Message-ID: <AANLkTimWFTUZTh2-3Bavmwcum2joQluMEDmXx3xjyHqQ@mail.gmail.com>
Date:	Thu, 15 Jul 2010 23:50:09 +0300
From:	Pekka Enberg <penberg@...helsinki.fi>
To:	Zeno Davatz <zdavatz@...il.com>
Cc:	Damien Wyart <damien.wyart@...e.fr>,
	Catalin Marinas <catalin.marinas@....com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	"x86@...nel.org" <x86@...nel.org>, "mingo@...e.hu" <mingo@...e.hu>,
	"yinghai@...nel.org" <yinghai@...nel.org>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>
Subject: Re: kmemleak, cpu usage jump out of nowhere

On Thu, Jul 15, 2010 at 11:38 PM, Zeno Davatz <zdavatz@...il.com> wrote:
> Am 15.07.2010 um 22:00 schrieb Damien Wyart <damien.wyart@...e.fr>:
>
>>>> For now, I can't reproduce the problem with CONFIG_NO_BOOTMEM disabled ;
>>>> with the option and rc5 the problem was happening quite quickly after
>>>> boot and normal use of the machine. So it seems I can confirme what Zeno
>>>> has seen and I hope this will give a hint to debug the problem. I guess
>>>> this has not been reported that much because many testers might not have
>>>> enabled CONFIG_NO_BOOTMEM... Maybe the scheduler folks could test their
>>>> benchmark with a kernel having this option enabled?
>>
>> * Pekka Enberg <penberg@...helsinki.fi> [2010-07-15 22:50]:
>>> To be honest, the bug is bit odd. It's related to boot-time memory
>>> allocator changes but yet it seems to manifest itself as a scheduling
>>> problem. So if you have some spare time and want to speed up the
>>> debugging process, please test v2.6.34 and v2.6.35-rc1 with
>>> CONFIG_NO_BOOTMEM and if former is good and latter is bad, try to see
>>> if you can identify the offending commit with "git bisect."
>>
>> Not sure I will have enough time in the coming days (doing that remotely
>> is fishy since ssh access is almost stuck when the problem occurs); if
>> Zeno can and would like to do it, maybe this could be done faster.
>>
>> As the scheduler is now very well instrumented (many debugging features
>> are available), reproducing the bug on a test platform (it happens quite
>> quickly for me) might also give some hints. So testers, if you have
>> time, please test 2.6.35-rc5 with CONFIG_NO_BOOTMEM on a Core i7 and see
>> if you can reproduce the problem!
>
> Will try to do so. Can you point me to the git bisect howto with the versions you want.

Cool. So like I said, you first want to test 2.6.34 to find a known
good version. Please remember to make sure you have CONFIG_NO_BOOTMEM
enabled. You can also try to speed up the process by testing
2.6.35-rc1 which is likely to include the offending commit. That's not
strictly necessary as long as you are sure that you have some
2.6.35-rc kernel that's bad.

After that, bisecting is as simple as:

  git bisect start
  git bisect good v2.6.34
  git bisect bad v2.6.31-rc1 # or some other kernel you know to be bad
  <compile, boot, and try to trigger the problem>

then

  git bisect bad # if you were able to trigger the problem

or

  git bisect good # if the problem doesn't exist

git will then find the next revision to test after which you do

  <compile, boot, and try to trigger the problem>

and repeat the "git bisect good/bad" step until git tells you it has
found the offending commit.

There's more information on the git bisect man pages:

http://www.kernel.org/pub/software/scm/git/docs/git-bisect.html

Let me know if you need more help with this.

                        Pekka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists