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

[alsa-devel] [BUG] ALSA: seq: a possible sleep-in-atomic-context bug in snd_virmidi_dev_receive_event()
by Jia-Ju Bai 17 Dec '19

17 Dec '19

The driver may sleep while holding a read lock. The function call path (from bottom to top) in Linux 4.19 is: sound/core/seq/seq_memory.c, 96: copy_from_user in snd_seq_dump_var_event sound/core/seq/seq_virmidi.c, 97: snd_seq_dump_var_event in snd_virmidi_dev_receive_event sound/core/seq/seq_virmidi.c, 88: _raw_read_lock in snd_virmidi_dev_receive_event copy_from_user() can sleep at runtime. I am not sure how to properly fix this possible bug, so I only report it. This bug is found by a static analysis tool STCheck written by myself. Best wishes, Jia-Ju Bai

3 3

[alsa-devel] [GIT PULL] ASoC fixes for v5.5
by Mark Brown 17 Dec '19

17 Dec '19

The following changes since commit 39870b0dec68ed7dd814beb697e541670975c7d8: ASoC: DMI long name - avoid to add board name if matches with product name (2019-11-22 19:54:50 +0000) are available in the Git repository at: https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git tags/asoc-fix-v5.5-rc2 for you to fetch changes up to 556672d75ff486e0b6786056da624131679e0576: ASoC: wm8962: fix lambda value (2019-12-16 11:47:28 +0000) ---------------------------------------------------------------- ASoC: Fixes for v5.5 A collection of fixes since the merge window, mostly driver specific but there's a few in the core that clean up fallout from the refactorings done in the last cycle. ---------------------------------------------------------------- Amery Song (1): ASoC: Intel: common: work-around incorrect ACPI HID for CML boards Curtis Malainey (3): ASoC: core: Init pcm runtime work early to avoid warnings ASoC: core: only flush inited work during free ASoC: SOF: Intel: split cht and byt debug window sizes Daniel Baluta (2): ASoC: SOF: topology: Fix unload for SAI/ESAI ASoC: simple-card: Don't create separate link when platform is present Dragos Tarcatu (2): ASoC: topology: Check return value for snd_soc_add_dai_link() ASoC: topology: Check return value for soc_tplg_pcm_create() Hans de Goede (1): ASoC: Intel: bytcr_rt5640: Update quirk for Teclast X89 Karol Trzcinski (2): ASoC: SOF: loader: snd_sof_fw_parse_ext_data log warning on unknown header ASoC: SOF: loader: fix snd_sof_fw_parse_ext_data Keyon Jie (1): ASoC: SOF: Intel: BYT: fix a copy/paste mistake in byt_dump() Michael Walle (1): ASoC: wm8904: fix automatic sysclk configuration Shengjiu Wang (1): ASoC: wm8962: fix lambda value Shuming Fan (1): ASoC: rt5682: fix i2c arbitration lost issue Tzung-Bi Shih (3): ASoC: max98090: remove msleep in PLL unlocked workaround ASoC: max98090: exit workaround earlier if PLL is locked ASoC: max98090: fix possible race conditions Yu-Hsuan Hsu (1): ASoC: AMD: Enable clk in startup intead of hw_params YueHaibing (2): ASoC: rt5677: Fix build error without CONFIG_SPI ASoC: Intel: sst: Add missing include <linux/io.h> include/sound/soc.h | 1 + sound/soc/amd/acp-da7219-max98357a.c | 46 +++++++---------------- sound/soc/codecs/max98090.c | 30 ++++++++++----- sound/soc/codecs/max98090.h | 1 - sound/soc/codecs/rt5677-spi.h | 16 ++++++++ sound/soc/codecs/rt5682.c | 2 + sound/soc/codecs/wm8904.c | 6 +++ sound/soc/codecs/wm8962.c | 4 +- sound/soc/generic/simple-card.c | 6 +++ sound/soc/intel/atom/sst/sst.c | 1 + sound/soc/intel/boards/bytcr_rt5640.c | 8 +++- sound/soc/intel/common/soc-acpi-intel-cml-match.c | 41 +++++++++++--------- sound/soc/soc-compress.c | 6 +-- sound/soc/soc-core.c | 13 ++++++- sound/soc/soc-pcm.c | 11 ++---- sound/soc/soc-topology.c | 27 +++++++++---- sound/soc/sof/intel/byt.c | 25 ++++++------ sound/soc/sof/loader.c | 7 +++- sound/soc/sof/topology.c | 4 +- 19 files changed, 158 insertions(+), 97 deletions(-)

2 1

[alsa-devel] [PATCH] ASoC: intel: skylake: Remove superfluous bus ops
by Takashi Iwai 17 Dec '19

17 Dec '19

The bus_core_ops values set in skl driver are same as the default values, so we can drop it and pass NULL to snd_hdac_ext_bus_init() instead. Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/soc/intel/skylake/skl.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/sound/soc/intel/skylake/skl.c b/sound/soc/intel/skylake/skl.c index 58ba3e9469ba..f755ca2484cf 100644 --- a/sound/soc/intel/skylake/skl.c +++ b/sound/soc/intel/skylake/skl.c @@ -773,11 +773,6 @@ static void skl_codec_create(struct hdac_bus *bus) } } -static const struct hdac_bus_ops bus_core_ops = { - .command = snd_hdac_bus_send_cmd, - .get_response = snd_hdac_bus_get_response, -}; - static int skl_i915_init(struct hdac_bus *bus) { int err; @@ -888,7 +883,7 @@ static int skl_create(struct pci_dev *pci, #if IS_ENABLED(CONFIG_SND_SOC_INTEL_SKYLAKE_HDAUDIO_CODEC) ext_ops = snd_soc_hdac_hda_get_ops(); #endif - snd_hdac_ext_bus_init(bus, &pci->dev, &bus_core_ops, ext_ops); + snd_hdac_ext_bus_init(bus, &pci->dev, NULL, ext_ops); bus->use_posbuf = 1; skl->pci = pci; INIT_WORK(&skl->probe_work, skl_probe_work); -- 2.16.4

2 1

[alsa-devel] Don't use 16 bit signed integers for sample rate conversion.
by GitHub issues - edited 17 Dec '19

17 Dec '19

alsa-project/alsa-plugins issue #5 was edited from crocket: Sample rate conversion should be done with the highest resolution possible. I recommend using 32bit signed integers for sample rate conversion. Currently, once sample rate conversion is done, using S32_LE is meaningless. Issue URL : https://github.com/alsa-project/alsa-plugins/issues/5 Repository URL: https://github.com/alsa-project/alsa-plugins

1 0

[alsa-devel] Don't use 16 bit signed integers for sample rate conversion.
by GitHub issues - edited 17 Dec '19

17 Dec '19

alsa-project/alsa-plugins issue #5 was edited from crocket: Sample rate conversion should be done with the highest resolution possible. Maybe, double, unsigned int, or unsigned int64? Once sample rate conversion is done, using S32_LE is meaningless. Does dmix also use 16bits internally? Issue URL : https://github.com/alsa-project/alsa-plugins/issues/5 Repository URL: https://github.com/alsa-project/alsa-plugins

1 0

[alsa-devel] [PATCH for-5.6 00/14] media: ALSA PCM API updates
by Takashi Iwai 17 Dec '19

17 Dec '19

Hi, this is a patch set to adapt the latest ALSA PCM API to media drivers. Basically these are merely cleanups, as shown in diffstat, and there should be no functional changes. As the corresponding ALSA PCM API change is found in 5.5-rc1, please apply these on 5.5-rc1 or later. Or if you prefer, I can merge them through sound tree, too. Let me know. thanks, Takashi === Takashi Iwai (14): media: solo6x10: Use managed buffer allocation media: tw686x: Use managed buffer allocation media: usbtv: Use managed buffer allocation media: cobalt: Clean up ALSA PCM API usages media: cx18: Clean up ALSA PCM API usages media: ivtv: Clean up ALSA PCM API usages media: cs231xx: Clean up ALSA PCM API usages media: em28xx: Clean up ALSA PCM API usages media: go7007: Clean up ALSA PCM API usages media: tm6000: Clean up ALSA PCM API usages media: cobalt: Drop superfluous ioctl PCM ops media: cx18: Drop superfluous ioctl PCM ops media: ivtv: Drop superfluous ioctl PCM ops media: Drop superfluous ioctl PCM ops drivers/media/pci/cobalt/cobalt-alsa-pcm.c | 69 ++---------------------- drivers/media/pci/cx18/cx18-alsa-pcm.c | 75 +------------------------- drivers/media/pci/cx23885/cx23885-alsa.c | 1 - drivers/media/pci/cx25821/cx25821-alsa.c | 1 - drivers/media/pci/cx88/cx88-alsa.c | 1 - drivers/media/pci/ivtv/ivtv-alsa-pcm.c | 76 +------------------------- drivers/media/pci/saa7134/saa7134-alsa.c | 1 - drivers/media/pci/solo6x10/solo6x10-g723.c | 24 ++------- drivers/media/pci/tw686x/tw686x-audio.c | 16 +----- drivers/media/usb/cx231xx/cx231xx-audio.c | 79 +-------------------------- drivers/media/usb/em28xx/em28xx-audio.c | 87 +----------------------------- drivers/media/usb/go7007/snd-go7007.c | 25 ++------- drivers/media/usb/tm6000/tm6000-alsa.c | 82 +--------------------------- drivers/media/usb/usbtv/usbtv-audio.c | 29 +--------- 14 files changed, 19 insertions(+), 547 deletions(-) -- 2.16.4

2 16

[alsa-devel] [PATCH v4 00/15] soundwire: intel: implement new ASoC interfaces
by Pierre-Louis Bossart 16 Dec '19

16 Dec '19

This patchset applies on top of soundwire/next, now that the interface definitions are merged. The changes are essentially a removal of the platform devices and the implementation of the new interfaces required to scan the ACPI tables, probe the links and start them. The missing prepare, trigger and setup ASoC callbacks are also implemented. The hw_params and free callbacks use the new interfaces as well. While there are quite a few lines of code changed, this is mostly about interface changes. The next series will contain more functional changes and deal with race conditions on probe, enumeration and suspend/resume issues. Changes since v3: One line change to re-add EXPORT_SYMBOL Add missing driver_registration Changes since v2: moved uevent handling to slave_type (Vinod) Changes since v1: fix typo (Vinod) removed uevent open for Master (Vinod) clarified commit messages (Cezary) no functionality change Bard Liao (1): soundwire: register master device driver Pierre-Louis Bossart (11): soundwire: renames to prepare support for master drivers/devices soundwire: rename dev_to_sdw_dev macro soundwire: rename drv_to_sdw_slave_driver macro soundwire: bus_type: rename sdw_drv_ to sdw_slave_drv soundwire: intel: rename res field as link_res soundwire: add support for sdw_slave_type soundwire: slave: move uevent handling to slave soundwire: add initial definitions for sdw_master_device soundwire: intel: remove platform devices and provide new interface soundwire: intel: free all resources on hw_free() soundwire: intel_init: add implementation of sdw_intel_enable_irq() Rander Wang (3): soundwire: intel: add prepare support in sdw dai driver soundwire: intel: add trigger support in sdw dai driver soundwire: intel: add sdw_stream_setup helper for .startup callback drivers/base/regmap/regmap-sdw.c | 4 +- drivers/soundwire/Makefile | 2 +- drivers/soundwire/bus.c | 2 +- drivers/soundwire/bus.h | 2 + drivers/soundwire/bus_type.c | 63 +++--- drivers/soundwire/intel.c | 282 +++++++++++++++++++++----- drivers/soundwire/intel.h | 8 +- drivers/soundwire/intel_init.c | 309 ++++++++++++++++++++++------- drivers/soundwire/master.c | 63 ++++++ drivers/soundwire/slave.c | 10 +- include/linux/soundwire/sdw.h | 39 +++- include/linux/soundwire/sdw_type.h | 34 +++- 12 files changed, 657 insertions(+), 161 deletions(-) create mode 100644 drivers/soundwire/master.c -- 2.20.1

2 33

[alsa-devel] alsa-info: lots of mostly syntactic changes
by GitHub pull_request - edited 16 Dec '19

16 Dec '19

alsa-project/alsa-utils pull request #29 was edited from E5ten: - remove all bashisms and change shebang to #!/bin/sh - replace use of which with command -v for portability - remove awk dependency - general cosmetic consistency: - indentation - always put then or do on the same line as if, for, or while - quote much more, especially when the variable refers to a file - replace many uses of >> $FILE with command grouping and single redirects of the command groups - simplify text processing pipelines, mostly by turning them into a single sed command Request URL : https://github.com/alsa-project/alsa-utils/pull/29 Patch URL : https://github.com/alsa-project/alsa-utils/pull/29.patch Repository URL: https://github.com/alsa-project/alsa-utils

1 0

[alsa-devel] Free-running MMAP Mode
by Fletcher Woodruff 16 Dec '19

16 Dec '19

Hi alsa-devel, I'm working on implementing memory-mapped audio for a project that uses the alsa user-space API. My understanding is that typically, mmap mode requires explicit calls to mmap_begin/mmap_commit before and after every read/write in order to notify the driver of newly read/written frames. Is there any way for a user to request a sort of "free-running" MMAP, where alsa just assumes that the data it needs has been written to the circular buffer and doesn't expect those begin/commit calls? Thanks, Fletcher

1 0

[alsa-devel] [PATCH v2 0/2] Add the broadmobi BM818
by Angus Ainslie (Purism) 16 Dec '19

16 Dec '19

he broadmobi uses slightly different parameters from the option modems so add the paramters and document them. Changes since v1: Cut back the CC list. Use data from compatible match instead of explicit compatible match. Angus Ainslie (Purism) (2): sound: codecs: gtm601: add Broadmobi bm818 sound profile dt-bindings: sound: gtm601: add the broadmobi interface .../devicetree/bindings/sound/gtm601.txt | 10 +++- sound/soc/codecs/gtm601.c | 46 +++++++++++++++---- 2 files changed, 46 insertions(+), 10 deletions(-) -- 2.17.1

3 4