Messages in this thread | | | Subject | Re: REGRESSION: Performance regressions from switching anon_vma->lock to mutex | From | Peter Zijlstra <> | Date | Wed, 15 Jun 2011 14:49:36 +0200 |
| |
On Wed, 2011-06-15 at 13:52 +0200, Peter Zijlstra wrote: > /me goes re-test without the sirq-rcu bits mixed in just to make sure.
I switched from PREEMPT=n to PREEMPT_VOLUNTARY=y, which seemed to make a difference:
.39 257.651 messages/sec/core -tip 254.976 messages/sec/core +linus 258.03 messages/sec/core +sirq 265.951 messages/sec/core
| |