diff options
author | Johannes Weiner <hannes@cmpxchg.org> | 2013-09-12 18:13:44 -0400 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2013-09-12 18:38:02 -0400 |
commit | 3812c8c8f3953921ef18544110dafc3505c1ac62 (patch) | |
tree | 8e5efc15fec4700644774df5fb5302f5c82f4a31 /include | |
parent | fb2a6fc56be66c169f8b80e07ed999ba453a2db2 (diff) |
mm: memcg: do not trap chargers with full callstack on OOM
The memcg OOM handling is incredibly fragile and can deadlock. When a
task fails to charge memory, it invokes the OOM killer and loops right
there in the charge code until it succeeds. Comparably, any other task
that enters the charge path at this point will go to a waitqueue right
then and there and sleep until the OOM situation is resolved. The problem
is that these tasks may hold filesystem locks and the mmap_sem; locks that
the selected OOM victim may need to exit.
For example, in one reported case, the task invoking the OOM killer was
about to charge a page cache page during a write(), which holds the
i_mutex. The OOM killer selected a task that was just entering truncate()
and trying to acquire the i_mutex:
OOM invoking task:
mem_cgroup_handle_oom+0x241/0x3b0
mem_cgroup_cache_charge+0xbe/0xe0
add_to_page_cache_locked+0x4c/0x140
add_to_page_cache_lru+0x22/0x50
grab_cache_page_write_begin+0x8b/0xe0
ext3_write_begin+0x88/0x270
generic_file_buffered_write+0x116/0x290
__generic_file_aio_write+0x27c/0x480
generic_file_aio_write+0x76/0xf0 # takes ->i_mutex
do_sync_write+0xea/0x130
vfs_write+0xf3/0x1f0
sys_write+0x51/0x90
system_call_fastpath+0x18/0x1d
OOM kill victim:
do_truncate+0x58/0xa0 # takes i_mutex
do_last+0x250/0xa30
path_openat+0xd7/0x440
do_filp_open+0x49/0xa0
do_sys_open+0x106/0x240
sys_open+0x20/0x30
system_call_fastpath+0x18/0x1d
The OOM handling task will retry the charge indefinitely while the OOM
killed task is not releasing any resources.
A similar scenario can happen when the kernel OOM killer for a memcg is
disabled and a userspace task is in charge of resolving OOM situations.
In this case, ALL tasks that enter the OOM path will be made to sleep on
the OOM waitqueue and wait for userspace to free resources or increase
the group's limit. But a userspace OOM handler is prone to deadlock
itself on the locks held by the waiting tasks. For example one of the
sleeping tasks may be stuck in a brk() call with the mmap_sem held for
writing but the userspace handler, in order to pick an optimal victim,
may need to read files from /proc/<pid>, which tries to acquire the same
mmap_sem for reading and deadlocks.
This patch changes the way tasks behave after detecting a memcg OOM and
makes sure nobody loops or sleeps with locks held:
1. When OOMing in a user fault, invoke the OOM killer and restart the
fault instead of looping on the charge attempt. This way, the OOM
victim can not get stuck on locks the looping task may hold.
2. When OOMing in a user fault but somebody else is handling it
(either the kernel OOM killer or a userspace handler), don't go to
sleep in the charge context. Instead, remember the OOMing memcg in
the task struct and then fully unwind the page fault stack with
-ENOMEM. pagefault_out_of_memory() will then call back into the
memcg code to check if the -ENOMEM came from the memcg, and then
either put the task to sleep on the memcg's OOM waitqueue or just
restart the fault. The OOM victim can no longer get stuck on any
lock a sleeping task may hold.
Debugged by Michal Hocko.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: azurIt <azurit@pobox.sk>
Acked-by: Michal Hocko <mhocko@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'include')
-rw-r--r-- | include/linux/memcontrol.h | 21 | ||||
-rw-r--r-- | include/linux/sched.h | 4 |
2 files changed, 25 insertions, 0 deletions
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 34ac6497d01a..89d576cfcc4c 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h | |||
@@ -157,6 +157,10 @@ extern void mem_cgroup_replace_page_cache(struct page *oldpage, | |||
157 | * | 157 | * |
158 | * Toggle whether a failed memcg charge should invoke the OOM killer | 158 | * Toggle whether a failed memcg charge should invoke the OOM killer |
159 | * or just return -ENOMEM. Returns the previous toggle state. | 159 | * or just return -ENOMEM. Returns the previous toggle state. |
160 | * | ||
161 | * NOTE: Any path that enables the OOM killer before charging must | ||
162 | * call mem_cgroup_oom_synchronize() afterward to finalize the | ||
163 | * OOM handling and clean up. | ||
160 | */ | 164 | */ |
161 | static inline bool mem_cgroup_toggle_oom(bool new) | 165 | static inline bool mem_cgroup_toggle_oom(bool new) |
162 | { | 166 | { |
@@ -182,6 +186,13 @@ static inline void mem_cgroup_disable_oom(void) | |||
182 | WARN_ON(old == false); | 186 | WARN_ON(old == false); |
183 | } | 187 | } |
184 | 188 | ||
189 | static inline bool task_in_memcg_oom(struct task_struct *p) | ||
190 | { | ||
191 | return p->memcg_oom.in_memcg_oom; | ||
192 | } | ||
193 | |||
194 | bool mem_cgroup_oom_synchronize(void); | ||
195 | |||
185 | #ifdef CONFIG_MEMCG_SWAP | 196 | #ifdef CONFIG_MEMCG_SWAP |
186 | extern int do_swap_account; | 197 | extern int do_swap_account; |
187 | #endif | 198 | #endif |
@@ -427,6 +438,16 @@ static inline void mem_cgroup_disable_oom(void) | |||
427 | { | 438 | { |
428 | } | 439 | } |
429 | 440 | ||
441 | static inline bool task_in_memcg_oom(struct task_struct *p) | ||
442 | { | ||
443 | return false; | ||
444 | } | ||
445 | |||
446 | static inline bool mem_cgroup_oom_synchronize(void) | ||
447 | { | ||
448 | return false; | ||
449 | } | ||
450 | |||
430 | static inline void mem_cgroup_inc_page_stat(struct page *page, | 451 | static inline void mem_cgroup_inc_page_stat(struct page *page, |
431 | enum mem_cgroup_page_stat_item idx) | 452 | enum mem_cgroup_page_stat_item idx) |
432 | { | 453 | { |
diff --git a/include/linux/sched.h b/include/linux/sched.h index 9ce1fa53031f..6682da36b293 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h | |||
@@ -1395,6 +1395,10 @@ struct task_struct { | |||
1395 | unsigned int memcg_kmem_skip_account; | 1395 | unsigned int memcg_kmem_skip_account; |
1396 | struct memcg_oom_info { | 1396 | struct memcg_oom_info { |
1397 | unsigned int may_oom:1; | 1397 | unsigned int may_oom:1; |
1398 | unsigned int in_memcg_oom:1; | ||
1399 | unsigned int oom_locked:1; | ||
1400 | int wakeups; | ||
1401 | struct mem_cgroup *wait_on_memcg; | ||
1398 | } memcg_oom; | 1402 | } memcg_oom; |
1399 | #endif | 1403 | #endif |
1400 | #ifdef CONFIG_UPROBES | 1404 | #ifdef CONFIG_UPROBES |