+zconf.lex.c
zoffset.h
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
-index 2fe6e76..df58221 100644
+index 2fe6e76..889ee23 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -976,6 +976,10 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
hashdist= [KNL,NUMA] Large hashes allocated during boot
are distributed across NUMA nodes. Defaults on
for 64-bit NUMA, off otherwise.
-@@ -2195,6 +2199,22 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
+@@ -1928,6 +1932,10 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
+ noexec=on: enable non-executable mappings (default)
+ noexec=off: disable non-executable mappings
+
++ nopcid [X86-64]
++ Disable PCID (Process-Context IDentifier) even if it
++ is supported by the processor.
++
+ nosmap [X86]
+ Disable SMAP (Supervisor Mode Access Prevention)
+ even if it is supported by processor.
+@@ -2195,6 +2203,25 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
the specified number of seconds. This is to be used if
your oopses keep scrolling off the screen.
+ Enable a very simple form of latent entropy extraction
+ from the first 4GB of memory as the bootmem allocator
+ passes the memory pages to the buddy allocator.
++
++ pax_weakuderef [X86-64] enables the weaker but faster form of UDEREF
++ when the processor supports PCID.
+
pcbit= [HW,ISDN]
pcd. [PARIDE]
diff --git a/Makefile b/Makefile
-index f8349d0..563a504 100644
+index 4b31d62..ac99d49 100644
--- a/Makefile
+++ b/Makefile
@@ -241,8 +241,9 @@ CONFIG_SHELL := $(shell if [ -x "$$BASH" ]; then echo $$BASH; \
/* Allow reads even for write-only mappings */
if (!(vma->vm_flags & (VM_READ | VM_WRITE)))
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
-index 136f263..f471277 100644
+index 18a9f5e..ca910b7 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1766,7 +1766,7 @@ config ALIGNMENT_TRAP
static inline void set_domain(unsigned val) { }
static inline void modify_domain(unsigned dom, unsigned type) { }
diff --git a/arch/arm/include/asm/elf.h b/arch/arm/include/asm/elf.h
-index 38050b1..9d90e8b 100644
+index 56211f2..17e8a25 100644
--- a/arch/arm/include/asm/elf.h
+++ b/arch/arm/include/asm/elf.h
@@ -116,7 +116,14 @@ int dump_task_regs(struct task_struct *t, elf_gregset_t *elfregs);
/* 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
-@@ -126,8 +133,4 @@ int dump_task_regs(struct task_struct *t, elf_gregset_t *elfregs);
+@@ -126,10 +133,6 @@ int dump_task_regs(struct task_struct *t, elf_gregset_t *elfregs);
extern void elf_set_personality(const struct elf32_hdr *);
#define SET_PERSONALITY(ex) elf_set_personality(&(ex))
-extern unsigned long arch_randomize_brk(struct mm_struct *mm);
-#define arch_randomize_brk arch_randomize_brk
-
- #endif
+ #ifdef CONFIG_MMU
+ #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1
+ struct linux_binprm;
diff --git a/arch/arm/include/asm/fncpy.h b/arch/arm/include/asm/fncpy.h
index de53547..52b9a28 100644
--- a/arch/arm/include/asm/fncpy.h
#ifdef CONFIG_OUTER_CACHE
diff --git a/arch/arm/include/asm/page.h b/arch/arm/include/asm/page.h
-index 812a494..71fc0b6 100644
+index cbdc7a2..32f44fe 100644
--- a/arch/arm/include/asm/page.h
+++ b/arch/arm/include/asm/page.h
@@ -114,7 +114,7 @@ struct cpu_user_fns {
#ifndef MULTI_CPU
extern void cpu_proc_init(void);
-diff --git a/arch/arm/include/asm/processor.h b/arch/arm/include/asm/processor.h
-index 06e7d50..8a8e251 100644
---- a/arch/arm/include/asm/processor.h
-+++ b/arch/arm/include/asm/processor.h
-@@ -65,9 +65,8 @@ struct thread_struct {
- regs->ARM_cpsr |= PSR_ENDSTATE; \
- regs->ARM_pc = pc & ~1; /* pc */ \
- regs->ARM_sp = sp; /* sp */ \
-- regs->ARM_r2 = stack[2]; /* r2 (envp) */ \
-- regs->ARM_r1 = stack[1]; /* r1 (argv) */ \
-- regs->ARM_r0 = stack[0]; /* r0 (argc) */ \
-+ /* r2 (envp), r1 (argv), r0 (argc) */ \
-+ (void)copy_from_user(®s->ARM_r0, (const char __user *)stack, 3 * sizeof(unsigned long)); \
- nommu_start_thread(regs); \
- })
-
diff --git a/arch/arm/include/asm/psci.h b/arch/arm/include/asm/psci.h
index ce0dbe7..c085b6f 100644
--- a/arch/arm/include/asm/psci.h
/*
* set platform specific SMP operations
diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h
-index 1995d1a..76693a2 100644
+index f00b569..aa5bb41 100644
--- a/arch/arm/include/asm/thread_info.h
+++ b/arch/arm/include/asm/thread_info.h
@@ -77,9 +77,9 @@ struct thread_info {
#define TIF_USING_IWMMXT 17
#define TIF_MEMDIE 18 /* is terminating due to OOM killer */
#define TIF_RESTORE_SIGMASK 20
-@@ -166,10 +170,11 @@ extern int vfp_restore_user_hwstate(struct user_vfp __user *,
+@@ -165,10 +169,11 @@ extern int vfp_restore_user_hwstate(struct user_vfp __user *,
#define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT)
#define _TIF_SECCOMP (1 << TIF_SECCOMP)
#define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT)
/*
* Change these and you break ASM code in entry-common.S
diff --git a/arch/arm/include/asm/uaccess.h b/arch/arm/include/asm/uaccess.h
-index 7e1f760..d42d7f8 100644
+index 7e1f760..de33b13 100644
--- a/arch/arm/include/asm/uaccess.h
+++ b/arch/arm/include/asm/uaccess.h
@@ -18,6 +18,7 @@
#define VERIFY_READ 0
#define VERIFY_WRITE 1
-@@ -63,11 +64,35 @@ extern int __put_user_bad(void);
+@@ -63,11 +64,38 @@ extern int __put_user_bad(void);
static inline void set_fs(mm_segment_t fs)
{
current_thread_info()->addr_limit = fs;
#define segment_eq(a,b) ((a) == (b))
++#define __HAVE_ARCH_PAX_OPEN_USERLAND
++#define __HAVE_ARCH_PAX_CLOSE_USERLAND
++
+static inline void pax_open_userland(void)
+{
+
#define __addr_ok(addr) ({ \
unsigned long flag; \
__asm__("cmp %2, %0; movlo %0, #0" \
-@@ -143,8 +168,12 @@ extern int __get_user_4(void *);
+@@ -143,8 +171,12 @@ extern int __get_user_4(void *);
#define get_user(x,p) \
({ \
})
extern int __put_user_1(void *, unsigned int);
-@@ -188,8 +217,12 @@ extern int __put_user_8(void *, unsigned long long);
+@@ -188,8 +220,12 @@ extern int __put_user_8(void *, unsigned long long);
#define put_user(x,p) \
({ \
})
#else /* CONFIG_MMU */
-@@ -230,13 +263,17 @@ static inline void set_fs(mm_segment_t fs)
+@@ -230,13 +266,17 @@ static inline void set_fs(mm_segment_t fs)
#define __get_user(x,ptr) \
({ \
long __gu_err = 0; \
(void) 0; \
})
-@@ -312,13 +349,17 @@ do { \
+@@ -312,13 +352,17 @@ do { \
#define __put_user(x,ptr) \
({ \
long __pu_err = 0; \
(void) 0; \
})
-@@ -418,11 +459,44 @@ do { \
+@@ -418,11 +462,44 @@ do { \
#ifdef CONFIG_MMU
#else
#define __copy_from_user(to,from,n) (memcpy(to, (void __force *)from, n), 0)
#define __copy_to_user(to,from,n) (memcpy((void __force *)to, from, n), 0)
-@@ -431,6 +505,9 @@ extern unsigned long __must_check __clear_user_std(void __user *addr, unsigned l
+@@ -431,6 +508,9 @@ extern unsigned long __must_check __clear_user_std(void __user *addr, unsigned l
static 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 /* security hole - plug it */
-@@ -440,6 +517,9 @@ static inline unsigned long __must_check copy_from_user(void *to, const void __u
+@@ -440,6 +520,9 @@ static inline unsigned long __must_check copy_from_user(void *to, const void __u
static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n)
{
EXPORT_SYMBOL(__get_user_1);
EXPORT_SYMBOL(__get_user_2);
diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S
-index 582b405..a78366b 100644
+index d43c7e5..257c050 100644
--- a/arch/arm/kernel/entry-armv.S
+++ b/arch/arm/kernel/entry-armv.S
@@ -47,6 +47,87 @@
sub sp, sp, #S_FRAME_SIZE
ARM( stmib sp, {r1 - r12} )
THUMB( stmia sp, {r0 - r12} )
-@@ -414,7 +511,9 @@ __und_usr:
+@@ -357,7 +454,8 @@ ENDPROC(__pabt_svc)
+ .endm
+
+ .macro kuser_cmpxchg_check
+-#if !defined(CONFIG_CPU_32v6K) && !defined(CONFIG_NEEDS_SYSCALL_FOR_CMPXCHG)
++#if !defined(CONFIG_CPU_32v6K) && defined(CONFIG_KUSER_HELPERS) && \
++ !defined(CONFIG_NEEDS_SYSCALL_FOR_CMPXCHG)
+ #ifndef CONFIG_MMU
+ #warning "NPTL on non MMU needs fixing"
+ #else
+@@ -414,7 +512,9 @@ __und_usr:
tst r3, #PSR_T_BIT @ Thumb mode?
bne __und_usr_thumb
sub r4, r2, #4 @ ARM instr at LR - 4
#ifdef CONFIG_CPU_ENDIAN_BE8
rev r0, r0 @ little endian instruction
#endif
-@@ -449,10 +548,14 @@ __und_usr_thumb:
+@@ -449,10 +549,14 @@ __und_usr_thumb:
*/
.arch armv6t2
#endif
add r2, r2, #2 @ r2 is PC + 2, make it PC + 4
str r2, [sp, #S_PC] @ it's a 2x16bit instr, update
orr r0, r0, r5, lsl #16
-@@ -481,7 +584,8 @@ ENDPROC(__und_usr)
+@@ -481,7 +585,8 @@ ENDPROC(__und_usr)
*/
.pushsection .fixup, "ax"
.align 2
.popsection
.pushsection __ex_table,"a"
.long 1b, 4b
-@@ -690,7 +794,7 @@ ENTRY(__switch_to)
+@@ -690,7 +795,7 @@ ENTRY(__switch_to)
THUMB( stmia ip!, {r4 - sl, fp} ) @ Store most regs on stack
THUMB( str sp, [ip], #4 )
THUMB( str lr, [ip], #4 )
ldr r6, [r2, #TI_CPU_DOMAIN]
#endif
set_tls r3, r4, r5
-@@ -699,7 +803,7 @@ ENTRY(__switch_to)
+@@ -699,7 +804,7 @@ ENTRY(__switch_to)
ldr r8, =__stack_chk_guard
ldr r7, [r7, #TSK_STACK_CANARY]
#endif
ldrd r0, r1, [sp, #S_LR] @ calling lr and pc
clrex @ clear the exclusive monitor
diff --git a/arch/arm/kernel/fiq.c b/arch/arm/kernel/fiq.c
-index 2adda11..7fbe958 100644
+index 25442f4..d4948fc 100644
--- a/arch/arm/kernel/fiq.c
+++ b/arch/arm/kernel/fiq.c
-@@ -82,7 +82,9 @@ void set_fiq_handler(void *start, unsigned int length)
- #if defined(CONFIG_CPU_USE_DOMAINS)
- memcpy((void *)0xffff001c, start, length);
- #else
+@@ -84,17 +84,16 @@ int show_fiq_list(struct seq_file *p, int prec)
+
+ void set_fiq_handler(void *start, unsigned int length)
+ {
+-#if defined(CONFIG_CPU_USE_DOMAINS)
+- void *base = (void *)0xffff0000;
+-#else
+ void *base = vectors_page;
+-#endif
+ unsigned offset = FIQ_OFFSET;
+
+ pax_open_kernel();
- memcpy(vectors_page + 0x1c, start, length);
+ memcpy(base + offset, start, length);
+ pax_close_kernel();
- #endif
- flush_icache_range(0xffff001c, 0xffff001c + length);
- if (!vectors_high())
++
++ if (!cache_is_vipt_nonaliasing())
++ flush_icache_range(base + offset, offset + length);
+ flush_icache_range(0xffff0000 + offset, 0xffff0000 + offset + length);
+- if (!vectors_high())
+- flush_icache_range(offset, offset + length);
+ }
+
+ int claim_fiq(struct fiq_handler *f)
diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S
index 8bac553..caee108 100644
--- a/arch/arm/kernel/head.S
flush_icache_range((uintptr_t)(addr),
(uintptr_t)(addr) + size);
+diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
+index e19edc6..e186ee1 100644
+--- a/arch/arm/kernel/perf_event.c
++++ b/arch/arm/kernel/perf_event.c
+@@ -56,7 +56,7 @@ armpmu_map_hw_event(const unsigned (*event_map)[PERF_COUNT_HW_MAX], u64 config)
+ int mapping;
+
+ if (config >= PERF_COUNT_HW_MAX)
+- return -ENOENT;
++ return -EINVAL;
+
+ mapping = (*event_map)[config];
+ return mapping == HW_OP_UNSUPPORTED ? -ENOENT : mapping;
diff --git a/arch/arm/kernel/perf_event_cpu.c b/arch/arm/kernel/perf_event_cpu.c
index 1f2740e..b36e225 100644
--- a/arch/arm/kernel/perf_event_cpu.c
};
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
-index 6e8931c..82ec6a5 100644
+index 5bc2615..dcd439f 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
-@@ -28,7 +28,6 @@
- #include <linux/tick.h>
- #include <linux/utsname.h>
- #include <linux/uaccess.h>
--#include <linux/random.h>
- #include <linux/hw_breakpoint.h>
- #include <linux/cpuidle.h>
- #include <linux/leds.h>
-@@ -223,6 +222,7 @@ void machine_power_off(void)
+@@ -223,6 +223,7 @@ void machine_power_off(void)
if (pm_power_off)
pm_power_off();
}
/*
-@@ -236,7 +236,7 @@ void machine_power_off(void)
+@@ -236,7 +237,7 @@ void machine_power_off(void)
* executing pre-reset code, and using RAM that the primary CPU's code wishes
* to use. Implementing such co-ordination would be essentially impossible.
*/
{
smp_send_stop();
-@@ -258,8 +258,8 @@ void __show_regs(struct pt_regs *regs)
+@@ -258,8 +259,8 @@ void __show_regs(struct pt_regs *regs)
show_regs_print_info(KERN_DEFAULT);
- print_symbol("PC is at %s\n", instruction_pointer(regs));
- print_symbol("LR is at %s\n", regs->ARM_lr);
-+ printk("PC is at %pA\n", instruction_pointer(regs));
-+ printk("LR is at %pA\n", regs->ARM_lr);
++ printk("PC is at %pA\n", (void *)instruction_pointer(regs));
++ printk("LR is at %pA\n", (void *)regs->ARM_lr);
printk("pc : [<%08lx>] lr : [<%08lx>] psr: %08lx\n"
"sp : %08lx ip : %08lx fp : %08lx\n",
regs->ARM_pc, regs->ARM_lr, regs->ARM_cpsr,
-@@ -426,12 +426,6 @@ unsigned long get_wchan(struct task_struct *p)
+@@ -426,12 +427,6 @@ unsigned long get_wchan(struct task_struct *p)
return 0;
}
-}
-
#ifdef CONFIG_MMU
+ #ifdef CONFIG_KUSER_HELPERS
/*
- * The vectors page is always readable from user space for the
-@@ -441,12 +435,12 @@ unsigned long arch_randomize_brk(struct mm_struct *mm)
- static struct vm_area_struct gate_vma = {
- .vm_start = 0xffff0000,
- .vm_end = 0xffff0000 + PAGE_SIZE,
-- .vm_flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC,
-+ .vm_flags = VM_NONE,
- };
+@@ -447,7 +442,7 @@ static struct vm_area_struct gate_vma = {
static int __init gate_vma_init(void)
{
- gate_vma.vm_page_prot = PAGE_READONLY_EXEC;
-+ gate_vma.vm_page_prot = vm_get_page_prot(gate_vma.vm_flags);
++ gate_vma.vm_page_prot = vm_get_page_prot(gate_vma.vm_flags);
return 0;
}
arch_initcall(gate_vma_init);
+@@ -466,48 +461,23 @@ int in_gate_area_no_mm(unsigned long addr)
+ {
+ return in_gate_area(NULL, addr);
+ }
+-#define is_gate_vma(vma) ((vma) = &gate_vma)
++#define is_gate_vma(vma) ((vma) == &gate_vma)
+ #else
+ #define is_gate_vma(vma) 0
+ #endif
+
+ const char *arch_vma_name(struct vm_area_struct *vma)
+ {
+- return is_gate_vma(vma) ? "[vectors]" :
+- (vma->vm_mm && vma->vm_start == vma->vm_mm->context.sigpage) ?
+- "[sigpage]" : NULL;
++ return is_gate_vma(vma) ? "[vectors]" : NULL;
+ }
+
+-static struct page *signal_page;
+-extern struct page *get_signal_page(void);
+-
+ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
+ {
+ struct mm_struct *mm = current->mm;
+- unsigned long addr;
+- int ret;
+-
+- if (!signal_page)
+- signal_page = get_signal_page();
+- if (!signal_page)
+- return -ENOMEM;
+
+ down_write(&mm->mmap_sem);
+- addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0);
+- if (IS_ERR_VALUE(addr)) {
+- ret = addr;
+- goto up_fail;
+- }
+-
+- ret = install_special_mapping(mm, addr, PAGE_SIZE,
+- VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC,
+- &signal_page);
+-
+- if (ret == 0)
+- mm->context.sigpage = addr;
+-
+- up_fail:
++ mm->context.sigpage = (PAGE_OFFSET + (get_random_int() % 0x3FFEFFE0)) & 0xFFFFFFFC;
+ up_write(&mm->mmap_sem);
+- return ret;
++ return 0;
+ }
+ #endif
diff --git a/arch/arm/kernel/psci.c b/arch/arm/kernel/psci.c
index 3653164..d83e55d 100644
--- a/arch/arm/kernel/psci.c
#ifdef MULTI_TLB
cpu_tlb = *list->tlb;
diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c
-index 296786b..a8d4dd5 100644
+index 5a42c12..a2bb7c6 100644
--- a/arch/arm/kernel/signal.c
+++ b/arch/arm/kernel/signal.c
-@@ -396,22 +396,14 @@ setup_return(struct pt_regs *regs, struct ksignal *ksig,
- __put_user(sigreturn_codes[idx+1], rc+1))
- return 1;
-
-- if (cpsr & MODE32_BIT) {
-- /*
-- * 32-bit code can use the new high-page
-- * signal return code support.
-- */
-- retcode = KERN_SIGRETURN_CODE + (idx << 2) + thumb;
-- } else {
-- /*
-- * Ensure that the instruction cache sees
-- * the return code written onto the stack.
-- */
-- flush_icache_range((unsigned long)rc,
-- (unsigned long)(rc + 2));
-+ /*
-+ * Ensure that the instruction cache sees
-+ * the return code written onto the stack.
-+ */
-+ flush_icache_range((unsigned long)rc,
-+ (unsigned long)(rc + 2));
-
-- retcode = ((unsigned long)rc) + thumb;
-- }
-+ retcode = ((unsigned long)rc) + thumb;
- }
+@@ -45,8 +45,6 @@ static const unsigned long sigreturn_codes[7] = {
+ MOV_R7_NR_RT_SIGRETURN, SWI_SYS_RT_SIGRETURN, SWI_THUMB_RT_SIGRETURN,
+ };
- regs->ARM_r0 = map_sig(ksig->sig);
+-static unsigned long signal_return_offset;
+-
+ #ifdef CONFIG_CRUNCH
+ static int preserve_crunch_context(struct crunch_sigframe __user *frame)
+ {
+@@ -406,8 +404,7 @@ setup_return(struct pt_regs *regs, struct ksignal *ksig,
+ * except when the MPU has protected the vectors
+ * page from PL0
+ */
+- retcode = mm->context.sigpage + signal_return_offset +
+- (idx << 2) + thumb;
++ retcode = mm->context.sigpage + (idx << 2) + thumb;
+ } else
+ #endif
+ {
+@@ -611,33 +608,3 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall)
+ } while (thread_flags & _TIF_WORK_MASK);
+ return 0;
+ }
+-
+-struct page *get_signal_page(void)
+-{
+- unsigned long ptr;
+- unsigned offset;
+- struct page *page;
+- void *addr;
+-
+- page = alloc_pages(GFP_KERNEL, 0);
+-
+- if (!page)
+- return NULL;
+-
+- addr = page_address(page);
+-
+- /* Give the signal return code some randomness */
+- offset = 0x200 + (get_random_int() & 0x7fc);
+- signal_return_offset = offset;
+-
+- /*
+- * Copy signal return handlers into the vector page, and
+- * set sigreturn to be a pointer to these.
+- */
+- memcpy(addr + offset, sigreturn_codes, sizeof(sigreturn_codes));
+-
+- ptr = (unsigned long)addr + offset;
+- flush_icache_range(ptr, ptr + sizeof(sigreturn_codes));
+-
+- return page;
+-}
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 5919eb4..b5d6dfe 100644
--- a/arch/arm/kernel/smp.c
void __init smp_set_ops(struct smp_operations *ops)
{
diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c
-index 18b32e8..b0c8dca 100644
+index 6b9567e..b8af2d6 100644
--- a/arch/arm/kernel/traps.c
+++ b/arch/arm/kernel/traps.c
-@@ -57,7 +57,7 @@ static void dump_mem(const char *, const char *, unsigned long, unsigned long);
+@@ -55,7 +55,7 @@ static void dump_mem(const char *, const char *, unsigned long, unsigned long);
void dump_backtrace_entry(unsigned long where, unsigned long from, unsigned long frame)
{
#ifdef CONFIG_KALLSYMS
#else
printk("Function entered at [<%08lx>] from [<%08lx>]\n", where, from);
#endif
-@@ -259,6 +259,8 @@ static arch_spinlock_t die_lock = __ARCH_SPIN_LOCK_UNLOCKED;
+@@ -257,6 +257,8 @@ static arch_spinlock_t die_lock = __ARCH_SPIN_LOCK_UNLOCKED;
static int die_owner = -1;
static unsigned int die_nest_count;
static unsigned long oops_begin(void)
{
int cpu;
-@@ -301,6 +303,9 @@ static void oops_end(unsigned long flags, struct pt_regs *regs, int signr)
+@@ -299,6 +301,9 @@ static void oops_end(unsigned long flags, struct pt_regs *regs, int signr)
panic("Fatal exception in interrupt");
if (panic_on_oops)
panic("Fatal exception");
if (signr)
do_exit(signr);
}
-@@ -594,7 +599,9 @@ asmlinkage int arm_syscall(int no, struct pt_regs *regs)
+@@ -592,7 +597,9 @@ asmlinkage int arm_syscall(int no, struct pt_regs *regs)
* The user helper at 0xffff0fe0 must be used instead.
* (see entry-armv.S for details)
*/
}
return 0;
-@@ -834,13 +841,10 @@ void __init early_trap_init(void *vectors_base)
- */
- kuser_get_tls_init(vectors);
+@@ -848,5 +855,9 @@ void __init early_trap_init(void *vectors_base)
+ kuser_init(vectors_base);
-- /*
-- * Copy signal return handlers into the vector page, and
-- * set sigreturn to be a pointer to these.
-- */
-- memcpy((void *)(vectors + KERN_SIGRETURN_CODE - CONFIG_VECTORS_BASE),
-- sigreturn_codes, sizeof(sigreturn_codes));
--
- flush_icache_range(vectors, vectors + PAGE_SIZE);
+ flush_icache_range(vectors, vectors + PAGE_SIZE * 2);
- modify_domain(DOMAIN_USER, DOMAIN_CLIENT);
+
+#ifndef CONFIG_PAX_MEMORY_UDEREF
+
}
diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
-index a871b8e..123b00a 100644
+index 33f2ea3..0b91824 100644
--- a/arch/arm/kernel/vmlinux.lds.S
+++ b/arch/arm/kernel/vmlinux.lds.S
@@ -8,7 +8,11 @@
#ifndef CONFIG_XIP_KERNEL
. = ALIGN(PAGE_SIZE);
-@@ -207,6 +220,11 @@ SECTIONS
+@@ -224,6 +237,11 @@ SECTIONS
. = PAGE_OFFSET + TEXT_OFFSET;
#else
__init_end = .;
extern void ux500_cpu_die(unsigned int cpu);
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
-index 35955b5..b475042 100644
+index 2950082..d0f0782 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
-@@ -432,7 +432,7 @@ config CPU_32v5
+@@ -436,7 +436,7 @@ config CPU_32v5
config CPU_32v6
bool
select TLS_REG_EMUL if !CPU_32v6K && !MMU
config CPU_32v6K
-@@ -581,6 +581,7 @@ config CPU_CP15_MPU
+@@ -585,6 +585,7 @@ config CPU_CP15_MPU
config CPU_USE_DOMAINS
bool
help
This option enables or disables the use of domain switching
via the set_fs() function.
+@@ -780,6 +781,7 @@ config NEED_KUSER_HELPERS
+ config KUSER_HELPERS
+ bool "Enable kuser helpers in vector page" if !NEED_KUSER_HELPERS
+ default y
++ depends on !(CPU_V6 || CPU_V6K || CPU_V7)
+ help
+ Warning: disabling this option may break user programs.
+
+@@ -790,7 +792,7 @@ config KUSER_HELPERS
+ run on ARMv4 through to ARMv7 without modification.
+
+ However, the fixed address nature of these helpers can be used
+- by ROP (return orientated programming) authors when creating
++ by ROP (Return Oriented Programming) authors when creating
+ exploits.
+
+ If all of the binaries and libraries which run on your platform
diff --git a/arch/arm/mm/alignment.c b/arch/arm/mm/alignment.c
index 6f4585b..7b6f52b 100644
--- a/arch/arm/mm/alignment.c
goto fault; \
} while (0)
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
-index 5dbf13f..1a60561 100644
+index 5dbf13f..ee1ec24 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -25,6 +25,7 @@
printk(KERN_ALERT "Unhandled fault: %s (0x%03x) at 0x%08lx\n",
inf->name, fsr, addr);
-@@ -575,9 +637,49 @@ do_PrefetchAbort(unsigned long addr, unsigned int ifsr, struct pt_regs *regs)
+@@ -569,15 +631,67 @@ hook_ifault_code(int nr, int (*fn)(unsigned long, unsigned int, struct pt_regs *
+ ifsr_info[nr].name = name;
+ }
+
++asmlinkage int sys_sigreturn(struct pt_regs *regs);
++asmlinkage int sys_rt_sigreturn(struct pt_regs *regs);
++
+ asmlinkage void __exception
+ do_PrefetchAbort(unsigned long addr, unsigned int ifsr, struct pt_regs *regs)
+ {
const struct fsr_info *inf = ifsr_info + fsr_fs(ifsr);
struct siginfo info;
+ if (user_mode(regs)) {
++ unsigned long sigpage = current->mm->context.sigpage;
++
++ if (sigpage <= addr && addr < sigpage + 7*4) {
++ if (addr < sigpage + 3*4)
++ sys_sigreturn(regs);
++ else
++ sys_rt_sigreturn(regs);
++ return;
++ }
+ if (addr == 0xffff0fe0UL) {
+ /*
+ * PaX: __kuser_get_tls emulation
mm->unmap_area = arch_unmap_area_topdown;
}
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
-index 4d409e6..f375351 100644
+index daf336f..4e6392c 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -36,6 +36,22 @@
.domain = DOMAIN_KERNEL,
},
#endif
-@@ -277,36 +301,65 @@ static struct mem_type mem_types[] = {
+@@ -277,36 +301,54 @@ static struct mem_type mem_types[] = {
.prot_pte = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
L_PTE_RDONLY,
.prot_l1 = PMD_TYPE_TABLE,
+ .domain = DOMAIN_VECTORS,
},
[MT_HIGH_VECTORS] = {
-- .prot_pte = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-- L_PTE_USER | L_PTE_RDONLY,
-+ /* we always want the vector page to be noaccess for userland on archs with
-+ XN where we can enforce some reasonable measure of security
-+ therefore, when kernexec is disabled, instead of L_PTE_USER | L_PTE_RDONLY
-+ which turns into supervisor rwx, userland rx, we instead omit that entirely,
-+ leaving it as supervisor rwx only
-+ */
-+#ifdef CONFIG_PAX_KERNEXEC
-+ .prot_pte = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY | L_PTE_RDONLY,
-+#elif __LINUX_ARM_ARCH__ >= 6
-+ .prot_pte = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY,
-+#else
-+ .prot_pte = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY | L_PTE_USER | L_PTE_RDONLY,
-+#endif
+ .prot_pte = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
+ L_PTE_USER | L_PTE_RDONLY,
.prot_l1 = PMD_TYPE_TABLE,
- .domain = DOMAIN_USER,
+ .domain = DOMAIN_VECTORS,
.domain = DOMAIN_KERNEL,
},
[MT_MEMORY_ITCM] = {
-@@ -316,10 +369,10 @@ static struct mem_type mem_types[] = {
+@@ -316,10 +358,10 @@ static struct mem_type mem_types[] = {
},
[MT_MEMORY_SO] = {
.prot_pte = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
.domain = DOMAIN_KERNEL,
},
[MT_MEMORY_DMA_READY] = {
-@@ -405,9 +458,35 @@ static void __init build_mem_type_table(void)
+@@ -405,9 +447,35 @@ static void __init build_mem_type_table(void)
* to prevent speculative instruction fetches.
*/
mem_types[MT_DEVICE].prot_sect |= PMD_SECT_XN;
}
if (cpu_arch >= CPU_ARCH_ARMv7 && (cr & CR_TRE)) {
/*
-@@ -468,6 +547,9 @@ static void __init build_mem_type_table(void)
+@@ -468,6 +536,9 @@ static void __init build_mem_type_table(void)
* from SVC mode and no access from userspace.
*/
mem_types[MT_ROM].prot_sect |= PMD_SECT_APX|PMD_SECT_AP_WRITE;
mem_types[MT_MINICLEAN].prot_sect |= PMD_SECT_APX|PMD_SECT_AP_WRITE;
mem_types[MT_CACHECLEAN].prot_sect |= PMD_SECT_APX|PMD_SECT_AP_WRITE;
#endif
-@@ -485,11 +567,17 @@ static void __init build_mem_type_table(void)
+@@ -485,11 +556,17 @@ static void __init build_mem_type_table(void)
mem_types[MT_DEVICE_WC].prot_pte |= L_PTE_SHARED;
mem_types[MT_DEVICE_CACHED].prot_sect |= PMD_SECT_S;
mem_types[MT_DEVICE_CACHED].prot_pte |= L_PTE_SHARED;
}
}
-@@ -500,15 +588,20 @@ static void __init build_mem_type_table(void)
+@@ -500,15 +577,20 @@ static void __init build_mem_type_table(void)
if (cpu_arch >= CPU_ARCH_ARMv6) {
if (cpu_arch >= CPU_ARCH_ARMv7 && (cr & CR_TRE)) {
/* Non-cacheable Normal is XCB = 001 */
}
#ifdef CONFIG_ARM_LPAE
-@@ -524,6 +617,8 @@ static void __init build_mem_type_table(void)
+@@ -524,6 +606,8 @@ static void __init build_mem_type_table(void)
vecs_pgprot |= PTE_EXT_AF;
#endif
for (i = 0; i < 16; i++) {
pteval_t v = pgprot_val(protection_map[i]);
protection_map[i] = __pgprot(v | user_pgprot);
-@@ -541,10 +636,15 @@ static void __init build_mem_type_table(void)
+@@ -541,10 +625,15 @@ static void __init build_mem_type_table(void)
mem_types[MT_LOW_VECTORS].prot_l1 |= ecc_mask;
mem_types[MT_HIGH_VECTORS].prot_l1 |= ecc_mask;
mem_types[MT_ROM].prot_sect |= cp->pmd;
switch (cp->pmd) {
-@@ -1166,18 +1266,15 @@ void __init arm_mm_memblock_reserve(void)
+@@ -1166,18 +1255,15 @@ void __init arm_mm_memblock_reserve(void)
* called function. This means you can't use any function or debugging
* method which may touch any device, otherwise the kernel _will_ crash.
*/
+
-+static char vectors[PAGE_SIZE] __read_only __aligned(PAGE_SIZE);
++static char vectors[PAGE_SIZE * 2] __read_only __aligned(PAGE_SIZE);
+
static void __init devicemaps_init(struct machine_desc *mdesc)
{
- /*
- * Allocate the vector page early.
- */
-- vectors = early_alloc(PAGE_SIZE);
+- vectors = early_alloc(PAGE_SIZE * 2);
-
- early_trap_init(vectors);
+ early_trap_init(&vectors);
for (addr = VMALLOC_START; addr; addr += PMD_SIZE)
pmd_clear(pmd_off_k(addr));
-@@ -1217,7 +1314,7 @@ static void __init devicemaps_init(struct machine_desc *mdesc)
+@@ -1217,7 +1303,7 @@ static void __init devicemaps_init(struct machine_desc *mdesc)
* location (0xffff0000). If we aren't using high-vectors, also
* create a mapping at the low-vectors virtual address.
*/
+ map.pfn = __phys_to_pfn(virt_to_phys(&vectors));
map.virtual = 0xffff0000;
map.length = PAGE_SIZE;
- map.type = MT_HIGH_VECTORS;
-@@ -1275,8 +1372,39 @@ static void __init map_lowmem(void)
+ #ifdef CONFIG_KUSER_HELPERS
+@@ -1287,8 +1373,39 @@ static void __init map_lowmem(void)
map.pfn = __phys_to_pfn(start);
map.virtual = __phys_to_virt(start);
map.length = end - start;
+#define arch_align_stack(x) ((x) & ~0xfUL)
#endif /* _ASM_EXEC_H */
+diff --git a/arch/mips/include/asm/local.h b/arch/mips/include/asm/local.h
+index d44622c..64990d2 100644
+--- a/arch/mips/include/asm/local.h
++++ b/arch/mips/include/asm/local.h
+@@ -12,15 +12,25 @@ typedef struct
+ atomic_long_t a;
+ } local_t;
+
++typedef struct {
++ atomic_long_unchecked_t a;
++} local_unchecked_t;
++
+ #define LOCAL_INIT(i) { ATOMIC_LONG_INIT(i) }
+
+ #define local_read(l) atomic_long_read(&(l)->a)
++#define local_read_unchecked(l) atomic_long_read_unchecked(&(l)->a)
+ #define local_set(l, i) atomic_long_set(&(l)->a, (i))
++#define local_set_unchecked(l, i) atomic_long_set_unchecked(&(l)->a, (i))
+
+ #define local_add(i, l) atomic_long_add((i), (&(l)->a))
++#define local_add_unchecked(i, l) atomic_long_add_unchecked((i), (&(l)->a))
+ #define local_sub(i, l) atomic_long_sub((i), (&(l)->a))
++#define local_sub_unchecked(i, l) atomic_long_sub_unchecked((i), (&(l)->a))
+ #define local_inc(l) atomic_long_inc(&(l)->a)
++#define local_inc_unchecked(l) atomic_long_inc_unchecked(&(l)->a)
+ #define local_dec(l) atomic_long_dec(&(l)->a)
++#define local_dec_unchecked(l) atomic_long_dec_unchecked(&(l)->a)
+
+ /*
+ * Same as above, but return the result value
+@@ -70,6 +80,51 @@ static __inline__ long local_add_return(long i, local_t * l)
+ return result;
+ }
+
++static __inline__ long local_add_return_unchecked(long i, local_unchecked_t * l)
++{
++ unsigned long result;
++
++ if (kernel_uses_llsc && R10000_LLSC_WAR) {
++ unsigned long temp;
++
++ __asm__ __volatile__(
++ " .set mips3 \n"
++ "1:" __LL "%1, %2 # local_add_return \n"
++ " addu %0, %1, %3 \n"
++ __SC "%0, %2 \n"
++ " beqzl %0, 1b \n"
++ " addu %0, %1, %3 \n"
++ " .set mips0 \n"
++ : "=&r" (result), "=&r" (temp), "=m" (l->a.counter)
++ : "Ir" (i), "m" (l->a.counter)
++ : "memory");
++ } else if (kernel_uses_llsc) {
++ unsigned long temp;
++
++ __asm__ __volatile__(
++ " .set mips3 \n"
++ "1:" __LL "%1, %2 # local_add_return \n"
++ " addu %0, %1, %3 \n"
++ __SC "%0, %2 \n"
++ " beqz %0, 1b \n"
++ " addu %0, %1, %3 \n"
++ " .set mips0 \n"
++ : "=&r" (result), "=&r" (temp), "=m" (l->a.counter)
++ : "Ir" (i), "m" (l->a.counter)
++ : "memory");
++ } else {
++ unsigned long flags;
++
++ local_irq_save(flags);
++ result = l->a.counter;
++ result += i;
++ l->a.counter = result;
++ local_irq_restore(flags);
++ }
++
++ return result;
++}
++
+ static __inline__ long local_sub_return(long i, local_t * l)
+ {
+ unsigned long result;
+@@ -117,6 +172,8 @@ static __inline__ long local_sub_return(long i, local_t * l)
+
+ #define local_cmpxchg(l, o, n) \
+ ((long)cmpxchg_local(&((l)->a.counter), (o), (n)))
++#define local_cmpxchg_unchecked(l, o, n) \
++ ((long)cmpxchg_local(&((l)->a.counter), (o), (n)))
+ #define local_xchg(l, n) (atomic_long_xchg((&(l)->a), (n)))
+
+ /**
diff --git a/arch/mips/include/asm/page.h b/arch/mips/include/asm/page.h
index f59552f..3abe9b9 100644
--- a/arch/mips/include/asm/page.h
#define _PAGE_NO_CACHE 0x020 /* I: cache inhibit */
#define _PAGE_WRITETHRU 0x040 /* W: cache write-through */
diff --git a/arch/powerpc/include/asm/reg.h b/arch/powerpc/include/asm/reg.h
-index 362142b..8b22c1b 100644
+index e1fb161..2290d1d 100644
--- a/arch/powerpc/include/asm/reg.h
+++ b/arch/powerpc/include/asm/reg.h
@@ -234,6 +234,7 @@
#define DSISR_ISSTORE 0x02000000 /* access was a store */
#define DSISR_DABRMATCH 0x00400000 /* hit data breakpoint */
diff --git a/arch/powerpc/include/asm/smp.h b/arch/powerpc/include/asm/smp.h
-index ffbaabe..eabe843 100644
+index 48cfc85..891382f 100644
--- a/arch/powerpc/include/asm/smp.h
+++ b/arch/powerpc/include/asm/smp.h
@@ -50,7 +50,7 @@ struct smp_ops_t {
ld r4,_DAR(r1)
bl .bad_page_fault
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
-index 4e00d22..b26abcc 100644
+index 902ca3c..e942155 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
-@@ -1356,10 +1356,10 @@ handle_page_fault:
+@@ -1357,10 +1357,10 @@ handle_page_fault:
11: ld r4,_DAR(r1)
ld r5,_DSISR(r1)
addi r3,r1,STACK_FRAME_OVERHEAD
/* Find this entry, or if that fails, the next avail. entry */
while (entry->jump[0]) {
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
-index 076d124..6cb2cbf 100644
+index 7baa27b..f6b394a 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
-@@ -874,8 +874,8 @@ void show_regs(struct pt_regs * regs)
+@@ -884,8 +884,8 @@ void show_regs(struct pt_regs * regs)
* Lookup NIP late so we have the best change of getting the
* above info out without failing
*/
#endif
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
printk("PACATMSCRATCH [%llx]\n", get_paca()->tm_scratch);
-@@ -1335,10 +1335,10 @@ void show_stack(struct task_struct *tsk, unsigned long *stack)
+@@ -1345,10 +1345,10 @@ void show_stack(struct task_struct *tsk, unsigned long *stack)
newsp = stack[0];
ip = stack[STACK_FRAME_LR_SAVE];
if (!firstframe || ip != lr) {
(void *)current->ret_stack[curr_frame].ret);
curr_frame--;
}
-@@ -1358,7 +1358,7 @@ void show_stack(struct task_struct *tsk, unsigned long *stack)
+@@ -1368,7 +1368,7 @@ void show_stack(struct task_struct *tsk, unsigned long *stack)
struct pt_regs *regs = (struct pt_regs *)
(sp + STACK_FRAME_OVERHEAD);
lr = regs->link;
regs->trap, (void *)regs->nip, (void *)lr);
firstframe = 1;
}
-@@ -1394,58 +1394,3 @@ void notrace __ppc64_runlatch_off(void)
+@@ -1404,58 +1404,3 @@ void notrace __ppc64_runlatch_off(void)
mtspr(SPRN_CTRLT, ctrl);
}
#endif /* CONFIG_PPC64 */
};
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
-index e4f205a..8bfffb8 100644
+index 88929b1..bece8f8 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
-@@ -143,6 +143,8 @@ static unsigned __kprobes long oops_begin(struct pt_regs *regs)
+@@ -141,6 +141,8 @@ static unsigned __kprobes long oops_begin(struct pt_regs *regs)
return flags;
}
static void __kprobes oops_end(unsigned long flags, struct pt_regs *regs,
int signr)
{
-@@ -192,6 +194,9 @@ static void __kprobes oops_end(unsigned long flags, struct pt_regs *regs,
+@@ -190,6 +192,9 @@ static void __kprobes oops_end(unsigned long flags, struct pt_regs *regs,
panic("Fatal exception in interrupt");
if (panic_on_oops)
panic("Fatal exception");
};
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
-index 2859a1f..74f9a6e 100644
+index cafad40..9cbc0fc 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
-@@ -919,7 +919,7 @@ static void __init *careful_zallocation(int nid, unsigned long size,
+@@ -920,7 +920,7 @@ static void __init *careful_zallocation(int nid, unsigned long size,
return ret;
}
* 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 --git a/arch/sparc/mm/hugetlbpage.c b/arch/sparc/mm/hugetlbpage.c
-index d2b5944..bd813f2 100644
+index d2b5944..d878f3c 100644
--- a/arch/sparc/mm/hugetlbpage.c
+++ b/arch/sparc/mm/hugetlbpage.c
-@@ -38,7 +38,7 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *filp,
+@@ -28,7 +28,8 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *filp,
+ unsigned long addr,
+ unsigned long len,
+ unsigned long pgoff,
+- unsigned long flags)
++ unsigned long flags,
++ unsigned long offset)
+ {
+ unsigned long task_size = TASK_SIZE;
+ struct vm_unmapped_area_info info;
+@@ -38,15 +39,22 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *filp,
info.flags = 0;
info.length = len;
info.high_limit = min(task_size, VA_EXCLUDE_START);
info.align_mask = PAGE_MASK & ~HPAGE_MASK;
info.align_offset = 0;
-@@ -47,6 +47,12 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *filp,
++ info.threadstack_offset = offset;
+ addr = vm_unmapped_area(&info);
+
if ((addr & ~PAGE_MASK) && task_size > VA_EXCLUDE_END) {
VM_BUG_ON(addr != -ENOMEM);
info.low_limit = VA_EXCLUDE_END;
info.high_limit = task_size;
addr = vm_unmapped_area(&info);
}
-@@ -85,6 +91,12 @@ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+@@ -58,7 +66,8 @@ static unsigned long
+ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ const unsigned long len,
+ const unsigned long pgoff,
+- const unsigned long flags)
++ const unsigned long flags,
++ const unsigned long offset)
+ {
+ struct mm_struct *mm = current->mm;
+ unsigned long addr = addr0;
+@@ -73,6 +82,7 @@ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ info.high_limit = mm->mmap_base;
+ info.align_mask = PAGE_MASK & ~HPAGE_MASK;
+ info.align_offset = 0;
++ info.threadstack_offset = offset;
+ addr = vm_unmapped_area(&info);
+
+ /*
+@@ -85,6 +95,12 @@ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
VM_BUG_ON(addr != -ENOMEM);
info.flags = 0;
info.low_limit = TASK_UNMAPPED_BASE;
info.high_limit = STACK_TOP32;
addr = vm_unmapped_area(&info);
}
-@@ -99,6 +111,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
+@@ -99,6 +115,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
struct mm_struct *mm = current->mm;
struct vm_area_struct *vma;
unsigned long task_size = TASK_SIZE;
if (test_thread_flag(TIF_32BIT))
task_size = STACK_TOP32;
-@@ -114,11 +127,14 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
+@@ -114,19 +131,22 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
return addr;
}
return addr;
}
if (mm->get_unmapped_area == arch_get_unmapped_area)
+ return hugetlb_get_unmapped_area_bottomup(file, addr, len,
+- pgoff, flags);
++ pgoff, flags, offset);
+ else
+ return hugetlb_get_unmapped_area_topdown(file, addr, len,
+- pgoff, flags);
++ pgoff, flags, offset);
+ }
+
+ pte_t *huge_pte_alloc(struct mm_struct *mm,
diff --git a/arch/tile/include/asm/atomic_64.h b/arch/tile/include/asm/atomic_64.h
index f4500c6..889656c 100644
--- a/arch/tile/include/asm/atomic_64.h
ret
ENDPROC(aesni_xts_crypt8)
+diff --git a/arch/x86/crypto/blowfish-avx2-asm_64.S b/arch/x86/crypto/blowfish-avx2-asm_64.S
+index 784452e..46982c7 100644
+--- a/arch/x86/crypto/blowfish-avx2-asm_64.S
++++ b/arch/x86/crypto/blowfish-avx2-asm_64.S
+@@ -221,6 +221,7 @@ __blowfish_enc_blk32:
+
+ write_block(RXl, RXr);
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(__blowfish_enc_blk32)
+
+@@ -250,6 +251,7 @@ __blowfish_dec_blk32:
+
+ write_block(RXl, RXr);
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(__blowfish_dec_blk32)
+
+@@ -284,6 +286,7 @@ ENTRY(blowfish_ecb_enc_32way)
+
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(blowfish_ecb_enc_32way)
+
+@@ -318,6 +321,7 @@ ENTRY(blowfish_ecb_dec_32way)
+
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(blowfish_ecb_dec_32way)
+
+@@ -365,6 +369,7 @@ ENTRY(blowfish_cbc_dec_32way)
+
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(blowfish_cbc_dec_32way)
+
+@@ -445,5 +450,6 @@ ENTRY(blowfish_ctr_32way)
+
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(blowfish_ctr_32way)
diff --git a/arch/x86/crypto/blowfish-x86_64-asm_64.S b/arch/x86/crypto/blowfish-x86_64-asm_64.S
index 246c670..4d1ed00 100644
--- a/arch/x86/crypto/blowfish-x86_64-asm_64.S
+ pax_force_retaddr 0, 1
ret;
ENDPROC(blowfish_dec_blk_4way)
+diff --git a/arch/x86/crypto/camellia-aesni-avx-asm_64.S b/arch/x86/crypto/camellia-aesni-avx-asm_64.S
+index ce71f92..2dd5b1e 100644
+--- a/arch/x86/crypto/camellia-aesni-avx-asm_64.S
++++ b/arch/x86/crypto/camellia-aesni-avx-asm_64.S
+@@ -16,6 +16,7 @@
+ */
+
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+
+ #define CAMELLIA_TABLE_BYTE_LEN 272
+
+@@ -191,6 +192,7 @@ roundsm16_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd:
+ roundsm16(%xmm0, %xmm1, %xmm2, %xmm3, %xmm4, %xmm5, %xmm6, %xmm7,
+ %xmm8, %xmm9, %xmm10, %xmm11, %xmm12, %xmm13, %xmm14, %xmm15,
+ %rcx, (%r9));
++ pax_force_retaddr_bts
+ ret;
+ ENDPROC(roundsm16_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
+
+@@ -199,6 +201,7 @@ roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab:
+ roundsm16(%xmm4, %xmm5, %xmm6, %xmm7, %xmm0, %xmm1, %xmm2, %xmm3,
+ %xmm12, %xmm13, %xmm14, %xmm15, %xmm8, %xmm9, %xmm10, %xmm11,
+ %rax, (%r9));
++ pax_force_retaddr_bts
+ ret;
+ ENDPROC(roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
+
+@@ -780,6 +783,7 @@ __camellia_enc_blk16:
+ %xmm8, %xmm9, %xmm10, %xmm11, %xmm12, %xmm13, %xmm14,
+ %xmm15, (key_table)(CTX, %r8, 8), (%rax), 1 * 16(%rax));
+
++ pax_force_retaddr_bts
+ ret;
+
+ .align 8
+@@ -865,6 +869,7 @@ __camellia_dec_blk16:
+ %xmm8, %xmm9, %xmm10, %xmm11, %xmm12, %xmm13, %xmm14,
+ %xmm15, (key_table)(CTX), (%rax), 1 * 16(%rax));
+
++ pax_force_retaddr_bts
+ ret;
+
+ .align 8
+@@ -904,6 +909,7 @@ ENTRY(camellia_ecb_enc_16way)
+ %xmm15, %xmm14, %xmm13, %xmm12, %xmm11, %xmm10, %xmm9,
+ %xmm8, %rsi);
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_ecb_enc_16way)
+
+@@ -932,6 +938,7 @@ ENTRY(camellia_ecb_dec_16way)
+ %xmm15, %xmm14, %xmm13, %xmm12, %xmm11, %xmm10, %xmm9,
+ %xmm8, %rsi);
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_ecb_dec_16way)
+
+@@ -981,6 +988,7 @@ ENTRY(camellia_cbc_dec_16way)
+ %xmm15, %xmm14, %xmm13, %xmm12, %xmm11, %xmm10, %xmm9,
+ %xmm8, %rsi);
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_cbc_dec_16way)
+
+@@ -1092,6 +1100,7 @@ ENTRY(camellia_ctr_16way)
+ %xmm15, %xmm14, %xmm13, %xmm12, %xmm11, %xmm10, %xmm9,
+ %xmm8, %rsi);
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_ctr_16way)
+
+@@ -1234,6 +1243,7 @@ camellia_xts_crypt_16way:
+ %xmm15, %xmm14, %xmm13, %xmm12, %xmm11, %xmm10, %xmm9,
+ %xmm8, %rsi);
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_xts_crypt_16way)
+
+diff --git a/arch/x86/crypto/camellia-aesni-avx2-asm_64.S b/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
+index 91a1878..bcf340a 100644
+--- a/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
++++ b/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
+@@ -11,6 +11,7 @@
+ */
+
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+
+ #define CAMELLIA_TABLE_BYTE_LEN 272
+
+@@ -212,6 +213,7 @@ roundsm32_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd:
+ roundsm32(%ymm0, %ymm1, %ymm2, %ymm3, %ymm4, %ymm5, %ymm6, %ymm7,
+ %ymm8, %ymm9, %ymm10, %ymm11, %ymm12, %ymm13, %ymm14, %ymm15,
+ %rcx, (%r9));
++ pax_force_retaddr_bts
+ ret;
+ ENDPROC(roundsm32_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
+
+@@ -220,6 +222,7 @@ roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab:
+ roundsm32(%ymm4, %ymm5, %ymm6, %ymm7, %ymm0, %ymm1, %ymm2, %ymm3,
+ %ymm12, %ymm13, %ymm14, %ymm15, %ymm8, %ymm9, %ymm10, %ymm11,
+ %rax, (%r9));
++ pax_force_retaddr_bts
+ ret;
+ ENDPROC(roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
+
+@@ -802,6 +805,7 @@ __camellia_enc_blk32:
+ %ymm8, %ymm9, %ymm10, %ymm11, %ymm12, %ymm13, %ymm14,
+ %ymm15, (key_table)(CTX, %r8, 8), (%rax), 1 * 32(%rax));
+
++ pax_force_retaddr_bts
+ ret;
+
+ .align 8
+@@ -887,6 +891,7 @@ __camellia_dec_blk32:
+ %ymm8, %ymm9, %ymm10, %ymm11, %ymm12, %ymm13, %ymm14,
+ %ymm15, (key_table)(CTX), (%rax), 1 * 32(%rax));
+
++ pax_force_retaddr_bts
+ ret;
+
+ .align 8
+@@ -930,6 +935,7 @@ ENTRY(camellia_ecb_enc_32way)
+
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_ecb_enc_32way)
+
+@@ -962,6 +968,7 @@ ENTRY(camellia_ecb_dec_32way)
+
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_ecb_dec_32way)
+
+@@ -1028,6 +1035,7 @@ ENTRY(camellia_cbc_dec_32way)
+
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_cbc_dec_32way)
+
+@@ -1166,6 +1174,7 @@ ENTRY(camellia_ctr_32way)
+
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_ctr_32way)
+
+@@ -1331,6 +1340,7 @@ camellia_xts_crypt_32way:
+
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(camellia_xts_crypt_32way)
+
diff --git a/arch/x86/crypto/camellia-x86_64-asm_64.S b/arch/x86/crypto/camellia-x86_64-asm_64.S
index 310319c..ce174a4 100644
--- a/arch/x86/crypto/camellia-x86_64-asm_64.S
+ pax_force_retaddr
ret;
ENDPROC(cast6_xts_dec_8way)
+diff --git a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S
+index dbc4339..3d868c5 100644
+--- a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S
++++ b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S
+@@ -45,6 +45,7 @@
+
+ #include <asm/inst.h>
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+
+ ## ISCSI CRC 32 Implementation with crc32 and pclmulqdq Instruction
+
+@@ -312,6 +313,7 @@ do_return:
+ popq %rsi
+ popq %rdi
+ popq %rbx
++ pax_force_retaddr 0, 1
+ ret
+
+ ################################################################
+diff --git a/arch/x86/crypto/ghash-clmulni-intel_asm.S b/arch/x86/crypto/ghash-clmulni-intel_asm.S
+index 586f41a..d02851e 100644
+--- a/arch/x86/crypto/ghash-clmulni-intel_asm.S
++++ b/arch/x86/crypto/ghash-clmulni-intel_asm.S
+@@ -18,6 +18,7 @@
+
+ #include <linux/linkage.h>
+ #include <asm/inst.h>
++#include <asm/alternative-asm.h>
+
+ .data
+
+@@ -93,6 +94,7 @@ __clmul_gf128mul_ble:
+ psrlq $1, T2
+ pxor T2, T1
+ pxor T1, DATA
++ pax_force_retaddr
+ ret
+ ENDPROC(__clmul_gf128mul_ble)
+
+@@ -105,6 +107,7 @@ ENTRY(clmul_ghash_mul)
+ call __clmul_gf128mul_ble
+ PSHUFB_XMM BSWAP DATA
+ movups DATA, (%rdi)
++ pax_force_retaddr
+ ret
+ ENDPROC(clmul_ghash_mul)
+
+@@ -132,6 +135,7 @@ ENTRY(clmul_ghash_update)
+ PSHUFB_XMM BSWAP DATA
+ movups DATA, (%rdi)
+ .Lupdate_just_ret:
++ pax_force_retaddr
+ ret
+ ENDPROC(clmul_ghash_update)
+
+@@ -157,5 +161,6 @@ ENTRY(clmul_ghash_setkey)
+ pand .Lpoly, %xmm1
+ pxor %xmm1, %xmm0
+ movups %xmm0, (%rdi)
++ pax_force_retaddr
+ ret
+ ENDPROC(clmul_ghash_setkey)
diff --git a/arch/x86/crypto/salsa20-x86_64-asm_64.S b/arch/x86/crypto/salsa20-x86_64-asm_64.S
index 9279e0b..9270820 100644
--- a/arch/x86/crypto/salsa20-x86_64-asm_64.S
+ pax_force_retaddr
ret;
ENDPROC(serpent_xts_dec_8way_avx)
+diff --git a/arch/x86/crypto/serpent-avx2-asm_64.S b/arch/x86/crypto/serpent-avx2-asm_64.S
+index b222085..abd483c 100644
+--- a/arch/x86/crypto/serpent-avx2-asm_64.S
++++ b/arch/x86/crypto/serpent-avx2-asm_64.S
+@@ -15,6 +15,7 @@
+ */
+
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+ #include "glue_helper-asm-avx2.S"
+
+ .file "serpent-avx2-asm_64.S"
+@@ -610,6 +611,7 @@ __serpent_enc_blk16:
+ write_blocks(RA1, RB1, RC1, RD1, RK0, RK1, RK2);
+ write_blocks(RA2, RB2, RC2, RD2, RK0, RK1, RK2);
+
++ pax_force_retaddr
+ ret;
+ ENDPROC(__serpent_enc_blk16)
+
+@@ -664,6 +666,7 @@ __serpent_dec_blk16:
+ write_blocks(RC1, RD1, RB1, RE1, RK0, RK1, RK2);
+ write_blocks(RC2, RD2, RB2, RE2, RK0, RK1, RK2);
+
++ pax_force_retaddr
+ ret;
+ ENDPROC(__serpent_dec_blk16)
+
+@@ -684,6 +687,7 @@ ENTRY(serpent_ecb_enc_16way)
+
+ vzeroupper;
+
++ pax_force_retaddr
+ ret;
+ ENDPROC(serpent_ecb_enc_16way)
+
+@@ -704,6 +708,7 @@ ENTRY(serpent_ecb_dec_16way)
+
+ vzeroupper;
+
++ pax_force_retaddr
+ ret;
+ ENDPROC(serpent_ecb_dec_16way)
+
+@@ -725,6 +730,7 @@ ENTRY(serpent_cbc_dec_16way)
+
+ vzeroupper;
+
++ pax_force_retaddr
+ ret;
+ ENDPROC(serpent_cbc_dec_16way)
+
+@@ -748,6 +754,7 @@ ENTRY(serpent_ctr_16way)
+
+ vzeroupper;
+
++ pax_force_retaddr
+ ret;
+ ENDPROC(serpent_ctr_16way)
+
+@@ -772,6 +779,7 @@ ENTRY(serpent_xts_enc_16way)
+
+ vzeroupper;
+
++ pax_force_retaddr
+ ret;
+ ENDPROC(serpent_xts_enc_16way)
+
+@@ -796,5 +804,6 @@ ENTRY(serpent_xts_dec_16way)
+
+ vzeroupper;
+
++ pax_force_retaddr
+ ret;
+ ENDPROC(serpent_xts_dec_16way)
diff --git a/arch/x86/crypto/serpent-sse2-x86_64-asm_64.S b/arch/x86/crypto/serpent-sse2-x86_64-asm_64.S
index acc066c..1559cc4 100644
--- a/arch/x86/crypto/serpent-sse2-x86_64-asm_64.S
ret
ENDPROC(\name)
+diff --git a/arch/x86/crypto/sha256-avx-asm.S b/arch/x86/crypto/sha256-avx-asm.S
+index 642f156..4ab07b9 100644
+--- a/arch/x86/crypto/sha256-avx-asm.S
++++ b/arch/x86/crypto/sha256-avx-asm.S
+@@ -49,6 +49,7 @@
+
+ #ifdef CONFIG_AS_AVX
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+
+ ## assume buffers not aligned
+ #define VMOVDQ vmovdqu
+@@ -460,6 +461,7 @@ done_hash:
+ popq %r13
+ popq %rbp
+ popq %rbx
++ pax_force_retaddr 0, 1
+ ret
+ ENDPROC(sha256_transform_avx)
+
+diff --git a/arch/x86/crypto/sha256-avx2-asm.S b/arch/x86/crypto/sha256-avx2-asm.S
+index 9e86944..2e7f95a 100644
+--- a/arch/x86/crypto/sha256-avx2-asm.S
++++ b/arch/x86/crypto/sha256-avx2-asm.S
+@@ -50,6 +50,7 @@
+
+ #ifdef CONFIG_AS_AVX2
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+
+ ## assume buffers not aligned
+ #define VMOVDQ vmovdqu
+@@ -720,6 +721,7 @@ done_hash:
+ popq %r12
+ popq %rbp
+ popq %rbx
++ pax_force_retaddr 0, 1
+ ret
+ ENDPROC(sha256_transform_rorx)
+
+diff --git a/arch/x86/crypto/sha256-ssse3-asm.S b/arch/x86/crypto/sha256-ssse3-asm.S
+index f833b74..c36ed14 100644
+--- a/arch/x86/crypto/sha256-ssse3-asm.S
++++ b/arch/x86/crypto/sha256-ssse3-asm.S
+@@ -47,6 +47,7 @@
+ ########################################################################
+
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+
+ ## assume buffers not aligned
+ #define MOVDQ movdqu
+@@ -471,6 +472,7 @@ done_hash:
+ popq %rbp
+ popq %rbx
+
++ pax_force_retaddr 0, 1
+ ret
+ ENDPROC(sha256_transform_ssse3)
+
+diff --git a/arch/x86/crypto/sha512-avx-asm.S b/arch/x86/crypto/sha512-avx-asm.S
+index 974dde9..4533d34 100644
+--- a/arch/x86/crypto/sha512-avx-asm.S
++++ b/arch/x86/crypto/sha512-avx-asm.S
+@@ -49,6 +49,7 @@
+
+ #ifdef CONFIG_AS_AVX
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+
+ .text
+
+@@ -364,6 +365,7 @@ updateblock:
+ mov frame_RSPSAVE(%rsp), %rsp
+
+ nowork:
++ pax_force_retaddr 0, 1
+ ret
+ ENDPROC(sha512_transform_avx)
+
+diff --git a/arch/x86/crypto/sha512-avx2-asm.S b/arch/x86/crypto/sha512-avx2-asm.S
+index 568b961..061ef1d 100644
+--- a/arch/x86/crypto/sha512-avx2-asm.S
++++ b/arch/x86/crypto/sha512-avx2-asm.S
+@@ -51,6 +51,7 @@
+
+ #ifdef CONFIG_AS_AVX2
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+
+ .text
+
+@@ -678,6 +679,7 @@ done_hash:
+
+ # Restore Stack Pointer
+ mov frame_RSPSAVE(%rsp), %rsp
++ pax_force_retaddr 0, 1
+ ret
+ ENDPROC(sha512_transform_rorx)
+
+diff --git a/arch/x86/crypto/sha512-ssse3-asm.S b/arch/x86/crypto/sha512-ssse3-asm.S
+index fb56855..e23914f 100644
+--- a/arch/x86/crypto/sha512-ssse3-asm.S
++++ b/arch/x86/crypto/sha512-ssse3-asm.S
+@@ -48,6 +48,7 @@
+ ########################################################################
+
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+
+ .text
+
+@@ -363,6 +364,7 @@ updateblock:
+ mov frame_RSPSAVE(%rsp), %rsp
+
+ nowork:
++ pax_force_retaddr 0, 1
+ ret
+ ENDPROC(sha512_transform_ssse3)
+
diff --git a/arch/x86/crypto/twofish-avx-x86_64-asm_64.S b/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
index 0505813..63b1d00 100644
--- a/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
+ pax_force_retaddr 0, 1
ret;
ENDPROC(twofish_xts_dec_8way)
+diff --git a/arch/x86/crypto/twofish-avx2-asm_64.S b/arch/x86/crypto/twofish-avx2-asm_64.S
+index e1a83b9..33006b9 100644
+--- a/arch/x86/crypto/twofish-avx2-asm_64.S
++++ b/arch/x86/crypto/twofish-avx2-asm_64.S
+@@ -11,6 +11,7 @@
+ */
+
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+ #include "glue_helper-asm-avx2.S"
+
+ .file "twofish-avx2-asm_64.S"
+@@ -422,6 +423,7 @@ __twofish_enc_blk16:
+ outunpack_enc16(RA, RB, RC, RD);
+ write_blocks16(RA, RB, RC, RD);
+
++ pax_force_retaddr_bts
+ ret;
+ ENDPROC(__twofish_enc_blk16)
+
+@@ -454,6 +456,7 @@ __twofish_dec_blk16:
+ outunpack_dec16(RA, RB, RC, RD);
+ write_blocks16(RA, RB, RC, RD);
+
++ pax_force_retaddr_bts
+ ret;
+ ENDPROC(__twofish_dec_blk16)
+
+@@ -476,6 +479,7 @@ ENTRY(twofish_ecb_enc_16way)
+ popq %r12;
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(twofish_ecb_enc_16way)
+
+@@ -498,6 +502,7 @@ ENTRY(twofish_ecb_dec_16way)
+ popq %r12;
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(twofish_ecb_dec_16way)
+
+@@ -521,6 +526,7 @@ ENTRY(twofish_cbc_dec_16way)
+ popq %r12;
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(twofish_cbc_dec_16way)
+
+@@ -546,6 +552,7 @@ ENTRY(twofish_ctr_16way)
+ popq %r12;
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(twofish_ctr_16way)
+
+@@ -574,6 +581,7 @@ twofish_xts_crypt_16way:
+ popq %r12;
+ vzeroupper;
+
++ pax_force_retaddr 0, 1
+ ret;
+ ENDPROC(twofish_xts_crypt_16way)
+
diff --git a/arch/x86/crypto/twofish-x86_64-asm_64-3way.S b/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
index 1c3b7ce..b365c5e 100644
--- a/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
set_fs(KERNEL_DS);
has_dumped = 1;
diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
-index cf1a471..3bc4cf8 100644
+index cf1a471..5ba2673 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -340,7 +340,7 @@ static void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs,
};
frame = get_sigframe(ksig, regs, sizeof(*frame), &fpstate);
-@@ -463,16 +463,18 @@ int ia32_setup_rt_frame(int sig, struct ksignal *ksig,
+@@ -459,20 +459,22 @@ int ia32_setup_rt_frame(int sig, struct ksignal *ksig,
+ else
+ put_user_ex(0, &frame->uc.uc_flags);
+ put_user_ex(0, &frame->uc.uc_link);
+- err |= __compat_save_altstack(&frame->uc.uc_stack, regs->sp);
++ __compat_save_altstack_ex(&frame->uc.uc_stack, regs->sp);
if (ksig->ka.sa.sa_flags & SA_RESTORER)
restorer = ksig->ka.sa.sa_restorer;
err |= copy_siginfo_to_user32(&frame->info, &ksig->info);
diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
-index 474dc1b..24aaa3e 100644
+index 474dc1b..9297c58 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -15,8 +15,10 @@
movl %ebp,%ebp /* zero extension */
pushq_cfi $__USER32_DS
/*CFI_REL_OFFSET ss,0*/
-@@ -135,24 +157,44 @@ ENTRY(ia32_sysenter_target)
+@@ -135,24 +157,49 @@ ENTRY(ia32_sysenter_target)
CFI_REL_OFFSET rsp,0
pushfq_cfi
/*CFI_REL_OFFSET rflags,0*/
32bit zero extended */
+
+#ifdef CONFIG_PAX_MEMORY_UDEREF
-+ mov pax_user_shadow_base,%r11
-+ add %r11,%rbp
++ addq pax_user_shadow_base,%rbp
++ ASM_PAX_OPEN_USERLAND
+#endif
+
ASM_STAC
ASM_CLAC
- orl $TS_COMPAT,TI_status+THREAD_INFO(%rsp,RIP-ARGOFFSET)
- testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags+THREAD_INFO(%rsp,RIP-ARGOFFSET)
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ ASM_PAX_CLOSE_USERLAND
++#endif
++
+ GET_THREAD_INFO(%r11)
+ orl $TS_COMPAT,TI_status(%r11)
+ testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r11)
CFI_REMEMBER_STATE
jnz sysenter_tracesys
cmpq $(IA32_NR_syscalls-1),%rax
-@@ -162,12 +204,15 @@ sysenter_do_call:
+@@ -162,12 +209,15 @@ sysenter_do_call:
sysenter_dispatch:
call *ia32_sys_call_table(,%rax,8)
movq %rax,RAX-ARGOFFSET(%rsp)
/* clear IF, that popfq doesn't enable interrupts early */
andl $~0x200,EFLAGS-R11(%rsp)
movl RIP-R11(%rsp),%edx /* User %eip */
-@@ -193,6 +238,9 @@ sysexit_from_sys_call:
+@@ -193,6 +243,9 @@ sysexit_from_sys_call:
movl %eax,%esi /* 2nd arg: syscall number */
movl $AUDIT_ARCH_I386,%edi /* 1st arg: audit arch */
call __audit_syscall_entry
movl RAX-ARGOFFSET(%rsp),%eax /* reload syscall number */
cmpq $(IA32_NR_syscalls-1),%rax
ja ia32_badsys
-@@ -204,7 +252,7 @@ sysexit_from_sys_call:
+@@ -204,7 +257,7 @@ sysexit_from_sys_call:
.endm
.macro auditsys_exit exit
jnz ia32_ret_from_sys_call
TRACE_IRQS_ON
ENABLE_INTERRUPTS(CLBR_NONE)
-@@ -215,11 +263,12 @@ sysexit_from_sys_call:
+@@ -215,11 +268,12 @@ sysexit_from_sys_call:
1: setbe %al /* 1 if error, 0 if not */
movzbl %al,%edi /* zero-extend that into %edi */
call __audit_syscall_exit
jz \exit
CLEAR_RREGS -ARGOFFSET
jmp int_with_check
-@@ -237,7 +286,7 @@ sysexit_audit:
+@@ -237,7 +291,7 @@ sysexit_audit:
sysenter_tracesys:
#ifdef CONFIG_AUDITSYSCALL
jz sysenter_auditsys
#endif
SAVE_REST
-@@ -249,6 +298,9 @@ sysenter_tracesys:
+@@ -249,6 +303,9 @@ sysenter_tracesys:
RESTORE_REST
cmpq $(IA32_NR_syscalls-1),%rax
ja int_ret_from_sys_call /* sysenter_tracesys has set RAX(%rsp) */
jmp sysenter_do_call
CFI_ENDPROC
ENDPROC(ia32_sysenter_target)
-@@ -276,19 +328,25 @@ ENDPROC(ia32_sysenter_target)
+@@ -276,19 +333,25 @@ ENDPROC(ia32_sysenter_target)
ENTRY(ia32_cstar_target)
CFI_STARTPROC32 simple
CFI_SIGNAL_FRAME
movl %eax,%eax /* zero extension */
movq %rax,ORIG_RAX-ARGOFFSET(%rsp)
movq %rcx,RIP-ARGOFFSET(%rsp)
-@@ -304,12 +362,19 @@ ENTRY(ia32_cstar_target)
+@@ -304,12 +367,25 @@ ENTRY(ia32_cstar_target)
/* no need to do an access_ok check here because r8 has been
32bit zero extended */
/* hardware stack frame is complete now */
+
+#ifdef CONFIG_PAX_MEMORY_UDEREF
-+ mov pax_user_shadow_base,%r11
-+ add %r11,%r8
++ ASM_PAX_OPEN_USERLAND
++ movq pax_user_shadow_base,%r8
++ addq RSP-ARGOFFSET(%rsp),%r8
+#endif
+
ASM_STAC
ASM_CLAC
- orl $TS_COMPAT,TI_status+THREAD_INFO(%rsp,RIP-ARGOFFSET)
- testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags+THREAD_INFO(%rsp,RIP-ARGOFFSET)
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ ASM_PAX_CLOSE_USERLAND
++#endif
++
+ GET_THREAD_INFO(%r11)
+ orl $TS_COMPAT,TI_status(%r11)
+ testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r11)
CFI_REMEMBER_STATE
jnz cstar_tracesys
cmpq $IA32_NR_syscalls-1,%rax
-@@ -319,12 +384,15 @@ cstar_do_call:
+@@ -319,12 +395,15 @@ cstar_do_call:
cstar_dispatch:
call *ia32_sys_call_table(,%rax,8)
movq %rax,RAX-ARGOFFSET(%rsp)
RESTORE_ARGS 0,-ARG_SKIP,0,0,0
movl RIP-ARGOFFSET(%rsp),%ecx
CFI_REGISTER rip,rcx
-@@ -352,7 +420,7 @@ sysretl_audit:
+@@ -352,7 +431,7 @@ sysretl_audit:
cstar_tracesys:
#ifdef CONFIG_AUDITSYSCALL
jz cstar_auditsys
#endif
xchgl %r9d,%ebp
-@@ -366,6 +434,9 @@ cstar_tracesys:
+@@ -366,11 +445,19 @@ cstar_tracesys:
xchgl %ebp,%r9d
cmpq $(IA32_NR_syscalls-1),%rax
ja int_ret_from_sys_call /* cstar_tracesys has set RAX(%rsp) */
jmp cstar_do_call
END(ia32_cstar_target)
-@@ -407,19 +478,26 @@ ENTRY(ia32_syscall)
+ ia32_badarg:
+ ASM_CLAC
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ ASM_PAX_CLOSE_USERLAND
++#endif
++
+ movq $-EFAULT,%rax
+ jmp ia32_sysret
+ CFI_ENDPROC
+@@ -407,19 +494,26 @@ ENTRY(ia32_syscall)
CFI_REL_OFFSET rip,RIP-RIP
PARAVIRT_ADJUST_EXCEPTION_FRAME
SWAPGS
jnz ia32_tracesys
cmpq $(IA32_NR_syscalls-1),%rax
ja ia32_badsys
-@@ -442,6 +520,9 @@ ia32_tracesys:
+@@ -442,6 +536,9 @@ ia32_tracesys:
RESTORE_REST
cmpq $(IA32_NR_syscalls-1),%rax
ja int_ret_from_sys_call /* ia32_tracesys has set RAX(%rsp) */
struct compat_timespec {
compat_time_t tv_sec;
diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h
-index e99ac27..e89e28c 100644
+index e99ac27..10d834e 100644
--- a/arch/x86/include/asm/cpufeature.h
+++ b/arch/x86/include/asm/cpufeature.h
+@@ -203,7 +203,7 @@
+ #define X86_FEATURE_DECODEASSISTS (8*32+12) /* AMD Decode Assists support */
+ #define X86_FEATURE_PAUSEFILTER (8*32+13) /* AMD filtered pause intercept */
+ #define X86_FEATURE_PFTHRESHOLD (8*32+14) /* AMD pause filter threshold */
+-
++#define X86_FEATURE_STRONGUDEREF (8*32+31) /* PaX PCID based strong UDEREF */
+
+ /* Intel-defined CPU features, CPUID level 0x00000007:0 (ebx), word 9 */
+ #define X86_FEATURE_FSGSBASE (9*32+ 0) /* {RD/WR}{FS/GS}BASE instructions*/
@@ -211,7 +211,7 @@
#define X86_FEATURE_BMI1 (9*32+ 3) /* 1st group bit manipulation extensions */
#define X86_FEATURE_HLE (9*32+ 4) /* Hardware Lock Elision */
#define X86_FEATURE_BMI2 (9*32+ 8) /* 2nd group bit manipulation extensions */
#define X86_FEATURE_ERMS (9*32+ 9) /* Enhanced REP MOVSB/STOSB */
#define X86_FEATURE_INVPCID (9*32+10) /* Invalidate Processor Context ID */
-@@ -394,7 +394,7 @@ static __always_inline __pure bool __static_cpu_has(u16 bit)
+@@ -353,6 +353,7 @@ extern const char * const x86_power_flags[32];
+ #undef cpu_has_centaur_mcr
+ #define cpu_has_centaur_mcr 0
+
++#define cpu_has_pcid boot_cpu_has(X86_FEATURE_PCID)
+ #endif /* CONFIG_X86_64 */
+
+ #if __GNUC__ >= 4
+@@ -394,7 +395,7 @@ static __always_inline __pure bool __static_cpu_has(u16 bit)
".section .discard,\"aw\",@progbits\n"
" .byte 0xff + (4f-3f) - (2b-1b)\n" /* size check */
".previous\n"
#endif /* _ASM_X86_EMERGENCY_RESTART_H */
diff --git a/arch/x86/include/asm/fpu-internal.h b/arch/x86/include/asm/fpu-internal.h
-index e25cc33..425d099 100644
+index e25cc33..7d3ec01 100644
--- a/arch/x86/include/asm/fpu-internal.h
+++ b/arch/x86/include/asm/fpu-internal.h
-@@ -127,7 +127,9 @@ static inline void sanitize_i387_state(struct task_struct *tsk)
+@@ -126,8 +126,11 @@ static inline void sanitize_i387_state(struct task_struct *tsk)
+ #define user_insn(insn, output, input...) \
({ \
int err; \
++ pax_open_userland(); \
asm volatile(ASM_STAC "\n" \
- "1:" #insn "\n\t" \
+ "1:" \
"2: " ASM_CLAC "\n" \
".section .fixup,\"ax\"\n" \
"3: movl $-1,%[err]\n" \
-@@ -300,7 +302,7 @@ static inline int restore_fpu_checking(struct task_struct *tsk)
+@@ -136,6 +139,7 @@ static inline void sanitize_i387_state(struct task_struct *tsk)
+ _ASM_EXTABLE(1b, 3b) \
+ : [err] "=r" (err), output \
+ : "0"(0), input); \
++ pax_close_userland(); \
+ err; \
+ })
+
+@@ -300,7 +304,7 @@ static inline int restore_fpu_checking(struct task_struct *tsk)
"emms\n\t" /* clear stack tags */
"fildl %P[addr]", /* set F?P to defined value */
X86_FEATURE_FXSAVE_LEAK,
return fpu_restore_checking(&tsk->thread.fpu);
}
diff --git a/arch/x86/include/asm/futex.h b/arch/x86/include/asm/futex.h
-index be27ba1..8f13ff9 100644
+index be27ba1..04a8801 100644
--- a/arch/x86/include/asm/futex.h
+++ b/arch/x86/include/asm/futex.h
@@ -12,6 +12,7 @@
: "r" (oparg), "i" (-EFAULT), "1" (0))
static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
-@@ -59,10 +61,10 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
+@@ -57,12 +59,13 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
+
+ pagefault_disable();
++ pax_open_userland();
switch (op) {
case FUTEX_OP_SET:
- __futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg);
uaddr, oparg);
break;
case FUTEX_OP_OR:
-@@ -116,14 +118,14 @@ static inline int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
+@@ -77,6 +80,7 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
+ default:
+ ret = -ENOSYS;
+ }
++ pax_close_userland();
+
+ pagefault_enable();
+
+@@ -115,18 +119,20 @@ static inline int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
+ if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))
return -EFAULT;
++ pax_open_userland();
asm volatile("\t" ASM_STAC "\n"
- "1:\t" LOCK_PREFIX "cmpxchgl %4, %2\n"
+ "1:\t" LOCK_PREFIX __copyuser_seg"cmpxchgl %4, %2\n"
: "i" (-EFAULT), "r" (newval), "1" (oldval)
: "memory"
);
++ pax_close_userland();
+
+ *uval = oldval;
+ return ret;
diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h
index 1da97ef..9c2ebff 100644
--- a/arch/x86/include/asm/hw_irq.h
#ifdef CONFIG_SMP
diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
-index cdbf367..adb37ac 100644
+index cdbf367..4c73c9e 100644
--- a/arch/x86/include/asm/mmu_context.h
+++ b/arch/x86/include/asm/mmu_context.h
-@@ -24,6 +24,18 @@ void destroy_context(struct mm_struct *mm);
+@@ -24,6 +24,20 @@ void destroy_context(struct mm_struct *mm);
static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
{
+
+#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
-+ unsigned int i;
-+ pgd_t *pgd;
++ if (!(static_cpu_has(X86_FEATURE_PCID))) {
++ unsigned int i;
++ pgd_t *pgd;
+
-+ pax_open_kernel();
-+ pgd = get_cpu_pgd(smp_processor_id());
-+ for (i = USER_PGD_PTRS; i < 2 * USER_PGD_PTRS; ++i)
-+ set_pgd_batched(pgd+i, native_make_pgd(0));
-+ pax_close_kernel();
++ pax_open_kernel();
++ pgd = get_cpu_pgd(smp_processor_id(), kernel);
++ for (i = USER_PGD_PTRS; i < 2 * USER_PGD_PTRS; ++i)
++ set_pgd_batched(pgd+i, native_make_pgd(0));
++ pax_close_kernel();
++ }
+#endif
+
#ifdef CONFIG_SMP
if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
this_cpu_write(cpu_tlbstate.state, TLBSTATE_LAZY);
-@@ -34,16 +46,30 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+@@ -34,16 +48,55 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
struct task_struct *tsk)
{
unsigned cpu = smp_processor_id();
/* Re-load page tables */
+#ifdef CONFIG_PAX_PER_CPU_PGD
+ pax_open_kernel();
-+ __clone_user_pgds(get_cpu_pgd(cpu), next->pgd);
-+ __shadow_user_pgds(get_cpu_pgd(cpu) + USER_PGD_PTRS, next->pgd);
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++ if (static_cpu_has(X86_FEATURE_PCID))
++ __clone_user_pgds(get_cpu_pgd(cpu, user), next->pgd);
++ else
++#endif
++
++ __clone_user_pgds(get_cpu_pgd(cpu, kernel), next->pgd);
++ __shadow_user_pgds(get_cpu_pgd(cpu, kernel) + USER_PGD_PTRS, next->pgd);
+ pax_close_kernel();
-+ load_cr3(get_cpu_pgd(cpu));
++ BUG_ON((__pa(get_cpu_pgd(cpu, kernel)) | PCID_KERNEL) != (read_cr3() & __PHYSICAL_MASK) && (__pa(get_cpu_pgd(cpu, user)) | PCID_USER) != (read_cr3() & __PHYSICAL_MASK));
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++ if (static_cpu_has(X86_FEATURE_PCID)) {
++ if (static_cpu_has(X86_FEATURE_INVPCID)) {
++ unsigned long descriptor[2];
++ descriptor[0] = PCID_USER;
++ asm volatile(__ASM_INVPCID : : "d"(&descriptor), "a"(INVPCID_SINGLE_CONTEXT) : "memory");
++ } else {
++ write_cr3(__pa(get_cpu_pgd(cpu, user)) | PCID_USER);
++ if (static_cpu_has(X86_FEATURE_STRONGUDEREF))
++ write_cr3(__pa(get_cpu_pgd(cpu, kernel)) | PCID_KERNEL | PCID_NOFLUSH);
++ else
++ write_cr3(__pa(get_cpu_pgd(cpu, kernel)) | PCID_KERNEL);
++ }
++ } else
++#endif
++
++ load_cr3(get_cpu_pgd(cpu, kernel));
+#else
load_cr3(next->pgd);
+#endif
/* stop flush ipis for the previous mm */
cpumask_clear_cpu(cpu, mm_cpumask(prev));
-@@ -53,9 +79,38 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+@@ -53,9 +106,63 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
*/
if (unlikely(prev->context.ldt != next->context.ldt))
load_LDT_nolock(&next->context);
+
+#ifdef CONFIG_PAX_PER_CPU_PGD
+ pax_open_kernel();
-+ __clone_user_pgds(get_cpu_pgd(cpu), next->pgd);
-+ __shadow_user_pgds(get_cpu_pgd(cpu) + USER_PGD_PTRS, next->pgd);
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++ if (static_cpu_has(X86_FEATURE_PCID))
++ __clone_user_pgds(get_cpu_pgd(cpu, user), next->pgd);
++ else
++#endif
++
++ __clone_user_pgds(get_cpu_pgd(cpu, kernel), next->pgd);
++ __shadow_user_pgds(get_cpu_pgd(cpu, kernel) + USER_PGD_PTRS, next->pgd);
+ pax_close_kernel();
-+ load_cr3(get_cpu_pgd(cpu));
++ BUG_ON((__pa(get_cpu_pgd(cpu, kernel)) | PCID_KERNEL) != (read_cr3() & __PHYSICAL_MASK) && (__pa(get_cpu_pgd(cpu, user)) | PCID_USER) != (read_cr3() & __PHYSICAL_MASK));
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++ if (static_cpu_has(X86_FEATURE_PCID)) {
++ if (static_cpu_has(X86_FEATURE_INVPCID)) {
++ unsigned long descriptor[2];
++ descriptor[0] = PCID_USER;
++ asm volatile(__ASM_INVPCID : : "d"(&descriptor), "a"(INVPCID_SINGLE_CONTEXT) : "memory");
++ } else {
++ write_cr3(__pa(get_cpu_pgd(cpu, user)) | PCID_USER);
++ if (static_cpu_has(X86_FEATURE_STRONGUDEREF))
++ write_cr3(__pa(get_cpu_pgd(cpu, kernel)) | PCID_KERNEL | PCID_NOFLUSH);
++ else
++ write_cr3(__pa(get_cpu_pgd(cpu, kernel)) | PCID_KERNEL);
++ }
++ } else
++#endif
++
++ load_cr3(get_cpu_pgd(cpu, kernel));
+#endif
+
+#ifdef CONFIG_SMP
this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK);
BUG_ON(this_cpu_read(cpu_tlbstate.active_mm) != next);
-@@ -64,11 +119,28 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+@@ -64,11 +171,28 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
* tlb flush IPI delivery. We must reload CR3
* to make sure to use no freed page tables.
*/
/*
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
-index 1e67223..dd6e7ea 100644
+index 1e67223..92a9585 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -44,6 +44,7 @@ extern struct mm_struct *pgd_page_get_mm(struct page *page);
}
static inline pte_t pte_mkdirty(pte_t pte)
-@@ -394,6 +459,15 @@ pte_t *populate_extra_pte(unsigned long vaddr);
+@@ -394,6 +459,16 @@ pte_t *populate_extra_pte(unsigned long vaddr);
#endif
#ifndef __ASSEMBLY__
+
+#ifdef CONFIG_PAX_PER_CPU_PGD
-+extern pgd_t cpu_pgd[NR_CPUS][PTRS_PER_PGD];
-+static inline pgd_t *get_cpu_pgd(unsigned int cpu)
++extern pgd_t cpu_pgd[NR_CPUS][2][PTRS_PER_PGD];
++enum cpu_pgd_type {kernel = 0, user = 1};
++static inline pgd_t *get_cpu_pgd(unsigned int cpu, enum cpu_pgd_type type)
+{
-+ return cpu_pgd[cpu];
++ return cpu_pgd[cpu][type];
+}
+#endif
+
#include <linux/mm_types.h>
#include <linux/log2.h>
-@@ -529,7 +603,7 @@ static inline unsigned long pud_page_vaddr(pud_t pud)
+@@ -529,7 +604,7 @@ static inline unsigned long pud_page_vaddr(pud_t pud)
* Currently stuck as a macro due to indirect forward reference to
* linux/mmzone.h's __section_mem_map_addr() definition:
*/
/* Find an entry in the second-level page table.. */
static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address)
-@@ -569,7 +643,7 @@ static inline unsigned long pgd_page_vaddr(pgd_t pgd)
+@@ -569,7 +644,7 @@ static inline unsigned long pgd_page_vaddr(pgd_t pgd)
* Currently stuck as a macro due to indirect forward reference to
* linux/mmzone.h's __section_mem_map_addr() definition:
*/
/* to find an entry in a page-table-directory. */
static inline unsigned long pud_index(unsigned long address)
-@@ -584,7 +658,7 @@ static inline pud_t *pud_offset(pgd_t *pgd, unsigned long address)
+@@ -584,7 +659,7 @@ static inline pud_t *pud_offset(pgd_t *pgd, unsigned long address)
static inline int pgd_bad(pgd_t pgd)
{
}
static inline int pgd_none(pgd_t pgd)
-@@ -607,7 +681,12 @@ static inline int pgd_none(pgd_t pgd)
+@@ -607,7 +682,12 @@ static inline int pgd_none(pgd_t pgd)
* pgd_offset() returns a (pgd_t *)
* pgd_index() is used get the offset into the pgd page's array of pgd_t's;
*/
+#define pgd_offset(mm, address) ((mm)->pgd + pgd_index(address))
+
+#ifdef CONFIG_PAX_PER_CPU_PGD
-+#define pgd_offset_cpu(cpu, address) (get_cpu_pgd(cpu) + pgd_index(address))
++#define pgd_offset_cpu(cpu, type, address) (get_cpu_pgd(cpu, type) + pgd_index(address))
+#endif
+
/*
* a shortcut which implies the use of the kernel's pgd, instead
* of a process's
-@@ -618,6 +697,22 @@ static inline int pgd_none(pgd_t pgd)
+@@ -618,6 +698,23 @@ static inline int pgd_none(pgd_t pgd)
#define KERNEL_PGD_BOUNDARY pgd_index(PAGE_OFFSET)
#define KERNEL_PGD_PTRS (PTRS_PER_PGD - KERNEL_PGD_BOUNDARY)
+#define pax_user_shadow_base pax_user_shadow_base(%rip)
+#else
+extern unsigned long pax_user_shadow_base;
++extern pgdval_t clone_pgd_mask;
+#endif
+#endif
+
#ifndef __ASSEMBLY__
extern int direct_gbpages;
-@@ -784,11 +879,24 @@ static inline void pmdp_set_wrprotect(struct mm_struct *mm,
+@@ -784,11 +881,24 @@ static inline void pmdp_set_wrprotect(struct mm_struct *mm,
* dst and src can be on the same page, but the range must not overlap,
* and must not cross a page boundary.
*/
#define pgprot_writecombine pgprot_writecombine
extern pgprot_t pgprot_writecombine(pgprot_t prot);
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
-index 22224b3..30c80ca 100644
+index 22224b3..b3a2f90 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
-@@ -282,7 +282,7 @@ struct tss_struct {
+@@ -198,9 +198,21 @@ static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
+ : "memory");
+ }
+
++/* invpcid (%rdx),%rax */
++#define __ASM_INVPCID ".byte 0x66,0x0f,0x38,0x82,0x02"
++
++#define INVPCID_SINGLE_ADDRESS 0UL
++#define INVPCID_SINGLE_CONTEXT 1UL
++#define INVPCID_ALL_GLOBAL 2UL
++#define INVPCID_ALL_MONGLOBAL 3UL
++
++#define PCID_KERNEL 0UL
++#define PCID_USER 1UL
++#define PCID_NOFLUSH (1UL << 63)
++
+ static inline void load_cr3(pgd_t *pgdir)
+ {
+- write_cr3(__pa(pgdir));
++ write_cr3(__pa(pgdir) | PCID_KERNEL);
+ }
+
+ #ifdef CONFIG_X86_32
+@@ -282,7 +294,7 @@ struct tss_struct {
} ____cacheline_aligned;
/*
* Save the original ist values for checking stack pointers during debugging
-@@ -823,11 +823,18 @@ static inline void spin_lock_prefetch(const void *x)
+@@ -452,6 +464,7 @@ struct thread_struct {
+ unsigned short ds;
+ unsigned short fsindex;
+ unsigned short gsindex;
++ unsigned short ss;
+ #endif
+ #ifdef CONFIG_X86_32
+ unsigned long ip;
+@@ -552,29 +565,8 @@ static inline void load_sp0(struct tss_struct *tss,
+ extern unsigned long mmu_cr4_features;
+ extern u32 *trampoline_cr4_features;
+
+-static inline void set_in_cr4(unsigned long mask)
+-{
+- unsigned long cr4;
+-
+- mmu_cr4_features |= mask;
+- if (trampoline_cr4_features)
+- *trampoline_cr4_features = mmu_cr4_features;
+- cr4 = read_cr4();
+- cr4 |= mask;
+- write_cr4(cr4);
+-}
+-
+-static inline void clear_in_cr4(unsigned long mask)
+-{
+- unsigned long cr4;
+-
+- mmu_cr4_features &= ~mask;
+- if (trampoline_cr4_features)
+- *trampoline_cr4_features = mmu_cr4_features;
+- cr4 = read_cr4();
+- cr4 &= ~mask;
+- write_cr4(cr4);
+-}
++extern void set_in_cr4(unsigned long mask);
++extern void clear_in_cr4(unsigned long mask);
+
+ typedef struct {
+ unsigned long seg;
+@@ -823,11 +815,18 @@ static inline void spin_lock_prefetch(const void *x)
*/
#define TASK_SIZE PAGE_OFFSET
#define TASK_SIZE_MAX TASK_SIZE
.vm86_info = NULL, \
.sysenter_cs = __KERNEL_CS, \
.io_bitmap_ptr = NULL, \
-@@ -841,7 +848,7 @@ static inline void spin_lock_prefetch(const void *x)
+@@ -841,7 +840,7 @@ static inline void spin_lock_prefetch(const void *x)
*/
#define INIT_TSS { \
.x86_tss = { \
.ss0 = __KERNEL_DS, \
.ss1 = __KERNEL_CS, \
.io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \
-@@ -852,11 +859,7 @@ static inline void spin_lock_prefetch(const void *x)
+@@ -852,11 +851,7 @@ static inline void spin_lock_prefetch(const void *x)
extern unsigned long thread_saved_pc(struct task_struct *tsk);
#define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long))
/*
* The below -8 is to reserve 8 bytes on top of the ring0 stack.
-@@ -871,7 +874,7 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
+@@ -871,7 +866,7 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
#define task_pt_regs(task) \
({ \
struct pt_regs *__regs__; \
__regs__ - 1; \
})
-@@ -881,13 +884,13 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
+@@ -881,13 +876,13 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
/*
* User space process size. 47bits minus one guard page.
*/
#define TASK_SIZE (test_thread_flag(TIF_ADDR32) ? \
IA32_PAGE_OFFSET : TASK_SIZE_MAX)
-@@ -898,11 +901,11 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
+@@ -898,11 +893,11 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
#define STACK_TOP_MAX TASK_SIZE_MAX
#define INIT_THREAD { \
}
/*
-@@ -930,6 +933,10 @@ extern void start_thread(struct pt_regs *regs, unsigned long new_ip,
+@@ -930,6 +925,10 @@ extern void start_thread(struct pt_regs *regs, unsigned long new_ip,
*/
#define TASK_UNMAPPED_BASE (PAGE_ALIGN(TASK_SIZE / 3))
#define KSTK_EIP(task) (task_pt_regs(task)->ip)
/* Get/set a process' ability to use the timestamp counter instruction */
-@@ -942,7 +949,8 @@ extern int set_tsc_mode(unsigned int val);
+@@ -942,7 +941,8 @@ extern int set_tsc_mode(unsigned int val);
extern u16 amd_get_nb_id(int cpu);
struct aperfmperf {
};
static inline void get_aperfmperf(struct aperfmperf *am)
-@@ -970,7 +978,7 @@ unsigned long calc_aperfmperf_ratio(struct aperfmperf *old,
+@@ -970,7 +970,7 @@ unsigned long calc_aperfmperf_ratio(struct aperfmperf *old,
return ratio;
}
extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
void default_idle(void);
-@@ -980,6 +988,6 @@ bool xen_set_default_idle(void);
+@@ -980,6 +980,6 @@ bool xen_set_default_idle(void);
#define xen_set_default_idle 0
#endif
#endif /* __KERNEL__ */
diff --git a/arch/x86/include/asm/segment.h b/arch/x86/include/asm/segment.h
-index c48a950..c6d7468 100644
+index c48a950..bc40804 100644
--- a/arch/x86/include/asm/segment.h
+++ b/arch/x86/include/asm/segment.h
@@ -64,10 +64,15 @@
#define GDT_ENTRY_TSS 8 /* needs two entries */
#define GDT_ENTRY_LDT 10 /* needs two entries */
#define GDT_ENTRY_TLS_MIN 12
-@@ -185,6 +200,7 @@
+@@ -173,6 +188,8 @@
+ #define GDT_ENTRY_PER_CPU 15 /* Abused to load per CPU data from limit */
+ #define __PER_CPU_SEG (GDT_ENTRY_PER_CPU * 8 + 3)
+
++#define GDT_ENTRY_UDEREF_KERNEL_DS 16
++
+ /* TLS indexes for 64bit - hardcoded in arch_prctl */
+ #define FS_TLS 0
+ #define GS_TLS 1
+@@ -180,12 +197,14 @@
+ #define GS_TLS_SEL ((GDT_ENTRY_TLS_MIN+GS_TLS)*8 + 3)
+ #define FS_TLS_SEL ((GDT_ENTRY_TLS_MIN+FS_TLS)*8 + 3)
+
+-#define GDT_ENTRIES 16
++#define GDT_ENTRIES 17
+
#endif
#define __KERNEL_CS (GDT_ENTRY_KERNEL_CS*8)
+#define __KERNEXEC_KERNEL_CS (GDT_ENTRY_KERNEXEC_KERNEL_CS*8)
#define __KERNEL_DS (GDT_ENTRY_KERNEL_DS*8)
++#define __UDEREF_KERNEL_DS (GDT_ENTRY_UDEREF_KERNEL_DS*8)
#define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS*8+3)
#define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS*8+3)
-@@ -265,7 +281,7 @@ static inline unsigned long get_limit(unsigned long segment)
+ #ifndef CONFIG_PARAVIRT
+@@ -265,7 +284,7 @@ static inline unsigned long get_limit(unsigned long segment)
{
unsigned long __limit;
asm("lsll %1,%0" : "=r" (__limit) : "r" (segment));
}
#endif /* !__ASSEMBLY__ */
+diff --git a/arch/x86/include/asm/smap.h b/arch/x86/include/asm/smap.h
+index 8d3120f..352b440 100644
+--- a/arch/x86/include/asm/smap.h
++++ b/arch/x86/include/asm/smap.h
+@@ -25,11 +25,40 @@
+
+ #include <asm/alternative-asm.h>
+
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++#define ASM_PAX_OPEN_USERLAND \
++ 661: jmp 663f; \
++ .pushsection .altinstr_replacement, "a" ; \
++ 662: pushq %rax; nop; \
++ .popsection ; \
++ .pushsection .altinstructions, "a" ; \
++ altinstruction_entry 661b, 662b, X86_FEATURE_STRONGUDEREF, 2, 2;\
++ .popsection ; \
++ call __pax_open_userland; \
++ popq %rax; \
++ 663:
++
++#define ASM_PAX_CLOSE_USERLAND \
++ 661: jmp 663f; \
++ .pushsection .altinstr_replacement, "a" ; \
++ 662: pushq %rax; nop; \
++ .popsection; \
++ .pushsection .altinstructions, "a" ; \
++ altinstruction_entry 661b, 662b, X86_FEATURE_STRONGUDEREF, 2, 2;\
++ .popsection; \
++ call __pax_close_userland; \
++ popq %rax; \
++ 663:
++#else
++#define ASM_PAX_OPEN_USERLAND
++#define ASM_PAX_CLOSE_USERLAND
++#endif
++
+ #ifdef CONFIG_X86_SMAP
+
+ #define ASM_CLAC \
+ 661: ASM_NOP3 ; \
+- .pushsection .altinstr_replacement, "ax" ; \
++ .pushsection .altinstr_replacement, "a" ; \
+ 662: __ASM_CLAC ; \
+ .popsection ; \
+ .pushsection .altinstructions, "a" ; \
+@@ -38,7 +67,7 @@
+
+ #define ASM_STAC \
+ 661: ASM_NOP3 ; \
+- .pushsection .altinstr_replacement, "ax" ; \
++ .pushsection .altinstr_replacement, "a" ; \
+ 662: __ASM_STAC ; \
+ .popsection ; \
+ .pushsection .altinstructions, "a" ; \
+@@ -56,6 +85,37 @@
+
+ #include <asm/alternative.h>
+
++#define __HAVE_ARCH_PAX_OPEN_USERLAND
++#define __HAVE_ARCH_PAX_CLOSE_USERLAND
++
++extern void __pax_open_userland(void);
++static __always_inline unsigned long pax_open_userland(void)
++{
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++ asm volatile(ALTERNATIVE(ASM_NOP5, "call %P[open]", X86_FEATURE_STRONGUDEREF)
++ :
++ : [open] "i" (__pax_open_userland)
++ : "memory", "rax");
++#endif
++
++ return 0;
++}
++
++extern void __pax_close_userland(void);
++static __always_inline unsigned long pax_close_userland(void)
++{
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++ asm volatile(ALTERNATIVE(ASM_NOP5, "call %P[close]", X86_FEATURE_STRONGUDEREF)
++ :
++ : [close] "i" (__pax_close_userland)
++ : "memory", "rax");
++#endif
++
++ return 0;
++}
++
+ #ifdef CONFIG_X86_SMAP
+
+ static __always_inline void clac(void)
diff --git a/arch/x86/include/asm/smp.h b/arch/x86/include/asm/smp.h
index b073aae..39f9bdd 100644
--- a/arch/x86/include/asm/smp.h
+
#endif
#endif /* _ASM_X86_THREAD_INFO_H */
+diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
+index 50a7fc0..7c437a7 100644
+--- a/arch/x86/include/asm/tlbflush.h
++++ b/arch/x86/include/asm/tlbflush.h
+@@ -17,18 +17,40 @@
+
+ static inline void __native_flush_tlb(void)
+ {
++ if (static_cpu_has(X86_FEATURE_INVPCID)) {
++ unsigned long descriptor[2];
++ asm volatile(__ASM_INVPCID : : "d"(&descriptor), "a"(INVPCID_ALL_MONGLOBAL) : "memory");
++ return;
++ }
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++ if (static_cpu_has(X86_FEATURE_PCID)) {
++ unsigned int cpu = raw_get_cpu();
++
++ native_write_cr3(__pa(get_cpu_pgd(cpu, user)) | PCID_USER);
++ native_write_cr3(__pa(get_cpu_pgd(cpu, kernel)) | PCID_KERNEL);
++ raw_put_cpu_no_resched();
++ return;
++ }
++#endif
++
+ native_write_cr3(native_read_cr3());
+ }
+
+ static inline void __native_flush_tlb_global_irq_disabled(void)
+ {
+- unsigned long cr4;
++ if (static_cpu_has(X86_FEATURE_INVPCID)) {
++ unsigned long descriptor[2];
++ asm volatile(__ASM_INVPCID : : "d"(&descriptor), "a"(INVPCID_ALL_GLOBAL) : "memory");
++ } else {
++ unsigned long cr4;
+
+- cr4 = native_read_cr4();
+- /* clear PGE */
+- native_write_cr4(cr4 & ~X86_CR4_PGE);
+- /* write old PGE again and flush TLBs */
+- native_write_cr4(cr4);
++ cr4 = native_read_cr4();
++ /* clear PGE */
++ native_write_cr4(cr4 & ~X86_CR4_PGE);
++ /* write old PGE again and flush TLBs */
++ native_write_cr4(cr4);
++ }
+ }
+
+ static inline void __native_flush_tlb_global(void)
+@@ -49,6 +71,42 @@ static inline void __native_flush_tlb_global(void)
+
+ static inline void __native_flush_tlb_single(unsigned long addr)
+ {
++
++ if (static_cpu_has(X86_FEATURE_INVPCID)) {
++ unsigned long descriptor[2];
++
++ descriptor[0] = PCID_KERNEL;
++ descriptor[1] = addr;
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++ if (!static_cpu_has(X86_FEATURE_STRONGUDEREF) || addr >= TASK_SIZE_MAX) {
++ if (addr < TASK_SIZE_MAX)
++ descriptor[1] += pax_user_shadow_base;
++ asm volatile(__ASM_INVPCID : : "d"(&descriptor), "a"(INVPCID_SINGLE_ADDRESS) : "memory");
++ }
++
++ descriptor[0] = PCID_USER;
++ descriptor[1] = addr;
++#endif
++
++ asm volatile(__ASM_INVPCID : : "d"(&descriptor), "a"(INVPCID_SINGLE_ADDRESS) : "memory");
++ return;
++ }
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++ if (static_cpu_has(X86_FEATURE_PCID)) {
++ unsigned int cpu = raw_get_cpu();
++
++ native_write_cr3(__pa(get_cpu_pgd(cpu, user)) | PCID_USER | PCID_NOFLUSH);
++ asm volatile("invlpg (%0)" ::"r" (addr) : "memory");
++ native_write_cr3(__pa(get_cpu_pgd(cpu, kernel)) | PCID_KERNEL | PCID_NOFLUSH);
++ raw_put_cpu_no_resched();
++
++ if (!static_cpu_has(X86_FEATURE_STRONGUDEREF) && addr < TASK_SIZE_MAX)
++ addr += pax_user_shadow_base;
++ }
++#endif
++
+ asm volatile("invlpg (%0)" ::"r" (addr) : "memory");
+ }
+
diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
-index 5ee2687..70d5895 100644
+index 5ee2687..74590b9 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -7,6 +7,7 @@
/*
* The exception table consists of pairs of addresses relative to the
-@@ -176,13 +207,21 @@ __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL))
+@@ -165,10 +196,12 @@ __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL))
+ register __inttype(*(ptr)) __val_gu asm("%edx"); \
+ __chk_user_ptr(ptr); \
+ might_fault(); \
++ pax_open_userland(); \
+ asm volatile("call __get_user_%P3" \
+ : "=a" (__ret_gu), "=r" (__val_gu) \
+ : "0" (ptr), "i" (sizeof(*(ptr)))); \
+ (x) = (__typeof__(*(ptr))) __val_gu; \
++ pax_close_userland(); \
+ __ret_gu; \
+ })
+
+@@ -176,13 +209,21 @@ __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL))
asm volatile("call __put_user_" #size : "=a" (__ret_pu) \
: "0" ((typeof(*(ptr)))(x)), "c" (ptr) : "ebx")
"3: " ASM_CLAC "\n" \
".section .fixup,\"ax\"\n" \
"4: movl %3,%0\n" \
-@@ -195,8 +234,8 @@ __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL))
+@@ -195,8 +236,8 @@ __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL))
#define __put_user_asm_ex_u64(x, addr) \
asm volatile(ASM_STAC "\n" \
"3: " ASM_CLAC "\n" \
_ASM_EXTABLE_EX(1b, 2b) \
_ASM_EXTABLE_EX(2b, 3b) \
-@@ -246,7 +285,7 @@ extern void __put_user_8(void);
+@@ -246,7 +287,8 @@ extern void __put_user_8(void);
__typeof__(*(ptr)) __pu_val; \
__chk_user_ptr(ptr); \
might_fault(); \
- __pu_val = x; \
+ __pu_val = (x); \
++ pax_open_userland(); \
switch (sizeof(*(ptr))) { \
case 1: \
__put_user_x(1, __pu_val, ptr, __ret_pu); \
-@@ -345,7 +384,7 @@ do { \
+@@ -264,6 +306,7 @@ extern void __put_user_8(void);
+ __put_user_x(X, __pu_val, ptr, __ret_pu); \
+ break; \
+ } \
++ pax_close_userland(); \
+ __ret_pu; \
+ })
+
+@@ -344,8 +387,10 @@ do { \
+ } while (0)
#define __get_user_asm(x, addr, err, itype, rtype, ltype, errret) \
++do { \
++ pax_open_userland(); \
asm volatile(ASM_STAC "\n" \
- "1: mov"itype" %2,%"rtype"1\n" \
+ "1: "__copyuser_seg"mov"itype" %2,%"rtype"1\n"\
"2: " ASM_CLAC "\n" \
".section .fixup,\"ax\"\n" \
"3: mov %3,%0\n" \
-@@ -353,7 +392,7 @@ do { \
+@@ -353,8 +398,10 @@ do { \
" 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))
++ : "m" (__m(addr)), "i" (errret), "0" (err)); \
++ pax_close_userland(); \
++} while (0)
#define __get_user_size_ex(x, ptr, size) \
-@@ -378,7 +417,7 @@ do { \
+ do { \
+@@ -378,7 +425,7 @@ do { \
} while (0)
#define __get_user_asm_ex(x, addr, itype, rtype, ltype) \
"2:\n" \
_ASM_EXTABLE_EX(1b, 2b) \
: ltype(x) : "m" (__m(addr)))
-@@ -395,13 +434,24 @@ do { \
+@@ -395,13 +442,24 @@ do { \
int __gu_err; \
unsigned long __gu_val; \
__get_user_size(__gu_val, (ptr), (size), __gu_err, -EFAULT); \
/*
* Tell gcc we read from memory instead of writing: this is because
-@@ -410,7 +460,7 @@ struct __large_struct { unsigned long buf[100]; };
+@@ -409,8 +467,10 @@ struct __large_struct { unsigned long buf[100]; };
+ * aliasing issues.
*/
#define __put_user_asm(x, addr, err, itype, rtype, ltype, errret) \
++do { \
++ pax_open_userland(); \
asm volatile(ASM_STAC "\n" \
- "1: mov"itype" %"rtype"1,%2\n" \
+ "1: "__copyuser_seg"mov"itype" %"rtype"1,%2\n"\
"2: " ASM_CLAC "\n" \
".section .fixup,\"ax\"\n" \
"3: mov %3,%0\n" \
-@@ -418,10 +468,10 @@ struct __large_struct { unsigned long buf[100]; };
+@@ -418,10 +478,12 @@ struct __large_struct { unsigned long buf[100]; };
".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))
++ : ltype (x), "m" (__m(addr)), "i" (errret), "0" (err));\
++ pax_close_userland(); \
++} while (0)
#define __put_user_asm_ex(x, addr, itype, rtype, ltype) \
- asm volatile("1: mov"itype" %"rtype"0,%1\n" \
"2:\n" \
_ASM_EXTABLE_EX(1b, 2b) \
: : ltype(x), "m" (__m(addr)))
-@@ -460,8 +510,12 @@ struct __large_struct { unsigned long buf[100]; };
+@@ -431,11 +493,13 @@ struct __large_struct { unsigned long buf[100]; };
+ */
+ #define uaccess_try do { \
+ current_thread_info()->uaccess_err = 0; \
++ pax_open_userland(); \
+ stac(); \
+ barrier();
+
+ #define uaccess_catch(err) \
+ clac(); \
++ pax_close_userland(); \
+ (err) |= (current_thread_info()->uaccess_err ? -EFAULT : 0); \
+ } while (0)
+
+@@ -460,8 +524,12 @@ struct __large_struct { unsigned long buf[100]; };
* On error, the variable @x is set to zero.
*/
/**
* __put_user: - Write a simple value into user space, with less checking.
-@@ -483,8 +537,12 @@ struct __large_struct { unsigned long buf[100]; };
+@@ -483,8 +551,12 @@ struct __large_struct { unsigned long buf[100]; };
* Returns zero on success, or -EFAULT on error.
*/
#define __get_user_unaligned __get_user
#define __put_user_unaligned __put_user
-@@ -502,7 +560,7 @@ struct __large_struct { unsigned long buf[100]; };
+@@ -502,7 +574,7 @@ struct __large_struct { unsigned long buf[100]; };
#define get_user_ex(x, ptr) do { \
unsigned long __gue_val; \
__get_user_size_ex((__gue_val), (ptr), (sizeof(*(ptr)))); \
} while (0)
#define put_user_try uaccess_try
-@@ -519,8 +577,8 @@ strncpy_from_user(char *dst, const char __user *src, long count);
+@@ -519,8 +591,8 @@ strncpy_from_user(char *dst, const char __user *src, long count);
extern __must_check long strlen_user(const char __user *str);
extern __must_check long strnlen_user(const char __user *str, long n);
extern struct x86_init_ops x86_init;
extern struct x86_cpuinit_ops x86_cpuinit;
diff --git a/arch/x86/include/asm/xsave.h b/arch/x86/include/asm/xsave.h
-index 0415cda..b43d877 100644
+index 0415cda..3b22adc 100644
--- a/arch/x86/include/asm/xsave.h
+++ b/arch/x86/include/asm/xsave.h
-@@ -71,7 +71,9 @@ static inline int xsave_user(struct xsave_struct __user *buf)
+@@ -70,8 +70,11 @@ static inline int xsave_user(struct xsave_struct __user *buf)
+ if (unlikely(err))
return -EFAULT;
++ pax_open_userland();
__asm__ __volatile__(ASM_STAC "\n"
- "1: .byte " REX_PREFIX "0x0f,0xae,0x27\n"
+ "1:"
"2: " ASM_CLAC "\n"
".section .fixup,\"ax\"\n"
"3: movl $-1,%[err]\n"
-@@ -87,12 +89,14 @@ static inline int xsave_user(struct xsave_struct __user *buf)
+@@ -81,18 +84,22 @@ static inline int xsave_user(struct xsave_struct __user *buf)
+ : [err] "=r" (err)
+ : "D" (buf), "a" (-1), "d" (-1), "0" (0)
+ : "memory");
++ pax_close_userland();
+ return err;
+ }
+
static inline int xrestore_user(struct xsave_struct __user *buf, u64 mask)
{
int err;
u32 lmask = mask;
u32 hmask = mask >> 32;
++ pax_open_userland();
__asm__ __volatile__(ASM_STAC "\n"
- "1: .byte " REX_PREFIX "0x0f,0xae,0x2f\n"
+ "1:"
"2: " ASM_CLAC "\n"
".section .fixup,\"ax\"\n"
"3: movl $-1,%[err]\n"
+@@ -102,6 +109,7 @@ static inline int xrestore_user(struct xsave_struct __user *buf, u64 mask)
+ : [err] "=r" (err)
+ : "D" (xstate), "a" (lmask), "d" (hmask), "0" (0)
+ : "memory"); /* memory required? */
++ pax_close_userland();
+ return err;
+ }
+
diff --git a/arch/x86/include/uapi/asm/e820.h b/arch/x86/include/uapi/asm/e820.h
index bbae024..e1528f9 100644
--- a/arch/x86/include/uapi/asm/e820.h
if (c->x86_model == 3 && c->x86_mask == 0)
size = 64;
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
-index 22018f7..bc6f5e3 100644
+index 22018f7..df77e23 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -88,60 +88,6 @@ static const struct cpu_dev __cpuinitconst default_cpu = {
static int __init x86_xsave_setup(char *s)
{
setup_clear_cpu_cap(X86_FEATURE_XSAVE);
-@@ -386,7 +332,7 @@ void switch_to_new_gdt(int cpu)
+@@ -288,6 +234,57 @@ static __always_inline void setup_smap(struct cpuinfo_x86 *c)
+ set_in_cr4(X86_CR4_SMAP);
+ }
+
++#ifdef CONFIG_X86_64
++static __init int setup_disable_pcid(char *arg)
++{
++ setup_clear_cpu_cap(X86_FEATURE_PCID);
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ if (clone_pgd_mask != ~(pgdval_t)0UL)
++ pax_user_shadow_base = 1UL << TASK_SIZE_MAX_SHIFT;
++#endif
++
++ return 1;
++}
++__setup("nopcid", setup_disable_pcid);
++
++static void setup_pcid(struct cpuinfo_x86 *c)
++{
++ if (!cpu_has(c, X86_FEATURE_PCID)) {
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ if (clone_pgd_mask != ~(pgdval_t)0UL) {
++ pax_open_kernel();
++ pax_user_shadow_base = 1UL << TASK_SIZE_MAX_SHIFT;
++ pax_close_kernel();
++ printk("PAX: slow and weak UDEREF enabled\n");
++ } else
++ printk("PAX: UDEREF disabled\n");
++#endif
++
++ return;
++ }
++
++ printk("PAX: PCID detected\n");
++ set_in_cr4(X86_CR4_PCIDE);
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ pax_open_kernel();
++ clone_pgd_mask = ~(pgdval_t)0UL;
++ pax_close_kernel();
++ if (pax_user_shadow_base)
++ printk("PAX: weak UDEREF enabled\n");
++ else {
++ set_cpu_cap(c, X86_FEATURE_STRONGUDEREF);
++ printk("PAX: strong UDEREF enabled\n");
++ }
++#endif
++
++ if (cpu_has(c, X86_FEATURE_INVPCID))
++ printk("PAX: INVPCID detected\n");
++}
++#endif
++
+ /*
+ * Some CPU features depend on higher CPUID levels, which may not always
+ * be available due to CPUID level capping or broken virtualization
+@@ -386,7 +383,7 @@ void switch_to_new_gdt(int cpu)
{
struct desc_ptr gdt_descr;
gdt_descr.size = GDT_SIZE - 1;
load_gdt(&gdt_descr);
/* Reload the per-cpu base */
-@@ -882,6 +828,10 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+@@ -874,6 +871,10 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+ setup_smep(c);
+ setup_smap(c);
+
++#ifdef CONFIG_X86_64
++ setup_pcid(c);
++#endif
++
+ /*
+ * The vendor-specific functions might have changed features.
+ * Now we do "generic changes."
+@@ -882,6 +883,10 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
/* Filter out anything that depends on CPUID levels we don't have */
filter_cpuid_features(c, true);
/* If the model name is still unset, do table lookup. */
if (!c->x86_model_id[0]) {
const char *p;
-@@ -1069,10 +1019,12 @@ static __init int setup_disablecpuid(char *arg)
+@@ -1069,10 +1074,12 @@ static __init int setup_disablecpuid(char *arg)
}
__setup("clearcpuid=", setup_disablecpuid);
DEFINE_PER_CPU_FIRST(union irq_stack_union,
irq_stack_union) __aligned(PAGE_SIZE);
-@@ -1086,7 +1038,7 @@ DEFINE_PER_CPU(struct task_struct *, current_task) ____cacheline_aligned =
+@@ -1086,7 +1093,7 @@ DEFINE_PER_CPU(struct task_struct *, current_task) ____cacheline_aligned =
EXPORT_PER_CPU_SYMBOL(current_task);
DEFINE_PER_CPU(unsigned long, kernel_stack) =
EXPORT_PER_CPU_SYMBOL(kernel_stack);
DEFINE_PER_CPU(char *, irq_stack_ptr) =
-@@ -1231,7 +1183,7 @@ void __cpuinit cpu_init(void)
+@@ -1231,7 +1238,7 @@ void __cpuinit cpu_init(void)
load_ucode_ap();
cpu = stack_smp_processor_id();
oist = &per_cpu(orig_ist, cpu);
#ifdef CONFIG_NUMA
-@@ -1257,7 +1209,7 @@ void __cpuinit cpu_init(void)
+@@ -1257,7 +1264,7 @@ void __cpuinit cpu_init(void)
switch_to_new_gdt(cpu);
loadsegment(fs, 0);
memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
syscall_init();
-@@ -1266,7 +1218,6 @@ void __cpuinit cpu_init(void)
+@@ -1266,7 +1273,6 @@ void __cpuinit cpu_init(void)
wrmsrl(MSR_KERNEL_GS_BASE, 0);
barrier();
enable_x2apic();
/*
-@@ -1318,7 +1269,7 @@ void __cpuinit cpu_init(void)
+@@ -1318,7 +1324,7 @@ void __cpuinit cpu_init(void)
{
int cpu = smp_processor_id();
struct task_struct *curr = current;
intel_ds_init();
diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.c b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
-index 52441a2..f94fae8 100644
+index 8aac56b..588fb13 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
@@ -3093,7 +3093,7 @@ static void __init uncore_types_exit(struct intel_uncore_type **types)
#include <asm/processor.h>
#include <asm/fcntl.h>
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
-index 8f3e2de..caecc4e 100644
+index 8f3e2de..6b71e39 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -177,13 +177,153 @@
ENTRY(simd_coprocessor_error)
RING0_INT_FRAME
+@@ -826,7 +1065,7 @@ ENTRY(simd_coprocessor_error)
+ .section .altinstructions,"a"
+ altinstruction_entry 661b, 663f, X86_FEATURE_XMM, 662b-661b, 664f-663f
+ .previous
+-.section .altinstr_replacement,"ax"
++.section .altinstr_replacement,"a"
+ 663: pushl $do_simd_coprocessor_error
+ 664:
+ .previous
@@ -835,7 +1074,7 @@ ENTRY(simd_coprocessor_error)
#endif
jmp error_code
/*
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
-index 7272089..6204f9c5 100644
+index 7272089..0b74104 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -59,6 +59,8 @@
#endif
-@@ -284,6 +293,309 @@ ENTRY(native_usergs_sysret64)
+@@ -284,6 +293,430 @@ ENTRY(native_usergs_sysret64)
ENDPROC(native_usergs_sysret64)
#endif /* CONFIG_PARAVIRT */
+
+ .macro pax_enter_kernel
+ pax_set_fptr_mask
-+#ifdef CONFIG_PAX_KERNEXEC
++#if defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF)
+ call pax_enter_kernel
+#endif
+ .endm
+
+ .macro pax_exit_kernel
-+#ifdef CONFIG_PAX_KERNEXEC
++#if defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF)
+ call pax_exit_kernel
+#endif
++
+ .endm
+
-+#ifdef CONFIG_PAX_KERNEXEC
++#if defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF)
+ENTRY(pax_enter_kernel)
+ pushq %rdi
+
+ PV_SAVE_REGS(CLBR_RDI)
+#endif
+
++#ifdef CONFIG_PAX_KERNEXEC
+ GET_CR0_INTO_RDI
+ bts $16,%rdi
+ jnc 3f
+ cmp $__KERNEL_CS,%edi
+ jnz 2f
+1:
++#endif
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ 661: jmp 111f
++ .pushsection .altinstr_replacement, "a"
++ 662: ASM_NOP2
++ .popsection
++ .pushsection .altinstructions, "a"
++ altinstruction_entry 661b, 662b, X86_FEATURE_PCID, 2, 2
++ .popsection
++ GET_CR3_INTO_RDI
++ cmp $0,%dil
++ jnz 112f
++ mov $__KERNEL_DS,%edi
++ mov %edi,%ss
++ jmp 111f
++112: cmp $1,%dil
++ jz 113f
++ ud2
++113: sub $4097,%rdi
++ bts $63,%rdi
++ SET_RDI_INTO_CR3
++ mov $__UDEREF_KERNEL_DS,%edi
++ mov %edi,%ss
++111:
++#endif
+
+#ifdef CONFIG_PARAVIRT
+ PV_RESTORE_REGS(CLBR_RDI)
+ pax_force_retaddr
+ retq
+
++#ifdef CONFIG_PAX_KERNEXEC
+2: ljmpq __KERNEL_CS,1b
+3: ljmpq __KERNEXEC_KERNEL_CS,4f
+4: SET_RDI_INTO_CR0
+ jmp 1b
++#endif
+ENDPROC(pax_enter_kernel)
+
+ENTRY(pax_exit_kernel)
+ PV_SAVE_REGS(CLBR_RDI)
+#endif
+
++#ifdef CONFIG_PAX_KERNEXEC
+ mov %cs,%rdi
+ cmp $__KERNEXEC_KERNEL_CS,%edi
+ jz 2f
+ bts $16,%rdi
+ jnc 4f
+1:
++#endif
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ 661: jmp 111f
++ .pushsection .altinstr_replacement, "a"
++ 662: ASM_NOP2
++ .popsection
++ .pushsection .altinstructions, "a"
++ altinstruction_entry 661b, 662b, X86_FEATURE_PCID, 2, 2
++ .popsection
++ mov %ss,%edi
++ cmp $__UDEREF_KERNEL_DS,%edi
++ jnz 111f
++ GET_CR3_INTO_RDI
++ cmp $0,%dil
++ jz 112f
++ ud2
++112: add $4097,%rdi
++ bts $63,%rdi
++ SET_RDI_INTO_CR3
++ mov $__KERNEL_DS,%edi
++ mov %edi,%ss
++111:
++#endif
+
+#ifdef CONFIG_PARAVIRT
+ PV_RESTORE_REGS(CLBR_RDI);
+ pax_force_retaddr
+ retq
+
++#ifdef CONFIG_PAX_KERNEXEC
+2: GET_CR0_INTO_RDI
+ btr $16,%rdi
+ jnc 4f
+ jmp 1b
+4: ud2
+ jmp 4b
++#endif
+ENDPROC(pax_exit_kernel)
+#endif
+
+ PV_SAVE_REGS(CLBR_RDI)
+#endif
+
++ 661: jmp 111f
++ .pushsection .altinstr_replacement, "a"
++ 662: ASM_NOP2
++ .popsection
++ .pushsection .altinstructions, "a"
++ altinstruction_entry 661b, 662b, X86_FEATURE_PCID, 2, 2
++ .popsection
++ GET_CR3_INTO_RDI
++ cmp $1,%dil
++ jnz 4f
++ sub $4097,%rdi
++ bts $63,%rdi
++ SET_RDI_INTO_CR3
++ jmp 3f
++111:
++
+ GET_CR3_INTO_RDI
+ mov %rdi,%rbx
+ add $__START_KERNEL_map,%rbx
+ i = i + 1
+ .endr
+
-+#ifdef CONFIG_PARAVIRT
-+2:
-+#endif
-+ SET_RDI_INTO_CR3
++2: SET_RDI_INTO_CR3
+
+#ifdef CONFIG_PAX_KERNEXEC
+ GET_CR0_INTO_RDI
+ SET_RDI_INTO_CR0
+#endif
+
++3:
++
+#ifdef CONFIG_PARAVIRT
+ PV_RESTORE_REGS(CLBR_RDI)
+#endif
+ popq %rdi
+ pax_force_retaddr
+ retq
++4: ud2
+ENDPROC(pax_enter_kernel_user)
+
+ENTRY(pax_exit_kernel_user)
+ PV_SAVE_REGS(CLBR_RDI)
+#endif
+
++ GET_CR3_INTO_RDI
++ 661: jmp 1f
++ .pushsection .altinstr_replacement, "a"
++ 662: ASM_NOP2
++ .popsection
++ .pushsection .altinstructions, "a"
++ altinstruction_entry 661b, 662b, X86_FEATURE_PCID, 2, 2
++ .popsection
++ cmp $0,%dil
++ jnz 3f
++ add $4097,%rdi
++ bts $63,%rdi
++ SET_RDI_INTO_CR3
++ jmp 2f
++1:
++
++ mov %rdi,%rbx
++
+#ifdef CONFIG_PAX_KERNEXEC
+ GET_CR0_INTO_RDI
+ btr $16,%rdi
+ SET_RDI_INTO_CR0
+#endif
+
-+ GET_CR3_INTO_RDI
-+ mov %rdi,%rbx
+ add $__START_KERNEL_map,%rbx
+ sub phys_base(%rip),%rbx
+
+ movb $0x67,i*8(%rbx)
+ i = i + 1
+ .endr
++2:
+
+#ifdef CONFIG_PARAVIRT
-+2: PV_RESTORE_REGS(CLBR_RDI)
++ PV_RESTORE_REGS(CLBR_RDI)
+#endif
+
+ popq %rbx
+ pax_force_retaddr
+ retq
+3: ud2
-+ jmp 3b
+ENDPROC(pax_exit_kernel_user)
+#endif
+
+ or $2,%ebx
+110:
+#endif
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ 661: jmp 111f
++ .pushsection .altinstr_replacement, "a"
++ 662: ASM_NOP2
++ .popsection
++ .pushsection .altinstructions, "a"
++ altinstruction_entry 661b, 662b, X86_FEATURE_PCID, 2, 2
++ .popsection
++ GET_CR3_INTO_RDI
++ cmp $0,%dil
++ jz 111f
++ sub $4097,%rdi
++ or $4,%ebx
++ bts $63,%rdi
++ SET_RDI_INTO_CR3
++ mov $__UDEREF_KERNEL_DS,%edi
++ mov %edi,%ss
++111:
++#endif
+ .endm
+
+ .macro pax_exit_kernel_nmi
+ SET_RDI_INTO_CR0
+110:
+#endif
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ btr $2,%ebx
++ jnc 111f
++ GET_CR3_INTO_RDI
++ add $4097,%rdi
++ bts $63,%rdi
++ SET_RDI_INTO_CR3
++ mov $__KERNEL_DS,%edi
++ mov %edi,%ss
++111:
++#endif
+ .endm
+
+ .macro pax_erase_kstack
.macro TRACE_IRQS_IRETQ offset=ARGOFFSET
#ifdef CONFIG_TRACE_IRQFLAGS
-@@ -375,8 +687,8 @@ ENDPROC(native_usergs_sysret64)
+@@ -375,8 +808,8 @@ ENDPROC(native_usergs_sysret64)
.endm
.macro UNFAKE_STACK_FRAME
.endm
/*
-@@ -463,7 +775,7 @@ ENDPROC(native_usergs_sysret64)
+@@ -463,7 +896,7 @@ ENDPROC(native_usergs_sysret64)
movq %rsp, %rsi
leaq -RBP(%rsp),%rdi /* arg1 for handler */
je 1f
SWAPGS
/*
-@@ -498,9 +810,10 @@ ENTRY(save_rest)
+@@ -498,9 +931,10 @@ ENTRY(save_rest)
movq_cfi r15, R15+16
movq %r11, 8(%rsp) /* return address */
FIXUP_TOP_OF_STACK %r11, 16
/* save complete stack frame */
.pushsection .kprobes.text, "ax"
-@@ -529,9 +842,10 @@ ENTRY(save_paranoid)
+@@ -529,9 +963,10 @@ ENTRY(save_paranoid)
js 1f /* negative -> in kernel */
SWAPGS
xorl %ebx,%ebx
.popsection
/*
-@@ -553,7 +867,7 @@ ENTRY(ret_from_fork)
+@@ -553,7 +988,7 @@ ENTRY(ret_from_fork)
RESTORE_REST
jz 1f
testl $_TIF_IA32, TI_flags(%rcx) # 32-bit compat task needs IRET
-@@ -571,7 +885,7 @@ ENTRY(ret_from_fork)
+@@ -571,7 +1006,7 @@ ENTRY(ret_from_fork)
RESTORE_REST
jmp int_ret_from_sys_call
CFI_ENDPROC
/*
* System call entry. Up to 6 arguments in registers are supported.
-@@ -608,7 +922,7 @@ END(ret_from_fork)
+@@ -608,7 +1043,7 @@ END(ret_from_fork)
ENTRY(system_call)
CFI_STARTPROC simple
CFI_SIGNAL_FRAME
CFI_REGISTER rip,rcx
/*CFI_REGISTER rflags,r11*/
SWAPGS_UNSAFE_STACK
-@@ -621,16 +935,23 @@ GLOBAL(system_call_after_swapgs)
+@@ -621,16 +1056,23 @@ GLOBAL(system_call_after_swapgs)
movq %rsp,PER_CPU_VAR(old_rsp)
movq PER_CPU_VAR(kernel_stack),%rsp
jnz tracesys
system_call_fastpath:
#if __SYSCALL_MASK == ~0
-@@ -640,7 +961,7 @@ system_call_fastpath:
+@@ -640,7 +1082,7 @@ system_call_fastpath:
cmpl $__NR_syscall_max,%eax
#endif
ja badsys
call *sys_call_table(,%rax,8) # XXX: rip relative
movq %rax,RAX-ARGOFFSET(%rsp)
/*
-@@ -654,10 +975,13 @@ sysret_check:
+@@ -654,10 +1096,13 @@ sysret_check:
LOCKDEP_SYS_EXIT
DISABLE_INTERRUPTS(CLBR_NONE)
TRACE_IRQS_OFF
/*
* sysretq will re-enable interrupts:
*/
-@@ -709,14 +1033,18 @@ badsys:
+@@ -709,14 +1154,18 @@ badsys:
* jump back to the normal fast path.
*/
auditsys:
jmp system_call_fastpath
/*
-@@ -737,7 +1065,7 @@ sysret_audit:
+@@ -737,7 +1186,7 @@ sysret_audit:
/* Do syscall tracing */
tracesys:
#ifdef CONFIG_AUDITSYSCALL
jz auditsys
#endif
SAVE_REST
-@@ -745,12 +1073,16 @@ tracesys:
+@@ -745,12 +1194,16 @@ tracesys:
FIXUP_TOP_OF_STACK %rdi
movq %rsp,%rdi
call syscall_trace_enter
RESTORE_REST
#if __SYSCALL_MASK == ~0
cmpq $__NR_syscall_max,%rax
-@@ -759,7 +1091,7 @@ tracesys:
+@@ -759,7 +1212,7 @@ tracesys:
cmpl $__NR_syscall_max,%eax
#endif
ja int_ret_from_sys_call /* RAX(%rsp) set to -ENOSYS above */
call *sys_call_table(,%rax,8)
movq %rax,RAX-ARGOFFSET(%rsp)
/* Use IRET because user could have changed frame */
-@@ -780,7 +1112,9 @@ GLOBAL(int_with_check)
+@@ -780,7 +1233,9 @@ GLOBAL(int_with_check)
andl %edi,%edx
jnz int_careful
andl $~TS_COMPAT,TI_status(%rcx)
/* Either reschedule or signal or syscall exit tracking needed. */
/* First do a reschedule test. */
-@@ -826,7 +1160,7 @@ int_restore_rest:
+@@ -826,7 +1281,7 @@ int_restore_rest:
TRACE_IRQS_OFF
jmp int_with_check
CFI_ENDPROC
.macro FORK_LIKE func
ENTRY(stub_\func)
-@@ -839,9 +1173,10 @@ ENTRY(stub_\func)
+@@ -839,9 +1294,10 @@ ENTRY(stub_\func)
DEFAULT_FRAME 0 8 /* offset 8: return address */
call sys_\func
RESTORE_TOP_OF_STACK %r11, 8
.endm
.macro FIXED_FRAME label,func
-@@ -851,9 +1186,10 @@ ENTRY(\label)
+@@ -851,9 +1307,10 @@ ENTRY(\label)
FIXUP_TOP_OF_STACK %r11, 8-ARGOFFSET
call \func
RESTORE_TOP_OF_STACK %r11, 8-ARGOFFSET
.endm
FORK_LIKE clone
-@@ -870,9 +1206,10 @@ ENTRY(ptregscall_common)
+@@ -870,9 +1327,10 @@ ENTRY(ptregscall_common)
movq_cfi_restore R12+8, r12
movq_cfi_restore RBP+8, rbp
movq_cfi_restore RBX+8, rbx
ENTRY(stub_execve)
CFI_STARTPROC
-@@ -885,7 +1222,7 @@ ENTRY(stub_execve)
+@@ -885,7 +1343,7 @@ ENTRY(stub_execve)
RESTORE_REST
jmp int_ret_from_sys_call
CFI_ENDPROC
/*
* sigreturn is special because it needs to restore all registers on return.
-@@ -902,7 +1239,7 @@ ENTRY(stub_rt_sigreturn)
+@@ -902,7 +1360,7 @@ ENTRY(stub_rt_sigreturn)
RESTORE_REST
jmp int_ret_from_sys_call
CFI_ENDPROC
#ifdef CONFIG_X86_X32_ABI
ENTRY(stub_x32_rt_sigreturn)
-@@ -916,7 +1253,7 @@ ENTRY(stub_x32_rt_sigreturn)
+@@ -916,7 +1374,7 @@ ENTRY(stub_x32_rt_sigreturn)
RESTORE_REST
jmp int_ret_from_sys_call
CFI_ENDPROC
ENTRY(stub_x32_execve)
CFI_STARTPROC
-@@ -930,7 +1267,7 @@ ENTRY(stub_x32_execve)
+@@ -930,7 +1388,7 @@ ENTRY(stub_x32_execve)
RESTORE_REST
jmp int_ret_from_sys_call
CFI_ENDPROC
#endif
-@@ -967,7 +1304,7 @@ vector=vector+1
+@@ -967,7 +1425,7 @@ vector=vector+1
2: jmp common_interrupt
.endr
CFI_ENDPROC
.previous
END(interrupt)
-@@ -987,6 +1324,16 @@ END(interrupt)
+@@ -987,6 +1445,16 @@ END(interrupt)
subq $ORIG_RAX-RBP, %rsp
CFI_ADJUST_CFA_OFFSET ORIG_RAX-RBP
SAVE_ARGS_IRQ
call \func
.endm
-@@ -1019,7 +1366,7 @@ ret_from_intr:
+@@ -1019,7 +1487,7 @@ ret_from_intr:
exit_intr:
GET_THREAD_INFO(%rcx)
je retint_kernel
/* Interrupt came from user space */
-@@ -1041,12 +1388,16 @@ retint_swapgs: /* return to user-space */
+@@ -1041,12 +1509,16 @@ retint_swapgs: /* return to user-space */
* The iretq could re-enable interrupts:
*/
DISABLE_INTERRUPTS(CLBR_ANY)
/*
* The iretq could re-enable interrupts:
*/
-@@ -1129,7 +1480,7 @@ ENTRY(retint_kernel)
+@@ -1129,7 +1601,7 @@ ENTRY(retint_kernel)
#endif
CFI_ENDPROC
/*
* End of kprobes section
*/
-@@ -1147,7 +1498,7 @@ ENTRY(\sym)
+@@ -1147,7 +1619,7 @@ ENTRY(\sym)
interrupt \do_sym
jmp ret_from_intr
CFI_ENDPROC
.endm
#ifdef CONFIG_SMP
-@@ -1208,12 +1559,22 @@ ENTRY(\sym)
+@@ -1208,12 +1680,22 @@ ENTRY(\sym)
CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
call error_entry
DEFAULT_FRAME 0
.endm
.macro paranoidzeroentry sym do_sym
-@@ -1226,15 +1587,25 @@ ENTRY(\sym)
+@@ -1226,15 +1708,25 @@ ENTRY(\sym)
CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
call save_paranoid
TRACE_IRQS_OFF
.macro paranoidzeroentry_ist sym do_sym ist
ENTRY(\sym)
INTR_FRAME
-@@ -1245,14 +1616,30 @@ ENTRY(\sym)
+@@ -1245,14 +1737,30 @@ ENTRY(\sym)
CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
call save_paranoid
TRACE_IRQS_OFF_DEBUG
.endm
.macro errorentry sym do_sym
-@@ -1264,13 +1651,23 @@ ENTRY(\sym)
+@@ -1264,13 +1772,23 @@ ENTRY(\sym)
CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
call error_entry
DEFAULT_FRAME 0
.endm
/* error code is on the stack already */
-@@ -1284,13 +1681,23 @@ ENTRY(\sym)
+@@ -1284,13 +1802,23 @@ ENTRY(\sym)
call save_paranoid
DEFAULT_FRAME 0
TRACE_IRQS_OFF
.endm
zeroentry divide_error do_divide_error
-@@ -1320,9 +1727,10 @@ gs_change:
+@@ -1320,9 +1848,10 @@ gs_change:
2: mfence /* workaround */
SWAPGS
popfq_cfi
_ASM_EXTABLE(gs_change,bad_gs)
.section .fixup,"ax"
-@@ -1350,9 +1758,10 @@ ENTRY(call_softirq)
+@@ -1350,9 +1879,10 @@ ENTRY(call_softirq)
CFI_DEF_CFA_REGISTER rsp
CFI_ADJUST_CFA_OFFSET -8
decl PER_CPU_VAR(irq_count)
#ifdef CONFIG_XEN
zeroentry xen_hypervisor_callback xen_do_hypervisor_callback
-@@ -1390,7 +1799,7 @@ ENTRY(xen_do_hypervisor_callback) # do_hypervisor_callback(struct *pt_regs)
+@@ -1390,7 +1920,7 @@ ENTRY(xen_do_hypervisor_callback) # do_hypervisor_callback(struct *pt_regs)
decl PER_CPU_VAR(irq_count)
jmp error_exit
CFI_ENDPROC
/*
* Hypervisor uses this for application faults while it executes.
-@@ -1449,7 +1858,7 @@ ENTRY(xen_failsafe_callback)
+@@ -1449,7 +1979,7 @@ ENTRY(xen_failsafe_callback)
SAVE_ALL
jmp error_exit
CFI_ENDPROC
apicinterrupt HYPERVISOR_CALLBACK_VECTOR \
xen_hvm_callback_vector xen_evtchn_do_upcall
-@@ -1501,18 +1910,33 @@ ENTRY(paranoid_exit)
+@@ -1501,18 +2031,33 @@ ENTRY(paranoid_exit)
DEFAULT_FRAME
DISABLE_INTERRUPTS(CLBR_NONE)
TRACE_IRQS_OFF_DEBUG
jmp irq_return
paranoid_userspace:
GET_THREAD_INFO(%rcx)
-@@ -1541,7 +1965,7 @@ paranoid_schedule:
+@@ -1541,7 +2086,7 @@ paranoid_schedule:
TRACE_IRQS_OFF
jmp paranoid_userspace
CFI_ENDPROC
/*
* Exception entry point. This expects an error code/orig_rax on the stack.
-@@ -1568,12 +1992,13 @@ ENTRY(error_entry)
+@@ -1568,12 +2113,13 @@ ENTRY(error_entry)
movq_cfi r14, R14+8
movq_cfi r15, R15+8
xorl %ebx,%ebx
ret
/*
-@@ -1600,7 +2025,7 @@ bstep_iret:
+@@ -1600,7 +2146,7 @@ bstep_iret:
movq %rcx,RIP+8(%rsp)
jmp error_swapgs
CFI_ENDPROC
/* ebx: no swapgs flag (1: don't need swapgs, 0: need it) */
-@@ -1611,7 +2036,7 @@ ENTRY(error_exit)
+@@ -1611,7 +2157,7 @@ ENTRY(error_exit)
DISABLE_INTERRUPTS(CLBR_NONE)
TRACE_IRQS_OFF
GET_THREAD_INFO(%rcx)
jne retint_kernel
LOCKDEP_SYS_EXIT_IRQ
movl TI_flags(%rcx),%edx
-@@ -1620,7 +2045,7 @@ ENTRY(error_exit)
+@@ -1620,7 +2166,7 @@ ENTRY(error_exit)
jnz retint_careful
jmp retint_swapgs
CFI_ENDPROC
/*
* Test if a given stack is an NMI stack or not.
-@@ -1678,9 +2103,11 @@ ENTRY(nmi)
+@@ -1678,9 +2224,11 @@ ENTRY(nmi)
* If %cs was not the kernel segment, then the NMI triggered in user
* space, which means it is definitely not nested.
*/
/*
* Check the special variable on the stack to see if NMIs are
* executing.
-@@ -1714,8 +2141,7 @@ nested_nmi:
+@@ -1714,8 +2262,7 @@ nested_nmi:
1:
/* Set up the interrupted NMIs stack to jump to repeat_nmi */
CFI_ADJUST_CFA_OFFSET 1*8
leaq -10*8(%rsp), %rdx
pushq_cfi $__KERNEL_DS
-@@ -1733,6 +2159,7 @@ nested_nmi_out:
+@@ -1733,6 +2280,7 @@ nested_nmi_out:
CFI_RESTORE rdx
/* No need to check faults here */
INTERRUPT_RETURN
CFI_RESTORE_STATE
-@@ -1849,6 +2276,8 @@ end_repeat_nmi:
+@@ -1849,6 +2397,8 @@ end_repeat_nmi:
*/
movq %cr2, %r12
/* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */
movq %rsp,%rdi
movq $-1,%rsi
-@@ -1861,26 +2290,31 @@ end_repeat_nmi:
+@@ -1861,26 +2411,31 @@ end_repeat_nmi:
movq %r12, %cr2
1:
init_level4_pgt[511] = early_level4_pgt[511];
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
-index 73afd11..d1670f5 100644
+index 73afd11..0ef46f2 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -26,6 +26,12 @@
+
+#ifdef CONFIG_PAX_PER_CPU_PGD
+ENTRY(cpu_pgd)
-+ .rept NR_CPUS
++ .rept 2*NR_CPUS
+ .fill 4,8,0
+ .endr
+#endif
+ .fill PAGE_SIZE_asm - GDT_SIZE,1,0
+ .endr
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
-index a836860..bdeb7a5 100644
+index a836860..1b5c665 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -20,6 +20,8 @@
+#ifdef CONFIG_PAX_PER_CPU_PGD
+NEXT_PAGE(cpu_pgd)
-+ .rept NR_CPUS
++ .rept 2*NR_CPUS
+ .fill 512,8,0
+ .endr
+#endif
NEXT_PAGE(level2_kernel_pgt)
/*
* 512 MB kernel mapping. We spend a full page on this pagetable
-@@ -488,39 +544,64 @@ NEXT_PAGE(level2_kernel_pgt)
+@@ -488,39 +544,70 @@ NEXT_PAGE(level2_kernel_pgt)
KERNEL_IMAGE_SIZE/PMD_SIZE)
NEXT_PAGE(level2_fixmap_pgt)
+ .quad 0x0000f40000000000 /* node/CPU stored in limit */
+ /* asm/segment.h:GDT_ENTRIES must match this */
+
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ .quad 0x00cf93000000ffff /* __UDEREF_KERNEL_DS */
++#else
++ .quad 0x0 /* unused */
++#endif
++
+ /* zero the remaining page */
+ .fill PAGE_SIZE / 8 - GDT_ENTRIES,8,0
+ .endr
- .skip PAGE_SIZE
+ .fill 512,8,0
diff --git a/arch/x86/kernel/i386_ksyms_32.c b/arch/x86/kernel/i386_ksyms_32.c
-index 0fa6912..37fce70 100644
+index 0fa6912..b37438b 100644
--- a/arch/x86/kernel/i386_ksyms_32.c
+++ b/arch/x86/kernel/i386_ksyms_32.c
@@ -20,8 +20,12 @@ extern void cmpxchg8b_emu(void);
EXPORT_SYMBOL(__get_user_1);
EXPORT_SYMBOL(__get_user_2);
-@@ -37,3 +41,7 @@ EXPORT_SYMBOL(strstr);
+@@ -37,3 +41,11 @@ EXPORT_SYMBOL(strstr);
EXPORT_SYMBOL(csum_partial);
EXPORT_SYMBOL(empty_zero_page);
+#ifdef CONFIG_PAX_KERNEXEC
+EXPORT_SYMBOL(__LOAD_PHYSICAL_ADDR);
+#endif
++
++#ifdef CONFIG_PAX_PER_CPU_PGD
++EXPORT_SYMBOL(cpu_pgd);
++#endif
diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
-index cb33909..1163b40 100644
+index f7ea30d..6318acc 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -51,7 +51,7 @@ static inline bool interrupted_kernel_fpu_idle(void)
}
-
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
-index 355ae06..4530766 100644
+index 355ae06..560fbbe 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -151,10 +151,11 @@ int copy_thread(unsigned long clone_flags, unsigned long sp,
set_tsk_thread_flag(p, TIF_FORK);
p->fpu_counter = 0;
p->thread.io_bitmap_ptr = NULL;
-@@ -273,7 +274,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+@@ -165,6 +166,8 @@ int copy_thread(unsigned long clone_flags, unsigned long sp,
+ p->thread.fs = p->thread.fsindex ? 0 : me->thread.fs;
+ savesegment(es, p->thread.es);
+ savesegment(ds, p->thread.ds);
++ savesegment(ss, p->thread.ss);
++ BUG_ON(p->thread.ss == __UDEREF_KERNEL_DS);
+ memset(p->thread.ptrace_bps, 0, sizeof(p->thread.ptrace_bps));
+
+ if (unlikely(p->flags & PF_KTHREAD)) {
+@@ -273,7 +276,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
struct thread_struct *prev = &prev_p->thread;
struct thread_struct *next = &next_p->thread;
int cpu = smp_processor_id();
unsigned fsindex, gsindex;
fpu_switch_t fpu;
-@@ -355,10 +356,9 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+@@ -296,6 +299,9 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+ if (unlikely(next->ds | prev->ds))
+ loadsegment(ds, next->ds);
+
++ savesegment(ss, prev->ss);
++ if (unlikely(next->ss != prev->ss))
++ loadsegment(ss, next->ss);
+
+ /* We must save %fs and %gs before load_TLS() because
+ * %fs and %gs may be cleared by load_TLS().
+@@ -355,10 +361,9 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
prev->usersp = this_cpu_read(old_rsp);
this_cpu_write(old_rsp, next->usersp);
this_cpu_write(current_task, next_p);
/*
* Now maybe reload the debug registers and handle I/O bitmaps
-@@ -427,12 +427,11 @@ unsigned long get_wchan(struct task_struct *p)
+@@ -427,12 +432,11 @@ unsigned long get_wchan(struct task_struct *p)
if (!p || p == current || p->state == TASK_RUNNING)
return 0;
stack = (unsigned long)task_stack_page(p);
1:
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
-index 56f7fcf..fa229f4 100644
+index 56f7fcf..2cfe4f1 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -110,6 +110,7 @@
/*
* max_low_pfn_mapped: highest direct mapped pfn under 4GB
-@@ -444,7 +445,7 @@ static void __init parse_setup_data(void)
+@@ -205,12 +206,50 @@ EXPORT_SYMBOL(boot_cpu_data);
+ #endif
+
+
+-#if !defined(CONFIG_X86_PAE) || defined(CONFIG_X86_64)
+-unsigned long mmu_cr4_features;
++#ifdef CONFIG_X86_64
++unsigned long mmu_cr4_features __read_only = X86_CR4_PSE | X86_CR4_PAE | X86_CR4_PGE;
++#elif defined(CONFIG_X86_PAE)
++unsigned long mmu_cr4_features __read_only = X86_CR4_PAE;
+ #else
+-unsigned long mmu_cr4_features = X86_CR4_PAE;
++unsigned long mmu_cr4_features __read_only;
+ #endif
+
++void set_in_cr4(unsigned long mask)
++{
++ unsigned long cr4 = read_cr4();
++
++ if ((cr4 & mask) == mask && cr4 == mmu_cr4_features)
++ return;
++
++ pax_open_kernel();
++ mmu_cr4_features |= mask;
++ pax_close_kernel();
++
++ if (trampoline_cr4_features)
++ *trampoline_cr4_features = mmu_cr4_features;
++ cr4 |= mask;
++ write_cr4(cr4);
++}
++EXPORT_SYMBOL(set_in_cr4);
++
++void clear_in_cr4(unsigned long mask)
++{
++ unsigned long cr4 = read_cr4();
++
++ if (!(cr4 & mask) && cr4 == mmu_cr4_features)
++ return;
++
++ pax_open_kernel();
++ mmu_cr4_features &= ~mask;
++ pax_close_kernel();
++
++ if (trampoline_cr4_features)
++ *trampoline_cr4_features = mmu_cr4_features;
++ cr4 &= ~mask;
++ write_cr4(cr4);
++}
++EXPORT_SYMBOL(clear_in_cr4);
++
+ /* Boot loader ID and version as integers, for the benefit of proc_dointvec */
+ int bootloader_type, bootloader_version;
+
+@@ -444,7 +483,7 @@ static void __init parse_setup_data(void)
switch (data->type) {
case SETUP_E820_EXT:
break;
case SETUP_DTB:
add_dtb(pa_data);
-@@ -771,7 +772,7 @@ static void __init trim_bios_range(void)
+@@ -771,7 +810,7 @@ static void __init trim_bios_range(void)
* area (640->1Mb) as ram even though it is not.
* take them out.
*/
sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
}
-@@ -779,7 +780,7 @@ static void __init trim_bios_range(void)
+@@ -779,7 +818,7 @@ static void __init trim_bios_range(void)
/* called before trim_bios_range() to spare extra sanitize */
static void __init e820_add_kernel_range(void)
{
u64 size = __pa_symbol(_end) - start;
/*
-@@ -841,8 +842,12 @@ static void __init trim_low_memory_range(void)
+@@ -841,8 +880,12 @@ static void __init trim_low_memory_range(void)
void __init setup_arch(char **cmdline_p)
{
early_reserve_initrd();
-@@ -934,14 +939,14 @@ void __init setup_arch(char **cmdline_p)
+@@ -934,14 +977,14 @@ void __init setup_arch(char **cmdline_p)
if (!boot_params.hdr.root_flags)
root_mountflags &= ~MS_RDONLY;
* Up to this point, the boot CPU has been using .init.data
* area. Reload any changed state for the boot CPU.
diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
-index 6956299..f20beae 100644
+index 6956299..18126ec4 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -196,7 +196,7 @@ static unsigned long align_sigframe(unsigned long sp)
if (err)
return -EFAULT;
-@@ -367,7 +367,10 @@ static int __setup_rt_frame(int sig, struct ksignal *ksig,
- err |= __save_altstack(&frame->uc.uc_stack, regs->sp);
+@@ -364,10 +364,13 @@ static int __setup_rt_frame(int sig, struct ksignal *ksig,
+ else
+ put_user_ex(0, &frame->uc.uc_flags);
+ put_user_ex(0, &frame->uc.uc_link);
+- err |= __save_altstack(&frame->uc.uc_stack, regs->sp);
++ __save_altstack_ex(&frame->uc.uc_stack, regs->sp);
/* Set up to return from userspace. */
- restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn);
} put_user_catch(err);
err |= copy_siginfo_to_user(&frame->info, &ksig->info);
+@@ -429,7 +432,7 @@ static int __setup_rt_frame(int sig, struct ksignal *ksig,
+ else
+ put_user_ex(0, &frame->uc.uc_flags);
+ put_user_ex(0, &frame->uc.uc_link);
+- err |= __save_altstack(&frame->uc.uc_stack, regs->sp);
++ __save_altstack_ex(&frame->uc.uc_stack, regs->sp);
+
+ /* Set up to return from userspace. If provided, use a stub
+ already in userspace. */
@@ -615,7 +618,12 @@ setup_rt_frame(struct ksignal *ksig, struct pt_regs *regs)
{
int usig = signr_convert(ksig->sig);
.smp_prepare_cpus = native_smp_prepare_cpus,
.smp_cpus_done = native_smp_cpus_done,
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
-index bfd348e..4816ad8 100644
+index bfd348e..914f323 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
-@@ -748,6 +748,7 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu, struct task_struct *idle)
+@@ -251,14 +251,18 @@ notrace static void __cpuinit start_secondary(void *unused)
+
+ enable_start_cpu0 = 0;
+
+-#ifdef CONFIG_X86_32
+- /* switch away from the initial page table */
+- load_cr3(swapper_pg_dir);
+- __flush_tlb_all();
+-#endif
+-
+ /* otherwise gcc will move up smp_processor_id before the cpu_init */
+ barrier();
++
++ /* switch away from the initial page table */
++#ifdef CONFIG_PAX_PER_CPU_PGD
++ load_cr3(get_cpu_pgd(smp_processor_id(), kernel));
++ __flush_tlb_all();
++#elif defined(CONFIG_X86_32)
++ load_cr3(swapper_pg_dir);
++ __flush_tlb_all();
++#endif
++
+ /*
+ * Check TSC synchronization with the BP:
+ */
+@@ -748,6 +752,7 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu, struct task_struct *idle)
idle->thread.sp = (unsigned long) (((struct pt_regs *)
(THREAD_SIZE + task_stack_page(idle))) - 1);
per_cpu(current_task, cpu) = idle;
#ifdef CONFIG_X86_32
/* Stack for startup_32 can be just as for start_secondary onwards */
-@@ -755,11 +756,13 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu, struct task_struct *idle)
+@@ -755,11 +760,13 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu, struct task_struct *idle)
#else
clear_tsk_thread_flag(idle, TIF_FORK);
initial_gs = per_cpu_offset(cpu);
initial_code = (unsigned long)start_secondary;
stack_start = idle->thread.sp;
-@@ -908,6 +911,15 @@ int __cpuinit native_cpu_up(unsigned int cpu, struct task_struct *tidle)
+@@ -908,6 +915,15 @@ int __cpuinit native_cpu_up(unsigned int cpu, struct task_struct *tidle)
/* the FPU context is blank, nobody can own it */
__cpu_disable_lazy_restore(cpu);
+#ifdef CONFIG_PAX_PER_CPU_PGD
-+ clone_pgd_range(get_cpu_pgd(cpu) + KERNEL_PGD_BOUNDARY,
++ clone_pgd_range(get_cpu_pgd(cpu, kernel) + KERNEL_PGD_BOUNDARY,
++ swapper_pg_dir + KERNEL_PGD_BOUNDARY,
++ KERNEL_PGD_PTRS);
++ clone_pgd_range(get_cpu_pgd(cpu, user) + KERNEL_PGD_BOUNDARY,
+ swapper_pg_dir + KERNEL_PGD_BOUNDARY,
+ KERNEL_PGD_PTRS);
+#endif
-+
-+ /* the FPU context is blank, nobody can own it */
-+ __cpu_disable_lazy_restore(cpu);
+
err = do_boot_cpu(apicid, cpu, tidle);
if (err) {
+ return arch_get_unmapped_area(filp, addr0, len, pgoff, flags);
+}
diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c
-index dbded5a..ace2781 100644
+index 48f8375..ace2781 100644
--- a/arch/x86/kernel/sys_x86_64.c
+++ b/arch/x86/kernel/sys_x86_64.c
@@ -81,8 +81,8 @@ out:
*begin = new_begin;
}
} else {
-- *begin = TASK_UNMAPPED_BASE;
+- *begin = mmap_legacy_base();
+ *begin = mm->mmap_base;
*end = TASK_SIZE;
}
(unsigned long)VSYSCALL_START);
diff --git a/arch/x86/kernel/x8664_ksyms_64.c b/arch/x86/kernel/x8664_ksyms_64.c
-index b014d94..6d6ca7b 100644
+index b014d94..e775258 100644
--- a/arch/x86/kernel/x8664_ksyms_64.c
+++ b/arch/x86/kernel/x8664_ksyms_64.c
@@ -34,8 +34,6 @@ EXPORT_SYMBOL(copy_user_generic_string);
EXPORT_SYMBOL(copy_page);
EXPORT_SYMBOL(clear_page);
+@@ -66,3 +64,7 @@ EXPORT_SYMBOL(empty_zero_page);
+ #ifndef CONFIG_PARAVIRT
+ EXPORT_SYMBOL(native_load_gs_index);
+ #endif
++
++#ifdef CONFIG_PAX_PER_CPU_PGD
++EXPORT_SYMBOL(cpu_pgd);
++#endif
diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c
index 45a14db..075bb9b 100644
--- a/arch/x86/kernel/x86_init.c
.byte (copy_page_rep - copy_page) - (2f - 1b) /* offset */
2:
diff --git a/arch/x86/lib/copy_user_64.S b/arch/x86/lib/copy_user_64.S
-index a30ca15..d25fab6 100644
+index a30ca15..6b3f4e1 100644
--- a/arch/x86/lib/copy_user_64.S
+++ b/arch/x86/lib/copy_user_64.S
-@@ -18,6 +18,7 @@
+@@ -18,31 +18,7 @@
#include <asm/alternative-asm.h>
#include <asm/asm.h>
#include <asm/smap.h>
+-
+-/*
+- * By placing feature2 after feature1 in altinstructions section, we logically
+- * implement:
+- * If CPU has feature2, jmp to alt2 is used
+- * else if CPU has feature1, jmp to alt1 is used
+- * else jmp to orig is used.
+- */
+- .macro ALTERNATIVE_JUMP feature1,feature2,orig,alt1,alt2
+-0:
+- .byte 0xe9 /* 32bit jump */
+- .long \orig-1f /* by default jump to orig */
+-1:
+- .section .altinstr_replacement,"ax"
+-2: .byte 0xe9 /* near jump with 32bit immediate */
+- .long \alt1-1b /* offset */ /* or alternatively to alt1 */
+-3: .byte 0xe9 /* near jump with 32bit immediate */
+- .long \alt2-1b /* offset */ /* or alternatively to alt2 */
+- .previous
+-
+- .section .altinstructions,"a"
+- altinstruction_entry 0b,2b,\feature1,5,5
+- altinstruction_entry 0b,3b,\feature2,5,5
+- .previous
+- .endm
+#include <asm/pgtable.h>
- /*
- * By placing feature2 after feature1 in altinstructions section, we logically
-@@ -31,7 +32,7 @@
- .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 \alt1-1b /* offset */ /* or alternatively to alt1 */
- 3: .byte 0xe9 /* near jump with 32bit immediate */
-@@ -70,47 +71,20 @@
+ .macro ALIGN_DESTINATION
+ #ifdef FIX_ALIGNMENT
+@@ -70,52 +46,6 @@
#endif
.endm
- CFI_ENDPROC
-ENDPROC(_copy_from_user)
-
- .section .fixup,"ax"
- /* must zero dest */
- ENTRY(bad_from_user)
- bad_from_user:
+- .section .fixup,"ax"
+- /* must zero dest */
+-ENTRY(bad_from_user)
+-bad_from_user:
+- CFI_STARTPROC
+- movl %edx,%ecx
+- xorl %eax,%eax
+- rep
+- stosb
+-bad_to_user:
+- movl %edx,%eax
+- ret
+- CFI_ENDPROC
+-ENDPROC(bad_from_user)
+- .previous
+-
+ /*
+ * copy_user_generic_unrolled - memory copy with exception handling.
+ * This version is for CPUs like P4 that don't have efficient micro
+@@ -131,6 +61,7 @@ ENDPROC(bad_from_user)
+ */
+ ENTRY(copy_user_generic_unrolled)
CFI_STARTPROC
-+ testl %edx,%edx
-+ js bad_to_user
- movl %edx,%ecx
- xorl %eax,%eax
- rep
- stosb
- bad_to_user:
- movl %edx,%eax
-+ pax_force_retaddr
- ret
- CFI_ENDPROC
- ENDPROC(bad_from_user)
-@@ -141,19 +115,19 @@ ENTRY(copy_user_generic_unrolled)
++ ASM_PAX_OPEN_USERLAND
+ ASM_STAC
+ cmpl $8,%edx
+ jb 20f /* less then 8 bytes, go to byte copy loop */
+@@ -141,19 +72,19 @@ ENTRY(copy_user_generic_unrolled)
jz 17f
1: movq (%rsi),%r8
2: movq 1*8(%rsi),%r9
16: movq %r11,7*8(%rdi)
leaq 64(%rsi),%rsi
leaq 64(%rdi),%rdi
-@@ -180,6 +154,7 @@ ENTRY(copy_user_generic_unrolled)
+@@ -180,6 +111,8 @@ ENTRY(copy_user_generic_unrolled)
jnz 21b
23: xor %eax,%eax
ASM_CLAC
++ ASM_PAX_CLOSE_USERLAND
+ pax_force_retaddr
ret
.section .fixup,"ax"
-@@ -251,6 +226,7 @@ ENTRY(copy_user_generic_string)
+@@ -235,6 +168,7 @@ ENDPROC(copy_user_generic_unrolled)
+ */
+ ENTRY(copy_user_generic_string)
+ CFI_STARTPROC
++ ASM_PAX_OPEN_USERLAND
+ ASM_STAC
+ andl %edx,%edx
+ jz 4f
+@@ -251,6 +185,8 @@ ENTRY(copy_user_generic_string)
movsb
4: xorl %eax,%eax
ASM_CLAC
++ ASM_PAX_CLOSE_USERLAND
+ pax_force_retaddr
ret
.section .fixup,"ax"
-@@ -286,6 +262,7 @@ ENTRY(copy_user_enhanced_fast_string)
+@@ -278,6 +214,7 @@ ENDPROC(copy_user_generic_string)
+ */
+ ENTRY(copy_user_enhanced_fast_string)
+ CFI_STARTPROC
++ ASM_PAX_OPEN_USERLAND
+ ASM_STAC
+ andl %edx,%edx
+ jz 2f
+@@ -286,6 +223,8 @@ ENTRY(copy_user_enhanced_fast_string)
movsb
2: xorl %eax,%eax
ASM_CLAC
++ ASM_PAX_CLOSE_USERLAND
+ pax_force_retaddr
ret
.section .fixup,"ax"
diff --git a/arch/x86/lib/copy_user_nocache_64.S b/arch/x86/lib/copy_user_nocache_64.S
-index 6a4f43c..f08b4a2 100644
+index 6a4f43c..55d26f2 100644
--- a/arch/x86/lib/copy_user_nocache_64.S
+++ b/arch/x86/lib/copy_user_nocache_64.S
@@ -8,6 +8,7 @@
.macro ALIGN_DESTINATION
#ifdef FIX_ALIGNMENT
-@@ -49,6 +51,15 @@
+@@ -49,6 +51,16 @@
*/
ENTRY(__copy_user_nocache)
CFI_STARTPROC
+1:
+#endif
+
++ ASM_PAX_OPEN_USERLAND
ASM_STAC
cmpl $8,%edx
jb 20f /* less then 8 bytes, go to byte copy loop */
-@@ -59,19 +70,19 @@ ENTRY(__copy_user_nocache)
+@@ -59,19 +71,19 @@ ENTRY(__copy_user_nocache)
jz 17f
1: movq (%rsi),%r8
2: movq 1*8(%rsi),%r9
16: movnti %r11,7*8(%rdi)
leaq 64(%rsi),%rsi
leaq 64(%rdi),%rdi
-@@ -99,6 +110,7 @@ ENTRY(__copy_user_nocache)
+@@ -98,7 +110,9 @@ ENTRY(__copy_user_nocache)
+ jnz 21b
23: xorl %eax,%eax
ASM_CLAC
++ ASM_PAX_CLOSE_USERLAND
sfence
+ pax_force_retaddr
ret
CFI_RESTORE_STATE
diff --git a/arch/x86/lib/csum-wrappers_64.c b/arch/x86/lib/csum-wrappers_64.c
-index 25b7ae8..169fafc 100644
+index 25b7ae8..c40113e 100644
--- a/arch/x86/lib/csum-wrappers_64.c
+++ b/arch/x86/lib/csum-wrappers_64.c
-@@ -52,7 +52,7 @@ csum_partial_copy_from_user(const void __user *src, void *dst,
+@@ -52,8 +52,12 @@ csum_partial_copy_from_user(const void __user *src, void *dst,
len -= 2;
}
}
- isum = csum_partial_copy_generic((__force const void *)src,
++ pax_open_userland();
++ stac();
+ isum = csum_partial_copy_generic((const void __force_kernel *)____m(src),
dst, len, isum, errp, NULL);
++ clac();
++ pax_close_userland();
if (unlikely(*errp))
goto out_err;
-@@ -105,7 +105,7 @@ csum_partial_copy_to_user(const void *src, void __user *dst,
+
+@@ -105,8 +109,13 @@ csum_partial_copy_to_user(const void *src, void __user *dst,
}
*errp = 0;
- return csum_partial_copy_generic(src, (void __force *)dst,
-+ return csum_partial_copy_generic(src, (void __force_kernel *)____m(dst),
++ pax_open_userland();
++ stac();
++ isum = csum_partial_copy_generic(src, (void __force_kernel *)____m(dst),
len, isum, NULL, errp);
++ clac();
++ pax_close_userland();
++ return isum;
}
EXPORT_SYMBOL(csum_partial_copy_to_user);
+
diff --git a/arch/x86/lib/getuser.S b/arch/x86/lib/getuser.S
index a451235..1daa956 100644
--- a/arch/x86/lib/getuser.S
CFI_ENDPROC
ENDPROC(__iowrite32_copy)
diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S
-index 56313a3..aa84a79 100644
+index 56313a3..9b59269 100644
--- a/arch/x86/lib/memcpy_64.S
+++ b/arch/x86/lib/memcpy_64.S
+@@ -24,7 +24,7 @@
+ * This gets patched over the unrolled variant (below) via the
+ * alternative instructions framework:
+ */
+- .section .altinstr_replacement, "ax", @progbits
++ .section .altinstr_replacement, "a", @progbits
+ .Lmemcpy_c:
+ movq %rdi, %rax
+ movq %rdx, %rcx
@@ -33,6 +33,7 @@
rep movsq
movl %edx, %ecx
ret
.Lmemcpy_e:
.previous
-@@ -49,6 +50,7 @@
+@@ -44,11 +45,12 @@
+ * This gets patched over the unrolled variant (below) via the
+ * alternative instructions framework:
+ */
+- .section .altinstr_replacement, "ax", @progbits
++ .section .altinstr_replacement, "a", @progbits
+ .Lmemcpy_c_e:
movq %rdi, %rax
movq %rdx, %rcx
rep movsb
CFI_ENDPROC
ENDPROC(memcpy)
diff --git a/arch/x86/lib/memmove_64.S b/arch/x86/lib/memmove_64.S
-index 65268a6..c9518d1 100644
+index 65268a6..5aa7815 100644
--- a/arch/x86/lib/memmove_64.S
+++ b/arch/x86/lib/memmove_64.S
@@ -61,13 +61,13 @@ ENTRY(memmove)
jmp 13f
12:
cmp $1, %rdx
-@@ -202,6 +202,7 @@ ENTRY(memmove)
+@@ -202,14 +202,16 @@ ENTRY(memmove)
movb (%rsi), %r11b
movb %r11b, (%rdi)
13:
retq
CFI_ENDPROC
-@@ -210,6 +211,7 @@ ENTRY(memmove)
+- .section .altinstr_replacement,"ax"
++ .section .altinstr_replacement,"a"
+ .Lmemmove_begin_forward_efs:
/* Forward moving data. */
movq %rdx, %rcx
rep movsb
.Lmemmove_end_forward_efs:
.previous
diff --git a/arch/x86/lib/memset_64.S b/arch/x86/lib/memset_64.S
-index 2dcb380..963660a 100644
+index 2dcb380..50a78bc 100644
--- a/arch/x86/lib/memset_64.S
+++ b/arch/x86/lib/memset_64.S
+@@ -16,7 +16,7 @@
+ *
+ * rax original destination
+ */
+- .section .altinstr_replacement, "ax", @progbits
++ .section .altinstr_replacement, "a", @progbits
+ .Lmemset_c:
+ movq %rdi,%r9
+ movq %rdx,%rcx
@@ -30,6 +30,7 @@
movl %edx,%ecx
rep stosb
ret
.Lmemset_e:
.previous
-@@ -52,6 +53,7 @@
+@@ -45,13 +46,14 @@
+ *
+ * rax original destination
+ */
+- .section .altinstr_replacement, "ax", @progbits
++ .section .altinstr_replacement, "a", @progbits
+ .Lmemset_c_e:
+ movq %rdi,%r9
+ movb %sil,%al
movq %rdx,%rcx
rep stosb
movq %r9,%rax
+EXPORT_SYMBOL(set_fs);
+#endif
diff --git a/arch/x86/lib/usercopy_64.c b/arch/x86/lib/usercopy_64.c
-index 906fea3..5646695 100644
+index 906fea3..0194a18 100644
--- a/arch/x86/lib/usercopy_64.c
+++ b/arch/x86/lib/usercopy_64.c
-@@ -39,7 +39,7 @@ unsigned long __clear_user(void __user *addr, unsigned long size)
+@@ -18,6 +18,7 @@ unsigned long __clear_user(void __user *addr, unsigned long size)
+ might_fault();
+ /* no memory constraint because it doesn't change any memory gcc knows
+ about */
++ pax_open_userland();
+ stac();
+ asm volatile(
+ " testq %[size8],%[size8]\n"
+@@ -39,9 +40,10 @@ unsigned long __clear_user(void __user *addr, unsigned long size)
_ASM_EXTABLE(0b,3b)
_ASM_EXTABLE(1b,2b)
: [size8] "=&c"(size), [dst] "=&D" (__d0)
+ : [size1] "r"(size & 7), "[size8]" (size / 8), "[dst]"(____m(addr)),
[zero] "r" (0UL), [eight] "r" (8UL));
clac();
++ pax_close_userland();
return size;
-@@ -54,12 +54,11 @@ unsigned long clear_user(void __user *to, unsigned long n)
+ }
+ EXPORT_SYMBOL(__clear_user);
+@@ -54,12 +56,11 @@ unsigned long clear_user(void __user *to, unsigned long n)
}
EXPORT_SYMBOL(clear_user);
}
EXPORT_SYMBOL(copy_in_user);
-@@ -69,7 +68,7 @@ EXPORT_SYMBOL(copy_in_user);
+@@ -69,11 +70,13 @@ EXPORT_SYMBOL(copy_in_user);
* it is not necessary to optimize tail handling.
*/
unsigned long
{
char c;
unsigned zero_len;
+
++ clac();
++ pax_close_userland();
+ for (; len; --len, to++) {
+ if (__get_user_nocheck(c, from++, sizeof(char)))
+ break;
+@@ -84,6 +87,5 @@ copy_user_handle_tail(char *to, char *from, unsigned len, unsigned zerorest)
+ for (c = 0, zero_len = len; zerorest && zero_len; --zero_len)
+ if (__put_user_nocheck(c, to++, sizeof(char)))
+ break;
+- clac();
+ return len;
+ }
+diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
+index 23d8e5f..9ccc13a 100644
+--- a/arch/x86/mm/Makefile
++++ b/arch/x86/mm/Makefile
+@@ -28,3 +28,7 @@ obj-$(CONFIG_ACPI_NUMA) += srat.o
+ obj-$(CONFIG_NUMA_EMU) += numa_emulation.o
+
+ obj-$(CONFIG_MEMTEST) += memtest.o
++
++quote:="
++obj-$(CONFIG_X86_64) += uderef_64.o
++CFLAGS_uderef_64.o := $(subst $(quote),,$(CONFIG_ARCH_HWEIGHT_CFLAGS))
diff --git a/arch/x86/mm/extable.c b/arch/x86/mm/extable.c
index 903ec1e..c4166b2 100644
--- a/arch/x86/mm/extable.c
}
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
-index 654be4a..d36985f 100644
+index 654be4a..a4a3da1 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -14,11 +14,18 @@
DEFINE_SPINLOCK(pgd_lock);
LIST_HEAD(pgd_list);
-@@ -232,10 +273,22 @@ void vmalloc_sync_all(void)
+@@ -232,10 +273,27 @@ void vmalloc_sync_all(void)
for (address = VMALLOC_START & PMD_MASK;
address >= TASK_SIZE && address < FIXADDR_TOP;
address += PMD_SIZE) {
+
+#ifdef CONFIG_PAX_PER_CPU_PGD
+ for (cpu = 0; cpu < nr_cpu_ids; ++cpu) {
-+ pgd_t *pgd = get_cpu_pgd(cpu);
++ pgd_t *pgd = get_cpu_pgd(cpu, user);
+ pmd_t *ret;
++
++ ret = vmalloc_sync_one(pgd, address);
++ if (!ret)
++ break;
++ pgd = get_cpu_pgd(cpu, kernel);
+#else
list_for_each_entry(page, &pgd_list, lru) {
+ pgd_t *pgd;
spinlock_t *pgt_lock;
pmd_t *ret;
-@@ -243,8 +296,14 @@ void vmalloc_sync_all(void)
+@@ -243,8 +301,14 @@ void vmalloc_sync_all(void)
pgt_lock = &pgd_page_get_mm(page)->page_table_lock;
spin_lock(pgt_lock);
if (!ret)
break;
-@@ -278,6 +337,11 @@ static noinline __kprobes int vmalloc_fault(unsigned long address)
+@@ -278,6 +342,12 @@ static noinline __kprobes int vmalloc_fault(unsigned long address)
* an interrupt in the middle of a task switch..
*/
pgd_paddr = read_cr3();
+
+#ifdef CONFIG_PAX_PER_CPU_PGD
-+ BUG_ON(__pa(get_cpu_pgd(smp_processor_id())) != (pgd_paddr & PHYSICAL_PAGE_MASK));
++ BUG_ON(__pa(get_cpu_pgd(smp_processor_id(), kernel)) != (pgd_paddr & __PHYSICAL_MASK));
++ vmalloc_sync_one(__va(pgd_paddr + PAGE_SIZE), address);
+#endif
+
pmd_k = vmalloc_sync_one(__va(pgd_paddr), address);
if (!pmd_k)
return -1;
-@@ -373,7 +437,14 @@ static noinline __kprobes int vmalloc_fault(unsigned long address)
+@@ -373,11 +443,25 @@ static noinline __kprobes int vmalloc_fault(unsigned long address)
* happen within a race in page table update. In the later
* case just flush:
*/
+- pgd = pgd_offset(current->active_mm, address);
+
+ pgd_ref = pgd_offset_k(address);
+ if (pgd_none(*pgd_ref))
+ return -1;
+
+#ifdef CONFIG_PAX_PER_CPU_PGD
-+ BUG_ON(__pa(get_cpu_pgd(smp_processor_id())) != (read_cr3() & PHYSICAL_PAGE_MASK));
-+ pgd = pgd_offset_cpu(smp_processor_id(), address);
++ BUG_ON(__pa(get_cpu_pgd(smp_processor_id(), kernel)) != (read_cr3() & __PHYSICAL_MASK));
++ pgd = pgd_offset_cpu(smp_processor_id(), user, address);
++ if (pgd_none(*pgd)) {
++ set_pgd(pgd, *pgd_ref);
++ arch_flush_lazy_mmu_mode();
++ } else {
++ BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref));
++ }
++ pgd = pgd_offset_cpu(smp_processor_id(), kernel, address);
+#else
- pgd = pgd_offset(current->active_mm, address);
++ pgd = pgd_offset(current->active_mm, address);
+#endif
+
- pgd_ref = pgd_offset_k(address);
- if (pgd_none(*pgd_ref))
- return -1;
-@@ -543,7 +614,7 @@ static int is_errata93(struct pt_regs *regs, unsigned long address)
+ if (pgd_none(*pgd)) {
+ set_pgd(pgd, *pgd_ref);
+ arch_flush_lazy_mmu_mode();
+@@ -543,7 +627,7 @@ static int is_errata93(struct pt_regs *regs, unsigned long address)
static int is_errata100(struct pt_regs *regs, unsigned long address)
{
#ifdef CONFIG_X86_64
return 1;
#endif
return 0;
-@@ -570,7 +641,7 @@ static int is_f00f_bug(struct pt_regs *regs, unsigned long address)
+@@ -570,7 +654,7 @@ static int is_f00f_bug(struct pt_regs *regs, unsigned long address)
}
static const char nx_warning[] = KERN_CRIT
static void
show_fault_oops(struct pt_regs *regs, unsigned long error_code,
-@@ -579,15 +650,27 @@ show_fault_oops(struct pt_regs *regs, unsigned long error_code,
+@@ -579,15 +663,27 @@ show_fault_oops(struct pt_regs *regs, unsigned long error_code,
if (!oops_may_print())
return;
printk(KERN_ALERT "BUG: unable to handle kernel ");
if (address < PAGE_SIZE)
printk(KERN_CONT "NULL pointer dereference");
-@@ -750,6 +833,22 @@ __bad_area_nosemaphore(struct pt_regs *regs, unsigned long error_code,
+@@ -750,6 +846,22 @@ __bad_area_nosemaphore(struct pt_regs *regs, unsigned long error_code,
return;
}
#endif
/* Kernel addresses are always protection faults: */
if (address >= TASK_SIZE)
error_code |= PF_PROT;
-@@ -835,7 +934,7 @@ do_sigbus(struct pt_regs *regs, unsigned long error_code, unsigned long address,
+@@ -835,7 +947,7 @@ do_sigbus(struct pt_regs *regs, unsigned long error_code, unsigned long address,
if (fault & (VM_FAULT_HWPOISON|VM_FAULT_HWPOISON_LARGE)) {
printk(KERN_ERR
"MCE: Killing %s:%d due to hardware memory corruption fault at %lx\n",
code = BUS_MCEERR_AR;
}
#endif
-@@ -898,6 +997,99 @@ static int spurious_fault_check(unsigned long error_code, pte_t *pte)
+@@ -898,6 +1010,99 @@ static int spurious_fault_check(unsigned long error_code, pte_t *pte)
return 1;
}
/*
* Handle a spurious fault caused by a stale TLB entry.
*
-@@ -964,6 +1156,9 @@ int show_unhandled_signals = 1;
+@@ -964,6 +1169,9 @@ int show_unhandled_signals = 1;
static inline int
access_error(unsigned long error_code, struct vm_area_struct *vma)
{
if (error_code & PF_WRITE) {
/* write, present and write, not present: */
if (unlikely(!(vma->vm_flags & VM_WRITE)))
-@@ -992,7 +1187,7 @@ static inline bool smap_violation(int error_code, struct pt_regs *regs)
+@@ -992,7 +1200,7 @@ static inline bool smap_violation(int error_code, struct pt_regs *regs)
if (error_code & PF_USER)
return false;
return false;
return true;
-@@ -1008,18 +1203,33 @@ __do_page_fault(struct pt_regs *regs, unsigned long error_code)
+@@ -1008,18 +1216,33 @@ __do_page_fault(struct pt_regs *regs, unsigned long error_code)
{
struct vm_area_struct *vma;
struct task_struct *tsk;
/*
* Detect and handle instructions that would cause a page fault for
-@@ -1080,7 +1290,7 @@ __do_page_fault(struct pt_regs *regs, unsigned long error_code)
+@@ -1080,7 +1303,7 @@ __do_page_fault(struct pt_regs *regs, unsigned long error_code)
* User-mode registers count as a user access even for any
* potential system fault or CPU buglet:
*/
local_irq_enable();
error_code |= PF_USER;
} else {
-@@ -1142,6 +1352,11 @@ retry:
+@@ -1142,6 +1365,11 @@ retry:
might_sleep();
}
vma = find_vma(mm, address);
if (unlikely(!vma)) {
bad_area(regs, error_code, address);
-@@ -1153,18 +1368,24 @@ retry:
+@@ -1153,18 +1381,24 @@ retry:
bad_area(regs, error_code, address);
return;
}
if (unlikely(expand_stack(vma, address))) {
bad_area(regs, error_code, address);
return;
-@@ -1230,3 +1451,292 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
+@@ -1230,3 +1464,292 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
__do_page_fault(regs, error_code);
exception_exit(prev_state);
}
#endif /*HAVE_ARCH_HUGETLB_UNMAPPED_AREA*/
diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c
-index 1f34e92..d252637 100644
+index 1f34e92..c97b98f 100644
--- a/arch/x86/mm/init.c
+++ b/arch/x86/mm/init.c
@@ -4,6 +4,7 @@
#include "mm_internal.h"
-@@ -465,7 +468,15 @@ void __init init_mem_mapping(void)
+@@ -465,7 +468,18 @@ void __init init_mem_mapping(void)
early_ioremap_page_table_range_init();
#endif
+#ifdef CONFIG_PAX_PER_CPU_PGD
-+ clone_pgd_range(get_cpu_pgd(0) + KERNEL_PGD_BOUNDARY,
++ clone_pgd_range(get_cpu_pgd(0, kernel) + KERNEL_PGD_BOUNDARY,
++ swapper_pg_dir + KERNEL_PGD_BOUNDARY,
++ KERNEL_PGD_PTRS);
++ clone_pgd_range(get_cpu_pgd(0, user) + KERNEL_PGD_BOUNDARY,
+ swapper_pg_dir + KERNEL_PGD_BOUNDARY,
+ KERNEL_PGD_PTRS);
-+ load_cr3(get_cpu_pgd(0));
++ load_cr3(get_cpu_pgd(0, kernel));
+#else
load_cr3(swapper_pg_dir);
+#endif
__flush_tlb_all();
early_memtest(0, max_pfn_mapped << PAGE_SHIFT);
-@@ -481,10 +492,40 @@ void __init init_mem_mapping(void)
+@@ -481,10 +495,40 @@ void __init init_mem_mapping(void)
* Access has to be given to non-kernel-ram areas as well, these contain the PCI
* mmio resources as well as potential bios/acpi data regions.
*/
if (iomem_is_exclusive(pagenr << PAGE_SHIFT))
return 0;
if (!page_is_ram(pagenr))
-@@ -538,8 +579,117 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
+@@ -538,8 +582,117 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
#endif
}
printk(KERN_INFO "Write protecting the kernel text: %luk\n",
size >> 10);
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
-index bb00c46..f31d2f0 100644
+index bb00c46..bf91a67 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -151,7 +151,7 @@ early_param("gbpages", parse_direct_gbpages_on);
EXPORT_SYMBOL_GPL(__supported_pte_mask);
int force_personality32;
-@@ -184,12 +184,22 @@ void sync_global_pgds(unsigned long start, unsigned long end)
+@@ -184,12 +184,29 @@ void sync_global_pgds(unsigned long start, unsigned long end)
for (address = start; address <= end; address += PGDIR_SIZE) {
const pgd_t *pgd_ref = pgd_offset_k(address);
+
+#ifdef CONFIG_PAX_PER_CPU_PGD
+ for (cpu = 0; cpu < nr_cpu_ids; ++cpu) {
-+ pgd_t *pgd = pgd_offset_cpu(cpu, address);
++ pgd_t *pgd = pgd_offset_cpu(cpu, user, address);
++
++ if (pgd_none(*pgd))
++ set_pgd(pgd, *pgd_ref);
++ else
++ BUG_ON(pgd_page_vaddr(*pgd)
++ != pgd_page_vaddr(*pgd_ref));
++ pgd = pgd_offset_cpu(cpu, kernel, address);
+#else
list_for_each_entry(page, &pgd_list, lru) {
pgd_t *pgd;
spinlock_t *pgt_lock;
-@@ -198,6 +208,7 @@ void sync_global_pgds(unsigned long start, unsigned long end)
+@@ -198,6 +215,7 @@ void sync_global_pgds(unsigned long start, unsigned long end)
/* the pgt_lock only for Xen */
pgt_lock = &pgd_page_get_mm(page)->page_table_lock;
spin_lock(pgt_lock);
if (pgd_none(*pgd))
set_pgd(pgd, *pgd_ref);
-@@ -205,7 +216,10 @@ void sync_global_pgds(unsigned long start, unsigned long end)
+@@ -205,7 +223,10 @@ void sync_global_pgds(unsigned long start, unsigned long end)
BUG_ON(pgd_page_vaddr(*pgd)
!= pgd_page_vaddr(*pgd_ref));
}
spin_unlock(&pgd_lock);
}
-@@ -238,7 +252,7 @@ static pud_t *fill_pud(pgd_t *pgd, unsigned long vaddr)
+@@ -238,7 +259,7 @@ static pud_t *fill_pud(pgd_t *pgd, unsigned long vaddr)
{
if (pgd_none(*pgd)) {
pud_t *pud = (pud_t *)spp_getpage();
if (pud != pud_offset(pgd, 0))
printk(KERN_ERR "PAGETABLE BUG #00! %p <-> %p\n",
pud, pud_offset(pgd, 0));
-@@ -250,7 +264,7 @@ static pmd_t *fill_pmd(pud_t *pud, unsigned long vaddr)
+@@ -250,7 +271,7 @@ static pmd_t *fill_pmd(pud_t *pud, unsigned long vaddr)
{
if (pud_none(*pud)) {
pmd_t *pmd = (pmd_t *) spp_getpage();
if (pmd != pmd_offset(pud, 0))
printk(KERN_ERR "PAGETABLE BUG #01! %p <-> %p\n",
pmd, pmd_offset(pud, 0));
-@@ -279,7 +293,9 @@ void set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte)
+@@ -279,7 +300,9 @@ void set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte)
pmd = fill_pmd(pud, vaddr);
pte = fill_pte(pmd, vaddr);
/*
* It's enough to flush this one mapping.
-@@ -338,14 +354,12 @@ static void __init __init_extra_mapping(unsigned long phys, unsigned long size,
+@@ -338,14 +361,12 @@ static void __init __init_extra_mapping(unsigned long phys, unsigned long size,
pgd = pgd_offset_k((unsigned long)__va(phys));
if (pgd_none(*pgd)) {
pud = (pud_t *) spp_getpage();
}
pmd = pmd_offset(pud, phys);
BUG_ON(!pmd_none(*pmd));
-@@ -586,7 +600,7 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end,
+@@ -586,7 +607,7 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end,
prot);
spin_lock(&init_mm.page_table_lock);
spin_unlock(&init_mm.page_table_lock);
}
__flush_tlb_all();
-@@ -627,7 +641,7 @@ kernel_physical_mapping_init(unsigned long start,
+@@ -627,7 +648,7 @@ kernel_physical_mapping_init(unsigned long start,
page_size_mask);
spin_lock(&init_mm.page_table_lock);
spin_unlock(&init_mm.page_table_lock);
pgd_changed = true;
}
-@@ -1221,8 +1235,8 @@ int kern_addr_valid(unsigned long addr)
+@@ -1221,8 +1242,8 @@ 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),
};
struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
-@@ -1256,7 +1270,7 @@ int in_gate_area_no_mm(unsigned long addr)
+@@ -1256,7 +1277,7 @@ int in_gate_area_no_mm(unsigned long addr)
const char *arch_vma_name(struct vm_area_struct *vma)
{
pte = kmemcheck_pte_lookup(address);
diff --git a/arch/x86/mm/mmap.c b/arch/x86/mm/mmap.c
-index 845df68..1d8d29f 100644
+index c1af323..4758dad 100644
--- a/arch/x86/mm/mmap.c
+++ b/arch/x86/mm/mmap.c
@@ -52,7 +52,7 @@ static unsigned int stack_maxrandom_size(void)
* 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)
+-unsigned long mmap_legacy_base(void)
++unsigned long mmap_legacy_base(struct mm_struct *mm)
{
- if (mmap_is_ia32())
+ if (mmap_is_ia32()) {
return TASK_UNMAPPED_BASE + mmap_rnd();
}
-@@ -113,11 +126,23 @@ static unsigned long mmap_legacy_base(void)
+@@ -113,11 +126,23 @@ unsigned long mmap_legacy_base(void)
void arch_pick_mmap_layout(struct mm_struct *mm)
{
if (mmap_is_legacy()) {
struct split_state {
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
-index bb32480..aef8278 100644
+index bb32480..75f2f5e 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -261,7 +261,7 @@ static inline pgprot_t static_protections(pgprot_t prot, unsigned long address,
+#ifdef CONFIG_PAX_PER_CPU_PGD
+ for (cpu = 0; cpu < nr_cpu_ids; ++cpu) {
-+ pgd_t *pgd = get_cpu_pgd(cpu);
++ pgd_t *pgd = get_cpu_pgd(cpu, kernel);
+#else
list_for_each_entry(page, &pgd_list, lru) {
- pgd_t *pgd;
p += get_opcode(p, &opcode);
for (i = 0; i < ARRAY_SIZE(imm_wop); i++)
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
-index 17fda6a..489c74a 100644
+index 17fda6a..f7d54a0 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
-@@ -91,10 +91,64 @@ static inline void pgd_list_del(pgd_t *pgd)
+@@ -91,10 +91,67 @@ static inline void pgd_list_del(pgd_t *pgd)
list_del(&page->lru);
}
+{
+ unsigned int count = USER_PGD_PTRS;
++ if (!pax_user_shadow_base)
++ return;
++
+ while (count--)
+ *dst++ = __pgd((pgd_val(*src++) | (_PAGE_NX & __supported_pte_mask)) & ~_PAGE_USER);
+}
static void pgd_set_mm(pgd_t *pgd, struct mm_struct *mm)
{
BUILD_BUG_ON(sizeof(virt_to_page(pgd)->index) < sizeof(mm));
-@@ -135,6 +189,7 @@ static void pgd_dtor(pgd_t *pgd)
+@@ -135,6 +192,7 @@ static void pgd_dtor(pgd_t *pgd)
pgd_list_del(pgd);
spin_unlock(&pgd_lock);
}
/*
* List of all pgd's needed for non-PAE so it can invalidate entries
-@@ -147,7 +202,7 @@ static void pgd_dtor(pgd_t *pgd)
+@@ -147,7 +205,7 @@ static void pgd_dtor(pgd_t *pgd)
* -- nyc
*/
/*
* In PAE mode, we need to do a cr3 reload (=tlb flush) when
* updating the top-level pagetable entries to guarantee the
-@@ -159,7 +214,7 @@ static void pgd_dtor(pgd_t *pgd)
+@@ -159,7 +217,7 @@ static void pgd_dtor(pgd_t *pgd)
* not shared between pagetables (!SHARED_KERNEL_PMDS), we allocate
* and initialize the kernel pmds here.
*/
void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd)
{
-@@ -177,36 +232,38 @@ void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd)
+@@ -177,36 +235,38 @@ void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd)
*/
flush_tlb_mm(mm);
}
return -ENOMEM;
}
-@@ -219,51 +276,55 @@ static int preallocate_pmds(pmd_t *pmds[])
+@@ -219,51 +279,55 @@ static int preallocate_pmds(pmd_t *pmds[])
* preallocate which never got a corresponding vma will need to be
* freed manually.
*/
pgd = (pgd_t *)__get_free_page(PGALLOC_GFP);
-@@ -272,11 +333,11 @@ pgd_t *pgd_alloc(struct mm_struct *mm)
+@@ -272,11 +336,11 @@ pgd_t *pgd_alloc(struct mm_struct *mm)
mm->pgd = pgd;
/*
* Make sure that pre-populating the pmds is atomic with
-@@ -286,14 +347,14 @@ pgd_t *pgd_alloc(struct mm_struct *mm)
+@@ -286,14 +350,14 @@ pgd_t *pgd_alloc(struct mm_struct *mm)
spin_lock(&pgd_lock);
pgd_ctor(mm, pgd);
out_free_pgd:
free_page((unsigned long)pgd);
out:
-@@ -302,7 +363,7 @@ out:
+@@ -302,7 +366,7 @@ out:
void pgd_free(struct mm_struct *mm, pgd_t *pgd)
{
}
}
EXPORT_SYMBOL_GPL(leave_mm);
+diff --git a/arch/x86/mm/uderef_64.c b/arch/x86/mm/uderef_64.c
+new file mode 100644
+index 0000000..dace51c
+--- /dev/null
++++ b/arch/x86/mm/uderef_64.c
+@@ -0,0 +1,37 @@
++#include <linux/mm.h>
++#include <asm/pgtable.h>
++#include <asm/uaccess.h>
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++/* PaX: due to the special call convention these functions must
++ * - remain leaf functions under all configurations,
++ * - never be called directly, only dereferenced from the wrappers.
++ */
++void __pax_open_userland(void)
++{
++ unsigned int cpu;
++
++ if (unlikely(!segment_eq(get_fs(), USER_DS)))
++ return;
++
++ cpu = raw_get_cpu();
++ BUG_ON((read_cr3() & ~PAGE_MASK) != PCID_KERNEL);
++ write_cr3(__pa(get_cpu_pgd(cpu, user)) | PCID_USER | PCID_NOFLUSH);
++ raw_put_cpu_no_resched();
++}
++EXPORT_SYMBOL(__pax_open_userland);
++
++void __pax_close_userland(void)
++{
++ unsigned int cpu;
++
++ if (unlikely(!segment_eq(get_fs(), USER_DS)))
++ return;
++
++ cpu = raw_get_cpu();
++ BUG_ON((read_cr3() & ~PAGE_MASK) != PCID_USER);
++ write_cr3(__pa(get_cpu_pgd(cpu, kernel)) | PCID_KERNEL | PCID_NOFLUSH);
++ raw_put_cpu_no_resched();
++}
++EXPORT_SYMBOL(__pax_close_userland);
++#endif
diff --git a/arch/x86/net/bpf_jit.S b/arch/x86/net/bpf_jit.S
index 877b9a1..a8ecf42 100644
--- a/arch/x86/net/bpf_jit.S
}
EXPORT_SYMBOL(pcibios_set_irq_routing);
diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c
-index 40e4469..0592924 100644
+index 40e4469..d915bf9 100644
--- a/arch/x86/platform/efi/efi_32.c
+++ b/arch/x86/platform/efi/efi_32.c
@@ -44,11 +44,22 @@ void efi_call_phys_prelog(void)
load_gdt(&gdt_descr);
+#ifdef CONFIG_PAX_PER_CPU_PGD
-+ load_cr3(get_cpu_pgd(smp_processor_id()));
++ load_cr3(get_cpu_pgd(smp_processor_id(), kernel));
+#else
load_cr3(swapper_pg_dir);
+#endif
local_irq_restore(efi_rt_eflags);
diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c
-index 39a0e7f1..ecc2f1e 100644
+index 39a0e7f1..872396e 100644
--- a/arch/x86/platform/efi/efi_64.c
+++ b/arch/x86/platform/efi/efi_64.c
@@ -76,6 +76,11 @@ void __init efi_call_phys_prelog(void)
kfree(save_pgd);
+
+#ifdef CONFIG_PAX_PER_CPU_PGD
-+ load_cr3(get_cpu_pgd(smp_processor_id()));
++ load_cr3(get_cpu_pgd(smp_processor_id(), kernel));
+#endif
+
__flush_tlb_all();
END(trampoline_header)
diff --git a/arch/x86/realmode/rm/trampoline_64.S b/arch/x86/realmode/rm/trampoline_64.S
-index bb360dc..3e5945f 100644
+index bb360dc..d0fd8f8 100644
--- a/arch/x86/realmode/rm/trampoline_64.S
+++ b/arch/x86/realmode/rm/trampoline_64.S
-@@ -107,7 +107,7 @@ ENTRY(startup_32)
+@@ -94,6 +94,7 @@ ENTRY(startup_32)
+ movl %edx, %gs
+
+ movl pa_tr_cr4, %eax
++ andl $~X86_CR4_PCIDE, %eax
+ movl %eax, %cr4 # Enable PAE mode
+
+ # Setup trampoline 4 level pagetables
+@@ -107,7 +108,7 @@ ENTRY(startup_32)
wrmsr
# Enable paging and in turn activate Long Mode
return 0;
}
diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
-index 1b456fe..2510242 100644
+index fc45567..fa2a590 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
-@@ -679,7 +679,7 @@ static ssize_t fill_readbuf(struct port *port, char *out_buf, size_t out_count,
+@@ -682,7 +682,7 @@ static ssize_t fill_readbuf(struct port *port, char *out_buf, size_t out_count,
if (to_user) {
ssize_t ret;
if (ret)
return -EFAULT;
} else {
-@@ -778,7 +778,7 @@ static ssize_t port_fops_read(struct file *filp, char __user *ubuf,
+@@ -785,7 +785,7 @@ static ssize_t port_fops_read(struct file *filp, char __user *ubuf,
if (!port_has_data(port) && !port->host_connected)
return 0;
.notifier_call = arch_timer_cpu_notify,
};
+diff --git a/drivers/clocksource/bcm_kona_timer.c b/drivers/clocksource/bcm_kona_timer.c
+index 350f493..489479e 100644
+--- a/drivers/clocksource/bcm_kona_timer.c
++++ b/drivers/clocksource/bcm_kona_timer.c
+@@ -199,7 +199,7 @@ static struct irqaction kona_timer_irq = {
+ .handler = kona_timer_interrupt,
+ };
+
+-static void __init kona_timer_init(void)
++static void __init kona_timer_init(struct device_node *np)
+ {
+ kona_timers_init();
+ kona_timer_clockevents_init();
diff --git a/drivers/clocksource/metag_generic.c b/drivers/clocksource/metag_generic.c
index ade7513..069445f 100644
--- a/drivers/clocksource/metag_generic.c
pr_debug("CPU%u - ACPI performance management activated.\n", cpu);
for (i = 0; i < perf->state_count; i++)
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
-index 178fe7a..5ee8501 100644
+index 6485547..477033e 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
-@@ -1853,7 +1853,7 @@ static int __cpuinit cpufreq_cpu_callback(struct notifier_block *nfb,
+@@ -1854,7 +1854,7 @@ static int __cpuinit cpufreq_cpu_callback(struct notifier_block *nfb,
return NOTIFY_OK;
}
.notifier_call = cpufreq_cpu_callback,
};
-@@ -1885,8 +1885,11 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
+@@ -1886,8 +1886,11 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
pr_debug("trying to register driver %s\n", driver_data->name);
write_lock_irqsave(&cpufreq_driver_lock, flags);
if (cpufreq_driver) {
diff --git a/drivers/cpufreq/cpufreq_governor.c b/drivers/cpufreq/cpufreq_governor.c
-index 5af40ad..ddf907b 100644
+index a86ff72..aad2b03 100644
--- a/drivers/cpufreq/cpufreq_governor.c
+++ b/drivers/cpufreq/cpufreq_governor.c
@@ -235,7 +235,7 @@ int cpufreq_governor_dbs(struct cpufreq_policy *policy,
cpufreq_unregister_notifier(cs_ops->notifier_block,
CPUFREQ_TRANSITION_NOTIFIER);
diff --git a/drivers/cpufreq/cpufreq_governor.h b/drivers/cpufreq/cpufreq_governor.h
-index e16a961..0e68927 100644
+index 0d9e6be..461fd3b 100644
--- a/drivers/cpufreq/cpufreq_governor.h
+++ b/drivers/cpufreq/cpufreq_governor.h
@@ -204,7 +204,7 @@ struct common_dbs_data {
/* Governer Per policy data */
diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
-index 93eb5cb..f8ab572 100644
+index c087347..dad6268 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -615,14 +615,18 @@ void od_register_powersave_bias_handler(unsigned int (*f)
if (IS_GEN6(dev) || IS_GEN7(dev)) {
seq_printf(m,
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
-index f968590..19115e35 100644
+index 17d9b0b..860e6d9 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1259,7 +1259,7 @@ static bool i915_switcheroo_can_switch(struct pci_dev *pdev)
iir = I915_READ(IIR);
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
-index e1f4e6e..c94a4b3 100644
+index eea5982..eeef407 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
-@@ -8933,13 +8933,13 @@ struct intel_quirk {
+@@ -8935,13 +8935,13 @@ struct intel_quirk {
int subsystem_vendor;
int subsystem_device;
void (*hook)(struct drm_device *dev);
static int intel_dmi_reverse_brightness(const struct dmi_system_id *id)
{
-@@ -8947,18 +8947,20 @@ static int intel_dmi_reverse_brightness(const struct dmi_system_id *id)
+@@ -8949,18 +8949,20 @@ static int intel_dmi_reverse_brightness(const struct dmi_system_id *id)
return 1;
}
vlan_req = (struct qlcnic_vlan_req *)&req->words[1];
vlan_req->vlan_id = cpu_to_le16(vlan_id);
+diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c
+index 887aebe..9095ff9 100644
+--- a/drivers/net/ethernet/realtek/8139cp.c
++++ b/drivers/net/ethernet/realtek/8139cp.c
+@@ -524,6 +524,7 @@ rx_status_loop:
+ PCI_DMA_FROMDEVICE);
+ if (dma_mapping_error(&cp->pdev->dev, new_mapping)) {
+ dev->stats.rx_dropped++;
++ kfree_skb(new_skb);
+ goto rx_next;
+ }
+
diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index 393f961..d343034 100644
--- a/drivers/net/ethernet/realtek/r8169.c
};
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
-index 2491eb2..694b2ec 100644
+index 2491eb2..1a453eb 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
-@@ -1869,7 +1869,7 @@ unlock:
+@@ -1076,8 +1076,9 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
+ u32 rxhash;
+
+ if (!(tun->flags & TUN_NO_PI)) {
+- if ((len -= sizeof(pi)) > total_len)
++ if (len < sizeof(pi))
+ return -EINVAL;
++ len -= sizeof(pi);
+
+ if (memcpy_fromiovecend((void *)&pi, iv, 0, sizeof(pi)))
+ return -EFAULT;
+@@ -1085,8 +1086,9 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
+ }
+
+ if (tun->flags & TUN_VNET_HDR) {
+- if ((len -= tun->vnet_hdr_sz) > total_len)
++ if (len < tun->vnet_hdr_sz)
+ return -EINVAL;
++ len -= tun->vnet_hdr_sz;
+
+ if (memcpy_fromiovecend((void *)&gso, iv, offset, sizeof(gso)))
+ return -EFAULT;
+@@ -1869,7 +1871,7 @@ unlock:
}
static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
{
struct tun_file *tfile = file->private_data;
struct tun_struct *tun;
-@@ -1881,6 +1881,9 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
+@@ -1881,6 +1883,9 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
int vnet_hdr_sz;
int ret;
static inline struct rt2x00_intf* vif_to_intf(struct ieee80211_vif *vif)
diff --git a/drivers/net/wireless/rt2x00/rt2x00queue.c b/drivers/net/wireless/rt2x00/rt2x00queue.c
-index 2c12311..7b77f24 100644
+index d955741..8730748 100644
--- a/drivers/net/wireless/rt2x00/rt2x00queue.c
+++ b/drivers/net/wireless/rt2x00/rt2x00queue.c
@@ -252,9 +252,9 @@ static void rt2x00queue_create_tx_descriptor_seq(struct rt2x00_dev *rt2x00dev,
ddb_entry->default_relogin_timeout =
(def_timeout > LOGIN_TOV) && (def_timeout < LOGIN_TOV * 10) ?
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
-index 3b1ea34..1583a72 100644
+index eaa808e..95f8841 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -661,7 +661,7 @@ int scsi_dispatch_cmd(struct scsi_cmnd *cmd)
pDevice->apdev->netdev_ops = &apdev_netdev_ops;
pDevice->apdev->type = ARPHRD_IEEE80211;
-diff --git a/drivers/staging/zcache/tmem.c b/drivers/staging/zcache/tmem.c
-index d7e51e4..d07eaab 100644
---- a/drivers/staging/zcache/tmem.c
-+++ b/drivers/staging/zcache/tmem.c
-@@ -51,7 +51,7 @@
- * A tmem host implementation must use this function to register callbacks
- * for memory allocation.
- */
--static struct tmem_hostops tmem_hostops;
-+static tmem_hostops_no_const tmem_hostops;
-
- static void tmem_objnode_tree_init(void);
-
-@@ -65,7 +65,7 @@ void tmem_register_hostops(struct tmem_hostops *m)
- * A tmem host implementation must use this function to register
- * callbacks for a page-accessible memory (PAM) implementation.
- */
--static struct tmem_pamops tmem_pamops;
-+static tmem_pamops_no_const tmem_pamops;
-
- void tmem_register_pamops(struct tmem_pamops *m)
- {
diff --git a/drivers/staging/zcache/tmem.h b/drivers/staging/zcache/tmem.h
-index d128ce2..a43980c 100644
+index d128ce2..fc1f9a1 100644
--- a/drivers/staging/zcache/tmem.h
+++ b/drivers/staging/zcache/tmem.h
-@@ -226,6 +226,7 @@ struct tmem_pamops {
+@@ -225,7 +225,7 @@ struct tmem_pamops {
+ bool (*is_remote)(void *);
int (*replace_in_obj)(void *, struct tmem_obj *);
#endif
- };
-+typedef struct tmem_pamops __no_const tmem_pamops_no_const;
+-};
++} __no_const;
extern void tmem_register_pamops(struct tmem_pamops *m);
/* memory allocation methods provided by the host implementation */
-@@ -235,6 +236,7 @@ struct tmem_hostops {
+@@ -234,7 +234,7 @@ struct tmem_hostops {
+ void (*obj_free)(struct tmem_obj *, struct tmem_pool *);
struct tmem_objnode *(*objnode_alloc)(struct tmem_pool *);
void (*objnode_free)(struct tmem_objnode *, struct tmem_pool *);
- };
-+typedef struct tmem_hostops __no_const tmem_hostops_no_const;
+-};
++} __no_const;
extern void tmem_register_hostops(struct tmem_hostops *m);
/* core tmem accessor functions */
/**
diff --git a/drivers/usb/wusbcore/wa-xfer.c b/drivers/usb/wusbcore/wa-xfer.c
-index 6ef94bc..1b41265 100644
+index 028fc83..65bb105 100644
--- a/drivers/usb/wusbcore/wa-xfer.c
+++ b/drivers/usb/wusbcore/wa-xfer.c
@@ -296,7 +296,7 @@ out:
int err;
u32 ftype;
struct ceph_mds_reply_info_parsed *rinfo;
+diff --git a/fs/ceph/super.c b/fs/ceph/super.c
+index 7d377c9..3fb6559 100644
+--- a/fs/ceph/super.c
++++ b/fs/ceph/super.c
+@@ -839,7 +839,7 @@ static int ceph_compare_super(struct super_block *sb, void *data)
+ /*
+ * construct our own bdi so we can control readahead, etc.
+ */
+-static atomic_long_t bdi_seq = ATOMIC_LONG_INIT(0);
++static atomic_long_unchecked_t bdi_seq = ATOMIC_LONG_INIT(0);
+
+ static int ceph_register_bdi(struct super_block *sb,
+ struct ceph_fs_client *fsc)
+@@ -856,7 +856,7 @@ static int ceph_register_bdi(struct super_block *sb,
+ default_backing_dev_info.ra_pages;
+
+ err = bdi_register(&fsc->backing_dev_info, NULL, "ceph-%ld",
+- atomic_long_inc_return(&bdi_seq));
++ atomic_long_inc_return_unchecked(&bdi_seq));
+ if (!err)
+ sb->s_bdi = &fsc->backing_dev_info;
+ return err;
diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c
index d597483..747901b 100644
--- a/fs/cifs/cifs_debug.c
atomic_set(&midCount, 0);
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
-index 4f07f6f..55de8ce 100644
+index ea3a0b3..0194e39 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
-@@ -751,35 +751,35 @@ struct cifs_tcon {
+@@ -752,35 +752,35 @@ struct cifs_tcon {
__u16 Flags; /* optional support bits */
enum statusEnum tidStatus;
#ifdef CONFIG_CIFS_STATS
} smb2_stats;
#endif /* CONFIG_CIFS_SMB2 */
} stats;
-@@ -1080,7 +1080,7 @@ convert_delimiter(char *path, char delim)
+@@ -1081,7 +1081,7 @@ convert_delimiter(char *path, char delim)
}
#ifdef CONFIG_CIFS_STATS
static inline void cifs_stats_bytes_written(struct cifs_tcon *tcon,
unsigned int bytes)
-@@ -1445,8 +1445,8 @@ GLOBAL_EXTERN atomic_t tconInfoReconnectCount;
+@@ -1446,8 +1446,8 @@ GLOBAL_EXTERN atomic_t tconInfoReconnectCount;
/* Various Debug counters */
GLOBAL_EXTERN atomic_t bufAllocCount; /* current number allocated */
#ifdef CONFIG_CIFS_STATS2
dcache_init();
inode_init();
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
-index 4888cb3..e0f7cf8 100644
+index c7c83ff..bda9461 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -415,7 +415,11 @@ EXPORT_SYMBOL_GPL(debugfs_create_file);
i += packet_length_size;
if (copy_to_user(&buf[i], msg_ctx->msg, msg_ctx->msg_size))
diff --git a/fs/exec.c b/fs/exec.c
-index ffd7a81..d95acf6 100644
+index 1f44670..3c84660 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -55,8 +55,20 @@
out:
if (bprm->mm) {
acct_arg_size(bprm, 0);
-@@ -1701,3 +1875,285 @@ asmlinkage long compat_sys_execve(const char __user * filename,
+@@ -1701,3 +1875,287 @@ asmlinkage long compat_sys_execve(const char __user * filename,
return error;
}
#endif
+ offset = vma_fault->vm_pgoff << PAGE_SHIFT;
+ if (vma_fault->vm_file)
+ path_fault = pax_get_path(&vma_fault->vm_file->f_path, buffer_fault, PAGE_SIZE);
-+ else if (pc >= mm->start_brk && pc < mm->brk)
++ else if ((unsigned long)pc >= mm->start_brk && (unsigned long)pc < mm->brk)
+ path_fault = "<heap>";
+ else if (vma_fault->vm_flags & (VM_GROWSDOWN | VM_GROWSUP))
+ path_fault = "<stack>";
+ printk(KERN_ERR "PAX: refcount overflow detected in: %s:%d, uid/euid: %u/%u\n", current->comm, task_pid_nr(current),
+ from_kuid_munged(&init_user_ns, current_uid()), from_kuid_munged(&init_user_ns, current_euid()));
+ print_symbol(KERN_ERR "PAX: refcount overflow occured at: %s\n", instruction_pointer(regs));
++ preempt_disable();
+ show_regs(regs);
++ preempt_enable();
+ force_sig_info(SIGKILL, SEND_SIG_FORCED, current);
+}
+#endif
else if (input->reserved_blocks > input->blocks_count / 5)
ext4_warning(sb, "Reserved blocks too high (%u)",
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
-index 6681c03..d88cd33 100644
+index 3f7c39e..227f24f 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1236,7 +1236,7 @@ static ext4_fsblk_t get_sb_block(void **data)
next->d_inode->i_ino,
dt_type(next->d_inode)) < 0)
diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
-index 9760ecb..9b838ef 100644
+index acd3947..1f896e2 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -36,11 +36,11 @@ static const struct rpc_call_ops nlmclnt_cancel_ops;
}
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
-index 6c80083..a1e6299 100644
+index 77cc85d..a1e6299 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
-@@ -122,6 +122,7 @@ static int fill_event_metadata(struct fsnotify_group *group,
- metadata->event_len = FAN_EVENT_METADATA_LEN;
- metadata->metadata_len = FAN_EVENT_METADATA_LEN;
- metadata->vers = FANOTIFY_METADATA_VERSION;
-+ metadata->reserved = 0;
- metadata->mask = event->mask & FAN_ALL_OUTGOING_EVENTS;
- metadata->pid = pid_vnr(event->tgid);
- if (unlikely(event->mask & FAN_Q_OVERFLOW))
-@@ -252,8 +253,8 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,
+@@ -253,8 +253,8 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,
fd = fanotify_event_metadata.fd;
ret = -EFAULT;
-const struct inode_operations ntfs_empty_inode_ops = {};
+const struct inode_operations ntfs_empty_inode_ops __read_only;
+diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
+index 20dfec7..e238cb7 100644
+--- a/fs/ocfs2/aops.c
++++ b/fs/ocfs2/aops.c
+@@ -1756,7 +1756,7 @@ try_again:
+ goto out;
+ } else if (ret == 1) {
+ clusters_need = wc->w_clen;
+- ret = ocfs2_refcount_cow(inode, filp, di_bh,
++ ret = ocfs2_refcount_cow(inode, di_bh,
+ wc->w_cpos, wc->w_clen, UINT_MAX);
+ if (ret) {
+ mlog_errno(ret);
+diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
+index ff54014..ff125fd 100644
+--- a/fs/ocfs2/file.c
++++ b/fs/ocfs2/file.c
+@@ -370,7 +370,7 @@ static int ocfs2_cow_file_pos(struct inode *inode,
+ if (!(ext_flags & OCFS2_EXT_REFCOUNTED))
+ goto out;
+
+- return ocfs2_refcount_cow(inode, NULL, fe_bh, cpos, 1, cpos+1);
++ return ocfs2_refcount_cow(inode, fe_bh, cpos, 1, cpos+1);
+
+ out:
+ return status;
+@@ -899,7 +899,7 @@ static int ocfs2_zero_extend_get_range(struct inode *inode,
+ zero_clusters = last_cpos - zero_cpos;
+
+ if (needs_cow) {
+- rc = ocfs2_refcount_cow(inode, NULL, di_bh, zero_cpos,
++ rc = ocfs2_refcount_cow(inode, di_bh, zero_cpos,
+ zero_clusters, UINT_MAX);
+ if (rc) {
+ mlog_errno(rc);
+@@ -2078,7 +2078,7 @@ static int ocfs2_prepare_inode_for_refcount(struct inode *inode,
+
+ *meta_level = 1;
+
+- ret = ocfs2_refcount_cow(inode, file, di_bh, cpos, clusters, UINT_MAX);
++ ret = ocfs2_refcount_cow(inode, di_bh, cpos, clusters, UINT_MAX);
+ if (ret)
+ mlog_errno(ret);
+ out:
diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c
index aebeacd..0dcdd26 100644
--- a/fs/ocfs2/localalloc.c
bail:
if (handle)
+diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c
+index f1fc172..452068b 100644
+--- a/fs/ocfs2/move_extents.c
++++ b/fs/ocfs2/move_extents.c
+@@ -69,7 +69,7 @@ static int __ocfs2_move_extent(handle_t *handle,
+ u64 ino = ocfs2_metadata_cache_owner(context->et.et_ci);
+ u64 old_blkno = ocfs2_clusters_to_blocks(inode->i_sb, p_cpos);
+
+- ret = ocfs2_duplicate_clusters_by_page(handle, context->file, cpos,
++ ret = ocfs2_duplicate_clusters_by_page(handle, inode, cpos,
+ p_cpos, new_p_cpos, len);
+ if (ret) {
+ mlog_errno(ret);
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index d355e6e..578d905 100644
--- a/fs/ocfs2/ocfs2.h
};
enum ocfs2_local_alloc_state
+diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
+index 998b17e..aefe414 100644
+--- a/fs/ocfs2/refcounttree.c
++++ b/fs/ocfs2/refcounttree.c
+@@ -49,7 +49,6 @@
+
+ struct ocfs2_cow_context {
+ struct inode *inode;
+- struct file *file;
+ u32 cow_start;
+ u32 cow_len;
+ struct ocfs2_extent_tree data_et;
+@@ -66,7 +65,7 @@ struct ocfs2_cow_context {
+ u32 *num_clusters,
+ unsigned int *extent_flags);
+ int (*cow_duplicate_clusters)(handle_t *handle,
+- struct file *file,
++ struct inode *inode,
+ u32 cpos, u32 old_cluster,
+ u32 new_cluster, u32 new_len);
+ };
+@@ -2922,14 +2921,12 @@ static int ocfs2_clear_cow_buffer(handle_t *handle, struct buffer_head *bh)
+ }
+
+ int ocfs2_duplicate_clusters_by_page(handle_t *handle,
+- struct file *file,
++ struct inode *inode,
+ u32 cpos, u32 old_cluster,
+ u32 new_cluster, u32 new_len)
+ {
+ int ret = 0, partial;
+- struct inode *inode = file_inode(file);
+- struct ocfs2_caching_info *ci = INODE_CACHE(inode);
+- struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
++ struct super_block *sb = inode->i_sb;
+ u64 new_block = ocfs2_clusters_to_blocks(sb, new_cluster);
+ struct page *page;
+ pgoff_t page_index;
+@@ -2973,13 +2970,6 @@ int ocfs2_duplicate_clusters_by_page(handle_t *handle,
+ if (PAGE_CACHE_SIZE <= OCFS2_SB(sb)->s_clustersize)
+ BUG_ON(PageDirty(page));
+
+- if (PageReadahead(page)) {
+- page_cache_async_readahead(mapping,
+- &file->f_ra, file,
+- page, page_index,
+- readahead_pages);
+- }
+-
+ if (!PageUptodate(page)) {
+ ret = block_read_full_page(page, ocfs2_get_block);
+ if (ret) {
+@@ -2999,7 +2989,8 @@ int ocfs2_duplicate_clusters_by_page(handle_t *handle,
+ }
+ }
+
+- ocfs2_map_and_dirty_page(inode, handle, from, to,
++ ocfs2_map_and_dirty_page(inode,
++ handle, from, to,
+ page, 0, &new_block);
+ mark_page_accessed(page);
+ unlock:
+@@ -3015,12 +3006,11 @@ unlock:
+ }
+
+ int ocfs2_duplicate_clusters_by_jbd(handle_t *handle,
+- struct file *file,
++ struct inode *inode,
+ u32 cpos, u32 old_cluster,
+ u32 new_cluster, u32 new_len)
+ {
+ int ret = 0;
+- struct inode *inode = file_inode(file);
+ struct super_block *sb = inode->i_sb;
+ struct ocfs2_caching_info *ci = INODE_CACHE(inode);
+ int i, blocks = ocfs2_clusters_to_blocks(sb, new_len);
+@@ -3145,7 +3135,7 @@ static int ocfs2_replace_clusters(handle_t *handle,
+
+ /*If the old clusters is unwritten, no need to duplicate. */
+ if (!(ext_flags & OCFS2_EXT_UNWRITTEN)) {
+- ret = context->cow_duplicate_clusters(handle, context->file,
++ ret = context->cow_duplicate_clusters(handle, context->inode,
+ cpos, old, new, len);
+ if (ret) {
+ mlog_errno(ret);
+@@ -3423,35 +3413,12 @@ static int ocfs2_replace_cow(struct ocfs2_cow_context *context)
+ return ret;
+ }
+
+-static void ocfs2_readahead_for_cow(struct inode *inode,
+- struct file *file,
+- u32 start, u32 len)
+-{
+- struct address_space *mapping;
+- pgoff_t index;
+- unsigned long num_pages;
+- int cs_bits = OCFS2_SB(inode->i_sb)->s_clustersize_bits;
+-
+- if (!file)
+- return;
+-
+- mapping = file->f_mapping;
+- num_pages = (len << cs_bits) >> PAGE_CACHE_SHIFT;
+- if (!num_pages)
+- num_pages = 1;
+-
+- index = ((loff_t)start << cs_bits) >> PAGE_CACHE_SHIFT;
+- page_cache_sync_readahead(mapping, &file->f_ra, file,
+- index, num_pages);
+-}
+-
+ /*
+ * Starting at cpos, try to CoW write_len clusters. Don't CoW
+ * past max_cpos. This will stop when it runs into a hole or an
+ * unrefcounted extent.
+ */
+ static int ocfs2_refcount_cow_hunk(struct inode *inode,
+- struct file *file,
+ struct buffer_head *di_bh,
+ u32 cpos, u32 write_len, u32 max_cpos)
+ {
+@@ -3480,8 +3447,6 @@ static int ocfs2_refcount_cow_hunk(struct inode *inode,
+
+ BUG_ON(cow_len == 0);
+
+- ocfs2_readahead_for_cow(inode, file, cow_start, cow_len);
+-
+ context = kzalloc(sizeof(struct ocfs2_cow_context), GFP_NOFS);
+ if (!context) {
+ ret = -ENOMEM;
+@@ -3503,7 +3468,6 @@ static int ocfs2_refcount_cow_hunk(struct inode *inode,
+ context->ref_root_bh = ref_root_bh;
+ context->cow_duplicate_clusters = ocfs2_duplicate_clusters_by_page;
+ context->get_clusters = ocfs2_di_get_clusters;
+- context->file = file;
+
+ ocfs2_init_dinode_extent_tree(&context->data_et,
+ INODE_CACHE(inode), di_bh);
+@@ -3532,7 +3496,6 @@ out:
+ * clusters between cpos and cpos+write_len are safe to modify.
+ */
+ int ocfs2_refcount_cow(struct inode *inode,
+- struct file *file,
+ struct buffer_head *di_bh,
+ u32 cpos, u32 write_len, u32 max_cpos)
+ {
+@@ -3552,7 +3515,7 @@ int ocfs2_refcount_cow(struct inode *inode,
+ num_clusters = write_len;
+
+ if (ext_flags & OCFS2_EXT_REFCOUNTED) {
+- ret = ocfs2_refcount_cow_hunk(inode, file, di_bh, cpos,
++ ret = ocfs2_refcount_cow_hunk(inode, di_bh, cpos,
+ num_clusters, max_cpos);
+ if (ret) {
+ mlog_errno(ret);
+diff --git a/fs/ocfs2/refcounttree.h b/fs/ocfs2/refcounttree.h
+index 7754608..6422bbcdb 100644
+--- a/fs/ocfs2/refcounttree.h
++++ b/fs/ocfs2/refcounttree.h
+@@ -53,7 +53,7 @@ int ocfs2_prepare_refcount_change_for_del(struct inode *inode,
+ int *credits,
+ int *ref_blocks);
+ int ocfs2_refcount_cow(struct inode *inode,
+- struct file *filep, struct buffer_head *di_bh,
++ struct buffer_head *di_bh,
+ u32 cpos, u32 write_len, u32 max_cpos);
+
+ typedef int (ocfs2_post_refcount_func)(struct inode *inode,
+@@ -85,11 +85,11 @@ int ocfs2_refcount_cow_xattr(struct inode *inode,
+ u32 cpos, u32 write_len,
+ struct ocfs2_post_refcount *post);
+ int ocfs2_duplicate_clusters_by_page(handle_t *handle,
+- struct file *file,
++ struct inode *inode,
+ u32 cpos, u32 old_cluster,
+ u32 new_cluster, u32 new_len);
+ int ocfs2_duplicate_clusters_by_jbd(handle_t *handle,
+- struct file *file,
++ struct inode *inode,
+ u32 cpos, u32 old_cluster,
+ u32 new_cluster, u32 new_len);
+ int ocfs2_cow_sync_writeback(struct super_block *sb,
diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index b7e74b5..19c6536 100644
--- a/fs/ocfs2/suballoc.c
kfree(s);
}
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
-index 3e636d8..83e3b71 100644
+index 65fc60a..350cc48 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -11,12 +11,19 @@
/* balance leaf returns 0 except if combining L R and S into
diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c
-index 33532f7..4846ade 100644
+index 1d48974..2f8f4e0 100644
--- a/fs/reiserfs/procfs.c
+++ b/fs/reiserfs/procfs.c
-@@ -112,7 +112,7 @@ static int show_super(struct seq_file *m, struct super_block *sb)
+@@ -114,7 +114,7 @@ static int show_super(struct seq_file *m, void *unused)
"SMALL_TAILS " : "NO_TAILS ",
replay_only(sb) ? "REPLAY_ONLY " : "",
convert_reiserfs(sb) ? "CONV " : "",
#endif /* CONFIG_MMU */
#endif /* !__ASSEMBLY__ */
+diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h
+index c184aa8..d049942 100644
+--- a/include/asm-generic/uaccess.h
++++ b/include/asm-generic/uaccess.h
+@@ -343,4 +343,12 @@ clear_user(void __user *to, unsigned long n)
+ return __clear_user(to, n);
+ }
+
++#ifndef __HAVE_ARCH_PAX_OPEN_USERLAND
++//static inline unsigned long pax_open_userland(void) { return 0; }
++#endif
++
++#ifndef __HAVE_ARCH_PAX_CLOSE_USERLAND
++//static inline unsigned long pax_close_userland(void) { return 0; }
++#endif
++
+ #endif /* __ASM_GENERIC_UACCESS_H */
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index eb58d2d..df131bf 100644
--- a/include/asm-generic/vmlinux.lds.h
/**
* struct clk_init_data - holds init data that's common to all clocks and is
diff --git a/include/linux/compat.h b/include/linux/compat.h
-index 7f0c1dd..b5729c6 100644
+index 7f0c1dd..206ac34 100644
--- a/include/linux/compat.h
+++ b/include/linux/compat.h
@@ -312,7 +312,7 @@ compat_sys_get_robust_list(int pid, compat_uptr_t __user *head_ptr,
asmlinkage long compat_sys_lookup_dcookie(u32, u32, char __user *, size_t);
/*
+@@ -669,6 +669,7 @@ asmlinkage long compat_sys_sigaltstack(const compat_stack_t __user *uss_ptr,
+
+ int compat_restore_altstack(const compat_stack_t __user *uss);
+ int __compat_save_altstack(compat_stack_t __user *, unsigned long);
++void __compat_save_altstack_ex(compat_stack_t __user *, unsigned long);
+
+ asmlinkage long compat_sys_sched_rr_get_interval(compat_pid_t pid,
+ struct compat_timespec __user *interval);
diff --git a/include/linux/compiler-gcc4.h b/include/linux/compiler-gcc4.h
index 842de22..7f3a41f 100644
--- a/include/linux/compiler-gcc4.h
void gic_init_bases(unsigned int, int, void __iomem *, void __iomem *,
u32 offset, struct device_node *);
diff --git a/include/linux/kallsyms.h b/include/linux/kallsyms.h
-index 6883e19..06992b1 100644
+index 6883e19..e854fcb 100644
--- a/include/linux/kallsyms.h
+++ b/include/linux/kallsyms.h
@@ -15,7 +15,8 @@
/* Lookup the address for a symbol. Returns 0 if not found. */
unsigned long kallsyms_lookup_name(const char *name);
-@@ -106,6 +107,17 @@ static inline int lookup_symbol_attrs(unsigned long addr, unsigned long *size, u
+@@ -106,6 +107,21 @@ static inline int lookup_symbol_attrs(unsigned long addr, unsigned long *size, u
/* Stupid that this does nothing, but I didn't create this mess. */
#define __print_symbol(fmt, addr)
#endif /*CONFIG_KALLSYMS*/
-+#else /* when included by kallsyms.c, vsnprintf.c, or
++#else /* when included by kallsyms.c, vsnprintf.c, kprobes.c, or
+ arch/x86/kernel/dumpstack.c, with HIDESYM enabled */
++extern unsigned long kallsyms_lookup_name(const char *name);
+extern void __print_symbol(const char *fmt, unsigned long address);
+extern int sprint_backtrace(char *buffer, unsigned long address);
+extern int sprint_symbol(char *buffer, unsigned long address);
+ unsigned long *symbolsize,
+ unsigned long *offset,
+ char **modname, char *namebuf);
++extern int kallsyms_lookup_size_offset(unsigned long addr,
++ unsigned long *symbolsize,
++ unsigned long *offset);
+#endif
/* This macro allows us to keep printk typechecking */
/*
* The return value from decompress routine is the length of the
+diff --git a/include/linux/preempt.h b/include/linux/preempt.h
+index f5d4723..a6ea2fa 100644
+--- a/include/linux/preempt.h
++++ b/include/linux/preempt.h
+@@ -18,8 +18,13 @@
+ # define sub_preempt_count(val) do { preempt_count() -= (val); } while (0)
+ #endif
+
++#define raw_add_preempt_count(val) do { preempt_count() += (val); } while (0)
++#define raw_sub_preempt_count(val) do { preempt_count() -= (val); } while (0)
++
+ #define inc_preempt_count() add_preempt_count(1)
++#define raw_inc_preempt_count() raw_add_preempt_count(1)
+ #define dec_preempt_count() sub_preempt_count(1)
++#define raw_dec_preempt_count() raw_sub_preempt_count(1)
+
+ #define preempt_count() (current_thread_info()->preempt_count)
+
+@@ -64,6 +69,12 @@ do { \
+ barrier(); \
+ } while (0)
+
++#define raw_preempt_disable() \
++do { \
++ raw_inc_preempt_count(); \
++ barrier(); \
++} while (0)
++
+ #define sched_preempt_enable_no_resched() \
+ do { \
+ barrier(); \
+@@ -72,6 +83,12 @@ do { \
+
+ #define preempt_enable_no_resched() sched_preempt_enable_no_resched()
+
++#define raw_preempt_enable_no_resched() \
++do { \
++ barrier(); \
++ raw_dec_preempt_count(); \
++} while (0)
++
+ #define preempt_enable() \
+ do { \
+ preempt_enable_no_resched(); \
+@@ -116,8 +133,10 @@ do { \
+ * region.
+ */
+ #define preempt_disable() barrier()
++#define raw_preempt_disable() barrier()
+ #define sched_preempt_enable_no_resched() barrier()
+ #define preempt_enable_no_resched() barrier()
++#define raw_preempt_enable_no_resched() barrier()
+ #define preempt_enable() barrier()
+
+ #define preempt_disable_notrace() barrier()
diff --git a/include/linux/printk.h b/include/linux/printk.h
index 22c7052..ad3fa0a 100644
--- a/include/linux/printk.h
static inline void anon_vma_merge(struct vm_area_struct *vma,
struct vm_area_struct *next)
diff --git a/include/linux/sched.h b/include/linux/sched.h
-index 178a8d9..450bf11 100644
+index 3aeb14b..73816a6 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -62,6 +62,7 @@ struct bio_list;
extern signed long schedule_timeout_interruptible(signed long timeout);
extern signed long schedule_timeout_killable(signed long timeout);
extern signed long schedule_timeout_uninterruptible(signed long timeout);
-@@ -314,6 +315,19 @@ struct nsproxy;
+@@ -314,7 +315,19 @@ struct nsproxy;
struct user_namespace;
#ifdef CONFIG_MMU
+-extern unsigned long mmap_legacy_base(void);
+
+#ifdef CONFIG_GRKERNSEC_RAND_THREADSTACK
+extern unsigned long gr_rand_threadstack_offset(const struct mm_struct *mm, const struct file *filp, unsigned long flags);
+
+extern bool check_heap_stack_gap(const struct vm_area_struct *vma, unsigned long addr, unsigned long len, unsigned long offset);
+extern unsigned long skip_heap_stack_gap(const struct vm_area_struct *vma, unsigned long len, unsigned long offset);
-+
++extern unsigned long mmap_legacy_base(struct mm_struct *mm);
extern void arch_pick_mmap_layout(struct mm_struct *mm);
extern unsigned long
arch_get_unmapped_area(struct file *, unsigned long, unsigned long,
-@@ -591,6 +605,17 @@ struct signal_struct {
+@@ -592,6 +605,17 @@ struct signal_struct {
#ifdef CONFIG_TASKSTATS
struct taskstats *stats;
#endif
#ifdef CONFIG_AUDIT
unsigned audit_tty;
unsigned audit_tty_log_passwd;
-@@ -671,6 +696,14 @@ struct user_struct {
+@@ -672,6 +696,14 @@ struct user_struct {
struct key *session_keyring; /* UID's default session keyring */
#endif
/* Hash table maintenance information */
struct hlist_node uidhash_node;
kuid_t uid;
-@@ -1158,8 +1191,8 @@ struct task_struct {
+@@ -1159,8 +1191,8 @@ struct task_struct {
struct list_head thread_group;
struct completion *vfork_done; /* for vfork() */
cputime_t utime, stime, utimescaled, stimescaled;
cputime_t gtime;
-@@ -1184,11 +1217,6 @@ struct task_struct {
+@@ -1185,11 +1217,6 @@ struct task_struct {
struct task_cputime cputime_expires;
struct list_head cpu_timers[3];
char comm[TASK_COMM_LEN]; /* executable name excluding path
- access with [gs]et_task_comm (which lock
it with task_lock())
-@@ -1205,6 +1233,10 @@ struct task_struct {
+@@ -1206,6 +1233,10 @@ struct task_struct {
#endif
/* CPU-specific state of this task */
struct thread_struct thread;
/* filesystem information */
struct fs_struct *fs;
/* open file information */
-@@ -1278,6 +1310,10 @@ struct task_struct {
+@@ -1279,6 +1310,10 @@ struct task_struct {
gfp_t lockdep_reclaim_gfp;
#endif
/* journalling filesystem info */
void *journal_info;
-@@ -1316,6 +1352,10 @@ struct task_struct {
+@@ -1317,6 +1352,10 @@ struct task_struct {
/* cg_list protected by css_set_lock and tsk->alloc_lock */
struct list_head cg_list;
#endif
#ifdef CONFIG_FUTEX
struct robust_list_head __user *robust_list;
#ifdef CONFIG_COMPAT
-@@ -1416,8 +1456,76 @@ struct task_struct {
+@@ -1417,8 +1456,76 @@ struct task_struct {
unsigned int sequential_io;
unsigned int sequential_io_avg;
#endif
/* Future-safe accessor for struct task_struct's cpus_allowed. */
#define tsk_cpus_allowed(tsk) (&(tsk)->cpus_allowed)
-@@ -1476,7 +1584,7 @@ struct pid_namespace;
+@@ -1477,7 +1584,7 @@ struct pid_namespace;
pid_t __task_pid_nr_ns(struct task_struct *task, enum pid_type type,
struct pid_namespace *ns);
{
return tsk->pid;
}
-@@ -1919,7 +2027,9 @@ void yield(void);
+@@ -1920,7 +2027,9 @@ void yield(void);
extern struct exec_domain default_exec_domain;
union thread_union {
unsigned long stack[THREAD_SIZE/sizeof(long)];
};
-@@ -1952,6 +2062,7 @@ extern struct pid_namespace init_pid_ns;
+@@ -1953,6 +2062,7 @@ extern struct pid_namespace init_pid_ns;
*/
extern struct task_struct *find_task_by_vpid(pid_t nr);
extern struct task_struct *find_task_by_pid_ns(pid_t nr,
struct pid_namespace *ns);
-@@ -2118,7 +2229,7 @@ extern void __cleanup_sighand(struct sighand_struct *);
+@@ -2119,7 +2229,7 @@ extern void __cleanup_sighand(struct sighand_struct *);
extern void exit_itimers(struct signal_struct *);
extern void flush_itimer_signals(void);
extern int allow_signal(int);
extern int disallow_signal(int);
-@@ -2309,9 +2420,9 @@ static inline unsigned long *end_of_stack(struct task_struct *p)
+@@ -2310,9 +2420,9 @@ static inline unsigned long *end_of_stack(struct task_struct *p)
#endif
};
/* shm_mode upper byte flags */
+diff --git a/include/linux/signal.h b/include/linux/signal.h
+index d897484..323ba98 100644
+--- a/include/linux/signal.h
++++ b/include/linux/signal.h
+@@ -433,6 +433,7 @@ void signals_init(void);
+
+ int restore_altstack(const stack_t __user *);
+ int __save_altstack(stack_t __user *, unsigned long);
++void __save_altstack_ex(stack_t __user *, unsigned long);
+
+ #ifdef CONFIG_PROC_FS
+ struct seq_file;
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index dec1748..112c1f9 100644
--- a/include/linux/skbuff.h
void *kmem_cache_alloc_node(struct kmem_cache *, gfp_t flags, int node);
#ifdef CONFIG_TRACING
+diff --git a/include/linux/smp.h b/include/linux/smp.h
+index c848876..11e8a84 100644
+--- a/include/linux/smp.h
++++ b/include/linux/smp.h
+@@ -221,7 +221,9 @@ static inline void kick_all_cpus_sync(void) { }
+ #endif
+
+ #define get_cpu() ({ preempt_disable(); smp_processor_id(); })
++#define raw_get_cpu() ({ raw_preempt_disable(); raw_smp_processor_id(); })
+ #define put_cpu() preempt_enable()
++#define raw_put_cpu_no_resched() raw_preempt_enable_no_resched()
+
+ /*
+ * Callback to arch code if there's nosmp or maxcpus=0 on the
diff --git a/include/linux/sock_diag.h b/include/linux/sock_diag.h
index 54f91d3..be2c379 100644
--- a/include/linux/sock_diag.h
extern dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
unsigned long offset, size_t size,
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
-index 4147d70..d356a10 100644
+index 84662ec..d8f8adb 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -97,8 +97,12 @@ struct sigaltstack;
#ifdef CONFIG_BLK_DEV_RAM
int fd;
diff --git a/init/main.c b/init/main.c
-index 9484f4b..4c01430 100644
+index 9484f4b..0eac7c3 100644
--- a/init/main.c
+++ b/init/main.c
@@ -100,6 +100,8 @@ static inline void mark_rodata_ro(void) { }
/*
* Debug helper: via this flag we know that we are in 'early bootup code'
* where only the boot processor is running with IRQ disabled. This means
-@@ -153,6 +155,64 @@ static int __init set_reset_devices(char *str)
+@@ -153,6 +155,74 @@ static int __init set_reset_devices(char *str)
__setup("reset_devices", set_reset_devices);
+#endif
+
+#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
-+unsigned long pax_user_shadow_base __read_only = 1UL << TASK_SIZE_MAX_SHIFT;
++unsigned long pax_user_shadow_base __read_only;
+EXPORT_SYMBOL(pax_user_shadow_base);
+extern char pax_enter_kernel_user[];
+extern char pax_exit_kernel_user[];
-+extern pgdval_t clone_pgd_mask;
+#endif
+
+#if defined(CONFIG_X86) && defined(CONFIG_PAX_MEMORY_UDEREF)
+ memcpy(pax_exit_kernel_user, (unsigned char []){0xc3}, 1);
+ clone_pgd_mask = ~(pgdval_t)0UL;
+ pax_user_shadow_base = 0UL;
++ setup_clear_cpu_cap(X86_FEATURE_PCID);
+#endif
+
+ return 0;
+}
+early_param("pax_nouderef", setup_pax_nouderef);
++
++#ifdef CONFIG_X86_64
++static int __init setup_pax_weakuderef(char *str)
++{
++ if (clone_pgd_mask != ~(pgdval_t)0UL)
++ pax_user_shadow_base = 1UL << TASK_SIZE_MAX_SHIFT;
++ return 1;
++}
++__setup("pax_weakuderef", setup_pax_weakuderef);
++#endif
+#endif
+
+#ifdef CONFIG_PAX_SOFTMODE
static const char * argv_init[MAX_INIT_ARGS+2] = { "init", NULL, };
const char * envp_init[MAX_INIT_ENVS+2] = { "HOME=/", "TERM=linux", NULL, };
static const char *panic_later, *panic_param;
-@@ -655,8 +715,6 @@ static void __init do_ctors(void)
+@@ -655,8 +725,6 @@ static void __init do_ctors(void)
bool initcall_debug;
core_param(initcall_debug, initcall_debug, bool, 0644);
static int __init_or_module do_one_initcall_debug(initcall_t fn)
{
ktime_t calltime, delta, rettime;
-@@ -679,23 +737,22 @@ int __init_or_module do_one_initcall(initcall_t fn)
+@@ -679,23 +747,22 @@ int __init_or_module do_one_initcall(initcall_t fn)
{
int count = preempt_count();
int ret;
return ret;
}
-@@ -748,8 +805,14 @@ static void __init do_initcall_level(int level)
+@@ -748,8 +815,14 @@ static void __init do_initcall_level(int level)
level, level,
&repair_env_string);
}
static void __init do_initcalls(void)
-@@ -783,8 +846,14 @@ static void __init do_pre_smp_initcalls(void)
+@@ -783,8 +856,14 @@ static void __init do_pre_smp_initcalls(void)
{
initcall_t *fn;
}
/*
-@@ -802,8 +871,8 @@ static int run_init_process(const char *init_filename)
+@@ -802,8 +881,8 @@ static int run_init_process(const char *init_filename)
{
argv_init[0] = init_filename;
return do_execve(init_filename,
}
static noinline void __init kernel_init_freeable(void);
-@@ -880,7 +949,7 @@ static noinline void __init kernel_init_freeable(void)
+@@ -880,7 +959,7 @@ static noinline void __init kernel_init_freeable(void)
do_basic_setup();
/* Open the /dev/console on the rootfs, this should never fail */
pr_err("Warning: unable to open an initial console.\n");
(void) sys_dup(0);
-@@ -893,11 +962,13 @@ static noinline void __init kernel_init_freeable(void)
+@@ -893,11 +972,13 @@ static noinline void __init kernel_init_freeable(void)
if (!ramdisk_execute_command)
ramdisk_execute_command = "/init";
+ return ns_capable_nolog(ns, cap) && kuid_has_mapping(ns, inode->i_uid);
+}
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
-index c6e77ef..af531a0 100644
+index 2e9b387..61817b1 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
-@@ -5391,7 +5391,7 @@ static int cgroup_css_links_read(struct cgroup *cont,
+@@ -5398,7 +5398,7 @@ static int cgroup_css_links_read(struct cgroup *cont,
struct css_set *cg = link->cg;
struct task_struct *task;
int count = 0;
{
struct signal_struct *sig = current->signal;
diff --git a/kernel/fork.c b/kernel/fork.c
-index 987b28a..11ee8a5 100644
+index ffbc090..08ceeee 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -319,7 +319,7 @@ static struct task_struct *dup_task_struct(struct task_struct *orig)
if (clone_flags & CLONE_VFORK) {
p->vfork_done = &vfork;
init_completion(&vfork);
-@@ -1723,7 +1785,7 @@ void __init proc_caches_init(void)
+@@ -1729,7 +1791,7 @@ void __init proc_caches_init(void)
mm_cachep = kmem_cache_create("mm_struct",
sizeof(struct mm_struct), ARCH_MIN_MMSTRUCT_ALIGN,
SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL);
mmap_init();
nsproxy_cache_init();
}
-@@ -1763,7 +1825,7 @@ static int unshare_fs(unsigned long unshare_flags, struct fs_struct **new_fsp)
+@@ -1769,7 +1831,7 @@ static int unshare_fs(unsigned long unshare_flags, struct fs_struct **new_fsp)
return 0;
/* don't need lock here; in the worst case we'll do useless copy */
return 0;
*new_fsp = copy_fs_struct(fs);
-@@ -1875,7 +1937,8 @@ SYSCALL_DEFINE1(unshare, unsigned long, unshare_flags)
+@@ -1881,7 +1943,8 @@ SYSCALL_DEFINE1(unshare, unsigned long, unshare_flags)
fs = current->fs;
spin_lock(&fs->lock);
current->fs = new_fs;
kernel_cap_t new_cap;
int err, i;
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
-index bddf3b2..07b90dd 100644
+index bddf3b2..233bf40 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
-@@ -185,7 +185,7 @@ static kprobe_opcode_t __kprobes *__get_insn_slot(struct kprobe_insn_cache *c)
+@@ -31,6 +31,9 @@
+ * <jkenisto@us.ibm.com> and Prasanna S Panchamukhi
+ * <prasanna@in.ibm.com> added function-return probes.
+ */
++#ifdef CONFIG_GRKERNSEC_HIDESYM
++#define __INCLUDED_BY_HIDESYM 1
++#endif
+ #include <linux/kprobes.h>
+ #include <linux/hash.h>
+ #include <linux/init.h>
+@@ -185,7 +188,7 @@ static kprobe_opcode_t __kprobes *__get_insn_slot(struct kprobe_insn_cache *c)
* kernel image and loaded module images reside. This is required
* so x86_64 can correctly handle the %rip-relative fixups.
*/
if (!kip->insns) {
kfree(kip);
return NULL;
-@@ -225,7 +225,7 @@ static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx)
+@@ -225,7 +228,7 @@ static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx)
*/
if (!list_is_singular(&kip->list)) {
list_del(&kip->list);
kfree(kip);
}
return 1;
-@@ -2083,7 +2083,7 @@ static int __init init_kprobes(void)
+@@ -2083,7 +2086,7 @@ static int __init init_kprobes(void)
{
int i, err = 0;
unsigned long offset = 0, size = 0;
const char *symbol_name;
void *addr;
struct kprobe_blackpoint *kb;
-@@ -2168,11 +2168,11 @@ static void __kprobes report_probe(struct seq_file *pi, struct kprobe *p,
+@@ -2168,11 +2171,11 @@ static void __kprobes report_probe(struct seq_file *pi, struct kprobe *p,
kprobe_type = "k";
if (sym)
p->addr, kprobe_type, p->addr);
if (!pp)
-@@ -2209,7 +2209,7 @@ static int __kprobes show_kprobe_addr(struct seq_file *pi, void *v)
+@@ -2209,7 +2212,7 @@ static int __kprobes show_kprobe_addr(struct seq_file *pi, void *v)
const char *sym = NULL;
unsigned int i = *(loff_t *) v;
unsigned long offset = 0;
.clock_get = thread_cpu_clock_get,
.timer_create = thread_cpu_timer_create,
diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
-index 424c2d4..a9194f7 100644
+index 424c2d4..679242f 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -43,6 +43,7 @@
}
static int common_timer_create(struct k_itimer *new_timer)
+@@ -597,7 +598,7 @@ SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock,
+ struct k_clock *kc = clockid_to_kclock(which_clock);
+ struct k_itimer *new_timer;
+ int error, new_timer_id;
+- sigevent_t event;
++ sigevent_t event = { };
+ int it_id_set = IT_ID_NOT_SET;
+
+ if (!kc)
@@ -1011,6 +1012,13 @@ SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock,
if (copy_from_user(&new_tp, tp, sizeof (*tp)))
return -EFAULT;
.priority = CPU_PRI_MIGRATION,
};
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
-index c61a614..d7f3d7e 100644
+index 03b73be..9422b9f 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -831,7 +831,7 @@ void task_numa_fault(int node, int pages, bool migrated)
p->mm->numa_scan_offset = 0;
}
-@@ -5686,7 +5686,7 @@ static void nohz_idle_balance(int this_cpu, enum cpu_idle_type idle) { }
+@@ -5687,7 +5687,7 @@ static void nohz_idle_balance(int this_cpu, enum cpu_idle_type idle) { }
* run_rebalance_domains is triggered when needed from the scheduler tick.
* Also triggered for nohz idle balancing (with nohz_balancing_kick set).
*/
#define sched_class_highest (&stop_sched_class)
#define for_each_class(class) \
diff --git a/kernel/signal.c b/kernel/signal.c
-index 113411b..17190e2 100644
+index 113411b..20d0a99 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -51,12 +51,12 @@ static struct kmem_cache *sigqueue_cachep;
if (p && (tgid <= 0 || task_tgid_vnr(p) == tgid)) {
error = check_kill_permission(sig, info, p);
/*
-@@ -3240,8 +3271,8 @@ COMPAT_SYSCALL_DEFINE2(sigaltstack,
+@@ -3219,6 +3250,16 @@ int __save_altstack(stack_t __user *uss, unsigned long sp)
+ __put_user(t->sas_ss_size, &uss->ss_size);
+ }
+
++#ifdef CONFIG_X86
++void __save_altstack_ex(stack_t __user *uss, unsigned long sp)
++{
++ struct task_struct *t = current;
++ put_user_ex((void __user *)t->sas_ss_sp, &uss->ss_sp);
++ put_user_ex(sas_ss_flags(sp), &uss->ss_flags);
++ put_user_ex(t->sas_ss_size, &uss->ss_size);
++}
++#endif
++
+ #ifdef CONFIG_COMPAT
+ COMPAT_SYSCALL_DEFINE2(sigaltstack,
+ const compat_stack_t __user *, uss_ptr,
+@@ -3240,8 +3281,8 @@ COMPAT_SYSCALL_DEFINE2(sigaltstack,
}
seg = get_fs();
set_fs(KERNEL_DS);
compat_user_stack_pointer());
set_fs(seg);
if (ret >= 0 && uoss_ptr) {
+@@ -3268,6 +3309,16 @@ int __compat_save_altstack(compat_stack_t __user *uss, unsigned long sp)
+ __put_user(sas_ss_flags(sp), &uss->ss_flags) |
+ __put_user(t->sas_ss_size, &uss->ss_size);
+ }
++
++#ifdef CONFIG_X86
++void __compat_save_altstack_ex(compat_stack_t __user *uss, unsigned long sp)
++{
++ struct task_struct *t = current;
++ put_user_ex(ptr_to_compat((void __user *)t->sas_ss_sp), &uss->ss_sp);
++ put_user_ex(sas_ss_flags(sp), &uss->ss_flags);
++ put_user_ex(t->sas_ss_size, &uss->ss_size);
++}
++#endif
+ #endif
+
+ #ifdef __ARCH_WANT_SYS_SIGPENDING
diff --git a/kernel/smp.c b/kernel/smp.c
index 4dba0f7..fe9f773 100644
--- a/kernel/smp.c
*data_page = bpage;
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
-index f7bc3ce..b8ef9b5 100644
+index 06a5bce..53ad6e7 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
-@@ -3303,7 +3303,7 @@ int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set)
+@@ -3347,7 +3347,7 @@ int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set)
return 0;
}
local_irq_save(flags);
diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c
-index d8c30db..f2f6af5 100644
+index 9064b91..1f5d2f8 100644
--- a/kernel/user_namespace.c
+++ b/kernel/user_namespace.c
-@@ -79,6 +79,21 @@ int create_user_ns(struct cred *new)
+@@ -82,6 +82,21 @@ int create_user_ns(struct cred *new)
!kgid_has_mapping(parent_ns, group))
return -EPERM;
ns = kmem_cache_zalloc(user_ns_cachep, GFP_KERNEL);
if (!ns)
return -ENOMEM;
-@@ -105,6 +120,7 @@ int create_user_ns(struct cred *new)
- int unshare_userns(unsigned long unshare_flags, struct cred **new_cred)
- {
- struct cred *cred;
-+ int err;
-
- if (!(unshare_flags & CLONE_NEWUSER))
- return 0;
-@@ -113,8 +129,12 @@ int unshare_userns(unsigned long unshare_flags, struct cred **new_cred)
- if (!cred)
- return -ENOMEM;
-
-- *new_cred = cred;
-- return create_user_ns(cred);
-+ err = create_user_ns(cred);
-+ if (err)
-+ put_cred(cred);
-+ else
-+ *new_cred = cred;
-+ return err;
- }
-
- void free_user_ns(struct user_namespace *ns)
-@@ -853,7 +873,7 @@ static int userns_install(struct nsproxy *nsproxy, void *ns)
+@@ -862,7 +877,7 @@ static int userns_install(struct nsproxy *nsproxy, void *ns)
if (atomic_read(¤t->mm->mm_users) > 1)
return -EINVAL;
.thread_should_run = watchdog_should_run,
.thread_fn = watchdog,
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
-index ee8e29a..410568e 100644
+index 6f01921..139869b 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
-@@ -4584,7 +4584,7 @@ static void rebind_workers(struct worker_pool *pool)
+@@ -4596,7 +4596,7 @@ static void rebind_workers(struct worker_pool *pool)
WARN_ON_ONCE(!(worker_flags & WORKER_UNBOUND));
worker_flags |= WORKER_REBOUND;
worker_flags &= ~WORKER_UNBOUND;
config NOMMU_INITIAL_TRIM_EXCESS
diff --git a/mm/backing-dev.c b/mm/backing-dev.c
-index 5025174..9fc1c5c 100644
+index 5025174..9d67dcd 100644
--- a/mm/backing-dev.c
+++ b/mm/backing-dev.c
+@@ -12,7 +12,7 @@
+ #include <linux/device.h>
+ #include <trace/events/writeback.h>
+
+-static atomic_long_t bdi_seq = ATOMIC_LONG_INIT(0);
++static atomic_long_unchecked_t bdi_seq = ATOMIC_LONG_INIT(0);
+
+ struct backing_dev_info default_backing_dev_info = {
+ .name = "default",
@@ -515,7 +515,6 @@ EXPORT_SYMBOL(bdi_destroy);
int bdi_setup_and_register(struct backing_dev_info *bdi, char *name,
unsigned int cap)
- sprintf(tmp, "%.28s%s", name, "-%d");
- err = bdi_register(bdi, NULL, tmp, atomic_long_inc_return(&bdi_seq));
-+ err = bdi_register(bdi, NULL, "%.28s-%ld", name, atomic_long_inc_return(&bdi_seq));
++ err = bdi_register(bdi, NULL, "%.28s-%ld", name, atomic_long_inc_return_unchecked(&bdi_seq));
if (err) {
bdi_destroy(bdi);
return err;
set_page_address(page, (void *)vaddr);
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
-index 5cf99bf..28634c8 100644
+index 7c5eb85..5c01c2f 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2022,15 +2022,17 @@ static int hugetlb_sysctl_handler_common(bool obey_mempolicy,
if (!ptep)
return VM_FAULT_OOM;
diff --git a/mm/internal.h b/mm/internal.h
-index 8562de0..7fdfe92 100644
+index 8562de0..92b2073 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -100,6 +100,7 @@ extern pmd_t *mm_find_pmd(struct mm_struct *mm, unsigned long address);
extern void prep_compound_page(struct page *page, unsigned long order);
#ifdef CONFIG_MEMORY_FAILURE
extern bool is_free_buddy_page(struct page *page);
+@@ -355,7 +356,7 @@ extern u32 hwpoison_filter_enable;
+
+ extern unsigned long vm_mmap_pgoff(struct file *, unsigned long,
+ unsigned long, unsigned long,
+- unsigned long, unsigned long);
++ unsigned long, unsigned long) __intentional_overflow(-1);
+
+ extern void set_pageblock_order(void);
+ unsigned long reclaim_clean_pages_from_list(struct zone *zone,
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index c8d7f31..2dbeffd 100644
--- a/mm/kmemleak.c
} else {
pr_info("soft offline: %#lx: isolation failed: %d, page count %d, type %lx\n",
diff --git a/mm/memory.c b/mm/memory.c
-index 5e50800..c47ba9a 100644
+index 5a35443..7c0340f 100644
--- a/mm/memory.c
+++ b/mm/memory.c
-@@ -429,6 +429,7 @@ static inline void free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
+@@ -428,6 +428,7 @@ static inline void free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
free_pte_range(tlb, pmd, addr);
} while (pmd++, addr = next, addr != end);
start &= PUD_MASK;
if (start < floor)
return;
-@@ -443,6 +444,8 @@ static inline void free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
+@@ -442,6 +443,8 @@ static inline void free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
pmd = pmd_offset(pud, start);
pud_clear(pud);
pmd_free_tlb(tlb, pmd, start);
}
static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
-@@ -462,6 +465,7 @@ static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
+@@ -461,6 +464,7 @@ static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
free_pmd_range(tlb, pud, addr, next, floor, ceiling);
} while (pud++, addr = next, addr != end);
start &= PGDIR_MASK;
if (start < floor)
return;
-@@ -476,6 +480,8 @@ static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
+@@ -475,6 +479,8 @@ static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
pud = pud_offset(pgd, start);
pgd_clear(pgd);
pud_free_tlb(tlb, pud, start);
}
/*
-@@ -1638,12 +1644,6 @@ no_page_table:
+@@ -1644,12 +1650,6 @@ no_page_table:
return page;
}
/**
* __get_user_pages() - pin user pages in memory
* @tsk: task_struct of target task
-@@ -1730,10 +1730,10 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
+@@ -1736,10 +1736,10 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
i = 0;
if (!vma && in_gate_area(mm, start)) {
unsigned long pg = start & PAGE_MASK;
pgd_t *pgd;
-@@ -1782,7 +1782,7 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
+@@ -1788,7 +1788,7 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
goto next_page;
}
(vma->vm_flags & (VM_IO | VM_PFNMAP)) ||
!(vm_flags & vma->vm_flags))
return i ? : -EFAULT;
-@@ -1811,11 +1811,6 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
+@@ -1817,11 +1817,6 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
int ret;
unsigned int fault_flags = 0;
if (foll_flags & FOLL_WRITE)
fault_flags |= FAULT_FLAG_WRITE;
if (nonblocking)
-@@ -1895,7 +1890,7 @@ next_page:
+@@ -1901,7 +1896,7 @@ next_page:
start += page_increm * PAGE_SIZE;
nr_pages -= page_increm;
} while (nr_pages && start < vma->vm_end);
return i;
}
EXPORT_SYMBOL(__get_user_pages);
-@@ -2102,6 +2097,10 @@ static int insert_page(struct vm_area_struct *vma, unsigned long addr,
+@@ -2108,6 +2103,10 @@ static int insert_page(struct vm_area_struct *vma, unsigned long addr,
page_add_file_rmap(page);
set_pte_at(mm, addr, pte, mk_pte(page, prot));
retval = 0;
pte_unmap_unlock(pte, ptl);
return retval;
-@@ -2146,9 +2145,21 @@ int vm_insert_page(struct vm_area_struct *vma, unsigned long addr,
+@@ -2152,9 +2151,21 @@ int vm_insert_page(struct vm_area_struct *vma, unsigned long addr,
if (!page_count(page))
return -EINVAL;
if (!(vma->vm_flags & VM_MIXEDMAP)) {
}
return insert_page(vma, addr, page, vma->vm_page_prot);
}
-@@ -2231,6 +2242,7 @@ int vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr,
+@@ -2237,6 +2248,7 @@ int vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr,
unsigned long pfn)
{
BUG_ON(!(vma->vm_flags & VM_MIXEDMAP));
if (addr < vma->vm_start || addr >= vma->vm_end)
return -EFAULT;
-@@ -2478,7 +2490,9 @@ static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
+@@ -2484,7 +2496,9 @@ static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
BUG_ON(pud_huge(*pud));
if (!pmd)
return -ENOMEM;
do {
-@@ -2498,7 +2512,9 @@ static int apply_to_pud_range(struct mm_struct *mm, pgd_t *pgd,
+@@ -2504,7 +2518,9 @@ static int apply_to_pud_range(struct mm_struct *mm, pgd_t *pgd,
unsigned long next;
int err;
if (!pud)
return -ENOMEM;
do {
-@@ -2586,6 +2602,186 @@ static inline void cow_user_page(struct page *dst, struct page *src, unsigned lo
+@@ -2592,6 +2608,186 @@ static inline void cow_user_page(struct page *dst, struct page *src, unsigned lo
copy_user_highpage(dst, src, va, vma);
}
/*
* This routine handles present pages, when users try to write
* to a shared page. It is done by copying the page to a new address
-@@ -2802,6 +2998,12 @@ gotten:
+@@ -2808,6 +3004,12 @@ gotten:
*/
page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
if (likely(pte_same(*page_table, orig_pte))) {
if (old_page) {
if (!PageAnon(old_page)) {
dec_mm_counter_fast(mm, MM_FILEPAGES);
-@@ -2853,6 +3055,10 @@ gotten:
+@@ -2859,6 +3061,10 @@ gotten:
page_remove_rmap(old_page);
}
/* Free the old page.. */
new_page = old_page;
ret |= VM_FAULT_WRITE;
-@@ -3128,6 +3334,11 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
+@@ -3134,6 +3340,11 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
swap_free(entry);
if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page))
try_to_free_swap(page);
unlock_page(page);
if (page != swapcache) {
/*
-@@ -3151,6 +3362,11 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
+@@ -3157,6 +3368,11 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
/* No need to invalidate - it was non-present before */
update_mmu_cache(vma, address, page_table);
unlock:
pte_unmap_unlock(page_table, ptl);
out:
-@@ -3170,40 +3386,6 @@ out_release:
+@@ -3176,40 +3392,6 @@ out_release:
}
/*
* We enter with non-exclusive mmap_sem (to exclude vma changes,
* but allow concurrent faults), and pte mapped but not yet locked.
* We return with mmap_sem still held, but pte unmapped and unlocked.
-@@ -3212,27 +3394,23 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
+@@ -3218,27 +3400,23 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
unsigned long address, pte_t *page_table, pmd_t *pmd,
unsigned int flags)
{
if (unlikely(anon_vma_prepare(vma)))
goto oom;
page = alloc_zeroed_user_highpage_movable(vma, address);
-@@ -3256,6 +3434,11 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
+@@ -3262,6 +3440,11 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
if (!pte_none(*page_table))
goto release;
inc_mm_counter_fast(mm, MM_ANONPAGES);
page_add_new_anon_rmap(page, vma, address);
setpte:
-@@ -3263,6 +3446,12 @@ setpte:
+@@ -3269,6 +3452,12 @@ setpte:
/* No need to invalidate - it was non-present before */
update_mmu_cache(vma, address, page_table);
unlock:
pte_unmap_unlock(page_table, ptl);
return 0;
-@@ -3406,6 +3595,12 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
+@@ -3412,6 +3601,12 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
*/
/* Only go through if we didn't race with anybody else... */
if (likely(pte_same(*page_table, orig_pte))) {
flush_icache_page(vma, page);
entry = mk_pte(page, vma->vm_page_prot);
if (flags & FAULT_FLAG_WRITE)
-@@ -3425,6 +3620,14 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
+@@ -3431,6 +3626,14 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
/* no need to invalidate: a not-present page won't be cached */
update_mmu_cache(vma, address, page_table);
} else {
if (cow_page)
mem_cgroup_uncharge_page(cow_page);
-@@ -3746,6 +3949,12 @@ int handle_pte_fault(struct mm_struct *mm,
+@@ -3752,6 +3955,12 @@ int handle_pte_fault(struct mm_struct *mm,
if (flags & FAULT_FLAG_WRITE)
flush_tlb_fix_spurious_fault(vma, address);
}
unlock:
pte_unmap_unlock(pte, ptl);
return 0;
-@@ -3762,6 +3971,10 @@ int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
+@@ -3768,6 +3977,10 @@ int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
pmd_t *pmd;
pte_t *pte;
__set_current_state(TASK_RUNNING);
count_vm_event(PGFAULT);
-@@ -3773,6 +3986,34 @@ int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
+@@ -3779,6 +3992,34 @@ int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
if (unlikely(is_vm_hugetlb_page(vma)))
return hugetlb_fault(mm, vma, address, flags);
retry:
pgd = pgd_offset(mm, address);
pud = pud_alloc(mm, pgd, address);
-@@ -3871,6 +4112,23 @@ int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address)
+@@ -3877,6 +4118,23 @@ int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address)
spin_unlock(&mm->page_table_lock);
return 0;
}
#endif /* __PAGETABLE_PUD_FOLDED */
#ifndef __PAGETABLE_PMD_FOLDED
-@@ -3901,6 +4159,30 @@ int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address)
+@@ -3907,6 +4165,30 @@ int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address)
spin_unlock(&mm->page_table_lock);
return 0;
}
#endif /* __PAGETABLE_PMD_FOLDED */
#if !defined(__HAVE_ARCH_GATE_AREA)
-@@ -3914,7 +4196,7 @@ static int __init gate_vma_init(void)
+@@ -3920,7 +4202,7 @@ 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;
return 0;
}
-@@ -4048,8 +4330,8 @@ out:
+@@ -4054,8 +4336,8 @@ out:
return ret;
}
{
resource_size_t phys_addr;
unsigned long prot = 0;
-@@ -4074,8 +4356,8 @@ int generic_access_phys(struct vm_area_struct *vma, unsigned long addr,
+@@ -4080,8 +4362,8 @@ int generic_access_phys(struct vm_area_struct *vma, unsigned long addr,
* Access another process' address space as given in mm. If non-NULL, use the
* given task for page fault accounting.
*/
{
struct vm_area_struct *vma;
void *old_buf = buf;
-@@ -4083,7 +4365,7 @@ static int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm,
+@@ -4089,7 +4371,7 @@ static int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm,
down_read(&mm->mmap_sem);
/* ignore errors, just check how much was successfully transferred */
while (len) {
void *maddr;
struct page *page = NULL;
-@@ -4142,8 +4424,8 @@ static int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm,
+@@ -4148,8 +4430,8 @@ static int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm,
*
* The caller must hold a reference on @mm.
*/
{
return __access_remote_vm(NULL, mm, addr, buf, len, write);
}
-@@ -4153,11 +4435,11 @@ int access_remote_vm(struct mm_struct *mm, unsigned long addr,
+@@ -4159,11 +4441,11 @@ int access_remote_vm(struct mm_struct *mm, unsigned long addr,
* Source/target buffer must be kernel space,
* Do not walk the page table directly, use get_user_pages
*/
capable(CAP_IPC_LOCK))
ret = do_mlockall(flags);
diff --git a/mm/mmap.c b/mm/mmap.c
-index 7dbe397..e84c411 100644
+index 8d25fdc..bfb7626 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -36,6 +36,7 @@
if (batadv_ogm_packet->flags & BATADV_DIRECTLINK)
has_directlink_flag = 1;
+diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
+index de27b31..7058bfe 100644
+--- a/net/batman-adv/bridge_loop_avoidance.c
++++ b/net/batman-adv/bridge_loop_avoidance.c
+@@ -1522,6 +1522,8 @@ out:
+ * in these cases, the skb is further handled by this function and
+ * returns 1, otherwise it returns 0 and the caller shall further
+ * process the skb.
++ *
++ * This call might reallocate skb data.
+ */
+ int batadv_bla_tx(struct batadv_priv *bat_priv, struct sk_buff *skb, short vid)
+ {
+diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
+index f105219..7614af3 100644
+--- a/net/batman-adv/gateway_client.c
++++ b/net/batman-adv/gateway_client.c
+@@ -508,6 +508,7 @@ out:
+ return 0;
+ }
+
++/* this call might reallocate skb data */
+ static bool batadv_is_type_dhcprequest(struct sk_buff *skb, int header_len)
+ {
+ int ret = false;
+@@ -568,6 +569,7 @@ out:
+ return ret;
+ }
+
++/* this call might reallocate skb data */
+ bool batadv_gw_is_dhcp_target(struct sk_buff *skb, unsigned int *header_len)
+ {
+ struct ethhdr *ethhdr;
+@@ -619,6 +621,12 @@ bool batadv_gw_is_dhcp_target(struct sk_buff *skb, unsigned int *header_len)
+
+ if (!pskb_may_pull(skb, *header_len + sizeof(*udphdr)))
+ return false;
++
++ /* skb->data might have been reallocated by pskb_may_pull() */
++ ethhdr = (struct ethhdr *)skb->data;
++ if (ntohs(ethhdr->h_proto) == ETH_P_8021Q)
++ ethhdr = (struct ethhdr *)(skb->data + VLAN_HLEN);
++
+ udphdr = (struct udphdr *)(skb->data + *header_len);
+ *header_len += sizeof(*udphdr);
+
+@@ -634,12 +642,14 @@ bool batadv_gw_is_dhcp_target(struct sk_buff *skb, unsigned int *header_len)
+ return true;
+ }
+
++/* this call might reallocate skb data */
+ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv,
+- struct sk_buff *skb, struct ethhdr *ethhdr)
++ struct sk_buff *skb)
+ {
+ struct batadv_neigh_node *neigh_curr = NULL, *neigh_old = NULL;
+ struct batadv_orig_node *orig_dst_node = NULL;
+ struct batadv_gw_node *curr_gw = NULL;
++ struct ethhdr *ethhdr;
+ bool ret, out_of_range = false;
+ unsigned int header_len = 0;
+ uint8_t curr_tq_avg;
+@@ -648,6 +658,7 @@ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv,
+ if (!ret)
+ goto out;
+
++ ethhdr = (struct ethhdr *)skb->data;
+ orig_dst_node = batadv_transtable_search(bat_priv, ethhdr->h_source,
+ ethhdr->h_dest);
+ if (!orig_dst_node)
+diff --git a/net/batman-adv/gateway_client.h b/net/batman-adv/gateway_client.h
+index 039902d..1037d75 100644
+--- a/net/batman-adv/gateway_client.h
++++ b/net/batman-adv/gateway_client.h
+@@ -34,7 +34,6 @@ void batadv_gw_node_delete(struct batadv_priv *bat_priv,
+ void batadv_gw_node_purge(struct batadv_priv *bat_priv);
+ int batadv_gw_client_seq_print_text(struct seq_file *seq, void *offset);
+ bool batadv_gw_is_dhcp_target(struct sk_buff *skb, unsigned int *header_len);
+-bool batadv_gw_out_of_range(struct batadv_priv *bat_priv,
+- struct sk_buff *skb, struct ethhdr *ethhdr);
++bool batadv_gw_out_of_range(struct batadv_priv *bat_priv, struct sk_buff *skb);
+
+ #endif /* _NET_BATMAN_ADV_GATEWAY_CLIENT_H_ */
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 522243a..b48c0ef 100644
--- a/net/batman-adv/hard-interface.c
return hard_iface;
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
-index 819dfb0..9a672d1 100644
+index 819dfb0..226bacd 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
-@@ -253,7 +253,7 @@ static int batadv_interface_tx(struct sk_buff *skb,
+@@ -180,6 +180,9 @@ static int batadv_interface_tx(struct sk_buff *skb,
+ if (batadv_bla_tx(bat_priv, skb, vid))
+ goto dropped;
+
++ /* skb->data might have been reallocated by batadv_bla_tx() */
++ ethhdr = (struct ethhdr *)skb->data;
++
+ /* Register the client MAC in the transtable */
+ if (!is_multicast_ether_addr(ethhdr->h_source))
+ batadv_tt_local_add(soft_iface, ethhdr->h_source, skb->skb_iif);
+@@ -220,6 +223,10 @@ static int batadv_interface_tx(struct sk_buff *skb,
+ default:
+ break;
+ }
++
++ /* reminder: ethhdr might have become unusable from here on
++ * (batadv_gw_is_dhcp_target() might have reallocated skb data)
++ */
+ }
+
+ /* ethernet packet should be broadcasted */
+@@ -253,7 +260,7 @@ static int batadv_interface_tx(struct sk_buff *skb,
primary_if->net_dev->dev_addr, ETH_ALEN);
/* set broadcast sequence number */
bcast_packet->seqno = htonl(seqno);
batadv_add_bcast_packet_to_list(bat_priv, skb, brd_delay);
-@@ -472,7 +472,7 @@ static int batadv_softif_init_late(struct net_device *dev)
+@@ -266,7 +273,7 @@ static int batadv_interface_tx(struct sk_buff *skb,
+ /* unicast packet */
+ } else {
+ if (atomic_read(&bat_priv->gw_mode) != BATADV_GW_MODE_OFF) {
+- ret = batadv_gw_out_of_range(bat_priv, skb, ethhdr);
++ ret = batadv_gw_out_of_range(bat_priv, skb);
+ if (ret)
+ goto dropped;
+ }
+@@ -472,7 +479,7 @@ static int batadv_softif_init_late(struct net_device *dev)
atomic_set(&bat_priv->batman_queue_left, BATADV_BATMAN_QUEUE_LEN);
atomic_set(&bat_priv->mesh_state, BATADV_MESH_INACTIVE);
atomic_t batman_queue_left;
char num_ifaces;
diff --git a/net/batman-adv/unicast.c b/net/batman-adv/unicast.c
-index 0bb3b59..ffcbf2f 100644
+index 0bb3b59..0e3052e 100644
--- a/net/batman-adv/unicast.c
+++ b/net/batman-adv/unicast.c
@@ -270,7 +270,7 @@ int batadv_frag_send_skb(struct sk_buff *skb, struct batadv_priv *bat_priv,
frag1->seqno = htons(seqno - 1);
frag2->seqno = htons(seqno);
+@@ -326,7 +326,9 @@ static bool batadv_unicast_push_and_fill_skb(struct sk_buff *skb, int hdr_size,
+ * @skb: the skb containing the payload to encapsulate
+ * @orig_node: the destination node
+ *
+- * Returns false if the payload could not be encapsulated or true otherwise
++ * Returns false if the payload could not be encapsulated or true otherwise.
++ *
++ * This call might reallocate skb data.
+ */
+ static bool batadv_unicast_prepare_skb(struct sk_buff *skb,
+ struct batadv_orig_node *orig_node)
+@@ -343,7 +345,9 @@ static bool batadv_unicast_prepare_skb(struct sk_buff *skb,
+ * @orig_node: the destination node
+ * @packet_subtype: the batman 4addr packet subtype to use
+ *
+- * Returns false if the payload could not be encapsulated or true otherwise
++ * Returns false if the payload could not be encapsulated or true otherwise.
++ *
++ * This call might reallocate skb data.
+ */
+ bool batadv_unicast_4addr_prepare_skb(struct batadv_priv *bat_priv,
+ struct sk_buff *skb,
+@@ -401,7 +405,7 @@ int batadv_unicast_generic_send_skb(struct batadv_priv *bat_priv,
+ struct batadv_neigh_node *neigh_node;
+ int data_len = skb->len;
+ int ret = NET_RX_DROP;
+- unsigned int dev_mtu;
++ unsigned int dev_mtu, header_len;
+
+ /* get routing information */
+ if (is_multicast_ether_addr(ethhdr->h_dest)) {
+@@ -429,10 +433,12 @@ find_router:
+ switch (packet_type) {
+ case BATADV_UNICAST:
+ batadv_unicast_prepare_skb(skb, orig_node);
++ header_len = sizeof(struct batadv_unicast_packet);
+ break;
+ case BATADV_UNICAST_4ADDR:
+ batadv_unicast_4addr_prepare_skb(bat_priv, skb, orig_node,
+ packet_subtype);
++ header_len = sizeof(struct batadv_unicast_4addr_packet);
+ break;
+ default:
+ /* this function supports UNICAST and UNICAST_4ADDR only. It
+@@ -441,6 +447,7 @@ find_router:
+ goto out;
+ }
+
++ ethhdr = (struct ethhdr *)(skb->data + header_len);
+ unicast_packet = (struct batadv_unicast_packet *)skb->data;
+
+ /* inform the destination node that we are still missing a correct route
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index ace5e55..a65a1c0 100644
--- a/net/bluetooth/hci_core.c
return error;
}
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
-index a08bd2b..4e8f43c 100644
+index a08bd2b..c59bd7c 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -58,7 +58,7 @@ struct rtnl_link {
}
EXPORT_SYMBOL_GPL(__rtnl_link_unregister);
+@@ -2374,7 +2377,7 @@ static int rtnl_bridge_getlink(struct sk_buff *skb, struct netlink_callback *cb)
+ struct nlattr *extfilt;
+ u32 filter_mask = 0;
+
+- extfilt = nlmsg_find_attr(cb->nlh, sizeof(struct rtgenmsg),
++ extfilt = nlmsg_find_attr(cb->nlh, sizeof(struct ifinfomsg),
+ IFLA_EXT_MASK);
+ if (extfilt)
+ filter_mask = nla_get_u32(extfilt);
diff --git a/net/core/scm.c b/net/core/scm.c
index 03795d0..eaf7368 100644
--- a/net/core/scm.c
#endif
if (dflt != &ipv4_devconf_dflt)
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
-index 4cfe34d..a6ba66e 100644
+index 4cfe34d..d2fac8a 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
+@@ -477,7 +477,7 @@ static u32 esp4_get_mtu(struct xfrm_state *x, int mtu)
+ }
+
+ return ((mtu - x->props.header_len - crypto_aead_authsize(esp->aead) -
+- net_adj) & ~(align - 1)) + (net_adj - 2);
++ net_adj) & ~(align - 1)) + net_adj - 2;
+ }
+
+ static void esp4_err(struct sk_buff *skb, u32 info)
@@ -503,7 +503,7 @@ static void esp4_err(struct sk_buff *skb, u32 info)
return;
return nh->nh_saddr;
}
+diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
+index 49616fe..6e8a13d 100644
+--- a/net/ipv4/fib_trie.c
++++ b/net/ipv4/fib_trie.c
+@@ -71,7 +71,6 @@
+ #include <linux/init.h>
+ #include <linux/list.h>
+ #include <linux/slab.h>
+-#include <linux/prefetch.h>
+ #include <linux/export.h>
+ #include <net/net_namespace.h>
+ #include <net/ip.h>
+@@ -1761,10 +1760,8 @@ static struct leaf *leaf_walk_rcu(struct tnode *p, struct rt_trie_node *c)
+ if (!c)
+ continue;
+
+- if (IS_LEAF(c)) {
+- prefetch(rcu_dereference_rtnl(p->child[idx]));
++ if (IS_LEAF(c))
+ return (struct leaf *) c;
+- }
+
+ /* Rescan start scanning in new node */
+ p = (struct tnode *) c;
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 6acb541..9ea617d 100644
--- a/net/ipv4/inet_connection_sock.c
return -ENOMEM;
}
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
-index 855004f..68e7458 100644
+index 855004f..9644112 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -115,7 +115,7 @@ static bool log_ecn_error = true;
static int ipgre_tunnel_init(struct net_device *dev);
static int ipgre_net_id __read_mostly;
+@@ -572,7 +572,7 @@ static int ipgre_header(struct sk_buff *skb, struct net_device *dev,
+ if (daddr)
+ memcpy(&iph->daddr, daddr, 4);
+ if (iph->daddr)
+- return t->hlen;
++ return t->hlen + sizeof(*iph);
+
+ return -(t->hlen + sizeof(*iph));
+ }
@@ -919,7 +919,7 @@ static const struct nla_policy ipgre_policy[IFLA_GRE_MAX + 1] = {
[IFLA_GRE_PMTUDISC] = { .type = NLA_U8 },
};
sizeof(net->ipv4.dev_addr_genid));
return 0;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
-index fa2f63f..6554815 100644
+index 3f25e75..3ae0f4d 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
-@@ -55,7 +55,7 @@ static int ipv4_local_port_range(ctl_table *table, int write,
+@@ -57,7 +57,7 @@ static int ipv4_local_port_range(ctl_table *table, int write,
{
int ret;
int range[2];
.data = &range,
.maxlen = sizeof(range),
.mode = table->mode,
-@@ -108,7 +108,7 @@ static int ipv4_ping_group_range(ctl_table *table, int write,
+@@ -110,7 +110,7 @@ static int ipv4_ping_group_range(ctl_table *table, int write,
int ret;
gid_t urange[2];
kgid_t low, high;
.data = &urange,
.maxlen = sizeof(urange),
.mode = table->mode,
-@@ -139,7 +139,7 @@ static int proc_tcp_congestion_control(ctl_table *ctl, int write,
+@@ -141,7 +141,7 @@ static int proc_tcp_congestion_control(ctl_table *ctl, int write,
void __user *buffer, size_t *lenp, loff_t *ppos)
{
char val[TCP_CA_NAME_MAX];
.data = val,
.maxlen = TCP_CA_NAME_MAX,
};
-@@ -158,7 +158,7 @@ static int proc_tcp_available_congestion_control(ctl_table *ctl,
+@@ -160,7 +160,7 @@ static int proc_tcp_available_congestion_control(ctl_table *ctl,
void __user *buffer, size_t *lenp,
loff_t *ppos)
{
int ret;
tbl.data = kmalloc(tbl.maxlen, GFP_USER);
-@@ -175,7 +175,7 @@ static int proc_allowed_congestion_control(ctl_table *ctl,
+@@ -177,7 +177,7 @@ static int proc_allowed_congestion_control(ctl_table *ctl,
void __user *buffer, size_t *lenp,
loff_t *ppos)
{
int ret;
tbl.data = kmalloc(tbl.maxlen, GFP_USER);
-@@ -201,15 +201,17 @@ static int ipv4_tcp_mem(ctl_table *ctl, int write,
+@@ -203,15 +203,17 @@ static int ipv4_tcp_mem(ctl_table *ctl, int write,
struct mem_cgroup *memcg;
#endif
}
ret = proc_doulongvec_minmax(&tmp, write, buffer, lenp, ppos);
-@@ -236,7 +238,7 @@ static int ipv4_tcp_mem(ctl_table *ctl, int write,
+@@ -238,7 +240,7 @@ static int ipv4_tcp_mem(ctl_table *ctl, int write,
static int proc_tcp_fastopen_key(ctl_table *ctl, int write, void __user *buffer,
size_t *lenp, loff_t *ppos)
{
struct tcp_fastopen_context *ctxt;
int ret;
u32 user_key[4]; /* 16 bytes, matching TCP_FASTOPEN_KEY_LENGTH */
-@@ -477,7 +479,7 @@ static struct ctl_table ipv4_table[] = {
+@@ -481,7 +483,7 @@ static struct ctl_table ipv4_table[] = {
},
{
.procname = "ip_local_reserved_ports",
.maxlen = 65536,
.mode = 0644,
.proc_handler = proc_do_large_bitmap,
-@@ -842,11 +844,10 @@ static struct ctl_table ipv4_net_table[] = {
+@@ -846,11 +848,10 @@ static struct ctl_table ipv4_net_table[] = {
static __net_init int ipv4_sysctl_init_net(struct net *net)
{
if (table == NULL)
goto err_alloc;
-@@ -881,15 +882,17 @@ static __net_init int ipv4_sysctl_init_net(struct net *net)
+@@ -885,15 +886,17 @@ static __net_init int ipv4_sysctl_init_net(struct net *net)
tcp_init_mem(net);
err_alloc:
return -ENOMEM;
}
-@@ -911,16 +914,6 @@ static __net_initdata struct pernet_operations ipv4_sysctl_ops = {
+@@ -915,16 +918,6 @@ static __net_initdata struct pernet_operations ipv4_sysctl_ops = {
static __init int sysctl_ipv4_init(void)
{
struct ctl_table_header *hdr;
int ret;
/*
+diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
+index 40ffd72..aeac0dc 100644
+--- a/net/ipv6/esp6.c
++++ b/net/ipv6/esp6.c
+@@ -425,7 +425,7 @@ static u32 esp6_get_mtu(struct xfrm_state *x, int mtu)
+ net_adj = 0;
+
+ return ((mtu - x->props.header_len - crypto_aead_authsize(esp->aead) -
+- net_adj) & ~(align - 1)) + (net_adj - 2);
++ net_adj) & ~(align - 1)) + net_adj - 2;
+ }
+
+ static void esp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index b4ff0a4..db9b764 100644
--- a/net/ipv6/icmp.c
};
diff --git a/net/key/af_key.c b/net/key/af_key.c
-index 9da8620..97070ad 100644
+index ab8bd2c..cd2d641 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
-@@ -3047,10 +3047,10 @@ static int pfkey_send_policy_notify(struct xfrm_policy *xp, int dir, const struc
+@@ -3048,10 +3048,10 @@ static int pfkey_send_policy_notify(struct xfrm_policy *xp, int dir, const struc
static u32 get_acqseq(void)
{
u32 res;
return res;
}
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
-index 4fdb306e..920086a 100644
+index ae36f8e..09d42ac 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
-@@ -804,7 +804,7 @@ static int ieee80211_set_monitor_channel(struct wiphy *wiphy,
+@@ -806,7 +806,7 @@ static int ieee80211_set_monitor_channel(struct wiphy *wiphy,
ret = ieee80211_vif_use_channel(sdata, chandef,
IEEE80211_CHANCTX_EXCLUSIVE);
}
local->_oper_chandef = *chandef;
ieee80211_hw_config(local, 0);
}
-@@ -2920,7 +2920,7 @@ static void ieee80211_mgmt_frame_register(struct wiphy *wiphy,
+@@ -2922,7 +2922,7 @@ static void ieee80211_mgmt_frame_register(struct wiphy *wiphy,
else
local->probe_req_reg--;
break;
ieee80211_queue_work(&local->hw, &local->reconfig_filter);
-@@ -3383,8 +3383,8 @@ static int ieee80211_cfg_get_channel(struct wiphy *wiphy,
+@@ -3385,8 +3385,8 @@ static int ieee80211_cfg_get_channel(struct wiphy *wiphy,
if (chanctx_conf) {
*chandef = chanctx_conf->def;
ret = 0;
/*
* Goal:
diff --git a/net/mac80211/pm.c b/net/mac80211/pm.c
-index 7fc5d0d..07ea536 100644
+index 3401262..d5cd68d 100644
--- a/net/mac80211/pm.c
+++ b/net/mac80211/pm.c
@@ -12,7 +12,7 @@ int __ieee80211_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan)
if (local->wowlan) {
int err = drv_suspend(local, wowlan);
if (err < 0) {
-@@ -113,7 +113,7 @@ int __ieee80211_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan)
+@@ -116,7 +116,7 @@ int __ieee80211_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan)
WARN_ON(!list_empty(&local->chanctx_list));
/* stop hardware - this must stop RX */
return -NF_ACCEPT;
}
+diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
+index 4d4d8f1..e0f9a32 100644
+--- a/net/netfilter/nf_conntrack_proto_tcp.c
++++ b/net/netfilter/nf_conntrack_proto_tcp.c
+@@ -526,7 +526,7 @@ static bool tcp_in_window(const struct nf_conn *ct,
+ const struct nf_conntrack_tuple *tuple = &ct->tuplehash[dir].tuple;
+ __u32 seq, ack, sack, end, win, swin;
+ s16 receiver_offset;
+- bool res;
++ bool res, in_recv_win;
+
+ /*
+ * Get the required data from the packet.
+@@ -649,14 +649,18 @@ static bool tcp_in_window(const struct nf_conn *ct,
+ receiver->td_end, receiver->td_maxend, receiver->td_maxwin,
+ receiver->td_scale);
+
++ /* Is the ending sequence in the receive window (if available)? */
++ in_recv_win = !receiver->td_maxwin ||
++ after(end, sender->td_end - receiver->td_maxwin - 1);
++
+ pr_debug("tcp_in_window: I=%i II=%i III=%i IV=%i\n",
+ before(seq, sender->td_maxend + 1),
+- after(end, sender->td_end - receiver->td_maxwin - 1),
++ (in_recv_win ? 1 : 0),
+ before(sack, receiver->td_end + 1),
+ after(sack, receiver->td_end - MAXACKWINDOW(sender) - 1));
+
+ if (before(seq, sender->td_maxend + 1) &&
+- after(end, sender->td_end - receiver->td_maxwin - 1) &&
++ in_recv_win &&
+ before(sack, receiver->td_end + 1) &&
+ after(sack, receiver->td_end - MAXACKWINDOW(sender) - 1)) {
+ /*
+@@ -725,7 +729,7 @@ static bool tcp_in_window(const struct nf_conn *ct,
+ nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL,
+ "nf_ct_tcp: %s ",
+ before(seq, sender->td_maxend + 1) ?
+- after(end, sender->td_end - receiver->td_maxwin - 1) ?
++ in_recv_win ?
+ before(sack, receiver->td_end + 1) ?
+ after(sack, receiver->td_end - MAXACKWINDOW(sender) - 1) ? "BUG"
+ : "ACK is under the lower bound (possible overly delayed ACK)"
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index bd700b4..4a3dc61 100644
--- a/net/netfilter/nf_conntrack_standalone.c
}
EXPORT_SYMBOL(nf_unregister_sockopt);
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
-index 962e979..d4ae2e9 100644
+index 962e979..e46f350 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -82,7 +82,7 @@ static int nfnl_log_net_id __read_mostly;
};
static struct nfnl_log_net *nfnl_log_pernet(struct net *net)
-@@ -559,7 +559,7 @@ __build_packet_message(struct nfnl_log_net *log,
+@@ -419,6 +419,7 @@ __build_packet_message(struct nfnl_log_net *log,
+ nfmsg->version = NFNETLINK_V0;
+ nfmsg->res_id = htons(inst->group_num);
+
++ memset(&pmsg, 0, sizeof(pmsg));
+ pmsg.hw_protocol = skb->protocol;
+ pmsg.hook = hooknum;
+
+@@ -498,7 +499,10 @@ __build_packet_message(struct nfnl_log_net *log,
+ if (indev && skb->dev &&
+ skb->mac_header != skb->network_header) {
+ struct nfulnl_msg_packet_hw phw;
+- int len = dev_parse_header(skb, phw.hw_addr);
++ int len;
++
++ memset(&phw, 0, sizeof(phw));
++ len = dev_parse_header(skb, phw.hw_addr);
+ if (len > 0) {
+ phw.hw_addrlen = htons(len);
+ if (nla_put(inst->skb, NFULA_HWADDR, sizeof(phw), &phw))
+@@ -559,7 +563,7 @@ __build_packet_message(struct nfnl_log_net *log,
/* global sequence number */
if ((inst->flags & NFULNL_CFG_F_SEQ_GLOBAL) &&
nla_put_be32(inst->skb, NFULA_SEQ_GLOBAL,
goto nla_put_failure;
if (data_len) {
+diff --git a/net/netfilter/nfnetlink_queue_core.c b/net/netfilter/nfnetlink_queue_core.c
+index 5352b2d..e0083ce 100644
+--- a/net/netfilter/nfnetlink_queue_core.c
++++ b/net/netfilter/nfnetlink_queue_core.c
+@@ -444,7 +444,10 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue,
+ if (indev && entskb->dev &&
+ entskb->mac_header != entskb->network_header) {
+ struct nfqnl_msg_packet_hw phw;
+- int len = dev_parse_header(entskb, phw.hw_addr);
++ int len;
++
++ memset(&phw, 0, sizeof(phw));
++ len = dev_parse_header(entskb, phw.hw_addr);
+ if (len) {
+ phw.hw_addrlen = htons(len);
+ if (nla_put(skb, NFQA_HWADDR, sizeof(phw), &phw))
+diff --git a/net/netfilter/xt_TCPMSS.c b/net/netfilter/xt_TCPMSS.c
+index 7011c71..6113cc7 100644
+--- a/net/netfilter/xt_TCPMSS.c
++++ b/net/netfilter/xt_TCPMSS.c
+@@ -52,7 +52,8 @@ tcpmss_mangle_packet(struct sk_buff *skb,
+ {
+ const struct xt_tcpmss_info *info = par->targinfo;
+ struct tcphdr *tcph;
+- unsigned int tcplen, i;
++ int len, tcp_hdrlen;
++ unsigned int i;
+ __be16 oldval;
+ u16 newmss;
+ u8 *opt;
+@@ -64,11 +65,14 @@ tcpmss_mangle_packet(struct sk_buff *skb,
+ if (!skb_make_writable(skb, skb->len))
+ return -1;
+
+- tcplen = skb->len - tcphoff;
++ len = skb->len - tcphoff;
++ if (len < (int)sizeof(struct tcphdr))
++ return -1;
++
+ tcph = (struct tcphdr *)(skb_network_header(skb) + tcphoff);
++ tcp_hdrlen = tcph->doff * 4;
+
+- /* Header cannot be larger than the packet */
+- if (tcplen < tcph->doff*4)
++ if (len < tcp_hdrlen)
+ return -1;
+
+ if (info->mss == XT_TCPMSS_CLAMP_PMTU) {
+@@ -87,9 +91,8 @@ tcpmss_mangle_packet(struct sk_buff *skb,
+ newmss = info->mss;
+
+ opt = (u_int8_t *)tcph;
+- for (i = sizeof(struct tcphdr); i < tcph->doff*4; i += optlen(opt, i)) {
+- if (opt[i] == TCPOPT_MSS && tcph->doff*4 - i >= TCPOLEN_MSS &&
+- opt[i+1] == TCPOLEN_MSS) {
++ for (i = sizeof(struct tcphdr); i <= tcp_hdrlen - TCPOLEN_MSS; i += optlen(opt, i)) {
++ if (opt[i] == TCPOPT_MSS && opt[i+1] == TCPOLEN_MSS) {
+ u_int16_t oldmss;
+
+ oldmss = (opt[i+2] << 8) | opt[i+3];
+@@ -112,9 +115,10 @@ tcpmss_mangle_packet(struct sk_buff *skb,
+ }
+
+ /* There is data after the header so the option can't be added
+- without moving it, and doing so may make the SYN packet
+- itself too large. Accept the packet unmodified instead. */
+- if (tcplen > tcph->doff*4)
++ * without moving it, and doing so may make the SYN packet
++ * itself too large. Accept the packet unmodified instead.
++ */
++ if (len > tcp_hdrlen)
+ return 0;
+
+ /*
+@@ -143,10 +147,10 @@ tcpmss_mangle_packet(struct sk_buff *skb,
+ newmss = min(newmss, (u16)1220);
+
+ opt = (u_int8_t *)tcph + sizeof(struct tcphdr);
+- memmove(opt + TCPOLEN_MSS, opt, tcplen - sizeof(struct tcphdr));
++ memmove(opt + TCPOLEN_MSS, opt, len - sizeof(struct tcphdr));
+
+ inet_proto_csum_replace2(&tcph->check, skb,
+- htons(tcplen), htons(tcplen + TCPOLEN_MSS), 1);
++ htons(len), htons(len + TCPOLEN_MSS), 1);
+ opt[0] = TCPOPT_MSS;
+ opt[1] = TCPOLEN_MSS;
+ opt[2] = (newmss & 0xff00) >> 8;
+diff --git a/net/netfilter/xt_TCPOPTSTRIP.c b/net/netfilter/xt_TCPOPTSTRIP.c
+index b68fa19..625fa1d 100644
+--- a/net/netfilter/xt_TCPOPTSTRIP.c
++++ b/net/netfilter/xt_TCPOPTSTRIP.c
+@@ -38,7 +38,7 @@ tcpoptstrip_mangle_packet(struct sk_buff *skb,
+ struct tcphdr *tcph;
+ u_int16_t n, o;
+ u_int8_t *opt;
+- int len;
++ int len, tcp_hdrlen;
+
+ /* This is a fragment, no TCP header is available */
+ if (par->fragoff != 0)
+@@ -52,7 +52,9 @@ tcpoptstrip_mangle_packet(struct sk_buff *skb,
+ return NF_DROP;
+
+ tcph = (struct tcphdr *)(skb_network_header(skb) + tcphoff);
+- if (tcph->doff * 4 > len)
++ tcp_hdrlen = tcph->doff * 4;
++
++ if (len < tcp_hdrlen)
+ return NF_DROP;
+
+ opt = (u_int8_t *)tcph;
+@@ -61,10 +63,10 @@ tcpoptstrip_mangle_packet(struct sk_buff *skb,
+ * Walk through all TCP options - if we find some option to remove,
+ * set all octets to %TCPOPT_NOP and adjust checksum.
+ */
+- for (i = sizeof(struct tcphdr); i < tcp_hdrlen(skb); i += optl) {
++ for (i = sizeof(struct tcphdr); i < tcp_hdrlen - 1; i += optl) {
+ optl = optlen(opt, i);
+
+- if (i + optl > tcp_hdrlen(skb))
++ if (i + optl > tcp_hdrlen)
+ break;
+
+ if (!tcpoptstrip_test_bit(info->strip_bmap, opt[i]))
diff --git a/net/netfilter/xt_gradm.c b/net/netfilter/xt_gradm.c
new file mode 100644
index 0000000..c566332
);
diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c
-index 2fd6dbe..1032269 100644
+index 1076fe1..f190285 100644
--- a/net/netlink/genetlink.c
+++ b/net/netlink/genetlink.c
@@ -310,18 +310,20 @@ int genl_register_ops(struct genl_family *family, struct genl_ops *ops)
_proto("Tx RESPONSE %%%u", ntohl(hdr->serial));
ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 3, len);
-diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
-index ca8e0a5..1f9c314 100644
---- a/net/sched/sch_atm.c
-+++ b/net/sched/sch_atm.c
-@@ -605,6 +605,7 @@ static int atm_tc_dump_class(struct Qdisc *sch, unsigned long cl,
- struct sockaddr_atmpvc pvc;
- int state;
-
-+ memset(&pvc, 0, sizeof(pvc));
- pvc.sap_family = AF_ATMPVC;
- pvc.sap_addr.itf = flow->vcc->dev ? flow->vcc->dev->number : -1;
- pvc.sap_addr.vpi = flow->vcc->vpi;
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index 391a245..296b3d7 100644
--- a/net/sctp/ipv6.c
int i;
table = kmemdup(sctp_net_table, sizeof(sctp_net_table), GFP_KERNEL);
+diff --git a/net/sctp/transport.c b/net/sctp/transport.c
+index 098f1d5f..60da2f7 100644
+--- a/net/sctp/transport.c
++++ b/net/sctp/transport.c
+@@ -178,12 +178,12 @@ static void sctp_transport_destroy(struct sctp_transport *transport)
+ {
+ SCTP_ASSERT(transport->dead, "Transport is not dead", return);
+
+- call_rcu(&transport->rcu, sctp_transport_destroy_rcu);
+-
+ sctp_packet_free(&transport->packet);
+
+ if (transport->asoc)
+ sctp_association_put(transport->asoc);
++
++ call_rcu(&transport->rcu, sctp_transport_destroy_rcu);
+ }
+
+ /* Start T3_rtx timer if it is not already running and update the heartbeat
diff --git a/net/socket.c b/net/socket.c
index 4ca1526..df83e47 100644
--- a/net/socket.c
set_fs(KERNEL_DS);
if (level == SOL_SOCKET)
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
-index 5a750b9..ca5d7af 100644
+index 426f8fc..1ef9c32 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -1288,7 +1288,9 @@ call_start(struct rpc_task *task)
table = kmemdup(unix_table, sizeof(unix_table), GFP_KERNEL);
if (table == NULL)
+diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
+index 3f77f42..662d89b 100644
+--- a/net/vmw_vsock/af_vsock.c
++++ b/net/vmw_vsock/af_vsock.c
+@@ -335,7 +335,7 @@ void vsock_for_each_connected_socket(void (*fn)(struct sock *sk))
+ for (i = 0; i < ARRAY_SIZE(vsock_connected_table); i++) {
+ struct vsock_sock *vsk;
+ list_for_each_entry(vsk, &vsock_connected_table[i],
+- connected_table);
++ connected_table)
+ fn(sk_vsock(vsk));
+ }
+
diff --git a/net/wireless/wext-core.c b/net/wireless/wext-core.c
index c8717c1..08539f5 100644
--- a/net/wireless/wext-core.c
shdr = (Elf_Shdr *)((char *)ehdr + _r(&ehdr->e_shoff));
shstrtab_sec = shdr + r2(&ehdr->e_shstrndx);
diff --git a/security/Kconfig b/security/Kconfig
-index e9c6ac7..80e2642 100644
+index e9c6ac7..3e3f362 100644
--- a/security/Kconfig
+++ b/security/Kconfig
-@@ -4,6 +4,954 @@
+@@ -4,6 +4,959 @@
menu "Security options"
+config PAX_NOEXEC
+ bool "Enforce non-executable pages"
+ default y if GRKERNSEC_CONFIG_AUTO
-+ depends on ALPHA || (ARM && (CPU_V6 || CPU_V7)) || IA64 || MIPS || PARISC || PPC || S390 || SPARC || X86
++ depends on ALPHA || (ARM && (CPU_V6 || CPU_V6K || CPU_V7)) || IA64 || MIPS || PARISC || PPC || S390 || SPARC || X86
+ help
+ By design some architectures do not allow for protecting memory
+ pages against execution or even if they do, Linux does not make
+config PAX_KERNEXEC
+ bool "Enforce non-executable kernel pages"
+ default y if GRKERNSEC_CONFIG_AUTO && (GRKERNSEC_CONFIG_VIRT_NONE || (GRKERNSEC_CONFIG_VIRT_EPT && GRKERNSEC_CONFIG_VIRT_GUEST) || (GRKERNSEC_CONFIG_VIRT_EPT && GRKERNSEC_CONFIG_VIRT_KVM))
-+ depends on (X86 || (ARM && (CPU_V6 || CPU_V7) && !(ARM_LPAE && MODULES))) && !XEN
++ depends on (X86 || (ARM && (CPU_V6 || CPU_V6K || CPU_V7) && !(ARM_LPAE && MODULES))) && !XEN
+ select PAX_PER_CPU_PGD if X86_64 || (X86_32 && X86_PAE)
+ select PAX_KERNEXEC_PLUGIN if X86_64
+ help
+config PAX_MEMORY_UDEREF
+ bool "Prevent invalid userland pointer dereference"
+ default y if GRKERNSEC_CONFIG_AUTO && !(X86_64 && GRKERNSEC_CONFIG_PRIORITY_PERF) && (GRKERNSEC_CONFIG_VIRT_NONE || GRKERNSEC_CONFIG_VIRT_EPT)
-+ depends on (X86 || (ARM && (CPU_V6 || CPU_V7) && !ARM_LPAE)) && !UML_X86 && !XEN
++ depends on (X86 || (ARM && (CPU_V6 || CPU_V6K || CPU_V7) && !ARM_LPAE)) && !UML_X86 && !XEN
+ select PAX_PER_CPU_PGD if X86_64
+ help
+ By saying Y here the kernel will be prevented from dereferencing
+ VMs running on CPUs without hardware virtualization support (i.e.,
+ the majority of IA-32 CPUs) will likely experience the slowdown.
+
++ On X86_64 the kernel will make use of PCID support when available
++ (Intel's Westmere, Sandy Bridge, etc) for better security (default)
++ or performance impact. Pass pax_weakuderef on the kernel command
++ line to choose the latter.
++
+config PAX_REFCOUNT
+ bool "Prevent various kernel object reference counter overflows"
+ default y if GRKERNSEC_CONFIG_AUTO
-+ depends on GRKERNSEC && ((ARM && (CPU_32v6 || CPU_32v6K || CPU_32v7)) || SPARC64 || X86)
++ depends on GRKERNSEC && ((ARM && (CPU_V6 || CPU_V6K || CPU_V7)) || SPARC64 || X86)
+ help
+ By saying Y here the kernel will detect and prevent overflowing
+ various (but not all) kinds of object reference counters. Such
source security/keys/Kconfig
config SECURITY_DMESG_RESTRICT
-@@ -103,7 +1051,7 @@ config INTEL_TXT
+@@ -103,7 +1056,7 @@ config INTEL_TXT
config LSM_MMAP_MIN_ADDR
int "Low address space for LSM to protect from user allocation"
depends on SECURITY && SECURITY_SELINUX
+
+ return 0;
+}
+diff --git a/tools/lib/lk/Makefile b/tools/lib/lk/Makefile
+index 926cbf3..b8403e0 100644
+--- a/tools/lib/lk/Makefile
++++ b/tools/lib/lk/Makefile
+@@ -10,7 +10,7 @@ LIB_OBJS += $(OUTPUT)debugfs.o
+
+ LIBFILE = liblk.a
+
+-CFLAGS = -ggdb3 -Wall -Wextra -std=gnu99 -Werror -O6 -D_FORTIFY_SOURCE=2 $(EXTRA_WARNINGS) $(EXTRA_CFLAGS) -fPIC
++CFLAGS = -ggdb3 -Wall -Wextra -std=gnu99 -Werror -O6 -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2 $(EXTRA_WARNINGS) $(EXTRA_CFLAGS) -fPIC
+ EXTLIBS = -lpthread -lrt -lelf -lm
+ ALL_CFLAGS = $(CFLAGS) $(BASIC_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
+ ALL_LDFLAGS = $(LDFLAGS)
+diff --git a/tools/perf/Makefile b/tools/perf/Makefile
+index b0f164b..63c9f7d 100644
+--- a/tools/perf/Makefile
++++ b/tools/perf/Makefile
+@@ -188,7 +188,7 @@ endif
+
+ ifndef PERF_DEBUG
+ ifeq ($(call try-cc,$(SOURCE_HELLO),$(CFLAGS) -D_FORTIFY_SOURCE=2,-D_FORTIFY_SOURCE=2),y)
+- CFLAGS := $(CFLAGS) -D_FORTIFY_SOURCE=2
++ CFLAGS := $(CFLAGS) -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2
+ endif
+ endif
+
diff --git a/tools/perf/util/include/asm/alternative-asm.h b/tools/perf/util/include/asm/alternative-asm.h
index 6789d78..4afd019e 100644
--- a/tools/perf/util/include/asm/alternative-asm.h
projects / people / ms / ipfire-3.x.git / commitdiff
