kernel 2.6 bug

kernel 2.6 bug

[Amaury Jacquot]

hi
I tried the 2.6 kernel today, and have a bug

using snd-usb-audio with the oss emulation I use a gstreamer based 
player (I know, I could use alsasink) :D

notes:
* uhci USB
* no modules support (everything in)
* preemptible kernel on

when preemptible is off, I only get the

"cannot submit datapipe for urb <blah>" messages as shown below

restarting the application gives sound again

the application works in 2.4.23 with alsa 1.0.0-rc2

I get the following trace :

bad: scheduling while atomic!
Call Trace:
  [<c011bdea>] schedule+0x70a/0x710
  [<c028ce43>] snd_pcm_oss_get_formats+0x23/0xc0
  [<c011a206>] recalc_task_prio+0xb6/0x220
  [<c011c0ec>] wait_for_completion+0x8c/0xf0
  [<c011be40>] default_wake_function+0x0/0x20
  [<c011be40>] default_wake_function+0x0/0x20
  [<c0261871>] hcd_unlink_urb+0x1a1/0x2b0
  [<c02a78e0>] snd_complete_urb+0x0/0xa0
  [<c02615c0>] hcd_submit_urb+0x110/0x190
  [<c026218a>] usb_unlink_urb+0x2a/0x30
  [<c02a7ad8>] deactivate_urbs+0xb8/0xc0
  [<c02a7d08>] snd_usb_pcm_trigger+0x58/0x60
  [<c02963b9>] snd_pcm_do_stop+0x19/0x20
  [<c0296002>] snd_pcm_action_single+0x22/0x50
  [<c0296120>] snd_pcm_action+0xf0/0x100
  [<c0296444>] snd_pcm_stop+0x14/0x20
  [<c029ac9d>] snd_pcm_update_hw_ptr+0x13d/0x1e0
  [<c029e785>] snd_pcm_lib_write1+0x5d5/0x5f0
  [<c0292d41>] resample_expand+0x271/0x340
  [<c0292dcf>] resample_expand+0x2ff/0x340
  [<c029e812>] snd_pcm_lib_write+0x72/0xa0
  [<c029e0f0>] snd_pcm_lib_write_transfer+0x0/0xc0
  [<c028c090>] snd_pcm_oss_write3+0x50/0xb0
  [<c0290721>] snd_pcm_plug_write_transfer+0x81/0xf0
  [<c028c421>] snd_pcm_oss_write2+0xb1/0x130
  [<c028c668>] snd_pcm_oss_write1+0x1c8/0x200
  [<c0299aa7>] snd_pcm_kernel_playback_ioctl+0x27/0x30
  [<c028e6c6>] snd_pcm_oss_write+0x36/0x50
  [<c0152f6c>] vfs_write+0x9c/0x100
  [<c015304d>] sys_write+0x2d/0x50
  [<c01092f7>] syscall_call+0x7/0xb

bad: scheduling while atomic!
Call Trace:
  [<c011bdea>] schedule+0x70a/0x710
  [<c011a206>] recalc_task_prio+0xb6/0x220
  [<c011c0ec>] wait_for_completion+0x8c/0xf0
  [<c011be40>] default_wake_function+0x0/0x20
  [<c011be40>] default_wake_function+0x0/0x20
  [<c0261871>] hcd_unlink_urb+0x1a1/0x2b0
  [<c02a78e0>] snd_complete_urb+0x0/0xa0
  [<c02615c0>] hcd_submit_urb+0x110/0x190
  [<c026218a>] usb_unlink_urb+0x2a/0x30
  [<c02a7ad8>] deactivate_urbs+0xb8/0xc0
  [<c02a7d08>] snd_usb_pcm_trigger+0x58/0x60
  [<c02963b9>] snd_pcm_do_stop+0x19/0x20
  [<c0296002>] snd_pcm_action_single+0x22/0x50
  [<c0296120>] snd_pcm_action+0xf0/0x100
  [<c0296444>] snd_pcm_stop+0x14/0x20
  [<c029ac9d>] snd_pcm_update_hw_ptr+0x13d/0x1e0
  [<c029e785>] snd_pcm_lib_write1+0x5d5/0x5f0
  [<c0292d41>] resample_expand+0x271/0x340
  [<c0292dcf>] resample_expand+0x2ff/0x340
  [<c029e812>] snd_pcm_lib_write+0x72/0xa0
  [<c029e0f0>] snd_pcm_lib_write_transfer+0x0/0xc0
  [<c028c090>] snd_pcm_oss_write3+0x50/0xb0
  [<c0290721>] snd_pcm_plug_write_transfer+0x81/0xf0
  [<c028c421>] snd_pcm_oss_write2+0xb1/0x130
  [<c028c668>] snd_pcm_oss_write1+0x1c8/0x200
  [<c0299aa7>] snd_pcm_kernel_playback_ioctl+0x27/0x30
  [<c028e6c6>] snd_pcm_oss_write+0x36/0x50
  [<c0152f6c>] vfs_write+0x9c/0x100
  [<c015304d>] sys_write+0x2d/0x50
  [<c01092f7>] syscall_call+0x7/0xb

cannot submit datapipe for urb 0, err = -22
cannot submit datapipe for urb 0, err = -22
cannot submit datapipe for urb 0, err = -22
cannot submit datapipe for urb 1, err = -22
cannot submit datapipe for urb 0, err = -22
cannot submit datapipe for urb 1, err = -22
cannot submit datapipe for urb 0, err = -22
cannot submit datapipe for urb 1, err = -22



-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click

Re: kernel 2.6 bug

[Clemens Ladisch]

Amaury Jacquot wrote:
> I tried the 2.6 kernel today, and have a bug
>
> using snd-usb-audio with the oss emulation
>
> bad: scheduling while atomic!
> Call Trace:
>   [<c011bdea>] schedule+0x70a/0x710
> ...
>   [<c026218a>] usb_unlink_urb+0x2a/0x30
>   [<c02a7ad8>] deactivate_urbs+0xb8/0xc0
> ...

This bug has been fixes in recent ALSA versions.

In usbaudio.c, replace the line

	/* #define SND_USE_ASYNC_UNLINK */

with

	#define SND_USB_ASYNC_UNLINK


HTH
Clemens




-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click

Re: kernel 2.6 bug

[Amaury Jacquot]

Clemens Ladisch wrote:
> Amaury Jacquot wrote:
> 
>>I tried the 2.6 kernel today, and have a bug
>>
>>using snd-usb-audio with the oss emulation
>>
>>bad: scheduling while atomic!
>>Call Trace:
>>  [<c011bdea>] schedule+0x70a/0x710
>>...
>>  [<c026218a>] usb_unlink_urb+0x2a/0x30
>>  [<c02a7ad8>] deactivate_urbs+0xb8/0xc0
>>...
> 
> 
> This bug has been fixes in recent ALSA versions.
> 
> In usbaudio.c, replace the line
> 
> 	/* #define SND_USE_ASYNC_UNLINK */
> 
> with
> 
> 	#define SND_USB_ASYNC_UNLINK

it corrected the crapping out. however, I still get tons of
"cannot submit datapipe for urb [0|1], err = -22"

> 
> HTH
> Clemens

it did, partially :D

Amaury
> 
> 




-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click