linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: Minchan Kim <minchan@kernel.org>
Cc: linux-mm <linux-mm@kvack.org>,
	LKML <linux-kernel@vger.kernel.org>,
	John Dias <joaodias@google.com>, Michal Hocko <mhocko@suse.com>,
	David Hildenbrand <david@redhat.com>,
	Jason Baron <jbaron@akamai.com>
Subject: Re: [PATCH v2] mm: page_alloc: dump migrate-failed pages
Date: Mon, 8 Mar 2021 16:21:28 -0800	[thread overview]
Message-ID: <20210308162128.9b4a7d4c1576a72fd4878bdb@linux-foundation.org> (raw)
In-Reply-To: <20210308202047.1903802-1-minchan@kernel.org>

On Mon,  8 Mar 2021 12:20:47 -0800 Minchan Kim <minchan@kernel.org> wrote:

> alloc_contig_range is usually used on cma area or movable zone.
> It's critical if the page migration fails on those areas so
> dump more debugging message.
> 
> page refcount, mapcount with page flags on dump_page are
> helpful information to deduce the culprit. Furthermore,
> dump_page_owner was super helpful to find long term pinner
> who initiated the page allocation.
> 
> Admin could enable the dump like this(by default, disabled)
> 
> 	echo "func dump_migrate_failure_pages +p" > control
> 
> Admin could disable it.
> 
> 	echo "func dump_migrate_failure_pages =_" > control
> 
> ...
>
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -8453,6 +8453,34 @@ static unsigned long pfn_max_align_up(unsigned long pfn)
>  				pageblock_nr_pages));
>  }
>  
> +#if defined(CONFIG_DYNAMIC_DEBUG) || \
> +	(defined(CONFIG_DYNAMIC_DEBUG_CORE) && defined(DYNAMIC_DEBUG_MODULE))
> +static DEFINE_RATELIMIT_STATE(alloc_contig_ratelimit_state,
> +		DEFAULT_RATELIMIT_INTERVAL, DEFAULT_RATELIMIT_BURST);
> +int alloc_contig_ratelimit(void)
> +{
> +	return __ratelimit(&alloc_contig_ratelimit_state);
> +}

Wow, that's an eyesore.  We're missing helpers in the ratelimit code. 
Can we do something like

/* description goes here */
#define RATELIMIT2(interval, burst)
({
	static DEFINE_RATELIMIT_STATE(_rs, interval, burst);

	__ratelimit(_rs);
})

#define RATELIMIT()
	RATELIMIT2(DEFAULT_RATELIMIT_INTERVAL, DEFAULT_RATELIMIT_BURST)

> +void dump_migrate_failure_pages(struct list_head *page_list)
> +{
> +	DEFINE_DYNAMIC_DEBUG_METADATA(descriptor,
> +			"migrate failure");
> +	if (DYNAMIC_DEBUG_BRANCH(descriptor) &&
> +			alloc_contig_ratelimit()) {
> +		struct page *page;
> +
> +		WARN(1, "failed callstack");
> +		list_for_each_entry(page, page_list, lru)
> +			dump_page(page, "migration failure");
> +	}
> +}

Then we can simply do

	if (DYNAMIC_DEBUG_BRANCH(descriptor) && RATELIMIT())



  parent reply	other threads:[~2021-03-09  0:22 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-08 20:20 [PATCH v2] mm: page_alloc: dump migrate-failed pages Minchan Kim
2021-03-08 21:29 ` kernel test robot
2021-03-08 22:29   ` Minchan Kim
2021-03-09  0:41     ` [kbuild-all] " Rong Chen
2021-03-09  2:23       ` Minchan Kim
2021-03-09  0:21 ` Andrew Morton [this message]
2021-03-09  2:21   ` Minchan Kim
2021-03-09  0:30 ` kernel test robot
2021-03-09  0:30 ` [RFC PATCH] mm: page_alloc: alloc_contig_ratelimit() can be static kernel test robot
2021-03-09  9:32 ` [PATCH v2] mm: page_alloc: dump migrate-failed pages Michal Hocko
2021-03-09 16:15   ` Minchan Kim
2021-03-09 16:32     ` Michal Hocko
2021-03-09 17:27       ` Minchan Kim
2021-03-10 13:04         ` Michal Hocko
2021-03-10 15:59           ` Minchan Kim
2021-03-10  7:42     ` Minchan Kim
2021-03-10  8:20       ` David Hildenbrand
2021-03-10 15:45         ` Minchan Kim
2021-03-10 13:07       ` Michal Hocko
2021-03-10 16:05         ` Minchan Kim
2021-03-10 16:46           ` Michal Hocko
2021-03-10 17:06             ` Minchan Kim
2021-03-10 18:07               ` Minchan Kim
2021-03-10 13:26   ` Matthew Wilcox
2021-03-10 13:54     ` Michal Hocko

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210308162128.9b4a7d4c1576a72fd4878bdb@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=david@redhat.com \
    --cc=jbaron@akamai.com \
    --cc=joaodias@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.com \
    --cc=minchan@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).