[CCing Johannes and Andrew]
On Sat 07-03-15 00:23:44, Balasubramani Vivekanandan wrote:
> If kernel panics due to oom, caused by a cgroup reaching its limit, when
> 'compulsory panic_on_oom' is enabled, then we will only see that the OOM
> happened because of "compulsory panic_on_oom is enabled" but this
> doesn't tell the difference between mempolicy and memcg. And dumping
> system wide information is plain wrong and more confusing. This patch
> provides the information of the cgroup whose limit triggerred panic
>
> Signed-off-by: Balasubramani Vivekanandan <balasubramani_vivekanandan@xxxxxxxxxx>
Acked-by: Michal Hocko <mhocko@xxxxxxx>
Thanks!
> ---
> include/linux/oom.h | 3 ++-
> mm/memcontrol.c | 16 +++++++++-------
> mm/oom_kill.c | 7 ++++---
> 3 files changed, 15 insertions(+), 11 deletions(-)
>
> diff --git a/include/linux/oom.h b/include/linux/oom.h
> index d5771bed..44b2f6f 100644
> --- a/include/linux/oom.h
> +++ b/include/linux/oom.h
> @@ -66,7 +66,8 @@ extern bool oom_zonelist_trylock(struct zonelist *zonelist, gfp_t gfp_flags);
> extern void oom_zonelist_unlock(struct zonelist *zonelist, gfp_t gfp_flags);
>
> extern void check_panic_on_oom(enum oom_constraint constraint, gfp_t gfp_mask,
> - int order, const nodemask_t *nodemask);
> + int order, const nodemask_t *nodemask,
> + struct mem_cgroup *memcg);
>
> extern enum oom_scan_t oom_scan_process_thread(struct task_struct *task,
> unsigned long totalpages, const nodemask_t *nodemask,
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 0c86945..fb884eb 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -1442,15 +1442,17 @@ void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p)
> struct mem_cgroup *iter;
> unsigned int i;
>
> - if (!p)
> - return;
> -
> mutex_lock(&oom_info_lock);
> rcu_read_lock();
>
> - pr_info("Task in ");
> - pr_cont_cgroup_path(task_cgroup(p, memory_cgrp_id));
> - pr_cont(" killed as a result of limit of ");
> + if (p) {
> + pr_info("Task in ");
> + pr_cont_cgroup_path(task_cgroup(p, memory_cgrp_id));
> + pr_cont(" killed as a result of limit of ");
> + } else {
> + pr_info("Memory limit reached of cgroup ");
> + }
> +
> pr_cont_cgroup_path(memcg->css.cgroup);
> pr_cont("\n");
>
> @@ -1537,7 +1539,7 @@ static void mem_cgroup_out_of_memory(struct mem_cgroup *memcg, gfp_t gfp_mask,
> return;
> }
>
> - check_panic_on_oom(CONSTRAINT_MEMCG, gfp_mask, order, NULL);
> + check_panic_on_oom(CONSTRAINT_MEMCG, gfp_mask, order, NULL, memcg);
> totalpages = mem_cgroup_get_limit(memcg) ? : 1;
> for_each_mem_cgroup_tree(iter, memcg) {
> struct css_task_iter it;
> diff --git a/mm/oom_kill.c b/mm/oom_kill.c
> index 642f38c..52628c8 100644
> --- a/mm/oom_kill.c
> +++ b/mm/oom_kill.c
> @@ -612,7 +612,8 @@ void oom_kill_process(struct task_struct *p, gfp_t gfp_mask, int order,
> * Determines whether the kernel must panic because of the panic_on_oom sysctl.
> */
> void check_panic_on_oom(enum oom_constraint constraint, gfp_t gfp_mask,
> - int order, const nodemask_t *nodemask)
> + int order, const nodemask_t *nodemask,
> + struct mem_cgroup *memcg)
> {
> if (likely(!sysctl_panic_on_oom))
> return;
> @@ -625,7 +626,7 @@ void check_panic_on_oom(enum oom_constraint constraint, gfp_t gfp_mask,
> if (constraint != CONSTRAINT_NONE)
> return;
> }
> - dump_header(NULL, gfp_mask, order, NULL, nodemask);
> + dump_header(NULL, gfp_mask, order, memcg, nodemask);
> panic("Out of memory: %s panic_on_oom is enabled\n",
> sysctl_panic_on_oom == 2 ? "compulsory" : "system-wide");
> }
> @@ -740,7 +741,7 @@ static void __out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask,
> constraint = constrained_alloc(zonelist, gfp_mask, nodemask,
> &totalpages);
> mpol_mask = (constraint == CONSTRAINT_MEMORY_POLICY) ? nodemask : NULL;
> - check_panic_on_oom(constraint, gfp_mask, order, mpol_mask);
> + check_panic_on_oom(constraint, gfp_mask, order, mpol_mask, NULL);
>
> if (sysctl_oom_kill_allocating_task && current->mm &&
> !oom_unkillable_task(current, NULL, nodemask) &&
> --
> 2.3.1
>
--
Michal Hocko
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe cgroups" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
