Skip to content

K
Kernel

openeuler / Kernel
12 个月 前同步成功

通知 5
Star 0
Fork 0
K
Kernel

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 15 4月, 2011 1 次提交
  3. 13 4月, 2011 1 次提交
  5. 14 1月, 2011 3 次提交
  7. 16 12月, 2010 1 次提交
    • T
      install_special_mapping skips security_file_mmap check. · 462e635e
      由 Tavis Ormandy 提交于 
      The install_special_mapping routine (used, for example, to setup the
vdso) skips the security check before insert_vm_struct, allowing a local
attacker to bypass the mmap_min_addr security restriction by limiting
the available pages for special mappings.

bprm_mm_init() also skips the check, and although I don't think this can
be used to bypass any restrictions, I don't see any reason not to have
the security check.

  $ uname -m
  x86_64
  $ cat /proc/sys/vm/mmap_min_addr
  65536
  $ cat install_special_mapping.s
  section .bss
      resb BSS_SIZE
  section .text
      global _start
      _start:
          mov     eax, __NR_pause
          int     0x80
  $ nasm -D__NR_pause=29 -DBSS_SIZE=0xfffed000 -f elf -o install_special_mapping.o install_special_mapping.s
  $ ld -m elf_i386 -Ttext=0x10000 -Tbss=0x11000 -o install_special_mapping install_special_mapping.o
  $ ./install_special_mapping &
  [1] 14303
  $ cat /proc/14303/maps
  0000f000-00010000 r-xp 00000000 00:00 0                                  [vdso]
  00010000-00011000 r-xp 00001000 00:19 2453665                            /home/taviso/install_special_mapping
  00011000-ffffe000 rwxp 00000000 00:00 0                                  [stack]

It's worth noting that Red Hat are shipping with mmap_min_addr set to
4096.
Signed-off-by: NTavis Ormandy <taviso@google.com>
Acked-by: NKees Cook <kees@ubuntu.com>
Acked-by: NRobert Swiecki <swiecki@google.com>
[ Changed to not drop the error code - akpm ]
Reviewed-by: NJames Morris <jmorris@namei.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      462e635e
  9. 30 10月, 2010 1 次提交
    • A
      audit mmap · 120a795d
      由 Al Viro 提交于 
      Normal syscall audit doesn't catch 5th argument of syscall.  It also
doesn't catch the contents of userland structures pointed to be
syscall argument, so for both old and new mmap(2) ABI it doesn't
record the descriptor we are mapping.  For old one it also misses
flags.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
      120a795d
  11. 23 9月, 2010 1 次提交
  13. 25 8月, 2010 1 次提交
  15. 21 8月, 2010 1 次提交
  17. 10 8月, 2010 4 次提交
  19. 09 6月, 2010 1 次提交
  21. 27 4月, 2010 1 次提交
  23. 13 4月, 2010 2 次提交
    • L
      vma_adjust: fix the copying of anon_vma chains · 287d97ac
      由 Linus Torvalds 提交于 
      When we move the boundaries between two vma's due to things like
mprotect, we need to make sure that the anon_vma of the pages that got
moved from one vma to another gets properly copied around.  And that was
not always the case, in this rather hard-to-follow code sequence.

Clarify the code, and fix it so that it copies the anon_vma from the
right source.
Reviewed-by: NRik van Riel <riel@redhat.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Tested-by: Borislav Petkov <bp@alien8.de> [ "Yeah, not so much this one either" ]
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      287d97ac
    • L
      Simplify and comment on anon_vma re-use for anon_vma_prepare() · d0e9fe17
      由 Linus Torvalds 提交于 
      This changes the anon_vma reuse case to require that we only reuse
simple anon_vma's - ie the case when the vma only has a single anon_vma
associated with it.

This means that a reuse of an anon_vma from an adjacent vma will always
guarantee that both vma's are associated not only with the same
anon_vma, they will also have the same anon_vma chain (of just a single
entry in this case).

And since anon_vma re-use was the only case where the same anon_vma
might be associated with different chains of anon_vma's, we now have the
case that every vma that shares the same anon_vma will always also have
the same chain.  That makes it much easier to think about merging vma's
that share the same anon_vma's: you can always just drop the other
anon_vma chain in anon_vma_merge() since you know that they are always
identical.

This also splits up the function to validate the anon_vma re-use, and
adds a lot of commentary about the possible races.
Reviewed-by: NRik van Riel <riel@redhat.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Tested-by: Borislav Petkov <bp@alien8.de> [ "That didn't fix it" ]
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      d0e9fe17
  25. 13 3月, 2010 1 次提交
  27. 07 3月, 2010 5 次提交
    • R
      mm: remove VM_LOCK_RMAP code · fc148a5f
      由 Rik van Riel 提交于 
      When a VMA is in an inconsistent state during setup or teardown, the worst
that can happen is that the rmap code will not be able to find the page.

The mapping is in the process of being torn down (PTEs just got
invalidated by munmap), or set up (no PTEs have been instantiated yet).

It is also impossible for the rmap code to follow a pointer to an already
freed VMA, because the rmap code holds the anon_vma->lock, which the VMA
teardown code needs to take before the VMA is removed from the anon_vma
chain.

Hence, we should not need the VM_LOCK_RMAP locking at all.
Signed-off-by: NRik van Riel <riel@redhat.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      fc148a5f
    • R
      mm: change anon_vma linking to fix multi-process server scalability issue · 5beb4930
      由 Rik van Riel 提交于 
      The old anon_vma code can lead to scalability issues with heavily forking
workloads.  Specifically, each anon_vma will be shared between the parent
process and all its child processes.

In a workload with 1000 child processes and a VMA with 1000 anonymous
pages per process that get COWed, this leads to a system with a million
anonymous pages in the same anon_vma, each of which is mapped in just one
of the 1000 processes.  However, the current rmap code needs to walk them
all, leading to O(N) scanning complexity for each page.

This can result in systems where one CPU is walking the page tables of
1000 processes in page_referenced_one, while all other CPUs are stuck on
the anon_vma lock.  This leads to catastrophic failure for a benchmark
like AIM7, where the total number of processes can reach in the tens of
thousands.  Real workloads are still a factor 10 less process intensive
than AIM7, but they are catching up.

This patch changes the way anon_vmas and VMAs are linked, which allows us
to associate multiple anon_vmas with a VMA.  At fork time, each child
process gets its own anon_vmas, in which its COWed pages will be
instantiated.  The parents' anon_vma is also linked to the VMA, because
non-COWed pages could be present in any of the children.

This reduces rmap scanning complexity to O(1) for the pages of the 1000
child processes, with O(N) complexity for at most 1/N pages in the system.
 This reduces the average scanning cost in heavily forking workloads from
O(N) to 2.

The only real complexity in this patch stems from the fact that linking a
VMA to anon_vmas now involves memory allocations.  This means vma_adjust
can fail, if it needs to attach a VMA to anon_vma structures.  This in
turn means error handling needs to be added to the calling functions.

A second source of complexity is that, because there can be multiple
anon_vmas, the anon_vma linking in vma_adjust can no longer be done under
"the" anon_vma lock.  To prevent the rmap code from walking up an
incomplete VMA, this patch introduces the VM_LOCK_RMAP VMA flag.  This bit
flag uses the same slot as the NOMMU VM_MAPPED_COPY, with an ifdef in mm.h
to make sure it is impossible to compile a kernel that needs both symbolic
values for the same bitflag.

Some test results:

Without the anon_vma changes, when AIM7 hits around 9.7k users (on a test
box with 16GB RAM and not quite enough IO), the system ends up running
>99% in system time, with every CPU on the same anon_vma lock in the
pageout code.

With these changes, AIM7 hits the cross-over point around 29.7k users.
This happens with ~99% IO wait time, there never seems to be any spike in
system time.  The anon_vma lock contention appears to be resolved.

[akpm@linux-foundation.org: cleanups]
Signed-off-by: NRik van Riel <riel@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      5beb4930
    • J
      mm: use rlimit helpers · 59e99e5b
      由 Jiri Slaby 提交于 
      Make sure compiler won't do weird things with limits.  E.g.  fetching them
twice may return 2 different values after writable limits are implemented.

I.e.  either use rlimit helpers added in
3e10e716 ("resource: add helpers for
fetching rlimits") or ACCESS_ONCE if not applicable.
Signed-off-by: NJiri Slaby <jslaby@suse.cz>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      59e99e5b
    • K
      mm: mlock_vma_pages_range() only return success or failure · 06f9d8c2
      由 KOSAKI Motohiro 提交于 
      Currently, mlock_vma_pages_range() only return len or 0.  then current
error handling of mmap_region() is meaningless complex.

This patch makes simplify and makes consist with brk() code.
Signed-off-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: KAMEZAWA Hiroyuki <kamewzawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      06f9d8c2
    • K
      mm: mlock_vma_pages_range() never return negative value · c58267c3
      由 KOSAKI Motohiro 提交于 
      Currently, mlock_vma_pages_range() never return negative value.  Then, we
can remove some worthless error check.
Signed-off-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: KAMEZAWA Hiroyuki <kamewzawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      c58267c3
  29. 31 12月, 2009 1 次提交
  31. 16 12月, 2009 2 次提交
    • M
      mm: uncached vma support with writenotify · c9d0bf24
      由 Magnus Damm 提交于 
      Modify the generic mmap() code to keep the cache attribute in
vma->vm_page_prot regardless if writenotify is enabled or not.  Without
this patch the cache configuration selected by f_op->mmap() is overwritten
if writenotify is enabled, making it impossible to keep the vma uncached.

Needed by drivers such as drivers/video/sh_mobile_lcdcfb.c which uses
deferred io together with uncached memory.
Signed-off-by: NMagnus Damm <damm@opensource.se>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Jaya Kumar <jayakumar.lkml@gmail.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      c9d0bf24
    • K
      mmap: don't return ENOMEM when mapcount is temporarily exceeded in munmap() · 659ace58
      由 KOSAKI Motohiro 提交于 
      On ia64, the following test program exit abnormally, because glibc thread
library called abort().

 ========================================================
 (gdb) bt
 #0  0xa000000000010620 in __kernel_syscall_via_break ()
 #1  0x20000000003208e0 in raise () from /lib/libc.so.6.1
 #2  0x2000000000324090 in abort () from /lib/libc.so.6.1
 #3  0x200000000027c3e0 in __deallocate_stack () from /lib/libpthread.so.0
 #4  0x200000000027f7c0 in start_thread () from /lib/libpthread.so.0
 #5  0x200000000047ef60 in __clone2 () from /lib/libc.so.6.1
 ========================================================

The fact is, glibc call munmap() when thread exitng time for freeing
stack, and it assume munlock() never fail.  However, munmap() often make
vma splitting and it with many mapcount make -ENOMEM.

Oh well, that's crazy, because stack unmapping never increase mapcount.
The maxcount exceeding is only temporary.  internal temporary exceeding
shouldn't make ENOMEM.

This patch does it.

 test_max_mapcount.c
 ==================================================================
  #include<stdio.h>
  #include<stdlib.h>
  #include<string.h>
  #include<pthread.h>
  #include<errno.h>
  #include<unistd.h>

  #define THREAD_NUM 30000
  #define MAL_SIZE (8*1024*1024)

 void *wait_thread(void *args)
 {
 	void *addr;

 	addr = malloc(MAL_SIZE);
 	sleep(10);

 	return NULL;
 }

 void *wait_thread2(void *args)
 {
 	sleep(60);

 	return NULL;
 }

 int main(int argc, char *argv[])
 {
 	int i;
 	pthread_t thread[THREAD_NUM], th;
 	int ret, count = 0;
 	pthread_attr_t attr;

 	ret = pthread_attr_init(&attr);
 	if(ret) {
 		perror("pthread_attr_init");
 	}

 	ret = pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_DETACHED);
 	if(ret) {
 		perror("pthread_attr_setdetachstate");
 	}

 	for (i = 0; i < THREAD_NUM; i++) {
 		ret = pthread_create(&th, &attr, wait_thread, NULL);
 		if(ret) {
 			fprintf(stderr, "[%d] ", count);
 			perror("pthread_create");
 		} else {
 			printf("[%d] create OK.\n", count);
 		}
 		count++;

 		ret = pthread_create(&thread[i], &attr, wait_thread2, NULL);
 		if(ret) {
 			fprintf(stderr, "[%d] ", count);
 			perror("pthread_create");
 		} else {
 			printf("[%d] create OK.\n", count);
 		}
 		count++;
 	}

 	sleep(3600);
 	return 0;
 }
 ==================================================================

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      659ace58
  33. 11 12月, 2009 3 次提交
  35. 25 10月, 2009 1 次提交
  37. 28 9月, 2009 1 次提交
  39. 22 9月, 2009 8 次提交
    • E
      hugetlb: add MAP_HUGETLB for mmaping pseudo-anonymous huge page regions · 4e52780d
      由 Eric B Munson 提交于 
      Add a flag for mmap that will be used to request a huge page region that
will look like anonymous memory to userspace.  This is accomplished by
using a file on the internal vfsmount.  MAP_HUGETLB is a modifier of
MAP_ANONYMOUS and so must be specified with it.  The region will behave
the same as a MAP_ANONYMOUS region using small pages.

[akpm@linux-foundation.org: fix arch definitions of MAP_HUGETLB]
Signed-off-by: NEric B Munson <ebmunson@us.ibm.com>
Acked-by: NDavid Rientjes <rientjes@google.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Adam Litke <agl@us.ibm.com>
Cc: David Gibson <david@gibson.dropbear.id.au>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      4e52780d
    • H
      mmap: save some cycles for the shared anonymous mapping · f8dbf0a7
      由 Huang Shijie 提交于 
      shmem_zero_setup() does not change vm_start, pgoff or vm_flags, only some
drivers change them (such as /driver/video/bfin-t350mcqb-fb.c).

Move these codes to a more proper place to save cycles for shared
anonymous mapping.
Signed-off-by: NHuang Shijie <shijie8@gmail.com>
Reviewed-by: NMinchan Kim <minchan.kim@gmail.com>
Acked-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      f8dbf0a7
    • L
      mmap: avoid unnecessary anon_vma lock acquisition in vma_adjust() · 252c5f94
      由 Lee Schermerhorn 提交于 
      We noticed very erratic behavior [throughput] with the AIM7 shared
workload running on recent distro [SLES11] and mainline kernels on an
8-socket, 32-core, 256GB x86_64 platform.  On the SLES11 kernel
[2.6.27.19+] with Barcelona processors, as we increased the load [10s of
thousands of tasks], the throughput would vary between two "plateaus"--one
at ~65K jobs per minute and one at ~130K jpm.  The simple patch below
causes the results to smooth out at the ~130k plateau.

But wait, there's more:

We do not see this behavior on smaller platforms--e.g., 4 socket/8 core.
This could be the result of the larger number of cpus on the larger
platform--a scalability issue--or it could be the result of the larger
number of interconnect "hops" between some nodes in this platform and how
the tasks for a given load end up distributed over the nodes' cpus and
memories--a stochastic NUMA effect.

The variability in the results are less pronounced [on the same platform]
with Shanghai processors and with mainline kernels.  With 31-rc6 on
Shanghai processors and 288 file systems on 288 fibre attached storage
volumes, the curves [jpm vs load] are both quite flat with the patched
kernel consistently producing ~3.9% better throughput [~80K jpm vs ~77K
jpm] than the unpatched kernel.

Profiling indicated that the "slow" runs were incurring high[er]
contention on an anon_vma lock in vma_adjust(), apparently called from the
sbrk() system call.

The patch:

A comment in mm/mmap.c:vma_adjust() suggests that we don't really need the
anon_vma lock when we're only adjusting the end of a vma, as is the case
for brk().  The comment questions whether it's worth while to optimize for
this case.  Apparently, on the newer, larger x86_64 platforms, with
interesting NUMA topologies, it is worth while--especially considering
that the patch [if correct!] is quite simple.

We can detect this condition--no overlap with next vma--by noting a NULL
"importer".  The anon_vma pointer will also be NULL in this case, so
simply avoid loading vma->anon_vma to avoid the lock.

However, we DO need to take the anon_vma lock when we're inserting a vma
['insert' non-NULL] even when we have no overlap [NULL "importer"], so we
need to check for 'insert', as well.  And Hugh points out that we should
also take it when adjusting vm_start (so that rmap.c can rely upon
vma_address() while it holds the anon_vma lock).

akpm: Zhang Yanmin reprts a 150% throughput improvement with aim7, so it
might be -stable material even though thiss isn't a regression: "this
issue is not clear on dual socket Nehalem machine (2*4*2 cpu), but is
severe on large machine (4*8*2 cpu)"

[hugh.dickins@tiscali.co.uk: test vma start too]
Signed-off-by: NLee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Eric Whitney <eric.whitney@hp.com>
Tested-by: N"Zhang, Yanmin" <yanmin_zhang@linux.intel.com>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      252c5f94
    • H
      mmap: remove unnecessary code · cdf7b341
      由 Huang Shijie 提交于 
      If (flags & MAP_LOCKED) is true, it means vm_flags has already contained
the bit VM_LOCKED which is set by calc_vm_flag_bits().

So there is no need to reset it again, just remove it.
Signed-off-by: NHuang Shijie <shijie8@gmail.com>
Acked-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      cdf7b341
    • H
      ksm: clean up obsolete references · a913e182
      由 Hugh Dickins 提交于 
      A few cleanups, given the munlock fix: the comment on ksm_test_exit() no
longer applies, and it can be made private to ksm.c; there's no more
reference to mmu_gather or tlb.h, and mmap.c doesn't need ksm.h.
Signed-off-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Acked-by: NIzik Eidus <ieidus@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      a913e182
    • H
      ksm: remove VM_MERGEABLE_FLAGS · 8314c4f2
      由 Hugh Dickins 提交于 
      KSM originally stood for Kernel Shared Memory: but the kernel has long
supported shared memory, and VM_SHARED and VM_MAYSHARE vmas, and KSM is
something else.  So we switched to saying "merge" instead of "share".

But Chris Wright points out that this is confusing where mmap.c merges
adjacent vmas: most especially in the name VM_MERGEABLE_FLAGS, used by
is_mergeable_vma() to let vmas be merged despite flags being different.

Call it VMA_MERGE_DESPITE_FLAGS?  Perhaps, but at present it consists
only of VM_CAN_NONLINEAR: so for now it's clearer on all sides to use
that directly, with a comment on it in is_mergeable_vma().
Signed-off-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Acked-by: NIzik Eidus <ieidus@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      8314c4f2
    • A
      ksm: fix deadlock with munlock in exit_mmap · 1c2fb7a4
      由 Andrea Arcangeli 提交于 
      Rawhide users have reported hang at startup when cryptsetup is run: the
same problem can be simply reproduced by running a program int main() {
mlockall(MCL_CURRENT | MCL_FUTURE); return 0; }

The problem is that exit_mmap() applies munlock_vma_pages_all() to
clean up VM_LOCKED areas, and its current implementation (stupidly)
tries to fault in absent pages, for example where PROT_NONE prevented
them being faulted in when mlocking.  Whereas the "ksm: fix oom
deadlock" patch, knowing there's a race by which KSM might try to fault
in pages after exit_mmap() had finally zapped the range, backs out of
such faults doing nothing when its ksm_test_exit() notices mm_users 0.

So revert that part of "ksm: fix oom deadlock" which moved the
ksm_exit() call from before exit_mmap() to the middle of exit_mmap();
and remove those ksm_test_exit() checks from the page fault paths, so
allowing the munlocking to proceed without interference.

ksm_exit, if there are rmap_items still chained on this mm slot, takes
mmap_sem write side: so preventing KSM from working on an mm while
exit_mmap runs.  And KSM will bail out as soon as it notices that
mm_users is already zero, thanks to its internal ksm_test_exit checks.
So that when a task is killed by OOM killer or the user, KSM will not
indefinitely prevent it from running exit_mmap to release its memory.

This does break a part of what "ksm: fix oom deadlock" was trying to
achieve.  When unmerging KSM (echo 2 >/sys/kernel/mm/ksm), and even
when ksmd itself has to cancel a KSM page, it is possible that the
first OOM-kill victim would be the KSM process being faulted: then its
memory won't be freed until a second victim has been selected (freeing
memory for the unmerging fault to complete).

But the OOM killer is already liable to kill a second victim once the
intended victim's p->mm goes to NULL: so there's not much point in
rejecting this KSM patch before fixing that OOM behaviour.  It is very
much more important to allow KSM users to boot up, than to haggle over
an unlikely and poorly supported OOM case.

We also intend to fix munlocking to not fault pages: at which point
this patch _could_ be reverted; though that would be controversial, so
we hope to find a better solution.
Signed-off-by: NAndrea Arcangeli <aarcange@redhat.com>
Acked-by: NJustin M. Forbes <jforbes@redhat.com>
Acked-for-now-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Izik Eidus <ieidus@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      1c2fb7a4
    • H
      ksm: fix oom deadlock · 9ba69294
      由 Hugh Dickins 提交于 
      There's a now-obvious deadlock in KSM's out-of-memory handling:
imagine ksmd or KSM_RUN_UNMERGE handling, holding ksm_thread_mutex,
trying to allocate a page to break KSM in an mm which becomes the
OOM victim (quite likely in the unmerge case): it's killed and goes
to exit, and hangs there waiting to acquire ksm_thread_mutex.

Clearly we must not require ksm_thread_mutex in __ksm_exit, simple
though that made everything else: perhaps use mmap_sem somehow?
And part of the answer lies in the comments on unmerge_ksm_pages:
__ksm_exit should also leave all the rmap_item removal to ksmd.

But there's a fundamental problem, that KSM relies upon mmap_sem to
guarantee the consistency of the mm it's dealing with, yet exit_mmap
tears down an mm without taking mmap_sem.  And bumping mm_users won't
help at all, that just ensures that the pages the OOM killer assumes
are on their way to being freed will not be freed.

The best answer seems to be, to move the ksm_exit callout from just
before exit_mmap, to the middle of exit_mmap: after the mm's pages
have been freed (if the mmu_gather is flushed), but before its page
tables and vma structures have been freed; and down_write,up_write
mmap_sem there to serialize with KSM's own reliance on mmap_sem.

But KSM then needs to be careful, whenever it downs mmap_sem, to
check that the mm is not already exiting: there's a danger of using
find_vma on a layout that's being torn apart, or writing into page
tables which have been freed for reuse; and even do_anonymous_page
and __do_fault need to check they're not being called by break_ksm
to reinstate a pte after zap_pte_range has zapped that page table.

Though it might be clearer to add an exiting flag, set while holding
mmap_sem in __ksm_exit, that wouldn't cover the issue of reinstating
a zapped pte.  All we need is to check whether mm_users is 0 - but
must remember that ksmd may detect that before __ksm_exit is reached.
So, ksm_test_exit(mm) added to comment such checks on mm->mm_users.

__ksm_exit now has to leave clearing up the rmap_items to ksmd,
that needs ksm_thread_mutex; but shift the exiting mm just after the
ksm_scan cursor so that it will soon be dealt with.  __ksm_enter raise
mm_count to hold the mm_struct, ksmd's exit processing (exactly like
its processing when it finds all VM_MERGEABLEs unmapped) mmdrop it,
similar procedure for KSM_RUN_UNMERGE (which has stopped ksmd).

But also give __ksm_exit a fast path: when there's no complication
(no rmap_items attached to mm and it's not at the ksm_scan cursor),
it can safely do all the exiting work itself.  This is not just an
optimization: when ksmd is not running, the raised mm_count would
otherwise leak mm_structs.
Signed-off-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Acked-by: NIzik Eidus <ieidus@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      9ba69294