Messages in this thread | | | Date | Sun, 26 May 2013 15:59:42 -0700 | Subject | Re: [PATCH 0/4] ipc/sem.c: Bug fixes, regression fixes, v3 | From | Linus Torvalds <> |
| |
On Sun, May 26, 2013 at 1:50 PM, Davidlohr Bueso <davidlohr.bueso@hp.com> wrote: > > In lack of getting my swingbench DSS environment back, I ran these > changes against the semop-multi program on my laptop. For 256 threads, > with Manfred's patchset the ops/sec suffers around -7.3%.
Hmm. That test program only tests simple ops, right? So the slowdown comes from either the code simply being that much slower due to extra complexity, or from the fair wakeups..
Anyway, I applied 1/4, we can continue to discuss the rest.
Linus
| |