Re: hackbench regression due to commit 9dfc6e68bfe6e

From: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com>
To: Tejun Heo <tj@kernel.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>,
	Christoph Lameter <cl@linux-foundation.org>,
	alex.shi@intel.com,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"Ma, Ling" <ling.ma@intel.com>,
	"Chen, Tim C" <tim.c.chen@intel.com>,
	Andrew Morton <akpm@linux-foundation.org>
Subject: Re: hackbench regression due to commit 9dfc6e68bfe6e
Date: Tue, 06 Apr 2010 16:28:17 +0800	[thread overview]
Message-ID: <1270542497.2078.123.camel@ymzhang.sh.intel.com> (raw)
In-Reply-To: <4BBA8DF9.8010409@kernel.org>

On Tue, 2010-04-06 at 10:27 +0900, Tejun Heo wrote:
> Hello,
> 
> On 04/06/2010 02:30 AM, Pekka Enberg wrote:
> >> Hmnmmm... The dynamic percpu areas use page tables and that data is used
> >> in the fast path. Maybe the high thread count causes tlb trashing?
> > 
> > Hmm indeed. I don't see anything particularly funny in the SLUB percpu
> > conversion so maybe this is a more issue with the new percpu
> > allocator?
> 
> By default, percpu allocator embeds the first chunk in the kernel
> linear mapping and accesses there shouldn't involve any TLB overhead.
> >From the second chunk on, they're mapped page-by-page into vmalloc
> area.  This can be updated to use larger page mapping but 2M page
> per-cpu is pretty large and the trade off hasn't been right yet.
> 
> The amount reserved for dynamic allocation in the first chunk is
> determined by PERCPU_DYNAMIC_RESERVE constant in
> include/linux/percpu.h.  It's currently 20k on 64bit machines and 12k
> on 32bit.  The intention was to size this such that most common stuff
> is allocated from this area.  The 20k and 12k are numbers that I
> pulled out of my ass :-) with the custom config I used.  Now that more
> stuff has been converted to dynamic percpu, it's quite possible that
> the area is too small.  Can you please try to increase the size of the
> area (say 2 or 4 times) and see whether the performance regression
> goes away?
Thanks. I tried 2 and 4 times and didn't see much improvement.
I checked /proc/vamallocinfo and it doesn't have item of pcpu_get_vm_areas
when I use 4 times of PERCPU_DYNAMIC_RESERVE.

I used perf to collect dtlb misses and LLC misses. dtlb miss data is not
stable. Sometimes, we have a bigger dtlb miss, but get a better result.

LLC misses data are more stable. Only LLC-load-misses is the clear sign now.
LLC-store-misses has no big difference.