Messages in this thread | | | Date | Wed, 20 Nov 2013 14:58:40 +0800 | From | Wei Yang <> | Subject | Re: [PATCH 1/3] percpu: stop the loop when a cpu belongs to a new group |
| |
On Wed, Nov 20, 2013 at 12:51:21AM -0500, Tejun Heo wrote: >Hello, > >On Wed, Nov 20, 2013 at 11:00:56AM +0800, Wei Yang wrote: >> What do you think about this one? >> >> > >> >From bd70498b9df47b25ff20054e24bb510c5430c0c3 Mon Sep 17 00:00:00 2001 >> >From: Wei Yang <weiyang@linux.vnet.ibm.com> >> >Date: Thu, 10 Oct 2013 09:42:14 +0800 >> >Subject: [PATCH] percpu: optimize group assignment when cpu_distance_fn is >> > NULL >> > >> >When cpu_distance_fn is NULL, all CPUs belongs to group 0. The original logic >> >will continue to go through each CPU and its predecessor. cpu_distance_fn is >> >always NULL when pcpu_build_alloc_info() is called from pcpu_page_first_chunk(). >> > >> >By applying this patch, the time complexity will drop to O(n) form O(n^2) in >> >case cpu_distance_fn is NULL. > >The test was put in the inner loop because the nesting was already too >deep and cpu_distance_fn is unlikely to be NULL on machines where the >number of CPUs is high enough to matter. If that O(n^2) loop is gonna >be a problem, it's gonna be a problem on large NUMA machines and we'll >have to do something about it for cases where cpu_distance_fn exists >anyway.
Tejun,
Yep, hope this will not bring some problem on a large NUMA machie when cpu_distance_fn is not NULL.
> >The patch is just extremely marginal. Ah well... why not? I'll apply >it once -rc1 drops. > >Thanks. > >-- >tejun
-- Richard Yang Help you, Help me
| |