Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Thu, 19 Nov 2020 10:37:03 +0100
From:   Michal Hocko <mhocko@suse.com>
To:     Rik van Riel <riel@surriel.com>
Cc:     hughd@google.com, xuyu@linux.alibaba.com,
        akpm@linux-foundation.org, mgorman@suse.de, aarcange@redhat.com,
        willy@infradead.org, linux-kernel@vger.kernel.org,
        kernel-team@fb.com, linux-mm@kvack.org, vbabka@suse.cz
Subject: Re: [PATCH 1/2] mm,thp,shmem: limit shmem THP alloc gfp_mask
Message-ID: <20201119093703.GB12284@dhcp22.suse.cz>
References: <20201105191508.1961686-1-riel@surriel.com>
 <20201105191508.1961686-2-riel@surriel.com>
 <20201112105258.GZ12240@dhcp22.suse.cz>
 <84effe90c3ee13fbfa6e732d2e3b3d9b557d1be1.camel@surriel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <84effe90c3ee13fbfa6e732d2e3b3d9b557d1be1.camel@surriel.com>
Precedence: bulk

On Fri 13-11-20 22:44:20, Rik van Riel wrote:
> On Thu, 2020-11-12 at 11:52 +0100, Michal Hocko wrote:
> > On Thu 05-11-20 14:15:07, Rik van Riel wrote:
> > > 
> > > This patch applies the same configurated limitation of THPs to
> > > shmem
> > > hugepage allocations, to prevent that from happening.
> > 
> > I believe you should also exaplain why we want to control defrag by
> > the
> > global knob while the enable logic is per mount.
> 
> I added that to the changelog for the next version of
> the patches.
> 
> > > This way a THP defrag setting of "never" or "defer+madvise" will
> > > result
> > > in quick allocation failures without direct reclaim when no 2MB
> > > free
> > > pages are available.
> > > 
> > > With this patch applied, THP allocations for tmpfs will be a little
> > > more aggressive than today for files mmapped with MADV_HUGEPAGE,
> > > and a little less aggressive for files that are not mmapped or
> > > mapped without that flag.
> > 
> > This begs some numbers. A little is rather bad unit of performance. I
> > do
> > agree that unifying those makes sense in general though.
> 
> The aggressiveness is in changes to the gfp_mask, eg by
> adding __GFP_NORETRY. How that translates into THP
> allocation success rates is entirely dependent on the
> workload and on what else is in memory at the time.

Yes and that is why I would argue about consistency with THP rather than
put claims that hard to back by numbers.

-- 
Michal Hocko
SUSE Labs