x86/mm: Handle mm_fault_error() in kernel space

mm_fault_error() should not execute oom-killer, if page fault
occurs in kernel space.  E.g. in copy_from_user()/copy_to_user().

This would happen if we find ourselves in OOM on a
copy_to_user(), or a copy_from_user() which faults.

Without this patch, the kernels hangs up in copy_from_user(),
because OOM killer sends SIG_KILL to current process, but it
can't handle a signal while in syscall, then the kernel returns
to copy_from_user(), reexcute current command and provokes
page_fault again.

With this patch the kernel return -EFAULT from copy_from_user().

The code, which checks that page fault occurred in kernel space,
has been copied from do_sigbus().

This situation is handled by the same way on powerpc, xtensa,
tile, ...

Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: <stable@kernel.org>
LKML-Reference: <201103092322.p29NMNPH001682@imap1.linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 7d90ceb882a41ec55f0d8aea7b2c40cd806afd81..ffc7be104fc2a955f33d2fb59101752ee7d1ac2d 100644 (file)
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -828,6 +828,13 @@ mm_fault_error(struct pt_regs *regs, unsigned long error_code,
               unsigned long address, unsigned int fault)
 {
        if (fault & VM_FAULT_OOM) {
+               /* Kernel mode? Handle exceptions or die: */
+               if (!(error_code & PF_USER)) {
+                       up_read(&current->mm->mmap_sem);
+                       no_context(regs, error_code, address);
+                       return;
+               }
+
                out_of_memory(regs, error_code, address);
        } else {
                if (fault & (VM_FAULT_SIGBUS|VM_FAULT_HWPOISON|