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

Re: [alsa-devel] linux-next: Tree for Aug 16 (mellanox/mlx5 & sound/oss/)
by Randy Dunlap 16 Aug '16

16 Aug '16

On 08/15/16 21:01, Stephen Rothwell wrote: > Hi all, > > Changes since 20160815: > on i386: sound/built-in.o:(.data+0x5c): multiple definition of `dev_list' drivers/built-in.o:(.data+0x6aee8): first defined here ld: Warning: size of symbol `dev_list' changed from 8 in drivers/built-in.o to 16 in sound/built-in.o sound/ is an old oss driver (soundcard.c): struct oss_minor_dev { unsigned short minor; unsigned int enabled; } dev_list[] = { { SND_DEV_DSP16 }, { SND_DEV_AUDIO }, }; ===== drivers/ is net/ethernet/mellanox/mlx5/core/mlx5_core.h: extern struct list_head dev_list; -- ~Randy

1 0

[alsa-devel] [PATCH] ASoC: compress: Fix leak of a widget list in soc_compr_open_fe
by Charles Keepax 16 Aug '16

16 Aug '16

After we have called dpcm_path_get we should make sure to call dpcm_path_put on all error paths. This was not happening causing the allocated widget list to be leaked, this patch corrects this by adding a dpcm_path_put to the error path. Signed-off-by: Charles Keepax <ckeepax(a)opensource.wolfsonmicro.com> --- sound/soc/soc-compress.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/sound/soc/soc-compress.c b/sound/soc/soc-compress.c index d2df46c..bf7b52f 100644 --- a/sound/soc/soc-compress.c +++ b/sound/soc/soc-compress.c @@ -121,7 +121,7 @@ static int soc_compr_open_fe(struct snd_compr_stream *cstream) dpcm_be_disconnect(fe, stream); fe->dpcm[stream].runtime = NULL; - goto fe_err; + goto path_err; } dpcm_clear_pending_state(fe, stream); @@ -136,6 +136,8 @@ static int soc_compr_open_fe(struct snd_compr_stream *cstream) return 0; +path_err: + dpcm_path_put(&list); fe_err: if (fe->dai_link->compr_ops && fe->dai_link->compr_ops->shutdown) fe->dai_link->compr_ops->shutdown(cstream); -- 2.1.4

2 1

[alsa-devel] [PATCH v2] ASoC: atmel_ssc_dai: Don't unconditionally reset SSC on stream startup
by Peter Meerwald-Stadler 16 Aug '16

16 Aug '16

From: Christoph Huber <c.huber(a)bct-electronic.com> commit cbaadf0f90d6 ("ASoC: atmel_ssc_dai: refactor the startup and shutdown") refactored code such that the SSC is reset on every startup; this breaks duplex audio (e.g. first start audio playback, then start record, causing the playback to stop/hang) Signed-off-by: Christoph Huber <c.huber(a)bct-electronic.com> Signed-off-by: Peter Meerwald-Stadler <p.meerwald(a)bct-electronic.com> Cc: Songjun Wu <songjun.wu(a)microchip.com> Cc: Nicolas Ferre <nicolas.ferre(a)atmel.com> --- v2 fixes formatting issues and rephrases the comment sound/soc/atmel/atmel_ssc_dai.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/sound/soc/atmel/atmel_ssc_dai.c b/sound/soc/atmel/atmel_ssc_dai.c index 54c09ac..16e459a 100644 --- a/sound/soc/atmel/atmel_ssc_dai.c +++ b/sound/soc/atmel/atmel_ssc_dai.c @@ -299,8 +299,9 @@ static int atmel_ssc_startup(struct snd_pcm_substream *substream, clk_enable(ssc_p->ssc->clk); ssc_p->mck_rate = clk_get_rate(ssc_p->ssc->clk); - /* Reset the SSC to keep it at a clean status */ - ssc_writel(ssc_p->ssc->regs, CR, SSC_BIT(CR_SWRST)); + /* Reset the SSC unless initialized to keep it in a clean state */ + if (!ssc_p->initialized) + ssc_writel(ssc_p->ssc->regs, CR, SSC_BIT(CR_SWRST)); if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) { dir = 0; -- 1.7.10.4

2 1

[alsa-devel] [PATCH] ASoC: max98371: Delete unnecessary assignment for the field "owner"
by SF Markus Elfring 16 Aug '16

16 Aug '16

From: Markus Elfring <elfring(a)users.sourceforge.net> Date: Mon, 15 Aug 2016 16:16:00 +0200 The field "owner" is set by the core. Thus delete an unneeded initialisation. Generated by: scripts/coccinelle/api/platform_no_drv_owner.cocci Signed-off-by: Markus Elfring <elfring(a)users.sourceforge.net> --- sound/soc/codecs/max98371.c | 1 - 1 file changed, 1 deletion(-) diff --git a/sound/soc/codecs/max98371.c b/sound/soc/codecs/max98371.c index cf0a39b..3053879 100644 --- a/sound/soc/codecs/max98371.c +++ b/sound/soc/codecs/max98371.c @@ -425,7 +425,6 @@ MODULE_DEVICE_TABLE(of, max98371_of_match); static struct i2c_driver max98371_i2c_driver = { .driver = { .name = "max98371", - .owner = THIS_MODULE, .pm = NULL, .of_match_table = of_match_ptr(max98371_of_match), }, -- 2.9.2

2 1

[alsa-devel] [PATCH 1149/1285] Replace numeric parameter like 0444 with macro
by Baole Ni 16 Aug '16

16 Aug '16

I find that the developers often just specified the numeric value when calling a macro which is defined with a parameter for access permission. As we know, these numeric value for access permission have had the corresponding macro, and that using macro can improve the robustness and readability of the code, thus, I suggest replacing the numeric parameter with the macro. Signed-off-by: Chuansheng Liu <chuansheng.liu(a)intel.com> Signed-off-by: Baole Ni <baolex.ni(a)intel.com> --- sound/core/oss/pcm_oss.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c index ebc9fdf..0ff0dd8 100644 --- a/sound/core/oss/pcm_oss.c +++ b/sound/core/oss/pcm_oss.c @@ -53,11 +53,11 @@ static bool nonblock_open = 1; MODULE_AUTHOR("Jaroslav Kysela <perex(a)perex.cz>, Abramo Bagnara <abramo(a)alsa-project.org>"); MODULE_DESCRIPTION("PCM OSS emulation for ALSA."); MODULE_LICENSE("GPL"); -module_param_array(dsp_map, int, NULL, 0444); +module_param_array(dsp_map, int, NULL, S_IRUSR | S_IRGRP | S_IROTH); MODULE_PARM_DESC(dsp_map, "PCM device number assigned to 1st OSS device."); -module_param_array(adsp_map, int, NULL, 0444); +module_param_array(adsp_map, int, NULL, S_IRUSR | S_IRGRP | S_IROTH); MODULE_PARM_DESC(adsp_map, "PCM device number assigned to 2nd OSS device."); -module_param(nonblock_open, bool, 0644); +module_param(nonblock_open, bool, S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH); MODULE_PARM_DESC(nonblock_open, "Don't block opening busy PCM devices."); MODULE_ALIAS_SNDRV_MINOR(SNDRV_MINOR_OSS_PCM); MODULE_ALIAS_SNDRV_MINOR(SNDRV_MINOR_OSS_PCM1); -- 2.9.2

2 1

[alsa-devel] [PATCH v1 00/14] clk/mfd/ASoC/ARM: OMAP4/5: McPDM/twl6040 pdmclk support
by Peter Ujfalusi 15 Aug '16

15 Aug '16

Hi, Changes since v1: - Added Rob's Acked-by to patch 5 and 14 - Check the return value of clk_prepare_enable() calls in the last patch this is something I wanted to do for a long time... First of all: sorry for the cross domain series. I did tested alone all the domain patches and they are not causing any regression. When they come together we will have ability to control the pdmclk needed by McPDM. CLK patches: The driver was w/o DT support and needed some cleanup so we can use the driver. I have also renamed it to twl6040-pdmclk from twl6040-clk to be precise. MFD patches: The regmap_sync() was not working since the twl6040 does not support bulk access and now we are going to create the needed child for the pdmclk. ARM patches: Updatding the board DTS files and add the needed lines for the pdmclk. ASoC patches: The machine driver needs to select the twl6040-clk driver as it will be used by the omap-mcpdm. I needed to do some cleanup and add also suspend/resume support before adding the code for pdmclk handling since I don't want to keep the twl6040 powered on when the board is suspended. At the moment it is not possible to do true dynamic twl6040 power up/down due to pop noises, but I will be looking at that later. The driver in the future will fail if it is not able to get the pdmclk to avoid kernel crash (McPDM registers are not accessible when pdmclk is not available). I have tested the patches by domain and also in all (I hope) permutation they could be. I have not experienced any regression. Regards, Peter --- Peter Ujfalusi (14): clk: twl6040: Correct clk_ops clk: twl6040: Register the clock as of_clk_provider clk: twl6040: Rename the driver and use consistent names in the code mfd: twl6040: The chip does not support bulk access mfd: twl6040: Register child device for twl6040-pdmclk ARM: dts: omap5-board-common: Add pdmclk binding for audio ARM: dts: omap4-panda-common: Add pdmclk binding for audio ARM: dts: omap4-sdp: Add pdmclk binding for audio ARM: dts: omap4-var-som-om44: Add pdmclk binding for audio ARM: dts: omap4-duovero: Add pdmclk binding for audio ASoC: omap: Kconfig: SND_OMAP_SOC_OMAP_ABE_TWL6040 to select CLK_TWL6040 ASoC: omap-mcpdm: Move the WD enable write inside omap_mcpdm_open_streams() ASoC: omap-mcpdm: Support for suspend resume ASoC: omap-mcpdm: Add support for pdmclk clock handling Documentation/devicetree/bindings/mfd/twl6040.txt | 1 + .../devicetree/bindings/sound/omap-mcpdm.txt | 10 +++ arch/arm/boot/dts/omap4-duovero.dtsi | 5 ++ arch/arm/boot/dts/omap4-panda-common.dtsi | 5 ++ arch/arm/boot/dts/omap4-sdp.dts | 5 ++ arch/arm/boot/dts/omap4-var-som-om44.dtsi | 5 ++ arch/arm/boot/dts/omap5-board-common.dtsi | 5 ++ drivers/clk/clk-twl6040.c | 80 ++++++++++++--------- drivers/mfd/twl6040.c | 6 ++ include/linux/mfd/twl6040.h | 2 +- sound/soc/omap/Kconfig | 1 + sound/soc/omap/omap-mcpdm.c | 84 ++++++++++++++++++++-- 12 files changed, 170 insertions(+), 39 deletions(-) -- 2.8.3

4 19

[alsa-devel] [PATCH v4 0/2] amidi delay option
by Felipe F. Tonello 15 Aug '16

15 Aug '16

Small improvement to amidi, adding useful functionality. changes from v3: * fixed a bug if memchr() returns NULL changes from v2: * only check of end of sysex (0xf7) * use memchr() instead of direct loop approach * fixed commit message changes from v1: * remove delay in between any MIDI message, only SysEx * added patch to use getopt_long() arguments Felipe F. Tonello (2): amidi: add delay option amidi: use GNU getopt_long() proper argument names amidi/amidi.1 | 14 +++++++++++++ amidi/amidi.c | 63 ++++++++++++++++++++++++++++++++++++++++++++--------------- 2 files changed, 61 insertions(+), 16 deletions(-) -- 2.9.3

2 3

[alsa-devel] [PATCH v3 0/2] amidi delay option
by Felipe F. Tonello 15 Aug '16

15 Aug '16

Small improvement to amidi, adding useful functionality. change from v2: * only check of end of sysex (0xf7) * use memchr() instead of direct loop approach * fixed commit message changes from v1: * remove delay in between any MIDI message, only SysEx * added patch to use getopt_long() arguments Felipe F. Tonello (2): amidi: add delay option amidi: use GNU getopt_long() proper argument names amidi/amidi.1 | 14 ++++++++++++++ amidi/amidi.c | 60 +++++++++++++++++++++++++++++++++++++++++++---------------- 2 files changed, 58 insertions(+), 16 deletions(-) -- 2.9.3

2 3

[alsa-devel] [PATCH v2 0/2] amidi delay option
by Felipe F. Tonello 15 Aug '16

15 Aug '16

Small improvement to amidi, adding useful functionality. changes from v1: * remove delay in between any MIDI message, only SysEx * added patch to use getopt_long() arguments Felipe F. Tonello (2): amidi: add delay option amidi: use GNU getopt_long() proper argument names amidi/amidi.1 | 14 +++++++++++++ amidi/amidi.c | 66 ++++++++++++++++++++++++++++++++++++++++++++--------------- 2 files changed, 64 insertions(+), 16 deletions(-) -- 2.9.2

3 4

[alsa-devel] [PATCH v8 0/5] ASoC: MT8173 HDMI jack detection
by Philipp Zabel 15 Aug '16

15 Aug '16

Hi, now that kernel v4.8-rc1 is released and includes the new CEC framework in drivers/staging/media/cec, I wonder how to proceed with this. To allow the ASoC drivers to present HDMI cable connection state as an ALSA jack, and to allow them to update sink capabilities when it signals an EDID change, we need a notification mechanism between CEC/HDMI, and ALSA drivers. I had used Russell's hdmi notification prototype to let the CEC/HDMI drivers notify the hdmi-codec and ASoC machine drivers of changes. The mtk_cec driver currently is just a placeholder that only handles the HPD interrupts. Should the notifications be integrated with the CEC framework for those devices that do have CEC support? Changes since v7 ("ASoC: MT8173 HDMI codec support"): - Dropped the already applied ELD control and machine driver patches - Dropped the N/CTS helper patch for now, as the helpers don't exist yet. - Lock ELD mutex while copying ELD in hdmi_eld_ctl_get regards Philipp Philipp Zabel (5): video: rmk's HDMI notification prototype ASoC: hdmi-codec: Use HDMI notifications to add jack support ASoC: mediatek: Add jack detection support to mt8173-rt5650-rt5676 machine driver ASoC: mediatek: Add jack detection support to the mt8173-rt5650 machine driver drm/mediatek: hdmi: issue notifications drivers/gpu/drm/mediatek/mtk_cec.c | 11 +++ drivers/gpu/drm/mediatek/mtk_hdmi.c | 3 + drivers/video/Kconfig | 3 + drivers/video/Makefile | 1 + drivers/video/hdmi-notifier.c | 61 ++++++++++++++++ include/linux/hdmi-notifier.h | 44 ++++++++++++ include/sound/hdmi-codec.h | 6 ++ sound/soc/codecs/Kconfig | 1 + sound/soc/codecs/hdmi-codec.c | 88 +++++++++++++++++++++--- sound/soc/mediatek/mt8173/mt8173-rt5650-rt5676.c | 21 ++++++ sound/soc/mediatek/mt8173/mt8173-rt5650.c | 21 ++++++ 11 files changed, 252 insertions(+), 8 deletions(-) create mode 100644 drivers/video/hdmi-notifier.c create mode 100644 include/linux/hdmi-notifier.h -- 2.8.1

3 10