linux-kernel - Re: [Bisected] Regression: Hang on boot in schedule_timeout

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

Message-ID: <20080728141533.GA3697@redhat.com>
Date:	Mon, 28 Jul 2008 11:15:33 -0300
From:	Glauber Costa <gcosta@...hat.com>
To:	Andrew Drake <zappacky.lists@...il.com>
Cc:	Glauber Costa <glommer@...il.com>, linux-kernel@...r.kernel.org
Subject: Re: [Bisected] Regression: Hang on boot in
	schedule_timeout_interruptible during ACPI init on SMP

On Sat, Jul 26, 2008 at 02:53:11PM -0400, Andrew Drake wrote:
> I just discovered; the nolapic_timer parameter causes the problem to go 
> away entirely. It doesn't seem to have any nasty side effects, so I'm 
> content to use it. Looking at the dmesg logs, however, I notice that the 
> "last-good" kernel doesn't even try to use the LAPIC timer, but the 
> "first-bad" does. I wasn't able to track down any discernible reason.
fine, but if you can still help us on this, it would be awesome ;-)

Actually, there is another nasty bug that depends, at least by a first
impression, on the lapic timer. I'm planning to save a time today to
look further on your problem.
>
> On a side note, that's the first I've ever heard/seen of that parameter, 
> I tried it because I noticed the problems started when the LAPIC timer 
> was successfully set up. I doubt the average user would notice this (you 
> can't even tell unless you pass apic=debug), so I doubt many others would 
> discover this fix/workaround.
>
> The question remains, though, why did it go from "don't try" to "try"?

Probably because the order of initialization of smp entities changed a
bit during this series. Probably the lapic timer is now initialized a
little bit earlier, before we have the chance to mark it as bad.

If you can send me at least your dmesgs and cpuinfo, it would be really really
awesome.

> Thanks,
>
> Andrew
>
--
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/