Messages in this thread | | | Date | Fri, 3 Apr 2009 14:59:03 +0200 | From | Stanislaw Gruszka <> | Subject | Re: [PATCH 0/2] itimers: periodic timers fixes |
| |
On Thu, 2 Apr 2009 18:57:53 +0200 Ingo Molnar <mingo@elte.hu> wrote:
> Converting those to GTOD sampling instead of jiffies sampling is a > worthwile change IMO and a good concept. > > The unificaton of ITIMER_PROF and ITIMER_VIRT is a nice observation > and a good patch. > > The second one, changing all the sampling from cputime to ktime_t is > nicely done too: > > We could do more though, there's still a bit of cputime legacies > around: > > + cputime_t cval, nval; > > Couldnt all of that go over into the ktime_t space as well, phasing > out cputime logic from the itimer code? > > The user ABI is struct timeval based, so there's no need to have > cputime anywhere. The scheduler does nanoseconds accurate stats so > it can be connected up there too.
Removing cputime stuff from itimers has probably only sense when utime, stime and related fields in task_struct would be represented as ktime or u64 variable in nanoseconds accurate. This mean a lot of work. I'm not sure if is worth to do in the meaning that as result we get better (faster and perhaps smaller) code.
I was thinking about removing cputime as whole, make utime and stime 64 bit variables and account them in nanoseconds resolution. Remove sum_exec_runtime from struct task_cputime and related CPUCLOCK_SCHED code as duplicate of nanosecond accounted stime and CPUCLOCK_PROF code. But that were too intrusive changes for me with unknown performance impact.
My primary goal is to improve periodic itimers accuracy (see https://bugzilla.redhat.com/show_bug.cgi?id=441134), these patches are just enough to achieve the goal.
Cheers Stanislaw Gruszka
| |