From: Ingo Molnar <mingo@elte.hu>
To: Mike Kravetz <kravetz@us.ibm.com>
Cc: linux-kernel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>,
John Stultz <johnstul@us.ibm.com>,
"Paul E. McKenney" <paulmck@us.ibm.com>,
Dipankar Sarma <dipankar@in.ibm.com>,
Arjan van de Ven <arjan@infradead.org>
Subject: Re: 2.6.18-rt1
Date: Wed, 27 Sep 2006 10:34:19 +0200 [thread overview]
Message-ID: <20060927083419.GC12149@elte.hu> (raw)
In-Reply-To: <20060925161215.GA19230@w-mikek2.ibm.com>
* Mike Kravetz <kravetz@us.ibm.com> wrote:
> On Wed, Sep 20, 2006 at 04:19:07PM +0200, Ingo Molnar wrote:
> > In particular, a nasty softirq performance bug has been fixed, which
> > caused the "5x slowdown under TCP" bug reported to lkml - those TCP
> > performance figures are now on par with vanilla performance.
>
> Just curious about the cause and fix for this issue. I tried
> searching the mail lists for discussion but came up empty. The patch
> it too big for me to determine what specific changes addressed this
> issue. If anyone can point me in the right direction, that would be
> appreciated.
i /think/ it's this bit of code commented out in kernel/softirq.c:
//if (!in_interrupt() && local_softirq_pending())
// invoke_softirq();
try to uncomment that. (this should i think do the trick on PREEMPT_RT -
but no guarantees. It might cause problems on !PREEMPT_RT configs, i had
to juggle around some code here. Maybe the easiest would be if you tried
to take the new softirq.c, sans the lockdep changes. Not sure how
feasible that is though.)
Ingo
next prev parent reply other threads:[~2006-09-27 8:42 UTC|newest]
Thread overview: 101+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-09-20 14:19 2.6.18-rt1 Ingo Molnar
2006-09-20 16:50 ` 2.6.18-rt1 Gene Heskett
2006-09-20 16:58 ` 2.6.18-rt1 Ingo Molnar
2006-09-20 17:33 ` 2.6.18-rt1 Gene Heskett
2006-09-20 18:34 ` 2.6.18-rt1 Gene Heskett
2006-09-20 17:00 ` 2.6.18-rt1 Daniel Walker
2006-09-20 17:38 ` 2.6.18-rt1 Paul E. McKenney
2006-09-20 17:41 ` 2.6.18-rt1 Daniel Walker
2006-09-20 18:23 ` 2.6.18-rt1 Gene Heskett
2006-09-20 18:25 ` 2.6.18-rt1 Paul E. McKenney
2006-09-20 18:34 ` 2.6.18-rt1 Daniel Walker
2006-09-20 20:06 ` 2.6.18-rt1 Paul E. McKenney
2006-09-20 21:38 ` 2.6.18-rt1 Gene Heskett
2006-09-20 20:17 ` 2.6.18-rt1 Gene Heskett
2006-09-20 18:36 ` 2.6.18-rt1 Gene Heskett
2006-09-20 18:47 ` 2.6.18-rt1 Thomas Gleixner
2006-09-20 19:20 ` 2.6.18-rt1 Thomas Gleixner
2006-09-20 19:46 ` 2.6.18-rt1 Ingo Molnar
2006-09-20 20:19 ` 2.6.18-rt1 Daniel Walker
2006-09-20 20:14 ` 2.6.18-rt1 Ingo Molnar
2006-09-20 20:31 ` 2.6.18-rt1 Daniel Walker
2006-09-21 19:02 ` 2.6.18-rt1 Ingo Molnar
2006-09-21 19:18 ` 2.6.18-rt1 Daniel Walker
2006-09-22 14:42 ` 2.6.18-rt1 Daniel Walker
2006-09-27 8:36 ` 2.6.18-rt1 Ingo Molnar
2006-09-21 8:04 ` 2.6.18-rt1 Deepak Saxena
2006-09-21 8:04 ` 2.6.18-rt1 Ingo Molnar
2006-09-21 8:24 ` 2.6.18-rt1 Deepak Saxena
2006-09-22 2:19 ` 2.6.18-rt1 john cooper
2006-09-22 6:36 ` 2.6.18-rt1 Lennert Buytenhek
2006-09-22 11:56 ` 2.6.18-rt1 Ingo Molnar
2006-09-27 13:10 ` 2.6.18-rt4 john cooper
2006-09-27 13:09 ` 2.6.18-rt4 Ingo Molnar
2006-09-20 18:56 ` 2.6.18-rt1 K.R. Foley
2006-09-20 19:49 ` 2.6.18-rt1 Ingo Molnar
2006-09-20 20:33 ` 2.6.18-rt1 K.R. Foley
2006-09-20 20:41 ` 2.6.18-rt1 Thomas Gleixner
2006-09-20 20:50 ` 2.6.18-rt1 K.R. Foley
2006-09-21 19:16 ` 2.6.18-rt1 john stultz
2006-09-22 2:18 ` 2.6.18-rt1 K.R. Foley
2006-09-22 11:58 ` 2.6.18-rt1 Ingo Molnar
2006-09-28 0:42 ` 2.6.18-rt1 john stultz
2006-09-28 22:48 ` 2.6.18-rt1 john stultz
2006-09-29 2:09 ` 2.6.18-rt1 K.R. Foley
2006-09-29 12:24 ` 2.6.18-rt1 Ingo Molnar
2006-09-29 12:40 ` 2.6.18-rt1 Ingo Molnar
2006-09-20 19:58 ` 2.6.18-rt1 Thomas Gleixner
2006-09-20 20:34 ` 2.6.18-rt1 K.R. Foley
2006-09-20 19:38 ` 2.6.18-rt1 Mark Knecht
2006-09-20 20:27 ` 2.6.18-rt1 Mark Knecht
2006-09-22 14:14 ` 2.6.18-rt1 Lee Revell
2006-09-20 20:54 ` 2.6.18-rt1 Michal Piotrowski
2006-09-20 22:07 ` 2.6.18-rt1 Michal Piotrowski
2006-09-20 22:26 ` 2.6.18-rt1 Michal Piotrowski
2006-09-21 6:56 ` [PATCH] move put_task_struct() reaping into a thread [Re: 2.6.18-rt1] Bill Huey
2006-09-21 6:54 ` Ingo Molnar
2006-09-21 7:18 ` Bill Huey
2006-09-21 7:16 ` Ingo Molnar
2006-09-21 7:32 ` Bill Huey
2006-09-21 7:29 ` Ingo Molnar
2006-09-21 7:48 ` Bill Huey
2006-09-21 7:56 ` Ingo Molnar
2006-09-21 8:13 ` Bill Huey
2006-09-21 12:23 ` Esben Nielsen
2006-09-21 12:56 ` Ingo Molnar
2006-09-21 7:27 ` Bill Huey
2006-09-21 7:22 ` Ingo Molnar
2006-09-21 7:35 ` Bill Huey
2006-09-21 7:31 ` Ingo Molnar
2006-09-21 7:52 ` Bill Huey
2006-09-27 2:55 ` Eric W. Biederman
2006-09-27 5:08 ` Bill Huey
2006-09-27 6:02 ` Eric W. Biederman
2006-09-27 6:34 ` Bill Huey
2006-09-27 7:29 ` Eric W. Biederman
2006-09-27 9:01 ` Ingo Molnar
2006-09-27 13:59 ` Eric W. Biederman
2006-09-27 14:06 ` Ingo Molnar
2006-09-27 16:18 ` Paul E. McKenney
2006-09-27 9:08 ` Ingo Molnar
2006-09-27 9:09 ` Bill Huey
2006-09-27 9:05 ` Ingo Molnar
2006-09-27 20:28 ` Esben Nielsen
2006-09-27 8:57 ` Ingo Molnar
2006-09-27 9:14 ` Bill Huey
2006-09-27 9:15 ` Ingo Molnar
2006-09-25 9:53 ` 2.6.18-rt1 Florian Schmidt
2006-09-26 7:57 ` 2.6.18-rt1 Florian Schmidt
2006-09-25 16:12 ` 2.6.18-rt1 Mike Kravetz
2006-09-27 8:34 ` Ingo Molnar [this message]
2006-09-30 18:06 ` 2.6.18-rt1 Lee Revell
2006-09-30 18:18 ` 2.6.18-rt1 Dipankar Sarma
2006-09-30 18:25 ` 2.6.18-rt1 Lee Revell
2006-10-13 21:18 ` 2.6.18-rt1 Karsten Wiese
2006-10-13 21:20 ` 2.6.18-rt1 Lee Revell
2006-10-13 21:24 ` 2.6.18-rt1 Dipankar Sarma
2006-10-13 22:12 ` 2.6.18-rt1 Lee Revell
2006-10-13 22:16 ` 2.6.18-rt1 Dipankar Sarma
2006-10-17 14:46 ` 2.6.18-rt1 Lee Revell
2006-10-18 8:34 ` 2.6.18-rt1 Ingo Molnar
2006-10-18 7:12 ` 2.6.18-rt1 Ingo Molnar
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=20060927083419.GC12149@elte.hu \
--to=mingo@elte.hu \
--cc=arjan@infradead.org \
--cc=dipankar@in.ibm.com \
--cc=johnstul@us.ibm.com \
--cc=kravetz@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=paulmck@us.ibm.com \
--cc=tglx@linutronix.de \
/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