Corey, Is this on PPC or X86?
On x86 (stop_save), the routine that checks for pending overflows and populate the povfl_pmds bitmask, only looks at the PMDs that are marked as used. So no risk of reporting an interrupt on a PMD not used by the set. Maybe you don't make that check in the PPC code. On Thu, Jul 17, 2008 at 9:17 PM, Corey Ashford <[EMAIL PROTECTED]> wrote: > Hello, > > I found that there's a problem with having counters in a set which are > unused in the current set, but have already overflowed. The overflowed > counter will cause an interrupt which is not recognized by the kernel since > it only looks at the used counters. > > To fix this problem, I propose zeroing out all of the interrupt-capable > counters during pfm_restore_pmds. I have tried out this code and it does > seem to work correctly (no more hangs during the PAPI multiplexing tests). > Attached is my suggested patch. What do you think? > > Thanks for your consideration, > > - Corey > > -- > 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 perfmon2-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/perfmon2-devel
