linux-kernel - RE: Possible CPU_IDLE bug [WAS: Re: Timer unstability on when using C2 and deeper sleep states (Dell Latitude XT)]

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <1218800811.4388.23.camel@localhost>
Date:	Fri, 15 Aug 2008 13:46:51 +0200
From:	Milan Plzik <milan.plzik@...il.com>
To:	"Pallipadi, Venkatesh" <venkatesh.pallipadi@...el.com>
Cc:	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"tglx@...utronix.de" <tglx@...utronix.de>
Subject: RE: Possible CPU_IDLE bug [WAS: Re: Timer unstability on
	when	using C2	and deeper sleep states (Dell Latitude XT)]

On St, 2008-08-13 at 14:22 -0700, Pallipadi, Venkatesh wrote:
> 
> >-----Original Message-----
> >From: Milan Plzik [mailto:milan.plzik@...il.com]
> >Sent: Wednesday, August 13, 2008 1:21 PM
> >To: Pallipadi, Venkatesh
> >Cc: linux-kernel@...r.kernel.org
> >Subject: RE: Possible CPU_IDLE bug [WAS: Re: Timer unstability
> >on when using C2 and deeper sleep states (Dell Latitude XT)]
> >

[...]

> Were all these tests with 2.6.26? Can you try with 2.6.27-rc3?
> 
> There is one bugfix patch that, IIRC, went in after 2.6.26.
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=b8f8c3cf0a4ac0632ec3f0e15e9dc0c29de917af

  I did some further testing on 2.6.27-rc3 today; unfortunately I found
out that the 'press key or I won't move' issue is still present,
although I have not noticed it first time.

  I tried to switch cpuidle governors on the fly and found out one
interesting thing -- if powertop shows ~ 10000 wakeups/sec and I just
play with the governors (I just echo "menu" or "ladder"
> /sys/devices/system/cpu/cpuidle/current_governor), with non-zero
probability (usually second attempt made it) the number of wakeups/sec
changes to more reasonable value, roughly 700. But this happened even if
I echoed the same name as the current governor is, which is a bit
strange, as this is (according to sources) basically NOOP -- it just
locks and unlocks the cpuidle_lock mutex.

  Interesting also is, that no matter in what state I was (menu governor
giving 10000 or 700), the 'ladder' governor usually yielded roughly 4000
wakeups per second. And, after starting and stopping X server (which
from some weird reason blocks C3 state even after terminating) I was
getting numbers below 300 wakeups/sec.

  Moreover, I previously ignored one fact, which might (or might not) be
related. Sometimes (quite rarely) the system crashes early on boot with
following message ("screenshot"):

http://image.dayphorum.eu/show/img-mhPMoxVfDZ.jpg

  I have no idea whether this is somehow related or not, anyway, the
boot doesn't continue. Unfortunately, VESA modes don't seem to work
properly (black screen), so I'm not able to get more than is visible on
the photo. Screenshot is from 2.6.26 kernel, but it has also happened on
2.6.27-rc3.

> 
> Thanks,
> Venki

  Thank you,
	Milan

--
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/