Re: [PATCH] anobjrmap 9 priority mjb tree

[Martin J. Bligh]

--Hugh Dickins <hugh@xxxxxxxxxxx> wrote (on Friday, April 09, 2004 22:51:03 +0100):

> On Fri, 9 Apr 2004, Martin J. Bligh wrote:
>> > This anobjrmap 9 (or anon_mm9) patch adds Rajesh's radix priority search
>> > tree on top of Martin's 2.6.5-rc3-mjb2 tree, making a priority mjb tree!
>> > Approximately equivalent to Andrea's 2.6.5-aa1, but using anonmm instead
>> > of anon_vma, and of course each tree has its own additional features.
>> 
>> This slows down kernel compile a little, but worse, it slows down SDET
>> by about 25% (on the 16x). I think you did something horrible to sem
>> contention ... presumably i_shared_sem, which SDET was fighting with
>> as it was anyway ;-(
>> 
>> Diffprofile shows:
>> 
>>     122626    15.7% total
>>      44129   790.0% __down
>>      20988     4.1% default_idle
> 
> Many thanks for the good news, Martin ;)
> Looks like I've done something very stupid, perhaps a mismerge.
> Not found it yet, I'll carry on looking tomorrow.

I don't think so ... I ran across a similar problems when I tried to convert
the vma list to a list-of-lists for what you're trying to use prio_tree
for, I think. The sorry fact is that the that thing is heavily contended,
and doing complex manipulations under it makes it worse ;-(

I vaguely considered being a smarty-pants and doing RCU list-of-lists, but
never got around to it. I'll shove in some backtraces and check out the
sem activity to be sure.

M.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/