From: Dave O. <ol...@us...> - 2002-02-22 18:47:25
|
Yes, this patch could increase dcache_lock contention. But it should also decrease cache line movement associated with the dcache_lock. So, the patch may be a win on two fronts. Reduce cache line traffic due to d_count movement, and so due to dcache_lock movement. The net result might be for the better despite the increased contention. > From lse...@li... Fri Feb 22 10:35:33 2002 > --On Friday, February 22, 2002 23:52:04 +0530 Dipankar Sarma <dip...@in...> wrote: > > > On Fri, Feb 22, 2002 at 09:18:41AM -0800, Hanna Linder wrote: > >> 3.1% 3.5% 0.6us(1047us) 3.2us( 979us)(0.07%) 25499542 96.5% 3.5% 0% dcache_lock > > > > IIUC, holding the dcache_lock while walking the path as long as > > we can find things in cache should decrease the number of times > > we grab dcache_lock, but it seems both 2.4.17 and fast-walk patch > > Well, I am actually adding holding the dcache_lock where > it was not held before (instead of incrementing d_count). So my > understanding is the dcache_lock contention should go up a little > bit. Although perhaps there is more tuning that needs to happen. > > Hanna |