Re: [GIT PULL] aio: fix sleeping while TASK_INTERRUPTIBLE

From: Jens Axboe <axboe@kernel.dk>
To: Peter Zijlstra <peterz@infradead.org>,
	Linus Torvalds <torvalds@linux-foundation.org>
Cc: Benjamin LaHaise <bcrl@kvack.org>,
	linux-aio@kvack.org, Linux Kernel <linux-kernel@vger.kernel.org>
Subject: Re: [GIT PULL] aio: fix sleeping while TASK_INTERRUPTIBLE
Date: Tue, 03 Feb 2015 16:24:55 -0700	[thread overview]
Message-ID: <54D158C7.6020606@kernel.dk> (raw)
In-Reply-To: <20150203115531.GI24151@twins.programming.kicks-ass.net>

On 02/03/2015 04:55 AM, Peter Zijlstra wrote:
> On Tue, Feb 03, 2015 at 12:33:48PM +0100, Peter Zijlstra wrote:
>>> block/bsg.c-    prepare_to_wait(&bd->wq_done, &wait, TASK_UNINTERRUPTIBLE);
>>> block/bsg.c-    spin_unlock_irq(&bd->lock);
>>> block/bsg.c:    io_schedule();
>>> block/bsg.c-    finish_wait(&bd->wq_done, &wait);
>>>
>>> Which is double buggy because:
>>>   1) it doesn't loop
>>>   2) it sets TASK_UNINTERRUPTIBLE _after_ testing for the sleep event.
>>
>> OK, actually had a look at this one; it might be ok.
>>
>> The spinlock might fully serialize the state so no fails, and the entire
>> function is called in a loop. Still seriously obtuse code.
>
> Jens, would something like the below work for you?

Yes, from a cursory look, that seems fine to me. Though I will hold the 
fact that you label my code as 'seriously obtuse' against you. Some day.

I can pull this in for testing for 3.20. Mind sending a properly 
formatted patch (signed off, commit message, all that stuff)?

-- 
Jens Axboe