2005-04-16 22:20:36 +00:00
|
|
|
/*
|
|
|
|
* linux/mm/oom_kill.c
|
|
|
|
*
|
|
|
|
* Copyright (C) 1998,2000 Rik van Riel
|
|
|
|
* Thanks go out to Claus Fischer for some serious inspiration and
|
|
|
|
* for goading me into coding this file...
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
* Copyright (C) 2010 Google, Inc.
|
|
|
|
* Rewritten by David Rientjes
|
2005-04-16 22:20:36 +00:00
|
|
|
*
|
|
|
|
* The routines in this file are used to kill a process when
|
[PATCH] cpusets: oom_kill tweaks
This patch series extends the use of the cpuset attribute 'mem_exclusive'
to support cpuset configurations that:
1) allow GFP_KERNEL allocations to come from a potentially larger
set of memory nodes than GFP_USER allocations, and
2) can constrain the oom killer to tasks running in cpusets in
a specified subtree of the cpuset hierarchy.
Here's an example usage scenario. For a few hours or more, a large NUMA
system at a University is to be divided in two halves, with a bunch of student
jobs running in half the system under some form of batch manager, and with a
big research project running in the other half. Each of the student jobs is
placed in a small cpuset, but should share the classic Unix time share
facilities, such as buffered pages of files in /bin and /usr/lib. The big
research project wants no interference whatsoever from the student jobs, and
has highly tuned, unusual memory and i/o patterns that intend to make full use
of all the main memory on the nodes available to it.
In this example, we have two big sibling cpusets, one of which is further
divided into a more dynamic set of child cpusets.
We want kernel memory allocations constrained by the two big cpusets, and user
allocations constrained by the smaller child cpusets where present. And we
require that the oom killer not operate across the two halves of this system,
or else the first time a student job runs amuck, the big research project will
likely be first inline to get shot.
Tweaking /proc/<pid>/oom_adj is not ideal -- if the big research project
really does run amuck allocating memory, it should be shot, not some other
task outside the research projects mem_exclusive cpuset.
I propose to extend the use of the 'mem_exclusive' flag of cpusets to manage
such scenarios. Let memory allocations for user space (GFP_USER) be
constrained by a tasks current cpuset, but memory allocations for kernel space
(GFP_KERNEL) by constrained by the nearest mem_exclusive ancestor of the
current cpuset, even though kernel space allocations will still _prefer_ to
remain within the current tasks cpuset, if memory is easily available.
Let the oom killer be constrained to consider only tasks that are in
overlapping mem_exclusive cpusets (it won't help much to kill a task that
normally cannot allocate memory on any of the same nodes as the ones on which
the current task can allocate.)
The current constraints imposed on setting mem_exclusive are unchanged. A
cpuset may only be mem_exclusive if its parent is also mem_exclusive, and a
mem_exclusive cpuset may not overlap any of its siblings memory nodes.
This patch was presented on linux-mm in early July 2005, though did not
generate much feedback at that time. It has been built for a variety of
arch's using cross tools, and built, booted and tested for function on SN2
(ia64).
There are 4 patches in this set:
1) Some minor cleanup, and some improvements to the code layout
of one routine to make subsequent patches cleaner.
2) Add another GFP flag - __GFP_HARDWALL. It marks memory
requests for USER space, which are tightly confined by the
current tasks cpuset.
3) Now memory requests (such as KERNEL) that not marked HARDWALL can
if short on memory, look in the potentially larger pool of memory
defined by the nearest mem_exclusive ancestor cpuset of the current
tasks cpuset.
4) Finally, modify the oom killer to skip any task whose mem_exclusive
cpuset doesn't overlap ours.
Patch (1), the one time I looked on an SN2 (ia64) build, actually saved 32
bytes of kernel text space. Patch (2) has no affect on the size of kernel
text space (it just adds a preprocessor flag). Patches (3) and (4) added
about 600 bytes each of kernel text space, mostly in kernel/cpuset.c, which
matters only if CONFIG_CPUSET is enabled.
This patch:
This patch applies a few comment and code cleanups to mm/oom_kill.c prior to
applying a few small patches to improve cpuset management of memory placement.
The comment changed in oom_kill.c was seriously misleading. The code layout
change in select_bad_process() makes room for adding another condition on
which a process can be spared the oom killer (see the subsequent
cpuset_nodes_overlap patch for this addition).
Also a couple typos and spellos that bugged me, while I was here.
This patch should have no material affect.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-09-06 22:18:09 +00:00
|
|
|
* we're seriously out of memory. This gets called from __alloc_pages()
|
|
|
|
* in mm/page_alloc.c when we really run out of memory.
|
2005-04-16 22:20:36 +00:00
|
|
|
*
|
|
|
|
* Since we won't call these routines often (on a well-configured
|
|
|
|
* machine) this file will double as a 'coding guide' and a signpost
|
|
|
|
* for newbie kernel hackers. It features several pointers to major
|
|
|
|
* kernel subsystems and hints as to where to find out what things do.
|
|
|
|
*/
|
|
|
|
|
2006-10-20 06:28:32 +00:00
|
|
|
#include <linux/oom.h>
|
2005-04-16 22:20:36 +00:00
|
|
|
#include <linux/mm.h>
|
2007-07-29 22:36:13 +00:00
|
|
|
#include <linux/err.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
|
|
|
#include <linux/gfp.h>
|
2005-04-16 22:20:36 +00:00
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/swap.h>
|
|
|
|
#include <linux/timex.h>
|
|
|
|
#include <linux/jiffies.h>
|
2005-09-06 22:18:13 +00:00
|
|
|
#include <linux/cpuset.h>
|
2011-10-16 06:01:52 +00:00
|
|
|
#include <linux/export.h>
|
2006-09-26 06:31:20 +00:00
|
|
|
#include <linux/notifier.h>
|
2008-02-07 08:13:58 +00:00
|
|
|
#include <linux/memcontrol.h>
|
2010-08-10 00:18:52 +00:00
|
|
|
#include <linux/mempolicy.h>
|
security: Fix setting of PF_SUPERPRIV by __capable()
Fix the setting of PF_SUPERPRIV by __capable() as it could corrupt the flags
the target process if that is not the current process and it is trying to
change its own flags in a different way at the same time.
__capable() is using neither atomic ops nor locking to protect t->flags. This
patch removes __capable() and introduces has_capability() that doesn't set
PF_SUPERPRIV on the process being queried.
This patch further splits security_ptrace() in two:
(1) security_ptrace_may_access(). This passes judgement on whether one
process may access another only (PTRACE_MODE_ATTACH for ptrace() and
PTRACE_MODE_READ for /proc), and takes a pointer to the child process.
current is the parent.
(2) security_ptrace_traceme(). This passes judgement on PTRACE_TRACEME only,
and takes only a pointer to the parent process. current is the child.
In Smack and commoncap, this uses has_capability() to determine whether
the parent will be permitted to use PTRACE_ATTACH if normal checks fail.
This does not set PF_SUPERPRIV.
Two of the instances of __capable() actually only act on current, and so have
been changed to calls to capable().
Of the places that were using __capable():
(1) The OOM killer calls __capable() thrice when weighing the killability of a
process. All of these now use has_capability().
(2) cap_ptrace() and smack_ptrace() were using __capable() to check to see
whether the parent was allowed to trace any process. As mentioned above,
these have been split. For PTRACE_ATTACH and /proc, capable() is now
used, and for PTRACE_TRACEME, has_capability() is used.
(3) cap_safe_nice() only ever saw current, so now uses capable().
(4) smack_setprocattr() rejected accesses to tasks other than current just
after calling __capable(), so the order of these two tests have been
switched and capable() is used instead.
(5) In smack_file_send_sigiotask(), we need to allow privileged processes to
receive SIGIO on files they're manipulating.
(6) In smack_task_wait(), we let a process wait for a privileged process,
whether or not the process doing the waiting is privileged.
I've tested this with the LTP SELinux and syscalls testscripts.
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
Acked-by: Andrew G. Morgan <morgan@kernel.org>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: James Morris <jmorris@namei.org>
2008-08-14 10:37:28 +00:00
|
|
|
#include <linux/security.h>
|
2011-03-22 23:30:12 +00:00
|
|
|
#include <linux/ptrace.h>
|
2011-11-01 00:07:07 +00:00
|
|
|
#include <linux/freezer.h>
|
2012-01-10 23:08:09 +00:00
|
|
|
#include <linux/ftrace.h>
|
2012-03-21 23:33:47 +00:00
|
|
|
#include <linux/ratelimit.h>
|
2012-01-10 23:08:09 +00:00
|
|
|
|
|
|
|
#define CREATE_TRACE_POINTS
|
|
|
|
#include <trace/events/oom.h>
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2006-06-23 09:03:13 +00:00
|
|
|
int sysctl_panic_on_oom;
|
2007-10-17 06:25:56 +00:00
|
|
|
int sysctl_oom_kill_allocating_task;
|
2010-08-10 00:18:53 +00:00
|
|
|
int sysctl_oom_dump_tasks = 1;
|
2009-01-06 22:39:00 +00:00
|
|
|
static DEFINE_SPINLOCK(zone_scan_lock);
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2011-11-01 00:07:18 +00:00
|
|
|
/*
|
|
|
|
* compare_swap_oom_score_adj() - compare and swap current's oom_score_adj
|
|
|
|
* @old_val: old oom_score_adj for compare
|
|
|
|
* @new_val: new oom_score_adj for swap
|
|
|
|
*
|
|
|
|
* Sets the oom_score_adj value for current to @new_val iff its present value is
|
|
|
|
* @old_val. Usually used to reinstate a previous value to prevent racing with
|
|
|
|
* userspacing tuning the value in the interim.
|
|
|
|
*/
|
|
|
|
void compare_swap_oom_score_adj(int old_val, int new_val)
|
|
|
|
{
|
|
|
|
struct sighand_struct *sighand = current->sighand;
|
|
|
|
|
|
|
|
spin_lock_irq(&sighand->siglock);
|
|
|
|
if (current->signal->oom_score_adj == old_val)
|
|
|
|
current->signal->oom_score_adj = new_val;
|
2012-01-10 23:08:09 +00:00
|
|
|
trace_oom_score_adj_update(current);
|
2011-11-01 00:07:18 +00:00
|
|
|
spin_unlock_irq(&sighand->siglock);
|
|
|
|
}
|
|
|
|
|
2011-05-25 00:11:40 +00:00
|
|
|
/**
|
|
|
|
* test_set_oom_score_adj() - set current's oom_score_adj and return old value
|
|
|
|
* @new_val: new oom_score_adj value
|
|
|
|
*
|
|
|
|
* Sets the oom_score_adj value for current to @new_val with proper
|
|
|
|
* synchronization and returns the old value. Usually used to temporarily
|
|
|
|
* set a value, save the old value in the caller, and then reinstate it later.
|
|
|
|
*/
|
|
|
|
int test_set_oom_score_adj(int new_val)
|
|
|
|
{
|
|
|
|
struct sighand_struct *sighand = current->sighand;
|
|
|
|
int old_val;
|
|
|
|
|
|
|
|
spin_lock_irq(&sighand->siglock);
|
|
|
|
old_val = current->signal->oom_score_adj;
|
2011-11-01 00:07:15 +00:00
|
|
|
current->signal->oom_score_adj = new_val;
|
2012-01-10 23:08:09 +00:00
|
|
|
trace_oom_score_adj_update(current);
|
2011-05-25 00:11:40 +00:00
|
|
|
spin_unlock_irq(&sighand->siglock);
|
|
|
|
|
|
|
|
return old_val;
|
|
|
|
}
|
|
|
|
|
2010-08-10 00:18:52 +00:00
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
/**
|
|
|
|
* has_intersects_mems_allowed() - check task eligiblity for kill
|
|
|
|
* @tsk: task struct of which task to consider
|
|
|
|
* @mask: nodemask passed to page allocator for mempolicy ooms
|
|
|
|
*
|
|
|
|
* Task eligibility is determined by whether or not a candidate task, @tsk,
|
|
|
|
* shares the same mempolicy nodes as current if it is bound by such a policy
|
|
|
|
* and whether or not it has the same set of allowed cpuset nodes.
|
2009-09-22 00:03:14 +00:00
|
|
|
*/
|
2010-08-10 00:18:52 +00:00
|
|
|
static bool has_intersects_mems_allowed(struct task_struct *tsk,
|
|
|
|
const nodemask_t *mask)
|
2009-09-22 00:03:14 +00:00
|
|
|
{
|
2010-08-10 00:18:52 +00:00
|
|
|
struct task_struct *start = tsk;
|
2009-09-22 00:03:14 +00:00
|
|
|
|
|
|
|
do {
|
2010-08-10 00:18:52 +00:00
|
|
|
if (mask) {
|
|
|
|
/*
|
|
|
|
* If this is a mempolicy constrained oom, tsk's
|
|
|
|
* cpuset is irrelevant. Only return true if its
|
|
|
|
* mempolicy intersects current, otherwise it may be
|
|
|
|
* needlessly killed.
|
|
|
|
*/
|
|
|
|
if (mempolicy_nodemask_intersects(tsk, mask))
|
|
|
|
return true;
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* This is not a mempolicy constrained oom, so only
|
|
|
|
* check the mems of tsk's cpuset.
|
|
|
|
*/
|
|
|
|
if (cpuset_mems_allowed_intersects(current, tsk))
|
|
|
|
return true;
|
|
|
|
}
|
2010-08-10 00:19:39 +00:00
|
|
|
} while_each_thread(start, tsk);
|
|
|
|
|
2010-08-10 00:18:52 +00:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
static bool has_intersects_mems_allowed(struct task_struct *tsk,
|
|
|
|
const nodemask_t *mask)
|
|
|
|
{
|
|
|
|
return true;
|
2009-09-22 00:03:14 +00:00
|
|
|
}
|
2010-08-10 00:18:52 +00:00
|
|
|
#endif /* CONFIG_NUMA */
|
2009-09-22 00:03:14 +00:00
|
|
|
|
2010-08-10 00:18:52 +00:00
|
|
|
/*
|
|
|
|
* The process p may have detached its own ->mm while exiting or through
|
|
|
|
* use_mm(), but one or more of its subthreads may still have a valid
|
|
|
|
* pointer. Return p, or any of its subthreads with a valid ->mm, with
|
|
|
|
* task_lock() held.
|
|
|
|
*/
|
2010-08-11 01:03:00 +00:00
|
|
|
struct task_struct *find_lock_task_mm(struct task_struct *p)
|
oom: introduce find_lock_task_mm() to fix !mm false positives
Almost all ->mm == NULL checks in oom_kill.c are wrong.
The current code assumes that the task without ->mm has already released
its memory and ignores the process. However this is not necessarily true
when this process is multithreaded, other live sub-threads can use this
->mm.
- Remove the "if (!p->mm)" check in select_bad_process(), it is
just wrong.
- Add the new helper, find_lock_task_mm(), which finds the live
thread which uses the memory and takes task_lock() to pin ->mm
- change oom_badness() to use this helper instead of just checking
->mm != NULL.
- As David pointed out, select_bad_process() must never choose the
task without ->mm, but no matter what oom_badness() returns the
task can be chosen if nothing else has been found yet.
Change oom_badness() to return int, change it to return -1 if
find_lock_task_mm() fails, and change select_bad_process() to
check points >= 0.
Note! This patch is not enough, we need more changes.
- oom_badness() was fixed, but oom_kill_task() still ignores
the task without ->mm
- oom_forkbomb_penalty() should use find_lock_task_mm() too,
and it also needs other changes to actually find the first
first-descendant children
This will be addressed later.
[kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()]
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: 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>
2010-08-10 00:18:45 +00:00
|
|
|
{
|
|
|
|
struct task_struct *t = p;
|
|
|
|
|
|
|
|
do {
|
|
|
|
task_lock(t);
|
|
|
|
if (likely(t->mm))
|
|
|
|
return t;
|
|
|
|
task_unlock(t);
|
|
|
|
} while_each_thread(p, t);
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2010-08-10 00:19:35 +00:00
|
|
|
/* return true if the task is not adequate as candidate victim task. */
|
2010-09-22 20:05:10 +00:00
|
|
|
static bool oom_unkillable_task(struct task_struct *p,
|
2012-01-13 01:18:32 +00:00
|
|
|
const struct mem_cgroup *memcg, const nodemask_t *nodemask)
|
2010-08-10 00:19:35 +00:00
|
|
|
{
|
|
|
|
if (is_global_init(p))
|
|
|
|
return true;
|
|
|
|
if (p->flags & PF_KTHREAD)
|
|
|
|
return true;
|
|
|
|
|
|
|
|
/* When mem_cgroup_out_of_memory() and p is not member of the group */
|
2012-01-13 01:18:32 +00:00
|
|
|
if (memcg && !task_in_mem_cgroup(p, memcg))
|
2010-08-10 00:19:35 +00:00
|
|
|
return true;
|
|
|
|
|
|
|
|
/* p may not have freeable memory in nodemask */
|
|
|
|
if (!has_intersects_mems_allowed(p, nodemask))
|
|
|
|
return true;
|
|
|
|
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2005-04-16 22:20:36 +00:00
|
|
|
/**
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
* oom_badness - heuristic function to determine which candidate task to kill
|
2005-04-16 22:20:36 +00:00
|
|
|
* @p: task struct of which task we should calculate
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
* @totalpages: total present RAM allowed for page allocation
|
2005-04-16 22:20:36 +00:00
|
|
|
*
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
* The heuristic for determining which task to kill is made to be as simple and
|
|
|
|
* predictable as possible. The goal is to return the highest value for the
|
|
|
|
* task consuming the most memory to avoid subsequent oom failures.
|
2005-04-16 22:20:36 +00:00
|
|
|
*/
|
2012-05-29 22:06:47 +00:00
|
|
|
unsigned long oom_badness(struct task_struct *p, struct mem_cgroup *memcg,
|
|
|
|
const nodemask_t *nodemask, unsigned long totalpages)
|
2005-04-16 22:20:36 +00:00
|
|
|
{
|
2012-06-08 20:21:26 +00:00
|
|
|
long points;
|
2012-06-20 19:52:58 +00:00
|
|
|
long adj;
|
oom: move oom_adj value from task_struct to signal_struct
Currently, OOM logic callflow is here.
__out_of_memory()
select_bad_process() for each task
badness() calculate badness of one task
oom_kill_process() search child
oom_kill_task() kill target task and mm shared tasks with it
example, process-A have two thread, thread-A and thread-B and it have very
fat memory and each thread have following oom_adj and oom_score.
thread-A: oom_adj = OOM_DISABLE, oom_score = 0
thread-B: oom_adj = 0, oom_score = very-high
Then, select_bad_process() select thread-B, but oom_kill_task() refuse
kill the task because thread-A have OOM_DISABLE. Thus __out_of_memory()
call select_bad_process() again. but select_bad_process() select the same
task. It mean kernel fall in livelock.
The fact is, select_bad_process() must select killable task. otherwise
OOM logic go into livelock.
And root cause is, oom_adj shouldn't be per-thread value. it should be
per-process value because OOM-killer kill a process, not thread. Thus
This patch moves oomkilladj (now more appropriately named oom_adj) from
struct task_struct to struct signal_struct. it naturally prevent
select_bad_process() choose wrong task.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Paul Menage <menage@google.com>
Cc: David Rientjes <rientjes@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 00:03:13 +00:00
|
|
|
|
2012-01-13 01:18:32 +00:00
|
|
|
if (oom_unkillable_task(p, memcg, nodemask))
|
2010-08-10 00:19:37 +00:00
|
|
|
return 0;
|
2005-04-16 22:20:36 +00:00
|
|
|
|
oom: introduce find_lock_task_mm() to fix !mm false positives
Almost all ->mm == NULL checks in oom_kill.c are wrong.
The current code assumes that the task without ->mm has already released
its memory and ignores the process. However this is not necessarily true
when this process is multithreaded, other live sub-threads can use this
->mm.
- Remove the "if (!p->mm)" check in select_bad_process(), it is
just wrong.
- Add the new helper, find_lock_task_mm(), which finds the live
thread which uses the memory and takes task_lock() to pin ->mm
- change oom_badness() to use this helper instead of just checking
->mm != NULL.
- As David pointed out, select_bad_process() must never choose the
task without ->mm, but no matter what oom_badness() returns the
task can be chosen if nothing else has been found yet.
Change oom_badness() to return int, change it to return -1 if
find_lock_task_mm() fails, and change select_bad_process() to
check points >= 0.
Note! This patch is not enough, we need more changes.
- oom_badness() was fixed, but oom_kill_task() still ignores
the task without ->mm
- oom_forkbomb_penalty() should use find_lock_task_mm() too,
and it also needs other changes to actually find the first
first-descendant children
This will be addressed later.
[kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()]
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: 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>
2010-08-10 00:18:45 +00:00
|
|
|
p = find_lock_task_mm(p);
|
|
|
|
if (!p)
|
2005-04-16 22:20:36 +00:00
|
|
|
return 0;
|
|
|
|
|
2012-06-20 19:52:58 +00:00
|
|
|
adj = p->signal->oom_score_adj;
|
|
|
|
if (adj == OOM_SCORE_ADJ_MIN) {
|
2011-11-15 22:36:07 +00:00
|
|
|
task_unlock(p);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-04-16 22:20:36 +00:00
|
|
|
/*
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
* The baseline for the badness score is the proportion of RAM that each
|
2011-04-27 22:26:50 +00:00
|
|
|
* task's rss, pagetable and swap space use.
|
2005-04-16 22:20:36 +00:00
|
|
|
*/
|
2012-05-29 22:06:47 +00:00
|
|
|
points = get_mm_rss(p->mm) + p->mm->nr_ptes +
|
|
|
|
get_mm_counter(p->mm, MM_SWAPENTS);
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
task_unlock(p);
|
2005-04-16 22:20:36 +00:00
|
|
|
|
|
|
|
/*
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
* Root processes get 3% bonus, just like the __vm_enough_memory()
|
|
|
|
* implementation used by LSMs.
|
2005-04-16 22:20:36 +00:00
|
|
|
*/
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
if (has_capability_noaudit(p, CAP_SYS_ADMIN))
|
2012-06-20 19:52:58 +00:00
|
|
|
adj -= 30;
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2012-06-20 19:52:58 +00:00
|
|
|
/* Normalize to oom_score_adj units */
|
|
|
|
adj *= totalpages / 1000;
|
|
|
|
points += adj;
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2010-09-22 20:04:52 +00:00
|
|
|
/*
|
2012-05-29 22:06:47 +00:00
|
|
|
* Never return 0 for an eligible task regardless of the root bonus and
|
|
|
|
* oom_score_adj (oom_score_adj can't be OOM_SCORE_ADJ_MIN here).
|
2010-09-22 20:04:52 +00:00
|
|
|
*/
|
2012-06-08 20:21:26 +00:00
|
|
|
return points > 0 ? points : 1;
|
2005-04-16 22:20:36 +00:00
|
|
|
}
|
|
|
|
|
2006-02-21 02:27:52 +00:00
|
|
|
/*
|
|
|
|
* Determine the type of allocation constraint.
|
|
|
|
*/
|
|
|
|
#ifdef CONFIG_NUMA
|
2009-12-16 00:45:33 +00:00
|
|
|
static enum oom_constraint constrained_alloc(struct zonelist *zonelist,
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
gfp_t gfp_mask, nodemask_t *nodemask,
|
|
|
|
unsigned long *totalpages)
|
2009-12-16 00:45:33 +00:00
|
|
|
{
|
2008-04-28 09:12:16 +00:00
|
|
|
struct zone *zone;
|
2008-04-28 09:12:17 +00:00
|
|
|
struct zoneref *z;
|
2008-04-28 09:12:16 +00:00
|
|
|
enum zone_type high_zoneidx = gfp_zone(gfp_mask);
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
bool cpuset_limited = false;
|
|
|
|
int nid;
|
2006-02-21 02:27:52 +00:00
|
|
|
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
/* Default to all available memory */
|
|
|
|
*totalpages = totalram_pages + total_swap_pages;
|
|
|
|
|
|
|
|
if (!zonelist)
|
|
|
|
return CONSTRAINT_NONE;
|
2009-12-16 00:45:33 +00:00
|
|
|
/*
|
|
|
|
* Reach here only when __GFP_NOFAIL is used. So, we should avoid
|
|
|
|
* to kill current.We have to random task kill in this case.
|
|
|
|
* Hopefully, CONSTRAINT_THISNODE...but no way to handle it, now.
|
|
|
|
*/
|
|
|
|
if (gfp_mask & __GFP_THISNODE)
|
|
|
|
return CONSTRAINT_NONE;
|
2006-02-21 02:27:52 +00:00
|
|
|
|
2009-12-16 00:45:33 +00:00
|
|
|
/*
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
* This is not a __GFP_THISNODE allocation, so a truncated nodemask in
|
|
|
|
* the page allocator means a mempolicy is in effect. Cpuset policy
|
|
|
|
* is enforced in get_page_from_freelist().
|
2009-12-16 00:45:33 +00:00
|
|
|
*/
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
if (nodemask && !nodes_subset(node_states[N_HIGH_MEMORY], *nodemask)) {
|
|
|
|
*totalpages = total_swap_pages;
|
|
|
|
for_each_node_mask(nid, *nodemask)
|
|
|
|
*totalpages += node_spanned_pages(nid);
|
2006-02-21 02:27:52 +00:00
|
|
|
return CONSTRAINT_MEMORY_POLICY;
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
}
|
2009-12-16 00:45:33 +00:00
|
|
|
|
|
|
|
/* Check this allocation failure is caused by cpuset's wall function */
|
|
|
|
for_each_zone_zonelist_nodemask(zone, z, zonelist,
|
|
|
|
high_zoneidx, nodemask)
|
|
|
|
if (!cpuset_zone_allowed_softwall(zone, gfp_mask))
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
cpuset_limited = true;
|
2006-02-21 02:27:52 +00:00
|
|
|
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
if (cpuset_limited) {
|
|
|
|
*totalpages = total_swap_pages;
|
|
|
|
for_each_node_mask(nid, cpuset_current_mems_allowed)
|
|
|
|
*totalpages += node_spanned_pages(nid);
|
|
|
|
return CONSTRAINT_CPUSET;
|
|
|
|
}
|
2006-02-21 02:27:52 +00:00
|
|
|
return CONSTRAINT_NONE;
|
|
|
|
}
|
2009-12-16 00:45:33 +00:00
|
|
|
#else
|
|
|
|
static enum oom_constraint constrained_alloc(struct zonelist *zonelist,
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
gfp_t gfp_mask, nodemask_t *nodemask,
|
|
|
|
unsigned long *totalpages)
|
2009-12-16 00:45:33 +00:00
|
|
|
{
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
*totalpages = totalram_pages + total_swap_pages;
|
2009-12-16 00:45:33 +00:00
|
|
|
return CONSTRAINT_NONE;
|
|
|
|
}
|
|
|
|
#endif
|
2006-02-21 02:27:52 +00:00
|
|
|
|
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: Sha Zhengju <handai.szj@taobao.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 23:43:44 +00:00
|
|
|
enum oom_scan_t oom_scan_process_thread(struct task_struct *task,
|
|
|
|
unsigned long totalpages, const nodemask_t *nodemask,
|
|
|
|
bool force_kill)
|
2012-07-31 23:43:40 +00:00
|
|
|
{
|
|
|
|
if (task->exit_state)
|
|
|
|
return OOM_SCAN_CONTINUE;
|
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: Sha Zhengju <handai.szj@taobao.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 23:43:44 +00:00
|
|
|
if (oom_unkillable_task(task, NULL, nodemask))
|
2012-07-31 23:43:40 +00:00
|
|
|
return OOM_SCAN_CONTINUE;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This task already has access to memory reserves and is being killed.
|
|
|
|
* Don't allow any other task to have access to the reserves.
|
|
|
|
*/
|
|
|
|
if (test_tsk_thread_flag(task, TIF_MEMDIE)) {
|
|
|
|
if (unlikely(frozen(task)))
|
|
|
|
__thaw_task(task);
|
|
|
|
if (!force_kill)
|
|
|
|
return OOM_SCAN_ABORT;
|
|
|
|
}
|
|
|
|
if (!task->mm)
|
|
|
|
return OOM_SCAN_CONTINUE;
|
|
|
|
|
|
|
|
if (task->flags & PF_EXITING) {
|
|
|
|
/*
|
|
|
|
* If task is current and is in the process of releasing memory,
|
|
|
|
* allow the "kill" to set TIF_MEMDIE, which will allow it to
|
|
|
|
* access memory reserves. Otherwise, it may stall forever.
|
|
|
|
*
|
|
|
|
* The iteration isn't broken here, however, in case other
|
|
|
|
* threads are found to have already been oom killed.
|
|
|
|
*/
|
|
|
|
if (task == current)
|
|
|
|
return OOM_SCAN_SELECT;
|
|
|
|
else if (!force_kill) {
|
|
|
|
/*
|
|
|
|
* If this task is not being ptraced on exit, then wait
|
|
|
|
* for it to finish before killing some other task
|
|
|
|
* unnecessarily.
|
|
|
|
*/
|
|
|
|
if (!(task->group_leader->ptrace & PT_TRACE_EXIT))
|
|
|
|
return OOM_SCAN_ABORT;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return OOM_SCAN_OK;
|
|
|
|
}
|
|
|
|
|
2005-04-16 22:20:36 +00:00
|
|
|
/*
|
|
|
|
* Simple selection loop. We chose the process with the highest
|
2012-07-31 23:43:45 +00:00
|
|
|
* number of 'points'.
|
2005-04-16 22:20:36 +00:00
|
|
|
*
|
|
|
|
* (not docbooked, we don't want this one cluttering up the manual)
|
|
|
|
*/
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
static struct task_struct *select_bad_process(unsigned int *ppoints,
|
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: Sha Zhengju <handai.szj@taobao.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 23:43:44 +00:00
|
|
|
unsigned long totalpages, const nodemask_t *nodemask,
|
|
|
|
bool force_kill)
|
2005-04-16 22:20:36 +00:00
|
|
|
{
|
oom: prevent unnecessary oom kills or kernel panics
This patch prevents unnecessary oom kills or kernel panics by reverting
two commits:
495789a5 (oom: make oom_score to per-process value)
cef1d352 (oom: multi threaded process coredump don't make deadlock)
First, 495789a5 (oom: make oom_score to per-process value) ignores the
fact that all threads in a thread group do not necessarily exit at the
same time.
It is imperative that select_bad_process() detect threads that are in the
exit path, specifically those with PF_EXITING set, to prevent needlessly
killing additional tasks. If a process is oom killed and the thread group
leader exits, select_bad_process() cannot detect the other threads that
are PF_EXITING by iterating over only processes. Thus, it currently
chooses another task unnecessarily for oom kill or panics the machine when
nothing else is eligible.
By iterating over threads instead, it is possible to detect threads that
are exiting and nominate them for oom kill so they get access to memory
reserves.
Second, cef1d352 (oom: multi threaded process coredump don't make
deadlock) erroneously avoids making the oom killer a no-op when an
eligible thread other than current isfound to be exiting. We want to
detect this situation so that we may allow that exiting thread time to
exit and free its memory; if it is able to exit on its own, that should
free memory so current is no loner oom. If it is not able to exit on its
own, the oom killer will nominate it for oom kill which, in this case,
only means it will get access to memory reserves.
Without this change, it is easy for the oom killer to unnecessarily target
tasks when all threads of a victim don't exit before the thread group
leader or, in the worst case, panic the machine.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Andrey Vagin <avagin@openvz.org>
Cc: <stable@kernel.org> [2.6.38.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-22 23:30:09 +00:00
|
|
|
struct task_struct *g, *p;
|
2005-04-16 22:20:36 +00:00
|
|
|
struct task_struct *chosen = NULL;
|
2012-05-29 22:06:47 +00:00
|
|
|
unsigned long chosen_points = 0;
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2012-07-31 23:43:45 +00:00
|
|
|
rcu_read_lock();
|
oom: prevent unnecessary oom kills or kernel panics
This patch prevents unnecessary oom kills or kernel panics by reverting
two commits:
495789a5 (oom: make oom_score to per-process value)
cef1d352 (oom: multi threaded process coredump don't make deadlock)
First, 495789a5 (oom: make oom_score to per-process value) ignores the
fact that all threads in a thread group do not necessarily exit at the
same time.
It is imperative that select_bad_process() detect threads that are in the
exit path, specifically those with PF_EXITING set, to prevent needlessly
killing additional tasks. If a process is oom killed and the thread group
leader exits, select_bad_process() cannot detect the other threads that
are PF_EXITING by iterating over only processes. Thus, it currently
chooses another task unnecessarily for oom kill or panics the machine when
nothing else is eligible.
By iterating over threads instead, it is possible to detect threads that
are exiting and nominate them for oom kill so they get access to memory
reserves.
Second, cef1d352 (oom: multi threaded process coredump don't make
deadlock) erroneously avoids making the oom killer a no-op when an
eligible thread other than current isfound to be exiting. We want to
detect this situation so that we may allow that exiting thread time to
exit and free its memory; if it is able to exit on its own, that should
free memory so current is no loner oom. If it is not able to exit on its
own, the oom killer will nominate it for oom kill which, in this case,
only means it will get access to memory reserves.
Without this change, it is easy for the oom killer to unnecessarily target
tasks when all threads of a victim don't exit before the thread group
leader or, in the worst case, panic the machine.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Andrey Vagin <avagin@openvz.org>
Cc: <stable@kernel.org> [2.6.38.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-22 23:30:09 +00:00
|
|
|
do_each_thread(g, p) {
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
unsigned int points;
|
[PATCH] cpusets: oom_kill tweaks
This patch series extends the use of the cpuset attribute 'mem_exclusive'
to support cpuset configurations that:
1) allow GFP_KERNEL allocations to come from a potentially larger
set of memory nodes than GFP_USER allocations, and
2) can constrain the oom killer to tasks running in cpusets in
a specified subtree of the cpuset hierarchy.
Here's an example usage scenario. For a few hours or more, a large NUMA
system at a University is to be divided in two halves, with a bunch of student
jobs running in half the system under some form of batch manager, and with a
big research project running in the other half. Each of the student jobs is
placed in a small cpuset, but should share the classic Unix time share
facilities, such as buffered pages of files in /bin and /usr/lib. The big
research project wants no interference whatsoever from the student jobs, and
has highly tuned, unusual memory and i/o patterns that intend to make full use
of all the main memory on the nodes available to it.
In this example, we have two big sibling cpusets, one of which is further
divided into a more dynamic set of child cpusets.
We want kernel memory allocations constrained by the two big cpusets, and user
allocations constrained by the smaller child cpusets where present. And we
require that the oom killer not operate across the two halves of this system,
or else the first time a student job runs amuck, the big research project will
likely be first inline to get shot.
Tweaking /proc/<pid>/oom_adj is not ideal -- if the big research project
really does run amuck allocating memory, it should be shot, not some other
task outside the research projects mem_exclusive cpuset.
I propose to extend the use of the 'mem_exclusive' flag of cpusets to manage
such scenarios. Let memory allocations for user space (GFP_USER) be
constrained by a tasks current cpuset, but memory allocations for kernel space
(GFP_KERNEL) by constrained by the nearest mem_exclusive ancestor of the
current cpuset, even though kernel space allocations will still _prefer_ to
remain within the current tasks cpuset, if memory is easily available.
Let the oom killer be constrained to consider only tasks that are in
overlapping mem_exclusive cpusets (it won't help much to kill a task that
normally cannot allocate memory on any of the same nodes as the ones on which
the current task can allocate.)
The current constraints imposed on setting mem_exclusive are unchanged. A
cpuset may only be mem_exclusive if its parent is also mem_exclusive, and a
mem_exclusive cpuset may not overlap any of its siblings memory nodes.
This patch was presented on linux-mm in early July 2005, though did not
generate much feedback at that time. It has been built for a variety of
arch's using cross tools, and built, booted and tested for function on SN2
(ia64).
There are 4 patches in this set:
1) Some minor cleanup, and some improvements to the code layout
of one routine to make subsequent patches cleaner.
2) Add another GFP flag - __GFP_HARDWALL. It marks memory
requests for USER space, which are tightly confined by the
current tasks cpuset.
3) Now memory requests (such as KERNEL) that not marked HARDWALL can
if short on memory, look in the potentially larger pool of memory
defined by the nearest mem_exclusive ancestor cpuset of the current
tasks cpuset.
4) Finally, modify the oom killer to skip any task whose mem_exclusive
cpuset doesn't overlap ours.
Patch (1), the one time I looked on an SN2 (ia64) build, actually saved 32
bytes of kernel text space. Patch (2) has no affect on the size of kernel
text space (it just adds a preprocessor flag). Patches (3) and (4) added
about 600 bytes each of kernel text space, mostly in kernel/cpuset.c, which
matters only if CONFIG_CPUSET is enabled.
This patch:
This patch applies a few comment and code cleanups to mm/oom_kill.c prior to
applying a few small patches to improve cpuset management of memory placement.
The comment changed in oom_kill.c was seriously misleading. The code layout
change in select_bad_process() makes room for adding another condition on
which a process can be spared the oom killer (see the subsequent
cpuset_nodes_overlap patch for this addition).
Also a couple typos and spellos that bugged me, while I was here.
This patch should have no material affect.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-09-06 22:18:09 +00:00
|
|
|
|
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: Sha Zhengju <handai.szj@taobao.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 23:43:44 +00:00
|
|
|
switch (oom_scan_process_thread(p, totalpages, nodemask,
|
2012-07-31 23:43:40 +00:00
|
|
|
force_kill)) {
|
|
|
|
case OOM_SCAN_SELECT:
|
|
|
|
chosen = p;
|
|
|
|
chosen_points = ULONG_MAX;
|
|
|
|
/* fall through */
|
|
|
|
case OOM_SCAN_CONTINUE:
|
2011-07-30 14:35:02 +00:00
|
|
|
continue;
|
2012-07-31 23:43:40 +00:00
|
|
|
case OOM_SCAN_ABORT:
|
2012-07-31 23:43:45 +00:00
|
|
|
rcu_read_unlock();
|
2012-07-31 23:43:40 +00:00
|
|
|
return ERR_PTR(-1UL);
|
|
|
|
case OOM_SCAN_OK:
|
|
|
|
break;
|
|
|
|
};
|
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: Sha Zhengju <handai.szj@taobao.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 23:43:44 +00:00
|
|
|
points = oom_badness(p, NULL, nodemask, totalpages);
|
2012-05-29 22:06:47 +00:00
|
|
|
if (points > chosen_points) {
|
[PATCH] cpusets: oom_kill tweaks
This patch series extends the use of the cpuset attribute 'mem_exclusive'
to support cpuset configurations that:
1) allow GFP_KERNEL allocations to come from a potentially larger
set of memory nodes than GFP_USER allocations, and
2) can constrain the oom killer to tasks running in cpusets in
a specified subtree of the cpuset hierarchy.
Here's an example usage scenario. For a few hours or more, a large NUMA
system at a University is to be divided in two halves, with a bunch of student
jobs running in half the system under some form of batch manager, and with a
big research project running in the other half. Each of the student jobs is
placed in a small cpuset, but should share the classic Unix time share
facilities, such as buffered pages of files in /bin and /usr/lib. The big
research project wants no interference whatsoever from the student jobs, and
has highly tuned, unusual memory and i/o patterns that intend to make full use
of all the main memory on the nodes available to it.
In this example, we have two big sibling cpusets, one of which is further
divided into a more dynamic set of child cpusets.
We want kernel memory allocations constrained by the two big cpusets, and user
allocations constrained by the smaller child cpusets where present. And we
require that the oom killer not operate across the two halves of this system,
or else the first time a student job runs amuck, the big research project will
likely be first inline to get shot.
Tweaking /proc/<pid>/oom_adj is not ideal -- if the big research project
really does run amuck allocating memory, it should be shot, not some other
task outside the research projects mem_exclusive cpuset.
I propose to extend the use of the 'mem_exclusive' flag of cpusets to manage
such scenarios. Let memory allocations for user space (GFP_USER) be
constrained by a tasks current cpuset, but memory allocations for kernel space
(GFP_KERNEL) by constrained by the nearest mem_exclusive ancestor of the
current cpuset, even though kernel space allocations will still _prefer_ to
remain within the current tasks cpuset, if memory is easily available.
Let the oom killer be constrained to consider only tasks that are in
overlapping mem_exclusive cpusets (it won't help much to kill a task that
normally cannot allocate memory on any of the same nodes as the ones on which
the current task can allocate.)
The current constraints imposed on setting mem_exclusive are unchanged. A
cpuset may only be mem_exclusive if its parent is also mem_exclusive, and a
mem_exclusive cpuset may not overlap any of its siblings memory nodes.
This patch was presented on linux-mm in early July 2005, though did not
generate much feedback at that time. It has been built for a variety of
arch's using cross tools, and built, booted and tested for function on SN2
(ia64).
There are 4 patches in this set:
1) Some minor cleanup, and some improvements to the code layout
of one routine to make subsequent patches cleaner.
2) Add another GFP flag - __GFP_HARDWALL. It marks memory
requests for USER space, which are tightly confined by the
current tasks cpuset.
3) Now memory requests (such as KERNEL) that not marked HARDWALL can
if short on memory, look in the potentially larger pool of memory
defined by the nearest mem_exclusive ancestor cpuset of the current
tasks cpuset.
4) Finally, modify the oom killer to skip any task whose mem_exclusive
cpuset doesn't overlap ours.
Patch (1), the one time I looked on an SN2 (ia64) build, actually saved 32
bytes of kernel text space. Patch (2) has no affect on the size of kernel
text space (it just adds a preprocessor flag). Patches (3) and (4) added
about 600 bytes each of kernel text space, mostly in kernel/cpuset.c, which
matters only if CONFIG_CPUSET is enabled.
This patch:
This patch applies a few comment and code cleanups to mm/oom_kill.c prior to
applying a few small patches to improve cpuset management of memory placement.
The comment changed in oom_kill.c was seriously misleading. The code layout
change in select_bad_process() makes room for adding another condition on
which a process can be spared the oom killer (see the subsequent
cpuset_nodes_overlap patch for this addition).
Also a couple typos and spellos that bugged me, while I was here.
This patch should have no material affect.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-09-06 22:18:09 +00:00
|
|
|
chosen = p;
|
2012-05-29 22:06:47 +00:00
|
|
|
chosen_points = points;
|
2005-04-16 22:20:36 +00:00
|
|
|
}
|
oom: prevent unnecessary oom kills or kernel panics
This patch prevents unnecessary oom kills or kernel panics by reverting
two commits:
495789a5 (oom: make oom_score to per-process value)
cef1d352 (oom: multi threaded process coredump don't make deadlock)
First, 495789a5 (oom: make oom_score to per-process value) ignores the
fact that all threads in a thread group do not necessarily exit at the
same time.
It is imperative that select_bad_process() detect threads that are in the
exit path, specifically those with PF_EXITING set, to prevent needlessly
killing additional tasks. If a process is oom killed and the thread group
leader exits, select_bad_process() cannot detect the other threads that
are PF_EXITING by iterating over only processes. Thus, it currently
chooses another task unnecessarily for oom kill or panics the machine when
nothing else is eligible.
By iterating over threads instead, it is possible to detect threads that
are exiting and nominate them for oom kill so they get access to memory
reserves.
Second, cef1d352 (oom: multi threaded process coredump don't make
deadlock) erroneously avoids making the oom killer a no-op when an
eligible thread other than current isfound to be exiting. We want to
detect this situation so that we may allow that exiting thread time to
exit and free its memory; if it is able to exit on its own, that should
free memory so current is no loner oom. If it is not able to exit on its
own, the oom killer will nominate it for oom kill which, in this case,
only means it will get access to memory reserves.
Without this change, it is easy for the oom killer to unnecessarily target
tasks when all threads of a victim don't exit before the thread group
leader or, in the worst case, panic the machine.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Andrey Vagin <avagin@openvz.org>
Cc: <stable@kernel.org> [2.6.38.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-22 23:30:09 +00:00
|
|
|
} while_each_thread(g, p);
|
2012-07-31 23:43:45 +00:00
|
|
|
if (chosen)
|
|
|
|
get_task_struct(chosen);
|
|
|
|
rcu_read_unlock();
|
2006-09-29 09:01:12 +00:00
|
|
|
|
2012-05-29 22:06:47 +00:00
|
|
|
*ppoints = chosen_points * 1000 / totalpages;
|
2005-04-16 22:20:36 +00:00
|
|
|
return chosen;
|
|
|
|
}
|
|
|
|
|
oom: add sysctl to enable task memory dump
Adds a new sysctl, 'oom_dump_tasks', that enables the kernel to produce a
dump of all system tasks (excluding kernel threads) when performing an
OOM-killing. Information includes pid, uid, tgid, vm size, rss, cpu,
oom_adj score, and name.
This is helpful for determining why there was an OOM condition and which
rogue task caused it.
It is configurable so that large systems, such as those with several
thousand tasks, do not incur a performance penalty associated with dumping
data they may not desire.
If an OOM was triggered as a result of a memory controller, the tasklist
shall be filtered to exclude tasks that are not a member of the same
cgroup.
Cc: Andrea Arcangeli <andrea@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-07 08:14:07 +00:00
|
|
|
/**
|
2008-03-20 00:00:42 +00:00
|
|
|
* dump_tasks - dump current memory state of all system tasks
|
2012-06-20 19:53:01 +00:00
|
|
|
* @memcg: current's memory controller, if constrained
|
2010-09-22 20:05:10 +00:00
|
|
|
* @nodemask: nodemask passed to page allocator for mempolicy ooms
|
2008-03-20 00:00:42 +00:00
|
|
|
*
|
2010-09-22 20:05:10 +00:00
|
|
|
* Dumps the current memory state of all eligible tasks. Tasks not in the same
|
|
|
|
* memcg, not in the same cpuset, or bound to a disjoint set of mempolicy nodes
|
|
|
|
* are not shown.
|
2012-07-31 23:42:56 +00:00
|
|
|
* State information includes task's pid, uid, tgid, vm size, rss, nr_ptes,
|
|
|
|
* swapents, oom_score_adj value, and name.
|
oom: add sysctl to enable task memory dump
Adds a new sysctl, 'oom_dump_tasks', that enables the kernel to produce a
dump of all system tasks (excluding kernel threads) when performing an
OOM-killing. Information includes pid, uid, tgid, vm size, rss, cpu,
oom_adj score, and name.
This is helpful for determining why there was an OOM condition and which
rogue task caused it.
It is configurable so that large systems, such as those with several
thousand tasks, do not incur a performance penalty associated with dumping
data they may not desire.
If an OOM was triggered as a result of a memory controller, the tasklist
shall be filtered to exclude tasks that are not a member of the same
cgroup.
Cc: Andrea Arcangeli <andrea@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-07 08:14:07 +00:00
|
|
|
*/
|
2012-01-13 01:18:32 +00:00
|
|
|
static void dump_tasks(const struct mem_cgroup *memcg, const nodemask_t *nodemask)
|
oom: add sysctl to enable task memory dump
Adds a new sysctl, 'oom_dump_tasks', that enables the kernel to produce a
dump of all system tasks (excluding kernel threads) when performing an
OOM-killing. Information includes pid, uid, tgid, vm size, rss, cpu,
oom_adj score, and name.
This is helpful for determining why there was an OOM condition and which
rogue task caused it.
It is configurable so that large systems, such as those with several
thousand tasks, do not incur a performance penalty associated with dumping
data they may not desire.
If an OOM was triggered as a result of a memory controller, the tasklist
shall be filtered to exclude tasks that are not a member of the same
cgroup.
Cc: Andrea Arcangeli <andrea@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-07 08:14:07 +00:00
|
|
|
{
|
2010-08-10 00:18:46 +00:00
|
|
|
struct task_struct *p;
|
|
|
|
struct task_struct *task;
|
oom: add sysctl to enable task memory dump
Adds a new sysctl, 'oom_dump_tasks', that enables the kernel to produce a
dump of all system tasks (excluding kernel threads) when performing an
OOM-killing. Information includes pid, uid, tgid, vm size, rss, cpu,
oom_adj score, and name.
This is helpful for determining why there was an OOM condition and which
rogue task caused it.
It is configurable so that large systems, such as those with several
thousand tasks, do not incur a performance penalty associated with dumping
data they may not desire.
If an OOM was triggered as a result of a memory controller, the tasklist
shall be filtered to exclude tasks that are not a member of the same
cgroup.
Cc: Andrea Arcangeli <andrea@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-07 08:14:07 +00:00
|
|
|
|
2012-07-31 23:42:56 +00:00
|
|
|
pr_info("[ pid ] uid tgid total_vm rss nr_ptes swapents oom_score_adj name\n");
|
2012-07-31 23:43:45 +00:00
|
|
|
rcu_read_lock();
|
2010-08-10 00:18:46 +00:00
|
|
|
for_each_process(p) {
|
2012-01-13 01:18:32 +00:00
|
|
|
if (oom_unkillable_task(p, memcg, nodemask))
|
2008-11-06 20:53:29 +00:00
|
|
|
continue;
|
oom: add sysctl to enable task memory dump
Adds a new sysctl, 'oom_dump_tasks', that enables the kernel to produce a
dump of all system tasks (excluding kernel threads) when performing an
OOM-killing. Information includes pid, uid, tgid, vm size, rss, cpu,
oom_adj score, and name.
This is helpful for determining why there was an OOM condition and which
rogue task caused it.
It is configurable so that large systems, such as those with several
thousand tasks, do not incur a performance penalty associated with dumping
data they may not desire.
If an OOM was triggered as a result of a memory controller, the tasklist
shall be filtered to exclude tasks that are not a member of the same
cgroup.
Cc: Andrea Arcangeli <andrea@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-07 08:14:07 +00:00
|
|
|
|
2010-08-10 00:18:46 +00:00
|
|
|
task = find_lock_task_mm(p);
|
|
|
|
if (!task) {
|
2009-05-28 21:34:19 +00:00
|
|
|
/*
|
2010-08-10 00:18:46 +00:00
|
|
|
* This is a kthread or all of p's threads have already
|
|
|
|
* detached their mm's. There's no need to report
|
2010-08-10 00:18:46 +00:00
|
|
|
* them; they can't be oom killed anyway.
|
2009-05-28 21:34:19 +00:00
|
|
|
*/
|
|
|
|
continue;
|
|
|
|
}
|
2010-08-10 00:18:46 +00:00
|
|
|
|
2012-07-31 23:42:56 +00:00
|
|
|
pr_info("[%5d] %5d %5d %8lu %8lu %7lu %8lu %5d %s\n",
|
2012-02-08 15:00:08 +00:00
|
|
|
task->pid, from_kuid(&init_user_ns, task_uid(task)),
|
|
|
|
task->tgid, task->mm->total_vm, get_mm_rss(task->mm),
|
2012-07-31 23:42:56 +00:00
|
|
|
task->mm->nr_ptes,
|
|
|
|
get_mm_counter(task->mm, MM_SWAPENTS),
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
task->signal->oom_score_adj, task->comm);
|
2010-08-10 00:18:46 +00:00
|
|
|
task_unlock(task);
|
|
|
|
}
|
2012-07-31 23:43:45 +00:00
|
|
|
rcu_read_unlock();
|
oom: add sysctl to enable task memory dump
Adds a new sysctl, 'oom_dump_tasks', that enables the kernel to produce a
dump of all system tasks (excluding kernel threads) when performing an
OOM-killing. Information includes pid, uid, tgid, vm size, rss, cpu,
oom_adj score, and name.
This is helpful for determining why there was an OOM condition and which
rogue task caused it.
It is configurable so that large systems, such as those with several
thousand tasks, do not incur a performance penalty associated with dumping
data they may not desire.
If an OOM was triggered as a result of a memory controller, the tasklist
shall be filtered to exclude tasks that are not a member of the same
cgroup.
Cc: Andrea Arcangeli <andrea@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-07 08:14:07 +00:00
|
|
|
}
|
|
|
|
|
2009-12-16 00:47:12 +00:00
|
|
|
static void dump_header(struct task_struct *p, gfp_t gfp_mask, int order,
|
2012-01-13 01:18:32 +00:00
|
|
|
struct mem_cgroup *memcg, const nodemask_t *nodemask)
|
2009-12-15 01:57:47 +00:00
|
|
|
{
|
2010-08-10 00:18:51 +00:00
|
|
|
task_lock(current);
|
2009-12-15 01:57:47 +00:00
|
|
|
pr_warning("%s invoked oom-killer: gfp_mask=0x%x, order=%d, "
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
"oom_adj=%d, oom_score_adj=%d\n",
|
|
|
|
current->comm, gfp_mask, order, current->signal->oom_adj,
|
|
|
|
current->signal->oom_score_adj);
|
2009-12-15 01:57:47 +00:00
|
|
|
cpuset_print_task_mems_allowed(current);
|
|
|
|
task_unlock(current);
|
|
|
|
dump_stack();
|
2012-01-13 01:18:32 +00:00
|
|
|
mem_cgroup_print_oom_info(memcg, p);
|
2011-03-24 22:18:15 +00:00
|
|
|
show_mem(SHOW_MEM_FILTER_NODES);
|
2009-12-15 01:57:47 +00:00
|
|
|
if (sysctl_oom_dump_tasks)
|
2012-01-13 01:18:32 +00:00
|
|
|
dump_tasks(memcg, nodemask);
|
2009-12-15 01:57:47 +00:00
|
|
|
}
|
|
|
|
|
2009-12-16 00:45:32 +00:00
|
|
|
#define K(x) ((x) << (PAGE_SHIFT-10))
|
2012-07-31 23:43:45 +00:00
|
|
|
/*
|
|
|
|
* Must be called while holding a reference to p, which will be released upon
|
|
|
|
* returning.
|
|
|
|
*/
|
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: Sha Zhengju <handai.szj@taobao.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 23:43:44 +00:00
|
|
|
void oom_kill_process(struct task_struct *p, gfp_t gfp_mask, int order,
|
|
|
|
unsigned int points, unsigned long totalpages,
|
|
|
|
struct mem_cgroup *memcg, nodemask_t *nodemask,
|
|
|
|
const char *message)
|
2005-04-16 22:20:36 +00:00
|
|
|
{
|
2011-03-14 22:17:07 +00:00
|
|
|
struct task_struct *victim = p;
|
2010-08-10 00:18:51 +00:00
|
|
|
struct task_struct *child;
|
2011-03-14 22:17:07 +00:00
|
|
|
struct task_struct *t = p;
|
2012-03-21 23:33:46 +00:00
|
|
|
struct mm_struct *mm;
|
2011-03-14 22:17:07 +00:00
|
|
|
unsigned int victim_points = 0;
|
2012-03-21 23:33:47 +00:00
|
|
|
static DEFINE_RATELIMIT_STATE(oom_rs, DEFAULT_RATELIMIT_INTERVAL,
|
|
|
|
DEFAULT_RATELIMIT_BURST);
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2006-09-26 06:31:29 +00:00
|
|
|
/*
|
|
|
|
* If the task is already exiting, don't alarm the sysadmin or kill
|
|
|
|
* its children or threads, just set TIF_MEMDIE so it can die quickly
|
|
|
|
*/
|
2009-08-18 21:11:10 +00:00
|
|
|
if (p->flags & PF_EXITING) {
|
2010-08-10 00:18:49 +00:00
|
|
|
set_tsk_thread_flag(p, TIF_MEMDIE);
|
2012-07-31 23:43:45 +00:00
|
|
|
put_task_struct(p);
|
2012-03-21 23:33:46 +00:00
|
|
|
return;
|
2006-09-26 06:31:29 +00:00
|
|
|
}
|
|
|
|
|
2012-03-21 23:33:47 +00:00
|
|
|
if (__ratelimit(&oom_rs))
|
2012-03-21 23:33:47 +00:00
|
|
|
dump_header(p, gfp_mask, order, memcg, nodemask);
|
|
|
|
|
2010-08-10 00:18:51 +00:00
|
|
|
task_lock(p);
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
pr_err("%s: Kill process %d (%s) score %d or sacrifice child\n",
|
2010-08-10 00:18:51 +00:00
|
|
|
message, task_pid_nr(p), p->comm, points);
|
|
|
|
task_unlock(p);
|
2006-12-07 04:31:51 +00:00
|
|
|
|
2010-08-10 00:18:51 +00:00
|
|
|
/*
|
|
|
|
* If any of p's children has a different mm and is eligible for kill,
|
2011-07-26 00:12:17 +00:00
|
|
|
* the one with the highest oom_badness() score is sacrificed for its
|
2010-08-10 00:18:51 +00:00
|
|
|
* parent. This attempts to lose the minimal amount of work done while
|
|
|
|
* still freeing memory.
|
|
|
|
*/
|
2012-07-31 23:43:45 +00:00
|
|
|
read_lock(&tasklist_lock);
|
oom: introduce find_lock_task_mm() to fix !mm false positives
Almost all ->mm == NULL checks in oom_kill.c are wrong.
The current code assumes that the task without ->mm has already released
its memory and ignores the process. However this is not necessarily true
when this process is multithreaded, other live sub-threads can use this
->mm.
- Remove the "if (!p->mm)" check in select_bad_process(), it is
just wrong.
- Add the new helper, find_lock_task_mm(), which finds the live
thread which uses the memory and takes task_lock() to pin ->mm
- change oom_badness() to use this helper instead of just checking
->mm != NULL.
- As David pointed out, select_bad_process() must never choose the
task without ->mm, but no matter what oom_badness() returns the
task can be chosen if nothing else has been found yet.
Change oom_badness() to return int, change it to return -1 if
find_lock_task_mm() fails, and change select_bad_process() to
check points >= 0.
Note! This patch is not enough, we need more changes.
- oom_badness() was fixed, but oom_kill_task() still ignores
the task without ->mm
- oom_forkbomb_penalty() should use find_lock_task_mm() too,
and it also needs other changes to actually find the first
first-descendant children
This will be addressed later.
[kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()]
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: 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>
2010-08-10 00:18:45 +00:00
|
|
|
do {
|
2010-08-10 00:18:51 +00:00
|
|
|
list_for_each_entry(child, &t->children, sibling) {
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
unsigned int child_points;
|
2010-08-10 00:18:51 +00:00
|
|
|
|
2011-03-22 23:30:12 +00:00
|
|
|
if (child->mm == p->mm)
|
|
|
|
continue;
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
/*
|
|
|
|
* oom_badness() returns 0 if the thread is unkillable
|
|
|
|
*/
|
2012-01-13 01:18:32 +00:00
|
|
|
child_points = oom_badness(child, memcg, nodemask,
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
totalpages);
|
2010-08-10 00:18:51 +00:00
|
|
|
if (child_points > victim_points) {
|
2012-07-31 23:43:45 +00:00
|
|
|
put_task_struct(victim);
|
2010-08-10 00:18:51 +00:00
|
|
|
victim = child;
|
|
|
|
victim_points = child_points;
|
2012-07-31 23:43:45 +00:00
|
|
|
get_task_struct(victim);
|
2010-08-10 00:18:51 +00:00
|
|
|
}
|
oom: introduce find_lock_task_mm() to fix !mm false positives
Almost all ->mm == NULL checks in oom_kill.c are wrong.
The current code assumes that the task without ->mm has already released
its memory and ignores the process. However this is not necessarily true
when this process is multithreaded, other live sub-threads can use this
->mm.
- Remove the "if (!p->mm)" check in select_bad_process(), it is
just wrong.
- Add the new helper, find_lock_task_mm(), which finds the live
thread which uses the memory and takes task_lock() to pin ->mm
- change oom_badness() to use this helper instead of just checking
->mm != NULL.
- As David pointed out, select_bad_process() must never choose the
task without ->mm, but no matter what oom_badness() returns the
task can be chosen if nothing else has been found yet.
Change oom_badness() to return int, change it to return -1 if
find_lock_task_mm() fails, and change select_bad_process() to
check points >= 0.
Note! This patch is not enough, we need more changes.
- oom_badness() was fixed, but oom_kill_task() still ignores
the task without ->mm
- oom_forkbomb_penalty() should use find_lock_task_mm() too,
and it also needs other changes to actually find the first
first-descendant children
This will be addressed later.
[kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()]
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: 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>
2010-08-10 00:18:45 +00:00
|
|
|
}
|
|
|
|
} while_each_thread(p, t);
|
2012-07-31 23:43:45 +00:00
|
|
|
read_unlock(&tasklist_lock);
|
oom: introduce find_lock_task_mm() to fix !mm false positives
Almost all ->mm == NULL checks in oom_kill.c are wrong.
The current code assumes that the task without ->mm has already released
its memory and ignores the process. However this is not necessarily true
when this process is multithreaded, other live sub-threads can use this
->mm.
- Remove the "if (!p->mm)" check in select_bad_process(), it is
just wrong.
- Add the new helper, find_lock_task_mm(), which finds the live
thread which uses the memory and takes task_lock() to pin ->mm
- change oom_badness() to use this helper instead of just checking
->mm != NULL.
- As David pointed out, select_bad_process() must never choose the
task without ->mm, but no matter what oom_badness() returns the
task can be chosen if nothing else has been found yet.
Change oom_badness() to return int, change it to return -1 if
find_lock_task_mm() fails, and change select_bad_process() to
check points >= 0.
Note! This patch is not enough, we need more changes.
- oom_badness() was fixed, but oom_kill_task() still ignores
the task without ->mm
- oom_forkbomb_penalty() should use find_lock_task_mm() too,
and it also needs other changes to actually find the first
first-descendant children
This will be addressed later.
[kosaki.motohiro@jp.fujitsu.com: use in badness(), __oom_kill_task()]
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: 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>
2010-08-10 00:18:45 +00:00
|
|
|
|
2012-07-31 23:43:45 +00:00
|
|
|
rcu_read_lock();
|
|
|
|
p = find_lock_task_mm(victim);
|
|
|
|
if (!p) {
|
|
|
|
rcu_read_unlock();
|
|
|
|
put_task_struct(victim);
|
2012-03-21 23:33:46 +00:00
|
|
|
return;
|
2012-07-31 23:43:45 +00:00
|
|
|
} else if (victim != p) {
|
|
|
|
get_task_struct(p);
|
|
|
|
put_task_struct(victim);
|
|
|
|
victim = p;
|
|
|
|
}
|
2012-03-21 23:33:46 +00:00
|
|
|
|
|
|
|
/* mm cannot safely be dereferenced after task_unlock(victim) */
|
|
|
|
mm = victim->mm;
|
|
|
|
pr_err("Killed process %d (%s) total-vm:%lukB, anon-rss:%lukB, file-rss:%lukB\n",
|
|
|
|
task_pid_nr(victim), victim->comm, K(victim->mm->total_vm),
|
|
|
|
K(get_mm_counter(victim->mm, MM_ANONPAGES)),
|
|
|
|
K(get_mm_counter(victim->mm, MM_FILEPAGES)));
|
|
|
|
task_unlock(victim);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Kill all user processes sharing victim->mm in other thread groups, if
|
|
|
|
* any. They don't get access to memory reserves, though, to avoid
|
|
|
|
* depletion of all memory. This prevents mm->mmap_sem livelock when an
|
|
|
|
* oom killed thread cannot exit because it requires the semaphore and
|
|
|
|
* its contended by another thread trying to allocate memory itself.
|
|
|
|
* That thread will now get access to memory reserves since it has a
|
|
|
|
* pending fatal signal.
|
|
|
|
*/
|
|
|
|
for_each_process(p)
|
|
|
|
if (p->mm == mm && !same_thread_group(p, victim) &&
|
|
|
|
!(p->flags & PF_KTHREAD)) {
|
|
|
|
if (p->signal->oom_score_adj == OOM_SCORE_ADJ_MIN)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
task_lock(p); /* Protect ->comm from prctl() */
|
|
|
|
pr_err("Kill process %d (%s) sharing same memory\n",
|
|
|
|
task_pid_nr(p), p->comm);
|
|
|
|
task_unlock(p);
|
2012-03-23 22:02:45 +00:00
|
|
|
do_send_sig_info(SIGKILL, SEND_SIG_FORCED, p, true);
|
2012-03-21 23:33:46 +00:00
|
|
|
}
|
2012-07-31 23:43:45 +00:00
|
|
|
rcu_read_unlock();
|
2012-03-21 23:33:46 +00:00
|
|
|
|
|
|
|
set_tsk_thread_flag(victim, TIF_MEMDIE);
|
2012-03-23 22:02:45 +00:00
|
|
|
do_send_sig_info(SIGKILL, SEND_SIG_FORCED, victim, true);
|
2012-07-31 23:43:45 +00:00
|
|
|
put_task_struct(victim);
|
2005-04-16 22:20:36 +00:00
|
|
|
}
|
2012-03-21 23:33:46 +00:00
|
|
|
#undef K
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2010-08-10 00:18:54 +00:00
|
|
|
/*
|
|
|
|
* Determines whether the kernel must panic because of the panic_on_oom sysctl.
|
|
|
|
*/
|
|
|
|
static void check_panic_on_oom(enum oom_constraint constraint, gfp_t gfp_mask,
|
2010-09-22 20:05:10 +00:00
|
|
|
int order, const nodemask_t *nodemask)
|
2010-08-10 00:18:54 +00:00
|
|
|
{
|
|
|
|
if (likely(!sysctl_panic_on_oom))
|
|
|
|
return;
|
|
|
|
if (sysctl_panic_on_oom != 2) {
|
|
|
|
/*
|
|
|
|
* panic_on_oom == 1 only affects CONSTRAINT_NONE, the kernel
|
|
|
|
* does not panic for cpuset, mempolicy, or memcg allocation
|
|
|
|
* failures.
|
|
|
|
*/
|
|
|
|
if (constraint != CONSTRAINT_NONE)
|
|
|
|
return;
|
|
|
|
}
|
2010-09-22 20:05:10 +00:00
|
|
|
dump_header(NULL, gfp_mask, order, NULL, nodemask);
|
2010-08-10 00:18:54 +00:00
|
|
|
panic("Out of memory: %s panic_on_oom is enabled\n",
|
|
|
|
sysctl_panic_on_oom == 2 ? "compulsory" : "system-wide");
|
|
|
|
}
|
|
|
|
|
2012-07-31 23:43:02 +00:00
|
|
|
#ifdef CONFIG_MEMCG
|
2012-03-21 23:34:10 +00:00
|
|
|
void mem_cgroup_out_of_memory(struct mem_cgroup *memcg, gfp_t gfp_mask,
|
|
|
|
int order)
|
2008-02-07 08:13:58 +00:00
|
|
|
{
|
2011-03-23 23:42:44 +00:00
|
|
|
/*
|
|
|
|
* If current has a pending SIGKILL, then automatically select it. The
|
|
|
|
* goal is to allow it to allocate so that it may quickly exit and free
|
|
|
|
* its memory.
|
|
|
|
*/
|
|
|
|
if (fatal_signal_pending(current)) {
|
|
|
|
set_thread_flag(TIF_MEMDIE);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2012-03-21 23:34:10 +00:00
|
|
|
check_panic_on_oom(CONSTRAINT_MEMCG, gfp_mask, order, NULL);
|
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: Sha Zhengju <handai.szj@taobao.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 23:43:44 +00:00
|
|
|
__mem_cgroup_out_of_memory(memcg, gfp_mask, order);
|
2008-02-07 08:13:58 +00:00
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2006-09-26 06:31:20 +00:00
|
|
|
static BLOCKING_NOTIFIER_HEAD(oom_notify_list);
|
|
|
|
|
|
|
|
int register_oom_notifier(struct notifier_block *nb)
|
|
|
|
{
|
|
|
|
return blocking_notifier_chain_register(&oom_notify_list, nb);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(register_oom_notifier);
|
|
|
|
|
|
|
|
int unregister_oom_notifier(struct notifier_block *nb)
|
|
|
|
{
|
|
|
|
return blocking_notifier_chain_unregister(&oom_notify_list, nb);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(unregister_oom_notifier);
|
|
|
|
|
2007-10-17 06:25:55 +00:00
|
|
|
/*
|
|
|
|
* Try to acquire the OOM killer lock for the zones in zonelist. Returns zero
|
|
|
|
* if a parallel OOM killing is already taking place that includes a zone in
|
|
|
|
* the zonelist. Otherwise, locks all zones in the zonelist and returns 1.
|
|
|
|
*/
|
2010-08-10 00:18:57 +00:00
|
|
|
int try_set_zonelist_oom(struct zonelist *zonelist, gfp_t gfp_mask)
|
2007-10-17 06:25:55 +00:00
|
|
|
{
|
2008-04-28 09:12:17 +00:00
|
|
|
struct zoneref *z;
|
|
|
|
struct zone *zone;
|
2007-10-17 06:25:55 +00:00
|
|
|
int ret = 1;
|
|
|
|
|
2009-01-06 22:39:00 +00:00
|
|
|
spin_lock(&zone_scan_lock);
|
2008-04-28 09:12:17 +00:00
|
|
|
for_each_zone_zonelist(zone, z, zonelist, gfp_zone(gfp_mask)) {
|
|
|
|
if (zone_is_oom_locked(zone)) {
|
2007-10-17 06:25:55 +00:00
|
|
|
ret = 0;
|
|
|
|
goto out;
|
|
|
|
}
|
2008-04-28 09:12:17 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
for_each_zone_zonelist(zone, z, zonelist, gfp_zone(gfp_mask)) {
|
|
|
|
/*
|
2009-01-06 22:39:00 +00:00
|
|
|
* Lock each zone in the zonelist under zone_scan_lock so a
|
2010-08-10 00:18:57 +00:00
|
|
|
* parallel invocation of try_set_zonelist_oom() doesn't succeed
|
2008-04-28 09:12:17 +00:00
|
|
|
* when it shouldn't.
|
|
|
|
*/
|
|
|
|
zone_set_flag(zone, ZONE_OOM_LOCKED);
|
|
|
|
}
|
2007-10-17 06:25:55 +00:00
|
|
|
|
|
|
|
out:
|
2009-01-06 22:39:00 +00:00
|
|
|
spin_unlock(&zone_scan_lock);
|
2007-10-17 06:25:55 +00:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Clears the ZONE_OOM_LOCKED flag for all zones in the zonelist so that failed
|
|
|
|
* allocation attempts with zonelists containing them may now recall the OOM
|
|
|
|
* killer, if necessary.
|
|
|
|
*/
|
2008-04-28 09:12:17 +00:00
|
|
|
void clear_zonelist_oom(struct zonelist *zonelist, gfp_t gfp_mask)
|
2007-10-17 06:25:55 +00:00
|
|
|
{
|
2008-04-28 09:12:17 +00:00
|
|
|
struct zoneref *z;
|
|
|
|
struct zone *zone;
|
2007-10-17 06:25:55 +00:00
|
|
|
|
2009-01-06 22:39:00 +00:00
|
|
|
spin_lock(&zone_scan_lock);
|
2008-04-28 09:12:17 +00:00
|
|
|
for_each_zone_zonelist(zone, z, zonelist, gfp_zone(gfp_mask)) {
|
|
|
|
zone_clear_flag(zone, ZONE_OOM_LOCKED);
|
|
|
|
}
|
2009-01-06 22:39:00 +00:00
|
|
|
spin_unlock(&zone_scan_lock);
|
2007-10-17 06:25:55 +00:00
|
|
|
}
|
|
|
|
|
2010-08-10 00:18:55 +00:00
|
|
|
/*
|
|
|
|
* Try to acquire the oom killer lock for all system zones. Returns zero if a
|
|
|
|
* parallel oom killing is taking place, otherwise locks all zones and returns
|
|
|
|
* non-zero.
|
|
|
|
*/
|
|
|
|
static int try_set_system_oom(void)
|
|
|
|
{
|
|
|
|
struct zone *zone;
|
|
|
|
int ret = 1;
|
|
|
|
|
|
|
|
spin_lock(&zone_scan_lock);
|
|
|
|
for_each_populated_zone(zone)
|
|
|
|
if (zone_is_oom_locked(zone)) {
|
|
|
|
ret = 0;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
for_each_populated_zone(zone)
|
|
|
|
zone_set_flag(zone, ZONE_OOM_LOCKED);
|
|
|
|
out:
|
|
|
|
spin_unlock(&zone_scan_lock);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Clears ZONE_OOM_LOCKED for all system zones so that failed allocation
|
|
|
|
* attempts or page faults may now recall the oom killer, if necessary.
|
|
|
|
*/
|
|
|
|
static void clear_system_oom(void)
|
|
|
|
{
|
|
|
|
struct zone *zone;
|
|
|
|
|
|
|
|
spin_lock(&zone_scan_lock);
|
|
|
|
for_each_populated_zone(zone)
|
|
|
|
zone_clear_flag(zone, ZONE_OOM_LOCKED);
|
|
|
|
spin_unlock(&zone_scan_lock);
|
|
|
|
}
|
|
|
|
|
2005-04-16 22:20:36 +00:00
|
|
|
/**
|
2006-06-23 09:03:13 +00:00
|
|
|
* out_of_memory - kill the "best" process when we run out of memory
|
2008-03-20 00:00:42 +00:00
|
|
|
* @zonelist: zonelist pointer
|
|
|
|
* @gfp_mask: memory allocation flags
|
|
|
|
* @order: amount of memory being requested as a power of 2
|
2010-08-10 00:18:52 +00:00
|
|
|
* @nodemask: nodemask passed to page allocator
|
2012-03-21 23:34:04 +00:00
|
|
|
* @force_kill: true if a task must be killed, even if others are exiting
|
2005-04-16 22:20:36 +00:00
|
|
|
*
|
|
|
|
* If we run out of memory, we have the choice between either
|
|
|
|
* killing a random task (bad), letting the system crash (worse)
|
|
|
|
* OR try to be smart about which process to kill. Note that we
|
|
|
|
* don't have to be perfect here, we just have to be good.
|
|
|
|
*/
|
2009-12-16 00:45:33 +00:00
|
|
|
void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask,
|
2012-03-21 23:34:04 +00:00
|
|
|
int order, nodemask_t *nodemask, bool force_kill)
|
2005-04-16 22:20:36 +00:00
|
|
|
{
|
2010-09-22 20:05:10 +00:00
|
|
|
const nodemask_t *mpol_mask;
|
2010-08-10 00:18:59 +00:00
|
|
|
struct task_struct *p;
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
unsigned long totalpages;
|
2006-09-26 06:31:20 +00:00
|
|
|
unsigned long freed = 0;
|
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: Sha Zhengju <handai.szj@taobao.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 23:43:44 +00:00
|
|
|
unsigned int uninitialized_var(points);
|
2010-08-10 00:18:55 +00:00
|
|
|
enum oom_constraint constraint = CONSTRAINT_NONE;
|
2010-08-19 21:13:39 +00:00
|
|
|
int killed = 0;
|
2006-09-26 06:31:20 +00:00
|
|
|
|
|
|
|
blocking_notifier_call_chain(&oom_notify_list, 0, &freed);
|
|
|
|
if (freed > 0)
|
|
|
|
/* Got some memory back in the last second. */
|
|
|
|
return;
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2010-08-10 00:18:48 +00:00
|
|
|
/*
|
|
|
|
* If current has a pending SIGKILL, then automatically select it. The
|
|
|
|
* goal is to allow it to allocate so that it may quickly exit and free
|
|
|
|
* its memory.
|
|
|
|
*/
|
|
|
|
if (fatal_signal_pending(current)) {
|
|
|
|
set_thread_flag(TIF_MEMDIE);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2006-02-21 02:27:52 +00:00
|
|
|
/*
|
|
|
|
* Check if there were limitations on the allocation (only relevant for
|
|
|
|
* NUMA) that may require different handling.
|
|
|
|
*/
|
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 00:19:46 +00:00
|
|
|
constraint = constrained_alloc(zonelist, gfp_mask, nodemask,
|
|
|
|
&totalpages);
|
2010-09-22 20:05:10 +00:00
|
|
|
mpol_mask = (constraint == CONSTRAINT_MEMORY_POLICY) ? nodemask : NULL;
|
|
|
|
check_panic_on_oom(constraint, gfp_mask, order, mpol_mask);
|
2010-08-10 00:18:59 +00:00
|
|
|
|
2012-07-31 23:42:55 +00:00
|
|
|
if (sysctl_oom_kill_allocating_task && current->mm &&
|
2010-08-10 00:19:38 +00:00
|
|
|
!oom_unkillable_task(current, NULL, nodemask) &&
|
2012-07-31 23:42:55 +00:00
|
|
|
current->signal->oom_score_adj != OOM_SCORE_ADJ_MIN) {
|
2012-07-31 23:43:45 +00:00
|
|
|
get_task_struct(current);
|
2012-03-21 23:33:46 +00:00
|
|
|
oom_kill_process(current, gfp_mask, order, 0, totalpages, NULL,
|
|
|
|
nodemask,
|
|
|
|
"Out of memory (oom_kill_allocating_task)");
|
|
|
|
goto out;
|
2010-08-10 00:18:59 +00:00
|
|
|
}
|
|
|
|
|
mm, memcg: introduce own oom handler to iterate only over its own threads
The global oom killer is serialized by the per-zonelist
try_set_zonelist_oom() which is used in the page allocator. Concurrent
oom kills are thus a rare event and only occur in systems using
mempolicies and with a large number of nodes.
Memory controller oom kills, however, can frequently be concurrent since
there is no serialization once the oom killer is called for oom conditions
in several different memcgs in parallel.
This creates a massive contention on tasklist_lock since the oom killer
requires the readside for the tasklist iteration. If several memcgs are
calling the oom killer, this lock can be held for a substantial amount of
time, especially if threads continue to enter it as other threads are
exiting.
Since the exit path grabs the writeside of the lock with irqs disabled in
a few different places, this can cause a soft lockup on cpus as a result
of tasklist_lock starvation.
The kernel lacks unfair writelocks, and successful calls to the oom killer
usually result in at least one thread entering the exit path, so an
alternative solution is needed.
This patch introduces a seperate oom handler for memcgs so that they do
not require tasklist_lock for as much time. Instead, it iterates only
over the threads attached to the oom memcg and grabs a reference to the
selected thread before calling oom_kill_process() to ensure it doesn't
prematurely exit.
This still requires tasklist_lock for the tasklist dump, iterating
children of the selected process, and killing all other threads on the
system sharing the same memory as the selected victim. So while this
isn't a complete solution to tasklist_lock starvation, it significantly
reduces the amount of time that it is held.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: Sha Zhengju <handai.szj@taobao.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 23:43:44 +00:00
|
|
|
p = select_bad_process(&points, totalpages, mpol_mask, force_kill);
|
2010-08-10 00:18:59 +00:00
|
|
|
/* Found nothing?!?! Either we hang forever, or we panic. */
|
|
|
|
if (!p) {
|
2010-09-22 20:05:10 +00:00
|
|
|
dump_header(NULL, gfp_mask, order, NULL, mpol_mask);
|
2010-08-10 00:18:59 +00:00
|
|
|
panic("Out of memory and no killable processes...\n");
|
|
|
|
}
|
2012-03-21 23:33:46 +00:00
|
|
|
if (PTR_ERR(p) != -1UL) {
|
|
|
|
oom_kill_process(p, gfp_mask, order, points, totalpages, NULL,
|
|
|
|
nodemask, "Out of memory");
|
|
|
|
killed = 1;
|
|
|
|
}
|
2010-08-19 21:13:39 +00:00
|
|
|
out:
|
2005-04-16 22:20:36 +00:00
|
|
|
/*
|
2012-07-31 23:42:37 +00:00
|
|
|
* Give the killed threads a good chance of exiting before trying to
|
|
|
|
* allocate memory again.
|
2005-04-16 22:20:36 +00:00
|
|
|
*/
|
2012-07-31 23:42:37 +00:00
|
|
|
if (killed)
|
|
|
|
schedule_timeout_killable(1);
|
2005-04-16 22:20:36 +00:00
|
|
|
}
|
2010-08-10 00:18:55 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* The pagefault handler calls here because it is out of memory, so kill a
|
|
|
|
* memory-hogging task. If a populated zone has ZONE_OOM_LOCKED set, a parallel
|
|
|
|
* oom killing is already in progress so do nothing. If a task is found with
|
|
|
|
* TIF_MEMDIE set, it has been killed so do nothing and allow it to exit.
|
|
|
|
*/
|
|
|
|
void pagefault_out_of_memory(void)
|
|
|
|
{
|
|
|
|
if (try_set_system_oom()) {
|
2012-03-21 23:34:04 +00:00
|
|
|
out_of_memory(NULL, 0, 0, NULL, false);
|
2010-08-10 00:18:55 +00:00
|
|
|
clear_system_oom();
|
|
|
|
}
|
2012-07-31 23:42:37 +00:00
|
|
|
schedule_timeout_killable(1);
|
2010-08-10 00:18:55 +00:00
|
|
|
}
|