From: Andrew Morton <akpm@linux-foundation.org>
To: Wu Fengguang <fengguang.wu@intel.com>
Cc: Minchan Kim <minchan.kim@gmail.com>,
Dave Young <hidave.darkstar@gmail.com>,
linux-mm <linux-mm@kvack.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Mel Gorman <mel@linux.vnet.ibm.com>
Subject: Re: readahead and oom
Date: Tue, 26 Apr 2011 12:47:43 -0700 [thread overview]
Message-ID: <20110426124743.e58d9746.akpm@linux-foundation.org> (raw)
In-Reply-To: <20110426092029.GA27053@localhost>
On Tue, 26 Apr 2011 17:20:29 +0800
Wu Fengguang <fengguang.wu@intel.com> wrote:
> Pass __GFP_NORETRY|__GFP_NOWARN for readahead page allocations.
>
> readahead page allocations are completely optional. They are OK to
> fail and in particular shall not trigger OOM on themselves.
I have distinct recollections of trying this many years ago, finding
that it caused problems then deciding not to do it. But I can't find
an email trail and I don't remember the reasons :(
If the system is so stressed for memory that the oom-killer might get
involved then the readahead pages may well be getting reclaimed before
the application actually gets to use them. But that's just an aside.
Ho hum. The patch *seems* good (as it did 5-10 years ago ;)) but there
may be surprising side-effects which could be exposed under heavy
testing. Testing which I'm sure hasn't been performed...
WARNING: multiple messages have this Message-ID (diff)
From: Andrew Morton <akpm@linux-foundation.org>
To: Wu Fengguang <fengguang.wu@intel.com>
Cc: Minchan Kim <minchan.kim@gmail.com>,
Dave Young <hidave.darkstar@gmail.com>,
linux-mm <linux-mm@kvack.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Mel Gorman <mel@linux.vnet.ibm.com>
Subject: Re: readahead and oom
Date: Tue, 26 Apr 2011 12:47:43 -0700 [thread overview]
Message-ID: <20110426124743.e58d9746.akpm@linux-foundation.org> (raw)
In-Reply-To: <20110426092029.GA27053@localhost>
On Tue, 26 Apr 2011 17:20:29 +0800
Wu Fengguang <fengguang.wu@intel.com> wrote:
> Pass __GFP_NORETRY|__GFP_NOWARN for readahead page allocations.
>
> readahead page allocations are completely optional. They are OK to
> fail and in particular shall not trigger OOM on themselves.
I have distinct recollections of trying this many years ago, finding
that it caused problems then deciding not to do it. But I can't find
an email trail and I don't remember the reasons :(
If the system is so stressed for memory that the oom-killer might get
involved then the readahead pages may well be getting reclaimed before
the application actually gets to use them. But that's just an aside.
Ho hum. The patch *seems* good (as it did 5-10 years ago ;)) but there
may be surprising side-effects which could be exposed under heavy
testing. Testing which I'm sure hasn't been performed...
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2011-04-26 19:51 UTC|newest]
Thread overview: 84+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-26 5:49 readahead and oom Dave Young
2011-04-26 5:49 ` Dave Young
2011-04-26 5:55 ` Wu Fengguang
2011-04-26 5:55 ` Wu Fengguang
2011-04-26 6:05 ` Dave Young
2011-04-26 6:05 ` Dave Young
2011-04-26 6:07 ` Dave Young
2011-04-26 6:07 ` Dave Young
2011-04-26 6:25 ` Wu Fengguang
2011-04-26 6:25 ` Wu Fengguang
2011-04-26 6:29 ` Dave Young
2011-04-26 6:29 ` Dave Young
2011-04-26 6:34 ` Wu Fengguang
2011-04-26 6:34 ` Wu Fengguang
2011-04-26 6:50 ` KOSAKI Motohiro
2011-04-26 6:50 ` KOSAKI Motohiro
2011-04-26 7:41 ` Minchan Kim
2011-04-26 7:41 ` Minchan Kim
2011-04-26 9:20 ` Wu Fengguang
2011-04-26 9:20 ` Wu Fengguang
2011-04-26 9:28 ` Minchan Kim
2011-04-26 9:28 ` Minchan Kim
2011-04-26 10:18 ` Pekka Enberg
2011-04-26 10:18 ` Pekka Enberg
2011-04-26 19:47 ` Andrew Morton [this message]
2011-04-26 19:47 ` Andrew Morton
2011-04-28 4:19 ` Wu Fengguang
2011-04-28 13:36 ` [RFC][PATCH] mm: cut down __GFP_NORETRY page allocation failures Wu Fengguang
2011-04-28 13:36 ` Wu Fengguang
2011-04-28 13:38 ` [patch] vmstat: account " Wu Fengguang
2011-04-28 13:38 ` Wu Fengguang
2011-04-28 13:50 ` KOSAKI Motohiro
2011-04-28 13:50 ` KOSAKI Motohiro
2011-04-29 2:28 ` [RFC][PATCH] mm: cut down __GFP_NORETRY " Wu Fengguang
2011-04-29 2:28 ` Wu Fengguang
2011-04-29 2:58 ` Wu Fengguang
2011-04-29 2:58 ` Wu Fengguang
2011-04-30 14:17 ` Wu Fengguang
2011-04-30 14:17 ` Wu Fengguang
2011-05-01 16:35 ` Minchan Kim
2011-05-01 16:35 ` Minchan Kim
2011-05-01 16:37 ` Minchan Kim
2011-05-01 16:37 ` Minchan Kim
2011-05-02 10:14 ` KOSAKI Motohiro
2011-05-02 10:14 ` KOSAKI Motohiro
2011-05-03 0:53 ` Minchan Kim
2011-05-03 0:53 ` Minchan Kim
2011-05-03 1:25 ` KOSAKI Motohiro
2011-05-03 1:25 ` KOSAKI Motohiro
2011-05-02 10:29 ` Wu Fengguang
2011-05-02 11:08 ` Wu Fengguang
2011-05-02 11:08 ` Wu Fengguang
2011-05-03 0:49 ` Minchan Kim
2011-05-03 0:49 ` Minchan Kim
2011-05-03 3:51 ` Wu Fengguang
2011-05-03 3:51 ` Wu Fengguang
2011-05-03 4:17 ` Minchan Kim
2011-05-03 4:17 ` Minchan Kim
2011-05-02 13:29 ` Wu Fengguang
2011-05-02 13:29 ` Wu Fengguang
2011-05-02 13:49 ` Wu Fengguang
2011-05-02 13:49 ` Wu Fengguang
2011-05-03 0:27 ` Satoru Moriya
2011-05-03 0:27 ` Satoru Moriya
2011-05-03 2:49 ` Wu Fengguang
2011-05-03 2:49 ` Wu Fengguang
2011-05-04 1:56 ` Dave Young
2011-05-04 1:56 ` Dave Young
2011-05-04 2:32 ` Dave Young
2011-05-04 2:32 ` Dave Young
2011-05-04 2:56 ` Wu Fengguang
2011-05-04 2:56 ` Wu Fengguang
2011-05-04 4:23 ` Wu Fengguang
2011-05-04 4:23 ` Wu Fengguang
2011-05-04 4:00 ` Wu Fengguang
2011-05-04 4:00 ` Wu Fengguang
2011-05-04 7:33 ` Dave Young
2011-05-04 7:33 ` Dave Young
2011-04-26 6:13 ` readahead and oom Wu Fengguang
2011-04-26 6:13 ` Wu Fengguang
2011-04-26 6:23 ` Dave Young
2011-04-26 6:23 ` Dave Young
2011-04-26 9:37 ` [PATCH] mm: readahead page allocations are OK to fail Wu Fengguang
2011-04-26 9:37 ` Wu Fengguang
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=20110426124743.e58d9746.akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=fengguang.wu@intel.com \
--cc=hidave.darkstar@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mel@linux.vnet.ibm.com \
--cc=minchan.kim@gmail.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.