Messages in this thread | | | Date | Thu, 3 Jun 2010 14:42:21 +0200 | From | Frederic Weisbecker <> | Subject | Re: [GIT PULL] perf crash fix |
| |
On Thu, Jun 03, 2010 at 02:40:31PM +0200, Peter Zijlstra wrote: > On Thu, 2010-06-03 at 14:35 +0200, Frederic Weisbecker wrote: > > > I wanted to, but I thought we could avoid two indirect calls on each > > ticks and I was also afraid of breaking start/stop original semantics, > > more especially the role of perf_event_stop/start > > Right, so you lost the fallback to ->enable/->disable, which would break > hardware PMU implementations that didn't actually implement > ->start/->stop (pretty much all of them except x86).
Indeed.
| |