Messages in this thread | | | Date | Fri, 24 Nov 2006 21:04:19 +0100 | From | Jens Axboe <> | Subject | Re: [patch] cpufreq: mark cpufreq_tsc() as core_initcall_sync |
| |
On Fri, Nov 24 2006, Oleg Nesterov wrote: > Ok, synchronize_xxx() passed 1 hour rcutorture test on dual P-III. > > It behaves the same as srcu but optimized for writers. The fast path > for synchronize_xxx() is mutex_lock() + atomic_read() + mutex_unlock(). > The slow path is __wait_event(), no polling. However, the reader does > atomic inc/dec on lock/unlock, and the counters are not per-cpu. > > Jens, is it ok for you? Alan, Paul, what is your opinion?
This looks good from my end, much more appropriate than the current SRCU code. Even if I could avoid synchronize_srcu() for most cases, when I did have to issue it, the 3x synchronize_sched() was a performance killer.
Thanks Oleg! And Alan and Paul for your excellent ideas.
-- Jens Axboe
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |