linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Lameter <cl@linux.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Nitin Gupta <ngupta@vflare.org>,
	Pekka Enberg <penberg@cs.helsinki.fi>,
	Minchan Kim <minchan.kim@gmail.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Greg KH <greg@kroah.com>,
	Linux Driver Project <devel@driverdev.osuosl.org>,
	linux-mm <linux-mm@kvack.org>,
	linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 03/10] Use percpu stats
Date: Tue, 31 Aug 2010 16:35:37 -0500 (CDT)	[thread overview]
Message-ID: <alpine.DEB.2.00.1008311635100.867@router.home> (raw)
In-Reply-To: <1283290106.2198.26.camel@edumazet-laptop>

On Tue, 31 Aug 2010, Eric Dumazet wrote:

> > Yes, this_cpu_add() seems sufficient. I can't recall why I used u64_stats_*
> > but if it's not required for atomic access to 64-bit then why was it added to
> > the mainline in the first place?
>
> Because we wanted to have fast 64bit counters, even on 32bit arches, and
> this has litle to do with 'atomic' on one entity, but a group of
> counters. (check drivers/net/loopback.c, lines 91-94). No lock prefix
> used in fast path.
>
> We also wanted readers to read correct values, not a value being changed
> by a writer, with inconsistent 32bit halves. SNMP applications want
> monotonically increasing counters.
>
> this_cpu_add()/this_cpu_read() doesnt fit.
>
> Even for single counter, this_cpu_read(64bit) is not using an RMW
> (cmpxchg8) instruction, so you can get very strange results when low
> order 32bit wraps.

How about fixing it so that everyone benefits?

--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2010-08-31 21:35 UTC|newest]

Thread overview: 67+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-09 17:26 [PATCH 00/10] zram: various improvements and cleanups Nitin Gupta
2010-08-09 17:26 ` [PATCH 01/10] Replace ioctls with sysfs interface Nitin Gupta
2010-08-09 18:34   ` Pekka Enberg
2010-08-10  3:06     ` Nitin Gupta
2010-08-31 23:06   ` Dave Hansen
2010-08-09 17:26 ` [PATCH 02/10] Remove need for explicit device initialization Nitin Gupta
2010-08-09 18:36   ` Pekka Enberg
2010-08-10  3:38     ` Nitin Gupta
2010-08-09 17:26 ` [PATCH 03/10] Use percpu stats Nitin Gupta
2010-08-09 18:44   ` Pekka Enberg
2010-08-10  4:34   ` Andrew Morton
2010-08-11 16:39     ` Nitin Gupta
2010-08-11 17:18       ` Andrew Morton
2010-08-30 16:20   ` Christoph Lameter
2010-08-31 20:31     ` Nitin Gupta
2010-08-31 21:28       ` Eric Dumazet
2010-08-31 21:35         ` Christoph Lameter [this message]
2010-08-31 21:41           ` Eric Dumazet
2010-09-01 20:05             ` Christoph Lameter
2010-09-01 20:38               ` Eric Dumazet
2010-09-02  0:34                 ` Christoph Lameter
2010-08-31  5:36   ` Anton Blanchard
2010-09-01  3:41     ` Anton Blanchard
2010-09-01  3:51   ` Anton Blanchard
2010-09-17 20:59     ` Andrew Morton
2010-08-09 17:26 ` [PATCH 04/10] Use percpu buffers Nitin Gupta
2010-08-09 18:57   ` Pekka Enberg
2010-08-10  4:47     ` Nitin Gupta
2010-08-10  5:05       ` Pekka Enberg
2010-08-10  5:32         ` Nitin Gupta
2010-08-10  7:36           ` Pekka Enberg
2010-08-09 17:26 ` [PATCH 05/10] Reduce per table entry overhead by 4 bytes Nitin Gupta
2010-08-09 18:59   ` Pekka Enberg
2010-08-10  4:55     ` Nitin Gupta
2010-08-09 17:26 ` [PATCH 06/10] Block discard support Nitin Gupta
2010-08-09 19:03   ` Pekka Enberg
2010-08-10  2:23     ` Jens Axboe
2010-08-10  4:54       ` Nitin Gupta
2010-08-10 15:54         ` Jens Axboe
2010-08-09 17:26 ` [PATCH 07/10] Increase compressed page size threshold Nitin Gupta
2010-08-09 18:32   ` Pekka Enberg
2010-08-09 17:26 ` [PATCH 08/10] Some cleanups Nitin Gupta
2010-08-09 19:02   ` Pekka Enberg
2010-08-09 17:26 ` [PATCH 09/10] Update zram documentation Nitin Gupta
2010-08-09 17:26 ` [PATCH 10/10] Document sysfs entries Nitin Gupta
2010-08-09 19:02   ` Pekka Enberg
2010-08-31 22:37 ` [PATCH 00/10] zram: various improvements and cleanups Greg KH
2010-09-01  3:32 ` Anton Blanchard
2010-09-09 17:24 ` OOM panics with zram Dave Hansen
2010-09-09 19:07   ` [patch -rc] oom: always return a badness score of non-zero for eligible tasks David Rientjes
2010-09-09 19:48     ` Dave Hansen
2010-09-09 21:00       ` David Rientjes
2010-09-09 21:10         ` Dave Hansen
2010-09-09 21:40           ` David Rientjes
2010-10-03 18:41   ` OOM panics with zram Nitin Gupta
2010-10-03 19:27     ` Dave Hansen
2010-10-03 19:40       ` Nitin Gupta
2010-10-04 11:08         ` Ed Tomlinson
2010-10-05 23:43     ` Greg KH
2010-10-06  2:29       ` Nitin Gupta
2010-10-06  2:36         ` Greg KH
2010-10-06  4:30           ` Nitin Gupta
2010-10-06  7:38             ` Pekka Enberg
2010-10-06 14:03               ` Greg KH
2010-10-06 14:16                 ` Pekka Enberg
2010-10-06 14:53                   ` Nitin Gupta
2010-10-06 14:02             ` Greg KH

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=alpine.DEB.2.00.1008311635100.867@router.home \
    --to=cl@linux.com \
    --cc=akpm@linux-foundation.org \
    --cc=devel@driverdev.osuosl.org \
    --cc=eric.dumazet@gmail.com \
    --cc=greg@kroah.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=minchan.kim@gmail.com \
    --cc=ngupta@vflare.org \
    --cc=penberg@cs.helsinki.fi \
    /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