linux-kernel - [2.6.21] BUG: clocksource_watchdog isn't reset after resume

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

Date:	Sun, 29 Apr 2007 18:20:12 +0200 (CEST)
From:	"Indan Zupancic" <indan@....nu>
To:	tglx@...utronix.de, johnstul@...ibm.com,
	linux-kernel@...r.kernel.org
Subject: [2.6.21] BUG: clocksource_watchdog isn't reset after resume

Hello,

After s2ram, basically the following happens, which shouldn't happen:

[    0.880234] Clocksource tsc unstable (delta = 449782067799 ns)
[    0.881221] Time: pit clocksource has been installed.

Looking at the source, it seems that clocksource_watchdog isn't reset
after resume, while the clock source used is, and hence it thinks that
the clock is unstable.

So it appears that timekeeping_resume and/or timekeeping_suspend should
do something to prevent this from happening.

Maybe by somehow updating 'watchdog_last' after resume, but it's unclear
how that should happen, as the watchdog doesn't export any functions and
the only thing that's reachable which calls anything watchdog related is
clocksource_register().

This is a UP machine with no frequency changing CPU, so the tsc is good
to use (before someone complaints that I should use the pm timer).

An alternative would be to not select CONFIG_CLOCKSOURCE_WATCHDOG, but
it's unclear how or by what that's selected.

Greetings,

Indan

-
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/