In some cases it may happen that pmd_none_or_clear_bad() is called with
the mmap_sem hold in read mode. In those cases the huge page faults can
allocate hugepmds under pmd_none_or_clear_bad() and that can trigger a
false positive from pmd_bad() that will not like to see a pmd
materializing as trans huge.
It's not khugepaged causing the problem, khugepaged holds the mmap_sem
in write mode (and all those sites must hold the mmap_sem in read mode
to prevent pagetables to go away from under them, during code review it
seems vm86 mode on 32bit kernels requires that too unless it's
restricted to 1 thread per process or UP builds). The race is only with
the huge pagefaults that can convert a pmd_none() into a
pmd_trans_huge().
Effectively all these pmd_none_or_clear_bad() sites running with
mmap_sem in read mode are somewhat speculative with the page faults, and
the result is always undefined when they run simultaneously. This is
probably why it wasn't common to run into this. For example if the
madvise(MADV_DONTNEED) runs zap_page_range() shortly before the page
fault, the hugepage will not be zapped, if the page fault runs first it
will be zapped.
Altering pmd_bad() not to error out if it finds hugepmds won't be enough
to fix this, because zap_pmd_range would then proceed to call
zap_pte_range (which would be incorrect if the pmd become a
pmd_trans_huge()).
The simplest way to fix this is to read the pmd in the local stack
(regardless of what we read, no need of actual CPU barriers, only
compiler barrier needed), and be sure it is not changing under the code
that computes its value. Even if the real pmd is changing under the
value we hold on the stack, we don't care. If we actually end up in
zap_pte_range it means the pmd was not none already and it was not huge,
and it can't become huge from under us (khugepaged locking explained
above).
All we need is to enforce that there is no way anymore that in a code
path like below, pmd_trans_huge can be false, but pmd_none_or_clear_bad
can run into a hugepmd. The overhead of a barrier() is just a compiler
tweak and should not be measurable (I only added it for THP builds). I
don't exclude different compiler versions may have prevented the race
too by caching the value of *pmd on the stack (that hasn't been
verified, but it wouldn't be impossible considering
pmd_none_or_clear_bad, pmd_bad, pmd_trans_huge, pmd_none are all inlines
and there's no external function called in between pmd_trans_huge and
pmd_none_or_clear_bad).
if (pmd_trans_huge(*pmd)) {
if (next-addr != HPAGE_PMD_SIZE) {
VM_BUG_ON(!rwsem_is_locked(&tlb->mm->mmap_sem));
split_huge_page_pmd(vma->vm_mm, pmd);
} else if (zap_huge_pmd(tlb, vma, pmd, addr))
continue;
/* fall through */
}
if (pmd_none_or_clear_bad(pmd))
Because this race condition could be exercised without special
privileges this was reported in CVE-2012-1179.
The race was identified and fully explained by Ulrich who debugged it.
I'm quoting his accurate explanation below, for reference.
====== start quote =======
mapcount 0 page_mapcount 1
kernel BUG at mm/huge_memory.c:1384!
At some point prior to the panic, a "bad pmd ..." message similar to the
following is logged on the console:
mm/memory.c:145: bad pmd ffff8800376e1f98(80000000314000e7).
The "bad pmd ..." message is logged by pmd_clear_bad() before it clears
the page's PMD table entry.
143 void pmd_clear_bad(pmd_t *pmd)
144 {
-> 145 pmd_ERROR(*pmd);
146 pmd_clear(pmd);
147 }
After the PMD table entry has been cleared, there is an inconsistency
between the actual number of PMD table entries that are mapping the page
and the page's map count (_mapcount field in struct page). When the page
is subsequently reclaimed, __split_huge_page() detects this inconsistency.
1381 if (mapcount != page_mapcount(page))
1382 printk(KERN_ERR "mapcount %d page_mapcount %d\n",
1383 mapcount, page_mapcount(page));
-> 1384 BUG_ON(mapcount != page_mapcount(page));
The root cause of the problem is a race of two threads in a multithreaded
process. Thread B incurs a page fault on a virtual address that has never
been accessed (PMD entry is zero) while Thread A is executing an madvise()
system call on a virtual address within the same 2 MB (huge page) range.
virtual address space
.---------------------.
| |
| |
.-|---------------------|
| | |
| | |<-- B(fault)
| | |
2 MB | |/////////////////////|-.
huge < |/////////////////////| > A(range)
page | |/////////////////////|-'
| | |
| | |
'-|---------------------|
| |
| |
'---------------------'
- Thread A is executing an madvise(..., MADV_DONTNEED) system call
on the virtual address range "A(range)" shown in the picture.
sys_madvise
// Acquire the semaphore in shared mode.
down_read(¤t->mm->mmap_sem)
...
madvise_vma
switch (behavior)
case MADV_DONTNEED:
madvise_dontneed
zap_page_range
unmap_vmas
unmap_page_range
zap_pud_range
zap_pmd_range
//
// Assume that this huge page has never been accessed.
// I.e. content of the PMD entry is zero (not mapped).
//
if (pmd_trans_huge(*pmd)) {
// We don't get here due to the above assumption.
}
//
// Assume that Thread B incurred a page fault and
.---------> // sneaks in here as shown below.
| //
| if (pmd_none_or_clear_bad(pmd))
| {
| if (unlikely(pmd_bad(*pmd)))
| pmd_clear_bad
| {
| pmd_ERROR
| // Log "bad pmd ..." message here.
| pmd_clear
| // Clear the page's PMD entry.
| // Thread B incremented the map count
| // in page_add_new_anon_rmap(), but
| // now the page is no longer mapped
| // by a PMD entry (-> inconsistency).
| }
| }
|
v
- Thread B is handling a page fault on virtual address "B(fault)" shown
in the picture.
...
do_page_fault
__do_page_fault
// Acquire the semaphore in shared mode.
down_read_trylock(&mm->mmap_sem)
...
handle_mm_fault
if (pmd_none(*pmd) && transparent_hugepage_enabled(vma))
// We get here due to the above assumption (PMD entry is zero).
do_huge_pmd_anonymous_page
alloc_hugepage_vma
// Allocate a new transparent huge page here.
...
__do_huge_pmd_anonymous_page
...
spin_lock(&mm->page_table_lock)
...
page_add_new_anon_rmap
// Here we increment the page's map count (starts at -1).
atomic_set(&page->_mapcount, 0)
set_pmd_at
// Here we set the page's PMD entry which will be cleared
// when Thread A calls pmd_clear_bad().
...
spin_unlock(&mm->page_table_lock)
The mmap_sem does not prevent the race because both threads are acquiring
it in shared mode (down_read). Thread B holds the page_table_lock while
the page's map count and PMD table entry are updated. However, Thread A
does not synchronize on that lock.
====== end quote =======
[akpm@linux-foundation.org: checkpatch fixes]
Reported-by: Ulrich Obergfell <uobergfe@redhat.com>
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: Dave Jones <davej@redhat.com>
Acked-by: Larry Woodman <lwoodman@redhat.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: <stable@vger.kernel.org> [2.6.38+]
Cc: Mark Salter <msalter@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
JONGMAN HEO reports:
With current linus git (commit a25a2b84), I got following build error,
arch/x86/kernel/vm86_32.c: In function 'do_sys_vm86':
arch/x86/kernel/vm86_32.c:340: error: implicit declaration of function '__audit_syscall_exit'
make[3]: *** [arch/x86/kernel/vm86_32.o] Error 1
OK, I can reproduce it (32bit allmodconfig with AUDIT=y, AUDITSYSCALL=n)
It's due to commit d7e7528bcd: "Audit: push audit success and retcode
into arch ptrace.h".
Reported-by: JONGMAN HEO <jongman.heo@samsung.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The audit system previously expected arches calling to audit_syscall_exit to
supply as arguments if the syscall was a success and what the return code was.
Audit also provides a helper AUDITSC_RESULT which was supposed to simplify things
by converting from negative retcodes to an audit internal magic value stating
success or failure. This helper was wrong and could indicate that a valid
pointer returned to userspace was a failed syscall. The fix is to fix the
layering foolishness. We now pass audit_syscall_exit a struct pt_reg and it
in turns calls back into arch code to collect the return value and to
determine if the syscall was a success or failure. We also define a generic
is_syscall_success() macro which determines success/failure based on if the
value is < -MAX_ERRNO. This works for arches like x86 which do not use a
separate mechanism to indicate syscall failure.
We make both the is_syscall_success() and regs_return_value() static inlines
instead of macros. The reason is because the audit function must take a void*
for the regs. (uml calls theirs struct uml_pt_regs instead of just struct
pt_regs so audit_syscall_exit can't take a struct pt_regs). Since the audit
function takes a void* we need to use static inlines to cast it back to the
arch correct structure to dereference it.
The other major change is that on some arches, like ia64, MIPS and ppc, we
change regs_return_value() to give us the negative value on syscall failure.
THE only other user of this macro, kretprobe_example.c, won't notice and it
makes the value signed consistently for the audit functions across all archs.
In arch/sh/kernel/ptrace_64.c I see that we were using regs[9] in the old
audit code as the return value. But the ptrace_64.h code defined the macro
regs_return_value() as regs[3]. I have no idea which one is correct, but this
patch now uses the regs_return_value() function, so it now uses regs[3].
For powerpc we previously used regs->result but now use the
regs_return_value() function which uses regs->gprs[3]. regs->gprs[3] is
always positive so the regs_return_value(), much like ia64 makes it negative
before calling the audit code when appropriate.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: H. Peter Anvin <hpa@zytor.com> [for x86 portion]
Acked-by: Tony Luck <tony.luck@intel.com> [for ia64]
Acked-by: Richard Weinberger <richard@nod.at> [for uml]
Acked-by: David S. Miller <davem@davemloft.net> [for sparc]
Acked-by: Ralf Baechle <ralf@linux-mips.org> [for mips]
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> [for ppc]
split_huge_page_pmd compat code. Each one of those would need to be
expanded to hundred of lines of complex code without a fully reliable
split_huge_page_pmd design.
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Impact: fix kernel bug such as:
BUG: scheduling while atomic: dosemu.bin/19680/0x00000004
See also Ubuntu bug 455067 at
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/455067
Commits 4915a35e35
("Use preempt_conditional_sti/cli in do_int3, like on x86_64.")
and 3d2a71a596
("x86, traps: converge do_debug handlers")
started disabling preemption in int1 and int3 handlers on i386.
The problem with vm86 is that the call to handle_vm86_trap() may jump
straight to entry_32.S and never returns so preempt is never enabled
again, and there is an imbalance in the preempt count.
Commit be716615fe ("x86, vm86:
fix preemption bug"), which was later (accidentally?) reverted by commit
08d68323d1 ("hw-breakpoints: modifying
generic debug exception to use thread-specific debug registers")
fixed the problem for debug exceptions but not for breakpoints.
There are three solutions to this problem.
1. Reenable preemption before calling handle_vm86_trap(). This
was the approach that was later reverted.
2. Do not disable preemption for i386 in breakpoint and debug handlers.
This was the situation before October 2008. As far as I understand
preemption only needs to be disabled on x86_64 because a seperate stack is
used, but it's nice to have things work the same way on
i386 and x86_64.
3. Let handle_vm86_trap() return instead of jumping to assembly code.
By setting a flag in _TIF_WORK_MASK, either TIF_IRET or TIF_NOTIFY_RESUME,
the code in entry_32.S is instructed to return to 32 bit mode from
V86 mode. The logic in entry_32.S was already present to handle signals.
(I chose TIF_IRET because it's slightly more efficient in
do_notify_resume() in signal.c, but in fact TIF_IRET can probably be
replaced by TIF_NOTIFY_RESUME everywhere.)
I'm submitting approach 3, because I believe it is the most elegant
and prevents future confusion. Still, an obvious
preempt_conditional_cli(regs); is necessary in traps.c to correct the
bug.
[ hpa: This is technically a regression, but because:
1. the regression is so old,
2. the patch seems relatively high risk, justifying more testing, and
3. we're late in the 2.6.36-rc cycle,
I'm queuing it up for the 2.6.37 merge window. It might, however,
justify as a -stable backport at a latter time, hence Cc: stable. ]
Signed-off-by: Bart Oldeman <bartoldeman@users.sourceforge.net>
LKML-Reference: <alpine.DEB.2.00.1009231312330.4732@localhost.localdomain>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Alexander van Heukelum <heukelum@fastmail.fm>
Cc: <stable@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Convert these to new PTREGSCALL stubs.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
LKML-Reference: <1260403316-5679-6-git-send-email-brgerst@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86: Clear TS in irq_ts_save() when in an atomic section
x86: Detect use of extended APIC ID for AMD CPUs
x86: memtest: remove 64-bit division
x86, UV: Fix macros for multiple coherency domains
x86: Fix non-lazy GS handling in sys_vm86()
x86: Add quirk for reboot stalls on a Dell Optiplex 360
x86: Fix UV BAU activation descriptor init
This fixes a stack corruption panic or null dereference oops
due to a bad GS in resume_userspace() when returning from
sys_vm86() and calling lockdep_sys_exit().
Only a problem when CONFIG_LOCKDEP and CONFIG_CC_STACKPROTECTOR
enabled.
Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
Cc: H. Peter Anvin <hpa@zytor.com>
LKML-Reference: <1244384628.2323.4.camel@bimbo>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This code has apparently used "0" and not VM86_SIGNAL since Linux
1.1.9, when Linus added VM86_SIGNAL to vm86.h. This patch changes the
code to use the symbolic name.
The magic 0 tripped me up in trying to extend the vm86(2) manpage to
actually explain vm86()'s interface -- my greps for VM86_SIGNAL came up
fruitless.
[ Impact: cleanup; no object code change ]
Signed-off-by: Samuel Bronson <naesten@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Some syscalls need to access the pt_regs structure, either to copy
user register state or to modifiy it. This patch adds stubs to load
the address of the pt_regs struct into the %eax register, and changes
the syscalls to take the pointer as an argument instead of relying on
the assumption that the pt_regs structure overlaps the function
arguments.
Drop the use of regparm(1) due to concern about gcc bugs, and to move
in the direction of the eventual removal of regparm(0) for asmlinkage.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Some syscalls need to access the pt_regs structure, either to copy
user register state or to modifiy it. This patch adds stubs to load
the address of the pt_regs struct into the %eax register, and changes
the syscalls to regparm(1) to receive the pt_regs pointer as the
first argument.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
On x86_32, %gs is handled lazily. It's not saved and restored on
kernel entry/exit but only when necessary which usually is during task
switch but there are few other places. Currently, it's done by
calling savesegment() and loadsegment() explicitly. Define
get_user_gs(), set_user_gs() and task_user_gs() and use them instead.
While at it, clean up register access macros in signal.c.
This cleans up code a bit and will help future changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch renames VM_MASK to X86_VM_MASK (which
in turn defined as alias to X86_EFLAGS_VM) to better
distinguish from virtual memory flags. We can't just
use X86_EFLAGS_VM instead because it is also used
for conditional compilation
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use force_sig in handle_vm86_trap like other machine traps do.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
fastcall is always defined to be empty, remove it from arch/x86
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This changes size-specific register names (eip/rip, esp/rsp, etc.) to
generic names in the thread and tss structures.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
We have a lot of code which differs only by the naming of specific
members of structures that contain registers. In order to enable
additional unifications, this patch drops the e- or r- size prefix
from the register names in struct pt_regs, and drops the x- prefixes
for segment registers on the 32-bit side.
This patch also performs the equivalent renames in some additional
places that might be candidates for unification in the future.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Since the x86 merge, lots of files that referenced their own filenames
are no longer correct. Rather than keep them up to date, just delete
them, as they add no real value.
Additionally:
- fix up comment formatting in scx200_32.c
- Remove a credit from myself in setup_64.c from a time when we had no SCM
- remove longwinded history from tsc_32.c which can be figured out from
git.
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>