Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing MQ IO

From	Tejun Heo <tj@kernel.org>
Newsgroups	linux.kernel
Subject	Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing MQ IO
Date	2016-04-26 19:50 +0200
Message-ID	<rsfhw-3db-29@gated-at.bofh.it> (permalink)
References	<rrQCw-89O-45@gated-at.bofh.it> <rrQVP-8hY-3@gated-at.bofh.it> <rrZZ8-7nE-29@gated-at.bofh.it> <rscWm-1qW-29@gated-at.bofh.it> <rseYa-34w-27@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

Hello, Peter.

On Tue, Apr 26, 2016 at 10:27:59AM -0700, Peter Hurley wrote:
> > It's unlikely to make any measureable difference.  Is xchg() actually
> > cheaper than store + rmb?
> 
> store + mfence (full barrier), yes. Roughly 2x faster.
> 
> https://lkml.org/lkml/2015/11/2/607

Ah, didn't know that.  Thanks for the pointer.

> > I'm not necessarily against making all clearings of
> > PENDING to be followed by a rmb or use xhcg.  Reasons 2-4 are pretty
> > weak tho.
> 
> I agree 2 and 3 are not the best reasons.
> Actually, it looks that I'm in the minority anyway, and that style-wise,
> naked barrier is preferred.

As long as what's happening is clearly documented, I think either is
fine.  I'm gonna go with Roman's mb patch for -stable fix but think
it'd be nice to have a separate patch to consolidate the paths which
clear PENDING and make them use xchg.  If you can spin up a patch for
that, I'd be happy to apply it to wq/for-3.7.

Thanks.

-- 
tejun

Thread

[PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing MQ IO Roman Pen <roman.penyaev@profitbricks.com> - 2016-04-25 17:30 +0200
  Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Tejun Heo <tj@kernel.org> - 2016-04-25 17:50 +0200
    Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Tejun Heo <tj@kernel.org> - 2016-04-25 18:10 +0200
      Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Roman Penyaev <roman.penyaev@profitbricks.com> - 2016-04-25 18:50 +0200
    Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Roman Penyaev <roman.penyaev@profitbricks.com> - 2016-04-25 18:40 +0200
      Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Tejun Heo <tj@kernel.org> - 2016-04-25 19:10 +0200
        Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Roman Penyaev <roman.penyaev@profitbricks.com> - 2016-04-25 19:50 +0200
          Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Tejun Heo <tj@kernel.org> - 2016-04-25 20:00 +0200
    Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Peter Hurley <peter@hurleysoftware.com> - 2016-04-26 03:30 +0200
      Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Tejun Heo <tj@kernel.org> - 2016-04-26 17:20 +0200
        Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Peter Hurley <peter@hurleysoftware.com> - 2016-04-26 19:30 +0200
          Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Tejun Heo <tj@kernel.org> - 2016-04-26 19:50 +0200
            Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Peter Hurley <peter@hurleysoftware.com> - 2016-04-26 22:10 +0200
            Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Hannes Reinecke <hare@suse.de> - 2016-04-27 08:00 +0200
              Re: [PATCH 1/1] [RFC] workqueue: fix ghost PENDING flag while doing  MQ IO Tejun Heo <tj@kernel.org> - 2016-04-27 21:10 +0200

csiph-web