Messages in this thread | | | Date | Tue, 23 Jul 2013 11:51:24 +0200 | From | Peter Zijlstra <> | Subject | Re: Performance regression from switching lock to rw-sem for anon-vma tree |
| |
On Tue, Jul 23, 2013 at 11:45:13AM +0200, Ingo Molnar wrote: > Why not just try the delayed addition approach first? The spinning is time > limited AFAICS, so we don't _have to_ recognize those as writers per se, > only if the spinning fails and it wants to go on the waitlist. Am I > missing something? > > It will change patterns, it might even change the fairness balance - but > is a legit change otherwise, especially if it helps performance.
Be very careful here. Some people (XFS) have very specific needs. Walken and dchinner had a longish discussion on this a while back.
| |