On Thu 22-10-15 06:49:01, Hongjie Fang (方洪杰) wrote:
>
> The oom_adj has been replaced by oom_score_adj in kernel,
> but the /proc/pid/oom_adj is provided for legacy purposes.
> When write/read a value into/from /proc/pid/oom_adj,
> there is a transformation between oom_adj and oom_score_adj.
>
> After writing a new value into /proc/pid/oom_adj, then read it.
> The return value is a different value than you wrote.
> Fix this by adding a adjustment factor.
... when printing the value.
Your previous patch has changed the stored value while this one only
changes the presented value. The previous one was more correct IMO
but in reality the difference (+-1) in oom_score_adj should be hardly
noticeable and nobody has complained about the current scaling for years
so this approach is probably more conservative.
> Signed-off-by: Hongjie Fang <hongjie.f...@spreadtrum.com>
Acked-by: Michal Hocko <mho...@suse.com>
> ---
> Encountered the problem when I changed a task's oom_adj on
> an Android smart phone. As follows,
> 1. # cat /proc/1450/oom_adj
> 15
> 2. # echo 10 > /proc/1450/oom_adj
> 3. # cat /proc/1450/oom_adj
> 9
>
>
> diff --git a/fs/proc/base.c b/fs/proc/base.c
> index b25eee4..2312e43 100644
> --- a/fs/proc/base.c
> +++ b/fs/proc/base.c
> @@ -1019,15 +1019,19 @@ static ssize_t oom_adj_read(struct file *file, char
> __user *buf, size_t count,
> int oom_adj = OOM_ADJUST_MIN;
> size_t len;
> unsigned long flags;
> + int adjust;
>
> if (!task)
> return -ESRCH;
> if (lock_task_sighand(task, &flags)) {
> - if (task->signal->oom_score_adj == OOM_SCORE_ADJ_MAX)
> + if (task->signal->oom_score_adj == OOM_SCORE_ADJ_MAX) {
> oom_adj = OOM_ADJUST_MAX;
> - else
> - oom_adj = (task->signal->oom_score_adj * -OOM_DISABLE) /
> + } else {
> + adjust = task->signal->oom_score_adj > 0 ?
> + (OOM_SCORE_ADJ_MAX-1) : -(OOM_SCORE_ADJ_MAX-1);
> + oom_adj = (task->signal->oom_score_adj * -OOM_DISABLE +
> adjust) /
> OOM_SCORE_ADJ_MAX;
> + }
> unlock_task_sighand(task, &flags);
> }
> put_task_struct(task);
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
Michal Hocko
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
