linux-kernel - Re: TSC problems with Acer TravelMate 5735Z

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: <20200528135659.GE706495@hirez.programming.kicks-ass.net>
Date:   Thu, 28 May 2020 15:56:59 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     Paul Menzel <pmenzel@...gen.mpg.de>, x86@...nel.org,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: TSC problems with Acer TravelMate 5735Z

On Thu, May 28, 2020 at 02:06:17PM +0200, Thomas Gleixner wrote:
> Peter Zijlstra <peterz@...radead.org> writes:
> > On Sun, May 24, 2020 at 07:09:18PM +0200, Paul Menzel wrote:
> >> Dear Linux folks,
> >> 
> >> 
> >> I got my hands on an old Acer TravelMate 5735Z (Intel GM45/Cantiga) and
> >> installed Debian Sid/unstable with Linux 5.6.7 on it.
> >> 
> >> Booting the system it takes a long time, and the systemd units fail to
> >> start. The logs contain that the TSC is unstable. Adding `tsc=unstable` to
> >> the Linux kernel command line fixes this.
> >
> > It fixes nothing; it just doesn't get you the warning because you told
> > it upfront.
> 
> Well, it prevents the kernel to use TSC upfront and not wait until it's
> discovered to be crap. That might make a difference.

IIRC my Core2 machines bail at boot with the message that TSC stops in
idle. Which is pretty deterministic and avoids userspace wreckage after
the fact, like with the BIOS 'goodness'.