linux-kernel - Re: [PATCH/RFC] wait_for_completion

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20131118234209.GO16796@laptop.programming.kicks-ass.net>
Date:	Tue, 19 Nov 2013 00:42:09 +0100
From:	Peter Zijlstra <peterz@...radead.org>
To:	Andrew Morton <akpm@...ux-foundation.org>
Cc:	NeilBrown <neilb@...e.de>, Ingo Molnar <mingo@...hat.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	lkml <linux-kernel@...r.kernel.org>,
	"Dr. H. Nikolaus Schaller" <hns@...delico.com>,
	Marek Belisko <marek@...delico.com>,
	Mark Brown <broonie@...nsource.wolfsonmicro.com>
Subject: Re: [PATCH/RFC] wait_for_completion_timeout() considered harmful.

On Mon, Nov 18, 2013 at 03:27:46PM -0800, Andrew Morton wrote:
> On Sun, 17 Nov 2013 08:06:03 +1100 NeilBrown <neilb@...e.de> wrote:
> 
> > It would be reasonable to assume that
> > 
> >       wait_for_completion_timeout(&wm8350->auxadc_done, msecs_to_jiffies(5));
> > 
> > would wait at least 5 msecs for the auxadc_done to complete.  But it does not.
> > With a HZ of 200 or less, msecs_to_jiffies(5) has value '1', and so this
> > will only wait until the next "timer tick", which could happen immediately.
> > 
> > This can lead to incorrect results - and has done so in out-of-tree patches
> > for drivers/misc/bmp085.c which uses a very similar construct to enable interrupt
> > based result collection.
> > 
> > The documentation for several *_timeout* functions claim they will wait for
> > "timeout jiffies" to have elapsed where this is not the case.  They will
> > actually wait for "timeout" jiffies to have started implying an elapsed time
> > between (timeout-1) and (timeout).
> > 
> > This patch corrects some of this documentation, and adds a collection of
> >   wait_for_completion*_msecs()
> > interfaces which wait at least the given number of milliseconds for the
> > completion (or a signal).
> 
> Mutter.  wait_for_x(..., 5ms) should wait for a minimum of 5ms, no matter
> what.
> 
> So I'd suggest we make that happen, rather than adding some new interfaces?

Yeah, also the completion interface is just one of many that's buggered
this way.

I briefly talked to Thomas about this earlier today and we need to fix
this at a lower level -- the quick 'n dirty solution is to add 1 jiffy
down in the timer-wheel when we enqueue these things.

And yes, we very much don't want to create new interfaces with similar
but slightly different semantics, that's just asking for trouble.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/