Corey,
Patch applied to both perfmon2/perfmon3 Thanks for you perseverance. On Wed, Jan 21, 2009 at 12:01 AM, Corey Ashford <cjash...@linux.vnet.ibm.com> wrote: > Hi Stephane, > > Here's version 4 of the patch. It contains two changes: > > 1) The optimization to get rid of the per-cpu ctx_arch variable. To create > this patch, I pulled down your latest git tree, and this time when I added > the __get_cpu_var(pmu_ctx) call, I did not get an error because of the lack > of EXPORT_PER_CPU_SYMBOL. Did you change something? Anyway, it appears to > work correctly without any the addition of the macro. > > 2) I changed the callback mode of the hrtimer from > HRTIMER_CB_IRQSAFE_UNLOCKED to HRTIMER_CB_IRQSAFE_PERCPU because we need to > be guaranteed that the callback runs on the same CPU where it it was > started. From my reading of the description in hrtimers.h, the > HRTIMER_CB_IRQSAFE_*UNLOCKED* mode is not needed for perfmon_power6's use of > the timer. > > Please let me know if there are issues with this patch, and thanks for your > consideration. > > Regards, > > - Corey > > Corey Ashford > Software Engineer > IBM Linux Technology Center, Linux Toolchain > Beaverton, OR > 503-578-3507 > cjash...@us.ibm.com > ------------------------------------------------------------------------------ This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword _______________________________________________ perfmon2-devel mailing list perfmon2-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/perfmon2-devel
