On 04/08/2010 12:39 PM, Avi Kivity wrote:
On 04/08/2010 04:50 AM, Andrea Arcangeli wrote:
Hello,
I merged memory compaction v7 from Mel, plus his latest incremental
updates
into my tree.
A quick benchmark, running 'sort -b 1200M /tmp/largerand >
/dev/null'. The file is 1GB in size, consisting of 15-char base64
encoded random string + newline records.
That's -S, not -b.
I'll try running this with a kernel build in parallel.
Results here are less than stellar. While khugepaged is pulling pages
together, something is breaking them apart. Even after memory pressure
is removed, this behaviour continues. Can it be that compaction is
tearing down huge pages?
After about 80 seconds of sort, we only have 40-50MB worth of hugepages,
even though khugepaged is collapsing 50-100 per second (which should
have got the job done in 5-10 seconds).
After dropping pagecache, we get about 800MB of large pages, so it looks
like antifrag/compaction isn't able to deal with pagecache well.
Dropping dcache adds a further 100MB.
All this with mem=2G.
--
error compiling committee.c: too many arguments to function
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>