Messages in this thread | | | Date | Thu, 14 Apr 2011 17:09:26 +0200 | From | Andrea Arcangeli <> | Subject | Re: 2.6.38 sbrk regression |
| |
On Thu, Apr 14, 2011 at 02:49:14PM +0300, raz ben yehuda wrote: > Hey Andrea > Me again. I managed to ftrace ( function graph ) the two kernels. I used > 2.6.37 and 2.6.39-rc3. The bellow is example for sys_brk calls traces > from each kernel. As you can see, there is no "single smoking gun" > here. > > The vm functions durations increased as a whole. > I repeated the tests from sha1 4e9f64c42d0ba5eb0c78569435ada4c224332ce4 > compared to sha1 152c9ccb75548c027fa3103efa4fa4e19a345449 and it is > consistent. ~13% performance decrease. >
> Can you see any relation to thp that might causes this degradation ?
With compaction and THP off I don't see how it could change anything.
But can you try the THP-33 tag of my aa.git tree, that was based on 2.3.37-rc5 so it'll rule out the whole THP patchset if it doesn't regress compared to 2.6.37-rc5 vanilla.
git clone --reference linux-2.6 git://git.kernel.org/pub/scm/linux/kernel/git/andrea/aa.git git checkout THP-33
Thanks, Andrea
| |