Re: [Patch 3/3] prepopulate/cache cleared pages

[Nick Piggin <nickpiggin@yahoo.com.au>]

Arjan van de Ven wrote:
>>Arjan, just to get an idea of your workload: obviously it is a mix of
>>read and write on the mmap_sem (read only will not really benefit from
>>reducing lock width because cacheline transfers will still be there).
> 
> 
> yeah it's threads that each allocate, use and then free memory with
> mmap()
> 
> 
>>Is it coming from brk() from the allocator? Someone told me a while ago
>>that glibc doesn't have a decent amount of hysteresis in its allocator
>>and tends to enter the kernel quite a lot... that might be something
>>to look into.
> 
> 
> we already are working on that angle; I just posted the kernel stuff as
> a side effect basically 
> 

OK.

[aside]
Actually I have a scalability improvement for rwsems, that moves the
actual task wakeups out from underneath the rwsem spinlock in the up()
paths. This was useful exactly on a mixed read+write workload on mmap_sem.

The difference was quite large for the "generic rwsem" algorithm because
it uses the spinlock in fastpaths a lot more than the xadd algorithm. I
think x86-64 uses the former, which is what I presume you're testing with?

Obviously this is a slightly different issue from the one you're trying to
address here, but I'll dig out patch when I get some time and you can see
if it helps.

-- 
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com