July 2016 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH] ASoC: rt5514-spi: Convert to use devm_* API
by Axel Lin 14 Jul '16

14 Jul '16

Use devm_* API to simplify the code. This patch also fixes the return value in probe error paths. Signed-off-by: Axel Lin <axel.lin(a)ingics.com> --- sound/soc/codecs/rt5514-spi.c | 25 ++++++------------------- 1 file changed, 6 insertions(+), 19 deletions(-) diff --git a/sound/soc/codecs/rt5514-spi.c b/sound/soc/codecs/rt5514-spi.c index 8a9382e..e1b326d 100644 --- a/sound/soc/codecs/rt5514-spi.c +++ b/sound/soc/codecs/rt5514-spi.c @@ -409,33 +409,21 @@ static int rt5514_spi_probe(struct spi_device *spi) rt5514_spi = spi; - ret = snd_soc_register_platform(&spi->dev, &rt5514_spi_platform); + ret = devm_snd_soc_register_platform(&spi->dev, &rt5514_spi_platform); if (ret < 0) { dev_err(&spi->dev, "Failed to register platform.\n"); - goto err_plat; + return ret; } - ret = snd_soc_register_component(&spi->dev, &rt5514_spi_dai_component, - &rt5514_spi_dai, 1); + ret = devm_snd_soc_register_component(&spi->dev, + &rt5514_spi_dai_component, + &rt5514_spi_dai, 1); if (ret < 0) { dev_err(&spi->dev, "Failed to register component.\n"); - goto err_comp; + return ret; } return 0; -err_comp: - snd_soc_unregister_platform(&spi->dev); -err_plat: - - return 0; -} - -static int rt5514_spi_remove(struct spi_device *spi) -{ - snd_soc_unregister_component(&spi->dev); - snd_soc_unregister_platform(&spi->dev); - - return 0; } static const struct of_device_id rt5514_of_match[] = { @@ -450,7 +438,6 @@ static struct spi_driver rt5514_spi_driver = { .of_match_table = of_match_ptr(rt5514_of_match), }, .probe = rt5514_spi_probe, - .remove = rt5514_spi_remove, }; module_spi_driver(rt5514_spi_driver); -- 2.5.0

2 1

[alsa-devel] [PATCH] ASoC: Intel: Skylake: Fix fw reload failure
by Vinod Koul 14 Jul '16

14 Jul '16

From: Senthilnathan Veppur <senthilnathanx.veppur(a)intel.com> FW reload had two issues: - We need to disable the core 0 on when fw fails - Before loading firmware mark boot flag as false This patch fixes these two Signed-off-by: Senthilnathan Veppur <senthilnathanx.veppur(a)intel.com> Signed-off-by: Vinod Koul <vinod.koul(a)intel.com> --- sound/soc/intel/skylake/bxt-sst.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/sound/soc/intel/skylake/bxt-sst.c b/sound/soc/intel/skylake/bxt-sst.c index 9c3750f49c21..16e2ed97d71a 100644 --- a/sound/soc/intel/skylake/bxt-sst.c +++ b/sound/soc/intel/skylake/bxt-sst.c @@ -139,7 +139,7 @@ static int sst_bxt_prepare_fw(struct sst_dsp *ctx, base_fw_load_failed: ctx->dsp_ops.cleanup(ctx->dev, &ctx->dmab, stream_tag); skl_dsp_core_power_down(ctx, SKL_DSP_CORE_MASK(1)); - skl_dsp_disable_core(ctx, SKL_DSP_CORE_MASK(1)); + skl_dsp_disable_core(ctx, SKL_DSP_CORE0_MASK); return ret; } @@ -232,6 +232,7 @@ static int bxt_set_dsp_D0(struct sst_dsp *ctx, unsigned int core_id) unsigned int core_mask = SKL_DSP_CORE_MASK(core_id); if (skl->fw_loaded == false) { + skl->boot_complete = false; ret = bxt_load_base_firmware(ctx); if (ret < 0) dev_err(ctx->dev, "reload fw failed: %d\n", ret); -- 1.9.1

2 1

[alsa-devel] [PATCH 0/5] ASoC: Intel: More Kabylake machine support
by Vinod Koul 14 Jul '16

14 Jul '16

We have two more kbl machines, one with nau88l25_ssm4567 and second one with nau88l25_max98357a. This is similar to skylake which had the same combinations. To add kbl IDs to existing machines we need to add names in id table. But current names are more than 20 characters which is limit for platform device id name names. So first truncate skl machines to less than 20 chars and then add new kbl names. Vinod Koul (5): ASoC: Intel: Skylake: reduce machine name for skl_nau88l25_ssm4567 ASoC: Intel: Skylake: reduce machine name for skl_nau88l25_max98357a ASoC: Intel: Kbl: add kabylake additional machine entries ASoC: Intel: board: add kabylake nau88l25_max98357a machine id ASoC: Intel: board: add kabylake nau88l25_ssm4567 machine id sound/soc/intel/boards/skl_nau88l25_max98357a.c | 12 ++++++++++-- sound/soc/intel/boards/skl_nau88l25_ssm4567.c | 12 ++++++++++-- sound/soc/intel/skylake/skl.c | 6 ++++-- 3 files changed, 24 insertions(+), 6 deletions(-) Thanks -- ~Vinod

2 10

[alsa-devel] [PATCH] ASoC: Intel: Fix conflicting pcm dev drvdata on haswell
by Ben Zhang 14 Jul '16

14 Jul '16

soc-core sets the snd_soc_pcm_runtime->dev drvdata to snd_soc_pcm_runtime in soc_post_component_init, and access it in places like codec_reg_show. hsw_pcm_open overwrites the drvdata to point to hsw_pcm_data, confusing soc-core, and causing crashes when cat /sys/devices/pci0000:00/INT3438:00/.../System PCM/codec_reg This patch removes the set in hsw_pcm_open since it's no longer used. commit 7ff9d6714a5c ("ASoC: Intel: Split hsw_pcm_data for playback and capture") already removed all calls to snd_soc_pcm_get_drvdata(rtd). Signed-off-by: Ben Zhang <benzh(a)chromium.org> --- sound/soc/intel/haswell/sst-haswell-pcm.c | 1 - 1 file changed, 1 deletion(-) diff --git a/sound/soc/intel/haswell/sst-haswell-pcm.c b/sound/soc/intel/haswell/sst-haswell-pcm.c index 994256b..3154525 100644 --- a/sound/soc/intel/haswell/sst-haswell-pcm.c +++ b/sound/soc/intel/haswell/sst-haswell-pcm.c @@ -819,7 +819,6 @@ static int hsw_pcm_open(struct snd_pcm_substream *substream) mutex_lock(&pcm_data->mutex); pm_runtime_get_sync(pdata->dev); - snd_soc_pcm_set_drvdata(rtd, pcm_data); pcm_data->substream = substream; snd_soc_set_runtime_hwparams(substream, &hsw_pcm_hardware); -- 2.8.0.rc3.226.g39d4020

2 1

[alsa-devel] [PATCH -next] ASoC: mediatek: mt2701: cHECK-fix non static symbol warning
by weiyj_lk＠163.com 14 Jul '16

14 Jul '16

From: Wei Yongjun <yongjun_wei(a)trendmicro.com.cn> Fixes the following sparse warning: sound/soc/mediatek/mt2701/mt2701-afe-pcm.c:72:5: warning: symbol 'mt2701_dai_num_to_i2s' was not declared. Should it be static? Signed-off-by: Wei Yongjun <yongjun_wei(a)trendmicro.com.cn> --- sound/soc/mediatek/mt2701/mt2701-afe-pcm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/mediatek/mt2701/mt2701-afe-pcm.c b/sound/soc/mediatek/mt2701/mt2701-afe-pcm.c index 15522c0..34a6123 100644 --- a/sound/soc/mediatek/mt2701/mt2701-afe-pcm.c +++ b/sound/soc/mediatek/mt2701/mt2701-afe-pcm.c @@ -69,7 +69,7 @@ static const struct mt2701_afe_rate mt2701_afe_i2s_rates[] = { { .rate = 352800, .regvalue = 24 }, }; -int mt2701_dai_num_to_i2s(struct mtk_base_afe *afe, int num) +static int mt2701_dai_num_to_i2s(struct mtk_base_afe *afe, int num) { int val = num - MT2701_IO_I2S;

3 3

[alsa-devel] [PATCH v8] ASoC: mediatek: Add HDMI dai-links to the mt8173-rt5650 machine driver
by Philipp Zabel 14 Jul '16

14 Jul '16

From: PC Liao <pc.liao(a)mediatek.com> This patch adds HDMI audio output support to the MT8173 RT5650 machine driver. Signed-off-by: PC Liao <pc.liao(a)mediatek.com> Signed-off-by: Philipp Zabel <p.zabel(a)pengutronix.de> --- Changes since v7: - Rebased onto asoc/for-next --- .../devicetree/bindings/sound/mt8173-rt5650.txt | 5 +++-- sound/soc/mediatek/Kconfig | 1 + sound/soc/mediatek/mt8173/mt8173-rt5650.c | 26 ++++++++++++++++++++++ 3 files changed, 30 insertions(+), 2 deletions(-) diff --git a/Documentation/devicetree/bindings/sound/mt8173-rt5650.txt b/Documentation/devicetree/bindings/sound/mt8173-rt5650.txt index f250fc7..29dce2a 100644 --- a/Documentation/devicetree/bindings/sound/mt8173-rt5650.txt +++ b/Documentation/devicetree/bindings/sound/mt8173-rt5650.txt @@ -1,8 +1,9 @@ -MT8173 with RT5650 CODECS +MT8173 with RT5650 CODECS and HDMI via I2S Required properties: - compatible : "mediatek,mt8173-rt5650" - mediatek,audio-codec: the phandles of rt5650 codecs + and of the hdmi encoder node - mediatek,platform: the phandle of MT8173 ASoC platform Optional subnodes: @@ -20,7 +21,7 @@ Example: sound { compatible = "mediatek,mt8173-rt5650"; - mediatek,audio-codec = <&rt5650>; + mediatek,audio-codec = <&rt5650 &hdmi0>; mediatek,platform = <&afe>; mediatek,mclk = <0>; codec-capture { diff --git a/sound/soc/mediatek/Kconfig b/sound/soc/mediatek/Kconfig index 2fbe543..05cf809 100644 --- a/sound/soc/mediatek/Kconfig +++ b/sound/soc/mediatek/Kconfig @@ -46,6 +46,7 @@ config SND_SOC_MT8173_RT5650 tristate "ASoC Audio driver for MT8173 with RT5650 codec" depends on SND_SOC_MT8173 && I2C select SND_SOC_RT5645 + select SND_SOC_HDMI_CODEC help This adds ASoC driver for Mediatek MT8173 boards with the RT5650 audio codec. diff --git a/sound/soc/mediatek/mt8173/mt8173-rt5650.c b/sound/soc/mediatek/mt8173/mt8173-rt5650.c index d478976..ba65f41 100644 --- a/sound/soc/mediatek/mt8173/mt8173-rt5650.c +++ b/sound/soc/mediatek/mt8173/mt8173-rt5650.c @@ -169,7 +169,9 @@ static struct snd_soc_dai_link_component mt8173_rt5650_codecs[] = { enum { DAI_LINK_PLAYBACK, DAI_LINK_CAPTURE, + DAI_LINK_HDMI, DAI_LINK_CODEC_I2S, + DAI_LINK_HDMI_I2S, }; /* Digital audio interface glue - connects codec <---> CPU */ @@ -195,6 +197,16 @@ static struct snd_soc_dai_link mt8173_rt5650_dais[] = { .dynamic = 1, .dpcm_capture = 1, }, + [DAI_LINK_HDMI] = { + .name = "HDMI", + .stream_name = "HDMI PCM", + .cpu_dai_name = "HDMI", + .codec_name = "snd-soc-dummy", + .codec_dai_name = "snd-soc-dummy-dai", + .trigger = {SND_SOC_DPCM_TRIGGER_POST, SND_SOC_DPCM_TRIGGER_POST}, + .dynamic = 1, + .dpcm_playback = 1, + }, /* Back End DAI links */ [DAI_LINK_CODEC_I2S] = { .name = "Codec", @@ -210,6 +222,13 @@ static struct snd_soc_dai_link mt8173_rt5650_dais[] = { .dpcm_playback = 1, .dpcm_capture = 1, }, + [DAI_LINK_HDMI_I2S] = { + .name = "HDMI BE", + .cpu_dai_name = "HDMIO", + .no_pcm = 1, + .codec_dai_name = "i2s-hifi", + .dpcm_playback = 1, + }, }; static struct snd_soc_card mt8173_rt5650_card = { @@ -284,6 +303,13 @@ static int mt8173_rt5650_dev_probe(struct platform_device *pdev) } } + mt8173_rt5650_dais[DAI_LINK_HDMI_I2S].codec_of_node = + of_parse_phandle(pdev->dev.of_node, "mediatek,audio-codec", 1); + if (!mt8173_rt5650_dais[DAI_LINK_HDMI_I2S].codec_of_node) { + dev_err(&pdev->dev, + "Property 'audio-codec' missing or invalid\n"); + return -EINVAL; + } card->dev = &pdev->dev; platform_set_drvdata(pdev, card); -- 2.8.1

2 1

[alsa-devel] [alsa-lib][PATCH 00/34] pcm/conf/alisp: replace usage of alloca() with automatic variables
by Takashi Sakamoto 14 Jul '16

14 Jul '16

Hi, A batch of this patchset is to save storage consumption and to improve execution performance, just for a little. Currently, ALSA userspace library doesn't disclose layouts of major structures to applications. The applications call some APIs in the library to get memory objects for the structures, then operates to corresponding opaque pointers. In most cases, the objects are kept on stack by calls of alloca(3). On the other hand, inner the library, the layouts are decided in pre-compile or compile time. Therefore, there's little needs to call of alloca(3). However, inner the library, we can see some usages of alloca(3). This brings two disadvantages to system. One is to execute more instructions to use stack than automatic variables, for the same task. Another is to enlarge size of shared object. Although both disadvantages are not so remarkable, it's better to improve them. This patchset replaces usages of alloca() with automatic variables just to raw structures. Additionally, this commit improves code formats in related functions. I can see below effect on my environment (Ubuntu 16.04, amd64, gcc) $ git checkout master $ git show HEAD commit 941bd150bef2560f3e38a3bf74d546447e3126d1 ... $ ./gitcompile --with-plugindir=/usr/lib/x86_64-linux-gnu/alsa-lib/ ... $ ls -l src/.libs/libasound.so.2.0.0 -rwxrwxr-x 1 mocchi mocchi 4794528 7月 14 22:22 src/.libs/libasound.so.2.0.0 $ git checkout remove-alloca-asound $ make ... $ ls -l src/.libs/libasound.so.2.0.0 -rwxrwxr-x 1 mocchi mocchi 4783712 7月 14 22:21 src/.libs/libasound.so.2.0.0 $ gcc -v ... gcc version 5.4.0 20160609 (Ubuntu 5.4.0-6ubuntu1~16.04.1) Takashi Sakamoto (34): pcm: change code formatting for snd_pcm_set_params() pcm: remove alloca() from snd_pcm_set_params() pcm: change code formatting for snd_pcm_get_params() pcm: remove alloca() from snd_pcm_get_params pcm: change code formatting for snd_pcm_direct_initialize_slave() pcm: remove alloca() from snd_pcm_direct_initialize_slave pcm: change code formatting for snd_pcm_direct_initialize_poll_fd() pcm: remove alloca() from snd_pcm_direct_initialize_poll_fd() pcm: change code formatting for snd_pcm_direct_set_timer_params() pcm: remove alloca() from snd_pcm_direct_set_timer_params pcm: remove alloca() from _snd_pcm_hook_ctl_elems_install() pcm: change code formatting for snd_pcm_hw_change_timer() pcm: remove alloca() from snd_pcm_hw_change_timer() pcm: remove alloca() from snd_pcm_query_chmaps_from_hw() pcm: remove alloca() from snd_pcm_hw_get_chmap() pcm: remove alloca() from snd_pcm_hw_set_chmap() pcm: remove alloca() from snd_spcm_init() pcm: remove alloca() from snd_spcm_init_duplex() pcm: change code formatting for softvol_load_control() pcm: remove alloca() from softvol_load_control() pcm: change code formatting for _snd_pcm_softvol_open() pcm: remove alloca() from _snd_pcm_softvol_open() conf: remove alloca() from snd_determine_driver() conf: remove alloca() from snd_func_card_id() conf: remove alloca() from snd_func_card_name() conf: remove alloca() from snd_func_pcm_id() conf: remove alloca() from snd_func_pcm_args_by_class() conf: remove alloca() from snd_func_private_pcm_subdevice() alisp: remove alloca() from FA_card_info() alisp: remove alloca() from FA_hctl_find_elem() alisp: remove alloca() from FA_hctl_elem_info() alisp: remove alloca() from FA_hctl_elem_read() alisp: remove alloca() from FA_hctl_elem_write() alisp: remove alloca() from FA_pcm_info() src/alisp/alisp_snd.c | 153 ++++++++++++++++----------------- src/confmisc.c | 51 +++++------ src/pcm/pcm.c | 232 ++++++++++++++++++++++++++++---------------------- src/pcm/pcm_direct.c | 132 ++++++++++++++++------------ src/pcm/pcm_hooks.c | 8 +- src/pcm/pcm_hw.c | 74 +++++++++------- src/pcm/pcm_simple.c | 22 ++--- src/pcm/pcm_softvol.c | 69 ++++++++------- 8 files changed, 393 insertions(+), 348 deletions(-) -- 2.7.4

2 35

[alsa-devel] [PATCH RESEND] ASoC: rt5514-spi: Convert to use devm_* API
by Axel Lin 14 Jul '16

14 Jul '16

Use devm_* API to simplify the code. This patch also fixes the return value in probe error paths. Signed-off-by: Axel Lin <axel.lin(a)ingics.com> --- sound/soc/codecs/rt5514-spi.c | 25 ++++++------------------- 1 file changed, 6 insertions(+), 19 deletions(-) diff --git a/sound/soc/codecs/rt5514-spi.c b/sound/soc/codecs/rt5514-spi.c index 8a9382e..e1b326d 100644 --- a/sound/soc/codecs/rt5514-spi.c +++ b/sound/soc/codecs/rt5514-spi.c @@ -409,33 +409,21 @@ static int rt5514_spi_probe(struct spi_device *spi) rt5514_spi = spi; - ret = snd_soc_register_platform(&spi->dev, &rt5514_spi_platform); + ret = devm_snd_soc_register_platform(&spi->dev, &rt5514_spi_platform); if (ret < 0) { dev_err(&spi->dev, "Failed to register platform.\n"); - goto err_plat; + return ret; } - ret = snd_soc_register_component(&spi->dev, &rt5514_spi_dai_component, - &rt5514_spi_dai, 1); + ret = devm_snd_soc_register_component(&spi->dev, + &rt5514_spi_dai_component, + &rt5514_spi_dai, 1); if (ret < 0) { dev_err(&spi->dev, "Failed to register component.\n"); - goto err_comp; + return ret; } return 0; -err_comp: - snd_soc_unregister_platform(&spi->dev); -err_plat: - - return 0; -} - -static int rt5514_spi_remove(struct spi_device *spi) -{ - snd_soc_unregister_component(&spi->dev); - snd_soc_unregister_platform(&spi->dev); - - return 0; } static const struct of_device_id rt5514_of_match[] = { @@ -450,7 +438,6 @@ static struct spi_driver rt5514_spi_driver = { .of_match_table = of_match_ptr(rt5514_of_match), }, .probe = rt5514_spi_probe, - .remove = rt5514_spi_remove, }; module_spi_driver(rt5514_spi_driver); -- 2.5.0

1 0

[alsa-devel] [PATCH] pcm: fix return value of snd_pcm_ioplug_sw_params()
by Takashi Sakamoto 13 Jul '16

13 Jul '16

In former commits for thread-safe PCM APIs, snd_pcm_ioplug_sw_params() got 0 as its return value, against the original implementation. This commit fixes it. Fixes: 54931e5a5455('pcm: Add thread-safety to PCM API') Signed-off-by: Takashi Sakamoto <o-takashi(a)sakamocchi.jp> --- src/pcm/pcm_ioplug.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/src/pcm/pcm_ioplug.c b/src/pcm/pcm_ioplug.c index 115d89b..1dc198e 100644 --- a/src/pcm/pcm_ioplug.c +++ b/src/pcm/pcm_ioplug.c @@ -434,14 +434,16 @@ static int snd_pcm_ioplug_hw_free(snd_pcm_t *pcm) static int snd_pcm_ioplug_sw_params(snd_pcm_t *pcm, snd_pcm_sw_params_t *params) { ioplug_priv_t *io = pcm->private_data; - int err = 0; + int err; - if (io->data->callback->sw_params) { - snd_pcm_unlock(pcm); /* to avoid deadlock */ - err = io->data->callback->sw_params(io->data, params); - snd_pcm_lock(pcm); - } - return 0; + if (!io->data->callback->sw_params) + return 0; + + snd_pcm_unlock(pcm); /* to avoid deadlock */ + err = io->data->callback->sw_params(io->data, params); + snd_pcm_lock(pcm); + + return err; } -- 2.7.4

2 1

[alsa-devel] [PATCH] pcm: fix missing return value of snd_pcm_status()
by Takashi Sakamoto 13 Jul '16

13 Jul '16

In former commits for thread-safe PCM APIs, snd_pcm_status() lost its return value. This commit regain it. Fixes: 54931e5a5455('pcm: Add thread-safety to PCM API') Signed-off-by: Takashi Sakamoto <o-takashi(a)sakamocchi.jp> --- src/pcm/pcm.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/pcm/pcm.c b/src/pcm/pcm.c index a66e8af..6c34719 100644 --- a/src/pcm/pcm.c +++ b/src/pcm/pcm.c @@ -953,6 +953,8 @@ int snd_pcm_status(snd_pcm_t *pcm, snd_pcm_status_t *status) snd_pcm_lock(pcm); err = pcm->fast_ops->status(pcm->fast_op_arg, status); snd_pcm_unlock(pcm); + + return err; } /** -- 2.7.4

2 1