July 2019 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] bug in playing 6-channel sound
by Julian Bradfield 05 Jul '19

05 Jul '19

Some five years ago, the following post https://mailman.alsa-project.org/pipermail/alsa-devel/2014-February/072265.… reported a failure to play six-channel sound, that could be worked around by setting the prealloc value for the device to various values *other than* the default 64. I have just run into this playing a movie from my laptop (Thinkpad Yoga X1, Intel HDA PCH sound card outputting through HDMI) running Xubuntu 18.04, up to date as at the time of posting. I see exactly the same problem as in that post, and the same workaround works for me - I tried prealloc values 63,65,32,128, and any of them is ok. I'm a bit baffled that such a problem has survived for several years. If any developer would like me to provide debugging information, please get in touch - I have some time available at present, but not enough to learn ALSA and debug it myself! I will remain subscribed to the list for a few weeks to read any followups.

2 1

[alsa-devel] [GIT PULL]: soundwire updates for v5.3-rc1
by Vinod Koul 05 Jul '19

05 Jul '19

Hello Greg, Please pull to receive the updates for soundwire for v5.3. The following changes since commit a188339ca5a396acc588e5851ed7e19f66b0ebd9: Linux 5.2-rc1 (2019-05-19 15:47:09 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire.git tags/soundwire-5.3-rc1 for you to fetch changes up to 2aeac95d1a4cc85aae57ab842d5c3340df0f817f: soundwire: add module_sdw_driver helper macro (2019-07-01 11:52:43 +0530) ---------------------------------------------------------------- soundwire updates for v5.3-rc1 Updates for 5.3 include: - module_sdw_driver macro for drivers - Documentation updates for code-blocks - Improvement from Pierre on intel and cadence driver - Clarification of DisCo properties and updates ---------------------------------------------------------------- Jan Kotas (1): soundwire: cdns: Fix compilation error on arm64 Mauro Carvalho Chehab (1): docs: soundwire: locking: fix tags for a code-block Pierre-Louis Bossart (15): soundwire: intel: filter SoundWire controller device search soundwire: mipi_disco: fix master/link error soundwire: add port-related definitions soundwire: remove master data port properties soundwire: mipi-disco: remove master_count property for masters soundwire: rename 'freq' fields soundwire: mipi-disco: fix clock stop modes soundwire: clarify comment soundwire: rename/clarify MIPI DisCo properties soundwire: cadence_master: use rate_limited dynamic debug soundwire: cadence_master: log Slave status mask on errors soundwire: cadence_master: check the number of bidir PDIs soundwire: Intel: add log for number of PCM and PDM PDIs soundwire: fix typo in comments soundwire: intel_init: add checks on link numbers Srinivas Kandagatla (1): soundwire: add module_sdw_driver helper macro Documentation/driver-api/soundwire/locking.rst | 4 +- drivers/soundwire/bus.c | 6 +- drivers/soundwire/cadence_master.c | 30 +++++---- drivers/soundwire/intel.c | 17 +++-- drivers/soundwire/intel.h | 2 +- drivers/soundwire/intel_init.c | 25 +++++++- drivers/soundwire/mipi_disco.c | 35 +++++----- drivers/soundwire/stream.c | 8 +-- include/linux/soundwire/sdw.h | 88 ++++++++++++++++++++------ include/linux/soundwire/sdw_type.h | 11 ++++ 10 files changed, 162 insertions(+), 64 deletions(-) -- ~Vinod

2 1

[alsa-devel] [PATCH] sound: pci: asihpi: Remove unneeded variable change
by Hariprasad Kelam 05 Jul '19

05 Jul '19

this patch fixes below issue reported by coccicheck sound/pci/asihpi/asihpi.c:1558:5-11: Unneeded variable: "change". Return "1" on line 1564 Signed-off-by: Hariprasad Kelam <hariprasad.kelam(a)gmail.com> --- sound/pci/asihpi/asihpi.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/sound/pci/asihpi/asihpi.c b/sound/pci/asihpi/asihpi.c index e7234f3..2a21a3d 100644 --- a/sound/pci/asihpi/asihpi.c +++ b/sound/pci/asihpi/asihpi.c @@ -1519,7 +1519,6 @@ static int snd_asihpi_volume_get(struct snd_kcontrol *kcontrol, static int snd_asihpi_volume_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { - int change; u32 h_control = kcontrol->private_value; short an_gain_mB[HPI_MAX_CHANNELS]; @@ -1530,9 +1529,8 @@ static int snd_asihpi_volume_put(struct snd_kcontrol *kcontrol, /* change = asihpi->mixer_volume[addr][0] != left || asihpi->mixer_volume[addr][1] != right; */ - change = 1; hpi_handle_error(hpi_volume_set_gain(h_control, an_gain_mB)); - return change; + return 1; } static const DECLARE_TLV_DB_SCALE(db_scale_100, -10000, VOL_STEP_mB, 0); @@ -1555,13 +1553,12 @@ static int snd_asihpi_volume_mute_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { u32 h_control = kcontrol->private_value; - int change = 1; /* HPI currently only supports all or none muting of multichannel volume ALSA Switch element has opposite sense to HPI mute: on==unmuted, off=muted */ int mute = ucontrol->value.integer.value[0] ? 0 : HPI_BITMASK_ALL_CHANNELS; hpi_handle_error(hpi_volume_set_mute(h_control, mute)); - return change; + return 1; } static int snd_asihpi_volume_add(struct snd_card_asihpi *asihpi, -- 2.7.4

2 1

[alsa-devel] audio lost from speaker after reboot from windows on the device ALC295
by He, Bo 05 Jul '19

05 Jul '19

Hi, patch_realtek.c maintainer: I see one issue that reboot from windows and boot to ubuntu, the audio lost from speaker, I suspect there are some bugs in patch_realtek.c drivers, the device is ALC295 and the device id is 0x10ec0295. I have done the below experiments: 1. reboot from windows to windows, the audio is persist . 2. reboot from windows to ubuntu, the audio lost from speaker, but can hear if I hotplug one earphone. 3. if the issue reproduce after reboot from windows, reboot the ubuntu can't restore the audio, I suspect it's warm reset. 4. if I write the port 0xcf9 with 0xe to do cold reset, the audio can restore. 5. if I do suspend/resume, the audio can restore, I suspect do cold boot and suspend will trigger the platform reset to reset the ALC295. 6. if I do double function reset (write the verb 0x7ff in alc_init), the audio is still can't restore. snd_hda_codec_write(codec, 0x01, 0, AC_VERB_SET_CODEC_RESET, 0); /* Function reset */ snd_hda_codec_write(codec, 0x01, 0, AC_VERB_SET_CODEC_RESET, 0); /* double Function reset */ 7. the issue is first found on kernel 4.19.50, I still see the issue with the latest kernel 5.2-rc2, is it possible windows change some default registers, but ALC295 don't initialize the register?

3 4

[alsa-devel] [PATCH 0/7] ASoC: Intel: Skylake: Driver fundaments overhaul
by Cezary Rojewski 04 Jul '19

04 Jul '19

Skylake driver is divided into two modules: - snd_soc_skl - snd_soc_skl_ipc and nothing would be wrong if not for the fact that both cannot exist without one another. IPC module is not some kind of extension, as it is the case for snd_hda_ext_core which is separated from snd_hda_core - legacy hda interface. It's as much core Skylake module as snd_soc_skl is. Statement backup by existence of circular dependency between this two. To eliminate said problem, struct skl_sst has been created. From that momment, Skylake has been plagued by header errors (incomplete sturcts, unknown references etc.) whenever something new is to be added or code is cleaned up. Fix this flawed design by merging snd_soc_skl and snd_soc_skl_ipc. Also, do not forget about struct skl_sst redundancy. Followup changes address harmful assumptions and false logic which driver currently implements e.g.: attempt to take role of master for DSP scheduling when in fact entire control takes place in DSP. Amadeusz Sławiński (2): ASoC: Intel: Skylake: Combine snd_soc_skl_ipc and snd_soc_skl ASoC: Intel: Skylake: Do not disable FW notifications Cezary Rojewski (5): ASoC: Intel: Skylake: Merge skl_sst and skl into skl_dev struct ASoC: Intel: Skylake: Remove MCPS available check ASoC: Intel: Skylake: Remove memory available check ASoC: Intel: Skylake: Make MCPS and CPS params obsolete ASoC: Intel: Skylake: Cleanup skl_module_cfg declaration sound/soc/intel/common/sst-ipc.h | 1 + sound/soc/intel/skylake/Makefile | 12 +- sound/soc/intel/skylake/bxt-sst.c | 50 +-- sound/soc/intel/skylake/cnl-sst-dsp.h | 7 +- sound/soc/intel/skylake/cnl-sst.c | 37 +- sound/soc/intel/skylake/skl-debug.c | 14 +- sound/soc/intel/skylake/skl-messages.c | 245 ++++++------- sound/soc/intel/skylake/skl-nhlt.c | 18 +- sound/soc/intel/skylake/skl-pcm.c | 74 ++-- sound/soc/intel/skylake/skl-ssp-clk.c | 4 +- sound/soc/intel/skylake/skl-sst-dsp.c | 10 +- sound/soc/intel/skylake/skl-sst-dsp.h | 29 +- sound/soc/intel/skylake/skl-sst-ipc.c | 8 +- sound/soc/intel/skylake/skl-sst-ipc.h | 52 +-- sound/soc/intel/skylake/skl-sst-utils.c | 37 +- sound/soc/intel/skylake/skl-sst.c | 51 +-- sound/soc/intel/skylake/skl-topology.c | 441 ++++++++---------------- sound/soc/intel/skylake/skl-topology.h | 43 +-- sound/soc/intel/skylake/skl.c | 54 +-- sound/soc/intel/skylake/skl.h | 102 ++++-- 20 files changed, 546 insertions(+), 743 deletions(-) -- 2.17.1

1 7

[alsa-devel] [PATCH v2] ASoC: wcd9335: remove multiple defines.
by Srinivas Kandagatla 04 Jul '19

04 Jul '19

Found during review that there are multiple defines of same constants. This patch removes them! Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla(a)linaro.org> --- Changes since v1: rebased on top of kernel/git/broonie/sound.git for-next sound/soc/codecs/wcd9335.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/sound/soc/codecs/wcd9335.c b/sound/soc/codecs/wcd9335.c index 85737fe54474..1bbbe421b999 100644 --- a/sound/soc/codecs/wcd9335.c +++ b/sound/soc/codecs/wcd9335.c @@ -86,11 +86,6 @@ #define WCD9335_DEC_PWR_LVL_HP 0x04 #define WCD9335_DEC_PWR_LVL_DF 0x00 -#define TX_HPF_CUT_OFF_FREQ_MASK 0x60 -#define CF_MIN_3DB_4HZ 0x0 -#define CF_MIN_3DB_75HZ 0x1 -#define CF_MIN_3DB_150HZ 0x2 - #define WCD9335_SLIM_RX_CH(p) \ {.port = p + WCD9335_RX_START, .shift = p,} -- 2.21.0

1 0

[alsa-devel] [PATCH] ASoC: wcd9335: remove multiple defines.
by Srinivas Kandagatla 04 Jul '19

04 Jul '19

Found during review that there are multiple defines of same constants. This patch removes them! Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla(a)linaro.org> --- sound/soc/codecs/wcd9335.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/sound/soc/codecs/wcd9335.c b/sound/soc/codecs/wcd9335.c index 6f205638a8b8..da6c50ccecd9 100644 --- a/sound/soc/codecs/wcd9335.c +++ b/sound/soc/codecs/wcd9335.c @@ -87,11 +87,6 @@ #define WCD9335_DEC_PWR_LVL_HP 0x04 #define WCD9335_DEC_PWR_LVL_DF 0x00 -#define TX_HPF_CUT_OFF_FREQ_MASK 0x60 -#define CF_MIN_3DB_4HZ 0x0 -#define CF_MIN_3DB_75HZ 0x1 -#define CF_MIN_3DB_150HZ 0x2 - #define WCD9335_MBHC_MAX_BUTTONS (8) #define WCD9335_SLIM_RX_CH(p) \ -- 2.21.0

2 1

[alsa-devel] [PATCH] ALSA: usb-audio: Fix parse of UAC2 Extension Units
by Takashi Iwai 04 Jul '19

04 Jul '19

Extension Unit (XU) is used to have a compatible layout with Processing Unit (PU) on UAC1, and the usb-audio driver code assumed it for parsing the descriptors. Meanwhile, on UAC2, XU became slightly incompatible with PU; namely, XU has a one-byte bmControls bitmap while PU has two bytes bmControls bitmap. This incompatibility results in the read of a wrong address for the last iExtension field, which ended up with an incorrect string for the mixer element name, as recently reported for Focusrite Scarlett 18i20 device. This patch corrects the present misalignment by introducing a couple of new macros and calling them depending on the descriptor type. Reported-by: Stefan Sauer <ensonic(a)hora-obscura.de> Cc: <stable(a)vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- include/uapi/linux/usb/audio.h | 37 +++++++++++++++++++++++++++++++++++++ sound/usb/mixer.c | 16 ++++++++++------ 2 files changed, 47 insertions(+), 6 deletions(-) diff --git a/include/uapi/linux/usb/audio.h b/include/uapi/linux/usb/audio.h index ddc5396800aa..76b7c3f6cd0d 100644 --- a/include/uapi/linux/usb/audio.h +++ b/include/uapi/linux/usb/audio.h @@ -450,6 +450,43 @@ static inline __u8 *uac_processing_unit_specific(struct uac_processing_unit_desc } } +/* + * Extension Unit (XU) has almost compatible layout with Processing Unit, but + * on UAC2, it has a different bmControls size (bControlSize); it's 1 byte for + * XU while 2 bytes for PU. The last iExtension field is a one-byte index as + * well as iProcessing field of PU. + */ +static inline __u8 uac_extension_unit_bControlSize(struct uac_processing_unit_descriptor *desc, + int protocol) +{ + switch (protocol) { + case UAC_VERSION_1: + return desc->baSourceID[desc->bNrInPins + 4]; + case UAC_VERSION_2: + return 1; /* in UAC2, this value is constant */ + case UAC_VERSION_3: + return 4; /* in UAC3, this value is constant */ + default: + return 1; + } +} + +static inline __u8 uac_extension_unit_iExtension(struct uac_processing_unit_descriptor *desc, + int protocol) +{ + __u8 control_size = uac_extension_unit_bControlSize(desc, protocol); + + switch (protocol) { + case UAC_VERSION_1: + case UAC_VERSION_2: + default: + return *(uac_processing_unit_bmControls(desc, protocol) + + control_size); + case UAC_VERSION_3: + return 0; /* UAC3 does not have this field */ + } +} + /* 4.5.2 Class-Specific AS Interface Descriptor */ struct uac1_as_header_descriptor { __u8 bLength; /* in bytes: 7 */ diff --git a/sound/usb/mixer.c b/sound/usb/mixer.c index e003b5e7b01a..ac121b10c51c 100644 --- a/sound/usb/mixer.c +++ b/sound/usb/mixer.c @@ -2318,7 +2318,7 @@ static struct procunit_info extunits[] = { */ static int build_audio_procunit(struct mixer_build *state, int unitid, void *raw_desc, struct procunit_info *list, - char *name) + bool extension_unit) { struct uac_processing_unit_descriptor *desc = raw_desc; int num_ins; @@ -2335,6 +2335,8 @@ static int build_audio_procunit(struct mixer_build *state, int unitid, static struct procunit_info default_info = { 0, NULL, default_value_info }; + const char *name = extension_unit ? + "Extension Unit" : "Processing Unit"; if (desc->bLength < 13) { usb_audio_err(state->chip, "invalid %s descriptor (id %d)\n", name, unitid); @@ -2448,7 +2450,10 @@ static int build_audio_procunit(struct mixer_build *state, int unitid, } else if (info->name) { strlcpy(kctl->id.name, info->name, sizeof(kctl->id.name)); } else { - nameid = uac_processing_unit_iProcessing(desc, state->mixer->protocol); + if (extension_unit) + nameid = uac_extension_unit_iExtension(desc, state->mixer->protocol); + else + nameid = uac_processing_unit_iProcessing(desc, state->mixer->protocol); len = 0; if (nameid) len = snd_usb_copy_string_desc(state->chip, @@ -2481,10 +2486,10 @@ static int parse_audio_processing_unit(struct mixer_build *state, int unitid, case UAC_VERSION_2: default: return build_audio_procunit(state, unitid, raw_desc, - procunits, "Processing Unit"); + procunits, false); case UAC_VERSION_3: return build_audio_procunit(state, unitid, raw_desc, - uac3_procunits, "Processing Unit"); + uac3_procunits, false); } } @@ -2495,8 +2500,7 @@ static int parse_audio_extension_unit(struct mixer_build *state, int unitid, * Note that we parse extension units with processing unit descriptors. * That's ok as the layout is the same. */ - return build_audio_procunit(state, unitid, raw_desc, - extunits, "Extension Unit"); + return build_audio_procunit(state, unitid, raw_desc, extunits, true); } /* -- 2.16.4

1 0

[alsa-devel] [PATCH] ALSA: hda/ca0132 - remove redundant assignment to variable 'changed'
by Colin King 04 Jul '19

04 Jul '19

From: Colin Ian King <colin.king(a)canonical.com> The variable 'changed' is being initialized with a value that is never read and it is being updated later with a new value. The initialization is redundant and can be removed. Addresses-Coverity: ("Unused value") Signed-off-by: Colin Ian King <colin.king(a)canonical.com> --- sound/pci/hda/patch_ca0132.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/hda/patch_ca0132.c b/sound/pci/hda/patch_ca0132.c index c0565437ffed..0d51823d7270 100644 --- a/sound/pci/hda/patch_ca0132.c +++ b/sound/pci/hda/patch_ca0132.c @@ -5980,7 +5980,7 @@ static int ca0132_alt_volume_put(struct snd_kcontrol *kcontrol, int ch = get_amp_channels(kcontrol); long *valp = ucontrol->value.integer.value; hda_nid_t vnid = 0; - int changed = 1; + int changed; switch (nid) { case 0x02: -- 2.20.1

2 1

[alsa-devel] [PATCH v2 33/35] sound/pci: Use kmemdup rather than duplicating its implementation
by Fuqian Huang 04 Jul '19

04 Jul '19

kmemdup is introduced to duplicate a region of memory in a neat way. Rather than kmalloc/kzalloc + memcpy, which the programmer needs to write the size twice (sometimes lead to mistakes), kmemdup improves readability, leads to smaller code and also reduce the chances of mistakes. Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy. Signed-off-by: Fuqian Huang <huangfq.daxian(a)gmail.com> --- Changes in v2: - Fix a typo in commit message (memset -> memcpy) sound/pci/echoaudio/echoaudio.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/sound/pci/echoaudio/echoaudio.c b/sound/pci/echoaudio/echoaudio.c index b612a536a5a1..35bd3e7c8ce1 100644 --- a/sound/pci/echoaudio/echoaudio.c +++ b/sound/pci/echoaudio/echoaudio.c @@ -2189,11 +2189,10 @@ static int snd_echo_resume(struct device *dev) u32 pipe_alloc_mask; int err; - commpage_bak = kmalloc(sizeof(*commpage), GFP_KERNEL); + commpage_bak = kmemdup(commpage, sizeof(*commpage), GFP_KERNEL); if (commpage_bak == NULL) return -ENOMEM; commpage = chip->comm_page; - memcpy(commpage_bak, commpage, sizeof(*commpage)); err = init_hw(chip, chip->pci->device, chip->pci->subsystem_device); if (err < 0) { -- 2.11.0

2 1