Re: [PATCH v2 3/3] percpu: improve allocation success rate for non-GFP_KERNEL callers

Path	csiph.com!news.mixmin.net!news.unit0.net!fu-berlin.de!bofh.it!news.nic.it!robomod
From	Tejun Heo <tj@kernel.org>
Newsgroups	linux.kernel
Subject	Re: [PATCH v2 3/3] percpu: improve allocation success rate for non-GFP_KERNEL callers
Date	Mon, 27 Feb 2017 23:40:02 +0100
Message-ID	<tfAS5-2SI-5@gated-at.bofh.it> (permalink)
References	<teVce-7pj-3@gated-at.bofh.it> <teYWt-1zx-1@gated-at.bofh.it> <tfqzo-491-15@gated-at.bofh.it> <tftdU-5Yo-25@gated-at.bofh.it> <tfA5I-2i1-17@gated-at.bofh.it> <tfA5I-2i1-15@gated-at.bofh.it> <tfAfo-2lV-13@gated-at.bofh.it> <tfAp4-2r5-21@gated-at.bofh.it>
Dkim-Signature	v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=sender:date:from:to:cc:subject:message-id:references:mime-version :content-disposition:in-reply-to:user-agent; bh=RXPyVtCSwj8MBkGL5nFkE1BQoY7SL7PcWevVIidAWGY=; b=Gu49sVrFh41AN6fvpTnHnjY7wfRxLPxC8DistSYolutOBNUz8mExyM6fs9W/F/bj9a v2JbtKOdl3ECAac/ckHo4PM919Urhrhv7pzCTNP+vIL8Q3HqPzlifK15a3o2TazVDVRD QO4y0bog6qOS0a4Bo+OY5lYlvDistDoZH5HeqsPwu77ejPUHnPLy+io1vIJ1TjqjL5aL CVPEREZKpelaXlomiiLq8l8gi/aTOlkvykO+Bn77vdARqBuA1AlAl/4rydPdIDLFy946 +/5lO3jhythUCCfSjFGO4T902x8HQxZf1kLzWTjEaqKYAxSMQrircNR9ArmM+xK1P3dl +7yw==
X-Google-Dkim-Signature	v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:sender:date:from:to:cc:subject:message-id :references:mime-version:content-disposition:in-reply-to:user-agent; bh=RXPyVtCSwj8MBkGL5nFkE1BQoY7SL7PcWevVIidAWGY=; b=MeJArM+hh/k9w1V8IKn8ZY6cI/6Fovzygfc8DHT1z2o6Kw3WhdfTzDBj49lLYUyd4K ya0elAou4/Fm1hf2ASS52+rE0MgtxAH1zJP+TCDSzjkcRb7dz7AsO88QFu/cx3Gb5hwF siQFGUVh0pmTC78J63AXg1HoSqVMVcWyJqKSjGBALJfk0zeVkj27JvDZ3yb/StiSnmF0 qwoBLOjyDaz7jahjTP/LkdkOcQDjVLatQeBCWMCdAJlAtupkDaGgk0lb8Qb/EAREXknB 4TadtpMfjWlN1UpVsMoE4jfXqWRrItQ6NzIxgRDNjxyP4llYHKwA/4MWsMECg5JYv0w5 bCFQ==
X-Gm-Message-State	AMke39lgvKK+GkCar5By09WZFyr+TZVygOKX8emoljHEhbugotu2z5IVBJX5hLZNUIgS9A==
X-Received	by 10.13.239.7 with SMTP id y7mr12814883ywe.33.1488228328978; Mon, 27 Feb 2017 12:45:28 -0800 (PST)
MIME-Version	1.0
Content-Type	text/plain; charset=us-ascii
Content-Disposition	inline
User-Agent	Mutt/1.7.1 (2016-10-04)
Sender	robomod@news.nic.it
List-ID	<linux-kernel.vger.kernel.org>
X-Mailing-List	linux-kernel@vger.kernel.org
Approved	robomod@news.nic.it
Lines	32
Organization	linux.* mail to news gateway
X-Original-Cc	Michal Hocko <mhocko@kernel.org>, Christoph Lameter <cl@linux.com>, Andrew Morton <akpm@linux-foundation.org>, Chris Wilson <chris@chris-wilson.co.uk>, Andrey Ryabinin <aryabinin@virtuozzo.com>, Roman Pen <r.peniaev@gmail.com>, Joonas Lahtinen <joonas.lahtinen@linux.intel.com>, zijun_hu <zijun_hu@htc.com>, Joonsoo Kim <iamjoonsoo.kim@lge.com>, David Rientjes <rientjes@google.com>, linux-mm@kvack.org, linux-kernel@vger.kernel.org
X-Original-Date	Mon, 27 Feb 2017 15:45:27 -0500
X-Original-Message-ID	<20170227204527.GG8707@htj.duckdns.org>
X-Original-References	<201702260805.zhem8KFI%fengguang.wu@intel.com> <20170226043829.14270-1-tahsin@google.com> <20170227095258.GG14029@dhcp22.suse.cz> <CAAeU0aMaGa63Nj=JvZKKy82FftAT9dF56=gZsufDvrkqDSGUrw@mail.gmail.com> <20170227195126.GC8707@htj.duckdns.org> <CAAeU0aORY=N0e0gMKu-CBAEF=HLuHUNV6KWy27th1rwuPMcTMg@mail.gmail.com> <20170227202906.GF8707@htj.duckdns.org> <CAAeU0aMnz-nsXGy44mwBfzwfFJtVWNRQiAE0UAonBQA3iDJBqg@mail.gmail.com>
X-Original-Sender	linux-kernel-owner@vger.kernel.org
Xref	csiph.com linux.kernel:1589011

Show key headers only | View raw

Hello, Tahsin.

On Mon, Feb 27, 2017 at 12:37:59PM -0800, Tahsin Erdogan wrote:
> > Ah, absolutely, that's a stupid failure but we should be able to fix
> > that by making the blkg functions take gfp mask and allocate
> > accordingly, right?  It'll probably take preallocation tricks because
> > of locking but should be doable.
> 
> My initial goal was to allow calls to vmalloc(), but I now see the
> challenges in that
> approach.

I'd love to see that working too but this is a different issue.  Even
GFP_ATOMIC can fail under pressure and it's kinda wrong to depend on
that for userspace interactions.

> Doing preallocations would probably work but not sure if that can be
> done without
> complicating code too much. Could you describe what you have in mind?

So, blkg_create() already takes @new_blkg argument which is the
preallocated blkg and used during q init.  Wouldn't it work to make
blkg_lookup_create() take @new_blkg too and pass it down to
blkg_create() (and also free it if it doesn't get used)?  Then,
blkg_conf_prep() can always (or after a failure with -ENOMEM) allocate
a new blkg before calling into blkg_lookup_create().  I don't think
it'll complicate the code path that much.

Thanks.

-- 
tejun

Thread

[PATCH 3/3] percpu: improve allocation success rate for non-GFP_KERNEL callers Tahsin Erdogan <tahsin@google.com> - 2017-02-25 22:10 +0100
  Re: [PATCH 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers kbuild test robot <lkp@intel.com> - 2017-02-26 01:00 +0100
  Re: [PATCH 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers kbuild test robot <lkp@intel.com> - 2017-02-26 01:50 +0100
    [PATCH v2 3/3] percpu: improve allocation success rate for non-GFP_KERNEL callers Tahsin Erdogan <tahsin@google.com> - 2017-02-26 05:50 +0100
      Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Michal Hocko <mhocko@kernel.org> - 2017-02-27 11:20 +0100
        Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tahsin Erdogan <tahsin@google.com> - 2017-02-27 14:10 +0100
          Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Michal Hocko <mhocko@kernel.org> - 2017-02-27 16:30 +0100
            Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Michal Hocko <mhocko@kernel.org> - 2017-02-27 18:10 +0100
              Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Michal Hocko <mhocko@kernel.org> - 2017-02-27 18:40 +0100
                Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tahsin Erdogan <tahsin@google.com> - 2017-02-27 20:40 +0100
                Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Michal Hocko <mhocko@kernel.org> - 2017-02-27 20:50 +0100
            Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tahsin Erdogan <tahsin@google.com> - 2017-02-27 18:10 +0100
          Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tahsin Erdogan <tahsin@google.com> - 2017-02-27 21:30 +0100
            Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tejun Heo <tj@kernel.org> - 2017-02-27 21:40 +0100
              Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tahsin Erdogan <tahsin@google.com> - 2017-02-27 21:50 +0100
                Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tahsin Erdogan <tahsin@google.com> - 2017-02-27 22:20 +0100
                Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tejun Heo <tj@kernel.org> - 2017-02-27 22:40 +0100
                Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tejun Heo <tj@kernel.org> - 2017-02-27 23:40 +0100
          Re: [PATCH v2 3/3] percpu: improve allocation success rate for  non-GFP_KERNEL callers Tejun Heo <tj@kernel.org> - 2017-02-27 23:40 +0100

csiph-web