RE: [patch] mm, slab: faster active and free stats

2016-12-02 Thread
Wednesday, November 30, 2016 9:57 AM To: Joonsoo Kim Cc: Andrew Morton ; Greg Thelen ; Aruna Ramakrishna ; Christoph Lameter ; linux-kernel@vger.kernel.org; linux...@kvack.org Subject: Re: [patch] mm, slab: faster active and free stats On Mon, 28 Nov 2016, Joonsoo Kim wrote: > Hello, >

Re: [patch] mm, slab: faster active and free stats

2016-11-29 Thread David Rientjes
On Mon, 28 Nov 2016, Joonsoo Kim wrote: > Hello, > > Sorry for long delay. > I agree that this improvement is needed. Could you try the approach > that maintains n->num_slabs and n->free_slabs? I guess that it would be > simpler than this patch so more maintainable. > Ok, what do you think abou

Re: [patch] mm, slab: faster active and free stats

2016-11-27 Thread Joonsoo Kim
On Fri, Nov 11, 2016 at 02:30:39AM -0800, David Rientjes wrote: > On Fri, 11 Nov 2016, Joonsoo Kim wrote: > > > Hello, David. > > > > Maintaining acitve/free_slab counters looks so complex. And, I think > > that we don't need to maintain these counters for faster slabinfo. > > Key point is to rem

Re: [patch] mm, slab: faster active and free stats

2016-11-11 Thread David Rientjes
On Fri, 11 Nov 2016, Joonsoo Kim wrote: > Hello, David. > > Maintaining acitve/free_slab counters looks so complex. And, I think > that we don't need to maintain these counters for faster slabinfo. > Key point is to remove iterating n->slabs_partial list. > > We can calculate active slab/object

Re: [patch] mm, slab: faster active and free stats

2016-11-10 Thread Joonsoo Kim
On Wed, Nov 09, 2016 at 04:38:08PM -0800, David Rientjes wrote: > On Tue, 8 Nov 2016, Andrew Morton wrote: > > > > Reading /proc/slabinfo or monitoring slabtop(1) can become very expensive > > > if there are many slab caches and if there are very lengthy per-node > > > partial and/or free lists. >

Re: [patch] mm, slab: faster active and free stats

2016-11-09 Thread David Rientjes
On Tue, 8 Nov 2016, Andrew Morton wrote: > > Reading /proc/slabinfo or monitoring slabtop(1) can become very expensive > > if there are many slab caches and if there are very lengthy per-node > > partial and/or free lists. > > > > Commit 07a63c41fa1f ("mm/slab: improve performance of gathering sl

Re: [patch] mm, slab: faster active and free stats

2016-11-08 Thread Andrew Morton
On Tue, 8 Nov 2016 15:06:45 -0800 (PST) David Rientjes wrote: > Reading /proc/slabinfo or monitoring slabtop(1) can become very expensive > if there are many slab caches and if there are very lengthy per-node > partial and/or free lists. > > Commit 07a63c41fa1f ("mm/slab: improve performance of