December 2011 - Alsa-devel - mailman.alsa-project.org

Re: [alsa-devel] [PATCH] ALSA HDA: add support for AMD SI audio
by Takashi Iwai 13 Dec '11

13 Dec '11

At Tue, 13 Dec 2011 16:42:12 +0800, Andiry Xu wrote: > > This patch adds support for AMD Southern Island HD audio. Could you give some more background for this patch? That is, it's enabling non-snooping mode for which reason, and why new driver_type is needed, etc. thanks, Takashi > > Signed-off-by: Andiry Xu <andiry.xu(a)amd.com> > --- > sound/pci/hda/hda_intel.c | 14 ++++++++++++++ > 1 files changed, 14 insertions(+), 0 deletions(-) > > diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c > index d1582dd..ae6b68c 100644 > --- a/sound/pci/hda/hda_intel.c > +++ b/sound/pci/hda/hda_intel.c > @@ -487,6 +487,7 @@ enum { > AZX_DRIVER_SCH, > AZX_DRIVER_ATI, > AZX_DRIVER_ATIHDMI, > + AZX_DRIVER_SI_ATIHDMI, > AZX_DRIVER_VIA, > AZX_DRIVER_SIS, > AZX_DRIVER_ULI, > @@ -533,6 +534,7 @@ static char *driver_short_names[] __devinitdata = { > [AZX_DRIVER_SCH] = "HDA Intel MID", > [AZX_DRIVER_ATI] = "HDA ATI SB", > [AZX_DRIVER_ATIHDMI] = "HDA ATI HDMI", > + [AZX_DRIVER_SI_ATIHDMI] = "HDA ATI SI HDMI", > [AZX_DRIVER_VIA] = "HDA VIA VT82xx", > [AZX_DRIVER_SIS] = "HDA SIS966", > [AZX_DRIVER_ULI] = "HDA ULI M5461", > @@ -2678,6 +2680,9 @@ static int __devinit azx_create(struct snd_card *card, struct pci_dev *pci, > > chip->single_cmd = single_cmd; > chip->snoop = hda_snoop; > + /* Default non-snoop mode for ATI SI HDMI AUDIO */ > + if (chip->driver_type == AZX_DRIVER_SI_ATIHDMI) > + chip->snoop = 0; > > if (bdl_pos_adj[dev] < 0) { > switch (chip->driver_type) { > @@ -2776,6 +2781,7 @@ static int __devinit azx_create(struct snd_card *card, struct pci_dev *pci, > chip->capture_streams = ULI_NUM_CAPTURE; > break; > case AZX_DRIVER_ATIHDMI: > + case AZX_DRIVER_SI_ATIHDMI: > chip->playback_streams = ATIHDMI_NUM_PLAYBACK; > chip->capture_streams = ATIHDMI_NUM_CAPTURE; > break; > @@ -3037,6 +3043,14 @@ static DEFINE_PCI_DEVICE_TABLE(azx_ids) = { > .driver_data = AZX_DRIVER_ATIHDMI | AZX_DCAPS_PRESET_ATI_HDMI }, > { PCI_DEVICE(0x1002, 0xaa48), > .driver_data = AZX_DRIVER_ATIHDMI | AZX_DCAPS_PRESET_ATI_HDMI }, > + { PCI_DEVICE(0x1002, 0x9902), > + .driver_data = AZX_DRIVER_SI_ATIHDMI | AZX_DCAPS_PRESET_ATI_HDMI }, > + { PCI_DEVICE(0x1002, 0xaaa0), > + .driver_data = AZX_DRIVER_SI_ATIHDMI | AZX_DCAPS_PRESET_ATI_HDMI }, > + { PCI_DEVICE(0x1002, 0xaaa8), > + .driver_data = AZX_DRIVER_SI_ATIHDMI | AZX_DCAPS_PRESET_ATI_HDMI }, > + { PCI_DEVICE(0x1002, 0xaab0), > + .driver_data = AZX_DRIVER_SI_ATIHDMI | AZX_DCAPS_PRESET_ATI_HDMI }, > /* VIA VT8251/VT8237A */ > { PCI_DEVICE(0x1106, 0x3288), > .driver_data = AZX_DRIVER_VIA | AZX_DCAPS_POSFIX_VIA }, > -- > 1.7.4.1 > >

1 0

[alsa-devel] ASoC updates for 3.2
by Mark Brown 13 Dec '11

13 Dec '11

The following changes since commit cefcc03ffc9527dde56807339edb1719c8dbae5f: ASoC: Provide a more complete DMA driver stub (2011-12-06 10:05:07 +0000) are available in the git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-3.2 The usual story - nothing major fixed here but all small, useful fixes. Axel Lin (5): ASoC: Include linux/io.h for jz4740 codec ASoC: Fix build dependency for SND_SOC_JZ4740_CODEC ASoC: Fix a typo in jive_wm8750 ASoC: Include linux/module.h for smdk2443_wm9710 ASoC: Fix hx4700 error handling to free gpios if snd_soc_register_card fails Jonathan Neuschäfer (1): ASoC: WM8958: correctly show firmware magic on mismatch Lothar Waßmann (2): ASoC: mxs: Add missing MODULE_LICENSE("GPL") ASoC: mxs: Add appropriate MODULE_ALIAS() Mark Brown (1): ASoC: Fix WM8996 24.576MHz clock operation sound/soc/codecs/Kconfig | 2 +- sound/soc/codecs/jz4740.c | 1 + sound/soc/codecs/wm8958-dsp2.c | 2 ++ sound/soc/codecs/wm8996.c | 1 + sound/soc/mxs/mxs-pcm.c | 3 +++ sound/soc/mxs/mxs-sgtl5000.c | 1 + sound/soc/pxa/hx4700.c | 5 +++-- sound/soc/samsung/jive_wm8750.c | 3 +-- sound/soc/samsung/smdk2443_wm9710.c | 1 + 9 files changed, 14 insertions(+), 5 deletions(-)

2 2

Re: [alsa-devel] [PATCH 06/19] mx27vis-aic32x4: only register when running on the right machine
by Mark Brown 13 Dec '11

13 Dec '11

On Tue, Dec 13, 2011 at 07:31:38AM +0100, Eric Bénard wrote: > without this fix, a kernel compiled with imx_v4_v5_defconfig > and running on a cpuimx25sd leads to : I'll apply this but a better fix is to convert the driver to be a normal platform driver and only register the device on the appropraite boards.

1 0

[alsa-devel] [PATCH] ASoC: Rename ALC5632 MICBIAS to common name convention.
by Leon Romanovsky 13 Dec '11

13 Dec '11

Signed-off-by: Leon Romanovsky <leon(a)leon.nu> --- sound/soc/codecs/alc5632.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/codecs/alc5632.c b/sound/soc/codecs/alc5632.c index 08613c7..390e437 100644 --- a/sound/soc/codecs/alc5632.c +++ b/sound/soc/codecs/alc5632.c @@ -379,8 +379,8 @@ SND_SOC_DAPM_PGA("MIC1 PGA", ALC5632_PWR_MANAG_ADD3, 3, 0, NULL, 0), SND_SOC_DAPM_PGA("MIC2 PGA", ALC5632_PWR_MANAG_ADD3, 2, 0, NULL, 0), SND_SOC_DAPM_PGA("MIC1 Pre Amp", ALC5632_PWR_MANAG_ADD3, 1, 0, NULL, 0), SND_SOC_DAPM_PGA("MIC2 Pre Amp", ALC5632_PWR_MANAG_ADD3, 0, 0, NULL, 0), -SND_SOC_DAPM_SUPPLY("Mic Bias1", ALC5632_PWR_MANAG_ADD1, 3, 0, NULL, 0), -SND_SOC_DAPM_SUPPLY("Mic Bias2", ALC5632_PWR_MANAG_ADD1, 2, 0, NULL, 0), +SND_SOC_DAPM_SUPPLY("MICBIAS1", ALC5632_PWR_MANAG_ADD1, 3, 0, NULL, 0), +SND_SOC_DAPM_SUPPLY("MICBIAS2", ALC5632_PWR_MANAG_ADD1, 2, 0, NULL, 0), SND_SOC_DAPM_PGA_E("D Amp", ALC5632_PWR_MANAG_ADD2, 14, 0, NULL, 0, amp_mixer_event, SND_SOC_DAPM_PRE_PMU | SND_SOC_DAPM_POST_PMD), -- 1.7.3.4

2 3

Re: [alsa-devel] [PATCH 3/3] sound/soc/lapis: add platform driver
by Mark Brown 13 Dec '11

13 Dec '11

On Mon, Nov 21, 2011 at 01:08:52PM +0900, Tomoya MORINAGA wrote: > +/* =================== I2S CH0 config =================== */ > +#define I2S_CH0_MCLK (12288000) /* Master Clock Frequency[Hz] */ This looks like it should be board specific? > + #if IOH_I2S_USE_PARAM > + #define I2S_CH0_FS_16000 16000 > + #define I2S_CH0_FS_32000 32000 > + #define I2S_CH0_FS_48000 48000 > + #else > + #define I2S_CH0_FS_8000 8000 > + #define I2S_CH0_FS_11025 11025 > + #define I2S_CH0_FS_22050 22050 > + #define I2S_CH0_FS_44100 44100 > + #endif What are these ifdefs for? This should be configured at runtime. > +/* select master or slave. The value is ioh_mssel_t */ > +#define I2S_CH0_MSSEL (ioh_mssel_master) > +/* select MCLK or MLBCLK into Master Clock. The value is enum ioh_masterclk_t */ > +#define I2S_CH0_MASTERCLKSEL (ioh_masterclksel_mclk) This and most of the following defines also look like compile time configuration for things that should be runtime configured. As with the previous patches in the series the big picture issue here is that you need to update your code to reflect the practices of modern mainline Linux code - here the major issue is that you appear to be doing all your configuration with #defines rather than by using runtime configuration from the machine driver as is normal for ASoC code. I got through a bunch more defines but stopped reviewing at some point as there was so much of this issue.

2 4

[alsa-devel] [PATCH 1/3] ASoC: Staticise rx51_aux_dev
by Axel Lin 13 Dec '11

13 Dec '11

Signed-off-by: Axel Lin <axel.lin(a)gmail.com> --- sound/soc/omap/rx51.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/sound/soc/omap/rx51.c b/sound/soc/omap/rx51.c index 4cabb74..ad16db5 100644 --- a/sound/soc/omap/rx51.c +++ b/sound/soc/omap/rx51.c @@ -365,7 +365,7 @@ static struct snd_soc_dai_link rx51_dai[] = { }, }; -struct snd_soc_aux_dev rx51_aux_dev[] = { +static struct snd_soc_aux_dev rx51_aux_dev[] = { { .name = "TLV320AIC34b", .codec_name = "tlv320aic3x-codec.2-0019", -- 1.7.5.4

2 3

[alsa-devel] [PATCH v2 1/6] ASoC: Allow device tree to specify a card's name
by Stephen Warren 13 Dec '11

13 Dec '11

If a card's device was instantiated from device tree, and the device tree has a "user-visible-name" property, use that as the card's name. Signed-off-by: Stephen Warren <swarren(a)nvidia.com> --- v2: New patch implementing new functionality Re: the binding documentation: * "SoC" here refers to the fact this is a binding oriented at System-on- chip audio complexes, rather than having to do with "ASoC"; both names were derived from the same root. * Do we need a compatible property for this "base class" binding at all? I think it's a good idea, even though the code doesn't actually rely on it. * Should the vendor field in the compatible property be "generic", "linux", or absent? I've tried to make these bindings generic and applicable to other OSs, so "linux," seems wrong. * Should the property "user-visible-name" have a "generic," prefix or similar? .../bindings/sound/soc-audio-complex.txt | 14 ++++++++ sound/soc/soc-core.c | 35 ++++++++++++++++++- 2 files changed, 47 insertions(+), 2 deletions(-) create mode 100644 Documentation/devicetree/bindings/sound/soc-audio-complex.txt diff --git a/Documentation/devicetree/bindings/sound/soc-audio-complex.txt b/Documentation/devicetree/bindings/sound/soc-audio-complex.txt new file mode 100644 index 0000000..1cc7059 --- /dev/null +++ b/Documentation/devicetree/bindings/sound/soc-audio-complex.txt @@ -0,0 +1,14 @@ +SoC Audio Complex + +Required properties: +- compatible : "generic,soc-audio-complex". Other compatible values + indicating the specific HW model will typically be present too. +- user-visible-name : The user-visible name of this sound complex. + +Example: + +sound { + compatible = "nvidia,tegra-audio-wm8903", + "generic,soc-audio-complex"; + user-visible-name = "tegra-wm8903-harmony"; +}; diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index 5195f06..56d1bc5 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -32,6 +32,7 @@ #include <linux/platform_device.h> #include <linux/ctype.h> #include <linux/slab.h> +#include <linux/of.h> #include <sound/ac97_codec.h> #include <sound/core.h> #include <sound/jack.h> @@ -2809,6 +2810,25 @@ int snd_soc_dai_digital_mute(struct snd_soc_dai *dai, int mute) } EXPORT_SYMBOL_GPL(snd_soc_dai_digital_mute); +/* Retrieve a card's name from device tree */ +static int snd_soc_of_parse_card_name(struct snd_soc_card *card) +{ + struct device_node *np = card->dev->of_node; + int ret; + + ret = of_property_read_string_index(np, "user-visible-name", 0, + &card->name); + /* + * EINVAL means the property does not exist. This is fine providing + * card->name was previously set, which is checked later in + * snd_soc_register_card. + */ + if (ret < 0 && ret != -EINVAL) + return ret; + + return 0; +} + /** * snd_soc_register_card - Register a card with the ASoC core * @@ -2817,11 +2837,22 @@ EXPORT_SYMBOL_GPL(snd_soc_dai_digital_mute); */ int snd_soc_register_card(struct snd_soc_card *card) { - int i; + int i, ret; - if (!card->name || !card->dev) + if (!card->dev) return -EINVAL; + if (card->dev->of_node) { + ret = snd_soc_of_parse_card_name(card); + if (ret < 0) + return ret; + } + + if (!card->name) { + dev_err(card->dev, "Card name is not set\n"); + return -EINVAL; + } + dev_set_drvdata(card->dev, card); snd_soc_initialize_card_lists(card); -- 1.7.0.4

3 18

[alsa-devel] QasTools 0.16.0 released
by Sebastian H. 13 Dec '11

13 Dec '11

QasTools version 0.16.0 is now available. QasTools is a collection of dektop applications for ALSA. Changes: * QasTools replaces separate QasConfig and QasMixer packages * QasHctl: QasHctl forked out of QasMixer into separate application * QasHctl: Wrong pixmap in switch widgets fixed * QasMixer: Command line arguments -D and -c fixed * QasMixer: Command line arguments descriptions added in manpage * All: Info dialog adapted to QasTools Homepage with more information http://xwmw.org/qastools Project page http://sourceforge.net/projects/qastools/ Happy volume changing! -- Sebastian Holtermann

1 0

[alsa-devel] [PATCH] plug: Move fast_ops copy after slave configuration.
by Reilly Grant 12 Dec '11

12 Dec '11

The plug plugin copies the fast_ops structure and fast_op_arg from its slave PCM on initialization and after inserting a new plugin. This works unless the slave PCM is also an instance of the plug plugin because plug can change its fast_ops structure during _snd_pcm_hw_params. Instead of copying fast_ops in snd_pcm_plug_insert_plugins wait until the end of snd_pcm_plug_hw_params when the slave has been fully initialized. This fixes a crash when two instances of plug are instantiated back to back and the format conversion plugin inserted by the second is skipped by the first because the fast_ops pointer is not properly updated. This is the case for applications requesting plug:default on Intel HDA chips. --- src/pcm/pcm_plug.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/src/pcm/pcm_plug.c b/src/pcm/pcm_plug.c index e9d2923..72456d5 100644 --- a/src/pcm/pcm_plug.c +++ b/src/pcm/pcm_plug.c @@ -659,8 +659,6 @@ static int snd_pcm_plug_insert_plugins(snd_pcm_t *pcm, } if (err) { plug->gen.slave = new; - pcm->fast_ops = new->fast_ops; - pcm->fast_op_arg = new->fast_op_arg; } k++; } @@ -1049,6 +1047,9 @@ static int snd_pcm_plug_hw_params(snd_pcm_t *pcm, snd_pcm_hw_params_t *params) } snd_pcm_unlink_hw_ptr(pcm, plug->req_slave); snd_pcm_unlink_appl_ptr(pcm, plug->req_slave); + + pcm->fast_ops = slave->fast_ops; + pcm->fast_op_arg = slave->fast_op_arg; snd_pcm_link_hw_ptr(pcm, slave); snd_pcm_link_appl_ptr(pcm, slave); return 0; -- 1.7.3.4

2 2

[alsa-devel] Problem calling "snd_pcm_hw_params_set_channels" with a RME 9652 (the same call -with more channels- works OK on a MADI one)
by Natanael Olaiz 12 Dec '11

12 Dec '11

Hi all, I have a problem with an application that worked OK on a RME Madi but now is failing with a 9652. It fails calling to snd_pcm_hw_params_set_channels . With the 9652 i'm getting an error -22 "invalid argument", with any number of channels (I understand it should be 24 ?) Any idea? Alsa ver. : 1.0.24 Kernel: 3.0.4 /proc/asound/cards : 0 [DSP ]: H-DSP - Hammerfall DSP RME Hammerfall HDSP 9652 at 0xfebf0000, irq 23 Thanks in advance, Natanael.

1 0