Re: Linux 2.6.10-rc2

Re: Linux 2.6.10-rc2

[Lee Revell]

Please report ALSA issues to alsa-devel@lists.sourceforge.net.  I have
added them to the cc:.

Lee

On Mon, 2004-11-15 at 15:23 -0500, Paul Blazejowski wrote:
> Starting up ut2004 or doom3 games produces ALSA segfaults, dmesg prints:
> 
> scheduling while atomic: ut2004-bin/0x00000001/8390
>  [<c0331508>] schedule+0x508/0x510
>  [<c03319a3>] schedule_timeout+0x63/0xc0
>  [<c0122c90>] process_timeout+0x0/0x10
>  [<c012306f>] msleep+0x2f/0x40
>  [<f8b9e02f>] snd_intel8x0_setup_pcm_out+0xbf/0x150 [snd_intel8x0]
>  [<f8b9e14c>] snd_intel8x0_pcm_prepare+0x8c/0xb0 [snd_intel8x0]
>  [<f8bfda84>] snd_pcm_do_prepare+0x14/0x40 [snd_pcm]
>  [<f8bfcdd8>] snd_pcm_action_single+0x38/0x80 [snd_pcm]
>  [<f8bfcfd0>] snd_pcm_action_nonatomic+0x80/0x90 [snd_pcm]
>  [<f8bfdb37>] snd_pcm_prepare+0x57/0x80 [snd_pcm]
>  [<f8c001b2>] snd_pcm_playback_ioctl1+0x52/0x310 [snd_pcm]
>  [<f8d25ae9>] snd_pcm_oss_poll+0x49/0x1a0 [snd_pcm_oss]
>  [<c016901a>] poll_freewait+0x3a/0x50
>  [<f8c00828>] snd_pcm_kernel_playback_ioctl+0x38/0x50 [snd_pcm]
>  [<f8d23026>] snd_pcm_oss_prepare+0x26/0x60 [snd_pcm_oss]
>  [<f8d2309d>] snd_pcm_oss_make_ready+0x3d/0x60 [snd_pcm_oss]
>  [<f8d2358d>] snd_pcm_oss_write1+0x3d/0x210 [snd_pcm_oss]
>  [<f8d259d0>] snd_pcm_oss_write+0x40/0x60 [snd_pcm_oss]
>  [<f8d25990>] snd_pcm_oss_write+0x0/0x60 [snd_pcm_oss]
>  [<c01563e8>] vfs_write+0xb8/0x130
>  [<c0156531>] sys_write+0x51/0x80
>  [<c010317b>] syscall_call+0x7/0xb
> 
> .config attached
> 
> Regards,
> 
> Paul B.
-- 
Lee Revell <rlrevell@joe-job.com>



-------------------------------------------------------
This SF.Net email is sponsored by: InterSystems CACHE
FREE OODBMS DOWNLOAD - A multidimensional database that combines
robust object and relational technologies, making it a perfect match
for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8

Re: Re: Linux 2.6.10-rc2

[Takashi Iwai]

At Mon, 15 Nov 2004 16:16:31 -0500,
Lee Revell wrote:
> 
> Please report ALSA issues to alsa-devel@lists.sourceforge.net.  I have
> added them to the cc:.

The attached patch should fix this.


Takashi

==
Summary: [ALSA] fix sleep in atomic during prepare callback

Fixed the sleep in spinlock during prepare callback.
This happened only on Nforce chips.

Signed-off-by: Takashi Iwai <tiwai@suse.de>

--- linux/sound/pci/intel8x0.c	15 Nov 2004 14:19:52 -0000	1.173
+++ linux/sound/pci/intel8x0.c	16 Nov 2004 09:41:47 -0000
@@ -1020,7 +1020,9 @@ static void snd_intel8x0_setup_pcm_out(i
 			 */
 			if (cnt & ICH_PCM_246_MASK) {
 				iputdword(chip, ICHREG(GLOB_CNT), cnt & ~ICH_PCM_246_MASK);
+				spin_unlock_irq(&chip->reg_lock);
 				msleep(50); /* grrr... */
+				spin_lock_irq(&chip->reg_lock);
 			}
 		} else if (chip->device_type == DEVICE_INTEL_ICH4) {
 			if (runtime->sample_bits > 16)


-------------------------------------------------------
This SF.Net email is sponsored by: InterSystems CACHE
FREE OODBMS DOWNLOAD - A multidimensional database that combines
robust object and relational technologies, making it a perfect match
for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8

Re: Re: Linux 2.6.10-rc2

[Bill Davidsen]

Takashi Iwai wrote:
> At Mon, 15 Nov 2004 16:16:31 -0500,
> Lee Revell wrote:
> 
>>Please report ALSA issues to alsa-devel@lists.sourceforge.net.  I have
>>added them to the cc:.
> 
> 
> The attached patch should fix this.
> 
> 
> Takashi
> 
> ==
> Summary: [ALSA] fix sleep in atomic during prepare callback
> 
> Fixed the sleep in spinlock during prepare callback.
> This happened only on Nforce chips.
> 
> Signed-off-by: Takashi Iwai <tiwai@suse.de>
> 
> --- linux/sound/pci/intel8x0.c	15 Nov 2004 14:19:52 -0000	1.173
> +++ linux/sound/pci/intel8x0.c	16 Nov 2004 09:41:47 -0000
> @@ -1020,7 +1020,9 @@ static void snd_intel8x0_setup_pcm_out(i
>  			 */
>  			if (cnt & ICH_PCM_246_MASK) {
>  				iputdword(chip, ICHREG(GLOB_CNT), cnt & ~ICH_PCM_246_MASK);
> +				spin_unlock_irq(&chip->reg_lock);
>  				msleep(50); /* grrr... */
> +				spin_lock_irq(&chip->reg_lock);
>  			}
>  		} else if (chip->device_type == DEVICE_INTEL_ICH4) {
>  			if (runtime->sample_bits > 16)

Might this make it into 2.6.10?

-- 
    -bill davidsen (davidsen@tmr.com)
"The secret to procrastination is to put things off until the
  last possible moment - but no longer"  -me


-------------------------------------------------------
This SF.Net email is sponsored by: InterSystems CACHE
FREE OODBMS DOWNLOAD - A multidimensional database that combines
robust object and relational technologies, making it a perfect match
for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8