[PATCH] usx2y jack driver

[PATCH] usx2y jack driver

[Karsten Wiese]

[-- Attachment #1: Type: text/plain, Size: 1071 bytes --]

hi,

this adds a hw_dep device to snd-usb-usx2y to communicate pcm in native usb 
chuncs with userspace.
it also adds a usx2y driver to jack, so you can play with it.

works fine here.
I can reliably use 128frames 2periods now on kernel 2.6.9-rc2-mm1-S1.
Without crackles of course.
(On a 500MHz PIII with UHCI-usb. Problems with OHCI may still be there, don't 
know.)

The tar.bz2 file is made of 4 parts:

Apply snd-usb-usx2y.patch.0.7.3+hwdep_pcm to alsa-kernel/usb, 
snd-usb-usx2y.patch.0.7.3+hwdep_pcm_asound.h to alsa-kernel/include/asound.h.

Apply jack_usx2y_driver.patch to jack and cp usx2y to jack/drivers.

To build Jack, you'll have to call autogen.sh and then configure.
The tweaked snd-usb-usx2y will propably only build in linux-kernel. I tried 
nothing else.
By the way: is there a script to cp alsa-kernel source to the linux kernel 
source?

Please comment.

Best regards,
Karsten

P.S.:
This snd-usb-usx2y keeps on working exactly as before, if it is used as 
before.
New behaviour will only show, if you start jackd like this:
	jackd -dusx2y -dhw:x


[-- Attachment #2: alsa+jack_usx2y_driver.tar.bz2 --]
[-- Type: application/x-tbz, Size: 28070 bytes --]

Re: [PATCH] usx2y jack driver

[Rui Nuno Capela]

Hi Karsten,
>
> this adds a hw_dep device to snd-usb-usx2y to communicate pcm in native
> usb chuncs with userspace. it also adds a usx2y driver to jack, so you
> can play with it.
>
> works fine here.
> I can reliably use 128frames 2periods now on kernel 2.6.9-rc2-mm1-S1.
> Without crackles of course. (On a 500MHz PIII with UHCI-usb. Problems
> with OHCI may still be there, don't know.)
>

Amazing! I will check this out soon on my US-224. I'm a bit busy right
now, so can't afford to trash my alsa/jack environment, at least for the
time being.

BTW, that OHCI issue is really still open and quite ugly (cf. alsa-driver
bug #0000425). It makes my US-224 almost useless on my laptop :( Every
time I try to open the pcm device I get a fatal kernel oops. Midi is ok
however.

I must remember you that this all used to work satisfactorily around
alsa-driver-1.0.5a inclusive. IIRC it was that "no sleeping in ALSA's pcm
START/STOP callback" patch (as of CVS circa July 12-13) that broke
something, unfortunately I'm sure.

If you find the patience/time to lead me on debugging this, I'll
appreciate a lot.

Cheers,
-- 
rncbc aka Rui Nuno Capela
rncbc@rncbc.org



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

usx2y jack driver

[Rui Nuno Capela]

Hi,

>Karsten Wiese wrote:
>>
>> this adds a hw_dep device to snd-usb-usx2y to communicate pcm in native
>> usb chuncs with userspace. it also adds a usx2y driver to jack, so you
>> can play with it.
>>
>> works fine here.
>> I can reliably use 128frames 2periods now on kernel 2.6.9-rc2-mm1-S1.
>> Without crackles of course. (On a 500MHz PIII with UHCI-usb. Problems
>> with OHCI may still be there, don't know.)
>>
>
> Amazing! I will check this out soon on my US-224. I'm a bit busy right
> now, so can't afford to trash my alsa/jack environment, at least for the
> time being.
>

That's right.

I've already worked out this new usx2y jack driver with my US-224 and the
results are really excellent! Kudos to Karten!

No crackles at all, while starting jackd -R -d usx2y -p 128 -n 2. A real
breakthrough achievement!

Real good stuff, considering that while on these settings with the stock
alsa jack driver my ears could not stand the annoyance of all the boiling
:) The minimum alsa settings were -p256 -n2, even barely crackling.

This has been just tested on my SUSE 9.1 box, P4 2.8GHz HT, happy now with
a 2.6.9-rc2-mm3-S6 SMP/HT (full VP), UHCI USB controller.

I'm so happy that even qjackctl has already experimental support for this
usx2y driver thing (qjackctl 0.2.11.9). You can check it out from CVS
(sf.net lag time applies here :)

I have also setup a web directory where I'll keep related packages files
(e.g. jack RPMS for SUSE 9.1 are in there).

It's here:

    http://www.rncbc.org/usx2y/

Please note that this will be only maintained while being experimental.

So here it goes, my lobbying but serious vote to include this usx2y driver
in the official JACK CVS tree.

>
> BTW, that OHCI issue is really still open and quite ugly (cf.
> alsa-driver bug #0000425). It makes my US-224 almost useless on my
> laptop :( Every time I try to open the pcm device I get a fatal kernel
> oops. Midi is ok however.
>
> I must remember you that this all used to work satisfactorily around
> alsa-driver-1.0.5a inclusive. IIRC it was that "no sleeping in ALSA's
> pcm START/STOP callback" patch (as of CVS circa July 12-13) that broke
> something, unfortunately I'm sure.
>
> If you find the patience/time to lead me on debugging this, I'll
> appreciate a lot.
>

I hadn't the time yet to test the new usx2y kernel and jack code on this
yet. This OHCI issue is probably still roaring its ugly head.

Bye now.
-- 
rncbc aka Rui Nuno Capela
rncbc@rncbc.org



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: [Jackit-devel] usx2y jack driver

[Lee Revell]

On Fri, 2004-09-24 at 18:31, Rui Nuno Capela wrote:
> I've already worked out this new usx2y jack driver with my US-224 and the

Maybe I am missing something.  Why does this device need a specific jack
driver, as opposed to using the ALSA driver?  Wouldn't it be better to
improve the ALSA driver so that it can be used with jack?

Lee



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: usx2y jack driver

[Karsten Wiese]

Am Samstag 25 September 2004 04:15 schrieb Lee Revell:
> On Fri, 2004-09-24 at 18:31, Rui Nuno Capela wrote:
> > I've already worked out this new usx2y jack driver with my US-224 and the
>
> Maybe I am missing something.  Why does this device need a specific jack
> driver, as opposed to using the ALSA driver?  Wouldn't it be better to
> improve the ALSA driver so that it can be used with jack?
>
Its usb's unableness to atomically handle power of 2 period sized data chuncs,
what led to: 
This usx2y-alsa-jack driver pair uses a hardware dependant alsa-device for 
mmaped pcm transport.
Advantage achieved:
	The usb_hcd places reads/writes pcm data into dma-memory.
	That memory is 	mmaped by jack's usx2y driver.
	Jack's usx2y driver is the first/last to read/write pcm data.
	Read/write is a combination of power of 2 period shaping and
	float/int conversation.
	Compared to mainline alsa/jack we leave out power of 2 period shaping inside
	snd-usb-* which needs memcpy() and additional buffers.
	As a side effect possible unwanted pcm-data coruption resulting of
	plain alsa's snd-usb-usx2y period shaping scheme falls away.
	Result is sane jack operation at buffering schemes down to 128frames,
	2 periods.
	plain usx2y alsa mode is able to achieve 64frames, 4periods, but only at the
	cost of easier triggered i.e. aeolus xruns (128 or 256frames,
	2periods works but is useless cause of crackling).

New functionality would possibly be better integrated into mainlines at more 
apropriate places.
Target could be: jack mmaping power of 2 period sized float-pcm buffers and 
letting alsa (snd-usb-lib - alsa-lib - plugin) do power of 2 period shaping 
and float/int conversation.

erm, but I'd rather make some sounds with this stuff before i dig into that 
deeper. And there is still that ohci issue.

Best,
Karsten


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: [Jackit-devel] usx2y jack driver

[Rui Nuno Capela]

Karsten Wiese wrote:

> Lee Revell wrote:
>>
>> Maybe I am missing something.  Why does this device need a specific
>> jack driver, as opposed to using the ALSA driver?  Wouldn't it be
>> better to improve the ALSA driver so that it can be used with jack?
>>

One thing I'm sure as my ears can't lie: the generic jack_alsa driver
crackles quite badly (no xruns however), the jack_usx2y don't. Period. :)

>
> Its usb's unableness to atomically handle power of 2 period sized data
> chuncs, what led to:
> This usx2y-alsa-jack driver pair uses a hardware dependant alsa-device
> for mmaped pcm transport.
> Advantage achieved:
> 	The usb_hcd places reads/writes pcm data into dma-memory.
> 	That memory is 	mmaped by jack's usx2y driver.
> 	Jack's usx2y driver is the first/last to read/write pcm data.
> 	Read/write is a combination of power of 2 period shaping and
> 	  float/int conversation.
> 	Compared to mainline alsa/jack we leave out power of 2 period
>         shaping inside snd-usb-* which needs memcpy() and additional
>         buffers.
> 	As a side effect possible unwanted pcm-data coruption resulting
>         of plain alsa's snd-usb-usx2y period shaping scheme falls away.
> 	Result is sane jack operation at buffering schemes down to
>         128frames, 2 periods.
> 	plain usx2y alsa mode is able to achieve 64frames, 4periods, but
>         only at the cost of easier triggered i.e. aeolus xruns (128 or
>         256frames, 2periods works but is useless cause of crackling).
>
> New functionality would possibly be better integrated into mainlines at
> more apropriate places. Target could be: jack mmaping power of 2 period
> sized float-pcm buffers and letting alsa (snd-usb-lib - alsa-lib -
> plugin) do power of 2 period shaping and float/int conversation.
>
> erm, but I'd rather make some sounds with this stuff before i dig into
> that deeper. And there is still that ohci issue.
>

FYI, on my laptop, that OHCI issue is still a serious showstopper. I am
collecting some log/trace data to file and append on alsa's bugtracker
(bug #0000425).

One thing I've noticed, but just related to VP, is that while having
hardirq-preempt=1 and /proc/irq/10/ohci_hcd/threaded=1, the kernel doesn't
crash when opening the PCM hw device (either via jackd or aplay). But
still useless or zombie anyway. It does crash however when I unplug my
US-224 (switch its power down).

With hardirq-preempt=0 or /proc/irq/10/ohci_hcd/threaded=0, I get the same
old bad behavior: a kernel oops somewhere on the ohci_irq handling, almost
immediately when the PCM device is open (e.g. by jackd or aplay).

If someone have an idea on how to dig on this, like some debugging
silver-bullet :) I'll very thankful. Ingo Molnar has taught me the serial
console method but, guess what, this is a post-modern laptop: it doesn't
have a serial port. Duh?

Cheers.
-- 
rncbc aka Rui Nuno Capela
rncbc@rncbc.org



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: usx2y jack driver

[Lee Revell]

On Sat, 2004-09-25 at 11:10, Rui Nuno Capela wrote:
> Karsten Wiese wrote:

Thanks for the excellent description of the driver.

> If someone have an idea on how to dig on this, like some debugging
> silver-bullet :) I'll very thankful. Ingo Molnar has taught me the serial
> console method but, guess what, this is a post-modern laptop: it doesn't
> have a serial port. Duh?
> 

This is a really good point, I asked about it on LKML.  I don't think an
equivalent procedure is currently available for a legacy-free machine.  
This will become a serious problem as such machines become more common,
as the serial console is the only way to diagnose some bugs.

For now you have to get by with magic sysrq.

Lee



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: usx2y jack driver

[Rui Nuno Capela]

Lee Revell wrote:
>
> For now you have to get by with magic sysrq.
>

That would be real magic to get SysRq+T to show up anything. It doesn't.
Nor any other SysRq key comination. And just in case you'll ask next, I do
have that magic included by kernel config, rest assured ;)

I guess the only chance I have right now, as kindly suggested by Karsten,
is about taking a photo of the kernel oops screens (which I have done
already, a couple of times), feeding it into some ocr, and disassemble
that with ksymoops. Just thinking about it makes me feel the creeps.... 
:)

Oh my!
-- 
rncbc aka Rui Nuno Capela
rncbc@rncbc.org




-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: usx2y jack driver

[Lee Revell]

On Sat, 2004-09-25 at 15:56, Rui Nuno Capela wrote:
> Lee Revell wrote:
> >
> > For now you have to get by with magic sysrq.
> >
> 
> That would be real magic to get SysRq+T to show up anything. It doesn't.
> Nor any other SysRq key comination. And just in case you'll ask next, I do
> have that magic included by kernel config, rest assured ;)
> 
> I guess the only chance I have right now, as kindly suggested by Karsten,
> is about taking a photo of the kernel oops screens (which I have done
> already, a couple of times), feeding it into some ocr, and disassemble
> that with ksymoops. Just thinking about it makes me feel the creeps.... 
> :)
> 

You tried Ingo's suggestion of making the keyboard aka i8042 IRQ
non-threaded, and still sysrq is not magic?

Lee



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Re: [Jackit-devel] usx2y jack driver

[Lee Revell]

On Sat, 2004-09-25 at 11:10, Rui Nuno Capela wrote:
> Karsten Wiese wrote:
> > erm, but I'd rather make some sounds with this stuff before i dig into
> > that deeper. And there is still that ohci issue.
> >
> 
> FYI, on my laptop, that OHCI issue is still a serious showstopper. I am
> collecting some log/trace data to file and append on alsa's bugtracker
> (bug #0000425).

One other idea.  Do you have CONFIG_DEBUG_SPINLOCK_SLEEP enabled?  If
this problem is indeed related to the sleeping in PCM start/stop
callbacks, and you enable this option, then I believe you would get a
"Bad: scheduling while atomic" error in the syslog before the machine
hangs.

Lee



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: usx2y jack driver

[Rui Nuno Capela]

Lee Revell wrote:
> Rui Nuno Capela wrote:
>>
>> FYI, on my laptop, that OHCI issue is still a serious showstopper. I am
>> collecting some log/trace data to file and append on alsa's bugtracker
>> (bug #0000425).
>
> One other idea.  Do you have CONFIG_DEBUG_SPINLOCK_SLEEP enabled?  If
> this problem is indeed related to the sleeping in PCM start/stop
> callbacks, and you enable this option, then I believe you would get a
> "Bad: scheduling while atomic" error in the syslog before the machine
> hangs.
>

Yes I have CONFIG_DEBUG_SPINLOCK_SLEEP enabled and nothing shows up on the
logs, at least as you say. I have already conducted my tests with
netconsole as you and Ingo suggested, and posted some early results on
alsa's bugtracker.

Hope to light up some bulb on Karsten's head ;)

Thank.
-- 
rncbc aka Rui Nuno Capela
rncbc@rncbc.org



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

usx2y jack driver: missing usx2yhwdeppcm.h

[Rui Nuno Capela]

Hi Karsten,

>
> The tar.bz2 file is made of 4 parts:
>
> Apply snd-usb-usx2y.patch.0.7.3+hwdep_pcm to alsa-kernel/usb,
> snd-usb-usx2y.patch.0.7.3+hwdep_pcm_asound.h to
> alsa-kernel/include/asound.h.
>
> Apply jack_usx2y_driver.patch to jack and cp usx2y to jack/drivers.
>
> To build Jack, you'll have to call autogen.sh and then configure.
> The tweaked snd-usb-usx2y will propably only build in linux-kernel. I
> tried nothing else.

I got the snd-usb-usx2y kernel module part done OK.

The jack usx2y driver part is missing the "usx2yhwdeppcm.h" header file.
Apparently it was included in the tarball as a symlink to some local file
of yours. Can't build jack without it.

Would you please send it to us?

Cheers.
-- 
rncbc aka Rui Nuno Capela
rncbc@rncbc.org



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: usx2y jack driver: missing usx2yhwdeppcm.h

[Karsten Wiese]

Am Freitag 24 September 2004 17:28 schrieb Rui Nuno Capela:
> Hi Karsten,
>
> > The tar.bz2 file is made of 4 parts:
> >
> > Apply snd-usb-usx2y.patch.0.7.3+hwdep_pcm to alsa-kernel/usb,
> > snd-usb-usx2y.patch.0.7.3+hwdep_pcm_asound.h to
> > alsa-kernel/include/asound.h.
> >
> > Apply jack_usx2y_driver.patch to jack and cp usx2y to jack/drivers.
> >
> > To build Jack, you'll have to call autogen.sh and then configure.
> > The tweaked snd-usb-usx2y will propably only build in linux-kernel. I
> > tried nothing else.
>
> I got the snd-usb-usx2y kernel module part done OK.
>
> The jack usx2y driver part is missing the "usx2yhwdeppcm.h" header file.
> Apparently it was included in the tarball as a symlink to some local file
> of yours. Can't build jack without it.
>
> Would you please send it to us?

you propably already have it: its exactly the same as in 
alsa-kernel/usb/usx2y. make the symlink point to where 
alsa-kernel/usb/usx2y/usx2yhwdeppcm.h is on your system.

best,
karsten


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

jack crackles using ice1712 or es1968 recently intoduced? was: usx2y jack driver

[Karsten Wiese]

<alsa and jack readers may skip the first lines>
Am Sonntag 26 September 2004 11:12 schrieben Sie:
> > P.S.: Will you make aeolus stops data persistent somewhen?
>
> I'm not sure what exactly you mean by 'stops data' --- the
> definition of the stops themselves or the stop combinations
> you use to play.
>
I mean the data calculated at aeolus startup or on retune.

> Both can be saved iff the stops directory and everything
> inside it is writeable for the user. If you are using a
> packaged distribution (e.g. CCRMA), the stops dir will be
> in a system-wide place and then you can't save anything.
> Just put it in your home dir and use the -S option or
> put
>
> Aeolus.stops: /home/<your_login>/stops-0.0.9
does it have to be that name stops-0.0.9?
>
> in your ~/.aeolusrc.
>
> The stop combination memories and midi routing are then
> saved when you click 'Save' in the main window.
>
> (All this assuming you are using aeolus-0.2.1)

My setup looks alike, but aeolus still takes its minute to boot,
which it shurely deserves ;-)

Just rechecked aeolus - jack - snd_es1968 at 128frames/2periods:
Crackles en masse :-/
I thought I have used it not too long ok without Crackles.
Everything OK with playback only:
/usr/bin/jackd -R -dalsa -Phw:0 -p128 -n2

The same with zynaddsubfx: Crackles with input & output.
output only ok.

using this (standard snd-usb-usx2y, 2ins 2outs, 4periods)
/usr/bin/jackd -R --unmlock -dalsa -dhw:1 -i2 -n4 -p64 -r48000
is still ok.
Are there some subtle changes in alsa/jack relating ice1712 and es1968 
recently introduced?
like in alsa pcm device syncing code, which jackd uses to simultaneously 
start/stop capture and playback?

Best,
Karsten



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: jack crackles using ice1712 or es1968 recently intoduced? was: usx2y jack driver

[Fons Adriaensen]

On Sun, Sep 26, 2004 at 05:24:53PM +0200, Karsten Wiese wrote:
<alsa and jack readers may skip the first lines>

> I mean the data calculated at aeolus startup or on retune.

The actual wavetables. These depend on four things:

1. stop definitions
2. tuning
3. temperament
4. sample frequency

So there are a lot of possible combinations. Most users will
use the same one most of the time, so it would make sense to
cache the wavetables somehow. It's one of the things I may
implement some rainy day... 

> My setup looks alike, but aeolus still takes its minute to boot,
> which it shurely deserves ;-)

:-) A real organ takes some time to build up the air pressure :-)
(agreed, less than a minute)

> Are there some subtle changes in alsa/jack relating ice1712 and es1968 
> recently introduced?
> like in alsa pcm device syncing code, which jackd uses to simultaneously 
> start/stop capture and playback?

I strongly suspect there are, but it's difficult to point out what
exactly has changed. I recently upgraded to SL9.1. This didn't work
at all (for audio). After an on-line update of the kernel (now
2.6.5-7.104-default) and instaling a more recent JACK and qjackctl
things are usable again. I will upgrade the kernel again when the
dust settles down.

I think your suspicion of the syncing code is warranted. Also the
snd-usb-audio remapping from 48 frames to 2^n is suspect, but that's
of course one of the things usx2y is meant to solve (haven't tried
it yet).

Besides the Terratec (ice1712) I also have an Edirol UA-5, and this
actually worked better (with snd-usb-audio) before the upgrade than after.

Meanwhile I found out some more about the xruns reported in my previous
message. The ones I have when Aeolus is calculating its wavetables with
-p64 or -p128 are due to memory allocation in a non-RT thread (blocks of
about one Mbyte at a time). Probably an -mm kernel would fix that. 
The cracks that remain afterwards are triggered by qjackctl updating its
connection diplay every 10 seconds. No idea what's the mechanism here.

But all by all, the system I have now works quite well with -p256, and
that's about all I need ATM.

-- 
Fons







-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php