From: Tejun Heo <tj@kernel.org>
To: Christoph Lameter <cl@linux.com>
Cc: akpm@linux-foundation.org, Pekka Enberg <penberg@cs.helsinki.fi>,
linux-kernel@vger.kernel.org,
Eric Dumazet <eric.dumazet@gmail.com>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Subject: Re: [thiscpuops upgrade 05/10] x86: Use this_cpu_inc_return for nmi counter
Date: Fri, 26 Nov 2010 18:05:58 +0100 [thread overview]
Message-ID: <4CEFE8F6.5050109@kernel.org> (raw)
In-Reply-To: <alpine.DEB.2.00.1011261047460.13524@router.home>
On 11/26/2010 06:02 PM, Christoph Lameter wrote:
> On Fri, 26 Nov 2010, Tejun Heo wrote:
>
>>> - __this_cpu_inc(alert_counter);
>>> - if (__this_cpu_read(alert_counter) == 5 * nmi_hz)
>>> + if (__this_cpu_inc_return(alert_counter) == 5 * nmi_hz)
>>
>> Hmmm... one worry I have is that xadd, being not a very popular
>> operation, might be slower than add and read. Using it for atomicity
>> would probably be beneficial in most cases but have you checked this
>> actually is cheaper?
>
> XADD takes 3 uops. INC 1 and MOV 1 uop. So there is an additiona uop.
>
> However, a memory fetch from l1 takes a mininum 4 cycles. Doing that twice
> already ends up with at least 8 cycles.
Thanks for the explanation. It might be beneficial to note
performance characteristics on top of the x86 implementation?
Anyways, for this and the following simple conversion patches.
Reviewed-by: Tejun Heo <tj@kernel.org>
--
tejun
next prev parent reply other threads:[~2010-11-26 17:06 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-23 23:51 [thiscpuops upgrade 00/10] Upgrade of this_cpu_ops Christoph Lameter
2010-11-23 23:51 ` [thiscpuops upgrade 01/10] percpucounter: Optimize __percpu_counter_add a bit through the use of this_cpu() options Christoph Lameter
2010-11-24 7:07 ` Pekka Enberg
2010-11-26 15:43 ` Tejun Heo
2010-11-23 23:51 ` [thiscpuops upgrade 02/10] vmstat: Optimize zone counter modifications through the use of this cpu operations Christoph Lameter
2010-11-26 16:25 ` Tejun Heo
2010-11-23 23:51 ` [thiscpuops upgrade 03/10] percpu: Generic support for this_cpu_add,sub,dec,inc_return Christoph Lameter
2010-11-26 16:31 ` Tejun Heo
2010-11-26 16:37 ` Christoph Lameter
2010-11-26 16:39 ` Tejun Heo
2010-11-23 23:51 ` [thiscpuops upgrade 04/10] x86: Support " Christoph Lameter
2010-11-26 16:33 ` Tejun Heo
2010-11-23 23:51 ` [thiscpuops upgrade 05/10] x86: Use this_cpu_inc_return for nmi counter Christoph Lameter
2010-11-26 16:35 ` Tejun Heo
2010-11-26 17:02 ` Christoph Lameter
2010-11-26 17:05 ` Tejun Heo [this message]
2010-11-23 23:51 ` [thiscpuops upgrade 06/10] vmstat: Use this_cpu_inc_return for vm statistics Christoph Lameter
2010-11-23 23:51 ` [thiscpuops upgrade 07/10] highmem: Use this_cpu_xx_return() operations Christoph Lameter
2010-11-23 23:51 ` [thiscpuops upgrade 08/10] percpu: generic this_cpu_cmpxchg() and this_cpu_cmpxchg_double support Christoph Lameter
2010-11-26 16:51 ` Tejun Heo
2010-11-26 16:56 ` Eric Dumazet
2010-11-26 16:58 ` Tejun Heo
2010-11-26 17:01 ` Eric Dumazet
2010-11-26 17:07 ` Tejun Heo
2010-11-26 17:16 ` Eric Dumazet
2010-11-23 23:51 ` [thiscpuops upgrade 09/10] x86: this_cpu_cmpxchg and this_cpu_cmpxchg_double operations Christoph Lameter
2010-11-24 0:41 ` Eric Dumazet
2010-11-24 3:11 ` Christoph Lameter
2010-11-24 7:05 ` Pekka Enberg
2010-11-24 0:44 ` Mathieu Desnoyers
2010-11-23 23:51 ` [thiscpuops upgrade 10/10] Lockless (and preemptless) fastpaths for slub Christoph Lameter
2010-11-24 0:22 ` Eric Dumazet
2010-11-24 3:13 ` Christoph Lameter
2010-11-24 4:37 ` Christoph Lameter
2010-11-24 1:02 ` Mathieu Desnoyers
2010-11-24 1:05 ` Mathieu Desnoyers
2010-11-24 3:09 ` Christoph Lameter
2010-11-24 7:16 ` Pekka Enberg
2010-11-24 16:17 ` Christoph Lameter
2010-11-24 16:37 ` Pekka Enberg
2010-11-24 16:45 ` Christoph Lameter
2010-11-24 16:47 ` Pekka Enberg
2010-11-24 16:55 ` Christoph Lameter
2010-11-24 19:37 ` Jeremy Fitzhardinge
2010-11-24 19:53 ` Christoph Lameter
2010-11-24 20:01 ` Jeremy Fitzhardinge
2010-11-24 19:56 ` Mathieu Desnoyers
2010-11-24 8:15 ` Peter Zijlstra
2010-11-24 16:14 ` Christoph Lameter
2010-11-24 17:26 ` Peter Zijlstra
2010-11-24 18:08 ` Christoph Lameter
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=4CEFE8F6.5050109@kernel.org \
--to=tj@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=cl@linux.com \
--cc=eric.dumazet@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--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 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.