August 2008 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] HDA Power management.
by Tobin Davis 06 Aug '08

06 Aug '08

Ok, I've been studying the code for a couple of weeks now, and testing on a mobile type development platform with a Realtek ALC262, but I have not been able to get the audio codec to enter low power mode (D3). I can see it switch when running that other OS. When running, the chip draws ~400mw, but drops to 40mw after switching to low power mode. How do I get it to do the same with the current driver? I'm using the git tip for alsa, and 2.6.26 final kernel. I have tried setting the power mode to D3 using the hda-tool from Takashi, but the codec reverts to D0 (full power) after a second or two, even though there is nothing touching the audio in my test environment. -- Tobin Davis The beauty of a pun is in the "Oy!" of the beholder.

2 1

[alsa-devel] [PATCH] ALSA: wss_lib: rework snd_ad1848_probe()
by Rene Herman 06 Aug '08

06 Aug '08

Hi Takashi. This reworks snd_ad1848_probe(), making for IMO significantly more straightforward reading. The original had a somewhat odd try-loop where it not only waited for the codec to come out of init but also retried a failed basic detect which seems to not make sense. Except for that, the detection logic should be unchanged. I am by the way rather unsure what it is that the spin_lock is trying to protect us from here. What's the error scenario if we just not take it? We aren't live yet... Next step is folding this into snd_wss_probe() directly. Rene.

2 2

[alsa-devel] [PATCH] ALSA: ASoC V2: Improve pause/unpause performance in Freescale 8610 drivers
by Timur Tabi 06 Aug '08

06 Aug '08

Add support for true pause and unpause. Without this, mplayer will drop some audio (less than one second, but still noticeable) when pausing playback. Remove support for PM suspend and resume from the trigger function, since the driver doesn't support PM anyway. Optimize the delay after starting capture. Instead of delaying 1ms, the driver now polls the hardware. The new delay is shorter by over 90% yet still effective. Signed-off-by: Timur Tabi <timur(a)freescale.com> --- This patch is for ASoC V2 only. sound/soc/fsl/fsl_dma.c | 3 ++- sound/soc/fsl/fsl_ssi.c | 17 +++++++++-------- 2 files changed, 11 insertions(+), 9 deletions(-) diff --git a/sound/soc/fsl/fsl_dma.c b/sound/soc/fsl/fsl_dma.c index 83f9c1d..649eac6 100644 --- a/sound/soc/fsl/fsl_dma.c +++ b/sound/soc/fsl/fsl_dma.c @@ -122,7 +122,8 @@ struct fsl_dma_private { static const struct snd_pcm_hardware fsl_dma_hardware = { .info = SNDRV_PCM_INFO_INTERLEAVED | SNDRV_PCM_INFO_MMAP | - SNDRV_PCM_INFO_MMAP_VALID, + SNDRV_PCM_INFO_MMAP_VALID | + SNDRV_PCM_INFO_PAUSE, .formats = FSLDMA_PCM_FORMATS, .rates = FSLDMA_PCM_RATES, .rate_min = 5512, diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 7ccd37d..916743f 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -409,28 +409,29 @@ static int fsl_ssi_trigger(struct snd_pcm_substream *substream, int cmd, switch (cmd) { case SNDRV_PCM_TRIGGER_START: - case SNDRV_PCM_TRIGGER_RESUME: + clrbits32(&ssi->scr, CCSR_SSI_SCR_SSIEN); case SNDRV_PCM_TRIGGER_PAUSE_RELEASE: if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) { - clrbits32(&ssi->scr, CCSR_SSI_SCR_SSIEN); setbits32(&ssi->scr, CCSR_SSI_SCR_SSIEN | CCSR_SSI_SCR_TE); } else { - clrbits32(&ssi->scr, CCSR_SSI_SCR_SSIEN); setbits32(&ssi->scr, CCSR_SSI_SCR_SSIEN | CCSR_SSI_SCR_RE); /* - * I think we need this delay to allow time for the SSI - * to put data into its FIFO. Without it, ALSA starts - * to complain about overruns. + * Wait until the SSI has filled its FIFO. Without this + * delay, ALSA complains about overruns. When the FIFO + * is full, the DMA controller initiates its first + * transfer. Until then, however, the DMA's DAR + * register is zero, which translates to an + * out-of-bounds pointer. This makes ALSA think an + * overrun has occurred. */ - mdelay(1); + while (!(in_be32(&ssi->sisr) & CCSR_SSI_SISR_RFF0)); } break; case SNDRV_PCM_TRIGGER_STOP: - case SNDRV_PCM_TRIGGER_SUSPEND: case SNDRV_PCM_TRIGGER_PAUSE_PUSH: if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) clrbits32(&ssi->scr, CCSR_SSI_SCR_TE); -- 1.5.5

2 1

Re: [alsa-devel] [PATCH] ALSA: wss_lib: rework snd_ad1848_probe()
by krzysztof.h1＠poczta.fm 06 Aug '08

06 Aug '08

> Hi Takashi. > > This reworks snd_ad1848_probe(), making for IMO significantly more > straightforward reading. > > The original had a somewhat odd try-loop where it not only waited for > the codec to come out of init but also retried a failed basic detect > which seems to not make sense. Except for that, the detection logic > should be unchanged. > Acked-by: Krzysztof Helt <krzysztof.h1(a)wp.pl> > I am by the way rather unsure what it is that the spin_lock is trying to > protect us from here. What's the error scenario if we just not take it? > We aren't live yet... > > Next step is folding this into snd_wss_probe() directly. > I would like to leave it separate as 1. End of this function is good check point of correct detection for ad1848 chips, 2. The snd_wss_probe() is long enough. But I won't oppose it fiercely. Krzysztof ---------------------------------------------------------------------- Tani Telefon! Sprawdz >>> http://link.interia.pl/f1ec0

1 0

[PATCH] ALSA: wss_lib: rework snd_ad1848_probe()
by Rene Herman 06 Aug '08

06 Aug '08

Make snd_ad1848_probe() easier to follow. With an exception for only trying once as soon as the codec is out of init (which should be all that's needed) the detection logic should be unchanged. Signed-off-by: Rene Herman <rene.herman(a)gmail.com> --- sound/isa/wss/wss_lib.c | 133 ++++++++++++++++++++++++---------------------- 1 files changed, 69 insertions(+), 64 deletions(-) diff --git a/sound/isa/wss/wss_lib.c b/sound/isa/wss/wss_lib.c index 955db45..cbe9565 100644 --- a/sound/isa/wss/wss_lib.c +++ b/sound/isa/wss/wss_lib.c @@ -1147,79 +1147,84 @@ static snd_pcm_uframes_t snd_wss_capture_pointer(struct snd_pcm_substream *subst static int snd_ad1848_probe(struct snd_wss *chip) { + unsigned long timeout = jiffies + msecs_to_jiffies(1000); unsigned long flags; - int i, id, rev, ad1847; + unsigned char r; + unsigned short hardware = 0; + int err = 0; + int i; - id = 0; - ad1847 = 0; - for (i = 0; i < 1000; i++) { - mb(); - if (inb(chip->port + CS4231P(REGSEL)) & CS4231_INIT) - msleep(1); - else { - spin_lock_irqsave(&chip->reg_lock, flags); - snd_wss_out(chip, CS4231_MISC_INFO, 0x00); - snd_wss_out(chip, CS4231_LEFT_INPUT, 0xaa); - snd_wss_out(chip, CS4231_RIGHT_INPUT, 0x45); - rev = snd_wss_in(chip, CS4231_RIGHT_INPUT); - if (rev == 0x65) { - spin_unlock_irqrestore(&chip->reg_lock, flags); - id = 1; - ad1847 = 1; - break; - } - if (rev == 0x45) { - rev = snd_wss_in(chip, CS4231_LEFT_INPUT); - if (rev == 0xaa || rev == 0x8a) { - spin_unlock_irqrestore(&chip->reg_lock, flags); - id = 1; - break; - } - } - spin_unlock_irqrestore(&chip->reg_lock, flags); - } + while (wss_inb(chip, CS4231P(REGSEL)) & CS4231_INIT) { + if (time_after(jiffies, timeout)) + return -ENODEV; + cond_resched(); } - if (id != 1) - return -ENODEV; /* no valid device found */ - id = 0; - if (chip->hardware == WSS_HW_DETECT) - id = ad1847 ? WSS_HW_AD1847 : WSS_HW_AD1848; - spin_lock_irqsave(&chip->reg_lock, flags); - inb(chip->port + CS4231P(STATUS)); /* clear any pendings IRQ */ - outb(0, chip->port + CS4231P(STATUS)); - mb(); - if (id == WSS_HW_AD1848) { - /* check if there are more than 16 registers */ - rev = snd_wss_in(chip, CS4231_MISC_INFO); - snd_wss_out(chip, CS4231_MISC_INFO, 0x40); - for (i = 0; i < 16; ++i) { - if (snd_wss_in(chip, i) != snd_wss_in(chip, i + 16)) { - id = WSS_HW_CMI8330; - break; - } + + /* set CS423x MODE 1 */ + snd_wss_out(chip, CS4231_MISC_INFO, 0); + + snd_wss_out(chip, CS4231_RIGHT_INPUT, 0x45); /* 0x55 & ~0x10 */ + r = snd_wss_in(chip, CS4231_RIGHT_INPUT); + if (r != 0x45) { + /* RMGE always high on AD1847 */ + if ((r & ~CS4231_ENABLE_MIC_GAIN) != 0x45) { + err = -ENODEV; + goto out; + } + hardware = WSS_HW_AD1847; + } else { + snd_wss_out(chip, CS4231_LEFT_INPUT, 0xaa); + r = snd_wss_in(chip, CS4231_LEFT_INPUT); + /* L/RMGE always low on AT2320 */ + if ((r | CS4231_ENABLE_MIC_GAIN) != 0xaa) { + err = -ENODEV; + goto out; } - snd_wss_out(chip, CS4231_MISC_INFO, 0x00); - if (id != WSS_HW_CMI8330 && (rev & 0x80)) - id = WSS_HW_CS4248; - if (id == WSS_HW_CMI8330 && (rev & 0x0f) != 0x0a) - id = 0; } - if (id == WSS_HW_CMI8330) { - /* verify it is not CS4231 by changing the version register */ - /* on CMI8330 it is volume control register and can be set 0 */ - snd_wss_out(chip, CS4231_MISC_INFO, CS4231_MODE2); - snd_wss_dout(chip, CS4231_VERSION, 0x00); - rev = snd_wss_in(chip, CS4231_VERSION) & 0xe7; - if (rev) - id = 0; - snd_wss_out(chip, CS4231_MISC_INFO, 0); + + /* clear pending IRQ */ + wss_inb(chip, CS4231P(STATUS)); + wss_outb(chip, CS4231P(STATUS), 0); + mb(); + + if ((chip->hardware & WSS_HW_TYPE_MASK) != WSS_HW_DETECT) + goto out; + + if (hardware) { + chip->hardware = hardware; + goto out; } - if (id) - chip->hardware = id; + r = snd_wss_in(chip, CS4231_MISC_INFO); + + /* set CS423x MODE 2 */ + snd_wss_out(chip, CS4231_MISC_INFO, CS4231_MODE2); + for (i = 0; i < 16; i++) { + if (snd_wss_in(chip, i) != snd_wss_in(chip, 16 + i)) { + /* we have more than 16 registers: check ID */ + if ((r & 0xf) != 0xa) + goto out_mode; + /* + * on CMI8330, CS4231_VERSION is volume control and + * can be set to 0 + */ + snd_wss_dout(chip, CS4231_VERSION, 0); + r = snd_wss_in(chip, CS4231_VERSION) & 0xe7; + if (!r) + chip->hardware = WSS_HW_CMI8330; + goto out_mode; + } + } + if (r & 0x80) + chip->hardware = WSS_HW_CS4248; + else + chip->hardware = WSS_HW_AD1848; +out_mode: + snd_wss_out(chip, CS4231_MISC_INFO, 0); +out: spin_unlock_irqrestore(&chip->reg_lock, flags); - return 0; /* all things are ok.. */ + return err; } static int snd_wss_probe(struct snd_wss *chip) -- 1.5.5 --------------080803020607030707060103--

1 0

[PATCH] ALSA: wss_lib: rework snd_ad1848_probe()
by Rene Herman 06 Aug '08

06 Aug '08

Make snd_ad1848_probe() easier to follow. With an exception for only trying once as soon as the codec is out of init (which should be all that's needed) the detection logic should be unchanged. Signed-off-by: Rene Herman <rene.herman(a)gmail.com> --- sound/isa/wss/wss_lib.c | 131 ++++++++++++++++++++++++----------------------- 1 files changed, 67 insertions(+), 64 deletions(-) diff --git a/sound/isa/wss/wss_lib.c b/sound/isa/wss/wss_lib.c index 955db45..802474b 100644 --- a/sound/isa/wss/wss_lib.c +++ b/sound/isa/wss/wss_lib.c @@ -1147,79 +1147,82 @@ static snd_pcm_uframes_t snd_wss_capture_pointer(struct snd_pcm_substream *subst static int snd_ad1848_probe(struct snd_wss *chip) { + unsigned long timeout = jiffies + msecs_to_jiffies(1000); unsigned long flags; - int i, id, rev, ad1847; + unsigned char r; + unsigned short hardware = 0; + int err = 0; + int i; - id = 0; - ad1847 = 0; - for (i = 0; i < 1000; i++) { - mb(); - if (inb(chip->port + CS4231P(REGSEL)) & CS4231_INIT) - msleep(1); - else { - spin_lock_irqsave(&chip->reg_lock, flags); - snd_wss_out(chip, CS4231_MISC_INFO, 0x00); - snd_wss_out(chip, CS4231_LEFT_INPUT, 0xaa); - snd_wss_out(chip, CS4231_RIGHT_INPUT, 0x45); - rev = snd_wss_in(chip, CS4231_RIGHT_INPUT); - if (rev == 0x65) { - spin_unlock_irqrestore(&chip->reg_lock, flags); - id = 1; - ad1847 = 1; - break; - } - if (rev == 0x45) { - rev = snd_wss_in(chip, CS4231_LEFT_INPUT); - if (rev == 0xaa || rev == 0x8a) { - spin_unlock_irqrestore(&chip->reg_lock, flags); - id = 1; - break; - } - } - spin_unlock_irqrestore(&chip->reg_lock, flags); - } + while (wss_inb(chip, CS4231P(REGSEL)) & CS4231_INIT) { + if (time_after(jiffies, timeout)) + return -ENODEV; + cond_resched(); } - if (id != 1) - return -ENODEV; /* no valid device found */ - id = 0; - if (chip->hardware == WSS_HW_DETECT) - id = ad1847 ? WSS_HW_AD1847 : WSS_HW_AD1848; - spin_lock_irqsave(&chip->reg_lock, flags); - inb(chip->port + CS4231P(STATUS)); /* clear any pendings IRQ */ - outb(0, chip->port + CS4231P(STATUS)); - mb(); - if (id == WSS_HW_AD1848) { - /* check if there are more than 16 registers */ - rev = snd_wss_in(chip, CS4231_MISC_INFO); - snd_wss_out(chip, CS4231_MISC_INFO, 0x40); - for (i = 0; i < 16; ++i) { - if (snd_wss_in(chip, i) != snd_wss_in(chip, i + 16)) { - id = WSS_HW_CMI8330; - break; - } + + /* set CS423x MODE 1 */ + snd_wss_out(chip, CS4231_MISC_INFO, 0); + + snd_wss_out(chip, CS4231_RIGHT_INPUT, 0x45); /* 0x55 & ~0x10 */ + r = snd_wss_in(chip, CS4231_RIGHT_INPUT); + if (r != 0x45) { + /* RMGE always high on AD1847 */ + if ((r & ~CS4231_ENABLE_MIC_GAIN) != 0x45) { + err = -ENODEV; + goto out; + } + hardware = WSS_HW_AD1847; + } else { + snd_wss_out(chip, CS4231_LEFT_INPUT, 0xaa); + r = snd_wss_in(chip, CS4231_LEFT_INPUT); + /* L/RMGE always low on AT2320 */ + if ((r | CS4231_ENABLE_MIC_GAIN) != 0xaa) { + err = -ENODEV; + goto out; } - snd_wss_out(chip, CS4231_MISC_INFO, 0x00); - if (id != WSS_HW_CMI8330 && (rev & 0x80)) - id = WSS_HW_CS4248; - if (id == WSS_HW_CMI8330 && (rev & 0x0f) != 0x0a) - id = 0; } - if (id == WSS_HW_CMI8330) { - /* verify it is not CS4231 by changing the version register */ - /* on CMI8330 it is volume control register and can be set 0 */ - snd_wss_out(chip, CS4231_MISC_INFO, CS4231_MODE2); - snd_wss_dout(chip, CS4231_VERSION, 0x00); - rev = snd_wss_in(chip, CS4231_VERSION) & 0xe7; - if (rev) - id = 0; - snd_wss_out(chip, CS4231_MISC_INFO, 0); + + /* clear pending IRQ */ + wss_inb(chip, CS4231P(STATUS)); + wss_outb(chip, CS4231P(STATUS), 0); + mb(); + + if ((chip->hardware & WSS_HW_TYPE_MASK) != WSS_HW_DETECT) + goto out; + + if (hardware) { + chip->hardware = hardware; + goto out; } - if (id) - chip->hardware = id; + r = snd_wss_in(chip, CS4231_MISC_INFO); + + /* set CS423x MODE 2 */ + snd_wss_out(chip, CS4231_MISC_INFO, CS4231_MODE2); + for (i = 0; i < 16; i++) { + if (snd_wss_in(chip, i) != snd_wss_in(chip, 16 + i)) { + /* we have more than 16 registers: check ID */ + if ((r & 0xf) != 0xa) + goto out; + /* + * on CMI8330, CS4231_VERSION is volume control and + * can be set to 0 + */ + snd_wss_dout(chip, CS4231_VERSION, 0); + r = snd_wss_in(chip, CS4231_VERSION) & 0xe7; + if (!r) + chip->hardware = WSS_HW_CMI8330; + goto out; + } + } + if (r & 0x80) + chip->hardware = WSS_HW_CS4248; + else + chip->hardware = WSS_HW_AD1848; +out: spin_unlock_irqrestore(&chip->reg_lock, flags); - return 0; /* all things are ok.. */ + return err; } static int snd_wss_probe(struct snd_wss *chip) -- 1.5.5 --------------060807000500060608070809--

1 0

[alsa-devel] dmix in a forked process
by Bankim Bhavsar 05 Aug '08

05 Aug '08

Don't know why the earlier message got truncated so here it is again... Reading the source alsa-lib I see that dmix runs in a forked process *without* exec. file: src/pcm/pcm_direct.c function: snd_pcm_direct_server_create ret = fork(); if (ret < 0) { close(dmix->server_fd); return ret; } else if (ret == 0) { ret = fork(); if (ret == 0) server_job(dmix); _exit(EXIT_SUCCESS); } else { waitpid(ret, NULL, 0); } _From what I understand, this is a concern since if our app is the first to start dmix then a copy of the process (app) will remain till the system is shutdown ? Also I see that from alsa-lib 1.0.12 a check is performed before starting dmix. file: src/pcm/pcm_dmix.c function: snd_pcm_dmix_open if (dmix->shmptr->use_server) { dmix->server_free = dmix_server_free; ret = snd_pcm_direct_server_create(dmix); if (ret < 0) { SNDERR("unable to create server"); goto _err; } } When I checked how dmix->shmptr->use_server is set file: src/pcm/pcm_dmix.c function: snd_pcm_direct_initialize_slave int ver = 0; ioctl(spcm->poll_fd, SNDRV_PCM_IOCTL_PVERSION, &ver); if (ver < SNDRV_PROTOCOL_VERSION(2, 0, 8)) dmix->shmptr->use_server = 1; So if sndrv version is older than 2.0.8, use_server is set! So even using alsa-lib greater than 1.0.12 doesn't guarantee that the forked process (without exec) code path will be avoided. I'm currently porting the sound code of our app from OSS to ALSA but this problem is preventing me from doing so. Is there any way to get around this problem? Help here will be greatly appreciated. If this is an incorrect list then kindly fwd it to the appropriate one. Thanks, Bankim. PS: I've tried using the ioctl mentioned above after copying the necessary macros from kernel headers but ioctl returns error "Inappropriate ioctl for this device" on my Ubuntu 7.04 host.

3 4

[alsa-devel] [PATCH] ac97 mono-out w/internal speaker
by george pee 05 Aug '08

05 Aug '08

I have a wireless tablet (webdt310) http://dtresearch.com/product/wt310/index.htm It has cs5536 audio and a realtek alc203 codec. The headphone jack works perfectly, but the internal speaker was not. After a bit a twiddling and reading the ALC203 data sheet, I finally figured out that the function of Pin 37 had to be changed to MONO-OUT to get sound out of the internal speaker. I'm not sure that this is the best approach (maybe there could be some pin sensing?) but I added this as an ac97 quirk. Here's the patch. diff -rub linux-2.6.26.orig/sound/pci/ac97/ac97_codec.c linux-2.6.26.new/sound/pci/ac97/ac97_codec.c --- linux-2.6.26.orig/sound/pci/ac97/ac97_codec.c 2008-07-29 15:30:57.000000000 -0500 +++ linux-2.6.26.new/sound/pci/ac97/ac97_codec.c 2008-08-04 10:58:12.000000000 -0500 @@ -2709,6 +2710,12 @@ return snd_ctl_add(ac97->bus->card, snd_ac97_cnew(&snd_ac97_alc_jack_detect, ac97)); } +/* ac97 tune: turn on mono-out */ +static int tune_mono_out(struct snd_ac97 *ac97) +{ + snd_ac97_update_bits(ac97, 0x7a, 0x400, 0x400); + return 1; +} + /* ac97 tune: inversed EAPD bit */ static int tune_inv_eapd(struct snd_ac97 *ac97) { @@ -2801,6 +2808,7 @@ { "inv_eapd", tune_inv_eapd }, { "mute_led", tune_mute_led }, { "hp_mute_led", tune_hp_mute_led }, + { "mono_out", tune_mono_out }, };

2 3

Re: [alsa-devel] [PATCH 07/11] wss_lib: use CS4231P instead of AD1848P (kill the AD1848P)
by krzysztof.h1＠poczta.fm 05 Aug '08

05 Aug '08

> On 31-07-08 21:07, Krzysztof Helt wrote: > > > From: Krzysztof Helt <krzysztof.h1(a)wp.pl> > > > > Use CS4231P instead of AD1848P (kill the AD1848P). > > > > Signed-off-by: Krzysztof Helt <krzysztof.h1(a)wp.pl> > > --- > > Changes since previous version: > > > > 1. No reformatting of the snd_ad1848_debug() > > as the file is going to be killed. > > 1. Nicer formatting of WSS_SINGLE and WSS_DOUBLE > > macros. > > I believe 1 no. 2 is a cut & paste error? There are no such changes in > this patch. > Right. Regards, Krzysztof ---------------------------------------------------------------------- Wymien zeszyty na notebooka! Sprawdz >>> http://link.interia.pl/f1eab

1 0

Re: [alsa-devel] ultra small alsa library
by Marco Schuster 05 Aug '08

05 Aug '08

The problem I have is: As soon as I try to play a mp3 via mpg123, in dmesg comes the following error: Division by zero in kernel. [<c0026e2c>] (dump_stack+0x0/0x14) from [<c0026e58>] (__div0+0x18/0x20) [<c0026e40>] (__div0+0x0/0x20) from [<c011c65c>] (Ldiv0+0x8/0x10) [<bf0267ec>] (iso_stream_find+0x0/0x340 [ehci_hcd]) from [<bf028140>] (ehci_urb_enqueue+0x848/0xdf4 [ehci_hcd]) [<bf0278f8>] (ehci_urb_enqueue+0x0/0xdf4 [ehci_hcd]) from [<bf005228>] (usb_hcd_submit_urb+0x660/0x77c [usbcore]) [<bf004bc8>] (usb_hcd_submit_urb+0x0/0x77c [usbcore]) from [<bf0057fc>] (usb_submit_urb+0x228/0x264 [usbcore]) [<bf0055d4>] (usb_submit_urb+0x0/0x264 [usbcore]) from [<bf1c2820>] (start_urbs+0x1cc/0x308 [snd_usb_audio]) r6:00000000 r5:c0c2dc10 r4:00000010 [<bf1c2654>] (start_urbs+0x0/0x308 [snd_usb_audio]) from [<bf1c4d78>] (snd_usb_pcm_prepare+0xa4/0xb8 [snd_usb_audio]) r7:00000000 r6:00000000 r5:c0c2b800 r4:c0c2dc10 [<bf1c4cd4>] (snd_usb_pcm_prepare+0x0/0xb8 [snd_usb_audio]) from [<bf19bdc4>] (snd_pcm_do_prepare+0x1c/0x3c [snd_pcm]) r6:00000802 r5:c13d02c0 r4:c13d02c0 [<bf19bda8>] (snd_pcm_do_prepare+0x0/0x3c [snd_pcm]) from [<bf19b96c>] (snd_pcm_action_single+0x40/0x7c [snd_pcm]) r4:bf1acd74 [<bf19b92c>] (snd_pcm_action_single+0x0/0x7c [snd_pcm]) from [<bf19c774>] (snd_pcm_action_nonatomic+0x58/0x70 [snd_pcm]) r7:00044930 r6:bf1acd74 r5:00000802 r4:c13d02c0 [<bf19c71c>] (snd_pcm_action_nonatomic+0x0/0x70 [snd_pcm]) from [<bf19eba8>] (snd_pcm_common_ioctl1+0x664/0xf7c [snd_pcm]) r6:c13d02c0 r5:00044930 r4:00044930 [<bf19e544>] (snd_pcm_common_ioctl1+0x0/0xf7c [snd_pcm]) from [<bf19fe18>] (snd_pcm_playback_ioctl1+0x48c/0x4b0 [snd_pcm]) [<bf19f98c>] (snd_pcm_playback_ioctl1+0x0/0x4b0 [snd_pcm]) from [<bf19ff20>] (snd_pcm_playback_ioctl+0x34/0x40 [snd_pcm]) [<bf19feec>] (snd_pcm_playback_ioctl+0x0/0x40 [snd_pcm]) from [<c0097e70>] (do_ioctl+0x34/0x78) [<c0097e3c>] (do_ioctl+0x0/0x78) from [<c0098114>] (vfs_ioctl+0x260/0x290) r5:00044930 r4:c0fbcbc0 [<c0097eb4>] (vfs_ioctl+0x0/0x290) from [<c009819c>] (sys_ioctl+0x58/0x78) r7:00004140 r6:c0fbcbc0 r5:00044930 r4:00000000 [<c0098144>] (sys_ioctl+0x0/0x78) from [<c0022de0>] (ret_fast_syscall+0x0/0x2c) r8:c0022f84 r7:00000036 r6:beebb3a0 r5:beebb708 r4:00043598 Can anyone point me what goes wrong here? Kernel is debian 2.6.24-1-ixp4xx (armv5tel). Thanks, Marco

2 1