From nox@saturn.kn-bremen.de  Mon Feb 28 21:40:19 2000
Return-Path: <nox@saturn.kn-bremen.de>
Received: from blaubaer.kn-bremen.de (blaubaer.kn-bremen.de [195.37.179.254])
	by hub.freebsd.org (Postfix) with ESMTP id 7D45637C37C
	for <FreeBSD-gnats-submit@freebsd.org>; Mon, 28 Feb 2000 21:39:09 -0800 (PST)
	(envelope-from nox@saturn.kn-bremen.de)
Received: from saturn.kn-bremen.de (uucp@localhost)
	by blaubaer.kn-bremen.de (8.9.1/8.9.1) with UUCP id GAA01398;
	Tue, 29 Feb 2000 06:33:24 +0100
Received: (from nox@localhost)
	by saturn.kn-bremen.de (8.9.3/8.8.5) id GAA98094;
	Tue, 29 Feb 2000 06:34:42 +0100 (CET)
Message-Id: <200002290534.GAA98094@saturn.kn-bremen.de>
Date: Tue, 29 Feb 2000 06:34:42 +0100 (CET)
From: Juergen Lock <nox@jelal.kn-bremen.de>
Sender: nox@saturn.kn-bremen.de
Reply-To: nox@jelal.kn-bremen.de
To: FreeBSD-gnats-submit@freebsd.org
Cc: esk@ira.uka.de
Subject: audio/xmms chewing CPU (and skipping too)
X-Send-Pr-Version: 3.2

>Number:         17066
>Category:       ports
>Synopsis:       audio/xmms chewing CPU (and skipping too)
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    freebsd-ports
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Mon Feb 28 21:50:00 PST 2000
>Closed-Date:    Sun Sep 17 10:32:46 PDT 2000
>Last-Modified:  Sun Sep 17 10:33:39 PDT 2000
>Originator:     Juergen Lock
>Release:        FreeBSD 3.4-STABLE i386
>Organization:
me?  organized??
>Environment:

	gettext-0.10.35, glib-1.2.7, gtk-1.2.7, libxml-1.8.6

>Description:

	Even when playing a .wav (so nothing to decompress...)
	xmms chews up all cpu cycles it can get, even hitting
	pause doesn't change that...  and it also doesn't seem
	to do any buffering at all, it keeps skipping (even when
	playing just a .wav) in situations where e.g. bplay_gramo
	didn't skip _once_.

	It looks like a really nice program but these two things make
	it so much less useful than it could be... :(

	And right now as i type this it just died like this:

	Fatal error 'Cannot set scheduling timer' at line ? in file /dnes4/usr/w/usr/home5/cvs/src31/src/lib/libc_r/uthread/uthread_kern.c (errno = ?)

	Oh and it could also stat() at least .wav files to get the
	length instead of trusting the header for that because
	thats not always right...

>How-To-Repeat:

	just start it and look at `top'.  and then listen to it
	when the system is loaded...

>Fix:

	i with i knew.  (well other than running it on a spare box
	thats always idle enough, if you have one...)

>Release-Note:
>Audit-Trail:

From: Mark Ovens <mark@ukug.uk.freebsd.org>
To: Juergen Lock <nox@jelal.kn-bremen.de>
Cc: FreeBSD-gnats-submit@freebsd.org, esk@ira.uka.de
Subject: Re: ports/17066: audio/xmms chewing CPU (and skipping too)
Date: Tue, 29 Feb 2000 17:55:54 +0000

 On Tue, Feb 29, 2000 at 06:34:42AM +0100, Juergen Lock wrote:
 > 
 > >Number:         17066
 > >Category:       ports
 > >Synopsis:       audio/xmms chewing CPU (and skipping too)
 > >Confidential:   no
 > >Severity:       non-critical
 > >Priority:       low
 > >Responsible:    freebsd-ports
 > >State:          open
 > >Quarter:        
 > >Keywords:       
 > >Date-Required:
 > >Class:          sw-bug
 > >Submitter-Id:   current-users
 > >Arrival-Date:   Mon Feb 28 21:50:00 PST 2000
 > >Closed-Date:
 > >Last-Modified:
 > >Originator:     Juergen Lock
 > >Release:        FreeBSD 3.4-STABLE i386
 > >Organization:
 > me?  organized??
 > >Environment:
 > 
 > 	gettext-0.10.35, glib-1.2.7, gtk-1.2.7, libxml-1.8.6
 > 
 > >Description:
 > 
 > 	Even when playing a .wav (so nothing to decompress...)
 > 	xmms chews up all cpu cycles it can get, even hitting
 > 	pause doesn't change that...  and it also doesn't seem
 > 	to do any buffering at all, it keeps skipping (even when
 > 	playing just a .wav) in situations where e.g. bplay_gramo
 > 	didn't skip _once_.
 > 
 > 	It looks like a really nice program but these two things make
 > 	it so much less useful than it could be... :(
 > 
 > 	And right now as i type this it just died like this:
 > 
 > 	Fatal error 'Cannot set scheduling timer' at line ? in file /dnes4/usr/w/usr/home5/cvs/src31/src/lib/libc_r/uthread/uthread_kern.c (errno = ?)
 > 
 > 	Oh and it could also stat() at least .wav files to get the
 > 	length instead of trusting the header for that because
 > 	thats not always right...
 > 
 > >How-To-Repeat:
 > 
 > 	just start it and look at `top'.  and then listen to it
 > 	when the system is loaded...
 > 
 > >Fix:
 > 
 > 	i with i knew.  (well other than running it on a spare box
 > 	thats always idle enough, if you have one...)
 > 
 
 Have you tried applying the patch in kern/16709? That fixed a whole
 load of speed related stuff for me (that only appeared in 1.0.1, 0.9.5
 was fine).
 
 > >Release-Note:
 > >Audit-Trail:
 > >Unformatted:
 > 
 > 
 > To Unsubscribe: send mail to majordomo@FreeBSD.org
 > with "unsubscribe freebsd-ports" in the body of the message
 
 -- 
 	Microsoft: Where do you want to go today?
 	Linux:     Where do you want to go tomorrow?
 	BSD:       Are you guys coming, or what?
 			              -Poster at LinuxWorld 2000
 ________________________________________________________________
       FreeBSD - The Power To Serve http://www.freebsd.org
       My Webpage http://ukug.uk.freebsd.org/~mark/
 mailto:mark@ukug.uk.freebsd.org             http://www.radan.com
 
 

From: Juergen Lock <nox@jelal.kn-bremen.de>
To: Mark Ovens <mark@dogma.freebsd-uk.eu.org>
Cc: FreeBSD-gnats-submit@freebsd.org, esk@ira.uka.de
Subject: Re: ports/17066: audio/xmms chewing CPU (and skipping too)
Date: Fri, 3 Mar 2000 19:37:34 +0100

 On Tue, Feb 29, 2000 at 05:55:54PM +0000, Mark Ovens wrote:
 > On Tue, Feb 29, 2000 at 06:34:42AM +0100, Juergen Lock wrote:
 > > 
 > > >Number:         17066
 > > >Category:       ports
 > > >Synopsis:       audio/xmms chewing CPU (and skipping too)
 >...
 
 > Have you tried applying the patch in kern/16709? That fixed a whole
 > load of speed related stuff for me (that only appeared in 1.0.1, 0.9.5
 > was fine).
 
 Ok i've looked at that patch and it won't help as the soundcard in
 that box uses a different driver (and that one does have select() btw.)
 I also tested it with the other driver (its a isa sb16 where both the
 old and new drivers work), but it made no real difference.
 
  And in the meantime i've also chatted with a linux user who said
 xmms doesn't eat as much cpu on his box (just a few % there), so i'm
 beginning to suspect a pthreads problem or something...  That would
 also explain why an entirely different audio player (freeamp) is
 chewing cpu too, tho that one doen't eat _all_ available cycles when
 its idle, just (imho) way too much.
 
  Soo, anyone who knows `our' pthreads implementation wants to take a
 look?  and if you want the unfinished cpu chewing freeamp port just
 mail me...
 
  Thanx + Regards,
 -- 
 Juergen Lock <nox.foo@jelal.kn-bremen.de>
 (remove dot foo from address to reply)
 
State-Changed-From-To: open->feedback 
State-Changed-By: sobomax 
State-Changed-When: Fri Jul 14 06:38:11 PDT 2000 
State-Changed-Why:  
Please try to verify if the problem still persists with newest xmms and let us 
know. 

Thanks! 

http://www.freebsd.org/cgi/query-pr.cgi?pr=17066 
State-Changed-From-To: feedback->closed 
State-Changed-By: ade 
State-Changed-When: Sun Sep 17 10:32:46 PDT 2000 
State-Changed-Why:  
Timeout (2 months) in feedback.  Please update your source and 
ports trees and file a new PR if the issue is still occurring 
(it doesn't here on 4.1-STABLE 9/11) 

http://www.freebsd.org/cgi/query-pr.cgi?pr=17066 
>Unformatted:
