Messages in this thread |  | | | Date | Wed, 7 May 2008 18:20:12 +0200 | | From | Ingo Molnar <> | | Subject | Re: AIM7 40% regression with 2.6.26-rc1 | |
* Linus Torvalds <torvalds@linux-foundation.org> wrote:
> I think turning the BKL into a semaphore was fine per se, but that was
> when semaphores were fast.
hm, do we know it for a fact that the 40% AIM regression is due to the
fastpath overhead of the BKL? It would be extraordinary if so.
I think it is far more likely that it's due to the different scheduling
and wakeup behavior of the new kernel/semaphore.c code. So the fix would
be to restore the old scheduling behavior - that's what Yanmin's manual
revert did and that's what got him back the previous AIM7 performance.
Ingo
|  |