Re: [PATCH 4/6] Record how many zones can be safely skipped in the zonelist

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
From: Mel Gorman
Date: Tuesday, August 21, 2007 - 1:58 am

On (17/08/07 14:03), Christoph Lameter didst pronounce:

It's a small gain on a few machines. Where I thought it was more likely
to be a win is on x86-64 NUMA machines particularly if the zonelist
ordering was zone order as there would be potentially many nodes to
skip.

Kernbench didn't show up any regressions for the other machines but the
userspace portion of that workload is unlikely to notice the loss of a
cache line.

-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab
-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
Re: [PATCH 2/6] Use one zonelist that is filtered instead ..., Christoph Lameter, (Fri Aug 17, 1:59 pm)
Re: [PATCH 3/6] Embed zone_id information within the zonel ..., Christoph Lameter, (Fri Aug 17, 2:02 pm)
Re: [PATCH 4/6] Record how many zones can be safely skippe ..., Christoph Lameter, (Fri Aug 17, 2:03 pm)
Re: [PATCH 6/6] Do not use FASTCALL for __alloc_pages_node ..., Christoph Lameter, (Fri Aug 17, 2:07 pm)
Re: [PATCH 5/6] Filter based on a nodemask as well as a gf ..., Christoph Lameter, (Fri Aug 17, 2:29 pm)
Re: [PATCH 4/6] Record how many zones can be safely skippe ..., Mel Gorman, (Tue Aug 21, 1:58 am)