diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2013-07-03 17:12:13 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2013-07-03 17:12:13 -0700 |
commit | 7f0ef0267e20d62d45d527911a993b1e998f4968 (patch) | |
tree | de51abc7da5903f59d83e23937f22420164c9477 /mm/memcontrol.c | |
parent | 862f0012549110d6f2586bf54b52ed4540cbff3a (diff) | |
parent | 9307c29524502c21f0e8a6d96d850b2f5bc0bd9a (diff) | |
download | linux-7f0ef0267e20d62d45d527911a993b1e998f4968.tar.bz2 |
Merge branch 'akpm' (updates from Andrew Morton)
Merge first patch-bomb from Andrew Morton:
- various misc bits
- I'm been patchmonkeying ocfs2 for a while, as Joel and Mark have been
distracted. There has been quite a bit of activity.
- About half the MM queue
- Some backlight bits
- Various lib/ updates
- checkpatch updates
- zillions more little rtc patches
- ptrace
- signals
- exec
- procfs
- rapidio
- nbd
- aoe
- pps
- memstick
- tools/testing/selftests updates
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (445 commits)
tools/testing/selftests: don't assume the x bit is set on scripts
selftests: add .gitignore for kcmp
selftests: fix clean target in kcmp Makefile
selftests: add .gitignore for vm
selftests: add hugetlbfstest
self-test: fix make clean
selftests: exit 1 on failure
kernel/resource.c: remove the unneeded assignment in function __find_resource
aio: fix wrong comment in aio_complete()
drivers/w1/slaves/w1_ds2408.c: add magic sequence to disable P0 test mode
drivers/memstick/host/r592.c: convert to module_pci_driver
drivers/memstick/host/jmb38x_ms: convert to module_pci_driver
pps-gpio: add device-tree binding and support
drivers/pps/clients/pps-gpio.c: convert to module_platform_driver
drivers/pps/clients/pps-gpio.c: convert to devm_* helpers
drivers/parport/share.c: use kzalloc
Documentation/accounting/getdelays.c: avoid strncpy in accounting tool
aoe: update internal version number to v83
aoe: update copyright date
aoe: perform I/O completions in parallel
...
Diffstat (limited to 'mm/memcontrol.c')
-rw-r--r-- | mm/memcontrol.c | 97 |
1 files changed, 63 insertions, 34 deletions
diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 194721839cf5..2e851f453814 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1148,6 +1148,58 @@ skip_node: return NULL; } +static void mem_cgroup_iter_invalidate(struct mem_cgroup *root) +{ + /* + * When a group in the hierarchy below root is destroyed, the + * hierarchy iterator can no longer be trusted since it might + * have pointed to the destroyed group. Invalidate it. + */ + atomic_inc(&root->dead_count); +} + +static struct mem_cgroup * +mem_cgroup_iter_load(struct mem_cgroup_reclaim_iter *iter, + struct mem_cgroup *root, + int *sequence) +{ + struct mem_cgroup *position = NULL; + /* + * A cgroup destruction happens in two stages: offlining and + * release. They are separated by a RCU grace period. + * + * If the iterator is valid, we may still race with an + * offlining. The RCU lock ensures the object won't be + * released, tryget will fail if we lost the race. + */ + *sequence = atomic_read(&root->dead_count); + if (iter->last_dead_count == *sequence) { + smp_rmb(); + position = iter->last_visited; + if (position && !css_tryget(&position->css)) + position = NULL; + } + return position; +} + +static void mem_cgroup_iter_update(struct mem_cgroup_reclaim_iter *iter, + struct mem_cgroup *last_visited, + struct mem_cgroup *new_position, + int sequence) +{ + if (last_visited) + css_put(&last_visited->css); + /* + * We store the sequence count from the time @last_visited was + * loaded successfully instead of rereading it here so that we + * don't lose destruction events in between. We could have + * raced with the destruction of @new_position after all. + */ + iter->last_visited = new_position; + smp_wmb(); + iter->last_dead_count = sequence; +} + /** * mem_cgroup_iter - iterate over memory cgroup hierarchy * @root: hierarchy root @@ -1171,7 +1223,6 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root, { struct mem_cgroup *memcg = NULL; struct mem_cgroup *last_visited = NULL; - unsigned long uninitialized_var(dead_count); if (mem_cgroup_disabled()) return NULL; @@ -1191,6 +1242,7 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root, rcu_read_lock(); while (!memcg) { struct mem_cgroup_reclaim_iter *uninitialized_var(iter); + int uninitialized_var(seq); if (reclaim) { int nid = zone_to_nid(reclaim->zone); @@ -1204,37 +1256,13 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root, goto out_unlock; } - /* - * If the dead_count mismatches, a destruction - * has happened or is happening concurrently. - * If the dead_count matches, a destruction - * might still happen concurrently, but since - * we checked under RCU, that destruction - * won't free the object until we release the - * RCU reader lock. Thus, the dead_count - * check verifies the pointer is still valid, - * css_tryget() verifies the cgroup pointed to - * is alive. - */ - dead_count = atomic_read(&root->dead_count); - if (dead_count == iter->last_dead_count) { - smp_rmb(); - last_visited = iter->last_visited; - if (last_visited && - !css_tryget(&last_visited->css)) - last_visited = NULL; - } + last_visited = mem_cgroup_iter_load(iter, root, &seq); } memcg = __mem_cgroup_iter_next(root, last_visited); if (reclaim) { - if (last_visited) - css_put(&last_visited->css); - - iter->last_visited = memcg; - smp_wmb(); - iter->last_dead_count = dead_count; + mem_cgroup_iter_update(iter, last_visited, memcg, seq); if (!memcg) iter->generation++; @@ -1448,11 +1476,12 @@ static bool mem_cgroup_same_or_subtree(const struct mem_cgroup *root_memcg, return ret; } -int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *memcg) +bool task_in_mem_cgroup(struct task_struct *task, + const struct mem_cgroup *memcg) { - int ret; struct mem_cgroup *curr = NULL; struct task_struct *p; + bool ret; p = find_lock_task_mm(task); if (p) { @@ -1464,14 +1493,14 @@ int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *memcg) * killer still needs to detect if they have already been oom * killed to prevent needlessly killing additional tasks. */ - task_lock(task); + rcu_read_lock(); curr = mem_cgroup_from_task(task); if (curr) css_get(&curr->css); - task_unlock(task); + rcu_read_unlock(); } if (!curr) - return 0; + return false; /* * We should check use_hierarchy of "memcg" not "curr". Because checking * use_hierarchy of "curr" here make this function true if hierarchy is @@ -6317,14 +6346,14 @@ static void mem_cgroup_invalidate_reclaim_iterators(struct mem_cgroup *memcg) struct mem_cgroup *parent = memcg; while ((parent = parent_mem_cgroup(parent))) - atomic_inc(&parent->dead_count); + mem_cgroup_iter_invalidate(parent); /* * if the root memcg is not hierarchical we have to check it * explicitely. */ if (!root_mem_cgroup->use_hierarchy) - atomic_inc(&root_mem_cgroup->dead_count); + mem_cgroup_iter_invalidate(root_mem_cgroup); } static void mem_cgroup_css_offline(struct cgroup *cont) |