March 2013 - Alsa-devel - mailman.alsa-project.org

Re: [alsa-devel] [PATCH] documentation: alsa: Fix typo in Documentation/sound
by Takashi Iwai 17 Mar '13

17 Mar '13

At Sun, 17 Mar 2013 02:57:28 +0900, Masanari Iida wrote: > > Correct spelling typos in Documentation/sound/alsa > > Signed-off-by: Masanari Iida <standby24x7(a)gmail.com> Thanks, applied. Takashi > --- > Documentation/sound/alsa/ALSA-Configuration.txt | 2 +- > Documentation/sound/alsa/seq_oss.html | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/Documentation/sound/alsa/ALSA-Configuration.txt b/Documentation/sound/alsa/ALSA-Configuration.txt > index ce6581c..4499bd9 100644 > --- a/Documentation/sound/alsa/ALSA-Configuration.txt > +++ b/Documentation/sound/alsa/ALSA-Configuration.txt > @@ -912,7 +912,7 @@ Prior to version 0.9.0rc4 options had a 'snd_' prefix. This was removed. > models depending on the codec chip. The list of available models > is found in HD-Audio-Models.txt > > - The model name "genric" is treated as a special case. When this > + The model name "generic" is treated as a special case. When this > model is given, the driver uses the generic codec parser without > "codec-patch". It's sometimes good for testing and debugging. > > diff --git a/Documentation/sound/alsa/seq_oss.html b/Documentation/sound/alsa/seq_oss.html > index d9776cf..9663b45 100644 > --- a/Documentation/sound/alsa/seq_oss.html > +++ b/Documentation/sound/alsa/seq_oss.html > @@ -285,7 +285,7 @@ sample data. > <H4> > 7.2.4 Close Callback</H4> > The <TT>close</TT> callback is called when this device is closed by the > -applicaion. If any private data was allocated in open callback, it must > +application. If any private data was allocated in open callback, it must > be released in the close callback. The deletion of ALSA port should be > done here, too. This callback must not be NULL. > <H4> > -- > 1.8.2 >

1 0

[alsa-devel] [PATCH] ASoC: core: Add support for platform and CODEC drivers on same device
by Charles Keepax 15 Mar '13

15 Mar '13

Currently DAI playback and capture widgets are created during soc_probe_codec and soc_probe_platform, using the device associated with the DAI to check which widgets should be created. If a device registers both a CODEC and platform driver this leads the CODEC playback and capture widgets being overwritten by the widgets created by the platform probe. It is more sensible to retain the CODEC widgets as the most common use case for registering both a CODEC and platform driver on the same chip is a CODEC which contains a DSP for compressed playback. In this situation it is more sensible to attach the routing information to the CODEC and add a thin platform driver interface to link into the compressed API. So this patch will check for existing widgets during soc_probe_platform and only create new widgets if no existing ones exist. Signed-off-by: Charles Keepax <ckeepax(a)opensource.wolfsonmicro.com> --- sound/soc/soc-core.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index 04af2a6..53efe1d 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -1159,7 +1159,8 @@ static int soc_probe_platform(struct snd_soc_card *card, /* Create DAPM widgets for each DAI stream */ list_for_each_entry(dai, &dai_list, list) { - if (dai->dev != platform->dev) + if (dai->dev != platform->dev || + dai->playback_widget || dai->capture_widget) continue; snd_soc_dapm_new_dai_widgets(&platform->dapm, dai); -- 1.7.2.5

2 6

[alsa-devel] [PATCH V2 0/2] ASoC: OMAP2+: Update Audio IP with sDMA binding for DT boot
by Sebastien Guiriec 15 Mar '13

15 Mar '13

This patch serie is converting OMAP Audio IPs in order to use OMAP sDMA DT binding in case of DT boot on OMAP2+ platforms. It is depending on the next patches: [1] http://www.spinics.net/lists/arm-kernel/msg227836.html [2] http://www.mail-archive.com/linux-omap@vger.kernel.org/msg84668.html It has been tested on: - OMAP5 (sEVM/panda)in DT boot (McPDM/DMIC). - OMAP4 (SDP4430/Panda) in both DT/non DT boot (McPDM/DMIC) - OMAP3 (Beagle XM) in both DT/non DT boot (McBSP) Update since v1: - Put DTS patch in first. - Squash all OMAP IPs patch together. Sebastien Guiriec (2): ARM: dts: OMAP2+: Add SDMA Audio IPs bindings ASoC: OMAP2+: Move McBSP, DMIC and McPDM to generic DMA DT binding arch/arm/boot/dts/omap2420.dtsi | 6 ++++++ arch/arm/boot/dts/omap2430.dtsi | 15 +++++++++++++++ arch/arm/boot/dts/omap3.dtsi | 15 +++++++++++++++ arch/arm/boot/dts/omap4.dtsi | 17 +++++++++++++++++ arch/arm/boot/dts/omap5.dtsi | 14 ++++++++++++++ sound/soc/omap/mcbsp.c | 26 ++++++++++++++++---------- sound/soc/omap/omap-dmic.c | 15 +++++++++------ sound/soc/omap/omap-mcpdm.c | 22 ++++++++++++++-------- sound/soc/omap/omap-pcm.c | 12 ++++++++++-- sound/soc/omap/omap-pcm.h | 1 + 10 files changed, 117 insertions(+), 26 deletions(-) -- 1.7.10.4

6 16

[alsa-devel] [PATCH] ALSA: at73c213: Use dev_pm_ops
by Lars-Peter Clausen 15 Mar '13

15 Mar '13

Use dev_pm_ops instead of the deprecated legacy suspend/resume callbacks. Signed-off-by: Lars-Peter Clausen <lars(a)metafoo.de> --- sound/spi/at73c213.c | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 deletions(-) diff --git a/sound/spi/at73c213.c b/sound/spi/at73c213.c index 4dd60d8..a1a24b9 100644 --- a/sound/spi/at73c213.c +++ b/sound/spi/at73c213.c @@ -1075,10 +1075,11 @@ out: return 0; } -#ifdef CONFIG_PM -static int snd_at73c213_suspend(struct spi_device *spi, pm_message_t msg) +#ifdef CONFIG_PM_SLEEP + +static int snd_at73c213_suspend(struct device *dev) { - struct snd_card *card = dev_get_drvdata(&spi->dev); + struct snd_card *card = dev_get_drvdata(dev); struct snd_at73c213 *chip = card->private_data; ssc_writel(chip->ssc->regs, CR, SSC_BIT(CR_TXDIS)); @@ -1087,9 +1088,9 @@ static int snd_at73c213_suspend(struct spi_device *spi, pm_message_t msg) return 0; } -static int snd_at73c213_resume(struct spi_device *spi) +static int snd_at73c213_resume(struct device *dev) { - struct snd_card *card = dev_get_drvdata(&spi->dev); + struct snd_card *card = dev_get_drvdata(dev); struct snd_at73c213 *chip = card->private_data; clk_enable(chip->board->dac_clk); @@ -1097,18 +1098,21 @@ static int snd_at73c213_resume(struct spi_device *spi) return 0; } + +static SIMPLE_DEV_PM_OPS(at73c213_pm_ops, snd_at73c213_suspend, + snd_at73c213_resume); +#define AT73C213_PM_OPS (&at73c213_pm_ops) + #else -#define snd_at73c213_suspend NULL -#define snd_at73c213_resume NULL +#define AT73C213_PM_OPS NULL #endif static struct spi_driver at73c213_driver = { .driver = { .name = "at73c213", + .pm = AT73C213_PM_OPS, }, .probe = snd_at73c213_probe, - .suspend = snd_at73c213_suspend, - .resume = snd_at73c213_resume, .remove = snd_at73c213_remove, }; -- 1.8.0

1 0

Re: [alsa-devel] [PATCH 08/10] ASoC: dmaengine_pcm: Add open function for DT DMA request
by Shawn Guo 15 Mar '13

15 Mar '13

On Thu, Mar 14, 2013 at 02:08:32PM +0100, Markus Pargmann wrote: > On Wed, Mar 13, 2013 at 10:18:29AM +0800, Shawn Guo wrote: > > On Tue, Mar 12, 2013 at 07:02:07PM +0000, Mark Brown wrote: > > > On Sun, Mar 10, 2013 at 07:33:09PM +0100, Markus Pargmann wrote: > > > > Add a function to open a DMA PCM substream using devicetree data > > > > provided via the client device node. The patch introduces a public > > > > function and a private subfunction that is called by both open > > > > functions. > > > > > > Someone (I think it was Shawn) sent a very similar patch just recently > > > which appears to have fallen out of my inbox unfortunately - can you > > > please check what's going on there and coordinate with them? Let me > > > know if you can't find the patch and I'll dig it out. > > > > It's here, Markus. > > > > http://thread.gmane.org/gmane.linux.alsa.devel/106027/focus=106031 > > > > Thanks. Do you already have a patch for the non-generic function which I > could use instead? > I do not follow. With my patch in place, there will be two dmaengine_pcm APIs. int snd_dmaengine_pcm_open(struct snd_pcm_substream *substream, dma_filter_fn filter_fn, void *filter_data) int snd_dmaengine_generic_pcm_open(struct snd_pcm_substream *substream, struct device *dev, const char *name) The first one is for users that their dmaengine driver hasn't converted to generic device tree bindings, and the second one is for users that their dmaengine driver is converted to generic device tree bindings. Shawn

2 1

[alsa-devel] [PATCH 0/3] ALSA: hda/ca0132 - Better handle a failed DSP firmware load.
by Dylan Reid 15 Mar '13

15 Mar '13

Enabling CONFIG_SND_HDA_CODEC_CA0132_DSP on the Chromebook Pixel leads to 10+ a second delay between running aplay and hearing audio. This was caused by the driver thinking the DSP was loaded when it had actually failed. The first two patches avoid that situation, the DSP still fails to load every time, but at least the system continues without a painful delay. The third patch is just a cleanup that touches the same state variable. Dylan Reid (3): ALSA: hda/ca0132 - Check if dspload_image succeeded. ALSA: hda/ca0132 - Check download state of DSP. ALSA: hda/ca0132 - Remove extra setting of dsp_state. sound/pci/hda/patch_ca0132.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) -- 1.8.1.3.605.g02339dd

2 4

[alsa-devel] [patch] sound: sequencer: cap array index in seq_chn_common_event()
by Dan Carpenter 15 Mar '13

15 Mar '13

"chn" here is a number between 0 and 255, but ->chn_info[] only has 16 elements so there is a potential write beyond the end of the array. If the seq_mode isn't SEQ_2 then we let the individual drivers (either opl3.c or midi_synth.c) handle it. Those functions all do a bounds check on "chn" so I haven't changed anything here. The opl3.c driver has up to 18 channels and not 16. Signed-off-by: Dan Carpenter <dan.carpenter(a)oracle.com> --- Static checker stuff. diff --git a/sound/oss/sequencer.c b/sound/oss/sequencer.c index 30bcfe4..4ff60a6 100644 --- a/sound/oss/sequencer.c +++ b/sound/oss/sequencer.c @@ -545,6 +545,9 @@ static void seq_chn_common_event(unsigned char *event_rec) case MIDI_PGM_CHANGE: if (seq_mode == SEQ_2) { + if (chn > 15) + break; + synth_devs[dev]->chn_info[chn].pgm_num = p1; if ((int) dev >= num_synths) synth_devs[dev]->set_instr(dev, chn, p1); @@ -596,6 +599,9 @@ static void seq_chn_common_event(unsigned char *event_rec) case MIDI_PITCH_BEND: if (seq_mode == SEQ_2) { + if (chn > 15) + break; + synth_devs[dev]->chn_info[chn].bender_value = w14; if ((int) dev < num_synths)

2 1

[alsa-devel] [PATCH 1/2] ASoC: cs4271: switch to mute_stream
by Daniel Mack 15 Mar '13

15 Mar '13

Use the newly introduced mute_stream DAI operation, and don't mute the codec if it's called for the _CAPTURE stream. Signed-off-by: Daniel Mack <zonque(a)gmail.com> --- sound/soc/codecs/cs4271.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/sound/soc/codecs/cs4271.c b/sound/soc/codecs/cs4271.c index ac0d3b4..03036b3 100644 --- a/sound/soc/codecs/cs4271.c +++ b/sound/soc/codecs/cs4271.c @@ -388,7 +388,7 @@ static int cs4271_hw_params(struct snd_pcm_substream *substream, return cs4271_set_deemph(codec); } -static int cs4271_digital_mute(struct snd_soc_dai *dai, int mute) +static int cs4271_mute_stream(struct snd_soc_dai *dai, int mute, int stream) { struct snd_soc_codec *codec = dai->codec; struct cs4271_private *cs4271 = snd_soc_codec_get_drvdata(codec); @@ -396,6 +396,9 @@ static int cs4271_digital_mute(struct snd_soc_dai *dai, int mute) int val_a = 0; int val_b = 0; + if (stream != SNDRV_PCM_STREAM_PLAYBACK) + return 0; + if (mute) { val_a = CS4271_VOLA_MUTE; val_b = CS4271_VOLB_MUTE; @@ -442,7 +445,7 @@ static const struct snd_soc_dai_ops cs4271_dai_ops = { .hw_params = cs4271_hw_params, .set_sysclk = cs4271_set_dai_sysclk, .set_fmt = cs4271_set_dai_fmt, - .digital_mute = cs4271_digital_mute, + .mute_stream = cs4271_mute_stream, }; static struct snd_soc_dai_driver cs4271_dai = { -- 1.8.1.4

1 1

[alsa-devel] [PATCH] ASoC: dapm: Fix pointer dereference in is_connected_output_ep()
by Peter Ujfalusi 15 Mar '13

15 Mar '13

*path is not yet initialized when we check if the widget is connected. The compiler also warns about this: sound/soc/soc-dapm.c: In function 'is_connected_output_ep': sound/soc/soc-dapm.c:824:18: warning: 'path' may be used uninitialized in this function Signed-off-by: Peter Ujfalusi <peter.ujfalusi(a)ti.com> --- Hi Mark, Liam, This patch is for the for-next branch. I got NULL pointer dereference because of this depending on the used toolchain. Regards, Peter sound/soc/soc-dapm.c | 1 - 1 file changed, 1 deletion(-) diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index eaf9b74..33acd8b 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -821,7 +821,6 @@ static int is_connected_output_ep(struct snd_soc_dapm_widget *widget, (widget->id == snd_soc_dapm_line && !list_empty(&widget->sources))) { widget->outputs = snd_soc_dapm_suspend_check(widget); - path->walking = 0; return widget->outputs; } } -- 1.8.1.5

2 1

[alsa-devel] [PATCH 0/2] ASoC: add new struct snd_soc_chip
by Kuninori Morimoto 15 Mar '13

15 Mar '13

Hi Mark, Liam, Stephen, Lars In my understanding on ASoC, - We will clean-up ASoC code. - CODEC/CPU will be common struct (platform too ??) As 1st trigger for this common struct, I created very basic struct snd_soc_chip. I hope it is accepted, and updated to be more useful. Kuninori Morimoto (2): ASoC: add snd_soc_register_chip() ASoC: fsi: use snd_soc_register_chip() instead of snd_soc_register_dais() include/sound/soc.h | 19 +++++++++++++ sound/soc/sh/fsi.c | 11 +++++--- sound/soc/soc-core.c | 77 ++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 103 insertions(+), 4 deletions(-)

4 18