[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20201112111436.c5deeadd3578877fc0b844a1@linux-foundation.org>
Date: Thu, 12 Nov 2020 11:14:36 -0800
From: Andrew Morton <akpm@...ux-foundation.org>
To: Georgi Djakov <georgi.djakov@...aro.org>
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
sudaraja@...eaurora.org, pratikp@...eaurora.org,
lmark@...eaurora.org
Subject: Re: [PATCH] mm/page_owner: Record timestamp and pid
On Thu, 12 Nov 2020 20:41:06 +0200 Georgi Djakov <georgi.djakov@...aro.org> wrote:
> From: Liam Mark <lmark@...eaurora.org>
>
> Collect the time for each allocation recorded in page owner so that
> allocation "surges" can be measured.
>
> Record the pid for each allocation recorded in page owner so that
> the source of allocation "surges" can be better identified.
Please provide a description of why this is considered useful. What
has it been used for, what problems has it been used to solve?
Are there userspace tools which aid in the processing of this new
information?
Can/should Documentation/vm/page_owner.rst be updated?
> --- a/mm/page_owner.c
> +++ b/mm/page_owner.c
> @@ -10,6 +10,7 @@
> #include <linux/migrate.h>
> #include <linux/stackdepot.h>
> #include <linux/seq_file.h>
> +#include <linux/sched/clock.h>
>
> #include "internal.h"
>
> @@ -25,6 +26,8 @@ struct page_owner {
> gfp_t gfp_mask;
> depot_stack_handle_t handle;
> depot_stack_handle_t free_handle;
> + u64 ts_nsec;
> + int pid;
pid_t would be nicer?
Powered by blists - more mailing lists