Corey,

I looked at the patch, and here is a recommendation:

+static struct hrtimer pmc5_6_update[NR_CPUS];

Use per-cpu variable instead of this. It is more the Linux-way of things
and it is also more efficient as all hrtimers will be in a different cachelines.

static DEFINE_PER_CPU(struct hrtimer, pmc5_6_update);

then use __get_cpu_var(pmc5_6_update) to get/set value.


On Tue, Jan 6, 2009 at 1:59 AM, Corey Ashford
<cjash...@linux.vnet.ibm.com> wrote:
> Hello,
>
> Last month, I got an email from Paul Mackerras (the maintainer of the Power
> arch for Linux) saying that the timer used for sampling the free-running PMU
> counters 5 & 6 was causing a problem with spurious kernel hangs.  Paul said
> that we ought to switch to using high resolution timers instead, since they
> avoid deadlock problems when the handlers are executed in an interrupts-off
> context.
>
> I have created and tried this change and it appears to work correctly. Can
> you please review it and let me know if there any issues with it? It patches
> just a single source file (the one for power6).
>
> Thanks for your consideration,
>
> - Corey
>
> Corey Ashford
> Software Engineer
> IBM Linux Technology Center, Linux Toolchain
> Beaverton, OR
> 503-578-3507
> cjash...@us.ibm.com
>
> ------------------------------------------------------------------------------
>
> _______________________________________________
> perfmon2-devel mailing list
> perfmon2-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/perfmon2-devel
>
>

------------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
_______________________________________________
perfmon2-devel mailing list
perfmon2-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/perfmon2-devel

Reply via email to