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

[alsa-devel] [PATCH 0/4] topology: Add support for BE DAIs
by mengdong.lin＠linux.intel.com 22 Jul '16

22 Jul '16

From: Mengdong Lin <mengdong.lin(a)linux.intel.com> This series is to work with the kernel series 'ASoC: topology: Add support for BE DAIs' This series adds support for BE (Back End) DAIs in user space: - Define the type and ABI struct for Backend DAIs in UAPI header file. - User can define BE DAIs either by the text conf file or by C API. There is topology ABI udpate in this series. Topology core can check size of each ABI objects to detect version mismatch between user space and kernel. There will be 2 pairs of kernel & user space series for PCM & BE links later. Guneshwor Singh (3): topology: Parse BE DAIs in text conf file topology: Support adding BE DAIs by C API topology: Export BE DAIs to the binary for kernel Mengdong Lin (1): topology: ABI - Add the types for BE DAI include/sound/asoc.h | 34 ++++++- include/topology.h | 44 +++++++++ src/topology/builder.c | 11 +++ src/topology/data.c | 4 + src/topology/elem.c | 4 + src/topology/parser.c | 16 ++++ src/topology/pcm.c | 237 ++++++++++++++++++++++++++++++++++++++++++++++ src/topology/tplg_local.h | 7 ++ 8 files changed, 355 insertions(+), 2 deletions(-) -- 2.5.0

1 4

[alsa-devel] [PATCH] topology: Fix the missing referenced elem ptr when merging private data
by mengdong.lin＠linux.intel.com 22 Jul '16

22 Jul '16

From: Mengdong Lin <mengdong.lin(a)linux.intel.com> tplg_copy_data() should set the valid referenced data element pointer. The caller will double check this pointer for all kinds of references, including TLV, controls, text and data. Also fix an inaccurate dmesg. Signed-off-by: Mengdong Lin <mengdong.lin(a)linux.intel.com> diff --git a/src/topology/dapm.c b/src/topology/dapm.c index 4d343b2..e3c90d8 100644 --- a/src/topology/dapm.c +++ b/src/topology/dapm.c @@ -201,7 +201,7 @@ static int tplg_build_widget(snd_tplg_t *tplg, } if (!ref->elem) { - SNDERR("error: cannot find control '%s'" + SNDERR("error: cannot find '%s'" " referenced by widget '%s'\n", ref->id, elem->id); return -EINVAL; diff --git a/src/topology/data.c b/src/topology/data.c index 768fc27..397f6a5 100644 --- a/src/topology/data.c +++ b/src/topology/data.c @@ -1050,6 +1050,7 @@ int tplg_copy_data(snd_tplg_t *tplg, struct tplg_elem *elem, " element '%s'\n", ref->id, elem->id); return -EINVAL; } + ref->elem = ref_elem; tplg_dbg("Data '%s' used by '%s'\n", ref->id, elem->id); /* overlook empty private data */ -- 2.5.0

3 3

Re: [alsa-devel] Could I put a patch to 4.7 final?
by Takashi Iwai 21 Jul '16

21 Jul '16

On Thu, 21 Jul 2016 05:12:10 +0200, Takashi Sakamoto wrote: > > Iwai-san, > > Yesterday, I found an user of additional OXFW device, and we're under > testing a patch to support it. > https://sourceforge.net/p/ffado/mailman/message/35229260/ > > I need one day more to post corresponding patch. The prototype is: > https://github.com/takaswie/snd-firewire-improve/commit/690a2419fc95766826c… > > This week is final week for 4.7, but do I have a rest to add the patch > to it in time of your work for mainline? If the patch is a regression fix or a stable fix, I can queue it for 4.7-final. Others will go to 4.8-rc1. I'm going to send a pull request tomorrow, so please submit in time. thanks, Takashi

2 1

[alsa-devel] [PATCH 0/2] ASoC: samsung: Deferred probing support fixes
by Sylwester Nawrocki 21 Jul '16

21 Jul '16

This patch series is an attempt to fix sound card initialization isssues observed in Odroid XU3 board after conversion of the AUDSS clock controller driver to proper loadable module. Related patch can be found here: https://patchwork.kernel.org/patch/9221051, it's also already in -next. It would be good to have these two patches merged in 4.8-rc cycle so as to avoid regressions on on boards using ADMA. The sound breakage happens only when ADMA is used, as ADMA gets apb_pclk clock from the AUDSS clock controller. Previously the clock controller was initialized early in core_initcall which ensured the ADMA PL330 driver initialization before the I2S probe. Probably as a followup the SND_DMAENGINE_PCM_FLAG_CUSTOM_CHANNEL_NAME could be dropped and it's usage replaced with struct snd_dmaengine_pcm_config ::dma_names, I didn't want to make too invasive changes for a bug fix patch and don't have the older s3c non-dt boards for testing. Sylwester Nawrocki (2): ASoC: samsung: Fix error paths in the I2S driver's probe() ASoC: samsung: Specify DMA channels through struct snd_dmaengine_pcm_config sound/soc/samsung/ac97.c | 3 ++- sound/soc/samsung/dma.h | 9 ++++++--- sound/soc/samsung/dmaengine.c | 31 ++++++++++++++++++++----------- sound/soc/samsung/i2s.c | 28 ++++++++++++++++++++++------ sound/soc/samsung/pcm.c | 3 ++- sound/soc/samsung/s3c2412-i2s.c | 3 ++- sound/soc/samsung/s3c24xx-i2s.c | 3 ++- sound/soc/samsung/spdif.c | 3 ++- 8 files changed, 58 insertions(+), 25 deletions(-) -- 1.9.1

4 8

[alsa-devel] [PATCH 1/2] ASoC: cs53l30: Fix a bug for TDM slot location validation
by Nicolin Chen 21 Jul '16

21 Jul '16

The maximum slot number of CS53L30 is 4 while it should support the situation that's less than 4 channels based on the rx_mask. So when the driver validates the last slot location, it should check the last active slot instead of always the 4th one. Signed-off-by: Nicolin Chen <nicoleotsuka(a)gmail.com> --- sound/soc/codecs/cs53l30.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/codecs/cs53l30.c b/sound/soc/codecs/cs53l30.c index 5988b5c..fd5502e 100644 --- a/sound/soc/codecs/cs53l30.c +++ b/sound/soc/codecs/cs53l30.c @@ -809,8 +809,8 @@ static int cs53l30_set_dai_tdm_slot(struct snd_soc_dai *dai, return -EINVAL; } - /* Validate the last CS53L30 slot */ - slot_next = loc[CS53L30_TDM_SLOT_MAX - 1] + slot_step - 1; + /* Validate the last active CS53L30 slot */ + slot_next = loc[i - 1] + slot_step - 1; if (slot_next > 47) { dev_err(dai->dev, "slot selection out of bounds: %u\n", slot_next); -- 2.1.4

2 3

[alsa-devel] [PATCH v2 0/2] ALSA - hda: Add support for link audio time reporting
by Vinod Koul 21 Jul '16

21 Jul '16

Skylake onwards we support GTS capability and thus can report audio link time. This series adds support for parsing the capability and then reports link time. Guneshwor Singh (2): ALSA - hda: Add support for parsing new HDA capabilities ALSA - hda: Add support for link audio time reporting include/sound/hda_register.h | 23 ++++ sound/pci/hda/hda_controller.c | 244 ++++++++++++++++++++++++++++++++++++++++- sound/pci/hda/hda_controller.h | 27 +++++ sound/pci/hda/hda_intel.c | 12 ++ 4 files changed, 305 insertions(+), 1 deletion(-) -- 1.9.1

2 7

[alsa-devel] [PATCH] ASoC: rockchip: correct the spdif clk
by Chris Zhong 20 Jul '16

20 Jul '16

The spdif mclk should be 128 times of sample rate, and there is a internal divider, the real rate of spdif mclk is mclk / (div + 1). Hence, the original driver always get the good frequency for 48000/96000/44100/192000. But for 32000, the mclk is incorrect, it should be 32000*128, but get 48000*128. Do not use the internal divider here, just set all mclk to 128 * sample rate directly. Signed-off-by: Chris Zhong <zyw(a)rock-chips.com> --- sound/soc/rockchip/rockchip_spdif.c | 17 +---------------- 1 file changed, 1 insertion(+), 16 deletions(-) diff --git a/sound/soc/rockchip/rockchip_spdif.c b/sound/soc/rockchip/rockchip_spdif.c index 100781e..4ca2657 100644 --- a/sound/soc/rockchip/rockchip_spdif.c +++ b/sound/soc/rockchip/rockchip_spdif.c @@ -101,21 +101,7 @@ static int rk_spdif_hw_params(struct snd_pcm_substream *substream, int ret; srate = params_rate(params); - switch (srate) { - case 32000: - case 48000: - case 96000: - mclk = 96000 * 128; /* 12288000 hz */ - break; - case 44100: - mclk = 44100 * 256; /* 11289600 hz */ - break; - case 192000: - mclk = 192000 * 128; /* 24576000 hz */ - break; - default: - return -EINVAL; - } + mclk = srate * 128; switch (params_format(params)) { case SNDRV_PCM_FORMAT_S16_LE: @@ -139,7 +125,6 @@ static int rk_spdif_hw_params(struct snd_pcm_substream *substream, return ret; } - val |= SPDIF_CFGR_CLK_DIV(mclk/(srate * 256)); ret = regmap_update_bits(spdif->regmap, SPDIF_CFGR, SPDIF_CFGR_CLK_DIV_MASK | SPDIF_CFGR_HALFWORD_ENABLE | SDPIF_CFGR_VDW_MASK, -- 2.6.3

2 1

[alsa-devel] [PATCH 0/3] ASoC: Intel: Add support for surface3 tablet
by Vinod Koul 20 Jul '16

20 Jul '16

Surface3 tablet is CHT based device which used RT5645 codec. But the ACPI ID is 10EC5640 which belong to RT5640 codec :( So we add quirk support to load cht-machine driver for surface tablet combination, add the ACPI id and DMI fix for jack detect in codec and finally machine update. This fixes Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=98001 Please verify and send your tested-by. Vinod Koul (3): ASoC: Intel: Atom: Add quirk for Surface 3 ASoC: rt5645: Add ACPI ID 10EC5640 ASoC: Intel: Add surface3 entry in CHT-RT5645 machine sound/soc/codecs/rt5645.c | 7 ++++++ sound/soc/intel/atom/sst/sst_acpi.c | 44 ++++++++++++++++++++++++++++++++- sound/soc/intel/boards/cht_bsw_rt5645.c | 20 ++++++++++++++- sound/soc/intel/common/sst-acpi.h | 2 +- 4 files changed, 70 insertions(+), 3 deletions(-) -- 1.9.1

5 13

[alsa-devel] [alsa-lib][PATCH] ctl: use condition statements instead of assert() for new APIs to add an element set
by Takashi Sakamoto 20 Jul '16

20 Jul '16

Usage of assert() is not better practice of programming as shared library APIs. They should return appropriate error code to promote applications to handle error state. This commit applies condition statements with return value of -EINVAL, instead of assert(). As a backward compatibility for existent applications, old APIs still call assert(). Signed-off-by: Takashi Sakamoto <o-takashi(a)sakamocchi.jp> --- src/control/control.c | 24 +++++++++++++++++++----- 1 file changed, 19 insertions(+), 5 deletions(-) diff --git a/src/control/control.c b/src/control/control.c index aa05130..6c00b8e 100644 --- a/src/control/control.c +++ b/src/control/control.c @@ -382,7 +382,8 @@ int snd_ctl_add_integer_elem_set(snd_ctl_t *ctl, snd_ctl_elem_info_t *info, unsigned int numid; int err; - assert(ctl && info && info->id.name[0]); + if (ctl == NULL || info == NULL || info->id.name[0] == '\0') + return -EINVAL; info->type = SND_CTL_ELEM_TYPE_INTEGER; info->access = SNDRV_CTL_ELEM_ACCESS_READWRITE | @@ -471,7 +472,8 @@ int snd_ctl_add_integer64_elem_set(snd_ctl_t *ctl, snd_ctl_elem_info_t *info, unsigned int numid; int err; - assert(ctl && info && info->id.name[0]); + if (ctl == NULL || info == NULL || info->id.name[0] == '\0') + return -EINVAL; info->type = SND_CTL_ELEM_TYPE_INTEGER64; info->access = SNDRV_CTL_ELEM_ACCESS_READWRITE | @@ -549,7 +551,8 @@ int snd_ctl_add_boolean_elem_set(snd_ctl_t *ctl, snd_ctl_elem_info_t *info, unsigned int element_count, unsigned int member_count) { - assert(ctl && info && info->id.name[0]); + if (ctl == NULL || info == NULL || info->id.name[0] == '\0') + return -EINVAL; info->type = SND_CTL_ELEM_TYPE_BOOLEAN; info->access = SNDRV_CTL_ELEM_ACCESS_READWRITE | @@ -619,7 +622,9 @@ int snd_ctl_add_enumerated_elem_set(snd_ctl_t *ctl, snd_ctl_elem_info_t *info, char *buf, *p; int err; - assert(ctl && info && info->id.name[0] && labels); + if (ctl == NULL || info == NULL || info->id.name[0] == '\0' || + labels == NULL) + return -EINVAL; info->type = SND_CTL_ELEM_TYPE_ENUMERATED; info->access = SNDRV_CTL_ELEM_ACCESS_READWRITE | @@ -698,7 +703,8 @@ int snd_ctl_add_bytes_elem_set(snd_ctl_t *ctl, snd_ctl_elem_info_t *info, unsigned int element_count, unsigned int member_count) { - assert(ctl && info && info->id.name[0]); + if (ctl == NULL || info == NULL || info->id.name[0] == '\0') + return -EINVAL; info->type = SND_CTL_ELEM_TYPE_BYTES; info->access = SNDRV_CTL_ELEM_ACCESS_READWRITE | @@ -726,6 +732,8 @@ int snd_ctl_elem_add_integer(snd_ctl_t *ctl, const snd_ctl_elem_id_t *id, { snd_ctl_elem_info_t info = {0}; + assert(ctl && id && id->name[0]); + info.id = *id; return snd_ctl_add_integer_elem_set(ctl, &info, 1, member_count, @@ -745,6 +753,8 @@ int snd_ctl_elem_add_integer64(snd_ctl_t *ctl, const snd_ctl_elem_id_t *id, { snd_ctl_elem_info_t info = {0}; + assert(ctl && id && id->name[0]); + info.id = *id; return snd_ctl_add_integer64_elem_set(ctl, &info, 1, member_count, @@ -763,6 +773,8 @@ int snd_ctl_elem_add_boolean(snd_ctl_t *ctl, const snd_ctl_elem_id_t *id, { snd_ctl_elem_info_t info = {0}; + assert(ctl && id && id->name[0]); + info.id = *id; return snd_ctl_add_boolean_elem_set(ctl, &info, 1, member_count); @@ -783,6 +795,8 @@ int snd_ctl_elem_add_enumerated(snd_ctl_t *ctl, const snd_ctl_elem_id_t *id, { snd_ctl_elem_info_t info = {0}; + assert(ctl && id && id->name[0] && labels); + info.id = *id; return snd_ctl_add_enumerated_elem_set(ctl, &info, 1, member_count, -- 2.7.4

2 1

[alsa-devel] problem with copy function
by Shahram Shabpareh 20 Jul '16

20 Jul '16

Hi I have a board with ti am3358 processor. it has a sound port called McASP and I've used IDT821034 as codec. The Linux kernel source package that is included by board company is 3.2.0 and I wrote a basic driver for IDT821034 codec and changed the code of davinci-mcasp.c . Now the basic functionality of system is ok and i have signals on oscilloscope. Another thing is i have to setup a specific configuration for this system. the target system must have 32 tdm slots in each frame and 32 PCM channels and each slot is 8 bits. this is also done in kernel source code. currently the problem is i want that data of channel 1 be mapped to slot 1, channel 2 be mapped to slot 2, .... and channel 32 be mapped to slot 32. but when i run aplay to send to only channel 1, data will be send to all slots! i want that all slots be 0 except slot 1. How can i do that? I searched in source tree and found the function that i need to override this behaviour is copy that is defined in snd_pcm_ops structure. I defined a function called davinci_pcm_copy in davinci-pcm.c file and added the function to structure like below: static struct snd_pcm_ops davinci_pcm_ops = { .... .copy = davinci_pcm_copy, }; the problem is, this copy function is not called at all by kernel pcm subsystem. Again i searched the kernel code and found that copy function has a calling path as below: alsa-lib -> snd_pcm_playback_ioctl -> snd_pcm_playback_ioctl1 -> snd_pcm_lib_write -> snd_pcm_lib_write1 -> snd_pcm_lib_write_transfer -> copy but my calling path is: alsa-lib -> snd_pcm_playback_ioctl -> snd_pcm_playback_ioctl1 -> snd_pcm_common_ioctl1 -> snd_pcm_sync_ptr I created a wav file from a mp3 file with this command: ffmpeg -i file.mp3 -ac 1 -ar 8000 -ab 64k -f u8 file.wav and played with this command: aplay -vv -c 1 file.wav Strange thing is when i run this aplay command in my PC the first calling path is executed but when i run the same aplay with same file in my board the second calling path is executed. Is it related to software and hardware params of sound devices? if yes, what should i do to copy function being called in my board? thanks

2 2