto out; ++ ++ au_xino_lock_dir(sb, file, &ldir); ++ /* mnt_want_write() is unnecessary here */ ++ new_xino = au_xino_create2(file, file); ++ au_xino_unlock_dir(&ldir); ++ err = PTR_ERR(new_xino); ++ if (IS_ERR(new_xino)) ++ goto out; ++ err = 0; ++ fput(file); ++ br->br_xino.xi_file = new_xino; ++ ++ h_sb = br->br_mnt->mnt_sb; ++ for (bi = 0; bi <= bend; bi++) { ++ if (unlikely(bi == bindex)) ++ continue; ++ br = au_sbr(sb, bi); ++ if (br->br_mnt->mnt_sb != h_sb) ++ continue; ++ ++ fput(br->br_xino.xi_file); ++ br->br_xino.xi_file = new_xino; ++ get_file(new_xino); ++ } ++ ++ out: ++ return err; ++} ++ ++struct xino_do_trunc_args { ++ struct super_block *sb; ++ struct au_branch *br; ++}; ++ ++static void xino_do_trunc(void *_args) ++{ ++ struct xino_do_trunc_args *args = _args; ++ struct super_block *sb; ++ struct au_branch *br; ++ struct inode *dir; ++ int err; ++ aufs_bindex_t bindex; ++ ++ err = 0; ++ sb = args->sb; ++ dir = sb->s_root->d_inode; ++ br = args->br; ++ ++ si_noflush_write_lock(sb); ++ ii_read_lock_parent(dir); ++ bindex = au_br_index(sb, br->br_id); ++ err = au_xino_trunc(sb, bindex); ++ if (!err ++ && br->br_xino.xi_file->f_dentry->d_inode->i_blocks ++ >= br->br_xino_upper) ++ br->br_xino_upper += AUFS_XINO_TRUNC_STEP; ++ ++ ii_read_unlock(dir); ++ if (unlikely(err)) ++ AuWarn("err b%d, (%d)\n", bindex, err); ++ atomic_dec(&br->br_xino_running); ++ atomic_dec(&br->br_count); ++ au_nwt_done(&au_sbi(sb)->si_nowait); ++ si_write_unlock(sb); ++ kfree(args); ++} ++ ++static void xino_try_trunc(struct super_block *sb, struct au_branch *br) ++{ ++ struct xino_do_trunc_args *args; ++ int wkq_err; ++ ++ if (br->br_xino.xi_file->f_dentry->d_inode->i_blocks ++ < br->br_xino_upper) ++ return; ++ ++ if (atomic_inc_return(&br->br_xino_running) > 1) ++ goto out; ++ ++ /* lock and kfree() will be called in trunc_xino() */ ++ args = kmalloc(sizeof(*args), GFP_NOFS); ++ if (unlikely(!args)) { ++ AuErr1("no memory\n"); ++ goto out_args; ++ } ++ ++ atomic_inc_return(&br->br_count); ++ args->sb = sb; ++ args->br = br; ++ wkq_err = au_wkq_nowait(xino_do_trunc, args, sb); ++ if (!wkq_err) ++ return; /* success */ ++ ++ AuErr("wkq %d\n", wkq_err); ++ atomic_dec_return(&br->br_count); ++ ++ out_args: ++ kfree(args); ++ out: ++ atomic_dec_return(&br->br_xino_running); ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++static int au_xino_do_write(au_writef_t write, struct file *file, ++ ino_t h_ino, ino_t ino) ++{ ++ loff_t pos; ++ ssize_t sz; ++ ++ pos = h_ino; ++ if (unlikely(au_loff_max / sizeof(ino) - 1 < pos)) { ++ AuIOErr1("too large hi%lu\n", (unsigned long)h_ino); ++ return -EFBIG; ++ } ++ pos *= sizeof(ino); ++ sz = xino_fwrite(write, file, &ino, sizeof(ino), &pos); ++ if (sz == sizeof(ino)) ++ return 0; /* success */ ++ ++ AuIOErr("write failed (%zd)\n", sz); ++ return -EIO; ++} ++ ++/* ++ * write @ino to the xinofile for the specified branch{@sb, @bindex} ++ * at the position of @h_ino. ++ * even if @ino is zero, it is written to the xinofile and means no entry. ++ * if the size of the xino file on a specific filesystem exceeds the watermark, ++ * try truncating it. ++ */ ++int au_xino_write(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, ++ ino_t ino) ++{ ++ int err; ++ unsigned int mnt_flags; ++ struct au_branch *br; ++ ++ BUILD_BUG_ON(sizeof(long long) != sizeof(au_loff_max) ++ || ((loff_t)-1) > 0); ++ SiMustAnyLock(sb); ++ ++ mnt_flags = au_mntflags(sb); ++ if (!au_opt_test(mnt_flags, XINO)) ++ return 0; ++ ++ br = au_sbr(sb, bindex); ++ err = au_xino_do_write(au_sbi(sb)->si_xwrite, br->br_xino.xi_file, ++ h_ino, ino); ++ if (!err) { ++ if (au_opt_test(mnt_flags, TRUNC_XINO) ++ && au_test_fs_trunc_xino(br->br_mnt->mnt_sb)) ++ xino_try_trunc(sb, br); ++ return 0; /* success */ ++ } ++ ++ AuIOErr("write failed (%d)\n", err); ++ return -EIO; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* aufs inode number bitmap */ ++ ++static const int page_bits = (int)PAGE_SIZE * BITS_PER_BYTE; ++static ino_t xib_calc_ino(unsigned long pindex, int bit) ++{ ++ ino_t ino; ++ ++ AuDebugOn(bit < 0 || page_bits <= bit); ++ ino = AUFS_FIRST_INO + pindex * page_bits + bit; ++ return ino; ++} ++ ++static void xib_calc_bit(ino_t ino, unsigned long *pindex, int *bit) ++{ ++ AuDebugOn(ino < AUFS_FIRST_INO); ++ ino -= AUFS_FIRST_INO; ++ *pindex = ino / page_bits; ++ *bit = ino % page_bits; ++} ++ ++static int xib_pindex(struct super_block *sb, unsigned long pindex) ++{ ++ int err; ++ loff_t pos; ++ ssize_t sz; ++ struct au_sbinfo *sbinfo; ++ struct file *xib; ++ unsigned long *p; ++ ++ sbinfo = au_sbi(sb); ++ MtxMustLock(&sbinfo->si_xib_mtx); ++ AuDebugOn(pindex > ULONG_MAX / PAGE_SIZE ++ || !au_opt_test(sbinfo->si_mntflags, XINO)); ++ ++ if (pindex == sbinfo->si_xib_last_pindex) ++ return 0; ++ ++ xib = sbinfo->si_xib; ++ p = sbinfo->si_xib_buf; ++ pos = sbinfo->si_xib_last_pindex; ++ pos *= PAGE_SIZE; ++ sz = xino_fwrite(sbinfo->si_xwrite, xib, p, PAGE_SIZE, &pos); ++ if (unlikely(sz != PAGE_SIZE)) ++ goto out; ++ ++ pos = pindex; ++ pos *= PAGE_SIZE; ++ if (i_size_read(xib->f_dentry->d_inode) >= pos + PAGE_SIZE) ++ sz = xino_fread(sbinfo->si_xread, xib, p, PAGE_SIZE, &pos); ++ else { ++ memset(p, 0, PAGE_SIZE); ++ sz = xino_fwrite(sbinfo->si_xwrite, xib, p, PAGE_SIZE, &pos); ++ } ++ if (sz == PAGE_SIZE) { ++ sbinfo->si_xib_last_pindex = pindex; ++ return 0; /* success */ ++ } ++ ++ out: ++ AuIOErr1("write failed (%zd)\n", sz); ++ err = sz; ++ if (sz >= 0) ++ err = -EIO; ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int au_xino_write0(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, ++ ino_t ino) ++{ ++ int err, bit; ++ unsigned long pindex; ++ struct au_sbinfo *sbinfo; ++ ++ if (!au_opt_test(au_mntflags(sb), XINO)) ++ return 0; ++ ++ err = 0; ++ if (ino) { ++ sbinfo = au_sbi(sb); ++ xib_calc_bit(ino, &pindex, &bit); ++ AuDebugOn(page_bits <= bit); ++ mutex_lock(&sbinfo->si_xib_mtx); ++ err = xib_pindex(sb, pindex); ++ if (!err) { ++ clear_bit(bit, sbinfo->si_xib_buf); ++ sbinfo->si_xib_next_bit = bit; ++ } ++ mutex_unlock(&sbinfo->si_xib_mtx); ++ } ++ ++ if (!err) ++ err = au_xino_write(sb, bindex, h_ino, 0); ++ return err; ++} ++ ++/* get an unused inode number from bitmap */ ++ino_t au_xino_new_ino(struct super_block *sb) ++{ ++ ino_t ino; ++ unsigned long *p, pindex, ul, pend; ++ struct au_sbinfo *sbinfo; ++ struct file *file; ++ int free_bit, err; ++ ++ if (!au_opt_test(au_mntflags(sb), XINO)) ++ return iunique(sb, AUFS_FIRST_INO); ++ ++ sbinfo = au_sbi(sb); ++ mutex_lock(&sbinfo->si_xib_mtx); ++ p = sbinfo->si_xib_buf; ++ free_bit = sbinfo->si_xib_next_bit; ++ if (free_bit < page_bits && !test_bit(free_bit, p)) ++ goto out; /* success */ ++ free_bit = find_first_zero_bit(p, page_bits); ++ if (free_bit < page_bits) ++ goto out; /* success */ ++ ++ pindex = sbinfo->si_xib_last_pindex; ++ for (ul = pindex - 1; ul < ULONG_MAX; ul--) { ++ err = xib_pindex(sb, ul); ++ if (unlikely(err)) ++ goto out_err; ++ free_bit = find_first_zero_bit(p, page_bits); ++ if (free_bit < page_bits) ++ goto out; /* success */ ++ } ++ ++ file = sbinfo->si_xib; ++ pend = i_size_read(file->f_dentry->d_inode) / PAGE_SIZE; ++ for (ul = pindex + 1; ul <= pend; ul++) { ++ err = xib_pindex(sb, ul); ++ if (unlikely(err)) ++ goto out_err; ++ free_bit = find_first_zero_bit(p, page_bits); ++ if (free_bit < page_bits) ++ goto out; /* success */ ++ } ++ BUG(); ++ ++ out: ++ set_bit(free_bit, p); ++ sbinfo->si_xib_next_bit++; ++ pindex = sbinfo->si_xib_last_pindex; ++ mutex_unlock(&sbinfo->si_xib_mtx); ++ ino = xib_calc_ino(pindex, free_bit); ++ AuDbg("i%lu\n", (unsigned long)ino); ++ return ino; ++ out_err: ++ mutex_unlock(&sbinfo->si_xib_mtx); ++ AuDbg("i0\n"); ++ return 0; ++} ++ ++/* ++ * read @ino from xinofile for the specified branch{@sb, @bindex} ++ * at the position of @h_ino. ++ * if @ino does not exist and @do_new is true, get new one. ++ */ ++int au_xino_read(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino, ++ ino_t *ino) ++{ ++ int err; ++ ssize_t sz; ++ loff_t pos; ++ struct file *file; ++ struct au_sbinfo *sbinfo; ++ ++ *ino = 0; ++ if (!au_opt_test(au_mntflags(sb), XINO)) ++ return 0; /* no xino */ ++ ++ err = 0; ++ sbinfo = au_sbi(sb); ++ pos = h_ino; ++ if (unlikely(au_loff_max / sizeof(*ino) - 1 < pos)) { ++ AuIOErr1("too large hi%lu\n", (unsigned long)h_ino); ++ return -EFBIG; ++ } ++ pos *= sizeof(*ino); ++ ++ file = au_sbr(sb, bindex)->br_xino.xi_file; ++ if (i_size_read(file->f_dentry->d_inode) < pos + sizeof(*ino)) ++ return 0; /* no ino */ ++ ++ sz = xino_fread(sbinfo->si_xread, file, ino, sizeof(*ino), &pos); ++ if (sz == sizeof(*ino)) ++ return 0; /* success */ ++ ++ err = sz; ++ if (unlikely(sz >= 0)) { ++ err = -EIO; ++ AuIOErr("xino read error (%zd)\n", sz); ++ } ++ ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* create and set a new xino file */ ++ ++struct file *au_xino_create(struct super_block *sb, char *fname, int silent) ++{ ++ struct file *file; ++ struct dentry *h_parent, *d; ++ struct inode *h_dir; ++ int err; ++ ++ /* ++ * at mount-time, and the xino file is the default path, ++ * hinotify is disabled so we have no inotify events to ignore. ++ * when a user specified the xino, we cannot get au_hdir to be ignored. ++ */ ++ file = vfsub_filp_open(fname, O_RDWR | O_CREAT | O_EXCL | O_LARGEFILE, ++ S_IRUGO | S_IWUGO); ++ if (IS_ERR(file)) { ++ if (!silent) ++ AuErr("open %s(%ld)\n", fname, PTR_ERR(file)); ++ return file; ++ } ++ ++ /* keep file count */ ++ h_parent = dget_parent(file->f_dentry); ++ h_dir = h_parent->d_inode; ++ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT); ++ /* mnt_want_write() is unnecessary here */ ++ err = vfsub_unlink(h_dir, &file->f_path, /*force*/0); ++ mutex_unlock(&h_dir->i_mutex); ++ dput(h_parent); ++ if (unlikely(err)) { ++ if (!silent) ++ AuErr("unlink %s(%d)\n", fname, err); ++ goto out; ++ } ++ ++ err = -EINVAL; ++ d = file->f_dentry; ++ if (unlikely(sb == d->d_sb)) { ++ if (!silent) ++ AuErr("%s must be outside\n", fname); ++ goto out; ++ } ++ if (unlikely(au_test_fs_bad_xino(d->d_sb))) { ++ if (!silent) ++ AuErr("xino doesn't support %s(%s)\n", ++ fname, au_sbtype(d->d_sb)); ++ goto out; ++ } ++ return file; /* success */ ++ ++ out: ++ fput(file); ++ file = ERR_PTR(err); ++ return file; ++} ++ ++/* ++ * find another branch who is on the same filesystem of the specified ++ * branch{@btgt}. search until @bend. ++ */ ++static int is_sb_shared(struct super_block *sb, aufs_bindex_t btgt, ++ aufs_bindex_t bend) ++{ ++ aufs_bindex_t bindex; ++ struct super_block *tgt_sb = au_sbr_sb(sb, btgt); ++ ++ for (bindex = 0; bindex < btgt; bindex++) ++ if (unlikely(tgt_sb == au_sbr_sb(sb, bindex))) ++ return bindex; ++ for (bindex++; bindex <= bend; bindex++) ++ if (unlikely(tgt_sb == au_sbr_sb(sb, bindex))) ++ return bindex; ++ return -1; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * initialize the xinofile for the specified branch @br ++ * at the place/path where @base_file indicates. ++ * test whether another branch is on the same filesystem or not, ++ * if @do_test is true. ++ */ ++int au_xino_br(struct super_block *sb, struct au_branch *br, ino_t h_ino, ++ struct file *base_file, int do_test) ++{ ++ int err; ++ ino_t ino; ++ aufs_bindex_t bend, bindex; ++ struct au_branch *shared_br, *b; ++ struct file *file; ++ struct super_block *tgt_sb; ++ ++ shared_br = NULL; ++ bend = au_sbend(sb); ++ if (do_test) { ++ tgt_sb = br->br_mnt->mnt_sb; ++ for (bindex = 0; bindex <= bend; bindex++) { ++ b = au_sbr(sb, bindex); ++ if (tgt_sb == b->br_mnt->mnt_sb) { ++ shared_br = b; ++ break; ++ } ++ } ++ } ++ ++ if (!shared_br || !shared_br->br_xino.xi_file) { ++ struct au_xino_lock_dir ldir; ++ ++ au_xino_lock_dir(sb, base_file, &ldir); ++ /* mnt_want_write() is unnecessary here */ ++ file = au_xino_create2(base_file, NULL); ++ au_xino_unlock_dir(&ldir); ++ err = PTR_ERR(file); ++ if (IS_ERR(file)) ++ goto out; ++ br->br_xino.xi_file = file; ++ } else { ++ br->br_xino.xi_file = shared_br->br_xino.xi_file; ++ get_file(br->br_xino.xi_file); ++ } ++ ++ ino = AUFS_ROOT_INO; ++ err = au_xino_do_write(au_sbi(sb)->si_xwrite, br->br_xino.xi_file, ++ h_ino, ino); ++ if (!err) ++ return 0; /* success */ ++ ++ ++ out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* trucate a xino bitmap file */ ++ ++/* todo: slow */ ++static int do_xib_restore(struct super_block *sb, struct file *file, void *page) ++{ ++ int err, bit; ++ ssize_t sz; ++ unsigned long pindex; ++ loff_t pos, pend; ++ struct au_sbinfo *sbinfo; ++ au_readf_t func; ++ ino_t *ino; ++ unsigned long *p; ++ ++ err = 0; ++ sbinfo = au_sbi(sb); ++ MtxMustLock(&sbinfo->si_xib_mtx); ++ p = sbinfo->si_xib_buf; ++ func = sbinfo->si_xread; ++ pend = i_size_read(file->f_dentry->d_inode); ++ pos = 0; ++ while (pos < pend) { ++ sz = xino_fread(func, file, page, PAGE_SIZE, &pos); ++ err = sz; ++ if (unlikely(sz <= 0)) ++ goto out; ++ ++ err = 0; ++ for (ino = page; sz > 0; ino++, sz -= sizeof(ino)) { ++ if (unlikely(*ino < AUFS_FIRST_INO)) ++ continue; ++ ++ xib_calc_bit(*ino, &pindex, &bit); ++ AuDebugOn(page_bits <= bit); ++ err = xib_pindex(sb, pindex); ++ if (!err) ++ set_bit(bit, p); ++ else ++ goto out; ++ } ++ } ++ ++ out: ++ return err; ++} ++ ++static int xib_restore(struct super_block *sb) ++{ ++ int err; ++ aufs_bindex_t bindex, bend; ++ void *page; ++ ++ err = -ENOMEM; ++ page = (void *)__get_free_page(GFP_NOFS); ++ if (unlikely(!page)) ++ goto out; ++ ++ err = 0; ++ bend = au_sbend(sb); ++ for (bindex = 0; !err && bindex <= bend; bindex++) ++ if (!bindex || is_sb_shared(sb, bindex, bindex - 1) < 0) ++ err = do_xib_restore ++ (sb, au_sbr(sb, bindex)->br_xino.xi_file, page); ++ else ++ AuDbg("b%d\n", bindex); ++ free_page((unsigned long)page); ++ ++ out: ++ return err; ++} ++ ++int au_xib_trunc(struct super_block *sb) ++{ ++ int err; ++ ssize_t sz; ++ loff_t pos; ++ struct au_xino_lock_dir ldir; ++ struct au_sbinfo *sbinfo; ++ unsigned long *p; ++ struct file *file; ++ ++ SiMustWriteLock(sb); ++ ++ err = 0; ++ sbinfo = au_sbi(sb); ++ if (!au_opt_test(sbinfo->si_mntflags, XINO)) ++ goto out; ++ ++ file = sbinfo->si_xib; ++ if (i_size_read(file->f_dentry->d_inode) <= PAGE_SIZE) ++ goto out; ++ ++ au_xino_lock_dir(sb, file, &ldir); ++ /* mnt_want_write() is unnecessary here */ ++ file = au_xino_create2(sbinfo->si_xib, NULL); ++ au_xino_unlock_dir(&ldir); ++ err = PTR_ERR(file); ++ if (IS_ERR(file)) ++ goto out; ++ fput(sbinfo->si_xib); ++ sbinfo->si_xib = file; ++ ++ p = sbinfo->si_xib_buf; ++ memset(p, 0, PAGE_SIZE); ++ pos = 0; ++ sz = xino_fwrite(sbinfo->si_xwrite, sbinfo->si_xib, p, PAGE_SIZE, &pos); ++ if (unlikely(sz != PAGE_SIZE)) { ++ err = sz; ++ AuIOErr("err %d\n", err); ++ if (sz >= 0) ++ err = -EIO; ++ goto out; ++ } ++ ++ mutex_lock(&sbinfo->si_xib_mtx); ++ /* mnt_want_write() is unnecessary here */ ++ err = xib_restore(sb); ++ mutex_unlock(&sbinfo->si_xib_mtx); ++ ++out: ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * xino mount option handlers ++ */ ++static au_readf_t find_readf(struct file *h_file) ++{ ++ const struct file_operations *fop = h_file->f_op; ++ ++ if (fop) { ++ if (fop->read) ++ return fop->read; ++ if (fop->aio_read) ++ return do_sync_read; ++ } ++ return ERR_PTR(-ENOSYS); ++} ++ ++static au_writef_t find_writef(struct file *h_file) ++{ ++ const struct file_operations *fop = h_file->f_op; ++ ++ if (fop) { ++ if (fop->write) ++ return fop->write; ++ if (fop->aio_write) ++ return do_sync_write; ++ } ++ return ERR_PTR(-ENOSYS); ++} ++ ++/* xino bitmap */ ++static void xino_clear_xib(struct super_block *sb) ++{ ++ struct au_sbinfo *sbinfo; ++ ++ SiMustWriteLock(sb); ++ ++ sbinfo = au_sbi(sb); ++ sbinfo->si_xread = NULL; ++ sbinfo->si_xwrite = NULL; ++ if (sbinfo->si_xib) ++ fput(sbinfo->si_xib); ++ sbinfo->si_xib = NULL; ++ free_page((unsigned long)sbinfo->si_xib_buf); ++ sbinfo->si_xib_buf = NULL; ++} ++ ++static int au_xino_set_xib(struct super_block *sb, struct file *base) ++{ ++ int err; ++ loff_t pos; ++ struct au_sbinfo *sbinfo; ++ struct file *file; ++ ++ SiMustWriteLock(sb); ++ ++ sbinfo = au_sbi(sb); ++ file = au_xino_create2(base, sbinfo->si_xib); ++ err = PTR_ERR(file); ++ if (IS_ERR(file)) ++ goto out; ++ if (sbinfo->si_xib) ++ fput(sbinfo->si_xib); ++ sbinfo->si_xib = file; ++ sbinfo->si_xread = find_readf(file); ++ sbinfo->si_xwrite = find_writef(file); ++ ++ err = -ENOMEM; ++ if (!sbinfo->si_xib_buf) ++ sbinfo->si_xib_buf = (void *)get_zeroed_page(GFP_NOFS); ++ if (unlikely(!sbinfo->si_xib_buf)) ++ goto out_unset; ++ ++ sbinfo->si_xib_last_pindex = 0; ++ sbinfo->si_xib_next_bit = 0; ++ if (i_size_read(file->f_dentry->d_inode) < PAGE_SIZE) { ++ pos = 0; ++ err = xino_fwrite(sbinfo->si_xwrite, file, sbinfo->si_xib_buf, ++ PAGE_SIZE, &pos); ++ if (unlikely(err != PAGE_SIZE)) ++ goto out_free; ++ } ++ err = 0; ++ goto out; /* success */ ++ ++ out_free: ++ free_page((unsigned long)sbinfo->si_xib_buf); ++ sbinfo->si_xib_buf = NULL; ++ if (err >= 0) ++ err = -EIO; ++ out_unset: ++ fput(sbinfo->si_xib); ++ sbinfo->si_xib = NULL; ++ sbinfo->si_xread = NULL; ++ sbinfo->si_xwrite = NULL; ++ out: ++ return err; ++} ++ ++/* xino for each branch */ ++static void xino_clear_br(struct super_block *sb) ++{ ++ aufs_bindex_t bindex, bend; ++ struct au_branch *br; ++ ++ bend = au_sbend(sb); ++ for (bindex = 0; bindex <= bend; bindex++) { ++ br = au_sbr(sb, bindex); ++ if (!br || !br->br_xino.xi_file) ++ continue; ++ ++ fput(br->br_xino.xi_file); ++ br->br_xino.xi_file = NULL; ++ } ++} ++ ++static int au_xino_set_br(struct super_block *sb, struct file *base) ++{ ++ int err; ++ ino_t ino; ++ aufs_bindex_t bindex, bend, bshared; ++ struct { ++ struct file *old, *new; ++ } *fpair, *p; ++ struct au_branch *br; ++ struct inode *inode; ++ au_writef_t writef; ++ ++ SiMustWriteLock(sb); ++ ++ err = -ENOMEM; ++ bend = au_sbend(sb); ++ fpair = kcalloc(bend + 1, sizeof(*fpair), GFP_NOFS); ++ if (unlikely(!fpair)) ++ goto out; ++ ++ inode = sb->s_root->d_inode; ++ ino = AUFS_ROOT_INO; ++ writef = au_sbi(sb)->si_xwrite; ++ for (bindex = 0, p = fpair; bindex <= bend; bindex++, p++) { ++ br = au_sbr(sb, bindex); ++ bshared = is_sb_shared(sb, bindex, bindex - 1); ++ if (bshared >= 0) { ++ /* shared xino */ ++ *p = fpair[bshared]; ++ get_file(p->new); ++ } ++ ++ if (!p->new) { ++ /* new xino */ ++ p->old = br->br_xino.xi_file; ++ p->new = au_xino_create2(base, br->br_xino.xi_file); ++ err = PTR_ERR(p->new); ++ if (IS_ERR(p->new)) { ++ p->new = NULL; ++ goto out_pair; ++ } ++ } ++ ++ err = au_xino_do_write(writef, p->new, ++ au_h_iptr(inode, bindex)->i_ino, ino); ++ if (unlikely(err)) ++ goto out_pair; ++ } ++ ++ for (bindex = 0, p = fpair; bindex <= bend; bindex++, p++) { ++ br = au_sbr(sb, bindex); ++ if (br->br_xino.xi_file) ++ fput(br->br_xino.xi_file); ++ get_file(p->new); ++ br->br_xino.xi_file = p->new; ++ } ++ ++ out_pair: ++ for (bindex = 0, p = fpair; bindex <= bend; bindex++, p++) ++ if (p->new) ++ fput(p->new); ++ else ++ break; ++ kfree(fpair); ++ out: ++ return err; ++} ++ ++void au_xino_clr(struct super_block *sb) ++{ ++ struct au_sbinfo *sbinfo; ++ ++ au_xigen_clr(sb); ++ xino_clear_xib(sb); ++ xino_clear_br(sb); ++ sbinfo = au_sbi(sb); ++ /* lvalue, do not call au_mntflags() */ ++ au_opt_clr(sbinfo->si_mntflags, XINO); ++} ++ ++int au_xino_set(struct super_block *sb, struct au_opt_xino *xino, int remount) ++{ ++ int err, skip; ++ struct dentry *parent, *cur_parent; ++ struct qstr *dname, *cur_name; ++ struct file *cur_xino; ++ struct inode *dir; ++ struct au_sbinfo *sbinfo; ++ ++ SiMustWriteLock(sb); ++ ++ err = 0; ++ sbinfo = au_sbi(sb); ++ parent = dget_parent(xino->file->f_dentry); ++ if (remount) { ++ skip = 0; ++ dname = &xino->file->f_dentry->d_name; ++ cur_xino = sbinfo->si_xib; ++ if (cur_xino) { ++ cur_parent = dget_parent(cur_xino->f_dentry); ++ cur_name = &cur_xino->f_dentry->d_name; ++ skip = (cur_parent == parent ++ && dname->len == cur_name->len ++ && !memcmp(dname->name, cur_name->name, ++ dname->len)); ++ dput(cur_parent); ++ } ++ if (skip) ++ goto out; ++ } ++ ++ au_opt_set(sbinfo->si_mntflags, XINO); ++ dir = parent->d_inode; ++ mutex_lock_nested(&dir->i_mutex, AuLsc_I_PARENT); ++ /* mnt_want_write() is unnecessary here */ ++ err = au_xino_set_xib(sb, xino->file); ++ if (!err) ++ err = au_xigen_set(sb, xino->file); ++ if (!err) ++ err = au_xino_set_br(sb, xino->file); ++ mutex_unlock(&dir->i_mutex); ++ if (!err) ++ goto out; /* success */ ++ ++ /* reset all */ ++ AuIOErr("failed creating xino(%d).\n", err); ++ ++ out: ++ dput(parent); ++ return err; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ++ * create a xinofile at the default place/path. ++ */ ++struct file *au_xino_def(struct super_block *sb) ++{ ++ struct file *file; ++ char *page, *p; ++ struct au_branch *br; ++ struct super_block *h_sb; ++ struct path path; ++ aufs_bindex_t bend, bindex, bwr; ++ ++ br = NULL; ++ bend = au_sbend(sb); ++ bwr = -1; ++ for (bindex = 0; bindex <= bend; bindex++) { ++ br = au_sbr(sb, bindex); ++ if (au_br_writable(br->br_perm) ++ && !au_test_fs_bad_xino(br->br_mnt->mnt_sb)) { ++ bwr = bindex; ++ break; ++ } ++ } ++ ++ if (bwr >= 0) { ++ file = ERR_PTR(-ENOMEM); ++ page = __getname(); ++ if (unlikely(!page)) ++ goto out; ++ path.mnt = br->br_mnt; ++ path.dentry = au_h_dptr(sb->s_root, bwr); ++ p = d_path(&path, page, PATH_MAX - sizeof(AUFS_XINO_FNAME)); ++ file = (void *)p; ++ if (!IS_ERR(p)) { ++ strcat(p, "/" AUFS_XINO_FNAME); ++ AuDbg("%s\n", p); ++ file = au_xino_create(sb, p, /*silent*/0); ++ if (!IS_ERR(file)) ++ au_xino_brid_set(sb, br->br_id); ++ } ++ __putname(page); ++ } else { ++ file = au_xino_create(sb, AUFS_XINO_DEFPATH, /*silent*/0); ++ if (IS_ERR(file)) ++ goto out; ++ h_sb = file->f_dentry->d_sb; ++ if (unlikely(au_test_fs_bad_xino(h_sb))) { ++ AuErr("xino doesn't support %s(%s)\n", ++ AUFS_XINO_DEFPATH, au_sbtype(h_sb)); ++ fput(file); ++ file = ERR_PTR(-EINVAL); ++ } ++ if (!IS_ERR(file)) ++ au_xino_brid_set(sb, -1); ++ } ++ ++ out: ++ return file; ++} ++ ++/* ---------------------------------------------------------------------- */ ++ ++int au_xino_path(struct seq_file *seq, struct file *file) ++{ ++ int err; ++ ++ err = au_seq_path(seq, &file->f_path); ++ if (unlikely(err < 0)) ++ goto out; ++ ++ err = 0; ++#define Deleted "\\040(deleted)" ++ seq->count -= sizeof(Deleted) - 1; ++ AuDebugOn(memcmp(seq->buf + seq->count, Deleted, ++ sizeof(Deleted) - 1)); ++#undef Deleted ++ ++ out: ++ return err; ++} +diff -Nur linux-2.6.31-vanilla/fs/Kconfig linux-2.6.31/fs/Kconfig +--- linux-2.6.31-vanilla/fs/Kconfig 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/fs/Kconfig 2009-09-16 13:55:56.000000000 +0200 +@@ -187,6 +187,7 @@ + source "fs/ufs/Kconfig" + source "fs/exofs/Kconfig" + source "fs/nilfs2/Kconfig" ++source "fs/aufs/Kconfig" + + endif # MISC_FILESYSTEMS + +diff -Nur linux-2.6.31-vanilla/fs/Makefile linux-2.6.31/fs/Makefile +--- linux-2.6.31-vanilla/fs/Makefile 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/fs/Makefile 2009-09-16 13:55:56.000000000 +0200 +@@ -124,3 +124,4 @@ + obj-$(CONFIG_BTRFS_FS) += btrfs/ + obj-$(CONFIG_GFS2_FS) += gfs2/ + obj-$(CONFIG_EXOFS_FS) += exofs/ ++obj-$(CONFIG_AUFS_FS) += aufs/ +diff -Nur linux-2.6.31-vanilla/fs/namei.c linux-2.6.31/fs/namei.c +--- linux-2.6.31-vanilla/fs/namei.c 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/fs/namei.c 2009-09-16 13:55:49.000000000 +0200 +@@ -337,6 +337,7 @@ + + return 0; + } ++EXPORT_SYMBOL(deny_write_access); + + /** + * path_get - get a reference to a path +@@ -1219,7 +1220,7 @@ + * needs parent already locked. Doesn't follow mounts. + * SMP-safe. + */ +-static struct dentry *lookup_hash(struct nameidata *nd) ++struct dentry *lookup_hash(struct nameidata *nd) + { + int err; + +@@ -1228,8 +1229,9 @@ + return ERR_PTR(err); + return __lookup_hash(&nd->last, nd->path.dentry, nd); + } ++EXPORT_SYMBOL(lookup_hash); + +-static int __lookup_one_len(const char *name, struct qstr *this, ++int __lookup_one_len(const char *name, struct qstr *this, + struct dentry *base, int len) + { + unsigned long hash; +@@ -1250,6 +1252,7 @@ + this->hash = end_name_hash(hash); + return 0; + } ++EXPORT_SYMBOL(__lookup_one_len); + + /** + * lookup_one_len - filesystem helper to lookup single pathname component +diff -Nur linux-2.6.31-vanilla/fs/namespace.c linux-2.6.31/fs/namespace.c +--- linux-2.6.31-vanilla/fs/namespace.c 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/fs/namespace.c 2009-09-16 13:55:49.000000000 +0200 +@@ -39,6 +39,7 @@ + + /* spinlock for vfsmount related operations, inplace of dcache_lock */ + __cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock); ++EXPORT_SYMBOL(vfsmount_lock); + + static int event; + static DEFINE_IDA(mnt_id_ida); +diff -Nur linux-2.6.31-vanilla/fs/open.c linux-2.6.31/fs/open.c +--- linux-2.6.31-vanilla/fs/open.c 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/fs/open.c 2009-09-16 13:55:49.000000000 +0200 +@@ -221,6 +221,7 @@ + mutex_unlock(&dentry->d_inode->i_mutex); + return err; + } ++EXPORT_SYMBOL(do_truncate); + + static long do_sys_truncate(const char __user *pathname, loff_t length) + { +diff -Nur linux-2.6.31-vanilla/fs/splice.c linux-2.6.31/fs/splice.c +--- linux-2.6.31-vanilla/fs/splice.c 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/fs/splice.c 2009-09-16 13:55:49.000000000 +0200 +@@ -1057,8 +1057,8 @@ + /* + * Attempt to initiate a splice from pipe to file. + */ +-static long do_splice_from(struct pipe_inode_info *pipe, struct file *out, +- loff_t *ppos, size_t len, unsigned int flags) ++long do_splice_from(struct pipe_inode_info *pipe, struct file *out, ++ loff_t *ppos, size_t len, unsigned int flags) + { + ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, + loff_t *, size_t, unsigned int); +@@ -1080,13 +1080,14 @@ + + return splice_write(pipe, out, ppos, len, flags); + } ++EXPORT_SYMBOL(do_splice_from); + + /* + * Attempt to initiate a splice from a file to a pipe. + */ +-static long do_splice_to(struct file *in, loff_t *ppos, +- struct pipe_inode_info *pipe, size_t len, +- unsigned int flags) ++long do_splice_to(struct file *in, loff_t *ppos, ++ struct pipe_inode_info *pipe, size_t len, ++ unsigned int flags) + { + ssize_t (*splice_read)(struct file *, loff_t *, + struct pipe_inode_info *, size_t, unsigned int); +@@ -1105,6 +1106,7 @@ + + return splice_read(in, ppos, pipe, len, flags); + } ++EXPORT_SYMBOL(do_splice_to); + + /** + * splice_direct_to_actor - splices data directly between two non-pipes +diff -Nur linux-2.6.31-vanilla/include/linux/aufs_type.h linux-2.6.31/include/linux/aufs_type.h +--- linux-2.6.31-vanilla/include/linux/aufs_type.h 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/include/linux/aufs_type.h 2009-09-16 13:55:30.000000000 +0200 +@@ -0,0 +1,184 @@ ++/* ++ * Copyright (C) 2005-2009 Junjiro R. Okajima ++ * ++ * This program, aufs is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License as published by ++ * the Free Software Foundation; either version 2 of the License, or ++ * (at your option) any later version. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public License ++ * along with this program; if not, write to the Free Software ++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA ++ */ ++ ++#ifndef __AUFS_TYPE_H__ ++#define __AUFS_TYPE_H__ ++ ++#include ++#include ++ ++#define AUFS_VERSION "2-standalone.tree-20090914" ++ ++/* todo? move this to linux-2.6.19/include/magic.h */ ++#define AUFS_SUPER_MAGIC ('a' << 24 | 'u' << 16 | 'f' << 8 | 's') ++ ++/* ---------------------------------------------------------------------- */ ++ ++#ifdef CONFIG_AUFS_BRANCH_MAX_127 ++typedef __s8 aufs_bindex_t; ++#define AUFS_BRANCH_MAX 127 ++#else ++typedef __s16 aufs_bindex_t; ++#ifdef CONFIG_AUFS_BRANCH_MAX_511 ++#define AUFS_BRANCH_MAX 511 ++#elif defined(CONFIG_AUFS_BRANCH_MAX_1023) ++#define AUFS_BRANCH_MAX 1023 ++#elif defined(CONFIG_AUFS_BRANCH_MAX_32767) ++#define AUFS_BRANCH_MAX 32767 ++#endif ++#endif ++ ++#ifdef __KERNEL__ ++#ifndef AUFS_BRANCH_MAX ++#error unknown CONFIG_AUFS_BRANCH_MAX value ++#endif ++#endif /* __KERNEL__ */ ++ ++/* ---------------------------------------------------------------------- */ ++ ++#define AUFS_NAME "aufs" ++#define AUFS_FSTYPE AUFS_NAME ++ ++#define AUFS_ROOT_INO 2 ++#define AUFS_FIRST_INO 11 ++ ++#define AUFS_WH_PFX ".wh." ++#define AUFS_WH_PFX_LEN ((int)sizeof(AUFS_WH_PFX) - 1) ++#define AUFS_XINO_FNAME "." AUFS_NAME ".xino" ++#define AUFS_XINO_DEFPATH "/tmp/" AUFS_XINO_FNAME ++#define AUFS_XINO_TRUNC_INIT 64 /* blocks */ ++#define AUFS_XINO_TRUNC_STEP 4 /* blocks */ ++#define AUFS_DIRWH_DEF 3 ++#define AUFS_RDCACHE_DEF 10 /* seconds */ ++#define AUFS_RDBLK_DEF 512 /* bytes */ ++#define AUFS_RDHASH_DEF 32 ++#define AUFS_WKQ_NAME AUFS_NAME "d" ++#define AUFS_NWKQ_DEF 4 ++#define AUFS_MFS_SECOND_DEF 30 /* seconds */ ++#define AUFS_PLINK_WARN 100 /* number of plinks */ ++ ++#define AUFS_DIROPQ_NAME AUFS_WH_PFX ".opq" /* whiteouted doubly */ ++#define AUFS_WH_DIROPQ AUFS_WH_PFX AUFS_DIROPQ_NAME ++ ++#define AUFS_BASE_NAME AUFS_WH_PFX AUFS_NAME ++#define AUFS_PLINKDIR_NAME AUFS_WH_PFX "plnk" ++#define AUFS_ORPHDIR_NAME AUFS_WH_PFX "orph" ++ ++/* doubly whiteouted */ ++#define AUFS_WH_BASE AUFS_WH_PFX AUFS_BASE_NAME ++#define AUFS_WH_PLINKDIR AUFS_WH_PFX AUFS_PLINKDIR_NAME ++#define AUFS_WH_ORPHDIR AUFS_WH_PFX AUFS_ORPHDIR_NAME ++ ++/* branch permission */ ++#define AUFS_BRPERM_RW "rw" ++#define AUFS_BRPERM_RO "ro" ++#define AUFS_BRPERM_RR "rr" ++#define AUFS_BRPERM_WH "wh" ++#define AUFS_BRPERM_NLWH "nolwh" ++#define AUFS_BRPERM_ROWH AUFS_BRPERM_RO "+" AUFS_BRPERM_WH ++#define AUFS_BRPERM_RRWH AUFS_BRPERM_RR "+" AUFS_BRPERM_WH ++#define AUFS_BRPERM_RWNLWH AUFS_BRPERM_RW "+" AUFS_BRPERM_NLWH ++ ++/* ---------------------------------------------------------------------- */ ++ ++/* ioctl */ ++enum { ++ AuCtl_PLINK_MAINT, ++ AuCtl_PLINK_CLEAN, ++ ++ /* readdir in userspace */ ++ AuCtl_RDU, ++ AuCtl_RDU_INO ++}; ++ ++/* borrowed from linux/include/linux/kernel.h */ ++#ifndef ALIGN ++#define ALIGN(x, a) __ALIGN_MASK(x, (typeof(x))(a)-1) ++#define __ALIGN_MASK(x, mask) (((x)+(mask))&~(mask)) ++#endif ++ ++/* borrowed from linux/include/linux/compiler-gcc3.h */ ++#ifndef __aligned ++#define __aligned(x) __attribute__((aligned(x))) ++#define __packed __attribute__((packed)) ++#endif ++ ++struct au_rdu_cookie { ++ __u64 h_pos; ++ __s16 bindex; ++ __u8 flags; ++ __u8 pad; ++ __u32 generation; ++} __aligned(8); ++ ++struct au_rdu_ent { ++ __u64 ino; ++ __s16 bindex; ++ __u8 type; ++ __u8 nlen; ++ __u8 wh; ++ char name[0]; ++} __aligned(8); ++ ++static inline int au_rdu_len(int nlen) ++{ ++ /* include the terminating NULL */ ++ return ALIGN(sizeof(struct au_rdu_ent) + nlen + 1, ++ sizeof(__u64)); ++} ++ ++union au_rdu_ent_ul { ++ struct au_rdu_ent __user *e; ++ unsigned long ul; ++}; ++ ++enum { ++ AufsCtlRduV_SZ, ++ AufsCtlRduV_SZ_PTR, ++ AufsCtlRduV_End ++}; ++ ++struct aufs_rdu { ++ /* input */ ++ union { ++ __u64 sz; /* AuCtl_RDU */ ++ __u64 nent; /* AuCtl_RDU_INO */ ++ }; ++ union au_rdu_ent_ul ent; ++ __u16 verify[AufsCtlRduV_End]; ++ ++ /* input/output */ ++ __u32 blk; ++ ++ /* output */ ++ union au_rdu_ent_ul tail; ++ /* number of entries which were added in a single call */ ++ __u64 rent; ++ __u8 full; ++ __u8 shwh; ++ ++ struct au_rdu_cookie cookie; ++} __aligned(8); ++ ++#define AuCtlType 'A' ++#define AUFS_CTL_PLINK_MAINT _IO(AuCtlType, AuCtl_PLINK_MAINT) ++#define AUFS_CTL_PLINK_CLEAN _IO(AuCtlType, AuCtl_PLINK_CLEAN) ++#define AUFS_CTL_RDU _IOWR(AuCtlType, AuCtl_RDU, struct aufs_rdu) ++#define AUFS_CTL_RDU_INO _IOWR(AuCtlType, AuCtl_RDU_INO, struct aufs_rdu) ++ ++#endif /* __AUFS_TYPE_H__ */ +diff -Nur linux-2.6.31-vanilla/include/linux/Kbuild linux-2.6.31/include/linux/Kbuild +--- linux-2.6.31-vanilla/include/linux/Kbuild 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/include/linux/Kbuild 2009-09-16 13:55:56.000000000 +0200 +@@ -34,6 +34,7 @@ + header-y += atmsap.h + header-y += atmsvc.h + header-y += atm_zatm.h ++header-y += aufs_type.h + header-y += auto_fs4.h + header-y += ax25.h + header-y += b1lli.h +diff -Nur linux-2.6.31-vanilla/include/linux/namei.h linux-2.6.31/include/linux/namei.h +--- linux-2.6.31-vanilla/include/linux/namei.h 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/include/linux/namei.h 2009-09-16 13:55:46.000000000 +0200 +@@ -75,6 +75,9 @@ + extern struct file *nameidata_to_filp(struct nameidata *nd, int flags); + extern void release_open_intent(struct nameidata *); + ++extern struct dentry *lookup_hash(struct nameidata *nd); ++extern int __lookup_one_len(const char *name, struct qstr *this, ++ struct dentry *base, int len); + extern struct dentry *lookup_one_len(const char *, struct dentry *, int); + extern struct dentry *lookup_one_noperm(const char *, struct dentry *); + +diff -Nur linux-2.6.31-vanilla/include/linux/splice.h linux-2.6.31/include/linux/splice.h +--- linux-2.6.31-vanilla/include/linux/splice.h 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/include/linux/splice.h 2009-09-16 13:55:46.000000000 +0200 +@@ -82,4 +82,10 @@ + extern ssize_t splice_direct_to_actor(struct file *, struct splice_desc *, + splice_direct_actor *); + ++extern long do_splice_from(struct pipe_inode_info *pipe, struct file *out, ++ loff_t *ppos, size_t len, unsigned int flags); ++extern long do_splice_to(struct file *in, loff_t *ppos, ++ struct pipe_inode_info *pipe, size_t len, ++ unsigned int flags); ++ + #endif +diff -Nur linux-2.6.31-vanilla/security/device_cgroup.c linux-2.6.31/security/device_cgroup.c +--- linux-2.6.31-vanilla/security/device_cgroup.c 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/security/device_cgroup.c 2009-09-16 13:55:49.000000000 +0200 +@@ -513,6 +513,7 @@ + + return -EPERM; + } ++EXPORT_SYMBOL(devcgroup_inode_permission); + + int devcgroup_inode_mknod(int mode, dev_t dev) + { +diff -Nur linux-2.6.31-vanilla/security/integrity/ima/ima_main.c linux-2.6.31/security/integrity/ima/ima_main.c +--- linux-2.6.31-vanilla/security/integrity/ima/ima_main.c 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/security/integrity/ima/ima_main.c 2009-09-16 13:55:49.000000000 +0200 +@@ -324,6 +324,7 @@ + MAY_EXEC, FILE_MMAP); + return 0; + } ++EXPORT_SYMBOL(ima_file_mmap); + + /** + * ima_bprm_check - based on policy, collect/store measurement. +diff -Nur linux-2.6.31-vanilla/security/integrity/ima/ima_main.c.orig linux-2.6.31/security/integrity/ima/ima_main.c.orig +--- linux-2.6.31-vanilla/security/integrity/ima/ima_main.c.orig 1970-01-01 01:00:00.000000000 +0100 ++++ linux-2.6.31/security/integrity/ima/ima_main.c.orig 2009-09-10 00:13:59.000000000 +0200 +@@ -0,0 +1,368 @@ ++/* ++ * Copyright (C) 2005,2006,2007,2008 IBM Corporation ++ * ++ * Authors: ++ * Reiner Sailer ++ * Serge Hallyn ++ * Kylene Hall ++ * Mimi Zohar ++ * ++ * This program is free software; you can redistribute it and/or ++ * modify it under the terms of the GNU General Public License as ++ * published by the Free Software Foundation, version 2 of the ++ * License. ++ * ++ * File: ima_main.c ++ * implements the IMA hooks: ima_bprm_check, ima_file_mmap, ++ * and ima_path_check. ++ */ ++#include ++#include ++#include ++#include ++#include ++ ++#include "ima.h" ++ ++int ima_initialized; ++ ++char *ima_hash = "sha1"; ++static int __init hash_setup(char *str) ++{ ++ if (strncmp(str, "md5", 3) == 0) ++ ima_hash = "md5"; ++ return 1; ++} ++__setup("ima_hash=", hash_setup); ++ ++/** ++ * ima_file_free - called on __fput() ++ * @file: pointer to file structure being freed ++ * ++ * Flag files that changed, based on i_version; ++ * and decrement the iint readcount/writecount. ++ */ ++void ima_file_free(struct file *file) ++{ ++ struct inode *inode = file->f_dentry->d_inode; ++ struct ima_iint_cache *iint; ++ ++ if (!ima_initialized || !S_ISREG(inode->i_mode)) ++ return; ++ iint = ima_iint_find_get(inode); ++ if (!iint) ++ return; ++ ++ mutex_lock(&iint->mutex); ++ if (iint->opencount <= 0) { ++ printk(KERN_INFO ++ "%s: %s open/free imbalance (r:%ld w:%ld o:%ld f:%ld)\n", ++ __FUNCTION__, file->f_dentry->d_name.name, ++ iint->readcount, iint->writecount, ++ iint->opencount, atomic_long_read(&file->f_count)); ++ if (!(iint->flags & IMA_IINT_DUMP_STACK)) { ++ dump_stack(); ++ iint->flags |= IMA_IINT_DUMP_STACK; ++ } ++ } ++ iint->opencount--; ++ ++ if ((file->f_mode & (FMODE_READ | FMODE_WRITE)) == FMODE_READ) ++ iint->readcount--; ++ ++ if (file->f_mode & FMODE_WRITE) { ++ iint->writecount--; ++ if (iint->writecount == 0) { ++ if (iint->version != inode->i_version) ++ iint->flags &= ~IMA_MEASURED; ++ } ++ } ++ mutex_unlock(&iint->mutex); ++ kref_put(&iint->refcount, iint_free); ++} ++ ++/* ima_read_write_check - reflect possible reading/writing errors in the PCR. ++ * ++ * When opening a file for read, if the file is already open for write, ++ * the file could change, resulting in a file measurement error. ++ * ++ * Opening a file for write, if the file is already open for read, results ++ * in a time of measure, time of use (ToMToU) error. ++ * ++ * In either case invalidate the PCR. ++ */ ++enum iint_pcr_error { TOMTOU, OPEN_WRITERS }; ++static void ima_read_write_check(enum iint_pcr_error error, ++ struct ima_iint_cache *iint, ++ struct inode *inode, ++ const unsigned char *filename) ++{ ++ switch (error) { ++ case TOMTOU: ++ if (iint->readcount > 0) ++ ima_add_violation(inode, filename, "invalid_pcr", ++ "ToMToU"); ++ break; ++ case OPEN_WRITERS: ++ if (iint->writecount > 0) ++ ima_add_violation(inode, filename, "invalid_pcr", ++ "open_writers"); ++ break; ++ } ++} ++ ++static int get_path_measurement(struct ima_iint_cache *iint, struct file *file, ++ const unsigned char *filename) ++{ ++ int rc = 0; ++ ++ iint->opencount++; ++ iint->readcount++; ++ ++ rc = ima_collect_measurement(iint, file); ++ if (!rc) ++ ima_store_measurement(iint, file, filename); ++ return rc; ++} ++ ++static void ima_update_counts(struct ima_iint_cache *iint, int mask) ++{ ++ iint->opencount++; ++ if ((mask & MAY_WRITE) || (mask == 0)) ++ iint->writecount++; ++ else if (mask & (MAY_READ | MAY_EXEC)) ++ iint->readcount++; ++} ++ ++/** ++ * ima_path_check - based on policy, collect/store measurement. ++ * @path: contains a pointer to the path to be measured ++ * @mask: contains MAY_READ, MAY_WRITE or MAY_EXECUTE ++ * ++ * Measure the file being open for readonly, based on the ++ * ima_must_measure() policy decision. ++ * ++ * Keep read/write counters for all files, but only ++ * invalidate the PCR for measured files: ++ * - Opening a file for write when already open for read, ++ * results in a time of measure, time of use (ToMToU) error. ++ * - Opening a file for read when already open for write, ++ * could result in a file measurement error. ++ * ++ * Always return 0 and audit dentry_open failures. ++ * (Return code will be based upon measurement appraisal.) ++ */ ++int ima_path_check(struct path *path, int mask, int update_counts) ++{ ++ struct inode *inode = path->dentry->d_inode; ++ struct ima_iint_cache *iint; ++ struct file *file = NULL; ++ int rc; ++ ++ if (!ima_initialized || !S_ISREG(inode->i_mode)) ++ return 0; ++ iint = ima_iint_find_insert_get(inode); ++ if (!iint) ++ return 0; ++ ++ mutex_lock(&iint->mutex); ++ if (update_counts) ++ ima_update_counts(iint, mask); ++ ++ rc = ima_must_measure(iint, inode, MAY_READ, PATH_CHECK); ++ if (rc < 0) ++ goto out; ++ ++ if ((mask & MAY_WRITE) || (mask == 0)) ++ ima_read_write_check(TOMTOU, iint, inode, ++ path->dentry->d_name.name); ++ ++ if ((mask & (MAY_WRITE | MAY_READ | MAY_EXEC)) != MAY_READ) ++ goto out; ++ ++ ima_read_write_check(OPEN_WRITERS, iint, inode, ++ path->dentry->d_name.name); ++ if (!(iint->flags & IMA_MEASURED)) { ++ struct dentry *dentry = dget(path->dentry); ++ struct vfsmount *mnt = mntget(path->mnt); ++ ++ file = dentry_open(dentry, mnt, O_RDONLY | O_LARGEFILE, ++ current_cred()); ++ if (IS_ERR(file)) { ++ int audit_info = 0; ++ ++ integrity_audit_msg(AUDIT_INTEGRITY_PCR, inode, ++ dentry->d_name.name, ++ "add_measurement", ++ "dentry_open failed", ++ 1, audit_info); ++ file = NULL; ++ goto out; ++ } ++ rc = get_path_measurement(iint, file, dentry->d_name.name); ++ } ++out: ++ mutex_unlock(&iint->mutex); ++ if (file) ++ fput(file); ++ kref_put(&iint->refcount, iint_free); ++ return 0; ++} ++EXPORT_SYMBOL_GPL(ima_path_check); ++ ++static int process_measurement(struct file *file, const unsigned char *filename, ++ int mask, int function) ++{ ++ struct inode *inode = file->f_dentry->d_inode; ++ struct ima_iint_cache *iint; ++ int rc; ++ ++ if (!ima_initialized || !S_ISREG(inode->i_mode)) ++ return 0; ++ iint = ima_iint_find_insert_get(inode); ++ if (!iint) ++ return -ENOMEM; ++ ++ mutex_lock(&iint->mutex); ++ rc = ima_must_measure(iint, inode, mask, function); ++ if (rc != 0) ++ goto out; ++ ++ rc = ima_collect_measurement(iint, file); ++ if (!rc) ++ ima_store_measurement(iint, file, filename); ++out: ++ mutex_unlock(&iint->mutex); ++ kref_put(&iint->refcount, iint_free); ++ return rc; ++} ++ ++/* ++ * ima_counts_put - decrement file counts ++ * ++ * File counts are incremented in ima_path_check. On file open ++ * error, such as ETXTBSY, decrement the counts to prevent ++ * unnecessary imbalance messages. ++ */ ++void ima_counts_put(struct path *path, int mask) ++{ ++ struct inode *inode = path->dentry->d_inode; ++ struct ima_iint_cache *iint; ++ ++ /* The inode may already have been freed, freeing the iint ++ * with it. Verify the inode is not NULL before dereferencing ++ * it. ++ */ ++ if (!ima_initialized || !inode || !S_ISREG(inode->i_mode)) ++ return; ++ iint = ima_iint_find_insert_get(inode); ++ if (!iint) ++ return; ++ ++ mutex_lock(&iint->mutex); ++ iint->opencount--; ++ if ((mask & MAY_WRITE) || (mask == 0)) ++ iint->writecount--; ++ else if (mask & (MAY_READ | MAY_EXEC)) ++ iint->readcount--; ++ mutex_unlock(&iint->mutex); ++ ++ kref_put(&iint->refcount, iint_free); ++} ++ ++/* ++ * ima_counts_get - increment file counts ++ * ++ * - for IPC shm and shmat file. ++ * - for nfsd exported files. ++ * ++ * Increment the counts for these files to prevent unnecessary ++ * imbalance messages. ++ */ ++void ima_counts_get(struct file *file) ++{ ++ struct inode *inode = file->f_dentry->d_inode; ++ struct ima_iint_cache *iint; ++ ++ if (!ima_initialized || !S_ISREG(inode->i_mode)) ++ return; ++ iint = ima_iint_find_insert_get(inode); ++ if (!iint) ++ return; ++ mutex_lock(&iint->mutex); ++ iint->opencount++; ++ if ((file->f_mode & (FMODE_READ | FMODE_WRITE)) == FMODE_READ) ++ iint->readcount++; ++ ++ if (file->f_mode & FMODE_WRITE) ++ iint->writecount++; ++ mutex_unlock(&iint->mutex); ++ ++ kref_put(&iint->refcount, iint_free); ++} ++EXPORT_SYMBOL_GPL(ima_counts_get); ++ ++/** ++ * ima_file_mmap - based on policy, collect/store measurement. ++ * @file: pointer to the file to be measured (May be NULL) ++ * @prot: contains the protection that will be applied by the kernel. ++ * ++ * Measure files being mmapped executable based on the ima_must_measure() ++ * policy decision. ++ * ++ * Return 0 on success, an error code on failure. ++ * (Based on the results of appraise_measurement().) ++ */ ++int ima_file_mmap(struct file *file, unsigned long prot) ++{ ++ int rc; ++ ++ if (!file) ++ return 0; ++ if (prot & PROT_EXEC) ++ rc = process_measurement(file, file->f_dentry->d_name.name, ++ MAY_EXEC, FILE_MMAP); ++ return 0; ++} ++ ++/** ++ * ima_bprm_check - based on policy, collect/store measurement. ++ * @bprm: contains the linux_binprm structure ++ * ++ * The OS protects against an executable file, already open for write, ++ * from being executed in deny_write_access() and an executable file, ++ * already open for execute, from being modified in get_write_access(). ++ * So we can be certain that what we verify and measure here is actually ++ * what is being executed. ++ * ++ * Return 0 on success, an error code on failure. ++ * (Based on the results of appraise_measurement().) ++ */ ++int ima_bprm_check(struct linux_binprm *bprm) ++{ ++ int rc; ++ ++ rc = process_measurement(bprm->file, bprm->filename, ++ MAY_EXEC, BPRM_CHECK); ++ return 0; ++} ++ ++static int __init init_ima(void) ++{ ++ int error; ++ ++ ima_iintcache_init(); ++ error = ima_init(); ++ ima_initialized = 1; ++ return error; ++} ++ ++static void __exit cleanup_ima(void) ++{ ++ ima_cleanup(); ++} ++ ++late_initcall(init_ima); /* Start IMA after the TPM is available */ ++ ++MODULE_DESCRIPTION("Integrity Measurement Architecture"); ++MODULE_LICENSE("GPL"); +diff -Nur linux-2.6.31-vanilla/security/security.c linux-2.6.31/security/security.c +--- linux-2.6.31-vanilla/security/security.c 2009-09-10 00:13:59.000000000 +0200 ++++ linux-2.6.31/security/security.c 2009-09-16 13:55:49.000000000 +0200 +@@ -386,6 +386,7 @@ + return 0; + return security_ops->path_mkdir(path, dentry, mode); + } ++EXPORT_SYMBOL(security_path_mkdir); + + int security_path_rmdir(struct path *path, struct dentry *dentry) + { +@@ -393,6 +394,7 @@ + return 0; + return security_ops->path_rmdir(path, dentry); + } ++EXPORT_SYMBOL(security_path_rmdir); + + int security_path_unlink(struct path *path, struct dentry *dentry) + { +@@ -400,6 +402,7 @@ + return 0; + return security_ops->path_unlink(path, dentry); + } ++EXPORT_SYMBOL(security_path_unlink); + + int security_path_symlink(struct path *path, struct dentry *dentry, + const char *old_name) +@@ -408,6 +411,7 @@ + return 0; + return security_ops->path_symlink(path, dentry, old_name); + } ++EXPORT_SYMBOL(security_path_symlink); + + int security_path_link(struct dentry *old_dentry, struct path *new_dir, + struct dentry *new_dentry) +@@ -416,6 +420,7 @@ + return 0; + return security_ops->path_link(old_dentry, new_dir, new_dentry); + } ++EXPORT_SYMBOL(security_path_link); + + int security_path_rename(struct path *old_dir, struct dentry *old_dentry, + struct path *new_dir, struct dentry *new_dentry) +@@ -426,6 +431,7 @@ + return security_ops->path_rename(old_dir, old_dentry, new_dir, + new_dentry); + } ++EXPORT_SYMBOL(security_path_rename); + + int security_path_truncate(struct path *path, loff_t length, + unsigned int time_attrs) +@@ -434,6 +440,7 @@ + return 0; + return security_ops->path_truncate(path, length, time_attrs); + } ++EXPORT_SYMBOL(security_path_truncate); + #endif + + int security_inode_create(struct inode *dir, struct dentry *dentry, int mode) +@@ -505,6 +512,7 @@ + return 0; + return security_ops->inode_readlink(dentry); + } ++EXPORT_SYMBOL(security_inode_readlink); + + int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd) + { +@@ -519,6 +527,7 @@ + return 0; + return security_ops->inode_permission(inode, mask); + } ++EXPORT_SYMBOL(security_inode_permission); + + int security_inode_setattr(struct dentry *dentry, struct iattr *attr) + { +@@ -619,6 +628,7 @@ + { + return security_ops->file_permission(file, mask); + } ++EXPORT_SYMBOL(security_file_permission); + + int security_file_alloc(struct file *file) + { diff --git a/pkgs/core/kernel/patches/grsecurity-2.1.14-2.6.31.1-200910012153.patch b/pkgs/core/kernel/patches/grsecurity-2.1.14-2.6.31.1-200910012153.patch new file mode 100644 index 000000000..ae34c8401 --- /dev/null +++ b/pkgs/core/kernel/patches/grsecurity-2.1.14-2.6.31.1-200910012153.patch @@ -0,0 +1,47055 @@ +diff -urNp linux-2.6.31.1/arch/alpha/include/asm/atomic.h linux-2.6.31.1/arch/alpha/include/asm/atomic.h +--- linux-2.6.31.1/arch/alpha/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/alpha/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -18,9 +18,11 @@ + #define ATOMIC64_INIT(i) ( (atomic64_t) { (i) } ) + + #define atomic_read(v) ((v)->counter + 0) ++#define atomic_read_unchecked(v) ((v)->counter + 0) + #define atomic64_read(v) ((v)->counter + 0) + + #define atomic_set(v,i) ((v)->counter = (i)) ++#define atomic_set_unchecked(v,i) ((v)->counter = (i)) + #define atomic64_set(v,i) ((v)->counter = (i)) + + /* +@@ -44,6 +46,11 @@ static __inline__ void atomic_add(int i, + :"Ir" (i), "m" (v->counter)); + } + ++static __inline__ void atomic_add_unchecked(int i, atomic_unchecked_t * v) ++{ ++ atomic_add(i, (atomic_t *)v); ++} ++ + static __inline__ void atomic64_add(long i, atomic64_t * v) + { + unsigned long temp; +@@ -74,6 +81,11 @@ static __inline__ void atomic_sub(int i, + :"Ir" (i), "m" (v->counter)); + } + ++static __inline__ void atomic_sub_unchecked(int i, atomic_unchecked_t * v) ++{ ++ atomic_sub(i, (atomic_t *)v); ++} ++ + static __inline__ void atomic64_sub(long i, atomic64_t * v) + { + unsigned long temp; +@@ -246,6 +258,7 @@ static __inline__ int atomic64_add_unles + #define atomic64_dec_and_test(v) (atomic64_sub_return(1, (v)) == 0) + + #define atomic_inc(v) atomic_add(1,(v)) ++#define atomic_inc_unchecked(v) atomic_add_unchecked(1,(v)) + #define atomic64_inc(v) atomic64_add(1,(v)) + + #define atomic_dec(v) atomic_sub(1,(v)) +diff -urNp linux-2.6.31.1/arch/alpha/include/asm/elf.h linux-2.6.31.1/arch/alpha/include/asm/elf.h +--- linux-2.6.31.1/arch/alpha/include/asm/elf.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/alpha/include/asm/elf.h 2009-10-01 20:12:42.000000000 -0400 +@@ -91,6 +91,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_N + + #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x1000000) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (current->personality & ADDR_LIMIT_32BIT ? 0x10000 : 0x120000000UL) ++ ++#define PAX_DELTA_MMAP_LEN (current->personality & ADDR_LIMIT_32BIT ? 14 : 28) ++#define PAX_DELTA_STACK_LEN (current->personality & ADDR_LIMIT_32BIT ? 14 : 19) ++#endif ++ + /* $0 is set by ld.so to a pointer to a function which might be + registered using atexit. This provides a mean for the dynamic + linker to call DT_FINI functions for shared libraries that have +diff -urNp linux-2.6.31.1/arch/alpha/include/asm/pgtable.h linux-2.6.31.1/arch/alpha/include/asm/pgtable.h +--- linux-2.6.31.1/arch/alpha/include/asm/pgtable.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/alpha/include/asm/pgtable.h 2009-10-01 20:12:42.000000000 -0400 +@@ -101,6 +101,17 @@ struct vm_area_struct; + #define PAGE_SHARED __pgprot(_PAGE_VALID | __ACCESS_BITS) + #define PAGE_COPY __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW) + #define PAGE_READONLY __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++# define PAGE_SHARED_NOEXEC __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOE) ++# define PAGE_COPY_NOEXEC __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW | _PAGE_FOE) ++# define PAGE_READONLY_NOEXEC __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW | _PAGE_FOE) ++#else ++# define PAGE_SHARED_NOEXEC PAGE_SHARED ++# define PAGE_COPY_NOEXEC PAGE_COPY ++# define PAGE_READONLY_NOEXEC PAGE_READONLY ++#endif ++ + #define PAGE_KERNEL __pgprot(_PAGE_VALID | _PAGE_ASM | _PAGE_KRE | _PAGE_KWE) + + #define _PAGE_NORMAL(x) __pgprot(_PAGE_VALID | __ACCESS_BITS | (x)) +diff -urNp linux-2.6.31.1/arch/alpha/kernel/module.c linux-2.6.31.1/arch/alpha/kernel/module.c +--- linux-2.6.31.1/arch/alpha/kernel/module.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/alpha/kernel/module.c 2009-10-01 20:12:42.000000000 -0400 +@@ -182,7 +182,7 @@ apply_relocate_add(Elf64_Shdr *sechdrs, + + /* The small sections were sorted to the end of the segment. + The following should definitely cover them. */ +- gp = (u64)me->module_core + me->core_size - 0x8000; ++ gp = (u64)me->module_core_rw + me->core_size_rw - 0x8000; + got = sechdrs[me->arch.gotsecindex].sh_addr; + + for (i = 0; i < n; i++) { +diff -urNp linux-2.6.31.1/arch/alpha/kernel/osf_sys.c linux-2.6.31.1/arch/alpha/kernel/osf_sys.c +--- linux-2.6.31.1/arch/alpha/kernel/osf_sys.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/alpha/kernel/osf_sys.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1212,6 +1212,10 @@ arch_get_unmapped_area(struct file *filp + merely specific addresses, but regions of memory -- perhaps + this feature should be incorporated into all ports? */ + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(current->mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + addr = arch_get_unmapped_area_1 (PAGE_ALIGN(addr), len, limit); + if (addr != (unsigned long) -ENOMEM) +@@ -1219,8 +1223,8 @@ arch_get_unmapped_area(struct file *filp + } + + /* Next, try allocating at TASK_UNMAPPED_BASE. */ +- addr = arch_get_unmapped_area_1 (PAGE_ALIGN(TASK_UNMAPPED_BASE), +- len, limit); ++ addr = arch_get_unmapped_area_1 (PAGE_ALIGN(current->mm->mmap_base), len, limit); ++ + if (addr != (unsigned long) -ENOMEM) + return addr; + +diff -urNp linux-2.6.31.1/arch/alpha/mm/fault.c linux-2.6.31.1/arch/alpha/mm/fault.c +--- linux-2.6.31.1/arch/alpha/mm/fault.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/alpha/mm/fault.c 2009-10-01 20:12:42.000000000 -0400 +@@ -54,6 +54,124 @@ __load_new_mm_context(struct mm_struct * + __reload_thread(pcb); + } + ++#ifdef CONFIG_PAX_PAGEEXEC ++/* ++ * PaX: decide what to do with offenders (regs->pc = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when patched PLT trampoline was detected ++ * 3 when unpatched PLT trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ ++#ifdef CONFIG_PAX_EMUPLT ++ int err; ++ ++ do { /* PaX: patched PLT emulation #1 */ ++ unsigned int ldah, ldq, jmp; ++ ++ err = get_user(ldah, (unsigned int *)regs->pc); ++ err |= get_user(ldq, (unsigned int *)(regs->pc+4)); ++ err |= get_user(jmp, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((ldah & 0xFFFF0000U) == 0x277B0000U && ++ (ldq & 0xFFFF0000U) == 0xA77B0000U && ++ jmp == 0x6BFB0000U) ++ { ++ unsigned long r27, addr; ++ unsigned long addrh = (ldah | 0xFFFFFFFFFFFF0000UL) << 16; ++ unsigned long addrl = ldq | 0xFFFFFFFFFFFF0000UL; ++ ++ addr = regs->r27 + ((addrh ^ 0x80000000UL) + 0x80000000UL) + ((addrl ^ 0x8000UL) + 0x8000UL); ++ err = get_user(r27, (unsigned long *)addr); ++ if (err) ++ break; ++ ++ regs->r27 = r27; ++ regs->pc = r27; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: patched PLT emulation #2 */ ++ unsigned int ldah, lda, br; ++ ++ err = get_user(ldah, (unsigned int *)regs->pc); ++ err |= get_user(lda, (unsigned int *)(regs->pc+4)); ++ err |= get_user(br, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((ldah & 0xFFFF0000U) == 0x277B0000U && ++ (lda & 0xFFFF0000U) == 0xA77B0000U && ++ (br & 0xFFE00000U) == 0xC3E00000U) ++ { ++ unsigned long addr = br | 0xFFFFFFFFFFE00000UL; ++ unsigned long addrh = (ldah | 0xFFFFFFFFFFFF0000UL) << 16; ++ unsigned long addrl = lda | 0xFFFFFFFFFFFF0000UL; ++ ++ regs->r27 += ((addrh ^ 0x80000000UL) + 0x80000000UL) + ((addrl ^ 0x8000UL) + 0x8000UL); ++ regs->pc += 12 + (((addr ^ 0x00100000UL) + 0x00100000UL) << 2); ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: unpatched PLT emulation */ ++ unsigned int br; ++ ++ err = get_user(br, (unsigned int *)regs->pc); ++ ++ if (!err && (br & 0xFFE00000U) == 0xC3800000U) { ++ unsigned int br2, ldq, nop, jmp; ++ unsigned long addr = br | 0xFFFFFFFFFFE00000UL, resolver; ++ ++ addr = regs->pc + 4 + (((addr ^ 0x00100000UL) + 0x00100000UL) << 2); ++ err = get_user(br2, (unsigned int *)addr); ++ err |= get_user(ldq, (unsigned int *)(addr+4)); ++ err |= get_user(nop, (unsigned int *)(addr+8)); ++ err |= get_user(jmp, (unsigned int *)(addr+12)); ++ err |= get_user(resolver, (unsigned long *)(addr+16)); ++ ++ if (err) ++ break; ++ ++ if (br2 == 0xC3600000U && ++ ldq == 0xA77B000CU && ++ nop == 0x47FF041FU && ++ jmp == 0x6B7B0000U) ++ { ++ regs->r28 = regs->pc+4; ++ regs->r27 = addr+16; ++ regs->pc = resolver; ++ return 3; ++ } ++ } ++ } while (0); ++#endif ++ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif + + /* + * This routine handles page faults. It determines the address, +@@ -131,8 +249,29 @@ do_page_fault(unsigned long address, uns + good_area: + si_code = SEGV_ACCERR; + if (cause < 0) { +- if (!(vma->vm_flags & VM_EXEC)) ++ if (!(vma->vm_flags & VM_EXEC)) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!(mm->pax_flags & MF_PAX_PAGEEXEC) || address != regs->pc) ++ goto bad_area; ++ ++ up_read(&mm->mmap_sem); ++ switch (pax_handle_fetch_fault(regs)) { ++ ++#ifdef CONFIG_PAX_EMUPLT ++ case 2: ++ case 3: ++ return; ++#endif ++ ++ } ++ pax_report_fault(regs, (void *)regs->pc, (void *)rdusp()); ++ do_group_exit(SIGKILL); ++#else + goto bad_area; ++#endif ++ ++ } + } else if (!cause) { + /* Allow reads even for write-only mappings */ + if (!(vma->vm_flags & (VM_READ | VM_WRITE))) +diff -urNp linux-2.6.31.1/arch/arm/include/asm/atomic.h linux-2.6.31.1/arch/arm/include/asm/atomic.h +--- linux-2.6.31.1/arch/arm/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/arm/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -20,6 +20,7 @@ + #ifdef __KERNEL__ + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + + #if __LINUX_ARM_ARCH__ >= 6 + +@@ -44,6 +45,11 @@ static inline void atomic_set(atomic_t * + : "cc"); + } + ++static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i) ++{ ++ atomic_set((atomic_t *)v, i); ++} ++ + static inline void atomic_add(int i, atomic_t *v) + { + unsigned long tmp; +@@ -60,6 +66,11 @@ static inline void atomic_add(int i, ato + : "cc"); + } + ++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_add(i, (atomic_t *)v); ++} ++ + static inline int atomic_add_return(int i, atomic_t *v) + { + unsigned long tmp; +@@ -98,6 +109,11 @@ static inline void atomic_sub(int i, ato + : "cc"); + } + ++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_sub(i, (atomic_t *)v); ++} ++ + static inline int atomic_sub_return(int i, atomic_t *v) + { + unsigned long tmp; +@@ -164,6 +180,7 @@ static inline void atomic_clear_mask(uns + #endif + + #define atomic_set(v,i) (((v)->counter) = (i)) ++#define atomic_set_unchecked(v,i) (((v)->counter) = (i)) + + static inline int atomic_add_return(int i, atomic_t *v) + { +@@ -232,6 +249,7 @@ static inline int atomic_add_unless(atom + #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) + + #define atomic_inc(v) atomic_add(1, v) ++#define atomic_inc_unchecked(v) atomic_add_unchecked(1, v) + #define atomic_dec(v) atomic_sub(1, v) + + #define atomic_inc_and_test(v) (atomic_add_return(1, v) == 0) +diff -urNp linux-2.6.31.1/arch/arm/include/asm/elf.h linux-2.6.31.1/arch/arm/include/asm/elf.h +--- linux-2.6.31.1/arch/arm/include/asm/elf.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/arm/include/asm/elf.h 2009-10-01 20:12:42.000000000 -0400 +@@ -103,7 +103,14 @@ extern int arm_elf_read_implies_exec(con + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. */ + +-#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) ++#define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) ++ ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE 0x00008000UL ++ ++#define PAX_DELTA_MMAP_LEN ((current->personality == PER_LINUX_32BIT) ? 16 : 10) ++#define PAX_DELTA_STACK_LEN ((current->personality == PER_LINUX_32BIT) ? 16 : 10) ++#endif + + /* When the program starts, a1 contains a pointer to a function to be + registered with atexit, as per the SVR4 ABI. A value of 0 means we +diff -urNp linux-2.6.31.1/arch/arm/include/asm/kmap_types.h linux-2.6.31.1/arch/arm/include/asm/kmap_types.h +--- linux-2.6.31.1/arch/arm/include/asm/kmap_types.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/arm/include/asm/kmap_types.h 2009-10-01 20:12:42.000000000 -0400 +@@ -19,6 +19,7 @@ enum km_type { + KM_SOFTIRQ0, + KM_SOFTIRQ1, + KM_L2_CACHE, ++ KM_CLEARPAGE, + KM_TYPE_NR + }; + +diff -urNp linux-2.6.31.1/arch/arm/include/asm/uaccess.h linux-2.6.31.1/arch/arm/include/asm/uaccess.h +--- linux-2.6.31.1/arch/arm/include/asm/uaccess.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/arm/include/asm/uaccess.h 2009-10-01 20:12:42.000000000 -0400 +@@ -400,6 +400,9 @@ extern unsigned long __must_check __strn + + static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + if (access_ok(VERIFY_READ, from, n)) + n = __copy_from_user(to, from, n); + else /* security hole - plug it */ +@@ -409,6 +412,9 @@ static inline unsigned long __must_check + + static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + if (access_ok(VERIFY_WRITE, to, n)) + n = __copy_to_user(to, from, n); + return n; +diff -urNp linux-2.6.31.1/arch/arm/mach-ns9xxx/clock.c linux-2.6.31.1/arch/arm/mach-ns9xxx/clock.c +--- linux-2.6.31.1/arch/arm/mach-ns9xxx/clock.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/arm/mach-ns9xxx/clock.c 2009-10-01 20:12:42.000000000 -0400 +@@ -195,7 +195,7 @@ static int clk_debugfs_open(struct inode + return single_open(file, clk_debugfs_show, NULL); + } + +-static struct file_operations clk_debugfs_operations = { ++static const struct file_operations clk_debugfs_operations = { + .open = clk_debugfs_open, + .read = seq_read, + .llseek = seq_lseek, +diff -urNp linux-2.6.31.1/arch/arm/mm/mmap.c linux-2.6.31.1/arch/arm/mm/mmap.c +--- linux-2.6.31.1/arch/arm/mm/mmap.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/arm/mm/mmap.c 2009-10-01 20:12:42.000000000 -0400 +@@ -62,6 +62,10 @@ arch_get_unmapped_area(struct file *filp + if (len > TASK_SIZE) + return -ENOMEM; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + if (do_align) + addr = COLOUR_ALIGN(addr, pgoff); +@@ -74,10 +78,10 @@ arch_get_unmapped_area(struct file *filp + return addr; + } + if (len > mm->cached_hole_size) { +- start_addr = addr = mm->free_area_cache; ++ start_addr = addr = mm->free_area_cache; + } else { +- start_addr = addr = TASK_UNMAPPED_BASE; +- mm->cached_hole_size = 0; ++ start_addr = addr = mm->mmap_base; ++ mm->cached_hole_size = 0; + } + + full_search: +@@ -93,8 +97,8 @@ full_search: + * Start a new search - just in case we missed + * some holes. + */ +- if (start_addr != TASK_UNMAPPED_BASE) { +- start_addr = addr = TASK_UNMAPPED_BASE; ++ if (start_addr != mm->mmap_base) { ++ start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + goto full_search; + } +diff -urNp linux-2.6.31.1/arch/avr32/include/asm/atomic.h linux-2.6.31.1/arch/avr32/include/asm/atomic.h +--- linux-2.6.31.1/arch/avr32/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/avr32/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -20,7 +20,9 @@ + #define ATOMIC_INIT(i) { (i) } + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + #define atomic_set(v, i) (((v)->counter) = i) ++#define atomic_set_unchecked(v, i) (((v)->counter) = i) + + /* + * atomic_sub_return - subtract the atomic variable +@@ -48,6 +50,18 @@ static inline int atomic_sub_return(int + } + + /* ++ * atomic_sub_return_unchecked - subtract the atomic variable ++ * @i: integer value to subtract ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically subtracts @i from @v. Returns the resulting value. ++ */ ++static inline int atomic_sub_return_unchecked(int i, atomic_unchecked_t *v) ++{ ++ return atomic_sub_return(i, (atomic_t *)v); ++} ++ ++/* + * atomic_add_return - add integer to atomic variable + * @i: integer value to add + * @v: pointer of type atomic_t +@@ -76,6 +90,18 @@ static inline int atomic_add_return(int + } + + /* ++ * atomic_add_return_unchecked - add integer to atomic variable ++ * @i: integer value to add ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically adds @i to @v. Returns the resulting value. ++ */ ++static inline int atomic_add_return_unchecked(int i, atomic_unchecked_t *v) ++{ ++ return atomic_add_return(i, (atomic_t *)v); ++} ++ ++/* + * atomic_sub_unless - sub unless the number is a given value + * @v: pointer of type atomic_t + * @a: the amount to add to v... +@@ -176,9 +202,12 @@ static inline int atomic_sub_if_positive + #define atomic_cmpxchg(v, o, n) (cmpxchg(&((v)->counter), (o), (n))) + + #define atomic_sub(i, v) (void)atomic_sub_return(i, v) ++#define atomic_sub_unchecked(i, v) (void)atomic_sub_return_unchecked(i, v) + #define atomic_add(i, v) (void)atomic_add_return(i, v) ++#define atomic_add_unchecked(i, v) (void)atomic_add_return_unchecked(i, v) + #define atomic_dec(v) atomic_sub(1, (v)) + #define atomic_inc(v) atomic_add(1, (v)) ++#define atomic_inc_unchecked(v) (void)atomic_add_return_unchecked(1, (v)) + + #define atomic_dec_return(v) atomic_sub_return(1, v) + #define atomic_inc_return(v) atomic_add_return(1, v) +diff -urNp linux-2.6.31.1/arch/avr32/include/asm/elf.h linux-2.6.31.1/arch/avr32/include/asm/elf.h +--- linux-2.6.31.1/arch/avr32/include/asm/elf.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/avr32/include/asm/elf.h 2009-10-01 20:12:42.000000000 -0400 +@@ -85,8 +85,14 @@ typedef struct user_fpu_struct elf_fpreg + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. */ + +-#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) ++#define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE 0x00001000UL ++ ++#define PAX_DELTA_MMAP_LEN 15 ++#define PAX_DELTA_STACK_LEN 15 ++#endif + + /* This yields a mask that user programs can use to figure out what + instruction set this CPU supports. This could be done in user space, +diff -urNp linux-2.6.31.1/arch/avr32/include/asm/kmap_types.h linux-2.6.31.1/arch/avr32/include/asm/kmap_types.h +--- linux-2.6.31.1/arch/avr32/include/asm/kmap_types.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/avr32/include/asm/kmap_types.h 2009-10-01 20:12:42.000000000 -0400 +@@ -22,7 +22,8 @@ D(10) KM_IRQ0, + D(11) KM_IRQ1, + D(12) KM_SOFTIRQ0, + D(13) KM_SOFTIRQ1, +-D(14) KM_TYPE_NR ++D(14) KM_CLEARPAGE, ++D(15) KM_TYPE_NR + }; + + #undef D +diff -urNp linux-2.6.31.1/arch/avr32/mm/fault.c linux-2.6.31.1/arch/avr32/mm/fault.c +--- linux-2.6.31.1/arch/avr32/mm/fault.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/avr32/mm/fault.c 2009-10-01 20:12:42.000000000 -0400 +@@ -41,6 +41,23 @@ static inline int notify_page_fault(stru + + int exception_trace = 1; + ++#ifdef CONFIG_PAX_PAGEEXEC ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 20; i++) { ++ unsigned char c; ++ if (get_user(c, (unsigned char *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%02x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + /* + * This routine handles page faults. It determines the address and the + * problem, and then passes it off to one of the appropriate routines. +@@ -157,6 +174,16 @@ bad_area: + up_read(&mm->mmap_sem); + + if (user_mode(regs)) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (mm->pax_flags & MF_PAX_PAGEEXEC) { ++ if (ecr == ECR_PROTECTION_X || ecr == ECR_TLB_MISS_X) { ++ pax_report_fault(regs, (void *)regs->pc, (void *)regs->sp); ++ do_group_exit(SIGKILL); ++ } ++ } ++#endif ++ + if (exception_trace && printk_ratelimit()) + printk("%s%s[%d]: segfault at %08lx pc %08lx " + "sp %08lx ecr %lu\n", +diff -urNp linux-2.6.31.1/arch/blackfin/include/asm/atomic.h linux-2.6.31.1/arch/blackfin/include/asm/atomic.h +--- linux-2.6.31.1/arch/blackfin/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/blackfin/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -15,8 +15,10 @@ + + #define ATOMIC_INIT(i) { (i) } + #define atomic_set(v, i) (((v)->counter) = i) ++#define atomic_set_unchecked(v, i) (((v)->counter) = i) + + #define atomic_read(v) __raw_uncached_fetch_asm(&(v)->counter) ++#define atomic_read_unchecked(v) __raw_uncached_fetch_asm(&(v)->counter) + + asmlinkage int __raw_uncached_fetch_asm(const volatile int *ptr); + +@@ -35,11 +37,21 @@ static inline void atomic_add(int i, ato + __raw_atomic_update_asm(&v->counter, i); + } + ++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_add(i, (atomic_t *)v); ++} ++ + static inline void atomic_sub(int i, atomic_t *v) + { + __raw_atomic_update_asm(&v->counter, -i); + } + ++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_sub(i, (atomic_t *)v); ++} ++ + static inline int atomic_add_return(int i, atomic_t *v) + { + return __raw_atomic_update_asm(&v->counter, i); +@@ -55,6 +67,11 @@ static inline void atomic_inc(volatile a + __raw_atomic_update_asm(&v->counter, 1); + } + ++static inline void atomic_inc_unchecked(volatile atomic_unchecked_t *v) ++{ ++ atomic_inc((atomic_t *)v); ++} ++ + static inline void atomic_dec(volatile atomic_t *v) + { + __raw_atomic_update_asm(&v->counter, -1); +diff -urNp linux-2.6.31.1/arch/blackfin/mach-bf561/coreb.c linux-2.6.31.1/arch/blackfin/mach-bf561/coreb.c +--- linux-2.6.31.1/arch/blackfin/mach-bf561/coreb.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/blackfin/mach-bf561/coreb.c 2009-10-01 20:12:42.000000000 -0400 +@@ -48,7 +48,7 @@ coreb_ioctl(struct inode *inode, struct + return ret; + } + +-static struct file_operations coreb_fops = { ++static const struct file_operations coreb_fops = { + .owner = THIS_MODULE, + .ioctl = coreb_ioctl, + }; +diff -urNp linux-2.6.31.1/arch/cris/arch-v10/drivers/sync_serial.c linux-2.6.31.1/arch/cris/arch-v10/drivers/sync_serial.c +--- linux-2.6.31.1/arch/cris/arch-v10/drivers/sync_serial.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/cris/arch-v10/drivers/sync_serial.c 2009-10-01 20:12:42.000000000 -0400 +@@ -244,7 +244,7 @@ static unsigned sync_serial_prescale_sha + + #define NUMBER_OF_PORTS 2 + +-static struct file_operations sync_serial_fops = { ++static const struct file_operations sync_serial_fops = { + .owner = THIS_MODULE, + .write = sync_serial_write, + .read = sync_serial_read, +diff -urNp linux-2.6.31.1/arch/cris/arch-v32/drivers/mach-fs/gpio.c linux-2.6.31.1/arch/cris/arch-v32/drivers/mach-fs/gpio.c +--- linux-2.6.31.1/arch/cris/arch-v32/drivers/mach-fs/gpio.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/cris/arch-v32/drivers/mach-fs/gpio.c 2009-10-01 20:12:42.000000000 -0400 +@@ -855,7 +855,7 @@ gpio_leds_ioctl(unsigned int cmd, unsign + return 0; + } + +-struct file_operations gpio_fops = { ++struct struct file_operations gpio_fops = { + .owner = THIS_MODULE, + .poll = gpio_poll, + .ioctl = gpio_ioctl, +diff -urNp linux-2.6.31.1/arch/cris/include/asm/atomic.h linux-2.6.31.1/arch/cris/include/asm/atomic.h +--- linux-2.6.31.1/arch/cris/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/cris/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -16,7 +16,9 @@ + #define ATOMIC_INIT(i) { (i) } + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + #define atomic_set(v,i) (((v)->counter) = (i)) ++#define atomic_set_unchecked(v,i) (((v)->counter) = (i)) + + /* These should be written in asm but we do it in C for now. */ + +@@ -28,6 +30,11 @@ static inline void atomic_add(int i, vol + cris_atomic_restore(v, flags); + } + ++static inline void atomic_add_unchecked(int i, volatile atomic_unchecked_t *v) ++{ ++ atomic_add(i, (volatile atomic_t *)v); ++} ++ + static inline void atomic_sub(int i, volatile atomic_t *v) + { + unsigned long flags; +@@ -36,6 +43,11 @@ static inline void atomic_sub(int i, vol + cris_atomic_restore(v, flags); + } + ++static inline void atomic_sub_unchecked(int i, volatile atomic_unchecked_t *v) ++{ ++ atomic_sub(i, (volatile atomic_t *)v); ++} ++ + static inline int atomic_add_return(int i, volatile atomic_t *v) + { + unsigned long flags; +@@ -76,6 +88,11 @@ static inline void atomic_inc(volatile a + cris_atomic_restore(v, flags); + } + ++static inline void atomic_inc_unchecked(volatile atomic_unchecked_t *v) ++{ ++ atomic_inc((volatile atomic_t *)v); ++} ++ + static inline void atomic_dec(volatile atomic_t *v) + { + unsigned long flags; +diff -urNp linux-2.6.31.1/arch/frv/include/asm/atomic.h linux-2.6.31.1/arch/frv/include/asm/atomic.h +--- linux-2.6.31.1/arch/frv/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/frv/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -37,7 +37,9 @@ + + #define ATOMIC_INIT(i) { (i) } + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + #define atomic_set(v, i) (((v)->counter) = (i)) ++#define atomic_set_unchecked(v, i) (((v)->counter) = (i)) + + #ifndef CONFIG_FRV_OUTOFLINE_ATOMIC_OPS + static inline int atomic_add_return(int i, atomic_t *v) +@@ -99,16 +101,31 @@ static inline void atomic_add(int i, ato + atomic_add_return(i, v); + } + ++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_add_return(i, (atomic_t *)v); ++} ++ + static inline void atomic_sub(int i, atomic_t *v) + { + atomic_sub_return(i, v); + } + ++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_sub_return(i, (atomic_t *)v); ++} ++ + static inline void atomic_inc(atomic_t *v) + { + atomic_add_return(1, v); + } + ++static inline void atomic_inc_unchecked(atomic_unchecked_t *v) ++{ ++ atomic_add_return(1, (atomic_t *)v); ++} ++ + static inline void atomic_dec(atomic_t *v) + { + atomic_sub_return(1, v); +diff -urNp linux-2.6.31.1/arch/frv/include/asm/kmap_types.h linux-2.6.31.1/arch/frv/include/asm/kmap_types.h +--- linux-2.6.31.1/arch/frv/include/asm/kmap_types.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/frv/include/asm/kmap_types.h 2009-10-01 20:12:42.000000000 -0400 +@@ -23,6 +23,7 @@ enum km_type { + KM_IRQ1, + KM_SOFTIRQ0, + KM_SOFTIRQ1, ++ KM_CLEARPAGE, + KM_TYPE_NR + }; + +diff -urNp linux-2.6.31.1/arch/h8300/include/asm/atomic.h linux-2.6.31.1/arch/h8300/include/asm/atomic.h +--- linux-2.6.31.1/arch/h8300/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/h8300/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -11,7 +11,9 @@ + #define ATOMIC_INIT(i) { (i) } + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + #define atomic_set(v, i) (((v)->counter) = i) ++#define atomic_set_unchecked(v, i) (((v)->counter) = i) + + #include + #include +@@ -25,7 +27,13 @@ static __inline__ int atomic_add_return( + return ret; + } + ++static __inline__ int atomic_add_return_unchecked(int i, atomic_unchecked_t *v) ++{ ++ return atomic_add_return(i, (atomic_t *)v); ++} ++ + #define atomic_add(i, v) atomic_add_return(i, v) ++#define atomic_add_unchecked(i, v) atomic_add_return_unchecked(i, v) + #define atomic_add_negative(a, v) (atomic_add_return((a), (v)) < 0) + + static __inline__ int atomic_sub_return(int i, atomic_t *v) +@@ -37,7 +45,13 @@ static __inline__ int atomic_sub_return( + return ret; + } + ++static __inline__ int atomic_sub_return_unchecked(int i, atomic_unchecked_t *v) ++{ ++ return atomic_sub_return(i, (atomic_t *)v); ++} ++ + #define atomic_sub(i, v) atomic_sub_return(i, v) ++#define atomic_sub_unchecked(i, v) atomic_sub_return_unchecked(i, v) + #define atomic_sub_and_test(i,v) (atomic_sub_return(i, v) == 0) + + static __inline__ int atomic_inc_return(atomic_t *v) +@@ -50,7 +64,13 @@ static __inline__ int atomic_inc_return( + return ret; + } + ++static __inline__ int atomic_inc_return_unchecked(atomic_unchecked_t *v) ++{ ++ return atomic_inc_return((atomic_t *)v); ++} ++ + #define atomic_inc(v) atomic_inc_return(v) ++#define atomic_inc_unchecked(v) atomic_inc_return_unchecked(v) + + /* + * atomic_inc_and_test - increment and test +diff -urNp linux-2.6.31.1/arch/ia64/ia32/binfmt_elf32.c linux-2.6.31.1/arch/ia64/ia32/binfmt_elf32.c +--- linux-2.6.31.1/arch/ia64/ia32/binfmt_elf32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/ia32/binfmt_elf32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -45,6 +45,13 @@ randomize_stack_top(unsigned long stack_ + + #define elf_read_implies_exec(ex, have_pt_gnu_stack) (!(have_pt_gnu_stack)) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (current->personality == PER_LINUX32 ? 0x08048000UL : 0x4000000000000000UL) ++ ++#define PAX_DELTA_MMAP_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) ++#define PAX_DELTA_STACK_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) ++#endif ++ + /* Ugly but avoids duplication */ + #include "../../../fs/binfmt_elf.c" + +@@ -69,11 +76,11 @@ ia32_install_gate_page (struct vm_area_s + } + + +-static struct vm_operations_struct ia32_shared_page_vm_ops = { ++static const struct vm_operations_struct ia32_shared_page_vm_ops = { + .fault = ia32_install_shared_page + }; + +-static struct vm_operations_struct ia32_gate_page_vm_ops = { ++static const struct vm_operations_struct ia32_gate_page_vm_ops = { + .fault = ia32_install_gate_page + }; + +diff -urNp linux-2.6.31.1/arch/ia64/ia32/ia32priv.h linux-2.6.31.1/arch/ia64/ia32/ia32priv.h +--- linux-2.6.31.1/arch/ia64/ia32/ia32priv.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/ia32/ia32priv.h 2009-10-01 20:12:42.000000000 -0400 +@@ -296,7 +296,14 @@ typedef struct compat_siginfo { + #define ELF_DATA ELFDATA2LSB + #define ELF_ARCH EM_386 + +-#define IA32_STACK_TOP IA32_PAGE_OFFSET ++#ifdef CONFIG_PAX_RANDUSTACK ++#define __IA32_DELTA_STACK (current->mm->delta_stack) ++#else ++#define __IA32_DELTA_STACK 0UL ++#endif ++ ++#define IA32_STACK_TOP (IA32_PAGE_OFFSET - __IA32_DELTA_STACK) ++ + #define IA32_GATE_OFFSET IA32_PAGE_OFFSET + #define IA32_GATE_END IA32_PAGE_OFFSET + PAGE_SIZE + +diff -urNp linux-2.6.31.1/arch/ia64/include/asm/atomic.h linux-2.6.31.1/arch/ia64/include/asm/atomic.h +--- linux-2.6.31.1/arch/ia64/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -22,9 +22,11 @@ + #define ATOMIC64_INIT(i) ((atomic64_t) { (i) }) + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + #define atomic64_read(v) ((v)->counter) + + #define atomic_set(v,i) (((v)->counter) = (i)) ++#define atomic_set_unchecked(v,i) (((v)->counter) = (i)) + #define atomic64_set(v,i) (((v)->counter) = (i)) + + static __inline__ int +@@ -201,8 +203,11 @@ atomic64_add_negative (__s64 i, atomic64 + #define atomic64_inc_and_test(v) (atomic64_add_return(1, (v)) == 0) + + #define atomic_add(i,v) atomic_add_return((i), (v)) ++#define atomic_add_unchecked(i,v) atomic_add((i), (atomic_t *)(v)) + #define atomic_sub(i,v) atomic_sub_return((i), (v)) ++#define atomic_sub_unchecked(i,v) atomic_sub((i), (atomic_t *)(v)) + #define atomic_inc(v) atomic_add(1, (v)) ++#define atomic_inc_unchecked(v) atomic_inc((atomic_t *)(v)) + #define atomic_dec(v) atomic_sub(1, (v)) + + #define atomic64_add(i,v) atomic64_add_return((i), (v)) +diff -urNp linux-2.6.31.1/arch/ia64/include/asm/elf.h linux-2.6.31.1/arch/ia64/include/asm/elf.h +--- linux-2.6.31.1/arch/ia64/include/asm/elf.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/include/asm/elf.h 2009-10-01 20:12:42.000000000 -0400 +@@ -43,6 +43,13 @@ + */ + #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x800000000UL) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (current->personality == PER_LINUX32 ? 0x08048000UL : 0x4000000000000000UL) ++ ++#define PAX_DELTA_MMAP_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) ++#define PAX_DELTA_STACK_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) ++#endif ++ + #define PT_IA_64_UNWIND 0x70000001 + + /* IA-64 relocations: */ +diff -urNp linux-2.6.31.1/arch/ia64/include/asm/pgtable.h linux-2.6.31.1/arch/ia64/include/asm/pgtable.h +--- linux-2.6.31.1/arch/ia64/include/asm/pgtable.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/include/asm/pgtable.h 2009-10-01 20:12:42.000000000 -0400 +@@ -143,6 +143,17 @@ + #define PAGE_READONLY __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) + #define PAGE_COPY __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) + #define PAGE_COPY_EXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_RX) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++# define PAGE_SHARED_NOEXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_RW) ++# define PAGE_READONLY_NOEXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) ++# define PAGE_COPY_NOEXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) ++#else ++# define PAGE_SHARED_NOEXEC PAGE_SHARED ++# define PAGE_READONLY_NOEXEC PAGE_READONLY ++# define PAGE_COPY_NOEXEC PAGE_COPY ++#endif ++ + #define PAGE_GATE __pgprot(__ACCESS_BITS | _PAGE_PL_0 | _PAGE_AR_X_RX) + #define PAGE_KERNEL __pgprot(__DIRTY_BITS | _PAGE_PL_0 | _PAGE_AR_RWX) + #define PAGE_KERNELRX __pgprot(__ACCESS_BITS | _PAGE_PL_0 | _PAGE_AR_RX) +diff -urNp linux-2.6.31.1/arch/ia64/include/asm/uaccess.h linux-2.6.31.1/arch/ia64/include/asm/uaccess.h +--- linux-2.6.31.1/arch/ia64/include/asm/uaccess.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/include/asm/uaccess.h 2009-10-01 20:12:42.000000000 -0400 +@@ -257,7 +257,7 @@ __copy_from_user (void *to, const void _ + const void *__cu_from = (from); \ + long __cu_len = (n); \ + \ +- if (__access_ok(__cu_to, __cu_len, get_fs())) \ ++ if (__cu_len > 0 && __cu_len <= INT_MAX && __access_ok(__cu_to, __cu_len, get_fs())) \ + __cu_len = __copy_user(__cu_to, (__force void __user *) __cu_from, __cu_len); \ + __cu_len; \ + }) +@@ -269,7 +269,7 @@ __copy_from_user (void *to, const void _ + long __cu_len = (n); \ + \ + __chk_user_ptr(__cu_from); \ +- if (__access_ok(__cu_from, __cu_len, get_fs())) \ ++ if (__cu_len > 0 && __cu_len <= INT_MAX && __access_ok(__cu_from, __cu_len, get_fs())) \ + __cu_len = __copy_user((__force void __user *) __cu_to, __cu_from, __cu_len); \ + __cu_len; \ + }) +diff -urNp linux-2.6.31.1/arch/ia64/kernel/module.c linux-2.6.31.1/arch/ia64/kernel/module.c +--- linux-2.6.31.1/arch/ia64/kernel/module.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/kernel/module.c 2009-10-01 20:12:42.000000000 -0400 +@@ -315,8 +315,7 @@ module_alloc (unsigned long size) + void + module_free (struct module *mod, void *module_region) + { +- if (mod && mod->arch.init_unw_table && +- module_region == mod->module_init) { ++ if (mod && mod->arch.init_unw_table && module_region == mod->module_init_rx) { + unw_remove_unwind_table(mod->arch.init_unw_table); + mod->arch.init_unw_table = NULL; + } +@@ -502,15 +501,39 @@ module_frob_arch_sections (Elf_Ehdr *ehd + } + + static inline int ++in_init_rx (const struct module *mod, uint64_t addr) ++{ ++ return addr - (uint64_t) mod->module_init_rx < mod->init_size_rx; ++} ++ ++static inline int ++in_init_rw (const struct module *mod, uint64_t addr) ++{ ++ return addr - (uint64_t) mod->module_init_rw < mod->init_size_rw; ++} ++ ++static inline int + in_init (const struct module *mod, uint64_t addr) + { +- return addr - (uint64_t) mod->module_init < mod->init_size; ++ return in_init_rx(mod, addr) || in_init_rw(mod, addr); ++} ++ ++static inline int ++in_core_rx (const struct module *mod, uint64_t addr) ++{ ++ return addr - (uint64_t) mod->module_core_rx < mod->core_size_rx; ++} ++ ++static inline int ++in_core_rw (const struct module *mod, uint64_t addr) ++{ ++ return addr - (uint64_t) mod->module_core_rw < mod->core_size_rw; + } + + static inline int + in_core (const struct module *mod, uint64_t addr) + { +- return addr - (uint64_t) mod->module_core < mod->core_size; ++ return in_core_rx(mod, addr) || in_core_rw(mod, addr); + } + + static inline int +@@ -693,7 +716,14 @@ do_reloc (struct module *mod, uint8_t r_ + break; + + case RV_BDREL: +- val -= (uint64_t) (in_init(mod, val) ? mod->module_init : mod->module_core); ++ if (in_init_rx(mod, val)) ++ val -= (uint64_t) mod->module_init_rx; ++ else if (in_init_rw(mod, val)) ++ val -= (uint64_t) mod->module_init_rw; ++ else if (in_core_rx(mod, val)) ++ val -= (uint64_t) mod->module_core_rx; ++ else if (in_core_rw(mod, val)) ++ val -= (uint64_t) mod->module_core_rw; + break; + + case RV_LTV: +@@ -828,15 +858,15 @@ apply_relocate_add (Elf64_Shdr *sechdrs, + * addresses have been selected... + */ + uint64_t gp; +- if (mod->core_size > MAX_LTOFF) ++ if (mod->core_size_rx + mod->core_size_rw > MAX_LTOFF) + /* + * This takes advantage of fact that SHF_ARCH_SMALL gets allocated + * at the end of the module. + */ +- gp = mod->core_size - MAX_LTOFF / 2; ++ gp = mod->core_size_rx + mod->core_size_rw - MAX_LTOFF / 2; + else +- gp = mod->core_size / 2; +- gp = (uint64_t) mod->module_core + ((gp + 7) & -8); ++ gp = (mod->core_size_rx + mod->core_size_rw) / 2; ++ gp = (uint64_t) mod->module_core_rx + ((gp + 7) & -8); + mod->arch.gp = gp; + DEBUGP("%s: placing gp at 0x%lx\n", __func__, gp); + } +diff -urNp linux-2.6.31.1/arch/ia64/kernel/sys_ia64.c linux-2.6.31.1/arch/ia64/kernel/sys_ia64.c +--- linux-2.6.31.1/arch/ia64/kernel/sys_ia64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/kernel/sys_ia64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -43,6 +43,13 @@ arch_get_unmapped_area (struct file *fil + if (REGION_NUMBER(addr) == RGN_HPAGE) + addr = 0; + #endif ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ addr = mm->free_area_cache; ++ else ++#endif ++ + if (!addr) + addr = mm->free_area_cache; + +@@ -61,9 +68,9 @@ arch_get_unmapped_area (struct file *fil + for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { + /* At this point: (!vma || addr < vma->vm_end). */ + if (TASK_SIZE - len < addr || RGN_MAP_LIMIT - len < REGION_OFFSET(addr)) { +- if (start_addr != TASK_UNMAPPED_BASE) { ++ if (start_addr != mm->mmap_base) { + /* Start a new search --- just in case we missed some holes. */ +- addr = TASK_UNMAPPED_BASE; ++ addr = mm->mmap_base; + goto full_search; + } + return -ENOMEM; +diff -urNp linux-2.6.31.1/arch/ia64/mm/fault.c linux-2.6.31.1/arch/ia64/mm/fault.c +--- linux-2.6.31.1/arch/ia64/mm/fault.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/mm/fault.c 2009-10-01 20:12:42.000000000 -0400 +@@ -72,6 +72,23 @@ mapped_kernel_page_is_present (unsigned + return pte_present(pte); + } + ++#ifdef CONFIG_PAX_PAGEEXEC ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 8; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + void __kprobes + ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *regs) + { +@@ -145,9 +162,23 @@ ia64_do_page_fault (unsigned long addres + mask = ( (((isr >> IA64_ISR_X_BIT) & 1UL) << VM_EXEC_BIT) + | (((isr >> IA64_ISR_W_BIT) & 1UL) << VM_WRITE_BIT)); + +- if ((vma->vm_flags & mask) != mask) ++ if ((vma->vm_flags & mask) != mask) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!(vma->vm_flags & VM_EXEC) && (mask & VM_EXEC)) { ++ if (!(mm->pax_flags & MF_PAX_PAGEEXEC) || address != regs->cr_iip) ++ goto bad_area; ++ ++ up_read(&mm->mmap_sem); ++ pax_report_fault(regs, (void *)regs->cr_iip, (void *)regs->r12); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ + goto bad_area; + ++ } ++ + survive: + /* + * If for any reason at all we couldn't handle the fault, make +diff -urNp linux-2.6.31.1/arch/ia64/mm/init.c linux-2.6.31.1/arch/ia64/mm/init.c +--- linux-2.6.31.1/arch/ia64/mm/init.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/ia64/mm/init.c 2009-10-01 20:12:42.000000000 -0400 +@@ -122,6 +122,19 @@ ia64_init_addr_space (void) + vma->vm_start = current->thread.rbs_bot & PAGE_MASK; + vma->vm_end = vma->vm_start + PAGE_SIZE; + vma->vm_flags = VM_DATA_DEFAULT_FLAGS|VM_GROWSUP|VM_ACCOUNT; ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (current->mm->pax_flags & MF_PAX_PAGEEXEC) { ++ vma->vm_flags &= ~VM_EXEC; ++ ++#ifdef CONFIG_PAX_MPROTECT ++ if (current->mm->pax_flags & MF_PAX_MPROTECT) ++ vma->vm_flags &= ~VM_MAYEXEC; ++#endif ++ ++ } ++#endif ++ + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + down_write(¤t->mm->mmap_sem); + if (insert_vm_struct(current->mm, vma)) { +diff -urNp linux-2.6.31.1/arch/m32r/include/asm/atomic.h linux-2.6.31.1/arch/m32r/include/asm/atomic.h +--- linux-2.6.31.1/arch/m32r/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/m32r/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -29,6 +29,14 @@ + #define atomic_read(v) ((v)->counter) + + /** ++ * atomic_read_unchecked - read atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically reads the value of @v. ++ */ ++#define atomic_read_unchecked(v) ((v)->counter) ++ ++/** + * atomic_set - set atomic variable + * @v: pointer of type atomic_t + * @i: required value +@@ -38,6 +46,15 @@ + #define atomic_set(v,i) (((v)->counter) = (i)) + + /** ++ * atomic_set_unchecked - set atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * @i: required value ++ * ++ * Atomically sets the value of @v to @i. ++ */ ++#define atomic_set_unchecked(v,i) (((v)->counter) = (i)) ++ ++/** + * atomic_add_return - add integer to atomic variable and return it + * @i: integer value to add + * @v: pointer of type atomic_t +@@ -308,6 +325,10 @@ static __inline__ void atomic_set_mask(u + local_irq_restore(flags); + } + ++#define atomic_inc_unchecked(v) atomic_inc((atomic_t *)(v)) ++#define atomic_add_unchecked(i,v) atomic_add((i),(atomic_t *)(v)) ++#define atomic_sub_unchecked(i,v) atomic_sub((i),(atomic_t *)(v)) ++ + /* Atomic operations are already serializing on m32r */ + #define smp_mb__before_atomic_dec() barrier() + #define smp_mb__after_atomic_dec() barrier() +diff -urNp linux-2.6.31.1/arch/m32r/lib/usercopy.c linux-2.6.31.1/arch/m32r/lib/usercopy.c +--- linux-2.6.31.1/arch/m32r/lib/usercopy.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/m32r/lib/usercopy.c 2009-10-01 20:12:42.000000000 -0400 +@@ -14,6 +14,9 @@ + unsigned long + __generic_copy_to_user(void __user *to, const void *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + prefetch(from); + if (access_ok(VERIFY_WRITE, to, n)) + __copy_user(to,from,n); +@@ -23,6 +26,9 @@ __generic_copy_to_user(void __user *to, + unsigned long + __generic_copy_from_user(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + prefetchw(to); + if (access_ok(VERIFY_READ, from, n)) + __copy_user_zeroing(to,from,n); +diff -urNp linux-2.6.31.1/arch/m68k/include/asm/atomic_mm.h linux-2.6.31.1/arch/m68k/include/asm/atomic_mm.h +--- linux-2.6.31.1/arch/m68k/include/asm/atomic_mm.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/m68k/include/asm/atomic_mm.h 2009-10-01 20:12:42.000000000 -0400 +@@ -16,23 +16,40 @@ + #define ATOMIC_INIT(i) { (i) } + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + #define atomic_set(v, i) (((v)->counter) = i) ++#define atomic_set_unchecked(v, i) (((v)->counter) = i) + + static inline void atomic_add(int i, atomic_t *v) + { + __asm__ __volatile__("addl %1,%0" : "+m" (*v) : "id" (i)); + } + ++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_add(i, (atomic_t *)v); ++} ++ + static inline void atomic_sub(int i, atomic_t *v) + { + __asm__ __volatile__("subl %1,%0" : "+m" (*v) : "id" (i)); + } + ++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_sub(i, (atomic_t *)v); ++} ++ + static inline void atomic_inc(atomic_t *v) + { + __asm__ __volatile__("addql #1,%0" : "+m" (*v)); + } + ++static inline void atomic_inc_unchecked(atomic_unchecked_t *v) ++{ ++ atomic_inc((atomic_t *)v); ++} ++ + static inline void atomic_dec(atomic_t *v) + { + __asm__ __volatile__("subql #1,%0" : "+m" (*v)); +diff -urNp linux-2.6.31.1/arch/m68k/include/asm/atomic_no.h linux-2.6.31.1/arch/m68k/include/asm/atomic_no.h +--- linux-2.6.31.1/arch/m68k/include/asm/atomic_no.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/m68k/include/asm/atomic_no.h 2009-10-01 20:12:42.000000000 -0400 +@@ -16,7 +16,9 @@ + #define ATOMIC_INIT(i) { (i) } + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + #define atomic_set(v, i) (((v)->counter) = i) ++#define atomic_set_unchecked(v, i) (((v)->counter) = i) + + static __inline__ void atomic_add(int i, atomic_t *v) + { +@@ -27,6 +29,11 @@ static __inline__ void atomic_add(int i, + #endif + } + ++static __inline__ void atomic_add_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_add(i, (atomic_t *)v); ++} ++ + static __inline__ void atomic_sub(int i, atomic_t *v) + { + #ifdef CONFIG_COLDFIRE +@@ -36,6 +43,11 @@ static __inline__ void atomic_sub(int i, + #endif + } + ++static __inline__ void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_sub(i, (atomic_t *)v); ++} ++ + static __inline__ int atomic_sub_and_test(int i, atomic_t * v) + { + char c; +@@ -56,6 +68,11 @@ static __inline__ void atomic_inc(volati + __asm__ __volatile__("addql #1,%0" : "+m" (*v)); + } + ++static __inline__ void atomic_inc_unchecked(volatile atomic_unchecked_t *v) ++{ ++ atomic_inc((volatile atomic_t *)v); ++} ++ + /* + * atomic_inc_and_test - increment and test + * @v: pointer of type atomic_t +diff -urNp linux-2.6.31.1/arch/mips/include/asm/atomic.h linux-2.6.31.1/arch/mips/include/asm/atomic.h +--- linux-2.6.31.1/arch/mips/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mips/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -32,6 +32,14 @@ + #define atomic_read(v) ((v)->counter) + + /* ++ * atomic_read_unchecked - read atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically reads the value of @v. ++ */ ++#define atomic_read_unchecked(v) ((v)->counter) ++ ++/* + * atomic_set - set atomic variable + * @v: pointer of type atomic_t + * @i: required value +@@ -41,6 +49,15 @@ + #define atomic_set(v, i) ((v)->counter = (i)) + + /* ++ * atomic_set_unchecked - set atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * @i: required value ++ * ++ * Atomically sets the value of @v to @i. ++ */ ++#define atomic_set_unchecked(v, i) ((v)->counter = (i)) ++ ++/* + * atomic_add - add integer to atomic variable + * @i: integer value to add + * @v: pointer of type atomic_t +@@ -381,6 +398,9 @@ static __inline__ int atomic_add_unless( + * Atomically increments @v by 1. + */ + #define atomic_inc(v) atomic_add(1, (v)) ++#define atomic_inc_unchecked(v) atomic_inc((atomic_t *)(v)) ++#define atomic_add_unchecked(i, v) atomic_add((i), (atomic_t *)(v)) ++#define atomic_sub_unchecked(i, v) atomic_sub((i), (atomic_t *)(v)) + + /* + * atomic_dec - decrement and test +diff -urNp linux-2.6.31.1/arch/mips/include/asm/elf.h linux-2.6.31.1/arch/mips/include/asm/elf.h +--- linux-2.6.31.1/arch/mips/include/asm/elf.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mips/include/asm/elf.h 2009-10-01 20:12:42.000000000 -0400 +@@ -368,4 +368,11 @@ extern int dump_task_fpu(struct task_str + #define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) + #endif + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT_ADDR) ? 0x00400000UL : 0x00400000UL) ++ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#endif ++ + #endif /* _ASM_ELF_H */ +diff -urNp linux-2.6.31.1/arch/mips/include/asm/page.h linux-2.6.31.1/arch/mips/include/asm/page.h +--- linux-2.6.31.1/arch/mips/include/asm/page.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mips/include/asm/page.h 2009-10-01 20:12:42.000000000 -0400 +@@ -92,7 +92,7 @@ extern void copy_user_highpage(struct pa + #ifdef CONFIG_CPU_MIPS32 + typedef struct { unsigned long pte_low, pte_high; } pte_t; + #define pte_val(x) ((x).pte_low | ((unsigned long long)(x).pte_high << 32)) +- #define __pte(x) ({ pte_t __pte = {(x), ((unsigned long long)(x)) >> 32}; __pte; }) ++ #define __pte(x) ({ pte_t __pte = {(x), (x) >> 32}; __pte; }) + #else + typedef struct { unsigned long long pte; } pte_t; + #define pte_val(x) ((x).pte) +diff -urNp linux-2.6.31.1/arch/mips/include/asm/system.h linux-2.6.31.1/arch/mips/include/asm/system.h +--- linux-2.6.31.1/arch/mips/include/asm/system.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mips/include/asm/system.h 2009-10-01 20:12:42.000000000 -0400 +@@ -217,6 +217,6 @@ extern void per_cpu_trap_init(void); + */ + #define __ARCH_WANT_UNLOCKED_CTXSW + +-extern unsigned long arch_align_stack(unsigned long sp); ++#define arch_align_stack(x) ((x) & ALMASK) + + #endif /* _ASM_SYSTEM_H */ +diff -urNp linux-2.6.31.1/arch/mips/kernel/binfmt_elfn32.c linux-2.6.31.1/arch/mips/kernel/binfmt_elfn32.c +--- linux-2.6.31.1/arch/mips/kernel/binfmt_elfn32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mips/kernel/binfmt_elfn32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -50,6 +50,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_N + #undef ELF_ET_DYN_BASE + #define ELF_ET_DYN_BASE (TASK32_SIZE / 3 * 2) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT_ADDR) ? 0x00400000UL : 0x00400000UL) ++ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#endif ++ + #include + #include + #include +diff -urNp linux-2.6.31.1/arch/mips/kernel/binfmt_elfo32.c linux-2.6.31.1/arch/mips/kernel/binfmt_elfo32.c +--- linux-2.6.31.1/arch/mips/kernel/binfmt_elfo32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mips/kernel/binfmt_elfo32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -52,6 +52,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_N + #undef ELF_ET_DYN_BASE + #define ELF_ET_DYN_BASE (TASK32_SIZE / 3 * 2) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT_ADDR) ? 0x00400000UL : 0x00400000UL) ++ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#endif ++ + #include + + /* +diff -urNp linux-2.6.31.1/arch/mips/kernel/process.c linux-2.6.31.1/arch/mips/kernel/process.c +--- linux-2.6.31.1/arch/mips/kernel/process.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mips/kernel/process.c 2009-10-01 20:12:42.000000000 -0400 +@@ -470,15 +470,3 @@ unsigned long get_wchan(struct task_stru + out: + return pc; + } +- +-/* +- * Don't forget that the stack pointer must be aligned on a 8 bytes +- * boundary for 32-bits ABI and 16 bytes for 64-bits ABI. +- */ +-unsigned long arch_align_stack(unsigned long sp) +-{ +- if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) +- sp -= get_random_int() & ~PAGE_MASK; +- +- return sp & ALMASK; +-} +diff -urNp linux-2.6.31.1/arch/mips/kernel/syscall.c linux-2.6.31.1/arch/mips/kernel/syscall.c +--- linux-2.6.31.1/arch/mips/kernel/syscall.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mips/kernel/syscall.c 2009-10-01 20:12:42.000000000 -0400 +@@ -99,6 +99,11 @@ unsigned long arch_get_unmapped_area(str + do_color_align = 0; + if (filp || (flags & MAP_SHARED)) + do_color_align = 1; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(current->mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + if (do_color_align) + addr = COLOUR_ALIGN(addr, pgoff); +@@ -109,7 +114,7 @@ unsigned long arch_get_unmapped_area(str + (!vmm || addr + len <= vmm->vm_start)) + return addr; + } +- addr = TASK_UNMAPPED_BASE; ++ addr = current->mm->mmap_base; + if (do_color_align) + addr = COLOUR_ALIGN(addr, pgoff); + else +diff -urNp linux-2.6.31.1/arch/mips/mm/fault.c linux-2.6.31.1/arch/mips/mm/fault.c +--- linux-2.6.31.1/arch/mips/mm/fault.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mips/mm/fault.c 2009-10-01 20:12:42.000000000 -0400 +@@ -26,6 +26,23 @@ + #include + #include /* For VMALLOC_END */ + ++#ifdef CONFIG_PAX_PAGEEXEC ++void pax_report_insns(void *pc) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + /* + * This routine handles page faults. It determines the address, + * and the problem, and then passes it off to one of the appropriate +diff -urNp linux-2.6.31.1/arch/mn10300/include/asm/atomic.h linux-2.6.31.1/arch/mn10300/include/asm/atomic.h +--- linux-2.6.31.1/arch/mn10300/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mn10300/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -34,6 +34,15 @@ + #define atomic_read(v) ((v)->counter) + + /** ++ * atomic_read_unchecked - read atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically reads the value of @v. Note that the guaranteed ++ * useful range of an atomic_unchecked_t is only 24 bits. ++ */ ++#define atomic_read_unchecked(v) ((v)->counter) ++ ++/** + * atomic_set - set atomic variable + * @v: pointer of type atomic_t + * @i: required value +@@ -43,6 +52,16 @@ + */ + #define atomic_set(v, i) (((v)->counter) = (i)) + ++/** ++ * atomic_set_unchecked - set atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * @i: required value ++ * ++ * Atomically sets the value of @v to @i. Note that the guaranteed ++ * useful range of an atomic_unchecked_t is only 24 bits. ++ */ ++#define atomic_set_unchecked(v, i) (((v)->counter) = (i)) ++ + #include + + /** +@@ -99,16 +118,31 @@ static inline void atomic_add(int i, ato + atomic_add_return(i, v); + } + ++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_add_return(i, (atomic_t *)v); ++} ++ + static inline void atomic_sub(int i, atomic_t *v) + { + atomic_sub_return(i, v); + } + ++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_sub_return(i, (atomic_t *)v); ++} ++ + static inline void atomic_inc(atomic_t *v) + { + atomic_add_return(1, v); + } + ++static inline void atomic_inc_unchecked(atomic_unchecked_t *v) ++{ ++ atomic_add_return(1, (atomic_t *)v); ++} ++ + static inline void atomic_dec(atomic_t *v) + { + atomic_sub_return(1, v); +diff -urNp linux-2.6.31.1/arch/mn10300/kernel/setup.c linux-2.6.31.1/arch/mn10300/kernel/setup.c +--- linux-2.6.31.1/arch/mn10300/kernel/setup.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/mn10300/kernel/setup.c 2009-10-01 20:12:42.000000000 -0400 +@@ -285,7 +285,7 @@ static void c_stop(struct seq_file *m, v + { + } + +-struct seq_operations cpuinfo_op = { ++const struct seq_operations cpuinfo_op = { + .start = c_start, + .next = c_next, + .stop = c_stop, +diff -urNp linux-2.6.31.1/arch/parisc/include/asm/atomic.h linux-2.6.31.1/arch/parisc/include/asm/atomic.h +--- linux-2.6.31.1/arch/parisc/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/parisc/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -177,6 +177,18 @@ static __inline__ int __atomic_add_retur + return ret; + } + ++static __inline__ int __atomic_add_return_unchecked(int i, atomic_unchecked_t *v) ++{ ++ int ret; ++ unsigned long flags; ++ _atomic_spin_lock_irqsave(v, flags); ++ ++ ret = (v->counter += i); ++ ++ _atomic_spin_unlock_irqrestore(v, flags); ++ return ret; ++} ++ + static __inline__ void atomic_set(atomic_t *v, int i) + { + unsigned long flags; +@@ -187,11 +199,26 @@ static __inline__ void atomic_set(atomic + _atomic_spin_unlock_irqrestore(v, flags); + } + ++static __inline__ void atomic_set_unchecked(atomic_unchecked_t *v, int i) ++{ ++ unsigned long flags; ++ _atomic_spin_lock_irqsave(v, flags); ++ ++ v->counter = i; ++ ++ _atomic_spin_unlock_irqrestore(v, flags); ++} ++ + static __inline__ int atomic_read(const atomic_t *v) + { + return v->counter; + } + ++static __inline__ int atomic_read_unchecked(const atomic_unchecked_t *v) ++{ ++ return v->counter; ++} ++ + /* exported interface */ + #define atomic_cmpxchg(v, o, n) (cmpxchg(&((v)->counter), (o), (n))) + #define atomic_xchg(v, new) (xchg(&((v)->counter), new)) +@@ -223,8 +250,11 @@ static __inline__ int atomic_add_unless( + #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) + + #define atomic_add(i,v) ((void)(__atomic_add_return( (i),(v)))) ++#define atomic_add_unchecked(i,v) ((void)(__atomic_add_return_unchecked( ((i),(v)))) + #define atomic_sub(i,v) ((void)(__atomic_add_return(-(i),(v)))) ++#define atomic_sub_unchecked(i,v) ((void)(__atomic_add_return_unchecked(-(i),(v)))) + #define atomic_inc(v) ((void)(__atomic_add_return( 1,(v)))) ++#define atomic_inc_unchecked(v) ((void)(__atomic_add_return_unchecked( 1,(v)))) + #define atomic_dec(v) ((void)(__atomic_add_return( -1,(v)))) + + #define atomic_add_return(i,v) (__atomic_add_return( (i),(v))) +diff -urNp linux-2.6.31.1/arch/parisc/include/asm/elf.h linux-2.6.31.1/arch/parisc/include/asm/elf.h +--- linux-2.6.31.1/arch/parisc/include/asm/elf.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/parisc/include/asm/elf.h 2009-10-01 20:12:42.000000000 -0400 +@@ -343,6 +343,13 @@ struct pt_regs; /* forward declaration.. + + #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x01000000) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE 0x10000UL ++ ++#define PAX_DELTA_MMAP_LEN 16 ++#define PAX_DELTA_STACK_LEN 16 ++#endif ++ + /* This yields a mask that user programs can use to figure out what + instruction set this CPU supports. This could be done in user space, + but it's not easy, and we've already done it here. */ +diff -urNp linux-2.6.31.1/arch/parisc/include/asm/pgtable.h linux-2.6.31.1/arch/parisc/include/asm/pgtable.h +--- linux-2.6.31.1/arch/parisc/include/asm/pgtable.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/parisc/include/asm/pgtable.h 2009-10-01 20:12:42.000000000 -0400 +@@ -207,6 +207,17 @@ + #define PAGE_EXECREAD __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_EXEC |_PAGE_ACCESSED) + #define PAGE_COPY PAGE_EXECREAD + #define PAGE_RWX __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_WRITE | _PAGE_EXEC |_PAGE_ACCESSED) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++# define PAGE_SHARED_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_WRITE | _PAGE_ACCESSED) ++# define PAGE_COPY_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_ACCESSED) ++# define PAGE_READONLY_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_ACCESSED) ++#else ++# define PAGE_SHARED_NOEXEC PAGE_SHARED ++# define PAGE_COPY_NOEXEC PAGE_COPY ++# define PAGE_READONLY_NOEXEC PAGE_READONLY ++#endif ++ + #define PAGE_KERNEL __pgprot(_PAGE_KERNEL) + #define PAGE_KERNEL_RO __pgprot(_PAGE_KERNEL & ~_PAGE_WRITE) + #define PAGE_KERNEL_UNC __pgprot(_PAGE_KERNEL | _PAGE_NO_CACHE) +diff -urNp linux-2.6.31.1/arch/parisc/kernel/module.c linux-2.6.31.1/arch/parisc/kernel/module.c +--- linux-2.6.31.1/arch/parisc/kernel/module.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/parisc/kernel/module.c 2009-10-01 20:12:42.000000000 -0400 +@@ -95,16 +95,38 @@ + + /* three functions to determine where in the module core + * or init pieces the location is */ ++static inline int in_init_rx(struct module *me, void *loc) ++{ ++ return (loc >= me->module_init_rx && ++ loc < (me->module_init_rx + me->init_size_rx)); ++} ++ ++static inline int in_init_rw(struct module *me, void *loc) ++{ ++ return (loc >= me->module_init_rw && ++ loc < (me->module_init_rw + me->init_size_rw)); ++} ++ + static inline int in_init(struct module *me, void *loc) + { +- return (loc >= me->module_init && +- loc <= (me->module_init + me->init_size)); ++ return in_init_rx(me, loc) || in_init_rw(me, loc); ++} ++ ++static inline int in_core_rx(struct module *me, void *loc) ++{ ++ return (loc >= me->module_core_rx && ++ loc < (me->module_core_rx + me->core_size_rx)); ++} ++ ++static inline int in_core_rw(struct module *me, void *loc) ++{ ++ return (loc >= me->module_core_rw && ++ loc < (me->module_core_rw + me->core_size_rw)); + } + + static inline int in_core(struct module *me, void *loc) + { +- return (loc >= me->module_core && +- loc <= (me->module_core + me->core_size)); ++ return in_core_rx(me, loc) || in_core_rw(me, loc); + } + + static inline int in_local(struct module *me, void *loc) +@@ -364,13 +386,13 @@ int module_frob_arch_sections(CONST Elf_ + } + + /* align things a bit */ +- me->core_size = ALIGN(me->core_size, 16); +- me->arch.got_offset = me->core_size; +- me->core_size += gots * sizeof(struct got_entry); +- +- me->core_size = ALIGN(me->core_size, 16); +- me->arch.fdesc_offset = me->core_size; +- me->core_size += fdescs * sizeof(Elf_Fdesc); ++ me->core_size_rw = ALIGN(me->core_size_rw, 16); ++ me->arch.got_offset = me->core_size_rw; ++ me->core_size_rw += gots * sizeof(struct got_entry); ++ ++ me->core_size_rw = ALIGN(me->core_size_rw, 16); ++ me->arch.fdesc_offset = me->core_size_rw; ++ me->core_size_rw += fdescs * sizeof(Elf_Fdesc); + + me->arch.got_max = gots; + me->arch.fdesc_max = fdescs; +@@ -388,7 +410,7 @@ static Elf64_Word get_got(struct module + + BUG_ON(value == 0); + +- got = me->module_core + me->arch.got_offset; ++ got = me->module_core_rw + me->arch.got_offset; + for (i = 0; got[i].addr; i++) + if (got[i].addr == value) + goto out; +@@ -406,7 +428,7 @@ static Elf64_Word get_got(struct module + #ifdef CONFIG_64BIT + static Elf_Addr get_fdesc(struct module *me, unsigned long value) + { +- Elf_Fdesc *fdesc = me->module_core + me->arch.fdesc_offset; ++ Elf_Fdesc *fdesc = me->module_core_rw + me->arch.fdesc_offset; + + if (!value) { + printk(KERN_ERR "%s: zero OPD requested!\n", me->name); +@@ -424,7 +446,7 @@ static Elf_Addr get_fdesc(struct module + + /* Create new one */ + fdesc->addr = value; +- fdesc->gp = (Elf_Addr)me->module_core + me->arch.got_offset; ++ fdesc->gp = (Elf_Addr)me->module_core_rw + me->arch.got_offset; + return (Elf_Addr)fdesc; + } + #endif /* CONFIG_64BIT */ +@@ -848,7 +870,7 @@ register_unwind_table(struct module *me, + + table = (unsigned char *)sechdrs[me->arch.unwind_section].sh_addr; + end = table + sechdrs[me->arch.unwind_section].sh_size; +- gp = (Elf_Addr)me->module_core + me->arch.got_offset; ++ gp = (Elf_Addr)me->module_core_rw + me->arch.got_offset; + + DEBUGP("register_unwind_table(), sect = %d at 0x%p - 0x%p (gp=0x%lx)\n", + me->arch.unwind_section, table, end, gp); +diff -urNp linux-2.6.31.1/arch/parisc/kernel/sys_parisc.c linux-2.6.31.1/arch/parisc/kernel/sys_parisc.c +--- linux-2.6.31.1/arch/parisc/kernel/sys_parisc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/parisc/kernel/sys_parisc.c 2009-10-01 20:12:42.000000000 -0400 +@@ -98,7 +98,7 @@ unsigned long arch_get_unmapped_area(str + if (flags & MAP_FIXED) + return addr; + if (!addr) +- addr = TASK_UNMAPPED_BASE; ++ addr = current->mm->mmap_base; + + if (filp) { + addr = get_shared_area(filp->f_mapping, addr, len, pgoff); +diff -urNp linux-2.6.31.1/arch/parisc/kernel/traps.c linux-2.6.31.1/arch/parisc/kernel/traps.c +--- linux-2.6.31.1/arch/parisc/kernel/traps.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/parisc/kernel/traps.c 2009-10-01 20:12:42.000000000 -0400 +@@ -733,9 +733,7 @@ void notrace handle_interruption(int cod + + down_read(¤t->mm->mmap_sem); + vma = find_vma(current->mm,regs->iaoq[0]); +- if (vma && (regs->iaoq[0] >= vma->vm_start) +- && (vma->vm_flags & VM_EXEC)) { +- ++ if (vma && (regs->iaoq[0] >= vma->vm_start)) { + fault_address = regs->iaoq[0]; + fault_space = regs->iasq[0]; + +diff -urNp linux-2.6.31.1/arch/parisc/mm/fault.c linux-2.6.31.1/arch/parisc/mm/fault.c +--- linux-2.6.31.1/arch/parisc/mm/fault.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/parisc/mm/fault.c 2009-10-01 20:12:42.000000000 -0400 +@@ -15,6 +15,7 @@ + #include + #include + #include ++#include + + #include + #include +@@ -52,7 +53,7 @@ DEFINE_PER_CPU(struct exception_data, ex + static unsigned long + parisc_acctyp(unsigned long code, unsigned int inst) + { +- if (code == 6 || code == 16) ++ if (code == 6 || code == 7 || code == 16) + return VM_EXEC; + + switch (inst & 0xf0000000) { +@@ -138,6 +139,116 @@ parisc_acctyp(unsigned long code, unsign + } + #endif + ++#ifdef CONFIG_PAX_PAGEEXEC ++/* ++ * PaX: decide what to do with offenders (instruction_pointer(regs) = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when rt_sigreturn trampoline was detected ++ * 3 when unpatched PLT trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ ++#ifdef CONFIG_PAX_EMUPLT ++ int err; ++ ++ do { /* PaX: unpatched PLT emulation */ ++ unsigned int bl, depwi; ++ ++ err = get_user(bl, (unsigned int *)instruction_pointer(regs)); ++ err |= get_user(depwi, (unsigned int *)(instruction_pointer(regs)+4)); ++ ++ if (err) ++ break; ++ ++ if (bl == 0xEA9F1FDDU && depwi == 0xD6801C1EU) { ++ unsigned int ldw, bv, ldw2, addr = instruction_pointer(regs)-12; ++ ++ err = get_user(ldw, (unsigned int *)addr); ++ err |= get_user(bv, (unsigned int *)(addr+4)); ++ err |= get_user(ldw2, (unsigned int *)(addr+8)); ++ ++ if (err) ++ break; ++ ++ if (ldw == 0x0E801096U && ++ bv == 0xEAC0C000U && ++ ldw2 == 0x0E881095U) ++ { ++ unsigned int resolver, map; ++ ++ err = get_user(resolver, (unsigned int *)(instruction_pointer(regs)+8)); ++ err |= get_user(map, (unsigned int *)(instruction_pointer(regs)+12)); ++ if (err) ++ break; ++ ++ regs->gr[20] = instruction_pointer(regs)+8; ++ regs->gr[21] = map; ++ regs->gr[22] = resolver; ++ regs->iaoq[0] = resolver | 3UL; ++ regs->iaoq[1] = regs->iaoq[0] + 4; ++ return 3; ++ } ++ } ++ } while (0); ++#endif ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ ++#ifndef CONFIG_PAX_EMUSIGRT ++ if (!(current->mm->pax_flags & MF_PAX_EMUTRAMP)) ++ return 1; ++#endif ++ ++ do { /* PaX: rt_sigreturn emulation */ ++ unsigned int ldi1, ldi2, bel, nop; ++ ++ err = get_user(ldi1, (unsigned int *)instruction_pointer(regs)); ++ err |= get_user(ldi2, (unsigned int *)(instruction_pointer(regs)+4)); ++ err |= get_user(bel, (unsigned int *)(instruction_pointer(regs)+8)); ++ err |= get_user(nop, (unsigned int *)(instruction_pointer(regs)+12)); ++ ++ if (err) ++ break; ++ ++ if ((ldi1 == 0x34190000U || ldi1 == 0x34190002U) && ++ ldi2 == 0x3414015AU && ++ bel == 0xE4008200U && ++ nop == 0x08000240U) ++ { ++ regs->gr[25] = (ldi1 & 2) >> 1; ++ regs->gr[20] = __NR_rt_sigreturn; ++ regs->gr[31] = regs->iaoq[1] + 16; ++ regs->sr[0] = regs->iasq[1]; ++ regs->iaoq[0] = 0x100UL; ++ regs->iaoq[1] = regs->iaoq[0] + 4; ++ regs->iasq[0] = regs->sr[2]; ++ regs->iasq[1] = regs->sr[2]; ++ return 2; ++ } ++ } while (0); ++#endif ++ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + int fixup_exception(struct pt_regs *regs) + { + const struct exception_table_entry *fix; +@@ -192,8 +303,33 @@ good_area: + + acc_type = parisc_acctyp(code,regs->iir); + +- if ((vma->vm_flags & acc_type) != acc_type) ++ if ((vma->vm_flags & acc_type) != acc_type) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && (acc_type & VM_EXEC) && ++ (address & ~3UL) == instruction_pointer(regs)) ++ { ++ up_read(&mm->mmap_sem); ++ switch (pax_handle_fetch_fault(regs)) { ++ ++#ifdef CONFIG_PAX_EMUPLT ++ case 3: ++ return; ++#endif ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ case 2: ++ return; ++#endif ++ ++ } ++ pax_report_fault(regs, (void *)instruction_pointer(regs), (void *)regs->gr[30]); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ + goto bad_area; ++ } + + /* + * If for any reason at all we couldn't handle the fault, make +diff -urNp linux-2.6.31.1/arch/powerpc/include/asm/atomic.h linux-2.6.31.1/arch/powerpc/include/asm/atomic.h +--- linux-2.6.31.1/arch/powerpc/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -24,11 +24,21 @@ static __inline__ int atomic_read(const + return t; + } + ++static __inline__ int atomic_read_unchecked(const atomic_unchecked_t *v) ++{ ++ return atomic_read((const atomic_t *)v); ++} ++ + static __inline__ void atomic_set(atomic_t *v, int i) + { + __asm__ __volatile__("stw%U0%X0 %1,%0" : "=m"(v->counter) : "r"(i)); + } + ++static __inline__ void atomic_set_unchecked(atomic_unchecked_t *v, int i) ++{ ++ atomic_set((atomic_t *)v, i); ++} ++ + static __inline__ void atomic_add(int a, atomic_t *v) + { + int t; +@@ -44,6 +54,11 @@ static __inline__ void atomic_add(int a, + : "cc"); + } + ++static __inline__ void atomic_add_unchecked(int a, atomic_unchecked_t *v) ++{ ++ atomic_add(a, (atomic_t *)v); ++} ++ + static __inline__ int atomic_add_return(int a, atomic_t *v) + { + int t; +@@ -80,6 +95,11 @@ static __inline__ void atomic_sub(int a, + : "cc"); + } + ++static __inline__ void atomic_sub_unchecked(int a, atomic_unchecked_t *v) ++{ ++ atomic_sub(a, (atomic_t *)v); ++} ++ + static __inline__ int atomic_sub_return(int a, atomic_t *v) + { + int t; +@@ -114,6 +134,11 @@ static __inline__ void atomic_inc(atomic + : "cc", "xer"); + } + ++static __inline__ void atomic_inc_unchecked(atomic_unchecked_t *v) ++{ ++ atomic_inc((atomic_t *)v); ++} ++ + static __inline__ int atomic_inc_return(atomic_t *v) + { + int t; +diff -urNp linux-2.6.31.1/arch/powerpc/include/asm/elf.h linux-2.6.31.1/arch/powerpc/include/asm/elf.h +--- linux-2.6.31.1/arch/powerpc/include/asm/elf.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/include/asm/elf.h 2009-10-01 20:12:42.000000000 -0400 +@@ -179,8 +179,19 @@ typedef elf_fpreg_t elf_vsrreghalf_t32[E + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. */ + +-extern unsigned long randomize_et_dyn(unsigned long base); +-#define ELF_ET_DYN_BASE (randomize_et_dyn(0x20000000)) ++#define ELF_ET_DYN_BASE (0x20000000) ++ ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (0x10000000UL) ++ ++#ifdef __powerpc64__ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT) ? 16 : 28) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT) ? 16 : 28) ++#else ++#define PAX_DELTA_MMAP_LEN 15 ++#define PAX_DELTA_STACK_LEN 15 ++#endif ++#endif + + /* + * Our registers are always unsigned longs, whether we're a 32 bit +@@ -279,9 +290,6 @@ extern int arch_setup_additional_pages(s + (0x7ff >> (PAGE_SHIFT - 12)) : \ + (0x3ffff >> (PAGE_SHIFT - 12))) + +-extern unsigned long arch_randomize_brk(struct mm_struct *mm); +-#define arch_randomize_brk arch_randomize_brk +- + #endif /* __KERNEL__ */ + + /* +diff -urNp linux-2.6.31.1/arch/powerpc/include/asm/kmap_types.h linux-2.6.31.1/arch/powerpc/include/asm/kmap_types.h +--- linux-2.6.31.1/arch/powerpc/include/asm/kmap_types.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/include/asm/kmap_types.h 2009-10-01 20:12:42.000000000 -0400 +@@ -26,6 +26,7 @@ enum km_type { + KM_SOFTIRQ1, + KM_PPC_SYNC_PAGE, + KM_PPC_SYNC_ICACHE, ++ KM_CLEARPAGE, + KM_TYPE_NR + }; + +diff -urNp linux-2.6.31.1/arch/powerpc/include/asm/page_64.h linux-2.6.31.1/arch/powerpc/include/asm/page_64.h +--- linux-2.6.31.1/arch/powerpc/include/asm/page_64.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/include/asm/page_64.h 2009-10-01 20:12:42.000000000 -0400 +@@ -170,15 +170,18 @@ do { \ + * stack by default, so in the absense of a PT_GNU_STACK program header + * we turn execute permission off. + */ +-#define VM_STACK_DEFAULT_FLAGS32 (VM_READ | VM_WRITE | VM_EXEC | \ +- VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) ++#define VM_STACK_DEFAULT_FLAGS32 \ ++ (((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \ ++ VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + + #define VM_STACK_DEFAULT_FLAGS64 (VM_READ | VM_WRITE | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + ++#ifndef CONFIG_PAX_PAGEEXEC + #define VM_STACK_DEFAULT_FLAGS \ + (test_thread_flag(TIF_32BIT) ? \ + VM_STACK_DEFAULT_FLAGS32 : VM_STACK_DEFAULT_FLAGS64) ++#endif + + #include + +diff -urNp linux-2.6.31.1/arch/powerpc/include/asm/page.h linux-2.6.31.1/arch/powerpc/include/asm/page.h +--- linux-2.6.31.1/arch/powerpc/include/asm/page.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/include/asm/page.h 2009-10-01 20:12:42.000000000 -0400 +@@ -116,8 +116,9 @@ extern phys_addr_t kernstart_addr; + * and needs to be executable. This means the whole heap ends + * up being executable. + */ +-#define VM_DATA_DEFAULT_FLAGS32 (VM_READ | VM_WRITE | VM_EXEC | \ +- VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) ++#define VM_DATA_DEFAULT_FLAGS32 \ ++ (((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \ ++ VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + + #define VM_DATA_DEFAULT_FLAGS64 (VM_READ | VM_WRITE | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) +diff -urNp linux-2.6.31.1/arch/powerpc/include/asm/pte-common.h linux-2.6.31.1/arch/powerpc/include/asm/pte-common.h +--- linux-2.6.31.1/arch/powerpc/include/asm/pte-common.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/include/asm/pte-common.h 2009-10-01 20:12:42.000000000 -0400 +@@ -121,11 +121,11 @@ extern unsigned long bad_call_to_PMD_PAG + */ + #define PAGE_NONE __pgprot(_PAGE_BASE) + #define PAGE_SHARED __pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_RW) +-#define PAGE_SHARED_X __pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_RW | _PAGE_EXEC) ++#define PAGE_SHARED_X __pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_RW | _PAGE_EXEC | _PAGE_HWEXEC) + #define PAGE_COPY __pgprot(_PAGE_BASE | _PAGE_USER) +-#define PAGE_COPY_X __pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_EXEC) ++#define PAGE_COPY_X __pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_EXEC | _PAGE_HWEXEC) + #define PAGE_READONLY __pgprot(_PAGE_BASE | _PAGE_USER) +-#define PAGE_READONLY_X __pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_EXEC) ++#define PAGE_READONLY_X __pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_EXEC | _PAGE_HWEXEC) + + #define __P000 PAGE_NONE + #define __P001 PAGE_READONLY +diff -urNp linux-2.6.31.1/arch/powerpc/include/asm/pte-hash32.h linux-2.6.31.1/arch/powerpc/include/asm/pte-hash32.h +--- linux-2.6.31.1/arch/powerpc/include/asm/pte-hash32.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/include/asm/pte-hash32.h 2009-10-01 20:12:42.000000000 -0400 +@@ -21,6 +21,7 @@ + #define _PAGE_FILE 0x004 /* when !present: nonlinear file mapping */ + #define _PAGE_USER 0x004 /* usermode access allowed */ + #define _PAGE_GUARDED 0x008 /* G: prohibit speculative access */ ++#define _PAGE_HWEXEC _PAGE_GUARDED + #define _PAGE_COHERENT 0x010 /* M: enforce memory coherence (SMP systems) */ + #define _PAGE_NO_CACHE 0x020 /* I: cache inhibit */ + #define _PAGE_WRITETHRU 0x040 /* W: cache write-through */ +diff -urNp linux-2.6.31.1/arch/powerpc/include/asm/reg.h linux-2.6.31.1/arch/powerpc/include/asm/reg.h +--- linux-2.6.31.1/arch/powerpc/include/asm/reg.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/include/asm/reg.h 2009-10-01 20:12:42.000000000 -0400 +@@ -195,6 +195,7 @@ + #define SPRN_DBCR 0x136 /* e300 Data Breakpoint Control Reg */ + #define SPRN_DSISR 0x012 /* Data Storage Interrupt Status Register */ + #define DSISR_NOHPTE 0x40000000 /* no translation found */ ++#define DSISR_GUARDED 0x10000000 /* fetch from guarded storage */ + #define DSISR_PROTFAULT 0x08000000 /* protection fault */ + #define DSISR_ISSTORE 0x02000000 /* access was a store */ + #define DSISR_DABRMATCH 0x00400000 /* hit data breakpoint */ +diff -urNp linux-2.6.31.1/arch/powerpc/include/asm/uaccess.h linux-2.6.31.1/arch/powerpc/include/asm/uaccess.h +--- linux-2.6.31.1/arch/powerpc/include/asm/uaccess.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/include/asm/uaccess.h 2009-10-01 20:12:42.000000000 -0400 +@@ -327,52 +327,6 @@ do { \ + extern unsigned long __copy_tofrom_user(void __user *to, + const void __user *from, unsigned long size); + +-#ifndef __powerpc64__ +- +-static inline unsigned long copy_from_user(void *to, +- const void __user *from, unsigned long n) +-{ +- unsigned long over; +- +- if (access_ok(VERIFY_READ, from, n)) +- return __copy_tofrom_user((__force void __user *)to, from, n); +- if ((unsigned long)from < TASK_SIZE) { +- over = (unsigned long)from + n - TASK_SIZE; +- return __copy_tofrom_user((__force void __user *)to, from, +- n - over) + over; +- } +- return n; +-} +- +-static inline unsigned long copy_to_user(void __user *to, +- const void *from, unsigned long n) +-{ +- unsigned long over; +- +- if (access_ok(VERIFY_WRITE, to, n)) +- return __copy_tofrom_user(to, (__force void __user *)from, n); +- if ((unsigned long)to < TASK_SIZE) { +- over = (unsigned long)to + n - TASK_SIZE; +- return __copy_tofrom_user(to, (__force void __user *)from, +- n - over) + over; +- } +- return n; +-} +- +-#else /* __powerpc64__ */ +- +-#define __copy_in_user(to, from, size) \ +- __copy_tofrom_user((to), (from), (size)) +- +-extern unsigned long copy_from_user(void *to, const void __user *from, +- unsigned long n); +-extern unsigned long copy_to_user(void __user *to, const void *from, +- unsigned long n); +-extern unsigned long copy_in_user(void __user *to, const void __user *from, +- unsigned long n); +- +-#endif /* __powerpc64__ */ +- + static inline unsigned long __copy_from_user_inatomic(void *to, + const void __user *from, unsigned long n) + { +@@ -396,6 +350,9 @@ static inline unsigned long __copy_from_ + if (ret == 0) + return 0; + } ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ + return __copy_tofrom_user((__force void __user *)to, from, n); + } + +@@ -422,6 +379,9 @@ static inline unsigned long __copy_to_us + if (ret == 0) + return 0; + } ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); ++ + return __copy_tofrom_user(to, (__force const void __user *)from, n); + } + +@@ -439,6 +399,97 @@ static inline unsigned long __copy_to_us + return __copy_to_user_inatomic(to, from, size); + } + ++#ifndef __powerpc64__ ++ ++static inline unsigned long __must_check copy_from_user(void *to, ++ const void __user *from, unsigned long n) ++{ ++ unsigned long over; ++ ++ if (((long)n < 0) || (n > INT_MAX)) ++ return n; ++ ++ if (access_ok(VERIFY_READ, from, n)) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ ++ return __copy_tofrom_user((__force void __user *)to, from, n); ++ } ++ if ((unsigned long)from < TASK_SIZE) { ++ over = (unsigned long)from + n - TASK_SIZE; ++ if (!__builtin_constant_p(n - over)) ++ check_object_size(to, n - over, false); ++ return __copy_tofrom_user((__force void __user *)to, from, ++ n - over) + over; ++ } ++ return n; ++} ++ ++static inline unsigned long __must_check copy_to_user(void __user *to, ++ const void *from, unsigned long n) ++{ ++ unsigned long over; ++ ++ if (((long)n < 0) || (n > INT_MAX)) ++ return n; ++ ++ if (access_ok(VERIFY_WRITE, to, n)) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); ++ return __copy_tofrom_user(to, (__force void __user *)from, n); ++ } ++ if ((unsigned long)to < TASK_SIZE) { ++ over = (unsigned long)to + n - TASK_SIZE; ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n - over, true); ++ return __copy_tofrom_user(to, (__force void __user *)from, ++ n - over) + over; ++ } ++ return n; ++} ++ ++#else /* __powerpc64__ */ ++ ++#define __copy_in_user(to, from, size) \ ++ __copy_tofrom_user((to), (from), (size)) ++ ++static inline unsigned long __must_check copy_from_user(void *to, ++ const void __user *from, unsigned long n) ++{ ++ if (unlikely(((long)n < 0) || (n > INT_MAX))) ++ return n; ++ ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ ++ if (likely(access_ok(VERIFY_READ, from, n))) ++ n = __copy_from_user(to, from, n); ++ else ++ memset(to, 0, n); ++ ++ return n; ++} ++ ++static inline unsigned long __must_check copy_to_user(void __user *to, ++ const void *from, unsigned long n) ++{ ++ if (unlikely(((long)n < 0) || (n > INT_MAX))) ++ return n; ++ ++ if (likely(access_ok(VERIFY_WRITE, to, n))) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); ++ n = __copy_to_user(to, from, n); ++ } ++ ++ return n; ++} ++ ++extern unsigned long copy_in_user(void __user *to, const void __user *from, ++ unsigned long n); ++ ++#endif /* __powerpc64__ */ ++ + extern unsigned long __clear_user(void __user *addr, unsigned long size); + + static inline unsigned long clear_user(void __user *addr, unsigned long size) +diff -urNp linux-2.6.31.1/arch/powerpc/kernel/module_32.c linux-2.6.31.1/arch/powerpc/kernel/module_32.c +--- linux-2.6.31.1/arch/powerpc/kernel/module_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/kernel/module_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -162,7 +162,7 @@ int module_frob_arch_sections(Elf32_Ehdr + me->arch.core_plt_section = i; + } + if (!me->arch.core_plt_section || !me->arch.init_plt_section) { +- printk("Module doesn't contain .plt or .init.plt sections.\n"); ++ printk("Module %s doesn't contain .plt or .init.plt sections.\n", me->name); + return -ENOEXEC; + } + +@@ -203,11 +203,16 @@ static uint32_t do_plt_call(void *locati + + DEBUGP("Doing plt for call to 0x%x at 0x%x\n", val, (unsigned int)location); + /* Init, or core PLT? */ +- if (location >= mod->module_core +- && location < mod->module_core + mod->core_size) ++ if ((location >= mod->module_core_rx && location < mod->module_core_rx + mod->core_size_rx) || ++ (location >= mod->module_core_rw && location < mod->module_core_rw + mod->core_size_rw)) + entry = (void *)sechdrs[mod->arch.core_plt_section].sh_addr; +- else ++ else if ((location >= mod->module_init_rx && location < mod->module_init_rx + mod->init_size_rx) || ++ (location >= mod->module_init_rw && location < mod->module_init_rw + mod->init_size_rw)) + entry = (void *)sechdrs[mod->arch.init_plt_section].sh_addr; ++ else { ++ printk(KERN_ERR "%s: invalid R_PPC_REL24 entry found\n", mod->name); ++ return ~0UL; ++ } + + /* Find this entry, or if that fails, the next avail. entry */ + while (entry->jump[0]) { +diff -urNp linux-2.6.31.1/arch/powerpc/kernel/process.c linux-2.6.31.1/arch/powerpc/kernel/process.c +--- linux-2.6.31.1/arch/powerpc/kernel/process.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/kernel/process.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1147,36 +1147,3 @@ unsigned long arch_align_stack(unsigned + sp -= get_random_int() & ~PAGE_MASK; + return sp & ~0xf; + } +- +-static inline unsigned long brk_rnd(void) +-{ +- unsigned long rnd = 0; +- +- /* 8MB for 32bit, 1GB for 64bit */ +- if (is_32bit_task()) +- rnd = (long)(get_random_int() % (1<<(23-PAGE_SHIFT))); +- else +- rnd = (long)(get_random_int() % (1<<(30-PAGE_SHIFT))); +- +- return rnd << PAGE_SHIFT; +-} +- +-unsigned long arch_randomize_brk(struct mm_struct *mm) +-{ +- unsigned long ret = PAGE_ALIGN(mm->brk + brk_rnd()); +- +- if (ret < mm->brk) +- return mm->brk; +- +- return ret; +-} +- +-unsigned long randomize_et_dyn(unsigned long base) +-{ +- unsigned long ret = PAGE_ALIGN(base + brk_rnd()); +- +- if (ret < base) +- return base; +- +- return ret; +-} +diff -urNp linux-2.6.31.1/arch/powerpc/kernel/setup-common.c linux-2.6.31.1/arch/powerpc/kernel/setup-common.c +--- linux-2.6.31.1/arch/powerpc/kernel/setup-common.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/kernel/setup-common.c 2009-10-01 20:12:42.000000000 -0400 +@@ -328,7 +328,7 @@ static void c_stop(struct seq_file *m, v + { + } + +-struct seq_operations cpuinfo_op = { ++const struct seq_operations cpuinfo_op = { + .start =c_start, + .next = c_next, + .stop = c_stop, +diff -urNp linux-2.6.31.1/arch/powerpc/kernel/signal_32.c linux-2.6.31.1/arch/powerpc/kernel/signal_32.c +--- linux-2.6.31.1/arch/powerpc/kernel/signal_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/kernel/signal_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -857,7 +857,7 @@ int handle_rt_signal32(unsigned long sig + /* Save user registers on the stack */ + frame = &rt_sf->uc.uc_mcontext; + addr = frame; +- if (vdso32_rt_sigtramp && current->mm->context.vdso_base) { ++ if (vdso32_rt_sigtramp && current->mm->context.vdso_base != ~0UL) { + if (save_user_regs(regs, frame, 0, 1)) + goto badframe; + regs->link = current->mm->context.vdso_base + vdso32_rt_sigtramp; +diff -urNp linux-2.6.31.1/arch/powerpc/kernel/signal_64.c linux-2.6.31.1/arch/powerpc/kernel/signal_64.c +--- linux-2.6.31.1/arch/powerpc/kernel/signal_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/kernel/signal_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -429,7 +429,7 @@ int handle_rt_signal64(int signr, struct + current->thread.fpscr.val = 0; + + /* Set up to return from userspace. */ +- if (vdso64_rt_sigtramp && current->mm->context.vdso_base) { ++ if (vdso64_rt_sigtramp && current->mm->context.vdso_base != ~0UL) { + regs->link = current->mm->context.vdso_base + vdso64_rt_sigtramp; + } else { + err |= setup_trampoline(__NR_rt_sigreturn, &frame->tramp[0]); +diff -urNp linux-2.6.31.1/arch/powerpc/kernel/sys_ppc32.c linux-2.6.31.1/arch/powerpc/kernel/sys_ppc32.c +--- linux-2.6.31.1/arch/powerpc/kernel/sys_ppc32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/kernel/sys_ppc32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -552,10 +552,10 @@ asmlinkage long compat_sys_sysctl(struct + if (oldlenp) { + if (!error) { + if (get_user(oldlen, oldlenp) || +- put_user(oldlen, (compat_size_t __user *)compat_ptr(tmp.oldlenp))) ++ put_user(oldlen, (compat_size_t __user *)compat_ptr(tmp.oldlenp)) || ++ copy_to_user(args->__unused, tmp.__unused, sizeof(tmp.__unused))) + error = -EFAULT; + } +- copy_to_user(args->__unused, tmp.__unused, sizeof(tmp.__unused)); + } + return error; + } +diff -urNp linux-2.6.31.1/arch/powerpc/kernel/vdso.c linux-2.6.31.1/arch/powerpc/kernel/vdso.c +--- linux-2.6.31.1/arch/powerpc/kernel/vdso.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/kernel/vdso.c 2009-10-01 20:12:42.000000000 -0400 +@@ -35,6 +35,7 @@ + #include + #include + #include ++#include + + #include "setup.h" + +@@ -211,7 +212,7 @@ int arch_setup_additional_pages(struct l + vdso_base = VDSO32_MBASE; + #endif + +- current->mm->context.vdso_base = 0; ++ current->mm->context.vdso_base = ~0UL; + + /* vDSO has a problem and was disabled, just don't "enable" it for the + * process +@@ -228,7 +229,7 @@ int arch_setup_additional_pages(struct l + */ + down_write(&mm->mmap_sem); + vdso_base = get_unmapped_area(NULL, vdso_base, +- vdso_pages << PAGE_SHIFT, 0, 0); ++ vdso_pages << PAGE_SHIFT, 0, MAP_PRIVATE | MAP_EXECUTABLE); + if (IS_ERR_VALUE(vdso_base)) { + rc = vdso_base; + goto fail_mmapsem; +diff -urNp linux-2.6.31.1/arch/powerpc/kvm/timing.c linux-2.6.31.1/arch/powerpc/kvm/timing.c +--- linux-2.6.31.1/arch/powerpc/kvm/timing.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/kvm/timing.c 2009-10-01 20:12:42.000000000 -0400 +@@ -201,7 +201,7 @@ static int kvmppc_exit_timing_open(struc + return single_open(file, kvmppc_exit_timing_show, inode->i_private); + } + +-static struct file_operations kvmppc_exit_timing_fops = { ++static const struct file_operations kvmppc_exit_timing_fops = { + .owner = THIS_MODULE, + .open = kvmppc_exit_timing_open, + .read = seq_read, +diff -urNp linux-2.6.31.1/arch/powerpc/lib/usercopy_64.c linux-2.6.31.1/arch/powerpc/lib/usercopy_64.c +--- linux-2.6.31.1/arch/powerpc/lib/usercopy_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/lib/usercopy_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -9,22 +9,6 @@ + #include + #include + +-unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) +-{ +- if (likely(access_ok(VERIFY_READ, from, n))) +- n = __copy_from_user(to, from, n); +- else +- memset(to, 0, n); +- return n; +-} +- +-unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) +-{ +- if (likely(access_ok(VERIFY_WRITE, to, n))) +- n = __copy_to_user(to, from, n); +- return n; +-} +- + unsigned long copy_in_user(void __user *to, const void __user *from, + unsigned long n) + { +@@ -35,7 +19,5 @@ unsigned long copy_in_user(void __user * + return n; + } + +-EXPORT_SYMBOL(copy_from_user); +-EXPORT_SYMBOL(copy_to_user); + EXPORT_SYMBOL(copy_in_user); + +diff -urNp linux-2.6.31.1/arch/powerpc/mm/fault.c linux-2.6.31.1/arch/powerpc/mm/fault.c +--- linux-2.6.31.1/arch/powerpc/mm/fault.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/mm/fault.c 2009-10-01 20:12:42.000000000 -0400 +@@ -30,6 +30,10 @@ + #include + #include + #include ++#include ++#include ++#include ++#include + + #include + #include +@@ -40,6 +44,7 @@ + #include + #include + #include ++#include + + + #ifdef CONFIG_KPROBES +@@ -64,6 +69,33 @@ static inline int notify_page_fault(stru + } + #endif + ++#ifdef CONFIG_PAX_PAGEEXEC ++/* ++ * PaX: decide what to do with offenders (regs->nip = fault address) ++ * ++ * returns 1 when task should be killed ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + /* + * Check whether the instruction at regs->nip is a store using + * an update addressing form which will update r1. +@@ -134,7 +166,7 @@ int __kprobes do_page_fault(struct pt_re + * indicate errors in DSISR but can validly be set in SRR1. + */ + if (trap == 0x400) +- error_code &= 0x48200000; ++ error_code &= 0x58200000; + else + is_write = error_code & DSISR_ISSTORE; + #else +@@ -250,7 +282,7 @@ good_area: + * "undefined". Of those that can be set, this is the only + * one which seems bad. + */ +- if (error_code & 0x10000000) ++ if (error_code & DSISR_GUARDED) + /* Guarded storage error. */ + goto bad_area; + #endif /* CONFIG_8xx */ +@@ -265,7 +297,7 @@ good_area: + * processors use the same I/D cache coherency mechanism + * as embedded. + */ +- if (error_code & DSISR_PROTFAULT) ++ if (error_code & (DSISR_PROTFAULT | DSISR_GUARDED)) + goto bad_area; + #endif /* CONFIG_PPC_STD_MMU */ + +@@ -335,6 +367,23 @@ bad_area: + bad_area_nosemaphore: + /* User mode accesses cause a SIGSEGV */ + if (user_mode(regs)) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (mm->pax_flags & MF_PAX_PAGEEXEC) { ++#ifdef CONFIG_PPC_STD_MMU ++ if (is_exec && (error_code & (DSISR_PROTFAULT | DSISR_GUARDED))) { ++#else ++ if (is_exec && regs->nip == address) { ++#endif ++ switch (pax_handle_fetch_fault(regs)) { ++ } ++ ++ pax_report_fault(regs, (void *)regs->nip, (void *)regs->gpr[PT_R1]); ++ do_group_exit(SIGKILL); ++ } ++ } ++#endif ++ + _exception(SIGSEGV, regs, code, address); + return 0; + } +diff -urNp linux-2.6.31.1/arch/powerpc/mm/mmap_64.c linux-2.6.31.1/arch/powerpc/mm/mmap_64.c +--- linux-2.6.31.1/arch/powerpc/mm/mmap_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/mm/mmap_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -99,10 +99,22 @@ void arch_pick_mmap_layout(struct mm_str + */ + if (mmap_is_legacy()) { + mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area; + mm->unmap_area = arch_unmap_area; + } else { + mm->mmap_base = mmap_base(); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base -= mm->delta_mmap + mm->delta_stack; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area_topdown; + mm->unmap_area = arch_unmap_area_topdown; + } +diff -urNp linux-2.6.31.1/arch/powerpc/mm/slice.c linux-2.6.31.1/arch/powerpc/mm/slice.c +--- linux-2.6.31.1/arch/powerpc/mm/slice.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/mm/slice.c 2009-10-01 20:12:42.000000000 -0400 +@@ -426,6 +426,11 @@ unsigned long slice_get_unmapped_area(un + if (fixed && addr > (mm->task_size - len)) + return -EINVAL; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!fixed && (mm->pax_flags & MF_PAX_RANDMMAP)) ++ addr = 0; ++#endif ++ + /* If hint, make sure it matches our alignment restrictions */ + if (!fixed && addr) { + addr = _ALIGN_UP(addr, 1ul << pshift); +diff -urNp linux-2.6.31.1/arch/powerpc/platforms/cell/spufs/file.c linux-2.6.31.1/arch/powerpc/platforms/cell/spufs/file.c +--- linux-2.6.31.1/arch/powerpc/platforms/cell/spufs/file.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/platforms/cell/spufs/file.c 2009-10-01 20:12:42.000000000 -0400 +@@ -147,7 +147,7 @@ static int __fops ## _open(struct inode + __simple_attr_check_format(__fmt, 0ull); \ + return spufs_attr_open(inode, file, __get, __set, __fmt); \ + } \ +-static struct file_operations __fops = { \ ++static const struct file_operations __fops = { \ + .owner = THIS_MODULE, \ + .open = __fops ## _open, \ + .release = spufs_attr_release, \ +@@ -309,7 +309,7 @@ static int spufs_mem_mmap_access(struct + return len; + } + +-static struct vm_operations_struct spufs_mem_mmap_vmops = { ++static const struct vm_operations_struct spufs_mem_mmap_vmops = { + .fault = spufs_mem_mmap_fault, + .access = spufs_mem_mmap_access, + }; +@@ -436,7 +436,7 @@ static int spufs_cntl_mmap_fault(struct + return spufs_ps_fault(vma, vmf, 0x4000, SPUFS_CNTL_MAP_SIZE); + } + +-static struct vm_operations_struct spufs_cntl_mmap_vmops = { ++static const struct vm_operations_struct spufs_cntl_mmap_vmops = { + .fault = spufs_cntl_mmap_fault, + }; + +@@ -1143,7 +1143,7 @@ spufs_signal1_mmap_fault(struct vm_area_ + #endif + } + +-static struct vm_operations_struct spufs_signal1_mmap_vmops = { ++static const struct vm_operations_struct spufs_signal1_mmap_vmops = { + .fault = spufs_signal1_mmap_fault, + }; + +@@ -1279,7 +1279,7 @@ spufs_signal2_mmap_fault(struct vm_area_ + #endif + } + +-static struct vm_operations_struct spufs_signal2_mmap_vmops = { ++static const struct vm_operations_struct spufs_signal2_mmap_vmops = { + .fault = spufs_signal2_mmap_fault, + }; + +@@ -1397,7 +1397,7 @@ spufs_mss_mmap_fault(struct vm_area_stru + return spufs_ps_fault(vma, vmf, 0x0000, SPUFS_MSS_MAP_SIZE); + } + +-static struct vm_operations_struct spufs_mss_mmap_vmops = { ++static const struct vm_operations_struct spufs_mss_mmap_vmops = { + .fault = spufs_mss_mmap_fault, + }; + +@@ -1458,7 +1458,7 @@ spufs_psmap_mmap_fault(struct vm_area_st + return spufs_ps_fault(vma, vmf, 0x0000, SPUFS_PS_MAP_SIZE); + } + +-static struct vm_operations_struct spufs_psmap_mmap_vmops = { ++static const struct vm_operations_struct spufs_psmap_mmap_vmops = { + .fault = spufs_psmap_mmap_fault, + }; + +@@ -1517,7 +1517,7 @@ spufs_mfc_mmap_fault(struct vm_area_stru + return spufs_ps_fault(vma, vmf, 0x3000, SPUFS_MFC_MAP_SIZE); + } + +-static struct vm_operations_struct spufs_mfc_mmap_vmops = { ++static const struct vm_operations_struct spufs_mfc_mmap_vmops = { + .fault = spufs_mfc_mmap_fault, + }; + +diff -urNp linux-2.6.31.1/arch/powerpc/platforms/pseries/dtl.c linux-2.6.31.1/arch/powerpc/platforms/pseries/dtl.c +--- linux-2.6.31.1/arch/powerpc/platforms/pseries/dtl.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/platforms/pseries/dtl.c 2009-10-01 20:12:42.000000000 -0400 +@@ -209,7 +209,7 @@ static ssize_t dtl_file_read(struct file + return n_read * sizeof(struct dtl_entry); + } + +-static struct file_operations dtl_fops = { ++static const struct file_operations dtl_fops = { + .open = dtl_file_open, + .release = dtl_file_release, + .read = dtl_file_read, +diff -urNp linux-2.6.31.1/arch/powerpc/platforms/pseries/hvCall_inst.c linux-2.6.31.1/arch/powerpc/platforms/pseries/hvCall_inst.c +--- linux-2.6.31.1/arch/powerpc/platforms/pseries/hvCall_inst.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/powerpc/platforms/pseries/hvCall_inst.c 2009-10-01 20:12:42.000000000 -0400 +@@ -71,7 +71,7 @@ static int hc_show(struct seq_file *m, v + return 0; + } + +-static struct seq_operations hcall_inst_seq_ops = { ++static const struct seq_operations hcall_inst_seq_ops = { + .start = hc_start, + .next = hc_next, + .stop = hc_stop, +diff -urNp linux-2.6.31.1/arch/s390/hypfs/inode.c linux-2.6.31.1/arch/s390/hypfs/inode.c +--- linux-2.6.31.1/arch/s390/hypfs/inode.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/s390/hypfs/inode.c 2009-10-01 20:12:42.000000000 -0400 +@@ -41,7 +41,7 @@ struct hypfs_sb_info { + + static const struct file_operations hypfs_file_ops; + static struct file_system_type hypfs_type; +-static struct super_operations hypfs_s_ops; ++static const struct super_operations hypfs_s_ops; + + /* start of list of all dentries, which have to be deleted on update */ + static struct dentry *hypfs_last_dentry; +@@ -476,7 +476,7 @@ static struct file_system_type hypfs_typ + .kill_sb = hypfs_kill_super + }; + +-static struct super_operations hypfs_s_ops = { ++static const struct super_operations hypfs_s_ops = { + .statfs = simple_statfs, + .drop_inode = hypfs_drop_inode, + .show_options = hypfs_show_options, +diff -urNp linux-2.6.31.1/arch/s390/include/asm/atomic.h linux-2.6.31.1/arch/s390/include/asm/atomic.h +--- linux-2.6.31.1/arch/s390/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/s390/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -71,19 +71,31 @@ static inline int atomic_read(const atom + return v->counter; + } + ++static inline int atomic_read_unchecked(const atomic_unchecked_t *v) ++{ ++ return atomic_read((const atomic_t *)v); ++} ++ + static inline void atomic_set(atomic_t *v, int i) + { + v->counter = i; + barrier(); + } + ++static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i) ++{ ++ atomic_set((atomic_t *)v, i); ++} ++ + static __inline__ int atomic_add_return(int i, atomic_t * v) + { + return __CS_LOOP(v, i, "ar"); + } + #define atomic_add(_i, _v) atomic_add_return(_i, _v) ++#define atomic_add_unchecked(_i, _v) atomic_add((_i), (atomic_t *)(_v)) + #define atomic_add_negative(_i, _v) (atomic_add_return(_i, _v) < 0) + #define atomic_inc(_v) atomic_add_return(1, _v) ++#define atomic_inc_unchecked(_v) atomic_inc((atomic_t *)(_v)) + #define atomic_inc_return(_v) atomic_add_return(1, _v) + #define atomic_inc_and_test(_v) (atomic_add_return(1, _v) == 0) + +@@ -92,6 +104,7 @@ static __inline__ int atomic_sub_return( + return __CS_LOOP(v, i, "sr"); + } + #define atomic_sub(_i, _v) atomic_sub_return(_i, _v) ++#define atomic_sub_unchecked(_i, _v) atomic_sub((_i), (atomic_t *)(_v)) + #define atomic_sub_and_test(_i, _v) (atomic_sub_return(_i, _v) == 0) + #define atomic_dec(_v) atomic_sub_return(1, _v) + #define atomic_dec_return(_v) atomic_sub_return(1, _v) +diff -urNp linux-2.6.31.1/arch/s390/include/asm/uaccess.h linux-2.6.31.1/arch/s390/include/asm/uaccess.h +--- linux-2.6.31.1/arch/s390/include/asm/uaccess.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/s390/include/asm/uaccess.h 2009-10-01 20:12:42.000000000 -0400 +@@ -232,6 +232,10 @@ static inline unsigned long __must_check + copy_to_user(void __user *to, const void *from, unsigned long n) + { + might_fault(); ++ ++ if ((long)n < 0) ++ return n; ++ + if (access_ok(VERIFY_WRITE, to, n)) + n = __copy_to_user(to, from, n); + return n; +@@ -257,6 +261,9 @@ copy_to_user(void __user *to, const void + static inline unsigned long __must_check + __copy_from_user(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + if (__builtin_constant_p(n) && (n <= 256)) + return uaccess.copy_from_user_small(n, from, to); + else +@@ -283,6 +290,10 @@ static inline unsigned long __must_check + copy_from_user(void *to, const void __user *from, unsigned long n) + { + might_fault(); ++ ++ if ((long)n < 0) ++ return n; ++ + if (access_ok(VERIFY_READ, from, n)) + n = __copy_from_user(to, from, n); + else +diff -urNp linux-2.6.31.1/arch/s390/kernel/module.c linux-2.6.31.1/arch/s390/kernel/module.c +--- linux-2.6.31.1/arch/s390/kernel/module.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/s390/kernel/module.c 2009-10-01 20:12:42.000000000 -0400 +@@ -164,11 +164,11 @@ module_frob_arch_sections(Elf_Ehdr *hdr, + + /* Increase core size by size of got & plt and set start + offsets for got and plt. */ +- me->core_size = ALIGN(me->core_size, 4); +- me->arch.got_offset = me->core_size; +- me->core_size += me->arch.got_size; +- me->arch.plt_offset = me->core_size; +- me->core_size += me->arch.plt_size; ++ me->core_size_rw = ALIGN(me->core_size_rw, 4); ++ me->arch.got_offset = me->core_size_rw; ++ me->core_size_rw += me->arch.got_size; ++ me->arch.plt_offset = me->core_size_rx; ++ me->core_size_rx += me->arch.plt_size; + return 0; + } + +@@ -254,7 +254,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + if (info->got_initialized == 0) { + Elf_Addr *gotent; + +- gotent = me->module_core + me->arch.got_offset + ++ gotent = me->module_core_rw + me->arch.got_offset + + info->got_offset; + *gotent = val; + info->got_initialized = 1; +@@ -278,7 +278,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + else if (r_type == R_390_GOTENT || + r_type == R_390_GOTPLTENT) + *(unsigned int *) loc = +- (val + (Elf_Addr) me->module_core - loc) >> 1; ++ (val + (Elf_Addr) me->module_core_rw - loc) >> 1; + else if (r_type == R_390_GOT64 || + r_type == R_390_GOTPLT64) + *(unsigned long *) loc = val; +@@ -292,7 +292,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + case R_390_PLTOFF64: /* 16 bit offset from GOT to PLT. */ + if (info->plt_initialized == 0) { + unsigned int *ip; +- ip = me->module_core + me->arch.plt_offset + ++ ip = me->module_core_rx + me->arch.plt_offset + + info->plt_offset; + #ifndef CONFIG_64BIT + ip[0] = 0x0d105810; /* basr 1,0; l 1,6(1); br 1 */ +@@ -317,7 +317,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + val - loc + 0xffffUL < 0x1ffffeUL) || + (r_type == R_390_PLT32DBL && + val - loc + 0xffffffffULL < 0x1fffffffeULL))) +- val = (Elf_Addr) me->module_core + ++ val = (Elf_Addr) me->module_core_rx + + me->arch.plt_offset + + info->plt_offset; + val += rela->r_addend - loc; +@@ -339,7 +339,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + case R_390_GOTOFF32: /* 32 bit offset to GOT. */ + case R_390_GOTOFF64: /* 64 bit offset to GOT. */ + val = val + rela->r_addend - +- ((Elf_Addr) me->module_core + me->arch.got_offset); ++ ((Elf_Addr) me->module_core_rw + me->arch.got_offset); + if (r_type == R_390_GOTOFF16) + *(unsigned short *) loc = val; + else if (r_type == R_390_GOTOFF32) +@@ -349,7 +349,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + break; + case R_390_GOTPC: /* 32 bit PC relative offset to GOT. */ + case R_390_GOTPCDBL: /* 32 bit PC rel. off. to GOT shifted by 1. */ +- val = (Elf_Addr) me->module_core + me->arch.got_offset + ++ val = (Elf_Addr) me->module_core_rw + me->arch.got_offset + + rela->r_addend - loc; + if (r_type == R_390_GOTPC) + *(unsigned int *) loc = val; +diff -urNp linux-2.6.31.1/arch/sh/include/asm/atomic.h linux-2.6.31.1/arch/sh/include/asm/atomic.h +--- linux-2.6.31.1/arch/sh/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sh/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -14,7 +14,9 @@ + #define ATOMIC_INIT(i) ( (atomic_t) { (i) } ) + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + #define atomic_set(v,i) ((v)->counter = (i)) ++#define atomic_set_unchecked(v,i) ((v)->counter = (i)) + + #if defined(CONFIG_GUSA_RB) + #include +@@ -43,6 +45,9 @@ + #define atomic_dec_and_test(v) (atomic_sub_return(1, (v)) == 0) + + #define atomic_inc(v) atomic_add(1,(v)) ++#define atomic_inc_unchecked(v) atomic_inc((atomic_t *)(v)) ++#define atomic_add_unchecked(i,v) atomic_add((i),(atomic_t *)(v)) ++#define atomic_sub_unchecked(i,v) atomic_sub((i),(atomic_t *)(v)) + #define atomic_dec(v) atomic_sub(1,(v)) + + #if !defined(CONFIG_GUSA_RB) && !defined(CONFIG_CPU_SH4A) +diff -urNp linux-2.6.31.1/arch/sparc/include/asm/atomic_32.h linux-2.6.31.1/arch/sparc/include/asm/atomic_32.h +--- linux-2.6.31.1/arch/sparc/include/asm/atomic_32.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/include/asm/atomic_32.h 2009-10-01 20:12:42.000000000 -0400 +@@ -24,12 +24,17 @@ extern int atomic_cmpxchg(atomic_t *, in + #define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + extern int atomic_add_unless(atomic_t *, int, int); + extern void atomic_set(atomic_t *, int); ++extern void atomic_set_unchecked(atomic_unchecked_t *, int); + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + + #define atomic_add(i, v) ((void)__atomic_add_return( (int)(i), (v))) ++#define atomic_add_unchecked(i, v) atomic_add((i), (atomic_t *)(v)) + #define atomic_sub(i, v) ((void)__atomic_add_return(-(int)(i), (v))) ++#define atomic_sub_unchecked(i, v) atomic_sub((i), (atomic_t *)(v)) + #define atomic_inc(v) ((void)__atomic_add_return( 1, (v))) ++#define atomic_inc_unchecked(v) atomic_inc((atomic_t *)(v)) + #define atomic_dec(v) ((void)__atomic_add_return( -1, (v))) + + #define atomic_add_return(i, v) (__atomic_add_return( (int)(i), (v))) +diff -urNp linux-2.6.31.1/arch/sparc/include/asm/atomic_64.h linux-2.6.31.1/arch/sparc/include/asm/atomic_64.h +--- linux-2.6.31.1/arch/sparc/include/asm/atomic_64.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/include/asm/atomic_64.h 2009-10-01 20:12:42.000000000 -0400 +@@ -14,14 +14,18 @@ + #define ATOMIC64_INIT(i) { (i) } + + #define atomic_read(v) ((v)->counter) ++#define atomic_read_unchecked(v) ((v)->counter) + #define atomic64_read(v) ((v)->counter) + + #define atomic_set(v, i) (((v)->counter) = i) ++#define atomic_set_unchecked(v, i) (((v)->counter) = i) + #define atomic64_set(v, i) (((v)->counter) = i) + + extern void atomic_add(int, atomic_t *); ++extern void atomic_add_unchecked(int, atomic_unchecked_t *); + extern void atomic64_add(int, atomic64_t *); + extern void atomic_sub(int, atomic_t *); ++extern void atomic_sub_unchecked(int, atomic_unchecked_t *); + extern void atomic64_sub(int, atomic64_t *); + + extern int atomic_add_ret(int, atomic_t *); +@@ -59,6 +63,7 @@ extern int atomic64_sub_ret(int, atomic6 + #define atomic64_dec_and_test(v) (atomic64_sub_ret(1, v) == 0) + + #define atomic_inc(v) atomic_add(1, v) ++#define atomic_inc_unchecked(v) atomic_add_unchecked(1, v) + #define atomic64_inc(v) atomic64_add(1, v) + + #define atomic_dec(v) atomic_sub(1, v) +@@ -72,17 +77,28 @@ extern int atomic64_sub_ret(int, atomic6 + + static inline int atomic_add_unless(atomic_t *v, int a, int u) + { +- int c, old; ++ int c, old, new; + c = atomic_read(v); + for (;;) { +- if (unlikely(c == (u))) ++ if (unlikely(c == u)) + break; +- old = atomic_cmpxchg((v), c, c + (a)); ++ ++ asm volatile("addcc %2, %0, %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "tvs %%icc, 6\n" ++#endif ++ ++ : "=r" (new) ++ : "0" (c), "ir" (a) ++ : "cc"); ++ ++ old = atomic_cmpxchg(v, c, new); + if (likely(old == c)) + break; + c = old; + } +- return c != (u); ++ return c != u; + } + + #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) +@@ -93,17 +109,28 @@ static inline int atomic_add_unless(atom + + static inline int atomic64_add_unless(atomic64_t *v, long a, long u) + { +- long c, old; ++ long c, old, new; + c = atomic64_read(v); + for (;;) { +- if (unlikely(c == (u))) ++ if (unlikely(c == u)) + break; +- old = atomic64_cmpxchg((v), c, c + (a)); ++ ++ asm volatile("addcc %2, %0, %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "tvs %%xcc, 6\n" ++#endif ++ ++ : "=r" (new) ++ : "0" (c), "ir" (a) ++ : "cc"); ++ ++ old = atomic64_cmpxchg(v, c, new); + if (likely(old == c)) + break; + c = old; + } +- return c != (u); ++ return c != u; + } + + #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0) +diff -urNp linux-2.6.31.1/arch/sparc/include/asm/elf_32.h linux-2.6.31.1/arch/sparc/include/asm/elf_32.h +--- linux-2.6.31.1/arch/sparc/include/asm/elf_32.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/include/asm/elf_32.h 2009-10-01 20:12:42.000000000 -0400 +@@ -116,6 +116,13 @@ typedef struct { + + #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE 0x10000UL ++ ++#define PAX_DELTA_MMAP_LEN 16 ++#define PAX_DELTA_STACK_LEN 16 ++#endif ++ + /* This yields a mask that user programs can use to figure out what + instruction set this cpu supports. This can NOT be done in userspace + on Sparc. */ +diff -urNp linux-2.6.31.1/arch/sparc/include/asm/elf_64.h linux-2.6.31.1/arch/sparc/include/asm/elf_64.h +--- linux-2.6.31.1/arch/sparc/include/asm/elf_64.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/include/asm/elf_64.h 2009-10-01 20:12:42.000000000 -0400 +@@ -163,6 +163,12 @@ typedef struct { + #define ELF_ET_DYN_BASE 0x0000010000000000UL + #define COMPAT_ELF_ET_DYN_BASE 0x0000000070000000UL + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT) ? 0x10000UL : 0x100000UL) ++ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT) ? 14 : 28 ) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT) ? 15 : 29 ) ++#endif + + /* This yields a mask that user programs can use to figure out what + instruction set this cpu supports. */ +diff -urNp linux-2.6.31.1/arch/sparc/include/asm/pgtable_32.h linux-2.6.31.1/arch/sparc/include/asm/pgtable_32.h +--- linux-2.6.31.1/arch/sparc/include/asm/pgtable_32.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/include/asm/pgtable_32.h 2009-10-01 20:12:42.000000000 -0400 +@@ -43,6 +43,13 @@ BTFIXUPDEF_SIMM13(user_ptrs_per_pgd) + BTFIXUPDEF_INT(page_none) + BTFIXUPDEF_INT(page_copy) + BTFIXUPDEF_INT(page_readonly) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++BTFIXUPDEF_INT(page_shared_noexec) ++BTFIXUPDEF_INT(page_copy_noexec) ++BTFIXUPDEF_INT(page_readonly_noexec) ++#endif ++ + BTFIXUPDEF_INT(page_kernel) + + #define PMD_SHIFT SUN4C_PMD_SHIFT +@@ -64,6 +71,16 @@ extern pgprot_t PAGE_SHARED; + #define PAGE_COPY __pgprot(BTFIXUP_INT(page_copy)) + #define PAGE_READONLY __pgprot(BTFIXUP_INT(page_readonly)) + ++#ifdef CONFIG_PAX_PAGEEXEC ++extern pgprot_t PAGE_SHARED_NOEXEC; ++# define PAGE_COPY_NOEXEC __pgprot(BTFIXUP_INT(page_copy_noexec)) ++# define PAGE_READONLY_NOEXEC __pgprot(BTFIXUP_INT(page_readonly_noexec)) ++#else ++# define PAGE_SHARED_NOEXEC PAGE_SHARED ++# define PAGE_COPY_NOEXEC PAGE_COPY ++# define PAGE_READONLY_NOEXEC PAGE_READONLY ++#endif ++ + extern unsigned long page_kernel; + + #ifdef MODULE +diff -urNp linux-2.6.31.1/arch/sparc/include/asm/pgtsrmmu.h linux-2.6.31.1/arch/sparc/include/asm/pgtsrmmu.h +--- linux-2.6.31.1/arch/sparc/include/asm/pgtsrmmu.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/include/asm/pgtsrmmu.h 2009-10-01 20:12:42.000000000 -0400 +@@ -115,6 +115,13 @@ + SRMMU_EXEC | SRMMU_REF) + #define SRMMU_PAGE_RDONLY __pgprot(SRMMU_VALID | SRMMU_CACHE | \ + SRMMU_EXEC | SRMMU_REF) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++#define SRMMU_PAGE_SHARED_NOEXEC __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_WRITE | SRMMU_REF) ++#define SRMMU_PAGE_COPY_NOEXEC __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_REF) ++#define SRMMU_PAGE_RDONLY_NOEXEC __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_REF) ++#endif ++ + #define SRMMU_PAGE_KERNEL __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_PRIV | \ + SRMMU_DIRTY | SRMMU_REF) + +diff -urNp linux-2.6.31.1/arch/sparc/include/asm/spinlock_64.h linux-2.6.31.1/arch/sparc/include/asm/spinlock_64.h +--- linux-2.6.31.1/arch/sparc/include/asm/spinlock_64.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/include/asm/spinlock_64.h 2009-10-01 20:12:42.000000000 -0400 +@@ -99,7 +99,12 @@ static void inline __read_lock(raw_rwloc + __asm__ __volatile__ ( + "1: ldsw [%2], %0\n" + " brlz,pn %0, 2f\n" +-"4: add %0, 1, %1\n" ++"4: addcc %0, 1, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++" tvs %%icc, 6\n" ++#endif ++ + " cas [%2], %0, %1\n" + " cmp %0, %1\n" + " bne,pn %%icc, 1b\n" +@@ -112,7 +117,7 @@ static void inline __read_lock(raw_rwloc + " .previous" + : "=&r" (tmp1), "=&r" (tmp2) + : "r" (lock) +- : "memory"); ++ : "memory", "cc"); + } + + static int inline __read_trylock(raw_rwlock_t *lock) +@@ -123,7 +128,12 @@ static int inline __read_trylock(raw_rwl + "1: ldsw [%2], %0\n" + " brlz,a,pn %0, 2f\n" + " mov 0, %0\n" +-" add %0, 1, %1\n" ++" addcc %0, 1, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++" tvs %%icc, 6\n" ++#endif ++ + " cas [%2], %0, %1\n" + " cmp %0, %1\n" + " bne,pn %%icc, 1b\n" +@@ -142,7 +152,12 @@ static void inline __read_unlock(raw_rwl + + __asm__ __volatile__( + "1: lduw [%2], %0\n" +-" sub %0, 1, %1\n" ++" subcc %0, 1, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++" tvs %%icc, 6\n" ++#endif ++ + " cas [%2], %0, %1\n" + " cmp %0, %1\n" + " bne,pn %%xcc, 1b\n" +diff -urNp linux-2.6.31.1/arch/sparc/include/asm/uaccess_32.h linux-2.6.31.1/arch/sparc/include/asm/uaccess_32.h +--- linux-2.6.31.1/arch/sparc/include/asm/uaccess_32.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/include/asm/uaccess_32.h 2009-10-01 20:12:42.000000000 -0400 +@@ -249,27 +249,49 @@ extern unsigned long __copy_user(void __ + + static inline unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) + { +- if (n && __access_ok((unsigned long) to, n)) ++ if ((long)n < 0) ++ return n; ++ ++ if (n && __access_ok((unsigned long) to, n)) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); + return __copy_user(to, (__force void __user *) from, n); +- else ++ } else + return n; + } + + static inline unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); ++ + return __copy_user(to, (__force void __user *) from, n); + } + + static inline unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) + { +- if (n && __access_ok((unsigned long) from, n)) ++ if ((long)n < 0) ++ return n; ++ ++ if (n && __access_ok((unsigned long) from, n)) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); + return __copy_user((__force void __user *) to, from, n); +- else ++ } else + return n; + } + + static inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ + return __copy_user((__force void __user *) to, from, n); + } + +diff -urNp linux-2.6.31.1/arch/sparc/include/asm/uaccess_64.h linux-2.6.31.1/arch/sparc/include/asm/uaccess_64.h +--- linux-2.6.31.1/arch/sparc/include/asm/uaccess_64.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/include/asm/uaccess_64.h 2009-10-01 20:12:42.000000000 -0400 +@@ -212,7 +212,15 @@ extern unsigned long copy_from_user_fixu + static inline unsigned long __must_check + copy_from_user(void *to, const void __user *from, unsigned long size) + { +- unsigned long ret = ___copy_from_user(to, from, size); ++ unsigned long ret; ++ ++ if (unlikely(((long)size > INT_MAX) || ((long)size < 0))) ++ return size; ++ ++ if (!__builtin_constant_p(size)) ++ check_object_size(to, size, false); ++ ++ ret = ___copy_from_user(to, from, size); + + if (unlikely(ret)) + ret = copy_from_user_fixup(to, from, size); +@@ -228,7 +236,15 @@ extern unsigned long copy_to_user_fixup( + static inline unsigned long __must_check + copy_to_user(void __user *to, const void *from, unsigned long size) + { +- unsigned long ret = ___copy_to_user(to, from, size); ++ unsigned long ret; ++ ++ if (unlikely(((long)size > INT_MAX) || ((long)size < 0))) ++ return size; ++ ++ if (!__builtin_constant_p(size)) ++ check_object_size(from, size, true); ++ ++ ret = ___copy_to_user(to, from, size); + + if (unlikely(ret)) + ret = copy_to_user_fixup(to, from, size); +diff -urNp linux-2.6.31.1/arch/sparc/kernel/Makefile linux-2.6.31.1/arch/sparc/kernel/Makefile +--- linux-2.6.31.1/arch/sparc/kernel/Makefile 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/kernel/Makefile 2009-10-01 20:12:42.000000000 -0400 +@@ -3,7 +3,7 @@ + # + + asflags-y := -ansi +-ccflags-y := -Werror ++#ccflags-y := -Werror + + extra-y := head_$(BITS).o + extra-y += init_task.o +diff -urNp linux-2.6.31.1/arch/sparc/kernel/sys_sparc_32.c linux-2.6.31.1/arch/sparc/kernel/sys_sparc_32.c +--- linux-2.6.31.1/arch/sparc/kernel/sys_sparc_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/kernel/sys_sparc_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -56,7 +56,7 @@ unsigned long arch_get_unmapped_area(str + if (ARCH_SUN4C && len > 0x20000000) + return -ENOMEM; + if (!addr) +- addr = TASK_UNMAPPED_BASE; ++ addr = current->mm->mmap_base; + + if (flags & MAP_SHARED) + addr = COLOUR_ALIGN(addr); +diff -urNp linux-2.6.31.1/arch/sparc/kernel/sys_sparc_64.c linux-2.6.31.1/arch/sparc/kernel/sys_sparc_64.c +--- linux-2.6.31.1/arch/sparc/kernel/sys_sparc_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/kernel/sys_sparc_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -125,7 +125,7 @@ unsigned long arch_get_unmapped_area(str + /* We do not accept a shared mapping if it would violate + * cache aliasing constraints. + */ +- if ((flags & MAP_SHARED) && ++ if ((filp || (flags & MAP_SHARED)) && + ((addr - (pgoff << PAGE_SHIFT)) & (SHMLBA - 1))) + return -EINVAL; + return addr; +@@ -140,6 +140,10 @@ unsigned long arch_get_unmapped_area(str + if (filp || (flags & MAP_SHARED)) + do_color_align = 1; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + if (do_color_align) + addr = COLOUR_ALIGN(addr, pgoff); +@@ -153,9 +157,9 @@ unsigned long arch_get_unmapped_area(str + } + + if (len > mm->cached_hole_size) { +- start_addr = addr = mm->free_area_cache; ++ start_addr = addr = mm->free_area_cache; + } else { +- start_addr = addr = TASK_UNMAPPED_BASE; ++ start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + } + +@@ -175,8 +179,8 @@ full_search: + vma = find_vma(mm, VA_EXCLUDE_END); + } + if (unlikely(task_size < addr)) { +- if (start_addr != TASK_UNMAPPED_BASE) { +- start_addr = addr = TASK_UNMAPPED_BASE; ++ if (start_addr != mm->mmap_base) { ++ start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + goto full_search; + } +@@ -216,7 +220,7 @@ arch_get_unmapped_area_topdown(struct fi + /* We do not accept a shared mapping if it would violate + * cache aliasing constraints. + */ +- if ((flags & MAP_SHARED) && ++ if ((filp || (flags & MAP_SHARED)) && + ((addr - (pgoff << PAGE_SHIFT)) & (SHMLBA - 1))) + return -EINVAL; + return addr; +@@ -380,6 +384,12 @@ void arch_pick_mmap_layout(struct mm_str + current->signal->rlim[RLIMIT_STACK].rlim_cur == RLIM_INFINITY || + sysctl_legacy_va_layout) { + mm->mmap_base = TASK_UNMAPPED_BASE + random_factor; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area; + mm->unmap_area = arch_unmap_area; + } else { +@@ -394,6 +404,12 @@ void arch_pick_mmap_layout(struct mm_str + gap = (task_size / 6 * 5); + + mm->mmap_base = PAGE_ALIGN(task_size - gap - random_factor); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base -= mm->delta_mmap + mm->delta_stack; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area_topdown; + mm->unmap_area = arch_unmap_area_topdown; + } +diff -urNp linux-2.6.31.1/arch/sparc/kernel/traps_64.c linux-2.6.31.1/arch/sparc/kernel/traps_64.c +--- linux-2.6.31.1/arch/sparc/kernel/traps_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/kernel/traps_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -93,6 +93,12 @@ void bad_trap(struct pt_regs *regs, long + + lvl -= 0x100; + if (regs->tstate & TSTATE_PRIV) { ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ if (lvl == 6) ++ pax_report_refcount_overflow(regs); ++#endif ++ + sprintf(buffer, "Kernel bad sw trap %lx", lvl); + die_if_kernel(buffer, regs); + } +@@ -111,11 +117,16 @@ void bad_trap(struct pt_regs *regs, long + void bad_trap_tl1(struct pt_regs *regs, long lvl) + { + char buffer[32]; +- ++ + if (notify_die(DIE_TRAP_TL1, "bad trap tl1", regs, + 0, lvl, SIGTRAP) == NOTIFY_STOP) + return; + ++#ifdef CONFIG_PAX_REFCOUNT ++ if (lvl == 6) ++ pax_report_refcount_overflow(regs); ++#endif ++ + dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); + + sprintf (buffer, "Bad trap %lx at tl>0", lvl); +diff -urNp linux-2.6.31.1/arch/sparc/lib/atomic32.c linux-2.6.31.1/arch/sparc/lib/atomic32.c +--- linux-2.6.31.1/arch/sparc/lib/atomic32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/lib/atomic32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -80,6 +80,12 @@ void atomic_set(atomic_t *v, int i) + } + EXPORT_SYMBOL(atomic_set); + ++void atomic_set_unchecked(atomic_unchecked_t *v, int i) ++{ ++ atomic_set((atomic_t *)v, i); ++} ++EXPORT_SYMBOL(atomic_set_unchecked); ++ + unsigned long ___set_bit(unsigned long *addr, unsigned long mask) + { + unsigned long old, flags; +diff -urNp linux-2.6.31.1/arch/sparc/lib/atomic_64.S linux-2.6.31.1/arch/sparc/lib/atomic_64.S +--- linux-2.6.31.1/arch/sparc/lib/atomic_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/lib/atomic_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -18,7 +18,12 @@ + atomic_add: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: lduw [%o1], %g1 +- add %g1, %o0, %g7 ++ addcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, 2f +@@ -28,12 +33,32 @@ atomic_add: /* %o0 = increment, %o1 = at + 2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic_add, .-atomic_add + ++ .globl atomic_add_unchecked ++ .type atomic_add_unchecked,#function ++atomic_add_unchecked: /* %o0 = increment, %o1 = atomic_ptr */ ++ BACKOFF_SETUP(%o2) ++1: lduw [%o1], %g1 ++ add %g1, %o0, %g7 ++ cas [%o1], %g1, %g7 ++ cmp %g1, %g7 ++ bne,pn %icc, 2f ++ nop ++ retl ++ nop ++2: BACKOFF_SPIN(%o2, %o3, 1b) ++ .size atomic_add_unchecked, .-atomic_add_unchecked ++ + .globl atomic_sub + .type atomic_sub,#function + atomic_sub: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: lduw [%o1], %g1 +- sub %g1, %o0, %g7 ++ subcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, 2f +@@ -43,12 +68,32 @@ atomic_sub: /* %o0 = decrement, %o1 = at + 2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic_sub, .-atomic_sub + ++ .globl atomic_sub_unchecked ++ .type atomic_sub_unchecked,#function ++atomic_sub_unchecked: /* %o0 = decrement, %o1 = atomic_ptr */ ++ BACKOFF_SETUP(%o2) ++1: lduw [%o1], %g1 ++ sub %g1, %o0, %g7 ++ cas [%o1], %g1, %g7 ++ cmp %g1, %g7 ++ bne,pn %icc, 2f ++ nop ++ retl ++ nop ++2: BACKOFF_SPIN(%o2, %o3, 1b) ++ .size atomic_sub_unchecked, .-atomic_sub_unchecked ++ + .globl atomic_add_ret + .type atomic_add_ret,#function + atomic_add_ret: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: lduw [%o1], %g1 +- add %g1, %o0, %g7 ++ addcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, 2f +@@ -64,7 +109,12 @@ atomic_add_ret: /* %o0 = increment, %o1 + atomic_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: lduw [%o1], %g1 +- sub %g1, %o0, %g7 ++ subcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, 2f +@@ -80,7 +130,12 @@ atomic_sub_ret: /* %o0 = decrement, %o1 + atomic64_add: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: ldx [%o1], %g1 +- add %g1, %o0, %g7 ++ addcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %xcc, 6 ++#endif ++ + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, 2f +@@ -95,7 +150,12 @@ atomic64_add: /* %o0 = increment, %o1 = + atomic64_sub: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: ldx [%o1], %g1 +- sub %g1, %o0, %g7 ++ subcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %xcc, 6 ++#endif ++ + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, 2f +@@ -110,7 +170,12 @@ atomic64_sub: /* %o0 = decrement, %o1 = + atomic64_add_ret: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: ldx [%o1], %g1 +- add %g1, %o0, %g7 ++ addcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %xcc, 6 ++#endif ++ + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, 2f +@@ -126,7 +191,12 @@ atomic64_add_ret: /* %o0 = increment, %o + atomic64_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: ldx [%o1], %g1 +- sub %g1, %o0, %g7 ++ subcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %xcc, 6 ++#endif ++ + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, 2f +diff -urNp linux-2.6.31.1/arch/sparc/lib/ksyms.c linux-2.6.31.1/arch/sparc/lib/ksyms.c +--- linux-2.6.31.1/arch/sparc/lib/ksyms.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/lib/ksyms.c 2009-10-01 20:12:42.000000000 -0400 +@@ -144,8 +144,10 @@ EXPORT_SYMBOL(__downgrade_write); + + /* Atomic counter implementation. */ + EXPORT_SYMBOL(atomic_add); ++EXPORT_SYMBOL(atomic_add_unchecked); + EXPORT_SYMBOL(atomic_add_ret); + EXPORT_SYMBOL(atomic_sub); ++EXPORT_SYMBOL(atomic_sub_unchecked); + EXPORT_SYMBOL(atomic_sub_ret); + EXPORT_SYMBOL(atomic64_add); + EXPORT_SYMBOL(atomic64_add_ret); +diff -urNp linux-2.6.31.1/arch/sparc/lib/rwsem_64.S linux-2.6.31.1/arch/sparc/lib/rwsem_64.S +--- linux-2.6.31.1/arch/sparc/lib/rwsem_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/lib/rwsem_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -11,7 +11,12 @@ + .globl __down_read + __down_read: + 1: lduw [%o0], %g1 +- add %g1, 1, %g7 ++ addcc %g1, 1, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o0], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, 1b +@@ -33,7 +38,12 @@ __down_read: + .globl __down_read_trylock + __down_read_trylock: + 1: lduw [%o0], %g1 +- add %g1, 1, %g7 ++ addcc %g1, 1, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cmp %g7, 0 + bl,pn %icc, 2f + mov 0, %o1 +@@ -51,7 +61,12 @@ __down_write: + or %g1, %lo(RWSEM_ACTIVE_WRITE_BIAS), %g1 + 1: + lduw [%o0], %g3 +- add %g3, %g1, %g7 ++ addcc %g3, %g1, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o0], %g3, %g7 + cmp %g3, %g7 + bne,pn %icc, 1b +@@ -77,7 +92,12 @@ __down_write_trylock: + cmp %g3, 0 + bne,pn %icc, 2f + mov 0, %o1 +- add %g3, %g1, %g7 ++ addcc %g3, %g1, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o0], %g3, %g7 + cmp %g3, %g7 + bne,pn %icc, 1b +@@ -90,7 +110,12 @@ __down_write_trylock: + __up_read: + 1: + lduw [%o0], %g1 +- sub %g1, 1, %g7 ++ subcc %g1, 1, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o0], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, 1b +@@ -118,7 +143,12 @@ __up_write: + or %g1, %lo(RWSEM_ACTIVE_WRITE_BIAS), %g1 + 1: + lduw [%o0], %g3 +- sub %g3, %g1, %g7 ++ subcc %g3, %g1, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o0], %g3, %g7 + cmp %g3, %g7 + bne,pn %icc, 1b +@@ -143,7 +173,12 @@ __downgrade_write: + or %g1, %lo(RWSEM_WAITING_BIAS), %g1 + 1: + lduw [%o0], %g3 +- sub %g3, %g1, %g7 ++ subcc %g3, %g1, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o0], %g3, %g7 + cmp %g3, %g7 + bne,pn %icc, 1b +diff -urNp linux-2.6.31.1/arch/sparc/Makefile linux-2.6.31.1/arch/sparc/Makefile +--- linux-2.6.31.1/arch/sparc/Makefile 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/Makefile 2009-10-01 20:12:42.000000000 -0400 +@@ -81,7 +81,7 @@ drivers-$(CONFIG_OPROFILE) += arch/sparc + # Export what is needed by arch/sparc/boot/Makefile + export VMLINUX_INIT VMLINUX_MAIN + VMLINUX_INIT := $(head-y) $(init-y) +-VMLINUX_MAIN := $(core-y) kernel/ mm/ fs/ ipc/ security/ crypto/ block/ ++VMLINUX_MAIN := $(core-y) kernel/ mm/ fs/ ipc/ security/ crypto/ block/ grsecurity/ + VMLINUX_MAIN += $(patsubst %/, %/lib.a, $(libs-y)) $(libs-y) + VMLINUX_MAIN += $(drivers-y) $(net-y) + +diff -urNp linux-2.6.31.1/arch/sparc/mm/fault_32.c linux-2.6.31.1/arch/sparc/mm/fault_32.c +--- linux-2.6.31.1/arch/sparc/mm/fault_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/mm/fault_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -21,6 +21,9 @@ + #include + #include + #include ++#include ++#include ++#include + + #include + #include +@@ -167,6 +170,267 @@ static unsigned long compute_si_addr(str + return safe_compute_effective_address(regs, insn); + } + ++#ifdef CONFIG_PAX_PAGEEXEC ++#ifdef CONFIG_PAX_DLRESOLVE ++void pax_emuplt_close(struct vm_area_struct *vma) ++{ ++ vma->vm_mm->call_dl_resolve = 0UL; ++} ++ ++static int pax_emuplt_fault(struct vm_area_struct *vma, struct vm_fault *vmf) ++{ ++ unsigned int *kaddr; ++ ++ vmf->page = alloc_page(GFP_HIGHUSER); ++ if (!vmf->page) ++ return VM_FAULT_OOM; ++ ++ kaddr = kmap(vmf->page); ++ memset(kaddr, 0, PAGE_SIZE); ++ kaddr[0] = 0x9DE3BFA8U; /* save */ ++ flush_dcache_page(vmf->page); ++ kunmap(vmf->page); ++ return VM_FAULT_MAJOR; ++} ++ ++static const struct vm_operations_struct pax_vm_ops = { ++ .close = pax_emuplt_close, ++ .fault = pax_emuplt_fault ++}; ++ ++static int pax_insert_vma(struct vm_area_struct *vma, unsigned long addr) ++{ ++ int ret; ++ ++ vma->vm_mm = current->mm; ++ vma->vm_start = addr; ++ vma->vm_end = addr + PAGE_SIZE; ++ vma->vm_flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC; ++ vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); ++ vma->vm_ops = &pax_vm_ops; ++ ++ ret = insert_vm_struct(current->mm, vma); ++ if (ret) ++ return ret; ++ ++ ++current->mm->total_vm; ++ return 0; ++} ++#endif ++ ++/* ++ * PaX: decide what to do with offenders (regs->pc = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when patched PLT trampoline was detected ++ * 3 when unpatched PLT trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ ++#ifdef CONFIG_PAX_EMUPLT ++ int err; ++ ++ do { /* PaX: patched PLT emulation #1 */ ++ unsigned int sethi1, sethi2, jmpl; ++ ++ err = get_user(sethi1, (unsigned int *)regs->pc); ++ err |= get_user(sethi2, (unsigned int *)(regs->pc+4)); ++ err |= get_user(jmpl, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi1 & 0xFFC00000U) == 0x03000000U && ++ (sethi2 & 0xFFC00000U) == 0x03000000U && ++ (jmpl & 0xFFFFE000U) == 0x81C06000U) ++ { ++ unsigned int addr; ++ ++ regs->u_regs[UREG_G1] = (sethi2 & 0x003FFFFFU) << 10; ++ addr = regs->u_regs[UREG_G1]; ++ addr += (((jmpl | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U); ++ regs->pc = addr; ++ regs->npc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ { /* PaX: patched PLT emulation #2 */ ++ unsigned int ba; ++ ++ err = get_user(ba, (unsigned int *)regs->pc); ++ ++ if (!err && (ba & 0xFFC00000U) == 0x30800000U) { ++ unsigned int addr; ++ ++ addr = regs->pc + ((((ba | 0xFFC00000U) ^ 0x00200000U) + 0x00200000U) << 2); ++ regs->pc = addr; ++ regs->npc = addr+4; ++ return 2; ++ } ++ } ++ ++ do { /* PaX: patched PLT emulation #3 */ ++ unsigned int sethi, jmpl, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->pc); ++ err |= get_user(jmpl, (unsigned int *)(regs->pc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ (jmpl & 0xFFFFE000U) == 0x81C06000U && ++ nop == 0x01000000U) ++ { ++ unsigned int addr; ++ ++ addr = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G1] = addr; ++ addr += (((jmpl | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U); ++ regs->pc = addr; ++ regs->npc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: unpatched PLT emulation step 1 */ ++ unsigned int sethi, ba, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->pc); ++ err |= get_user(ba, (unsigned int *)(regs->pc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ ((ba & 0xFFC00000U) == 0x30800000U || (ba & 0xFFF80000U) == 0x30680000U) && ++ nop == 0x01000000U) ++ { ++ unsigned int addr, save, call; ++ ++ if ((ba & 0xFFC00000U) == 0x30800000U) ++ addr = regs->pc + 4 + ((((ba | 0xFFC00000U) ^ 0x00200000U) + 0x00200000U) << 2); ++ else ++ addr = regs->pc + 4 + ((((ba | 0xFFF80000U) ^ 0x00040000U) + 0x00040000U) << 2); ++ ++ err = get_user(save, (unsigned int *)addr); ++ err |= get_user(call, (unsigned int *)(addr+4)); ++ err |= get_user(nop, (unsigned int *)(addr+8)); ++ if (err) ++ break; ++ ++#ifdef CONFIG_PAX_DLRESOLVE ++ if (save == 0x9DE3BFA8U && ++ (call & 0xC0000000U) == 0x40000000U && ++ nop == 0x01000000U) ++ { ++ struct vm_area_struct *vma; ++ unsigned long call_dl_resolve; ++ ++ down_read(¤t->mm->mmap_sem); ++ call_dl_resolve = current->mm->call_dl_resolve; ++ up_read(¤t->mm->mmap_sem); ++ if (likely(call_dl_resolve)) ++ goto emulate; ++ ++ vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); ++ ++ down_write(¤t->mm->mmap_sem); ++ if (current->mm->call_dl_resolve) { ++ call_dl_resolve = current->mm->call_dl_resolve; ++ up_write(¤t->mm->mmap_sem); ++ if (vma) ++ kmem_cache_free(vm_area_cachep, vma); ++ goto emulate; ++ } ++ ++ call_dl_resolve = get_unmapped_area(NULL, 0UL, PAGE_SIZE, 0UL, MAP_PRIVATE); ++ if (!vma || (call_dl_resolve & ~PAGE_MASK)) { ++ up_write(¤t->mm->mmap_sem); ++ if (vma) ++ kmem_cache_free(vm_area_cachep, vma); ++ return 1; ++ } ++ ++ if (pax_insert_vma(vma, call_dl_resolve)) { ++ up_write(¤t->mm->mmap_sem); ++ kmem_cache_free(vm_area_cachep, vma); ++ return 1; ++ } ++ ++ current->mm->call_dl_resolve = call_dl_resolve; ++ up_write(¤t->mm->mmap_sem); ++ ++emulate: ++ regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; ++ regs->pc = call_dl_resolve; ++ regs->npc = addr+4; ++ return 3; ++ } ++#endif ++ ++ /* PaX: glibc 2.4+ generates sethi/jmpl instead of save/call */ ++ if ((save & 0xFFC00000U) == 0x05000000U && ++ (call & 0xFFFFE000U) == 0x85C0A000U && ++ nop == 0x01000000U) ++ { ++ regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G2] = addr + 4; ++ addr = (save & 0x003FFFFFU) << 10; ++ addr += (((call | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U); ++ regs->pc = addr; ++ regs->npc = addr+4; ++ return 3; ++ } ++ } ++ } while (0); ++ ++ do { /* PaX: unpatched PLT emulation step 2 */ ++ unsigned int save, call, nop; ++ ++ err = get_user(save, (unsigned int *)(regs->pc-4)); ++ err |= get_user(call, (unsigned int *)regs->pc); ++ err |= get_user(nop, (unsigned int *)(regs->pc+4)); ++ if (err) ++ break; ++ ++ if (save == 0x9DE3BFA8U && ++ (call & 0xC0000000U) == 0x40000000U && ++ nop == 0x01000000U) ++ { ++ unsigned int dl_resolve = regs->pc + ((((call | 0xC0000000U) ^ 0x20000000U) + 0x20000000U) << 2); ++ ++ regs->u_regs[UREG_RETPC] = regs->pc; ++ regs->pc = dl_resolve; ++ regs->npc = dl_resolve+4; ++ return 3; ++ } ++ } while (0); ++#endif ++ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + asmlinkage void do_sparc_fault(struct pt_regs *regs, int text_fault, int write, + unsigned long address) + { +@@ -231,6 +495,24 @@ good_area: + if(!(vma->vm_flags & VM_WRITE)) + goto bad_area; + } else { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && text_fault && !(vma->vm_flags & VM_EXEC)) { ++ up_read(&mm->mmap_sem); ++ switch (pax_handle_fetch_fault(regs)) { ++ ++#ifdef CONFIG_PAX_EMUPLT ++ case 2: ++ case 3: ++ return; ++#endif ++ ++ } ++ pax_report_fault(regs, (void *)regs->pc, (void *)regs->u_regs[UREG_FP]); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ + /* Allow reads even for write-only mappings */ + if(!(vma->vm_flags & (VM_READ | VM_EXEC))) + goto bad_area; +diff -urNp linux-2.6.31.1/arch/sparc/mm/fault_64.c linux-2.6.31.1/arch/sparc/mm/fault_64.c +--- linux-2.6.31.1/arch/sparc/mm/fault_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/mm/fault_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -20,6 +20,9 @@ + #include + #include + #include ++#include ++#include ++#include + + #include + #include +@@ -249,6 +252,416 @@ static void noinline bogus_32bit_fault_a + show_regs(regs); + } + ++#ifdef CONFIG_PAX_PAGEEXEC ++#ifdef CONFIG_PAX_DLRESOLVE ++static void pax_emuplt_close(struct vm_area_struct *vma) ++{ ++ vma->vm_mm->call_dl_resolve = 0UL; ++} ++ ++static int pax_emuplt_fault(struct vm_area_struct *vma, struct vm_fault *vmf) ++{ ++ unsigned int *kaddr; ++ ++ vmf->page = alloc_page(GFP_HIGHUSER); ++ if (!vmf->page) ++ return VM_FAULT_OOM; ++ ++ kaddr = kmap(vmf->page); ++ memset(kaddr, 0, PAGE_SIZE); ++ kaddr[0] = 0x9DE3BFA8U; /* save */ ++ flush_dcache_page(vmf->page); ++ kunmap(vmf->page); ++ return VM_FAULT_MAJOR; ++} ++ ++static const struct vm_operations_struct pax_vm_ops = { ++ .close = pax_emuplt_close, ++ .fault = pax_emuplt_fault ++}; ++ ++static int pax_insert_vma(struct vm_area_struct *vma, unsigned long addr) ++{ ++ int ret; ++ ++ vma->vm_mm = current->mm; ++ vma->vm_start = addr; ++ vma->vm_end = addr + PAGE_SIZE; ++ vma->vm_flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC; ++ vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); ++ vma->vm_ops = &pax_vm_ops; ++ ++ ret = insert_vm_struct(current->mm, vma); ++ if (ret) ++ return ret; ++ ++ ++current->mm->total_vm; ++ return 0; ++} ++#endif ++ ++/* ++ * PaX: decide what to do with offenders (regs->tpc = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when patched PLT trampoline was detected ++ * 3 when unpatched PLT trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ ++#ifdef CONFIG_PAX_EMUPLT ++ int err; ++ ++ do { /* PaX: patched PLT emulation #1 */ ++ unsigned int sethi1, sethi2, jmpl; ++ ++ err = get_user(sethi1, (unsigned int *)regs->tpc); ++ err |= get_user(sethi2, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(jmpl, (unsigned int *)(regs->tpc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi1 & 0xFFC00000U) == 0x03000000U && ++ (sethi2 & 0xFFC00000U) == 0x03000000U && ++ (jmpl & 0xFFFFE000U) == 0x81C06000U) ++ { ++ unsigned long addr; ++ ++ regs->u_regs[UREG_G1] = (sethi2 & 0x003FFFFFU) << 10; ++ addr = regs->u_regs[UREG_G1]; ++ addr += (((jmpl | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ { /* PaX: patched PLT emulation #2 */ ++ unsigned int ba; ++ ++ err = get_user(ba, (unsigned int *)regs->tpc); ++ ++ if (!err && (ba & 0xFFC00000U) == 0x30800000U) { ++ unsigned long addr; ++ ++ addr = regs->tpc + ((((ba | 0xFFFFFFFFFFC00000UL) ^ 0x00200000UL) + 0x00200000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } ++ ++ do { /* PaX: patched PLT emulation #3 */ ++ unsigned int sethi, jmpl, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->tpc); ++ err |= get_user(jmpl, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ (jmpl & 0xFFFFE000U) == 0x81C06000U && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ ++ addr = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G1] = addr; ++ addr += (((jmpl | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: patched PLT emulation #4 */ ++ unsigned int mov1, call, mov2; ++ ++ err = get_user(mov1, (unsigned int *)regs->tpc); ++ err |= get_user(call, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(mov2, (unsigned int *)(regs->tpc+8)); ++ ++ if (err) ++ break; ++ ++ if (mov1 == 0x8210000FU && ++ (call & 0xC0000000U) == 0x40000000U && ++ mov2 == 0x9E100001U) ++ { ++ unsigned long addr; ++ ++ regs->u_regs[UREG_G1] = regs->u_regs[UREG_RETPC]; ++ addr = regs->tpc + 4 + ((((call | 0xFFFFFFFFC0000000UL) ^ 0x20000000UL) + 0x20000000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: patched PLT emulation #5 */ ++ unsigned int sethi1, sethi2, or1, or2, sllx, jmpl, nop; ++ ++ err = get_user(sethi1, (unsigned int *)regs->tpc); ++ err |= get_user(sethi2, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(or1, (unsigned int *)(regs->tpc+8)); ++ err |= get_user(or2, (unsigned int *)(regs->tpc+12)); ++ err |= get_user(sllx, (unsigned int *)(regs->tpc+16)); ++ err |= get_user(jmpl, (unsigned int *)(regs->tpc+20)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+24)); ++ ++ if (err) ++ break; ++ ++ if ((sethi1 & 0xFFC00000U) == 0x03000000U && ++ (sethi2 & 0xFFC00000U) == 0x0B000000U && ++ (or1 & 0xFFFFE000U) == 0x82106000U && ++ (or2 & 0xFFFFE000U) == 0x8A116000U && ++ sllx == 0x83287020 && ++ jmpl == 0x81C04005U && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ ++ regs->u_regs[UREG_G1] = ((sethi1 & 0x003FFFFFU) << 10) | (or1 & 0x000003FFU); ++ regs->u_regs[UREG_G1] <<= 32; ++ regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or2 & 0x000003FFU); ++ addr = regs->u_regs[UREG_G1] + regs->u_regs[UREG_G5]; ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: patched PLT emulation #6 */ ++ unsigned int sethi1, sethi2, sllx, or, jmpl, nop; ++ ++ err = get_user(sethi1, (unsigned int *)regs->tpc); ++ err |= get_user(sethi2, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(sllx, (unsigned int *)(regs->tpc+8)); ++ err |= get_user(or, (unsigned int *)(regs->tpc+12)); ++ err |= get_user(jmpl, (unsigned int *)(regs->tpc+16)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+20)); ++ ++ if (err) ++ break; ++ ++ if ((sethi1 & 0xFFC00000U) == 0x03000000U && ++ (sethi2 & 0xFFC00000U) == 0x0B000000U && ++ sllx == 0x83287020 && ++ (or & 0xFFFFE000U) == 0x8A116000U && ++ jmpl == 0x81C04005U && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ ++ regs->u_regs[UREG_G1] = (sethi1 & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G1] <<= 32; ++ regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or & 0x3FFU); ++ addr = regs->u_regs[UREG_G1] + regs->u_regs[UREG_G5]; ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: unpatched PLT emulation step 1 */ ++ unsigned int sethi, ba, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->tpc); ++ err |= get_user(ba, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ ((ba & 0xFFC00000U) == 0x30800000U || (ba & 0xFFF80000U) == 0x30680000U) && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ unsigned int save, call; ++ ++ if ((ba & 0xFFC00000U) == 0x30800000U) ++ addr = regs->tpc + 4 + ((((ba | 0xFFFFFFFFFFC00000UL) ^ 0x00200000UL) + 0x00200000UL) << 2); ++ else ++ addr = regs->tpc + 4 + ((((ba | 0xFFFFFFFFFFF80000UL) ^ 0x00040000UL) + 0x00040000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ err = get_user(save, (unsigned int *)addr); ++ err |= get_user(call, (unsigned int *)(addr+4)); ++ err |= get_user(nop, (unsigned int *)(addr+8)); ++ if (err) ++ break; ++ ++#ifdef CONFIG_PAX_DLRESOLVE ++ if (save == 0x9DE3BFA8U && ++ (call & 0xC0000000U) == 0x40000000U && ++ nop == 0x01000000U) ++ { ++ struct vm_area_struct *vma; ++ unsigned long call_dl_resolve; ++ ++ down_read(¤t->mm->mmap_sem); ++ call_dl_resolve = current->mm->call_dl_resolve; ++ up_read(¤t->mm->mmap_sem); ++ if (likely(call_dl_resolve)) ++ goto emulate; ++ ++ vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); ++ ++ down_write(¤t->mm->mmap_sem); ++ if (current->mm->call_dl_resolve) { ++ call_dl_resolve = current->mm->call_dl_resolve; ++ up_write(¤t->mm->mmap_sem); ++ if (vma) ++ kmem_cache_free(vm_area_cachep, vma); ++ goto emulate; ++ } ++ ++ call_dl_resolve = get_unmapped_area(NULL, 0UL, PAGE_SIZE, 0UL, MAP_PRIVATE); ++ if (!vma || (call_dl_resolve & ~PAGE_MASK)) { ++ up_write(¤t->mm->mmap_sem); ++ if (vma) ++ kmem_cache_free(vm_area_cachep, vma); ++ return 1; ++ } ++ ++ if (pax_insert_vma(vma, call_dl_resolve)) { ++ up_write(¤t->mm->mmap_sem); ++ kmem_cache_free(vm_area_cachep, vma); ++ return 1; ++ } ++ ++ current->mm->call_dl_resolve = call_dl_resolve; ++ up_write(¤t->mm->mmap_sem); ++ ++emulate: ++ regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; ++ regs->tpc = call_dl_resolve; ++ regs->tnpc = addr+4; ++ return 3; ++ } ++#endif ++ ++ /* PaX: glibc 2.4+ generates sethi/jmpl instead of save/call */ ++ if ((save & 0xFFC00000U) == 0x05000000U && ++ (call & 0xFFFFE000U) == 0x85C0A000U && ++ nop == 0x01000000U) ++ { ++ regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G2] = addr + 4; ++ addr = (save & 0x003FFFFFU) << 10; ++ addr += (((call | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 3; ++ } ++ } ++ } while (0); ++ ++#ifdef CONFIG_PAX_DLRESOLVE ++ do { /* PaX: unpatched PLT emulation step 2 */ ++ unsigned int save, call, nop; ++ ++ err = get_user(save, (unsigned int *)(regs->tpc-4)); ++ err |= get_user(call, (unsigned int *)regs->tpc); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+4)); ++ if (err) ++ break; ++ ++ if (save == 0x9DE3BFA8U && ++ (call & 0xC0000000U) == 0x40000000U && ++ nop == 0x01000000U) ++ { ++ unsigned long dl_resolve = regs->tpc + ((((call | 0xFFFFFFFFC0000000UL) ^ 0x20000000UL) + 0x20000000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ dl_resolve &= 0xFFFFFFFFUL; ++ ++ regs->u_regs[UREG_RETPC] = regs->tpc; ++ regs->tpc = dl_resolve; ++ regs->tnpc = dl_resolve+4; ++ return 3; ++ } ++ } while (0); ++#endif ++ ++ do { /* PaX: patched PLT emulation #7, must be AFTER the unpatched PLT emulation */ ++ unsigned int sethi, ba, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->tpc); ++ err |= get_user(ba, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ (ba & 0xFFF00000U) == 0x30600000U && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ ++ addr = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G1] = addr; ++ addr = regs->tpc + ((((ba | 0xFFFFFFFFFFF80000UL) ^ 0x00040000UL) + 0x00040000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++#endif ++ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) + { + struct mm_struct *mm = current->mm; +@@ -315,6 +728,29 @@ asmlinkage void __kprobes do_sparc64_fau + if (!vma) + goto bad_area; + ++#ifdef CONFIG_PAX_PAGEEXEC ++ /* PaX: detect ITLB misses on non-exec pages */ ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && vma->vm_start <= address && ++ !(vma->vm_flags & VM_EXEC) && (fault_code & FAULT_CODE_ITLB)) ++ { ++ if (address != regs->tpc) ++ goto good_area; ++ ++ up_read(&mm->mmap_sem); ++ switch (pax_handle_fetch_fault(regs)) { ++ ++#ifdef CONFIG_PAX_EMUPLT ++ case 2: ++ case 3: ++ return; ++#endif ++ ++ } ++ pax_report_fault(regs, (void *)regs->tpc, (void *)(regs->u_regs[UREG_FP] + STACK_BIAS)); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ + /* Pure DTLB misses do not tell us whether the fault causing + * load/store/atomic was a write or not, it only says that there + * was no match. So in such a case we (carefully) read the +diff -urNp linux-2.6.31.1/arch/sparc/mm/init_32.c linux-2.6.31.1/arch/sparc/mm/init_32.c +--- linux-2.6.31.1/arch/sparc/mm/init_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/mm/init_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -316,6 +316,9 @@ extern void device_scan(void); + pgprot_t PAGE_SHARED __read_mostly; + EXPORT_SYMBOL(PAGE_SHARED); + ++pgprot_t PAGE_SHARED_NOEXEC __read_mostly; ++EXPORT_SYMBOL(PAGE_SHARED_NOEXEC); ++ + void __init paging_init(void) + { + switch(sparc_cpu_model) { +@@ -341,17 +344,17 @@ void __init paging_init(void) + + /* Initialize the protection map with non-constant, MMU dependent values. */ + protection_map[0] = PAGE_NONE; +- protection_map[1] = PAGE_READONLY; +- protection_map[2] = PAGE_COPY; +- protection_map[3] = PAGE_COPY; ++ protection_map[1] = PAGE_READONLY_NOEXEC; ++ protection_map[2] = PAGE_COPY_NOEXEC; ++ protection_map[3] = PAGE_COPY_NOEXEC; + protection_map[4] = PAGE_READONLY; + protection_map[5] = PAGE_READONLY; + protection_map[6] = PAGE_COPY; + protection_map[7] = PAGE_COPY; + protection_map[8] = PAGE_NONE; +- protection_map[9] = PAGE_READONLY; +- protection_map[10] = PAGE_SHARED; +- protection_map[11] = PAGE_SHARED; ++ protection_map[9] = PAGE_READONLY_NOEXEC; ++ protection_map[10] = PAGE_SHARED_NOEXEC; ++ protection_map[11] = PAGE_SHARED_NOEXEC; + protection_map[12] = PAGE_READONLY; + protection_map[13] = PAGE_READONLY; + protection_map[14] = PAGE_SHARED; +diff -urNp linux-2.6.31.1/arch/sparc/mm/Makefile linux-2.6.31.1/arch/sparc/mm/Makefile +--- linux-2.6.31.1/arch/sparc/mm/Makefile 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/mm/Makefile 2009-10-01 20:12:42.000000000 -0400 +@@ -2,7 +2,7 @@ + # + + asflags-y := -ansi +-ccflags-y := -Werror ++#ccflags-y := -Werror + + obj-$(CONFIG_SPARC64) += ultra.o tlb.o tsb.o + obj-y += fault_$(BITS).o +diff -urNp linux-2.6.31.1/arch/sparc/mm/srmmu.c linux-2.6.31.1/arch/sparc/mm/srmmu.c +--- linux-2.6.31.1/arch/sparc/mm/srmmu.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/sparc/mm/srmmu.c 2009-10-01 20:12:42.000000000 -0400 +@@ -2149,6 +2149,13 @@ void __init ld_mmu_srmmu(void) + PAGE_SHARED = pgprot_val(SRMMU_PAGE_SHARED); + BTFIXUPSET_INT(page_copy, pgprot_val(SRMMU_PAGE_COPY)); + BTFIXUPSET_INT(page_readonly, pgprot_val(SRMMU_PAGE_RDONLY)); ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ PAGE_SHARED_NOEXEC = pgprot_val(SRMMU_PAGE_SHARED_NOEXEC); ++ BTFIXUPSET_INT(page_copy_noexec, pgprot_val(SRMMU_PAGE_COPY_NOEXEC)); ++ BTFIXUPSET_INT(page_readonly_noexec, pgprot_val(SRMMU_PAGE_RDONLY_NOEXEC)); ++#endif ++ + BTFIXUPSET_INT(page_kernel, pgprot_val(SRMMU_PAGE_KERNEL)); + page_kernel = pgprot_val(SRMMU_PAGE_KERNEL); + +diff -urNp linux-2.6.31.1/arch/um/include/asm/kmap_types.h linux-2.6.31.1/arch/um/include/asm/kmap_types.h +--- linux-2.6.31.1/arch/um/include/asm/kmap_types.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/um/include/asm/kmap_types.h 2009-10-01 20:12:42.000000000 -0400 +@@ -23,6 +23,7 @@ enum km_type { + KM_IRQ1, + KM_SOFTIRQ0, + KM_SOFTIRQ1, ++ KM_CLEARPAGE, + KM_TYPE_NR + }; + +diff -urNp linux-2.6.31.1/arch/um/include/asm/page.h linux-2.6.31.1/arch/um/include/asm/page.h +--- linux-2.6.31.1/arch/um/include/asm/page.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/um/include/asm/page.h 2009-10-01 20:12:42.000000000 -0400 +@@ -14,6 +14,9 @@ + #define PAGE_SIZE (_AC(1, UL) << PAGE_SHIFT) + #define PAGE_MASK (~(PAGE_SIZE-1)) + ++#define ktla_ktva(addr) (addr) ++#define ktva_ktla(addr) (addr) ++ + #ifndef __ASSEMBLY__ + + struct page; +diff -urNp linux-2.6.31.1/arch/um/sys-i386/syscalls.c linux-2.6.31.1/arch/um/sys-i386/syscalls.c +--- linux-2.6.31.1/arch/um/sys-i386/syscalls.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/um/sys-i386/syscalls.c 2009-10-01 20:12:42.000000000 -0400 +@@ -11,6 +11,21 @@ + #include "asm/uaccess.h" + #include "asm/unistd.h" + ++int i386_mmap_check(unsigned long addr, unsigned long len, unsigned long flags) ++{ ++ unsigned long pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (current->mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ if (len > pax_task_size || addr > pax_task_size - len) ++ return -EINVAL; ++ ++ return 0; ++} ++ + /* + * Perform the select(nd, in, out, ex, tv) and mmap() system + * calls. Linux/i386 didn't use to be able to handle more than +diff -urNp linux-2.6.31.1/arch/x86/boot/bitops.h linux-2.6.31.1/arch/x86/boot/bitops.h +--- linux-2.6.31.1/arch/x86/boot/bitops.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/bitops.h 2009-10-01 20:12:42.000000000 -0400 +@@ -26,7 +26,7 @@ static inline int variable_test_bit(int + u8 v; + const u32 *p = (const u32 *)addr; + +- asm("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr)); ++ asm volatile("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr)); + return v; + } + +@@ -37,7 +37,7 @@ static inline int variable_test_bit(int + + static inline void set_bit(int nr, void *addr) + { +- asm("btsl %1,%0" : "+m" (*(u32 *)addr) : "Ir" (nr)); ++ asm volatile("btsl %1,%0" : "+m" (*(u32 *)addr) : "Ir" (nr)); + } + + #endif /* BOOT_BITOPS_H */ +diff -urNp linux-2.6.31.1/arch/x86/boot/boot.h linux-2.6.31.1/arch/x86/boot/boot.h +--- linux-2.6.31.1/arch/x86/boot/boot.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/boot.h 2009-10-01 20:12:42.000000000 -0400 +@@ -82,7 +82,7 @@ static inline void io_delay(void) + static inline u16 ds(void) + { + u16 seg; +- asm("movw %%ds,%0" : "=rm" (seg)); ++ asm volatile("movw %%ds,%0" : "=rm" (seg)); + return seg; + } + +@@ -178,7 +178,7 @@ static inline void wrgs32(u32 v, addr_t + static inline int memcmp(const void *s1, const void *s2, size_t len) + { + u8 diff; +- asm("repe; cmpsb; setnz %0" ++ asm volatile("repe; cmpsb; setnz %0" + : "=qm" (diff), "+D" (s1), "+S" (s2), "+c" (len)); + return diff; + } +diff -urNp linux-2.6.31.1/arch/x86/boot/compressed/head_32.S linux-2.6.31.1/arch/x86/boot/compressed/head_32.S +--- linux-2.6.31.1/arch/x86/boot/compressed/head_32.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/compressed/head_32.S 2009-10-01 20:12:42.000000000 -0400 +@@ -75,7 +75,7 @@ ENTRY(startup_32) + notl %eax + andl %eax, %ebx + #else +- movl $LOAD_PHYSICAL_ADDR, %ebx ++ movl $____LOAD_PHYSICAL_ADDR, %ebx + #endif + + /* Target address to relocate to for decompression */ +@@ -148,7 +148,7 @@ relocated: + * and where it was actually loaded. + */ + movl %ebp, %ebx +- subl $LOAD_PHYSICAL_ADDR, %ebx ++ subl $____LOAD_PHYSICAL_ADDR, %ebx + jz 2f /* Nothing to be done if loaded at compiled addr. */ + /* + * Process relocations. +@@ -156,8 +156,7 @@ relocated: + + 1: subl $4, %edi + movl (%edi), %ecx +- testl %ecx, %ecx +- jz 2f ++ jecxz 2f + addl %ebx, -__PAGE_OFFSET(%ebx, %ecx) + jmp 1b + 2: +diff -urNp linux-2.6.31.1/arch/x86/boot/compressed/head_64.S linux-2.6.31.1/arch/x86/boot/compressed/head_64.S +--- linux-2.6.31.1/arch/x86/boot/compressed/head_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/compressed/head_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -90,7 +90,7 @@ ENTRY(startup_32) + notl %eax + andl %eax, %ebx + #else +- movl $LOAD_PHYSICAL_ADDR, %ebx ++ movl $____LOAD_PHYSICAL_ADDR, %ebx + #endif + + /* Target address to relocate to for decompression */ +@@ -233,7 +233,7 @@ ENTRY(startup_64) + notq %rax + andq %rax, %rbp + #else +- movq $LOAD_PHYSICAL_ADDR, %rbp ++ movq $____LOAD_PHYSICAL_ADDR, %rbp + #endif + + /* Target address to relocate to for decompression */ +diff -urNp linux-2.6.31.1/arch/x86/boot/compressed/misc.c linux-2.6.31.1/arch/x86/boot/compressed/misc.c +--- linux-2.6.31.1/arch/x86/boot/compressed/misc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/compressed/misc.c 2009-10-01 20:12:42.000000000 -0400 +@@ -288,7 +288,7 @@ static void parse_elf(void *output) + case PT_LOAD: + #ifdef CONFIG_RELOCATABLE + dest = output; +- dest += (phdr->p_paddr - LOAD_PHYSICAL_ADDR); ++ dest += (phdr->p_paddr - ____LOAD_PHYSICAL_ADDR); + #else + dest = (void *)(phdr->p_paddr); + #endif +@@ -335,7 +335,7 @@ asmlinkage void decompress_kernel(void * + error("Destination address too large"); + #endif + #ifndef CONFIG_RELOCATABLE +- if ((unsigned long)output != LOAD_PHYSICAL_ADDR) ++ if ((unsigned long)output != ____LOAD_PHYSICAL_ADDR) + error("Wrong destination address"); + #endif + +diff -urNp linux-2.6.31.1/arch/x86/boot/compressed/mkpiggy.c linux-2.6.31.1/arch/x86/boot/compressed/mkpiggy.c +--- linux-2.6.31.1/arch/x86/boot/compressed/mkpiggy.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/compressed/mkpiggy.c 2009-10-01 20:12:42.000000000 -0400 +@@ -74,7 +74,7 @@ int main(int argc, char *argv[]) + + offs = (olen > ilen) ? olen - ilen : 0; + offs += olen >> 12; /* Add 8 bytes for each 32K block */ +- offs += 32*1024 + 18; /* Add 32K + 18 bytes slack */ ++ offs += 64*1024; /* Add 64K bytes slack */ + offs = (offs+4095) & ~4095; /* Round to a 4K boundary */ + + printf(".section \".rodata.compressed\",\"a\",@progbits\n"); +diff -urNp linux-2.6.31.1/arch/x86/boot/compressed/relocs.c linux-2.6.31.1/arch/x86/boot/compressed/relocs.c +--- linux-2.6.31.1/arch/x86/boot/compressed/relocs.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/compressed/relocs.c 2009-10-01 20:12:42.000000000 -0400 +@@ -10,8 +10,11 @@ + #define USE_BSD + #include + ++#include "../../../../include/linux/autoconf.h" ++ + #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) + static Elf32_Ehdr ehdr; ++static Elf32_Phdr *phdr; + static unsigned long reloc_count, reloc_idx; + static unsigned long *relocs; + +@@ -37,7 +40,7 @@ static const char* safe_abs_relocs[] = { + + static int is_safe_abs_reloc(const char* sym_name) + { +- int i; ++ unsigned int i; + + for (i = 0; i < ARRAY_SIZE(safe_abs_relocs); i++) { + if (!strcmp(sym_name, safe_abs_relocs[i])) +@@ -245,9 +248,39 @@ static void read_ehdr(FILE *fp) + } + } + ++static void read_phdrs(FILE *fp) ++{ ++ unsigned int i; ++ ++ phdr = calloc(ehdr.e_phnum, sizeof(Elf32_Phdr)); ++ if (!phdr) { ++ die("Unable to allocate %d program headers\n", ++ ehdr.e_phnum); ++ } ++ if (fseek(fp, ehdr.e_phoff, SEEK_SET) < 0) { ++ die("Seek to %d failed: %s\n", ++ ehdr.e_phoff, strerror(errno)); ++ } ++ if (fread(phdr, sizeof(*phdr), ehdr.e_phnum, fp) != ehdr.e_phnum) { ++ die("Cannot read ELF program headers: %s\n", ++ strerror(errno)); ++ } ++ for(i = 0; i < ehdr.e_phnum; i++) { ++ phdr[i].p_type = elf32_to_cpu(phdr[i].p_type); ++ phdr[i].p_offset = elf32_to_cpu(phdr[i].p_offset); ++ phdr[i].p_vaddr = elf32_to_cpu(phdr[i].p_vaddr); ++ phdr[i].p_paddr = elf32_to_cpu(phdr[i].p_paddr); ++ phdr[i].p_filesz = elf32_to_cpu(phdr[i].p_filesz); ++ phdr[i].p_memsz = elf32_to_cpu(phdr[i].p_memsz); ++ phdr[i].p_flags = elf32_to_cpu(phdr[i].p_flags); ++ phdr[i].p_align = elf32_to_cpu(phdr[i].p_align); ++ } ++ ++} ++ + static void read_shdrs(FILE *fp) + { +- int i; ++ unsigned int i; + Elf32_Shdr shdr; + + secs = calloc(ehdr.e_shnum, sizeof(struct section)); +@@ -282,7 +315,7 @@ static void read_shdrs(FILE *fp) + + static void read_strtabs(FILE *fp) + { +- int i; ++ unsigned int i; + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + if (sec->shdr.sh_type != SHT_STRTAB) { +@@ -307,7 +340,7 @@ static void read_strtabs(FILE *fp) + + static void read_symtabs(FILE *fp) + { +- int i,j; ++ unsigned int i,j; + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + if (sec->shdr.sh_type != SHT_SYMTAB) { +@@ -340,7 +373,9 @@ static void read_symtabs(FILE *fp) + + static void read_relocs(FILE *fp) + { +- int i,j; ++ unsigned int i,j; ++ uint32_t base; ++ + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + if (sec->shdr.sh_type != SHT_REL) { +@@ -360,9 +395,18 @@ static void read_relocs(FILE *fp) + die("Cannot read symbol table: %s\n", + strerror(errno)); + } ++ base = 0; ++ for (j = 0; j < ehdr.e_phnum; j++) { ++ if (phdr[j].p_type != PT_LOAD ) ++ continue; ++ if (secs[sec->shdr.sh_info].shdr.sh_offset < phdr[j].p_offset || secs[sec->shdr.sh_info].shdr.sh_offset >= phdr[j].p_offset + phdr[j].p_filesz) ++ continue; ++ base = CONFIG_PAGE_OFFSET + phdr[j].p_paddr - phdr[j].p_vaddr; ++ break; ++ } + for (j = 0; j < sec->shdr.sh_size/sizeof(Elf32_Rel); j++) { + Elf32_Rel *rel = &sec->reltab[j]; +- rel->r_offset = elf32_to_cpu(rel->r_offset); ++ rel->r_offset = elf32_to_cpu(rel->r_offset) + base; + rel->r_info = elf32_to_cpu(rel->r_info); + } + } +@@ -371,14 +415,14 @@ static void read_relocs(FILE *fp) + + static void print_absolute_symbols(void) + { +- int i; ++ unsigned int i; + printf("Absolute symbols\n"); + printf(" Num: Value Size Type Bind Visibility Name\n"); + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + char *sym_strtab; + Elf32_Sym *sh_symtab; +- int j; ++ unsigned int j; + + if (sec->shdr.sh_type != SHT_SYMTAB) { + continue; +@@ -406,14 +450,14 @@ static void print_absolute_symbols(void) + + static void print_absolute_relocs(void) + { +- int i, printed = 0; ++ unsigned int i, printed = 0; + + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + struct section *sec_applies, *sec_symtab; + char *sym_strtab; + Elf32_Sym *sh_symtab; +- int j; ++ unsigned int j; + if (sec->shdr.sh_type != SHT_REL) { + continue; + } +@@ -474,13 +518,13 @@ static void print_absolute_relocs(void) + + static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym)) + { +- int i; ++ unsigned int i; + /* Walk through the relocations */ + for (i = 0; i < ehdr.e_shnum; i++) { + char *sym_strtab; + Elf32_Sym *sh_symtab; + struct section *sec_applies, *sec_symtab; +- int j; ++ unsigned int j; + struct section *sec = &secs[i]; + + if (sec->shdr.sh_type != SHT_REL) { +@@ -504,6 +548,19 @@ static void walk_relocs(void (*visit)(El + if (sym->st_shndx == SHN_ABS) { + continue; + } ++ /* Don't relocate actual per-cpu variables, they are absolute indices, not addresses */ ++ if (!strcmp(sec_name(sym->st_shndx), ".data.percpu") && strcmp(sym_name(sym_strtab, sym), "__per_cpu_load")) ++ continue; ++ ++#if defined(CONFIG_PAX_KERNEXEC) && defined(CONFIG_X86_32) ++ /* Don't relocate actual code, they are relocated implicitly by the base address of KERNEL_CS */ ++ if (!strcmp(sec_name(sym->st_shndx), ".init.text")) ++ continue; ++ if (!strcmp(sec_name(sym->st_shndx), ".exit.text")) ++ continue; ++ if (!strcmp(sec_name(sym->st_shndx), ".text") && strcmp(sym_name(sym_strtab, sym), "__LOAD_PHYSICAL_ADDR")) ++ continue; ++#endif + if (r_type == R_386_NONE || r_type == R_386_PC32) { + /* + * NONE can be ignored and and PC relative +@@ -541,7 +598,7 @@ static int cmp_relocs(const void *va, co + + static void emit_relocs(int as_text) + { +- int i; ++ unsigned int i; + /* Count how many relocations I have and allocate space for them. */ + reloc_count = 0; + walk_relocs(count_reloc); +@@ -634,6 +691,7 @@ int main(int argc, char **argv) + fname, strerror(errno)); + } + read_ehdr(fp); ++ read_phdrs(fp); + read_shdrs(fp); + read_strtabs(fp); + read_symtabs(fp); +diff -urNp linux-2.6.31.1/arch/x86/boot/cpucheck.c linux-2.6.31.1/arch/x86/boot/cpucheck.c +--- linux-2.6.31.1/arch/x86/boot/cpucheck.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/cpucheck.c 2009-10-01 20:12:42.000000000 -0400 +@@ -74,7 +74,7 @@ static int has_fpu(void) + u16 fcw = -1, fsw = -1; + u32 cr0; + +- asm("movl %%cr0,%0" : "=r" (cr0)); ++ asm volatile("movl %%cr0,%0" : "=r" (cr0)); + if (cr0 & (X86_CR0_EM|X86_CR0_TS)) { + cr0 &= ~(X86_CR0_EM|X86_CR0_TS); + asm volatile("movl %0,%%cr0" : : "r" (cr0)); +@@ -90,7 +90,7 @@ static int has_eflag(u32 mask) + { + u32 f0, f1; + +- asm("pushfl ; " ++ asm volatile("pushfl ; " + "pushfl ; " + "popl %0 ; " + "movl %0,%1 ; " +@@ -115,7 +115,7 @@ static void get_flags(void) + set_bit(X86_FEATURE_FPU, cpu.flags); + + if (has_eflag(X86_EFLAGS_ID)) { +- asm("cpuid" ++ asm volatile("cpuid" + : "=a" (max_intel_level), + "=b" (cpu_vendor[0]), + "=d" (cpu_vendor[1]), +@@ -124,7 +124,7 @@ static void get_flags(void) + + if (max_intel_level >= 0x00000001 && + max_intel_level <= 0x0000ffff) { +- asm("cpuid" ++ asm volatile("cpuid" + : "=a" (tfms), + "=c" (cpu.flags[4]), + "=d" (cpu.flags[0]) +@@ -136,7 +136,7 @@ static void get_flags(void) + cpu.model += ((tfms >> 16) & 0xf) << 4; + } + +- asm("cpuid" ++ asm volatile("cpuid" + : "=a" (max_amd_level) + : "a" (0x80000000) + : "ebx", "ecx", "edx"); +@@ -144,7 +144,7 @@ static void get_flags(void) + if (max_amd_level >= 0x80000001 && + max_amd_level <= 0x8000ffff) { + u32 eax = 0x80000001; +- asm("cpuid" ++ asm volatile("cpuid" + : "+a" (eax), + "=c" (cpu.flags[6]), + "=d" (cpu.flags[1]) +@@ -203,9 +203,9 @@ int check_cpu(int *cpu_level_ptr, int *r + u32 ecx = MSR_K7_HWCR; + u32 eax, edx; + +- asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); ++ asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); + eax &= ~(1 << 15); +- asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); ++ asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); + + get_flags(); /* Make sure it really did something */ + err = check_flags(); +@@ -218,9 +218,9 @@ int check_cpu(int *cpu_level_ptr, int *r + u32 ecx = MSR_VIA_FCR; + u32 eax, edx; + +- asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); ++ asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); + eax |= (1<<1)|(1<<7); +- asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); ++ asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); + + set_bit(X86_FEATURE_CX8, cpu.flags); + err = check_flags(); +@@ -231,12 +231,12 @@ int check_cpu(int *cpu_level_ptr, int *r + u32 eax, edx; + u32 level = 1; + +- asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); +- asm("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx)); +- asm("cpuid" ++ asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); ++ asm volatile("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx)); ++ asm volatile("cpuid" + : "+a" (level), "=d" (cpu.flags[0]) + : : "ecx", "ebx"); +- asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); ++ asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); + + err = check_flags(); + } +diff -urNp linux-2.6.31.1/arch/x86/boot/header.S linux-2.6.31.1/arch/x86/boot/header.S +--- linux-2.6.31.1/arch/x86/boot/header.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/header.S 2009-10-01 20:12:42.000000000 -0400 +@@ -224,7 +224,7 @@ setup_data: .quad 0 # 64-bit physical + # single linked list of + # struct setup_data + +-pref_address: .quad LOAD_PHYSICAL_ADDR # preferred load addr ++pref_address: .quad ____LOAD_PHYSICAL_ADDR # preferred load addr + + #define ZO_INIT_SIZE (ZO__end - ZO_startup_32 + ZO_z_extract_offset) + #define VO_INIT_SIZE (VO__end - VO__text) +diff -urNp linux-2.6.31.1/arch/x86/boot/video-vesa.c linux-2.6.31.1/arch/x86/boot/video-vesa.c +--- linux-2.6.31.1/arch/x86/boot/video-vesa.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/boot/video-vesa.c 2009-10-01 20:12:42.000000000 -0400 +@@ -205,6 +205,7 @@ static void vesa_store_pm_info(void) + + boot_params.screen_info.vesapm_seg = oreg.es; + boot_params.screen_info.vesapm_off = oreg.di; ++ boot_params.screen_info.vesapm_size = oreg.cx; + } + + /* +diff -urNp linux-2.6.31.1/arch/x86/ia32/ia32_signal.c linux-2.6.31.1/arch/x86/ia32/ia32_signal.c +--- linux-2.6.31.1/arch/x86/ia32/ia32_signal.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/ia32/ia32_signal.c 2009-10-01 20:12:42.000000000 -0400 +@@ -403,7 +403,7 @@ static void __user *get_sigframe(struct + sp -= frame_size; + /* Align the stack pointer according to the i386 ABI, + * i.e. so that on function entry ((sp + 4) & 15) == 0. */ +- sp = ((sp + 4) & -16ul) - 4; ++ sp = ((sp - 12) & -16ul) - 4; + return (void __user *) sp; + } + +@@ -503,7 +503,7 @@ int ia32_setup_rt_frame(int sig, struct + 0xb8, + __NR_ia32_rt_sigreturn, + 0x80cd, +- 0, ++ 0 + }; + + frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate); +diff -urNp linux-2.6.31.1/arch/x86/include/asm/alternative.h linux-2.6.31.1/arch/x86/include/asm/alternative.h +--- linux-2.6.31.1/arch/x86/include/asm/alternative.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/alternative.h 2009-10-01 20:12:42.000000000 -0400 +@@ -87,7 +87,7 @@ const unsigned char *const *find_nop_tab + " .byte 662b-661b\n" /* sourcelen */ \ + " .byte 664f-663f\n" /* replacementlen */ \ + ".previous\n" \ +- ".section .altinstr_replacement, \"ax\"\n" \ ++ ".section .altinstr_replacement, \"a\"\n" \ + "663:\n\t" newinstr "\n664:\n" /* replacement */ \ + ".previous" + +diff -urNp linux-2.6.31.1/arch/x86/include/asm/apm.h linux-2.6.31.1/arch/x86/include/asm/apm.h +--- linux-2.6.31.1/arch/x86/include/asm/apm.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/apm.h 2009-10-01 20:12:42.000000000 -0400 +@@ -34,7 +34,7 @@ static inline void apm_bios_call_asm(u32 + __asm__ __volatile__(APM_DO_ZERO_SEGS + "pushl %%edi\n\t" + "pushl %%ebp\n\t" +- "lcall *%%cs:apm_bios_entry\n\t" ++ "lcall *%%ss:apm_bios_entry\n\t" + "setc %%al\n\t" + "popl %%ebp\n\t" + "popl %%edi\n\t" +@@ -58,7 +58,7 @@ static inline u8 apm_bios_call_simple_as + __asm__ __volatile__(APM_DO_ZERO_SEGS + "pushl %%edi\n\t" + "pushl %%ebp\n\t" +- "lcall *%%cs:apm_bios_entry\n\t" ++ "lcall *%%ss:apm_bios_entry\n\t" + "setc %%bl\n\t" + "popl %%ebp\n\t" + "popl %%edi\n\t" +diff -urNp linux-2.6.31.1/arch/x86/include/asm/atomic_32.h linux-2.6.31.1/arch/x86/include/asm/atomic_32.h +--- linux-2.6.31.1/arch/x86/include/asm/atomic_32.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/atomic_32.h 2009-10-01 20:12:42.000000000 -0400 +@@ -25,6 +25,17 @@ static inline int atomic_read(const atom + } + + /** ++ * atomic_read_unchecked - read atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically reads the value of @v. ++ */ ++static inline int atomic_read_unchecked(const atomic_unchecked_t *v) ++{ ++ return v->counter; ++} ++ ++/** + * atomic_set - set atomic variable + * @v: pointer of type atomic_t + * @i: required value +@@ -37,6 +48,18 @@ static inline void atomic_set(atomic_t * + } + + /** ++ * atomic_set_unchecked - set atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * @i: required value ++ * ++ * Atomically sets the value of @v to @i. ++ */ ++static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i) ++{ ++ v->counter = i; ++} ++ ++/** + * atomic_add - add integer to atomic variable + * @i: integer value to add + * @v: pointer of type atomic_t +@@ -45,7 +68,29 @@ static inline void atomic_set(atomic_t * + */ + static inline void atomic_add(int i, atomic_t *v) + { +- asm volatile(LOCK_PREFIX "addl %1,%0" ++ asm volatile(LOCK_PREFIX "addl %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subl %1,%0\n" ++ "into\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "+m" (v->counter) ++ : "ir" (i)); ++} ++ ++/** ++ * atomic_add_unchecked - add integer to atomic variable ++ * @i: integer value to add ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically adds @i to @v. ++ */ ++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "addl %1,%0\n" + : "+m" (v->counter) + : "ir" (i)); + } +@@ -59,7 +104,29 @@ static inline void atomic_add(int i, ato + */ + static inline void atomic_sub(int i, atomic_t *v) + { +- asm volatile(LOCK_PREFIX "subl %1,%0" ++ asm volatile(LOCK_PREFIX "subl %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addl %1,%0\n" ++ "into\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "+m" (v->counter) ++ : "ir" (i)); ++} ++ ++/** ++ * atomic_sub_unchecked - subtract integer from atomic variable ++ * @i: integer value to subtract ++ * @v: pointer of type atomic_t ++ * ++ * Atomically subtracts @i from @v. ++ */ ++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "subl %1,%0\n" + : "+m" (v->counter) + : "ir" (i)); + } +@@ -77,7 +144,16 @@ static inline int atomic_sub_and_test(in + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "subl %2,%0; sete %1" ++ asm volatile(LOCK_PREFIX "subl %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addl %2,%0\n" ++ "into\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "+m" (v->counter), "=qm" (c) + : "ir" (i) : "memory"); + return c; +@@ -91,7 +167,30 @@ static inline int atomic_sub_and_test(in + */ + static inline void atomic_inc(atomic_t *v) + { +- asm volatile(LOCK_PREFIX "incl %0" ++ asm volatile(LOCK_PREFIX "incl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "into\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ LOCK_PREFIX "decl %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ : "+m" (v->counter)); ++} ++ ++/** ++ * atomic_inc_unchecked - increment atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically increments @v by 1. ++ */ ++static inline void atomic_inc_unchecked(atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "incl %0\n" + : "+m" (v->counter)); + } + +@@ -103,7 +202,18 @@ static inline void atomic_inc(atomic_t * + */ + static inline void atomic_dec(atomic_t *v) + { +- asm volatile(LOCK_PREFIX "decl %0" ++ asm volatile(LOCK_PREFIX "decl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "into\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1: \n" ++ LOCK_PREFIX "incl %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "+m" (v->counter)); + } + +@@ -119,7 +229,19 @@ static inline int atomic_dec_and_test(at + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "decl %0; sete %1" ++ asm volatile(LOCK_PREFIX "decl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "into\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1: \n" ++ LOCK_PREFIX "incl %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sete %1\n" + : "+m" (v->counter), "=qm" (c) + : : "memory"); + return c != 0; +@@ -137,7 +259,19 @@ static inline int atomic_inc_and_test(at + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "incl %0; sete %1" ++ asm volatile(LOCK_PREFIX "incl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "into\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1: \n" ++ LOCK_PREFIX "decl %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sete %1\n" + : "+m" (v->counter), "=qm" (c) + : : "memory"); + return c != 0; +@@ -156,7 +290,16 @@ static inline int atomic_add_negative(in + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "addl %2,%0; sets %1" ++ asm volatile(LOCK_PREFIX "addl %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subl %2,%0\n" ++ "into\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sets %1\n" + : "+m" (v->counter), "=qm" (c) + : "ir" (i) : "memory"); + return c; +@@ -179,7 +322,15 @@ static inline int atomic_add_return(int + #endif + /* Modern 486+ processor */ + __i = i; +- asm volatile(LOCK_PREFIX "xaddl %0, %1" ++ asm volatile(LOCK_PREFIX "xaddl %0, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "movl %0, %1\n" ++ "into\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+r" (i), "+m" (v->counter) + : : "memory"); + return i + __i; +@@ -227,17 +378,28 @@ static inline int atomic_xchg(atomic_t * + */ + static inline int atomic_add_unless(atomic_t *v, int a, int u) + { +- int c, old; ++ int c, old, new; + c = atomic_read(v); + for (;;) { +- if (unlikely(c == (u))) ++ if (unlikely(c == u)) + break; +- old = atomic_cmpxchg((v), c, c + (a)); ++ ++ asm volatile("addl %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "into\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=r" (new) ++ : "0" (c), "ir" (a)); ++ ++ old = atomic_cmpxchg(v, c, new); + if (likely(old == c)) + break; + c = old; + } +- return c != (u); ++ return c != u; + } + + #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) +diff -urNp linux-2.6.31.1/arch/x86/include/asm/atomic_64.h linux-2.6.31.1/arch/x86/include/asm/atomic_64.h +--- linux-2.6.31.1/arch/x86/include/asm/atomic_64.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/atomic_64.h 2009-10-01 20:12:42.000000000 -0400 +@@ -24,6 +24,17 @@ static inline int atomic_read(const atom + } + + /** ++ * atomic_read_unchecked - read atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically reads the value of @v. ++ */ ++static inline int atomic_read_unchecked(const atomic_unchecked_t *v) ++{ ++ return v->counter; ++} ++ ++/** + * atomic_set - set atomic variable + * @v: pointer of type atomic_t + * @i: required value +@@ -36,6 +47,18 @@ static inline void atomic_set(atomic_t * + } + + /** ++ * atomic_set_unchecked - set atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * @i: required value ++ * ++ * Atomically sets the value of @v to @i. ++ */ ++static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i) ++{ ++ v->counter = i; ++} ++ ++/** + * atomic_add - add integer to atomic variable + * @i: integer value to add + * @v: pointer of type atomic_t +@@ -44,7 +67,29 @@ static inline void atomic_set(atomic_t * + */ + static inline void atomic_add(int i, atomic_t *v) + { +- asm volatile(LOCK_PREFIX "addl %1,%0" ++ asm volatile(LOCK_PREFIX "addl %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subl %1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=m" (v->counter) ++ : "ir" (i), "m" (v->counter)); ++} ++ ++/** ++ * atomic_add_unchecked - add integer to atomic variable ++ * @i: integer value to add ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically adds @i to @v. ++ */ ++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "addl %1,%0\n" + : "=m" (v->counter) + : "ir" (i), "m" (v->counter)); + } +@@ -58,7 +103,29 @@ static inline void atomic_add(int i, ato + */ + static inline void atomic_sub(int i, atomic_t *v) + { +- asm volatile(LOCK_PREFIX "subl %1,%0" ++ asm volatile(LOCK_PREFIX "subl %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addl %1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=m" (v->counter) ++ : "ir" (i), "m" (v->counter)); ++} ++ ++/** ++ * atomic_sub_unchecked - subtract the atomic variable ++ * @i: integer value to subtract ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically subtracts @i from @v. ++ */ ++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "subl %1,%0\n" + : "=m" (v->counter) + : "ir" (i), "m" (v->counter)); + } +@@ -76,7 +143,16 @@ static inline int atomic_sub_and_test(in + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "subl %2,%0; sete %1" ++ asm volatile(LOCK_PREFIX "subl %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addl %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "ir" (i), "m" (v->counter) : "memory"); + return c; +@@ -90,7 +166,32 @@ static inline int atomic_sub_and_test(in + */ + static inline void atomic_inc(atomic_t *v) + { +- asm volatile(LOCK_PREFIX "incl %0" ++ asm volatile(LOCK_PREFIX "incl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ LOCK_PREFIX "decl %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ : "=m" (v->counter) ++ : "m" (v->counter)); ++} ++ ++/** ++ * atomic_inc_unchecked - increment atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically increments @v by 1. ++ */ ++static inline void atomic_inc_unchecked(atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "incl %0\n" + : "=m" (v->counter) + : "m" (v->counter)); + } +@@ -103,7 +204,19 @@ static inline void atomic_inc(atomic_t * + */ + static inline void atomic_dec(atomic_t *v) + { +- asm volatile(LOCK_PREFIX "decl %0" ++ asm volatile(LOCK_PREFIX "decl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1: \n" ++ LOCK_PREFIX "incl %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "=m" (v->counter) + : "m" (v->counter)); + } +@@ -120,7 +233,20 @@ static inline int atomic_dec_and_test(at + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "decl %0; sete %1" ++ asm volatile(LOCK_PREFIX "decl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1: \n" ++ LOCK_PREFIX "incl %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "m" (v->counter) : "memory"); + return c != 0; +@@ -138,7 +264,20 @@ static inline int atomic_inc_and_test(at + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "incl %0; sete %1" ++ asm volatile(LOCK_PREFIX "incl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1: \n" ++ LOCK_PREFIX "decl %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "m" (v->counter) : "memory"); + return c != 0; +@@ -157,7 +296,16 @@ static inline int atomic_add_negative(in + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "addl %2,%0; sets %1" ++ asm volatile(LOCK_PREFIX "addl %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subl %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sets %1\n" + : "=m" (v->counter), "=qm" (c) + : "ir" (i), "m" (v->counter) : "memory"); + return c; +@@ -173,7 +321,15 @@ static inline int atomic_add_negative(in + static inline int atomic_add_return(int i, atomic_t *v) + { + int __i = i; +- asm volatile(LOCK_PREFIX "xaddl %0, %1" ++ asm volatile(LOCK_PREFIX "xaddl %0, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "movl %0, %1\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+r" (i), "+m" (v->counter) + : : "memory"); + return i + __i; +@@ -224,7 +380,15 @@ static inline void atomic64_set(atomic64 + */ + static inline void atomic64_add(long i, atomic64_t *v) + { +- asm volatile(LOCK_PREFIX "addq %1,%0" ++ asm volatile(LOCK_PREFIX "addq %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subq %1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "=m" (v->counter) + : "er" (i), "m" (v->counter)); + } +@@ -238,7 +402,15 @@ static inline void atomic64_add(long i, + */ + static inline void atomic64_sub(long i, atomic64_t *v) + { +- asm volatile(LOCK_PREFIX "subq %1,%0" ++ asm volatile(LOCK_PREFIX "subq %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addq %1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "=m" (v->counter) + : "er" (i), "m" (v->counter)); + } +@@ -256,7 +428,16 @@ static inline int atomic64_sub_and_test( + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "subq %2,%0; sete %1" ++ asm volatile(LOCK_PREFIX "subq %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addq %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "er" (i), "m" (v->counter) : "memory"); + return c; +@@ -270,7 +451,19 @@ static inline int atomic64_sub_and_test( + */ + static inline void atomic64_inc(atomic64_t *v) + { +- asm volatile(LOCK_PREFIX "incq %0" ++ asm volatile(LOCK_PREFIX "incq %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ LOCK_PREFIX "decq %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "=m" (v->counter) + : "m" (v->counter)); + } +@@ -283,7 +476,19 @@ static inline void atomic64_inc(atomic64 + */ + static inline void atomic64_dec(atomic64_t *v) + { +- asm volatile(LOCK_PREFIX "decq %0" ++ asm volatile(LOCK_PREFIX "decq %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1: \n" ++ LOCK_PREFIX "incq %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "=m" (v->counter) + : "m" (v->counter)); + } +@@ -300,7 +505,20 @@ static inline int atomic64_dec_and_test( + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "decq %0; sete %1" ++ asm volatile(LOCK_PREFIX "decq %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1: \n" ++ LOCK_PREFIX "incq %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "m" (v->counter) : "memory"); + return c != 0; +@@ -318,7 +536,20 @@ static inline int atomic64_inc_and_test( + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "incq %0; sete %1" ++ asm volatile(LOCK_PREFIX "incq %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ ".pushsection .fixup,\"ax\"\n" ++ "1: \n" ++ LOCK_PREFIX "decq %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "m" (v->counter) : "memory"); + return c != 0; +@@ -337,7 +568,16 @@ static inline int atomic64_add_negative( + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "addq %2,%0; sets %1" ++ asm volatile(LOCK_PREFIX "addq %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subq %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sets %1\n" + : "=m" (v->counter), "=qm" (c) + : "er" (i), "m" (v->counter) : "memory"); + return c; +@@ -353,7 +593,15 @@ static inline int atomic64_add_negative( + static inline long atomic64_add_return(long i, atomic64_t *v) + { + long __i = i; +- asm volatile(LOCK_PREFIX "xaddq %0, %1;" ++ asm volatile(LOCK_PREFIX "xaddq %0, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "movq %0, %1\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+r" (i), "+m" (v->counter) + : : "memory"); + return i + __i; +@@ -398,17 +646,29 @@ static inline long atomic_xchg(atomic_t + */ + static inline int atomic_add_unless(atomic_t *v, int a, int u) + { +- int c, old; ++ int c, old, new; + c = atomic_read(v); + for (;;) { +- if (unlikely(c == (u))) ++ if (unlikely(c == u)) + break; +- old = atomic_cmpxchg((v), c, c + (a)); ++ ++ asm volatile("addl %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=r" (new) ++ : "0" (c), "ir" (a)); ++ ++ old = atomic_cmpxchg(v, c, new); + if (likely(old == c)) + break; + c = old; + } +- return c != (u); ++ return c != u; + } + + #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) +@@ -424,17 +684,29 @@ static inline int atomic_add_unless(atom + */ + static inline int atomic64_add_unless(atomic64_t *v, long a, long u) + { +- long c, old; ++ long c, old, new; + c = atomic64_read(v); + for (;;) { +- if (unlikely(c == (u))) ++ if (unlikely(c == u)) + break; +- old = atomic64_cmpxchg((v), c, c + (a)); ++ ++ asm volatile("addq %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=r" (new) ++ : "0" (c), "er" (a)); ++ ++ old = atomic64_cmpxchg((v), c, new); + if (likely(old == c)) + break; + c = old; + } +- return c != (u); ++ return c != u; + } + + /** +diff -urNp linux-2.6.31.1/arch/x86/include/asm/boot.h linux-2.6.31.1/arch/x86/include/asm/boot.h +--- linux-2.6.31.1/arch/x86/include/asm/boot.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/boot.h 2009-10-01 20:12:42.000000000 -0400 +@@ -11,10 +11,15 @@ + #include + + /* Physical address where kernel should be loaded. */ +-#define LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \ ++#define ____LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \ + + (CONFIG_PHYSICAL_ALIGN - 1)) \ + & ~(CONFIG_PHYSICAL_ALIGN - 1)) + ++#ifndef __ASSEMBLY__ ++extern unsigned char __LOAD_PHYSICAL_ADDR[]; ++#define LOAD_PHYSICAL_ADDR ((unsigned long)__LOAD_PHYSICAL_ADDR) ++#endif ++ + /* Minimum kernel alignment, as a power of two */ + #ifdef CONFIG_X86_64 + #define MIN_KERNEL_ALIGN_LG2 PMD_SHIFT +diff -urNp linux-2.6.31.1/arch/x86/include/asm/cache.h linux-2.6.31.1/arch/x86/include/asm/cache.h +--- linux-2.6.31.1/arch/x86/include/asm/cache.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/cache.h 2009-10-01 20:12:42.000000000 -0400 +@@ -6,6 +6,7 @@ + #define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT) + + #define __read_mostly __attribute__((__section__(".data.read_mostly"))) ++#define __read_only __attribute__((__section__(".data.read_only"))) + + #ifdef CONFIG_X86_VSMP + /* vSMP Internode cacheline shift */ +diff -urNp linux-2.6.31.1/arch/x86/include/asm/checksum_32.h linux-2.6.31.1/arch/x86/include/asm/checksum_32.h +--- linux-2.6.31.1/arch/x86/include/asm/checksum_32.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/checksum_32.h 2009-10-01 20:12:42.000000000 -0400 +@@ -31,6 +31,14 @@ asmlinkage __wsum csum_partial_copy_gene + int len, __wsum sum, + int *src_err_ptr, int *dst_err_ptr); + ++asmlinkage __wsum csum_partial_copy_generic_to_user(const void *src, void *dst, ++ int len, __wsum sum, ++ int *src_err_ptr, int *dst_err_ptr); ++ ++asmlinkage __wsum csum_partial_copy_generic_from_user(const void *src, void *dst, ++ int len, __wsum sum, ++ int *src_err_ptr, int *dst_err_ptr); ++ + /* + * Note: when you get a NULL pointer exception here this means someone + * passed in an incorrect kernel address to one of these functions. +@@ -50,7 +58,7 @@ static inline __wsum csum_partial_copy_f + int *err_ptr) + { + might_sleep(); +- return csum_partial_copy_generic((__force void *)src, dst, ++ return csum_partial_copy_generic_from_user((__force void *)src, dst, + len, sum, err_ptr, NULL); + } + +@@ -177,7 +185,7 @@ static inline __wsum csum_and_copy_to_us + { + might_sleep(); + if (access_ok(VERIFY_WRITE, dst, len)) +- return csum_partial_copy_generic(src, (__force void *)dst, ++ return csum_partial_copy_generic_to_user(src, (__force void *)dst, + len, sum, NULL, err_ptr); + + if (len) +diff -urNp linux-2.6.31.1/arch/x86/include/asm/desc.h linux-2.6.31.1/arch/x86/include/asm/desc.h +--- linux-2.6.31.1/arch/x86/include/asm/desc.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/desc.h 2009-10-01 20:12:42.000000000 -0400 +@@ -15,6 +15,7 @@ static inline void fill_ldt(struct desc_ + desc->base1 = (info->base_addr & 0x00ff0000) >> 16; + desc->type = (info->read_exec_only ^ 1) << 1; + desc->type |= info->contents << 2; ++ desc->type |= info->seg_not_present ^ 1; + desc->s = 1; + desc->dpl = 0x3; + desc->p = info->seg_not_present ^ 1; +@@ -31,16 +32,12 @@ static inline void fill_ldt(struct desc_ + } + + extern struct desc_ptr idt_descr; +-extern gate_desc idt_table[]; +- +-struct gdt_page { +- struct desc_struct gdt[GDT_ENTRIES]; +-} __attribute__((aligned(PAGE_SIZE))); +-DECLARE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page); ++extern gate_desc idt_table[256]; + ++extern struct desc_struct cpu_gdt_table[NR_CPUS][PAGE_SIZE / sizeof(struct desc_struct)]; + static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu) + { +- return per_cpu(gdt_page, cpu).gdt; ++ return cpu_gdt_table[cpu]; + } + + #ifdef CONFIG_X86_64 +@@ -115,19 +112,48 @@ static inline void paravirt_free_ldt(str + static inline void native_write_idt_entry(gate_desc *idt, int entry, + const gate_desc *gate) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + memcpy(&idt[entry], gate, sizeof(*gate)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + static inline void native_write_ldt_entry(struct desc_struct *ldt, int entry, + const void *desc) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + memcpy(&ldt[entry], desc, 8); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + static inline void native_write_gdt_entry(struct desc_struct *gdt, int entry, + const void *desc, int type) + { + unsigned int size; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + switch (type) { + case DESC_TSS: + size = sizeof(tss_desc); +@@ -139,7 +165,17 @@ static inline void native_write_gdt_entr + size = sizeof(struct desc_struct); + break; + } ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + memcpy(&gdt[entry], desc, size); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + static inline void pack_descriptor(struct desc_struct *desc, unsigned long base, +@@ -211,7 +247,19 @@ static inline void native_set_ldt(const + + static inline void native_load_tr_desc(void) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + static inline void native_load_gdt(const struct desc_ptr *dtr) +@@ -246,8 +294,19 @@ static inline void native_load_tls(struc + unsigned int i; + struct desc_struct *gdt = get_cpu_gdt_table(cpu); + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++) + gdt[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i]; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + #define _LDT_empty(info) \ +@@ -379,4 +438,16 @@ static inline void set_system_intr_gate_ + _set_gate(n, GATE_INTERRUPT, addr, 0x3, ist, __KERNEL_CS); + } + ++#ifdef CONFIG_X86_32 ++static inline void set_user_cs(unsigned long base, unsigned long limit, int cpu) ++{ ++ struct desc_struct d; ++ ++ if (likely(limit)) ++ limit = (limit - 1UL) >> PAGE_SHIFT; ++ pack_descriptor(&d, base, limit, 0xFB, 0xC); ++ write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_DEFAULT_USER_CS, &d, DESCTYPE_S); ++} ++#endif ++ + #endif /* _ASM_X86_DESC_H */ +diff -urNp linux-2.6.31.1/arch/x86/include/asm/e820.h linux-2.6.31.1/arch/x86/include/asm/e820.h +--- linux-2.6.31.1/arch/x86/include/asm/e820.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/e820.h 2009-10-01 20:12:42.000000000 -0400 +@@ -135,7 +135,7 @@ extern char *memory_setup(void); + #define ISA_END_ADDRESS 0x100000 + #define is_ISA_range(s, e) ((s) >= ISA_START_ADDRESS && (e) < ISA_END_ADDRESS) + +-#define BIOS_BEGIN 0x000a0000 ++#define BIOS_BEGIN 0x000c0000 + #define BIOS_END 0x00100000 + + #ifdef __KERNEL__ +diff -urNp linux-2.6.31.1/arch/x86/include/asm/elf.h linux-2.6.31.1/arch/x86/include/asm/elf.h +--- linux-2.6.31.1/arch/x86/include/asm/elf.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/elf.h 2009-10-01 20:12:42.000000000 -0400 +@@ -263,7 +263,25 @@ extern int force_personality32; + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. */ + ++#ifdef CONFIG_PAX_SEGMEXEC ++#define ELF_ET_DYN_BASE ((current->mm->pax_flags & MF_PAX_SEGMEXEC) ? SEGMEXEC_TASK_SIZE/3*2 : TASK_SIZE/3*2) ++#else + #define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) ++#endif ++ ++#ifdef CONFIG_PAX_ASLR ++#ifdef CONFIG_X86_32 ++#define PAX_ELF_ET_DYN_BASE 0x10000000UL ++ ++#define PAX_DELTA_MMAP_LEN (current->mm->pax_flags & MF_PAX_SEGMEXEC ? 15 : 16) ++#define PAX_DELTA_STACK_LEN (current->mm->pax_flags & MF_PAX_SEGMEXEC ? 15 : 16) ++#else ++#define PAX_ELF_ET_DYN_BASE 0x400000UL ++ ++#define PAX_DELTA_MMAP_LEN ((test_thread_flag(TIF_IA32)) ? 16 : 32) ++#define PAX_DELTA_STACK_LEN ((test_thread_flag(TIF_IA32)) ? 16 : 32) ++#endif ++#endif + + /* This yields a mask that user programs can use to figure out what + instruction set this CPU supports. This could be done in user space, +@@ -315,8 +333,7 @@ do { \ + #define ARCH_DLINFO \ + do { \ + if (vdso_enabled) \ +- NEW_AUX_ENT(AT_SYSINFO_EHDR, \ +- (unsigned long)current->mm->context.vdso); \ ++ NEW_AUX_ENT(AT_SYSINFO_EHDR, current->mm->context.vdso);\ + } while (0) + + #define AT_SYSINFO 32 +@@ -327,7 +344,7 @@ do { \ + + #endif /* !CONFIG_X86_32 */ + +-#define VDSO_CURRENT_BASE ((unsigned long)current->mm->context.vdso) ++#define VDSO_CURRENT_BASE (current->mm->context.vdso) + + #define VDSO_ENTRY \ + ((unsigned long)VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall)) +@@ -341,7 +358,4 @@ extern int arch_setup_additional_pages(s + extern int syscall32_setup_pages(struct linux_binprm *, int exstack); + #define compat_arch_setup_additional_pages syscall32_setup_pages + +-extern unsigned long arch_randomize_brk(struct mm_struct *mm); +-#define arch_randomize_brk arch_randomize_brk +- + #endif /* _ASM_X86_ELF_H */ +diff -urNp linux-2.6.31.1/arch/x86/include/asm/futex.h linux-2.6.31.1/arch/x86/include/asm/futex.h +--- linux-2.6.31.1/arch/x86/include/asm/futex.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/futex.h 2009-10-01 20:12:42.000000000 -0400 +@@ -11,6 +11,40 @@ + #include + #include + ++#ifdef CONFIG_X86_32 ++#define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg) \ ++ asm volatile( \ ++ "movw\t%w6, %%ds\n" \ ++ "1:\t" insn "\n" \ ++ "2:\tpushl\t%%ss\n" \ ++ "\tpopl\t%%ds\n" \ ++ "\t.section .fixup,\"ax\"\n" \ ++ "3:\tmov\t%3, %1\n" \ ++ "\tjmp\t2b\n" \ ++ "\t.previous\n" \ ++ _ASM_EXTABLE(1b, 3b) \ ++ : "=r" (oldval), "=r" (ret), "+m" (*uaddr) \ ++ : "i" (-EFAULT), "0" (oparg), "1" (0), "r" (__USER_DS)) ++ ++#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg) \ ++ asm volatile("movw\t%w7, %%es\n" \ ++ "1:\tmovl\t%%es:%2, %0\n" \ ++ "\tmovl\t%0, %3\n" \ ++ "\t" insn "\n" \ ++ "2:\t" LOCK_PREFIX "cmpxchgl %3, %%es:%2\n"\ ++ "\tjnz\t1b\n" \ ++ "3:\tpushl\t%%ss\n" \ ++ "\tpopl\t%%es\n" \ ++ "\t.section .fixup,\"ax\"\n" \ ++ "4:\tmov\t%5, %1\n" \ ++ "\tjmp\t3b\n" \ ++ "\t.previous\n" \ ++ _ASM_EXTABLE(1b, 4b) \ ++ _ASM_EXTABLE(2b, 4b) \ ++ : "=&a" (oldval), "=&r" (ret), \ ++ "+m" (*uaddr), "=&r" (tem) \ ++ : "r" (oparg), "i" (-EFAULT), "1" (0), "r" (__USER_DS)) ++#else + #define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg) \ + asm volatile("1:\t" insn "\n" \ + "2:\t.section .fixup,\"ax\"\n" \ +@@ -36,8 +70,9 @@ + : "=&a" (oldval), "=&r" (ret), \ + "+m" (*uaddr), "=&r" (tem) \ + : "r" (oparg), "i" (-EFAULT), "1" (0)) ++#endif + +-static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) ++static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr) + { + int op = (encoded_op >> 28) & 7; + int cmp = (encoded_op >> 24) & 15; +@@ -61,11 +96,20 @@ static inline int futex_atomic_op_inuser + + switch (op) { + case FUTEX_OP_SET: ++#ifdef CONFIG_X86_32 ++ __futex_atomic_op1("xchgl %0, %%ds:%2", ret, oldval, uaddr, oparg); ++#else + __futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg); ++#endif + break; + case FUTEX_OP_ADD: ++#ifdef CONFIG_X86_32 ++ __futex_atomic_op1(LOCK_PREFIX "xaddl %0, %%ds:%2", ret, oldval, ++ uaddr, oparg); ++#else + __futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret, oldval, + uaddr, oparg); ++#endif + break; + case FUTEX_OP_OR: + __futex_atomic_op2("orl %4, %3", ret, oldval, uaddr, oparg); +@@ -109,7 +153,7 @@ static inline int futex_atomic_op_inuser + return ret; + } + +-static inline int futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, ++static inline int futex_atomic_cmpxchg_inatomic(u32 __user *uaddr, int oldval, + int newval) + { + +@@ -122,14 +166,27 @@ static inline int futex_atomic_cmpxchg_i + if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int))) + return -EFAULT; + +- asm volatile("1:\t" LOCK_PREFIX "cmpxchgl %3, %1\n" ++ asm volatile( ++#ifdef CONFIG_X86_32 ++ "\tmovw %w5, %%ds\n" ++ "1:\t" LOCK_PREFIX "cmpxchgl %3, %1\n" ++ "2:\tpushl %%ss\n" ++ "\tpopl %%ds\n" ++ "\t.section .fixup, \"ax\"\n" ++#else ++ "1:\t" LOCK_PREFIX "cmpxchgl %3, %1\n" + "2:\t.section .fixup, \"ax\"\n" ++#endif + "3:\tmov %2, %0\n" + "\tjmp 2b\n" + "\t.previous\n" + _ASM_EXTABLE(1b, 3b) + : "=a" (oldval), "+m" (*uaddr) ++#ifdef CONFIG_X86_32 ++ : "i" (-EFAULT), "r" (newval), "0" (oldval), "r" (__USER_DS) ++#else + : "i" (-EFAULT), "r" (newval), "0" (oldval) ++#endif + : "memory" + ); + +diff -urNp linux-2.6.31.1/arch/x86/include/asm/i387.h linux-2.6.31.1/arch/x86/include/asm/i387.h +--- linux-2.6.31.1/arch/x86/include/asm/i387.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/i387.h 2009-10-01 20:12:42.000000000 -0400 +@@ -194,13 +194,8 @@ static inline int fxrstor_checking(struc + } + + /* We need a safe address that is cheap to find and that is already +- in L1 during context switch. The best choices are unfortunately +- different for UP and SMP */ +-#ifdef CONFIG_SMP +-#define safe_address (__per_cpu_offset[0]) +-#else +-#define safe_address (kstat_cpu(0).cpustat.user) +-#endif ++ in L1 during context switch. */ ++#define safe_address (init_tss[smp_processor_id()].x86_tss.sp0) + + /* + * These must be called with preempt disabled +diff -urNp linux-2.6.31.1/arch/x86/include/asm/io_64.h linux-2.6.31.1/arch/x86/include/asm/io_64.h +--- linux-2.6.31.1/arch/x86/include/asm/io_64.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/io_64.h 2009-10-01 20:12:42.000000000 -0400 +@@ -140,6 +140,17 @@ __OUTS(l) + + #include + ++#define ARCH_HAS_VALID_PHYS_ADDR_RANGE ++static inline int valid_phys_addr_range (unsigned long addr, size_t count) ++{ ++ return ((addr + count + PAGE_SIZE - 1) >> PAGE_SHIFT) < (1 << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0; ++} ++ ++static inline int valid_mmap_phys_addr_range (unsigned long pfn, size_t count) ++{ ++ return (pfn + (count >> PAGE_SHIFT)) < (1 << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0; ++} ++ + #include + + void __memcpy_fromio(void *, unsigned long, unsigned); +diff -urNp linux-2.6.31.1/arch/x86/include/asm/irqflags.h linux-2.6.31.1/arch/x86/include/asm/irqflags.h +--- linux-2.6.31.1/arch/x86/include/asm/irqflags.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/irqflags.h 2009-10-01 20:12:42.000000000 -0400 +@@ -147,6 +147,8 @@ static inline unsigned long __raw_local_ + #define INTERRUPT_RETURN iret + #define ENABLE_INTERRUPTS_SYSEXIT sti; sysexit + #define GET_CR0_INTO_EAX movl %cr0, %eax ++#define GET_CR0_INTO_EDX movl %cr0, %edx ++#define SET_CR0_FROM_EDX movl %edx, %cr0 + #endif + + +diff -urNp linux-2.6.31.1/arch/x86/include/asm/kvm_host.h linux-2.6.31.1/arch/x86/include/asm/kvm_host.h +--- linux-2.6.31.1/arch/x86/include/asm/kvm_host.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/kvm_host.h 2009-10-01 20:12:42.000000000 -0400 +@@ -528,7 +528,7 @@ struct kvm_x86_ops { + u64 (*get_mt_mask)(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio); + }; + +-extern struct kvm_x86_ops *kvm_x86_ops; ++extern const struct kvm_x86_ops *kvm_x86_ops; + + int kvm_mmu_module_init(void); + void kvm_mmu_module_exit(void); +diff -urNp linux-2.6.31.1/arch/x86/include/asm/local.h linux-2.6.31.1/arch/x86/include/asm/local.h +--- linux-2.6.31.1/arch/x86/include/asm/local.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/local.h 2009-10-01 20:12:42.000000000 -0400 +@@ -18,26 +18,90 @@ typedef struct { + + static inline void local_inc(local_t *l) + { +- asm volatile(_ASM_INC "%0" ++ asm volatile(_ASM_INC "%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ _ASM_DEC "%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "+m" (l->a.counter)); + } + + static inline void local_dec(local_t *l) + { +- asm volatile(_ASM_DEC "%0" ++ asm volatile(_ASM_DEC "%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ _ASM_INC "%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "+m" (l->a.counter)); + } + + static inline void local_add(long i, local_t *l) + { +- asm volatile(_ASM_ADD "%1,%0" ++ asm volatile(_ASM_ADD "%1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ _ASM_SUB "%1,%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "+m" (l->a.counter) + : "ir" (i)); + } + + static inline void local_sub(long i, local_t *l) + { +- asm volatile(_ASM_SUB "%1,%0" ++ asm volatile(_ASM_SUB "%1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ _ASM_ADD "%1,%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "+m" (l->a.counter) + : "ir" (i)); + } +@@ -55,7 +119,24 @@ static inline int local_sub_and_test(lon + { + unsigned char c; + +- asm volatile(_ASM_SUB "%2,%0; sete %1" ++ asm volatile(_ASM_SUB "%2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ _ASM_ADD "%2,%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sete %1\n" + : "+m" (l->a.counter), "=qm" (c) + : "ir" (i) : "memory"); + return c; +@@ -73,7 +154,24 @@ static inline int local_dec_and_test(loc + { + unsigned char c; + +- asm volatile(_ASM_DEC "%0; sete %1" ++ asm volatile(_ASM_DEC "%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ _ASM_INC "%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sete %1\n" + : "+m" (l->a.counter), "=qm" (c) + : : "memory"); + return c != 0; +@@ -91,7 +189,24 @@ static inline int local_inc_and_test(loc + { + unsigned char c; + +- asm volatile(_ASM_INC "%0; sete %1" ++ asm volatile(_ASM_INC "%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ _ASM_DEC "%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sete %1\n" + : "+m" (l->a.counter), "=qm" (c) + : : "memory"); + return c != 0; +@@ -110,7 +225,24 @@ static inline int local_add_negative(lon + { + unsigned char c; + +- asm volatile(_ASM_ADD "%2,%0; sets %1" ++ asm volatile(_ASM_ADD "%2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ _ASM_SUB "%2,%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "sets %1\n" + : "+m" (l->a.counter), "=qm" (c) + : "ir" (i) : "memory"); + return c; +@@ -133,7 +265,23 @@ static inline long local_add_return(long + #endif + /* Modern 486+ processor */ + __i = i; +- asm volatile(_ASM_XADD "%0, %1;" ++ asm volatile(_ASM_XADD "%0, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ _ASM_MOV "%0,%1\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "+r" (i), "+m" (l->a.counter) + : : "memory"); + return i + __i; +diff -urNp linux-2.6.31.1/arch/x86/include/asm/mman.h linux-2.6.31.1/arch/x86/include/asm/mman.h +--- linux-2.6.31.1/arch/x86/include/asm/mman.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/mman.h 2009-10-01 20:12:42.000000000 -0400 +@@ -17,4 +17,14 @@ + #define MCL_CURRENT 1 /* lock all current mappings */ + #define MCL_FUTURE 2 /* lock all future mappings */ + ++#ifdef __KERNEL__ ++#ifndef __ASSEMBLY__ ++#ifdef CONFIG_X86_32 ++#define arch_mmap_check i386_mmap_check ++int i386_mmap_check(unsigned long addr, unsigned long len, ++ unsigned long flags); ++#endif ++#endif ++#endif ++ + #endif /* _ASM_X86_MMAN_H */ +diff -urNp linux-2.6.31.1/arch/x86/include/asm/mmu_context.h linux-2.6.31.1/arch/x86/include/asm/mmu_context.h +--- linux-2.6.31.1/arch/x86/include/asm/mmu_context.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/mmu_context.h 2009-10-01 20:12:42.000000000 -0400 +@@ -34,11 +34,17 @@ static inline void switch_mm(struct mm_s + struct task_struct *tsk) + { + unsigned cpu = smp_processor_id(); ++#if defined(CONFIG_X86_32) && defined(CONFIG_SMP) ++ int tlbstate = TLBSTATE_OK; ++#endif + + if (likely(prev != next)) { + /* stop flush ipis for the previous mm */ + cpu_clear(cpu, prev->cpu_vm_mask); + #ifdef CONFIG_SMP ++#ifdef CONFIG_X86_32 ++ tlbstate = percpu_read(cpu_tlbstate.state); ++#endif + percpu_write(cpu_tlbstate.state, TLBSTATE_OK); + percpu_write(cpu_tlbstate.active_mm, next); + #endif +@@ -52,6 +58,26 @@ static inline void switch_mm(struct mm_s + */ + if (unlikely(prev->context.ldt != next->context.ldt)) + load_LDT_nolock(&next->context); ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP) ++ if (!nx_enabled) { ++ smp_mb__before_clear_bit(); ++ cpu_clear(cpu, prev->context.cpu_user_cs_mask); ++ smp_mb__after_clear_bit(); ++ cpu_set(cpu, next->context.cpu_user_cs_mask); ++ } ++#endif ++ ++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)) ++ if (unlikely(prev->context.user_cs_base != next->context.user_cs_base || ++ prev->context.user_cs_limit != next->context.user_cs_limit ++#ifdef CONFIG_SMP ++ || tlbstate != TLBSTATE_OK ++#endif ++ )) ++ set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu); ++#endif ++ + } + #ifdef CONFIG_SMP + else { +@@ -65,6 +91,19 @@ static inline void switch_mm(struct mm_s + */ + load_cr3(next->pgd); + load_LDT_nolock(&next->context); ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) ++ if (!nx_enabled) ++ cpu_set(cpu, next->context.cpu_user_cs_mask); ++#endif ++ ++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)) ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!((next->pax_flags & MF_PAX_PAGEEXEC) && nx_enabled)) ++#endif ++ set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu); ++#endif ++ + } + } + #endif +diff -urNp linux-2.6.31.1/arch/x86/include/asm/mmu.h linux-2.6.31.1/arch/x86/include/asm/mmu.h +--- linux-2.6.31.1/arch/x86/include/asm/mmu.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/mmu.h 2009-10-01 20:12:42.000000000 -0400 +@@ -9,10 +9,23 @@ + * we put the segment information here. + */ + typedef struct { +- void *ldt; ++ struct desc_struct *ldt; + int size; + struct mutex lock; +- void *vdso; ++ unsigned long vdso; ++ ++#ifdef CONFIG_X86_32 ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ unsigned long user_cs_base; ++ unsigned long user_cs_limit; ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP) ++ cpumask_t cpu_user_cs_mask; ++#endif ++ ++#endif ++#endif ++ + } mm_context_t; + + #ifdef CONFIG_SMP +diff -urNp linux-2.6.31.1/arch/x86/include/asm/module.h linux-2.6.31.1/arch/x86/include/asm/module.h +--- linux-2.6.31.1/arch/x86/include/asm/module.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/module.h 2009-10-01 20:12:42.000000000 -0400 +@@ -74,7 +74,12 @@ struct mod_arch_specific {}; + # else + # define MODULE_STACKSIZE "" + # endif +-# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE ++# ifdef CONFIG_GRKERNSEC ++# define MODULE_GRSEC "GRSECURITY " ++# else ++# define MODULE_GRSEC "" ++# endif ++# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE MODULE_GRSEC + #endif + + #endif /* _ASM_X86_MODULE_H */ +diff -urNp linux-2.6.31.1/arch/x86/include/asm/page_32_types.h linux-2.6.31.1/arch/x86/include/asm/page_32_types.h +--- linux-2.6.31.1/arch/x86/include/asm/page_32_types.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/page_32_types.h 2009-10-01 20:12:42.000000000 -0400 +@@ -15,6 +15,10 @@ + */ + #define __PAGE_OFFSET _AC(CONFIG_PAGE_OFFSET, UL) + ++#ifdef CONFIG_PAX_PAGEEXEC ++#define CONFIG_ARCH_TRACK_EXEC_LIMIT 1 ++#endif ++ + #ifdef CONFIG_4KSTACKS + #define THREAD_ORDER 0 + #else +diff -urNp linux-2.6.31.1/arch/x86/include/asm/page_64_types.h linux-2.6.31.1/arch/x86/include/asm/page_64_types.h +--- linux-2.6.31.1/arch/x86/include/asm/page_64_types.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/page_64_types.h 2009-10-01 20:12:42.000000000 -0400 +@@ -39,6 +39,9 @@ + #define __START_KERNEL (__START_KERNEL_map + __PHYSICAL_START) + #define __START_KERNEL_map _AC(0xffffffff80000000, UL) + ++#define ktla_ktva(addr) (addr) ++#define ktva_ktla(addr) (addr) ++ + /* See Documentation/x86/x86_64/mm.txt for a description of the memory map. */ + #define __PHYSICAL_MASK_SHIFT 46 + #define __VIRTUAL_MASK_SHIFT 47 +diff -urNp linux-2.6.31.1/arch/x86/include/asm/paravirt.h linux-2.6.31.1/arch/x86/include/asm/paravirt.h +--- linux-2.6.31.1/arch/x86/include/asm/paravirt.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/paravirt.h 2009-10-01 20:12:42.000000000 -0400 +@@ -1688,7 +1688,7 @@ static inline unsigned long __raw_local_ + + #define PARA_PATCH(struct, off) ((PARAVIRT_PATCH_##struct + (off)) / 4) + #define PARA_SITE(ptype, clobbers, ops) _PVSITE(ptype, clobbers, ops, .long, 4) +-#define PARA_INDIRECT(addr) *%cs:addr ++#define PARA_INDIRECT(addr) *%ss:addr + #endif + + #define INTERRUPT_RETURN \ +@@ -1718,6 +1718,18 @@ static inline unsigned long __raw_local_ + call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0); \ + pop %edx; pop %ecx + ++#define GET_CR0_INTO_EDX \ ++ push %eax; push %ecx; \ ++ call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0); \ ++ mov %eax, %edx; \ ++ pop %ecx; pop %eax ++ ++#define SET_CR0_FROM_EDX \ ++ push %eax; push %ecx; \ ++ mov %edx, %eax; \ ++ call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0);\ ++ pop %ecx; pop %eax ++ + #define ENABLE_INTERRUPTS_SYSEXIT \ + PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_irq_enable_sysexit), \ + CLBR_NONE, \ +diff -urNp linux-2.6.31.1/arch/x86/include/asm/pgalloc.h linux-2.6.31.1/arch/x86/include/asm/pgalloc.h +--- linux-2.6.31.1/arch/x86/include/asm/pgalloc.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/pgalloc.h 2009-10-01 20:12:42.000000000 -0400 +@@ -58,6 +58,13 @@ static inline void pmd_populate_kernel(s + pmd_t *pmd, pte_t *pte) + { + paravirt_alloc_pte(mm, __pa(pte) >> PAGE_SHIFT); ++ set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE)); ++} ++ ++static inline void pmd_populate_user(struct mm_struct *mm, ++ pmd_t *pmd, pte_t *pte) ++{ ++ paravirt_alloc_pte(mm, __pa(pte) >> PAGE_SHIFT); + set_pmd(pmd, __pmd(__pa(pte) | _PAGE_TABLE)); + } + +diff -urNp linux-2.6.31.1/arch/x86/include/asm/pgtable-2level.h linux-2.6.31.1/arch/x86/include/asm/pgtable-2level.h +--- linux-2.6.31.1/arch/x86/include/asm/pgtable-2level.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/pgtable-2level.h 2009-10-01 20:12:42.000000000 -0400 +@@ -18,7 +18,19 @@ static inline void native_set_pte(pte_t + + static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + *pmdp = pmd; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte) +diff -urNp linux-2.6.31.1/arch/x86/include/asm/pgtable_32.h linux-2.6.31.1/arch/x86/include/asm/pgtable_32.h +--- linux-2.6.31.1/arch/x86/include/asm/pgtable_32.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/pgtable_32.h 2009-10-01 20:12:42.000000000 -0400 +@@ -26,8 +26,6 @@ + struct mm_struct; + struct vm_area_struct; + +-extern pgd_t swapper_pg_dir[1024]; +- + static inline void pgtable_cache_init(void) { } + static inline void check_pgt_cache(void) { } + void paging_init(void); +@@ -48,6 +46,11 @@ extern void set_pmd_pfn(unsigned long, u + # include + #endif + ++extern pgd_t swapper_pg_dir[PTRS_PER_PGD]; ++#ifdef CONFIG_X86_PAE ++extern pmd_t swapper_pm_dir[PTRS_PER_PGD][PTRS_PER_PMD]; ++#endif ++ + #if defined(CONFIG_HIGHPTE) + #define __KM_PTE \ + (in_nmi() ? KM_NMI_PTE : \ +@@ -84,6 +87,9 @@ do { \ + + #endif /* !__ASSEMBLY__ */ + ++#define HAVE_ARCH_UNMAPPED_AREA ++#define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN ++ + /* + * kern_addr_valid() is (1) for FLATMEM and (0) for + * SPARSEMEM and DISCONTIGMEM +diff -urNp linux-2.6.31.1/arch/x86/include/asm/pgtable_32_types.h linux-2.6.31.1/arch/x86/include/asm/pgtable_32_types.h +--- linux-2.6.31.1/arch/x86/include/asm/pgtable_32_types.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/pgtable_32_types.h 2009-10-01 20:12:42.000000000 -0400 +@@ -8,7 +8,7 @@ + */ + #ifdef CONFIG_X86_PAE + # include +-# define PMD_SIZE (1UL << PMD_SHIFT) ++# define PMD_SIZE (_AC(1, UL) << PMD_SHIFT) + # define PMD_MASK (~(PMD_SIZE - 1)) + #else + # include +@@ -46,6 +46,19 @@ extern bool __vmalloc_start_set; /* set + # define VMALLOC_END (FIXADDR_START - 2 * PAGE_SIZE) + #endif + ++#ifdef CONFIG_PAX_KERNEXEC ++#ifndef __ASSEMBLY__ ++extern unsigned char MODULES_EXEC_VADDR[]; ++extern unsigned char MODULES_EXEC_END[]; ++#endif ++#include ++#define ktla_ktva(addr) (addr + LOAD_PHYSICAL_ADDR + PAGE_OFFSET) ++#define ktva_ktla(addr) (addr - LOAD_PHYSICAL_ADDR - PAGE_OFFSET) ++#else ++#define ktla_ktva(addr) (addr) ++#define ktva_ktla(addr) (addr) ++#endif ++ + #define MODULES_VADDR VMALLOC_START + #define MODULES_END VMALLOC_END + #define MODULES_LEN (MODULES_VADDR - MODULES_END) +diff -urNp linux-2.6.31.1/arch/x86/include/asm/pgtable-3level.h linux-2.6.31.1/arch/x86/include/asm/pgtable-3level.h +--- linux-2.6.31.1/arch/x86/include/asm/pgtable-3level.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/pgtable-3level.h 2009-10-01 20:12:42.000000000 -0400 +@@ -38,12 +38,36 @@ static inline void native_set_pte_atomic + + static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + set_64bit((unsigned long long *)(pmdp), native_pmd_val(pmd)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + static inline void native_set_pud(pud_t *pudp, pud_t pud) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + set_64bit((unsigned long long *)(pudp), native_pud_val(pud)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + /* +diff -urNp linux-2.6.31.1/arch/x86/include/asm/pgtable_64.h linux-2.6.31.1/arch/x86/include/asm/pgtable_64.h +--- linux-2.6.31.1/arch/x86/include/asm/pgtable_64.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/pgtable_64.h 2009-10-01 20:12:42.000000000 -0400 +@@ -16,9 +16,12 @@ + + extern pud_t level3_kernel_pgt[512]; + extern pud_t level3_ident_pgt[512]; ++extern pud_t level3_vmalloc_pgt[512]; ++extern pud_t level3_vmemmap_pgt[512]; ++extern pud_t level2_vmemmap_pgt[512]; + extern pmd_t level2_kernel_pgt[512]; + extern pmd_t level2_fixmap_pgt[512]; +-extern pmd_t level2_ident_pgt[512]; ++extern pmd_t level2_ident_pgt[512*4]; + extern pgd_t init_level4_pgt[]; + + #define swapper_pg_dir init_level4_pgt +@@ -74,7 +77,19 @@ static inline pte_t native_ptep_get_and_ + + static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + *pmdp = pmd; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + static inline void native_pmd_clear(pmd_t *pmd) +diff -urNp linux-2.6.31.1/arch/x86/include/asm/pgtable.h linux-2.6.31.1/arch/x86/include/asm/pgtable.h +--- linux-2.6.31.1/arch/x86/include/asm/pgtable.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/pgtable.h 2009-10-01 20:12:42.000000000 -0400 +@@ -90,6 +90,11 @@ static inline void __init paravirt_paget + * The following only work if pte_present() is true. + * Undefined behaviour if not.. + */ ++static inline int pte_user(pte_t pte) ++{ ++ return pte_val(pte) & _PAGE_USER; ++} ++ + static inline int pte_dirty(pte_t pte) + { + return pte_flags(pte) & _PAGE_DIRTY; +@@ -172,9 +177,29 @@ static inline pte_t pte_wrprotect(pte_t + return pte_clear_flags(pte, _PAGE_RW); + } + ++static inline pte_t pte_mkread(pte_t pte) ++{ ++ return __pte(pte_val(pte) | _PAGE_USER); ++} ++ + static inline pte_t pte_mkexec(pte_t pte) + { +- return pte_clear_flags(pte, _PAGE_NX); ++#ifdef CONFIG_X86_PAE ++ if (__supported_pte_mask & _PAGE_NX) ++ return pte_clear_flags(pte, _PAGE_NX); ++ else ++#endif ++ return pte_set_flags(pte, _PAGE_USER); ++} ++ ++static inline pte_t pte_exprotect(pte_t pte) ++{ ++#ifdef CONFIG_X86_PAE ++ if (__supported_pte_mask & _PAGE_NX) ++ return pte_set_flags(pte, _PAGE_NX); ++ else ++#endif ++ return pte_clear_flags(pte, _PAGE_USER); + } + + static inline pte_t pte_mkdirty(pte_t pte) +@@ -482,7 +507,7 @@ static inline pud_t *pud_offset(pgd_t *p + + static inline int pgd_bad(pgd_t pgd) + { +- return (pgd_flags(pgd) & ~_PAGE_USER) != _KERNPG_TABLE; ++ return (pgd_flags(pgd) & ~(_PAGE_USER | _PAGE_NX)) != _KERNPG_TABLE; + } + + static inline int pgd_none(pgd_t pgd) +@@ -623,7 +648,19 @@ static inline void ptep_set_wrprotect(st + */ + static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count) + { +- memcpy(dst, src, count * sizeof(pgd_t)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ ++ memcpy(dst, src, count * sizeof(pgd_t)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + +diff -urNp linux-2.6.31.1/arch/x86/include/asm/pgtable_types.h linux-2.6.31.1/arch/x86/include/asm/pgtable_types.h +--- linux-2.6.31.1/arch/x86/include/asm/pgtable_types.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/pgtable_types.h 2009-10-01 20:12:42.000000000 -0400 +@@ -16,12 +16,11 @@ + #define _PAGE_BIT_PSE 7 /* 4 MB (or 2MB) page */ + #define _PAGE_BIT_PAT 7 /* on 4KB pages */ + #define _PAGE_BIT_GLOBAL 8 /* Global TLB entry PPro+ */ +-#define _PAGE_BIT_UNUSED1 9 /* available for programmer */ ++#define _PAGE_BIT_SPECIAL 9 /* special mappings, no associated struct page */ + #define _PAGE_BIT_IOMAP 10 /* flag used to indicate IO mapping */ + #define _PAGE_BIT_HIDDEN 11 /* hidden by kmemcheck */ + #define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */ +-#define _PAGE_BIT_SPECIAL _PAGE_BIT_UNUSED1 +-#define _PAGE_BIT_CPA_TEST _PAGE_BIT_UNUSED1 ++#define _PAGE_BIT_CPA_TEST _PAGE_BIT_SPECIAL + #define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */ + + /* If _PAGE_BIT_PRESENT is clear, we use these: */ +@@ -39,7 +38,6 @@ + #define _PAGE_DIRTY (_AT(pteval_t, 1) << _PAGE_BIT_DIRTY) + #define _PAGE_PSE (_AT(pteval_t, 1) << _PAGE_BIT_PSE) + #define _PAGE_GLOBAL (_AT(pteval_t, 1) << _PAGE_BIT_GLOBAL) +-#define _PAGE_UNUSED1 (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED1) + #define _PAGE_IOMAP (_AT(pteval_t, 1) << _PAGE_BIT_IOMAP) + #define _PAGE_PAT (_AT(pteval_t, 1) << _PAGE_BIT_PAT) + #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE) +@@ -55,8 +53,10 @@ + + #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) + #define _PAGE_NX (_AT(pteval_t, 1) << _PAGE_BIT_NX) +-#else ++#elif defined(CONFIG_KMEMCHECK) + #define _PAGE_NX (_AT(pteval_t, 0)) ++#else ++#define _PAGE_NX (_AT(pteval_t, 1) << _PAGE_BIT_HIDDEN) + #endif + + #define _PAGE_FILE (_AT(pteval_t, 1) << _PAGE_BIT_FILE) +@@ -93,6 +93,9 @@ + #define PAGE_READONLY_EXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | \ + _PAGE_ACCESSED) + ++#define PAGE_READONLY_NOEXEC PAGE_READONLY ++#define PAGE_SHARED_NOEXEC PAGE_SHARED ++ + #define __PAGE_KERNEL_EXEC \ + (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_GLOBAL) + #define __PAGE_KERNEL (__PAGE_KERNEL_EXEC | _PAGE_NX) +@@ -103,8 +106,8 @@ + #define __PAGE_KERNEL_WC (__PAGE_KERNEL | _PAGE_CACHE_WC) + #define __PAGE_KERNEL_NOCACHE (__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT) + #define __PAGE_KERNEL_UC_MINUS (__PAGE_KERNEL | _PAGE_PCD) +-#define __PAGE_KERNEL_VSYSCALL (__PAGE_KERNEL_RX | _PAGE_USER) +-#define __PAGE_KERNEL_VSYSCALL_NOCACHE (__PAGE_KERNEL_VSYSCALL | _PAGE_PCD | _PAGE_PWT) ++#define __PAGE_KERNEL_VSYSCALL (__PAGE_KERNEL_RO | _PAGE_USER) ++#define __PAGE_KERNEL_VSYSCALL_NOCACHE (__PAGE_KERNEL_RO | _PAGE_PCD | _PAGE_PWT | _PAGE_USER) + #define __PAGE_KERNEL_LARGE (__PAGE_KERNEL | _PAGE_PSE) + #define __PAGE_KERNEL_LARGE_NOCACHE (__PAGE_KERNEL | _PAGE_CACHE_UC | _PAGE_PSE) + #define __PAGE_KERNEL_LARGE_EXEC (__PAGE_KERNEL_EXEC | _PAGE_PSE) +@@ -163,8 +166,8 @@ + * bits are combined, this will alow user to access the high address mapped + * VDSO in the presence of CONFIG_COMPAT_VDSO + */ +-#define PTE_IDENT_ATTR 0x003 /* PRESENT+RW */ +-#define PDE_IDENT_ATTR 0x067 /* PRESENT+RW+USER+DIRTY+ACCESSED */ ++#define PTE_IDENT_ATTR 0x063 /* PRESENT+RW+DIRTY+ACCESSED */ ++#define PDE_IDENT_ATTR 0x063 /* PRESENT+RW+DIRTY+ACCESSED */ + #define PGD_IDENT_ATTR 0x001 /* PRESENT (no other attributes) */ + #endif + +@@ -277,7 +280,15 @@ static inline pteval_t pte_flags(pte_t p + typedef struct page *pgtable_t; + + extern pteval_t __supported_pte_mask; ++#ifdef CONFIG_X86_32 ++#ifdef CONFIG_X86_PAE + extern int nx_enabled; ++#else ++#define nx_enabled (0) ++#endif ++#else ++#define nx_enabled (1) ++#endif + + #define pgprot_writecombine pgprot_writecombine + extern pgprot_t pgprot_writecombine(pgprot_t prot); +diff -urNp linux-2.6.31.1/arch/x86/include/asm/processor.h linux-2.6.31.1/arch/x86/include/asm/processor.h +--- linux-2.6.31.1/arch/x86/include/asm/processor.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/processor.h 2009-10-01 20:12:42.000000000 -0400 +@@ -271,7 +271,7 @@ struct tss_struct { + + } ____cacheline_aligned; + +-DECLARE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss); ++extern struct tss_struct init_tss[NR_CPUS]; + + /* + * Save the original ist values for checking stack pointers during debugging +@@ -900,8 +900,17 @@ static inline void spin_lock_prefetch(co + */ + #define TASK_SIZE PAGE_OFFSET + #define TASK_SIZE_MAX TASK_SIZE ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++#define SEGMEXEC_TASK_SIZE (TASK_SIZE / 2) ++#endif ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++#define STACK_TOP ((current->mm->pax_flags & MF_PAX_SEGMEXEC)?SEGMEXEC_TASK_SIZE:TASK_SIZE) ++#else + #define STACK_TOP TASK_SIZE +-#define STACK_TOP_MAX STACK_TOP ++#endif ++#define STACK_TOP_MAX TASK_SIZE + + #define INIT_THREAD { \ + .sp0 = sizeof(init_stack) + (long)&init_stack, \ +@@ -918,7 +927,7 @@ static inline void spin_lock_prefetch(co + */ + #define INIT_TSS { \ + .x86_tss = { \ +- .sp0 = sizeof(init_stack) + (long)&init_stack, \ ++ .sp0 = sizeof(init_stack) + (long)&init_stack - 8, \ + .ss0 = __KERNEL_DS, \ + .ss1 = __KERNEL_CS, \ + .io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \ +@@ -929,11 +938,7 @@ static inline void spin_lock_prefetch(co + extern unsigned long thread_saved_pc(struct task_struct *tsk); + + #define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long)) +-#define KSTK_TOP(info) \ +-({ \ +- unsigned long *__ptr = (unsigned long *)(info); \ +- (unsigned long)(&__ptr[THREAD_SIZE_LONGS]); \ +-}) ++#define KSTK_TOP(info) ((info)->task.thread.sp0) + + /* + * The below -8 is to reserve 8 bytes on top of the ring0 stack. +@@ -948,7 +953,7 @@ extern unsigned long thread_saved_pc(str + #define task_pt_regs(task) \ + ({ \ + struct pt_regs *__regs__; \ +- __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \ ++ __regs__ = (struct pt_regs *)((task)->thread.sp0); \ + __regs__ - 1; \ + }) + +@@ -964,7 +969,7 @@ extern unsigned long thread_saved_pc(str + * space during mmap's. + */ + #define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? \ +- 0xc0000000 : 0xFFFFe000) ++ 0xc0000000 : 0xFFFFf000) + + #define TASK_SIZE (test_thread_flag(TIF_IA32) ? \ + IA32_PAGE_OFFSET : TASK_SIZE_MAX) +@@ -1001,6 +1006,10 @@ extern void start_thread(struct pt_regs + */ + #define TASK_UNMAPPED_BASE (PAGE_ALIGN(TASK_SIZE / 3)) + ++#ifdef CONFIG_PAX_SEGMEXEC ++#define SEGMEXEC_TASK_UNMAPPED_BASE (PAGE_ALIGN(SEGMEXEC_TASK_SIZE / 3)) ++#endif ++ + #define KSTK_EIP(task) (task_pt_regs(task)->ip) + + /* Get/set a process' ability to use the timestamp counter instruction */ +diff -urNp linux-2.6.31.1/arch/x86/include/asm/ptrace.h linux-2.6.31.1/arch/x86/include/asm/ptrace.h +--- linux-2.6.31.1/arch/x86/include/asm/ptrace.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/ptrace.h 2009-10-01 20:12:42.000000000 -0400 +@@ -151,28 +151,29 @@ static inline unsigned long regs_return_ + } + + /* +- * user_mode_vm(regs) determines whether a register set came from user mode. ++ * user_mode(regs) determines whether a register set came from user mode. + * This is true if V8086 mode was enabled OR if the register set was from + * protected mode with RPL-3 CS value. This tricky test checks that with + * one comparison. Many places in the kernel can bypass this full check +- * if they have already ruled out V8086 mode, so user_mode(regs) can be used. ++ * if they have already ruled out V8086 mode, so user_mode_novm(regs) can ++ * be used. + */ +-static inline int user_mode(struct pt_regs *regs) ++static inline int user_mode_novm(struct pt_regs *regs) + { + #ifdef CONFIG_X86_32 + return (regs->cs & SEGMENT_RPL_MASK) == USER_RPL; + #else +- return !!(regs->cs & 3); ++ return !!(regs->cs & SEGMENT_RPL_MASK); + #endif + } + +-static inline int user_mode_vm(struct pt_regs *regs) ++static inline int user_mode(struct pt_regs *regs) + { + #ifdef CONFIG_X86_32 + return ((regs->cs & SEGMENT_RPL_MASK) | (regs->flags & X86_VM_MASK)) >= + USER_RPL; + #else +- return user_mode(regs); ++ return user_mode_novm(regs); + #endif + } + +diff -urNp linux-2.6.31.1/arch/x86/include/asm/reboot.h linux-2.6.31.1/arch/x86/include/asm/reboot.h +--- linux-2.6.31.1/arch/x86/include/asm/reboot.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/reboot.h 2009-10-01 20:12:42.000000000 -0400 +@@ -18,7 +18,7 @@ extern struct machine_ops machine_ops; + + void native_machine_crash_shutdown(struct pt_regs *regs); + void native_machine_shutdown(void); +-void machine_real_restart(const unsigned char *code, int length); ++void machine_real_restart(const unsigned char *code, unsigned int length); + + typedef void (*nmi_shootdown_cb)(int, struct die_args*); + void nmi_shootdown_cpus(nmi_shootdown_cb callback); +diff -urNp linux-2.6.31.1/arch/x86/include/asm/rwsem.h linux-2.6.31.1/arch/x86/include/asm/rwsem.h +--- linux-2.6.31.1/arch/x86/include/asm/rwsem.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/rwsem.h 2009-10-01 20:12:42.000000000 -0400 +@@ -106,10 +106,26 @@ static inline void __down_read(struct rw + { + asm volatile("# beginning down_read\n\t" + LOCK_PREFIX " incl (%%eax)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ LOCK_PREFIX "decl (%%eax)\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + /* adds 0x00000001, returns the old value */ +- " jns 1f\n" ++ " jns 2f\n" + " call call_rwsem_down_read_failed\n" +- "1:\n\t" ++ "2:\n\t" + "# ending down_read\n\t" + : "+m" (sem->count) + : "a" (sem) +@@ -124,13 +140,29 @@ static inline int __down_read_trylock(st + __s32 result, tmp; + asm volatile("# beginning __down_read_trylock\n\t" + " movl %0,%1\n\t" +- "1:\n\t" ++ "2:\n\t" + " movl %1,%2\n\t" + " addl %3,%2\n\t" +- " jle 2f\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ "subl %3,%2\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ " jle 3f\n\t" + LOCK_PREFIX " cmpxchgl %2,%0\n\t" +- " jnz 1b\n\t" +- "2:\n\t" ++ " jnz 2b\n\t" ++ "3:\n\t" + "# ending __down_read_trylock\n\t" + : "+m" (sem->count), "=&a" (result), "=&r" (tmp) + : "i" (RWSEM_ACTIVE_READ_BIAS) +@@ -148,12 +180,28 @@ static inline void __down_write_nested(s + tmp = RWSEM_ACTIVE_WRITE_BIAS; + asm volatile("# beginning down_write\n\t" + LOCK_PREFIX " xadd %%edx,(%%eax)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ "movl %%edx,(%%eax)\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + /* subtract 0x0000ffff, returns the old value */ + " testl %%edx,%%edx\n\t" + /* was the count 0 before? */ +- " jz 1f\n" ++ " jz 2f\n" + " call call_rwsem_down_write_failed\n" +- "1:\n" ++ "2:\n" + "# ending down_write" + : "+m" (sem->count), "=d" (tmp) + : "a" (sem), "1" (tmp) +@@ -186,10 +234,26 @@ static inline void __up_read(struct rw_s + __s32 tmp = -RWSEM_ACTIVE_READ_BIAS; + asm volatile("# beginning __up_read\n\t" + LOCK_PREFIX " xadd %%edx,(%%eax)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ "movl %%edx,(%%eax)\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + /* subtracts 1, returns the old value */ +- " jns 1f\n\t" ++ " jns 2f\n\t" + " call call_rwsem_wake\n" +- "1:\n" ++ "2:\n" + "# ending __up_read\n" + : "+m" (sem->count), "=d" (tmp) + : "a" (sem), "1" (tmp) +@@ -204,11 +268,27 @@ static inline void __up_write(struct rw_ + asm volatile("# beginning __up_write\n\t" + " movl %2,%%edx\n\t" + LOCK_PREFIX " xaddl %%edx,(%%eax)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ "movl %%edx,(%%eax)\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + /* tries to transition + 0xffff0001 -> 0x00000000 */ +- " jz 1f\n" ++ " jz 2f\n" + " call call_rwsem_wake\n" +- "1:\n\t" ++ "2:\n\t" + "# ending __up_write\n" + : "+m" (sem->count) + : "a" (sem), "i" (-RWSEM_ACTIVE_WRITE_BIAS) +@@ -222,10 +302,26 @@ static inline void __downgrade_write(str + { + asm volatile("# beginning __downgrade_write\n\t" + LOCK_PREFIX " addl %2,(%%eax)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ LOCK_PREFIX "subl %2,(%%eax)\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + /* transitions 0xZZZZ0001 -> 0xYYYY0001 */ +- " jns 1f\n\t" ++ " jns 2f\n\t" + " call call_rwsem_downgrade_wake\n" +- "1:\n\t" ++ "2:\n\t" + "# ending __downgrade_write\n" + : "+m" (sem->count) + : "a" (sem), "i" (-RWSEM_WAITING_BIAS) +@@ -237,7 +333,23 @@ static inline void __downgrade_write(str + */ + static inline void rwsem_atomic_add(int delta, struct rw_semaphore *sem) + { +- asm volatile(LOCK_PREFIX "addl %1,%0" ++ asm volatile(LOCK_PREFIX "addl %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ LOCK_PREFIX "subl %1,%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "+m" (sem->count) + : "ir" (delta)); + } +@@ -249,7 +361,23 @@ static inline int rwsem_atomic_update(in + { + int tmp = delta; + +- asm volatile(LOCK_PREFIX "xadd %0,%1" ++ asm volatile(LOCK_PREFIX "xadd %0,%1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ "movl %0,%1\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "+r" (tmp), "+m" (sem->count) + : : "memory"); + +diff -urNp linux-2.6.31.1/arch/x86/include/asm/segment.h linux-2.6.31.1/arch/x86/include/asm/segment.h +--- linux-2.6.31.1/arch/x86/include/asm/segment.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/segment.h 2009-10-01 20:12:42.000000000 -0400 +@@ -88,7 +88,7 @@ + #define GDT_ENTRY_ESPFIX_SS (GDT_ENTRY_KERNEL_BASE + 14) + #define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8) + +-#define GDT_ENTRY_PERCPU (GDT_ENTRY_KERNEL_BASE + 15) ++#define GDT_ENTRY_PERCPU (GDT_ENTRY_KERNEL_BASE + 15) + #ifdef CONFIG_SMP + #define __KERNEL_PERCPU (GDT_ENTRY_PERCPU * 8) + #else +@@ -102,6 +102,12 @@ + #define __KERNEL_STACK_CANARY 0 + #endif + ++#define GDT_ENTRY_PCIBIOS_CS (GDT_ENTRY_KERNEL_BASE + 17) ++#define __PCIBIOS_CS (GDT_ENTRY_PCIBIOS_CS * 8) ++ ++#define GDT_ENTRY_PCIBIOS_DS (GDT_ENTRY_KERNEL_BASE + 18) ++#define __PCIBIOS_DS (GDT_ENTRY_PCIBIOS_DS * 8) ++ + #define GDT_ENTRY_DOUBLEFAULT_TSS 31 + + /* +@@ -139,7 +145,7 @@ + */ + + /* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */ +-#define SEGMENT_IS_PNP_CODE(x) (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8) ++#define SEGMENT_IS_PNP_CODE(x) (((x) & 0xFFFCU) == PNP_CS32 || ((x) & 0xFFFCU) == PNP_CS16) + + + #else +diff -urNp linux-2.6.31.1/arch/x86/include/asm/spinlock.h linux-2.6.31.1/arch/x86/include/asm/spinlock.h +--- linux-2.6.31.1/arch/x86/include/asm/spinlock.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/spinlock.h 2009-10-01 20:12:42.000000000 -0400 +@@ -249,18 +249,50 @@ static inline int __raw_write_can_lock(r + static inline void __raw_read_lock(raw_rwlock_t *rw) + { + asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" +- "jns 1f\n" +- "call __read_lock_failed\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" + "1:\n" ++ LOCK_PREFIX " addl $1,(%0)\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "jns 2f\n" ++ "call __read_lock_failed\n\t" ++ "2:\n" + ::LOCK_PTR_REG (rw) : "memory"); + } + + static inline void __raw_write_lock(raw_rwlock_t *rw) + { + asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t" +- "jz 1f\n" +- "call __write_lock_failed\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" + "1:\n" ++ LOCK_PREFIX " addl %1,(%0)\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ "jz 2f\n" ++ "call __write_lock_failed\n\t" ++ "2:\n" + ::LOCK_PTR_REG (rw), "i" (RW_LOCK_BIAS) : "memory"); + } + +@@ -286,12 +318,45 @@ static inline int __raw_write_trylock(ra + + static inline void __raw_read_unlock(raw_rwlock_t *rw) + { +- asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory"); ++ asm volatile(LOCK_PREFIX "incl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ LOCK_PREFIX "decl %0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ ++ :"+m" (rw->lock) : : "memory"); + } + + static inline void __raw_write_unlock(raw_rwlock_t *rw) + { +- asm volatile(LOCK_PREFIX "addl %1, %0" ++ asm volatile(LOCK_PREFIX "addl %1, %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#ifdef CONFIG_X86_32 ++ "into\n0:\n" ++#else ++ "jno 0f\n" ++ "int $4\n0:\n" ++#endif ++ ".pushsection .fixup,\"ax\"\n" ++ "1:\n" ++ LOCK_PREFIX "subl %1,%0\n" ++ "jmp 0b\n" ++ ".popsection\n" ++ _ASM_EXTABLE(0b, 1b) ++#endif ++ + : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory"); + } + +diff -urNp linux-2.6.31.1/arch/x86/include/asm/system.h linux-2.6.31.1/arch/x86/include/asm/system.h +--- linux-2.6.31.1/arch/x86/include/asm/system.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/system.h 2009-10-01 20:12:42.000000000 -0400 +@@ -227,7 +227,7 @@ static inline unsigned long get_limit(un + { + unsigned long __limit; + asm("lsll %1,%0" : "=r" (__limit) : "r" (segment)); +- return __limit + 1; ++ return __limit; + } + + static inline void native_clts(void) +@@ -353,6 +353,23 @@ static inline void native_wbinvd(void) + + #define stts() write_cr0(read_cr0() | X86_CR0_TS) + ++#define pax_open_kernel(cr0) \ ++do { \ ++ typecheck(unsigned long, cr0); \ ++ preempt_disable(); \ ++ barrier(); \ ++ cr0 = read_cr0(); \ ++ write_cr0(cr0 & ~X86_CR0_WP); \ ++} while (0) ++ ++#define pax_close_kernel(cr0) \ ++do { \ ++ typecheck(unsigned long, cr0); \ ++ write_cr0(cr0); \ ++ barrier(); \ ++ preempt_enable_no_resched(); \ ++} while (0) ++ + #endif /* __KERNEL__ */ + + static inline void clflush(volatile void *__p) +@@ -367,7 +384,7 @@ void enable_hlt(void); + + void cpu_idle_wait(void); + +-extern unsigned long arch_align_stack(unsigned long sp); ++#define arch_align_stack(x) ((x) & ~0xfUL) + extern void free_init_pages(char *what, unsigned long begin, unsigned long end); + + void default_idle(void); +diff -urNp linux-2.6.31.1/arch/x86/include/asm/uaccess_32.h linux-2.6.31.1/arch/x86/include/asm/uaccess_32.h +--- linux-2.6.31.1/arch/x86/include/asm/uaccess_32.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/uaccess_32.h 2009-10-01 20:12:42.000000000 -0400 +@@ -44,6 +44,9 @@ unsigned long __must_check __copy_from_u + static __always_inline unsigned long __must_check + __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + if (__builtin_constant_p(n)) { + unsigned long ret; + +@@ -62,6 +65,8 @@ __copy_to_user_inatomic(void __user *to, + return ret; + } + } ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); + return __copy_to_user_ll(to, from, n); + } + +@@ -89,6 +94,9 @@ __copy_to_user(void __user *to, const vo + static __always_inline unsigned long + __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + /* Avoid zeroing the tail if the copy fails.. + * If 'n' is constant and 1, 2, or 4, we do still zero on a failure, + * but as the zeroing behaviour is only significant when n is not +@@ -138,6 +146,10 @@ static __always_inline unsigned long + __copy_from_user(void *to, const void __user *from, unsigned long n) + { + might_fault(); ++ ++ if ((long)n < 0) ++ return n; ++ + if (__builtin_constant_p(n)) { + unsigned long ret; + +@@ -153,6 +165,8 @@ __copy_from_user(void *to, const void __ + return ret; + } + } ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); + return __copy_from_user_ll(to, from, n); + } + +@@ -160,6 +174,10 @@ static __always_inline unsigned long __c + const void __user *from, unsigned long n) + { + might_fault(); ++ ++ if ((long)n < 0) ++ return n; ++ + if (__builtin_constant_p(n)) { + unsigned long ret; + +@@ -182,14 +200,62 @@ static __always_inline unsigned long + __copy_from_user_inatomic_nocache(void *to, const void __user *from, + unsigned long n) + { +- return __copy_from_user_ll_nocache_nozero(to, from, n); ++ if ((long)n < 0) ++ return n; ++ ++ return __copy_from_user_ll_nocache_nozero(to, from, n); ++} ++ ++/** ++ * copy_to_user: - Copy a block of data into user space. ++ * @to: Destination address, in user space. ++ * @from: Source address, in kernel space. ++ * @n: Number of bytes to copy. ++ * ++ * Context: User context only. This function may sleep. ++ * ++ * Copy data from kernel space to user space. ++ * ++ * Returns number of bytes that could not be copied. ++ * On success, this will be zero. ++ */ ++static __always_inline unsigned long __must_check ++copy_to_user(void __user *to, const void *from, unsigned long n) ++{ ++ if (access_ok(VERIFY_WRITE, to, n)) ++ n = __copy_to_user(to, from, n); ++ return n; ++} ++ ++/** ++ * copy_from_user: - Copy a block of data from user space. ++ * @to: Destination address, in kernel space. ++ * @from: Source address, in user space. ++ * @n: Number of bytes to copy. ++ * ++ * Context: User context only. This function may sleep. ++ * ++ * Copy data from user space to kernel space. ++ * ++ * Returns number of bytes that could not be copied. ++ * On success, this will be zero. ++ * ++ * If some data could not be copied, this function will pad the copied ++ * data to the requested size using zero bytes. ++ */ ++static __always_inline unsigned long __must_check ++copy_from_user(void *to, const void __user *from, unsigned long n) ++{ ++ if (access_ok(VERIFY_READ, from, n)) ++ n = __copy_from_user(to, from, n); ++ else if ((long)n > 0) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ memset(to, 0, n); ++ } ++ return n; + } + +-unsigned long __must_check copy_to_user(void __user *to, +- const void *from, unsigned long n); +-unsigned long __must_check copy_from_user(void *to, +- const void __user *from, +- unsigned long n); + long __must_check strncpy_from_user(char *dst, const char __user *src, + long count); + long __must_check __strncpy_from_user(char *dst, +diff -urNp linux-2.6.31.1/arch/x86/include/asm/uaccess_64.h linux-2.6.31.1/arch/x86/include/asm/uaccess_64.h +--- linux-2.6.31.1/arch/x86/include/asm/uaccess_64.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/uaccess_64.h 2009-10-01 20:12:42.000000000 -0400 +@@ -10,6 +10,8 @@ + #include + #include + ++#define set_fs(x) (current_thread_info()->addr_limit = (x)) ++ + /* + * Copy To/From Userspace + */ +@@ -19,20 +21,22 @@ __must_check unsigned long + copy_user_generic(void *to, const void *from, unsigned len); + + __must_check unsigned long +-copy_to_user(void __user *to, const void *from, unsigned len); +-__must_check unsigned long +-copy_from_user(void *to, const void __user *from, unsigned len); +-__must_check unsigned long + copy_in_user(void __user *to, const void __user *from, unsigned len); + + static __always_inline __must_check +-int __copy_from_user(void *dst, const void __user *src, unsigned size) ++unsigned long __copy_from_user(void *dst, const void __user *src, unsigned size) + { +- int ret = 0; ++ unsigned ret = 0; + + might_fault(); +- if (!__builtin_constant_p(size)) ++ ++ if ((int)size < 0) ++ return size; ++ ++ if (!__builtin_constant_p(size)) { ++ check_object_size(dst, size, false); + return copy_user_generic(dst, (__force void *)src, size); ++ } + switch (size) { + case 1:__get_user_asm(*(u8 *)dst, (u8 __user *)src, + ret, "b", "b", "=q", 1); +@@ -70,13 +74,19 @@ int __copy_from_user(void *dst, const vo + } + + static __always_inline __must_check +-int __copy_to_user(void __user *dst, const void *src, unsigned size) ++unsigned long __copy_to_user(void __user *dst, const void *src, unsigned size) + { +- int ret = 0; ++ unsigned ret = 0; + + might_fault(); +- if (!__builtin_constant_p(size)) ++ ++ if ((int)size < 0) ++ return size; ++ ++ if (!__builtin_constant_p(size)) { ++ check_object_size(src, size, true); + return copy_user_generic((__force void *)dst, src, size); ++ } + switch (size) { + case 1:__put_user_asm(*(u8 *)src, (u8 __user *)dst, + ret, "b", "b", "iq", 1); +@@ -114,11 +124,39 @@ int __copy_to_user(void __user *dst, con + } + + static __always_inline __must_check +-int __copy_in_user(void __user *dst, const void __user *src, unsigned size) ++unsigned long copy_to_user(void __user *to, const void *from, unsigned len) + { +- int ret = 0; ++ if (access_ok(VERIFY_WRITE, to, len)) ++ len = __copy_to_user(to, from, len); ++ return len; ++} ++ ++static __always_inline __must_check ++unsigned long copy_from_user(void *to, const void __user *from, unsigned len) ++{ ++ if ((int)len < 0) ++ return len; ++ ++ if (access_ok(VERIFY_READ, from, len)) ++ len = __copy_from_user(to, from, len); ++ else if ((int)len > 0) { ++ if (!__builtin_constant_p(len)) ++ check_object_size(to, len, false); ++ memset(to, 0, len); ++ } ++ return len; ++} ++ ++static __always_inline __must_check ++unsigned long __copy_in_user(void __user *dst, const void __user *src, unsigned size) ++{ ++ unsigned ret = 0; + + might_fault(); ++ ++ if ((int)size < 0) ++ return size; ++ + if (!__builtin_constant_p(size)) + return copy_user_generic((__force void *)dst, + (__force void *)src, size); +@@ -179,30 +217,38 @@ __must_check unsigned long __clear_user( + __must_check long __copy_from_user_inatomic(void *dst, const void __user *src, + unsigned size); + +-static __must_check __always_inline int ++static __must_check __always_inline unsigned long + __copy_to_user_inatomic(void __user *dst, const void *src, unsigned size) + { ++ if ((int)size < 0) ++ return size; ++ + return copy_user_generic((__force void *)dst, src, size); + } + +-extern long __copy_user_nocache(void *dst, const void __user *src, ++extern unsigned long __copy_user_nocache(void *dst, const void __user *src, + unsigned size, int zerorest); + +-static inline int +-__copy_from_user_nocache(void *dst, const void __user *src, unsigned size) ++static inline unsigned long __copy_from_user_nocache(void *dst, const void __user *src, unsigned size) + { + might_sleep(); ++ ++ if ((int)size < 0) ++ return size; ++ + return __copy_user_nocache(dst, src, size, 1); + } + +-static inline int +-__copy_from_user_inatomic_nocache(void *dst, const void __user *src, ++static inline unsigned long __copy_from_user_inatomic_nocache(void *dst, const void __user *src, + unsigned size) + { ++ if ((int)size < 0) ++ return size; ++ + return __copy_user_nocache(dst, src, size, 0); + } + +-unsigned long ++extern unsigned long + copy_user_handle_tail(char *to, char *from, unsigned len, unsigned zerorest); + + #endif /* _ASM_X86_UACCESS_64_H */ +diff -urNp linux-2.6.31.1/arch/x86/include/asm/uaccess.h linux-2.6.31.1/arch/x86/include/asm/uaccess.h +--- linux-2.6.31.1/arch/x86/include/asm/uaccess.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/uaccess.h 2009-10-01 20:12:42.000000000 -0400 +@@ -8,8 +8,11 @@ + #include + #include + #include ++#include ++#include + #include + #include ++#include + + #define VERIFY_READ 0 + #define VERIFY_WRITE 1 +@@ -29,7 +32,12 @@ + + #define get_ds() (KERNEL_DS) + #define get_fs() (current_thread_info()->addr_limit) ++#ifdef CONFIG_X86_32 ++void __set_fs(mm_segment_t x, int cpu); ++void set_fs(mm_segment_t x); ++#else + #define set_fs(x) (current_thread_info()->addr_limit = (x)) ++#endif + + #define segment_eq(a, b) ((a).seg == (b).seg) + +@@ -77,7 +85,26 @@ + * checks that the pointer is in the user space range - after calling + * this function, memory access functions may still return -EFAULT. + */ +-#define access_ok(type, addr, size) (likely(__range_not_ok(addr, size) == 0)) ++#define access_ok(type, addr, size) \ ++({ \ ++ bool __ret_ao = __range_not_ok(addr, size) == 0; \ ++ unsigned long __addr_ao = (unsigned long)addr & PAGE_MASK; \ ++ unsigned long __end_ao = (unsigned long)addr + size - 1; \ ++ if (__ret_ao && unlikely((__end_ao ^ __addr_ao) & PAGE_MASK)) { \ ++ for (; __addr_ao <= __end_ao; __addr_ao += PAGE_SIZE) { \ ++ char __c_ao; \ ++ if (size > PAGE_SIZE) \ ++ cond_resched(); \ ++ if (__get_user(__c_ao, (char __user *)__addr_ao))\ ++ break; \ ++ if (type != VERIFY_WRITE) \ ++ continue; \ ++ if (__put_user(__c_ao, (char __user *)__addr_ao))\ ++ break; \ ++ } \ ++ } \ ++ __ret_ao; \ ++}) + + /* + * The exception table consists of pairs of addresses: the first is the +@@ -183,13 +210,21 @@ extern int __get_user_bad(void); + asm volatile("call __put_user_" #size : "=a" (__ret_pu) \ + : "0" ((typeof(*(ptr)))(x)), "c" (ptr) : "ebx") + +- ++#ifdef CONFIG_X86_32 ++#define _ASM_LOAD_USER_DS(ds) "movw %w" #ds ",%%ds\n" ++#define _ASM_LOAD_KERNEL_DS "pushl %%ss; popl %%ds\n" ++#else ++#define _ASM_LOAD_USER_DS(ds) ++#define _ASM_LOAD_KERNEL_DS ++#endif + + #ifdef CONFIG_X86_32 + #define __put_user_asm_u64(x, addr, err, errret) \ +- asm volatile("1: movl %%eax,0(%2)\n" \ +- "2: movl %%edx,4(%2)\n" \ ++ asm volatile(_ASM_LOAD_USER_DS(5) \ ++ "1: movl %%eax,%%ds:0(%2)\n" \ ++ "2: movl %%edx,%%ds:4(%2)\n" \ + "3:\n" \ ++ _ASM_LOAD_KERNEL_DS \ + ".section .fixup,\"ax\"\n" \ + "4: movl %3,%0\n" \ + " jmp 3b\n" \ +@@ -197,15 +232,18 @@ extern int __get_user_bad(void); + _ASM_EXTABLE(1b, 4b) \ + _ASM_EXTABLE(2b, 4b) \ + : "=r" (err) \ +- : "A" (x), "r" (addr), "i" (errret), "0" (err)) ++ : "A" (x), "r" (addr), "i" (errret), "0" (err), \ ++ "r"(__USER_DS)) + + #define __put_user_asm_ex_u64(x, addr) \ +- asm volatile("1: movl %%eax,0(%1)\n" \ +- "2: movl %%edx,4(%1)\n" \ ++ asm volatile(_ASM_LOAD_USER_DS(2) \ ++ "1: movl %%eax,%%ds:0(%1)\n" \ ++ "2: movl %%edx,%%ds:4(%1)\n" \ + "3:\n" \ ++ _ASM_LOAD_KERNEL_DS \ + _ASM_EXTABLE(1b, 2b - 1b) \ + _ASM_EXTABLE(2b, 3b - 2b) \ +- : : "A" (x), "r" (addr)) ++ : : "A" (x), "r" (addr), "r"(__USER_DS)) + + #define __put_user_x8(x, ptr, __ret_pu) \ + asm volatile("call __put_user_8" : "=a" (__ret_pu) \ +@@ -374,16 +412,18 @@ do { \ + } while (0) + + #define __get_user_asm(x, addr, err, itype, rtype, ltype, errret) \ +- asm volatile("1: mov"itype" %2,%"rtype"1\n" \ ++ asm volatile(_ASM_LOAD_USER_DS(5) \ ++ "1: mov"itype" %%ds:%2,%"rtype"1\n" \ + "2:\n" \ ++ _ASM_LOAD_KERNEL_DS \ + ".section .fixup,\"ax\"\n" \ + "3: mov %3,%0\n" \ + " xor"itype" %"rtype"1,%"rtype"1\n" \ + " jmp 2b\n" \ + ".previous\n" \ + _ASM_EXTABLE(1b, 3b) \ +- : "=r" (err), ltype(x) \ +- : "m" (__m(addr)), "i" (errret), "0" (err)) ++ : "=r" (err), ltype (x) \ ++ : "m" (__m(addr)), "i" (errret), "0" (err), "r"(__USER_DS)) + + #define __get_user_size_ex(x, ptr, size) \ + do { \ +@@ -407,10 +447,12 @@ do { \ + } while (0) + + #define __get_user_asm_ex(x, addr, itype, rtype, ltype) \ +- asm volatile("1: mov"itype" %1,%"rtype"0\n" \ ++ asm volatile(_ASM_LOAD_USER_DS(2) \ ++ "1: mov"itype" %%ds:%1,%"rtype"0\n" \ + "2:\n" \ ++ _ASM_LOAD_KERNEL_DS \ + _ASM_EXTABLE(1b, 2b - 1b) \ +- : ltype(x) : "m" (__m(addr))) ++ : ltype(x) : "m" (__m(addr)), "r"(__USER_DS)) + + #define __put_user_nocheck(x, ptr, size) \ + ({ \ +@@ -438,21 +480,26 @@ struct __large_struct { unsigned long bu + * aliasing issues. + */ + #define __put_user_asm(x, addr, err, itype, rtype, ltype, errret) \ +- asm volatile("1: mov"itype" %"rtype"1,%2\n" \ ++ asm volatile(_ASM_LOAD_USER_DS(5) \ ++ "1: mov"itype" %"rtype"1,%%ds:%2\n" \ + "2:\n" \ ++ _ASM_LOAD_KERNEL_DS \ + ".section .fixup,\"ax\"\n" \ + "3: mov %3,%0\n" \ + " jmp 2b\n" \ + ".previous\n" \ + _ASM_EXTABLE(1b, 3b) \ + : "=r"(err) \ +- : ltype(x), "m" (__m(addr)), "i" (errret), "0" (err)) ++ : ltype (x), "m" (__m(addr)), "i" (errret), "0" (err),\ ++ "r"(__USER_DS)) + + #define __put_user_asm_ex(x, addr, itype, rtype, ltype) \ +- asm volatile("1: mov"itype" %"rtype"0,%1\n" \ ++ asm volatile(_ASM_LOAD_USER_DS(2) \ ++ "1: mov"itype" %"rtype"0,%%ds:%1\n" \ + "2:\n" \ ++ _ASM_LOAD_KERNEL_DS \ + _ASM_EXTABLE(1b, 2b - 1b) \ +- : : ltype(x), "m" (__m(addr))) ++ : : ltype(x), "m" (__m(addr)), "r"(__USER_DS)) + + /* + * uaccess_try and catch +@@ -567,6 +614,7 @@ extern struct movsl_mask { + + #define ARCH_HAS_NOCACHE_UACCESS 1 + ++#define ARCH_HAS_SORT_EXTABLE + #ifdef CONFIG_X86_32 + # include "uaccess_32.h" + #else +diff -urNp linux-2.6.31.1/arch/x86/include/asm/vgtod.h linux-2.6.31.1/arch/x86/include/asm/vgtod.h +--- linux-2.6.31.1/arch/x86/include/asm/vgtod.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/vgtod.h 2009-10-01 20:12:42.000000000 -0400 +@@ -14,6 +14,7 @@ struct vsyscall_gtod_data { + int sysctl_enabled; + struct timezone sys_tz; + struct { /* extract of a clocksource struct */ ++ char name[8]; + cycle_t (*vread)(void); + cycle_t cycle_last; + cycle_t mask; +diff -urNp linux-2.6.31.1/arch/x86/include/asm/vsyscall.h linux-2.6.31.1/arch/x86/include/asm/vsyscall.h +--- linux-2.6.31.1/arch/x86/include/asm/vsyscall.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/include/asm/vsyscall.h 2009-10-01 20:12:42.000000000 -0400 +@@ -15,9 +15,10 @@ enum vsyscall_num { + + #ifdef __KERNEL__ + #include ++#include ++#include + + #define __section_vgetcpu_mode __attribute__ ((unused, __section__ (".vgetcpu_mode"), aligned(16))) +-#define __section_jiffies __attribute__ ((unused, __section__ (".jiffies"), aligned(16))) + + /* Definitions for CONFIG_GENERIC_TIME definitions */ + #define __section_vsyscall_gtod_data __attribute__ \ +@@ -31,7 +32,6 @@ enum vsyscall_num { + #define VGETCPU_LSL 2 + + extern int __vgetcpu_mode; +-extern volatile unsigned long __jiffies; + + /* kernel space (writeable) */ + extern int vgetcpu_mode; +@@ -39,6 +39,9 @@ extern struct timezone sys_tz; + + extern void map_vsyscall(void); + ++extern int vgettimeofday(struct timeval * tv, struct timezone * tz); ++extern time_t vtime(time_t *t); ++extern long vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache); + #endif /* __KERNEL__ */ + + #endif /* _ASM_X86_VSYSCALL_H */ +diff -urNp linux-2.6.31.1/arch/x86/Kconfig linux-2.6.31.1/arch/x86/Kconfig +--- linux-2.6.31.1/arch/x86/Kconfig 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/Kconfig 2009-10-01 20:12:42.000000000 -0400 +@@ -1098,7 +1098,7 @@ config PAGE_OFFSET + hex + default 0xB0000000 if VMSPLIT_3G_OPT + default 0x80000000 if VMSPLIT_2G +- default 0x78000000 if VMSPLIT_2G_OPT ++ default 0x70000000 if VMSPLIT_2G_OPT + default 0x40000000 if VMSPLIT_1G + default 0xC0000000 + depends on X86_32 +@@ -1416,7 +1416,7 @@ config X86_PAT + + config EFI + bool "EFI runtime service support" +- depends on ACPI ++ depends on ACPI && !PAX_KERNEXEC + ---help--- + This enables the kernel to use EFI runtime services that are + available (such as the EFI variable services). +@@ -1602,9 +1602,10 @@ config HOTPLUG_CPU + Say N if you want to disable CPU hotplug. + + config COMPAT_VDSO +- def_bool y ++ def_bool n + prompt "Compat VDSO support" + depends on X86_32 || IA32_EMULATION ++ depends on !PAX_NOEXEC && !PAX_MEMORY_UDEREF + ---help--- + Map the 32-bit VDSO to the predictable old-style address too. + ---help--- +diff -urNp linux-2.6.31.1/arch/x86/Kconfig.cpu linux-2.6.31.1/arch/x86/Kconfig.cpu +--- linux-2.6.31.1/arch/x86/Kconfig.cpu 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/Kconfig.cpu 2009-10-01 20:12:42.000000000 -0400 +@@ -331,7 +331,7 @@ config X86_PPRO_FENCE + + config X86_F00F_BUG + def_bool y +- depends on M586MMX || M586TSC || M586 || M486 || M386 ++ depends on (M586MMX || M586TSC || M586 || M486 || M386) && !PAX_KERNEXEC + + config X86_WP_WORKS_OK + def_bool y +@@ -351,7 +351,7 @@ config X86_POPAD_OK + + config X86_ALIGNMENT_16 + def_bool y +- depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 ++ depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK8 || MK7 || MK6 || MCORE2 || MPENTIUM4 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 + + config X86_INTEL_USERCOPY + def_bool y +@@ -397,7 +397,7 @@ config X86_CMPXCHG64 + # generates cmov. + config X86_CMOV + def_bool y +- depends on (MK8 || MK7 || MCORE2 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64) ++ depends on (MK8 || MK7 || MCORE2 || MPSC || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64) + + config X86_MINIMUM_CPU_FAMILY + int +diff -urNp linux-2.6.31.1/arch/x86/Kconfig.debug linux-2.6.31.1/arch/x86/Kconfig.debug +--- linux-2.6.31.1/arch/x86/Kconfig.debug 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/Kconfig.debug 2009-10-01 20:12:42.000000000 -0400 +@@ -99,7 +99,7 @@ config X86_PTDUMP + config DEBUG_RODATA + bool "Write protect kernel read-only data structures" + default y +- depends on DEBUG_KERNEL ++ depends on DEBUG_KERNEL && BROKEN + ---help--- + Mark the kernel read-only data as write-protected in the pagetables, + in order to catch accidental (and incorrect) writes to such const +diff -urNp linux-2.6.31.1/arch/x86/kernel/acpi/boot.c linux-2.6.31.1/arch/x86/kernel/acpi/boot.c +--- linux-2.6.31.1/arch/x86/kernel/acpi/boot.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/acpi/boot.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1609,7 +1609,7 @@ static struct dmi_system_id __initdata a + DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq 6715b"), + }, + }, +- {} ++ { NULL, NULL, {{0, {0}}}, NULL} + }; + + /* +diff -urNp linux-2.6.31.1/arch/x86/kernel/acpi/realmode/wakeup.S linux-2.6.31.1/arch/x86/kernel/acpi/realmode/wakeup.S +--- linux-2.6.31.1/arch/x86/kernel/acpi/realmode/wakeup.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/acpi/realmode/wakeup.S 2009-10-01 20:12:42.000000000 -0400 +@@ -104,7 +104,7 @@ _start: + movl %eax, %ecx + orl %edx, %ecx + jz 1f +- movl $0xc0000080, %ecx ++ mov $MSR_EFER, %ecx + wrmsr + 1: + +diff -urNp linux-2.6.31.1/arch/x86/kernel/acpi/sleep.c linux-2.6.31.1/arch/x86/kernel/acpi/sleep.c +--- linux-2.6.31.1/arch/x86/kernel/acpi/sleep.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/acpi/sleep.c 2009-10-01 20:12:42.000000000 -0400 +@@ -11,11 +11,12 @@ + #include + #include + #include ++#include + + #include "realmode/wakeup.h" + #include "sleep.h" + +-unsigned long acpi_wakeup_address; ++unsigned long acpi_wakeup_address = 0x2000; + unsigned long acpi_realmode_flags; + + /* address in low memory of the wakeup routine. */ +@@ -37,6 +38,10 @@ int acpi_save_state_mem(void) + { + struct wakeup_header *header; + ++#if defined(CONFIG_64BIT) && defined(CONFIG_SMP) && defined(CONFIG_PAX_KERNEXEC) ++ unsigned long cr0; ++#endif ++ + if (!acpi_realmode) { + printk(KERN_ERR "Could not allocate memory during boot, " + "S3 disabled\n"); +@@ -99,8 +104,18 @@ int acpi_save_state_mem(void) + header->trampoline_segment = setup_trampoline() >> 4; + #ifdef CONFIG_SMP + stack_start.sp = temp_stack + sizeof(temp_stack); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + early_gdt_descr.address = + (unsigned long)get_cpu_gdt_table(smp_processor_id()); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + initial_gs = per_cpu_offset(smp_processor_id()); + #endif + initial_code = (unsigned long)wakeup_long64; +@@ -134,14 +149,8 @@ void __init acpi_reserve_bootmem(void) + return; + } + +- acpi_realmode = (unsigned long)alloc_bootmem_low(WAKEUP_SIZE); +- +- if (!acpi_realmode) { +- printk(KERN_ERR "ACPI: Cannot allocate lowmem, S3 disabled.\n"); +- return; +- } +- +- acpi_wakeup_address = virt_to_phys((void *)acpi_realmode); ++ reserve_early(acpi_wakeup_address, acpi_wakeup_address + WAKEUP_SIZE, "ACPI Wakeup Code"); ++ acpi_realmode = (unsigned long)__va(acpi_wakeup_address);; + } + + +diff -urNp linux-2.6.31.1/arch/x86/kernel/acpi/wakeup_32.S linux-2.6.31.1/arch/x86/kernel/acpi/wakeup_32.S +--- linux-2.6.31.1/arch/x86/kernel/acpi/wakeup_32.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/acpi/wakeup_32.S 2009-10-01 20:12:42.000000000 -0400 +@@ -30,13 +30,11 @@ wakeup_pmode_return: + # and restore the stack ... but you need gdt for this to work + movl saved_context_esp, %esp + +- movl %cs:saved_magic, %eax +- cmpl $0x12345678, %eax ++ cmpl $0x12345678, saved_magic + jne bogus_magic + + # jump to place where we left off +- movl saved_eip, %eax +- jmp *%eax ++ jmp *(saved_eip) + + bogus_magic: + jmp bogus_magic +diff -urNp linux-2.6.31.1/arch/x86/kernel/alternative.c linux-2.6.31.1/arch/x86/kernel/alternative.c +--- linux-2.6.31.1/arch/x86/kernel/alternative.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/alternative.c 2009-10-01 20:12:42.000000000 -0400 +@@ -400,7 +400,7 @@ void apply_paravirt(struct paravirt_patc + + BUG_ON(p->len > MAX_PATCH_LEN); + /* prep the buffer with the original instructions */ +- memcpy(insnbuf, p->instr, p->len); ++ memcpy(insnbuf, ktla_ktva(p->instr), p->len); + used = pv_init_ops.patch(p->instrtype, p->clobbers, insnbuf, + (unsigned long)p->instr, p->len); + +@@ -485,11 +485,26 @@ void __init alternative_instructions(voi + * instructions. And on the local CPU you need to be protected again NMI or MCE + * handlers seeing an inconsistent instruction while you patch. + */ +-void *text_poke_early(void *addr, const void *opcode, size_t len) ++void *__kprobes text_poke_early(void *addr, const void *opcode, size_t len) + { + unsigned long flags; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + local_irq_save(flags); +- memcpy(addr, opcode, len); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ ++ memcpy(ktla_ktva(addr), opcode, len); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + local_irq_restore(flags); + sync_core(); + /* Could also do a CLFLUSH here to speed up CPU recovery; but +@@ -512,35 +527,27 @@ void *text_poke_early(void *addr, const + */ + void *__kprobes text_poke(void *addr, const void *opcode, size_t len) + { +- unsigned long flags; +- char *vaddr; ++ unsigned char *vaddr = ktla_ktva(addr); + struct page *pages[2]; +- int i; ++ size_t i; ++ ++ if (!core_kernel_text((unsigned long)addr) + +- if (!core_kernel_text((unsigned long)addr)) { +- pages[0] = vmalloc_to_page(addr); +- pages[1] = vmalloc_to_page(addr + PAGE_SIZE); ++#if defined(CONFIG_X86_32) && defined(CONFIG_MODULES) && defined(CONFIG_PAX_KERNEXEC) ++ && (vaddr < MODULES_EXEC_VADDR || MODULES_EXEC_END < vaddr) ++#endif ++ ++ ) { ++ pages[0] = vmalloc_to_page(vaddr); ++ pages[1] = vmalloc_to_page(vaddr + PAGE_SIZE); + } else { +- pages[0] = virt_to_page(addr); ++ pages[0] = virt_to_page(vaddr); + WARN_ON(!PageReserved(pages[0])); +- pages[1] = virt_to_page(addr + PAGE_SIZE); ++ pages[1] = virt_to_page(vaddr + PAGE_SIZE); + } + BUG_ON(!pages[0]); +- local_irq_save(flags); +- set_fixmap(FIX_TEXT_POKE0, page_to_phys(pages[0])); +- if (pages[1]) +- set_fixmap(FIX_TEXT_POKE1, page_to_phys(pages[1])); +- vaddr = (char *)fix_to_virt(FIX_TEXT_POKE0); +- memcpy(&vaddr[(unsigned long)addr & ~PAGE_MASK], opcode, len); +- clear_fixmap(FIX_TEXT_POKE0); +- if (pages[1]) +- clear_fixmap(FIX_TEXT_POKE1); +- local_flush_tlb(); +- sync_core(); +- /* Could also do a CLFLUSH here to speed up CPU recovery; but +- that causes hangs on some VIA CPUs. */ ++ text_poke_early(addr, opcode, len); + for (i = 0; i < len; i++) +- BUG_ON(((char *)addr)[i] != ((char *)opcode)[i]); +- local_irq_restore(flags); ++ BUG_ON(((char *)vaddr)[i] != ((char *)opcode)[i]); + return addr; + } +diff -urNp linux-2.6.31.1/arch/x86/kernel/apm_32.c linux-2.6.31.1/arch/x86/kernel/apm_32.c +--- linux-2.6.31.1/arch/x86/kernel/apm_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/apm_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -403,7 +403,7 @@ static DECLARE_WAIT_QUEUE_HEAD(apm_waitq + static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue); + static struct apm_user *user_list; + static DEFINE_SPINLOCK(user_list_lock); +-static const struct desc_struct bad_bios_desc = { { { 0, 0x00409200 } } }; ++static const struct desc_struct bad_bios_desc = { { { 0, 0x00409300 } } }; + + static const char driver_version[] = "1.16ac"; /* no spaces */ + +@@ -576,12 +576,25 @@ static long __apm_bios_call(void *_call) + struct desc_struct *gdt; + struct apm_bios_call *call = _call; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + cpu = get_cpu(); + BUG_ON(cpu != 0); + gdt = get_cpu_gdt_table(cpu); + save_desc_40 = gdt[0x40 / 8]; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + gdt[0x40 / 8] = bad_bios_desc; + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + apm_irq_save(flags); + APM_DO_SAVE_SEGS; + apm_bios_call_asm(call->func, call->ebx, call->ecx, +@@ -589,7 +602,17 @@ static long __apm_bios_call(void *_call) + &call->esi); + APM_DO_RESTORE_SEGS; + apm_irq_restore(flags); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + gdt[0x40 / 8] = save_desc_40; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + put_cpu(); + + return call->eax & 0xff; +@@ -652,19 +675,42 @@ static long __apm_bios_call_simple(void + struct desc_struct *gdt; + struct apm_bios_call *call = _call; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + cpu = get_cpu(); + BUG_ON(cpu != 0); + gdt = get_cpu_gdt_table(cpu); + save_desc_40 = gdt[0x40 / 8]; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + gdt[0x40 / 8] = bad_bios_desc; + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + apm_irq_save(flags); + APM_DO_SAVE_SEGS; + error = apm_bios_call_simple_asm(call->func, call->ebx, call->ecx, + &call->eax); + APM_DO_RESTORE_SEGS; + apm_irq_restore(flags); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + gdt[0x40 / 8] = save_desc_40; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + put_cpu(); + return error; + } +@@ -967,7 +1013,7 @@ recalc: + + static void apm_power_off(void) + { +- unsigned char po_bios_call[] = { ++ const unsigned char po_bios_call[] = { + 0xb8, 0x00, 0x10, /* movw $0x1000,ax */ + 0x8e, 0xd0, /* movw ax,ss */ + 0xbc, 0x00, 0xf0, /* movw $0xf000,sp */ +@@ -1925,7 +1971,10 @@ static const struct file_operations apm_ + static struct miscdevice apm_device = { + APM_MINOR_DEV, + "apm_bios", +- &apm_bios_fops ++ &apm_bios_fops, ++ {NULL, NULL}, ++ NULL, ++ NULL + }; + + +@@ -2246,7 +2295,7 @@ static struct dmi_system_id __initdata a + { DMI_MATCH(DMI_SYS_VENDOR, "IBM"), }, + }, + +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL} + }; + + /* +@@ -2264,6 +2313,10 @@ static int __init apm_init(void) + struct desc_struct *gdt; + int err; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + dmi_check_system(apm_dmi_table); + + if (apm_info.bios.version == 0 || paravirt_enabled() || machine_is_olpc()) { +@@ -2337,9 +2390,18 @@ static int __init apm_init(void) + * This is for buggy BIOS's that refer to (real mode) segment 0x40 + * even though they are called in protected mode. + */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + set_base(bad_bios_desc, __va((unsigned long)0x40 << 4)); + _set_limit((char *)&bad_bios_desc, 4095 - (0x40 << 4)); + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + /* + * Set up the long jump entry point to the APM BIOS, which is called + * from inline assembly. +@@ -2358,6 +2420,11 @@ static int __init apm_init(void) + * code to that CPU. + */ + gdt = get_cpu_gdt_table(0); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + set_base(gdt[APM_CS >> 3], + __va((unsigned long)apm_info.bios.cseg << 4)); + set_base(gdt[APM_CS_16 >> 3], +@@ -2365,6 +2432,10 @@ static int __init apm_init(void) + set_base(gdt[APM_DS >> 3], + __va((unsigned long)apm_info.bios.dseg << 4)); + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + proc_create("apm", 0, NULL, &apm_file_ops); + + kapmd_task = kthread_create(apm, NULL, "kapmd"); +diff -urNp linux-2.6.31.1/arch/x86/kernel/asm-offsets_32.c linux-2.6.31.1/arch/x86/kernel/asm-offsets_32.c +--- linux-2.6.31.1/arch/x86/kernel/asm-offsets_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/asm-offsets_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -115,6 +115,7 @@ void foo(void) + OFFSET(PV_CPU_iret, pv_cpu_ops, iret); + OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit); + OFFSET(PV_CPU_read_cr0, pv_cpu_ops, read_cr0); ++ OFFSET(PV_CPU_write_cr0, pv_cpu_ops, write_cr0); + #endif + + #ifdef CONFIG_XEN +diff -urNp linux-2.6.31.1/arch/x86/kernel/asm-offsets_64.c linux-2.6.31.1/arch/x86/kernel/asm-offsets_64.c +--- linux-2.6.31.1/arch/x86/kernel/asm-offsets_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/asm-offsets_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -114,6 +114,7 @@ int main(void) + ENTRY(cr8); + BLANK(); + #undef ENTRY ++ DEFINE(TSS_size, sizeof(struct tss_struct)); + DEFINE(TSS_ist, offsetof(struct tss_struct, x86_tss.ist)); + BLANK(); + DEFINE(crypto_tfm_ctx_offset, offsetof(struct crypto_tfm, __crt_ctx)); +diff -urNp linux-2.6.31.1/arch/x86/kernel/cpu/common.c linux-2.6.31.1/arch/x86/kernel/cpu/common.c +--- linux-2.6.31.1/arch/x86/kernel/cpu/common.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/cpu/common.c 2009-10-01 20:12:42.000000000 -0400 +@@ -84,60 +84,6 @@ static const struct cpu_dev __cpuinitcon + + static const struct cpu_dev *this_cpu __cpuinitdata = &default_cpu; + +-DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { +-#ifdef CONFIG_X86_64 +- /* +- * We need valid kernel segments for data and code in long mode too +- * IRET will check the segment types kkeil 2000/10/28 +- * Also sysret mandates a special GDT layout +- * +- * TLS descriptors are currently at a different place compared to i386. +- * Hopefully nobody expects them at a fixed place (Wine?) +- */ +- [GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } }, +- [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } }, +- [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } }, +- [GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } }, +- [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } }, +- [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } }, +-#else +- [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00cf9a00 } } }, +- [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9200 } } }, +- [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00cffa00 } } }, +- [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff200 } } }, +- /* +- * Segments used for calling PnP BIOS have byte granularity. +- * They code segments and data segments have fixed 64k limits, +- * the transfer segment sizes are set at run time. +- */ +- /* 32-bit code */ +- [GDT_ENTRY_PNPBIOS_CS32] = { { { 0x0000ffff, 0x00409a00 } } }, +- /* 16-bit code */ +- [GDT_ENTRY_PNPBIOS_CS16] = { { { 0x0000ffff, 0x00009a00 } } }, +- /* 16-bit data */ +- [GDT_ENTRY_PNPBIOS_DS] = { { { 0x0000ffff, 0x00009200 } } }, +- /* 16-bit data */ +- [GDT_ENTRY_PNPBIOS_TS1] = { { { 0x00000000, 0x00009200 } } }, +- /* 16-bit data */ +- [GDT_ENTRY_PNPBIOS_TS2] = { { { 0x00000000, 0x00009200 } } }, +- /* +- * The APM segments have byte granularity and their bases +- * are set at run time. All have 64k limits. +- */ +- /* 32-bit code */ +- [GDT_ENTRY_APMBIOS_BASE] = { { { 0x0000ffff, 0x00409a00 } } }, +- /* 16-bit code */ +- [GDT_ENTRY_APMBIOS_BASE+1] = { { { 0x0000ffff, 0x00009a00 } } }, +- /* data */ +- [GDT_ENTRY_APMBIOS_BASE+2] = { { { 0x0000ffff, 0x00409200 } } }, +- +- [GDT_ENTRY_ESPFIX_SS] = { { { 0x0000ffff, 0x00cf9200 } } }, +- [GDT_ENTRY_PERCPU] = { { { 0x0000ffff, 0x00cf9200 } } }, +- GDT_STACK_CANARY_INIT +-#endif +-} }; +-EXPORT_PER_CPU_SYMBOL_GPL(gdt_page); +- + static int __init x86_xsave_setup(char *s) + { + setup_clear_cpu_cap(X86_FEATURE_XSAVE); +@@ -345,7 +291,7 @@ void switch_to_new_gdt(int cpu) + { + struct desc_ptr gdt_descr; + +- gdt_descr.address = (long)get_cpu_gdt_table(cpu); ++ gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu); + gdt_descr.size = GDT_SIZE - 1; + load_gdt(&gdt_descr); + /* Reload the per-cpu base */ +@@ -799,6 +745,10 @@ static void __cpuinit identify_cpu(struc + /* Filter out anything that depends on CPUID levels we don't have */ + filter_cpuid_features(c, true); + ++#if defined(CONFIG_PAX_SEGMEXEC) || defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF) ++ setup_clear_cpu_cap(X86_FEATURE_SEP); ++#endif ++ + /* If the model name is still unset, do table lookup. */ + if (!c->x86_model_id[0]) { + const char *p; +@@ -982,7 +932,7 @@ static __init int setup_disablecpuid(cha + __setup("clearcpuid=", setup_disablecpuid); + + #ifdef CONFIG_X86_64 +-struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table }; ++struct desc_ptr idt_descr __read_only = { 256 * 16 - 1, (unsigned long) idt_table }; + + DEFINE_PER_CPU_FIRST(union irq_stack_union, + irq_stack_union) __aligned(PAGE_SIZE); +@@ -1092,7 +1042,7 @@ void __cpuinit cpu_init(void) + int i; + + cpu = stack_smp_processor_id(); +- t = &per_cpu(init_tss, cpu); ++ t = init_tss + cpu; + orig_ist = &per_cpu(orig_ist, cpu); + + #ifdef CONFIG_NUMA +@@ -1190,7 +1140,7 @@ void __cpuinit cpu_init(void) + { + int cpu = smp_processor_id(); + struct task_struct *curr = current; +- struct tss_struct *t = &per_cpu(init_tss, cpu); ++ struct tss_struct *t = init_tss + cpu; + struct thread_struct *thread = &curr->thread; + + if (cpumask_test_and_set_cpu(cpu, cpu_initialized_mask)) { +diff -urNp linux-2.6.31.1/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c linux-2.6.31.1/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +--- linux-2.6.31.1/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c 2009-10-01 20:12:42.000000000 -0400 +@@ -586,7 +586,7 @@ static const struct dmi_system_id sw_any + DMI_MATCH(DMI_PRODUCT_NAME, "X6DLP"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + #endif + +diff -urNp linux-2.6.31.1/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c linux-2.6.31.1/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c +--- linux-2.6.31.1/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c 2009-10-01 20:12:42.000000000 -0400 +@@ -225,7 +225,7 @@ static struct cpu_model models[] = + { &cpu_ids[CPU_MP4HT_D0], NULL, 0, NULL }, + { &cpu_ids[CPU_MP4HT_E0], NULL, 0, NULL }, + +- { NULL, } ++ { NULL, NULL, 0, NULL} + }; + #undef _BANIAS + #undef BANIAS +diff -urNp linux-2.6.31.1/arch/x86/kernel/cpu/intel.c linux-2.6.31.1/arch/x86/kernel/cpu/intel.c +--- linux-2.6.31.1/arch/x86/kernel/cpu/intel.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/cpu/intel.c 2009-10-01 20:12:42.000000000 -0400 +@@ -140,7 +140,7 @@ static void __cpuinit trap_init_f00f_bug + * Update the IDT descriptor and reload the IDT so that + * it uses the read-only mapped virtual address. + */ +- idt_descr.address = fix_to_virt(FIX_F00F_IDT); ++ idt_descr.address = (struct desc_struct *)fix_to_virt(FIX_F00F_IDT); + load_idt(&idt_descr); + } + #endif +diff -urNp linux-2.6.31.1/arch/x86/kernel/cpu/Makefile linux-2.6.31.1/arch/x86/kernel/cpu/Makefile +--- linux-2.6.31.1/arch/x86/kernel/cpu/Makefile 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/cpu/Makefile 2009-10-01 20:12:42.000000000 -0400 +@@ -7,10 +7,6 @@ ifdef CONFIG_FUNCTION_TRACER + CFLAGS_REMOVE_common.o = -pg + endif + +-# Make sure load_percpu_segment has no stackprotector +-nostackp := $(call cc-option, -fno-stack-protector) +-CFLAGS_common.o := $(nostackp) +- + obj-y := intel_cacheinfo.o addon_cpuid_features.o + obj-y += proc.o capflags.o powerflags.o common.o + obj-y += vmware.o hypervisor.o +diff -urNp linux-2.6.31.1/arch/x86/kernel/cpu/mcheck/mce.c linux-2.6.31.1/arch/x86/kernel/cpu/mcheck/mce.c +--- linux-2.6.31.1/arch/x86/kernel/cpu/mcheck/mce.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/cpu/mcheck/mce.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1370,14 +1370,14 @@ void __cpuinit mcheck_init(struct cpuinf + */ + + static DEFINE_SPINLOCK(mce_state_lock); +-static int open_count; /* #times opened */ ++static atomic_t open_count; /* #times opened */ + static int open_exclu; /* already open exclusive? */ + + static int mce_open(struct inode *inode, struct file *file) + { + spin_lock(&mce_state_lock); + +- if (open_exclu || (open_count && (file->f_flags & O_EXCL))) { ++ if (open_exclu || (atomic_read(&open_count) && (file->f_flags & O_EXCL))) { + spin_unlock(&mce_state_lock); + + return -EBUSY; +@@ -1385,7 +1385,7 @@ static int mce_open(struct inode *inode, + + if (file->f_flags & O_EXCL) + open_exclu = 1; +- open_count++; ++ atomic_inc(&open_count); + + spin_unlock(&mce_state_lock); + +@@ -1396,7 +1396,7 @@ static int mce_release(struct inode *ino + { + spin_lock(&mce_state_lock); + +- open_count--; ++ atomic_dec(&open_count); + open_exclu = 0; + + spin_unlock(&mce_state_lock); +@@ -1536,6 +1536,7 @@ static struct miscdevice mce_log_device + MISC_MCELOG_MINOR, + "mcelog", + &mce_chrdev_ops, ++ {NULL, NULL}, NULL, NULL + }; + + /* +diff -urNp linux-2.6.31.1/arch/x86/kernel/cpu/mtrr/generic.c linux-2.6.31.1/arch/x86/kernel/cpu/mtrr/generic.c +--- linux-2.6.31.1/arch/x86/kernel/cpu/mtrr/generic.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/cpu/mtrr/generic.c 2009-10-01 20:12:42.000000000 -0400 +@@ -23,14 +23,14 @@ static struct fixed_range_block fixed_ra + { MSR_MTRRfix64K_00000, 1 }, /* one 64k MTRR */ + { MSR_MTRRfix16K_80000, 2 }, /* two 16k MTRRs */ + { MSR_MTRRfix4K_C0000, 8 }, /* eight 4k MTRRs */ +- {} ++ { 0, 0 } + }; + + static unsigned long smp_changes_mask; + static int mtrr_state_set; + u64 mtrr_tom2; + +-struct mtrr_state_type mtrr_state = {}; ++struct mtrr_state_type mtrr_state; + EXPORT_SYMBOL_GPL(mtrr_state); + + /** +diff -urNp linux-2.6.31.1/arch/x86/kernel/crash.c linux-2.6.31.1/arch/x86/kernel/crash.c +--- linux-2.6.31.1/arch/x86/kernel/crash.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/crash.c 2009-10-01 20:12:42.000000000 -0400 +@@ -42,7 +42,7 @@ static void kdump_nmi_callback(int cpu, + regs = args->regs; + + #ifdef CONFIG_X86_32 +- if (!user_mode_vm(regs)) { ++ if (!user_mode(regs)) { + crash_fixup_ss_esp(&fixed_regs, regs); + regs = &fixed_regs; + } +diff -urNp linux-2.6.31.1/arch/x86/kernel/doublefault_32.c linux-2.6.31.1/arch/x86/kernel/doublefault_32.c +--- linux-2.6.31.1/arch/x86/kernel/doublefault_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/doublefault_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -11,7 +11,7 @@ + + #define DOUBLEFAULT_STACKSIZE (1024) + static unsigned long doublefault_stack[DOUBLEFAULT_STACKSIZE]; +-#define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE) ++#define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE-2) + + #define ptr_ok(x) ((x) > PAGE_OFFSET && (x) < PAGE_OFFSET + MAXMEM) + +@@ -21,7 +21,7 @@ static void doublefault_fn(void) + unsigned long gdt, tss; + + store_gdt(&gdt_desc); +- gdt = gdt_desc.address; ++ gdt = (unsigned long)gdt_desc.address; + + printk(KERN_EMERG "PANIC: double fault, gdt at %08lx [%d bytes]\n", gdt, gdt_desc.size); + +@@ -60,10 +60,10 @@ struct tss_struct doublefault_tss __cach + /* 0x2 bit is always set */ + .flags = X86_EFLAGS_SF | 0x2, + .sp = STACK_START, +- .es = __USER_DS, ++ .es = __KERNEL_DS, + .cs = __KERNEL_CS, + .ss = __KERNEL_DS, +- .ds = __USER_DS, ++ .ds = __KERNEL_DS, + .fs = __KERNEL_PERCPU, + + .__cr3 = __pa_nodebug(swapper_pg_dir), +diff -urNp linux-2.6.31.1/arch/x86/kernel/dumpstack_32.c linux-2.6.31.1/arch/x86/kernel/dumpstack_32.c +--- linux-2.6.31.1/arch/x86/kernel/dumpstack_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/dumpstack_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -113,11 +113,12 @@ void show_registers(struct pt_regs *regs + * When in-kernel, we also print out the stack and code at the + * time of the fault.. + */ +- if (!user_mode_vm(regs)) { ++ if (!user_mode(regs)) { + unsigned int code_prologue = code_bytes * 43 / 64; + unsigned int code_len = code_bytes; + unsigned char c; + u8 *ip; ++ unsigned long cs_base = get_desc_base(&get_cpu_gdt_table(smp_processor_id())[(0xffff & regs->cs) >> 3]); + + printk(KERN_EMERG "Stack:\n"); + show_stack_log_lvl(NULL, regs, ®s->sp, +@@ -125,10 +126,10 @@ void show_registers(struct pt_regs *regs + + printk(KERN_EMERG "Code: "); + +- ip = (u8 *)regs->ip - code_prologue; ++ ip = (u8 *)regs->ip - code_prologue + cs_base; + if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { + /* try starting at IP */ +- ip = (u8 *)regs->ip; ++ ip = (u8 *)regs->ip + cs_base; + code_len = code_len - code_prologue + 1; + } + for (i = 0; i < code_len; i++, ip++) { +@@ -137,7 +138,7 @@ void show_registers(struct pt_regs *regs + printk(" Bad EIP value."); + break; + } +- if (ip == (u8 *)regs->ip) ++ if (ip == (u8 *)regs->ip + cs_base) + printk("<%02x> ", c); + else + printk("%02x ", c); +@@ -150,6 +151,7 @@ int is_valid_bugaddr(unsigned long ip) + { + unsigned short ud2; + ++ ip = ktla_ktva(ip); + if (ip < PAGE_OFFSET) + return 0; + if (probe_kernel_address((unsigned short *)ip, ud2)) +diff -urNp linux-2.6.31.1/arch/x86/kernel/dumpstack.c linux-2.6.31.1/arch/x86/kernel/dumpstack.c +--- linux-2.6.31.1/arch/x86/kernel/dumpstack.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/dumpstack.c 2009-10-01 20:12:42.000000000 -0400 +@@ -181,7 +181,7 @@ void dump_stack(void) + #endif + + printk("Pid: %d, comm: %.20s %s %s %.*s\n", +- current->pid, current->comm, print_tainted(), ++ task_pid_nr(current), current->comm, print_tainted(), + init_utsname()->release, + (int)strcspn(init_utsname()->version, " "), + init_utsname()->version); +@@ -242,7 +242,7 @@ void __kprobes oops_end(unsigned long fl + panic("Fatal exception in interrupt"); + if (panic_on_oops) + panic("Fatal exception"); +- do_exit(signr); ++ do_group_exit(signr); + } + + int __kprobes __die(const char *str, struct pt_regs *regs, long err) +@@ -296,7 +296,7 @@ void die(const char *str, struct pt_regs + unsigned long flags = oops_begin(); + int sig = SIGSEGV; + +- if (!user_mode_vm(regs)) ++ if (!user_mode(regs)) + report_bug(regs->ip, regs); + + if (__die(str, regs, err)) +diff -urNp linux-2.6.31.1/arch/x86/kernel/e820.c linux-2.6.31.1/arch/x86/kernel/e820.c +--- linux-2.6.31.1/arch/x86/kernel/e820.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/e820.c 2009-10-01 20:12:42.000000000 -0400 +@@ -733,7 +733,10 @@ struct early_res { + }; + static struct early_res early_res[MAX_EARLY_RES] __initdata = { + { 0, PAGE_SIZE, "BIOS data page" }, /* BIOS data page */ +- {} ++#ifdef CONFIG_VM86 ++ { PAGE_SIZE, ISA_START_ADDRESS, "V86 mode memory", 1 }, ++#endif ++ { 0, 0, {0}, 0 } + }; + + static int __init find_overlapped_early(u64 start, u64 end) +diff -urNp linux-2.6.31.1/arch/x86/kernel/efi_32.c linux-2.6.31.1/arch/x86/kernel/efi_32.c +--- linux-2.6.31.1/arch/x86/kernel/efi_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/efi_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -38,70 +38,38 @@ + */ + + static unsigned long efi_rt_eflags; +-static pgd_t efi_bak_pg_dir_pointer[2]; ++static pgd_t __initdata efi_bak_pg_dir_pointer[KERNEL_PGD_PTRS]; + +-void efi_call_phys_prelog(void) ++void __init efi_call_phys_prelog(void) + { +- unsigned long cr4; +- unsigned long temp; + struct desc_ptr gdt_descr; + + local_irq_save(efi_rt_eflags); + +- /* +- * If I don't have PAE, I should just duplicate two entries in page +- * directory. If I have PAE, I just need to duplicate one entry in +- * page directory. +- */ +- cr4 = read_cr4_safe(); + +- if (cr4 & X86_CR4_PAE) { +- efi_bak_pg_dir_pointer[0].pgd = +- swapper_pg_dir[pgd_index(0)].pgd; +- swapper_pg_dir[0].pgd = +- swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; +- } else { +- efi_bak_pg_dir_pointer[0].pgd = +- swapper_pg_dir[pgd_index(0)].pgd; +- efi_bak_pg_dir_pointer[1].pgd = +- swapper_pg_dir[pgd_index(0x400000)].pgd; +- swapper_pg_dir[pgd_index(0)].pgd = +- swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; +- temp = PAGE_OFFSET + 0x400000; +- swapper_pg_dir[pgd_index(0x400000)].pgd = +- swapper_pg_dir[pgd_index(temp)].pgd; +- } ++ clone_pgd_range(efi_bak_pg_dir_pointer, swapper_pg_dir, KERNEL_PGD_PTRS); ++ clone_pgd_range(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, ++ min_t(unsigned long, KERNEL_PGD_PTRS, KERNEL_PGD_BOUNDARY)); + + /* + * After the lock is released, the original page table is restored. + */ + __flush_tlb_all(); + +- gdt_descr.address = __pa(get_cpu_gdt_table(0)); ++ gdt_descr.address = (struct desc_struct *)__pa(get_cpu_gdt_table(0)); + gdt_descr.size = GDT_SIZE - 1; + load_gdt(&gdt_descr); + } + +-void efi_call_phys_epilog(void) ++void __init efi_call_phys_epilog(void) + { +- unsigned long cr4; + struct desc_ptr gdt_descr; + +- gdt_descr.address = (unsigned long)get_cpu_gdt_table(0); ++ gdt_descr.address = get_cpu_gdt_table(0); + gdt_descr.size = GDT_SIZE - 1; + load_gdt(&gdt_descr); + +- cr4 = read_cr4_safe(); +- +- if (cr4 & X86_CR4_PAE) { +- swapper_pg_dir[pgd_index(0)].pgd = +- efi_bak_pg_dir_pointer[0].pgd; +- } else { +- swapper_pg_dir[pgd_index(0)].pgd = +- efi_bak_pg_dir_pointer[0].pgd; +- swapper_pg_dir[pgd_index(0x400000)].pgd = +- efi_bak_pg_dir_pointer[1].pgd; +- } ++ clone_pgd_range(swapper_pg_dir, efi_bak_pg_dir_pointer, KERNEL_PGD_PTRS); + + /* + * After the lock is released, the original page table is restored. +diff -urNp linux-2.6.31.1/arch/x86/kernel/efi_stub_32.S linux-2.6.31.1/arch/x86/kernel/efi_stub_32.S +--- linux-2.6.31.1/arch/x86/kernel/efi_stub_32.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/efi_stub_32.S 2009-10-01 20:12:42.000000000 -0400 +@@ -6,6 +6,7 @@ + */ + + #include ++#include + #include + + /* +@@ -20,7 +21,7 @@ + * service functions will comply with gcc calling convention, too. + */ + +-.text ++__INIT + ENTRY(efi_call_phys) + /* + * 0. The function can only be called in Linux kernel. So CS has been +@@ -36,9 +37,7 @@ ENTRY(efi_call_phys) + * The mapping of lower virtual memory has been created in prelog and + * epilog. + */ +- movl $1f, %edx +- subl $__PAGE_OFFSET, %edx +- jmp *%edx ++ jmp 1f-__PAGE_OFFSET + 1: + + /* +@@ -47,14 +46,8 @@ ENTRY(efi_call_phys) + * parameter 2, ..., param n. To make things easy, we save the return + * address of efi_call_phys in a global variable. + */ +- popl %edx +- movl %edx, saved_return_addr +- /* get the function pointer into ECX*/ +- popl %ecx +- movl %ecx, efi_rt_function_ptr +- movl $2f, %edx +- subl $__PAGE_OFFSET, %edx +- pushl %edx ++ popl (saved_return_addr) ++ popl (efi_rt_function_ptr) + + /* + * 3. Clear PG bit in %CR0. +@@ -73,9 +66,8 @@ ENTRY(efi_call_phys) + /* + * 5. Call the physical function. + */ +- jmp *%ecx ++ call *(efi_rt_function_ptr-__PAGE_OFFSET) + +-2: + /* + * 6. After EFI runtime service returns, control will return to + * following instruction. We'd better readjust stack pointer first. +@@ -88,35 +80,28 @@ ENTRY(efi_call_phys) + movl %cr0, %edx + orl $0x80000000, %edx + movl %edx, %cr0 +- jmp 1f +-1: ++ + /* + * 8. Now restore the virtual mode from flat mode by + * adding EIP with PAGE_OFFSET. + */ +- movl $1f, %edx +- jmp *%edx ++ jmp 1f+__PAGE_OFFSET + 1: + + /* + * 9. Balance the stack. And because EAX contain the return value, + * we'd better not clobber it. + */ +- leal efi_rt_function_ptr, %edx +- movl (%edx), %ecx +- pushl %ecx ++ pushl (efi_rt_function_ptr) + + /* +- * 10. Push the saved return address onto the stack and return. ++ * 10. Return to the saved return address. + */ +- leal saved_return_addr, %edx +- movl (%edx), %ecx +- pushl %ecx +- ret ++ jmpl *(saved_return_addr) + ENDPROC(efi_call_phys) + .previous + +-.data ++__INITDATA + saved_return_addr: + .long 0 + efi_rt_function_ptr: +diff -urNp linux-2.6.31.1/arch/x86/kernel/entry_32.S linux-2.6.31.1/arch/x86/kernel/entry_32.S +--- linux-2.6.31.1/arch/x86/kernel/entry_32.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/entry_32.S 2009-10-01 20:12:42.000000000 -0400 +@@ -191,7 +191,7 @@ + + #endif /* CONFIG_X86_32_LAZY_GS */ + +-.macro SAVE_ALL ++.macro __SAVE_ALL _DS + cld + PUSH_GS + pushl %fs +@@ -224,7 +224,7 @@ + pushl %ebx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ebx, 0 +- movl $(__USER_DS), %edx ++ movl $\_DS, %edx + movl %edx, %ds + movl %edx, %es + movl $(__KERNEL_PERCPU), %edx +@@ -232,6 +232,21 @@ + SET_KERNEL_GS %edx + .endm + ++.macro SAVE_ALL ++#ifdef CONFIG_PAX_KERNEXEC ++ __SAVE_ALL __KERNEL_DS ++ GET_CR0_INTO_EDX; ++ movl %edx, %esi; ++ orl $X86_CR0_WP, %edx; ++ xorl %edx, %esi; ++ SET_CR0_FROM_EDX ++#elif defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF) ++ __SAVE_ALL __KERNEL_DS ++#else ++ __SAVE_ALL __USER_DS ++#endif ++.endm ++ + .macro RESTORE_INT_REGS + popl %ebx + CFI_ADJUST_CFA_OFFSET -4 +@@ -329,6 +344,11 @@ ENTRY(ret_from_fork) + CFI_ADJUST_CFA_OFFSET 4 + popfl + CFI_ADJUST_CFA_OFFSET -4 ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ xorl %esi, %esi ++#endif ++ + jmp syscall_exit + CFI_ENDPROC + END(ret_from_fork) +@@ -352,7 +372,17 @@ check_userspace: + movb PT_CS(%esp), %al + andl $(X86_EFLAGS_VM | SEGMENT_RPL_MASK), %eax + cmpl $USER_RPL, %eax ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ jae resume_userspace ++ ++ GET_CR0_INTO_EDX ++ xorl %esi, %edx ++ SET_CR0_FROM_EDX ++ jmp resume_kernel ++#else + jb resume_kernel # not returning to v8086 or userspace ++#endif + + ENTRY(resume_userspace) + LOCKDEP_SYS_EXIT +@@ -414,10 +444,9 @@ sysenter_past_esp: + /*CFI_REL_OFFSET cs, 0*/ + /* + * Push current_thread_info()->sysenter_return to the stack. +- * A tiny bit of offset fixup is necessary - 4*4 means the 4 words +- * pushed above; +8 corresponds to copy_thread's esp0 setting. + */ +- pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp) ++ GET_THREAD_INFO(%ebp) ++ pushl TI_sysenter_return(%ebp) + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET eip, 0 + +@@ -430,9 +459,19 @@ sysenter_past_esp: + * Load the potential sixth argument from user stack. + * Careful about security. + */ ++ movl PT_OLDESP(%esp),%ebp ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ mov PT_OLDSS(%esp),%ds ++1: movl %ds:(%ebp),%ebp ++ push %ss ++ pop %ds ++#else + cmpl $__PAGE_OFFSET-3,%ebp + jae syscall_fault + 1: movl (%ebp),%ebp ++#endif ++ + movl %ebp,PT_EBP(%esp) + .section __ex_table,"a" + .align 4 +@@ -455,12 +494,23 @@ sysenter_do_call: + testl $_TIF_ALLWORK_MASK, %ecx + jne sysexit_audit + sysenter_exit: ++ ++#ifdef CONFIG_PAX_RANDKSTACK ++ pushl %eax ++ CFI_ADJUST_CFA_OFFSET 4 ++ call pax_randomize_kstack ++ popl %eax ++ CFI_ADJUST_CFA_OFFSET -4 ++#endif ++ + /* if something modifies registers it must also disable sysexit */ + movl PT_EIP(%esp), %edx + movl PT_OLDESP(%esp), %ecx + xorl %ebp,%ebp + TRACE_IRQS_ON + 1: mov PT_FS(%esp), %fs ++2: mov PT_DS(%esp), %ds ++3: mov PT_ES(%esp), %es + PTGS_TO_GS + ENABLE_INTERRUPTS_SYSEXIT + +@@ -504,11 +554,17 @@ sysexit_audit: + + CFI_ENDPROC + .pushsection .fixup,"ax" +-2: movl $0,PT_FS(%esp) ++4: movl $0,PT_FS(%esp) ++ jmp 1b ++5: movl $0,PT_DS(%esp) ++ jmp 1b ++6: movl $0,PT_ES(%esp) + jmp 1b + .section __ex_table,"a" + .align 4 +- .long 1b,2b ++ .long 1b,4b ++ .long 2b,5b ++ .long 3b,6b + .popsection + PTGS_TO_GS_EX + ENDPROC(ia32_sysenter_target) +@@ -538,6 +594,10 @@ syscall_exit: + testl $_TIF_ALLWORK_MASK, %ecx # current->work + jne syscall_exit_work + ++#ifdef CONFIG_PAX_RANDKSTACK ++ call pax_randomize_kstack ++#endif ++ + restore_all: + TRACE_IRQS_IRET + restore_all_notrace: +@@ -602,7 +662,13 @@ ldt_ss: + mov PT_OLDESP(%esp), %eax /* load userspace esp */ + mov %dx, %ax /* eax: new kernel esp */ + sub %eax, %edx /* offset (low word is 0) */ +- PER_CPU(gdt_page, %ebx) ++#ifdef CONFIG_SMP ++ movl PER_CPU_VAR(cpu_number), %ebx ++ shll $PAGE_SHIFT_asm, %ebx ++ addl $cpu_gdt_table, %ebx ++#else ++ movl $cpu_gdt_table, %ebx ++#endif + shr $16, %edx + mov %dl, GDT_ENTRY_ESPFIX_SS * 8 + 4(%ebx) /* bits 16..23 */ + mov %dh, GDT_ENTRY_ESPFIX_SS * 8 + 7(%ebx) /* bits 24..31 */ +@@ -642,25 +708,19 @@ work_resched: + + work_notifysig: # deal with pending signals and + # notify-resume requests ++ movl %esp, %eax + #ifdef CONFIG_VM86 + testl $X86_EFLAGS_VM, PT_EFLAGS(%esp) +- movl %esp, %eax +- jne work_notifysig_v86 # returning to kernel-space or ++ jz 1f # returning to kernel-space or + # vm86-space +- xorl %edx, %edx +- call do_notify_resume +- jmp resume_userspace_sig + +- ALIGN +-work_notifysig_v86: + pushl %ecx # save ti_flags for do_notify_resume + CFI_ADJUST_CFA_OFFSET 4 + call save_v86_state # %eax contains pt_regs pointer + popl %ecx + CFI_ADJUST_CFA_OFFSET -4 + movl %eax, %esp +-#else +- movl %esp, %eax ++1: + #endif + xorl %edx, %edx + call do_notify_resume +@@ -695,6 +755,10 @@ END(syscall_exit_work) + + RING0_INT_FRAME # can't unwind into user space anyway + syscall_fault: ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ push %ss ++ pop %ds ++#endif + GET_THREAD_INFO(%ebp) + movl $-EFAULT,PT_EAX(%esp) + jmp resume_userspace +@@ -735,7 +799,13 @@ PTREGSCALL(vm86old) + * normal stack and adjusts ESP with the matching offset. + */ + /* fixup the stack */ +- PER_CPU(gdt_page, %ebx) ++#ifdef CONFIG_SMP ++ movl PER_CPU_VAR(cpu_number), %ebx ++ shll $PAGE_SHIFT_asm, %ebx ++ addl $cpu_gdt_table, %ebx ++#else ++ movl $cpu_gdt_table, %ebx ++#endif + mov GDT_ENTRY_ESPFIX_SS * 8 + 4(%ebx), %al /* bits 16..23 */ + mov GDT_ENTRY_ESPFIX_SS * 8 + 7(%ebx), %ah /* bits 24..31 */ + shl $16, %eax +@@ -1198,7 +1268,6 @@ return_to_handler: + ret + #endif + +-.section .rodata,"a" + #include "syscall_table_32.S" + + syscall_table_size=(.-sys_call_table) +@@ -1250,12 +1319,21 @@ error_code: + movl %ecx, %fs + UNWIND_ESPFIX_STACK + GS_TO_REG %ecx ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ GET_CR0_INTO_EDX ++ movl %edx, %esi ++ orl $X86_CR0_WP, %edx ++ xorl %edx, %esi ++ SET_CR0_FROM_EDX ++#endif ++ + movl PT_GS(%esp), %edi # get the function address + movl PT_ORIG_EAX(%esp), %edx # get the error code + movl $-1, PT_ORIG_EAX(%esp) # no syscall to restart + REG_TO_PTGS %ecx + SET_KERNEL_GS %ecx +- movl $(__USER_DS), %ecx ++ movl $(__KERNEL_DS), %ecx + movl %ecx, %ds + movl %ecx, %es + TRACE_IRQS_OFF +@@ -1351,6 +1429,13 @@ nmi_stack_correct: + xorl %edx,%edx # zero error code + movl %esp,%eax # pt_regs pointer + call do_nmi ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ GET_CR0_INTO_EDX ++ xorl %esi, %edx ++ SET_CR0_FROM_EDX ++#endif ++ + jmp restore_all_notrace + CFI_ENDPROC + +@@ -1391,6 +1476,13 @@ nmi_espfix_stack: + FIXUP_ESPFIX_STACK # %eax == %esp + xorl %edx,%edx # zero error code + call do_nmi ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ GET_CR0_INTO_EDX ++ xorl %esi, %edx ++ SET_CR0_FROM_EDX ++#endif ++ + RESTORE_REGS + lss 12+4(%esp), %esp # back to espfix stack + CFI_ADJUST_CFA_OFFSET -24 +diff -urNp linux-2.6.31.1/arch/x86/kernel/entry_64.S linux-2.6.31.1/arch/x86/kernel/entry_64.S +--- linux-2.6.31.1/arch/x86/kernel/entry_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/entry_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -1074,7 +1074,12 @@ ENTRY(\sym) + TRACE_IRQS_OFF + movq %rsp,%rdi /* pt_regs pointer */ + xorl %esi,%esi /* no error code */ +- PER_CPU(init_tss, %rbp) ++#ifdef CONFIG_SMP ++ imul $TSS_size, PER_CPU_VAR(cpu_number), %ebp ++ lea init_tss(%rbp), %rbp ++#else ++ lea init_tss(%rip), %rbp ++#endif + subq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%rbp) + call \do_sym + addq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%rbp) +diff -urNp linux-2.6.31.1/arch/x86/kernel/ftrace.c linux-2.6.31.1/arch/x86/kernel/ftrace.c +--- linux-2.6.31.1/arch/x86/kernel/ftrace.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/ftrace.c 2009-10-01 20:12:42.000000000 -0400 +@@ -284,9 +284,9 @@ int ftrace_update_ftrace_func(ftrace_fun + unsigned char old[MCOUNT_INSN_SIZE], *new; + int ret; + +- memcpy(old, &ftrace_call, MCOUNT_INSN_SIZE); ++ memcpy(old, (void *)ktla_ktva((unsigned long)ftrace_call), MCOUNT_INSN_SIZE); + new = ftrace_call_replace(ip, (unsigned long)func); +- ret = ftrace_modify_code(ip, old, new); ++ ret = ftrace_modify_code(ktla_ktva(ip), old, new); + + return ret; + } +diff -urNp linux-2.6.31.1/arch/x86/kernel/head32.c linux-2.6.31.1/arch/x86/kernel/head32.c +--- linux-2.6.31.1/arch/x86/kernel/head32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/head32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -13,12 +13,13 @@ + #include + #include + #include ++#include + + void __init i386_start_kernel(void) + { + reserve_trampoline_memory(); + +- reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); ++ reserve_early(LOAD_PHYSICAL_ADDR, __pa_symbol(&__bss_stop), "TEXT DATA BSS"); + + #ifdef CONFIG_BLK_DEV_INITRD + /* Reserve INITRD */ +diff -urNp linux-2.6.31.1/arch/x86/kernel/head_32.S linux-2.6.31.1/arch/x86/kernel/head_32.S +--- linux-2.6.31.1/arch/x86/kernel/head_32.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/head_32.S 2009-10-01 20:12:42.000000000 -0400 +@@ -19,6 +19,7 @@ + #include + #include + #include ++#include + + /* Physical address */ + #define pa(X) ((X) - __PAGE_OFFSET) +@@ -52,11 +53,7 @@ + * and small than max_low_pfn, otherwise will waste some page table entries + */ + +-#if PTRS_PER_PMD > 1 +-#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) + PTRS_PER_PGD) +-#else +-#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD) +-#endif ++#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PTE) + + /* Enough space to fit pagetables for the low memory linear map */ + MAPPING_BEYOND_END = \ +@@ -73,6 +70,12 @@ INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_P + RESERVE_BRK(pagetables, INIT_MAP_SIZE) + + /* ++ * Real beginning of normal "text" segment ++ */ ++ENTRY(stext) ++ENTRY(_stext) ++ ++/* + * 32-bit kernel entrypoint; only used by the boot CPU. On entry, + * %esi points to the real-mode code as a 32-bit pointer. + * CS and DS must be 4 GB flat segments, but we don't depend on +@@ -80,6 +83,13 @@ RESERVE_BRK(pagetables, INIT_MAP_SIZE) + * can. + */ + .section .text.head,"ax",@progbits ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ jmp startup_32 ++/* PaX: fill first page in .text with int3 to catch NULL derefs in kernel mode */ ++.fill PAGE_SIZE-5,1,0xcc ++#endif ++ + ENTRY(startup_32) + /* test KEEP_SEGMENTS flag to see if the bootloader is asking + us to not reload segments */ +@@ -97,6 +107,48 @@ ENTRY(startup_32) + movl %eax,%gs + 2: + ++#ifdef CONFIG_SMP ++ movl $pa(cpu_gdt_table),%edi ++ movl $__per_cpu_load,%eax ++ movw %ax,__KERNEL_PERCPU + 2(%edi) ++ rorl $16,%eax ++ movb %al,__KERNEL_PERCPU + 4(%edi) ++ movb %ah,__KERNEL_PERCPU + 7(%edi) ++ movl $__per_cpu_end - 1,%eax ++ subl $__per_cpu_load,%eax ++ movw %ax,__KERNEL_PERCPU + 0(%edi) ++#endif ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ movl $NR_CPUS,%ecx ++ movl $pa(cpu_gdt_table),%edi ++1: ++ movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c09700),GDT_ENTRY_KERNEL_DS * 8 + 4(%edi) ++ addl $PAGE_SIZE_asm,%edi ++ loop 1b ++#endif ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ movl $pa(boot_gdt),%edi ++ movl $__LOAD_PHYSICAL_ADDR + __PAGE_OFFSET,%eax ++ movw %ax,__BOOT_CS + 2(%edi) ++ rorl $16,%eax ++ movb %al,__BOOT_CS + 4(%edi) ++ movb %ah,__BOOT_CS + 7(%edi) ++ rorl $16,%eax ++ ++ movl $NR_CPUS,%ecx ++ movl $pa(cpu_gdt_table),%edi ++1: ++ movw %ax,__KERNEL_CS + 2(%edi) ++ rorl $16,%eax ++ movb %al,__KERNEL_CS + 4(%edi) ++ movb %ah,__KERNEL_CS + 7(%edi) ++ rorl $16,%eax ++ addl $PAGE_SIZE_asm,%edi ++ loop 1b ++#endif ++ + /* + * Clear BSS first so that there are no surprises... + */ +@@ -140,9 +192,7 @@ ENTRY(startup_32) + cmpl $num_subarch_entries, %eax + jae bad_subarch + +- movl pa(subarch_entries)(,%eax,4), %eax +- subl $__PAGE_OFFSET, %eax +- jmp *%eax ++ jmp *pa(subarch_entries)(,%eax,4) + + bad_subarch: + WEAK(lguest_entry) +@@ -154,9 +204,9 @@ WEAK(xen_entry) + __INITDATA + + subarch_entries: +- .long default_entry /* normal x86/PC */ +- .long lguest_entry /* lguest hypervisor */ +- .long xen_entry /* Xen hypervisor */ ++ .long pa(default_entry) /* normal x86/PC */ ++ .long pa(lguest_entry) /* lguest hypervisor */ ++ .long pa(xen_entry) /* Xen hypervisor */ + num_subarch_entries = (. - subarch_entries) / 4 + .previous + #endif /* CONFIG_PARAVIRT */ +@@ -217,8 +267,11 @@ default_entry: + movl %eax, pa(max_pfn_mapped) + + /* Do early initialization of the fixmap area */ +- movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax +- movl %eax,pa(swapper_pg_pmd+0x1000*KPMDS-8) ++#ifdef CONFIG_COMPAT_VDSO ++ movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR+_PAGE_USER,pa(swapper_pg_pmd+0x1000*KPMDS-8) ++#else ++ movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,pa(swapper_pg_pmd+0x1000*KPMDS-8) ++#endif + #else /* Not PAE */ + + page_pde_offset = (__PAGE_OFFSET >> 20); +@@ -248,8 +301,11 @@ page_pde_offset = (__PAGE_OFFSET >> 20); + movl %eax, pa(max_pfn_mapped) + + /* Do early initialization of the fixmap area */ +- movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax +- movl %eax,pa(swapper_pg_dir+0xffc) ++#ifdef CONFIG_COMPAT_VDSO ++ movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR+_PAGE_USER,pa(swapper_pg_dir+0xffc) ++#else ++ movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,pa(swapper_pg_dir+0xffc) ++#endif + #endif + jmp 3f + /* +@@ -296,6 +352,7 @@ ENTRY(startup_32_smp) + orl %edx,%eax + movl %eax,%cr4 + ++#ifdef CONFIG_X86_PAE + btl $5, %eax # check if PAE is enabled + jnc 6f + +@@ -311,13 +368,16 @@ ENTRY(startup_32_smp) + jnc 6f + + /* Setup EFER (Extended Feature Enable Register) */ +- movl $0xc0000080, %ecx ++ movl $MSR_EFER, %ecx + rdmsr + + btsl $11, %eax + /* Make changes effective */ + wrmsr + ++ btsl $_PAGE_BIT_NX-32,pa(__supported_pte_mask+4) ++ movl $1,pa(nx_enabled) ++#endif + 6: + + /* +@@ -343,9 +403,7 @@ ENTRY(startup_32_smp) + + #ifdef CONFIG_SMP + cmpb $0, ready +- jz 1f /* Initial CPU cleans BSS */ +- jmp checkCPUtype +-1: ++ jnz checkCPUtype /* Initial CPU cleans BSS */ + #endif /* CONFIG_SMP */ + + /* +@@ -423,7 +481,7 @@ is386: movl $2,%ecx # set MP + 1: movl $(__KERNEL_DS),%eax # reload all the segment registers + movl %eax,%ss # after changing gdt. + +- movl $(__USER_DS),%eax # DS/ES contains default USER segment ++# movl $(__KERNEL_DS),%eax # DS/ES contains default KERNEL segment + movl %eax,%ds + movl %eax,%es + +@@ -437,8 +495,11 @@ is386: movl $2,%ecx # set MP + */ + cmpb $0,ready + jne 1f +- movl $per_cpu__gdt_page,%eax ++ movl $cpu_gdt_table,%eax + movl $per_cpu__stack_canary,%ecx ++#ifdef CONFIG_SMP ++ addl $__per_cpu_load,%ecx ++#endif + movw %cx, 8 * GDT_ENTRY_STACK_CANARY + 2(%eax) + shrl $16, %ecx + movb %cl, 8 * GDT_ENTRY_STACK_CANARY + 4(%eax) +@@ -456,10 +517,6 @@ is386: movl $2,%ecx # set MP + #ifdef CONFIG_SMP + movb ready, %cl + movb $1, ready +- cmpb $0,%cl # the first CPU calls start_kernel +- je 1f +- movl (stack_start), %esp +-1: + #endif /* CONFIG_SMP */ + jmp *(initial_code) + +@@ -545,22 +602,22 @@ early_page_fault: + jmp early_fault + + early_fault: +- cld + #ifdef CONFIG_PRINTK ++ cmpl $1,%ss:early_recursion_flag ++ je hlt_loop ++ incl %ss:early_recursion_flag ++ cld + pusha + movl $(__KERNEL_DS),%eax + movl %eax,%ds + movl %eax,%es +- cmpl $2,early_recursion_flag +- je hlt_loop +- incl early_recursion_flag + movl %cr2,%eax + pushl %eax + pushl %edx /* trapno */ + pushl $fault_msg + call printk ++; call dump_stack + #endif +- call dump_stack + hlt_loop: + hlt + jmp hlt_loop +@@ -568,8 +625,11 @@ hlt_loop: + /* This is the default interrupt "handler" :-) */ + ALIGN + ignore_int: +- cld + #ifdef CONFIG_PRINTK ++ cmpl $2,%ss:early_recursion_flag ++ je hlt_loop ++ incl %ss:early_recursion_flag ++ cld + pushl %eax + pushl %ecx + pushl %edx +@@ -578,9 +638,6 @@ ignore_int: + movl $(__KERNEL_DS),%eax + movl %eax,%ds + movl %eax,%es +- cmpl $2,early_recursion_flag +- je hlt_loop +- incl early_recursion_flag + pushl 16(%esp) + pushl 24(%esp) + pushl 32(%esp) +@@ -607,27 +664,37 @@ ENTRY(initial_code) + /* + * BSS section + */ +-.section ".bss.page_aligned","wa" +- .align PAGE_SIZE_asm + #ifdef CONFIG_X86_PAE ++.section .swapper_pg_pmd,"a",@progbits + swapper_pg_pmd: + .fill 1024*KPMDS,4,0 + #else ++.section .swapper_pg_dir,"a",@progbits + ENTRY(swapper_pg_dir) + .fill 1024,4,0 + #endif ++ + swapper_pg_fixmap: + .fill 1024,4,0 ++ ++.section .empty_zero_page,"a",@progbits + ENTRY(empty_zero_page) + .fill 4096,1,0 + + /* ++ * The IDT has to be page-aligned to simplify the Pentium ++ * F0 0F bug workaround.. We have a special link segment ++ * for this. ++ */ ++.section .idt,"a",@progbits ++ENTRY(idt_table) ++ .fill 256,8,0 ++ ++/* + * This starts the data section. + */ + #ifdef CONFIG_X86_PAE +-.section ".data.page_aligned","wa" +- /* Page-aligned for the benefit of paravirt? */ +- .align PAGE_SIZE_asm ++.section .swapper_pg_dir,"a",@progbits + ENTRY(swapper_pg_dir) + .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 /* low identity map */ + # if KPMDS == 3 +@@ -650,11 +717,12 @@ ENTRY(swapper_pg_dir) + + .data + ENTRY(stack_start) +- .long init_thread_union+THREAD_SIZE ++ .long init_thread_union+THREAD_SIZE-8 + .long __BOOT_DS + + ready: .byte 0 + ++.section .rodata,"a",@progbits + early_recursion_flag: + .long 0 + +@@ -690,7 +758,7 @@ fault_msg: + .word 0 # 32 bit align gdt_desc.address + boot_gdt_descr: + .word __BOOT_DS+7 +- .long boot_gdt - __PAGE_OFFSET ++ .long pa(boot_gdt) + + .word 0 # 32-bit align idt_desc.address + idt_descr: +@@ -701,7 +769,7 @@ idt_descr: + .word 0 # 32 bit align gdt_desc.address + ENTRY(early_gdt_descr) + .word GDT_ENTRIES*8-1 +- .long per_cpu__gdt_page /* Overwritten for secondary CPUs */ ++ .long cpu_gdt_table /* Overwritten for secondary CPUs */ + + /* + * The boot_gdt must mirror the equivalent in setup.S and is +@@ -710,5 +778,59 @@ ENTRY(early_gdt_descr) + .align L1_CACHE_BYTES + ENTRY(boot_gdt) + .fill GDT_ENTRY_BOOT_CS,8,0 +- .quad 0x00cf9a000000ffff /* kernel 4GB code at 0x00000000 */ +- .quad 0x00cf92000000ffff /* kernel 4GB data at 0x00000000 */ ++ .quad 0x00cf9b000000ffff /* kernel 4GB code at 0x00000000 */ ++ .quad 0x00cf93000000ffff /* kernel 4GB data at 0x00000000 */ ++ ++ .align PAGE_SIZE_asm ++ENTRY(cpu_gdt_table) ++ .rept NR_CPUS ++ .quad 0x0000000000000000 /* NULL descriptor */ ++ .quad 0x0000000000000000 /* 0x0b reserved */ ++ .quad 0x0000000000000000 /* 0x13 reserved */ ++ .quad 0x0000000000000000 /* 0x1b reserved */ ++ .quad 0x0000000000000000 /* 0x20 unused */ ++ .quad 0x0000000000000000 /* 0x28 unused */ ++ .quad 0x0000000000000000 /* 0x33 TLS entry 1 */ ++ .quad 0x0000000000000000 /* 0x3b TLS entry 2 */ ++ .quad 0x0000000000000000 /* 0x43 TLS entry 3 */ ++ .quad 0x0000000000000000 /* 0x4b reserved */ ++ .quad 0x0000000000000000 /* 0x53 reserved */ ++ .quad 0x0000000000000000 /* 0x5b reserved */ ++ ++ .quad 0x00cf9b000000ffff /* 0x60 kernel 4GB code at 0x00000000 */ ++ .quad 0x00cf93000000ffff /* 0x68 kernel 4GB data at 0x00000000 */ ++ .quad 0x00cffb000000ffff /* 0x73 user 4GB code at 0x00000000 */ ++ .quad 0x00cff3000000ffff /* 0x7b user 4GB data at 0x00000000 */ ++ ++ .quad 0x0000000000000000 /* 0x80 TSS descriptor */ ++ .quad 0x0000000000000000 /* 0x88 LDT descriptor */ ++ ++ /* ++ * Segments used for calling PnP BIOS have byte granularity. ++ * The code segments and data segments have fixed 64k limits, ++ * the transfer segment sizes are set at run time. ++ */ ++ .quad 0x00409b000000ffff /* 0x90 32-bit code */ ++ .quad 0x00009b000000ffff /* 0x98 16-bit code */ ++ .quad 0x000093000000ffff /* 0xa0 16-bit data */ ++ .quad 0x0000930000000000 /* 0xa8 16-bit data */ ++ .quad 0x0000930000000000 /* 0xb0 16-bit data */ ++ ++ /* ++ * The APM segments have byte granularity and their bases ++ * are set at run time. All have 64k limits. ++ */ ++ .quad 0x00409b000000ffff /* 0xb8 APM CS code */ ++ .quad 0x00009b000000ffff /* 0xc0 APM CS 16 code (16 bit) */ ++ .quad 0x004093000000ffff /* 0xc8 APM DS data */ ++ ++ .quad 0x00c0930000000000 /* 0xd0 - ESPFIX SS */ ++ .quad 0x0040930000000000 /* 0xd8 - PERCPU */ ++ .quad 0x0040930000000018 /* 0xe0 - STACK_CANARY */ ++ .quad 0x0000000000000000 /* 0xe8 - PCIBIOS_CS */ ++ .quad 0x0000000000000000 /* 0xf0 - PCIBIOS_DS */ ++ .quad 0x0000000000000000 /* 0xf8 - GDT entry 31: double-fault TSS */ ++ ++ /* Be sure this is zeroed to avoid false validations in Xen */ ++ .fill PAGE_SIZE_asm - GDT_SIZE,1,0 ++ .endr +diff -urNp linux-2.6.31.1/arch/x86/kernel/head_64.S linux-2.6.31.1/arch/x86/kernel/head_64.S +--- linux-2.6.31.1/arch/x86/kernel/head_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/head_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -38,6 +38,10 @@ L4_PAGE_OFFSET = pgd_index(__PAGE_OFFSET + L3_PAGE_OFFSET = pud_index(__PAGE_OFFSET) + L4_START_KERNEL = pgd_index(__START_KERNEL_map) + L3_START_KERNEL = pud_index(__START_KERNEL_map) ++L4_VMALLOC_START = pgd_index(VMALLOC_START) ++L3_VMALLOC_START = pud_index(VMALLOC_START) ++L4_VMEMMAP_START = pgd_index(VMEMMAP_START) ++L3_VMEMMAP_START = pud_index(VMEMMAP_START) + + .text + .section .text.head +@@ -85,35 +89,22 @@ startup_64: + */ + addq %rbp, init_level4_pgt + 0(%rip) + addq %rbp, init_level4_pgt + (L4_PAGE_OFFSET*8)(%rip) ++ addq %rbp, init_level4_pgt + (L4_VMALLOC_START*8)(%rip) ++ addq %rbp, init_level4_pgt + (L4_VMEMMAP_START*8)(%rip) + addq %rbp, init_level4_pgt + (L4_START_KERNEL*8)(%rip) + + addq %rbp, level3_ident_pgt + 0(%rip) ++ addq %rbp, level3_ident_pgt + 8(%rip) ++ addq %rbp, level3_ident_pgt + 16(%rip) ++ addq %rbp, level3_ident_pgt + 24(%rip) + +- addq %rbp, level3_kernel_pgt + (510*8)(%rip) +- addq %rbp, level3_kernel_pgt + (511*8)(%rip) ++ addq %rbp, level3_vmemmap_pgt + (L3_VMEMMAP_START*8)(%rip) + +- addq %rbp, level2_fixmap_pgt + (506*8)(%rip) ++ addq %rbp, level3_kernel_pgt + (L3_START_KERNEL*8)(%rip) ++ addq %rbp, level3_kernel_pgt + (L3_START_KERNEL*8+8)(%rip) + +- /* Add an Identity mapping if I am above 1G */ +- leaq _text(%rip), %rdi +- andq $PMD_PAGE_MASK, %rdi +- +- movq %rdi, %rax +- shrq $PUD_SHIFT, %rax +- andq $(PTRS_PER_PUD - 1), %rax +- jz ident_complete +- +- leaq (level2_spare_pgt - __START_KERNEL_map + _KERNPG_TABLE)(%rbp), %rdx +- leaq level3_ident_pgt(%rip), %rbx +- movq %rdx, 0(%rbx, %rax, 8) +- +- movq %rdi, %rax +- shrq $PMD_SHIFT, %rax +- andq $(PTRS_PER_PMD - 1), %rax +- leaq __PAGE_KERNEL_IDENT_LARGE_EXEC(%rdi), %rdx +- leaq level2_spare_pgt(%rip), %rbx +- movq %rdx, 0(%rbx, %rax, 8) +-ident_complete: ++ addq %rbp, level2_fixmap_pgt + (506*8)(%rip) ++ addq %rbp, level2_fixmap_pgt + (507*8)(%rip) + + /* + * Fixup the kernel text+data virtual addresses. Note that +@@ -187,6 +178,10 @@ ENTRY(secondary_startup_64) + btl $20,%edi /* No Execute supported? */ + jnc 1f + btsl $_EFER_NX, %eax ++ leaq init_level4_pgt(%rip), %rdi ++ btsq $_PAGE_BIT_NX, 8*L4_PAGE_OFFSET(%rdi) ++ btsq $_PAGE_BIT_NX, 8*L4_VMALLOC_START(%rdi) ++ btsq $_PAGE_BIT_NX, 8*L4_VMEMMAP_START(%rdi) + 1: wrmsr /* Make changes effective */ + + /* Setup cr0 */ +@@ -262,16 +257,16 @@ ENTRY(secondary_startup_64) + .quad x86_64_start_kernel + ENTRY(initial_gs) + .quad INIT_PER_CPU_VAR(irq_stack_union) +- __FINITDATA + + ENTRY(stack_start) + .quad init_thread_union+THREAD_SIZE-8 + .word 0 ++ __FINITDATA + + bad_address: + jmp bad_address + +- .section ".init.text","ax" ++ __INIT + #ifdef CONFIG_EARLY_PRINTK + .globl early_idt_handlers + early_idt_handlers: +@@ -316,18 +311,23 @@ ENTRY(early_idt_handler) + #endif /* EARLY_PRINTK */ + 1: hlt + jmp 1b ++ .previous + + #ifdef CONFIG_EARLY_PRINTK ++ __INITDATA + early_recursion_flag: + .long 0 ++ .previous + ++ .section .rodata,"a",@progbits + early_idt_msg: + .asciz "PANIC: early exception %02lx rip %lx:%lx error %lx cr2 %lx\n" + early_idt_ripmsg: + .asciz "RIP %s\n" +-#endif /* CONFIG_EARLY_PRINTK */ + .previous ++#endif /* CONFIG_EARLY_PRINTK */ + ++ .section .rodata,"a",@progbits + #define NEXT_PAGE(name) \ + .balign PAGE_SIZE; \ + ENTRY(name) +@@ -350,13 +350,31 @@ NEXT_PAGE(init_level4_pgt) + .quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE + .org init_level4_pgt + L4_PAGE_OFFSET*8, 0 + .quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE ++ .org init_level4_pgt + L4_VMALLOC_START*8, 0 ++ .quad level3_vmalloc_pgt - __START_KERNEL_map + _KERNPG_TABLE ++ .org init_level4_pgt + L4_VMEMMAP_START*8, 0 ++ .quad level3_vmemmap_pgt - __START_KERNEL_map + _KERNPG_TABLE + .org init_level4_pgt + L4_START_KERNEL*8, 0 + /* (2^48-(2*1024*1024*1024))/(2^39) = 511 */ + .quad level3_kernel_pgt - __START_KERNEL_map + _PAGE_TABLE + + NEXT_PAGE(level3_ident_pgt) + .quad level2_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE ++#ifdef CONFIG_XEN + .fill 511,8,0 ++#else ++ .quad level2_ident_pgt + PAGE_SIZE - __START_KERNEL_map + _KERNPG_TABLE ++ .quad level2_ident_pgt + 2*PAGE_SIZE - __START_KERNEL_map + _KERNPG_TABLE ++ .quad level2_ident_pgt + 3*PAGE_SIZE - __START_KERNEL_map + _KERNPG_TABLE ++ .fill 508,8,0 ++#endif ++ ++NEXT_PAGE(level3_vmalloc_pgt) ++ .fill 512,8,0 ++ ++NEXT_PAGE(level3_vmemmap_pgt) ++ .fill L3_VMEMMAP_START,8,0 ++ .quad level2_vmemmap_pgt - __START_KERNEL_map + _KERNPG_TABLE + + NEXT_PAGE(level3_kernel_pgt) + .fill L3_START_KERNEL,8,0 +@@ -364,20 +382,23 @@ NEXT_PAGE(level3_kernel_pgt) + .quad level2_kernel_pgt - __START_KERNEL_map + _KERNPG_TABLE + .quad level2_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE + ++NEXT_PAGE(level2_vmemmap_pgt) ++ .fill 512,8,0 ++ + NEXT_PAGE(level2_fixmap_pgt) +- .fill 506,8,0 +- .quad level1_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE +- /* 8MB reserved for vsyscalls + a 2MB hole = 4 + 1 entries */ +- .fill 5,8,0 ++ .fill 507,8,0 ++ .quad level1_vsyscall_pgt - __START_KERNEL_map + _PAGE_TABLE ++ /* 6MB reserved for vsyscalls + a 2MB hole = 3 + 1 entries */ ++ .fill 4,8,0 + +-NEXT_PAGE(level1_fixmap_pgt) ++NEXT_PAGE(level1_vsyscall_pgt) + .fill 512,8,0 + +-NEXT_PAGE(level2_ident_pgt) +- /* Since I easily can, map the first 1G. ++ /* Since I easily can, map the first 4G. + * Don't set NX because code runs from these pages. + */ +- PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD) ++NEXT_PAGE(level2_ident_pgt) ++ PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, 4*PTRS_PER_PMD) + + NEXT_PAGE(level2_kernel_pgt) + /* +@@ -390,33 +411,49 @@ NEXT_PAGE(level2_kernel_pgt) + * If you want to increase this then increase MODULES_VADDR + * too.) + */ +- PMDS(0, __PAGE_KERNEL_LARGE_EXEC, +- KERNEL_IMAGE_SIZE/PMD_SIZE) +- +-NEXT_PAGE(level2_spare_pgt) +- .fill 512, 8, 0 ++ PMDS(0, __PAGE_KERNEL_LARGE_EXEC, KERNEL_IMAGE_SIZE/PMD_SIZE) + + #undef PMDS + #undef NEXT_PAGE + +- .data ++ .align PAGE_SIZE ++ENTRY(cpu_gdt_table) ++ .rept NR_CPUS ++ .quad 0x0000000000000000 /* NULL descriptor */ ++ .quad 0x00cf9b000000ffff /* __KERNEL32_CS */ ++ .quad 0x00af9b000000ffff /* __KERNEL_CS */ ++ .quad 0x00cf93000000ffff /* __KERNEL_DS */ ++ .quad 0x00cffb000000ffff /* __USER32_CS */ ++ .quad 0x00cff3000000ffff /* __USER_DS, __USER32_DS */ ++ .quad 0x00affb000000ffff /* __USER_CS */ ++ .quad 0x0 /* unused */ ++ .quad 0,0 /* TSS */ ++ .quad 0,0 /* LDT */ ++ .quad 0,0,0 /* three TLS descriptors */ ++ .quad 0x0000f40000000000 /* node/CPU stored in limit */ ++ /* asm/segment.h:GDT_ENTRIES must match this */ ++ ++ /* zero the remaining page */ ++ .fill PAGE_SIZE / 8 - GDT_ENTRIES,8,0 ++ .endr ++ + .align 16 + .globl early_gdt_descr + early_gdt_descr: + .word GDT_ENTRIES*8-1 + early_gdt_descr_base: +- .quad INIT_PER_CPU_VAR(gdt_page) ++ .quad cpu_gdt_table + + ENTRY(phys_base) + /* This must match the first entry in level2_kernel_pgt */ + .quad 0x0000000000000000 + + #include "../../x86/xen/xen-head.S" +- +- .section .bss, "aw", @nobits ++ ++ .section .rodata,"a",@progbits + .align L1_CACHE_BYTES + ENTRY(idt_table) +- .skip IDT_ENTRIES * 16 ++ .fill 512,8,0 + + .section .bss.page_aligned, "aw", @nobits + .align PAGE_SIZE +diff -urNp linux-2.6.31.1/arch/x86/kernel/i386_ksyms_32.c linux-2.6.31.1/arch/x86/kernel/i386_ksyms_32.c +--- linux-2.6.31.1/arch/x86/kernel/i386_ksyms_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/i386_ksyms_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -10,8 +10,12 @@ + EXPORT_SYMBOL(mcount); + #endif + ++EXPORT_SYMBOL_GPL(cpu_gdt_table); ++ + /* Networking helper routines. */ + EXPORT_SYMBOL(csum_partial_copy_generic); ++EXPORT_SYMBOL(csum_partial_copy_generic_to_user); ++EXPORT_SYMBOL(csum_partial_copy_generic_from_user); + + EXPORT_SYMBOL(__get_user_1); + EXPORT_SYMBOL(__get_user_2); +@@ -26,3 +30,7 @@ EXPORT_SYMBOL(strstr); + + EXPORT_SYMBOL(csum_partial); + EXPORT_SYMBOL(empty_zero_page); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++EXPORT_SYMBOL(__LOAD_PHYSICAL_ADDR); ++#endif +diff -urNp linux-2.6.31.1/arch/x86/kernel/init_task.c linux-2.6.31.1/arch/x86/kernel/init_task.c +--- linux-2.6.31.1/arch/x86/kernel/init_task.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/init_task.c 2009-10-01 20:12:42.000000000 -0400 +@@ -39,5 +39,5 @@ EXPORT_SYMBOL(init_task); + * section. Since TSS's are completely CPU-local, we want them + * on exact cacheline boundaries, to eliminate cacheline ping-pong. + */ +-DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss) = INIT_TSS; +- ++struct tss_struct init_tss[NR_CPUS] ____cacheline_internodealigned_in_smp = { [0 ... NR_CPUS-1] = INIT_TSS }; ++EXPORT_SYMBOL(init_tss); +diff -urNp linux-2.6.31.1/arch/x86/kernel/ioport.c linux-2.6.31.1/arch/x86/kernel/ioport.c +--- linux-2.6.31.1/arch/x86/kernel/ioport.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/ioport.c 2009-10-01 20:12:42.000000000 -0400 +@@ -6,6 +6,7 @@ + #include + #include + #include ++#include + #include + #include + #include +@@ -41,6 +42,12 @@ asmlinkage long sys_ioperm(unsigned long + + if ((from + num <= from) || (from + num > IO_BITMAP_BITS)) + return -EINVAL; ++#ifdef CONFIG_GRKERNSEC_IO ++ if (turn_on) { ++ gr_handle_ioperm(); ++ return -EPERM; ++ } ++#endif + if (turn_on && !capable(CAP_SYS_RAWIO)) + return -EPERM; + +@@ -67,7 +74,7 @@ asmlinkage long sys_ioperm(unsigned long + * because the ->io_bitmap_max value must match the bitmap + * contents: + */ +- tss = &per_cpu(init_tss, get_cpu()); ++ tss = init_tss + get_cpu(); + + set_bitmap(t->io_bitmap_ptr, from, num, !turn_on); + +@@ -111,8 +118,13 @@ static int do_iopl(unsigned int level, s + return -EINVAL; + /* Trying to gain more privileges? */ + if (level > old) { ++#ifdef CONFIG_GRKERNSEC_IO ++ gr_handle_iopl(); ++ return -EPERM; ++#else + if (!capable(CAP_SYS_RAWIO)) + return -EPERM; ++#endif + } + regs->flags = (regs->flags & ~X86_EFLAGS_IOPL) | (level << 12); + +diff -urNp linux-2.6.31.1/arch/x86/kernel/irq_32.c linux-2.6.31.1/arch/x86/kernel/irq_32.c +--- linux-2.6.31.1/arch/x86/kernel/irq_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/irq_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -94,7 +94,7 @@ execute_on_irq_stack(int overflow, struc + return 0; + + /* build the stack frame on the IRQ stack */ +- isp = (u32 *) ((char *)irqctx + sizeof(*irqctx)); ++ isp = (u32 *) ((char *)irqctx + sizeof(*irqctx) - 8); + irqctx->tinfo.task = curctx->tinfo.task; + irqctx->tinfo.previous_esp = current_stack_pointer; + +@@ -175,7 +175,7 @@ asmlinkage void do_softirq(void) + irqctx->tinfo.previous_esp = current_stack_pointer; + + /* build the stack frame on the softirq stack */ +- isp = (u32 *) ((char *)irqctx + sizeof(*irqctx)); ++ isp = (u32 *) ((char *)irqctx + sizeof(*irqctx) - 8); + + call_on_stack(__do_softirq, isp); + /* +diff -urNp linux-2.6.31.1/arch/x86/kernel/kprobes.c linux-2.6.31.1/arch/x86/kernel/kprobes.c +--- linux-2.6.31.1/arch/x86/kernel/kprobes.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/kprobes.c 2009-10-01 20:12:42.000000000 -0400 +@@ -166,9 +166,24 @@ static void __kprobes set_jmp_op(void *f + char op; + s32 raddr; + } __attribute__((packed)) * jop; +- jop = (struct __arch_jmp_op *)from; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ ++ jop = (struct __arch_jmp_op *)(ktla_ktva(from)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + jop->raddr = (s32)((long)(to) - ((long)(from) + 5)); + jop->op = RELATIVEJUMP_INSTRUCTION; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } + + /* +@@ -345,16 +360,29 @@ static void __kprobes fix_riprel(struct + + static void __kprobes arch_copy_kprobe(struct kprobe *p) + { +- memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ ++ memcpy(p->ainsn.insn, ktla_ktva(p->addr), MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif + + fix_riprel(p); + +- if (can_boost(p->addr)) ++ if (can_boost(ktla_ktva(p->addr))) + p->ainsn.boostable = 0; + else + p->ainsn.boostable = -1; + +- p->opcode = *p->addr; ++ p->opcode = *(ktla_ktva(p->addr)); + } + + int __kprobes arch_prepare_kprobe(struct kprobe *p) +@@ -432,7 +460,7 @@ static void __kprobes prepare_singlestep + if (p->opcode == BREAKPOINT_INSTRUCTION) + regs->ip = (unsigned long)p->addr; + else +- regs->ip = (unsigned long)p->ainsn.insn; ++ regs->ip = ktva_ktla((unsigned long)p->ainsn.insn); + } + + void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, +@@ -453,7 +481,7 @@ static void __kprobes setup_singlestep(s + if (p->ainsn.boostable == 1 && !p->post_handler) { + /* Boost up -- we can execute copied instructions directly */ + reset_current_kprobe(); +- regs->ip = (unsigned long)p->ainsn.insn; ++ regs->ip = ktva_ktla((unsigned long)p->ainsn.insn); + preempt_enable_no_resched(); + return; + } +@@ -523,7 +551,7 @@ static int __kprobes kprobe_handler(stru + struct kprobe_ctlblk *kcb; + + addr = (kprobe_opcode_t *)(regs->ip - sizeof(kprobe_opcode_t)); +- if (*addr != BREAKPOINT_INSTRUCTION) { ++ if (*(kprobe_opcode_t *)ktla_ktva((unsigned long)addr) != BREAKPOINT_INSTRUCTION) { + /* + * The breakpoint instruction was removed right + * after we hit it. Another cpu has removed +@@ -775,7 +803,7 @@ static void __kprobes resume_execution(s + struct pt_regs *regs, struct kprobe_ctlblk *kcb) + { + unsigned long *tos = stack_addr(regs); +- unsigned long copy_ip = (unsigned long)p->ainsn.insn; ++ unsigned long copy_ip = ktva_ktla((unsigned long)p->ainsn.insn); + unsigned long orig_ip = (unsigned long)p->addr; + kprobe_opcode_t *insn = p->ainsn.insn; + +@@ -958,7 +986,7 @@ int __kprobes kprobe_exceptions_notify(s + struct die_args *args = data; + int ret = NOTIFY_DONE; + +- if (args->regs && user_mode_vm(args->regs)) ++ if (args->regs && user_mode(args->regs)) + return ret; + + switch (val) { +diff -urNp linux-2.6.31.1/arch/x86/kernel/ldt.c linux-2.6.31.1/arch/x86/kernel/ldt.c +--- linux-2.6.31.1/arch/x86/kernel/ldt.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/ldt.c 2009-10-01 20:12:42.000000000 -0400 +@@ -66,13 +66,13 @@ static int alloc_ldt(mm_context_t *pc, i + if (reload) { + #ifdef CONFIG_SMP + preempt_disable(); +- load_LDT(pc); ++ load_LDT_nolock(pc); + if (!cpus_equal(current->mm->cpu_vm_mask, + cpumask_of_cpu(smp_processor_id()))) + smp_call_function(flush_ldt, current->mm, 1); + preempt_enable(); + #else +- load_LDT(pc); ++ load_LDT_nolock(pc); + #endif + } + if (oldsize) { +@@ -94,7 +94,7 @@ static inline int copy_ldt(mm_context_t + return err; + + for (i = 0; i < old->size; i++) +- write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE); ++ write_ldt_entry(new->ldt, i, old->ldt + i); + return 0; + } + +@@ -115,6 +115,24 @@ int init_new_context(struct task_struct + retval = copy_ldt(&mm->context, &old_mm->context); + mutex_unlock(&old_mm->context.lock); + } ++ ++ if (tsk == current) { ++ mm->context.vdso = ~0UL; ++ ++#ifdef CONFIG_X86_32 ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ mm->context.user_cs_base = 0UL; ++ mm->context.user_cs_limit = ~0UL; ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP) ++ cpus_clear(mm->context.cpu_user_cs_mask); ++#endif ++ ++#endif ++#endif ++ ++ } ++ + return retval; + } + +@@ -229,6 +247,13 @@ static int write_ldt(void __user *ptr, u + } + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && (ldt_info.contents & MODIFY_LDT_CONTENTS_CODE)) { ++ error = -EINVAL; ++ goto out_unlock; ++ } ++#endif ++ + fill_ldt(&ldt, &ldt_info); + if (oldmode) + ldt.avl = 0; +diff -urNp linux-2.6.31.1/arch/x86/kernel/machine_kexec_32.c linux-2.6.31.1/arch/x86/kernel/machine_kexec_32.c +--- linux-2.6.31.1/arch/x86/kernel/machine_kexec_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/machine_kexec_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -26,7 +26,7 @@ + #include + #include + +-static void set_idt(void *newidt, __u16 limit) ++static void set_idt(struct desc_struct *newidt, __u16 limit) + { + struct desc_ptr curidt; + +@@ -38,7 +38,7 @@ static void set_idt(void *newidt, __u16 + } + + +-static void set_gdt(void *newgdt, __u16 limit) ++static void set_gdt(struct desc_struct *newgdt, __u16 limit) + { + struct desc_ptr curgdt; + +@@ -217,7 +217,7 @@ void machine_kexec(struct kimage *image) + } + + control_page = page_address(image->control_code_page); +- memcpy(control_page, relocate_kernel, KEXEC_CONTROL_CODE_MAX_SIZE); ++ memcpy(control_page, (void *)ktla_ktva((unsigned long)relocate_kernel), KEXEC_CONTROL_CODE_MAX_SIZE); + + relocate_kernel_ptr = control_page; + page_list[PA_CONTROL_PAGE] = __pa(control_page); +diff -urNp linux-2.6.31.1/arch/x86/kernel/module.c linux-2.6.31.1/arch/x86/kernel/module.c +--- linux-2.6.31.1/arch/x86/kernel/module.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/module.c 2009-10-01 20:12:42.000000000 -0400 +@@ -27,6 +27,7 @@ + #include + #include + #include ++#include + + #if 0 + #define DEBUGP printk +@@ -34,7 +35,7 @@ + #define DEBUGP(fmt...) + #endif + +-void *module_alloc(unsigned long size) ++static void *__module_alloc(unsigned long size, pgprot_t prot) + { + struct vm_struct *area; + +@@ -48,9 +49,92 @@ void *module_alloc(unsigned long size) + if (!area) + return NULL; + +- return __vmalloc_area(area, GFP_KERNEL | __GFP_HIGHMEM, +- PAGE_KERNEL_EXEC); ++ return __vmalloc_area(area, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, prot); ++} ++ ++#ifdef CONFIG_PAX_KERNEXEC ++#ifdef CONFIG_X86_32 ++void *module_alloc(unsigned long size) ++{ ++ return __module_alloc(size, PAGE_KERNEL); ++} ++ ++void *module_alloc_exec(unsigned long size) ++{ ++ struct vm_struct *area; ++ ++ if (size == 0) ++ return NULL; ++ ++ area = __get_vm_area(size, VM_ALLOC, (unsigned long)&MODULES_EXEC_VADDR, (unsigned long)&MODULES_EXEC_END); ++ if (area) ++ return area->addr; ++ ++ return NULL; ++} ++EXPORT_SYMBOL(module_alloc_exec); ++ ++void module_free_exec(struct module *mod, void *module_region) ++{ ++ struct vm_struct **p, *tmp; ++ ++ if (!module_region) ++ return; ++ ++ if ((PAGE_SIZE-1) & (unsigned long)module_region) { ++ printk(KERN_ERR "Trying to module_free_exec() bad address (%p)\n", module_region); ++ WARN_ON(1); ++ return; ++ } ++ ++ write_lock(&vmlist_lock); ++ for (p = &vmlist; (tmp = *p) != NULL; p = &tmp->next) ++ if (tmp->addr == module_region) ++ break; ++ ++ if (tmp) { ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++ memset(tmp->addr, 0xCC, tmp->size); ++ pax_close_kernel(cr0); ++ ++ *p = tmp->next; ++ kfree(tmp); ++ } ++ write_unlock(&vmlist_lock); ++ ++ if (!tmp) { ++ printk(KERN_ERR "Trying to module_free_exec() nonexistent vm area (%p)\n", ++ module_region); ++ WARN_ON(1); ++ } ++} ++EXPORT_SYMBOL(module_free_exec); ++#else ++void *module_alloc(unsigned long size) ++{ ++ return __module_alloc(size, PAGE_KERNEL); ++} ++ ++void module_free_exec(struct module *mod, void *module_region) ++{ ++ module_free(mod, module_region); ++} ++EXPORT_SYMBOL(module_free_exec); ++ ++void *module_alloc_exec(unsigned long size) ++{ ++ return __module_alloc(size, PAGE_KERNEL_RX); ++} ++EXPORT_SYMBOL(module_alloc_exec); ++#endif ++#else ++void *module_alloc(unsigned long size) ++{ ++ return __module_alloc(size, PAGE_KERNEL_EXEC); + } ++#endif + + /* Free memory returned from module_alloc */ + void module_free(struct module *mod, void *module_region) +@@ -77,14 +161,20 @@ int apply_relocate(Elf32_Shdr *sechdrs, + unsigned int i; + Elf32_Rel *rel = (void *)sechdrs[relsec].sh_addr; + Elf32_Sym *sym; +- uint32_t *location; ++ uint32_t *plocation, location; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif + + DEBUGP("Applying relocate section %u to %u\n", relsec, + sechdrs[relsec].sh_info); + for (i = 0; i < sechdrs[relsec].sh_size / sizeof(*rel); i++) { + /* This is where to make the change */ +- location = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr +- + rel[i].r_offset; ++ plocation = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr + rel[i].r_offset; ++ location = (uint32_t)plocation; ++ if (sechdrs[sechdrs[relsec].sh_info].sh_flags & SHF_EXECINSTR) ++ plocation = ktla_ktva((void *)plocation); + /* This is the symbol it is referring to. Note that all + undefined symbols have been resolved. */ + sym = (Elf32_Sym *)sechdrs[symindex].sh_addr +@@ -93,11 +183,31 @@ int apply_relocate(Elf32_Shdr *sechdrs, + switch (ELF32_R_TYPE(rel[i].r_info)) { + case R_386_32: + /* We add the value into the location given */ +- *location += sym->st_value; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ ++ *plocation += sym->st_value; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + break; + case R_386_PC32: + /* Add the value, subtract its postition */ +- *location += sym->st_value - (uint32_t)location; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ ++ *plocation += sym->st_value - location; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + break; + default: + printk(KERN_ERR "module %s: Unknown relocation: %u\n", +@@ -131,6 +241,10 @@ int apply_relocate_add(Elf64_Shdr *sechd + void *loc; + u64 val; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + DEBUGP("Applying relocate section %u to %u\n", relsec, + sechdrs[relsec].sh_info); + for (i = 0; i < sechdrs[relsec].sh_size / sizeof(*rel); i++) { +@@ -153,21 +267,61 @@ int apply_relocate_add(Elf64_Shdr *sechd + case R_X86_64_NONE: + break; + case R_X86_64_64: ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + *(u64 *)loc = val; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + break; + case R_X86_64_32: ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + *(u32 *)loc = val; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + if (val != *(u32 *)loc) + goto overflow; + break; + case R_X86_64_32S: ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + *(s32 *)loc = val; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + if ((s64)val != *(s32 *)loc) + goto overflow; + break; + case R_X86_64_PC32: + val -= (u64)loc; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + *(u32 *)loc = val; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + #if 0 + if ((s64)val != *(s32 *)loc) + goto overflow; +diff -urNp linux-2.6.31.1/arch/x86/kernel/paravirt.c linux-2.6.31.1/arch/x86/kernel/paravirt.c +--- linux-2.6.31.1/arch/x86/kernel/paravirt.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/paravirt.c 2009-10-01 20:12:42.000000000 -0400 +@@ -54,7 +54,7 @@ u64 _paravirt_ident_64(u64 x) + return x; + } + +-static void __init default_banner(void) ++static void default_banner(void) + { + printk(KERN_INFO "Booting paravirtualized kernel on %s\n", + pv_info.name); +@@ -125,9 +125,9 @@ unsigned paravirt_patch_jmp(void *insnbu + + /* Neat trick to map patch type back to the call within the + * corresponding structure. */ +-static void *get_call_destination(u8 type) ++static const void *get_call_destination(u8 type) + { +- struct paravirt_patch_template tmpl = { ++ const struct paravirt_patch_template tmpl = { + .pv_init_ops = pv_init_ops, + .pv_time_ops = pv_time_ops, + .pv_cpu_ops = pv_cpu_ops, +@@ -138,13 +138,13 @@ static void *get_call_destination(u8 typ + .pv_lock_ops = pv_lock_ops, + #endif + }; +- return *((void **)&tmpl + type); ++ return *((const void **)&tmpl + type); + } + + unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf, + unsigned long addr, unsigned len) + { +- void *opfunc = get_call_destination(type); ++ const void *opfunc = get_call_destination(type); + unsigned ret; + + if (opfunc == NULL) +@@ -183,7 +183,7 @@ unsigned paravirt_patch_insns(void *insn + if (insn_len > len || start == NULL) + insn_len = len; + else +- memcpy(insnbuf, start, insn_len); ++ memcpy(insnbuf, ktla_ktva(start), insn_len); + + return insn_len; + } +@@ -311,21 +311,21 @@ void arch_flush_lazy_mmu_mode(void) + preempt_enable(); + } + +-struct pv_info pv_info = { ++struct pv_info pv_info __read_only = { + .name = "bare hardware", + .paravirt_enabled = 0, + .kernel_rpl = 0, + .shared_kernel_pmd = 1, /* Only used when CONFIG_X86_PAE is set */ + }; + +-struct pv_init_ops pv_init_ops = { ++struct pv_init_ops pv_init_ops __read_only = { + .patch = native_patch, + .banner = default_banner, + .arch_setup = paravirt_nop, + .memory_setup = machine_specific_memory_setup, + }; + +-struct pv_time_ops pv_time_ops = { ++struct pv_time_ops pv_time_ops __read_only = { + .time_init = hpet_time_init, + .get_wallclock = native_get_wallclock, + .set_wallclock = native_set_wallclock, +@@ -333,7 +333,7 @@ struct pv_time_ops pv_time_ops = { + .get_tsc_khz = native_calibrate_tsc, + }; + +-struct pv_irq_ops pv_irq_ops = { ++struct pv_irq_ops pv_irq_ops __read_only = { + .init_IRQ = native_init_IRQ, + .save_fl = __PV_IS_CALLEE_SAVE(native_save_fl), + .restore_fl = __PV_IS_CALLEE_SAVE(native_restore_fl), +@@ -346,7 +346,7 @@ struct pv_irq_ops pv_irq_ops = { + #endif + }; + +-struct pv_cpu_ops pv_cpu_ops = { ++struct pv_cpu_ops pv_cpu_ops __read_only = { + .cpuid = native_cpuid, + .get_debugreg = native_get_debugreg, + .set_debugreg = native_set_debugreg, +@@ -406,7 +406,7 @@ struct pv_cpu_ops pv_cpu_ops = { + .end_context_switch = paravirt_nop, + }; + +-struct pv_apic_ops pv_apic_ops = { ++struct pv_apic_ops pv_apic_ops __read_only = { + #ifdef CONFIG_X86_LOCAL_APIC + .setup_boot_clock = setup_boot_APIC_clock, + .setup_secondary_clock = setup_secondary_APIC_clock, +@@ -422,7 +422,7 @@ struct pv_apic_ops pv_apic_ops = { + #define PTE_IDENT __PV_IS_CALLEE_SAVE(_paravirt_ident_64) + #endif + +-struct pv_mmu_ops pv_mmu_ops = { ++struct pv_mmu_ops pv_mmu_ops __read_only = { + #ifndef CONFIG_X86_64 + .pagetable_setup_start = native_pagetable_setup_start, + .pagetable_setup_done = native_pagetable_setup_done, +diff -urNp linux-2.6.31.1/arch/x86/kernel/paravirt-spinlocks.c linux-2.6.31.1/arch/x86/kernel/paravirt-spinlocks.c +--- linux-2.6.31.1/arch/x86/kernel/paravirt-spinlocks.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/paravirt-spinlocks.c 2009-10-01 20:12:42.000000000 -0400 +@@ -13,7 +13,7 @@ default_spin_lock_flags(raw_spinlock_t * + __raw_spin_lock(lock); + } + +-struct pv_lock_ops pv_lock_ops = { ++struct pv_lock_ops pv_lock_ops __read_only = { + #ifdef CONFIG_SMP + .spin_is_locked = __ticket_spin_is_locked, + .spin_is_contended = __ticket_spin_is_contended, +diff -urNp linux-2.6.31.1/arch/x86/kernel/process_32.c linux-2.6.31.1/arch/x86/kernel/process_32.c +--- linux-2.6.31.1/arch/x86/kernel/process_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/process_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -70,6 +70,7 @@ EXPORT_PER_CPU_SYMBOL(current_task); + unsigned long thread_saved_pc(struct task_struct *tsk) + { + return ((unsigned long *)tsk->thread.sp)[3]; ++//XXX return tsk->thread.eip; + } + + #ifndef CONFIG_SMP +@@ -132,7 +133,7 @@ void __show_regs(struct pt_regs *regs, i + unsigned short ss, gs; + const char *board; + +- if (user_mode_vm(regs)) { ++ if (user_mode(regs)) { + sp = regs->sp; + ss = regs->ss & 0xffff; + gs = get_user_gs(regs); +@@ -213,8 +214,8 @@ int kernel_thread(int (*fn)(void *), voi + regs.bx = (unsigned long) fn; + regs.dx = (unsigned long) arg; + +- regs.ds = __USER_DS; +- regs.es = __USER_DS; ++ regs.ds = __KERNEL_DS; ++ regs.es = __KERNEL_DS; + regs.fs = __KERNEL_PERCPU; + regs.gs = __KERNEL_STACK_CANARY; + regs.orig_ax = -1; +@@ -250,7 +251,7 @@ int copy_thread(unsigned long clone_flag + struct task_struct *tsk; + int err; + +- childregs = task_pt_regs(p); ++ childregs = task_stack_page(p) + THREAD_SIZE - sizeof(struct pt_regs) - 8; + *childregs = *regs; + childregs->ax = 0; + childregs->sp = sp; +@@ -279,6 +280,7 @@ int copy_thread(unsigned long clone_flag + * Set a new TLS for the child thread? + */ + if (clone_flags & CLONE_SETTLS) ++//XXX needs set_fs()? + err = do_set_thread_area(p, -1, + (struct user_desc __user *)childregs->si, 0); + +@@ -349,7 +351,7 @@ __switch_to(struct task_struct *prev_p, + struct thread_struct *prev = &prev_p->thread, + *next = &next_p->thread; + int cpu = smp_processor_id(); +- struct tss_struct *tss = &per_cpu(init_tss, cpu); ++ struct tss_struct *tss = init_tss + cpu; + + /* never put a printk in __switch_to... printk() calls wake_up*() indirectly */ + +@@ -377,6 +379,11 @@ __switch_to(struct task_struct *prev_p, + */ + lazy_save_gs(prev->gs); + ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ if (!segment_eq(task_thread_info(prev_p)->addr_limit, task_thread_info(next_p)->addr_limit)) ++ __set_fs(task_thread_info(next_p)->addr_limit, cpu); ++#endif ++ + /* + * Load the per-thread Thread-Local Storage descriptor. + */ +@@ -495,3 +502,27 @@ unsigned long get_wchan(struct task_stru + return 0; + } + ++#ifdef CONFIG_PAX_RANDKSTACK ++asmlinkage void pax_randomize_kstack(void) ++{ ++ struct thread_struct *thread = ¤t->thread; ++ unsigned long time; ++ ++ if (!randomize_va_space) ++ return; ++ ++ rdtscl(time); ++ ++ /* P4 seems to return a 0 LSB, ignore it */ ++#ifdef CONFIG_MPENTIUM4 ++ time &= 0x1EUL; ++ time <<= 2; ++#else ++ time &= 0xFUL; ++ time <<= 3; ++#endif ++ ++ thread->sp0 ^= time; ++ load_sp0(init_tss + smp_processor_id(), thread); ++} ++#endif +diff -urNp linux-2.6.31.1/arch/x86/kernel/process_64.c linux-2.6.31.1/arch/x86/kernel/process_64.c +--- linux-2.6.31.1/arch/x86/kernel/process_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/process_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -94,7 +94,7 @@ static void __exit_idle(void) + void exit_idle(void) + { + /* idle loop has pid 0 */ +- if (current->pid) ++ if (task_pid_nr(current)) + return; + __exit_idle(); + } +@@ -173,7 +173,7 @@ void __show_regs(struct pt_regs *regs, i + if (!board) + board = ""; + printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s %s\n", +- current->pid, current->comm, print_tainted(), ++ task_pid_nr(current), current->comm, print_tainted(), + init_utsname()->release, + (int)strcspn(init_utsname()->version, " "), + init_utsname()->version, board); +@@ -384,7 +384,7 @@ __switch_to(struct task_struct *prev_p, + struct thread_struct *prev = &prev_p->thread; + struct thread_struct *next = &next_p->thread; + int cpu = smp_processor_id(); +- struct tss_struct *tss = &per_cpu(init_tss, cpu); ++ struct tss_struct *tss = init_tss + cpu; + unsigned fsindex, gsindex; + + /* we're going to use this soon, after a few expensive things */ +@@ -543,12 +543,11 @@ unsigned long get_wchan(struct task_stru + if (!p || p == current || p->state == TASK_RUNNING) + return 0; + stack = (unsigned long)task_stack_page(p); +- if (p->thread.sp < stack || p->thread.sp >= stack+THREAD_SIZE) ++ if (p->thread.sp < stack || p->thread.sp > stack+THREAD_SIZE-8-sizeof(u64)) + return 0; + fp = *(u64 *)(p->thread.sp); + do { +- if (fp < (unsigned long)stack || +- fp >= (unsigned long)stack+THREAD_SIZE) ++ if (fp < stack || fp > stack+THREAD_SIZE-8-sizeof(u64)) + return 0; + ip = *(u64 *)(fp+8); + if (!in_sched_functions(ip)) +diff -urNp linux-2.6.31.1/arch/x86/kernel/process.c linux-2.6.31.1/arch/x86/kernel/process.c +--- linux-2.6.31.1/arch/x86/kernel/process.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/process.c 2009-10-01 20:12:42.000000000 -0400 +@@ -76,7 +76,7 @@ void exit_thread(void) + unsigned long *bp = t->io_bitmap_ptr; + + if (bp) { +- struct tss_struct *tss = &per_cpu(init_tss, get_cpu()); ++ struct tss_struct *tss = init_tss + get_cpu(); + + t->io_bitmap_ptr = NULL; + clear_thread_flag(TIF_IO_BITMAP); +@@ -108,6 +108,9 @@ void flush_thread(void) + + clear_tsk_thread_flag(tsk, TIF_DEBUG); + ++#if defined(CONFIG_X86_32) && !defined(CONFIG_CC_STACKPROTECTOR) ++ loadsegment(gs, 0); ++#endif + tsk->thread.debugreg0 = 0; + tsk->thread.debugreg1 = 0; + tsk->thread.debugreg2 = 0; +@@ -611,17 +614,3 @@ static int __init idle_setup(char *str) + return 0; + } + early_param("idle", idle_setup); +- +-unsigned long arch_align_stack(unsigned long sp) +-{ +- if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) +- sp -= get_random_int() % 8192; +- return sp & ~0xf; +-} +- +-unsigned long arch_randomize_brk(struct mm_struct *mm) +-{ +- unsigned long range_end = mm->brk + 0x02000000; +- return randomize_range(mm->brk, range_end, 0) ? : mm->brk; +-} +- +diff -urNp linux-2.6.31.1/arch/x86/kernel/ptrace.c linux-2.6.31.1/arch/x86/kernel/ptrace.c +--- linux-2.6.31.1/arch/x86/kernel/ptrace.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/ptrace.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1454,7 +1454,7 @@ void send_sigtrap(struct task_struct *ts + info.si_code = si_code; + + /* User-mode ip? */ +- info.si_addr = user_mode_vm(regs) ? (void __user *) regs->ip : NULL; ++ info.si_addr = user_mode(regs) ? (void __user *) regs->ip : NULL; + + /* Send us the fake SIGTRAP */ + force_sig_info(SIGTRAP, &info, tsk); +diff -urNp linux-2.6.31.1/arch/x86/kernel/reboot.c linux-2.6.31.1/arch/x86/kernel/reboot.c +--- linux-2.6.31.1/arch/x86/kernel/reboot.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/reboot.c 2009-10-01 20:12:42.000000000 -0400 +@@ -31,7 +31,7 @@ void (*pm_power_off)(void); + EXPORT_SYMBOL(pm_power_off); + + static const struct desc_ptr no_idt = {}; +-static int reboot_mode; ++static unsigned short reboot_mode; + enum reboot_type reboot_type = BOOT_KBD; + int reboot_force; + +@@ -257,7 +257,7 @@ static struct dmi_system_id __initdata r + DMI_MATCH(DMI_PRODUCT_NAME, "SBC-FITPC2"), + }, + }, +- { } ++ { NULL, NULL, {{0, {0}}}, NULL} + }; + + static int __init reboot_init(void) +@@ -273,12 +273,12 @@ core_initcall(reboot_init); + controller to pulse the CPU reset line, which is more thorough, but + doesn't work with at least one type of 486 motherboard. It is easy + to stop this code working; hence the copious comments. */ +-static const unsigned long long +-real_mode_gdt_entries [3] = ++static struct desc_struct ++real_mode_gdt_entries [3] __read_only = + { +- 0x0000000000000000ULL, /* Null descriptor */ +- 0x00009b000000ffffULL, /* 16-bit real-mode 64k code at 0x00000000 */ +- 0x000093000100ffffULL /* 16-bit real-mode 64k data at 0x00000100 */ ++ {{{0x00000000, 0x00000000}}}, /* Null descriptor */ ++ {{{0x0000ffff, 0x00009b00}}}, /* 16-bit real-mode 64k code at 0x00000000 */ ++ {{{0x0100ffff, 0x00009300}}} /* 16-bit real-mode 64k data at 0x00000100 */ + }; + + static const struct desc_ptr +@@ -327,7 +327,7 @@ static const unsigned char jump_to_bios + * specified by the code and length parameters. + * We assume that length will aways be less that 100! + */ +-void machine_real_restart(const unsigned char *code, int length) ++void machine_real_restart(const unsigned char *code, unsigned int length) + { + local_irq_disable(); + +@@ -347,8 +347,8 @@ void machine_real_restart(const unsigned + /* Remap the kernel at virtual address zero, as well as offset zero + from the kernel segment. This assumes the kernel segment starts at + virtual address PAGE_OFFSET. */ +- memcpy(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, +- sizeof(swapper_pg_dir [0]) * KERNEL_PGD_PTRS); ++ clone_pgd_range(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, ++ min_t(unsigned long, KERNEL_PGD_PTRS, KERNEL_PGD_BOUNDARY)); + + /* + * Use `swapper_pg_dir' as our page directory. +@@ -360,16 +360,15 @@ void machine_real_restart(const unsigned + boot)". This seems like a fairly standard thing that gets set by + REBOOT.COM programs, and the previous reset routine did this + too. */ +- *((unsigned short *)0x472) = reboot_mode; ++ *(unsigned short *)(__va(0x472)) = reboot_mode; + + /* For the switch to real mode, copy some code to low memory. It has + to be in the first 64k because it is running in 16-bit mode, and it + has to have the same physical and virtual address, because it turns + off paging. Copy it near the end of the first page, out of the way + of BIOS variables. */ +- memcpy((void *)(0x1000 - sizeof(real_mode_switch) - 100), +- real_mode_switch, sizeof (real_mode_switch)); +- memcpy((void *)(0x1000 - 100), code, length); ++ memcpy(__va(0x1000 - sizeof (real_mode_switch) - 100), real_mode_switch, sizeof (real_mode_switch)); ++ memcpy(__va(0x1000 - 100), code, length); + + /* Set up the IDT for real mode. */ + load_idt(&real_mode_idt); +diff -urNp linux-2.6.31.1/arch/x86/kernel/setup.c linux-2.6.31.1/arch/x86/kernel/setup.c +--- linux-2.6.31.1/arch/x86/kernel/setup.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/setup.c 2009-10-01 20:12:42.000000000 -0400 +@@ -768,14 +768,14 @@ void __init setup_arch(char **cmdline_p) + + if (!boot_params.hdr.root_flags) + root_mountflags &= ~MS_RDONLY; +- init_mm.start_code = (unsigned long) _text; +- init_mm.end_code = (unsigned long) _etext; ++ init_mm.start_code = ktla_ktva((unsigned long) _text); ++ init_mm.end_code = ktla_ktva((unsigned long) _etext); + init_mm.end_data = (unsigned long) _edata; + init_mm.brk = _brk_end; + +- code_resource.start = virt_to_phys(_text); +- code_resource.end = virt_to_phys(_etext)-1; +- data_resource.start = virt_to_phys(_etext); ++ code_resource.start = virt_to_phys(ktla_ktva(_text)); ++ code_resource.end = virt_to_phys(ktla_ktva(_etext))-1; ++ data_resource.start = virt_to_phys(_sdata); + data_resource.end = virt_to_phys(_edata)-1; + bss_resource.start = virt_to_phys(&__bss_start); + bss_resource.end = virt_to_phys(&__bss_stop)-1; +diff -urNp linux-2.6.31.1/arch/x86/kernel/setup_percpu.c linux-2.6.31.1/arch/x86/kernel/setup_percpu.c +--- linux-2.6.31.1/arch/x86/kernel/setup_percpu.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/setup_percpu.c 2009-10-01 20:12:42.000000000 -0400 +@@ -25,19 +25,17 @@ + # define DBG(x...) + #endif + ++#ifdef CONFIG_SMP + DEFINE_PER_CPU(int, cpu_number); + EXPORT_PER_CPU_SYMBOL(cpu_number); ++#endif + +-#ifdef CONFIG_X86_64 + #define BOOT_PERCPU_OFFSET ((unsigned long)__per_cpu_load) +-#else +-#define BOOT_PERCPU_OFFSET 0 +-#endif + + DEFINE_PER_CPU(unsigned long, this_cpu_off) = BOOT_PERCPU_OFFSET; + EXPORT_PER_CPU_SYMBOL(this_cpu_off); + +-unsigned long __per_cpu_offset[NR_CPUS] __read_mostly = { ++unsigned long __per_cpu_offset[NR_CPUS] __read_only = { + [0 ... NR_CPUS-1] = BOOT_PERCPU_OFFSET, + }; + EXPORT_SYMBOL(__per_cpu_offset); +@@ -429,13 +427,15 @@ early_param("percpu_alloc", percpu_alloc + static inline void setup_percpu_segment(int cpu) + { + #ifdef CONFIG_X86_32 +- struct desc_struct gdt; +- +- pack_descriptor(&gdt, per_cpu_offset(cpu), 0xFFFFF, +- 0x2 | DESCTYPE_S, 0x8); +- gdt.s = 1; +- write_gdt_entry(get_cpu_gdt_table(cpu), +- GDT_ENTRY_PERCPU, &gdt, DESCTYPE_S); ++ struct desc_struct d, *gdt = get_cpu_gdt_table(cpu); ++ unsigned long base = per_cpu_offset(cpu); ++ const unsigned long limit = VMALLOC_END - base - 1; ++ ++ if (limit < 64*1024) ++ pack_descriptor(&d, base, limit, 0x80 | DESCTYPE_S | 0x3, 0x4); ++ else ++ pack_descriptor(&d, base, limit >> PAGE_SHIFT, 0x80 | DESCTYPE_S | 0x3, 0xC); ++ write_gdt_entry(gdt, GDT_ENTRY_PERCPU, &d, DESCTYPE_S); + #endif + } + +@@ -486,6 +486,11 @@ void __init setup_per_cpu_areas(void) + /* alrighty, percpu areas up and running */ + delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start; + for_each_possible_cpu(cpu) { ++#ifdef CONFIG_CC_STACKPROTECTOR ++#ifdef CONFIG_x86_32 ++ unsigned long canary = per_cpu(stack_canary, cpu); ++#endif ++#endif + per_cpu_offset(cpu) = delta + cpu * pcpu_unit_size; + per_cpu(this_cpu_off, cpu) = per_cpu_offset(cpu); + per_cpu(cpu_number, cpu) = cpu; +@@ -513,6 +518,12 @@ void __init setup_per_cpu_areas(void) + early_per_cpu_map(x86_cpu_to_node_map, cpu); + #endif + #endif ++#ifdef CONFIG_CC_STACKPROTECTOR ++#ifdef CONFIG_x86_32 ++ if (cpu == boot_cpu_id) ++ per_cpu(stack_canary, cpu) = canary; ++#endif ++#endif + /* + * Up to this point, the boot CPU has been using .data.init + * area. Reload any changed state for the boot CPU. +diff -urNp linux-2.6.31.1/arch/x86/kernel/signal.c linux-2.6.31.1/arch/x86/kernel/signal.c +--- linux-2.6.31.1/arch/x86/kernel/signal.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/signal.c 2009-10-01 20:12:42.000000000 -0400 +@@ -197,7 +197,7 @@ static unsigned long align_sigframe(unsi + * Align the stack pointer according to the i386 ABI, + * i.e. so that on function entry ((sp + 4) & 15) == 0. + */ +- sp = ((sp + 4) & -16ul) - 4; ++ sp = ((sp - 12) & -16ul) - 4; + #else /* !CONFIG_X86_32 */ + sp = round_down(sp, 16) - 8; + #endif +@@ -307,9 +307,9 @@ __setup_frame(int sig, struct k_sigactio + } + + if (current->mm->context.vdso) +- restorer = VDSO32_SYMBOL(current->mm->context.vdso, sigreturn); ++ restorer = (void __user *)VDSO32_SYMBOL(current->mm->context.vdso, sigreturn); + else +- restorer = &frame->retcode; ++ restorer = (void __user *)&frame->retcode; + if (ka->sa.sa_flags & SA_RESTORER) + restorer = ka->sa.sa_restorer; + +@@ -377,7 +377,7 @@ static int __setup_rt_frame(int sig, str + err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); + + /* Set up to return from userspace. */ +- restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); ++ restorer = (void __user *)VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); + if (ka->sa.sa_flags & SA_RESTORER) + restorer = ka->sa.sa_restorer; + put_user_ex(restorer, &frame->pretcode); +@@ -789,7 +789,7 @@ static void do_signal(struct pt_regs *re + * X86_32: vm86 regs switched out by assembly code before reaching + * here, so testing against kernel CS suffices. + */ +- if (!user_mode(regs)) ++ if (!user_mode_novm(regs)) + return; + + if (current_thread_info()->status & TS_RESTORE_SIGMASK) +diff -urNp linux-2.6.31.1/arch/x86/kernel/smpboot.c linux-2.6.31.1/arch/x86/kernel/smpboot.c +--- linux-2.6.31.1/arch/x86/kernel/smpboot.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/smpboot.c 2009-10-01 20:12:42.000000000 -0400 +@@ -685,6 +685,10 @@ static int __cpuinit do_boot_cpu(int api + .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done), + }; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + INIT_WORK(&c_idle.work, do_fork_idle); + + alternatives_smp_switch(1); +@@ -727,7 +731,17 @@ do_rest: + (unsigned long)task_stack_page(c_idle.idle) - + KERNEL_STACK_OFFSET + THREAD_SIZE; + #endif ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + initial_code = (unsigned long)start_secondary; + stack_start.sp = (void *) c_idle.idle->thread.sp; + +diff -urNp linux-2.6.31.1/arch/x86/kernel/step.c linux-2.6.31.1/arch/x86/kernel/step.c +--- linux-2.6.31.1/arch/x86/kernel/step.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/step.c 2009-10-01 20:12:42.000000000 -0400 +@@ -23,22 +23,20 @@ unsigned long convert_ip_to_linear(struc + * and APM bios ones we just ignore here. + */ + if ((seg & SEGMENT_TI_MASK) == SEGMENT_LDT) { +- u32 *desc; ++ struct desc_struct *desc; + unsigned long base; + +- seg &= ~7UL; ++ seg >>= 3; + + mutex_lock(&child->mm->context.lock); +- if (unlikely((seg >> 3) >= child->mm->context.size)) +- addr = -1L; /* bogus selector, access would fault */ ++ if (unlikely(seg >= child->mm->context.size)) ++ addr = -EINVAL; + else { +- desc = child->mm->context.ldt + seg; +- base = ((desc[0] >> 16) | +- ((desc[1] & 0xff) << 16) | +- (desc[1] & 0xff000000)); ++ desc = &child->mm->context.ldt[seg]; ++ base = (desc->a >> 16) | ((desc->b & 0xff) << 16) | (desc->b & 0xff000000); + + /* 16-bit code segment? */ +- if (!((desc[1] >> 22) & 1)) ++ if (!((desc->b >> 22) & 1)) + addr &= 0xffff; + addr += base; + } +@@ -54,6 +52,9 @@ static int is_setting_trap_flag(struct t + unsigned char opcode[15]; + unsigned long addr = convert_ip_to_linear(child, regs); + ++ if (addr == -EINVAL) ++ return 0; ++ + copied = access_process_vm(child, addr, opcode, sizeof(opcode), 0); + for (i = 0; i < copied; i++) { + switch (opcode[i]) { +@@ -75,7 +76,7 @@ static int is_setting_trap_flag(struct t + + #ifdef CONFIG_X86_64 + case 0x40 ... 0x4f: +- if (regs->cs != __USER_CS) ++ if ((regs->cs & 0xffff) != __USER_CS) + /* 32-bit mode: register increment */ + return 0; + /* 64-bit mode: REX prefix */ +diff -urNp linux-2.6.31.1/arch/x86/kernel/syscall_table_32.S linux-2.6.31.1/arch/x86/kernel/syscall_table_32.S +--- linux-2.6.31.1/arch/x86/kernel/syscall_table_32.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/syscall_table_32.S 2009-10-01 20:12:42.000000000 -0400 +@@ -1,3 +1,4 @@ ++.section .rodata,"a",@progbits + ENTRY(sys_call_table) + .long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */ + .long sys_exit +diff -urNp linux-2.6.31.1/arch/x86/kernel/sys_i386_32.c linux-2.6.31.1/arch/x86/kernel/sys_i386_32.c +--- linux-2.6.31.1/arch/x86/kernel/sys_i386_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/sys_i386_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -24,6 +24,21 @@ + + #include + ++int i386_mmap_check(unsigned long addr, unsigned long len, unsigned long flags) ++{ ++ unsigned long pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (current->mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ if (len > pax_task_size || addr > pax_task_size - len) ++ return -EINVAL; ++ ++ return 0; ++} ++ + asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long pgoff) +@@ -83,6 +98,205 @@ out: + return err; + } + ++unsigned long ++arch_get_unmapped_area(struct file *filp, unsigned long addr, ++ unsigned long len, unsigned long pgoff, unsigned long flags) ++{ ++ struct mm_struct *mm = current->mm; ++ struct vm_area_struct *vma; ++ unsigned long start_addr, pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ if (len > pax_task_size) ++ return -ENOMEM; ++ ++ if (flags & MAP_FIXED) ++ return addr; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ ++ if (addr) { ++ addr = PAGE_ALIGN(addr); ++ vma = find_vma(mm, addr); ++ if (pax_task_size - len >= addr && ++ (!vma || addr + len <= vma->vm_start)) ++ return addr; ++ } ++ if (len > mm->cached_hole_size) { ++ start_addr = addr = mm->free_area_cache; ++ } else { ++ start_addr = addr = mm->mmap_base; ++ mm->cached_hole_size = 0; ++ } ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!nx_enabled && (mm->pax_flags & MF_PAX_PAGEEXEC) && (flags & MAP_EXECUTABLE) && start_addr >= mm->mmap_base) { ++ start_addr = 0x00110000UL; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ start_addr += mm->delta_mmap & 0x03FFF000UL; ++#endif ++ ++ if (mm->start_brk <= start_addr && start_addr < mm->mmap_base) ++ start_addr = addr = mm->mmap_base; ++ else ++ addr = start_addr; ++ } ++#endif ++ ++full_search: ++ for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { ++ /* At this point: (!vma || addr < vma->vm_end). */ ++ if (pax_task_size - len < addr) { ++ /* ++ * Start a new search - just in case we missed ++ * some holes. ++ */ ++ if (start_addr != mm->mmap_base) { ++ start_addr = addr = mm->mmap_base; ++ mm->cached_hole_size = 0; ++ goto full_search; ++ } ++ return -ENOMEM; ++ } ++ if (!vma || addr + len <= vma->vm_start) { ++ /* ++ * Remember the place where we stopped the search: ++ */ ++ mm->free_area_cache = addr + len; ++ return addr; ++ } ++ if (addr + mm->cached_hole_size < vma->vm_start) ++ mm->cached_hole_size = vma->vm_start - addr; ++ addr = vma->vm_end; ++ if (mm->start_brk <= addr && addr < mm->mmap_base) { ++ start_addr = addr = mm->mmap_base; ++ mm->cached_hole_size = 0; ++ goto full_search; ++ } ++ } ++} ++ ++unsigned long ++arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, ++ const unsigned long len, const unsigned long pgoff, ++ const unsigned long flags) ++{ ++ struct vm_area_struct *vma; ++ struct mm_struct *mm = current->mm; ++ unsigned long base = mm->mmap_base, addr = addr0, pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ /* requested length too big for entire address space */ ++ if (len > pax_task_size) ++ return -ENOMEM; ++ ++ if (flags & MAP_FIXED) ++ return addr; ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!nx_enabled && (mm->pax_flags & MF_PAX_PAGEEXEC) && (flags & MAP_EXECUTABLE)) ++ goto bottomup; ++#endif ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ ++ /* requesting a specific address */ ++ if (addr) { ++ addr = PAGE_ALIGN(addr); ++ vma = find_vma(mm, addr); ++ if (pax_task_size - len >= addr && ++ (!vma || addr + len <= vma->vm_start)) ++ return addr; ++ } ++ ++ /* check if free_area_cache is useful for us */ ++ if (len <= mm->cached_hole_size) { ++ mm->cached_hole_size = 0; ++ mm->free_area_cache = mm->mmap_base; ++ } ++ ++ /* either no address requested or can't fit in requested address hole */ ++ addr = mm->free_area_cache; ++ ++ /* make sure it can fit in the remaining address space */ ++ if (addr > len) { ++ vma = find_vma(mm, addr-len); ++ if (!vma || addr <= vma->vm_start) ++ /* remember the address as a hint for next time */ ++ return (mm->free_area_cache = addr-len); ++ } ++ ++ if (mm->mmap_base < len) ++ goto bottomup; ++ ++ addr = mm->mmap_base-len; ++ ++ do { ++ /* ++ * Lookup failure means no vma is above this address, ++ * else if new region fits below vma->vm_start, ++ * return with success: ++ */ ++ vma = find_vma(mm, addr); ++ if (!vma || addr+len <= vma->vm_start) ++ /* remember the address as a hint for next time */ ++ return (mm->free_area_cache = addr); ++ ++ /* remember the largest hole we saw so far */ ++ if (addr + mm->cached_hole_size < vma->vm_start) ++ mm->cached_hole_size = vma->vm_start - addr; ++ ++ /* try just below the current vma->vm_start */ ++ addr = vma->vm_start-len; ++ } while (len < vma->vm_start); ++ ++bottomup: ++ /* ++ * A failed mmap() very likely causes application failure, ++ * so fall back to the bottom-up function here. This scenario ++ * can happen with large stack limits and large mmap() ++ * allocations. ++ */ ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ mm->mmap_base = SEGMEXEC_TASK_UNMAPPED_BASE; ++ else ++#endif ++ ++ mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ ++ mm->free_area_cache = mm->mmap_base; ++ mm->cached_hole_size = ~0UL; ++ addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); ++ /* ++ * Restore the topdown base: ++ */ ++ mm->mmap_base = base; ++ mm->free_area_cache = base; ++ mm->cached_hole_size = ~0UL; ++ ++ return addr; ++} + + struct sel_arg_struct { + unsigned long n; +diff -urNp linux-2.6.31.1/arch/x86/kernel/sys_x86_64.c linux-2.6.31.1/arch/x86/kernel/sys_x86_64.c +--- linux-2.6.31.1/arch/x86/kernel/sys_x86_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/sys_x86_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -47,8 +47,8 @@ out: + return error; + } + +-static void find_start_end(unsigned long flags, unsigned long *begin, +- unsigned long *end) ++static void find_start_end(struct mm_struct *mm, unsigned long flags, ++ unsigned long *begin, unsigned long *end) + { + if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) { + unsigned long new_begin; +@@ -67,7 +67,7 @@ static void find_start_end(unsigned long + *begin = new_begin; + } + } else { +- *begin = TASK_UNMAPPED_BASE; ++ *begin = mm->mmap_base; + *end = TASK_SIZE; + } + } +@@ -84,11 +84,15 @@ arch_get_unmapped_area(struct file *filp + if (flags & MAP_FIXED) + return addr; + +- find_start_end(flags, &begin, &end); ++ find_start_end(mm, flags, &begin, &end); + + if (len > end) + return -ENOMEM; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + addr = PAGE_ALIGN(addr); + vma = find_vma(mm, addr); +@@ -143,7 +147,7 @@ arch_get_unmapped_area_topdown(struct fi + { + struct vm_area_struct *vma; + struct mm_struct *mm = current->mm; +- unsigned long addr = addr0; ++ unsigned long base = mm->mmap_base, addr = addr0; + + /* requested length too big for entire address space */ + if (len > TASK_SIZE) +@@ -156,6 +160,10 @@ arch_get_unmapped_area_topdown(struct fi + if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) + goto bottomup; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + /* requesting a specific address */ + if (addr) { + addr = PAGE_ALIGN(addr); +@@ -213,13 +221,21 @@ bottomup: + * can happen with large stack limits and large mmap() + * allocations. + */ ++ mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ ++ mm->free_area_cache = mm->mmap_base; + mm->cached_hole_size = ~0UL; +- mm->free_area_cache = TASK_UNMAPPED_BASE; + addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); + /* + * Restore the topdown base: + */ +- mm->free_area_cache = mm->mmap_base; ++ mm->mmap_base = base; ++ mm->free_area_cache = base; + mm->cached_hole_size = ~0UL; + + return addr; +diff -urNp linux-2.6.31.1/arch/x86/kernel/time_32.c linux-2.6.31.1/arch/x86/kernel/time_32.c +--- linux-2.6.31.1/arch/x86/kernel/time_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/time_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -47,22 +47,32 @@ unsigned long profile_pc(struct pt_regs + unsigned long pc = instruction_pointer(regs); + + #ifdef CONFIG_SMP +- if (!user_mode_vm(regs) && in_lock_functions(pc)) { ++ if (!user_mode(regs) && in_lock_functions(pc)) { + #ifdef CONFIG_FRAME_POINTER +- return *(unsigned long *)(regs->bp + sizeof(long)); ++ return ktla_ktva(*(unsigned long *)(regs->bp + sizeof(long))); + #else + unsigned long *sp = (unsigned long *)®s->sp; + + /* Return address is either directly at stack pointer + or above a saved flags. Eflags has bits 22-31 zero, + kernel addresses don't. */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ return ktla_ktva(sp[0]); ++#else + if (sp[0] >> 22) + return sp[0]; + if (sp[1] >> 22) + return sp[1]; + #endif ++ ++#endif + } + #endif ++ ++ if (!user_mode(regs)) ++ pc = ktla_ktva(pc); ++ + return pc; + } + EXPORT_SYMBOL(profile_pc); +diff -urNp linux-2.6.31.1/arch/x86/kernel/time_64.c linux-2.6.31.1/arch/x86/kernel/time_64.c +--- linux-2.6.31.1/arch/x86/kernel/time_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/time_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -25,8 +25,6 @@ + #include + #include + +-volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES; +- + unsigned long profile_pc(struct pt_regs *regs) + { + unsigned long pc = instruction_pointer(regs); +@@ -34,7 +32,7 @@ unsigned long profile_pc(struct pt_regs + /* Assume the lock function has either no stack frame or a copy + of flags from PUSHF + Eflags always has bits 22 and up cleared unlike kernel addresses. */ +- if (!user_mode_vm(regs) && in_lock_functions(pc)) { ++ if (!user_mode(regs) && in_lock_functions(pc)) { + #ifdef CONFIG_FRAME_POINTER + return *(unsigned long *)(regs->bp + sizeof(long)); + #else +diff -urNp linux-2.6.31.1/arch/x86/kernel/tls.c linux-2.6.31.1/arch/x86/kernel/tls.c +--- linux-2.6.31.1/arch/x86/kernel/tls.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/tls.c 2009-10-01 20:12:42.000000000 -0400 +@@ -85,6 +85,11 @@ int do_set_thread_area(struct task_struc + if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX) + return -EINVAL; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((p->mm->pax_flags & MF_PAX_SEGMEXEC) && (info.contents & MODIFY_LDT_CONTENTS_CODE)) ++ return -EINVAL; ++#endif ++ + set_tls_desc(p, idx, &info, 1); + + return 0; +diff -urNp linux-2.6.31.1/arch/x86/kernel/traps.c linux-2.6.31.1/arch/x86/kernel/traps.c +--- linux-2.6.31.1/arch/x86/kernel/traps.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/traps.c 2009-10-01 20:12:42.000000000 -0400 +@@ -70,14 +70,6 @@ asmlinkage int system_call(void); + + /* Do we ignore FPU interrupts ? */ + char ignore_fpu_irq; +- +-/* +- * The IDT has to be page-aligned to simplify the Pentium +- * F0 0F bug workaround.. We have a special link segment +- * for this. +- */ +-gate_desc idt_table[256] +- __attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, }; + #endif + + DECLARE_BITMAP(used_vectors, NR_VECTORS); +@@ -115,7 +107,7 @@ static inline void preempt_conditional_c + static inline void + die_if_kernel(const char *str, struct pt_regs *regs, long err) + { +- if (!user_mode_vm(regs)) ++ if (!user_mode(regs)) + die(str, regs, err); + } + #endif +@@ -127,7 +119,7 @@ do_trap(int trapnr, int signr, char *str + struct task_struct *tsk = current; + + #ifdef CONFIG_X86_32 +- if (regs->flags & X86_VM_MASK) { ++ if (v8086_mode(regs)) { + /* + * traps 0, 1, 3, 4, and 5 should be forwarded to vm86. + * On nmi (interrupt 2), do_trap should not be called. +@@ -138,7 +130,7 @@ do_trap(int trapnr, int signr, char *str + } + #endif + +- if (!user_mode(regs)) ++ if (!user_mode_novm(regs)) + goto kernel_trap; + + #ifdef CONFIG_X86_32 +@@ -161,7 +153,7 @@ trap_signal: + printk_ratelimit()) { + printk(KERN_INFO + "%s[%d] trap %s ip:%lx sp:%lx error:%lx", +- tsk->comm, tsk->pid, str, ++ tsk->comm, task_pid_nr(tsk), str, + regs->ip, regs->sp, error_code); + print_vma_addr(" in ", regs->ip); + printk("\n"); +@@ -180,6 +172,12 @@ kernel_trap: + tsk->thread.trap_no = trapnr; + die(str, regs, error_code); + } ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ if (trapnr == 4) ++ pax_report_refcount_overflow(regs); ++#endif ++ + return; + + #ifdef CONFIG_X86_32 +@@ -268,14 +266,30 @@ do_general_protection(struct pt_regs *re + conditional_sti(regs); + + #ifdef CONFIG_X86_32 +- if (regs->flags & X86_VM_MASK) ++ if (v8086_mode(regs)) + goto gp_in_vm86; + #endif + + tsk = current; +- if (!user_mode(regs)) ++ if (!user_mode_novm(regs)) + goto gp_in_kernel; + ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) ++ if (!nx_enabled && tsk->mm && (tsk->mm->pax_flags & MF_PAX_PAGEEXEC)) { ++ struct mm_struct *mm = tsk->mm; ++ unsigned long limit; ++ ++ down_write(&mm->mmap_sem); ++ limit = mm->context.user_cs_limit; ++ if (limit < TASK_SIZE) { ++ track_exec_limit(mm, limit, TASK_SIZE, VM_EXEC); ++ up_write(&mm->mmap_sem); ++ return; ++ } ++ up_write(&mm->mmap_sem); ++ } ++#endif ++ + tsk->thread.error_code = error_code; + tsk->thread.trap_no = 13; + +@@ -308,6 +322,13 @@ gp_in_kernel: + if (notify_die(DIE_GPF, "general protection fault", regs, + error_code, 13, SIGSEGV) == NOTIFY_STOP) + return; ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ if ((regs->cs & 0xFFFF) == __KERNEL_CS) ++ die("PAX: suspicious general protection fault", regs, error_code); ++ else ++#endif ++ + die("general protection fault", regs, error_code); + } + +@@ -561,7 +582,7 @@ dotraplinkage void __kprobes do_debug(st + } + + #ifdef CONFIG_X86_32 +- if (regs->flags & X86_VM_MASK) ++ if (v8086_mode(regs)) + goto debug_vm86; + #endif + +@@ -573,7 +594,7 @@ dotraplinkage void __kprobes do_debug(st + * kernel space (but re-enable TF when returning to user mode). + */ + if (condition & DR_STEP) { +- if (!user_mode(regs)) ++ if (!user_mode_novm(regs)) + goto clear_TF_reenable; + } + +@@ -760,7 +781,7 @@ do_simd_coprocessor_error(struct pt_regs + * Handle strange cache flush from user space exception + * in all other cases. This is undocumented behaviour. + */ +- if (regs->flags & X86_VM_MASK) { ++ if (v8086_mode(regs)) { + handle_vm86_fault((struct kernel_vm86_regs *)regs, error_code); + return; + } +@@ -789,19 +810,14 @@ do_spurious_interrupt_bug(struct pt_regs + #ifdef CONFIG_X86_32 + unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp) + { +- struct desc_struct *gdt = get_cpu_gdt_table(smp_processor_id()); + unsigned long base = (kesp - uesp) & -THREAD_SIZE; + unsigned long new_kesp = kesp - base; + unsigned long lim_pages = (new_kesp | (THREAD_SIZE - 1)) >> PAGE_SHIFT; +- __u64 desc = *(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS]; ++ struct desc_struct ss; + + /* Set up base for espfix segment */ +- desc &= 0x00f0ff0000000000ULL; +- desc |= ((((__u64)base) << 16) & 0x000000ffffff0000ULL) | +- ((((__u64)base) << 32) & 0xff00000000000000ULL) | +- ((((__u64)lim_pages) << 32) & 0x000f000000000000ULL) | +- (lim_pages & 0xffff); +- *(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS] = desc; ++ pack_descriptor(&ss, base, lim_pages, 0x93, 0xC); ++ write_gdt_entry(get_cpu_gdt_table(smp_processor_id()), GDT_ENTRY_ESPFIX_SS, &ss, DESCTYPE_S); + + return new_kesp; + } +diff -urNp linux-2.6.31.1/arch/x86/kernel/tsc.c linux-2.6.31.1/arch/x86/kernel/tsc.c +--- linux-2.6.31.1/arch/x86/kernel/tsc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/tsc.c 2009-10-01 20:12:42.000000000 -0400 +@@ -790,7 +790,7 @@ static struct dmi_system_id __initdata b + DMI_MATCH(DMI_BOARD_NAME, "2635FA0"), + }, + }, +- {} ++ { NULL, NULL, {{0, {0}}}, NULL} + }; + + static void __init check_system_tsc_reliable(void) +diff -urNp linux-2.6.31.1/arch/x86/kernel/vm86_32.c linux-2.6.31.1/arch/x86/kernel/vm86_32.c +--- linux-2.6.31.1/arch/x86/kernel/vm86_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/vm86_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -148,7 +148,7 @@ struct pt_regs *save_v86_state(struct ke + do_exit(SIGSEGV); + } + +- tss = &per_cpu(init_tss, get_cpu()); ++ tss = init_tss + get_cpu(); + current->thread.sp0 = current->thread.saved_sp0; + current->thread.sysenter_cs = __KERNEL_CS; + load_sp0(tss, ¤t->thread); +@@ -324,7 +324,7 @@ static void do_sys_vm86(struct kernel_vm + tsk->thread.saved_fs = info->regs32->fs; + tsk->thread.saved_gs = get_user_gs(info->regs32); + +- tss = &per_cpu(init_tss, get_cpu()); ++ tss = init_tss + get_cpu(); + tsk->thread.sp0 = (unsigned long) &info->VM86_TSS_ESP0; + if (cpu_has_sep) + tsk->thread.sysenter_cs = 0; +diff -urNp linux-2.6.31.1/arch/x86/kernel/vmi_32.c linux-2.6.31.1/arch/x86/kernel/vmi_32.c +--- linux-2.6.31.1/arch/x86/kernel/vmi_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/vmi_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -102,18 +102,43 @@ static unsigned patch_internal(int call, + { + u64 reloc; + struct vmi_relocation_info *const rel = (struct vmi_relocation_info *)&reloc; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + reloc = call_vrom_long_func(vmi_rom, get_reloc, call); + switch(rel->type) { + case VMI_RELOCATION_CALL_REL: + BUG_ON(len < 5); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + *(char *)insnbuf = MNEM_CALL; + patch_offset(insnbuf, ip, (unsigned long)rel->eip); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + return 5; + + case VMI_RELOCATION_JUMP_REL: + BUG_ON(len < 5); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + *(char *)insnbuf = MNEM_JMP; + patch_offset(insnbuf, ip, (unsigned long)rel->eip); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + return 5; + + case VMI_RELOCATION_NOP: +@@ -404,13 +429,13 @@ static void vmi_set_pud(pud_t *pudp, pud + + static void vmi_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) + { +- const pte_t pte = { .pte = 0 }; ++ const pte_t pte = __pte(0ULL); + vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0)); + } + + static void vmi_pmd_clear(pmd_t *pmd) + { +- const pte_t pte = { .pte = 0 }; ++ const pte_t pte = __pte(0ULL); + vmi_ops.set_pte(pte, (pte_t *)pmd, VMI_PAGE_PD); + } + #endif +@@ -438,8 +463,8 @@ vmi_startup_ipi_hook(int phys_apicid, un + ap.ss = __KERNEL_DS; + ap.esp = (unsigned long) start_esp; + +- ap.ds = __USER_DS; +- ap.es = __USER_DS; ++ ap.ds = __KERNEL_DS; ++ ap.es = __KERNEL_DS; + ap.fs = __KERNEL_PERCPU; + ap.gs = __KERNEL_STACK_CANARY; + +@@ -640,12 +665,20 @@ static inline int __init activate_vmi(vo + u64 reloc; + const struct vmi_relocation_info *rel = (struct vmi_relocation_info *)&reloc; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + if (call_vrom_func(vmi_rom, vmi_init) != 0) { + printk(KERN_ERR "VMI ROM failed to initialize!"); + return 0; + } + savesegment(cs, kernel_cs); + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + pv_info.paravirt_enabled = 1; + pv_info.kernel_rpl = kernel_cs & SEGMENT_RPL_MASK; + pv_info.name = "vmi"; +@@ -836,6 +869,10 @@ static inline int __init activate_vmi(vo + + para_fill(pv_irq_ops.safe_halt, Halt); + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + /* + * Alternative instruction rewriting doesn't happen soon enough + * to convert VMI_IRET to a call instead of a jump; so we have +diff -urNp linux-2.6.31.1/arch/x86/kernel/vmlinux.lds.S linux-2.6.31.1/arch/x86/kernel/vmlinux.lds.S +--- linux-2.6.31.1/arch/x86/kernel/vmlinux.lds.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/vmlinux.lds.S 2009-10-01 20:12:42.000000000 -0400 +@@ -26,6 +26,22 @@ + #include + #include + #include ++#include ++ ++#undef PMD_SIZE ++#undef PMD_SHIFT ++#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) ++#define PMD_SHIFT 21 ++#else ++#define PMD_SHIFT 22 ++#endif ++#define PMD_SIZE (1 << PMD_SHIFT) ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++#define __KERNEL_TEXT_OFFSET (LOAD_OFFSET + ____LOAD_PHYSICAL_ADDR) ++#else ++#define __KERNEL_TEXT_OFFSET 0 ++#endif + + #undef i386 /* in case the preprocessor is a 32bit one */ + +@@ -34,46 +50,52 @@ OUTPUT_FORMAT(CONFIG_OUTPUT_FORMAT, CONF + #ifdef CONFIG_X86_32 + OUTPUT_ARCH(i386) + ENTRY(phys_startup_32) +-jiffies = jiffies_64; + #else + OUTPUT_ARCH(i386:x86-64) + ENTRY(phys_startup_64) +-jiffies_64 = jiffies; + #endif + ++jiffies = jiffies_64; ++ + PHDRS { + text PT_LOAD FLAGS(5); /* R_E */ +- data PT_LOAD FLAGS(7); /* RWE */ ++ rodata PT_LOAD FLAGS(4); /* R__ */ ++ data PT_LOAD FLAGS(6); /* RW_ */ + #ifdef CONFIG_X86_64 +- user PT_LOAD FLAGS(7); /* RWE */ ++ user PT_LOAD FLAGS(5); /* R_E */ ++#endif ++ init.begin PT_LOAD FLAGS(6); /* RW_ */ + #ifdef CONFIG_SMP +- percpu PT_LOAD FLAGS(7); /* RWE */ ++ percpu PT_LOAD FLAGS(6); /* RW_ */ + #endif ++ text.init PT_LOAD FLAGS(5); /* R_E */ ++ text.exit PT_LOAD FLAGS(5); /* R_E */ + init PT_LOAD FLAGS(7); /* RWE */ +-#endif + note PT_NOTE FLAGS(0); /* ___ */ + } + + SECTIONS + { + #ifdef CONFIG_X86_32 +- . = LOAD_OFFSET + LOAD_PHYSICAL_ADDR; +- phys_startup_32 = startup_32 - LOAD_OFFSET; ++ . = LOAD_OFFSET + ____LOAD_PHYSICAL_ADDR; + #else +- . = __START_KERNEL; +- phys_startup_64 = startup_64 - LOAD_OFFSET; ++ . = __START_KERNEL; + #endif + + /* Text and read-only data */ + +- /* bootstrapping code */ +- .text.head : AT(ADDR(.text.head) - LOAD_OFFSET) { ++ .text (. - __KERNEL_TEXT_OFFSET): AT(ADDR(.text) - LOAD_OFFSET + __KERNEL_TEXT_OFFSET) { ++ /* bootstrapping code */ ++#ifdef CONFIG_X86_32 ++ phys_startup_32 = startup_32 - LOAD_OFFSET + __KERNEL_TEXT_OFFSET; ++#else ++ phys_startup_64 = startup_64 - LOAD_OFFSET + __KERNEL_TEXT_OFFSET; ++#endif ++ __LOAD_PHYSICAL_ADDR = . - LOAD_OFFSET + __KERNEL_TEXT_OFFSET; + _text = .; + *(.text.head) +- } :text = 0x9090 + +- /* The rest of the text */ +- .text : AT(ADDR(.text) - LOAD_OFFSET) { ++ /* The rest of the text */ + #ifdef CONFIG_X86_32 + /* not really needed, already page aligned */ + . = ALIGN(PAGE_SIZE); +@@ -92,7 +114,10 @@ SECTIONS + _etext = .; + } :text = 0x9090 + +- NOTES :text :note ++ . += __KERNEL_TEXT_OFFSET; ++ ++ . = ALIGN(PAGE_SIZE); ++ NOTES :rodata :note + + /* Exception table */ + . = ALIGN(16); +@@ -100,22 +125,53 @@ SECTIONS + __start___ex_table = .; + *(__ex_table) + __stop___ex_table = .; +- } :text = 0x9090 ++ } :rodata + + RO_DATA(PAGE_SIZE) + ++#ifdef CONFIG_X86_32 ++ . = ALIGN(PAGE_SIZE); ++ .rodata.page_aligned : AT(ADDR(.rodata.page_aligned) - LOAD_OFFSET) { ++ *(.idt) ++ . = ALIGN(PAGE_SIZE); ++ *(.empty_zero_page) ++ *(.swapper_pg_pmd) ++ *(.swapper_pg_dir) ++ ++#if defined(CONFIG_PAX_KERNEXEC) && !defined(CONFIG_MODULES) ++ . = ALIGN(PMD_SIZE); ++#endif ++ ++ } ++ ++#if defined(CONFIG_PAX_KERNEXEC) && defined(CONFIG_MODULES) ++ . = ALIGN(PAGE_SIZE); ++ .module.text : AT(ADDR(.module.text) - LOAD_OFFSET) { ++ MODULES_EXEC_VADDR = .; ++ BYTE(0) ++ . += (8 * 1024 * 1024); ++ . = ALIGN(PMD_SIZE); ++ MODULES_EXEC_END = . - 1; ++ } ++#endif ++#endif ++ + /* Data */ + .data : AT(ADDR(.data) - LOAD_OFFSET) { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ . = ALIGN(PMD_SIZE); ++#else ++ . = ALIGN(PAGE_SIZE); ++#endif ++ + /* Start of data section */ + _sdata = .; + + /* init_task */ + INIT_TASK_DATA(THREAD_SIZE) + +-#ifdef CONFIG_X86_32 +- /* 32 bit has nosave before _edata */ + NOSAVE_DATA +-#endif + + PAGE_ALIGNED_DATA(PAGE_SIZE) + *(.data.idt) +@@ -182,12 +238,6 @@ SECTIONS + } + vgetcpu_mode = VVIRT(.vgetcpu_mode); + +- . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); +- .jiffies : AT(VLOAD(.jiffies)) { +- *(.jiffies) +- } +- jiffies = VVIRT(.jiffies); +- + .vsyscall_3 ADDR(.vsyscall_0) + 3072: AT(VLOAD(.vsyscall_3)) { + *(.vsyscall_3) + } +@@ -205,12 +255,19 @@ SECTIONS + #endif /* CONFIG_X86_64 */ + + /* Init code and data - will be freed after init */ +- . = ALIGN(PAGE_SIZE); + .init.begin : AT(ADDR(.init.begin) - LOAD_OFFSET) { ++ BYTE(0) ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ . = ALIGN(PMD_SIZE); ++#else ++ . = ALIGN(PAGE_SIZE); ++#endif ++ + __init_begin = .; /* paired with __init_end */ +- } ++ } :init.begin + +-#if defined(CONFIG_X86_64) && defined(CONFIG_SMP) ++#ifdef CONFIG_SMP + /* + * percpu offsets are zero-based on SMP. PERCPU_VADDR() changes the + * output PHDR, so the next output section - .init.text - should +@@ -219,18 +276,26 @@ SECTIONS + PERCPU_VADDR(0, :percpu) + #endif + +- .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) { ++ init_begin = .; ++ .init.text (. - __KERNEL_TEXT_OFFSET): AT(init_begin - LOAD_OFFSET) { + _sinittext = .; + INIT_TEXT + _einittext = .; +- } +-#ifdef CONFIG_X86_64 +- :init +-#endif ++ } :text.init ++ ++ /* ++ * .exit.text is discard at runtime, not link time, to deal with ++ * references from .altinstructions and .eh_frame ++ */ ++ .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { ++ EXIT_TEXT ++ . = ALIGN(16); ++ } :text.exit ++ . = init_begin + SIZEOF(.init.text) + SIZEOF(.exit.text); + + .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { + INIT_DATA +- } ++ } :init + + . = ALIGN(16); + .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { +@@ -276,14 +341,6 @@ SECTIONS + *(.altinstr_replacement) + } + +- /* +- * .exit.text is discard at runtime, not link time, to deal with +- * references from .altinstructions and .eh_frame +- */ +- .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { +- EXIT_TEXT +- } +- + .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { + EXIT_DATA + } +@@ -297,7 +354,7 @@ SECTIONS + } + #endif + +-#if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP) ++#ifndef CONFIG_SMP + PERCPU(PAGE_SIZE) + #endif + +@@ -320,12 +377,6 @@ SECTIONS + . = ALIGN(PAGE_SIZE); + } + +-#ifdef CONFIG_X86_64 +- .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { +- NOSAVE_DATA +- } +-#endif +- + /* BSS */ + . = ALIGN(PAGE_SIZE); + .bss : AT(ADDR(.bss) - LOAD_OFFSET) { +@@ -341,6 +392,7 @@ SECTIONS + __brk_base = .; + . += 64 * 1024; /* 64k alignment slop space */ + *(.brk_reservation) /* areas brk users have reserved */ ++ . = ALIGN(PMD_SIZE); + __brk_limit = .; + } + +@@ -369,13 +421,12 @@ SECTIONS + * for the boot processor. + */ + #define INIT_PER_CPU(x) init_per_cpu__##x = per_cpu__##x + __per_cpu_load +-INIT_PER_CPU(gdt_page); + INIT_PER_CPU(irq_stack_union); + + /* + * Build-time check on the image size: + */ +-. = ASSERT((_end - _text <= KERNEL_IMAGE_SIZE), ++. = ASSERT((_end - _text - __KERNEL_TEXT_OFFSET <= KERNEL_IMAGE_SIZE), + "kernel image bigger than KERNEL_IMAGE_SIZE"); + + #ifdef CONFIG_SMP +diff -urNp linux-2.6.31.1/arch/x86/kernel/vsyscall_64.c linux-2.6.31.1/arch/x86/kernel/vsyscall_64.c +--- linux-2.6.31.1/arch/x86/kernel/vsyscall_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/vsyscall_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -79,6 +79,7 @@ void update_vsyscall(struct timespec *wa + + write_seqlock_irqsave(&vsyscall_gtod_data.lock, flags); + /* copy vsyscall data */ ++ strlcpy(vsyscall_gtod_data.clock.name, clock->name, sizeof vsyscall_gtod_data.clock.name); + vsyscall_gtod_data.clock.vread = clock->vread; + vsyscall_gtod_data.clock.cycle_last = clock->cycle_last; + vsyscall_gtod_data.clock.mask = clock->mask; +@@ -201,7 +202,7 @@ vgetcpu(unsigned *cpu, unsigned *node, s + We do this here because otherwise user space would do it on + its own in a likely inferior way (no access to jiffies). + If you don't like it pass NULL. */ +- if (tcache && tcache->blob[0] == (j = __jiffies)) { ++ if (tcache && tcache->blob[0] == (j = jiffies)) { + p = tcache->blob[1]; + } else if (__vgetcpu_mode == VGETCPU_RDTSCP) { + /* Load per CPU data from RDTSCP */ +@@ -240,13 +241,13 @@ static ctl_table kernel_table2[] = { + .data = &vsyscall_gtod_data.sysctl_enabled, .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = vsyscall_sysctl_change }, +- {} ++ { 0, NULL, NULL, 0, 0, NULL, NULL, NULL, NULL, NULL, NULL } + }; + + static ctl_table kernel_root_table2[] = { + { .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555, + .child = kernel_table2 }, +- {} ++ { 0, NULL, NULL, 0, 0, NULL, NULL, NULL, NULL, NULL, NULL } + }; + #endif + +diff -urNp linux-2.6.31.1/arch/x86/kernel/x8664_ksyms_64.c linux-2.6.31.1/arch/x86/kernel/x8664_ksyms_64.c +--- linux-2.6.31.1/arch/x86/kernel/x8664_ksyms_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kernel/x8664_ksyms_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -30,8 +30,6 @@ EXPORT_SYMBOL(__put_user_8); + + EXPORT_SYMBOL(copy_user_generic); + EXPORT_SYMBOL(__copy_user_nocache); +-EXPORT_SYMBOL(copy_from_user); +-EXPORT_SYMBOL(copy_to_user); + EXPORT_SYMBOL(__copy_from_user_inatomic); + + EXPORT_SYMBOL(copy_page); +diff -urNp linux-2.6.31.1/arch/x86/kvm/svm.c linux-2.6.31.1/arch/x86/kvm/svm.c +--- linux-2.6.31.1/arch/x86/kvm/svm.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kvm/svm.c 2009-10-01 20:12:42.000000000 -0400 +@@ -2289,7 +2289,19 @@ static void reload_tss(struct kvm_vcpu * + int cpu = raw_smp_processor_id(); + + struct svm_cpu_data *svm_data = per_cpu(svm_data, cpu); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + svm_data->tss_desc->type = 9; /* available 32/64-bit TSS */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + load_TR_desc(); + } + +@@ -2673,7 +2685,7 @@ static u64 svm_get_mt_mask(struct kvm_vc + return 0; + } + +-static struct kvm_x86_ops svm_x86_ops = { ++static const struct kvm_x86_ops svm_x86_ops = { + .cpu_has_kvm_support = has_svm, + .disabled_by_bios = is_disabled, + .hardware_setup = svm_hardware_setup, +diff -urNp linux-2.6.31.1/arch/x86/kvm/vmx.c linux-2.6.31.1/arch/x86/kvm/vmx.c +--- linux-2.6.31.1/arch/x86/kvm/vmx.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kvm/vmx.c 2009-10-01 20:12:42.000000000 -0400 +@@ -519,9 +519,23 @@ static void reload_tss(void) + struct descriptor_table gdt; + struct desc_struct *descs; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + kvm_get_gdt(&gdt); + descs = (void *)gdt.base; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + descs[GDT_ENTRY_TSS].type = 9; /* available TSS */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + load_TR_desc(); + } + +@@ -1321,6 +1335,11 @@ static __init int alloc_kvm_area(void) + + static __init int hardware_setup(void) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + if (setup_vmcs_config(&vmcs_config) < 0) + return -EIO; + +@@ -1336,8 +1355,19 @@ static __init int hardware_setup(void) + if (!cpu_has_vmx_flexpriority()) + flexpriority_enabled = 0; + +- if (!cpu_has_vmx_tpr_shadow()) +- kvm_x86_ops->update_cr8_intercept = NULL; ++ if (!cpu_has_vmx_tpr_shadow()) { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ ++ *(void **)&kvm_x86_ops->update_cr8_intercept = NULL; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ ++ } + + return alloc_kvm_area(); + } +@@ -2239,7 +2269,7 @@ static int vmx_vcpu_setup(struct vcpu_vm + vmcs_writel(HOST_IDTR_BASE, dt.base); /* 22.2.4 */ + + asm("mov $.Lkvm_vmx_return, %0" : "=r"(kvm_vmx_return)); +- vmcs_writel(HOST_RIP, kvm_vmx_return); /* 22.2.5 */ ++ vmcs_writel(HOST_RIP, ktla_ktva(kvm_vmx_return)); /* 22.2.5 */ + vmcs_write32(VM_EXIT_MSR_STORE_COUNT, 0); + vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, 0); + vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, 0); +@@ -3493,6 +3523,12 @@ static void vmx_vcpu_run(struct kvm_vcpu + "jmp .Lkvm_vmx_return \n\t" + ".Llaunched: " __ex(ASM_VMX_VMRESUME) "\n\t" + ".Lkvm_vmx_return: " ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ "ljmp %[cs],$.Lkvm_vmx_return2\n\t" ++ ".Lkvm_vmx_return2: " ++#endif ++ + /* Save guest registers, load host registers, keep flags */ + "xchg %0, (%%"R"sp) \n\t" + "mov %%"R"ax, %c[rax](%0) \n\t" +@@ -3539,6 +3575,11 @@ static void vmx_vcpu_run(struct kvm_vcpu + [r15]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R15])), + #endif + [cr2]"i"(offsetof(struct vcpu_vmx, vcpu.arch.cr2)) ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ ,[cs]"i"(__KERNEL_CS) ++#endif ++ + : "cc", "memory" + , R"bx", R"di", R"si" + #ifdef CONFIG_X86_64 +@@ -3555,7 +3596,7 @@ static void vmx_vcpu_run(struct kvm_vcpu + if (vmx->rmode.irq.pending) + fixup_rmode_irq(vmx); + +- asm("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS)); ++ asm("mov %0, %%ds; mov %0, %%es" : : "r"(__KERNEL_DS)); + vmx->launched = 1; + + vmx_complete_interrupts(vmx); +@@ -3698,7 +3739,7 @@ static u64 vmx_get_mt_mask(struct kvm_vc + return ret; + } + +-static struct kvm_x86_ops vmx_x86_ops = { ++static const struct kvm_x86_ops vmx_x86_ops = { + .cpu_has_kvm_support = cpu_has_kvm_support, + .disabled_by_bios = vmx_disabled_by_bios, + .hardware_setup = hardware_setup, +diff -urNp linux-2.6.31.1/arch/x86/kvm/x86.c linux-2.6.31.1/arch/x86/kvm/x86.c +--- linux-2.6.31.1/arch/x86/kvm/x86.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/kvm/x86.c 2009-10-01 20:12:42.000000000 -0400 +@@ -73,42 +73,42 @@ static int kvm_dev_ioctl_get_supported_c + struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu, + u32 function, u32 index); + +-struct kvm_x86_ops *kvm_x86_ops; ++const struct kvm_x86_ops *kvm_x86_ops; + EXPORT_SYMBOL_GPL(kvm_x86_ops); + + struct kvm_stats_debugfs_item debugfs_entries[] = { +- { "pf_fixed", VCPU_STAT(pf_fixed) }, +- { "pf_guest", VCPU_STAT(pf_guest) }, +- { "tlb_flush", VCPU_STAT(tlb_flush) }, +- { "invlpg", VCPU_STAT(invlpg) }, +- { "exits", VCPU_STAT(exits) }, +- { "io_exits", VCPU_STAT(io_exits) }, +- { "mmio_exits", VCPU_STAT(mmio_exits) }, +- { "signal_exits", VCPU_STAT(signal_exits) }, +- { "irq_window", VCPU_STAT(irq_window_exits) }, +- { "nmi_window", VCPU_STAT(nmi_window_exits) }, +- { "halt_exits", VCPU_STAT(halt_exits) }, +- { "halt_wakeup", VCPU_STAT(halt_wakeup) }, +- { "hypercalls", VCPU_STAT(hypercalls) }, +- { "request_irq", VCPU_STAT(request_irq_exits) }, +- { "irq_exits", VCPU_STAT(irq_exits) }, +- { "host_state_reload", VCPU_STAT(host_state_reload) }, +- { "efer_reload", VCPU_STAT(efer_reload) }, +- { "fpu_reload", VCPU_STAT(fpu_reload) }, +- { "insn_emulation", VCPU_STAT(insn_emulation) }, +- { "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) }, +- { "irq_injections", VCPU_STAT(irq_injections) }, +- { "nmi_injections", VCPU_STAT(nmi_injections) }, +- { "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) }, +- { "mmu_pte_write", VM_STAT(mmu_pte_write) }, +- { "mmu_pte_updated", VM_STAT(mmu_pte_updated) }, +- { "mmu_pde_zapped", VM_STAT(mmu_pde_zapped) }, +- { "mmu_flooded", VM_STAT(mmu_flooded) }, +- { "mmu_recycled", VM_STAT(mmu_recycled) }, +- { "mmu_cache_miss", VM_STAT(mmu_cache_miss) }, +- { "mmu_unsync", VM_STAT(mmu_unsync) }, +- { "remote_tlb_flush", VM_STAT(remote_tlb_flush) }, +- { "largepages", VM_STAT(lpages) }, ++ { "pf_fixed", VCPU_STAT(pf_fixed), NULL }, ++ { "pf_guest", VCPU_STAT(pf_guest), NULL }, ++ { "tlb_flush", VCPU_STAT(tlb_flush), NULL }, ++ { "invlpg", VCPU_STAT(invlpg), NULL }, ++ { "exits", VCPU_STAT(exits), NULL }, ++ { "io_exits", VCPU_STAT(io_exits), NULL }, ++ { "mmio_exits", VCPU_STAT(mmio_exits), NULL }, ++ { "signal_exits", VCPU_STAT(signal_exits), NULL }, ++ { "irq_window", VCPU_STAT(irq_window_exits), NULL }, ++ { "nmi_window", VCPU_STAT(nmi_window_exits), NULL }, ++ { "halt_exits", VCPU_STAT(halt_exits), NULL }, ++ { "halt_wakeup", VCPU_STAT(halt_wakeup), NULL }, ++ { "hypercalls", VCPU_STAT(hypercalls), NULL }, ++ { "request_irq", VCPU_STAT(request_irq_exits), NULL }, ++ { "irq_exits", VCPU_STAT(irq_exits), NULL }, ++ { "host_state_reload", VCPU_STAT(host_state_reload), NULL }, ++ { "efer_reload", VCPU_STAT(efer_reload), NULL }, ++ { "fpu_reload", VCPU_STAT(fpu_reload), NULL }, ++ { "insn_emulation", VCPU_STAT(insn_emulation), NULL }, ++ { "insn_emulation_fail", VCPU_STAT(insn_emulation_fail), NULL }, ++ { "irq_injections", VCPU_STAT(irq_injections), NULL }, ++ { "nmi_injections", VCPU_STAT(nmi_injections), NULL }, ++ { "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped), NULL }, ++ { "mmu_pte_write", VM_STAT(mmu_pte_write), NULL }, ++ { "mmu_pte_updated", VM_STAT(mmu_pte_updated), NULL }, ++ { "mmu_pde_zapped", VM_STAT(mmu_pde_zapped), NULL }, ++ { "mmu_flooded", VM_STAT(mmu_flooded), NULL }, ++ { "mmu_recycled", VM_STAT(mmu_recycled), NULL }, ++ { "mmu_cache_miss", VM_STAT(mmu_cache_miss), NULL }, ++ { "mmu_unsync", VM_STAT(mmu_unsync), NULL }, ++ { "remote_tlb_flush", VM_STAT(remote_tlb_flush), NULL }, ++ { "largepages", VM_STAT(lpages), NULL }, + { NULL } + }; + +@@ -1485,7 +1485,7 @@ static int kvm_vcpu_ioctl_set_lapic(stru + static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, + struct kvm_interrupt *irq) + { +- if (irq->irq < 0 || irq->irq >= 256) ++ if (irq->irq >= 256) + return -EINVAL; + if (irqchip_in_kernel(vcpu->kvm)) + return -ENXIO; +@@ -2810,10 +2810,10 @@ static struct notifier_block kvmclock_cp + .notifier_call = kvmclock_cpufreq_notifier + }; + +-int kvm_arch_init(void *opaque) ++int kvm_arch_init(const void *opaque) + { + int r, cpu; +- struct kvm_x86_ops *ops = (struct kvm_x86_ops *)opaque; ++ const struct kvm_x86_ops *ops = (const struct kvm_x86_ops *)opaque; + + if (kvm_x86_ops) { + printk(KERN_ERR "kvm: already loaded the other module\n"); +diff -urNp linux-2.6.31.1/arch/x86/lib/checksum_32.S linux-2.6.31.1/arch/x86/lib/checksum_32.S +--- linux-2.6.31.1/arch/x86/lib/checksum_32.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/checksum_32.S 2009-10-01 20:12:42.000000000 -0400 +@@ -28,7 +28,8 @@ + #include + #include + #include +- ++#include ++ + /* + * computes a partial checksum, e.g. for TCP/UDP fragments + */ +@@ -304,9 +305,22 @@ unsigned int csum_partial_copy_generic ( + + #define ARGBASE 16 + #define FP 12 +- +-ENTRY(csum_partial_copy_generic) ++ ++ENTRY(csum_partial_copy_generic_to_user) + CFI_STARTPROC ++ pushl $(__USER_DS) ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %es ++ CFI_ADJUST_CFA_OFFSET -4 ++ jmp csum_partial_copy_generic ++ ++ENTRY(csum_partial_copy_generic_from_user) ++ pushl $(__USER_DS) ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %ds ++ CFI_ADJUST_CFA_OFFSET -4 ++ ++ENTRY(csum_partial_copy_generic) + subl $4,%esp + CFI_ADJUST_CFA_OFFSET 4 + pushl %edi +@@ -331,7 +345,7 @@ ENTRY(csum_partial_copy_generic) + jmp 4f + SRC(1: movw (%esi), %bx ) + addl $2, %esi +-DST( movw %bx, (%edi) ) ++DST( movw %bx, %es:(%edi) ) + addl $2, %edi + addw %bx, %ax + adcl $0, %eax +@@ -343,30 +357,30 @@ DST( movw %bx, (%edi) ) + SRC(1: movl (%esi), %ebx ) + SRC( movl 4(%esi), %edx ) + adcl %ebx, %eax +-DST( movl %ebx, (%edi) ) ++DST( movl %ebx, %es:(%edi) ) + adcl %edx, %eax +-DST( movl %edx, 4(%edi) ) ++DST( movl %edx, %es:4(%edi) ) + + SRC( movl 8(%esi), %ebx ) + SRC( movl 12(%esi), %edx ) + adcl %ebx, %eax +-DST( movl %ebx, 8(%edi) ) ++DST( movl %ebx, %es:8(%edi) ) + adcl %edx, %eax +-DST( movl %edx, 12(%edi) ) ++DST( movl %edx, %es:12(%edi) ) + + SRC( movl 16(%esi), %ebx ) + SRC( movl 20(%esi), %edx ) + adcl %ebx, %eax +-DST( movl %ebx, 16(%edi) ) ++DST( movl %ebx, %es:16(%edi) ) + adcl %edx, %eax +-DST( movl %edx, 20(%edi) ) ++DST( movl %edx, %es:20(%edi) ) + + SRC( movl 24(%esi), %ebx ) + SRC( movl 28(%esi), %edx ) + adcl %ebx, %eax +-DST( movl %ebx, 24(%edi) ) ++DST( movl %ebx, %es:24(%edi) ) + adcl %edx, %eax +-DST( movl %edx, 28(%edi) ) ++DST( movl %edx, %es:28(%edi) ) + + lea 32(%esi), %esi + lea 32(%edi), %edi +@@ -380,7 +394,7 @@ DST( movl %edx, 28(%edi) ) + shrl $2, %edx # This clears CF + SRC(3: movl (%esi), %ebx ) + adcl %ebx, %eax +-DST( movl %ebx, (%edi) ) ++DST( movl %ebx, %es:(%edi) ) + lea 4(%esi), %esi + lea 4(%edi), %edi + dec %edx +@@ -392,12 +406,12 @@ DST( movl %ebx, (%edi) ) + jb 5f + SRC( movw (%esi), %cx ) + leal 2(%esi), %esi +-DST( movw %cx, (%edi) ) ++DST( movw %cx, %es:(%edi) ) + leal 2(%edi), %edi + je 6f + shll $16,%ecx + SRC(5: movb (%esi), %cl ) +-DST( movb %cl, (%edi) ) ++DST( movb %cl, %es:(%edi) ) + 6: addl %ecx, %eax + adcl $0, %eax + 7: +@@ -408,7 +422,7 @@ DST( movb %cl, (%edi) ) + + 6001: + movl ARGBASE+20(%esp), %ebx # src_err_ptr +- movl $-EFAULT, (%ebx) ++ movl $-EFAULT, %ss:(%ebx) + + # zero the complete destination - computing the rest + # is too much work +@@ -421,11 +435,19 @@ DST( movb %cl, (%edi) ) + + 6002: + movl ARGBASE+24(%esp), %ebx # dst_err_ptr +- movl $-EFAULT,(%ebx) ++ movl $-EFAULT,%ss:(%ebx) + jmp 5000b + + .previous + ++ pushl %ss ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %ds ++ CFI_ADJUST_CFA_OFFSET -4 ++ pushl %ss ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %es ++ CFI_ADJUST_CFA_OFFSET -4 + popl %ebx + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE ebx +@@ -439,26 +461,41 @@ DST( movb %cl, (%edi) ) + CFI_ADJUST_CFA_OFFSET -4 + ret + CFI_ENDPROC +-ENDPROC(csum_partial_copy_generic) ++ENDPROC(csum_partial_copy_generic_to_user) + + #else + + /* Version for PentiumII/PPro */ + + #define ROUND1(x) \ ++ nop; nop; nop; \ + SRC(movl x(%esi), %ebx ) ; \ + addl %ebx, %eax ; \ +- DST(movl %ebx, x(%edi) ) ; ++ DST(movl %ebx, %es:x(%edi)) ; + + #define ROUND(x) \ ++ nop; nop; nop; \ + SRC(movl x(%esi), %ebx ) ; \ + adcl %ebx, %eax ; \ +- DST(movl %ebx, x(%edi) ) ; ++ DST(movl %ebx, %es:x(%edi)) ; + + #define ARGBASE 12 +- +-ENTRY(csum_partial_copy_generic) ++ ++ENTRY(csum_partial_copy_generic_to_user) + CFI_STARTPROC ++ pushl $(__USER_DS) ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %es ++ CFI_ADJUST_CFA_OFFSET -4 ++ jmp csum_partial_copy_generic ++ ++ENTRY(csum_partial_copy_generic_from_user) ++ pushl $(__USER_DS) ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %ds ++ CFI_ADJUST_CFA_OFFSET -4 ++ ++ENTRY(csum_partial_copy_generic) + pushl %ebx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ebx, 0 +@@ -482,7 +519,7 @@ ENTRY(csum_partial_copy_generic) + subl %ebx, %edi + lea -1(%esi),%edx + andl $-32,%edx +- lea 3f(%ebx,%ebx), %ebx ++ lea 3f(%ebx,%ebx,2), %ebx + testl %esi, %esi + jmp *%ebx + 1: addl $64,%esi +@@ -503,19 +540,19 @@ ENTRY(csum_partial_copy_generic) + jb 5f + SRC( movw (%esi), %dx ) + leal 2(%esi), %esi +-DST( movw %dx, (%edi) ) ++DST( movw %dx, %es:(%edi) ) + leal 2(%edi), %edi + je 6f + shll $16,%edx + 5: + SRC( movb (%esi), %dl ) +-DST( movb %dl, (%edi) ) ++DST( movb %dl, %es:(%edi) ) + 6: addl %edx, %eax + adcl $0, %eax + 7: + .section .fixup, "ax" + 6001: movl ARGBASE+20(%esp), %ebx # src_err_ptr +- movl $-EFAULT, (%ebx) ++ movl $-EFAULT, %ss:(%ebx) + # zero the complete destination (computing the rest is too much work) + movl ARGBASE+8(%esp),%edi # dst + movl ARGBASE+12(%esp),%ecx # len +@@ -523,10 +560,18 @@ DST( movb %dl, (%edi) ) + rep; stosb + jmp 7b + 6002: movl ARGBASE+24(%esp), %ebx # dst_err_ptr +- movl $-EFAULT, (%ebx) ++ movl $-EFAULT, %ss:(%ebx) + jmp 7b + .previous + ++ pushl %ss ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %ds ++ CFI_ADJUST_CFA_OFFSET -4 ++ pushl %ss ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %es ++ CFI_ADJUST_CFA_OFFSET -4 + popl %esi + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE esi +@@ -538,7 +583,7 @@ DST( movb %dl, (%edi) ) + CFI_RESTORE ebx + ret + CFI_ENDPROC +-ENDPROC(csum_partial_copy_generic) ++ENDPROC(csum_partial_copy_generic_to_user) + + #undef ROUND + #undef ROUND1 +diff -urNp linux-2.6.31.1/arch/x86/lib/clear_page_64.S linux-2.6.31.1/arch/x86/lib/clear_page_64.S +--- linux-2.6.31.1/arch/x86/lib/clear_page_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/clear_page_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -43,7 +43,7 @@ ENDPROC(clear_page) + + #include + +- .section .altinstr_replacement,"ax" ++ .section .altinstr_replacement,"a" + 1: .byte 0xeb /* jmp */ + .byte (clear_page_c - clear_page) - (2f - 1b) /* offset */ + 2: +diff -urNp linux-2.6.31.1/arch/x86/lib/copy_page_64.S linux-2.6.31.1/arch/x86/lib/copy_page_64.S +--- linux-2.6.31.1/arch/x86/lib/copy_page_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/copy_page_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -104,7 +104,7 @@ ENDPROC(copy_page) + + #include + +- .section .altinstr_replacement,"ax" ++ .section .altinstr_replacement,"a" + 1: .byte 0xeb /* jmp */ + .byte (copy_page_c - copy_page) - (2f - 1b) /* offset */ + 2: +diff -urNp linux-2.6.31.1/arch/x86/lib/copy_user_64.S linux-2.6.31.1/arch/x86/lib/copy_user_64.S +--- linux-2.6.31.1/arch/x86/lib/copy_user_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/copy_user_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -21,7 +21,7 @@ + .byte 0xe9 /* 32bit jump */ + .long \orig-1f /* by default jump to orig */ + 1: +- .section .altinstr_replacement,"ax" ++ .section .altinstr_replacement,"a" + 2: .byte 0xe9 /* near jump with 32bit immediate */ + .long \alt-1b /* offset */ /* or alternatively to alt */ + .previous +@@ -64,32 +64,6 @@ + #endif + .endm + +-/* Standard copy_to_user with segment limit checking */ +-ENTRY(copy_to_user) +- CFI_STARTPROC +- GET_THREAD_INFO(%rax) +- movq %rdi,%rcx +- addq %rdx,%rcx +- jc bad_to_user +- cmpq TI_addr_limit(%rax),%rcx +- jae bad_to_user +- ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string +- CFI_ENDPROC +-ENDPROC(copy_to_user) +- +-/* Standard copy_from_user with segment limit checking */ +-ENTRY(copy_from_user) +- CFI_STARTPROC +- GET_THREAD_INFO(%rax) +- movq %rsi,%rcx +- addq %rdx,%rcx +- jc bad_from_user +- cmpq TI_addr_limit(%rax),%rcx +- jae bad_from_user +- ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string +- CFI_ENDPROC +-ENDPROC(copy_from_user) +- + ENTRY(copy_user_generic) + CFI_STARTPROC + ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string +@@ -107,6 +81,8 @@ ENDPROC(__copy_from_user_inatomic) + ENTRY(bad_from_user) + bad_from_user: + CFI_STARTPROC ++ testl %edx,%edx ++ js bad_to_user + movl %edx,%ecx + xorl %eax,%eax + rep +diff -urNp linux-2.6.31.1/arch/x86/lib/getuser.S linux-2.6.31.1/arch/x86/lib/getuser.S +--- linux-2.6.31.1/arch/x86/lib/getuser.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/getuser.S 2009-10-01 20:12:42.000000000 -0400 +@@ -33,6 +33,7 @@ + #include + #include + #include ++#include + + .text + ENTRY(__get_user_1) +@@ -40,7 +41,19 @@ ENTRY(__get_user_1) + GET_THREAD_INFO(%_ASM_DX) + cmp TI_addr_limit(%_ASM_DX),%_ASM_AX + jae bad_get_user ++ ++#ifdef CONFIG_X86_32 ++ pushl $(__USER_DS) ++ popl %ds ++#endif ++ + 1: movzb (%_ASM_AX),%edx ++ ++#ifdef CONFIG_X86_32 ++ pushl %ss ++ pop %ds ++#endif ++ + xor %eax,%eax + ret + CFI_ENDPROC +@@ -53,7 +66,19 @@ ENTRY(__get_user_2) + GET_THREAD_INFO(%_ASM_DX) + cmp TI_addr_limit(%_ASM_DX),%_ASM_AX + jae bad_get_user ++ ++#ifdef CONFIG_X86_32 ++ pushl $(__USER_DS) ++ popl %ds ++#endif ++ + 2: movzwl -1(%_ASM_AX),%edx ++ ++#ifdef CONFIG_X86_32 ++ pushl %ss ++ pop %ds ++#endif ++ + xor %eax,%eax + ret + CFI_ENDPROC +@@ -66,7 +91,19 @@ ENTRY(__get_user_4) + GET_THREAD_INFO(%_ASM_DX) + cmp TI_addr_limit(%_ASM_DX),%_ASM_AX + jae bad_get_user ++ ++#ifdef CONFIG_X86_32 ++ pushl $(__USER_DS) ++ popl %ds ++#endif ++ + 3: mov -3(%_ASM_AX),%edx ++ ++#ifdef CONFIG_X86_32 ++ pushl %ss ++ pop %ds ++#endif ++ + xor %eax,%eax + ret + CFI_ENDPROC +@@ -89,6 +126,12 @@ ENDPROC(__get_user_8) + + bad_get_user: + CFI_STARTPROC ++ ++#ifdef CONFIG_X86_32 ++ pushl %ss ++ pop %ds ++#endif ++ + xor %edx,%edx + mov $(-EFAULT),%_ASM_AX + ret +diff -urNp linux-2.6.31.1/arch/x86/lib/memcpy_64.S linux-2.6.31.1/arch/x86/lib/memcpy_64.S +--- linux-2.6.31.1/arch/x86/lib/memcpy_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/memcpy_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -128,7 +128,7 @@ ENDPROC(__memcpy) + * It is also a lot simpler. Use this when possible: + */ + +- .section .altinstr_replacement, "ax" ++ .section .altinstr_replacement, "a" + 1: .byte 0xeb /* jmp */ + .byte (memcpy_c - memcpy) - (2f - 1b) /* offset */ + 2: +diff -urNp linux-2.6.31.1/arch/x86/lib/memset_64.S linux-2.6.31.1/arch/x86/lib/memset_64.S +--- linux-2.6.31.1/arch/x86/lib/memset_64.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/memset_64.S 2009-10-01 20:12:42.000000000 -0400 +@@ -118,7 +118,7 @@ ENDPROC(__memset) + + #include + +- .section .altinstr_replacement,"ax" ++ .section .altinstr_replacement,"a" + 1: .byte 0xeb /* jmp */ + .byte (memset_c - memset) - (2f - 1b) /* offset */ + 2: +diff -urNp linux-2.6.31.1/arch/x86/lib/mmx_32.c linux-2.6.31.1/arch/x86/lib/mmx_32.c +--- linux-2.6.31.1/arch/x86/lib/mmx_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/mmx_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -29,6 +29,7 @@ void *_mmx_memcpy(void *to, const void * + { + void *p; + int i; ++ unsigned long cr0; + + if (unlikely(in_interrupt())) + return __memcpy(to, from, len); +@@ -39,44 +40,72 @@ void *_mmx_memcpy(void *to, const void * + kernel_fpu_begin(); + + __asm__ __volatile__ ( +- "1: prefetch (%0)\n" /* This set is 28 bytes */ +- " prefetch 64(%0)\n" +- " prefetch 128(%0)\n" +- " prefetch 192(%0)\n" +- " prefetch 256(%0)\n" ++ "1: prefetch (%1)\n" /* This set is 28 bytes */ ++ " prefetch 64(%1)\n" ++ " prefetch 128(%1)\n" ++ " prefetch 192(%1)\n" ++ " prefetch 256(%1)\n" + "2: \n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ "3: \n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" + _ASM_EXTABLE(1b, 3b) +- : : "r" (from)); ++ : "=&r" (cr0) : "r" (from) : "ax"); + + for ( ; i > 5; i--) { + __asm__ __volatile__ ( +- "1: prefetch 320(%0)\n" +- "2: movq (%0), %%mm0\n" +- " movq 8(%0), %%mm1\n" +- " movq 16(%0), %%mm2\n" +- " movq 24(%0), %%mm3\n" +- " movq %%mm0, (%1)\n" +- " movq %%mm1, 8(%1)\n" +- " movq %%mm2, 16(%1)\n" +- " movq %%mm3, 24(%1)\n" +- " movq 32(%0), %%mm0\n" +- " movq 40(%0), %%mm1\n" +- " movq 48(%0), %%mm2\n" +- " movq 56(%0), %%mm3\n" +- " movq %%mm0, 32(%1)\n" +- " movq %%mm1, 40(%1)\n" +- " movq %%mm2, 48(%1)\n" +- " movq %%mm3, 56(%1)\n" ++ "1: prefetch 320(%1)\n" ++ "2: movq (%1), %%mm0\n" ++ " movq 8(%1), %%mm1\n" ++ " movq 16(%1), %%mm2\n" ++ " movq 24(%1), %%mm3\n" ++ " movq %%mm0, (%2)\n" ++ " movq %%mm1, 8(%2)\n" ++ " movq %%mm2, 16(%2)\n" ++ " movq %%mm3, 24(%2)\n" ++ " movq 32(%1), %%mm0\n" ++ " movq 40(%1), %%mm1\n" ++ " movq 48(%1), %%mm2\n" ++ " movq 56(%1), %%mm3\n" ++ " movq %%mm0, 32(%2)\n" ++ " movq %%mm1, 40(%2)\n" ++ " movq %%mm2, 48(%2)\n" ++ " movq %%mm3, 56(%2)\n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ "3:\n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" + _ASM_EXTABLE(1b, 3b) +- : : "r" (from), "r" (to) : "memory"); ++ : "=&r" (cr0) : "r" (from), "r" (to) : "memory", "ax"); + + from += 64; + to += 64; +@@ -158,6 +187,7 @@ static void fast_clear_page(void *page) + static void fast_copy_page(void *to, void *from) + { + int i; ++ unsigned long cr0; + + kernel_fpu_begin(); + +@@ -166,42 +196,70 @@ static void fast_copy_page(void *to, voi + * but that is for later. -AV + */ + __asm__ __volatile__( +- "1: prefetch (%0)\n" +- " prefetch 64(%0)\n" +- " prefetch 128(%0)\n" +- " prefetch 192(%0)\n" +- " prefetch 256(%0)\n" ++ "1: prefetch (%1)\n" ++ " prefetch 64(%1)\n" ++ " prefetch 128(%1)\n" ++ " prefetch 192(%1)\n" ++ " prefetch 256(%1)\n" + "2: \n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ "3: \n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" +- _ASM_EXTABLE(1b, 3b) : : "r" (from)); ++ _ASM_EXTABLE(1b, 3b) : "=&r" (cr0) : "r" (from) : "ax"); + + for (i = 0; i < (4096-320)/64; i++) { + __asm__ __volatile__ ( +- "1: prefetch 320(%0)\n" +- "2: movq (%0), %%mm0\n" +- " movntq %%mm0, (%1)\n" +- " movq 8(%0), %%mm1\n" +- " movntq %%mm1, 8(%1)\n" +- " movq 16(%0), %%mm2\n" +- " movntq %%mm2, 16(%1)\n" +- " movq 24(%0), %%mm3\n" +- " movntq %%mm3, 24(%1)\n" +- " movq 32(%0), %%mm4\n" +- " movntq %%mm4, 32(%1)\n" +- " movq 40(%0), %%mm5\n" +- " movntq %%mm5, 40(%1)\n" +- " movq 48(%0), %%mm6\n" +- " movntq %%mm6, 48(%1)\n" +- " movq 56(%0), %%mm7\n" +- " movntq %%mm7, 56(%1)\n" ++ "1: prefetch 320(%1)\n" ++ "2: movq (%1), %%mm0\n" ++ " movntq %%mm0, (%2)\n" ++ " movq 8(%1), %%mm1\n" ++ " movntq %%mm1, 8(%2)\n" ++ " movq 16(%1), %%mm2\n" ++ " movntq %%mm2, 16(%2)\n" ++ " movq 24(%1), %%mm3\n" ++ " movntq %%mm3, 24(%2)\n" ++ " movq 32(%1), %%mm4\n" ++ " movntq %%mm4, 32(%2)\n" ++ " movq 40(%1), %%mm5\n" ++ " movntq %%mm5, 40(%2)\n" ++ " movq 48(%1), %%mm6\n" ++ " movntq %%mm6, 48(%2)\n" ++ " movq 56(%1), %%mm7\n" ++ " movntq %%mm7, 56(%2)\n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ "3:\n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" +- _ASM_EXTABLE(1b, 3b) : : "r" (from), "r" (to) : "memory"); ++ _ASM_EXTABLE(1b, 3b) : "=&r" (cr0) : "r" (from), "r" (to) : "memory", "ax"); + + from += 64; + to += 64; +@@ -280,47 +338,76 @@ static void fast_clear_page(void *page) + static void fast_copy_page(void *to, void *from) + { + int i; ++ unsigned long cr0; + + kernel_fpu_begin(); + + __asm__ __volatile__ ( +- "1: prefetch (%0)\n" +- " prefetch 64(%0)\n" +- " prefetch 128(%0)\n" +- " prefetch 192(%0)\n" +- " prefetch 256(%0)\n" ++ "1: prefetch (%1)\n" ++ " prefetch 64(%1)\n" ++ " prefetch 128(%1)\n" ++ " prefetch 192(%1)\n" ++ " prefetch 256(%1)\n" + "2: \n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ "3: \n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" +- _ASM_EXTABLE(1b, 3b) : : "r" (from)); ++ _ASM_EXTABLE(1b, 3b) : "=&r" (cr0) : "r" (from) : "ax"); + + for (i = 0; i < 4096/64; i++) { + __asm__ __volatile__ ( +- "1: prefetch 320(%0)\n" +- "2: movq (%0), %%mm0\n" +- " movq 8(%0), %%mm1\n" +- " movq 16(%0), %%mm2\n" +- " movq 24(%0), %%mm3\n" +- " movq %%mm0, (%1)\n" +- " movq %%mm1, 8(%1)\n" +- " movq %%mm2, 16(%1)\n" +- " movq %%mm3, 24(%1)\n" +- " movq 32(%0), %%mm0\n" +- " movq 40(%0), %%mm1\n" +- " movq 48(%0), %%mm2\n" +- " movq 56(%0), %%mm3\n" +- " movq %%mm0, 32(%1)\n" +- " movq %%mm1, 40(%1)\n" +- " movq %%mm2, 48(%1)\n" +- " movq %%mm3, 56(%1)\n" ++ "1: prefetch 320(%1)\n" ++ "2: movq (%1), %%mm0\n" ++ " movq 8(%1), %%mm1\n" ++ " movq 16(%1), %%mm2\n" ++ " movq 24(%1), %%mm3\n" ++ " movq %%mm0, (%2)\n" ++ " movq %%mm1, 8(%2)\n" ++ " movq %%mm2, 16(%2)\n" ++ " movq %%mm3, 24(%2)\n" ++ " movq 32(%1), %%mm0\n" ++ " movq 40(%1), %%mm1\n" ++ " movq 48(%1), %%mm2\n" ++ " movq 56(%1), %%mm3\n" ++ " movq %%mm0, 32(%2)\n" ++ " movq %%mm1, 40(%2)\n" ++ " movq %%mm2, 48(%2)\n" ++ " movq %%mm3, 56(%2)\n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ "3:\n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" + _ASM_EXTABLE(1b, 3b) +- : : "r" (from), "r" (to) : "memory"); ++ : "=&r" (cr0) : "r" (from), "r" (to) : "memory", "ax"); + + from += 64; + to += 64; +diff -urNp linux-2.6.31.1/arch/x86/lib/putuser.S linux-2.6.31.1/arch/x86/lib/putuser.S +--- linux-2.6.31.1/arch/x86/lib/putuser.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/putuser.S 2009-10-01 20:12:42.000000000 -0400 +@@ -15,6 +15,7 @@ + #include + #include + #include ++#include + + + /* +@@ -39,7 +40,19 @@ ENTRY(__put_user_1) + ENTER + cmp TI_addr_limit(%_ASM_BX),%_ASM_CX + jae bad_put_user ++ ++#ifdef CONFIG_X86_32 ++ pushl $(__USER_DS) ++ popl %ds ++#endif ++ + 1: movb %al,(%_ASM_CX) ++ ++#ifdef CONFIG_X86_32 ++ pushl %ss ++ popl %ds ++#endif ++ + xor %eax,%eax + EXIT + ENDPROC(__put_user_1) +@@ -50,7 +63,19 @@ ENTRY(__put_user_2) + sub $1,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jae bad_put_user ++ ++#ifdef CONFIG_X86_32 ++ pushl $(__USER_DS) ++ popl %ds ++#endif ++ + 2: movw %ax,(%_ASM_CX) ++ ++#ifdef CONFIG_X86_32 ++ pushl %ss ++ popl %ds ++#endif ++ + xor %eax,%eax + EXIT + ENDPROC(__put_user_2) +@@ -61,7 +86,19 @@ ENTRY(__put_user_4) + sub $3,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jae bad_put_user ++ ++#ifdef CONFIG_X86_32 ++ pushl $(__USER_DS) ++ popl %ds ++#endif ++ + 3: movl %eax,(%_ASM_CX) ++ ++#ifdef CONFIG_X86_32 ++ pushl %ss ++ popl %ds ++#endif ++ + xor %eax,%eax + EXIT + ENDPROC(__put_user_4) +@@ -72,16 +109,34 @@ ENTRY(__put_user_8) + sub $7,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jae bad_put_user ++ ++#ifdef CONFIG_X86_32 ++ pushl $(__USER_DS) ++ popl %ds ++#endif ++ + 4: mov %_ASM_AX,(%_ASM_CX) + #ifdef CONFIG_X86_32 + 5: movl %edx,4(%_ASM_CX) + #endif ++ ++#ifdef CONFIG_X86_32 ++ pushl %ss ++ popl %ds ++#endif ++ + xor %eax,%eax + EXIT + ENDPROC(__put_user_8) + + bad_put_user: + CFI_STARTPROC ++ ++#ifdef CONFIG_X86_32 ++ pushl %ss ++ popl %ds ++#endif ++ + movl $-EFAULT,%eax + EXIT + END(bad_put_user) +diff -urNp linux-2.6.31.1/arch/x86/lib/usercopy_32.c linux-2.6.31.1/arch/x86/lib/usercopy_32.c +--- linux-2.6.31.1/arch/x86/lib/usercopy_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/lib/usercopy_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -36,31 +36,38 @@ static inline int __movsl_is_ok(unsigned + * Copy a null terminated string from userspace. + */ + +-#define __do_strncpy_from_user(dst, src, count, res) \ +-do { \ +- int __d0, __d1, __d2; \ +- might_fault(); \ +- __asm__ __volatile__( \ +- " testl %1,%1\n" \ +- " jz 2f\n" \ +- "0: lodsb\n" \ +- " stosb\n" \ +- " testb %%al,%%al\n" \ +- " jz 1f\n" \ +- " decl %1\n" \ +- " jnz 0b\n" \ +- "1: subl %1,%0\n" \ +- "2:\n" \ +- ".section .fixup,\"ax\"\n" \ +- "3: movl %5,%0\n" \ +- " jmp 2b\n" \ +- ".previous\n" \ +- _ASM_EXTABLE(0b,3b) \ +- : "=&d"(res), "=&c"(count), "=&a" (__d0), "=&S" (__d1), \ +- "=&D" (__d2) \ +- : "i"(-EFAULT), "0"(count), "1"(count), "3"(src), "4"(dst) \ +- : "memory"); \ +-} while (0) ++static long __do_strncpy_from_user(char *dst, const char __user *src, long count) ++{ ++ int __d0, __d1, __d2; ++ long res = -EFAULT; ++ ++ might_fault(); ++ __asm__ __volatile__( ++ " movw %w10,%%ds\n" ++ " testl %1,%1\n" ++ " jz 2f\n" ++ "0: lodsb\n" ++ " stosb\n" ++ " testb %%al,%%al\n" ++ " jz 1f\n" ++ " decl %1\n" ++ " jnz 0b\n" ++ "1: subl %1,%0\n" ++ "2:\n" ++ " pushl %%ss\n" ++ " popl %%ds\n" ++ ".section .fixup,\"ax\"\n" ++ "3: movl %5,%0\n" ++ " jmp 2b\n" ++ ".previous\n" ++ _ASM_EXTABLE(0b,3b) ++ : "=&d"(res), "=&c"(count), "=&a" (__d0), "=&S" (__d1), ++ "=&D" (__d2) ++ : "i"(-EFAULT), "0"(count), "1"(count), "3"(src), "4"(dst), ++ "r"(__USER_DS) ++ : "memory"); ++ return res; ++} + + /** + * __strncpy_from_user: - Copy a NUL terminated string from userspace, with less checking. +@@ -85,9 +92,7 @@ do { \ + long + __strncpy_from_user(char *dst, const char __user *src, long count) + { +- long res; +- __do_strncpy_from_user(dst, src, count, res); +- return res; ++ return __do_strncpy_from_user(dst, src, count); + } + EXPORT_SYMBOL(__strncpy_from_user); + +@@ -114,7 +119,7 @@ strncpy_from_user(char *dst, const char + { + long res = -EFAULT; + if (access_ok(VERIFY_READ, src, 1)) +- __do_strncpy_from_user(dst, src, count, res); ++ res = __do_strncpy_from_user(dst, src, count); + return res; + } + EXPORT_SYMBOL(strncpy_from_user); +@@ -123,24 +128,30 @@ EXPORT_SYMBOL(strncpy_from_user); + * Zero Userspace + */ + +-#define __do_clear_user(addr,size) \ +-do { \ +- int __d0; \ +- might_fault(); \ +- __asm__ __volatile__( \ +- "0: rep; stosl\n" \ +- " movl %2,%0\n" \ +- "1: rep; stosb\n" \ +- "2:\n" \ +- ".section .fixup,\"ax\"\n" \ +- "3: lea 0(%2,%0,4),%0\n" \ +- " jmp 2b\n" \ +- ".previous\n" \ +- _ASM_EXTABLE(0b,3b) \ +- _ASM_EXTABLE(1b,2b) \ +- : "=&c"(size), "=&D" (__d0) \ +- : "r"(size & 3), "0"(size / 4), "1"(addr), "a"(0)); \ +-} while (0) ++static unsigned long __do_clear_user(void __user *addr, unsigned long size) ++{ ++ int __d0; ++ ++ might_fault(); ++ __asm__ __volatile__( ++ " movw %w6,%%es\n" ++ "0: rep; stosl\n" ++ " movl %2,%0\n" ++ "1: rep; stosb\n" ++ "2:\n" ++ " pushl %%ss\n" ++ " popl %%es\n" ++ ".section .fixup,\"ax\"\n" ++ "3: lea 0(%2,%0,4),%0\n" ++ " jmp 2b\n" ++ ".previous\n" ++ _ASM_EXTABLE(0b,3b) ++ _ASM_EXTABLE(1b,2b) ++ : "=&c"(size), "=&D" (__d0) ++ : "r"(size & 3), "0"(size / 4), "1"(addr), "a"(0), ++ "r"(__USER_DS)); ++ return size; ++} + + /** + * clear_user: - Zero a block of memory in user space. +@@ -157,7 +168,7 @@ clear_user(void __user *to, unsigned lon + { + might_fault(); + if (access_ok(VERIFY_WRITE, to, n)) +- __do_clear_user(to, n); ++ n = __do_clear_user(to, n); + return n; + } + EXPORT_SYMBOL(clear_user); +@@ -176,8 +187,7 @@ EXPORT_SYMBOL(clear_user); + unsigned long + __clear_user(void __user *to, unsigned long n) + { +- __do_clear_user(to, n); +- return n; ++ return __do_clear_user(to, n); + } + EXPORT_SYMBOL(__clear_user); + +@@ -200,14 +210,17 @@ long strnlen_user(const char __user *s, + might_fault(); + + __asm__ __volatile__( ++ " movw %w8,%%es\n" + " testl %0, %0\n" + " jz 3f\n" +- " andl %0,%%ecx\n" ++ " movl %0,%%ecx\n" + "0: repne; scasb\n" + " setne %%al\n" + " subl %%ecx,%0\n" + " addl %0,%%eax\n" + "1:\n" ++ " pushl %%ss\n" ++ " popl %%es\n" + ".section .fixup,\"ax\"\n" + "2: xorl %%eax,%%eax\n" + " jmp 1b\n" +@@ -219,7 +232,7 @@ long strnlen_user(const char __user *s, + " .long 0b,2b\n" + ".previous" + :"=&r" (n), "=&D" (s), "=&a" (res), "=&c" (tmp) +- :"0" (n), "1" (s), "2" (0), "3" (mask) ++ :"0" (n), "1" (s), "2" (0), "3" (mask), "r" (__USER_DS) + :"cc"); + return res & mask; + } +@@ -227,10 +240,121 @@ EXPORT_SYMBOL(strnlen_user); + + #ifdef CONFIG_X86_INTEL_USERCOPY + static unsigned long +-__copy_user_intel(void __user *to, const void *from, unsigned long size) ++__generic_copy_to_user_intel(void __user *to, const void *from, unsigned long size) ++{ ++ int d0, d1; ++ __asm__ __volatile__( ++ " movw %w6, %%es\n" ++ " .align 2,0x90\n" ++ "1: movl 32(%4), %%eax\n" ++ " cmpl $67, %0\n" ++ " jbe 3f\n" ++ "2: movl 64(%4), %%eax\n" ++ " .align 2,0x90\n" ++ "3: movl 0(%4), %%eax\n" ++ "4: movl 4(%4), %%edx\n" ++ "5: movl %%eax, %%es:0(%3)\n" ++ "6: movl %%edx, %%es:4(%3)\n" ++ "7: movl 8(%4), %%eax\n" ++ "8: movl 12(%4),%%edx\n" ++ "9: movl %%eax, %%es:8(%3)\n" ++ "10: movl %%edx, %%es:12(%3)\n" ++ "11: movl 16(%4), %%eax\n" ++ "12: movl 20(%4), %%edx\n" ++ "13: movl %%eax, %%es:16(%3)\n" ++ "14: movl %%edx, %%es:20(%3)\n" ++ "15: movl 24(%4), %%eax\n" ++ "16: movl 28(%4), %%edx\n" ++ "17: movl %%eax, %%es:24(%3)\n" ++ "18: movl %%edx, %%es:28(%3)\n" ++ "19: movl 32(%4), %%eax\n" ++ "20: movl 36(%4), %%edx\n" ++ "21: movl %%eax, %%es:32(%3)\n" ++ "22: movl %%edx, %%es:36(%3)\n" ++ "23: movl 40(%4), %%eax\n" ++ "24: movl 44(%4), %%edx\n" ++ "25: movl %%eax, %%es:40(%3)\n" ++ "26: movl %%edx, %%es:44(%3)\n" ++ "27: movl 48(%4), %%eax\n" ++ "28: movl 52(%4), %%edx\n" ++ "29: movl %%eax, %%es:48(%3)\n" ++ "30: movl %%edx, %%es:52(%3)\n" ++ "31: movl 56(%4), %%eax\n" ++ "32: movl 60(%4), %%edx\n" ++ "33: movl %%eax, %%es:56(%3)\n" ++ "34: movl %%edx, %%es:60(%3)\n" ++ " addl $-64, %0\n" ++ " addl $64, %4\n" ++ " addl $64, %3\n" ++ " cmpl $63, %0\n" ++ " ja 1b\n" ++ "35: movl %0, %%eax\n" ++ " shrl $2, %0\n" ++ " andl $3, %%eax\n" ++ " cld\n" ++ "99: rep; movsl\n" ++ "36: movl %%eax, %0\n" ++ "37: rep; movsb\n" ++ "100:\n" ++ " pushl %%ss\n" ++ " popl %%es\n" ++ ".section .fixup,\"ax\"\n" ++ "101: lea 0(%%eax,%0,4),%0\n" ++ " jmp 100b\n" ++ ".previous\n" ++ ".section __ex_table,\"a\"\n" ++ " .align 4\n" ++ " .long 1b,100b\n" ++ " .long 2b,100b\n" ++ " .long 3b,100b\n" ++ " .long 4b,100b\n" ++ " .long 5b,100b\n" ++ " .long 6b,100b\n" ++ " .long 7b,100b\n" ++ " .long 8b,100b\n" ++ " .long 9b,100b\n" ++ " .long 10b,100b\n" ++ " .long 11b,100b\n" ++ " .long 12b,100b\n" ++ " .long 13b,100b\n" ++ " .long 14b,100b\n" ++ " .long 15b,100b\n" ++ " .long 16b,100b\n" ++ " .long 17b,100b\n" ++ " .long 18b,100b\n" ++ " .long 19b,100b\n" ++ " .long 20b,100b\n" ++ " .long 21b,100b\n" ++ " .long 22b,100b\n" ++ " .long 23b,100b\n" ++ " .long 24b,100b\n" ++ " .long 25b,100b\n" ++ " .long 26b,100b\n" ++ " .long 27b,100b\n" ++ " .long 28b,100b\n" ++ " .long 29b,100b\n" ++ " .long 30b,100b\n" ++ " .long 31b,100b\n" ++ " .long 32b,100b\n" ++ " .long 33b,100b\n" ++ " .long 34b,100b\n" ++ " .long 35b,100b\n" ++ " .long 36b,100b\n" ++ " .long 37b,100b\n" ++ " .long 99b,101b\n" ++ ".previous" ++ : "=&c"(size), "=&D" (d0), "=&S" (d1) ++ : "1"(to), "2"(from), "0"(size), "r"(__USER_DS) ++ : "eax", "edx", "memory"); ++ return size; ++} ++ ++static unsigned long ++__generic_copy_from_user_intel(void *to, const void __user *from, unsigned long size) + { + int d0, d1; + __asm__ __volatile__( ++ " movw %w6, %%ds\n" + " .align 2,0x90\n" + "1: movl 32(%4), %%eax\n" + " cmpl $67, %0\n" +@@ -239,36 +363,36 @@ __copy_user_intel(void __user *to, const + " .align 2,0x90\n" + "3: movl 0(%4), %%eax\n" + "4: movl 4(%4), %%edx\n" +- "5: movl %%eax, 0(%3)\n" +- "6: movl %%edx, 4(%3)\n" ++ "5: movl %%eax, %%es:0(%3)\n" ++ "6: movl %%edx, %%es:4(%3)\n" + "7: movl 8(%4), %%eax\n" + "8: movl 12(%4),%%edx\n" +- "9: movl %%eax, 8(%3)\n" +- "10: movl %%edx, 12(%3)\n" ++ "9: movl %%eax, %%es:8(%3)\n" ++ "10: movl %%edx, %%es:12(%3)\n" + "11: movl 16(%4), %%eax\n" + "12: movl 20(%4), %%edx\n" +- "13: movl %%eax, 16(%3)\n" +- "14: movl %%edx, 20(%3)\n" ++ "13: movl %%eax, %%es:16(%3)\n" ++ "14: movl %%edx, %%es:20(%3)\n" + "15: movl 24(%4), %%eax\n" + "16: movl 28(%4), %%edx\n" +- "17: movl %%eax, 24(%3)\n" +- "18: movl %%edx, 28(%3)\n" ++ "17: movl %%eax, %%es:24(%3)\n" ++ "18: movl %%edx, %%es:28(%3)\n" + "19: movl 32(%4), %%eax\n" + "20: movl 36(%4), %%edx\n" +- "21: movl %%eax, 32(%3)\n" +- "22: movl %%edx, 36(%3)\n" ++ "21: movl %%eax, %%es:32(%3)\n" ++ "22: movl %%edx, %%es:36(%3)\n" + "23: movl 40(%4), %%eax\n" + "24: movl 44(%4), %%edx\n" +- "25: movl %%eax, 40(%3)\n" +- "26: movl %%edx, 44(%3)\n" ++ "25: movl %%eax, %%es:40(%3)\n" ++ "26: movl %%edx, %%es:44(%3)\n" + "27: movl 48(%4), %%eax\n" + "28: movl 52(%4), %%edx\n" +- "29: movl %%eax, 48(%3)\n" +- "30: movl %%edx, 52(%3)\n" ++ "29: movl %%eax, %%es:48(%3)\n" ++ "30: movl %%edx, %%es:52(%3)\n" + "31: movl 56(%4), %%eax\n" + "32: movl 60(%4), %%edx\n" +- "33: movl %%eax, 56(%3)\n" +- "34: movl %%edx, 60(%3)\n" ++ "33: movl %%eax, %%es:56(%3)\n" ++ "34: movl %%edx, %%es:60(%3)\n" + " addl $-64, %0\n" + " addl $64, %4\n" + " addl $64, %3\n" +@@ -282,6 +406,8 @@ __copy_user_intel(void __user *to, const + "36: movl %%eax, %0\n" + "37: rep; movsb\n" + "100:\n" ++ " pushl %%ss\n" ++ " popl %%ds\n" + ".section .fixup,\"ax\"\n" + "101: lea 0(%%eax,%0,4),%0\n" + " jmp 100b\n" +@@ -328,7 +454,7 @@ __copy_user_intel(void __user *to, const + " .long 99b,101b\n" + ".previous" + : "=&c"(size), "=&D" (d0), "=&S" (d1) +- : "1"(to), "2"(from), "0"(size) ++ : "1"(to), "2"(from), "0"(size), "r"(__USER_DS) + : "eax", "edx", "memory"); + return size; + } +@@ -338,6 +464,7 @@ __copy_user_zeroing_intel(void *to, cons + { + int d0, d1; + __asm__ __volatile__( ++ " movw %w6, %%ds\n" + " .align 2,0x90\n" + "0: movl 32(%4), %%eax\n" + " cmpl $67, %0\n" +@@ -346,36 +473,36 @@ __copy_user_zeroing_intel(void *to, cons + " .align 2,0x90\n" + "2: movl 0(%4), %%eax\n" + "21: movl 4(%4), %%edx\n" +- " movl %%eax, 0(%3)\n" +- " movl %%edx, 4(%3)\n" ++ " movl %%eax, %%es:0(%3)\n" ++ " movl %%edx, %%es:4(%3)\n" + "3: movl 8(%4), %%eax\n" + "31: movl 12(%4),%%edx\n" +- " movl %%eax, 8(%3)\n" +- " movl %%edx, 12(%3)\n" ++ " movl %%eax, %%es:8(%3)\n" ++ " movl %%edx, %%es:12(%3)\n" + "4: movl 16(%4), %%eax\n" + "41: movl 20(%4), %%edx\n" +- " movl %%eax, 16(%3)\n" +- " movl %%edx, 20(%3)\n" ++ " movl %%eax, %%es:16(%3)\n" ++ " movl %%edx, %%es:20(%3)\n" + "10: movl 24(%4), %%eax\n" + "51: movl 28(%4), %%edx\n" +- " movl %%eax, 24(%3)\n" +- " movl %%edx, 28(%3)\n" ++ " movl %%eax, %%es:24(%3)\n" ++ " movl %%edx, %%es:28(%3)\n" + "11: movl 32(%4), %%eax\n" + "61: movl 36(%4), %%edx\n" +- " movl %%eax, 32(%3)\n" +- " movl %%edx, 36(%3)\n" ++ " movl %%eax, %%es:32(%3)\n" ++ " movl %%edx, %%es:36(%3)\n" + "12: movl 40(%4), %%eax\n" + "71: movl 44(%4), %%edx\n" +- " movl %%eax, 40(%3)\n" +- " movl %%edx, 44(%3)\n" ++ " movl %%eax, %%es:40(%3)\n" ++ " movl %%edx, %%es:44(%3)\n" + "13: movl 48(%4), %%eax\n" + "81: movl 52(%4), %%edx\n" +- " movl %%eax, 48(%3)\n" +- " movl %%edx, 52(%3)\n" ++ " movl %%eax, %%es:48(%3)\n" ++ " movl %%edx, %%es:52(%3)\n" + "14: movl 56(%4), %%eax\n" + "91: movl 60(%4), %%edx\n" +- " movl %%eax, 56(%3)\n" +- " movl %%edx, 60(%3)\n" ++ " movl %%eax, %%es:56(%3)\n" ++ " movl %%edx, %%es:60(%3)\n" + " addl $-64, %0\n" + " addl $64, %4\n" + " addl $64, %3\n" +@@ -389,6 +516,8 @@ __copy_user_zeroing_intel(void *to, cons + " movl %%eax,%0\n" + "7: rep; movsb\n" + "8:\n" ++ " pushl %%ss\n" ++ " popl %%ds\n" + ".section .fixup,\"ax\"\n" + "9: lea 0(%%eax,%0,4),%0\n" + "16: pushl %0\n" +@@ -423,7 +552,7 @@ __copy_user_zeroing_intel(void *to, cons + " .long 7b,16b\n" + ".previous" + : "=&c"(size), "=&D" (d0), "=&S" (d1) +- : "1"(to), "2"(from), "0"(size) ++ : "1"(to), "2"(from), "0"(size), "r"(__USER_DS) + : "eax", "edx", "memory"); + return size; + } +@@ -439,6 +568,7 @@ static unsigned long __copy_user_zeroing + int d0, d1; + + __asm__ __volatile__( ++ " movw %w6, %%ds\n" + " .align 2,0x90\n" + "0: movl 32(%4), %%eax\n" + " cmpl $67, %0\n" +@@ -447,36 +577,36 @@ static unsigned long __copy_user_zeroing + " .align 2,0x90\n" + "2: movl 0(%4), %%eax\n" + "21: movl 4(%4), %%edx\n" +- " movnti %%eax, 0(%3)\n" +- " movnti %%edx, 4(%3)\n" ++ " movnti %%eax, %%es:0(%3)\n" ++ " movnti %%edx, %%es:4(%3)\n" + "3: movl 8(%4), %%eax\n" + "31: movl 12(%4),%%edx\n" +- " movnti %%eax, 8(%3)\n" +- " movnti %%edx, 12(%3)\n" ++ " movnti %%eax, %%es:8(%3)\n" ++ " movnti %%edx, %%es:12(%3)\n" + "4: movl 16(%4), %%eax\n" + "41: movl 20(%4), %%edx\n" +- " movnti %%eax, 16(%3)\n" +- " movnti %%edx, 20(%3)\n" ++ " movnti %%eax, %%es:16(%3)\n" ++ " movnti %%edx, %%es:20(%3)\n" + "10: movl 24(%4), %%eax\n" + "51: movl 28(%4), %%edx\n" +- " movnti %%eax, 24(%3)\n" +- " movnti %%edx, 28(%3)\n" ++ " movnti %%eax, %%es:24(%3)\n" ++ " movnti %%edx, %%es:28(%3)\n" + "11: movl 32(%4), %%eax\n" + "61: movl 36(%4), %%edx\n" +- " movnti %%eax, 32(%3)\n" +- " movnti %%edx, 36(%3)\n" ++ " movnti %%eax, %%es:32(%3)\n" ++ " movnti %%edx, %%es:36(%3)\n" + "12: movl 40(%4), %%eax\n" + "71: movl 44(%4), %%edx\n" +- " movnti %%eax, 40(%3)\n" +- " movnti %%edx, 44(%3)\n" ++ " movnti %%eax, %%es:40(%3)\n" ++ " movnti %%edx, %%es:44(%3)\n" + "13: movl 48(%4), %%eax\n" + "81: movl 52(%4), %%edx\n" +- " movnti %%eax, 48(%3)\n" +- " movnti %%edx, 52(%3)\n" ++ " movnti %%eax, %%es:48(%3)\n" ++ " movnti %%edx, %%es:52(%3)\n" + "14: movl 56(%4), %%eax\n" + "91: movl 60(%4), %%edx\n" +- " movnti %%eax, 56(%3)\n" +- " movnti %%edx, 60(%3)\n" ++ " movnti %%eax, %%es:56(%3)\n" ++ " movnti %%edx, %%es:60(%3)\n" + " addl $-64, %0\n" + " addl $64, %4\n" + " addl $64, %3\n" +@@ -491,6 +621,8 @@ static unsigned long __copy_user_zeroing + " movl %%eax,%0\n" + "7: rep; movsb\n" + "8:\n" ++ " pushl %%ss\n" ++ " popl %%ds\n" + ".section .fixup,\"ax\"\n" + "9: lea 0(%%eax,%0,4),%0\n" + "16: pushl %0\n" +@@ -525,7 +657,7 @@ static unsigned long __copy_user_zeroing + " .long 7b,16b\n" + ".previous" + : "=&c"(size), "=&D" (d0), "=&S" (d1) +- : "1"(to), "2"(from), "0"(size) ++ : "1"(to), "2"(from), "0"(size), "r"(__USER_DS) + : "eax", "edx", "memory"); + return size; + } +@@ -536,6 +668,7 @@ static unsigned long __copy_user_intel_n + int d0, d1; + + __asm__ __volatile__( ++ " movw %w6, %%ds\n" + " .align 2,0x90\n" + "0: movl 32(%4), %%eax\n" + " cmpl $67, %0\n" +@@ -544,36 +677,36 @@ static unsigned long __copy_user_intel_n + " .align 2,0x90\n" + "2: movl 0(%4), %%eax\n" + "21: movl 4(%4), %%edx\n" +- " movnti %%eax, 0(%3)\n" +- " movnti %%edx, 4(%3)\n" ++ " movnti %%eax, %%es:0(%3)\n" ++ " movnti %%edx, %%es:4(%3)\n" + "3: movl 8(%4), %%eax\n" + "31: movl 12(%4),%%edx\n" +- " movnti %%eax, 8(%3)\n" +- " movnti %%edx, 12(%3)\n" ++ " movnti %%eax, %%es:8(%3)\n" ++ " movnti %%edx, %%es:12(%3)\n" + "4: movl 16(%4), %%eax\n" + "41: movl 20(%4), %%edx\n" +- " movnti %%eax, 16(%3)\n" +- " movnti %%edx, 20(%3)\n" ++ " movnti %%eax, %%es:16(%3)\n" ++ " movnti %%edx, %%es:20(%3)\n" + "10: movl 24(%4), %%eax\n" + "51: movl 28(%4), %%edx\n" +- " movnti %%eax, 24(%3)\n" +- " movnti %%edx, 28(%3)\n" ++ " movnti %%eax, %%es:24(%3)\n" ++ " movnti %%edx, %%es:28(%3)\n" + "11: movl 32(%4), %%eax\n" + "61: movl 36(%4), %%edx\n" +- " movnti %%eax, 32(%3)\n" +- " movnti %%edx, 36(%3)\n" ++ " movnti %%eax, %%es:32(%3)\n" ++ " movnti %%edx, %%es:36(%3)\n" + "12: movl 40(%4), %%eax\n" + "71: movl 44(%4), %%edx\n" +- " movnti %%eax, 40(%3)\n" +- " movnti %%edx, 44(%3)\n" ++ " movnti %%eax, %%es:40(%3)\n" ++ " movnti %%edx, %%es:44(%3)\n" + "13: movl 48(%4), %%eax\n" + "81: movl 52(%4), %%edx\n" +- " movnti %%eax, 48(%3)\n" +- " movnti %%edx, 52(%3)\n" ++ " movnti %%eax, %%es:48(%3)\n" ++ " movnti %%edx, %%es:52(%3)\n" + "14: movl 56(%4), %%eax\n" + "91: movl 60(%4), %%edx\n" +- " movnti %%eax, 56(%3)\n" +- " movnti %%edx, 60(%3)\n" ++ " movnti %%eax, %%es:56(%3)\n" ++ " movnti %%edx, %%es:60(%3)\n" + " addl $-64, %0\n" + " addl $64, %4\n" + " addl $64, %3\n" +@@ -588,6 +721,8 @@ static unsigned long __copy_user_intel_n + " movl %%eax,%0\n" + "7: rep; movsb\n" + "8:\n" ++ " pushl %%ss\n" ++ " popl %%ds\n" + ".section .fixup,\"ax\"\n" + "9: lea 0(%%eax,%0,4),%0\n" + "16: jmp 8b\n" +@@ -616,7 +751,7 @@ static unsigned long __copy_user_intel_n + " .long 7b,16b\n" + ".previous" + : "=&c"(size), "=&D" (d0), "=&S" (d1) +- : "1"(to), "2"(from), "0"(size) ++ : "1"(to), "2"(from), "0"(size), "r"(__USER_DS) + : "eax", "edx", "memory"); + return size; + } +@@ -629,90 +764,146 @@ static unsigned long __copy_user_intel_n + */ + unsigned long __copy_user_zeroing_intel(void *to, const void __user *from, + unsigned long size); +-unsigned long __copy_user_intel(void __user *to, const void *from, ++unsigned long __generic_copy_to_user_intel(void __user *to, const void *from, ++ unsigned long size); ++unsigned long __generic_copy_from_user_intel(void *to, const void __user *from, + unsigned long size); + unsigned long __copy_user_zeroing_intel_nocache(void *to, + const void __user *from, unsigned long size); + #endif /* CONFIG_X86_INTEL_USERCOPY */ + + /* Generic arbitrary sized copy. */ +-#define __copy_user(to, from, size) \ +-do { \ +- int __d0, __d1, __d2; \ +- __asm__ __volatile__( \ +- " cmp $7,%0\n" \ +- " jbe 1f\n" \ +- " movl %1,%0\n" \ +- " negl %0\n" \ +- " andl $7,%0\n" \ +- " subl %0,%3\n" \ +- "4: rep; movsb\n" \ +- " movl %3,%0\n" \ +- " shrl $2,%0\n" \ +- " andl $3,%3\n" \ +- " .align 2,0x90\n" \ +- "0: rep; movsl\n" \ +- " movl %3,%0\n" \ +- "1: rep; movsb\n" \ +- "2:\n" \ +- ".section .fixup,\"ax\"\n" \ +- "5: addl %3,%0\n" \ +- " jmp 2b\n" \ +- "3: lea 0(%3,%0,4),%0\n" \ +- " jmp 2b\n" \ +- ".previous\n" \ +- ".section __ex_table,\"a\"\n" \ +- " .align 4\n" \ +- " .long 4b,5b\n" \ +- " .long 0b,3b\n" \ +- " .long 1b,2b\n" \ +- ".previous" \ +- : "=&c"(size), "=&D" (__d0), "=&S" (__d1), "=r"(__d2) \ +- : "3"(size), "0"(size), "1"(to), "2"(from) \ +- : "memory"); \ +-} while (0) +- +-#define __copy_user_zeroing(to, from, size) \ +-do { \ +- int __d0, __d1, __d2; \ +- __asm__ __volatile__( \ +- " cmp $7,%0\n" \ +- " jbe 1f\n" \ +- " movl %1,%0\n" \ +- " negl %0\n" \ +- " andl $7,%0\n" \ +- " subl %0,%3\n" \ +- "4: rep; movsb\n" \ +- " movl %3,%0\n" \ +- " shrl $2,%0\n" \ +- " andl $3,%3\n" \ +- " .align 2,0x90\n" \ +- "0: rep; movsl\n" \ +- " movl %3,%0\n" \ +- "1: rep; movsb\n" \ +- "2:\n" \ +- ".section .fixup,\"ax\"\n" \ +- "5: addl %3,%0\n" \ +- " jmp 6f\n" \ +- "3: lea 0(%3,%0,4),%0\n" \ +- "6: pushl %0\n" \ +- " pushl %%eax\n" \ +- " xorl %%eax,%%eax\n" \ +- " rep; stosb\n" \ +- " popl %%eax\n" \ +- " popl %0\n" \ +- " jmp 2b\n" \ +- ".previous\n" \ +- ".section __ex_table,\"a\"\n" \ +- " .align 4\n" \ +- " .long 4b,5b\n" \ +- " .long 0b,3b\n" \ +- " .long 1b,6b\n" \ +- ".previous" \ +- : "=&c"(size), "=&D" (__d0), "=&S" (__d1), "=r"(__d2) \ +- : "3"(size), "0"(size), "1"(to), "2"(from) \ +- : "memory"); \ +-} while (0) ++static unsigned long ++__generic_copy_to_user(void __user *to, const void *from, unsigned long size) ++{ ++ int __d0, __d1, __d2; ++ ++ __asm__ __volatile__( ++ " movw %w8,%%es\n" ++ " cmp $7,%0\n" ++ " jbe 1f\n" ++ " movl %1,%0\n" ++ " negl %0\n" ++ " andl $7,%0\n" ++ " subl %0,%3\n" ++ "4: rep; movsb\n" ++ " movl %3,%0\n" ++ " shrl $2,%0\n" ++ " andl $3,%3\n" ++ " .align 2,0x90\n" ++ "0: rep; movsl\n" ++ " movl %3,%0\n" ++ "1: rep; movsb\n" ++ "2:\n" ++ " pushl %%ss\n" ++ " popl %%es\n" ++ ".section .fixup,\"ax\"\n" ++ "5: addl %3,%0\n" ++ " jmp 2b\n" ++ "3: lea 0(%3,%0,4),%0\n" ++ " jmp 2b\n" ++ ".previous\n" ++ ".section __ex_table,\"a\"\n" ++ " .align 4\n" ++ " .long 4b,5b\n" ++ " .long 0b,3b\n" ++ " .long 1b,2b\n" ++ ".previous" ++ : "=&c"(size), "=&D" (__d0), "=&S" (__d1), "=r"(__d2) ++ : "3"(size), "0"(size), "1"(to), "2"(from), "r"(__USER_DS) ++ : "memory"); ++ return size; ++} ++ ++static unsigned long ++__generic_copy_from_user(void *to, const void __user *from, unsigned long size) ++{ ++ int __d0, __d1, __d2; ++ ++ __asm__ __volatile__( ++ " movw %w8,%%ds\n" ++ " cmp $7,%0\n" ++ " jbe 1f\n" ++ " movl %1,%0\n" ++ " negl %0\n" ++ " andl $7,%0\n" ++ " subl %0,%3\n" ++ "4: rep; movsb\n" ++ " movl %3,%0\n" ++ " shrl $2,%0\n" ++ " andl $3,%3\n" ++ " .align 2,0x90\n" ++ "0: rep; movsl\n" ++ " movl %3,%0\n" ++ "1: rep; movsb\n" ++ "2:\n" ++ " pushl %%ss\n" ++ " popl %%ds\n" ++ ".section .fixup,\"ax\"\n" ++ "5: addl %3,%0\n" ++ " jmp 2b\n" ++ "3: lea 0(%3,%0,4),%0\n" ++ " jmp 2b\n" ++ ".previous\n" ++ ".section __ex_table,\"a\"\n" ++ " .align 4\n" ++ " .long 4b,5b\n" ++ " .long 0b,3b\n" ++ " .long 1b,2b\n" ++ ".previous" ++ : "=&c"(size), "=&D" (__d0), "=&S" (__d1), "=r"(__d2) ++ : "3"(size), "0"(size), "1"(to), "2"(from), "r"(__USER_DS) ++ : "memory"); ++ return size; ++} ++ ++static unsigned long ++__copy_user_zeroing(void *to, const void __user *from, unsigned long size) ++{ ++ int __d0, __d1, __d2; ++ ++ __asm__ __volatile__( ++ " movw %w8,%%ds\n" ++ " cmp $7,%0\n" ++ " jbe 1f\n" ++ " movl %1,%0\n" ++ " negl %0\n" ++ " andl $7,%0\n" ++ " subl %0,%3\n" ++ "4: rep; movsb\n" ++ " movl %3,%0\n" ++ " shrl $2,%0\n" ++ " andl $3,%3\n" ++ " .align 2,0x90\n" ++ "0: rep; movsl\n" ++ " movl %3,%0\n" ++ "1: rep; movsb\n" ++ "2:\n" ++ " pushl %%ss\n" ++ " popl %%ds\n" ++ ".section .fixup,\"ax\"\n" ++ "5: addl %3,%0\n" ++ " jmp 6f\n" ++ "3: lea 0(%3,%0,4),%0\n" ++ "6: pushl %0\n" ++ " pushl %%eax\n" ++ " xorl %%eax,%%eax\n" ++ " rep; stosb\n" ++ " popl %%eax\n" ++ " popl %0\n" ++ " jmp 2b\n" ++ ".previous\n" ++ ".section __ex_table,\"a\"\n" ++ " .align 4\n" ++ " .long 4b,5b\n" ++ " .long 0b,3b\n" ++ " .long 1b,6b\n" ++ ".previous" ++ : "=&c"(size), "=&D" (__d0), "=&S" (__d1), "=r"(__d2) ++ : "3"(size), "0"(size), "1"(to), "2"(from), "r"(__USER_DS) ++ : "memory"); ++ return size; ++} + + unsigned long __copy_to_user_ll(void __user *to, const void *from, + unsigned long n) +@@ -775,9 +966,9 @@ survive: + } + #endif + if (movsl_is_ok(to, from, n)) +- __copy_user(to, from, n); ++ n = __generic_copy_to_user(to, from, n); + else +- n = __copy_user_intel(to, from, n); ++ n = __generic_copy_to_user_intel(to, from, n); + return n; + } + EXPORT_SYMBOL(__copy_to_user_ll); +@@ -786,7 +977,7 @@ unsigned long __copy_from_user_ll(void * + unsigned long n) + { + if (movsl_is_ok(to, from, n)) +- __copy_user_zeroing(to, from, n); ++ n = __copy_user_zeroing(to, from, n); + else + n = __copy_user_zeroing_intel(to, from, n); + return n; +@@ -797,10 +988,9 @@ unsigned long __copy_from_user_ll_nozero + unsigned long n) + { + if (movsl_is_ok(to, from, n)) +- __copy_user(to, from, n); ++ n = __generic_copy_from_user(to, from, n); + else +- n = __copy_user_intel((void __user *)to, +- (const void *)from, n); ++ n = __generic_copy_from_user_intel(to, from, n); + return n; + } + EXPORT_SYMBOL(__copy_from_user_ll_nozero); +@@ -812,9 +1002,9 @@ unsigned long __copy_from_user_ll_nocach + if (n > 64 && cpu_has_xmm2) + n = __copy_user_zeroing_intel_nocache(to, from, n); + else +- __copy_user_zeroing(to, from, n); ++ n = __copy_user_zeroing(to, from, n); + #else +- __copy_user_zeroing(to, from, n); ++ n = __copy_user_zeroing(to, from, n); + #endif + return n; + } +@@ -827,59 +1017,37 @@ unsigned long __copy_from_user_ll_nocach + if (n > 64 && cpu_has_xmm2) + n = __copy_user_intel_nocache(to, from, n); + else +- __copy_user(to, from, n); ++ n = __generic_copy_from_user(to, from, n); + #else +- __copy_user(to, from, n); ++ n = __generic_copy_from_user(to, from, n); + #endif + return n; + } + EXPORT_SYMBOL(__copy_from_user_ll_nocache_nozero); + +-/** +- * copy_to_user: - Copy a block of data into user space. +- * @to: Destination address, in user space. +- * @from: Source address, in kernel space. +- * @n: Number of bytes to copy. +- * +- * Context: User context only. This function may sleep. +- * +- * Copy data from kernel space to user space. +- * +- * Returns number of bytes that could not be copied. +- * On success, this will be zero. +- */ +-unsigned long +-copy_to_user(void __user *to, const void *from, unsigned long n) ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++void __set_fs(mm_segment_t x, int cpu) + { +- if (access_ok(VERIFY_WRITE, to, n)) +- n = __copy_to_user(to, from, n); +- return n; ++ unsigned long limit = x.seg; ++ struct desc_struct d; ++ ++ current_thread_info()->addr_limit = x; ++ if (likely(limit)) ++ limit = (limit - 1UL) >> PAGE_SHIFT; ++ pack_descriptor(&d, 0UL, limit, 0xF3, 0xC); ++ write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_DEFAULT_USER_DS, &d, DESCTYPE_S); + } +-EXPORT_SYMBOL(copy_to_user); + +-/** +- * copy_from_user: - Copy a block of data from user space. +- * @to: Destination address, in kernel space. +- * @from: Source address, in user space. +- * @n: Number of bytes to copy. +- * +- * Context: User context only. This function may sleep. +- * +- * Copy data from user space to kernel space. +- * +- * Returns number of bytes that could not be copied. +- * On success, this will be zero. +- * +- * If some data could not be copied, this function will pad the copied +- * data to the requested size using zero bytes. +- */ +-unsigned long +-copy_from_user(void *to, const void __user *from, unsigned long n) ++void set_fs(mm_segment_t x) + { +- if (access_ok(VERIFY_READ, from, n)) +- n = __copy_from_user(to, from, n); +- else +- memset(to, 0, n); +- return n; ++ __set_fs(x, get_cpu()); ++ put_cpu(); + } +-EXPORT_SYMBOL(copy_from_user); ++#else ++void set_fs(mm_segment_t x) ++{ ++ current_thread_info()->addr_limit = x; ++} ++#endif ++ ++EXPORT_SYMBOL(set_fs); +diff -urNp linux-2.6.31.1/arch/x86/Makefile linux-2.6.31.1/arch/x86/Makefile +--- linux-2.6.31.1/arch/x86/Makefile 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/Makefile 2009-10-01 20:12:42.000000000 -0400 +@@ -188,3 +188,12 @@ define archhelp + echo ' FDARGS="..." arguments for the booted kernel' + echo ' FDINITRD=file initrd for the booted kernel' + endef ++ ++define OLD_LD ++ ++*** ${VERSION}.${PATCHLEVEL} PaX kernels no longer build correctly with old versions of binutils. ++*** Please upgrade your binutils to 2.18 or newer ++endef ++ ++archprepare: ++ $(if $(LDFLAGS_BUILD_ID),,$(error $(OLD_LD))) +diff -urNp linux-2.6.31.1/arch/x86/mm/extable.c linux-2.6.31.1/arch/x86/mm/extable.c +--- linux-2.6.31.1/arch/x86/mm/extable.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/extable.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1,14 +1,81 @@ + #include + #include ++#include + #include + ++/* ++ * The exception table needs to be sorted so that the binary ++ * search that we use to find entries in it works properly. ++ * This is used both for the kernel exception table and for ++ * the exception tables of modules that get loaded. ++ */ ++static int cmp_ex(const void *a, const void *b) ++{ ++ const struct exception_table_entry *x = a, *y = b; ++ ++ /* avoid overflow */ ++ if (x->insn > y->insn) ++ return 1; ++ if (x->insn < y->insn) ++ return -1; ++ return 0; ++} ++ ++static void swap_ex(void *a, void *b, int size) ++{ ++ struct exception_table_entry t, *x = a, *y = b; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ ++ t = *x; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ ++ *x = *y; ++ *y = t; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ ++} ++ ++void sort_extable(struct exception_table_entry *start, ++ struct exception_table_entry *finish) ++{ ++ sort(start, finish - start, sizeof(struct exception_table_entry), ++ cmp_ex, swap_ex); ++} ++ ++#ifdef CONFIG_MODULES ++/* ++ * If the exception table is sorted, any referring to the module init ++ * will be at the beginning or the end. ++ */ ++void trim_init_extable(struct module *m) ++{ ++ /*trim the beginning*/ ++ while (m->num_exentries && within_module_init(m->extable[0].insn, m)) { ++ m->extable++; ++ m->num_exentries--; ++ } ++ /*trim the end*/ ++ while (m->num_exentries && ++ within_module_init(m->extable[m->num_exentries-1].insn, m)) ++ m->num_exentries--; ++} ++#endif /* CONFIG_MODULES */ + + int fixup_exception(struct pt_regs *regs) + { + const struct exception_table_entry *fixup; + + #ifdef CONFIG_PNPBIOS +- if (unlikely(SEGMENT_IS_PNP_CODE(regs->cs))) { ++ if (unlikely(!v8086_mode(regs) && SEGMENT_IS_PNP_CODE(regs->cs))) { + extern u32 pnp_bios_fault_eip, pnp_bios_fault_esp; + extern u32 pnp_bios_is_utter_crap; + pnp_bios_is_utter_crap = 1; +diff -urNp linux-2.6.31.1/arch/x86/mm/fault.c linux-2.6.31.1/arch/x86/mm/fault.c +--- linux-2.6.31.1/arch/x86/mm/fault.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/fault.c 2009-10-01 20:12:42.000000000 -0400 +@@ -11,10 +11,14 @@ + #include /* __kprobes, ... */ + #include /* kmmio_handler, ... */ + #include /* perf_swcounter_event */ ++#include ++#include + + #include /* dotraplinkage, ... */ + #include /* pgd_*(), ... */ + #include /* kmemcheck_*(), ... */ ++#include ++#include + + /* + * Page fault error code bits: +@@ -51,7 +55,7 @@ static inline int notify_page_fault(stru + int ret = 0; + + /* kprobe_running() needs smp_processor_id() */ +- if (kprobes_built_in() && !user_mode_vm(regs)) { ++ if (kprobes_built_in() && !user_mode(regs)) { + preempt_disable(); + if (kprobe_running() && kprobe_fault_handler(regs, 14)) + ret = 1; +@@ -171,6 +175,30 @@ force_sig_info_fault(int si_signo, int s + force_sig_info(si_signo, &info, tsk); + } + ++#ifdef CONFIG_PAX_EMUTRAMP ++static int pax_handle_fetch_fault(struct pt_regs *regs); ++#endif ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++static inline pmd_t * pax_get_pmd(struct mm_struct *mm, unsigned long address) ++{ ++ pgd_t *pgd; ++ pud_t *pud; ++ pmd_t *pmd; ++ ++ pgd = pgd_offset(mm, address); ++ if (!pgd_present(*pgd)) ++ return NULL; ++ pud = pud_offset(pgd, address); ++ if (!pud_present(*pud)) ++ return NULL; ++ pmd = pmd_offset(pud, address); ++ if (!pmd_present(*pmd)) ++ return NULL; ++ return pmd; ++} ++#endif ++ + DEFINE_SPINLOCK(pgd_lock); + LIST_HEAD(pgd_list); + +@@ -543,7 +571,7 @@ static int is_errata93(struct pt_regs *r + static int is_errata100(struct pt_regs *regs, unsigned long address) + { + #ifdef CONFIG_X86_64 +- if ((regs->cs == __USER32_CS || (regs->cs & (1<<2))) && (address >> 32)) ++ if ((regs->cs == __USER32_CS || (regs->cs & SEGMENT_LDT)) && (address >> 32)) + return 1; + #endif + return 0; +@@ -570,7 +598,7 @@ static int is_f00f_bug(struct pt_regs *r + } + + static const char nx_warning[] = KERN_CRIT +-"kernel tried to execute NX-protected page - exploit attempt? (uid: %d)\n"; ++"kernel tried to execute NX-protected page - exploit attempt? (uid: %d, task: %s, pid: %d)\n"; + + static void + show_fault_oops(struct pt_regs *regs, unsigned long error_code, +@@ -579,15 +607,31 @@ show_fault_oops(struct pt_regs *regs, un + if (!oops_may_print()) + return; + +- if (error_code & PF_INSTR) { ++ if (nx_enabled && (error_code & PF_INSTR)) { + unsigned int level; + + pte_t *pte = lookup_address(address, &level); + + if (pte && pte_present(*pte) && !pte_exec(*pte)) +- printk(nx_warning, current_uid()); ++ printk(nx_warning, current_uid(), current->comm, task_pid_nr(current)); + } + ++#ifdef CONFIG_PAX_KERNEXEC ++#if defined(CONFIG_x86_32) && defined(CONFIG_MODULES) ++ if (init_mm.start_code <= address && address < (unsigned long)&MODULES_EXEC_END) ++#else ++ if (init_mm.start_code <= address && address < init_mm.end_code) ++#endif ++ { ++ if (current->signal->curr_ip) ++ printk(KERN_ERR "PAX: From %u.%u.%u.%u: %s:%d, uid/euid: %u/%u, attempted to modify kernel code\n", ++ NIPQUAD(current->signal->curr_ip), current->comm, task_pid_nr(current), current_uid(), current_euid()); ++ else ++ printk(KERN_ERR "PAX: %s:%d, uid/euid: %u/%u, attempted to modify kernel code\n", ++ current->comm, task_pid_nr(current), current_uid(), current_euid()); ++ } ++#endif ++ + printk(KERN_ALERT "BUG: unable to handle kernel "); + if (address < PAGE_SIZE) + printk(KERN_CONT "NULL pointer dereference"); +@@ -712,6 +756,68 @@ __bad_area_nosemaphore(struct pt_regs *r + unsigned long address, int si_code) + { + struct task_struct *tsk = current; ++ struct mm_struct *mm = tsk->mm; ++ ++#ifdef CONFIG_X86_64 ++ if (mm && (error_code & PF_INSTR)) { ++ if (regs->ip == (unsigned long)vgettimeofday) { ++ regs->ip = (unsigned long)VDSO64_SYMBOL(mm->context.vdso, fallback_gettimeofday); ++ return; ++ } else if (regs->ip == (unsigned long)vtime) { ++ regs->ip = (unsigned long)VDSO64_SYMBOL(mm->context.vdso, fallback_time); ++ return; ++ } else if (regs->ip == (unsigned long)vgetcpu) { ++ regs->ip = (unsigned long)VDSO64_SYMBOL(mm->context.vdso, getcpu); ++ return; ++ } ++ } ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ if (mm && (error_code & PF_USER)) { ++ unsigned long ip = regs->ip; ++ ++ if (v8086_mode(regs)) ++ ip = ((regs->cs & 0xffff) << 4) + (regs->ip & 0xffff); ++ ++ /* ++ * It's possible to have interrupts off here: ++ */ ++ local_irq_enable(); ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && ++ ((nx_enabled && (error_code & PF_INSTR)) || (!(error_code & (PF_PROT | PF_WRITE)) && regs->ip == address))) { ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ switch (pax_handle_fetch_fault(regs)) { ++ case 2: ++ return; ++ } ++#endif ++ ++ pax_report_fault(regs, (void *)regs->ip, (void *)regs->sp); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && !(error_code & (PF_PROT | PF_WRITE)) && (regs->ip + SEGMEXEC_TASK_SIZE == address)) { ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ switch (pax_handle_fetch_fault(regs)) { ++ case 2: ++ return; ++ } ++#endif ++ ++ pax_report_fault(regs, (void *)regs->ip, (void *)regs->sp); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ ++ } ++#endif + + /* User mode accesses just cause a SIGSEGV */ + if (error_code & PF_USER) { +@@ -846,6 +952,106 @@ static int spurious_fault_check(unsigned + return 1; + } + ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) ++static int pax_handle_pageexec_fault(struct pt_regs *regs, struct mm_struct *mm, unsigned long address, unsigned long error_code) ++{ ++ pte_t *pte; ++ pmd_t *pmd; ++ spinlock_t *ptl; ++ unsigned char pte_mask; ++ ++ if (nx_enabled || (error_code & (PF_PROT|PF_USER)) != (PF_PROT|PF_USER) || v8086_mode(regs) || ++ !(mm->pax_flags & MF_PAX_PAGEEXEC)) ++ return 0; ++ ++ /* PaX: it's our fault, let's handle it if we can */ ++ ++ /* PaX: take a look at read faults before acquiring any locks */ ++ if (unlikely(!(error_code & PF_WRITE) && (regs->ip == address))) { ++ /* instruction fetch attempt from a protected page in user mode */ ++ up_read(&mm->mmap_sem); ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ switch (pax_handle_fetch_fault(regs)) { ++ case 2: ++ return 1; ++ } ++#endif ++ ++ pax_report_fault(regs, (void *)regs->ip, (void *)regs->sp); ++ do_group_exit(SIGKILL); ++ } ++ ++ pmd = pax_get_pmd(mm, address); ++ if (unlikely(!pmd)) ++ return 0; ++ ++ pte = pte_offset_map_lock(mm, pmd, address, &ptl); ++ if (unlikely(!(pte_val(*pte) & _PAGE_PRESENT) || pte_user(*pte))) { ++ pte_unmap_unlock(pte, ptl); ++ return 0; ++ } ++ ++ if (unlikely((error_code & PF_WRITE) && !pte_write(*pte))) { ++ /* write attempt to a protected page in user mode */ ++ pte_unmap_unlock(pte, ptl); ++ return 0; ++ } ++ ++#ifdef CONFIG_SMP ++ if (likely(address > get_limit(regs->cs) && cpu_isset(smp_processor_id(), mm->context.cpu_user_cs_mask))) ++#else ++ if (likely(address > get_limit(regs->cs))) ++#endif ++ { ++ set_pte(pte, pte_mkread(*pte)); ++ __flush_tlb_one(address); ++ pte_unmap_unlock(pte, ptl); ++ up_read(&mm->mmap_sem); ++ return 1; ++ } ++ ++ pte_mask = _PAGE_ACCESSED | _PAGE_USER | ((error_code & PF_WRITE) << (_PAGE_BIT_DIRTY-1)); ++ ++ /* ++ * PaX: fill DTLB with user rights and retry ++ */ ++ __asm__ __volatile__ ( ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ "movw %w4,%%es\n" ++#endif ++ "orb %2,(%1)\n" ++#if defined(CONFIG_M586) || defined(CONFIG_M586TSC) ++/* ++ * PaX: let this uncommented 'invlpg' remind us on the behaviour of Intel's ++ * (and AMD's) TLBs. namely, they do not cache PTEs that would raise *any* ++ * page fault when examined during a TLB load attempt. this is true not only ++ * for PTEs holding a non-present entry but also present entries that will ++ * raise a page fault (such as those set up by PaX, or the copy-on-write ++ * mechanism). in effect it means that we do *not* need to flush the TLBs ++ * for our target pages since their PTEs are simply not in the TLBs at all. ++ ++ * the best thing in omitting it is that we gain around 15-20% speed in the ++ * fast path of the page fault handler and can get rid of tracing since we ++ * can no longer flush unintended entries. ++ */ ++ "invlpg (%0)\n" ++#endif ++ "testb $0,%%es:(%0)\n" ++ "xorb %3,(%1)\n" ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ "pushl %%ss\n" ++ "popl %%es\n" ++#endif ++ : ++ : "r" (address), "r" (pte), "q" (pte_mask), "i" (_PAGE_USER), "r" (__USER_DS) ++ : "memory", "cc"); ++ pte_unmap_unlock(pte, ptl); ++ up_read(&mm->mmap_sem); ++ return 1; ++} ++#endif ++ + /* + * Handle a spurious fault caused by a stale TLB entry. + * +@@ -912,6 +1118,9 @@ int show_unhandled_signals = 1; + static inline int + access_error(unsigned long error_code, int write, struct vm_area_struct *vma) + { ++ if (nx_enabled && (error_code & PF_INSTR) && !(vma->vm_flags & VM_EXEC)) ++ return 1; ++ + if (write) { + /* write, present and write, not present: */ + if (unlikely(!(vma->vm_flags & VM_WRITE))) +@@ -945,17 +1154,16 @@ do_page_fault(struct pt_regs *regs, unsi + { + struct vm_area_struct *vma; + struct task_struct *tsk; +- unsigned long address; + struct mm_struct *mm; + int write; + int fault; + ++ /* Get the faulting address: */ ++ const unsigned long address = read_cr2(); ++ + tsk = current; + mm = tsk->mm; + +- /* Get the faulting address: */ +- address = read_cr2(); +- + /* + * Detect and handle instructions that would cause a page fault for + * both a tracked kernel page and a userspace page. +@@ -1015,7 +1223,7 @@ do_page_fault(struct pt_regs *regs, unsi + * User-mode registers count as a user access even for any + * potential system fault or CPU buglet: + */ +- if (user_mode_vm(regs)) { ++ if (user_mode(regs)) { + local_irq_enable(); + error_code |= PF_USER; + } else { +@@ -1069,6 +1277,11 @@ do_page_fault(struct pt_regs *regs, unsi + might_sleep(); + } + ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) ++ if (pax_handle_pageexec_fault(regs, mm, address, error_code)) ++ return; ++#endif ++ + vma = find_vma(mm, address); + if (unlikely(!vma)) { + bad_area(regs, error_code, address); +@@ -1080,18 +1293,24 @@ do_page_fault(struct pt_regs *regs, unsi + bad_area(regs, error_code, address); + return; + } +- if (error_code & PF_USER) { +- /* +- * Accessing the stack below %sp is always a bug. +- * The large cushion allows instructions like enter +- * and pusha to work. ("enter $65535, $31" pushes +- * 32 pointers and then decrements %sp by 65535.) +- */ +- if (unlikely(address + 65536 + 32 * sizeof(unsigned long) < regs->sp)) { +- bad_area(regs, error_code, address); +- return; +- } ++ /* ++ * Accessing the stack below %sp is always a bug. ++ * The large cushion allows instructions like enter ++ * and pusha to work. ("enter $65535, $31" pushes ++ * 32 pointers and then decrements %sp by 65535.) ++ */ ++ if (unlikely(address + 65536 + 32 * sizeof(unsigned long) < task_pt_regs(tsk)->sp)) { ++ bad_area(regs, error_code, address); ++ return; ++ } ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (unlikely((mm->pax_flags & MF_PAX_SEGMEXEC) && vma->vm_end - SEGMEXEC_TASK_SIZE - 1 < address - SEGMEXEC_TASK_SIZE - 1)) { ++ bad_area(regs, error_code, address); ++ return; + } ++#endif ++ + if (unlikely(expand_stack(vma, address))) { + bad_area(regs, error_code, address); + return; +@@ -1135,3 +1354,174 @@ good_area: + + up_read(&mm->mmap_sem); + } ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++static int pax_handle_fetch_fault_32(struct pt_regs *regs) ++{ ++ int err; ++ ++ do { /* PaX: gcc trampoline emulation #1 */ ++ unsigned char mov1, mov2; ++ unsigned short jmp; ++ unsigned int addr1, addr2; ++ ++#ifdef CONFIG_X86_64 ++ if ((regs->ip + 11) >> 32) ++ break; ++#endif ++ ++ err = get_user(mov1, (unsigned char __user *)regs->ip); ++ err |= get_user(addr1, (unsigned int __user *)(regs->ip + 1)); ++ err |= get_user(mov2, (unsigned char __user *)(regs->ip + 5)); ++ err |= get_user(addr2, (unsigned int __user *)(regs->ip + 6)); ++ err |= get_user(jmp, (unsigned short __user *)(regs->ip + 10)); ++ ++ if (err) ++ break; ++ ++ if (mov1 == 0xB9 && mov2 == 0xB8 && jmp == 0xE0FF) { ++ regs->cx = addr1; ++ regs->ax = addr2; ++ regs->ip = addr2; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: gcc trampoline emulation #2 */ ++ unsigned char mov, jmp; ++ unsigned int addr1, addr2; ++ ++#ifdef CONFIG_X86_64 ++ if ((regs->ip + 9) >> 32) ++ break; ++#endif ++ ++ err = get_user(mov, (unsigned char __user *)regs->ip); ++ err |= get_user(addr1, (unsigned int __user *)(regs->ip + 1)); ++ err |= get_user(jmp, (unsigned char __user *)(regs->ip + 5)); ++ err |= get_user(addr2, (unsigned int __user *)(regs->ip + 6)); ++ ++ if (err) ++ break; ++ ++ if (mov == 0xB9 && jmp == 0xE9) { ++ regs->cx = addr1; ++ regs->ip = (unsigned int)(regs->ip + addr2 + 10); ++ return 2; ++ } ++ } while (0); ++ ++ return 1; /* PaX in action */ ++} ++ ++#ifdef CONFIG_X86_64 ++static int pax_handle_fetch_fault_64(struct pt_regs *regs) ++{ ++ int err; ++ ++ do { /* PaX: gcc trampoline emulation #1 */ ++ unsigned short mov1, mov2, jmp1; ++ unsigned char jmp2; ++ unsigned int addr1; ++ unsigned long addr2; ++ ++ err = get_user(mov1, (unsigned short __user *)regs->ip); ++ err |= get_user(addr1, (unsigned int __user *)(regs->ip + 2)); ++ err |= get_user(mov2, (unsigned short __user *)(regs->ip + 6)); ++ err |= get_user(addr2, (unsigned long __user *)(regs->ip + 8)); ++ err |= get_user(jmp1, (unsigned short __user *)(regs->ip + 16)); ++ err |= get_user(jmp2, (unsigned char __user *)(regs->ip + 18)); ++ ++ if (err) ++ break; ++ ++ if (mov1 == 0xBB41 && mov2 == 0xBA49 && jmp1 == 0xFF49 && jmp2 == 0xE3) { ++ regs->r11 = addr1; ++ regs->r10 = addr2; ++ regs->ip = addr1; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: gcc trampoline emulation #2 */ ++ unsigned short mov1, mov2, jmp1; ++ unsigned char jmp2; ++ unsigned long addr1, addr2; ++ ++ err = get_user(mov1, (unsigned short __user *)regs->ip); ++ err |= get_user(addr1, (unsigned long __user *)(regs->ip + 2)); ++ err |= get_user(mov2, (unsigned short __user *)(regs->ip + 10)); ++ err |= get_user(addr2, (unsigned long __user *)(regs->ip + 12)); ++ err |= get_user(jmp1, (unsigned short __user *)(regs->ip + 20)); ++ err |= get_user(jmp2, (unsigned char __user *)(regs->ip + 22)); ++ ++ if (err) ++ break; ++ ++ if (mov1 == 0xBB49 && mov2 == 0xBA49 && jmp1 == 0xFF49 && jmp2 == 0xE3) { ++ regs->r11 = addr1; ++ regs->r10 = addr2; ++ regs->ip = addr1; ++ return 2; ++ } ++ } while (0); ++ ++ return 1; /* PaX in action */ ++} ++#endif ++ ++/* ++ * PaX: decide what to do with offenders (regs->ip = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when gcc trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ if (v8086_mode(regs)) ++ return 1; ++ ++ if (!(current->mm->pax_flags & MF_PAX_EMUTRAMP)) ++ return 1; ++ ++#ifdef CONFIG_X86_32 ++ return pax_handle_fetch_fault_32(regs); ++#else ++ if (regs->cs == __USER32_CS || (regs->cs & SEGMENT_LDT)) ++ return pax_handle_fetch_fault_32(regs); ++ else ++ return pax_handle_fetch_fault_64(regs); ++#endif ++} ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++void pax_report_insns(void *pc, void *sp) ++{ ++ long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 20; i++) { ++ unsigned char c; ++ if (get_user(c, (unsigned char __user *)pc+i)) ++ printk(KERN_CONT "?? "); ++ else ++ printk(KERN_CONT "%02x ", c); ++ } ++ printk("\n"); ++ ++ printk(KERN_ERR "PAX: bytes at SP-%lu: ", (unsigned long)sizeof(long)); ++ for (i = -1; i < 80 / sizeof(long); i++) { ++ unsigned long c; ++ if (get_user(c, (unsigned long __user *)sp+i)) ++#ifdef CONFIG_X86_32 ++ printk(KERN_CONT "???????? "); ++#else ++ printk(KERN_CONT "???????????????? "); ++#endif ++ else ++ printk(KERN_CONT "%0*lx ", 2 * (int)sizeof(long), c); ++ } ++ printk("\n"); ++} ++#endif +diff -urNp linux-2.6.31.1/arch/x86/mm/highmem_32.c linux-2.6.31.1/arch/x86/mm/highmem_32.c +--- linux-2.6.31.1/arch/x86/mm/highmem_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/highmem_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -32,6 +32,10 @@ void *kmap_atomic_prot(struct page *page + enum fixed_addresses idx; + unsigned long vaddr; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */ + pagefault_disable(); + +@@ -43,8 +47,17 @@ void *kmap_atomic_prot(struct page *page + idx = type + KM_TYPE_NR*smp_processor_id(); + vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); + BUG_ON(!pte_none(*(kmap_pte-idx))); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + set_pte(kmap_pte-idx, mk_pte(page, prot)); + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + return (void *)vaddr; + } + +@@ -58,15 +71,29 @@ void kunmap_atomic(void *kvaddr, enum km + unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK; + enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id(); + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + /* + * Force other mappings to Oops if they'll try to access this pte + * without first remap it. Keeping stale mappings around is a bad idea + * also, in case the page changes cacheability attributes or becomes + * a protected page in a hypervisor. + */ +- if (vaddr == __fix_to_virt(FIX_KMAP_BEGIN+idx)) ++ if (vaddr == __fix_to_virt(FIX_KMAP_BEGIN+idx)) { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + kpte_clear_flush(kmap_pte-idx, vaddr); +- else { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ ++ } else { + #ifdef CONFIG_DEBUG_HIGHMEM + BUG_ON(vaddr < PAGE_OFFSET); + BUG_ON(vaddr >= (unsigned long)high_memory); +diff -urNp linux-2.6.31.1/arch/x86/mm/hugetlbpage.c linux-2.6.31.1/arch/x86/mm/hugetlbpage.c +--- linux-2.6.31.1/arch/x86/mm/hugetlbpage.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/hugetlbpage.c 2009-10-01 20:12:42.000000000 -0400 +@@ -267,13 +267,18 @@ static unsigned long hugetlb_get_unmappe + struct hstate *h = hstate_file(file); + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; +- unsigned long start_addr; ++ unsigned long start_addr, pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif + + if (len > mm->cached_hole_size) { +- start_addr = mm->free_area_cache; ++ start_addr = mm->free_area_cache; + } else { +- start_addr = TASK_UNMAPPED_BASE; +- mm->cached_hole_size = 0; ++ start_addr = mm->mmap_base; ++ mm->cached_hole_size = 0; + } + + full_search: +@@ -281,13 +286,13 @@ full_search: + + for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { + /* At this point: (!vma || addr < vma->vm_end). */ +- if (TASK_SIZE - len < addr) { ++ if (pax_task_size - len < addr) { + /* + * Start a new search - just in case we missed + * some holes. + */ +- if (start_addr != TASK_UNMAPPED_BASE) { +- start_addr = TASK_UNMAPPED_BASE; ++ if (start_addr != mm->mmap_base) { ++ start_addr = mm->mmap_base; + mm->cached_hole_size = 0; + goto full_search; + } +@@ -310,9 +315,8 @@ static unsigned long hugetlb_get_unmappe + struct hstate *h = hstate_file(file); + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma, *prev_vma; +- unsigned long base = mm->mmap_base, addr = addr0; ++ unsigned long base = mm->mmap_base, addr; + unsigned long largest_hole = mm->cached_hole_size; +- int first_time = 1; + + /* don't allow allocations above current base */ + if (mm->free_area_cache > base) +@@ -322,7 +326,7 @@ static unsigned long hugetlb_get_unmappe + largest_hole = 0; + mm->free_area_cache = base; + } +-try_again: ++ + /* make sure it can fit in the remaining address space */ + if (mm->free_area_cache < len) + goto fail; +@@ -364,22 +368,26 @@ try_again: + + fail: + /* +- * if hint left us with no space for the requested +- * mapping then try again: +- */ +- if (first_time) { +- mm->free_area_cache = base; +- largest_hole = 0; +- first_time = 0; +- goto try_again; +- } +- /* + * A failed mmap() very likely causes application failure, + * so fall back to the bottom-up function here. This scenario + * can happen with large stack limits and large mmap() + * allocations. + */ +- mm->free_area_cache = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ mm->mmap_base = SEGMEXEC_TASK_UNMAPPED_BASE; ++ else ++#endif ++ ++ mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ ++ mm->free_area_cache = mm->mmap_base; + mm->cached_hole_size = ~0UL; + addr = hugetlb_get_unmapped_area_bottomup(file, addr0, + len, pgoff, flags); +@@ -387,6 +395,7 @@ fail: + /* + * Restore the topdown base: + */ ++ mm->mmap_base = base; + mm->free_area_cache = base; + mm->cached_hole_size = ~0UL; + +@@ -400,10 +409,17 @@ hugetlb_get_unmapped_area(struct file *f + struct hstate *h = hstate_file(file); + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; ++ unsigned long pax_task_size = TASK_SIZE; + + if (len & ~huge_page_mask(h)) + return -EINVAL; +- if (len > TASK_SIZE) ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ if (len > pax_task_size) + return -ENOMEM; + + if (flags & MAP_FIXED) { +@@ -415,7 +431,7 @@ hugetlb_get_unmapped_area(struct file *f + if (addr) { + addr = ALIGN(addr, huge_page_size(h)); + vma = find_vma(mm, addr); +- if (TASK_SIZE - len >= addr && ++ if (pax_task_size - len >= addr && + (!vma || addr + len <= vma->vm_start)) + return addr; + } +diff -urNp linux-2.6.31.1/arch/x86/mm/init_32.c linux-2.6.31.1/arch/x86/mm/init_32.c +--- linux-2.6.31.1/arch/x86/mm/init_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/init_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -51,6 +51,7 @@ + #include + #include + #include ++#include + + unsigned long highstart_pfn, highend_pfn; + +@@ -72,36 +73,6 @@ static __init void *alloc_low_page(void) + } + + /* +- * Creates a middle page table and puts a pointer to it in the +- * given global directory entry. This only returns the gd entry +- * in non-PAE compilation mode, since the middle layer is folded. +- */ +-static pmd_t * __init one_md_table_init(pgd_t *pgd) +-{ +- pud_t *pud; +- pmd_t *pmd_table; +- +-#ifdef CONFIG_X86_PAE +- if (!(pgd_val(*pgd) & _PAGE_PRESENT)) { +- if (after_bootmem) +- pmd_table = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE); +- else +- pmd_table = (pmd_t *)alloc_low_page(); +- paravirt_alloc_pmd(&init_mm, __pa(pmd_table) >> PAGE_SHIFT); +- set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT)); +- pud = pud_offset(pgd, 0); +- BUG_ON(pmd_table != pmd_offset(pud, 0)); +- +- return pmd_table; +- } +-#endif +- pud = pud_offset(pgd, 0); +- pmd_table = pmd_offset(pud, 0); +- +- return pmd_table; +-} +- +-/* + * Create a page table and place a pointer to it in a middle page + * directory entry: + */ +@@ -121,13 +92,28 @@ static pte_t * __init one_page_table_ini + page_table = (pte_t *)alloc_low_page(); + + paravirt_alloc_pte(&init_mm, __pa(page_table) >> PAGE_SHIFT); ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ set_pmd(pmd, __pmd(__pa(page_table) | _KERNPG_TABLE)); ++#else + set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE)); ++#endif + BUG_ON(page_table != pte_offset_kernel(pmd, 0)); + } + + return pte_offset_kernel(pmd, 0); + } + ++static pmd_t * __init one_md_table_init(pgd_t *pgd) ++{ ++ pud_t *pud; ++ pmd_t *pmd_table; ++ ++ pud = pud_offset(pgd, 0); ++ pmd_table = pmd_offset(pud, 0); ++ ++ return pmd_table; ++} ++ + pmd_t * __init populate_extra_pmd(unsigned long vaddr) + { + int pgd_idx = pgd_index(vaddr); +@@ -201,6 +187,7 @@ page_table_range_init(unsigned long star + int pgd_idx, pmd_idx; + unsigned long vaddr; + pgd_t *pgd; ++ pud_t *pud; + pmd_t *pmd; + pte_t *pte = NULL; + +@@ -210,8 +197,13 @@ page_table_range_init(unsigned long star + pgd = pgd_base + pgd_idx; + + for ( ; (pgd_idx < PTRS_PER_PGD) && (vaddr != end); pgd++, pgd_idx++) { +- pmd = one_md_table_init(pgd); +- pmd = pmd + pmd_index(vaddr); ++ pud = pud_offset(pgd, vaddr); ++ pmd = pmd_offset(pud, vaddr); ++ ++#ifdef CONFIG_X86_PAE ++ paravirt_alloc_pmd(&init_mm, __pa(pmd) >> PAGE_SHIFT); ++#endif ++ + for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); + pmd++, pmd_idx++) { + pte = page_table_kmap_check(one_page_table_init(pmd), +@@ -223,11 +215,23 @@ page_table_range_init(unsigned long star + } + } + +-static inline int is_kernel_text(unsigned long addr) ++static inline int is_kernel_text(unsigned long start, unsigned long end) + { +- if (addr >= PAGE_OFFSET && addr <= (unsigned long)__init_end) +- return 1; +- return 0; ++ unsigned long etext; ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_PAX_KERNEXEC) ++ etext = ktva_ktla((unsigned long)&MODULES_EXEC_END); ++#else ++ etext = (unsigned long)&_etext; ++#endif ++ ++ if ((start > ktla_ktva(etext) || ++ end <= ktla_ktva((unsigned long)_stext)) && ++ (start > ktla_ktva((unsigned long)_einittext) || ++ end <= ktla_ktva((unsigned long)_sinittext)) && ++ (start > (unsigned long)__va(0xfffff) || end <= (unsigned long)__va(0xc0000))) ++ return 0; ++ return 1; + } + + /* +@@ -243,9 +247,10 @@ kernel_physical_mapping_init(unsigned lo + int use_pse = page_size_mask == (1<> PAGE_SHIFT); ++#endif + + if (pfn >= end_pfn) + continue; +@@ -291,14 +301,13 @@ repeat: + #endif + for (; pmd_idx < PTRS_PER_PMD && pfn < end_pfn; + pmd++, pmd_idx++) { +- unsigned int addr = pfn * PAGE_SIZE + PAGE_OFFSET; ++ unsigned long address = pfn * PAGE_SIZE + PAGE_OFFSET; + + /* + * Map with big pages if possible, otherwise + * create normal page tables: + */ + if (use_pse) { +- unsigned int addr2; + pgprot_t prot = PAGE_KERNEL_LARGE; + /* + * first pass will use the same initial +@@ -308,11 +317,7 @@ repeat: + __pgprot(PTE_IDENT_ATTR | + _PAGE_PSE); + +- addr2 = (pfn + PTRS_PER_PTE-1) * PAGE_SIZE + +- PAGE_OFFSET + PAGE_SIZE-1; +- +- if (is_kernel_text(addr) || +- is_kernel_text(addr2)) ++ if (is_kernel_text(address, address + PMD_SIZE)) + prot = PAGE_KERNEL_LARGE_EXEC; + + pages_2m++; +@@ -329,7 +334,7 @@ repeat: + pte_ofs = pte_index((pfn<> 10, + +- (unsigned long)&_etext, (unsigned long)&_edata, +- ((unsigned long)&_edata - (unsigned long)&_etext) >> 10, ++ (unsigned long)&_sdata, (unsigned long)&_edata, ++ ((unsigned long)&_edata - (unsigned long)&_sdata) >> 10, + +- (unsigned long)&_text, (unsigned long)&_etext, ++ ktla_ktva((unsigned long)&_text), ktla_ktva((unsigned long)&_etext), + ((unsigned long)&_etext - (unsigned long)&_text) >> 10); + + /* +diff -urNp linux-2.6.31.1/arch/x86/mm/init_64.c linux-2.6.31.1/arch/x86/mm/init_64.c +--- linux-2.6.31.1/arch/x86/mm/init_64.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/init_64.c 2009-10-01 20:12:42.000000000 -0400 +@@ -159,12 +159,24 @@ void set_pte_vaddr_pud(pud_t *pud_page, + pmd_t *pmd; + pte_t *pte; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + pud = pud_page + pud_index(vaddr); + pmd = fill_pmd(pud, vaddr); + pte = fill_pte(pmd, vaddr); + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + set_pte(pte, new_pte); + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + /* + * It's enough to flush this one mapping. + * (PGE mappings get flushed as well) +@@ -222,14 +234,12 @@ static void __init __init_extra_mapping( + pgd = pgd_offset_k((unsigned long)__va(phys)); + if (pgd_none(*pgd)) { + pud = (pud_t *) spp_getpage(); +- set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE | +- _PAGE_USER)); ++ set_pgd(pgd, __pgd(__pa(pud) | _PAGE_TABLE)); + } + pud = pud_offset(pgd, (unsigned long)__va(phys)); + if (pud_none(*pud)) { + pmd = (pmd_t *) spp_getpage(); +- set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | +- _PAGE_USER)); ++ set_pud(pud, __pud(__pa(pmd) | _PAGE_TABLE)); + } + pmd = pmd_offset(pud, phys); + BUG_ON(!pmd_none(*pmd)); +@@ -848,8 +858,8 @@ int kern_addr_valid(unsigned long addr) + static struct vm_area_struct gate_vma = { + .vm_start = VSYSCALL_START, + .vm_end = VSYSCALL_START + (VSYSCALL_MAPPED_PAGES * PAGE_SIZE), +- .vm_page_prot = PAGE_READONLY_EXEC, +- .vm_flags = VM_READ | VM_EXEC ++ .vm_page_prot = PAGE_READONLY, ++ .vm_flags = VM_READ + }; + + struct vm_area_struct *get_gate_vma(struct task_struct *tsk) +@@ -883,7 +893,7 @@ int in_gate_area_no_task(unsigned long a + + const char *arch_vma_name(struct vm_area_struct *vma) + { +- if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso) ++ if (vma->vm_mm && vma->vm_start == vma->vm_mm->context.vdso) + return "[vdso]"; + if (vma == &gate_vma) + return "[vsyscall]"; +diff -urNp linux-2.6.31.1/arch/x86/mm/init.c linux-2.6.31.1/arch/x86/mm/init.c +--- linux-2.6.31.1/arch/x86/mm/init.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/init.c 2009-10-01 20:12:42.000000000 -0400 +@@ -28,11 +28,10 @@ int direct_gbpages + #endif + ; + ++#if defined(CONFIG_X86_32) && defined(CONFIG_X86_PAE) + int nx_enabled; + +-#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) +-static int disable_nx __cpuinitdata; +- ++#ifndef CONFIG_PAX_PAGEEXEC + /* + * noexec = on|off + * +@@ -46,32 +45,26 @@ static int __init noexec_setup(char *str + if (!str) + return -EINVAL; + if (!strncmp(str, "on", 2)) { +- __supported_pte_mask |= _PAGE_NX; +- disable_nx = 0; ++ nx_enabled = 1; + } else if (!strncmp(str, "off", 3)) { +- disable_nx = 1; +- __supported_pte_mask &= ~_PAGE_NX; ++ nx_enabled = 0; + } + return 0; + } + early_param("noexec", noexec_setup); + #endif ++#endif + + #ifdef CONFIG_X86_PAE + static void __init set_nx(void) + { +- unsigned int v[4], l, h; +- +- if (cpu_has_pae && (cpuid_eax(0x80000000) > 0x80000001)) { +- cpuid(0x80000001, &v[0], &v[1], &v[2], &v[3]); ++ if (!nx_enabled && cpu_has_nx) { ++ unsigned l, h; + +- if ((v[3] & (1 << 20)) && !disable_nx) { +- rdmsr(MSR_EFER, l, h); +- l |= EFER_NX; +- wrmsr(MSR_EFER, l, h); +- nx_enabled = 1; +- __supported_pte_mask |= _PAGE_NX; +- } ++ __supported_pte_mask &= ~_PAGE_NX; ++ rdmsr(MSR_EFER, l, h); ++ l &= ~EFER_NX; ++ wrmsr(MSR_EFER, l, h); + } + } + #else +@@ -86,7 +79,7 @@ void __cpuinit check_efer(void) + unsigned long efer; + + rdmsrl(MSR_EFER, efer); +- if (!(efer & EFER_NX) || disable_nx) ++ if (!(efer & EFER_NX) || !nx_enabled) + __supported_pte_mask &= ~_PAGE_NX; + } + #endif +@@ -394,7 +387,13 @@ unsigned long __init_refok init_memory_m + */ + int devmem_is_allowed(unsigned long pagenr) + { +- if (pagenr <= 256) ++ if (!pagenr) ++ return 1; ++#ifdef CONFIG_VM86 ++ if (pagenr < (ISA_START_ADDRESS >> PAGE_SHIFT)) ++ return 1; ++#endif ++ if ((ISA_START_ADDRESS >> PAGE_SHIFT) <= pagenr && pagenr < (ISA_END_ADDRESS >> PAGE_SHIFT)) + return 1; + if (iomem_is_exclusive(pagenr << PAGE_SHIFT)) + return 0; +@@ -442,6 +441,76 @@ void free_init_pages(char *what, unsigne + + void free_initmem(void) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pgd_t *pgd; ++ pud_t *pud; ++ pmd_t *pmd; ++ ++#ifdef CONFIG_X86_32 ++ /* PaX: limit KERNEL_CS to actual size */ ++ unsigned long addr, limit; ++ struct desc_struct d; ++ int cpu; ++ ++#ifdef CONFIG_MODULES ++ limit = ktva_ktla((unsigned long)&MODULES_EXEC_END); ++#else ++ limit = (unsigned long)&_etext; ++#endif ++ limit = (limit - 1UL) >> PAGE_SHIFT; ++ ++ memset(__LOAD_PHYSICAL_ADDR + PAGE_OFFSET, POISON_FREE_INITMEM, PAGE_SIZE); ++ for (cpu = 0; cpu < NR_CPUS; cpu++) { ++ pack_descriptor(&d, get_desc_base(&get_cpu_gdt_table(cpu)[GDT_ENTRY_KERNEL_CS]), limit, 0x9B, 0xC); ++ write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_KERNEL_CS, &d, DESCTYPE_S); ++ } ++ ++ /* PaX: make KERNEL_CS read-only */ ++ for (addr = ktla_ktva((unsigned long)&_text); addr < (unsigned long)&_sdata; addr += PMD_SIZE) { ++ pgd = pgd_offset_k(addr); ++ pud = pud_offset(pgd, addr); ++ pmd = pmd_offset(pud, addr); ++ set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_RW)); ++ } ++#ifdef CONFIG_X86_PAE ++ for (addr = (unsigned long)&__init_begin; addr < (unsigned long)&__init_end; addr += PMD_SIZE) { ++ pgd = pgd_offset_k(addr); ++ pud = pud_offset(pgd, addr); ++ pmd = pmd_offset(pud, addr); ++ set_pmd(pmd, __pmd(pmd_val(*pmd) | (_PAGE_NX & __supported_pte_mask))); ++ } ++#endif ++#else ++ unsigned long addr, end; ++ ++ /* PaX: make kernel code/rodata read-only, rest non-executable */ ++ for (addr = __START_KERNEL_map; addr < __START_KERNEL_map + KERNEL_IMAGE_SIZE; addr += PMD_SIZE) { ++ pgd = pgd_offset_k(addr); ++ pud = pud_offset(pgd, addr); ++ pmd = pmd_offset(pud, addr); ++ if ((unsigned long)_text <= addr && addr < (unsigned long)_sdata) ++ set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_RW)); ++ else ++ set_pmd(pmd, __pmd(pmd_val(*pmd) | (_PAGE_NX & __supported_pte_mask))); ++ } ++ ++ addr = (unsigned long)__va(__pa(__START_KERNEL_map)); ++ end = addr + KERNEL_IMAGE_SIZE; ++ for (; addr < end; addr += PMD_SIZE) { ++ pgd = pgd_offset_k(addr); ++ pud = pud_offset(pgd, addr); ++ pmd = pmd_offset(pud, addr); ++ if ((unsigned long)__va(__pa(_text)) <= addr && addr < (unsigned long)__va(__pa(_sdata))) ++ set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_RW)); ++ else ++ set_pmd(pmd, __pmd(pmd_val(*pmd) | (_PAGE_NX & __supported_pte_mask))); ++ } ++#endif ++ ++ flush_tlb_all(); ++#endif ++ + free_init_pages("unused kernel memory", + (unsigned long)(&__init_begin), + (unsigned long)(&__init_end)); +diff -urNp linux-2.6.31.1/arch/x86/mm/iomap_32.c linux-2.6.31.1/arch/x86/mm/iomap_32.c +--- linux-2.6.31.1/arch/x86/mm/iomap_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/iomap_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -37,12 +37,26 @@ void *kmap_atomic_prot_pfn(unsigned long + enum fixed_addresses idx; + unsigned long vaddr; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + pagefault_disable(); + + debug_kmap_atomic(type); + idx = type + KM_TYPE_NR * smp_processor_id(); + vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + set_pte(kmap_pte - idx, pfn_pte(pfn, prot)); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + arch_flush_lazy_mmu_mode(); + + return (void *)vaddr; +diff -urNp linux-2.6.31.1/arch/x86/mm/ioremap.c linux-2.6.31.1/arch/x86/mm/ioremap.c +--- linux-2.6.31.1/arch/x86/mm/ioremap.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/ioremap.c 2009-10-01 20:12:42.000000000 -0400 +@@ -111,8 +111,8 @@ int page_is_ram(unsigned long pagenr) + * Second special case: Some BIOSen report the PC BIOS + * area (640->1Mb) as ram even though it is not. + */ +- if (pagenr >= (BIOS_BEGIN >> PAGE_SHIFT) && +- pagenr < (BIOS_END >> PAGE_SHIFT)) ++ if (pagenr >= (ISA_START_ADDRESS >> PAGE_SHIFT) && ++ pagenr < (ISA_END_ADDRESS >> PAGE_SHIFT)) + return 0; + + for (i = 0; i < e820.nr_map; i++) { +@@ -207,10 +207,7 @@ static void __iomem *__ioremap_caller(re + /* + * Don't allow anybody to remap normal RAM that we're using.. + */ +- for (pfn = phys_addr >> PAGE_SHIFT; +- (pfn << PAGE_SHIFT) < (last_addr & PAGE_MASK); +- pfn++) { +- ++ for (pfn = phys_addr >> PAGE_SHIFT; ((resource_size_t)pfn << PAGE_SHIFT) < (last_addr & PAGE_MASK); pfn++) { + int is_ram = page_is_ram(pfn); + + if (is_ram && pfn_valid(pfn) && !PageReserved(pfn_to_page(pfn))) +@@ -272,6 +269,8 @@ static void __iomem *__ioremap_caller(re + break; + } + ++ prot = canon_pgprot(prot); ++ + /* + * Ok, go for it.. + */ +@@ -489,7 +488,7 @@ static int __init early_ioremap_debug_se + early_param("early_ioremap_debug", early_ioremap_debug_setup); + + static __initdata int after_paging_init; +-static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)] __page_aligned_bss; ++static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)] __read_only __aligned(PAGE_SIZE); + + static inline pmd_t * __init early_ioremap_pmd(unsigned long addr) + { +@@ -521,8 +520,7 @@ void __init early_ioremap_init(void) + slot_virt[i] = __fix_to_virt(FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*i); + + pmd = early_ioremap_pmd(fix_to_virt(FIX_BTMAP_BEGIN)); +- memset(bm_pte, 0, sizeof(bm_pte)); +- pmd_populate_kernel(&init_mm, pmd, bm_pte); ++ pmd_populate_user(&init_mm, pmd, bm_pte); + + /* + * The boot-ioremap range spans multiple pmds, for which +diff -urNp linux-2.6.31.1/arch/x86/mm/mmap.c linux-2.6.31.1/arch/x86/mm/mmap.c +--- linux-2.6.31.1/arch/x86/mm/mmap.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/mmap.c 2009-10-01 20:12:42.000000000 -0400 +@@ -36,7 +36,7 @@ + * Leave an at least ~128 MB hole. + */ + #define MIN_GAP (128*1024*1024) +-#define MAX_GAP (TASK_SIZE/6*5) ++#define MAX_GAP (pax_task_size/6*5) + + /* + * True on X86_32 or when emulating IA32 on X86_64 +@@ -81,27 +81,40 @@ static unsigned long mmap_rnd(void) + return rnd << PAGE_SHIFT; + } + +-static unsigned long mmap_base(void) ++static unsigned long mmap_base(struct mm_struct *mm) + { + unsigned long gap = current->signal->rlim[RLIMIT_STACK].rlim_cur; ++ unsigned long pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif + + if (gap < MIN_GAP) + gap = MIN_GAP; + else if (gap > MAX_GAP) + gap = MAX_GAP; + +- return PAGE_ALIGN(TASK_SIZE - gap - mmap_rnd()); ++ return PAGE_ALIGN(pax_task_size - gap - mmap_rnd()); + } + + /* + * Bottom-up (legacy) layout on X86_32 did not support randomization, X86_64 + * does, but not when emulating X86_32 + */ +-static unsigned long mmap_legacy_base(void) ++static unsigned long mmap_legacy_base(struct mm_struct *mm) + { +- if (mmap_is_ia32()) ++ if (mmap_is_ia32()) { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ return SEGMEXEC_TASK_UNMAPPED_BASE; ++ else ++#endif ++ + return TASK_UNMAPPED_BASE; +- else ++ } else + return TASK_UNMAPPED_BASE + mmap_rnd(); + } + +@@ -112,11 +125,23 @@ static unsigned long mmap_legacy_base(vo + void arch_pick_mmap_layout(struct mm_struct *mm) + { + if (mmap_is_legacy()) { +- mm->mmap_base = mmap_legacy_base(); ++ mm->mmap_base = mmap_legacy_base(mm); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area; + mm->unmap_area = arch_unmap_area; + } else { +- mm->mmap_base = mmap_base(); ++ mm->mmap_base = mmap_base(mm); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base -= mm->delta_mmap + mm->delta_stack; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area_topdown; + mm->unmap_area = arch_unmap_area_topdown; + } +diff -urNp linux-2.6.31.1/arch/x86/mm/numa_32.c linux-2.6.31.1/arch/x86/mm/numa_32.c +--- linux-2.6.31.1/arch/x86/mm/numa_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/numa_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -98,7 +98,6 @@ unsigned long node_memmap_size_bytes(int + } + #endif + +-extern unsigned long find_max_low_pfn(void); + extern unsigned long highend_pfn, highstart_pfn; + + #define LARGE_PAGE_BYTES (PTRS_PER_PTE * PAGE_SIZE) +diff -urNp linux-2.6.31.1/arch/x86/mm/pageattr.c linux-2.6.31.1/arch/x86/mm/pageattr.c +--- linux-2.6.31.1/arch/x86/mm/pageattr.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/pageattr.c 2009-10-01 20:12:42.000000000 -0400 +@@ -22,6 +22,7 @@ + #include + #include + #include ++#include + + /* + * The current flushing context - we pass it instead of 5 arguments: +@@ -266,9 +267,10 @@ static inline pgprot_t static_protection + * Does not cover __inittext since that is gone later on. On + * 64bit we do not enforce !NX on the low mapping + */ +- if (within(address, (unsigned long)_text, (unsigned long)_etext)) ++ if (within(address, ktla_ktva((unsigned long)_text), ktla_ktva((unsigned long)_etext))) + pgprot_val(forbidden) |= _PAGE_NX; + ++#ifdef CONFIG_DEBUG_RODATA + /* + * The .rodata section needs to be read-only. Using the pfn + * catches all aliases. +@@ -276,6 +278,7 @@ static inline pgprot_t static_protection + if (within(pfn, __pa((unsigned long)__start_rodata) >> PAGE_SHIFT, + __pa((unsigned long)__end_rodata) >> PAGE_SHIFT)) + pgprot_val(forbidden) |= _PAGE_RW; ++#endif + + prot = __pgprot(pgprot_val(prot) & ~pgprot_val(forbidden)); + +@@ -328,8 +331,20 @@ EXPORT_SYMBOL_GPL(lookup_address); + */ + static void __set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++ ++ pax_open_kernel(cr0); ++#endif ++ + /* change init_mm */ + set_pte_atomic(kpte, pte); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + #ifdef CONFIG_X86_32 + if (!SHARED_KERNEL_PMD) { + struct page *page; +diff -urNp linux-2.6.31.1/arch/x86/mm/pageattr-test.c linux-2.6.31.1/arch/x86/mm/pageattr-test.c +--- linux-2.6.31.1/arch/x86/mm/pageattr-test.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/pageattr-test.c 2009-10-01 20:12:42.000000000 -0400 +@@ -36,7 +36,7 @@ enum { + + static int pte_testbit(pte_t pte) + { +- return pte_flags(pte) & _PAGE_UNUSED1; ++ return pte_flags(pte) & _PAGE_CPA_TEST; + } + + struct split_state { +diff -urNp linux-2.6.31.1/arch/x86/mm/pat.c linux-2.6.31.1/arch/x86/mm/pat.c +--- linux-2.6.31.1/arch/x86/mm/pat.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/pat.c 2009-10-01 20:12:42.000000000 -0400 +@@ -213,7 +213,7 @@ chk_conflict(struct memtype *new, struct + + conflict: + printk(KERN_INFO "%s:%d conflicting memory types " +- "%Lx-%Lx %s<->%s\n", current->comm, current->pid, new->start, ++ "%Lx-%Lx %s<->%s\n", current->comm, task_pid_nr(current), new->start, + new->end, cattr_name(new->type), cattr_name(entry->type)); + return -EBUSY; + } +@@ -487,7 +487,7 @@ int free_memtype(u64 start, u64 end) + + if (err) { + printk(KERN_INFO "%s:%d freeing invalid memtype %Lx-%Lx\n", +- current->comm, current->pid, start, end); ++ current->comm, task_pid_nr(current), start, end); + } + + dprintk("free_memtype request 0x%Lx-0x%Lx\n", start, end); +@@ -588,7 +588,7 @@ int kernel_map_sync_memtype(u64 base, un + printk(KERN_INFO + "%s:%d ioremap_change_attr failed %s " + "for %Lx-%Lx\n", +- current->comm, current->pid, ++ current->comm, task_pid_nr(current), + cattr_name(flags), + base, (unsigned long long)(base + size)); + return -EINVAL; +@@ -628,7 +628,7 @@ static int reserve_pfn_range(u64 paddr, + free_memtype(paddr, paddr + size); + printk(KERN_ERR "%s:%d map pfn expected mapping type %s" + " for %Lx-%Lx, got %s\n", +- current->comm, current->pid, ++ current->comm, task_pid_nr(current), + cattr_name(want_flags), + (unsigned long long)paddr, + (unsigned long long)(paddr + size), +@@ -827,7 +827,7 @@ static int memtype_seq_show(struct seq_f + return 0; + } + +-static struct seq_operations memtype_seq_ops = { ++static const struct seq_operations memtype_seq_ops = { + .start = memtype_seq_start, + .next = memtype_seq_next, + .stop = memtype_seq_stop, +diff -urNp linux-2.6.31.1/arch/x86/mm/pgtable_32.c linux-2.6.31.1/arch/x86/mm/pgtable_32.c +--- linux-2.6.31.1/arch/x86/mm/pgtable_32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/pgtable_32.c 2009-10-01 20:12:42.000000000 -0400 +@@ -33,6 +33,10 @@ void set_pte_vaddr(unsigned long vaddr, + pmd_t *pmd; + pte_t *pte; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + pgd = swapper_pg_dir + pgd_index(vaddr); + if (pgd_none(*pgd)) { + BUG(); +@@ -49,11 +53,20 @@ void set_pte_vaddr(unsigned long vaddr, + return; + } + pte = pte_offset_kernel(pmd, vaddr); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + if (pte_val(pteval)) + set_pte_at(&init_mm, vaddr, pte, pteval); + else + pte_clear(&init_mm, vaddr, pte); + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + /* + * It's enough to flush this one mapping. + * (PGE mappings get flushed as well) +diff -urNp linux-2.6.31.1/arch/x86/mm/tlb.c linux-2.6.31.1/arch/x86/mm/tlb.c +--- linux-2.6.31.1/arch/x86/mm/tlb.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/mm/tlb.c 2009-10-01 20:12:42.000000000 -0400 +@@ -12,7 +12,7 @@ + #include + + DEFINE_PER_CPU_SHARED_ALIGNED(struct tlb_state, cpu_tlbstate) +- = { &init_mm, 0, }; ++ = { &init_mm, 0 }; + + /* + * Smarter SMP flushing macros. +diff -urNp linux-2.6.31.1/arch/x86/oprofile/backtrace.c linux-2.6.31.1/arch/x86/oprofile/backtrace.c +--- linux-2.6.31.1/arch/x86/oprofile/backtrace.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/oprofile/backtrace.c 2009-10-01 20:12:42.000000000 -0400 +@@ -37,7 +37,7 @@ static void backtrace_address(void *data + unsigned int *depth = data; + + if ((*depth)--) +- oprofile_add_trace(addr); ++ oprofile_add_trace(ktla_ktva(addr)); + } + + static struct stacktrace_ops backtrace_ops = { +@@ -77,7 +77,7 @@ x86_backtrace(struct pt_regs * const reg + { + struct frame_head *head = (struct frame_head *)frame_pointer(regs); + +- if (!user_mode_vm(regs)) { ++ if (!user_mode(regs)) { + unsigned long stack = kernel_stack_pointer(regs); + if (depth) + dump_trace(NULL, regs, (unsigned long *)stack, 0, +diff -urNp linux-2.6.31.1/arch/x86/oprofile/op_model_p4.c linux-2.6.31.1/arch/x86/oprofile/op_model_p4.c +--- linux-2.6.31.1/arch/x86/oprofile/op_model_p4.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/oprofile/op_model_p4.c 2009-10-01 20:12:42.000000000 -0400 +@@ -48,7 +48,7 @@ static inline void setup_num_counters(vo + #endif + } + +-static int inline addr_increment(void) ++static inline int addr_increment(void) + { + #ifdef CONFIG_SMP + return smp_num_siblings == 2 ? 2 : 1; +diff -urNp linux-2.6.31.1/arch/x86/pci/common.c linux-2.6.31.1/arch/x86/pci/common.c +--- linux-2.6.31.1/arch/x86/pci/common.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/pci/common.c 2009-10-01 20:12:42.000000000 -0400 +@@ -370,7 +370,7 @@ static const struct dmi_system_id __devi + DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL585 G2"), + }, + }, +- {} ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL} + }; + + void __init dmi_check_pciprobe(void) +diff -urNp linux-2.6.31.1/arch/x86/pci/fixup.c linux-2.6.31.1/arch/x86/pci/fixup.c +--- linux-2.6.31.1/arch/x86/pci/fixup.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/pci/fixup.c 2009-10-01 20:12:42.000000000 -0400 +@@ -364,7 +364,7 @@ static const struct dmi_system_id __devi + DMI_MATCH(DMI_PRODUCT_NAME, "MS-6702E"), + }, + }, +- {} ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + /* +@@ -435,7 +435,7 @@ static const struct dmi_system_id __devi + DMI_MATCH(DMI_PRODUCT_VERSION, "PSA40U"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + static void __devinit pci_pre_fixup_toshiba_ohci1394(struct pci_dev *dev) +diff -urNp linux-2.6.31.1/arch/x86/pci/i386.c linux-2.6.31.1/arch/x86/pci/i386.c +--- linux-2.6.31.1/arch/x86/pci/i386.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/pci/i386.c 2009-10-01 20:12:42.000000000 -0400 +@@ -266,7 +266,7 @@ void pcibios_set_master(struct pci_dev * + pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat); + } + +-static struct vm_operations_struct pci_mmap_ops = { ++static const struct vm_operations_struct pci_mmap_ops = { + .access = generic_access_phys, + }; + +diff -urNp linux-2.6.31.1/arch/x86/pci/irq.c linux-2.6.31.1/arch/x86/pci/irq.c +--- linux-2.6.31.1/arch/x86/pci/irq.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/pci/irq.c 2009-10-01 20:12:42.000000000 -0400 +@@ -543,7 +543,7 @@ static __init int intel_router_probe(str + static struct pci_device_id __initdata pirq_440gx[] = { + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_0) }, + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_2) }, +- { }, ++ { PCI_DEVICE(0, 0) } + }; + + /* 440GX has a proprietary PIRQ router -- don't use it */ +@@ -1107,7 +1107,7 @@ static struct dmi_system_id __initdata p + DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 360"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + int __init pcibios_irq_init(void) +diff -urNp linux-2.6.31.1/arch/x86/pci/pcbios.c linux-2.6.31.1/arch/x86/pci/pcbios.c +--- linux-2.6.31.1/arch/x86/pci/pcbios.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/pci/pcbios.c 2009-10-01 20:12:42.000000000 -0400 +@@ -56,50 +56,120 @@ union bios32 { + static struct { + unsigned long address; + unsigned short segment; +-} bios32_indirect = { 0, __KERNEL_CS }; ++} bios32_indirect __read_only = { 0, __PCIBIOS_CS }; + + /* + * Returns the entry point for the given service, NULL on error + */ + +-static unsigned long bios32_service(unsigned long service) ++static unsigned long __devinit bios32_service(unsigned long service) + { + unsigned char return_code; /* %al */ + unsigned long address; /* %ebx */ + unsigned long length; /* %ecx */ + unsigned long entry; /* %edx */ + unsigned long flags; ++ struct desc_struct d, *gdt; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif + + local_irq_save(flags); +- __asm__("lcall *(%%edi); cld" ++ ++ gdt = get_cpu_gdt_table(smp_processor_id()); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ ++ pack_descriptor(&d, 0UL, 0xFFFFFUL, 0x9B, 0xC); ++ write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_CS, &d, DESCTYPE_S); ++ pack_descriptor(&d, 0UL, 0xFFFFFUL, 0x93, 0xC); ++ write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_DS, &d, DESCTYPE_S); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ ++ __asm__("movw %w7, %%ds; lcall *(%%edi); push %%ss; pop %%ds; cld" + : "=a" (return_code), + "=b" (address), + "=c" (length), + "=d" (entry) + : "0" (service), + "1" (0), +- "D" (&bios32_indirect)); ++ "D" (&bios32_indirect), ++ "r"(__PCIBIOS_DS) ++ : "memory"); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ ++ gdt[GDT_ENTRY_PCIBIOS_CS].a = 0; ++ gdt[GDT_ENTRY_PCIBIOS_CS].b = 0; ++ gdt[GDT_ENTRY_PCIBIOS_DS].a = 0; ++ gdt[GDT_ENTRY_PCIBIOS_DS].b = 0; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + local_irq_restore(flags); + + switch (return_code) { +- case 0: +- return address + entry; +- case 0x80: /* Not present */ +- printk(KERN_WARNING "bios32_service(0x%lx): not present\n", service); +- return 0; +- default: /* Shouldn't happen */ +- printk(KERN_WARNING "bios32_service(0x%lx): returned 0x%x -- BIOS bug!\n", +- service, return_code); ++ case 0: { ++ int cpu; ++ unsigned char flags; ++ ++ printk(KERN_INFO "bios32_service: base:%08lx length:%08lx entry:%08lx\n", address, length, entry); ++ if (address >= 0xFFFF0 || length > 0x100000 - address || length <= entry) { ++ printk(KERN_WARNING "bios32_service: not valid\n"); + return 0; ++ } ++ address = address + PAGE_OFFSET; ++ length += 16UL; /* some BIOSs underreport this... */ ++ flags = 4; ++ if (length >= 64*1024*1024) { ++ length >>= PAGE_SHIFT; ++ flags |= 8; ++ } ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ ++ for (cpu = 0; cpu < NR_CPUS; cpu++) { ++ gdt = get_cpu_gdt_table(cpu); ++ pack_descriptor(&d, address, length, 0x9b, flags); ++ write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_CS, &d, DESCTYPE_S); ++ pack_descriptor(&d, address, length, 0x93, flags); ++ write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_DS, &d, DESCTYPE_S); ++ } ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ ++ return entry; ++ } ++ case 0x80: /* Not present */ ++ printk(KERN_WARNING "bios32_service(0x%lx): not present\n", service); ++ return 0; ++ default: /* Shouldn't happen */ ++ printk(KERN_WARNING "bios32_service(0x%lx): returned 0x%x -- BIOS bug!\n", ++ service, return_code); ++ return 0; + } + } + + static struct { + unsigned long address; + unsigned short segment; +-} pci_indirect = { 0, __KERNEL_CS }; ++} pci_indirect __read_only = { 0, __PCIBIOS_CS }; + +-static int pci_bios_present; ++static int pci_bios_present __read_only; + + static int __devinit check_pcibios(void) + { +@@ -108,11 +178,13 @@ static int __devinit check_pcibios(void) + unsigned long flags, pcibios_entry; + + if ((pcibios_entry = bios32_service(PCI_SERVICE))) { +- pci_indirect.address = pcibios_entry + PAGE_OFFSET; ++ pci_indirect.address = pcibios_entry; + + local_irq_save(flags); +- __asm__( +- "lcall *(%%edi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%edi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -121,7 +193,8 @@ static int __devinit check_pcibios(void) + "=b" (ebx), + "=c" (ecx) + : "1" (PCIBIOS_PCI_BIOS_PRESENT), +- "D" (&pci_indirect) ++ "D" (&pci_indirect), ++ "r" (__PCIBIOS_DS) + : "memory"); + local_irq_restore(flags); + +@@ -165,7 +238,10 @@ static int pci_bios_read(unsigned int se + + switch (len) { + case 1: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -174,7 +250,8 @@ static int pci_bios_read(unsigned int se + : "1" (PCIBIOS_READ_CONFIG_BYTE), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + /* + * Zero-extend the result beyond 8 bits, do not trust the + * BIOS having done it: +@@ -182,7 +259,10 @@ static int pci_bios_read(unsigned int se + *value &= 0xff; + break; + case 2: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -191,7 +271,8 @@ static int pci_bios_read(unsigned int se + : "1" (PCIBIOS_READ_CONFIG_WORD), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + /* + * Zero-extend the result beyond 16 bits, do not trust the + * BIOS having done it: +@@ -199,7 +280,10 @@ static int pci_bios_read(unsigned int se + *value &= 0xffff; + break; + case 4: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -208,7 +292,8 @@ static int pci_bios_read(unsigned int se + : "1" (PCIBIOS_READ_CONFIG_DWORD), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + break; + } + +@@ -231,7 +316,10 @@ static int pci_bios_write(unsigned int s + + switch (len) { + case 1: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -240,10 +328,14 @@ static int pci_bios_write(unsigned int s + "c" (value), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + break; + case 2: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -252,10 +344,14 @@ static int pci_bios_write(unsigned int s + "c" (value), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + break; + case 4: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -264,7 +360,8 @@ static int pci_bios_write(unsigned int s + "c" (value), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + break; + } + +@@ -368,10 +465,13 @@ struct irq_routing_table * pcibios_get_i + + DBG("PCI: Fetching IRQ routing table... "); + __asm__("push %%es\n\t" ++ "movw %w8, %%ds\n\t" + "push %%ds\n\t" + "pop %%es\n\t" +- "lcall *(%%esi); cld\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" + "pop %%es\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -382,7 +482,8 @@ struct irq_routing_table * pcibios_get_i + "1" (0), + "D" ((long) &opt), + "S" (&pci_indirect), +- "m" (opt) ++ "m" (opt), ++ "r" (__PCIBIOS_DS) + : "memory"); + DBG("OK ret=%d, size=%d, map=%x\n", ret, opt.size, map); + if (ret & 0xff00) +@@ -406,7 +507,10 @@ int pcibios_set_irq_routing(struct pci_d + { + int ret; + +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w5, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -414,7 +518,8 @@ int pcibios_set_irq_routing(struct pci_d + : "0" (PCIBIOS_SET_PCI_HW_INT), + "b" ((dev->bus->number << 8) | dev->devfn), + "c" ((irq << 8) | (pin + 10)), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + return !(ret & 0xff00); + } + EXPORT_SYMBOL(pcibios_set_irq_routing); +diff -urNp linux-2.6.31.1/arch/x86/power/cpu.c linux-2.6.31.1/arch/x86/power/cpu.c +--- linux-2.6.31.1/arch/x86/power/cpu.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/power/cpu.c 2009-10-01 20:12:42.000000000 -0400 +@@ -126,7 +126,11 @@ static void do_fpu_end(void) + static void fix_processor_context(void) + { + int cpu = smp_processor_id(); +- struct tss_struct *t = &per_cpu(init_tss, cpu); ++ struct tss_struct *t = init_tss + cpu; ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_KERNEXEC) ++ unsigned long cr0; ++#endif + + set_tss_desc(cpu, t); /* + * This just modifies memory; should not be +@@ -136,8 +140,17 @@ static void fix_processor_context(void) + */ + + #ifdef CONFIG_X86_64 ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + get_cpu_gdt_table(cpu)[GDT_ENTRY_TSS].type = 9; + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + syscall_init(); /* This sets MSR_*STAR and related */ + #endif + load_TR_desc(); /* This does ltr */ +diff -urNp linux-2.6.31.1/arch/x86/vdso/Makefile linux-2.6.31.1/arch/x86/vdso/Makefile +--- linux-2.6.31.1/arch/x86/vdso/Makefile 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/vdso/Makefile 2009-10-01 20:12:42.000000000 -0400 +@@ -122,7 +122,7 @@ quiet_cmd_vdso = VDSO $@ + $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ + -Wl,-T,$(filter %.lds,$^) $(filter %.o,$^) + +-VDSO_LDFLAGS = -fPIC -shared $(call ld-option, -Wl$(comma)--hash-style=sysv) ++VDSO_LDFLAGS = -fPIC -shared --no-undefined $(call ld-option, -Wl$(comma)--hash-style=sysv) + GCOV_PROFILE := n + + # +diff -urNp linux-2.6.31.1/arch/x86/vdso/vclock_gettime.c linux-2.6.31.1/arch/x86/vdso/vclock_gettime.c +--- linux-2.6.31.1/arch/x86/vdso/vclock_gettime.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/vdso/vclock_gettime.c 2009-10-01 20:12:42.000000000 -0400 +@@ -22,24 +22,48 @@ + #include + #include + #include ++#include + #include "vextern.h" + + #define gtod vdso_vsyscall_gtod_data + ++notrace noinline long __vdso_fallback_time(long *t) ++{ ++ long secs; ++ asm volatile("syscall" ++ : "=a" (secs) ++ : "0" (__NR_time),"D" (t) : "r11", "cx", "memory"); ++ return secs; ++} ++ + notrace static long vdso_fallback_gettime(long clock, struct timespec *ts) + { + long ret; + asm("syscall" : "=a" (ret) : +- "0" (__NR_clock_gettime),"D" (clock), "S" (ts) : "memory"); ++ "0" (__NR_clock_gettime),"D" (clock), "S" (ts) : "r11", "cx", "memory"); + return ret; + } + ++notrace static inline cycle_t __vdso_vread_hpet(void) ++{ ++ return readl((const void __iomem *)fix_to_virt(VSYSCALL_HPET) + 0xf0); ++} ++ ++notrace static inline cycle_t __vdso_vread_tsc(void) ++{ ++ cycle_t ret = (cycle_t)vget_cycles(); ++ ++ return ret >= gtod->clock.cycle_last ? ret : gtod->clock.cycle_last; ++} ++ + notrace static inline long vgetns(void) + { + long v; +- cycles_t (*vread)(void); +- vread = gtod->clock.vread; +- v = (vread() - gtod->clock.cycle_last) & gtod->clock.mask; ++ if (gtod->clock.name[0] == 't' && gtod->clock.name[1] == 's' && gtod->clock.name[2] == 'c' && !gtod->clock.name[3]) ++ v = __vdso_vread_tsc(); ++ else ++ v = __vdso_vread_hpet(); ++ v = (v - gtod->clock.cycle_last) & gtod->clock.mask; + return (v * gtod->clock.mult) >> gtod->clock.shift; + } + +@@ -88,7 +112,9 @@ notrace static noinline int do_monotonic + + notrace int __vdso_clock_gettime(clockid_t clock, struct timespec *ts) + { +- if (likely(gtod->sysctl_enabled && gtod->clock.vread)) ++ if (likely(gtod->sysctl_enabled && ++ ((gtod->clock.name[0] == 'h' && gtod->clock.name[1] == 'p' && gtod->clock.name[2] == 'e' && gtod->clock.name[3] == 't' && !gtod->clock.name[4]) || ++ (gtod->clock.name[0] == 't' && gtod->clock.name[1] == 's' && gtod->clock.name[2] == 'c' && !gtod->clock.name[3])))) + switch (clock) { + case CLOCK_REALTIME: + return do_realtime(ts); +@@ -100,10 +126,20 @@ notrace int __vdso_clock_gettime(clockid + int clock_gettime(clockid_t, struct timespec *) + __attribute__((weak, alias("__vdso_clock_gettime"))); + +-notrace int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz) ++notrace noinline int __vdso_fallback_gettimeofday(struct timeval *tv, struct timezone *tz) + { + long ret; +- if (likely(gtod->sysctl_enabled && gtod->clock.vread)) { ++ asm("syscall" : "=a" (ret) : ++ "0" (__NR_gettimeofday), "D" (tv), "S" (tz) : "r11", "cx", "memory"); ++ return ret; ++} ++ ++notrace int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz) ++{ ++ if (likely(gtod->sysctl_enabled && ++ ((gtod->clock.name[0] == 'h' && gtod->clock.name[1] == 'p' && gtod->clock.name[2] == 'e' && gtod->clock.name[3] == 't' && !gtod->clock.name[4]) || ++ (gtod->clock.name[0] == 't' && gtod->clock.name[1] == 's' && gtod->clock.name[2] == 'c' && !gtod->clock.name[3])))) ++ { + if (likely(tv != NULL)) { + BUILD_BUG_ON(offsetof(struct timeval, tv_usec) != + offsetof(struct timespec, tv_nsec) || +@@ -118,9 +154,7 @@ notrace int __vdso_gettimeofday(struct t + } + return 0; + } +- asm("syscall" : "=a" (ret) : +- "0" (__NR_gettimeofday), "D" (tv), "S" (tz) : "memory"); +- return ret; ++ return __vdso_fallback_gettimeofday(tv, tz); + } + int gettimeofday(struct timeval *, struct timezone *) + __attribute__((weak, alias("__vdso_gettimeofday"))); +diff -urNp linux-2.6.31.1/arch/x86/vdso/vdso32-setup.c linux-2.6.31.1/arch/x86/vdso/vdso32-setup.c +--- linux-2.6.31.1/arch/x86/vdso/vdso32-setup.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/vdso/vdso32-setup.c 2009-10-01 20:12:42.000000000 -0400 +@@ -25,6 +25,7 @@ + #include + #include + #include ++#include + + enum { + VDSO_DISABLED = 0, +@@ -226,7 +227,7 @@ static inline void map_compat_vdso(int m + void enable_sep_cpu(void) + { + int cpu = get_cpu(); +- struct tss_struct *tss = &per_cpu(init_tss, cpu); ++ struct tss_struct *tss = init_tss + cpu; + + if (!boot_cpu_has(X86_FEATURE_SEP)) { + put_cpu(); +@@ -249,7 +250,7 @@ static int __init gate_vma_init(void) + gate_vma.vm_start = FIXADDR_USER_START; + gate_vma.vm_end = FIXADDR_USER_END; + gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; +- gate_vma.vm_page_prot = __P101; ++ gate_vma.vm_page_prot = vm_get_page_prot(gate_vma.vm_flags); + /* + * Make sure the vDSO gets into every core dump. + * Dumping its contents makes post-mortem fully interpretable later +@@ -331,14 +332,14 @@ int arch_setup_additional_pages(struct l + if (compat) + addr = VDSO_HIGH_BASE; + else { +- addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0); ++ addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, MAP_EXECUTABLE); + if (IS_ERR_VALUE(addr)) { + ret = addr; + goto up_fail; + } + } + +- current->mm->context.vdso = (void *)addr; ++ current->mm->context.vdso = addr; + + if (compat_uses_vma || !compat) { + /* +@@ -365,7 +366,7 @@ int arch_setup_additional_pages(struct l + + up_fail: + if (ret) +- current->mm->context.vdso = NULL; ++ current->mm->context.vdso = 0; + + up_write(&mm->mmap_sem); + +@@ -388,7 +389,7 @@ static ctl_table abi_table2[] = { + .mode = 0644, + .proc_handler = proc_dointvec + }, +- {} ++ { 0, NULL, NULL, 0, 0, NULL, NULL, NULL, NULL, NULL, NULL } + }; + + static ctl_table abi_root_table2[] = { +@@ -398,7 +399,7 @@ static ctl_table abi_root_table2[] = { + .mode = 0555, + .child = abi_table2 + }, +- {} ++ { 0, NULL, NULL, 0, 0, NULL, NULL, NULL, NULL, NULL, NULL } + }; + + static __init int ia32_binfmt_init(void) +@@ -413,8 +414,14 @@ __initcall(ia32_binfmt_init); + + const char *arch_vma_name(struct vm_area_struct *vma) + { +- if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso) ++ if (vma->vm_mm && vma->vm_start == vma->vm_mm->context.vdso) + return "[vdso]"; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma->vm_mm && vma->vm_mirror && vma->vm_mirror->vm_start == vma->vm_mm->context.vdso) ++ return "[vdso]"; ++#endif ++ + return NULL; + } + +@@ -423,7 +430,7 @@ struct vm_area_struct *get_gate_vma(stru + struct mm_struct *mm = tsk->mm; + + /* Check to see if this task was created in compat vdso mode */ +- if (mm && mm->context.vdso == (void *)VDSO_HIGH_BASE) ++ if (mm && mm->context.vdso == VDSO_HIGH_BASE) + return &gate_vma; + return NULL; + } +diff -urNp linux-2.6.31.1/arch/x86/vdso/vdso.lds.S linux-2.6.31.1/arch/x86/vdso/vdso.lds.S +--- linux-2.6.31.1/arch/x86/vdso/vdso.lds.S 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/vdso/vdso.lds.S 2009-10-01 20:12:42.000000000 -0400 +@@ -35,3 +35,9 @@ VDSO64_PRELINK = VDSO_PRELINK; + #define VEXTERN(x) VDSO64_ ## x = vdso_ ## x; + #include "vextern.h" + #undef VEXTERN ++ ++#define VEXTERN(x) VDSO64_ ## x = __vdso_ ## x; ++VEXTERN(fallback_gettimeofday) ++VEXTERN(fallback_time) ++VEXTERN(getcpu) ++#undef VEXTERN +diff -urNp linux-2.6.31.1/arch/x86/vdso/vextern.h linux-2.6.31.1/arch/x86/vdso/vextern.h +--- linux-2.6.31.1/arch/x86/vdso/vextern.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/vdso/vextern.h 2009-10-01 20:12:42.000000000 -0400 +@@ -11,6 +11,5 @@ + put into vextern.h and be referenced as a pointer with vdso prefix. + The main kernel later fills in the values. */ + +-VEXTERN(jiffies) + VEXTERN(vgetcpu_mode) + VEXTERN(vsyscall_gtod_data) +diff -urNp linux-2.6.31.1/arch/x86/vdso/vma.c linux-2.6.31.1/arch/x86/vdso/vma.c +--- linux-2.6.31.1/arch/x86/vdso/vma.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/vdso/vma.c 2009-10-01 20:12:42.000000000 -0400 +@@ -57,7 +57,7 @@ static int __init init_vdso_vars(void) + if (!vbase) + goto oom; + +- if (memcmp(vbase, "\177ELF", 4)) { ++ if (memcmp(vbase, ELFMAG, SELFMAG)) { + printk("VDSO: I'm broken; not ELF\n"); + vdso_enabled = 0; + } +@@ -66,6 +66,7 @@ static int __init init_vdso_vars(void) + *(typeof(__ ## x) **) var_ref(VDSO64_SYMBOL(vbase, x), #x) = &__ ## x; + #include "vextern.h" + #undef VEXTERN ++ vunmap(vbase); + return 0; + + oom: +@@ -116,7 +117,7 @@ int arch_setup_additional_pages(struct l + goto up_fail; + } + +- current->mm->context.vdso = (void *)addr; ++ current->mm->context.vdso = addr; + + ret = install_special_mapping(mm, addr, vdso_size, + VM_READ|VM_EXEC| +@@ -124,7 +125,7 @@ int arch_setup_additional_pages(struct l + VM_ALWAYSDUMP, + vdso_pages); + if (ret) { +- current->mm->context.vdso = NULL; ++ current->mm->context.vdso = 0; + goto up_fail; + } + +@@ -132,10 +133,3 @@ up_fail: + up_write(&mm->mmap_sem); + return ret; + } +- +-static __init int vdso_setup(char *s) +-{ +- vdso_enabled = simple_strtoul(s, NULL, 0); +- return 0; +-} +-__setup("vdso=", vdso_setup); +diff -urNp linux-2.6.31.1/arch/x86/xen/debugfs.c linux-2.6.31.1/arch/x86/xen/debugfs.c +--- linux-2.6.31.1/arch/x86/xen/debugfs.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/xen/debugfs.c 2009-10-01 20:12:42.000000000 -0400 +@@ -100,7 +100,7 @@ static int xen_array_release(struct inod + return 0; + } + +-static struct file_operations u32_array_fops = { ++static const struct file_operations u32_array_fops = { + .owner = THIS_MODULE, + .open = u32_array_open, + .release= xen_array_release, +diff -urNp linux-2.6.31.1/arch/x86/xen/enlighten.c linux-2.6.31.1/arch/x86/xen/enlighten.c +--- linux-2.6.31.1/arch/x86/xen/enlighten.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/xen/enlighten.c 2009-10-01 20:12:42.000000000 -0400 +@@ -69,8 +69,6 @@ EXPORT_SYMBOL_GPL(xen_start_info); + + struct shared_info xen_dummy_shared_info; + +-void *xen_initial_gdt; +- + /* + * Point at some empty memory to start with. We map the real shared_info + * page as soon as fixmap is up and running. +@@ -490,7 +488,7 @@ static void xen_write_idt_entry(gate_des + + preempt_disable(); + +- start = __get_cpu_var(idt_desc).address; ++ start = (unsigned long)__get_cpu_var(idt_desc).address; + end = start + __get_cpu_var(idt_desc).size + 1; + + xen_mc_flush(); +@@ -1010,13 +1008,6 @@ asmlinkage void __init xen_start_kernel( + + machine_ops = xen_machine_ops; + +- /* +- * The only reliable way to retain the initial address of the +- * percpu gdt_page is to remember it here, so we can go and +- * mark it RW later, when the initial percpu area is freed. +- */ +- xen_initial_gdt = &per_cpu(gdt_page, 0); +- + xen_smp_init(); + + /* Get mfn list */ +diff -urNp linux-2.6.31.1/arch/x86/xen/mmu.c linux-2.6.31.1/arch/x86/xen/mmu.c +--- linux-2.6.31.1/arch/x86/xen/mmu.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/xen/mmu.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1707,6 +1707,8 @@ __init pgd_t *xen_setup_kernel_pagetable + convert_pfn_mfn(init_level4_pgt); + convert_pfn_mfn(level3_ident_pgt); + convert_pfn_mfn(level3_kernel_pgt); ++ convert_pfn_mfn(level3_vmalloc_pgt); ++ convert_pfn_mfn(level3_vmemmap_pgt); + + l3 = m2v(pgd[pgd_index(__START_KERNEL_map)].pgd); + l2 = m2v(l3[pud_index(__START_KERNEL_map)].pud); +@@ -1725,7 +1727,10 @@ __init pgd_t *xen_setup_kernel_pagetable + set_page_prot(init_level4_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_ident_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_kernel_pgt, PAGE_KERNEL_RO); ++ set_page_prot(level3_vmalloc_pgt, PAGE_KERNEL_RO); ++ set_page_prot(level3_vmemmap_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_user_vsyscall, PAGE_KERNEL_RO); ++ set_page_prot(level2_vmemmap_pgt, PAGE_KERNEL_RO); + set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO); + set_page_prot(level2_fixmap_pgt, PAGE_KERNEL_RO); + +diff -urNp linux-2.6.31.1/arch/x86/xen/smp.c linux-2.6.31.1/arch/x86/xen/smp.c +--- linux-2.6.31.1/arch/x86/xen/smp.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/xen/smp.c 2009-10-01 20:12:42.000000000 -0400 +@@ -167,11 +167,6 @@ static void __init xen_smp_prepare_boot_ + { + BUG_ON(smp_processor_id() != 0); + native_smp_prepare_boot_cpu(); +- +- /* We've switched to the "real" per-cpu gdt, so make sure the +- old memory can be recycled */ +- make_lowmem_page_readwrite(xen_initial_gdt); +- + xen_setup_vcpu_info_placement(); + } + +@@ -231,8 +226,8 @@ cpu_initialize_context(unsigned int cpu, + gdt = get_cpu_gdt_table(cpu); + + ctxt->flags = VGCF_IN_KERNEL; +- ctxt->user_regs.ds = __USER_DS; +- ctxt->user_regs.es = __USER_DS; ++ ctxt->user_regs.ds = __KERNEL_DS; ++ ctxt->user_regs.es = __KERNEL_DS; + ctxt->user_regs.ss = __KERNEL_DS; + #ifdef CONFIG_X86_32 + ctxt->user_regs.fs = __KERNEL_PERCPU; +diff -urNp linux-2.6.31.1/arch/x86/xen/xen-ops.h linux-2.6.31.1/arch/x86/xen/xen-ops.h +--- linux-2.6.31.1/arch/x86/xen/xen-ops.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/x86/xen/xen-ops.h 2009-10-01 20:12:42.000000000 -0400 +@@ -10,8 +10,6 @@ + extern const char xen_hypervisor_callback[]; + extern const char xen_failsafe_callback[]; + +-extern void *xen_initial_gdt; +- + struct trap_info; + void xen_copy_trap_info(struct trap_info *traps); + +diff -urNp linux-2.6.31.1/arch/xtensa/include/asm/atomic.h linux-2.6.31.1/arch/xtensa/include/asm/atomic.h +--- linux-2.6.31.1/arch/xtensa/include/asm/atomic.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/arch/xtensa/include/asm/atomic.h 2009-10-01 20:12:42.000000000 -0400 +@@ -49,6 +49,14 @@ + #define atomic_read(v) ((v)->counter) + + /** ++ * atomic_read_unchecked - read atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically reads the value of @v. ++ */ ++#define atomic_read_unchecked(v) ((v)->counter) ++ ++/** + * atomic_set - set atomic variable + * @v: pointer of type atomic_t + * @i: required value +@@ -58,6 +66,15 @@ + #define atomic_set(v,i) ((v)->counter = (i)) + + /** ++ * atomic_set_unchecked - set atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * @i: required value ++ * ++ * Atomically sets the value of @v to @i. ++ */ ++#define atomic_set_unchecked(v,i) ((v)->counter = (i)) ++ ++/** + * atomic_add - add integer to atomic variable + * @i: integer value to add + * @v: pointer of type atomic_t +@@ -81,6 +98,11 @@ static inline void atomic_add(int i, ato + ); + } + ++static inline void atomic_add_unchecked(int i, atomic_unchecked_t * v) ++{ ++ atomic_add(i, (atomic_t *)v); ++} ++ + /** + * atomic_sub - subtract the atomic variable + * @i: integer value to subtract +@@ -105,6 +127,11 @@ static inline void atomic_sub(int i, ato + ); + } + ++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ atomic_sub(i, (atomic_t *)v); ++} ++ + /* + * We use atomic_{add|sub}_return to define other functions. + */ +@@ -165,6 +192,7 @@ static inline int atomic_sub_return(int + * Atomically increments @v by 1. + */ + #define atomic_inc(v) atomic_add(1,(v)) ++#define atomic_inc_unchecked(v) atomic_add_unchecked(1,(v)) + + /** + * atomic_inc - increment atomic variable +diff -urNp linux-2.6.31.1/crypto/lrw.c linux-2.6.31.1/crypto/lrw.c +--- linux-2.6.31.1/crypto/lrw.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/crypto/lrw.c 2009-10-01 20:12:42.000000000 -0400 +@@ -60,7 +60,7 @@ static int setkey(struct crypto_tfm *par + struct priv *ctx = crypto_tfm_ctx(parent); + struct crypto_cipher *child = ctx->child; + int err, i; +- be128 tmp = { 0 }; ++ be128 tmp = { 0, 0 }; + int bsize = crypto_cipher_blocksize(child); + + crypto_cipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); +diff -urNp linux-2.6.31.1/Documentation/dontdiff linux-2.6.31.1/Documentation/dontdiff +--- linux-2.6.31.1/Documentation/dontdiff 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/Documentation/dontdiff 2009-10-01 20:12:42.000000000 -0400 +@@ -3,6 +3,7 @@ + *.bin + *.cpio + *.csp ++*.dbg + *.dsp + *.dvi + *.elf +@@ -49,11 +50,16 @@ + 53c700_d.h + CVS + ChangeSet ++GPATH ++GRTAGS ++GSYMS ++GTAGS + Image + Kerntypes + Module.markers + Module.symvers + PENDING ++PERF* + SCCS + System.map* + TAGS +@@ -76,7 +82,9 @@ btfixupprep + build + bvmlinux + bzImage* ++capflags.c + classlist.h* ++common-cmds.h + comp*.log + compile.h* + conf +@@ -103,13 +111,14 @@ gen_crc32table + gen_init_cpio + genksyms + *_gray256.c ++hash + ihex2fw + ikconfig.h* + initramfs_data.cpio ++initramfs_data.cpio.bz2 + initramfs_data.cpio.gz + initramfs_list + kallsyms +-kconfig + keywords.c + ksym.c* + ksym.h* +@@ -133,6 +142,7 @@ mkboot + mkbugboot + mkcpustr + mkdep ++mkpiggy + mkprep + mktables + mktree +@@ -149,6 +159,7 @@ patches* + pca200e.bin + pca200e_ecd.bin2 + piggy.gz ++piggy.S + piggyback + pnmtologo + ppc_defs.h* +@@ -164,6 +175,7 @@ setup + setup.bin + setup.elf + sImage ++slabinfo + sm_tbl* + split-include + syscalltab.h +@@ -187,14 +199,20 @@ version.h* + vmlinux + vmlinux-* + vmlinux.aout ++vmlinux.bin.all ++vmlinux.bin.bz2 + vmlinux.lds ++vmlinux.relocs ++voffset.h + vsyscall.lds + vsyscall_32.lds + wanxlfw.inc + uImage + unifdef ++utsrelease.h + wakeup.bin + wakeup.elf + wakeup.lds + zImage* + zconf.hash.c ++zoffset.h +diff -urNp linux-2.6.31.1/Documentation/kernel-parameters.txt linux-2.6.31.1/Documentation/kernel-parameters.txt +--- linux-2.6.31.1/Documentation/kernel-parameters.txt 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/Documentation/kernel-parameters.txt 2009-10-01 20:12:42.000000000 -0400 +@@ -1776,6 +1776,12 @@ and is between 256 and 4096 characters. + the specified number of seconds. This is to be used if + your oopses keep scrolling off the screen. + ++ pax_nouderef [X86-32] disables UDEREF. Most likely needed under certain ++ virtualization environments that don't cope well with the ++ expand down segment used by UDEREF on X86-32. ++ ++ pax_softmode= [X86-32] 0/1 to disable/enable PaX softmode on boot already. ++ + pcbit= [HW,ISDN] + + pcd. [PARIDE] +diff -urNp linux-2.6.31.1/drivers/acpi/blacklist.c linux-2.6.31.1/drivers/acpi/blacklist.c +--- linux-2.6.31.1/drivers/acpi/blacklist.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/acpi/blacklist.c 2009-10-01 20:12:42.000000000 -0400 +@@ -71,7 +71,7 @@ static struct acpi_blacklist_item acpi_b + {"IBM ", "TP600E ", 0x00000105, ACPI_SIG_DSDT, less_than_or_equal, + "Incorrect _ADR", 1}, + +- {""} ++ {"", "", 0, 0, 0, all_versions, 0} + }; + + #if CONFIG_ACPI_BLACKLIST_YEAR +diff -urNp linux-2.6.31.1/drivers/acpi/osl.c linux-2.6.31.1/drivers/acpi/osl.c +--- linux-2.6.31.1/drivers/acpi/osl.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/acpi/osl.c 2009-10-01 20:12:42.000000000 -0400 +@@ -521,6 +521,8 @@ acpi_os_read_memory(acpi_physical_addres + void __iomem *virt_addr; + + virt_addr = ioremap(phys_addr, width); ++ if (!virt_addr) ++ return AE_NO_MEMORY; + if (!value) + value = &dummy; + +@@ -549,6 +551,8 @@ acpi_os_write_memory(acpi_physical_addre + void __iomem *virt_addr; + + virt_addr = ioremap(phys_addr, width); ++ if (!virt_addr) ++ return AE_NO_MEMORY; + + switch (width) { + case 8: +diff -urNp linux-2.6.31.1/drivers/acpi/processor_core.c linux-2.6.31.1/drivers/acpi/processor_core.c +--- linux-2.6.31.1/drivers/acpi/processor_core.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/acpi/processor_core.c 2009-10-01 20:12:42.000000000 -0400 +@@ -712,7 +712,7 @@ static int __cpuinit acpi_processor_star + return 0; + } + +- BUG_ON((pr->id >= nr_cpu_ids) || (pr->id < 0)); ++ BUG_ON(pr->id >= nr_cpu_ids); + + /* + * Buggy BIOS check +diff -urNp linux-2.6.31.1/drivers/acpi/processor_idle.c linux-2.6.31.1/drivers/acpi/processor_idle.c +--- linux-2.6.31.1/drivers/acpi/processor_idle.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/acpi/processor_idle.c 2009-10-01 20:12:42.000000000 -0400 +@@ -108,7 +108,7 @@ static struct dmi_system_id __cpuinitdat + DMI_MATCH(DMI_BIOS_VENDOR,"Phoenix Technologies LTD"), + DMI_MATCH(DMI_BIOS_VERSION,"SHE845M0.86C.0013.D.0302131307")}, + (void *)2}, +- {}, ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL}, + }; + + +diff -urNp linux-2.6.31.1/drivers/acpi/video.c linux-2.6.31.1/drivers/acpi/video.c +--- linux-2.6.31.1/drivers/acpi/video.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/acpi/video.c 2009-10-01 20:12:42.000000000 -0400 +@@ -283,7 +283,7 @@ static int acpi_video_device_brightness_ + struct file *file); + static ssize_t acpi_video_device_write_brightness(struct file *file, + const char __user *buffer, size_t count, loff_t *data); +-static struct file_operations acpi_video_device_brightness_fops = { ++static const struct file_operations acpi_video_device_brightness_fops = { + .owner = THIS_MODULE, + .open = acpi_video_device_brightness_open_fs, + .read = seq_read, +diff -urNp linux-2.6.31.1/drivers/ata/ahci.c linux-2.6.31.1/drivers/ata/ahci.c +--- linux-2.6.31.1/drivers/ata/ahci.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ata/ahci.c 2009-10-01 20:12:42.000000000 -0400 +@@ -629,7 +629,7 @@ static const struct pci_device_id ahci_p + { PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, + PCI_CLASS_STORAGE_SATA_AHCI, 0xffffff, board_ahci }, + +- { } /* terminate list */ ++ { 0, 0, 0, 0, 0, 0, 0 } /* terminate list */ + }; + + +diff -urNp linux-2.6.31.1/drivers/ata/ata_piix.c linux-2.6.31.1/drivers/ata/ata_piix.c +--- linux-2.6.31.1/drivers/ata/ata_piix.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ata/ata_piix.c 2009-10-01 20:12:42.000000000 -0400 +@@ -291,7 +291,7 @@ static const struct pci_device_id piix_p + { 0x8086, 0x3b2d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich8_2port_sata }, + /* SATA Controller IDE (PCH) */ + { 0x8086, 0x3b2e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich8_sata }, +- { } /* terminate list */ ++ { 0, 0, 0, 0, 0, 0, 0 } /* terminate list */ + }; + + static struct pci_driver piix_pci_driver = { +@@ -608,7 +608,7 @@ static const struct ich_laptop ich_lapto + { 0x2653, 0x1043, 0x82D8 }, /* ICH6M on Asus Eee 701 */ + { 0x27df, 0x104d, 0x900e }, /* ICH7 on Sony TZ-90 */ + /* end marker */ +- { 0, } ++ { 0, 0, 0 } + }; + + /** +@@ -1086,7 +1086,7 @@ static int piix_broken_suspend(void) + }, + }, + +- { } /* terminate list */ ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } /* terminate list */ + }; + static const char *oemstrs[] = { + "Tecra M3,", +diff -urNp linux-2.6.31.1/drivers/ata/libata-core.c linux-2.6.31.1/drivers/ata/libata-core.c +--- linux-2.6.31.1/drivers/ata/libata-core.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ata/libata-core.c 2009-10-01 20:12:42.000000000 -0400 +@@ -896,7 +896,7 @@ static const struct ata_xfer_ent { + { ATA_SHIFT_PIO, ATA_NR_PIO_MODES, XFER_PIO_0 }, + { ATA_SHIFT_MWDMA, ATA_NR_MWDMA_MODES, XFER_MW_DMA_0 }, + { ATA_SHIFT_UDMA, ATA_NR_UDMA_MODES, XFER_UDMA_0 }, +- { -1, }, ++ { -1, 0, 0 } + }; + + /** +@@ -3141,7 +3141,7 @@ static const struct ata_timing ata_timin + { XFER_UDMA_5, 0, 0, 0, 0, 0, 0, 0, 0, 20 }, + { XFER_UDMA_6, 0, 0, 0, 0, 0, 0, 0, 0, 15 }, + +- { 0xFF } ++ { 0xFF, 0, 0, 0, 0, 0, 0, 0, 0 } + }; + + #define ENOUGH(v, unit) (((v)-1)/(unit)+1) +@@ -4339,7 +4339,7 @@ static const struct ata_blacklist_entry + { "PIONEER DVD-RW DVRTD08", "1.00", ATA_HORKAGE_NOSETXFER }, + + /* End Marker */ +- { } ++ { NULL, NULL, 0 } + }; + + static int strn_pattern_cmp(const char *patt, const char *name, int wildchar) +diff -urNp linux-2.6.31.1/drivers/atm/adummy.c linux-2.6.31.1/drivers/atm/adummy.c +--- linux-2.6.31.1/drivers/atm/adummy.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/adummy.c 2009-10-01 20:12:42.000000000 -0400 +@@ -77,7 +77,7 @@ adummy_send(struct atm_vcc *vcc, struct + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + + return 0; + } +diff -urNp linux-2.6.31.1/drivers/atm/ambassador.c linux-2.6.31.1/drivers/atm/ambassador.c +--- linux-2.6.31.1/drivers/atm/ambassador.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/ambassador.c 2009-10-01 20:12:42.000000000 -0400 +@@ -453,7 +453,7 @@ static void tx_complete (amb_dev * dev, + PRINTD (DBG_FLOW|DBG_TX, "tx_complete %p %p", dev, tx); + + // VC layer stats +- atomic_inc(&ATM_SKB(skb)->vcc->stats->tx); ++ atomic_inc_unchecked(&ATM_SKB(skb)->vcc->stats->tx); + + // free the descriptor + kfree (tx_descr); +@@ -494,7 +494,7 @@ static void rx_complete (amb_dev * dev, + dump_skb ("<<<", vc, skb); + + // VC layer stats +- atomic_inc(&atm_vcc->stats->rx); ++ atomic_inc_unchecked(&atm_vcc->stats->rx); + __net_timestamp(skb); + // end of our responsability + atm_vcc->push (atm_vcc, skb); +@@ -509,7 +509,7 @@ static void rx_complete (amb_dev * dev, + } else { + PRINTK (KERN_INFO, "dropped over-size frame"); + // should we count this? +- atomic_inc(&atm_vcc->stats->rx_drop); ++ atomic_inc_unchecked(&atm_vcc->stats->rx_drop); + } + + } else { +@@ -1349,7 +1349,7 @@ static int amb_send (struct atm_vcc * at + } + + if (check_area (skb->data, skb->len)) { +- atomic_inc(&atm_vcc->stats->tx_err); ++ atomic_inc_unchecked(&atm_vcc->stats->tx_err); + return -ENOMEM; // ? + } + +diff -urNp linux-2.6.31.1/drivers/atm/atmtcp.c linux-2.6.31.1/drivers/atm/atmtcp.c +--- linux-2.6.31.1/drivers/atm/atmtcp.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/atmtcp.c 2009-10-01 20:12:42.000000000 -0400 +@@ -206,7 +206,7 @@ static int atmtcp_v_send(struct atm_vcc + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb(skb); + if (dev_data) return 0; +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -ENOLINK; + } + size = skb->len+sizeof(struct atmtcp_hdr); +@@ -214,7 +214,7 @@ static int atmtcp_v_send(struct atm_vcc + if (!new_skb) { + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -ENOBUFS; + } + hdr = (void *) skb_put(new_skb,sizeof(struct atmtcp_hdr)); +@@ -225,8 +225,8 @@ static int atmtcp_v_send(struct atm_vcc + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb(skb); + out_vcc->push(out_vcc,new_skb); +- atomic_inc(&vcc->stats->tx); +- atomic_inc(&out_vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->tx); ++ atomic_inc_unchecked(&out_vcc->stats->rx); + return 0; + } + +@@ -300,7 +300,7 @@ static int atmtcp_c_send(struct atm_vcc + out_vcc = find_vcc(dev, ntohs(hdr->vpi), ntohs(hdr->vci)); + read_unlock(&vcc_sklist_lock); + if (!out_vcc) { +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + goto done; + } + skb_pull(skb,sizeof(struct atmtcp_hdr)); +@@ -312,8 +312,8 @@ static int atmtcp_c_send(struct atm_vcc + __net_timestamp(new_skb); + skb_copy_from_linear_data(skb, skb_put(new_skb, skb->len), skb->len); + out_vcc->push(out_vcc,new_skb); +- atomic_inc(&vcc->stats->tx); +- atomic_inc(&out_vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->tx); ++ atomic_inc_unchecked(&out_vcc->stats->rx); + done: + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb(skb); +diff -urNp linux-2.6.31.1/drivers/atm/eni.c linux-2.6.31.1/drivers/atm/eni.c +--- linux-2.6.31.1/drivers/atm/eni.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/eni.c 2009-10-01 20:12:42.000000000 -0400 +@@ -525,7 +525,7 @@ static int rx_aal0(struct atm_vcc *vcc) + DPRINTK(DEV_LABEL "(itf %d): trashing empty cell\n", + vcc->dev->number); + length = 0; +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + } + else { + length = ATM_CELL_SIZE-1; /* no HEC */ +@@ -580,7 +580,7 @@ static int rx_aal5(struct atm_vcc *vcc) + size); + } + eff = length = 0; +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + } + else { + size = (descr & MID_RED_COUNT)*(ATM_CELL_PAYLOAD >> 2); +@@ -597,7 +597,7 @@ static int rx_aal5(struct atm_vcc *vcc) + "(VCI=%d,length=%ld,size=%ld (descr 0x%lx))\n", + vcc->dev->number,vcc->vci,length,size << 2,descr); + length = eff = 0; +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + } + } + skb = eff ? atm_alloc_charge(vcc,eff << 2,GFP_ATOMIC) : NULL; +@@ -770,7 +770,7 @@ rx_dequeued++; + vcc->push(vcc,skb); + pushed++; + } +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + wake_up(&eni_dev->rx_wait); + } +@@ -1227,7 +1227,7 @@ static void dequeue_tx(struct atm_dev *d + PCI_DMA_TODEVICE); + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb_irq(skb); +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + wake_up(&eni_dev->tx_wait); + dma_complete++; + } +diff -urNp linux-2.6.31.1/drivers/atm/firestream.c linux-2.6.31.1/drivers/atm/firestream.c +--- linux-2.6.31.1/drivers/atm/firestream.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/firestream.c 2009-10-01 20:12:42.000000000 -0400 +@@ -748,7 +748,7 @@ static void process_txdone_queue (struct + } + } + +- atomic_inc(&ATM_SKB(skb)->vcc->stats->tx); ++ atomic_inc_unchecked(&ATM_SKB(skb)->vcc->stats->tx); + + fs_dprintk (FS_DEBUG_TXMEM, "i"); + fs_dprintk (FS_DEBUG_ALLOC, "Free t-skb: %p\n", skb); +@@ -815,7 +815,7 @@ static void process_incoming (struct fs_ + #endif + skb_put (skb, qe->p1 & 0xffff); + ATM_SKB(skb)->vcc = atm_vcc; +- atomic_inc(&atm_vcc->stats->rx); ++ atomic_inc_unchecked(&atm_vcc->stats->rx); + __net_timestamp(skb); + fs_dprintk (FS_DEBUG_ALLOC, "Free rec-skb: %p (pushed)\n", skb); + atm_vcc->push (atm_vcc, skb); +@@ -836,12 +836,12 @@ static void process_incoming (struct fs_ + kfree (pe); + } + if (atm_vcc) +- atomic_inc(&atm_vcc->stats->rx_drop); ++ atomic_inc_unchecked(&atm_vcc->stats->rx_drop); + break; + case 0x1f: /* Reassembly abort: no buffers. */ + /* Silently increment error counter. */ + if (atm_vcc) +- atomic_inc(&atm_vcc->stats->rx_drop); ++ atomic_inc_unchecked(&atm_vcc->stats->rx_drop); + break; + default: /* Hmm. Haven't written the code to handle the others yet... -- REW */ + printk (KERN_WARNING "Don't know what to do with RX status %x: %s.\n", +diff -urNp linux-2.6.31.1/drivers/atm/fore200e.c linux-2.6.31.1/drivers/atm/fore200e.c +--- linux-2.6.31.1/drivers/atm/fore200e.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/fore200e.c 2009-10-01 20:12:42.000000000 -0400 +@@ -931,9 +931,9 @@ fore200e_tx_irq(struct fore200e* fore200 + #endif + /* check error condition */ + if (*entry->status & STATUS_ERROR) +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + else +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + } + } + +@@ -1082,7 +1082,7 @@ fore200e_push_rpd(struct fore200e* fore2 + if (skb == NULL) { + DPRINTK(2, "unable to alloc new skb, rx PDU length = %d\n", pdu_len); + +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + return -ENOMEM; + } + +@@ -1125,14 +1125,14 @@ fore200e_push_rpd(struct fore200e* fore2 + + dev_kfree_skb_any(skb); + +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + return -ENOMEM; + } + + ASSERT(atomic_read(&sk_atm(vcc)->sk_wmem_alloc) >= 0); + + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + ASSERT(atomic_read(&sk_atm(vcc)->sk_wmem_alloc) >= 0); + +@@ -1210,7 +1210,7 @@ fore200e_rx_irq(struct fore200e* fore200 + DPRINTK(2, "damaged PDU on %d.%d.%d\n", + fore200e->atm_dev->number, + entry->rpd->atm_header.vpi, entry->rpd->atm_header.vci); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + } + } + +@@ -1655,7 +1655,7 @@ fore200e_send(struct atm_vcc *vcc, struc + goto retry_here; + } + +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + + fore200e->tx_sat++; + DPRINTK(2, "tx queue of device %s is saturated, PDU dropped - heartbeat is %08x\n", +diff -urNp linux-2.6.31.1/drivers/atm/he.c linux-2.6.31.1/drivers/atm/he.c +--- linux-2.6.31.1/drivers/atm/he.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/he.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1728,7 +1728,7 @@ he_service_rbrq(struct he_dev *he_dev, i + + if (RBRQ_HBUF_ERR(he_dev->rbrq_head)) { + hprintk("HBUF_ERR! (cid 0x%x)\n", cid); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + goto return_host_buffers; + } + +@@ -1761,7 +1761,7 @@ he_service_rbrq(struct he_dev *he_dev, i + RBRQ_LEN_ERR(he_dev->rbrq_head) + ? "LEN_ERR" : "", + vcc->vpi, vcc->vci); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto return_host_buffers; + } + +@@ -1820,7 +1820,7 @@ he_service_rbrq(struct he_dev *he_dev, i + vcc->push(vcc, skb); + spin_lock(&he_dev->global_lock); + +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + return_host_buffers: + ++pdus_assembled; +@@ -2165,7 +2165,7 @@ __enqueue_tpd(struct he_dev *he_dev, str + tpd->vcc->pop(tpd->vcc, tpd->skb); + else + dev_kfree_skb_any(tpd->skb); +- atomic_inc(&tpd->vcc->stats->tx_err); ++ atomic_inc_unchecked(&tpd->vcc->stats->tx_err); + } + pci_pool_free(he_dev->tpd_pool, tpd, TPD_ADDR(tpd->status)); + return; +@@ -2577,7 +2577,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -EINVAL; + } + +@@ -2588,7 +2588,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -EINVAL; + } + #endif +@@ -2600,7 +2600,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + spin_unlock_irqrestore(&he_dev->global_lock, flags); + return -ENOMEM; + } +@@ -2642,7 +2642,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + spin_unlock_irqrestore(&he_dev->global_lock, flags); + return -ENOMEM; + } +@@ -2673,7 +2673,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + __enqueue_tpd(he_dev, tpd, cid); + spin_unlock_irqrestore(&he_dev->global_lock, flags); + +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + + return 0; + } +diff -urNp linux-2.6.31.1/drivers/atm/horizon.c linux-2.6.31.1/drivers/atm/horizon.c +--- linux-2.6.31.1/drivers/atm/horizon.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/horizon.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1033,7 +1033,7 @@ static void rx_schedule (hrz_dev * dev, + { + struct atm_vcc * vcc = ATM_SKB(skb)->vcc; + // VC layer stats +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + __net_timestamp(skb); + // end of our responsability + vcc->push (vcc, skb); +@@ -1185,7 +1185,7 @@ static void tx_schedule (hrz_dev * const + dev->tx_iovec = NULL; + + // VC layer stats +- atomic_inc(&ATM_SKB(skb)->vcc->stats->tx); ++ atomic_inc_unchecked(&ATM_SKB(skb)->vcc->stats->tx); + + // free the skb + hrz_kfree_skb (skb); +diff -urNp linux-2.6.31.1/drivers/atm/idt77252.c linux-2.6.31.1/drivers/atm/idt77252.c +--- linux-2.6.31.1/drivers/atm/idt77252.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/idt77252.c 2009-10-01 20:12:42.000000000 -0400 +@@ -810,7 +810,7 @@ drain_scq(struct idt77252_dev *card, str + else + dev_kfree_skb(skb); + +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + } + + atomic_dec(&scq->used); +@@ -1073,13 +1073,13 @@ dequeue_rx(struct idt77252_dev *card, st + if ((sb = dev_alloc_skb(64)) == NULL) { + printk("%s: Can't allocate buffers for aal0.\n", + card->name); +- atomic_add(i, &vcc->stats->rx_drop); ++ atomic_add_unchecked(i, &vcc->stats->rx_drop); + break; + } + if (!atm_charge(vcc, sb->truesize)) { + RXPRINTK("%s: atm_charge() dropped aal0 packets.\n", + card->name); +- atomic_add(i - 1, &vcc->stats->rx_drop); ++ atomic_add_unchecked(i - 1, &vcc->stats->rx_drop); + dev_kfree_skb(sb); + break; + } +@@ -1096,7 +1096,7 @@ dequeue_rx(struct idt77252_dev *card, st + ATM_SKB(sb)->vcc = vcc; + __net_timestamp(sb); + vcc->push(vcc, sb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + cell += ATM_CELL_PAYLOAD; + } +@@ -1133,13 +1133,13 @@ dequeue_rx(struct idt77252_dev *card, st + "(CDC: %08x)\n", + card->name, len, rpp->len, readl(SAR_REG_CDC)); + recycle_rx_pool_skb(card, rpp); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + return; + } + if (stat & SAR_RSQE_CRC) { + RXPRINTK("%s: AAL5 CRC error.\n", card->name); + recycle_rx_pool_skb(card, rpp); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + return; + } + if (skb_queue_len(&rpp->queue) > 1) { +@@ -1150,7 +1150,7 @@ dequeue_rx(struct idt77252_dev *card, st + RXPRINTK("%s: Can't alloc RX skb.\n", + card->name); + recycle_rx_pool_skb(card, rpp); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + return; + } + if (!atm_charge(vcc, skb->truesize)) { +@@ -1169,7 +1169,7 @@ dequeue_rx(struct idt77252_dev *card, st + __net_timestamp(skb); + + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + return; + } +@@ -1191,7 +1191,7 @@ dequeue_rx(struct idt77252_dev *card, st + __net_timestamp(skb); + + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + if (skb->truesize > SAR_FB_SIZE_3) + add_rx_skb(card, 3, SAR_FB_SIZE_3, 1); +@@ -1303,14 +1303,14 @@ idt77252_rx_raw(struct idt77252_dev *car + if (vcc->qos.aal != ATM_AAL0) { + RPRINTK("%s: raw cell for non AAL0 vc %u.%u\n", + card->name, vpi, vci); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + goto drop; + } + + if ((sb = dev_alloc_skb(64)) == NULL) { + printk("%s: Can't allocate buffers for AAL0.\n", + card->name); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto drop; + } + +@@ -1329,7 +1329,7 @@ idt77252_rx_raw(struct idt77252_dev *car + ATM_SKB(sb)->vcc = vcc; + __net_timestamp(sb); + vcc->push(vcc, sb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + drop: + skb_pull(queue, 64); +@@ -1954,13 +1954,13 @@ idt77252_send_skb(struct atm_vcc *vcc, s + + if (vc == NULL) { + printk("%s: NULL connection in send().\n", card->name); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return -EINVAL; + } + if (!test_bit(VCF_TX, &vc->flags)) { + printk("%s: Trying to transmit on a non-tx VC.\n", card->name); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return -EINVAL; + } +@@ -1972,14 +1972,14 @@ idt77252_send_skb(struct atm_vcc *vcc, s + break; + default: + printk("%s: Unsupported AAL: %d\n", card->name, vcc->qos.aal); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return -EINVAL; + } + + if (skb_shinfo(skb)->nr_frags != 0) { + printk("%s: No scatter-gather yet.\n", card->name); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return -EINVAL; + } +@@ -1987,7 +1987,7 @@ idt77252_send_skb(struct atm_vcc *vcc, s + + err = queue_skb(card, vc, skb, oam); + if (err) { +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return err; + } +@@ -2010,7 +2010,7 @@ idt77252_send_oam(struct atm_vcc *vcc, v + skb = dev_alloc_skb(64); + if (!skb) { + printk("%s: Out of memory in send_oam().\n", card->name); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -ENOMEM; + } + atomic_add(skb->truesize, &sk_atm(vcc)->sk_wmem_alloc); +diff -urNp linux-2.6.31.1/drivers/atm/iphase.c linux-2.6.31.1/drivers/atm/iphase.c +--- linux-2.6.31.1/drivers/atm/iphase.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/iphase.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1123,7 +1123,7 @@ static int rx_pkt(struct atm_dev *dev) + status = (u_short) (buf_desc_ptr->desc_mode); + if (status & (RX_CER | RX_PTE | RX_OFL)) + { +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + IF_ERR(printk("IA: bad packet, dropping it");) + if (status & RX_CER) { + IF_ERR(printk(" cause: packet CRC error\n");) +@@ -1146,7 +1146,7 @@ static int rx_pkt(struct atm_dev *dev) + len = dma_addr - buf_addr; + if (len > iadev->rx_buf_sz) { + printk("Over %d bytes sdu received, dropped!!!\n", iadev->rx_buf_sz); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto out_free_desc; + } + +@@ -1296,7 +1296,7 @@ static void rx_dle_intr(struct atm_dev * + ia_vcc = INPH_IA_VCC(vcc); + if (ia_vcc == NULL) + { +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + dev_kfree_skb_any(skb); + atm_return(vcc, atm_guess_pdu2truesize(len)); + goto INCR_DLE; +@@ -1308,7 +1308,7 @@ static void rx_dle_intr(struct atm_dev * + if ((length > iadev->rx_buf_sz) || (length > + (skb->len - sizeof(struct cpcs_trailer)))) + { +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + IF_ERR(printk("rx_dle_intr: Bad AAL5 trailer %d (skb len %d)", + length, skb->len);) + dev_kfree_skb_any(skb); +@@ -1324,7 +1324,7 @@ static void rx_dle_intr(struct atm_dev * + + IF_RX(printk("rx_dle_intr: skb push");) + vcc->push(vcc,skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + iadev->rx_pkt_cnt++; + } + INCR_DLE: +@@ -2806,15 +2806,15 @@ static int ia_ioctl(struct atm_dev *dev, + { + struct k_sonet_stats *stats; + stats = &PRIV(_ia_dev[board])->sonet_stats; +- printk("section_bip: %d\n", atomic_read(&stats->section_bip)); +- printk("line_bip : %d\n", atomic_read(&stats->line_bip)); +- printk("path_bip : %d\n", atomic_read(&stats->path_bip)); +- printk("line_febe : %d\n", atomic_read(&stats->line_febe)); +- printk("path_febe : %d\n", atomic_read(&stats->path_febe)); +- printk("corr_hcs : %d\n", atomic_read(&stats->corr_hcs)); +- printk("uncorr_hcs : %d\n", atomic_read(&stats->uncorr_hcs)); +- printk("tx_cells : %d\n", atomic_read(&stats->tx_cells)); +- printk("rx_cells : %d\n", atomic_read(&stats->rx_cells)); ++ printk("section_bip: %d\n", atomic_read_unchecked(&stats->section_bip)); ++ printk("line_bip : %d\n", atomic_read_unchecked(&stats->line_bip)); ++ printk("path_bip : %d\n", atomic_read_unchecked(&stats->path_bip)); ++ printk("line_febe : %d\n", atomic_read_unchecked(&stats->line_febe)); ++ printk("path_febe : %d\n", atomic_read_unchecked(&stats->path_febe)); ++ printk("corr_hcs : %d\n", atomic_read_unchecked(&stats->corr_hcs)); ++ printk("uncorr_hcs : %d\n", atomic_read_unchecked(&stats->uncorr_hcs)); ++ printk("tx_cells : %d\n", atomic_read_unchecked(&stats->tx_cells)); ++ printk("rx_cells : %d\n", atomic_read_unchecked(&stats->rx_cells)); + } + ia_cmds.status = 0; + break; +@@ -2919,7 +2919,7 @@ static int ia_pkt_tx (struct atm_vcc *vc + if ((desc == 0) || (desc > iadev->num_tx_desc)) + { + IF_ERR(printk(DEV_LABEL "invalid desc for send: %d\n", desc);) +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + if (vcc->pop) + vcc->pop(vcc, skb); + else +@@ -3024,14 +3024,14 @@ static int ia_pkt_tx (struct atm_vcc *vc + ATM_DESC(skb) = vcc->vci; + skb_queue_tail(&iadev->tx_dma_q, skb); + +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + iadev->tx_pkt_cnt++; + /* Increment transaction counter */ + writel(2, iadev->dma+IPHASE5575_TX_COUNTER); + + #if 0 + /* add flow control logic */ +- if (atomic_read(&vcc->stats->tx) % 20 == 0) { ++ if (atomic_read_unchecked(&vcc->stats->tx) % 20 == 0) { + if (iavcc->vc_desc_cnt > 10) { + vcc->tx_quota = vcc->tx_quota * 3 / 4; + printk("Tx1: vcc->tx_quota = %d \n", (u32)vcc->tx_quota ); +diff -urNp linux-2.6.31.1/drivers/atm/lanai.c linux-2.6.31.1/drivers/atm/lanai.c +--- linux-2.6.31.1/drivers/atm/lanai.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/lanai.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1305,7 +1305,7 @@ static void lanai_send_one_aal5(struct l + vcc_tx_add_aal5_trailer(lvcc, skb->len, 0, 0); + lanai_endtx(lanai, lvcc); + lanai_free_skb(lvcc->tx.atmvcc, skb); +- atomic_inc(&lvcc->tx.atmvcc->stats->tx); ++ atomic_inc_unchecked(&lvcc->tx.atmvcc->stats->tx); + } + + /* Try to fill the buffer - don't call unless there is backlog */ +@@ -1428,7 +1428,7 @@ static void vcc_rx_aal5(struct lanai_vcc + ATM_SKB(skb)->vcc = lvcc->rx.atmvcc; + __net_timestamp(skb); + lvcc->rx.atmvcc->push(lvcc->rx.atmvcc, skb); +- atomic_inc(&lvcc->rx.atmvcc->stats->rx); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx); + out: + lvcc->rx.buf.ptr = end; + cardvcc_write(lvcc, endptr, vcc_rxreadptr); +@@ -1670,7 +1670,7 @@ static int handle_service(struct lanai_d + DPRINTK("(itf %d) got RX service entry 0x%X for non-AAL5 " + "vcc %d\n", lanai->number, (unsigned int) s, vci); + lanai->stats.service_rxnotaal5++; +- atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); + return 0; + } + if (likely(!(s & (SERVICE_TRASH | SERVICE_STREAM | SERVICE_CRCERR)))) { +@@ -1682,7 +1682,7 @@ static int handle_service(struct lanai_d + int bytes; + read_unlock(&vcc_sklist_lock); + DPRINTK("got trashed rx pdu on vci %d\n", vci); +- atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); + lvcc->stats.x.aal5.service_trash++; + bytes = (SERVICE_GET_END(s) * 16) - + (((unsigned long) lvcc->rx.buf.ptr) - +@@ -1694,7 +1694,7 @@ static int handle_service(struct lanai_d + } + if (s & SERVICE_STREAM) { + read_unlock(&vcc_sklist_lock); +- atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); + lvcc->stats.x.aal5.service_stream++; + printk(KERN_ERR DEV_LABEL "(itf %d): Got AAL5 stream " + "PDU on VCI %d!\n", lanai->number, vci); +@@ -1702,7 +1702,7 @@ static int handle_service(struct lanai_d + return 0; + } + DPRINTK("got rx crc error on vci %d\n", vci); +- atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); + lvcc->stats.x.aal5.service_rxcrc++; + lvcc->rx.buf.ptr = &lvcc->rx.buf.start[SERVICE_GET_END(s) * 4]; + cardvcc_write(lvcc, SERVICE_GET_END(s), vcc_rxreadptr); +diff -urNp linux-2.6.31.1/drivers/atm/nicstar.c linux-2.6.31.1/drivers/atm/nicstar.c +--- linux-2.6.31.1/drivers/atm/nicstar.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/nicstar.c 2009-10-01 20:12:42.000000000 -0400 +@@ -1723,7 +1723,7 @@ static int ns_send(struct atm_vcc *vcc, + if ((vc = (vc_map *) vcc->dev_data) == NULL) + { + printk("nicstar%d: vcc->dev_data == NULL on ns_send().\n", card->index); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EINVAL; + } +@@ -1731,7 +1731,7 @@ static int ns_send(struct atm_vcc *vcc, + if (!vc->tx) + { + printk("nicstar%d: Trying to transmit on a non-tx VC.\n", card->index); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EINVAL; + } +@@ -1739,7 +1739,7 @@ static int ns_send(struct atm_vcc *vcc, + if (vcc->qos.aal != ATM_AAL5 && vcc->qos.aal != ATM_AAL0) + { + printk("nicstar%d: Only AAL0 and AAL5 are supported.\n", card->index); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EINVAL; + } +@@ -1747,7 +1747,7 @@ static int ns_send(struct atm_vcc *vcc, + if (skb_shinfo(skb)->nr_frags != 0) + { + printk("nicstar%d: No scatter-gather yet.\n", card->index); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EINVAL; + } +@@ -1792,11 +1792,11 @@ static int ns_send(struct atm_vcc *vcc, + + if (push_scqe(card, vc, scq, &scqe, skb) != 0) + { +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EIO; + } +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + + return 0; + } +@@ -2111,14 +2111,14 @@ static void dequeue_rx(ns_dev *card, ns_ + { + printk("nicstar%d: Can't allocate buffers for aal0.\n", + card->index); +- atomic_add(i,&vcc->stats->rx_drop); ++ atomic_add_unchecked(i,&vcc->stats->rx_drop); + break; + } + if (!atm_charge(vcc, sb->truesize)) + { + RXPRINTK("nicstar%d: atm_charge() dropped aal0 packets.\n", + card->index); +- atomic_add(i-1,&vcc->stats->rx_drop); /* already increased by 1 */ ++ atomic_add_unchecked(i-1,&vcc->stats->rx_drop); /* already increased by 1 */ + dev_kfree_skb_any(sb); + break; + } +@@ -2133,7 +2133,7 @@ static void dequeue_rx(ns_dev *card, ns_ + ATM_SKB(sb)->vcc = vcc; + __net_timestamp(sb); + vcc->push(vcc, sb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + cell += ATM_CELL_PAYLOAD; + } + +@@ -2152,7 +2152,7 @@ static void dequeue_rx(ns_dev *card, ns_ + if (iovb == NULL) + { + printk("nicstar%d: Out of iovec buffers.\n", card->index); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + recycle_rx_buf(card, skb); + return; + } +@@ -2182,7 +2182,7 @@ static void dequeue_rx(ns_dev *card, ns_ + else if (NS_SKB(iovb)->iovcnt >= NS_MAX_IOVECS) + { + printk("nicstar%d: received too big AAL5 SDU.\n", card->index); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + recycle_iovec_rx_bufs(card, (struct iovec *) iovb->data, NS_MAX_IOVECS); + NS_SKB(iovb)->iovcnt = 0; + iovb->len = 0; +@@ -2202,7 +2202,7 @@ static void dequeue_rx(ns_dev *card, ns_ + printk("nicstar%d: Expected a small buffer, and this is not one.\n", + card->index); + which_list(card, skb); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + recycle_rx_buf(card, skb); + vc->rx_iov = NULL; + recycle_iov_buf(card, iovb); +@@ -2216,7 +2216,7 @@ static void dequeue_rx(ns_dev *card, ns_ + printk("nicstar%d: Expected a large buffer, and this is not one.\n", + card->index); + which_list(card, skb); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + recycle_iovec_rx_bufs(card, (struct iovec *) iovb->data, + NS_SKB(iovb)->iovcnt); + vc->rx_iov = NULL; +@@ -2240,7 +2240,7 @@ static void dequeue_rx(ns_dev *card, ns_ + printk(" - PDU size mismatch.\n"); + else + printk(".\n"); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + recycle_iovec_rx_bufs(card, (struct iovec *) iovb->data, + NS_SKB(iovb)->iovcnt); + vc->rx_iov = NULL; +@@ -2256,7 +2256,7 @@ static void dequeue_rx(ns_dev *card, ns_ + if (!atm_charge(vcc, skb->truesize)) + { + push_rxbufs(card, skb); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + } + else + { +@@ -2268,7 +2268,7 @@ static void dequeue_rx(ns_dev *card, ns_ + ATM_SKB(skb)->vcc = vcc; + __net_timestamp(skb); + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + } + else if (NS_SKB(iovb)->iovcnt == 2) /* One small plus one large buffer */ +@@ -2283,7 +2283,7 @@ static void dequeue_rx(ns_dev *card, ns_ + if (!atm_charge(vcc, sb->truesize)) + { + push_rxbufs(card, sb); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + } + else + { +@@ -2295,7 +2295,7 @@ static void dequeue_rx(ns_dev *card, ns_ + ATM_SKB(sb)->vcc = vcc; + __net_timestamp(sb); + vcc->push(vcc, sb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + + push_rxbufs(card, skb); +@@ -2306,7 +2306,7 @@ static void dequeue_rx(ns_dev *card, ns_ + if (!atm_charge(vcc, skb->truesize)) + { + push_rxbufs(card, skb); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + } + else + { +@@ -2320,7 +2320,7 @@ static void dequeue_rx(ns_dev *card, ns_ + ATM_SKB(skb)->vcc = vcc; + __net_timestamp(skb); + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + + push_rxbufs(card, sb); +@@ -2342,7 +2342,7 @@ static void dequeue_rx(ns_dev *card, ns_ + if (hb == NULL) + { + printk("nicstar%d: Out of huge buffers.\n", card->index); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + recycle_iovec_rx_bufs(card, (struct iovec *) iovb->data, + NS_SKB(iovb)->iovcnt); + vc->rx_iov = NULL; +@@ -2393,7 +2393,7 @@ static void dequeue_rx(ns_dev *card, ns_ + } + else + dev_kfree_skb_any(hb); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + } + else + { +@@ -2427,7 +2427,7 @@ static void dequeue_rx(ns_dev *card, ns_ + #endif /* NS_USE_DESTRUCTORS */ + __net_timestamp(hb); + vcc->push(vcc, hb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + } + +diff -urNp linux-2.6.31.1/drivers/atm/solos-pci.c linux-2.6.31.1/drivers/atm/solos-pci.c +--- linux-2.6.31.1/drivers/atm/solos-pci.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/solos-pci.c 2009-10-01 20:12:42.000000000 -0400 +@@ -663,7 +663,7 @@ void solos_bh(unsigned long card_arg) + } + atm_charge(vcc, skb->truesize); + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + break; + + case PKT_STATUS: +@@ -966,7 +966,7 @@ static uint32_t fpga_tx(struct solos_car + vcc = SKB_CB(oldskb)->vcc; + + if (vcc) { +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + solos_pop(vcc, oldskb); + } else + dev_kfree_skb_irq(oldskb); +diff -urNp linux-2.6.31.1/drivers/atm/suni.c linux-2.6.31.1/drivers/atm/suni.c +--- linux-2.6.31.1/drivers/atm/suni.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/suni.c 2009-10-01 20:12:42.000000000 -0400 +@@ -49,8 +49,8 @@ static DEFINE_SPINLOCK(sunis_lock); + + + #define ADD_LIMITED(s,v) \ +- atomic_add((v),&stats->s); \ +- if (atomic_read(&stats->s) < 0) atomic_set(&stats->s,INT_MAX); ++ atomic_add_unchecked((v),&stats->s); \ ++ if (atomic_read_unchecked(&stats->s) < 0) atomic_set_unchecked(&stats->s,INT_MAX); + + + static void suni_hz(unsigned long from_timer) +diff -urNp linux-2.6.31.1/drivers/atm/uPD98402.c linux-2.6.31.1/drivers/atm/uPD98402.c +--- linux-2.6.31.1/drivers/atm/uPD98402.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/uPD98402.c 2009-10-01 20:12:42.000000000 -0400 +@@ -41,7 +41,7 @@ static int fetch_stats(struct atm_dev *d + struct sonet_stats tmp; + int error = 0; + +- atomic_add(GET(HECCT),&PRIV(dev)->sonet_stats.uncorr_hcs); ++ atomic_add_unchecked(GET(HECCT),&PRIV(dev)->sonet_stats.uncorr_hcs); + sonet_copy_stats(&PRIV(dev)->sonet_stats,&tmp); + if (arg) error = copy_to_user(arg,&tmp,sizeof(tmp)); + if (zero && !error) { +@@ -160,9 +160,9 @@ static int uPD98402_ioctl(struct atm_dev + + + #define ADD_LIMITED(s,v) \ +- { atomic_add(GET(v),&PRIV(dev)->sonet_stats.s); \ +- if (atomic_read(&PRIV(dev)->sonet_stats.s) < 0) \ +- atomic_set(&PRIV(dev)->sonet_stats.s,INT_MAX); } ++ { atomic_add_unchecked(GET(v),&PRIV(dev)->sonet_stats.s); \ ++ if (atomic_read_unchecked(&PRIV(dev)->sonet_stats.s) < 0) \ ++ atomic_set_unchecked(&PRIV(dev)->sonet_stats.s,INT_MAX); } + + + static void stat_event(struct atm_dev *dev) +@@ -193,7 +193,7 @@ static void uPD98402_int(struct atm_dev + if (reason & uPD98402_INT_PFM) stat_event(dev); + if (reason & uPD98402_INT_PCO) { + (void) GET(PCOCR); /* clear interrupt cause */ +- atomic_add(GET(HECCT), ++ atomic_add_unchecked(GET(HECCT), + &PRIV(dev)->sonet_stats.uncorr_hcs); + } + if ((reason & uPD98402_INT_RFO) && +@@ -221,9 +221,9 @@ static int uPD98402_start(struct atm_dev + PUT(~(uPD98402_INT_PFM | uPD98402_INT_ALM | uPD98402_INT_RFO | + uPD98402_INT_LOS),PIMR); /* enable them */ + (void) fetch_stats(dev,NULL,1); /* clear kernel counters */ +- atomic_set(&PRIV(dev)->sonet_stats.corr_hcs,-1); +- atomic_set(&PRIV(dev)->sonet_stats.tx_cells,-1); +- atomic_set(&PRIV(dev)->sonet_stats.rx_cells,-1); ++ atomic_set_unchecked(&PRIV(dev)->sonet_stats.corr_hcs,-1); ++ atomic_set_unchecked(&PRIV(dev)->sonet_stats.tx_cells,-1); ++ atomic_set_unchecked(&PRIV(dev)->sonet_stats.rx_cells,-1); + return 0; + } + +diff -urNp linux-2.6.31.1/drivers/atm/zatm.c linux-2.6.31.1/drivers/atm/zatm.c +--- linux-2.6.31.1/drivers/atm/zatm.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/atm/zatm.c 2009-10-01 20:12:42.000000000 -0400 +@@ -458,7 +458,7 @@ printk("dummy: 0x%08lx, 0x%08lx\n",dummy + } + if (!size) { + dev_kfree_skb_irq(skb); +- if (vcc) atomic_inc(&vcc->stats->rx_err); ++ if (vcc) atomic_inc_unchecked(&vcc->stats->rx_err); + continue; + } + if (!atm_charge(vcc,skb->truesize)) { +@@ -468,7 +468,7 @@ printk("dummy: 0x%08lx, 0x%08lx\n",dummy + skb->len = size; + ATM_SKB(skb)->vcc = vcc; + vcc->push(vcc,skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + zout(pos & 0xffff,MTA(mbx)); + #if 0 /* probably a stupid idea */ +@@ -732,7 +732,7 @@ if (*ZATM_PRV_DSC(skb) != (uPD98401_TXPD + skb_queue_head(&zatm_vcc->backlog,skb); + break; + } +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + wake_up(&zatm_vcc->tx_wait); + } + +diff -urNp linux-2.6.31.1/drivers/block/cciss.c linux-2.6.31.1/drivers/block/cciss.c +--- linux-2.6.31.1/drivers/block/cciss.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/block/cciss.c 2009-10-01 20:12:42.000000000 -0400 +@@ -363,7 +363,7 @@ static void cciss_seq_stop(struct seq_fi + h->busy_configuring = 0; + } + +-static struct seq_operations cciss_seq_ops = { ++static const struct seq_operations cciss_seq_ops = { + .start = cciss_seq_start, + .show = cciss_seq_show, + .next = cciss_seq_next, +@@ -426,7 +426,7 @@ out: + return err; + } + +-static struct file_operations cciss_proc_fops = { ++static const struct file_operations cciss_proc_fops = { + .owner = THIS_MODULE, + .open = cciss_seq_open, + .read = seq_read, +diff -urNp linux-2.6.31.1/drivers/char/agp/agp.h linux-2.6.31.1/drivers/char/agp/agp.h +--- linux-2.6.31.1/drivers/char/agp/agp.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/agp/agp.h 2009-10-01 20:12:42.000000000 -0400 +@@ -126,7 +126,7 @@ struct agp_bridge_driver { + struct agp_bridge_data { + const struct agp_version *version; + const struct agp_bridge_driver *driver; +- struct vm_operations_struct *vm_ops; ++ const struct vm_operations_struct *vm_ops; + void *previous_size; + void *current_size; + void *dev_private_data; +diff -urNp linux-2.6.31.1/drivers/char/agp/alpha-agp.c linux-2.6.31.1/drivers/char/agp/alpha-agp.c +--- linux-2.6.31.1/drivers/char/agp/alpha-agp.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/agp/alpha-agp.c 2009-10-01 20:12:42.000000000 -0400 +@@ -40,7 +40,7 @@ static struct aper_size_info_fixed alpha + { 0, 0, 0 }, /* filled in by alpha_core_agp_setup */ + }; + +-struct vm_operations_struct alpha_core_agp_vm_ops = { ++const struct vm_operations_struct alpha_core_agp_vm_ops = { + .fault = alpha_core_agp_vm_fault, + }; + +diff -urNp linux-2.6.31.1/drivers/char/agp/frontend.c linux-2.6.31.1/drivers/char/agp/frontend.c +--- linux-2.6.31.1/drivers/char/agp/frontend.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/agp/frontend.c 2009-10-01 20:12:42.000000000 -0400 +@@ -824,7 +824,7 @@ static int agpioc_reserve_wrap(struct ag + if (copy_from_user(&reserve, arg, sizeof(struct agp_region))) + return -EFAULT; + +- if ((unsigned) reserve.seg_count >= ~0U/sizeof(struct agp_segment)) ++ if ((unsigned) reserve.seg_count >= ~0U/sizeof(struct agp_segment_priv)) + return -EFAULT; + + client = agp_find_client_by_pid(reserve.pid); +diff -urNp linux-2.6.31.1/drivers/char/agp/intel-agp.c linux-2.6.31.1/drivers/char/agp/intel-agp.c +--- linux-2.6.31.1/drivers/char/agp/intel-agp.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/agp/intel-agp.c 2009-10-01 20:12:42.000000000 -0400 +@@ -2395,7 +2395,7 @@ static struct pci_device_id agp_intel_pc + ID(PCI_DEVICE_ID_INTEL_IGDNG_D_HB), + ID(PCI_DEVICE_ID_INTEL_IGDNG_M_HB), + ID(PCI_DEVICE_ID_INTEL_IGDNG_MA_HB), +- { } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, agp_intel_pci_table); +diff -urNp linux-2.6.31.1/drivers/char/apm-emulation.c linux-2.6.31.1/drivers/char/apm-emulation.c +--- linux-2.6.31.1/drivers/char/apm-emulation.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/apm-emulation.c 2009-10-01 20:12:42.000000000 -0400 +@@ -393,7 +393,7 @@ static int apm_open(struct inode * inode + return as ? 0 : -ENOMEM; + } + +-static struct file_operations apm_bios_fops = { ++static const struct file_operations apm_bios_fops = { + .owner = THIS_MODULE, + .read = apm_read, + .poll = apm_poll, +diff -urNp linux-2.6.31.1/drivers/char/bfin-otp.c linux-2.6.31.1/drivers/char/bfin-otp.c +--- linux-2.6.31.1/drivers/char/bfin-otp.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/bfin-otp.c 2009-10-01 20:12:42.000000000 -0400 +@@ -133,7 +133,7 @@ static ssize_t bfin_otp_write(struct fil + # define bfin_otp_write NULL + #endif + +-static struct file_operations bfin_otp_fops = { ++static const struct file_operations bfin_otp_fops = { + .owner = THIS_MODULE, + .read = bfin_otp_read, + .write = bfin_otp_write, +diff -urNp linux-2.6.31.1/drivers/char/hpet.c linux-2.6.31.1/drivers/char/hpet.c +--- linux-2.6.31.1/drivers/char/hpet.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/hpet.c 2009-10-01 20:12:42.000000000 -0400 +@@ -995,7 +995,7 @@ static struct acpi_driver hpet_acpi_driv + }, + }; + +-static struct miscdevice hpet_misc = { HPET_MINOR, "hpet", &hpet_fops }; ++static struct miscdevice hpet_misc = { HPET_MINOR, "hpet", &hpet_fops, {NULL, NULL}, NULL, NULL }; + + static int __init hpet_init(void) + { +diff -urNp linux-2.6.31.1/drivers/char/hvcs.c linux-2.6.31.1/drivers/char/hvcs.c +--- linux-2.6.31.1/drivers/char/hvcs.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/hvcs.c 2009-10-01 20:12:42.000000000 -0400 +@@ -269,7 +269,7 @@ struct hvcs_struct { + unsigned int index; + + struct tty_struct *tty; +- int open_count; ++ atomic_t open_count; + + /* + * Used to tell the driver kernel_thread what operations need to take +@@ -419,7 +419,7 @@ static ssize_t hvcs_vterm_state_store(st + + spin_lock_irqsave(&hvcsd->lock, flags); + +- if (hvcsd->open_count > 0) { ++ if (atomic_read(&hvcsd->open_count) > 0) { + spin_unlock_irqrestore(&hvcsd->lock, flags); + printk(KERN_INFO "HVCS: vterm state unchanged. " + "The hvcs device node is still in use.\n"); +@@ -1135,7 +1135,7 @@ static int hvcs_open(struct tty_struct * + if ((retval = hvcs_partner_connect(hvcsd))) + goto error_release; + +- hvcsd->open_count = 1; ++ atomic_set(&hvcsd->open_count, 1); + hvcsd->tty = tty; + tty->driver_data = hvcsd; + +@@ -1169,7 +1169,7 @@ fast_open: + + spin_lock_irqsave(&hvcsd->lock, flags); + kref_get(&hvcsd->kref); +- hvcsd->open_count++; ++ atomic_inc(&hvcsd->open_count); + hvcsd->todo_mask |= HVCS_SCHED_READ; + spin_unlock_irqrestore(&hvcsd->lock, flags); + +@@ -1213,7 +1213,7 @@ static void hvcs_close(struct tty_struct + hvcsd = tty->driver_data; + + spin_lock_irqsave(&hvcsd->lock, flags); +- if (--hvcsd->open_count == 0) { ++ if (atomic_dec_and_test(&hvcsd->open_count)) { + + vio_disable_interrupts(hvcsd->vdev); + +@@ -1239,10 +1239,10 @@ static void hvcs_close(struct tty_struct + free_irq(irq, hvcsd); + kref_put(&hvcsd->kref, destroy_hvcs_struct); + return; +- } else if (hvcsd->open_count < 0) { ++ } else if (atomic_read(&hvcsd->open_count) < 0) { + printk(KERN_ERR "HVCS: vty-server@%X open_count: %d" + " is missmanaged.\n", +- hvcsd->vdev->unit_address, hvcsd->open_count); ++ hvcsd->vdev->unit_address, atomic_read(&hvcsd->open_count)); + } + + spin_unlock_irqrestore(&hvcsd->lock, flags); +@@ -1258,7 +1258,7 @@ static void hvcs_hangup(struct tty_struc + + spin_lock_irqsave(&hvcsd->lock, flags); + /* Preserve this so that we know how many kref refs to put */ +- temp_open_count = hvcsd->open_count; ++ temp_open_count = atomic_read(&hvcsd->open_count); + + /* + * Don't kref put inside the spinlock because the destruction +@@ -1273,7 +1273,7 @@ static void hvcs_hangup(struct tty_struc + hvcsd->tty->driver_data = NULL; + hvcsd->tty = NULL; + +- hvcsd->open_count = 0; ++ atomic_set(&hvcsd->open_count, 0); + + /* This will drop any buffered data on the floor which is OK in a hangup + * scenario. */ +@@ -1344,7 +1344,7 @@ static int hvcs_write(struct tty_struct + * the middle of a write operation? This is a crummy place to do this + * but we want to keep it all in the spinlock. + */ +- if (hvcsd->open_count <= 0) { ++ if (atomic_read(&hvcsd->open_count) <= 0) { + spin_unlock_irqrestore(&hvcsd->lock, flags); + return -ENODEV; + } +@@ -1418,7 +1418,7 @@ static int hvcs_write_room(struct tty_st + { + struct hvcs_struct *hvcsd = tty->driver_data; + +- if (!hvcsd || hvcsd->open_count <= 0) ++ if (!hvcsd || atomic_read(&hvcsd->open_count) <= 0) + return 0; + + return HVCS_BUFF_LEN - hvcsd->chars_in_buffer; +diff -urNp linux-2.6.31.1/drivers/char/ipmi/ipmi_msghandler.c linux-2.6.31.1/drivers/char/ipmi/ipmi_msghandler.c +--- linux-2.6.31.1/drivers/char/ipmi/ipmi_msghandler.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/ipmi/ipmi_msghandler.c 2009-10-01 20:12:42.000000000 -0400 +@@ -413,7 +413,7 @@ struct ipmi_smi { + struct proc_dir_entry *proc_dir; + char proc_dir_name[10]; + +- atomic_t stats[IPMI_NUM_STATS]; ++ atomic_unchecked_t stats[IPMI_NUM_STATS]; + + /* + * run_to_completion duplicate of smb_info, smi_info +@@ -446,9 +446,9 @@ static DEFINE_MUTEX(smi_watchers_mutex); + + + #define ipmi_inc_stat(intf, stat) \ +- atomic_inc(&(intf)->stats[IPMI_STAT_ ## stat]) ++ atomic_inc_unchecked(&(intf)->stats[IPMI_STAT_ ## stat]) + #define ipmi_get_stat(intf, stat) \ +- ((unsigned int) atomic_read(&(intf)->stats[IPMI_STAT_ ## stat])) ++ ((unsigned int) atomic_read_unchecked(&(intf)->stats[IPMI_STAT_ ## stat])) + + static int is_lan_addr(struct ipmi_addr *addr) + { +@@ -2807,7 +2807,7 @@ int ipmi_register_smi(struct ipmi_smi_ha + INIT_LIST_HEAD(&intf->cmd_rcvrs); + init_waitqueue_head(&intf->waitq); + for (i = 0; i < IPMI_NUM_STATS; i++) +- atomic_set(&intf->stats[i], 0); ++ atomic_set_unchecked(&intf->stats[i], 0); + + intf->proc_dir = NULL; + +diff -urNp linux-2.6.31.1/drivers/char/ipmi/ipmi_si_intf.c linux-2.6.31.1/drivers/char/ipmi/ipmi_si_intf.c +--- linux-2.6.31.1/drivers/char/ipmi/ipmi_si_intf.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/ipmi/ipmi_si_intf.c 2009-10-01 20:12:42.000000000 -0400 +@@ -277,7 +277,7 @@ struct smi_info { + unsigned char slave_addr; + + /* Counters and things for the proc filesystem. */ +- atomic_t stats[SI_NUM_STATS]; ++ atomic_unchecked_t stats[SI_NUM_STATS]; + + struct task_struct *thread; + +@@ -285,9 +285,9 @@ struct smi_info { + }; + + #define smi_inc_stat(smi, stat) \ +- atomic_inc(&(smi)->stats[SI_STAT_ ## stat]) ++ atomic_inc_unchecked(&(smi)->stats[SI_STAT_ ## stat]) + #define smi_get_stat(smi, stat) \ +- ((unsigned int) atomic_read(&(smi)->stats[SI_STAT_ ## stat])) ++ ((unsigned int) atomic_read_unchecked(&(smi)->stats[SI_STAT_ ## stat])) + + #define SI_MAX_PARMS 4 + +@@ -2926,7 +2926,7 @@ static int try_smi_init(struct smi_info + atomic_set(&new_smi->req_events, 0); + new_smi->run_to_completion = 0; + for (i = 0; i < SI_NUM_STATS; i++) +- atomic_set(&new_smi->stats[i], 0); ++ atomic_set_unchecked(&new_smi->stats[i], 0); + + new_smi->interrupt_disabled = 0; + atomic_set(&new_smi->stop_operation, 0); +diff -urNp linux-2.6.31.1/drivers/char/keyboard.c linux-2.6.31.1/drivers/char/keyboard.c +--- linux-2.6.31.1/drivers/char/keyboard.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/keyboard.c 2009-10-01 20:12:42.000000000 -0400 +@@ -635,6 +635,16 @@ static void k_spec(struct vc_data *vc, u + kbd->kbdmode == VC_MEDIUMRAW) && + value != KVAL(K_SAK)) + return; /* SAK is allowed even in raw mode */ ++ ++#if defined(CONFIG_GRKERNSEC_PROC) || defined(CONFIG_GRKERNSEC_PROC_MEMMAP) ++ { ++ void *func = fn_handler[value]; ++ if (func == fn_show_state || func == fn_show_ptregs || ++ func == fn_show_mem) ++ return; ++ } ++#endif ++ + fn_handler[value](vc); + } + +@@ -1386,7 +1396,7 @@ static const struct input_device_id kbd_ + .evbit = { BIT_MASK(EV_SND) }, + }, + +- { }, /* Terminating entry */ ++ { 0 }, /* Terminating entry */ + }; + + MODULE_DEVICE_TABLE(input, kbd_ids); +diff -urNp linux-2.6.31.1/drivers/char/mem.c linux-2.6.31.1/drivers/char/mem.c +--- linux-2.6.31.1/drivers/char/mem.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/mem.c 2009-10-01 20:12:42.000000000 -0400 +@@ -18,6 +18,7 @@ + #include + #include + #include ++#include + #include + #include + #include +@@ -35,6 +36,10 @@ + # include + #endif + ++#if defined(CONFIG_GRKERNSEC) && !defined(CONFIG_GRKERNSEC_NO_RBAC) ++extern struct file_operations grsec_fops; ++#endif ++ + /* + * Architectures vary in how they handle caching for addresses + * outside of main memory. +@@ -192,6 +197,11 @@ static ssize_t write_mem(struct file * f + if (!valid_phys_addr_range(p, count)) + return -EFAULT; + ++#ifdef CONFIG_GRKERNSEC_KMEM ++ gr_handle_mem_write(); ++ return -EPERM; ++#endif ++ + written = 0; + + #ifdef __ARCH_HAS_NO_PAGE_ZERO_MAPPED +@@ -301,7 +311,7 @@ static inline int private_mapping_ok(str + } + #endif + +-static struct vm_operations_struct mmap_mem_ops = { ++static const struct vm_operations_struct mmap_mem_ops = { + #ifdef CONFIG_HAVE_IOREMAP_PROT + .access = generic_access_phys + #endif +@@ -324,6 +334,11 @@ static int mmap_mem(struct file * file, + &vma->vm_page_prot)) + return -EINVAL; + ++#ifdef CONFIG_GRKERNSEC_KMEM ++ if (gr_handle_mem_mmap(vma->vm_pgoff << PAGE_SHIFT, vma)) ++ return -EPERM; ++#endif ++ + vma->vm_page_prot = phys_mem_access_prot(file, vma->vm_pgoff, + size, + vma->vm_page_prot); +@@ -558,6 +573,11 @@ static ssize_t write_kmem(struct file * + ssize_t written; + char * kbuf; /* k-addr because vwrite() takes vmlist_lock rwlock */ + ++#ifdef CONFIG_GRKERNSEC_KMEM ++ gr_handle_kmem_write(); ++ return -EPERM; ++#endif ++ + if (p < (unsigned long) high_memory) { + + wrote = count; +@@ -763,6 +783,16 @@ static loff_t memory_lseek(struct file * + + static int open_port(struct inode * inode, struct file * filp) + { ++#ifdef CONFIG_GRKERNSEC_KMEM ++ gr_handle_open_port(); ++ return -EPERM; ++#endif ++ ++ return capable(CAP_SYS_RAWIO) ? 0 : -EPERM; ++} ++ ++static int open_mem(struct inode * inode, struct file * filp) ++{ + return capable(CAP_SYS_RAWIO) ? 0 : -EPERM; + } + +@@ -770,7 +800,6 @@ static int open_port(struct inode * inod + #define full_lseek null_lseek + #define write_zero write_null + #define read_full read_zero +-#define open_mem open_port + #define open_kmem open_mem + #define open_oldmem open_mem + +@@ -888,6 +917,9 @@ static const struct { + #ifdef CONFIG_CRASH_DUMP + {12,"oldmem", S_IRUSR | S_IWUSR | S_IRGRP, &oldmem_fops, NULL}, + #endif ++#if defined(CONFIG_GRKERNSEC) && !defined(CONFIG_GRKERNSEC_NO_RBAC) ++ {13,"grsec", S_IRUSR | S_IWUGO, &grsec_fops}, ++#endif + }; + + static int memory_open(struct inode *inode, struct file *filp) +diff -urNp linux-2.6.31.1/drivers/char/misc.c linux-2.6.31.1/drivers/char/misc.c +--- linux-2.6.31.1/drivers/char/misc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/misc.c 2009-10-01 20:12:42.000000000 -0400 +@@ -91,7 +91,7 @@ static int misc_seq_show(struct seq_file + } + + +-static struct seq_operations misc_seq_ops = { ++static const struct seq_operations misc_seq_ops = { + .start = misc_seq_start, + .next = misc_seq_next, + .stop = misc_seq_stop, +diff -urNp linux-2.6.31.1/drivers/char/mspec.c linux-2.6.31.1/drivers/char/mspec.c +--- linux-2.6.31.1/drivers/char/mspec.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/mspec.c 2009-10-01 20:12:43.000000000 -0400 +@@ -239,7 +239,7 @@ mspec_fault(struct vm_area_struct *vma, + return VM_FAULT_NOPAGE; + } + +-static struct vm_operations_struct mspec_vm_ops = { ++static const struct vm_operations_struct mspec_vm_ops = { + .open = mspec_open, + .close = mspec_close, + .fault = mspec_fault, +diff -urNp linux-2.6.31.1/drivers/char/nvram.c linux-2.6.31.1/drivers/char/nvram.c +--- linux-2.6.31.1/drivers/char/nvram.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/nvram.c 2009-10-01 20:12:43.000000000 -0400 +@@ -429,7 +429,10 @@ static const struct file_operations nvra + static struct miscdevice nvram_dev = { + NVRAM_MINOR, + "nvram", +- &nvram_fops ++ &nvram_fops, ++ {NULL, NULL}, ++ NULL, ++ NULL + }; + + static int __init nvram_init(void) +diff -urNp linux-2.6.31.1/drivers/char/pcmcia/ipwireless/tty.c linux-2.6.31.1/drivers/char/pcmcia/ipwireless/tty.c +--- linux-2.6.31.1/drivers/char/pcmcia/ipwireless/tty.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/pcmcia/ipwireless/tty.c 2009-10-01 20:12:43.000000000 -0400 +@@ -51,7 +51,7 @@ struct ipw_tty { + int tty_type; + struct ipw_network *network; + struct tty_struct *linux_tty; +- int open_count; ++ atomic_t open_count; + unsigned int control_lines; + struct mutex ipw_tty_mutex; + int tx_bytes_queued; +@@ -127,10 +127,10 @@ static int ipw_open(struct tty_struct *l + mutex_unlock(&tty->ipw_tty_mutex); + return -ENODEV; + } +- if (tty->open_count == 0) ++ if (atomic_read(&tty->open_count) == 0) + tty->tx_bytes_queued = 0; + +- tty->open_count++; ++ atomic_inc(&tty->open_count); + + tty->linux_tty = linux_tty; + linux_tty->driver_data = tty; +@@ -146,9 +146,7 @@ static int ipw_open(struct tty_struct *l + + static void do_ipw_close(struct ipw_tty *tty) + { +- tty->open_count--; +- +- if (tty->open_count == 0) { ++ if (atomic_dec_return(&tty->open_count) == 0) { + struct tty_struct *linux_tty = tty->linux_tty; + + if (linux_tty != NULL) { +@@ -169,7 +167,7 @@ static void ipw_hangup(struct tty_struct + return; + + mutex_lock(&tty->ipw_tty_mutex); +- if (tty->open_count == 0) { ++ if (atomic_read(&tty->open_count) == 0) { + mutex_unlock(&tty->ipw_tty_mutex); + return; + } +@@ -198,7 +196,7 @@ void ipwireless_tty_received(struct ipw_ + return; + } + +- if (!tty->open_count) { ++ if (!atomic_read(&tty->open_count)) { + mutex_unlock(&tty->ipw_tty_mutex); + return; + } +@@ -240,7 +238,7 @@ static int ipw_write(struct tty_struct * + return -ENODEV; + + mutex_lock(&tty->ipw_tty_mutex); +- if (!tty->open_count) { ++ if (!atomic_read(&tty->open_count)) { + mutex_unlock(&tty->ipw_tty_mutex); + return -EINVAL; + } +@@ -280,7 +278,7 @@ static int ipw_write_room(struct tty_str + if (!tty) + return -ENODEV; + +- if (!tty->open_count) ++ if (!atomic_read(&tty->open_count)) + return -EINVAL; + + room = IPWIRELESS_TX_QUEUE_SIZE - tty->tx_bytes_queued; +@@ -322,7 +320,7 @@ static int ipw_chars_in_buffer(struct tt + if (!tty) + return 0; + +- if (!tty->open_count) ++ if (!atomic_read(&tty->open_count)) + return 0; + + return tty->tx_bytes_queued; +@@ -403,7 +401,7 @@ static int ipw_tiocmget(struct tty_struc + if (!tty) + return -ENODEV; + +- if (!tty->open_count) ++ if (!atomic_read(&tty->open_count)) + return -EINVAL; + + return get_control_lines(tty); +@@ -419,7 +417,7 @@ ipw_tiocmset(struct tty_struct *linux_tt + if (!tty) + return -ENODEV; + +- if (!tty->open_count) ++ if (!atomic_read(&tty->open_count)) + return -EINVAL; + + return set_control_lines(tty, set, clear); +@@ -433,7 +431,7 @@ static int ipw_ioctl(struct tty_struct * + if (!tty) + return -ENODEV; + +- if (!tty->open_count) ++ if (!atomic_read(&tty->open_count)) + return -EINVAL; + + /* FIXME: Exactly how is the tty object locked here .. */ +@@ -591,7 +589,7 @@ void ipwireless_tty_free(struct ipw_tty + against a parallel ioctl etc */ + mutex_lock(&ttyj->ipw_tty_mutex); + } +- while (ttyj->open_count) ++ while (atomic_read(&ttyj->open_count)) + do_ipw_close(ttyj); + ipwireless_disassociate_network_ttys(network, + ttyj->channel_idx); +diff -urNp linux-2.6.31.1/drivers/char/random.c linux-2.6.31.1/drivers/char/random.c +--- linux-2.6.31.1/drivers/char/random.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/random.c 2009-10-01 20:12:43.000000000 -0400 +@@ -253,8 +253,13 @@ + /* + * Configuration information + */ ++#ifdef CONFIG_GRKERNSEC_RANDNET ++#define INPUT_POOL_WORDS 512 ++#define OUTPUT_POOL_WORDS 128 ++#else + #define INPUT_POOL_WORDS 128 + #define OUTPUT_POOL_WORDS 32 ++#endif + #define SEC_XFER_SIZE 512 + + /* +@@ -291,10 +296,17 @@ static struct poolinfo { + int poolwords; + int tap1, tap2, tap3, tap4, tap5; + } poolinfo_table[] = { ++#ifdef CONFIG_GRKERNSEC_RANDNET ++ /* x^512 + x^411 + x^308 + x^208 +x^104 + x + 1 -- 225 */ ++ { 512, 411, 308, 208, 104, 1 }, ++ /* x^128 + x^103 + x^76 + x^51 + x^25 + x + 1 -- 105 */ ++ { 128, 103, 76, 51, 25, 1 }, ++#else + /* x^128 + x^103 + x^76 + x^51 +x^25 + x + 1 -- 105 */ + { 128, 103, 76, 51, 25, 1 }, + /* x^32 + x^26 + x^20 + x^14 + x^7 + x + 1 -- 15 */ + { 32, 26, 20, 14, 7, 1 }, ++#endif + #if 0 + /* x^2048 + x^1638 + x^1231 + x^819 + x^411 + x + 1 -- 115 */ + { 2048, 1638, 1231, 819, 411, 1 }, +@@ -1204,7 +1216,7 @@ EXPORT_SYMBOL(generate_random_uuid); + #include + + static int min_read_thresh = 8, min_write_thresh; +-static int max_read_thresh = INPUT_POOL_WORDS * 32; ++static int max_read_thresh = OUTPUT_POOL_WORDS * 32; + static int max_write_thresh = INPUT_POOL_WORDS * 32; + static char sysctl_bootid[16]; + +diff -urNp linux-2.6.31.1/drivers/char/sonypi.c linux-2.6.31.1/drivers/char/sonypi.c +--- linux-2.6.31.1/drivers/char/sonypi.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/sonypi.c 2009-10-01 20:12:43.000000000 -0400 +@@ -490,7 +490,7 @@ static struct sonypi_device { + spinlock_t fifo_lock; + wait_queue_head_t fifo_proc_list; + struct fasync_struct *fifo_async; +- int open_count; ++ atomic_t open_count; + int model; + struct input_dev *input_jog_dev; + struct input_dev *input_key_dev; +@@ -894,7 +894,7 @@ static int sonypi_misc_fasync(int fd, st + static int sonypi_misc_release(struct inode *inode, struct file *file) + { + mutex_lock(&sonypi_device.lock); +- sonypi_device.open_count--; ++ atomic_dec(&sonypi_device.open_count); + mutex_unlock(&sonypi_device.lock); + return 0; + } +@@ -904,9 +904,9 @@ static int sonypi_misc_open(struct inode + lock_kernel(); + mutex_lock(&sonypi_device.lock); + /* Flush input queue on first open */ +- if (!sonypi_device.open_count) ++ if (!atomic_read(&sonypi_device.open_count)) + kfifo_reset(sonypi_device.fifo); +- sonypi_device.open_count++; ++ atomic_inc(&sonypi_device.open_count); + mutex_unlock(&sonypi_device.lock); + unlock_kernel(); + return 0; +diff -urNp linux-2.6.31.1/drivers/char/tpm/tpm_bios.c linux-2.6.31.1/drivers/char/tpm/tpm_bios.c +--- linux-2.6.31.1/drivers/char/tpm/tpm_bios.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/tpm/tpm_bios.c 2009-10-01 20:12:43.000000000 -0400 +@@ -343,14 +343,14 @@ static int tpm_ascii_bios_measurements_s + return 0; + } + +-static struct seq_operations tpm_ascii_b_measurments_seqops = { ++static const struct seq_operations tpm_ascii_b_measurments_seqops = { + .start = tpm_bios_measurements_start, + .next = tpm_bios_measurements_next, + .stop = tpm_bios_measurements_stop, + .show = tpm_ascii_bios_measurements_show, + }; + +-static struct seq_operations tpm_binary_b_measurments_seqops = { ++static const struct seq_operations tpm_binary_b_measurments_seqops = { + .start = tpm_bios_measurements_start, + .next = tpm_bios_measurements_next, + .stop = tpm_bios_measurements_stop, +diff -urNp linux-2.6.31.1/drivers/char/tty_ldisc.c linux-2.6.31.1/drivers/char/tty_ldisc.c +--- linux-2.6.31.1/drivers/char/tty_ldisc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/tty_ldisc.c 2009-10-01 20:12:43.000000000 -0400 +@@ -73,7 +73,7 @@ static void put_ldisc(struct tty_ldisc * + if (atomic_dec_and_lock(&ld->users, &tty_ldisc_lock)) { + struct tty_ldisc_ops *ldo = ld->ops; + +- ldo->refcount--; ++ atomic_dec(&ldo->refcount); + module_put(ldo->owner); + spin_unlock_irqrestore(&tty_ldisc_lock, flags); + +@@ -107,7 +107,7 @@ int tty_register_ldisc(int disc, struct + spin_lock_irqsave(&tty_ldisc_lock, flags); + tty_ldiscs[disc] = new_ldisc; + new_ldisc->num = disc; +- new_ldisc->refcount = 0; ++ atomic_set(&new_ldisc->refcount, 0); + spin_unlock_irqrestore(&tty_ldisc_lock, flags); + + return ret; +@@ -135,7 +135,7 @@ int tty_unregister_ldisc(int disc) + return -EINVAL; + + spin_lock_irqsave(&tty_ldisc_lock, flags); +- if (tty_ldiscs[disc]->refcount) ++ if (atomic_read(&tty_ldiscs[disc]->refcount)) + ret = -EBUSY; + else + tty_ldiscs[disc] = NULL; +@@ -175,7 +175,7 @@ static struct tty_ldisc *tty_ldisc_try_g + err = -EAGAIN; + else { + /* lock it */ +- ldops->refcount++; ++ atomic_inc(&ldops->refcount); + ld->ops = ldops; + atomic_set(&ld->users, 1); + err = 0; +diff -urNp linux-2.6.31.1/drivers/char/vt_ioctl.c linux-2.6.31.1/drivers/char/vt_ioctl.c +--- linux-2.6.31.1/drivers/char/vt_ioctl.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/vt_ioctl.c 2009-10-01 20:12:43.000000000 -0400 +@@ -97,6 +97,12 @@ do_kdsk_ioctl(int cmd, struct kbentry __ + case KDSKBENT: + if (!perm) + return -EPERM; ++ ++#ifdef CONFIG_GRKERNSEC ++ if (!capable(CAP_SYS_TTY_CONFIG)) ++ return -EPERM; ++#endif ++ + if (!i && v == K_NOSUCHMAP) { + /* deallocate map */ + key_map = key_maps[s]; +@@ -237,6 +243,13 @@ do_kdgkb_ioctl(int cmd, struct kbsentry + goto reterr; + } + ++#ifdef CONFIG_GRKERNSEC ++ if (!capable(CAP_SYS_TTY_CONFIG)) { ++ ret = -EPERM; ++ goto reterr; ++ } ++#endif ++ + q = func_table[i]; + first_free = funcbufptr + (funcbufsize - funcbufleft); + for (j = i+1; j < MAX_NR_FUNC && !func_table[j]; j++) +diff -urNp linux-2.6.31.1/drivers/char/xilinx_hwicap/xilinx_hwicap.c linux-2.6.31.1/drivers/char/xilinx_hwicap/xilinx_hwicap.c +--- linux-2.6.31.1/drivers/char/xilinx_hwicap/xilinx_hwicap.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/char/xilinx_hwicap/xilinx_hwicap.c 2009-10-01 20:12:43.000000000 -0400 +@@ -559,7 +559,7 @@ static int hwicap_release(struct inode * + return status; + } + +-static struct file_operations hwicap_fops = { ++static const struct file_operations hwicap_fops = { + .owner = THIS_MODULE, + .write = hwicap_write, + .read = hwicap_read, +diff -urNp linux-2.6.31.1/drivers/edac/edac_core.h linux-2.6.31.1/drivers/edac/edac_core.h +--- linux-2.6.31.1/drivers/edac/edac_core.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/edac/edac_core.h 2009-10-01 20:12:43.000000000 -0400 +@@ -99,11 +99,11 @@ extern int edac_debug_level; + + #else /* !CONFIG_EDAC_DEBUG */ + +-#define debugf0( ... ) +-#define debugf1( ... ) +-#define debugf2( ... ) +-#define debugf3( ... ) +-#define debugf4( ... ) ++#define debugf0( ... ) do {} while (0) ++#define debugf1( ... ) do {} while (0) ++#define debugf2( ... ) do {} while (0) ++#define debugf3( ... ) do {} while (0) ++#define debugf4( ... ) do {} while (0) + + #endif /* !CONFIG_EDAC_DEBUG */ + +diff -urNp linux-2.6.31.1/drivers/firmware/dmi_scan.c linux-2.6.31.1/drivers/firmware/dmi_scan.c +--- linux-2.6.31.1/drivers/firmware/dmi_scan.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/firmware/dmi_scan.c 2009-10-01 20:12:43.000000000 -0400 +@@ -391,11 +391,6 @@ void __init dmi_scan_machine(void) + } + } + else { +- /* +- * no iounmap() for that ioremap(); it would be a no-op, but +- * it's so early in setup that sucker gets confused into doing +- * what it shouldn't if we actually call it. +- */ + p = dmi_ioremap(0xF0000, 0x10000); + if (p == NULL) + goto error; +diff -urNp linux-2.6.31.1/drivers/gpio/gpiolib.c linux-2.6.31.1/drivers/gpio/gpiolib.c +--- linux-2.6.31.1/drivers/gpio/gpiolib.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpio/gpiolib.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1244,7 +1244,7 @@ static int gpiolib_open(struct inode *in + return single_open(file, gpiolib_show, NULL); + } + +-static struct file_operations gpiolib_operations = { ++static const struct file_operations gpiolib_operations = { + .open = gpiolib_open, + .read = seq_read, + .llseek = seq_lseek, +diff -urNp linux-2.6.31.1/drivers/gpu/drm/drm_drv.c linux-2.6.31.1/drivers/gpu/drm/drm_drv.c +--- linux-2.6.31.1/drivers/gpu/drm/drm_drv.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/drm_drv.c 2009-10-01 20:12:43.000000000 -0400 +@@ -417,7 +417,7 @@ int drm_ioctl(struct inode *inode, struc + char *kdata = NULL; + + atomic_inc(&dev->ioctl_count); +- atomic_inc(&dev->counts[_DRM_STAT_IOCTLS]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_IOCTLS]); + ++file_priv->ioctl_count; + + DRM_DEBUG("pid=%d, cmd=0x%02x, nr=0x%02x, dev 0x%lx, auth=%d\n", +diff -urNp linux-2.6.31.1/drivers/gpu/drm/drm_fops.c linux-2.6.31.1/drivers/gpu/drm/drm_fops.c +--- linux-2.6.31.1/drivers/gpu/drm/drm_fops.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/drm_fops.c 2009-10-01 20:12:43.000000000 -0400 +@@ -66,7 +66,7 @@ static int drm_setup(struct drm_device * + } + + for (i = 0; i < ARRAY_SIZE(dev->counts); i++) +- atomic_set(&dev->counts[i], 0); ++ atomic_set_unchecked(&dev->counts[i], 0); + + dev->sigdata.lock = NULL; + +@@ -130,9 +130,9 @@ int drm_open(struct inode *inode, struct + + retcode = drm_open_helper(inode, filp, dev); + if (!retcode) { +- atomic_inc(&dev->counts[_DRM_STAT_OPENS]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_OPENS]); + spin_lock(&dev->count_lock); +- if (!dev->open_count++) { ++ if (atomic_inc_return(&dev->open_count) == 1) { + spin_unlock(&dev->count_lock); + retcode = drm_setup(dev); + goto out; +@@ -433,7 +433,7 @@ int drm_release(struct inode *inode, str + + lock_kernel(); + +- DRM_DEBUG("open_count = %d\n", dev->open_count); ++ DRM_DEBUG("open_count = %d\n", atomic_read(&dev->open_count)); + + if (dev->driver->preclose) + dev->driver->preclose(dev, file_priv); +@@ -445,7 +445,7 @@ int drm_release(struct inode *inode, str + DRM_DEBUG("pid = %d, device = 0x%lx, open_count = %d\n", + task_pid_nr(current), + (long)old_encode_dev(file_priv->minor->device), +- dev->open_count); ++ atomic_read(&dev->open_count)); + + /* if the master has gone away we can't do anything with the lock */ + if (file_priv->minor->master) +@@ -522,9 +522,9 @@ int drm_release(struct inode *inode, str + * End inline drm_release + */ + +- atomic_inc(&dev->counts[_DRM_STAT_CLOSES]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_CLOSES]); + spin_lock(&dev->count_lock); +- if (!--dev->open_count) { ++ if (atomic_dec_and_test(&dev->open_count)) { + if (atomic_read(&dev->ioctl_count)) { + DRM_ERROR("Device busy: %d\n", + atomic_read(&dev->ioctl_count)); +diff -urNp linux-2.6.31.1/drivers/gpu/drm/drm_ioctl.c linux-2.6.31.1/drivers/gpu/drm/drm_ioctl.c +--- linux-2.6.31.1/drivers/gpu/drm/drm_ioctl.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/drm_ioctl.c 2009-10-01 20:12:43.000000000 -0400 +@@ -283,7 +283,7 @@ int drm_getstats(struct drm_device *dev, + stats->data[i].value = + (file_priv->master->lock.hw_lock ? file_priv->master->lock.hw_lock->lock : 0); + else +- stats->data[i].value = atomic_read(&dev->counts[i]); ++ stats->data[i].value = atomic_read_unchecked(&dev->counts[i]); + stats->data[i].type = dev->types[i]; + } + +diff -urNp linux-2.6.31.1/drivers/gpu/drm/drm_lock.c linux-2.6.31.1/drivers/gpu/drm/drm_lock.c +--- linux-2.6.31.1/drivers/gpu/drm/drm_lock.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/drm_lock.c 2009-10-01 20:12:43.000000000 -0400 +@@ -87,7 +87,7 @@ int drm_lock(struct drm_device *dev, voi + if (drm_lock_take(&master->lock, lock->context)) { + master->lock.file_priv = file_priv; + master->lock.lock_time = jiffies; +- atomic_inc(&dev->counts[_DRM_STAT_LOCKS]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_LOCKS]); + break; /* Got lock */ + } + +@@ -165,7 +165,7 @@ int drm_unlock(struct drm_device *dev, v + return -EINVAL; + } + +- atomic_inc(&dev->counts[_DRM_STAT_UNLOCKS]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_UNLOCKS]); + + /* kernel_context_switch isn't used by any of the x86 drm + * modules but is required by the Sparc driver. +diff -urNp linux-2.6.31.1/drivers/gpu/drm/drm_vm.c linux-2.6.31.1/drivers/gpu/drm/drm_vm.c +--- linux-2.6.31.1/drivers/gpu/drm/drm_vm.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/drm_vm.c 2009-10-01 20:12:43.000000000 -0400 +@@ -369,28 +369,28 @@ static int drm_vm_sg_fault(struct vm_are + } + + /** AGP virtual memory operations */ +-static struct vm_operations_struct drm_vm_ops = { ++static const struct vm_operations_struct drm_vm_ops = { + .fault = drm_vm_fault, + .open = drm_vm_open, + .close = drm_vm_close, + }; + + /** Shared virtual memory operations */ +-static struct vm_operations_struct drm_vm_shm_ops = { ++static const struct vm_operations_struct drm_vm_shm_ops = { + .fault = drm_vm_shm_fault, + .open = drm_vm_open, + .close = drm_vm_shm_close, + }; + + /** DMA virtual memory operations */ +-static struct vm_operations_struct drm_vm_dma_ops = { ++static const struct vm_operations_struct drm_vm_dma_ops = { + .fault = drm_vm_dma_fault, + .open = drm_vm_open, + .close = drm_vm_close, + }; + + /** Scatter-gather virtual memory operations */ +-static struct vm_operations_struct drm_vm_sg_ops = { ++static const struct vm_operations_struct drm_vm_sg_ops = { + .fault = drm_vm_sg_fault, + .open = drm_vm_open, + .close = drm_vm_close, +diff -urNp linux-2.6.31.1/drivers/gpu/drm/i810/i810_dma.c linux-2.6.31.1/drivers/gpu/drm/i810/i810_dma.c +--- linux-2.6.31.1/drivers/gpu/drm/i810/i810_dma.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/i810/i810_dma.c 2009-10-01 20:12:43.000000000 -0400 +@@ -952,8 +952,8 @@ static int i810_dma_vertex(struct drm_de + dma->buflist[vertex->idx], + vertex->discard, vertex->used); + +- atomic_add(vertex->used, &dev->counts[_DRM_STAT_SECONDARY]); +- atomic_inc(&dev->counts[_DRM_STAT_DMA]); ++ atomic_add_unchecked(vertex->used, &dev->counts[_DRM_STAT_SECONDARY]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_DMA]); + sarea_priv->last_enqueue = dev_priv->counter - 1; + sarea_priv->last_dispatch = (int)hw_status[5]; + +@@ -1115,8 +1115,8 @@ static int i810_dma_mc(struct drm_device + i810_dma_dispatch_mc(dev, dma->buflist[mc->idx], mc->used, + mc->last_render); + +- atomic_add(mc->used, &dev->counts[_DRM_STAT_SECONDARY]); +- atomic_inc(&dev->counts[_DRM_STAT_DMA]); ++ atomic_add_unchecked(mc->used, &dev->counts[_DRM_STAT_SECONDARY]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_DMA]); + sarea_priv->last_enqueue = dev_priv->counter - 1; + sarea_priv->last_dispatch = (int)hw_status[5]; + +diff -urNp linux-2.6.31.1/drivers/gpu/drm/i915/i915_drv.c linux-2.6.31.1/drivers/gpu/drm/i915/i915_drv.c +--- linux-2.6.31.1/drivers/gpu/drm/i915/i915_drv.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/i915/i915_drv.c 2009-10-01 20:12:43.000000000 -0400 +@@ -154,7 +154,7 @@ i915_pci_resume(struct pci_dev *pdev) + return i915_resume(dev); + } + +-static struct vm_operations_struct i915_gem_vm_ops = { ++static const struct vm_operations_struct i915_gem_vm_ops = { + .fault = i915_gem_fault, + .open = drm_gem_vm_open, + .close = drm_gem_vm_close, +diff -urNp linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_atombios.c linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_atombios.c +--- linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_atombios.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_atombios.c 2009-10-01 20:12:43.000000000 -0400 +@@ -425,13 +425,13 @@ bool radeon_get_atom_connector_info_from + return true; + } + +-struct bios_connector { ++static struct bios_connector { + bool valid; + uint8_t line_mux; + uint16_t devices; + int connector_type; + struct radeon_i2c_bus_rec ddc_bus; +-}; ++} bios_connectors[ATOM_MAX_SUPPORTED_DEVICE];; + + bool radeon_get_atom_connector_info_from_supported_devices_table(struct + drm_device +@@ -447,7 +447,6 @@ bool radeon_get_atom_connector_info_from + uint8_t dac; + union atom_supported_devices *supported_devices; + int i, j; +- struct bios_connector bios_connectors[ATOM_MAX_SUPPORTED_DEVICE]; + + atom_parse_data_header(ctx, index, &size, &frev, &crev, &data_offset); + +diff -urNp linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_state.c linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_state.c +--- linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_state.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_state.c 2009-10-01 20:12:43.000000000 -0400 +@@ -3007,7 +3007,7 @@ static int radeon_cp_getparam(struct drm + { + drm_radeon_private_t *dev_priv = dev->dev_private; + drm_radeon_getparam_t *param = data; +- int value; ++ int value = 0; + + DRM_DEBUG("pid=%d\n", DRM_CURRENTPID); + +diff -urNp linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_ttm.c linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_ttm.c +--- linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_ttm.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/radeon/radeon_ttm.c 2009-10-01 20:12:43.000000000 -0400 +@@ -500,27 +500,10 @@ void radeon_ttm_fini(struct radeon_devic + DRM_INFO("radeon: ttm finalized\n"); + } + +-static struct vm_operations_struct radeon_ttm_vm_ops; +-static struct vm_operations_struct *ttm_vm_ops = NULL; +- +-static int radeon_ttm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +-{ +- struct ttm_buffer_object *bo; +- int r; +- +- bo = (struct ttm_buffer_object *)vma->vm_private_data; +- if (bo == NULL) { +- return VM_FAULT_NOPAGE; +- } +- r = ttm_vm_ops->fault(vma, vmf); +- return r; +-} +- + int radeon_mmap(struct file *filp, struct vm_area_struct *vma) + { + struct drm_file *file_priv; + struct radeon_device *rdev; +- int r; + + if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET)) { + return drm_mmap(filp, vma); +@@ -528,20 +511,9 @@ int radeon_mmap(struct file *filp, struc + + file_priv = (struct drm_file *)filp->private_data; + rdev = file_priv->minor->dev->dev_private; +- if (rdev == NULL) { ++ if (!rdev) + return -EINVAL; +- } +- r = ttm_bo_mmap(filp, vma, &rdev->mman.bdev); +- if (unlikely(r != 0)) { +- return r; +- } +- if (unlikely(ttm_vm_ops == NULL)) { +- ttm_vm_ops = vma->vm_ops; +- radeon_ttm_vm_ops = *ttm_vm_ops; +- radeon_ttm_vm_ops.fault = &radeon_ttm_fault; +- } +- vma->vm_ops = &radeon_ttm_vm_ops; +- return 0; ++ return ttm_bo_mmap(filp, vma, &rdev->mman.bdev); + } + + +diff -urNp linux-2.6.31.1/drivers/gpu/drm/ttm/ttm_bo_vm.c linux-2.6.31.1/drivers/gpu/drm/ttm/ttm_bo_vm.c +--- linux-2.6.31.1/drivers/gpu/drm/ttm/ttm_bo_vm.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/gpu/drm/ttm/ttm_bo_vm.c 2009-10-01 20:12:43.000000000 -0400 +@@ -73,7 +73,7 @@ static int ttm_bo_vm_fault(struct vm_are + { + struct ttm_buffer_object *bo = (struct ttm_buffer_object *) + vma->vm_private_data; +- struct ttm_bo_device *bdev = bo->bdev; ++ struct ttm_bo_device *bdev; + unsigned long bus_base; + unsigned long bus_offset; + unsigned long bus_size; +@@ -88,6 +88,10 @@ static int ttm_bo_vm_fault(struct vm_are + unsigned long address = (unsigned long)vmf->virtual_address; + int retval = VM_FAULT_NOPAGE; + ++ if (!bo) ++ return VM_FAULT_NOPAGE; ++ bdev = bo->bdev; ++ + /* + * Work around locking order reversal in fault / nopfn + * between mmap_sem and bo_reserve: Perform a trylock operation +@@ -228,7 +232,7 @@ static void ttm_bo_vm_close(struct vm_ar + vma->vm_private_data = NULL; + } + +-static struct vm_operations_struct ttm_bo_vm_ops = { ++static const struct vm_operations_struct ttm_bo_vm_ops = { + .fault = ttm_bo_vm_fault, + .open = ttm_bo_vm_open, + .close = ttm_bo_vm_close +diff -urNp linux-2.6.31.1/drivers/hwmon/fschmd.c linux-2.6.31.1/drivers/hwmon/fschmd.c +--- linux-2.6.31.1/drivers/hwmon/fschmd.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/hwmon/fschmd.c 2009-10-01 20:12:43.000000000 -0400 +@@ -915,7 +915,7 @@ static int watchdog_ioctl(struct inode * + return ret; + } + +-static struct file_operations watchdog_fops = { ++static const struct file_operations watchdog_fops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .open = watchdog_open, +diff -urNp linux-2.6.31.1/drivers/hwmon/fscpos.c linux-2.6.31.1/drivers/hwmon/fscpos.c +--- linux-2.6.31.1/drivers/hwmon/fscpos.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/hwmon/fscpos.c 2009-10-01 20:12:43.000000000 -0400 +@@ -240,7 +240,6 @@ static ssize_t set_pwm(struct i2c_client + unsigned long v = simple_strtoul(buf, NULL, 10); + + /* Range: 0..255 */ +- if (v < 0) v = 0; + if (v > 255) v = 255; + + mutex_lock(&data->update_lock); +diff -urNp linux-2.6.31.1/drivers/hwmon/k8temp.c linux-2.6.31.1/drivers/hwmon/k8temp.c +--- linux-2.6.31.1/drivers/hwmon/k8temp.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/hwmon/k8temp.c 2009-10-01 20:12:43.000000000 -0400 +@@ -138,7 +138,7 @@ static DEVICE_ATTR(name, S_IRUGO, show_n + + static struct pci_device_id k8temp_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) }, +- { 0 }, ++ { 0, 0, 0, 0, 0, 0, 0 }, + }; + + MODULE_DEVICE_TABLE(pci, k8temp_ids); +diff -urNp linux-2.6.31.1/drivers/hwmon/sis5595.c linux-2.6.31.1/drivers/hwmon/sis5595.c +--- linux-2.6.31.1/drivers/hwmon/sis5595.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/hwmon/sis5595.c 2009-10-01 20:12:43.000000000 -0400 +@@ -699,7 +699,7 @@ static struct sis5595_data *sis5595_upda + + static struct pci_device_id sis5595_pci_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_503) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, sis5595_pci_ids); +diff -urNp linux-2.6.31.1/drivers/hwmon/via686a.c linux-2.6.31.1/drivers/hwmon/via686a.c +--- linux-2.6.31.1/drivers/hwmon/via686a.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/hwmon/via686a.c 2009-10-01 20:12:43.000000000 -0400 +@@ -769,7 +769,7 @@ static struct via686a_data *via686a_upda + + static struct pci_device_id via686a_pci_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, via686a_pci_ids); +diff -urNp linux-2.6.31.1/drivers/hwmon/vt8231.c linux-2.6.31.1/drivers/hwmon/vt8231.c +--- linux-2.6.31.1/drivers/hwmon/vt8231.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/hwmon/vt8231.c 2009-10-01 20:12:43.000000000 -0400 +@@ -699,7 +699,7 @@ static struct platform_driver vt8231_dri + + static struct pci_device_id vt8231_pci_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8231_4) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, vt8231_pci_ids); +diff -urNp linux-2.6.31.1/drivers/hwmon/w83791d.c linux-2.6.31.1/drivers/hwmon/w83791d.c +--- linux-2.6.31.1/drivers/hwmon/w83791d.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/hwmon/w83791d.c 2009-10-01 20:12:43.000000000 -0400 +@@ -330,8 +330,8 @@ static int w83791d_detect(struct i2c_cli + struct i2c_board_info *info); + static int w83791d_remove(struct i2c_client *client); + +-static int w83791d_read(struct i2c_client *client, u8 register); +-static int w83791d_write(struct i2c_client *client, u8 register, u8 value); ++static int w83791d_read(struct i2c_client *client, u8 reg); ++static int w83791d_write(struct i2c_client *client, u8 reg, u8 value); + static struct w83791d_data *w83791d_update_device(struct device *dev); + + #ifdef DEBUG +diff -urNp linux-2.6.31.1/drivers/i2c/busses/i2c-i801.c linux-2.6.31.1/drivers/i2c/busses/i2c-i801.c +--- linux-2.6.31.1/drivers/i2c/busses/i2c-i801.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/i2c/busses/i2c-i801.c 2009-10-01 20:12:43.000000000 -0400 +@@ -578,7 +578,7 @@ static struct pci_device_id i801_ids[] = + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH10_4) }, + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH10_5) }, + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_PCH_SMBUS) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE (pci, i801_ids); +diff -urNp linux-2.6.31.1/drivers/i2c/busses/i2c-piix4.c linux-2.6.31.1/drivers/i2c/busses/i2c-piix4.c +--- linux-2.6.31.1/drivers/i2c/busses/i2c-piix4.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/i2c/busses/i2c-piix4.c 2009-10-01 20:12:43.000000000 -0400 +@@ -123,7 +123,7 @@ static struct dmi_system_id __devinitdat + .ident = "IBM", + .matches = { DMI_MATCH(DMI_SYS_VENDOR, "IBM"), }, + }, +- { }, ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, NULL)}, NULL }, + }; + + static int __devinit piix4_setup(struct pci_dev *PIIX4_dev, +@@ -489,7 +489,7 @@ static struct pci_device_id piix4_ids[] + PCI_DEVICE_ID_SERVERWORKS_HT1000SB) }, + { PCI_DEVICE(PCI_VENDOR_ID_SERVERWORKS, + PCI_DEVICE_ID_SERVERWORKS_HT1100LD) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE (pci, piix4_ids); +diff -urNp linux-2.6.31.1/drivers/i2c/busses/i2c-sis630.c linux-2.6.31.1/drivers/i2c/busses/i2c-sis630.c +--- linux-2.6.31.1/drivers/i2c/busses/i2c-sis630.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/i2c/busses/i2c-sis630.c 2009-10-01 20:12:43.000000000 -0400 +@@ -471,7 +471,7 @@ static struct i2c_adapter sis630_adapter + static struct pci_device_id sis630_ids[] __devinitdata = { + { PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_503) }, + { PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_LPC) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE (pci, sis630_ids); +diff -urNp linux-2.6.31.1/drivers/i2c/busses/i2c-sis96x.c linux-2.6.31.1/drivers/i2c/busses/i2c-sis96x.c +--- linux-2.6.31.1/drivers/i2c/busses/i2c-sis96x.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/i2c/busses/i2c-sis96x.c 2009-10-01 20:12:43.000000000 -0400 +@@ -247,7 +247,7 @@ static struct i2c_adapter sis96x_adapter + + static struct pci_device_id sis96x_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_SMBUS) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE (pci, sis96x_ids); +diff -urNp linux-2.6.31.1/drivers/ieee1394/dma.c linux-2.6.31.1/drivers/ieee1394/dma.c +--- linux-2.6.31.1/drivers/ieee1394/dma.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ieee1394/dma.c 2009-10-01 20:12:43.000000000 -0400 +@@ -247,7 +247,7 @@ static int dma_region_pagefault(struct v + return 0; + } + +-static struct vm_operations_struct dma_region_vm_ops = { ++static const struct vm_operations_struct dma_region_vm_ops = { + .fault = dma_region_pagefault, + }; + +diff -urNp linux-2.6.31.1/drivers/ieee1394/dv1394.c linux-2.6.31.1/drivers/ieee1394/dv1394.c +--- linux-2.6.31.1/drivers/ieee1394/dv1394.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ieee1394/dv1394.c 2009-10-01 20:12:43.000000000 -0400 +@@ -739,7 +739,7 @@ static void frame_prepare(struct video_c + based upon DIF section and sequence + */ + +-static void inline ++static inline void + frame_put_packet (struct frame *f, struct packet *p) + { + int section_type = p->data[0] >> 5; /* section type is in bits 5 - 7 */ +@@ -2178,7 +2178,7 @@ static const struct ieee1394_device_id d + .specifier_id = AVC_UNIT_SPEC_ID_ENTRY & 0xffffff, + .version = AVC_SW_VERSION_ENTRY & 0xffffff + }, +- { } ++ { 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(ieee1394, dv1394_id_table); +diff -urNp linux-2.6.31.1/drivers/ieee1394/eth1394.c linux-2.6.31.1/drivers/ieee1394/eth1394.c +--- linux-2.6.31.1/drivers/ieee1394/eth1394.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ieee1394/eth1394.c 2009-10-01 20:12:43.000000000 -0400 +@@ -445,7 +445,7 @@ static const struct ieee1394_device_id e + .specifier_id = ETHER1394_GASP_SPECIFIER_ID, + .version = ETHER1394_GASP_VERSION, + }, +- {} ++ { 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(ieee1394, eth1394_id_table); +diff -urNp linux-2.6.31.1/drivers/ieee1394/hosts.c linux-2.6.31.1/drivers/ieee1394/hosts.c +--- linux-2.6.31.1/drivers/ieee1394/hosts.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ieee1394/hosts.c 2009-10-01 20:12:43.000000000 -0400 +@@ -78,6 +78,7 @@ static int dummy_isoctl(struct hpsb_iso + } + + static struct hpsb_host_driver dummy_driver = { ++ .name = "dummy", + .transmit_packet = dummy_transmit_packet, + .devctl = dummy_devctl, + .isoctl = dummy_isoctl +diff -urNp linux-2.6.31.1/drivers/ieee1394/ohci1394.c linux-2.6.31.1/drivers/ieee1394/ohci1394.c +--- linux-2.6.31.1/drivers/ieee1394/ohci1394.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ieee1394/ohci1394.c 2009-10-01 20:12:43.000000000 -0400 +@@ -147,9 +147,9 @@ printk(level "%s: " fmt "\n" , OHCI1394_ + printk(level "%s: fw-host%d: " fmt "\n" , OHCI1394_DRIVER_NAME, ohci->host->id , ## args) + + /* Module Parameters */ +-static int phys_dma = 1; ++static int phys_dma; + module_param(phys_dma, int, 0444); +-MODULE_PARM_DESC(phys_dma, "Enable physical DMA (default = 1)."); ++MODULE_PARM_DESC(phys_dma, "Enable physical DMA (default = 0)."); + + static void dma_trm_tasklet(unsigned long data); + static void dma_trm_reset(struct dma_trm_ctx *d); +@@ -3449,7 +3449,7 @@ static struct pci_device_id ohci1394_pci + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + }, +- { 0, }, ++ { 0, 0, 0, 0, 0, 0, 0 }, + }; + + MODULE_DEVICE_TABLE(pci, ohci1394_pci_tbl); +diff -urNp linux-2.6.31.1/drivers/ieee1394/raw1394.c linux-2.6.31.1/drivers/ieee1394/raw1394.c +--- linux-2.6.31.1/drivers/ieee1394/raw1394.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ieee1394/raw1394.c 2009-10-01 20:12:43.000000000 -0400 +@@ -2999,7 +2999,7 @@ static const struct ieee1394_device_id r + .match_flags = IEEE1394_MATCH_SPECIFIER_ID | IEEE1394_MATCH_VERSION, + .specifier_id = CAMERA_UNIT_SPEC_ID_ENTRY & 0xffffff, + .version = (CAMERA_SW_VERSION_ENTRY + 2) & 0xffffff}, +- {} ++ { 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(ieee1394, raw1394_id_table); +diff -urNp linux-2.6.31.1/drivers/ieee1394/sbp2.c linux-2.6.31.1/drivers/ieee1394/sbp2.c +--- linux-2.6.31.1/drivers/ieee1394/sbp2.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ieee1394/sbp2.c 2009-10-01 20:12:43.000000000 -0400 +@@ -290,7 +290,7 @@ static const struct ieee1394_device_id s + .match_flags = IEEE1394_MATCH_SPECIFIER_ID | IEEE1394_MATCH_VERSION, + .specifier_id = SBP2_UNIT_SPEC_ID_ENTRY & 0xffffff, + .version = SBP2_SW_VERSION_ENTRY & 0xffffff}, +- {} ++ { 0, 0, 0, 0, 0, 0 } + }; + MODULE_DEVICE_TABLE(ieee1394, sbp2_id_table); + +@@ -2112,7 +2112,7 @@ MODULE_DESCRIPTION("IEEE-1394 SBP-2 prot + MODULE_SUPPORTED_DEVICE(SBP2_DEVICE_NAME); + MODULE_LICENSE("GPL"); + +-static int sbp2_module_init(void) ++static int __init sbp2_module_init(void) + { + int ret; + +diff -urNp linux-2.6.31.1/drivers/ieee1394/video1394.c linux-2.6.31.1/drivers/ieee1394/video1394.c +--- linux-2.6.31.1/drivers/ieee1394/video1394.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/ieee1394/video1394.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1310,7 +1310,7 @@ static const struct ieee1394_device_id v + .specifier_id = CAMERA_UNIT_SPEC_ID_ENTRY & 0xffffff, + .version = (CAMERA_SW_VERSION_ENTRY + 2) & 0xffffff + }, +- { } ++ { 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(ieee1394, video1394_id_table); +diff -urNp linux-2.6.31.1/drivers/infiniband/hw/ehca/ehca_uverbs.c linux-2.6.31.1/drivers/infiniband/hw/ehca/ehca_uverbs.c +--- linux-2.6.31.1/drivers/infiniband/hw/ehca/ehca_uverbs.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/infiniband/hw/ehca/ehca_uverbs.c 2009-10-01 20:12:43.000000000 -0400 +@@ -95,7 +95,7 @@ static void ehca_mm_close(struct vm_area + vma->vm_start, vma->vm_end, *count); + } + +-static struct vm_operations_struct vm_ops = { ++static const struct vm_operations_struct vm_ops = { + .open = ehca_mm_open, + .close = ehca_mm_close, + }; +diff -urNp linux-2.6.31.1/drivers/infiniband/hw/ipath/ipath_file_ops.c linux-2.6.31.1/drivers/infiniband/hw/ipath/ipath_file_ops.c +--- linux-2.6.31.1/drivers/infiniband/hw/ipath/ipath_file_ops.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/infiniband/hw/ipath/ipath_file_ops.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1151,7 +1151,7 @@ static int ipath_file_vma_fault(struct v + return 0; + } + +-static struct vm_operations_struct ipath_file_vm_ops = { ++static const struct vm_operations_struct ipath_file_vm_ops = { + .fault = ipath_file_vma_fault, + }; + +diff -urNp linux-2.6.31.1/drivers/infiniband/hw/ipath/ipath_mmap.c linux-2.6.31.1/drivers/infiniband/hw/ipath/ipath_mmap.c +--- linux-2.6.31.1/drivers/infiniband/hw/ipath/ipath_mmap.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/infiniband/hw/ipath/ipath_mmap.c 2009-10-01 20:12:43.000000000 -0400 +@@ -74,7 +74,7 @@ static void ipath_vma_close(struct vm_ar + kref_put(&ip->ref, ipath_release_mmap_info); + } + +-static struct vm_operations_struct ipath_vm_ops = { ++static const struct vm_operations_struct ipath_vm_ops = { + .open = ipath_vma_open, + .close = ipath_vma_close, + }; +diff -urNp linux-2.6.31.1/drivers/input/keyboard/atkbd.c linux-2.6.31.1/drivers/input/keyboard/atkbd.c +--- linux-2.6.31.1/drivers/input/keyboard/atkbd.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/input/keyboard/atkbd.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1188,7 +1188,7 @@ static struct serio_device_id atkbd_seri + .id = SERIO_ANY, + .extra = SERIO_ANY, + }, +- { 0 } ++ { 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(serio, atkbd_serio_ids); +diff -urNp linux-2.6.31.1/drivers/input/mouse/lifebook.c linux-2.6.31.1/drivers/input/mouse/lifebook.c +--- linux-2.6.31.1/drivers/input/mouse/lifebook.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/input/mouse/lifebook.c 2009-10-01 20:12:43.000000000 -0400 +@@ -116,7 +116,7 @@ static const struct dmi_system_id lifebo + DMI_MATCH(DMI_PRODUCT_NAME, "LifeBook B142"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL} + }; + + static psmouse_ret_t lifebook_process_byte(struct psmouse *psmouse) +diff -urNp linux-2.6.31.1/drivers/input/mouse/psmouse-base.c linux-2.6.31.1/drivers/input/mouse/psmouse-base.c +--- linux-2.6.31.1/drivers/input/mouse/psmouse-base.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/input/mouse/psmouse-base.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1380,7 +1380,7 @@ static struct serio_device_id psmouse_se + .id = SERIO_ANY, + .extra = SERIO_ANY, + }, +- { 0 } ++ { 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(serio, psmouse_serio_ids); +diff -urNp linux-2.6.31.1/drivers/input/mouse/synaptics.c linux-2.6.31.1/drivers/input/mouse/synaptics.c +--- linux-2.6.31.1/drivers/input/mouse/synaptics.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/input/mouse/synaptics.c 2009-10-01 20:12:43.000000000 -0400 +@@ -437,7 +437,7 @@ static void synaptics_process_packet(str + break; + case 2: + if (SYN_MODEL_PEN(priv->model_id)) +- ; /* Nothing, treat a pen as a single finger */ ++ break; /* Nothing, treat a pen as a single finger */ + break; + case 4 ... 15: + if (SYN_CAP_PALMDETECT(priv->capabilities)) +@@ -653,7 +653,7 @@ static const struct dmi_system_id toshib + DMI_MATCH(DMI_PRODUCT_NAME, "PORTEGE M300"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + #endif + +diff -urNp linux-2.6.31.1/drivers/input/mousedev.c linux-2.6.31.1/drivers/input/mousedev.c +--- linux-2.6.31.1/drivers/input/mousedev.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/input/mousedev.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1056,7 +1056,7 @@ static struct input_handler mousedev_han + + #ifdef CONFIG_INPUT_MOUSEDEV_PSAUX + static struct miscdevice psaux_mouse = { +- PSMOUSE_MINOR, "psaux", &mousedev_fops ++ PSMOUSE_MINOR, "psaux", &mousedev_fops, {NULL, NULL}, NULL, NULL + }; + static int psaux_registered; + #endif +diff -urNp linux-2.6.31.1/drivers/input/serio/i8042-x86ia64io.h linux-2.6.31.1/drivers/input/serio/i8042-x86ia64io.h +--- linux-2.6.31.1/drivers/input/serio/i8042-x86ia64io.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/input/serio/i8042-x86ia64io.h 2009-10-01 20:12:43.000000000 -0400 +@@ -167,7 +167,7 @@ static struct dmi_system_id __initdata i + DMI_MATCH(DMI_PRODUCT_VERSION, "Rev 1"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + /* +@@ -390,7 +390,7 @@ static struct dmi_system_id __initdata i + DMI_MATCH(DMI_PRODUCT_VERSION, "0100"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + static struct dmi_system_id __initdata i8042_dmi_reset_table[] = { +@@ -436,7 +436,7 @@ static struct dmi_system_id __initdata i + DMI_MATCH(DMI_PRODUCT_NAME, "N10"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + #ifdef CONFIG_PNP +@@ -455,7 +455,7 @@ static struct dmi_system_id __initdata i + DMI_MATCH(DMI_BOARD_VENDOR, "MICRO-STAR INTERNATIONAL CO., LTD"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + #endif + +@@ -522,7 +522,7 @@ static struct dmi_system_id __initdata i + DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 4280"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + #endif /* CONFIG_X86 */ +diff -urNp linux-2.6.31.1/drivers/input/serio/serio_raw.c linux-2.6.31.1/drivers/input/serio/serio_raw.c +--- linux-2.6.31.1/drivers/input/serio/serio_raw.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/input/serio/serio_raw.c 2009-10-01 20:12:43.000000000 -0400 +@@ -376,7 +376,7 @@ static struct serio_device_id serio_raw_ + .id = SERIO_ANY, + .extra = SERIO_ANY, + }, +- { 0 } ++ { 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(serio, serio_raw_serio_ids); +diff -urNp linux-2.6.31.1/drivers/isdn/capi/kcapi_proc.c linux-2.6.31.1/drivers/isdn/capi/kcapi_proc.c +--- linux-2.6.31.1/drivers/isdn/capi/kcapi_proc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/isdn/capi/kcapi_proc.c 2009-10-01 20:12:43.000000000 -0400 +@@ -89,14 +89,14 @@ static int contrstats_show(struct seq_fi + return 0; + } + +-static struct seq_operations seq_controller_ops = { ++static const struct seq_operations seq_controller_ops = { + .start = controller_start, + .next = controller_next, + .stop = controller_stop, + .show = controller_show, + }; + +-static struct seq_operations seq_contrstats_ops = { ++static const struct seq_operations seq_contrstats_ops = { + .start = controller_start, + .next = controller_next, + .stop = controller_stop, +@@ -194,14 +194,14 @@ applstats_show(struct seq_file *seq, voi + return 0; + } + +-static struct seq_operations seq_applications_ops = { ++static const struct seq_operations seq_applications_ops = { + .start = applications_start, + .next = applications_next, + .stop = applications_stop, + .show = applications_show, + }; + +-static struct seq_operations seq_applstats_ops = { ++static const struct seq_operations seq_applstats_ops = { + .start = applications_start, + .next = applications_next, + .stop = applications_stop, +@@ -264,7 +264,7 @@ static int capi_driver_show(struct seq_f + return 0; + } + +-static struct seq_operations seq_capi_driver_ops = { ++static const struct seq_operations seq_capi_driver_ops = { + .start = capi_driver_start, + .next = capi_driver_next, + .stop = capi_driver_stop, +diff -urNp linux-2.6.31.1/drivers/isdn/gigaset/common.c linux-2.6.31.1/drivers/isdn/gigaset/common.c +--- linux-2.6.31.1/drivers/isdn/gigaset/common.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/isdn/gigaset/common.c 2009-10-01 20:12:43.000000000 -0400 +@@ -665,7 +665,7 @@ struct cardstate *gigaset_initcs(struct + cs->commands_pending = 0; + cs->cur_at_seq = 0; + cs->gotfwver = -1; +- cs->open_count = 0; ++ atomic_set(&cs->open_count, 0); + cs->dev = NULL; + cs->tty = NULL; + cs->tty_dev = NULL; +diff -urNp linux-2.6.31.1/drivers/isdn/gigaset/gigaset.h linux-2.6.31.1/drivers/isdn/gigaset/gigaset.h +--- linux-2.6.31.1/drivers/isdn/gigaset/gigaset.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/isdn/gigaset/gigaset.h 2009-10-01 20:12:43.000000000 -0400 +@@ -446,7 +446,7 @@ struct cardstate { + spinlock_t cmdlock; + unsigned curlen, cmdbytes; + +- unsigned open_count; ++ atomic_t open_count; + struct tty_struct *tty; + struct tasklet_struct if_wake_tasklet; + unsigned control_state; +diff -urNp linux-2.6.31.1/drivers/isdn/gigaset/interface.c linux-2.6.31.1/drivers/isdn/gigaset/interface.c +--- linux-2.6.31.1/drivers/isdn/gigaset/interface.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/isdn/gigaset/interface.c 2009-10-01 20:12:43.000000000 -0400 +@@ -165,9 +165,7 @@ static int if_open(struct tty_struct *tt + return -ERESTARTSYS; // FIXME -EINTR? + tty->driver_data = cs; + +- ++cs->open_count; +- +- if (cs->open_count == 1) { ++ if (atomic_inc_return(&cs->open_count) == 1) { + spin_lock_irqsave(&cs->lock, flags); + cs->tty = tty; + spin_unlock_irqrestore(&cs->lock, flags); +@@ -195,10 +193,10 @@ static void if_close(struct tty_struct * + + if (!cs->connected) + gig_dbg(DEBUG_IF, "not connected"); /* nothing to do */ +- else if (!cs->open_count) ++ else if (!atomic_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else { +- if (!--cs->open_count) { ++ if (!atomic_dec_return(&cs->open_count)) { + spin_lock_irqsave(&cs->lock, flags); + cs->tty = NULL; + spin_unlock_irqrestore(&cs->lock, flags); +@@ -233,7 +231,7 @@ static int if_ioctl(struct tty_struct *t + if (!cs->connected) { + gig_dbg(DEBUG_IF, "not connected"); + retval = -ENODEV; +- } else if (!cs->open_count) ++ } else if (!atomic_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else { + retval = 0; +@@ -361,7 +359,7 @@ static int if_write(struct tty_struct *t + if (!cs->connected) { + gig_dbg(DEBUG_IF, "not connected"); + retval = -ENODEV; +- } else if (!cs->open_count) ++ } else if (!atomic_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else if (cs->mstate != MS_LOCKED) { + dev_warn(cs->dev, "can't write to unlocked device\n"); +@@ -395,7 +393,7 @@ static int if_write_room(struct tty_stru + if (!cs->connected) { + gig_dbg(DEBUG_IF, "not connected"); + retval = -ENODEV; +- } else if (!cs->open_count) ++ } else if (!atomic_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else if (cs->mstate != MS_LOCKED) { + dev_warn(cs->dev, "can't write to unlocked device\n"); +@@ -429,7 +427,7 @@ static int if_chars_in_buffer(struct tty + if (!cs->connected) { + gig_dbg(DEBUG_IF, "not connected"); + retval = -ENODEV; +- } else if (!cs->open_count) ++ } else if (!atomic_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else if (cs->mstate != MS_LOCKED) { + dev_warn(cs->dev, "can't write to unlocked device\n"); +@@ -458,7 +456,7 @@ static void if_throttle(struct tty_struc + + if (!cs->connected) + gig_dbg(DEBUG_IF, "not connected"); /* nothing to do */ +- else if (!cs->open_count) ++ else if (!atomic_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else { + //FIXME +@@ -483,7 +481,7 @@ static void if_unthrottle(struct tty_str + + if (!cs->connected) + gig_dbg(DEBUG_IF, "not connected"); /* nothing to do */ +- else if (!cs->open_count) ++ else if (!atomic_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else { + //FIXME +@@ -515,7 +513,7 @@ static void if_set_termios(struct tty_st + goto out; + } + +- if (!cs->open_count) { ++ if (!atomic_read(&cs->open_count)) { + dev_warn(cs->dev, "%s: device not opened\n", __func__); + goto out; + } +diff -urNp linux-2.6.31.1/drivers/lguest/core.c linux-2.6.31.1/drivers/lguest/core.c +--- linux-2.6.31.1/drivers/lguest/core.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/lguest/core.c 2009-10-01 20:12:43.000000000 -0400 +@@ -92,9 +92,17 @@ static __init int map_switcher(void) + * it's worked so far. The end address needs +1 because __get_vm_area + * allocates an extra guard page, so we need space for that. + */ ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ switcher_vma = __get_vm_area(TOTAL_SWITCHER_PAGES * PAGE_SIZE, ++ VM_ALLOC | VM_KERNEXEC, SWITCHER_ADDR, SWITCHER_ADDR ++ + (TOTAL_SWITCHER_PAGES+1) * PAGE_SIZE); ++#else + switcher_vma = __get_vm_area(TOTAL_SWITCHER_PAGES * PAGE_SIZE, + VM_ALLOC, SWITCHER_ADDR, SWITCHER_ADDR + + (TOTAL_SWITCHER_PAGES+1) * PAGE_SIZE); ++#endif ++ + if (!switcher_vma) { + err = -ENOMEM; + printk("lguest: could not map switcher pages high\n"); +diff -urNp linux-2.6.31.1/drivers/lguest/lguest_user.c linux-2.6.31.1/drivers/lguest/lguest_user.c +--- linux-2.6.31.1/drivers/lguest/lguest_user.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/lguest/lguest_user.c 2009-10-01 20:12:43.000000000 -0400 +@@ -508,7 +508,7 @@ static int close(struct inode *inode, st + * uses: reading and writing a character device called /dev/lguest. All the + * work happens in the read(), write() and close() routines: + */ +-static struct file_operations lguest_fops = { ++static const struct file_operations lguest_fops = { + .owner = THIS_MODULE, + .release = close, + .write = write, +diff -urNp linux-2.6.31.1/drivers/md/bitmap.c linux-2.6.31.1/drivers/md/bitmap.c +--- linux-2.6.31.1/drivers/md/bitmap.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/md/bitmap.c 2009-10-01 20:12:43.000000000 -0400 +@@ -58,7 +58,7 @@ + # if DEBUG > 0 + # define PRINTK(x...) printk(KERN_DEBUG x) + # else +-# define PRINTK(x...) ++# define PRINTK(x...) do {} while (0) + # endif + #endif + +diff -urNp linux-2.6.31.1/drivers/md/dm-table.c linux-2.6.31.1/drivers/md/dm-table.c +--- linux-2.6.31.1/drivers/md/dm-table.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/md/dm-table.c 2009-10-01 20:12:43.000000000 -0400 +@@ -359,7 +359,7 @@ static int device_area_is_invalid(struct + if (!dev_size) + return 0; + +- if ((start >= dev_size) || (start + len > dev_size)) { ++ if ((start >= dev_size) || (len > dev_size - start)) { + DMWARN("%s: %s too small for target: " + "start=%llu, len=%llu, dev_size=%llu", + dm_device_name(ti->table->md), bdevname(bdev, b), +diff -urNp linux-2.6.31.1/drivers/md/md.c linux-2.6.31.1/drivers/md/md.c +--- linux-2.6.31.1/drivers/md/md.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/md/md.c 2009-10-01 20:12:43.000000000 -0400 +@@ -5963,7 +5963,7 @@ static int md_seq_show(struct seq_file * + chunk_kb ? "KB" : "B"); + if (bitmap->file) { + seq_printf(seq, ", file: "); +- seq_path(seq, &bitmap->file->f_path, " \t\n"); ++ seq_path(seq, &bitmap->file->f_path, " \t\n\\"); + } + + seq_printf(seq, "\n"); +@@ -6057,7 +6057,7 @@ static int is_mddev_idle(mddev_t *mddev, + struct gendisk *disk = rdev->bdev->bd_contains->bd_disk; + curr_events = (int)part_stat_read(&disk->part0, sectors[0]) + + (int)part_stat_read(&disk->part0, sectors[1]) - +- atomic_read(&disk->sync_io); ++ atomic_read_unchecked(&disk->sync_io); + /* sync IO will cause sync_io to increase before the disk_stats + * as sync_io is counted when a request starts, and + * disk_stats is counted when it completes. +diff -urNp linux-2.6.31.1/drivers/md/md.h linux-2.6.31.1/drivers/md/md.h +--- linux-2.6.31.1/drivers/md/md.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/md/md.h 2009-10-01 20:12:43.000000000 -0400 +@@ -303,7 +303,7 @@ static inline void rdev_dec_pending(mdk_ + + static inline void md_sync_acct(struct block_device *bdev, unsigned long nr_sectors) + { +- atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io); ++ atomic_add_unchecked(nr_sectors, &bdev->bd_contains->bd_disk->sync_io); + } + + struct mdk_personality +diff -urNp linux-2.6.31.1/drivers/media/dvb/dvb-core/dmxdev.c linux-2.6.31.1/drivers/media/dvb/dvb-core/dmxdev.c +--- linux-2.6.31.1/drivers/media/dvb/dvb-core/dmxdev.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/dvb/dvb-core/dmxdev.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1086,7 +1086,7 @@ static unsigned int dvb_dvr_poll(struct + return mask; + } + +-static struct file_operations dvb_dvr_fops = { ++static const struct file_operations dvb_dvr_fops = { + .owner = THIS_MODULE, + .read = dvb_dvr_read, + .write = dvb_dvr_write, +diff -urNp linux-2.6.31.1/drivers/media/dvb/firewire/firedtv-ci.c linux-2.6.31.1/drivers/media/dvb/firewire/firedtv-ci.c +--- linux-2.6.31.1/drivers/media/dvb/firewire/firedtv-ci.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/dvb/firewire/firedtv-ci.c 2009-10-01 20:12:43.000000000 -0400 +@@ -215,7 +215,7 @@ static unsigned int fdtv_ca_io_poll(stru + return POLLIN; + } + +-static struct file_operations fdtv_ca_fops = { ++static const struct file_operations fdtv_ca_fops = { + .owner = THIS_MODULE, + .ioctl = dvb_generic_ioctl, + .open = dvb_generic_open, +diff -urNp linux-2.6.31.1/drivers/media/video/cafe_ccic.c linux-2.6.31.1/drivers/media/video/cafe_ccic.c +--- linux-2.6.31.1/drivers/media/video/cafe_ccic.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/cafe_ccic.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1326,7 +1326,7 @@ static void cafe_v4l_vm_close(struct vm_ + mutex_unlock(&sbuf->cam->s_mutex); + } + +-static struct vm_operations_struct cafe_v4l_vm_ops = { ++static const struct vm_operations_struct cafe_v4l_vm_ops = { + .open = cafe_v4l_vm_open, + .close = cafe_v4l_vm_close + }; +diff -urNp linux-2.6.31.1/drivers/media/video/et61x251/et61x251_core.c linux-2.6.31.1/drivers/media/video/et61x251/et61x251_core.c +--- linux-2.6.31.1/drivers/media/video/et61x251/et61x251_core.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/et61x251/et61x251_core.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1494,7 +1494,7 @@ static void et61x251_vm_close(struct vm_ + } + + +-static struct vm_operations_struct et61x251_vm_ops = { ++static const struct vm_operations_struct et61x251_vm_ops = { + .open = et61x251_vm_open, + .close = et61x251_vm_close, + }; +diff -urNp linux-2.6.31.1/drivers/media/video/gspca/gspca.c linux-2.6.31.1/drivers/media/video/gspca/gspca.c +--- linux-2.6.31.1/drivers/media/video/gspca/gspca.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/gspca/gspca.c 2009-10-01 20:12:43.000000000 -0400 +@@ -99,7 +99,7 @@ static void gspca_vm_close(struct vm_are + frame->v4l2_buf.flags &= ~V4L2_BUF_FLAG_MAPPED; + } + +-static struct vm_operations_struct gspca_vm_ops = { ++static const struct vm_operations_struct gspca_vm_ops = { + .open = gspca_vm_open, + .close = gspca_vm_close, + }; +diff -urNp linux-2.6.31.1/drivers/media/video/meye.c linux-2.6.31.1/drivers/media/video/meye.c +--- linux-2.6.31.1/drivers/media/video/meye.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/meye.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1589,7 +1589,7 @@ static void meye_vm_close(struct vm_area + meye.vma_use_count[idx]--; + } + +-static struct vm_operations_struct meye_vm_ops = { ++static const struct vm_operations_struct meye_vm_ops = { + .open = meye_vm_open, + .close = meye_vm_close, + }; +diff -urNp linux-2.6.31.1/drivers/media/video/sn9c102/sn9c102_core.c linux-2.6.31.1/drivers/media/video/sn9c102/sn9c102_core.c +--- linux-2.6.31.1/drivers/media/video/sn9c102/sn9c102_core.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/sn9c102/sn9c102_core.c 2009-10-01 20:12:43.000000000 -0400 +@@ -2075,7 +2075,7 @@ static void sn9c102_vm_close(struct vm_a + } + + +-static struct vm_operations_struct sn9c102_vm_ops = { ++static const struct vm_operations_struct sn9c102_vm_ops = { + .open = sn9c102_vm_open, + .close = sn9c102_vm_close, + }; +diff -urNp linux-2.6.31.1/drivers/media/video/stk-webcam.c linux-2.6.31.1/drivers/media/video/stk-webcam.c +--- linux-2.6.31.1/drivers/media/video/stk-webcam.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/stk-webcam.c 2009-10-01 20:12:43.000000000 -0400 +@@ -790,7 +790,7 @@ static void stk_v4l_vm_close(struct vm_a + if (sbuf->mapcount == 0) + sbuf->v4lbuf.flags &= ~V4L2_BUF_FLAG_MAPPED; + } +-static struct vm_operations_struct stk_v4l_vm_ops = { ++static const struct vm_operations_struct stk_v4l_vm_ops = { + .open = stk_v4l_vm_open, + .close = stk_v4l_vm_close + }; +diff -urNp linux-2.6.31.1/drivers/media/video/usbvideo/konicawc.c linux-2.6.31.1/drivers/media/video/usbvideo/konicawc.c +--- linux-2.6.31.1/drivers/media/video/usbvideo/konicawc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/usbvideo/konicawc.c 2009-10-01 20:12:43.000000000 -0400 +@@ -225,7 +225,7 @@ static void konicawc_register_input(stru + int error; + + usb_make_path(dev, cam->input_physname, sizeof(cam->input_physname)); +- strncat(cam->input_physname, "/input0", sizeof(cam->input_physname)); ++ strlcat(cam->input_physname, "/input0", sizeof(cam->input_physname)); + + cam->input = input_dev = input_allocate_device(); + if (!input_dev) { +diff -urNp linux-2.6.31.1/drivers/media/video/usbvideo/quickcam_messenger.c linux-2.6.31.1/drivers/media/video/usbvideo/quickcam_messenger.c +--- linux-2.6.31.1/drivers/media/video/usbvideo/quickcam_messenger.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/usbvideo/quickcam_messenger.c 2009-10-01 20:12:43.000000000 -0400 +@@ -89,7 +89,7 @@ static void qcm_register_input(struct qc + int error; + + usb_make_path(dev, cam->input_physname, sizeof(cam->input_physname)); +- strncat(cam->input_physname, "/input0", sizeof(cam->input_physname)); ++ strlcat(cam->input_physname, "/input0", sizeof(cam->input_physname)); + + cam->input = input_dev = input_allocate_device(); + if (!input_dev) { +diff -urNp linux-2.6.31.1/drivers/media/video/uvc/uvc_v4l2.c linux-2.6.31.1/drivers/media/video/uvc/uvc_v4l2.c +--- linux-2.6.31.1/drivers/media/video/uvc/uvc_v4l2.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/uvc/uvc_v4l2.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1063,7 +1063,7 @@ static void uvc_vm_close(struct vm_area_ + buffer->vma_use_count--; + } + +-static struct vm_operations_struct uvc_vm_ops = { ++static const struct vm_operations_struct uvc_vm_ops = { + .open = uvc_vm_open, + .close = uvc_vm_close, + }; +diff -urNp linux-2.6.31.1/drivers/media/video/videobuf-dma-contig.c linux-2.6.31.1/drivers/media/video/videobuf-dma-contig.c +--- linux-2.6.31.1/drivers/media/video/videobuf-dma-contig.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/videobuf-dma-contig.c 2009-10-01 20:12:43.000000000 -0400 +@@ -105,7 +105,7 @@ static void videobuf_vm_close(struct vm_ + } + } + +-static struct vm_operations_struct videobuf_vm_ops = { ++static const struct vm_operations_struct videobuf_vm_ops = { + .open = videobuf_vm_open, + .close = videobuf_vm_close, + }; +diff -urNp linux-2.6.31.1/drivers/media/video/vino.c linux-2.6.31.1/drivers/media/video/vino.c +--- linux-2.6.31.1/drivers/media/video/vino.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/vino.c 2009-10-01 20:12:43.000000000 -0400 +@@ -3858,7 +3858,7 @@ static void vino_vm_close(struct vm_area + dprintk("vino_vm_close(): count = %d\n", fb->map_count); + } + +-static struct vm_operations_struct vino_vm_ops = { ++static const struct vm_operations_struct vino_vm_ops = { + .open = vino_vm_open, + .close = vino_vm_close, + }; +diff -urNp linux-2.6.31.1/drivers/media/video/zc0301/zc0301_core.c linux-2.6.31.1/drivers/media/video/zc0301/zc0301_core.c +--- linux-2.6.31.1/drivers/media/video/zc0301/zc0301_core.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/zc0301/zc0301_core.c 2009-10-01 20:12:43.000000000 -0400 +@@ -933,7 +933,7 @@ static void zc0301_vm_close(struct vm_ar + } + + +-static struct vm_operations_struct zc0301_vm_ops = { ++static const struct vm_operations_struct zc0301_vm_ops = { + .open = zc0301_vm_open, + .close = zc0301_vm_close, + }; +diff -urNp linux-2.6.31.1/drivers/media/video/zoran/zoran_driver.c linux-2.6.31.1/drivers/media/video/zoran/zoran_driver.c +--- linux-2.6.31.1/drivers/media/video/zoran/zoran_driver.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/media/video/zoran/zoran_driver.c 2009-10-01 20:12:43.000000000 -0400 +@@ -3172,7 +3172,7 @@ zoran_vm_close (struct vm_area_struct *v + mutex_unlock(&zr->resource_lock); + } + +-static struct vm_operations_struct zoran_vm_ops = { ++static const struct vm_operations_struct zoran_vm_ops = { + .open = zoran_vm_open, + .close = zoran_vm_close, + }; +diff -urNp linux-2.6.31.1/drivers/message/i2o/i2o_proc.c linux-2.6.31.1/drivers/message/i2o/i2o_proc.c +--- linux-2.6.31.1/drivers/message/i2o/i2o_proc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/message/i2o/i2o_proc.c 2009-10-01 20:12:43.000000000 -0400 +@@ -259,13 +259,6 @@ static char *scsi_devices[] = { + "Array Controller Device" + }; + +-static char *chtostr(u8 * chars, int n) +-{ +- char tmp[256]; +- tmp[0] = 0; +- return strncat(tmp, (char *)chars, n); +-} +- + static int i2o_report_query_status(struct seq_file *seq, int block_status, + char *group) + { +@@ -842,8 +835,7 @@ static int i2o_seq_show_ddm_table(struct + + seq_printf(seq, "%-#7x", ddm_table.i2o_vendor_id); + seq_printf(seq, "%-#8x", ddm_table.module_id); +- seq_printf(seq, "%-29s", +- chtostr(ddm_table.module_name_version, 28)); ++ seq_printf(seq, "%-.28s", ddm_table.module_name_version); + seq_printf(seq, "%9d ", ddm_table.data_size); + seq_printf(seq, "%8d", ddm_table.code_size); + +@@ -944,8 +936,8 @@ static int i2o_seq_show_drivers_stored(s + + seq_printf(seq, "%-#7x", dst->i2o_vendor_id); + seq_printf(seq, "%-#8x", dst->module_id); +- seq_printf(seq, "%-29s", chtostr(dst->module_name_version, 28)); +- seq_printf(seq, "%-9s", chtostr(dst->date, 8)); ++ seq_printf(seq, "%-.28s", dst->module_name_version); ++ seq_printf(seq, "%-.8s", dst->date); + seq_printf(seq, "%8d ", dst->module_size); + seq_printf(seq, "%8d ", dst->mpb_size); + seq_printf(seq, "0x%04x", dst->module_flags); +@@ -1276,14 +1268,10 @@ static int i2o_seq_show_dev_identity(str + seq_printf(seq, "Device Class : %s\n", i2o_get_class_name(work16[0])); + seq_printf(seq, "Owner TID : %0#5x\n", work16[2]); + seq_printf(seq, "Parent TID : %0#5x\n", work16[3]); +- seq_printf(seq, "Vendor info : %s\n", +- chtostr((u8 *) (work32 + 2), 16)); +- seq_printf(seq, "Product info : %s\n", +- chtostr((u8 *) (work32 + 6), 16)); +- seq_printf(seq, "Description : %s\n", +- chtostr((u8 *) (work32 + 10), 16)); +- seq_printf(seq, "Product rev. : %s\n", +- chtostr((u8 *) (work32 + 14), 8)); ++ seq_printf(seq, "Vendor info : %.16s\n", (u8 *) (work32 + 2)); ++ seq_printf(seq, "Product info : %.16s\n", (u8 *) (work32 + 6)); ++ seq_printf(seq, "Description : %.16s\n", (u8 *) (work32 + 10)); ++ seq_printf(seq, "Product rev. : %.8s\n", (u8 *) (work32 + 14)); + + seq_printf(seq, "Serial number : "); + print_serial_number(seq, (u8 *) (work32 + 16), +@@ -1328,10 +1316,8 @@ static int i2o_seq_show_ddm_identity(str + } + + seq_printf(seq, "Registering DDM TID : 0x%03x\n", result.ddm_tid); +- seq_printf(seq, "Module name : %s\n", +- chtostr(result.module_name, 24)); +- seq_printf(seq, "Module revision : %s\n", +- chtostr(result.module_rev, 8)); ++ seq_printf(seq, "Module name : %.24s\n", result.module_name); ++ seq_printf(seq, "Module revision : %.8s\n", result.module_rev); + + seq_printf(seq, "Serial number : "); + print_serial_number(seq, result.serial_number, sizeof(result) - 36); +@@ -1362,14 +1348,10 @@ static int i2o_seq_show_uinfo(struct seq + return 0; + } + +- seq_printf(seq, "Device name : %s\n", +- chtostr(result.device_name, 64)); +- seq_printf(seq, "Service name : %s\n", +- chtostr(result.service_name, 64)); +- seq_printf(seq, "Physical name : %s\n", +- chtostr(result.physical_location, 64)); +- seq_printf(seq, "Instance number : %s\n", +- chtostr(result.instance_number, 4)); ++ seq_printf(seq, "Device name : %.64s\n", result.device_name); ++ seq_printf(seq, "Service name : %.64s\n", result.service_name); ++ seq_printf(seq, "Physical name : %.64s\n", result.physical_location); ++ seq_printf(seq, "Instance number : %.4s\n", result.instance_number); + + return 0; + } +diff -urNp linux-2.6.31.1/drivers/misc/ibmasm/ibmasmfs.c linux-2.6.31.1/drivers/misc/ibmasm/ibmasmfs.c +--- linux-2.6.31.1/drivers/misc/ibmasm/ibmasmfs.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/misc/ibmasm/ibmasmfs.c 2009-10-01 20:12:43.000000000 -0400 +@@ -97,7 +97,7 @@ static int ibmasmfs_get_super(struct fil + return get_sb_single(fst, flags, data, ibmasmfs_fill_super, mnt); + } + +-static struct super_operations ibmasmfs_s_ops = { ++static const struct super_operations ibmasmfs_s_ops = { + .statfs = simple_statfs, + .drop_inode = generic_delete_inode, + }; +diff -urNp linux-2.6.31.1/drivers/misc/phantom.c linux-2.6.31.1/drivers/misc/phantom.c +--- linux-2.6.31.1/drivers/misc/phantom.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/misc/phantom.c 2009-10-01 20:12:43.000000000 -0400 +@@ -271,7 +271,7 @@ static unsigned int phantom_poll(struct + return mask; + } + +-static struct file_operations phantom_file_ops = { ++static const struct file_operations phantom_file_ops = { + .open = phantom_open, + .release = phantom_release, + .unlocked_ioctl = phantom_ioctl, +diff -urNp linux-2.6.31.1/drivers/misc/sgi-gru/grufile.c linux-2.6.31.1/drivers/misc/sgi-gru/grufile.c +--- linux-2.6.31.1/drivers/misc/sgi-gru/grufile.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/misc/sgi-gru/grufile.c 2009-10-01 20:12:43.000000000 -0400 +@@ -53,7 +53,7 @@ struct gru_stats_s gru_stats; + /* Guaranteed user available resources on each node */ + static int max_user_cbrs, max_user_dsr_bytes; + +-static struct file_operations gru_fops; ++static const struct file_operations gru_fops; + static struct miscdevice gru_miscdev; + + +@@ -426,7 +426,7 @@ static void __exit gru_exit(void) + gru_proc_exit(); + } + +-static struct file_operations gru_fops = { ++static const struct file_operations gru_fops = { + .owner = THIS_MODULE, + .unlocked_ioctl = gru_file_unlocked_ioctl, + .mmap = gru_file_mmap, +@@ -438,7 +438,7 @@ static struct miscdevice gru_miscdev = { + .fops = &gru_fops, + }; + +-struct vm_operations_struct gru_vm_ops = { ++const struct vm_operations_struct gru_vm_ops = { + .close = gru_vma_close, + .fault = gru_fault, + }; +diff -urNp linux-2.6.31.1/drivers/misc/sgi-gru/grutables.h linux-2.6.31.1/drivers/misc/sgi-gru/grutables.h +--- linux-2.6.31.1/drivers/misc/sgi-gru/grutables.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/misc/sgi-gru/grutables.h 2009-10-01 20:12:43.000000000 -0400 +@@ -624,7 +624,7 @@ static inline int is_kernel_context(stru + */ + struct gru_unload_context_req; + +-extern struct vm_operations_struct gru_vm_ops; ++extern const struct vm_operations_struct gru_vm_ops; + extern struct device *grudev; + + extern struct gru_vma_data *gru_alloc_vma_data(struct vm_area_struct *vma, +diff -urNp linux-2.6.31.1/drivers/mmc/core/debugfs.c linux-2.6.31.1/drivers/mmc/core/debugfs.c +--- linux-2.6.31.1/drivers/mmc/core/debugfs.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/mmc/core/debugfs.c 2009-10-01 20:12:43.000000000 -0400 +@@ -240,7 +240,7 @@ static int mmc_ext_csd_release(struct in + return 0; + } + +-static struct file_operations mmc_dbg_ext_csd_fops = { ++static const struct file_operations mmc_dbg_ext_csd_fops = { + .open = mmc_ext_csd_open, + .read = mmc_ext_csd_read, + .release = mmc_ext_csd_release, +diff -urNp linux-2.6.31.1/drivers/mtd/devices/doc2000.c linux-2.6.31.1/drivers/mtd/devices/doc2000.c +--- linux-2.6.31.1/drivers/mtd/devices/doc2000.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/mtd/devices/doc2000.c 2009-10-01 20:12:43.000000000 -0400 +@@ -776,7 +776,7 @@ static int doc_write(struct mtd_info *mt + + /* The ECC will not be calculated correctly if less than 512 is written */ + /* DBB- +- if (len != 0x200 && eccbuf) ++ if (len != 0x200) + printk(KERN_WARNING + "ECC needs a full sector write (adr: %lx size %lx)\n", + (long) to, (long) len); +diff -urNp linux-2.6.31.1/drivers/mtd/devices/doc2001.c linux-2.6.31.1/drivers/mtd/devices/doc2001.c +--- linux-2.6.31.1/drivers/mtd/devices/doc2001.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/mtd/devices/doc2001.c 2009-10-01 20:12:43.000000000 -0400 +@@ -395,6 +395,8 @@ static int doc_read (struct mtd_info *mt + /* Don't allow read past end of device */ + if (from >= this->totlen) + return -EINVAL; ++ if (!len) ++ return -EINVAL; + + /* Don't allow a single read to cross a 512-byte block boundary */ + if (from + len > ((from | 0x1ff) + 1)) +diff -urNp linux-2.6.31.1/drivers/mtd/ubi/build.c linux-2.6.31.1/drivers/mtd/ubi/build.c +--- linux-2.6.31.1/drivers/mtd/ubi/build.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/mtd/ubi/build.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1257,7 +1257,7 @@ static int __init bytes_str_to_int(const + unsigned long result; + + result = simple_strtoul(str, &endp, 0); +- if (str == endp || result < 0) { ++ if (str == endp) { + printk(KERN_ERR "UBI error: incorrect bytes count: \"%s\"\n", + str); + return -EINVAL; +diff -urNp linux-2.6.31.1/drivers/net/irda/vlsi_ir.c linux-2.6.31.1/drivers/net/irda/vlsi_ir.c +--- linux-2.6.31.1/drivers/net/irda/vlsi_ir.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/net/irda/vlsi_ir.c 2009-10-01 20:12:43.000000000 -0400 +@@ -906,13 +906,12 @@ static int vlsi_hard_start_xmit(struct s + /* no race - tx-ring already empty */ + vlsi_set_baud(idev, iobase); + netif_wake_queue(ndev); +- } +- else +- ; ++ } else { + /* keep the speed change pending like it would + * for any len>0 packet. tx completion interrupt + * will apply it when the tx ring becomes empty. + */ ++ } + spin_unlock_irqrestore(&idev->lock, flags); + dev_kfree_skb_any(skb); + return 0; +diff -urNp linux-2.6.31.1/drivers/net/pcnet32.c linux-2.6.31.1/drivers/net/pcnet32.c +--- linux-2.6.31.1/drivers/net/pcnet32.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/net/pcnet32.c 2009-10-01 20:12:43.000000000 -0400 +@@ -78,7 +78,7 @@ static int cards_found; + /* + * VLB I/O addresses + */ +-static unsigned int pcnet32_portlist[] __initdata = ++static unsigned int pcnet32_portlist[] __devinitdata = + { 0x300, 0x320, 0x340, 0x360, 0 }; + + static int pcnet32_debug = 0; +diff -urNp linux-2.6.31.1/drivers/net/tg3.h linux-2.6.31.1/drivers/net/tg3.h +--- linux-2.6.31.1/drivers/net/tg3.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/net/tg3.h 2009-10-01 20:12:43.000000000 -0400 +@@ -89,6 +89,7 @@ + #define CHIPREV_ID_5750_A0 0x4000 + #define CHIPREV_ID_5750_A1 0x4001 + #define CHIPREV_ID_5750_A3 0x4003 ++#define CHIPREV_ID_5750_C1 0x4201 + #define CHIPREV_ID_5750_C2 0x4202 + #define CHIPREV_ID_5752_A0_HW 0x5000 + #define CHIPREV_ID_5752_A0 0x6000 +diff -urNp linux-2.6.31.1/drivers/net/usb/hso.c linux-2.6.31.1/drivers/net/usb/hso.c +--- linux-2.6.31.1/drivers/net/usb/hso.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/net/usb/hso.c 2009-10-01 20:12:43.000000000 -0400 +@@ -258,7 +258,7 @@ struct hso_serial { + + /* from usb_serial_port */ + struct tty_struct *tty; +- int open_count; ++ atomic_t open_count; + spinlock_t serial_lock; + + int (*write_data) (struct hso_serial *serial); +@@ -1179,7 +1179,7 @@ static void put_rxbuf_data_and_resubmit_ + struct urb *urb; + + urb = serial->rx_urb[0]; +- if (serial->open_count > 0) { ++ if (atomic_read(&serial->open_count) > 0) { + count = put_rxbuf_data(urb, serial); + if (count == -1) + return; +@@ -1215,7 +1215,7 @@ static void hso_std_serial_read_bulk_cal + DUMP1(urb->transfer_buffer, urb->actual_length); + + /* Anyone listening? */ +- if (serial->open_count == 0) ++ if (atomic_read(&serial->open_count) == 0) + return; + + if (status == 0) { +@@ -1310,8 +1310,7 @@ static int hso_serial_open(struct tty_st + spin_unlock_irq(&serial->serial_lock); + + /* check for port already opened, if not set the termios */ +- serial->open_count++; +- if (serial->open_count == 1) { ++ if (atomic_inc_return(&serial->open_count) == 1) { + tty->low_latency = 1; + serial->rx_state = RX_IDLE; + /* Force default termio settings */ +@@ -1324,7 +1323,7 @@ static int hso_serial_open(struct tty_st + result = hso_start_serial_device(serial->parent, GFP_KERNEL); + if (result) { + hso_stop_serial_device(serial->parent); +- serial->open_count--; ++ atomic_dec(&serial->open_count); + kref_put(&serial->parent->ref, hso_serial_ref_free); + } + } else { +@@ -1361,10 +1360,10 @@ static void hso_serial_close(struct tty_ + + /* reset the rts and dtr */ + /* do the actual close */ +- serial->open_count--; ++ atomic_dec(&serial->open_count); + kref_put(&serial->parent->ref, hso_serial_ref_free); +- if (serial->open_count <= 0) { +- serial->open_count = 0; ++ if (atomic_read(&serial->open_count) <= 0) { ++ atomic_set(&serial->open_count, 0); + spin_lock_irq(&serial->serial_lock); + if (serial->tty == tty) { + serial->tty->driver_data = NULL; +@@ -1444,7 +1443,7 @@ static void hso_serial_set_termios(struc + + /* the actual setup */ + spin_lock_irqsave(&serial->serial_lock, flags); +- if (serial->open_count) ++ if (atomic_read(&serial->open_count)) + _hso_serial_set_termios(tty, old); + else + tty->termios = old; +@@ -3087,7 +3086,7 @@ static int hso_resume(struct usb_interfa + /* Start all serial ports */ + for (i = 0; i < HSO_SERIAL_TTY_MINORS; i++) { + if (serial_table[i] && (serial_table[i]->interface == iface)) { +- if (dev2ser(serial_table[i])->open_count) { ++ if (atomic_read(&dev2ser(serial_table[i])->open_count)) { + result = + hso_start_serial_device(serial_table[i], GFP_NOIO); + hso_kick_transmit(dev2ser(serial_table[i])); +diff -urNp linux-2.6.31.1/drivers/oprofile/buffer_sync.c linux-2.6.31.1/drivers/oprofile/buffer_sync.c +--- linux-2.6.31.1/drivers/oprofile/buffer_sync.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/oprofile/buffer_sync.c 2009-10-01 20:12:43.000000000 -0400 +@@ -341,7 +341,7 @@ static void add_data(struct op_entry *en + if (cookie == NO_COOKIE) + offset = pc; + if (cookie == INVALID_COOKIE) { +- atomic_inc(&oprofile_stats.sample_lost_no_mapping); ++ atomic_inc_unchecked(&oprofile_stats.sample_lost_no_mapping); + offset = pc; + } + if (cookie != last_cookie) { +@@ -385,14 +385,14 @@ add_sample(struct mm_struct *mm, struct + /* add userspace sample */ + + if (!mm) { +- atomic_inc(&oprofile_stats.sample_lost_no_mm); ++ atomic_inc_unchecked(&oprofile_stats.sample_lost_no_mm); + return 0; + } + + cookie = lookup_dcookie(mm, s->eip, &offset); + + if (cookie == INVALID_COOKIE) { +- atomic_inc(&oprofile_stats.sample_lost_no_mapping); ++ atomic_inc_unchecked(&oprofile_stats.sample_lost_no_mapping); + return 0; + } + +@@ -561,7 +561,7 @@ void sync_buffer(int cpu) + /* ignore backtraces if failed to add a sample */ + if (state == sb_bt_start) { + state = sb_bt_ignore; +- atomic_inc(&oprofile_stats.bt_lost_no_mapping); ++ atomic_inc_unchecked(&oprofile_stats.bt_lost_no_mapping); + } + } + release_mm(mm); +diff -urNp linux-2.6.31.1/drivers/oprofile/event_buffer.c linux-2.6.31.1/drivers/oprofile/event_buffer.c +--- linux-2.6.31.1/drivers/oprofile/event_buffer.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/oprofile/event_buffer.c 2009-10-01 20:12:43.000000000 -0400 +@@ -42,7 +42,7 @@ static atomic_t buffer_ready = ATOMIC_IN + void add_event_entry(unsigned long value) + { + if (buffer_pos == buffer_size) { +- atomic_inc(&oprofile_stats.event_lost_overflow); ++ atomic_inc_unchecked(&oprofile_stats.event_lost_overflow); + return; + } + +diff -urNp linux-2.6.31.1/drivers/oprofile/oprofilefs.c linux-2.6.31.1/drivers/oprofile/oprofilefs.c +--- linux-2.6.31.1/drivers/oprofile/oprofilefs.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/oprofile/oprofilefs.c 2009-10-01 20:12:43.000000000 -0400 +@@ -35,7 +35,7 @@ static struct inode *oprofilefs_get_inod + } + + +-static struct super_operations s_ops = { ++static const struct super_operations s_ops = { + .statfs = simple_statfs, + .drop_inode = generic_delete_inode, + }; +@@ -187,7 +187,7 @@ static const struct file_operations atom + + + int oprofilefs_create_ro_atomic(struct super_block *sb, struct dentry *root, +- char const *name, atomic_t *val) ++ char const *name, atomic_unchecked_t *val) + { + struct dentry *d = __oprofilefs_create_file(sb, root, name, + &atomic_ro_fops, 0444); +diff -urNp linux-2.6.31.1/drivers/oprofile/oprofile_stats.c linux-2.6.31.1/drivers/oprofile/oprofile_stats.c +--- linux-2.6.31.1/drivers/oprofile/oprofile_stats.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/oprofile/oprofile_stats.c 2009-10-01 20:12:43.000000000 -0400 +@@ -30,10 +30,10 @@ void oprofile_reset_stats(void) + cpu_buf->sample_invalid_eip = 0; + } + +- atomic_set(&oprofile_stats.sample_lost_no_mm, 0); +- atomic_set(&oprofile_stats.sample_lost_no_mapping, 0); +- atomic_set(&oprofile_stats.event_lost_overflow, 0); +- atomic_set(&oprofile_stats.bt_lost_no_mapping, 0); ++ atomic_set_unchecked(&oprofile_stats.sample_lost_no_mm, 0); ++ atomic_set_unchecked(&oprofile_stats.sample_lost_no_mapping, 0); ++ atomic_set_unchecked(&oprofile_stats.event_lost_overflow, 0); ++ atomic_set_unchecked(&oprofile_stats.bt_lost_no_mapping, 0); + } + + +diff -urNp linux-2.6.31.1/drivers/oprofile/oprofile_stats.h linux-2.6.31.1/drivers/oprofile/oprofile_stats.h +--- linux-2.6.31.1/drivers/oprofile/oprofile_stats.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/oprofile/oprofile_stats.h 2009-10-01 20:12:43.000000000 -0400 +@@ -13,10 +13,10 @@ + #include + + struct oprofile_stat_struct { +- atomic_t sample_lost_no_mm; +- atomic_t sample_lost_no_mapping; +- atomic_t bt_lost_no_mapping; +- atomic_t event_lost_overflow; ++ atomic_unchecked_t sample_lost_no_mm; ++ atomic_unchecked_t sample_lost_no_mapping; ++ atomic_unchecked_t bt_lost_no_mapping; ++ atomic_unchecked_t event_lost_overflow; + }; + + extern struct oprofile_stat_struct oprofile_stats; +diff -urNp linux-2.6.31.1/drivers/pci/hotplug/cpqphp_nvram.c linux-2.6.31.1/drivers/pci/hotplug/cpqphp_nvram.c +--- linux-2.6.31.1/drivers/pci/hotplug/cpqphp_nvram.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/pci/hotplug/cpqphp_nvram.c 2009-10-01 20:12:43.000000000 -0400 +@@ -428,9 +428,13 @@ static u32 store_HRT (void __iomem *rom_ + + void compaq_nvram_init (void __iomem *rom_start) + { ++ ++#ifndef CONFIG_PAX_KERNEXEC + if (rom_start) { + compaq_int15_entry_point = (rom_start + ROM_INT15_PHY_ADDR - ROM_PHY_ADDR); + } ++#endif ++ + dbg("int15 entry = %p\n", compaq_int15_entry_point); + + /* initialize our int15 lock */ +diff -urNp linux-2.6.31.1/drivers/pci/pcie/portdrv_pci.c linux-2.6.31.1/drivers/pci/pcie/portdrv_pci.c +--- linux-2.6.31.1/drivers/pci/pcie/portdrv_pci.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/pci/pcie/portdrv_pci.c 2009-10-01 20:12:43.000000000 -0400 +@@ -249,7 +249,7 @@ static void pcie_portdrv_err_resume(stru + static const struct pci_device_id port_pci_ids[] = { { + /* handle any PCI-Express port */ + PCI_DEVICE_CLASS(((PCI_CLASS_BRIDGE_PCI << 8) | 0x00), ~0), +- }, { /* end: all zeroes */ } ++ }, { 0, 0, 0, 0, 0, 0, 0 } + }; + MODULE_DEVICE_TABLE(pci, port_pci_ids); + +diff -urNp linux-2.6.31.1/drivers/pci/proc.c linux-2.6.31.1/drivers/pci/proc.c +--- linux-2.6.31.1/drivers/pci/proc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/pci/proc.c 2009-10-01 20:12:43.000000000 -0400 +@@ -480,7 +480,16 @@ static const struct file_operations proc + static int __init pci_proc_init(void) + { + struct pci_dev *dev = NULL; ++ ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ proc_bus_pci_dir = proc_mkdir_mode("bus/pci", S_IRUSR | S_IXUSR, NULL); ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ proc_bus_pci_dir = proc_mkdir_mode("bus/pci", S_IRUSR | S_IXUSR | S_IRGRP | S_IXGRP, NULL); ++#endif ++#else + proc_bus_pci_dir = proc_mkdir("bus/pci", NULL); ++#endif + proc_create("devices", 0, proc_bus_pci_dir, + &proc_bus_pci_dev_operations); + proc_initialized = 1; +diff -urNp linux-2.6.31.1/drivers/pcmcia/ti113x.h linux-2.6.31.1/drivers/pcmcia/ti113x.h +--- linux-2.6.31.1/drivers/pcmcia/ti113x.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/pcmcia/ti113x.h 2009-10-01 20:12:43.000000000 -0400 +@@ -903,7 +903,7 @@ static struct pci_device_id ene_tune_tbl + DEVID(PCI_VENDOR_ID_MOTOROLA, 0x3410, 0xECC0, PCI_ANY_ID, + ENE_TEST_C9_TLTENABLE | ENE_TEST_C9_PFENABLE, ENE_TEST_C9_TLTENABLE), + +- {} ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + static void ene_tune_bridge(struct pcmcia_socket *sock, struct pci_bus *bus) +diff -urNp linux-2.6.31.1/drivers/pcmcia/yenta_socket.c linux-2.6.31.1/drivers/pcmcia/yenta_socket.c +--- linux-2.6.31.1/drivers/pcmcia/yenta_socket.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/pcmcia/yenta_socket.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1366,7 +1366,7 @@ static struct pci_device_id yenta_table + + /* match any cardbus bridge */ + CB_ID(PCI_ANY_ID, PCI_ANY_ID, DEFAULT), +- { /* all zeroes */ } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + MODULE_DEVICE_TABLE(pci, yenta_table); + +diff -urNp linux-2.6.31.1/drivers/pnp/pnpbios/bioscalls.c linux-2.6.31.1/drivers/pnp/pnpbios/bioscalls.c +--- linux-2.6.31.1/drivers/pnp/pnpbios/bioscalls.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/pnp/pnpbios/bioscalls.c 2009-10-01 20:12:43.000000000 -0400 +@@ -60,7 +60,7 @@ set_base(gdt[(selname) >> 3], (u32)(addr + set_limit(gdt[(selname) >> 3], size); \ + } while(0) + +-static struct desc_struct bad_bios_desc; ++static struct desc_struct bad_bios_desc __read_only; + + /* + * At some point we want to use this stack frame pointer to unwind +@@ -87,6 +87,10 @@ static inline u16 call_pnp_bios(u16 func + struct desc_struct save_desc_40; + int cpu; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + /* + * PnP BIOSes are generally not terribly re-entrant. + * Also, don't rely on them to save everything correctly. +@@ -96,8 +100,17 @@ static inline u16 call_pnp_bios(u16 func + + cpu = get_cpu(); + save_desc_40 = get_cpu_gdt_table(cpu)[0x40 / 8]; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + get_cpu_gdt_table(cpu)[0x40 / 8] = bad_bios_desc; + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + /* On some boxes IRQ's during PnP BIOS calls are deadly. */ + spin_lock_irqsave(&pnp_bios_lock, flags); + +@@ -134,7 +147,16 @@ static inline u16 call_pnp_bios(u16 func + :"memory"); + spin_unlock_irqrestore(&pnp_bios_lock, flags); + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + get_cpu_gdt_table(cpu)[0x40 / 8] = save_desc_40; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + put_cpu(); + + /* If we get here and this is set then the PnP BIOS faulted on us. */ +@@ -468,16 +490,24 @@ int pnp_bios_read_escd(char *data, u32 n + return status; + } + +-void pnpbios_calls_init(union pnp_bios_install_struct *header) ++void __init pnpbios_calls_init(union pnp_bios_install_struct *header) + { + int i; + ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long cr0; ++#endif ++ + spin_lock_init(&pnp_bios_lock); + pnp_bios_callpoint.offset = header->fields.pm16offset; + pnp_bios_callpoint.segment = PNP_CS16; + ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_open_kernel(cr0); ++#endif ++ + bad_bios_desc.a = 0; +- bad_bios_desc.b = 0x00409200; ++ bad_bios_desc.b = 0x00409300; + + set_base(bad_bios_desc, __va((unsigned long)0x40 << 4)); + _set_limit((char *)&bad_bios_desc, 4095 - (0x40 << 4)); +@@ -491,4 +521,9 @@ void pnpbios_calls_init(union pnp_bios_i + set_base(gdt[GDT_ENTRY_PNPBIOS_DS], + __va(header->fields.pm16dseg)); + } ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ pax_close_kernel(cr0); ++#endif ++ + } +diff -urNp linux-2.6.31.1/drivers/pnp/quirks.c linux-2.6.31.1/drivers/pnp/quirks.c +--- linux-2.6.31.1/drivers/pnp/quirks.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/pnp/quirks.c 2009-10-01 20:12:43.000000000 -0400 +@@ -327,7 +327,7 @@ static struct pnp_fixup pnp_fixups[] = { + /* PnP resources that might overlap PCI BARs */ + {"PNP0c01", quirk_system_pci_resources}, + {"PNP0c02", quirk_system_pci_resources}, +- {""} ++ {"", NULL} + }; + + void pnp_fixup_device(struct pnp_dev *dev) +diff -urNp linux-2.6.31.1/drivers/pnp/resource.c linux-2.6.31.1/drivers/pnp/resource.c +--- linux-2.6.31.1/drivers/pnp/resource.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/pnp/resource.c 2009-10-01 20:12:43.000000000 -0400 +@@ -355,7 +355,7 @@ int pnp_check_irq(struct pnp_dev *dev, s + return 1; + + /* check if the resource is valid */ +- if (*irq < 0 || *irq > 15) ++ if (*irq > 15) + return 0; + + /* check if the resource is reserved */ +@@ -419,7 +419,7 @@ int pnp_check_dma(struct pnp_dev *dev, s + return 1; + + /* check if the resource is valid */ +- if (*dma < 0 || *dma == 4 || *dma > 7) ++ if (*dma == 4 || *dma > 7) + return 0; + + /* check if the resource is reserved */ +diff -urNp linux-2.6.31.1/drivers/s390/cio/qdio_debug.c linux-2.6.31.1/drivers/s390/cio/qdio_debug.c +--- linux-2.6.31.1/drivers/s390/cio/qdio_debug.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/s390/cio/qdio_debug.c 2009-10-01 20:12:43.000000000 -0400 +@@ -144,7 +144,7 @@ static void remove_debugfs_entry(struct + } + } + +-static struct file_operations debugfs_fops = { ++static const struct file_operations debugfs_fops = { + .owner = THIS_MODULE, + .open = qstat_seq_open, + .read = seq_read, +diff -urNp linux-2.6.31.1/drivers/s390/cio/qdio_perf.c linux-2.6.31.1/drivers/s390/cio/qdio_perf.c +--- linux-2.6.31.1/drivers/s390/cio/qdio_perf.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/s390/cio/qdio_perf.c 2009-10-01 20:12:43.000000000 -0400 +@@ -84,7 +84,7 @@ static int qdio_perf_seq_open(struct ino + return single_open(filp, qdio_perf_proc_show, NULL); + } + +-static struct file_operations qdio_perf_proc_fops = { ++static const struct file_operations qdio_perf_proc_fops = { + .owner = THIS_MODULE, + .open = qdio_perf_seq_open, + .read = seq_read, +diff -urNp linux-2.6.31.1/drivers/scsi/libfc/fc_exch.c linux-2.6.31.1/drivers/scsi/libfc/fc_exch.c +--- linux-2.6.31.1/drivers/scsi/libfc/fc_exch.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/scsi/libfc/fc_exch.c 2009-10-01 20:12:43.000000000 -0400 +@@ -73,12 +73,12 @@ struct fc_exch_mgr { + * all together if not used XXX + */ + struct { +- atomic_t no_free_exch; +- atomic_t no_free_exch_xid; +- atomic_t xid_not_found; +- atomic_t xid_busy; +- atomic_t seq_not_found; +- atomic_t non_bls_resp; ++ atomic_unchecked_t no_free_exch; ++ atomic_unchecked_t no_free_exch_xid; ++ atomic_unchecked_t xid_not_found; ++ atomic_unchecked_t xid_busy; ++ atomic_unchecked_t seq_not_found; ++ atomic_unchecked_t non_bls_resp; + } stats; + struct fc_exch **exches; /* for exch pointers indexed by xid */ + }; +@@ -523,7 +523,7 @@ struct fc_exch *fc_exch_alloc(struct fc_ + /* allocate memory for exchange */ + ep = mempool_alloc(mp->ep_pool, GFP_ATOMIC); + if (!ep) { +- atomic_inc(&mp->stats.no_free_exch); ++ atomic_inc_unchecked(&mp->stats.no_free_exch); + goto out; + } + memset(ep, 0, sizeof(*ep)); +@@ -568,7 +568,7 @@ out: + return ep; + err: + spin_unlock_bh(&mp->em_lock); +- atomic_inc(&mp->stats.no_free_exch_xid); ++ atomic_inc_unchecked(&mp->stats.no_free_exch_xid); + mempool_free(ep, mp->ep_pool); + return NULL; + } +@@ -671,7 +671,7 @@ static enum fc_pf_rjt_reason fc_seq_look + xid = ntohs(fh->fh_ox_id); /* we originated exch */ + ep = fc_exch_find(mp, xid); + if (!ep) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + reject = FC_RJT_OX_ID; + goto out; + } +@@ -701,7 +701,7 @@ static enum fc_pf_rjt_reason fc_seq_look + ep = fc_exch_find(mp, xid); + if ((f_ctl & FC_FC_FIRST_SEQ) && fc_sof_is_init(fr_sof(fp))) { + if (ep) { +- atomic_inc(&mp->stats.xid_busy); ++ atomic_inc_unchecked(&mp->stats.xid_busy); + reject = FC_RJT_RX_ID; + goto rel; + } +@@ -712,7 +712,7 @@ static enum fc_pf_rjt_reason fc_seq_look + } + xid = ep->xid; /* get our XID */ + } else if (!ep) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + reject = FC_RJT_RX_ID; /* XID not found */ + goto out; + } +@@ -733,7 +733,7 @@ static enum fc_pf_rjt_reason fc_seq_look + } else { + sp = &ep->seq; + if (sp->id != fh->fh_seq_id) { +- atomic_inc(&mp->stats.seq_not_found); ++ atomic_inc_unchecked(&mp->stats.seq_not_found); + reject = FC_RJT_SEQ_ID; /* sequence/exch should exist */ + goto rel; + } +@@ -1145,22 +1145,22 @@ static void fc_exch_recv_seq_resp(struct + + ep = fc_exch_find(mp, ntohs(fh->fh_ox_id)); + if (!ep) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + goto out; + } + if (ep->esb_stat & ESB_ST_COMPLETE) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + goto out; + } + if (ep->rxid == FC_XID_UNKNOWN) + ep->rxid = ntohs(fh->fh_rx_id); + if (ep->sid != 0 && ep->sid != ntoh24(fh->fh_d_id)) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + goto rel; + } + if (ep->did != ntoh24(fh->fh_s_id) && + ep->did != FC_FID_FLOGI) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + goto rel; + } + sof = fr_sof(fp); +@@ -1171,7 +1171,7 @@ static void fc_exch_recv_seq_resp(struct + } else { + sp = &ep->seq; + if (sp->id != fh->fh_seq_id) { +- atomic_inc(&mp->stats.seq_not_found); ++ atomic_inc_unchecked(&mp->stats.seq_not_found); + goto rel; + } + } +@@ -1230,10 +1230,10 @@ static void fc_exch_recv_resp(struct fc_ + + sp = fc_seq_lookup_orig(mp, fp); /* doesn't hold sequence */ + if (!sp) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + FC_EM_DBG(mp, "seq lookup failed\n"); + } else { +- atomic_inc(&mp->stats.non_bls_resp); ++ atomic_inc_unchecked(&mp->stats.non_bls_resp); + FC_EM_DBG(mp, "non-BLS response to sequence"); + } + fc_frame_free(fp); +diff -urNp linux-2.6.31.1/drivers/scsi/scsi_logging.h linux-2.6.31.1/drivers/scsi/scsi_logging.h +--- linux-2.6.31.1/drivers/scsi/scsi_logging.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/scsi/scsi_logging.h 2009-10-01 20:12:43.000000000 -0400 +@@ -51,7 +51,7 @@ do { \ + } while (0); \ + } while (0) + #else +-#define SCSI_CHECK_LOGGING(SHIFT, BITS, LEVEL, CMD) ++#define SCSI_CHECK_LOGGING(SHIFT, BITS, LEVEL, CMD) do {} while (0) + #endif /* CONFIG_SCSI_LOGGING */ + + /* +diff -urNp linux-2.6.31.1/drivers/scsi/sg.c linux-2.6.31.1/drivers/scsi/sg.c +--- linux-2.6.31.1/drivers/scsi/sg.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/scsi/sg.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1185,7 +1185,7 @@ sg_vma_fault(struct vm_area_struct *vma, + return VM_FAULT_SIGBUS; + } + +-static struct vm_operations_struct sg_mmap_vm_ops = { ++static const struct vm_operations_struct sg_mmap_vm_ops = { + .fault = sg_vma_fault, + }; + +@@ -1317,7 +1317,7 @@ static void sg_rq_end_io(struct request + } + } + +-static struct file_operations sg_fops = { ++static const struct file_operations sg_fops = { + .owner = THIS_MODULE, + .read = sg_read, + .write = sg_write, +@@ -2194,8 +2194,11 @@ static int sg_proc_seq_show_int(struct s + static int sg_proc_single_open_adio(struct inode *inode, struct file *file); + static ssize_t sg_proc_write_adio(struct file *filp, const char __user *buffer, + size_t count, loff_t *off); +-static struct file_operations adio_fops = { +- /* .owner, .read and .llseek added in sg_proc_init() */ ++ ++static const struct file_operations adio_fops = { ++ .owner = THIS_MODULE, ++ .read = seq_read, ++ .llseek = seq_lseek, + .open = sg_proc_single_open_adio, + .write = sg_proc_write_adio, + .release = single_release, +@@ -2204,7 +2207,10 @@ static struct file_operations adio_fops + static int sg_proc_single_open_dressz(struct inode *inode, struct file *file); + static ssize_t sg_proc_write_dressz(struct file *filp, + const char __user *buffer, size_t count, loff_t *off); +-static struct file_operations dressz_fops = { ++static const struct file_operations dressz_fops = { ++ .owner = THIS_MODULE, ++ .read = seq_read, ++ .llseek = seq_lseek, + .open = sg_proc_single_open_dressz, + .write = sg_proc_write_dressz, + .release = single_release, +@@ -2212,14 +2218,20 @@ static struct file_operations dressz_fop + + static int sg_proc_seq_show_version(struct seq_file *s, void *v); + static int sg_proc_single_open_version(struct inode *inode, struct file *file); +-static struct file_operations version_fops = { ++static const struct file_operations version_fops = { ++ .owner = THIS_MODULE, ++ .read = seq_read, ++ .llseek = seq_lseek, + .open = sg_proc_single_open_version, + .release = single_release, + }; + + static int sg_proc_seq_show_devhdr(struct seq_file *s, void *v); + static int sg_proc_single_open_devhdr(struct inode *inode, struct file *file); +-static struct file_operations devhdr_fops = { ++static const struct file_operations devhdr_fops = { ++ .owner = THIS_MODULE, ++ .read = seq_read, ++ .llseek = seq_lseek, + .open = sg_proc_single_open_devhdr, + .release = single_release, + }; +@@ -2229,11 +2241,14 @@ static int sg_proc_open_dev(struct inode + static void * dev_seq_start(struct seq_file *s, loff_t *pos); + static void * dev_seq_next(struct seq_file *s, void *v, loff_t *pos); + static void dev_seq_stop(struct seq_file *s, void *v); +-static struct file_operations dev_fops = { ++static const struct file_operations dev_fops = { ++ .owner = THIS_MODULE, ++ .read = seq_read, ++ .llseek = seq_lseek, + .open = sg_proc_open_dev, + .release = seq_release, + }; +-static struct seq_operations dev_seq_ops = { ++static const struct seq_operations dev_seq_ops = { + .start = dev_seq_start, + .next = dev_seq_next, + .stop = dev_seq_stop, +@@ -2242,11 +2257,14 @@ static struct seq_operations dev_seq_ops + + static int sg_proc_seq_show_devstrs(struct seq_file *s, void *v); + static int sg_proc_open_devstrs(struct inode *inode, struct file *file); +-static struct file_operations devstrs_fops = { ++static const struct file_operations devstrs_fops = { ++ .owner = THIS_MODULE, ++ .read = seq_read, ++ .llseek = seq_lseek, + .open = sg_proc_open_devstrs, + .release = seq_release, + }; +-static struct seq_operations devstrs_seq_ops = { ++static const struct seq_operations devstrs_seq_ops = { + .start = dev_seq_start, + .next = dev_seq_next, + .stop = dev_seq_stop, +@@ -2255,11 +2273,14 @@ static struct seq_operations devstrs_seq + + static int sg_proc_seq_show_debug(struct seq_file *s, void *v); + static int sg_proc_open_debug(struct inode *inode, struct file *file); +-static struct file_operations debug_fops = { ++static const struct file_operations debug_fops = { ++ .owner = THIS_MODULE, ++ .read = seq_read, ++ .llseek = seq_lseek, + .open = sg_proc_open_debug, + .release = seq_release, + }; +-static struct seq_operations debug_seq_ops = { ++static const struct seq_operations debug_seq_ops = { + .start = dev_seq_start, + .next = dev_seq_next, + .stop = dev_seq_stop, +@@ -2269,7 +2290,7 @@ static struct seq_operations debug_seq_o + + struct sg_proc_leaf { + const char * name; +- struct file_operations * fops; ++ const struct file_operations * fops; + }; + + static struct sg_proc_leaf sg_proc_leaf_arr[] = { +@@ -2295,9 +2316,6 @@ sg_proc_init(void) + for (k = 0; k < num_leaves; ++k) { + leaf = &sg_proc_leaf_arr[k]; + mask = leaf->fops->write ? S_IRUGO | S_IWUSR : S_IRUGO; +- leaf->fops->owner = THIS_MODULE; +- leaf->fops->read = seq_read; +- leaf->fops->llseek = seq_lseek; + proc_create(leaf->name, mask, sg_proc_sgp, leaf->fops); + } + return 0; +diff -urNp linux-2.6.31.1/drivers/serial/8250_pci.c linux-2.6.31.1/drivers/serial/8250_pci.c +--- linux-2.6.31.1/drivers/serial/8250_pci.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/serial/8250_pci.c 2009-10-01 20:12:43.000000000 -0400 +@@ -3580,7 +3580,7 @@ static struct pci_device_id serial_pci_t + PCI_ANY_ID, PCI_ANY_ID, + PCI_CLASS_COMMUNICATION_MULTISERIAL << 8, + 0xffff00, pbn_default }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + static struct pci_driver serial_pci_driver = { +diff -urNp linux-2.6.31.1/drivers/spi/spidev.c linux-2.6.31.1/drivers/spi/spidev.c +--- linux-2.6.31.1/drivers/spi/spidev.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/spi/spidev.c 2009-10-01 20:12:43.000000000 -0400 +@@ -537,7 +537,7 @@ static int spidev_release(struct inode * + return status; + } + +-static struct file_operations spidev_fops = { ++static const struct file_operations spidev_fops = { + .owner = THIS_MODULE, + /* REVISIT switch to aio primitives, so that userspace + * gets more complete API coverage. It'll simplify things +diff -urNp linux-2.6.31.1/drivers/staging/android/binder.c linux-2.6.31.1/drivers/staging/android/binder.c +--- linux-2.6.31.1/drivers/staging/android/binder.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/staging/android/binder.c 2009-10-01 20:12:43.000000000 -0400 +@@ -2717,7 +2717,7 @@ static void binder_vma_close(struct vm_a + binder_defer_work(proc, BINDER_DEFERRED_PUT_FILES); + } + +-static struct vm_operations_struct binder_vm_ops = { ++static const struct vm_operations_struct binder_vm_ops = { + .open = binder_vma_open, + .close = binder_vma_close, + }; +diff -urNp linux-2.6.31.1/drivers/staging/b3dfg/b3dfg.c linux-2.6.31.1/drivers/staging/b3dfg/b3dfg.c +--- linux-2.6.31.1/drivers/staging/b3dfg/b3dfg.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/staging/b3dfg/b3dfg.c 2009-10-01 20:12:43.000000000 -0400 +@@ -454,7 +454,7 @@ static int b3dfg_vma_fault(struct vm_are + return VM_FAULT_NOPAGE; + } + +-static struct vm_operations_struct b3dfg_vm_ops = { ++static const struct vm_operations_struct b3dfg_vm_ops = { + .fault = b3dfg_vma_fault, + }; + +@@ -854,7 +854,7 @@ static int b3dfg_mmap(struct file *filp, + return r; + } + +-static struct file_operations b3dfg_fops = { ++static const struct file_operations b3dfg_fops = { + .owner = THIS_MODULE, + .open = b3dfg_open, + .release = b3dfg_release, +diff -urNp linux-2.6.31.1/drivers/staging/comedi/comedi_fops.c linux-2.6.31.1/drivers/staging/comedi/comedi_fops.c +--- linux-2.6.31.1/drivers/staging/comedi/comedi_fops.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/staging/comedi/comedi_fops.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1370,7 +1370,7 @@ void comedi_unmap(struct vm_area_struct + mutex_unlock(&dev->mutex); + } + +-static struct vm_operations_struct comedi_vm_ops = { ++static const struct vm_operations_struct comedi_vm_ops = { + .close = comedi_unmap, + }; + +diff -urNp linux-2.6.31.1/drivers/staging/cpc-usb/cpc-usb_drv.c linux-2.6.31.1/drivers/staging/cpc-usb/cpc-usb_drv.c +--- linux-2.6.31.1/drivers/staging/cpc-usb/cpc-usb_drv.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/staging/cpc-usb/cpc-usb_drv.c 2009-10-01 20:12:43.000000000 -0400 +@@ -104,7 +104,7 @@ static void cpcusb_read_interrupt_callba + + static int cpcusb_setup_intrep(CPC_USB_T *card); + +-static struct file_operations cpcusb_fops = { ++static const struct file_operations cpcusb_fops = { + /* + * The owner field is part of the module-locking + * mechanism. The idea is that the kernel knows +diff -urNp linux-2.6.31.1/drivers/staging/epl/EplApiLinuxKernel.c linux-2.6.31.1/drivers/staging/epl/EplApiLinuxKernel.c +--- linux-2.6.31.1/drivers/staging/epl/EplApiLinuxKernel.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/staging/epl/EplApiLinuxKernel.c 2009-10-01 20:12:43.000000000 -0400 +@@ -203,7 +203,7 @@ static int EplLinIoctl(struct inode *pDe + module_init(EplLinInit); + module_exit(EplLinExit); + +-static struct file_operations EplLinFileOps_g = { ++static const struct file_operations EplLinFileOps_g = { + .owner = THIS_MODULE, + .open = EplLinOpen, + .release = EplLinRelease, +diff -urNp linux-2.6.31.1/drivers/staging/go7007/go7007-v4l2.c linux-2.6.31.1/drivers/staging/go7007/go7007-v4l2.c +--- linux-2.6.31.1/drivers/staging/go7007/go7007-v4l2.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/staging/go7007/go7007-v4l2.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1717,7 +1717,7 @@ static int go7007_vm_fault(struct vm_are + return 0; + } + +-static struct vm_operations_struct go7007_vm_ops = { ++static const struct vm_operations_struct go7007_vm_ops = { + .open = go7007_vm_open, + .close = go7007_vm_close, + .fault = go7007_vm_fault, +diff -urNp linux-2.6.31.1/drivers/staging/panel/panel.c linux-2.6.31.1/drivers/staging/panel/panel.c +--- linux-2.6.31.1/drivers/staging/panel/panel.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/staging/panel/panel.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1263,7 +1263,7 @@ static int lcd_release(struct inode *ino + return 0; + } + +-static struct file_operations lcd_fops = { ++static const struct file_operations lcd_fops = { + .write = lcd_write, + .open = lcd_open, + .release = lcd_release, +@@ -1519,7 +1519,7 @@ static int keypad_release(struct inode * + return 0; + } + +-static struct file_operations keypad_fops = { ++static const struct file_operations keypad_fops = { + .read = keypad_read, /* read */ + .open = keypad_open, /* open */ + .release = keypad_release, /* close */ +diff -urNp linux-2.6.31.1/drivers/staging/poch/poch.c linux-2.6.31.1/drivers/staging/poch/poch.c +--- linux-2.6.31.1/drivers/staging/poch/poch.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/staging/poch/poch.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1056,7 +1056,7 @@ static int poch_ioctl(struct inode *inod + return 0; + } + +-static struct file_operations poch_fops = { ++static const struct file_operations poch_fops = { + .owner = THIS_MODULE, + .open = poch_open, + .release = poch_release, +diff -urNp linux-2.6.31.1/drivers/staging/rtl8192su/ieee80211/proc.c linux-2.6.31.1/drivers/staging/rtl8192su/ieee80211/proc.c +--- linux-2.6.31.1/drivers/staging/rtl8192su/ieee80211/proc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/staging/rtl8192su/ieee80211/proc.c 2009-10-01 20:12:43.000000000 -0400 +@@ -87,7 +87,7 @@ static int c_show(struct seq_file *m, vo + return 0; + } + +-static struct seq_operations crypto_seq_ops = { ++static const struct seq_operations crypto_seq_ops = { + .start = c_start, + .next = c_next, + .stop = c_stop, +@@ -99,7 +99,7 @@ static int crypto_info_open(struct inode + return seq_open(file, &crypto_seq_ops); + } + +-static struct file_operations proc_crypto_ops = { ++static const struct file_operations proc_crypto_ops = { + .open = crypto_info_open, + .read = seq_read, + .llseek = seq_lseek, +diff -urNp linux-2.6.31.1/drivers/uio/uio.c linux-2.6.31.1/drivers/uio/uio.c +--- linux-2.6.31.1/drivers/uio/uio.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/uio/uio.c 2009-10-01 20:12:43.000000000 -0400 +@@ -658,7 +658,7 @@ static int uio_vma_fault(struct vm_area_ + return 0; + } + +-static struct vm_operations_struct uio_vm_ops = { ++static const struct vm_operations_struct uio_vm_ops = { + .open = uio_vma_open, + .close = uio_vma_close, + .fault = uio_vma_fault, +diff -urNp linux-2.6.31.1/drivers/usb/atm/usbatm.c linux-2.6.31.1/drivers/usb/atm/usbatm.c +--- linux-2.6.31.1/drivers/usb/atm/usbatm.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/atm/usbatm.c 2009-10-01 20:12:43.000000000 -0400 +@@ -333,7 +333,7 @@ static void usbatm_extract_one_cell(stru + if (printk_ratelimit()) + atm_warn(instance, "%s: OAM not supported (vpi %d, vci %d)!\n", + __func__, vpi, vci); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + return; + } + +@@ -361,7 +361,7 @@ static void usbatm_extract_one_cell(stru + if (length > ATM_MAX_AAL5_PDU) { + atm_rldbg(instance, "%s: bogus length %u (vcc: 0x%p)!\n", + __func__, length, vcc); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto out; + } + +@@ -370,14 +370,14 @@ static void usbatm_extract_one_cell(stru + if (sarb->len < pdu_length) { + atm_rldbg(instance, "%s: bogus pdu_length %u (sarb->len: %u, vcc: 0x%p)!\n", + __func__, pdu_length, sarb->len, vcc); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto out; + } + + if (crc32_be(~0, skb_tail_pointer(sarb) - pdu_length, pdu_length) != 0xc704dd7b) { + atm_rldbg(instance, "%s: packet failed crc check (vcc: 0x%p)!\n", + __func__, vcc); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto out; + } + +@@ -387,7 +387,7 @@ static void usbatm_extract_one_cell(stru + if (printk_ratelimit()) + atm_err(instance, "%s: no memory for skb (length: %u)!\n", + __func__, length); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + goto out; + } + +@@ -412,7 +412,7 @@ static void usbatm_extract_one_cell(stru + + vcc->push(vcc, skb); + +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + out: + skb_trim(sarb, 0); + } +@@ -616,7 +616,7 @@ static void usbatm_tx_process(unsigned l + struct atm_vcc *vcc = UDSL_SKB(skb)->atm.vcc; + + usbatm_pop(vcc, skb); +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + + skb = skb_dequeue(&instance->sndqueue); + } +@@ -775,11 +775,11 @@ static int usbatm_atm_proc_read(struct a + if (!left--) + return sprintf(page, + "AAL5: tx %d ( %d err ), rx %d ( %d err, %d drop )\n", +- atomic_read(&atm_dev->stats.aal5.tx), +- atomic_read(&atm_dev->stats.aal5.tx_err), +- atomic_read(&atm_dev->stats.aal5.rx), +- atomic_read(&atm_dev->stats.aal5.rx_err), +- atomic_read(&atm_dev->stats.aal5.rx_drop)); ++ atomic_read_unchecked(&atm_dev->stats.aal5.tx), ++ atomic_read_unchecked(&atm_dev->stats.aal5.tx_err), ++ atomic_read_unchecked(&atm_dev->stats.aal5.rx), ++ atomic_read_unchecked(&atm_dev->stats.aal5.rx_err), ++ atomic_read_unchecked(&atm_dev->stats.aal5.rx_drop)); + + if (!left--) { + if (instance->disconnected) +diff -urNp linux-2.6.31.1/drivers/usb/class/cdc-acm.c linux-2.6.31.1/drivers/usb/class/cdc-acm.c +--- linux-2.6.31.1/drivers/usb/class/cdc-acm.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/class/cdc-acm.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1529,7 +1529,7 @@ static struct usb_device_id acm_ids[] = + USB_CDC_ACM_PROTO_AT_CDMA) }, + + /* NOTE: COMM/ACM/0xff is likely MSFT RNDIS ... NOT a modem!! */ +- { } ++ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(usb, acm_ids); +diff -urNp linux-2.6.31.1/drivers/usb/class/usblp.c linux-2.6.31.1/drivers/usb/class/usblp.c +--- linux-2.6.31.1/drivers/usb/class/usblp.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/class/usblp.c 2009-10-01 20:12:43.000000000 -0400 +@@ -228,7 +228,7 @@ static const struct quirk_printer_struct + { 0x0482, 0x0010, USBLP_QUIRK_BIDIR }, /* Kyocera Mita FS 820, by zut */ + { 0x04f9, 0x000d, USBLP_QUIRK_BIDIR }, /* Brother Industries, Ltd HL-1440 Laser Printer */ + { 0x04b8, 0x0202, USBLP_QUIRK_BAD_CLASS }, /* Seiko Epson Receipt Printer M129C */ +- { 0, 0 } ++ { 0, 0, 0 } + }; + + static int usblp_wwait(struct usblp *usblp, int nonblock); +@@ -1412,7 +1412,7 @@ static struct usb_device_id usblp_ids [] + { USB_INTERFACE_INFO(7, 1, 2) }, + { USB_INTERFACE_INFO(7, 1, 3) }, + { USB_DEVICE(0x04b8, 0x0202) }, /* Seiko Epson Receipt Printer M129C */ +- { } /* Terminating entry */ ++ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } /* Terminating entry */ + }; + + MODULE_DEVICE_TABLE (usb, usblp_ids); +diff -urNp linux-2.6.31.1/drivers/usb/class/usbtmc.c linux-2.6.31.1/drivers/usb/class/usbtmc.c +--- linux-2.6.31.1/drivers/usb/class/usbtmc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/class/usbtmc.c 2009-10-01 20:12:43.000000000 -0400 +@@ -956,7 +956,7 @@ static long usbtmc_ioctl(struct file *fi + return retval; + } + +-static struct file_operations fops = { ++static const struct file_operations fops = { + .owner = THIS_MODULE, + .read = usbtmc_read, + .write = usbtmc_write, +diff -urNp linux-2.6.31.1/drivers/usb/core/hub.c linux-2.6.31.1/drivers/usb/core/hub.c +--- linux-2.6.31.1/drivers/usb/core/hub.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/core/hub.c 2009-10-01 20:12:43.000000000 -0400 +@@ -3284,7 +3284,7 @@ static struct usb_device_id hub_id_table + .bDeviceClass = USB_CLASS_HUB}, + { .match_flags = USB_DEVICE_ID_MATCH_INT_CLASS, + .bInterfaceClass = USB_CLASS_HUB}, +- { } /* Terminating entry */ ++ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } /* Terminating entry */ + }; + + MODULE_DEVICE_TABLE (usb, hub_id_table); +diff -urNp linux-2.6.31.1/drivers/usb/core/inode.c linux-2.6.31.1/drivers/usb/core/inode.c +--- linux-2.6.31.1/drivers/usb/core/inode.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/core/inode.c 2009-10-01 20:12:43.000000000 -0400 +@@ -48,7 +48,7 @@ + #define USBFS_DEFAULT_BUSMODE (S_IXUGO | S_IRUGO) + #define USBFS_DEFAULT_LISTMODE S_IRUGO + +-static struct super_operations usbfs_ops; ++static const struct super_operations usbfs_ops; + static const struct file_operations default_file_operations; + static struct vfsmount *usbfs_mount; + static int usbfs_mount_count; /* = 0 */ +@@ -449,7 +449,7 @@ static const struct file_operations defa + .llseek = default_file_lseek, + }; + +-static struct super_operations usbfs_ops = { ++static const struct super_operations usbfs_ops = { + .statfs = simple_statfs, + .drop_inode = generic_delete_inode, + .remount_fs = remount, +diff -urNp linux-2.6.31.1/drivers/usb/core/message.c linux-2.6.31.1/drivers/usb/core/message.c +--- linux-2.6.31.1/drivers/usb/core/message.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/core/message.c 2009-10-01 20:12:43.000000000 -0400 +@@ -926,8 +926,8 @@ char *usb_cache_string(struct usb_device + buf = kmalloc(MAX_USB_STRING_SIZE, GFP_KERNEL); + if (buf) { + len = usb_string(udev, index, buf, MAX_USB_STRING_SIZE); +- if (len > 0) { +- smallbuf = kmalloc(++len, GFP_KERNEL); ++ if (len++ > 0) { ++ smallbuf = kmalloc(len, GFP_KERNEL); + if (!smallbuf) + return buf; + memcpy(smallbuf, buf, len); +diff -urNp linux-2.6.31.1/drivers/usb/gadget/inode.c linux-2.6.31.1/drivers/usb/gadget/inode.c +--- linux-2.6.31.1/drivers/usb/gadget/inode.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/gadget/inode.c 2009-10-01 20:12:43.000000000 -0400 +@@ -2033,7 +2033,7 @@ gadgetfs_create_file (struct super_block + return inode; + } + +-static struct super_operations gadget_fs_operations = { ++static const struct super_operations gadget_fs_operations = { + .statfs = simple_statfs, + .drop_inode = generic_delete_inode, + }; +diff -urNp linux-2.6.31.1/drivers/usb/gadget/printer.c linux-2.6.31.1/drivers/usb/gadget/printer.c +--- linux-2.6.31.1/drivers/usb/gadget/printer.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/gadget/printer.c 2009-10-01 20:12:43.000000000 -0400 +@@ -875,7 +875,7 @@ printer_ioctl(struct file *fd, unsigned + } + + /* used after endpoint configuration */ +-static struct file_operations printer_io_operations = { ++static const struct file_operations printer_io_operations = { + .owner = THIS_MODULE, + .open = printer_open, + .read = printer_read, +diff -urNp linux-2.6.31.1/drivers/usb/host/ehci-pci.c linux-2.6.31.1/drivers/usb/host/ehci-pci.c +--- linux-2.6.31.1/drivers/usb/host/ehci-pci.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/host/ehci-pci.c 2009-10-01 20:12:43.000000000 -0400 +@@ -416,7 +416,7 @@ static const struct pci_device_id pci_id + PCI_DEVICE_CLASS(PCI_CLASS_SERIAL_USB_EHCI, ~0), + .driver_data = (unsigned long) &ehci_pci_hc_driver, + }, +- { /* end: all zeroes */ } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + MODULE_DEVICE_TABLE(pci, pci_ids); + +diff -urNp linux-2.6.31.1/drivers/usb/host/uhci-hcd.c linux-2.6.31.1/drivers/usb/host/uhci-hcd.c +--- linux-2.6.31.1/drivers/usb/host/uhci-hcd.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/host/uhci-hcd.c 2009-10-01 20:12:43.000000000 -0400 +@@ -927,7 +927,7 @@ static const struct pci_device_id uhci_p + /* handle any USB UHCI controller */ + PCI_DEVICE_CLASS(PCI_CLASS_SERIAL_USB_UHCI, ~0), + .driver_data = (unsigned long) &uhci_driver, +- }, { /* end: all zeroes */ } ++ }, { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, uhci_pci_ids); +diff -urNp linux-2.6.31.1/drivers/usb/host/whci/debug.c linux-2.6.31.1/drivers/usb/host/whci/debug.c +--- linux-2.6.31.1/drivers/usb/host/whci/debug.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/host/whci/debug.c 2009-10-01 20:12:43.000000000 -0400 +@@ -134,7 +134,7 @@ static int pzl_open(struct inode *inode, + return single_open(file, pzl_print, inode->i_private); + } + +-static struct file_operations di_fops = { ++static const struct file_operations di_fops = { + .open = di_open, + .read = seq_read, + .llseek = seq_lseek, +@@ -142,7 +142,7 @@ static struct file_operations di_fops = + .owner = THIS_MODULE, + }; + +-static struct file_operations asl_fops = { ++static const struct file_operations asl_fops = { + .open = asl_open, + .read = seq_read, + .llseek = seq_lseek, +@@ -150,7 +150,7 @@ static struct file_operations asl_fops = + .owner = THIS_MODULE, + }; + +-static struct file_operations pzl_fops = { ++static const struct file_operations pzl_fops = { + .open = pzl_open, + .read = seq_read, + .llseek = seq_lseek, +diff -urNp linux-2.6.31.1/drivers/usb/mon/mon_bin.c linux-2.6.31.1/drivers/usb/mon/mon_bin.c +--- linux-2.6.31.1/drivers/usb/mon/mon_bin.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/mon/mon_bin.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1184,7 +1184,7 @@ static int mon_bin_vma_fault(struct vm_a + return 0; + } + +-static struct vm_operations_struct mon_bin_vm_ops = { ++static const struct vm_operations_struct mon_bin_vm_ops = { + .open = mon_bin_vma_open, + .close = mon_bin_vma_close, + .fault = mon_bin_vma_fault, +diff -urNp linux-2.6.31.1/drivers/usb/storage/debug.h linux-2.6.31.1/drivers/usb/storage/debug.h +--- linux-2.6.31.1/drivers/usb/storage/debug.h 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/storage/debug.h 2009-10-01 20:12:43.000000000 -0400 +@@ -54,9 +54,9 @@ void usb_stor_show_sense( unsigned char + #define US_DEBUGPX(x...) printk( x ) + #define US_DEBUG(x) x + #else +-#define US_DEBUGP(x...) +-#define US_DEBUGPX(x...) +-#define US_DEBUG(x) ++#define US_DEBUGP(x...) do {} while (0) ++#define US_DEBUGPX(x...) do {} while (0) ++#define US_DEBUG(x) do {} while (0) + #endif + + #endif +diff -urNp linux-2.6.31.1/drivers/usb/storage/usb.c linux-2.6.31.1/drivers/usb/storage/usb.c +--- linux-2.6.31.1/drivers/usb/storage/usb.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/storage/usb.c 2009-10-01 20:12:43.000000000 -0400 +@@ -118,7 +118,7 @@ MODULE_PARM_DESC(quirks, "supplemental l + + static struct us_unusual_dev us_unusual_dev_list[] = { + # include "unusual_devs.h" +- { } /* Terminating entry */ ++ { NULL, NULL, 0, 0, NULL } /* Terminating entry */ + }; + + #undef UNUSUAL_DEV +diff -urNp linux-2.6.31.1/drivers/usb/storage/usual-tables.c linux-2.6.31.1/drivers/usb/storage/usual-tables.c +--- linux-2.6.31.1/drivers/usb/storage/usual-tables.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/usb/storage/usual-tables.c 2009-10-01 20:12:43.000000000 -0400 +@@ -48,7 +48,7 @@ + + struct usb_device_id usb_storage_usb_ids[] = { + # include "unusual_devs.h" +- { } /* Terminating entry */ ++ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } /* Terminating entry */ + }; + EXPORT_SYMBOL_GPL(usb_storage_usb_ids); + +diff -urNp linux-2.6.31.1/drivers/uwb/uwb-debug.c linux-2.6.31.1/drivers/uwb/uwb-debug.c +--- linux-2.6.31.1/drivers/uwb/uwb-debug.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/uwb/uwb-debug.c 2009-10-01 20:12:43.000000000 -0400 +@@ -205,7 +205,7 @@ static ssize_t command_write(struct file + return ret < 0 ? ret : len; + } + +-static struct file_operations command_fops = { ++static const struct file_operations command_fops = { + .open = command_open, + .write = command_write, + .read = NULL, +@@ -255,7 +255,7 @@ static int reservations_open(struct inod + return single_open(file, reservations_print, inode->i_private); + } + +-static struct file_operations reservations_fops = { ++static const struct file_operations reservations_fops = { + .open = reservations_open, + .read = seq_read, + .llseek = seq_lseek, +@@ -283,7 +283,7 @@ static int drp_avail_open(struct inode * + return single_open(file, drp_avail_print, inode->i_private); + } + +-static struct file_operations drp_avail_fops = { ++static const struct file_operations drp_avail_fops = { + .open = drp_avail_open, + .read = seq_read, + .llseek = seq_lseek, +diff -urNp linux-2.6.31.1/drivers/uwb/wlp/messages.c linux-2.6.31.1/drivers/uwb/wlp/messages.c +--- linux-2.6.31.1/drivers/uwb/wlp/messages.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/uwb/wlp/messages.c 2009-10-01 20:12:43.000000000 -0400 +@@ -903,7 +903,7 @@ int wlp_parse_f0(struct wlp *wlp, struct + size_t len = skb->len; + size_t used; + ssize_t result; +- struct wlp_nonce enonce, rnonce; ++ struct wlp_nonce enonce = {{0}}, rnonce = {{0}}; + enum wlp_assc_error assc_err; + char enonce_buf[WLP_WSS_NONCE_STRSIZE]; + char rnonce_buf[WLP_WSS_NONCE_STRSIZE]; +diff -urNp linux-2.6.31.1/drivers/video/fb_defio.c linux-2.6.31.1/drivers/video/fb_defio.c +--- linux-2.6.31.1/drivers/video/fb_defio.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/video/fb_defio.c 2009-10-01 20:12:43.000000000 -0400 +@@ -125,7 +125,7 @@ page_already_added: + return 0; + } + +-static struct vm_operations_struct fb_deferred_io_vm_ops = { ++static const struct vm_operations_struct fb_deferred_io_vm_ops = { + .fault = fb_deferred_io_fault, + .page_mkwrite = fb_deferred_io_mkwrite, + }; +diff -urNp linux-2.6.31.1/drivers/video/fbmem.c linux-2.6.31.1/drivers/video/fbmem.c +--- linux-2.6.31.1/drivers/video/fbmem.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/video/fbmem.c 2009-10-01 20:12:43.000000000 -0400 +@@ -403,7 +403,7 @@ static void fb_do_show_logo(struct fb_in + image->dx += image->width + 8; + } + } else if (rotate == FB_ROTATE_UD) { +- for (x = 0; x < num && image->dx >= 0; x++) { ++ for (x = 0; x < num && (__s32)image->dx >= 0; x++) { + info->fbops->fb_imageblit(info, image); + image->dx -= image->width + 8; + } +@@ -415,7 +415,7 @@ static void fb_do_show_logo(struct fb_in + image->dy += image->height + 8; + } + } else if (rotate == FB_ROTATE_CCW) { +- for (x = 0; x < num && image->dy >= 0; x++) { ++ for (x = 0; x < num && (__s32)image->dy >= 0; x++) { + info->fbops->fb_imageblit(info, image); + image->dy -= image->height + 8; + } +@@ -1108,7 +1108,7 @@ static long do_fb_ioctl(struct fb_info * + return -EFAULT; + if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) + return -EINVAL; +- if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX) ++ if (con2fb.framebuffer >= FB_MAX) + return -EINVAL; + if (!registered_fb[con2fb.framebuffer]) + request_module("fb%d", con2fb.framebuffer); +diff -urNp linux-2.6.31.1/drivers/video/fbmon.c linux-2.6.31.1/drivers/video/fbmon.c +--- linux-2.6.31.1/drivers/video/fbmon.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/video/fbmon.c 2009-10-01 20:12:43.000000000 -0400 +@@ -45,7 +45,7 @@ + #ifdef DEBUG + #define DPRINTK(fmt, args...) printk(fmt,## args) + #else +-#define DPRINTK(fmt, args...) ++#define DPRINTK(fmt, args...) do {} while (0) + #endif + + #define FBMON_FIX_HEADER 1 +diff -urNp linux-2.6.31.1/drivers/video/i810/i810_accel.c linux-2.6.31.1/drivers/video/i810/i810_accel.c +--- linux-2.6.31.1/drivers/video/i810/i810_accel.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/video/i810/i810_accel.c 2009-10-01 20:12:43.000000000 -0400 +@@ -73,6 +73,7 @@ static inline int wait_for_space(struct + } + } + printk("ringbuffer lockup!!!\n"); ++ printk("head:%u tail:%u iring.size:%u space:%u\n", head, tail, par->iring.size, space); + i810_report_error(mmio); + par->dev_flags |= LOCKUP; + info->pixmap.scan_align = 1; +diff -urNp linux-2.6.31.1/drivers/video/i810/i810_main.c linux-2.6.31.1/drivers/video/i810/i810_main.c +--- linux-2.6.31.1/drivers/video/i810/i810_main.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/video/i810/i810_main.c 2009-10-01 20:12:43.000000000 -0400 +@@ -120,7 +120,7 @@ static struct pci_device_id i810fb_pci_t + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 4 }, + { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82815_CGC, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 5 }, +- { 0 }, ++ { 0, 0, 0, 0, 0, 0, 0 }, + }; + + static struct pci_driver i810fb_driver = { +diff -urNp linux-2.6.31.1/drivers/video/modedb.c linux-2.6.31.1/drivers/video/modedb.c +--- linux-2.6.31.1/drivers/video/modedb.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/video/modedb.c 2009-10-01 20:12:43.000000000 -0400 +@@ -38,240 +38,240 @@ static const struct fb_videomode modedb[ + { + /* 640x400 @ 70 Hz, 31.5 kHz hsync */ + NULL, 70, 640, 400, 39721, 40, 24, 39, 9, 96, 2, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 640x480 @ 60 Hz, 31.5 kHz hsync */ + NULL, 60, 640, 480, 39721, 40, 24, 32, 11, 96, 2, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 800x600 @ 56 Hz, 35.15 kHz hsync */ + NULL, 56, 800, 600, 27777, 128, 24, 22, 1, 72, 2, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1024x768 @ 87 Hz interlaced, 35.5 kHz hsync */ + NULL, 87, 1024, 768, 22271, 56, 24, 33, 8, 160, 8, +- 0, FB_VMODE_INTERLACED ++ 0, FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 640x400 @ 85 Hz, 37.86 kHz hsync */ + NULL, 85, 640, 400, 31746, 96, 32, 41, 1, 64, 3, +- FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 640x480 @ 72 Hz, 36.5 kHz hsync */ + NULL, 72, 640, 480, 31746, 144, 40, 30, 8, 40, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 640x480 @ 75 Hz, 37.50 kHz hsync */ + NULL, 75, 640, 480, 31746, 120, 16, 16, 1, 64, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 800x600 @ 60 Hz, 37.8 kHz hsync */ + NULL, 60, 800, 600, 25000, 88, 40, 23, 1, 128, 4, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 640x480 @ 85 Hz, 43.27 kHz hsync */ + NULL, 85, 640, 480, 27777, 80, 56, 25, 1, 56, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1152x864 @ 89 Hz interlaced, 44 kHz hsync */ + NULL, 89, 1152, 864, 15384, 96, 16, 110, 1, 216, 10, +- 0, FB_VMODE_INTERLACED ++ 0, FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 800x600 @ 72 Hz, 48.0 kHz hsync */ + NULL, 72, 800, 600, 20000, 64, 56, 23, 37, 120, 6, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1024x768 @ 60 Hz, 48.4 kHz hsync */ + NULL, 60, 1024, 768, 15384, 168, 8, 29, 3, 144, 6, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 640x480 @ 100 Hz, 53.01 kHz hsync */ + NULL, 100, 640, 480, 21834, 96, 32, 36, 8, 96, 6, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1152x864 @ 60 Hz, 53.5 kHz hsync */ + NULL, 60, 1152, 864, 11123, 208, 64, 16, 4, 256, 8, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 800x600 @ 85 Hz, 55.84 kHz hsync */ + NULL, 85, 800, 600, 16460, 160, 64, 36, 16, 64, 5, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1024x768 @ 70 Hz, 56.5 kHz hsync */ + NULL, 70, 1024, 768, 13333, 144, 24, 29, 3, 136, 6, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1280x1024 @ 87 Hz interlaced, 51 kHz hsync */ + NULL, 87, 1280, 1024, 12500, 56, 16, 128, 1, 216, 12, +- 0, FB_VMODE_INTERLACED ++ 0, FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 800x600 @ 100 Hz, 64.02 kHz hsync */ + NULL, 100, 800, 600, 14357, 160, 64, 30, 4, 64, 6, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1024x768 @ 76 Hz, 62.5 kHz hsync */ + NULL, 76, 1024, 768, 11764, 208, 8, 36, 16, 120, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1152x864 @ 70 Hz, 62.4 kHz hsync */ + NULL, 70, 1152, 864, 10869, 106, 56, 20, 1, 160, 10, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1280x1024 @ 61 Hz, 64.2 kHz hsync */ + NULL, 61, 1280, 1024, 9090, 200, 48, 26, 1, 184, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1400x1050 @ 60Hz, 63.9 kHz hsync */ + NULL, 60, 1400, 1050, 9259, 136, 40, 13, 1, 112, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1400x1050 @ 75,107 Hz, 82,392 kHz +hsync +vsync*/ + NULL, 75, 1400, 1050, 7190, 120, 56, 23, 10, 112, 13, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1400x1050 @ 60 Hz, ? kHz +hsync +vsync*/ + NULL, 60, 1400, 1050, 9259, 128, 40, 12, 0, 112, 3, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1024x768 @ 85 Hz, 70.24 kHz hsync */ + NULL, 85, 1024, 768, 10111, 192, 32, 34, 14, 160, 6, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1152x864 @ 78 Hz, 70.8 kHz hsync */ + NULL, 78, 1152, 864, 9090, 228, 88, 32, 0, 84, 12, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1280x1024 @ 70 Hz, 74.59 kHz hsync */ + NULL, 70, 1280, 1024, 7905, 224, 32, 28, 8, 160, 8, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1600x1200 @ 60Hz, 75.00 kHz hsync */ + NULL, 60, 1600, 1200, 6172, 304, 64, 46, 1, 192, 3, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1152x864 @ 84 Hz, 76.0 kHz hsync */ + NULL, 84, 1152, 864, 7407, 184, 312, 32, 0, 128, 12, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1280x1024 @ 74 Hz, 78.85 kHz hsync */ + NULL, 74, 1280, 1024, 7407, 256, 32, 34, 3, 144, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1024x768 @ 100Hz, 80.21 kHz hsync */ + NULL, 100, 1024, 768, 8658, 192, 32, 21, 3, 192, 10, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1280x1024 @ 76 Hz, 81.13 kHz hsync */ + NULL, 76, 1280, 1024, 7407, 248, 32, 34, 3, 104, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1600x1200 @ 70 Hz, 87.50 kHz hsync */ + NULL, 70, 1600, 1200, 5291, 304, 64, 46, 1, 192, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1152x864 @ 100 Hz, 89.62 kHz hsync */ + NULL, 100, 1152, 864, 7264, 224, 32, 17, 2, 128, 19, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1280x1024 @ 85 Hz, 91.15 kHz hsync */ + NULL, 85, 1280, 1024, 6349, 224, 64, 44, 1, 160, 3, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1600x1200 @ 75 Hz, 93.75 kHz hsync */ + NULL, 75, 1600, 1200, 4938, 304, 64, 46, 1, 192, 3, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1680x1050 @ 60 Hz, 65.191 kHz hsync */ + NULL, 60, 1680, 1050, 6848, 280, 104, 30, 3, 176, 6, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1600x1200 @ 85 Hz, 105.77 kHz hsync */ + NULL, 85, 1600, 1200, 4545, 272, 16, 37, 4, 192, 3, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1280x1024 @ 100 Hz, 107.16 kHz hsync */ + NULL, 100, 1280, 1024, 5502, 256, 32, 26, 7, 128, 15, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1800x1440 @ 64Hz, 96.15 kHz hsync */ + NULL, 64, 1800, 1440, 4347, 304, 96, 46, 1, 192, 3, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1800x1440 @ 70Hz, 104.52 kHz hsync */ + NULL, 70, 1800, 1440, 4000, 304, 96, 46, 1, 192, 3, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 512x384 @ 78 Hz, 31.50 kHz hsync */ + NULL, 78, 512, 384, 49603, 48, 16, 16, 1, 64, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 512x384 @ 85 Hz, 34.38 kHz hsync */ + NULL, 85, 512, 384, 45454, 48, 16, 16, 1, 64, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 320x200 @ 70 Hz, 31.5 kHz hsync, 8:5 aspect ratio */ + NULL, 70, 320, 200, 79440, 16, 16, 20, 4, 48, 1, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 320x240 @ 60 Hz, 31.5 kHz hsync, 4:3 aspect ratio */ + NULL, 60, 320, 240, 79440, 16, 16, 16, 5, 48, 1, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 320x240 @ 72 Hz, 36.5 kHz hsync */ + NULL, 72, 320, 240, 63492, 16, 16, 16, 4, 48, 2, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 400x300 @ 56 Hz, 35.2 kHz hsync, 4:3 aspect ratio */ + NULL, 56, 400, 300, 55555, 64, 16, 10, 1, 32, 1, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 400x300 @ 60 Hz, 37.8 kHz hsync */ + NULL, 60, 400, 300, 50000, 48, 16, 11, 1, 64, 2, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 400x300 @ 72 Hz, 48.0 kHz hsync */ + NULL, 72, 400, 300, 40000, 32, 24, 11, 19, 64, 3, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 480x300 @ 56 Hz, 35.2 kHz hsync, 8:5 aspect ratio */ + NULL, 56, 480, 300, 46176, 80, 16, 10, 1, 40, 1, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 480x300 @ 60 Hz, 37.8 kHz hsync */ + NULL, 60, 480, 300, 41858, 56, 16, 11, 1, 80, 2, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 480x300 @ 63 Hz, 39.6 kHz hsync */ + NULL, 63, 480, 300, 40000, 56, 16, 11, 1, 80, 2, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 480x300 @ 72 Hz, 48.0 kHz hsync */ + NULL, 72, 480, 300, 33386, 40, 24, 11, 19, 80, 3, +- 0, FB_VMODE_DOUBLE ++ 0, FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN + }, { + /* 1920x1200 @ 60 Hz, 74.5 Khz hsync */ + NULL, 60, 1920, 1200, 5177, 128, 336, 1, 38, 208, 3, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1152x768, 60 Hz, PowerBook G4 Titanium I and II */ + NULL, 60, 1152, 768, 14047, 158, 26, 29, 3, 136, 6, +- FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED ++ FB_SYNC_HOR_HIGH_ACT|FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1366x768, 60 Hz, 47.403 kHz hsync, WXGA 16:9 aspect ratio */ + NULL, 60, 1366, 768, 13806, 120, 10, 14, 3, 32, 5, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 1280x800, 60 Hz, 47.403 kHz hsync, WXGA 16:10 aspect ratio */ + NULL, 60, 1280, 800, 12048, 200, 64, 24, 1, 136, 3, +- 0, FB_VMODE_NONINTERLACED ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 720x576i @ 50 Hz, 15.625 kHz hsync (PAL RGB) */ + NULL, 50, 720, 576, 74074, 64, 16, 39, 5, 64, 5, +- 0, FB_VMODE_INTERLACED ++ 0, FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN + }, { + /* 800x520i @ 50 Hz, 15.625 kHz hsync (PAL RGB) */ + NULL, 50, 800, 520, 58823, 144, 64, 72, 28, 80, 5, +- 0, FB_VMODE_INTERLACED ++ 0, FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN + }, + }; + +diff -urNp linux-2.6.31.1/drivers/video/omap/dispc.c linux-2.6.31.1/drivers/video/omap/dispc.c +--- linux-2.6.31.1/drivers/video/omap/dispc.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/video/omap/dispc.c 2009-10-01 20:12:43.000000000 -0400 +@@ -1013,7 +1013,7 @@ static void mmap_user_close(struct vm_ar + atomic_dec(&dispc.map_count[plane]); + } + +-static struct vm_operations_struct mmap_user_ops = { ++static const struct vm_operations_struct mmap_user_ops = { + .open = mmap_user_open, + .close = mmap_user_close, + }; +diff -urNp linux-2.6.31.1/drivers/video/uvesafb.c linux-2.6.31.1/drivers/video/uvesafb.c +--- linux-2.6.31.1/drivers/video/uvesafb.c 2009-09-24 11:45:25.000000000 -0400 ++++ linux-2.6.31.1/drivers/video/uvesafb.c 2009-10-01 20:12:43.000000000 -0400 +@@ -18,6 +18,7 @@ + #include + #include + #include ++#include + #include