Messages in this thread | | | Date | Fri, 11 May 2007 17:56:21 +0200 | From | Ingo Molnar <> | Subject | Re: [PATCH 0/2] convert mmap_sem to a scalable rw_mutex |
| |
* Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
> I was toying with a scalable rw_mutex and found that it gives ~10% > reduction in system time on ebizzy runs (without the MADV_FREE patch). > > 2-way x86_64 pentium D box: > > 2.6.21 > > /usr/bin/time ./ebizzy -m -P > 59.49user 137.74system 1:49.22elapsed 180%CPU (0avgtext+0avgdata 0maxresident)k > 0inputs+0outputs (0major+33555877minor)pagefaults 0swaps > > 2.6.21-rw_mutex > > /usr/bin/time ./ebizzy -m -P > 57.85user 124.30system 1:42.99elapsed 176%CPU (0avgtext+0avgdata 0maxresident)k > 0inputs+0outputs (0major+33555877minor)pagefaults 0swaps
nice! This 6% runtime reduction on a 2-way box will i suspect get exponentially better on systems with more CPUs/cores.
i also like the design, alot: instead of doing a full new lock type (with per-arch changes, extra lockdep support, etc. etc) you layered the new abstraction ontop of mutexes. This makes this hard locking abstraction look really, really simple, while the percpu_counter trick makes it scale _perfectly_ for the reader case. Congratulations!
given how nice this looks already, have you considered completely replacing rwsems with this? I suspect you could test the correctness of that without doing a mass API changeover, by embedding struct rw_mutex in struct rwsem and implementing kernel/rwsem.c's API that way. (the real patch would just flip it all over to rw-mutexes)
Ingo - 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/
| |