Threading issues? [[e-mail]: ck Digest, Vol 3, Issue 16]

[Andreas Mohr]

Hi all,

this is a nice little posting from the Con Kolivas Linux interactivity
patches list.
Note that it is about Cedega, not Wine, but it might easily affect Wine, too.

Are we doing some big no-nos in the threading area here?
(or maybe it's just done the way it is because we cannot do it a
different, better way?)

For the whole discussion, please see
http://bhhdoa.org.au/pipermail/ck/2004-August/000428.html

Andreas Mohr

----- Forwarded message from ck-request(a)vds.kolivas.org -----

Message: 12
Date: Mon, 16 Aug 2004 21:31:59 +1000
From: Con Kolivas <kernel(a)kolivas.org>
Subject: Re: [ck] 2.6.8.1-ck1, cedega-cvs, ALSA sound (choppy)
To: greblus(a)poczta.onet.pl
Cc: ck(a)vds.kolivas.org
Message-ID: <41209B2F.4030207(a)kolivas.org>
Content-Type: text/plain; charset="us-ascii"

Wiktor Grebla wrote:
> On Mon, Aug 16, 2004 at 08:16:08PM +1000, Con Kolivas wrote:
> > It wont be a preempt threshold problem.
> > See if it's a priority issue on the part of wine*. Be absolutely certain 
> > that wineserver, wine and the game are all run the _same_ priority and 
> > not a better priority than X. Send me the output of 'top -b -n 1' during 
> > the sound being choppy. Then try running _all_ the wine things at nice 
> > +19. This is a simple sanity check to see where the problem lies.
>
> It's in fact as you say, all parts of wine have different priorities, and
> some of them are higher than X:
>
> greblus(a)greblus:~$ top -n 1 -b  | grep wine
>  1630 greblus   23   0  471m 230m 164m R 92.8 45.9   1:54.63 wine
>  1666 greblus   20   0  471m 230m 164m S  2.0 45.9   0:01.64 wine
>  1629 greblus   20   0  2400 1192 2176 S  0.0  0.2   0:00.00 winex
>  1633 greblus   20   0  6864 5056 6756 S  0.0  1.0   0:01.10 wineserver
>  1634 greblus   20   0  6864 5056 6756 S  0.0  1.0   0:00.15 wineserver
>  1659 greblus   20   0  471m 230m 164m S  0.0 45.9   0:00.00 wine
>  1660 greblus   21   0  471m 230m 164m S  0.0 45.9   0:00.00 wine
>  1662 greblus   20   0  471m 230m 164m S  0.0 45.9   0:00.06 wine
>  1663 greblus   33   0  471m 230m 164m R  0.0 45.9   0:00.41 wine
>
> greblus(a)greblus:~$ top -b -n 1 | grep X
>   621 root      20   0 88092  18m  75m S  0.0  3.8   1:02.96 X
>
> After renicing all of them there is no problem with choppy sound in all of time 
> wasting programs I've on my disk :>, and overall (subjective) performance is 
> better of course.
>
> Full output of top -b -n 1 is here:
> http://unixforge.org/~greblus/ck/top_-b_-n_1.txt
>
> Kindest Regards,
>   W.

Thanks for details. You do not have different nice levels; that is PRI 
that is different, which is the "dynamic" priority that constantly 
changes by the scheduler. It is as I suspected, though - if you renice 
everything +19 and the problem goes away it is bad programming on the 
wine developers' part that use what's called sched_yield instead of 
proper locking between their threads. Don't concern yourself over the 
details, but at least you have a simple workaround... sigh

Cheers,
Con

[Mike Hearn]

On Tue, 17 Aug 2004 10:06:46 +0200, Andreas Mohr wrote:
> Are we doing some big no-nos in the threading area here?
> (or maybe it's just done the way it is because we cannot do it a
> different, better way?)
>
> For the whole discussion, please see
> http://bhhdoa.org.au/pipermail/ck/2004-August/000428.html

Well, apparently we don't use sched_yield, so the problem must
lie somewhere else. Maybe Con can help us out here? Alexandre says he
doesn't know what the issue is either and somebody needs to investigate. I
guess we do need to concern ourselves over the details :)