4.9-stable patches

added patches:
	ceph-try-getting-buffer-capability-for-readahead-fadvise.patch

diff --git a/queue-4.9/ceph-try-getting-buffer-capability-for-readahead-fadvise.patch b/queue-4.9/ceph-try-getting-buffer-capability-for-readahead-fadvise.patch
new file mode 100644 (file)
index 0000000..b03dafa
--- /dev/null
+++ b/queue-4.9/ceph-try-getting-buffer-capability-for-readahead-fadvise.patch
@@ -0,0 +1,186 @@
+From 2b1ac852eb67a6e95595e576371d23519105559f Mon Sep 17 00:00:00 2001
+From: "Yan, Zheng" <zyan@redhat.com>
+Date: Tue, 25 Oct 2016 10:51:55 +0800
+Subject: ceph: try getting buffer capability for readahead/fadvise
+
+From: Yan, Zheng <zyan@redhat.com>
+
+commit 2b1ac852eb67a6e95595e576371d23519105559f upstream.
+
+For readahead/fadvise cases, caller of ceph_readpages does not
+hold buffer capability. Pages can be added to page cache while
+there is no buffer capability. This can cause data integrity
+issue.
+
+Signed-off-by: Yan, Zheng <zyan@redhat.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ fs/ceph/addr.c  |   58 ++++++++++++++++++++++++++++++++++++++++++++++----------
+ fs/ceph/caps.c  |   21 ++++++++++++++++++++
+ fs/ceph/file.c  |    3 +-
+ fs/ceph/super.h |    2 +
+ 4 files changed, 73 insertions(+), 11 deletions(-)
+
+--- a/fs/ceph/addr.c
++++ b/fs/ceph/addr.c
+@@ -315,7 +315,32 @@ static int start_read(struct inode *inod
+       struct page **pages;
+       pgoff_t next_index;
+       int nr_pages = 0;
+-      int ret;
++      int got = 0;
++      int ret = 0;
++
++      if (!current->journal_info) {
++              /* caller of readpages does not hold buffer and read caps
++               * (fadvise, madvise and readahead cases) */
++              int want = CEPH_CAP_FILE_CACHE;
++              ret = ceph_try_get_caps(ci, CEPH_CAP_FILE_RD, want, &got);
++              if (ret < 0) {
++                      dout("start_read %p, error getting cap\n", inode);
++              } else if (!(got & want)) {
++                      dout("start_read %p, no cache cap\n", inode);
++                      ret = 0;
++              }
++              if (ret <= 0) {
++                      if (got)
++                              ceph_put_cap_refs(ci, got);
++                      while (!list_empty(page_list)) {
++                              page = list_entry(page_list->prev,
++                                                struct page, lru);
++                              list_del(&page->lru);
++                              put_page(page);
++                      }
++                      return ret;
++              }
++      }
+ 
+       off = (u64) page_offset(page);
+ 
+@@ -338,15 +363,18 @@ static int start_read(struct inode *inod
+                                   CEPH_OSD_FLAG_READ, NULL,
+                                   ci->i_truncate_seq, ci->i_truncate_size,
+                                   false);
+-      if (IS_ERR(req))
+-              return PTR_ERR(req);
++      if (IS_ERR(req)) {
++              ret = PTR_ERR(req);
++              goto out;
++      }
+ 
+       /* build page vector */
+       nr_pages = calc_pages_for(0, len);
+       pages = kmalloc(sizeof(*pages) * nr_pages, GFP_KERNEL);
+-      ret = -ENOMEM;
+-      if (!pages)
+-              goto out;
++      if (!pages) {
++              ret = -ENOMEM;
++              goto out_put;
++      }
+       for (i = 0; i < nr_pages; ++i) {
+               page = list_entry(page_list->prev, struct page, lru);
+               BUG_ON(PageLocked(page));
+@@ -379,6 +407,12 @@ static int start_read(struct inode *inod
+       if (ret < 0)
+               goto out_pages;
+       ceph_osdc_put_request(req);
++
++      /* After adding locked pages to page cache, the inode holds cache cap.
++       * So we can drop our cap refs. */
++      if (got)
++              ceph_put_cap_refs(ci, got);
++
+       return nr_pages;
+ 
+ out_pages:
+@@ -387,8 +421,11 @@ out_pages:
+               unlock_page(pages[i]);
+       }
+       ceph_put_page_vector(pages, nr_pages, false);
+-out:
++out_put:
+       ceph_osdc_put_request(req);
++out:
++      if (got)
++              ceph_put_cap_refs(ci, got);
+       return ret;
+ }
+ 
+@@ -425,7 +462,6 @@ static int ceph_readpages(struct file *f
+               rc = start_read(inode, page_list, max);
+               if (rc < 0)
+                       goto out;
+-              BUG_ON(rc == 0);
+       }
+ out:
+       ceph_fscache_readpages_cancel(inode, page_list);
+@@ -1372,9 +1408,11 @@ static int ceph_filemap_fault(struct vm_
+            inode, off, (size_t)PAGE_SIZE, ceph_cap_string(got));
+ 
+       if ((got & (CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO)) ||
+-          ci->i_inline_version == CEPH_INLINE_NONE)
++          ci->i_inline_version == CEPH_INLINE_NONE) {
++              current->journal_info = vma->vm_file;
+               ret = filemap_fault(vma, vmf);
+-      else
++              current->journal_info = NULL;
++      } else
+               ret = -EAGAIN;
+ 
+       dout("filemap_fault %p %llu~%zd dropping cap refs on %s ret %d\n",
+--- a/fs/ceph/caps.c
++++ b/fs/ceph/caps.c
+@@ -2479,6 +2479,27 @@ static void check_max_size(struct inode
+               ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
+ }
+ 
++int ceph_try_get_caps(struct ceph_inode_info *ci, int need, int want, int *got)
++{
++      int ret, err = 0;
++
++      BUG_ON(need & ~CEPH_CAP_FILE_RD);
++      BUG_ON(want & ~(CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO));
++      ret = ceph_pool_perm_check(ci, need);
++      if (ret < 0)
++              return ret;
++
++      ret = try_get_cap_refs(ci, need, want, 0, true, got, &err);
++      if (ret) {
++              if (err == -EAGAIN) {
++                      ret = 0;
++              } else if (err < 0) {
++                      ret = err;
++              }
++      }
++      return ret;
++}
++
+ /*
+  * Wait for caps, and take cap references.  If we can't get a WR cap
+  * due to a small max_size, make sure we check_max_size (and possibly
+--- a/fs/ceph/file.c
++++ b/fs/ceph/file.c
+@@ -1249,8 +1249,9 @@ again:
+               dout("aio_read %p %llx.%llx %llu~%u got cap refs on %s\n",
+                    inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len,
+                    ceph_cap_string(got));
+-
++              current->journal_info = filp;
+               ret = generic_file_read_iter(iocb, to);
++              current->journal_info = NULL;
+       }
+       dout("aio_read %p %llx.%llx dropping cap refs on %s = %d\n",
+            inode, ceph_vinop(inode), ceph_cap_string(got), (int)ret);
+--- a/fs/ceph/super.h
++++ b/fs/ceph/super.h
+@@ -905,6 +905,8 @@ extern int ceph_encode_dentry_release(vo
+ 
+ extern int ceph_get_caps(struct ceph_inode_info *ci, int need, int want,
+                        loff_t endoff, int *got, struct page **pinned_page);
++extern int ceph_try_get_caps(struct ceph_inode_info *ci,
++                           int need, int want, int *got);
+ 
+ /* for counting open files by mode */
+ extern void __ceph_get_fmode(struct ceph_inode_info *ci, int mode);

diff --git a/queue-4.9/series b/queue-4.9/series
index 753e3e4865f008a454c7d93fbe25f9714a6f5e8f..93143cd3b798f132a54b4baa1d6d2dc8a53b782e 100644 (file)
--- a/queue-4.9/series
+++ b/queue-4.9/series
@@ -4,3 +4,4 @@ tpm_tis-use-default-timeout-value-if-chip-reports-it-as-zero.patch
 scsi-storvsc-workaround-for-virtual-dvd-scsi-version.patch
 hwmon-it87-avoid-registering-the-same-chip-on-both-sio-addresses.patch
 8250_pci-fix-potential-use-after-free-in-error-path.patch
+ceph-try-getting-buffer-capability-for-readahead-fadvise.patch