lkml.org 
[lkml]   [2011]   [May]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 1/4] oom: improve dump_tasks() show items
On Tue, 10 May 2011, KOSAKI Motohiro wrote:

> Recently, oom internal logic was dramatically changed. Thus
> dump_tasks() is no longer useful. it has some meaningless
> items and don't have some oom socre related items.
>

This changelog is inaccurate.

dump_tasks() is actually useful as it currently stands; there are things
that you may add or remove but saying that it is "no longer useful" is an
exaggeration.

> This patch adapt displaying fields to new oom logic.
>
> details
> ==========
> removed: pid (we always kill process. don't need thread id),
> mm->total_vm (we no longer uses virtual memory size)

Showing mm->total_vm is still interesting to know what the old heuristic
would have used rather than the new heuristic, I'd prefer if we kept it.

> signal->oom_adj (we no longer uses it internally)
> added: ppid (we often kill sacrifice child process)
> modify: RSS (account mm->nr_ptes too)

I'd prefer if ptes were shown independently from rss instead of adding it
to the thread's true rss usage and representing it as such.

I think the cpu should also be removed.

For the next version, could you show the old output and comparsion to new
output in the changelog?

>
> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> ---
>
> Strictly speaking. this is NOT a part of oom fixing patches. but it's
> necessary when I parse QAI's test result.
>
>
> mm/oom_kill.c | 14 ++++++++------
> 1 files changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/mm/oom_kill.c b/mm/oom_kill.c
> index f52e85c..118d958 100644
> --- a/mm/oom_kill.c
> +++ b/mm/oom_kill.c
> @@ -355,7 +355,7 @@ static void dump_tasks(const struct mem_cgroup *mem, const nodemask_t *nodemask)
> struct task_struct *p;
> struct task_struct *task;
>
> - pr_info("[ pid ] uid tgid total_vm rss cpu oom_adj oom_score_adj name\n");
> + pr_info("[ pid] ppid uid rss cpu score_adj name\n");
> for_each_process(p) {
> if (oom_unkillable_task(p, mem, nodemask))
> continue;
> @@ -370,11 +370,13 @@ static void dump_tasks(const struct mem_cgroup *mem, const nodemask_t *nodemask)
> continue;
> }
>
> - pr_info("[%5d] %5d %5d %8lu %8lu %3u %3d %5d %s\n",
> - task->pid, task_uid(task), task->tgid,
> - task->mm->total_vm, get_mm_rss(task->mm),
> - task_cpu(task), task->signal->oom_adj,
> - task->signal->oom_score_adj, task->comm);
> + pr_info("[%6d] %6d %5d %8lu %4u %9d %s\n",
> + task_tgid_nr(task), task_tgid_nr(task->real_parent),
> + task_uid(task),
> + get_mm_rss(task->mm) + p->mm->nr_ptes,
> + task_cpu(task),
> + task->signal->oom_score_adj,
> + task->comm);
> task_unlock(task);
> }
> }


\
 
 \ /
  Last update: 2011-05-11 01:31    [W:0.168 / U:1.168 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site