Re: [2.6.18-rt7] BUG: time warp detected!

[Mike Galbraith <efault@gmx.de>]

On Fri, 2006-10-27 at 06:08 +0000, Mike Galbraith wrote:
> On Fri, 2006-10-27 at 05:45 +0000, Mike Galbraith wrote:
> > On Thu, 2006-10-26 at 14:25 -0700, john stultz wrote:
> > > On Thu, 2006-10-26 at 07:20 +0000, Mike Galbraith wrote:
> > > > $subject happened on my single P4/HT box sometime after resume from
> > > > disk.  Hohum activity:  I had just read lkml and was retrieving latest
> > > > glibc snapshot when I noticed the trace.  I also noticed that the kernel
> > > > decided to use pit instead of tsc.
> > > 
> > > Huh. Was the PIT selected before or after the resume from disk?
> > 
> > Both.  If I don't specify tsc, it chooses pit.  I just removed freshly
> > added clocksource=tsc, rebooted, and I'm back on pit again.
> 
> (hm. virgin 2.6.18 selects tsc properly... will rummage through rt7)

#include <naughty_language.h>

#ifndef CONFIG_HIGH_RES_TIMERS
		/* lower the rating if we already know its unstable: */
		if (check_tsc_unstable())
			clocksource_tsc.rating = 50;
#else
		/*
		 * Mark TSC unsuitable for high resolution timers. TSC has so
		 * many pitfalls: frequency changes, stop in idle ...  When we
		 * switch to high resolution mode we can not longer detect a
		 * firmware caused frequency change, as the emulated tick uses
		 * TSC as reference. This results in a circular dependency.
		 * Switch only to high resolution mode, if pm_timer or such
		 * is available.
		 */
		clocksource_tsc.rating = 50;
		clocksource_tsc.is_continuous = 0;
#endif