November 2014 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH] ASoC: rt286: build warning of section mismatch
by Sudip Mukherjee 18 Nov '14

18 Nov '14

From: Sudip Mukherjee <sudip(a)vectorindia.org> while building we were getting the following build warning: Section mismatch in reference from the function rt286_i2c_probe() to the variable .init.data:force_combo_jack_table The function rt286_i2c_probe() references the variable __initdata force_combo_jack_table. This is often because rt286_i2c_probe lacks a __initdata annotation or the annotation of force_combo_jack_table is wrong. we were getting the warning as force_combo_jack_table was marked with __initdata Signed-off-by: Sudip Mukherjee <sudip(a)vectorindia.org> --- sound/soc/codecs/rt286.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/codecs/rt286.c b/sound/soc/codecs/rt286.c index 2e818aa..2cd4fe4 100644 --- a/sound/soc/codecs/rt286.c +++ b/sound/soc/codecs/rt286.c @@ -1206,7 +1206,7 @@ static const struct acpi_device_id rt286_acpi_match[] = { }; MODULE_DEVICE_TABLE(acpi, rt286_acpi_match); -static struct dmi_system_id force_combo_jack_table[] __initdata = { +static struct dmi_system_id force_combo_jack_table[] = { { .ident = "Intel Wilson Beach", .matches = { -- 1.8.1.2

2 1

[alsa-devel] [PATCH 00/11] ASoC: Untangle AC'97 support form the core
by Lars-Peter Clausen 18 Nov '14

18 Nov '14

The AC'97 support has had a special role within the ASoC framework with it being (in contrast to other audio buses) tightly integrated into the core. This patch series creates a clean separation between the core and the AC'97 support making AC'97 from the point of view of the core just another bus. Some of the changes in this series are required for the upcoming conversion of the AC'97 drivers to regmap, the other changes just fit nicely with the idea of componentization and having a generic ASoC core. The series starts with moving AC'97 support to its own file and doing a few general cleanups. After that it drops support of two features which were mainly responsible for the tight integration but are seemingly unused. This is a) Delaying the initialization and registration of the AC'97 device until after the card has been registered. There doesn't seem to be a good reason to delay to registration, all the information we need to register the device is available at the time when the device is created. Furthermore this will be needed for regmap. b) Setting platform data for the AC'97 via the CPU DAI. This hasn't been used in the last 5 years and doesn't seem to be that useful in general. If we should ever have a need for those features again they can still be implemented on top of the current code while keeping the clean separation. But for the time being it is simpler to just drop unused code rather than restructuring it. Finally the series pushes the pointer to the AC'97 device from the snd_soc_codec struct out to the driver private structs. Depends on asoc/topic/ad1980, asoc/topic/wm9705 and asoc/topic/wm971x - Lars Lars-Peter Clausen (11): ASoC: mpc5200_psc_ac97: Remove unused on-stack snd_ac97 device ASoC: mpc5200_dma: Don't overwrite ac97 device private_data ASoC: Properly handle AC'97 device lifetime management ASoC: Move AC'97 support to its own file ASoC: ac97: Use static ac97_bus ASoC: ac97: Merge soc_ac97_dev_{un,}register()/soc_{un,}register_ac97_codec() ASoC: ac97: Drop support for setting platform data via the CPU DAI ASoC: ac97: Drop delayed device registration ASoC: Drop ac97_control initialization from CODEC driver DAIs ASoC: Rename snd_soc_dai_driver struct ac97_control field to bus_control ASoC: ac97: Push snd_ac97 pointer to the driver level include/sound/soc-dai.h | 4 +- include/sound/soc.h | 25 ++- sound/soc/Makefile | 4 + sound/soc/au1x/ac97c.c | 2 +- sound/soc/au1x/psc-ac97.c | 2 +- sound/soc/blackfin/bf5xx-ac97.c | 2 +- sound/soc/cirrus/ep93xx-ac97.c | 2 +- sound/soc/codecs/ac97.c | 18 +- sound/soc/codecs/ad1980.c | 28 +-- sound/soc/codecs/stac9766.c | 40 +++-- sound/soc/codecs/wm9705.c | 32 ++-- sound/soc/codecs/wm9712.c | 33 ++-- sound/soc/codecs/wm9713.c | 32 ++-- sound/soc/fsl/fsl_ssi.c | 2 +- sound/soc/fsl/imx-ssi.c | 2 +- sound/soc/fsl/mpc5200_dma.c | 3 - sound/soc/fsl/mpc5200_psc_ac97.c | 6 +- sound/soc/nuc900/nuc900-ac97.c | 2 +- sound/soc/pxa/pxa2xx-ac97.c | 6 +- sound/soc/samsung/ac97.c | 4 +- sound/soc/sh/hac.c | 2 +- sound/soc/soc-ac97.c | 256 +++++++++++++++++++++++++++ sound/soc/soc-core.c | 372 ++------------------------------------- sound/soc/tegra/tegra20_ac97.c | 2 +- sound/soc/txx9/txx9aclc-ac97.c | 2 +- 25 files changed, 428 insertions(+), 455 deletions(-) create mode 100644 sound/soc/soc-ac97.c -- 1.8.0

3 13

[alsa-devel] [PATCH 1/2] ALSA: hda - move DELL_WMI_MIC_MUTE_LED to the tail in the quirk chain
by Hui Wang 18 Nov '14

18 Nov '14

We have one more Dell machine needs DELL_WMI_MIC_MUTE_LED quirk, but the machine uses alc293 instead of alc255. So if DELL_WMI_MIC_MUTE_LED still chain ALC255_FIXUP_DELL1_MIC_NO_PRESENCE, the machine can't use this quirk. To change this situation, let the DELL_WMI_MIC_MUTE_LED to be a standalone quirk, and let other quirks chain it. After this change, this quirk can be chained to any existing quirks, and as a result, it is possible that this quirk is applied to a non-Dell machine or a Dell machine without mic mute led on it, but it is still safe since alc_fixup_dell_wmi() will return an error in these situations. And remove the quirk for machine with subsystem id 0x6010 and 0x601f, these two machines will fall back to the quirk ALC255_FIXUP_DELL1_MIC_NO_PRESENCE-->ALC255_FIXUP_HEADSET_MODE--> ALC255_FIXUP_DELL_WMI_MIC_MUTE_LED through pin_fixup_tbl[]. BugLink: https://bugs.launchpad.net/bugs/1381856 Reported-and-Tested-by: Po-Hsu Lin <po-hsu.lin(a)canonical.com> Signed-off-by: Hui Wang <hui.wang(a)canonical.com> --- sound/pci/hda/patch_realtek.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 1723954..9da5798 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -4709,6 +4709,8 @@ static const struct hda_fixup alc269_fixups[] = { [ALC255_FIXUP_HEADSET_MODE] = { .type = HDA_FIXUP_FUNC, .v.func = alc_fixup_headset_mode_alc255, + .chained = true, + .chain_id = ALC255_FIXUP_DELL_WMI_MIC_MUTE_LED }, [ALC255_FIXUP_HEADSET_MODE_NO_HP_MIC] = { .type = HDA_FIXUP_FUNC, @@ -4744,8 +4746,6 @@ static const struct hda_fixup alc269_fixups[] = { [ALC255_FIXUP_DELL_WMI_MIC_MUTE_LED] = { .type = HDA_FIXUP_FUNC, .v.func = alc_fixup_dell_wmi, - .chained_before = true, - .chain_id = ALC255_FIXUP_DELL1_MIC_NO_PRESENCE }, [ALC282_FIXUP_ASPIRE_V5_PINS] = { .type = HDA_FIXUP_PINS, @@ -4783,10 +4783,8 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1028, 0x05f4, "Dell", ALC269_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x05f5, "Dell", ALC269_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x05f6, "Dell", ALC269_FIXUP_DELL1_MIC_NO_PRESENCE), - SND_PCI_QUIRK(0x1028, 0x0610, "Dell", ALC255_FIXUP_DELL_WMI_MIC_MUTE_LED), SND_PCI_QUIRK(0x1028, 0x0615, "Dell Vostro 5470", ALC290_FIXUP_SUBWOOFER_HSJACK), SND_PCI_QUIRK(0x1028, 0x0616, "Dell Vostro 5470", ALC290_FIXUP_SUBWOOFER_HSJACK), - SND_PCI_QUIRK(0x1028, 0x061f, "Dell", ALC255_FIXUP_DELL_WMI_MIC_MUTE_LED), SND_PCI_QUIRK(0x1028, 0x0638, "Dell Inspiron 5439", ALC290_FIXUP_MONO_SPEAKERS_HSJACK), SND_PCI_QUIRK(0x1028, 0x064a, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x064b, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), -- 1.9.1

2 3

[alsa-devel] [PATCH] ALSA: hda: separate stream tag for playback and capture
by Vinod Koul 18 Nov '14

18 Nov '14

From: Rafal Redzimski <rafal.f.redzimski(a)intel.com> According to hda specification stream tag must be unique throughout the input streams group, however an output stream might use a stream tag which is already in use by an input stream. Newer Intel HW has the capability to provides a total of more than 15 stream DMA engines which with the current implementation causes an overflow on STRM field (and the whole register) which results in usage of reserved value 0 in the STRM field which confuses HDA controller. So make stream_tag assignment separate for input and output streams Signed-off-by: Rafal Redzimski <rafal.f.redzimski(a)intel.com> Signed-off-by: Vinod Koul <vinod.koul(a)intel.com> --- sound/pci/hda/hda_controller.c | 14 ++++++++++++-- 1 files changed, 12 insertions(+), 2 deletions(-) diff --git a/sound/pci/hda/hda_controller.c b/sound/pci/hda/hda_controller.c index 8337645..703ac6e 100644 --- a/sound/pci/hda/hda_controller.c +++ b/sound/pci/hda/hda_controller.c @@ -1921,11 +1921,17 @@ int azx_mixer_create(struct azx *chip) } EXPORT_SYMBOL_GPL(azx_mixer_create); +static bool is_input_stream(struct azx *chip, unsigned char index) +{ + return index < chip->playback_index_offset; +} /* initialize SD streams */ int azx_init_stream(struct azx *chip) { int i; + int in_stream_tag = 0; + int out_stream_tag = 0; /* initialize each stream (aka device) * assign the starting bdl address to each stream (device) @@ -1938,9 +1944,13 @@ int azx_init_stream(struct azx *chip) azx_dev->sd_addr = chip->remap_addr + (0x20 * i + 0x80); /* int mask: SDI0=0x01, SDI1=0x02, ... SDO3=0x80 */ azx_dev->sd_int_sta_mask = 1 << i; - /* stream tag: must be non-zero and unique */ azx_dev->index = i; - azx_dev->stream_tag = i + 1; + + /* stream tag must be unique throughout the stream direction + * group, valid values 1...15 + */ + azx_dev->stream_tag = is_input_stream(chip, i) ? + ++in_stream_tag : ++out_stream_tag; } return 0; -- 1.7.0.4

2 1

[alsa-devel] ASoC: omap-mcbsp: Deletion of an unnecessary check before the function call "kfree"
by SF Markus Elfring 18 Nov '14

18 Nov '14

From: Markus Elfring <elfring(a)users.sourceforge.net> Date: Mon, 17 Nov 2014 14:05:27 +0100 The kfree() function tests whether its argument is NULL and then returns immediately. Thus the test around the call is not needed. This issue was detected by using the Coccinelle software. Signed-off-by: Markus Elfring <elfring(a)users.sourceforge.net> --- sound/soc/omap/mcbsp.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/sound/soc/omap/mcbsp.c b/sound/soc/omap/mcbsp.c index 86c7538..68a1252 100644 --- a/sound/soc/omap/mcbsp.c +++ b/sound/soc/omap/mcbsp.c @@ -621,8 +621,7 @@ void omap_mcbsp_free(struct omap_mcbsp *mcbsp) mcbsp->reg_cache = NULL; spin_unlock(&mcbsp->lock); - if (reg_cache) - kfree(reg_cache); + kfree(reg_cache); } /* -- 2.1.3

3 2

[alsa-devel] [PATCH] ASoC: fsl_ssi: Remove comment about SSI running only in slave mode
by Fabio Estevam 18 Nov '14

18 Nov '14

From: Fabio Estevam <fabio.estevam(a)freescale.com> Current driver can also run in I2S master mode, so remove the old comment. Signed-off-by: Fabio Estevam <fabio.estevam(a)freescale.com> --- sound/soc/fsl/fsl_ssi.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index ab05262..1ef0c84 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -67,8 +67,6 @@ /** * FSLSSI_I2S_FORMATS: audio formats supported by the SSI * - * This driver currently only supports the SSI running in I2S slave mode. - * * The SSI has a limitation in that the samples must be in the same byte * order as the host CPU. This is because when multiple bytes are written * to the STX register, the bytes and bits must be written in the same -- 1.9.1

3 2

[alsa-devel] [PATCH] ASoC: rt5677: Modify the default value of the MX-8E[4] for ASRC function
by Oder Chiou 18 Nov '14

18 Nov '14

Modify the default value of the MX-8E[4] to 1 for ASRC function. It could prevent the pop noise with ASRC function. Signed-off-by: Oder Chiou <oder_chiou(a)realtek.com> --- sound/soc/codecs/rt5677.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/sound/soc/codecs/rt5677.c b/sound/soc/codecs/rt5677.c index d659237..49ac9ba 100644 --- a/sound/soc/codecs/rt5677.c +++ b/sound/soc/codecs/rt5677.c @@ -55,7 +55,8 @@ static const struct regmap_range_cfg rt5677_ranges[] = { }; static const struct reg_default init_list[] = { - {RT5677_PR_BASE + 0x3d, 0x364d}, + {RT5677_ASRC_12, 0x0018}, + {RT5677_PR_BASE + 0x3d, 0x364d}, {RT5677_PR_BASE + 0x17, 0x4fc0}, {RT5677_PR_BASE + 0x13, 0x0312}, {RT5677_PR_BASE + 0x1e, 0x0000}, @@ -173,7 +174,7 @@ static const struct reg_default rt5677_reg[] = { {RT5677_ASRC_9 , 0x0000}, {RT5677_ASRC_10 , 0x0000}, {RT5677_ASRC_11 , 0x0000}, - {RT5677_ASRC_12 , 0x0008}, + {RT5677_ASRC_12 , 0x0018}, {RT5677_ASRC_13 , 0x0000}, {RT5677_ASRC_14 , 0x0000}, {RT5677_ASRC_15 , 0x0000}, -- 1.8.1.1.439.g50a6b54

2 1

[alsa-devel] ASoC fixes for v3.18
by Mark Brown 17 Nov '14

17 Nov '14

The following changes since commit fc14f9c1272f62c3e8d01300f52467c0d9af50f9: Linux 3.18-rc5 (2014-11-16 16:36:20 -0800) are available in the git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git tags/asoc-v3.18-rc5 for you to fetch changes up to cf9a7f7823c67243da44da2ac47ca944a3108282: Merge remote-tracking branches 'asoc/fix/rt5670', 'asoc/fix/samsung' and 'asoc/fix/sgtl5000' into asoc-linus (2014-11-17 16:41:11 +0000) ---------------------------------------------------------------- ASoC: Fixes for v3.18 As well as the usual driver fixes there's a few other things here: One is a fix for a race in DPCM which is unfortuantely a rather large diffstat, this is the result of growing usage of the mainline code and hence more detailed testing so I'm relatively happy. The other is a fix for non-DT machine driver matching following some of the componentization work which is much more focused. Both have had a while to cook in -next. ---------------------------------------------------------------- Andreas Färber (1): ASoC: samsung: Add MODULE_DEVICE_TABLE for Snow Axel Lin (1): ASoC: es8328-i2c: Fix i2c_device_id name field in es8328_id Bard Liao (2): ASoC: rt5670: correct the incorrect default values ASoC: rt5670: change dapm routes of PLL connection Charles Keepax (1): ASoC: wm_adsp: Avoid attempt to free buffers that might still be in use Dylan Reid (1): ASoC: max98090: Correct pclk divisor settings Fabio Estevam (1): ASoC: sgtl5000: Fix SMALL_POP bit definition Fang, Yang A (1): ASoC: rt5645: Mark RT5645_TDM_CTRL_3 as readable Jianqun (1): ASoC: rockchip-i2s: fix infinite loop in rockchip_snd_rxctrl Kuninori Morimoto (2): ASoC: fsi: remove unsupported PAUSE flag ASoC: rsnd: remove unsupported PAUSE flag Lars-Peter Clausen (1): ASoC: Fix snd_soc_find_dai() matching component by name Mark Brown (3): Merge remote-tracking branch 'asoc/fix/core' into asoc-linus Merge remote-tracking branches 'asoc/fix/adsp', 'asoc/fix/cs41l51', 'asoc/fix/dpcm', 'asoc/fix/es8328', 'asoc/fix/fsl-asrc', 'asoc/fix/max98090', 'asoc/fix/rcar', 'asoc/fix/rockchip' and 'asoc/fix/rt5645' into asoc-linus Merge remote-tracking branches 'asoc/fix/rt5670', 'asoc/fix/samsung' and 'asoc/fix/sgtl5000' into asoc-linus Nicolin Chen (1): ASoC: fsl_asrc: Add reg_defaults for regmap to fix kernel dump Takashi Iwai (1): ASoC: dpcm: Fix race between FE/BE updates and trigger Thomas Petazzoni (1): ASoC: cs42l51: re-hook of_match_table pointer include/sound/soc-dpcm.h | 2 ++ sound/soc/codecs/cs42l51-i2c.c | 1 + sound/soc/codecs/cs42l51.c | 4 ++- sound/soc/codecs/cs42l51.h | 1 + sound/soc/codecs/es8328-i2c.c | 2 +- sound/soc/codecs/max98090.c | 6 ++-- sound/soc/codecs/rt5645.c | 2 ++ sound/soc/codecs/rt5670.c | 36 ++++++++++---------- sound/soc/codecs/sgtl5000.c | 3 +- sound/soc/codecs/sgtl5000.h | 2 +- sound/soc/codecs/wm_adsp.c | 1 + sound/soc/fsl/fsl_asrc.c | 26 ++++++++++++++ sound/soc/rockchip/rockchip_i2s.c | 4 ++- sound/soc/samsung/snow.c | 1 + sound/soc/sh/fsi.c | 3 +- sound/soc/sh/rcar/core.c | 3 +- sound/soc/soc-core.c | 2 +- sound/soc/soc-pcm.c | 72 ++++++++++++++++++++++++++++++--------- 18 files changed, 123 insertions(+), 48 deletions(-)

2 1

[alsa-devel] [PATCH] ASoC: samsung-i2s: Add 192kHz config option
by Richard Fitzgerald 17 Nov '14

17 Nov '14

Adds QUIRK_SUPPORTS_192KHZ to allow 192kHz rate to be selected for hardware that supports it. Signed-off-by: Richard Fitzgerald <rf(a)opensource.wolfsonmicro.com> --- .../devicetree/bindings/sound/samsung-i2s.txt | 2 ++ include/linux/platform_data/asoc-s3c.h | 1 + sound/soc/samsung/i2s.c | 8 ++++++++ 3 files changed, 11 insertions(+), 0 deletions(-) diff --git a/Documentation/devicetree/bindings/sound/samsung-i2s.txt b/Documentation/devicetree/bindings/sound/samsung-i2s.txt index d188296..c3f3950 100644 --- a/Documentation/devicetree/bindings/sound/samsung-i2s.txt +++ b/Documentation/devicetree/bindings/sound/samsung-i2s.txt @@ -40,6 +40,8 @@ Optional SoC Specific Properties: sub system(used in secondary sound source). - pinctrl-0: Should specify pin control groups used for this controller. - pinctrl-names: Should contain only one value - "default". +- samsung,supports-192khz: specify this (without a value) if you want to allow + 192000 sample rate over the I2S link. Example: diff --git a/include/linux/platform_data/asoc-s3c.h b/include/linux/platform_data/asoc-s3c.h index 5e0bc77..6d0702d 100644 --- a/include/linux/platform_data/asoc-s3c.h +++ b/include/linux/platform_data/asoc-s3c.h @@ -28,6 +28,7 @@ struct samsung_i2s { #define QUIRK_NEED_RSTCLR (1 << 3) #define QUIRK_SUPPORTS_TDM (1 << 4) #define QUIRK_SUPPORTS_IDMA (1 << 5) +#define QUIRK_SUPPORTS_192KHZ (1 << 6) /* Quirks of the I2S controller */ u32 quirks; dma_addr_t idma_addr; diff --git a/sound/soc/samsung/i2s.c b/sound/soc/samsung/i2s.c index 947352d..f688e89 100644 --- a/sound/soc/samsung/i2s.c +++ b/sound/soc/samsung/i2s.c @@ -1219,6 +1219,9 @@ static int samsung_i2s_probe(struct platform_device *pdev) "specified"); } } + + if (of_find_property(np, "samsung,supports-192khz", NULL)) + quirks |= QUIRK_SUPPORTS_192KHZ; } res = platform_get_resource(pdev, IORESOURCE_MEM, 0); @@ -1247,6 +1250,11 @@ static int samsung_i2s_probe(struct platform_device *pdev) if (quirks & QUIRK_PRI_6CHAN) pri_dai->i2s_dai_drv.playback.channels_max = 6; + if (quirks & QUIRK_SUPPORTS_192KHZ) { + pri_dai->i2s_dai_drv.playback.rates |= SNDRV_PCM_RATE_192000; + pri_dai->i2s_dai_drv.capture.rates |= SNDRV_PCM_RATE_192000; + } + if (quirks & QUIRK_SEC_DAI) { sec_dai = i2s_alloc_dai(pdev, true); if (!sec_dai) { -- 1.7.2.5

2 1

[alsa-devel] [PATCH] ASoC: wm_adsp: Avoid attempt to free buffers that might still be in use
by Charles Keepax 17 Nov '14

17 Nov '14

We should not free any buffers associated with writing out coefficients to the DSP until all the async writes have completed. This patch updates the out of memory path when allocating a new buffer to include a call to regmap_async_complete. Reported-by: JS Park <aitdark.park(a)samsung.com> Signed-off-by: Charles Keepax <ckeepax(a)opensource.wolfsonmicro.com> --- sound/soc/codecs/wm_adsp.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/sound/soc/codecs/wm_adsp.c b/sound/soc/codecs/wm_adsp.c index 0a08ef5e..08d8ec3 100644 --- a/sound/soc/codecs/wm_adsp.c +++ b/sound/soc/codecs/wm_adsp.c @@ -1343,6 +1343,7 @@ static int wm_adsp_load_coeff(struct wm_adsp *dsp) file, blocks, pos - firmware->size); out_fw: + regmap_async_complete(regmap); release_firmware(firmware); wm_adsp_buf_free(&buf_list); out: -- 1.7.2.5

2 1