From: "Björn Steinbrink" <B.Steinbrink@gmx.de>
To: Mike Galbraith <efault@gmx.de>
Cc: Arjan van de Ven <arjan@infradead.org>,
linux-kernel@vger.kernel.org, danial_thom@yahoo.com,
Linus Torvalds <torvalds@osdl.org>, Andrew Morton <akpm@osdl.org>
Subject: Re: [PATCH] i386: Fix softirq accounting with 4K stacks
Date: Mon, 26 Jun 2006 19:58:44 +0200 [thread overview]
Message-ID: <20060626175844.GA3822@atjola.homenet> (raw)
In-Reply-To: <1151291114.7611.8.camel@Homer.TheSimpsons.net>
On 2006.06.26 05:05:14 +0200, Mike Galbraith wrote:
> On Mon, 2006-06-26 at 04:23 +0200, Mike Galbraith wrote:
>
> > Something is certainly still b0rken. I still get three different
> > answers to the question "what is my cpu usage" depending on
> > configuration. With stock UP kernel with no IO-APIC, interrupt load is
> > all hi. With your patch and IO-APIC, it's all si. SMP shows a mix of
> > both.
>
> I didn't say that quite right.
>
> The third case for my box is XT-PIC and your patch. Previously, top
> showed the 10% interrupt load of a flood ping as all hi. Now it's both
> hi and si when using XT-PIC, but it's no longer the 10% that agreed with
> the profile, it's 10% hi, but with an added ~7% si.
I see the following here, when a box is being ping flooded:
UP/SMP stack-size PIC-type hi si
-----------------------------------
UP 8K IO-APIC 0 16
UP 8K XT-PIC 7 7
UP 4K IO-APIC 0 16
UP 4K XT-PIC 7 7
SMP 8K IO-APIC 5 4 (forcedeth)
SMP 8K IO-APIC 0 11 (tg3)
The UP system is a Thinkpad which only has a tg3 driven NIC. The SMP
system is x86_64, so there's no 4K stacks to test with, maybe I'll fetch
a i386 live CD to do some more valid tests on SMP.
The UP tests seem to show the IO-APIC hardirq are completely deferred
to be handled in a softirq, as the sum of "hi" and "si" with XT-PIC is
about equal to the "si" value with IO-APIC (the numbers are guessed
averages of the observed values, so the 2% difference is not too be
taken too serious).
The results for the forcedeth driven NIC do not agree though, and your
results differ from what I see as well, right? So I'm kinda lost again.
Björn
next prev parent reply other threads:[~2006-06-26 17:59 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-06-22 15:26 Measuring tools - top and interrupts Danial Thom
2006-06-22 16:19 ` Randy.Dunlap
2006-06-22 17:16 ` Bernd Eckenfels
2006-06-22 16:21 ` Erik Mouw
2006-06-22 16:58 ` Danial Thom
2006-06-22 17:31 ` Erik Mouw
2006-06-22 23:37 ` Danial Thom
2006-06-23 8:32 ` Mike Galbraith
2006-06-23 20:14 ` Danial Thom
2006-06-23 22:51 ` Mike Galbraith
2006-06-23 9:02 ` Erik Mouw
2006-06-23 17:13 ` Alistair John Strachan
2006-06-23 18:16 ` Danial Thom
2006-06-23 5:34 ` sena seneviratne
2006-06-23 20:42 ` Danial Thom
2006-06-22 17:57 ` Francois Romieu
2006-06-22 22:47 ` Danial Thom
2006-06-22 23:53 ` Francois Romieu
2006-06-23 20:34 ` Danial Thom
2006-06-23 21:19 ` Francois Romieu
2006-06-24 2:07 ` Björn Steinbrink
2006-06-24 4:39 ` sena seneviratne
2006-06-24 5:59 ` Mike Galbraith
2006-06-24 6:26 ` Mike Galbraith
2006-06-24 9:21 ` Björn Steinbrink
2006-06-24 9:51 ` Mike Galbraith
2006-06-24 11:41 ` Mike Galbraith
2006-06-24 15:40 ` Björn Steinbrink
2006-06-24 16:23 ` Mike Galbraith
2006-06-24 19:25 ` Björn Steinbrink
2006-06-25 5:06 ` Mike Galbraith
2006-06-25 11:12 ` Björn Steinbrink
2006-06-25 14:24 ` [PATCH] i386: Fix softirq accounting with 4K stacks Björn Steinbrink
2006-06-25 15:15 ` Arjan van de Ven
2006-06-25 17:44 ` Mike Galbraith
2006-06-25 17:43 ` Arjan van de Ven
2006-06-25 18:42 ` Björn Steinbrink
2006-06-26 2:23 ` Mike Galbraith
2006-06-26 3:05 ` Mike Galbraith
2006-06-26 17:58 ` Björn Steinbrink [this message]
2006-06-27 10:09 ` Mike Galbraith
2006-06-25 20:45 ` Measuring tools - top and interrupts Danial Thom
2006-06-26 2:33 ` Mike Galbraith
2006-06-30 13:41 ` Danial Thom
2006-06-30 18:03 ` Mike Galbraith
2006-06-26 22:54 [PATCH] i386: Fix softirq accounting with 4K stacks Chuck Ebbert
2006-06-27 0:57 ` Björn Steinbrink
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=20060626175844.GA3822@atjola.homenet \
--to=b.steinbrink@gmx.de \
--cc=akpm@osdl.org \
--cc=arjan@infradead.org \
--cc=danial_thom@yahoo.com \
--cc=efault@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=torvalds@osdl.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).