Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 27 Nov 2018 19:21:37 +0100
From:   Michal Hocko <mhocko@kernel.org>
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     rong.a.chen@intel.com, Andrea Arcangeli <aarcange@redhat.com>,
        s.priebe@profihost.ag, alex.williamson@redhat.com,
        mgorman@techsingularity.net, zi.yan@cs.rutgers.edu,
        Vlastimil Babka <vbabka@suse.cz>, rientjes@google.com,
        kirill@shutemov.name, Andrew Morton <akpm@linux-foundation.org>,
        Linux List Kernel Mailing <linux-kernel@vger.kernel.org>,
        lkp@01.org
Subject: Re: [LKP] [mm] ac5b2c1891: vm-scalability.throughput -61.3%
 regression
Message-ID: <20181127182137.GE6923@dhcp22.suse.cz>
References: <20181127062503.GH6163@shao2-debian>
 <CAHk-=wiq9eTzenOPh9srdB=Y8NcP6NJxh7o8Y5SfSjfqZ20-RA@mail.gmail.com>
 <20181127181727.GD6923@dhcp22.suse.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20181127181727.GD6923@dhcp22.suse.cz>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue 27-11-18 19:17:27, Michal Hocko wrote:
> On Tue 27-11-18 09:08:50, Linus Torvalds wrote:
> > On Mon, Nov 26, 2018 at 10:24 PM kernel test robot
> > <rong.a.chen@intel.com> wrote:
> > >
> > > FYI, we noticed a -61.3% regression of vm-scalability.throughput due
> > > to commit ac5b2c18911f ("mm: thp: relax __GFP_THISNODE for
> > > MADV_HUGEPAGE mappings")
> > 
> > Well, that's certainly noticeable and not good.
> > 
> > Andrea, I suspect it might be causing fights with auto numa migration..
> > 
> > Lots more system time, but also look at this:
> > 
> > >    1122389 ?  9%     +17.2%    1315380 ?  4%  proc-vmstat.numa_hit
> > >     214722 ?  5%     +21.6%     261076 ?  3%  proc-vmstat.numa_huge_pte_updates
> > >    1108142 ?  9%     +17.4%    1300857 ?  4%  proc-vmstat.numa_local
> > >     145368 ? 48%     +63.1%     237050 ? 17%  proc-vmstat.numa_miss
> > >     159615 ? 44%     +57.6%     251573 ? 16%  proc-vmstat.numa_other
> > >     185.50 ? 81%   +8278.6%      15542 ? 40%  proc-vmstat.numa_pages_migrated
> > 
> > Should the commit be reverted? Or perhaps at least modified?
> 
> Well, the commit is trying to revert to the behavior before
> 5265047ac301 because there are real usecases that suffered from that
> change and bug reports as a result of that.
> 
> will-it-scale is certainly worth considering but it is an artificial
> testcase. A higher NUMA miss rate is an expected side effect of the
> patch because the fallback to a different NUMA node is more likely. The
> __GFP_THISNODE side effect is basically introducing node-reclaim
> behavior for THPages. Another thing is that there is no good behavior
> for everybody. Reclaim locally vs. THP on a remote node is hard to
> tell by default. We have discussed that at length and there were some
> conclusions. One of them is that we need a numa policy to tell whether
> a expensive localility is preferred over remote allocation.  Also we
> definitely need a better pro-active defragmentation to allow larger
> pages on a local node. This is a work in progress and this patch is a
> stop gap fix.

Btw. the associated discussion is http://lkml.kernel.org/r/20180925120326.24392-1-mhocko@kernel.org

-- 
Michal Hocko
SUSE Labs