September 2012 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [RFC 0/3] Initial stab at converting OMAP ASoC support to DMA engine
by Russell King - ARM Linux 04 Sep '12

04 Sep '12

The following series of three patches is an attempt to convert the OMAP ASoC backend to use the DMA engine support. I'll bring your attention to the comments in patch 3 which highlight some of the features lost in this process. Some questions need answering there (in particular the one concerning pause/resume) especially as it seems the present driver could well be buggy wrt comments recently on the mailing list about ALSA pause/resume requirements. This works for me, given the wind is in the right direction and if it's sunny outside. I find the audio support on OMAP to be rather flakey at best, requiring reboots if it doesn't work first time.

5 7

[alsa-devel] [PATCH 0/3] ALSA: snd-usb: Regression fixes for 3.5
by Daniel Mack 04 Sep '12

04 Sep '12

Here are 3 more regression fixes for snd-usb in 3.5. It's annoying they haven't been found earlier, but at least we're now back to the streaming quality we had before the streaming refactoring. Anyone with a USB device, please give it a quick shot at least. Thanks, Daniel Daniel Mack (3): ALSA: snd-usb: restore delay information ALSA: snd-usb: fix calls to next_packet_size ALSA: snd-usb: fix cross-interface streaming devices sound/usb/endpoint.c | 13 +------------ sound/usb/endpoint.h | 1 + sound/usb/pcm.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++---- 3 files changed, 49 insertions(+), 16 deletions(-) -- 1.7.11.4

3 8

[alsa-devel] [PATCH v3 0/3] introduce device_buffer
by Vinod Koul 04 Sep '12

04 Sep '12

Old wine in a new bottle with added flavour now. Based on our discussions in LPC, now we will report the combined delay value as sum of the buffering done in driver and hardware. This is used to report delay Additionally we tell pcm what is buffer in device/driver, this is used to check for error calulcations Vinod Koul (3): ALSA: pcm - introduce device_buffer ASoC: add device_buffer in asoc ASoC: mid-x86 - implement buffer query in sst_platform driver include/sound/pcm.h | 1 + include/sound/soc-dai.h | 6 ++++++ include/sound/soc.h | 6 ++++++ sound/core/pcm_lib.c | 10 +++++++++- sound/soc/mid-x86/sst_platform.c | 23 +++++++++++++++++++++++ sound/soc/mid-x86/sst_platform.h | 4 +++- sound/soc/soc-pcm.c | 13 ++++++++++++- 7 files changed, 60 insertions(+), 3 deletions(-) -- 1.7.9.5

4 10

[alsa-devel] [PATCH 0/3] ALSA: Add DSP firmware loader
by Ian Minett 04 Sep '12

04 Sep '12

From: Ian Minett <ian_minett(a)creativelabs.com> Hi Takashi, Thanks for the recent code samples, they were very useful. We've built on them to produce this latest patchset, incorporating the updates you provided to add the loader functions, and doing a minimal update to the CA0132 codec to use the new ops to perform the firmware transfer. Please let us know if any tweaks are needed to the update. Thanks, - Ian Signed-off-by: Ian Minett <ian_minett(a)creativelabs.com> --- 1: - memalloc.h - pcm.h - pcm_memory.c - sgbuf.c Include Takashi's updates: Make snd_sgbuf_get_{ptr|addr}() available for non-SG cases. Passing struct snd_dma_buffer pointer instead, so that they work no matter whether real SG buffer is used or not. 2: - hda_intel.c - hda_codec.h Include Takashi's updates: Pass DMA buffer pointers in calls to setup_bdle(). Add new callback routines to controller to handle loading firmware binaries. Add new switch to Kconfig to enable DSP firmware loader. 3: - patch_ca0132.c - ca0132_regs.h Add support for DSP firmware transfer to CA0132 codec: Add transfer handler functions. Call the new snd_hda_codec_load_dsp_xxx loader ops to control download. Add chip access routines.

2 3

[alsa-devel] [PATCH] ALSA: snd-usb: Add quirks for Playback Designs devices
by Daniel Mack 04 Sep '12

04 Sep '12

Playback Designs' USB devices have some hardware limitations on their USB interface. In particular: - They need a 20ms delay after each class compliant request as the hardware ACKs the USB packets before the device is actually ready for the next command. Sending data immediately will result in buffer overflows in the hardware. - The devices send bogus feedback data at the start of each stream which confuse the feedback format auto-detection. This patch introduces a new quirks hook that is called after each control packet and which adds a delay for all devices that match Playback Designs' USB VID for now. In addition, it adds a counter to snd_usb_endpoint to drop received packets on the floor. Another new quirks function that is called once an endpoint is started initializes that counter for these devices on their sync endpoint. Signed-off-by: Daniel Mack <zonque(a)gmail.com> Reported-and-tested-by: Andreas Koch <andreas(a)akdesigninc.com> Supported-by: Demian Martin <demianm_1(a)yahoo.com> --- sound/usb/card.h | 2 ++ sound/usb/endpoint.c | 8 ++++++++ sound/usb/helper.c | 5 +++++ sound/usb/quirks.c | 24 ++++++++++++++++++++++++ sound/usb/quirks.h | 10 ++++++++++ 5 files changed, 49 insertions(+) diff --git a/sound/usb/card.h b/sound/usb/card.h index 2b9ffff..d9d2b5a 100644 --- a/sound/usb/card.h +++ b/sound/usb/card.h @@ -92,6 +92,8 @@ struct snd_usb_endpoint { unsigned char silence_value; unsigned int stride; int iface, alt_idx; + int skip_packets; /* quirks for devices to ignore the first n packets + in a stream */ spinlock_t lock; struct list_head list; diff --git a/sound/usb/endpoint.c b/sound/usb/endpoint.c index d6e2bb4..a83a18d 100644 --- a/sound/usb/endpoint.c +++ b/sound/usb/endpoint.c @@ -31,6 +31,7 @@ #include "card.h" #include "endpoint.h" #include "pcm.h" +#include "quirks.h" #define EP_FLAG_ACTIVATED 0 #define EP_FLAG_RUNNING 1 @@ -170,6 +171,11 @@ static void retire_inbound_urb(struct snd_usb_endpoint *ep, { struct urb *urb = urb_ctx->urb; + if (unlikely(ep->skip_packets > 0)) { + ep->skip_packets--; + return; + } + if (ep->sync_slave) snd_usb_handle_sync_urb(ep->sync_slave, ep, urb); @@ -822,6 +828,8 @@ int snd_usb_endpoint_start(struct snd_usb_endpoint *ep, int can_sleep) ep->unlink_mask = 0; ep->phase = 0; + snd_usb_endpoint_start_quirk(ep); + /* * If this endpoint has a data endpoint as implicit feedback source, * don't start the urbs here. Instead, mark them all as available, diff --git a/sound/usb/helper.c b/sound/usb/helper.c index 9eed8f4..c1db28f 100644 --- a/sound/usb/helper.c +++ b/sound/usb/helper.c @@ -21,6 +21,7 @@ #include "usbaudio.h" #include "helper.h" +#include "quirks.h" /* * combine bytes and get an integer value @@ -97,6 +98,10 @@ int snd_usb_ctl_msg(struct usb_device *dev, unsigned int pipe, __u8 request, memcpy(data, buf, size); kfree(buf); } + + snd_usb_ctl_msg_quirk(dev, pipe, request, requesttype, + value, index, data, size); + return err; } diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index 2781726..1eaacf1 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -761,3 +761,27 @@ void snd_usb_set_format_quirk(struct snd_usb_substream *subs, } } +void snd_usb_endpoint_start_quirk(struct snd_usb_endpoint *ep) +{ + /* + * "Playback Design" products send bogus feedback data at the start + * of the stream. Ignore them. + */ + if ((le16_to_cpu(ep->chip->dev->descriptor.idVendor) == 0x23ba) && + ep->type == SND_USB_ENDPOINT_TYPE_SYNC) + ep->skip_packets = 4; +} + +void snd_usb_ctl_msg_quirk(struct usb_device *dev, unsigned int pipe, + __u8 request, __u8 requesttype, __u16 value, + __u16 index, void *data, __u16 size) +{ + /* + * "Playback Design" products need a 20ms delay after each + * class compliant request + */ + if ((le16_to_cpu(dev->descriptor.idVendor) == 0x23ba) && + (requesttype & USB_TYPE_MASK) == USB_TYPE_CLASS) + mdelay(20); +} + diff --git a/sound/usb/quirks.h b/sound/usb/quirks.h index 03e5e94..0ca9e91 100644 --- a/sound/usb/quirks.h +++ b/sound/usb/quirks.h @@ -1,6 +1,10 @@ #ifndef __USBAUDIO_QUIRKS_H #define __USBAUDIO_QUIRKS_H +struct audioformat; +struct snd_usb_endpoint; +struct snd_usb_substream; + int snd_usb_create_quirk(struct snd_usb_audio *chip, struct usb_interface *iface, struct usb_driver *driver, @@ -20,4 +24,10 @@ void snd_usb_set_format_quirk(struct snd_usb_substream *subs, int snd_usb_is_big_endian_format(struct snd_usb_audio *chip, struct audioformat *fp); +void snd_usb_endpoint_start_quirk(struct snd_usb_endpoint *ep); + +void snd_usb_ctl_msg_quirk(struct usb_device *dev, unsigned int pipe, + __u8 request, __u8 requesttype, __u16 value, + __u16 index, void *data, __u16 size); + #endif /* __USBAUDIO_QUIRKS_H */ -- 1.7.11.4

2 1

Re: [alsa-devel] [PATCH] ASoC: mc13783: Remove mono support
by Gaëtan Carlier 04 Sep '12

04 Sep '12

Hi, On 09/03/2012 06:04 PM, Fabio Estevam wrote: > Playing a mono track on a mc13783 codec results in incorrect playback rate. > > Remove mono support so that a mono track can be played correctly. > > Signed-off-by: Fabio Estevam <fabio.estevam(a)freescale.com> > --- > sound/soc/codecs/mc13783.c | 8 ++++---- > 1 file changed, 4 insertions(+), 4 deletions(-) > > diff --git a/sound/soc/codecs/mc13783.c b/sound/soc/codecs/mc13783.c > index 8f726c0..115a403 100644 > --- a/sound/soc/codecs/mc13783.c > +++ b/sound/soc/codecs/mc13783.c > @@ -659,7 +659,7 @@ static struct snd_soc_dai_driver mc13783_dai_async[] = { > .id = MC13783_ID_STEREO_DAC, > .playback = { > .stream_name = "Playback", > - .channels_min = 1, > + .channels_min = 2, > .channels_max = 2, > .rates = SNDRV_PCM_RATE_8000_96000, > .formats = MC13783_FORMATS, > @@ -670,7 +670,7 @@ static struct snd_soc_dai_driver mc13783_dai_async[] = { > .id = MC13783_ID_STEREO_CODEC, > .capture = { > .stream_name = "Capture", > - .channels_min = 1, > + .channels_min = 2, > .channels_max = 2, > .rates = MC13783_RATES_RECORD, > .formats = MC13783_FORMATS, > @@ -692,14 +692,14 @@ static struct snd_soc_dai_driver mc13783_dai_sync[] = { > .id = MC13783_ID_SYNC, > .playback = { > .stream_name = "Playback", > - .channels_min = 1, > + .channels_min = 2, > .channels_max = 2, > .rates = SNDRV_PCM_RATE_8000_96000, > .formats = MC13783_FORMATS, > }, > .capture = { > .stream_name = "Capture", > - .channels_min = 1, > + .channels_min = 2, > .channels_max = 2, > .rates = MC13783_RATES_RECORD, > .formats = MC13783_FORMATS, > On imx27_3ds board: Tested-by: Gaëtan Carlier <gcembed(a)gmail.com> Regards, Gaëtan Carlier

1 0

Re: [alsa-devel] [ASoC] Fix: Revert 'ASoC: imx-ssi: Remove mono support'.
by Gaëtan Carlier 04 Sep '12

04 Sep '12

Hi Javier, On 09/03/2012 10:27 AM, Javier Martin wrote: > The following commit should be reverted: 0865a75d4166bddc533fd50831829ceefb94f9b0 > > The bug this patch is meant to solve doesn't occur in Visstrim_M10 boards. > Furthermore, after applying this patch sound in Visstrim_M10 is played > at slower rates. > > Signed-off-by: Javier Martin <javier.martin(a)vista-silicon.com> > --- > diff --git b/sound/soc/fsl/imx-ssi.c a/sound/soc/fsl/imx-ssi.c > index ac337ac..28dd76c 100644 > --- b/sound/soc/fsl/imx-ssi.c > +++ a/sound/soc/fsl/imx-ssi.c > @@ -380,13 +380,13 @@ static int imx_ssi_dai_probe(struct snd_soc_dai *dai) > static struct snd_soc_dai_driver imx_ssi_dai = { > .probe = imx_ssi_dai_probe, > .playback = { > - .channels_min = 2, > + .channels_min = 1, > .channels_max = 2, > .rates = SNDRV_PCM_RATE_8000_96000, > .formats = SNDRV_PCM_FMTBIT_S16_LE, > }, > .capture = { > - .channels_min = 2, > + .channels_min = 1, > .channels_max = 2, > .rates = SNDRV_PCM_RATE_8000_96000, > .formats = SNDRV_PCM_FMTBIT_S16_LE, > When applied on linux-next-20120824, the patch failed because original imx-ssi.c file looks like : static struct snd_soc_dai_driver imx_ssi_dai = { .probe = imx_ssi_dai_probe, .playback = { /* The SSI does not support monaural audio. */ .channels_min = 2, .channels_max = 2, .rates = SNDRV_PCM_RATE_8000_96000, .formats = SNDRV_PCM_FMTBIT_S16_LE, }, .capture = { .channels_min = 2, .channels_max = 2, .rates = SNDRV_PCM_RATE_8000_96000, .formats = SNDRV_PCM_FMTBIT_S16_LE, }, .ops = &imx_ssi_pcm_dai_ops, }; The comment line is missing in your patch. Here is the right patch : @@ -378,18 +378,17 @@ static int imx_ssi_dai_probe(struct snd_soc_dai *dai) } static struct snd_soc_dai_driver imx_ssi_dai = { .probe = imx_ssi_dai_probe, .playback = { - /* The SSI does not support monaural audio. */ - .channels_min = 2, + .channels_min = 1, .channels_max = 2, .rates = SNDRV_PCM_RATE_8000_96000, .formats = SNDRV_PCM_FMTBIT_S16_LE, }, .capture = { - .channels_min = 2, + .channels_min = 1, .channels_max = 2, .rates = SNDRV_PCM_RATE_8000_96000, .formats = SNDRV_PCM_FMTBIT_S16_LE, }, .ops = &imx_ssi_pcm_dai_ops, Regards, Gaëtan Carlier.

1 0

[alsa-devel] [Patch] Support for (original) Xbox Communicator
by Marko Friedemann 03 Sep '12

03 Sep '12

Hey guys, attached is a patch for the USB quirks to support the original Xbox Communicator (aka XboxLive Headset). I had actually submitted this some time ago against an earlier version but wasn't subscribed to the list and it was lost in moderation. I present it again now, against alsa 1.0.25. Tested and working under ubuntu 12.04.1 with kernel 3.2.0-29-generic. Regards, Marko -- "Somit werden Zitate zu zählbaren Einheiten wissenschaftlicher Reputation." -- NeMa-News - Trends in der Wissenschaft

2 3

Re: [alsa-devel] [PATCH] sound: Remove the last mention of SNDRV_MAIN_OBJECT_FILE
by Takashi Iwai 03 Sep '12

03 Sep '12

At Sun, 2 Sep 2012 20:04:17 -0700, Josh Triplett wrote: > > SNDRV_MAIN_OBJECT_FILE hasn't done anything since the pre-git days, and > the only remaining reference occurs as a #define in sound/last.c. Drop > that last mention of it. > > Signed-off-by: Josh Triplett <josh(a)joshtriplett.org> Applied now. Thanks. Takashi > --- > sound/last.c | 1 - > 1 file changed, 1 deletion(-) > > diff --git a/sound/last.c b/sound/last.c > index 7ffc182..43f2228 100644 > --- a/sound/last.c > +++ b/sound/last.c > @@ -19,7 +19,6 @@ > * > */ > > -#define SNDRV_MAIN_OBJECT_FILE > #include <linux/init.h> > #include <sound/core.h> > > -- > 1.7.10.4 >

1 0

Re: [alsa-devel] [PATCH] ALSA: fix possible memory leak in snd_mixer_oss_build_input()
by Takashi Iwai 03 Sep '12

03 Sep '12

At Sun, 2 Sep 2012 22:10:27 +0800, Wei Yongjun wrote: > > From: Wei Yongjun <yongjun_wei(a)trendmicro.com.cn> > > uinfo has been allocated in this function and should be > freed before leaving from the error handling cases. > > spatch with a semantic match is used to found this problem. > (http://coccinelle.lip6.fr/) > > Signed-off-by: Wei Yongjun <yongjun_wei(a)trendmicro.com.cn> Applied now. Thanks. Takashi > --- > sound/core/oss/mixer_oss.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/sound/core/oss/mixer_oss.c b/sound/core/oss/mixer_oss.c > index 18297f7..29f6ded 100644 > --- a/sound/core/oss/mixer_oss.c > +++ b/sound/core/oss/mixer_oss.c > @@ -1046,6 +1046,7 @@ static int snd_mixer_oss_build_input(struct snd_mixer_oss *mixer, struct snd_mix > > if (kctl->info(kctl, uinfo)) { > up_read(&mixer->card->controls_rwsem); > + kfree(uinfo); > return 0; > } > strcpy(str, ptr->name); > @@ -1061,6 +1062,7 @@ static int snd_mixer_oss_build_input(struct snd_mixer_oss *mixer, struct snd_mix > uinfo->value.enumerated.item = slot.capture_item; > if (kctl->info(kctl, uinfo)) { > up_read(&mixer->card->controls_rwsem); > + kfree(uinfo); > return 0; > } > if (!strcmp(uinfo->value.enumerated.name, str)) { > >

1 0