Messages in this thread | | | From | "Rafael J. Wysocki" <> | Subject | Re: [Bug #12667] Badness at kernel/time/timekeeping.c:98 in pmud (timekeeping_suspended) | Date | Thu, 19 Feb 2009 22:23:04 +0100 |
| |
On Thursday 19 February 2009, Thomas Gleixner wrote: > On Thu, 19 Feb 2009, Benjamin Herrenschmidt wrote: > > > On Thu, 2009-02-19 at 21:27 +1300, Paul Collins wrote: > > > > Just for laughs I slapped together the following, which seems to do > > > the > > > > job, although not especially tidily. > > > > > > And it doesn't even do the job. Judging by this new trace, submitting > > > input events from the via-pmu resume function is still too early. > > > > > What's up Thomas ? We can't call gettimeofday() from a sysdev > > suspend/resume ? That's a little bit too harsh no ? > > Well, harsh or not is not the question here. > > Fact is that you call gettimeofday() _before_ the timekeeping code has > resumed. > > That's a simple ordering problem. timekeeping is in the sysdev class > as well and it's not the only sysdev which has explicit ordering > requirements.
Do we need suspend-resume priorities for sysdevs? Such that sysdevs with a higher priority will always be suspended earlier and resumed later than sysdevs with lower priority (or the other way around)?
Rafael
| |