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:   Fri, 22 Jun 2018 14:46:58 -0400 (EDT)
From:   Mikulas Patocka <mpatocka@redhat.com>
To:     Michal Hocko <mhocko@kernel.org>
cc:     jing xia <jing.xia.mail@gmail.com>,
        Mike Snitzer <snitzer@redhat.com>, agk@redhat.com,
        dm-devel@redhat.com, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org
Subject: Re: dm bufio: Reduce dm_bufio_lock contention
In-Reply-To: <20180622131013.GA10465@dhcp22.suse.cz>
Message-ID: <alpine.LRH.2.02.1806221445310.2717@file01.intranet.prod.int.rdu2.redhat.com>
References: <20180615073201.GB24039@dhcp22.suse.cz> <alpine.LRH.2.02.1806150724260.15022@file01.intranet.prod.int.rdu2.redhat.com> <20180615115547.GH24039@dhcp22.suse.cz> <alpine.LRH.2.02.1806150832100.26650@file01.intranet.prod.int.rdu2.redhat.com>
 <20180615130925.GI24039@dhcp22.suse.cz> <alpine.LRH.2.02.1806181003560.4201@file01.intranet.prod.int.rdu2.redhat.com> <20180619104312.GD13685@dhcp22.suse.cz> <alpine.LRH.2.02.1806191228110.25656@file01.intranet.prod.int.rdu2.redhat.com>
 <20180622090151.GS10465@dhcp22.suse.cz> <alpine.LRH.2.02.1806220828040.8072@file01.intranet.prod.int.rdu2.redhat.com> <20180622131013.GA10465@dhcp22.suse.cz>
User-Agent: Alpine 2.02 (LRH 1266 2009-07-14)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On Fri, 22 Jun 2018, Michal Hocko wrote:

> On Fri 22-06-18 08:44:52, Mikulas Patocka wrote:
> > On Fri, 22 Jun 2018, Michal Hocko wrote:
> [...]
> > > Why? How are you going to audit all the callers that the behavior makes
> > > sense and moreover how are you going to ensure that future usage will
> > > still make sense. The more subtle side effects gfp flags have the harder
> > > they are to maintain.
> > 
> > I did audit them - see the previous email in this thread: 
> > https://www.redhat.com/archives/dm-devel/2018-June/thread.html
> 
> I do not see any mention about throttling expectations for those users.
> You have focused only on the allocation failure fallback AFAIR

How should the callers be analyzed with respect to throttling?

Mikulas