Hi Stephane, I was not aware that interrupts are disabled during the call the pmd_sread call, thanks for catching that. Is that also the case for the enable_counters and disable_counters calls as well?
Regards, - Corey Corey Ashford Software Engineer IBM Linux Technology Center, Linux Toolchain Beaverton, OR 503-578-3507 [EMAIL PROTECTED] "stephane eranian" <[EMAIL PROTECTED]> 07/15/2008 03:24 AM Please respond to [EMAIL PROTECTED] To Corey J Ashford/Beaverton/[EMAIL PROTECTED] cc [email protected] Subject Re: [PATCH 5/5] fixes for full perfmon2 on POWER Corey, I don't understand this patch. You say: "This patch fixes a problem with accessing the array of cpu-specific PMU counters, where there was a race condition that could occur where we acquire the cpu number, and then have the cpu switch out from underneath us. Using get_cpu()/put_cpu() fixes this problem because get_cpu() disables preemption until put_cpu() is called again." What do you mean by "switch from underneath". AFAIK, pmd_sread() is called with interrupts masked. I thought, this was enough to prevent any preemption? On Mon, Jul 14, 2008 at 11:50 PM, Corey Ashford <[EMAIL PROTECTED]> wrote: > This fixes an issue on POWER4 and POWER6 where PMU exceptions need to be > disabled when the context is in masked mode. > > -- > Corey Ashford > Software Engineer > IBM Linux Technology Center, Linux Toolchain > Beaverton, OR > 503-578-3507 > [EMAIL PROTECTED] > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ perfmon2-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/perfmon2-devel
