September 2015 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH 0/9] ASoC: rsnd: Gen3 support
by Kuninori Morimoto 10 Sep '15

10 Sep '15

Hi Mark These are Gen3 support patches for Renesas sound IP. These are based on my previous cleanup patches [PATCH 0/4] ASoC: rsnd: cleanup for Gen3 support Date: Thu, 10 Sep 2015 15:44:13 +0900 Kuninori Morimoto (9): ASoC: rsnd: Add Gen3 initial support ASoC: rsnd: add rsnd_mod_get() macro and use it ASoC: rsnd: ADG uses mod base common method ASoC: rsnd: add common mod confirm method ASoC: rsnd: ADG initial setup on rsnd_adg_ssi_clk_init() ASoC: rsnd: tidyup ADG clock calculate method ASoC: rsnd: tidyup ADG debug message for clock selection ASoC: rsnd: adg: ignore undefined clock error ASoC: rsnd: add AUDIO_CLKOUT support Documentation/devicetree/bindings/sound/renesas,rsnd.txt | 5 ++ sound/soc/sh/rcar/adg.c | 302 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++----------------------- sound/soc/sh/rcar/core.c | 12 +++++ sound/soc/sh/rcar/ctu.c | 6 +-- sound/soc/sh/rcar/dvc.c | 6 +-- sound/soc/sh/rcar/mix.c | 6 +-- sound/soc/sh/rcar/rsnd.h | 12 +++++ sound/soc/sh/rcar/src.c | 6 +-- sound/soc/sh/rcar/ssi.c | 98 ++++++++++++++++++---------------- 9 files changed, 326 insertions(+), 127 deletions(-)

1 9

[alsa-devel] [PATCH 0/4] ASoC: rsnd: cleanup for Gen3 support
by Kuninori Morimoto 10 Sep '15

10 Sep '15

Hi Mark These are not so important, but cleanup for preparing of Gen3 support Kuninori Morimoto (4): ASoC: rsnd: tidyup rsnd_dma_to_xxx() position ASoC: rsnd: rename rsnd_src_pcm_new() to rsnd_src_pcm_new_gen2() ASoC: rsnd: remove unneeded sh_clk header ASoC: rsnd: add missing #sound-dai-cells explain on Document Documentation/devicetree/bindings/sound/renesas,rsnd.txt | 2 ++ include/sound/rcar_snd.h | 1 - sound/soc/sh/rcar/adg.c | 1 - sound/soc/sh/rcar/rsnd.h | 3 +-- sound/soc/sh/rcar/src.c | 11 ++--------- 5 files changed, 5 insertions(+), 13 deletions(-)

1 4

[alsa-devel] [PATCH 0/2] ASoC: ak4642: MCKO support
by Kuninori Morimoto 10 Sep '15

10 Sep '15

Hi Mark These patches adds MCKO support for ak4642. Kuninori Morimoto (2): ASoC: ak4642: use *dev on ak4642_i2c_probe() ASoC: ak4642: enable to use MCKO as fixed rate output pin on DT Documentation/devicetree/bindings/sound/ak4642.txt | 22 +++++++++++++++- sound/soc/codecs/ak4642.c | 152 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-------------------------------------- 2 files changed, 120 insertions(+), 54 deletions(-) Best regards --- Kuninori Morimoto

1 2

[alsa-devel] [PATCH 1/1] ASoC: sti: fix compilation warnings
by Arnaud Pouliquen 09 Sep '15

09 Sep '15

Add check on of_property_read_string return to avoid compilation warning. Signed-off-by: Arnaud Pouliquen <arnaud.pouliquen(a)st.com> --- sound/soc/sti/uniperif_player.c | 32 +++++++++++++++++++------------- sound/soc/sti/uniperif_reader.c | 6 +++++- 2 files changed, 24 insertions(+), 14 deletions(-) diff --git a/sound/soc/sti/uniperif_player.c b/sound/soc/sti/uniperif_player.c index f6eefe1..d90a3fb 100644 --- a/sound/soc/sti/uniperif_player.c +++ b/sound/soc/sti/uniperif_player.c @@ -989,8 +989,8 @@ static int uni_player_parse_dt(struct platform_device *pdev, if (!info) return -ENOMEM; - of_property_read_u32(pnode, "version", &player->ver); - if (player->ver == SND_ST_UNIPERIF_VERSION_UNKNOWN) { + if ((of_property_read_u32(pnode, "version", &player->ver)) || + (player->ver == SND_ST_UNIPERIF_VERSION_UNKNOWN)) { dev_err(dev, "Unknown uniperipheral version "); return -EINVAL; } @@ -998,19 +998,25 @@ static int uni_player_parse_dt(struct platform_device *pdev, if (player->ver >= SND_ST_UNIPERIF_VERSION_UNI_PLR_TOP_1_0) info->underflow_enabled = 1; - of_property_read_u32(pnode, "uniperiph-id", &info->id); + if (of_property_read_u32(pnode, "uniperiph-id", &info->id)) { + dev_err(dev, "uniperipheral id not defined"); + return -EINVAL; + } /* Read the device mode property */ - of_property_read_string(pnode, "mode", &mode); - - if (strcasecmp(mode, "hdmi") == 0) - info->player_type = SND_ST_UNIPERIF_PLAYER_TYPE_HDMI; - else if (strcasecmp(mode, "pcm") == 0) - info->player_type = SND_ST_UNIPERIF_PLAYER_TYPE_PCM; - else if (strcasecmp(mode, "spdif") == 0) - info->player_type = SND_ST_UNIPERIF_PLAYER_TYPE_SPDIF; - else - info->player_type = SND_ST_UNIPERIF_PLAYER_TYPE_NONE; + if (!of_property_read_string(pnode, "mode", &mode)) { + if (strcasecmp(mode, "hdmi") == 0) + info->player_type = SND_ST_UNIPERIF_PLAYER_TYPE_HDMI; + else if (strcasecmp(mode, "pcm") == 0) + info->player_type = SND_ST_UNIPERIF_PLAYER_TYPE_PCM; + else if (strcasecmp(mode, "spdif") == 0) + info->player_type = SND_ST_UNIPERIF_PLAYER_TYPE_SPDIF; + else + info->player_type = SND_ST_UNIPERIF_PLAYER_TYPE_NONE; + } else { + dev_err(dev, "uniperipheral mode not defined"); + return -EINVAL; + } /* Save the info structure */ player->info = info; diff --git a/sound/soc/sti/uniperif_reader.c b/sound/soc/sti/uniperif_reader.c index c502626..7ddbd14 100644 --- a/sound/soc/sti/uniperif_reader.c +++ b/sound/soc/sti/uniperif_reader.c @@ -316,7 +316,11 @@ static int uni_reader_parse_dt(struct platform_device *pdev, if (!info) return -ENOMEM; - of_property_read_u32(node, "version", &reader->ver); + if ((of_property_read_u32(node, "version", &reader->ver)) || + (reader->ver == SND_ST_UNIPERIF_VERSION_UNKNOWN)) { + dev_err(&pdev->dev, "Unknown uniperipheral version "); + return -EINVAL; + } /* Save the info structure */ reader->info = info; -- 1.9.1

2 1

[alsa-devel] [PATCH lib 0/5] alsa-lib topology trivial fixes
by Takashi Iwai 09 Sep '15

09 Sep '15

Hi, here is a series of trivial fixes that have been spotted by gcc warnings. Takashi Takashi Iwai (5): topology: builder: Fix possibly uninitialized variable in write_elem_block() topology: ctl: Fix access type checks topology: data: Fix wrong size check in tplg_parse_data_hex() topology: parser: Add missing return value to snd_tplg_set_manifest_data() topology: pcm: Remove unused variables src/topology/builder.c | 9 ++++----- src/topology/ctl.c | 2 +- src/topology/data.c | 4 ++-- src/topology/parser.c | 1 + src/topology/pcm.c | 3 +-- 5 files changed, 9 insertions(+), 10 deletions(-) -- 2.5.1

2 9

[alsa-devel] [PATCH v3 0/9] asoc: Add audio for sti platforms
by Arnaud Pouliquen 09 Sep '15

09 Sep '15

version 3 : - patch dependency: ASoC: simple card: set cpu-dai sysclk with mclk-fs (Arnaud Pouliquen) - Remark not applied: Use Russell King's patch "sound/core: add IEC958 channel status helper". => This allows to create settings based on runtime, but not compatible with iec958 controls exposed to userland (settings overwritten by snd_pcm_create_iec958_consumer) - main corrections: - rework DAI CPU declaration to use generic dma - Supress abstraction layer for uniperiph player/reader - use dapm for DAC powerup/down version 2: corrections after first review - bindings: . suppress useless dts declaration - uniperif_player.c: . Suppress standby mode in uniplayer ( will be part of a dedicated patch) . Deliver IEC and clock adjustement controls in separate patch - uniperif_reader.c: . kept as separate file as behavior and feature not same than uniplayer . minor corrections - codec/sti-sas.c . implement DAPM for DAC . rework registers declaration to separate device and driver part version 1: - First patches submission Arnaud Pouliquen (9): ASoC: sti: add binding for ASoC driver ASoC: sti: Add uniperipheral header file ASoC: sti: Add CPU DAI driver for playback ASoC: sti: Add CPU DAI driver for capture ASoC: sti: Add uniperipheral dai driver ASoC: Add ability to build sti drivers ASoC: Codec: Add sti platform codec ASoC: sti: Add clock adjustement control ASoC: sti: Add IEC control .../devicetree/bindings/sound/st,sti-asoc-card.txt | 155 +++ sound/soc/Kconfig | 1 + sound/soc/Makefile | 1 + sound/soc/codecs/Kconfig | 5 + sound/soc/codecs/Makefile | 2 + sound/soc/codecs/sti-sas.c | 627 ++++++++++ sound/soc/sti/Kconfig | 11 + sound/soc/sti/Makefile | 4 + sound/soc/sti/sti_uniperif.c | 254 ++++ sound/soc/sti/uniperif.h | 1229 ++++++++++++++++++++ sound/soc/sti/uniperif_player.c | 1109 ++++++++++++++++++ sound/soc/sti/uniperif_reader.c | 364 ++++++ 12 files changed, 3762 insertions(+) create mode 100644 Documentation/devicetree/bindings/sound/st,sti-asoc-card.txt create mode 100644 sound/soc/codecs/sti-sas.c create mode 100644 sound/soc/sti/Kconfig create mode 100644 sound/soc/sti/Makefile create mode 100644 sound/soc/sti/sti_uniperif.c create mode 100644 sound/soc/sti/uniperif.h create mode 100644 sound/soc/sti/uniperif_player.c create mode 100644 sound/soc/sti/uniperif_reader.c -- 1.9.1

4 27

[alsa-devel] [PATCH] [alsa-utils.git] Fix a strange regression in aplay's '-d' option
by Grond 09 Sep '15

09 Sep '15

Hello. (J. Random Hacker here) While using aplay/arecord (version 1.0.28) to generate some noise, I ran across a regression in the use of the '-d' option. Specifically: According to the man page: -d, --duration=# Interrupt after # seconds. A value of zero means infinity. The default is zero, so if this option is omitted then the arecord process will run until it is killed. [snip] arecord -d 10 -f cd -t wav -D copy foobar.wav will record foobar.wav as a 10-second, CD-quality wave file, using the PCM "copy" (which might be defined in the user's .asoundrc file as: pcm.copy { type plug slave { pcm hw } route_policy copy } [snip] So the behavior I expect to see (and have experienced in the past) is: $ arecord -d 3 -f cd dump.wav Recording WAVE 'dump.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo $ ls -l -rw------- 1 grond66 grond66 529244 Sep 5 00:34 dump.wav What actually happens is: $ arecord -d 3 -f cd dump.wav Recording WAVE 'dump.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo [ program hangs here until SIGINT ] ^CAborted by signal Interrupt... arecord: pcm_read:2031: read error: Interrupted system call $ ls -l -rw------- 1 grond66 grond66 526444 Sep 5 00:39 dump-01.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-02.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-03.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-04.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-05.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-06.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-07.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-08.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-09.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-100.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-101.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-102.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-103.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-104.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-105.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-106.wav -rw------- 1 grond66 grond66 44 Sep 5 00:39 dump-107.wav [ etcetera... ] $ aplay dump-01.wav [ three seconds of hiss, as expected ] $ less dump-02.wav RIFF$^@^@^@WAVEfmt ^P^@^@^@^A^@^B^@D<AC>^@^@^P<B1>^B^@^D^@^P^@data^@^@^@^@ $ So arecord is recording the requested audio and putting it into the first file, then spamming WAV headers into all the subsequent files. This is not terribly useful. The problem appears to be six lines of code in aplay/aplay.c, in the functions pcm_read() and pcm_readv(): 2006 static ssize_t pcm_read(u_char *data, size_t rcount) 2007 { 2008 ssize_t r; 2009 size_t result = 0; 2010 size_t count = rcount; 2011 )> 2012 if (count != chunk_size) { )> 2013 count = chunk_size; )> 2014 } 2015 2016 while (count > 0 && !in_aborting) { 2045 static ssize_t pcm_readv(u_char **data, unsigned int channels, size_t rcount) 2046 { 2047 ssize_t r; 2048 size_t result = 0; 2049 size_t count = rcount; 2050 )> 2051 if (count != chunk_size) { )> 2052 count = chunk_size; )> 2053 } 2054 2055 while (count > 0 && !in_aborting) { In this case, the 'rcount' arguments are the number of PCM integers to be read from the capture device. In theory. Because what lines 2012-2014 and 2051-2053 do is effectively: count = chunk_size; In other words, the argument is completely disregarded, and 'chunk_size' is used no matter what the functions are passed. But capture() and family use the following logic: "If pcm_read{,v}() returns with a value other than the number integers we told it to read off the wire, assume something went wrong": 2925 static void capture(char *orig_name) 2926 { 2927 int tostdout=0; /* boolean which describes output stream */ 2928 int filecount=0; /* number of files written */ 2929 char *name = orig_name; /* current filename */ 2930 char namebuf[PATH_MAX+1]; 2931 off64_t count, rest; /* number of bytes to capture */ [...] 2964 do { 2965 /* open a file to write */ 2966 if(!tostdout) { 2967 /* upon the second file we start the numbering scheme */ 2968 if (filecount || use_strftime) { 2969 filecount = new_capture_file(orig_name, namebuf, 2970 sizeof(namebuf), 2971 filecount); 2972 name = namebuf; 2973 } 2974 2975 /* open a new file */ 2976 remove(name); 2977 fd = safe_open(name); 2978 f (fd < 0) { 2979 perror(name); 2980 prg_exit(EXIT_FAILURE); 2981 } 2982 filecount++; 2983 } 2984 2985 rest = count; 2986 if (rest > fmt_rec_table[file_type].max_filesize) 2987 rest = fmt_rec_table[file_type].max_filesize; 2988 if (max_file_size && (rest > max_file_size)) 2989 rest = max_file_size; 2990 2991 /* setup sample header */ 2992 if (fmt_rec_table[file_type].start) 2993 fmt_rec_table[file_type].start(fd, rest); 2994 2995 /* capture */ 2996 fdcount = 0; 2997 while (rest > 0 && recycle_capture_file == 0 && !in_aborting) { )> 2998 size_t c = (rest <= (off64_t)chunk_bytes) ? )> 2999 (size_t)rest : chunk_bytes; )> 3000 size_t f = c * 8 / bits_per_frame; )> 3001 if (pcm_read(audiobuf, f) != f) )> 3002 break; 3003 if (write(fd, audiobuf, c) != c) { 3004 perror(name); 3005 prg_exit(EXIT_FAILURE); 3006 } 3007 count -= c; 3008 rest -= c; 3009 fdcount += c; 3010 } 3011 [...] 3027 /* repeat the loop when format is raw without timelimit or 3028 * requested counts of data are recorded 3029 */ 3030 } while ((file_type == FORMAT_RAW && !timelimit) || count > 0); 3031 } These two modes of operation are incompatible. However, if we nix the strange requirement that all calls to pcm_read{,v}() must be for exactly 'chunk_size' samples, the problem goes away. This is the purpose of the patch I've included below.

2 2

[alsa-devel] Getting notification of sound card load
by Mina Almasry 09 Sep '15

09 Sep '15

Hi, I'd like to get a notification when a new sound card becomes available in my system. What's the best way to do that? For example, when I 'cat /proc/asound/cards' I usually see one sound card available. But if I plug in USB speakers to my system and then run 'cat /proc/asound/cards', I see that the USB sound card is ready for the device I just plugged in. What's the best way via C API to get a notification that a new card has been loaded? Is there a better way that looking for changes on /proc/asound/cards? Thanks, Mina

2 2

[alsa-devel] [PATCH] ASoC: dapm: add kcontrol to switch regulator to regulated/bypass state
by Nikesh Oswal 08 Sep '15

08 Sep '15

When regulator is defined with SND_SOC_DAPM_REGULATOR_CONTROL_BYPASS flag, then a kcontrol will be created which can be used to switch regulator to regulated/bypass state. This will help to control the behaviour of the regulator based on a usecase. For example voice call may need a regulated voltage to acheive higher quality whereas voice trigger may need bypass voltage so as to save on power. Signed-off-by: Nikesh Oswal <nikesh(a)opensource.wolfsonmicro.com> --- include/sound/soc-dapm.h | 7 +- sound/soc/soc-dapm.c | 177 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 183 insertions(+), 1 deletion(-) diff --git a/include/sound/soc-dapm.h b/include/sound/soc-dapm.h index 15717b4..778b847 100644 --- a/include/sound/soc-dapm.h +++ b/include/sound/soc-dapm.h @@ -343,7 +343,12 @@ struct device; (e & (SND_SOC_DAPM_PRE_PMD | SND_SOC_DAPM_POST_PMD)) /* regulator widget flags */ -#define SND_SOC_DAPM_REGULATOR_BYPASS 0x1 /* bypass when disabled */ +/* bypass when disabled and regulated when enabled */ +#define SND_SOC_DAPM_REGULATOR_BYPASS 0x1 +/* bypass when disabled and regulated when enable by default and a + kcontrol is created to explicitly switch between bypass/regulated */ +#define SND_SOC_DAPM_REGULATOR_CONTROL_BYPASS \ + (SND_SOC_DAPM_REGULATOR_BYPASS | 0x2) struct snd_soc_dapm_widget; enum snd_soc_dapm_type; diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index 36ab9cb..2d77eb9 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -965,6 +965,180 @@ static int dapm_new_mux(struct snd_soc_dapm_widget *w) return 0; } +static int snd_soc_dapm_regulator_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct snd_soc_dapm_widget_list *wlist = + dapm_kcontrol_get_wlist(kcontrol); + struct snd_soc_dapm_widget *widget = wlist->widgets[0]; + struct snd_soc_card *card = widget->dapm->card; + struct soc_enum *e = (struct soc_enum *)kcontrol->private_value; + unsigned int *item = ucontrol->value.enumerated.item; + unsigned int new_val, val; + int ret; + bool bypass; + + if (item[0] >= e->items) + return -EINVAL; + + val = dapm_kcontrol_get_value(kcontrol); + new_val = item[0] == 0 ? SND_SOC_DAPM_REGULATOR_BYPASS : 0; + bypass = new_val == SND_SOC_DAPM_REGULATOR_BYPASS ? false : true; + + if (new_val != val) { + mutex_lock_nested(&card->dapm_mutex, + SND_SOC_DAPM_CLASS_RUNTIME); + if (regulator_is_enabled(widget->regulator)) { + ret = regulator_allow_bypass(widget->regulator, bypass); + if (ret != 0) + dev_warn(widget->dapm->dev, + "ASoC: Failed to change bypass %s: %d\n", + widget->name, ret); + } + dapm_kcontrol_set_value(kcontrol, new_val); + widget->on_val = new_val; + mutex_unlock(&card->dapm_mutex); + } + + return 0; +} + +static int snd_soc_dapm_regulator_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + unsigned int val; + + val = dapm_kcontrol_get_value(kcontrol); + + if (val == SND_SOC_DAPM_REGULATOR_BYPASS) + ucontrol->value.enumerated.item[0] = 0; + else + ucontrol->value.enumerated.item[0] = 1; + + return 0; +} + +static const char * const dapm_regulator_texts[] = { + "Regulated", + "Bypass", +}; + +/* create new dapm regulator control */ +static int dapm_new_regulator(struct snd_soc_dapm_widget *w) +{ + int ret = 0; + struct snd_soc_card *card = w->dapm->card; + unsigned long private_value; + struct snd_kcontrol *kcontrol; + struct snd_soc_dapm_path *path; + struct soc_enum regulator_enum[] = { + SOC_ENUM_SINGLE_VIRT(ARRAY_SIZE(dapm_regulator_texts), + dapm_regulator_texts), + }; + struct snd_kcontrol_new kcontrol_regulator[] = { + SOC_ENUM_EXT(NULL, regulator_enum[0], + snd_soc_dapm_regulator_get, + snd_soc_dapm_regulator_put), + }; + + + /* kcontrol creation is done only if client requests it */ + if (w->on_val != SND_SOC_DAPM_REGULATOR_CONTROL_BYPASS) + return 0; + + + /* create a kcontrol only if somebody is sourcing + from this regulator widget */ + if (list_empty(&w->edges[SND_SOC_DAPM_DIR_IN])) { + dev_err(w->dapm->dev, "ASoC: %s has no sinks\n", w->name); + return -EINVAL; + } + + w->num_kcontrols = 1; + + private_value = (unsigned long) devm_kmemdup(card->dev, + (void *)(kcontrol_regulator[0].private_value), + sizeof(struct soc_enum), GFP_KERNEL); + if (!private_value) { + dev_err(card->dev, "ASoC: Failed to create control for %s widget\n", + w->name); + ret = -ENOMEM; + goto err_out; + } + + kcontrol_regulator[0].private_value = private_value; + + w->kcontrol_news = devm_kmemdup(card->dev, &kcontrol_regulator[0], + sizeof(struct snd_kcontrol_new), GFP_KERNEL); + if (!(w->kcontrol_news)) { + dev_err(card->dev, "ASoC: Failed to create control for %s widget\n", + w->name); + ret = -ENOMEM; + goto err_private; + } + + + kcontrol = snd_soc_cnew(&w->kcontrol_news[0], NULL, + w->name, NULL); + + if (!kcontrol) { + ret = -ENOMEM; + goto err_kcontrol_news; + } + + kcontrol->private_free = dapm_kcontrol_free; + + ret = dapm_kcontrol_data_alloc(w, kcontrol); + if (ret) + goto err_kcontrol; + + + ret = snd_ctl_add(card->snd_card, kcontrol); + if (ret < 0) { + dev_err(w->dapm->dev, + "ASoC: failed to add widget %s dapm kcontrol %s: %d\n", + w->name, w->kcontrol_news[0].name, ret); + goto err_kcontrol; + } + + ret = dapm_kcontrol_add_widget(kcontrol, w); + if (ret) + goto err_kcontrol; + + + /* change on_val to remove the kcontrol creation bit + as kcontrol is already created */ + w->on_val = SND_SOC_DAPM_REGULATOR_BYPASS; + /* update the kcontrol value to reflect the initial value */ + dapm_kcontrol_set_value(kcontrol, w->on_val); + + w->kcontrols = kzalloc(w->num_kcontrols * + sizeof(struct snd_kcontrol *), + GFP_KERNEL); + if (!w->kcontrols) { + ret = -ENOMEM; + goto err_kcontrol; + } + + w->kcontrols[0] = kcontrol; + + snd_soc_dapm_widget_for_each_path(w, SND_SOC_DAPM_DIR_IN, path) { + if (path->name) + dapm_kcontrol_add_path(w->kcontrols[0], path); + } + + return 0; + +err_kcontrol: + snd_ctl_free_one(kcontrol); +err_kcontrol_news: + devm_kfree(card->dev, (void *)w->kcontrol_news); +err_private: + devm_kfree(card->dev, (void *)private_value); +err_out: + return ret; +} + /* create new dapm volume control */ static int dapm_new_pga(struct snd_soc_dapm_widget *w) { @@ -2922,6 +3096,9 @@ int snd_soc_dapm_new_widgets(struct snd_soc_card *card) case snd_soc_dapm_dai_link: dapm_new_dai_link(w); break; + case snd_soc_dapm_regulator_supply: + dapm_new_regulator(w); + break; default: break; } -- 1.7.9.5

2 1

[alsa-devel] [PATCH] ASoC: make dapm cache search depth configurable
by Nikesh Oswal 08 Sep '15

08 Sep '15

cache search depth will have a default value of 2 and can be modified by the respective component probe function Signed-off-by: Nikesh Oswal <nikesh(a)opensource.wolfsonmicro.com> --- include/sound/soc-dapm.h | 1 + sound/soc/soc-core.c | 2 ++ sound/soc/soc-dapm.c | 5 +++-- 3 files changed, 6 insertions(+), 2 deletions(-) diff --git a/include/sound/soc-dapm.h b/include/sound/soc-dapm.h index 5abba03..15717b4 100644 --- a/include/sound/soc-dapm.h +++ b/include/sound/soc-dapm.h @@ -629,6 +629,7 @@ struct snd_soc_dapm_context { struct snd_soc_dapm_wcache path_sink_cache; struct snd_soc_dapm_wcache path_source_cache; + unsigned int cache_search_depth; #ifdef CONFIG_DEBUG_FS struct dentry *debugfs_dapm; diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index 6173d15..1802883 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -1595,6 +1595,7 @@ static int snd_soc_instantiate_card(struct snd_soc_card *card) card->dapm.bias_level = SND_SOC_BIAS_OFF; card->dapm.dev = card->dev; card->dapm.card = card; + card->dapm.cache_search_depth = 2; list_add(&card->dapm.list, &card->dapm_list); #ifdef CONFIG_DEBUG_FS @@ -2665,6 +2666,7 @@ static int snd_soc_component_initialize(struct snd_soc_component *component, component->remove = component->driver->remove; dapm = &component->dapm; + dapm->cache_search_depth = 2; dapm->dev = dev; dapm->component = component; dapm->bias_level = SND_SOC_BIAS_OFF; diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index f4bf21a..36ab9cb 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -580,11 +580,12 @@ dapm_wcache_lookup(struct snd_soc_dapm_wcache *wcache, const char *name) { struct snd_soc_dapm_widget *w = wcache->widget; struct list_head *wlist; - const int depth = 2; + int depth; int i = 0; - if (w) { + if (w && w->dapm->cache_search_depth) { wlist = &w->dapm->card->widgets; + depth = w->dapm->cache_search_depth; list_for_each_entry_from(w, wlist, list) { if (!strcmp(name, w->name)) -- 1.7.9.5

2 1