linux-kernel - Re: [1/3] Bugfix: Don't use the TSC in sched

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: <200703041537.27581.ak@suse.de>
Date:	Sun, 4 Mar 2007 15:37:27 +0100
From:	Andi Kleen <ak@...e.de>
To:	"Guillaume Chazarain" <guichaz@...il.com>
Cc:	"Ingo Molnar" <mingo@...e.hu>, linux-kernel@...r.kernel.org
Subject: Re: [1/3] Bugfix: Don't use the TSC in sched_clock if unstable

On Saturday 03 March 2007 22:41, Guillaume Chazarain wrote:
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=f9690982b8c2f9a2c65acdc113e758ec356676a3
> caused a regression by letting sched_clock use the TSC even when cpufreq
> disabled it. This caused scheduling weirdnesses.

What weirdnesses? Did you actually see any?  Hard data please.

I don't like this because right now it will effectively disable TSC use
on anything with pstate invariant TSC and using cpufreq: that is all
modern AMD systems. This is IMHO a serious bug in the timer subsystem which
I intend to fix.

Also I don't see where your instability should come from.  AFAIK 
there are only minor hickups in the result which shouldn't be a problem IMHO.

-Andi
-
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/