Re: [Lhms-devel] [RFC] buddy allocator without bitmap [2/4]

[Hiroyuki KAMEZAWA <kamezawa.hiroyu@jp.fujitsu.com>]

Andrew Morton wrote:

> Certainly, executing an atomic op in a tight loop will show a lot of
> difference.  But that doesn't mean that making these operations non-atomic
> makes a significant difference to overall kernel performance!
> 
Thanks.
My test before positng patch is calling mmap()/munmap() with 4-16Mega bytes.
munmap with such Mega bytes causes many calls of __free_pages_bulk() and
many pages are coalesced at once.

This means atomic_ops in heavyly called tight loop
(I called it 3 times in the most inner loop...)

and my test shows bad performance ;).


> But whatever - it all adds up.  The microoptimisation is fine - let's go
> that way.
> 
I'd like to add macros and to get my codes clear.

> 
>>Result:
>>[root@kanex2 atomic]# nice -10 ./test-atomics
>>score 0 is            64011 note: cache hit, no atomic
>>score 1 is           543011 note: cache hit, atomic
>>score 2 is           303901 note: cache hit, mixture
>>score 3 is           344261 note: cache miss, no atomic
>>score 4 is          1131085 note: cache miss, atomic
>>score 5 is           593443 note: cache miss, mixture
>>score 6 is           118455 note: cache hit, dependency, noatomic
>>score 7 is           416195 note: cache hit, dependency, mixture
>>
>>smaller score is better.
>>score 0-2 shows set_bit/__set_bit performance during good cache hit rate.
>>score 3-5 shows set_bit/__set_bit performance during bad cache hit rate.
>>score 6-7 shows set_bit/__set_bit performance during good cache hit
>>but there is data dependency on each access in the tight loop.
> 
> 
> I _think_ the above means atomic ops are 10x more costly, yes?
> 
yes, when L2 cache hits, I think.




-- 
--the clue is these footmarks leading to the door.--
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>