Skip to contentSkip to conversation

snd-usb-usx2y 0.7.1 against cvs head: please test it!

PATCH Applied Karsten Wiese 5 participants 15 messages Jul 13, 2004 – Jul 15, 2004

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 snd-usb-usx2y 0.7.1 against cvs head: please test it!

Hi,

this module version sports "no sleeping" in ALSA's pcm START/STOP callbacks. It could need a little more testing, as the changes are many. Here it works well with kernels "planet CCRMA's 2.4.26" and "vanilla 2.6.7 all debug flags I found enabled".

Thanks & regards,
Karsten

P.S.: Please ignore the previously postet "[PATCH]snd-usb-usx2y 0.7"

Applied
#1 lore raw

At Tue, 13 Jul 2004 11:55:43 +0200,

Quoted 7 lines · Karsten Wiese

Hi,

this module version sports "no sleeping" in ALSA's pcm START/STOP callbacks. It could need a little more testing, as the changes are many. Here it works well with kernels "planet CCRMA's 2.4.26" and "vanilla 2.6.7 all debug flags I found enabled".

Applied to cvs. Thanks.

Takashi

-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training. Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital self defense, top technical experts, no vendor pitches, unmatched networking opportunities. Visit www.blackhat.com

#2 lore raw
Quoted 6 lines · Karsten Wiese

Hi,
this module version sports "no sleeping" in ALSA's pcm START/STOP callbacks.
It could need a little more testing, as the changes are many. Here it works well with kernels "planet CCRMA's 2.4.26" and "vanilla 2.6.7 all debug flags I found enabled".

Just built from cvs against a custom 2.6.7-cko6 kernel on a SUSE 9.1 SMP/HT box, and assured that the patch is already applied.

Erm... I get the following while modprobing snd-usb-usx2y:

WARNING: Error inserting snd_hwdep (/lib/modules/2.6.7-cko6.0smp/kernel/sound/acore/snd-hwdep.ko): Unknown symbol in module, or unknown parameter (see dmesg)
FATAL: Error inserting snd_usb_usx2y (/lib/modules/2.6.7-cko6.0smp/kernel/sound/snd-usb-usx2y.ko): Unknown symbol in module, or unknown parameter (see dmesg)

Then the relevant dmesg says:

snd_hwdep: Unknown symbol snd_kcalloc
snd_usb_usx2y: Unknown symbol snd_hwdep_new

So I guess something went wrong on my snd-usb-usx2y build. OTOH, the onboard snd-intel8x0 based is loaded and working without a glitch.

So, too pity I can't get my us224 onto the testing right now.

Any ideas?

Signature · 10 lines
rncbc aka Rui Nuno Capela
[email protected]



-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com
#3 lore raw

At Tue, 13 Jul 2004 23:12:26 +0100 (WEST),

Quoted 21 lines · Rui Nuno Capela
Quoted 6 lines · Karsten Wiese

Hi,
this module version sports "no sleeping" in ALSA's pcm START/STOP callbacks.
It could need a little more testing, as the changes are many. Here it works well with kernels "planet CCRMA's 2.4.26" and "vanilla 2.6.7 all debug flags I found enabled".

Just built from cvs against a custom 2.6.7-cko6 kernel on a SUSE 9.1 SMP/HT box, and assured that the patch is already applied.

Erm... I get the following while modprobing snd-usb-usx2y:

WARNING: Error inserting snd_hwdep (/lib/modules/2.6.7-cko6.0smp/kernel/sound/acore/snd-hwdep.ko): Unknown symbol in module, or unknown parameter (see dmesg)
FATAL: Error inserting snd_usb_usx2y (/lib/modules/2.6.7-cko6.0smp/kernel/sound/snd-usb-usx2y.ko): Unknown symbol in module, or unknown parameter (see dmesg)

Quoted 4 lines

Then the relevant dmesg says:

snd_hwdep: Unknown symbol snd_kcalloc
snd_usb_usx2y: Unknown symbol snd_hwdep_new

Did you run cvscompile? configure script must be regenerated.

Takashi

-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training. Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital self defense, top technical experts, no vendor pitches, unmatched networking opportunities. Visit www.blackhat.com

#4 lore raw
Quoted 11 lines · Takashi Iwai
Quoted 10 lines · Rui Nuno Capela

Erm... I get the following while modprobing snd-usb-usx2y:

WARNING: Error inserting snd_hwdep (/lib/modules/2.6.7-cko6.0smp/kernel/sound/acore/snd-hwdep.ko): Unknown symbol in module, or unknown parameter (see dmesg)
FATAL: Error inserting snd_usb_usx2y (/lib/modules/2.6.7-cko6.0smp/kernel/sound/snd-usb-usx2y.ko): Unknown symbol in module, or unknown parameter (see dmesg)

Quoted 8 lines
Quoted 4 lines

Then the relevant dmesg says:

snd_hwdep: Unknown symbol snd_kcalloc
snd_usb_usx2y: Unknown symbol snd_hwdep_new

Did you run cvscompile? configure script must be regenerated.

Yes, I've ran ./cvscompile. My steps were just the following:

cvs co alsa-kernel
cvs co alsa-driver
cd alsa-driver
export ALSAKERNELDIR=../alsa-kernel
./cvscompile
su -c "make install"

Something wrong?

Signature · 9 lines
rncbc aka Rui Nuno Capela
[email protected]


-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com
#5 lore raw

At Wed, 14 Jul 2004 12:37:55 +0100 (WEST),

Quoted 24 lines · Rui Nuno Capela
Quoted 11 lines · Takashi Iwai
Quoted 10 lines · Rui Nuno Capela

Erm... I get the following while modprobing snd-usb-usx2y:

WARNING: Error inserting snd_hwdep (/lib/modules/2.6.7-cko6.0smp/kernel/sound/acore/snd-hwdep.ko): Unknown symbol in module, or unknown parameter (see dmesg)
FATAL: Error inserting snd_usb_usx2y (/lib/modules/2.6.7-cko6.0smp/kernel/sound/snd-usb-usx2y.ko): Unknown symbol in module, or unknown parameter (see dmesg)

Quoted 8 lines
Quoted 4 lines

Then the relevant dmesg says:

snd_hwdep: Unknown symbol snd_kcalloc
snd_usb_usx2y: Unknown symbol snd_hwdep_new

Did you run cvscompile? configure script must be regenerated.

Yes, I've ran ./cvscompile. My steps were just the following:

Quoted 8 lines

cvs co alsa-kernel
cvs co alsa-driver
cd alsa-driver
export ALSAKERNELDIR=../alsa-kernel
./cvscompile
su -c "make install"

Something wrong?

Should be ok.

snd_kcalloc() is an obsolete function which was replaced recently with kcalloc(). I guess the modules are not installed correctly, or the detection of the kernel verion was wrong.

Takashi

-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training. Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital self defense, top technical experts, no vendor pitches, unmatched networking opportunities. Visit www.blackhat.com

#6 lore raw
Quoted 34 lines

At Wed, 14 Jul 2004 12:37:55 +0100 (WEST),

Quoted 8 lines · Rui Nuno Capela
Quoted 6 lines · Takashi Iwai
Quoted 5 lines · Rui Nuno Capela

Erm... I get the following while modprobing snd-usb-usx2y:

WARNING: Error inserting snd_hwdep (/lib/modules/2.6.7-cko6.0smp/kernel/sound/acore/snd-hwdep.ko):

Unknown

Quoted 23 lines
Quoted 5 lines
Quoted 5 lines

symbol in module, or unknown parameter (see dmesg)
FATAL: Error inserting snd_usb_usx2y (/lib/modules/2.6.7-cko6.0smp/kernel/sound/snd-usb-usx2y.ko): Unknown symbol in module, or unknown parameter (see dmesg)

Quoted 8 lines
Quoted 4 lines

Then the relevant dmesg says:

snd_hwdep: Unknown symbol snd_kcalloc
snd_usb_usx2y: Unknown symbol snd_hwdep_new

Did you run cvscompile? configure script must be regenerated.

Yes, I've ran ./cvscompile. My steps were just the following:

cvs co alsa-kernel
cvs co alsa-driver
cd alsa-driver
export ALSAKERNELDIR=../alsa-kernel
./cvscompile
su -c "make install"

Quoted 8 lines
Quoted 1 line

Something wrong?

Should be ok.

snd_kcalloc() is an obsolete function which was replaced recently with kcalloc(). I guess the modules are not installed correctly, or the detection of the kernel verion was wrong.

Checkout alsa-tools and rebuild your usx2yloader

martin

-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training. Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital self defense, top technical experts, no vendor pitches, unmatched networking opportunities. Visit www.blackhat.com

#7 lore raw
Quoted 34 lines
Quoted 33 lines · Martin Langer
Quoted 31 lines · Rui Nuno Capela
Quoted 11 lines · Takashi Iwai
Quoted 10 lines · Rui Nuno Capela

Erm... I get the following while modprobing snd-usb-usx2y:

WARNING: Error inserting snd_hwdep (/lib/modules/2.6.7-cko6.0smp/kernel/sound/acore/snd-hwdep.ko):
Unknown symbol in module, or unknown parameter (see dmesg)
FATAL: Error inserting snd_usb_usx2y (/lib/modules/2.6.7-cko6.0smp/kernel/sound/snd-usb-usx2y.ko):
Unknown symbol in module, or unknown parameter (see dmesg)

Quoted 8 lines
Quoted 4 lines

Then the relevant dmesg says:

snd_hwdep: Unknown symbol snd_kcalloc
snd_usb_usx2y: Unknown symbol snd_hwdep_new

Did you run cvscompile? configure script must be regenerated.

Yes, I've ran ./cvscompile. My steps were just the following:

cvs co alsa-kernel
cvs co alsa-driver
cd alsa-driver
export ALSAKERNELDIR=../alsa-kernel
./cvscompile
su -c "make install"

Something wrong?

Quoted 9 lines
Quoted 6 lines

Should be ok.

snd_kcalloc() is an obsolete function which was replaced recently with kcalloc(). I guess the modules are not installed correctly, or the detection of the kernel verion was wrong.

Checkout alsa-tools and rebuild your usx2yloader

OK. It was something I've screwed up. A second 'make install' and everything is up and running on my US-224, at least as it was before.

JACK seems to work fine, and MIDI also gets it thru. So chalk one for US-224 and 2.6.7-cko6.

Anything I should check deeper?

Signature · 10 lines
rncbc aka Rui Nuno Capela
[email protected]



-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
#8 lore raw
In reply to Takashi Iwai

testing on 2.6.7-bk20 i get the following message when starting JACK:

ALSA lib pcm_hw.c:549:(snd_pcm_hw_start) SNDRV_PCM_IOCTL_START failed: Broken pipe
could not start playback (Broken pipe)
jackd watchdog: timeout - killing jackd

the system log says:

Jul 14 18:40:53 t10 vmunix: 
ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:246: 
-28
Jul 14 18:40:53 t10 vmunix: 
ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:347: 
Sequence Error!(ep=10;nuc=1,frame=860)
Jul 14 18:40:53 t10 vmunix: 

ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:347:
Sequence Error!(ep=8;nuc=1,frame=860)

/werner

-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training. Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital self defense, top technical experts, no vendor pitches, unmatched networking opportunities. Visit www.blackhat.com

#9 lore raw
Quoted 8 lines

testing on 2.6.7-bk20 i get the following message when starting JACK:

ALSA lib pcm_hw.c:549:(snd_pcm_hw_start) SNDRV_PCM_IOCTL_START failed:
Broken pipe
could not start playback (Broken pipe)
jackd watchdog: timeout - killing jackd

Quoted 6 lines

the system log says:

Jul 14 18:40:53 t10 vmunix:

ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:246:
-28

#define ENOSPC 28 /* No space left on device */

If you didn't try yet,
please start Jack with the -S alsa driver option to use 16bit format. Using 24bit on kernels with "uhci debugging" I get lots of dmesg warnings complaining about overusing the usb - bandwidth although everything works fine. Using 16bit the kernel doesn't complain.
Maybe usb bandwidth checks are broken?

Quoted 10 lines
Jul 14 18:40:53 t10 vmunix:
ALSA
/home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:347:
Sequence Error!(ep=10;nuc=1,frame=860)
Jul 14 18:40:53 t10 vmunix:

ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:347:
Sequence Error!(ep=8;nuc=1,frame=860)

/werner

-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training. Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital self defense, top technical experts, no vendor pitches, unmatched networking opportunities. Visit www.blackhat.com

#10 lore raw
Quoted 15 lines · Karsten Wiese
Quoted 8 lines

testing on 2.6.7-bk20 i get the following message when starting JACK:

ALSA lib pcm_hw.c:549:(snd_pcm_hw_start) SNDRV_PCM_IOCTL_START failed:
Broken pipe
could not start playback (Broken pipe)
jackd watchdog: timeout - killing jackd

Quoted 6 lines

the system log says:

Jul 14 18:40:53 t10 vmunix:

ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:24 6: -28

Quoted 17 lines

#define ENOSPC 28 /* No space left on device */

If you didn't try yet,
please start Jack with the -S alsa driver option to use 16bit format. Using 24bit on kernels with "uhci debugging" I get lots of dmesg warnings complaining about overusing the usb - bandwidth although everything works fine. Using 16bit the kernel doesn't complain.
Maybe usb bandwidth checks are broken?

Quoted 8 lines
Jul 14 18:40:53 t10 vmunix:
ALSA
/home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:34
7: Sequence Error!(ep=10;nuc=1,frame=860)
Jul 14 18:40:53 t10 vmunix:

ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:34 7: Sequence Error!(ep=8;nuc=1,frame=860)

starting jackd with -S (force 16bit format) works!

/werner

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click

#11 lore raw
Quoted 34 lines · Werner Schweer
Quoted 15 lines · Karsten Wiese
Quoted 8 lines

testing on 2.6.7-bk20 i get the following message when starting JACK:

ALSA lib pcm_hw.c:549:(snd_pcm_hw_start) SNDRV_PCM_IOCTL_START failed:
Broken pipe
could not start playback (Broken pipe)
jackd watchdog: timeout - killing jackd

Quoted 6 lines

the system log says:

Jul 14 18:40:53 t10 vmunix:

ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:
24 6: -28

Quoted 17 lines

#define ENOSPC 28 /* No space left on device */

If you didn't try yet,
please start Jack with the -S alsa driver option to use 16bit format. Using 24bit on kernels with "uhci debugging" I get lots of dmesg warnings complaining about overusing the usb - bandwidth although everything works fine. Using 16bit the kernel doesn't complain.
Maybe usb bandwidth checks are broken?

Quoted 8 lines
Jul 14 18:40:53 t10 vmunix:
ALSA
/home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:
34 7: Sequence Error!(ep=10;nuc=1,frame=860)
Jul 14 18:40:53 t10 vmunix:

ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.c:
34 7: Sequence Error!(ep=8;nuc=1,frame=860)

Quoted 2 lines

starting jackd with -S (force 16bit format) works!

but only for about 20 minutes, then jackd dies with same error, same system error message :-(

/werner

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click

#12 lore raw

Am Mittwoch 14 Juli 2004 20:34 schrieb Werner Schweer:

Quoted 35 lines · Werner Schweer
Quoted 34 lines · Karsten Wiese
Quoted 32 lines
Quoted 8 lines

testing on 2.6.7-bk20 i get the following message when starting JACK:

ALSA lib pcm_hw.c:549:(snd_pcm_hw_start) SNDRV_PCM_IOCTL_START
failed: Broken pipe
could not start playback (Broken pipe)
jackd watchdog: timeout - killing jackd

Quoted 6 lines

the system log says:

Jul 14 18:40:53 t10 vmunix:

ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.
c: 24 6: -28

#define ENOSPC 28 /* No space left on device */

If you didn't try yet,
please start Jack with the -S alsa driver option to use 16bit format. Using 24bit on kernels with "uhci debugging" I get lots of dmesg warnings complaining about overusing the usb - bandwidth although everything works fine. Using 16bit the kernel doesn't complain. Maybe usb bandwidth checks are broken?

Quoted 8 lines
Jul 14 18:40:53 t10 vmunix:
ALSA
/home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.
c: 34 7: Sequence Error!(ep=10;nuc=1,frame=860)
Jul 14 18:40:53 t10 vmunix:

ALSA /home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudio.
c: 34 7: Sequence Error!(ep=8;nuc=1,frame=860)

Quoted 4 lines
Quoted 1 line

starting jackd with -S (force 16bit format) works!

but only for about 20 minutes, then jackd dies with same error, same system error message :-(

if your kernel is currently configured with CONFIG_USB_BANDWIDTH=y (and you like to spend some more time testing):
Please recompile the kernel without that option and retry. the relevant part of the .config would look like:

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

#
# Miscellaneous USB options
#
CONFIG_USB_DEVICEFS=y
# CONFIG_USB_BANDWIDTH is not set
# CONFIG_USB_DYNAMIC_MINORS is not set
<<<<<<

karsten

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click

#13 lore raw
Quoted 3 lines · Karsten Wiese

Am Mittwoch 14 Juli 2004 20:34 schrieb Werner Schweer:

Quoted 2 lines · Werner Schweer
Quoted 1 line · Karsten Wiese

...

Quoted 21 lines
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
Quoted 2 lines

/home/ws/projects/alsa/alsa-driver/kbuild/../usb/usx2y/usbusx2yaudi o. c: 34 7: Sequence Error!(ep=8;nuc=1,frame=860)

starting jackd with -S (force 16bit format) works!

but only for about 20 minutes, then jackd dies with same error, same system error message :-(

if your kernel is currently configured with CONFIG_USB_BANDWIDTH=y (and you like to spend some more time testing):
Please recompile the kernel without that option and retry. the relevant part of the .config would look like:

#
# Miscellaneous USB options
#
CONFIG_USB_DEVICEFS=y
# CONFIG_USB_BANDWIDTH is not set
# CONFIG_USB_DYNAMIC_MINORS is not set
<<<<<<

Quoted 1 line

karsten

without USB_BANDWIDTH enabled all works fine now! I tested JACK with period size 144 / 2 buffers / 24 samples and it now works very stable.

/werner

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click

#14 lore raw
In reply to Karsten Wiese
Quoted 6 lines · Karsten Wiese

Hi,

this module version sports "no sleeping" in ALSA's pcm START/STOP callbacks. It could need a little more testing, as the changes are many. Here it works well with kernels "planet CCRMA's 2.4.26" and "vanilla 2.6.7 all debug flags I found enabled".

No problems on debian with 2.4.25-rc1 using an us122 :)

martin

Signature · 9 lines
                                     Living on earth may be expensive,
                   but it includes an annual free trip around the sun. 


-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com