Messages in this thread | | | Date | Fri, 29 Jul 2011 17:23:13 +0900 | From | Minchan Kim <> | Subject | Re: [PATCH v4 00/10] Prevent LRU churning |
| |
On Wed, Jul 27, 2011 at 01:16:50PM -0700, Andrew Morton wrote: > On Mon, 4 Jul 2011 23:04:33 +0900 > Minchan Kim <minchan.kim@gmail.com> wrote: > > > Test result is following as. > > > > 1) Elapased time 10GB file decompressed. > > Old inorder inorder + pagevec flush[10/10] > > 01:47:50.88 01:43:16.16 01:40:27.18 > > > > 2) failure of inorder lru > > For test, it isolated 375756 pages. Only 45875 pages(12%) are put backed to > > out-of-order(ie, head of LRU) Others, 329963 pages(88%) are put backed to in-order > > (ie, position of old page in LRU). > > I'm getting more and more worried about how complex MM is becoming and > this patchset doesn't take us in a helpful direction :(
Hmm. I think it's not too complicated stuff. :( But I understand your concern enoughly.
> > But it's hard to argue with numbers like that. Please respin patches 6-10?
Of course, but it would be rather late due to my business and other interesting features. I will try to get a new data point in next version.
Thanks, Andrew.
> >
-- Kind regards, Minchan Kim
| |