This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: Unified tracing buffer
On Tue, 23 Sep 2008, Martin Bligh wrote:
> On Tue, Sep 23, 2008 at 12:36 PM, Thomas Gleixner <tglx@linutronix.de> wrote:
> > On Tue, 23 Sep 2008, Masami Hiramatsu wrote:
> >> # cat /sys/devices/system/cpu/cpu0/cpuidle/state3/usage
> >> 171210
> >
> > C3 stops the TSC. So depending on how many C3 entries you have on the
> > different cores, your TSCs will drift apart. Some BIOSes do even a
> > lousy job trying to fixup the TSCs on exit from C3, which makes things
> > even worse.
> >
> >> C1: type[C1] promotion[--] demotion[--] latency[001] usage[00000016] duration[00000000000000000000]
> >> C2: type[C2] promotion[--] demotion[--] latency[001] usage[00037969] duration[00000000000024288003]
> >> C3: type[C3] promotion[--] demotion[--] latency[057] usage[00171818] duration[00000000001881257636]
> >>
> >> Could these help you?
> >
> > Yup, explains your TSC observation. Nothing we can do about. Broken by
> > system design :( Welcome in the wonderful world of Inhell/BIOS/ACPI !
>
> We have linux patches that sync the TSC on exit_idle. I'll see if I can get
> Michael to send them out.
Are you sure that they sync it precicely enough that there is no user
space observable way of time going backwards between cores ?
Thanks,
tglx