linux-kernel - Re: constant_tsc and TSC unstable

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]

Date:	Fri, 30 Nov 2007 01:30:22 +0100
From:	Frans Pop <elendil@...net.nl>
To:	Paul Rolland <rol@...be.net> (Witbe.net)
Cc:	linux-kernel@...r.kernel.org, rol@...be.net
Subject: Re: constant_tsc and TSC unstable

Paul Rolland wrote:
> Total of 2 processors activated (6919.15 BogoMIPS).
> ENABLING IO-APIC IRQs
> ..TIMER: vector=0x31 apic1=0 pin1=2 apic2=-1 pin2=-1
> checking TSC synchronization [CPU#0 -> CPU#1]:
> Measured 3978592228 cycles TSC warp between CPUs, turning off TSC clock.
> Marking TSC unstable due to: check_tsc_sync_source failed.
> Brought up 2 CPUs
> ...

Not sure if this is related, but thought I'd contribute it anyway...

I've got a Pentium D system (dual core, single processor) and I on some
boots I get "Marking TSC unstable due to check_tsc_sync_source failed" with
some cycles warp between CPUs, while most boots are OK. This kind of
inconsistency seems more due to a failure in the kernel to deal with
differences between boots than with something inherent to the hardware.

I conclude that because basically I never have any problems with the system
once it has booted and the TSC has passed.

>From my kern.logs since Okt 26, I get the following data:
2.6.23+cfs:  2 passes
2.6.23.1:    1 pass;   1 failure  (48 cycles warp)
2.6.24-rc1: 15 passes
2.6.24-rc2: 13 passes; 1 failure  (8 cycles warp)
2.6.24-rc3:  5 passes; 3 failures (8, 8 and 16 cycles warp)

Note that this is not a new issue. For 2.6.21/2.6.23-RCx kernels I reported
similar data in http://lkml.org/lkml/2007/9/16/45.

Cheers,
FJP
-
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/