Message-ID: <5307849A.9050209@hurleysoftware.com>
Date: Fri, 21 Feb 2014 11:53:46 -0500
From: Peter Hurley <peter@hurleysoftware.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.3.0
MIME-Version: 1.0
To: Tejun Heo <tj@kernel.org>
CC: laijs@cn.fujitsu.com, linux-kernel@vger.kernel.org,
        Stefan Richter <stefanr@s5r6.in-berlin.de>,
        linux1394-devel@lists.sourceforge.net, Chris Boot <bootc@bootc.net>,
        linux-scsi@vger.kernel.org, target-devel@vger.kernel.org
Subject: Re: [PATCH 4/9] firewire: don't use PREPARE_DELAYED_WORK
References: <1392929071-16555-1-git-send-email-tj@kernel.org> <1392929071-16555-5-git-send-email-tj@kernel.org> <5306AF8E.3080006@hurleysoftware.com> <20140221015935.GF6897@htj.dyndns.org> <5306B4DF.4000901@hurleysoftware.com> <20140221021341.GG6897@htj.dyndns.org> <5306E06C.5020805@hurleysoftware.com> <20140221100301.GA14653@mtj.dyndns.org> <53074BE4.1020307@hurleysoftware.com> <20140221130614.GH6897@htj.dyndns.org>
In-Reply-To: <20140221130614.GH6897@htj.dyndns.org>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org

Hi Tejun,

On 02/21/2014 08:06 AM, Tejun Heo wrote:
> Hello,
>
> On Fri, Feb 21, 2014 at 07:51:48AM -0500, Peter Hurley wrote:
>> I think the vast majority of kernel code which uses the workqueue
>> assumes there is a memory ordering guarantee.
>
> Not really.  Workqueues haven't even guaranteed non-reentrancy until
> recently, forcing everybody to lock explicitly in the work function.
> I don't think there'd be many, if any, use cases which depend on
> ordering guarantee on duplicate queueing.

I added some in 3.12 :)

>> Another way to look at this problem is that process_one_work()
>> doesn't become the existing instance _until_ PENDING is cleared.
>
> Sure, having that guarantee definitely is nicer and all we need seems
> to be mb_after_unlock in the execution path.  Please feel free to
> submit a patch.

Ok, I can do that. But AFAIK it'll have to be an smp_rmb(); there is
no mb__after unlock.

[ After thinking about it some, I don't think preventing speculative
   writes before clearing PENDING if useful or necessary, so that's
   why I'm suggesting only the rmb. ]

>>> add such guarantee, not sure how much it'd matter but it's not like
>>> it's gonna cost a lot either.
>>>
>>> This doesn't have much to do with the current series tho.  In fact,
>>> PREPARE_WORK can't ever be made to give such guarantee.
>>
>> Yes, I agree that PREPARE_DELAYED_WORK was also broken usage with the
>> same problem. [And there are other bugs in that firewire device work
>> code which I'm working on.]
>>
>>> The function pointer has to fetched before clearing of PENDING.
>>
>> Why?
>>
>> As long as the load takes place within the pool->lock, I don't think
>> it matters (especially now PREPARE_WORK is removed).
>
> Hmmm... I was talking about PREPARE_WORK().  Clearing PENDING means
> that the work item is released from the worker context and may be
> freed or reused at any time (hmm... this may not be true anymore as
> non-syncing variants of cancel_work are gone), so clearing PENDING
> should be the last access to the work item and thus we can't use that
> as the barrier event for fetching its work function.

Yeah, it seems like the work item lifetime is at least guaranteed
while either PENDING is set _or_ while the pool->lock is held
after PENDING is cleared.

Regards,
Peter Hurley

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/