All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Andi Kleen <andi@firstfloor.org>
Cc: Ingo Molnar <mingo@elte.hu>, LKML <linux-kernel@vger.kernel.org>,
	x86@kernel.org, Nick Piggin <nickpiggin@yahoo.com.au>,
	Jens Axboe <jens.axboe@oracle.com>
Subject: Re: [PATCH 0 of 8] x86/smp function calls: convert x86 tlb flushes to use function calls
Date: Mon, 11 Aug 2008 16:57:26 -0700	[thread overview]
Message-ID: <48A0D1E6.5030609@goop.org> (raw)
In-Reply-To: <20080806033736.GB24801@one.firstfloor.org>

Andi Kleen wrote:
> Micro benchmark. Just stick RDTSC around the calls and trigger
> a few remote flushes by e.g. doing mmap/munmap on a multithreaded program
> running on different CPUs.
>   

On a HT 3GHz Xeon, I see a ~35% increase in time (1830ns -> 2500ns), 
averaged over 1000,000 interrupts.  Because it's only 2 cpu, it ends up 
always using the smp_call_function_single path (but via the 
smp_call_function_mask path).

I need to try getting a more-cpu machine running native to do a proper 
test, esp to see if putting kmalloc on that path is very expensive or 
not.  (Is it worth using a kmem_cache?)

    J

      reply	other threads:[~2008-08-11 23:57 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-08-02  0:14 [PATCH 0 of 8] x86/smp function calls: convert x86 tlb flushes to use function calls Jeremy Fitzhardinge
2008-08-02  0:14 ` [PATCH 1 of 8] x86-32: use smp_call_function_mask for SMP TLB invalidations Jeremy Fitzhardinge
2008-08-02  0:14 ` [PATCH 2 of 8] x86-64: " Jeremy Fitzhardinge
2008-08-02  0:14 ` [PATCH 3 of 8] x86: make tlb_32|64 closer Jeremy Fitzhardinge
2008-08-02  0:14 ` [PATCH 4 of 8] x86: unify tlb.c Jeremy Fitzhardinge
2008-08-02  0:14 ` [PATCH 5 of 8] smp_function_call: add multiple queues for scalability Jeremy Fitzhardinge
2008-08-02  0:14 ` [PATCH 6 of 8] x86: add multiple smp_call_function queues Jeremy Fitzhardinge
2008-08-02  0:14 ` [PATCH 7 of 8] x86: make number of smp_call_function queues truely configurable Jeremy Fitzhardinge
2008-08-02  0:14 ` [PATCH 8 of 8] smp function calls: add kernel parameter to disable multiple queues Jeremy Fitzhardinge
2008-08-06  3:22 ` [PATCH 0 of 8] x86/smp function calls: convert x86 tlb flushes to use function calls Andi Kleen
2008-08-06  3:33   ` Jeremy Fitzhardinge
2008-08-06  3:37     ` Andi Kleen
2008-08-11 23:57       ` Jeremy Fitzhardinge [this message]

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=48A0D1E6.5030609@goop.org \
    --to=jeremy@goop.org \
    --cc=andi@firstfloor.org \
    --cc=jens.axboe@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=nickpiggin@yahoo.com.au \
    --cc=x86@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 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.