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

[alsa-devel] [PATCH 0/3] ASoC: SOF: fix suspend ordering with runtime idle
by Kai Vehmanen 12 Jun '19

12 Jun '19

Hi all, this series addresses issues with ordering of HDA codec and controller suspends in the runtime PM flows. The implemented logic for SOF is similar to what has been used by the Intel AZX HDA driver. To implement this, first a fix is needed to hdac_hdmic ASoC codec driver. SOF framework also needs to be extended to allow SOF devices to implement a runtime_idle callback. Third, concrete implementation is in a separate patch for APL/CNL Intel hardware, for which strict ordering of codec-controller power down sequence needs to be maintained. As this extends the SOF device interface, Pierre asked me to send to the list for wider review. This series has been prereviewd at SOF github as: https://github.com/thesofproject/linux/pull/1003 Kai Vehmanen (3): ASoC: hdac_hdmi: report codec link up/down status to bus ASoC: SOF: add runtime idle callback ASoC: SOF: Intel: implement runtime idle for CNL/APL sound/soc/codecs/hdac_hdmi.c | 2 ++ sound/soc/sof/intel/apl.c | 1 + sound/soc/sof/intel/cnl.c | 1 + sound/soc/sof/intel/hda-dsp.c | 13 +++++++++++++ sound/soc/sof/intel/hda.h | 1 + sound/soc/sof/pm.c | 8 ++++++++ sound/soc/sof/sof-acpi-dev.c | 2 +- sound/soc/sof/sof-pci-dev.c | 2 +- sound/soc/sof/sof-priv.h | 2 ++ 9 files changed, 30 insertions(+), 2 deletions(-) -- 2.17.1

2 4

[alsa-devel] [PATCH] Revert "wcd9335: fix a incorrect use of kstrndup()"
by Ben Hutchings 12 Jun '19

12 Jun '19

This reverts commit a54988113985ca22e414e132054f234fc8a92604. The strings being duplicated are not fixed-length, so kstrndup() is the correct function to use. Signed-off-by: Ben Hutchings <ben.hutchings(a)codethink.co.uk> --- sound/soc/codecs/wcd9335.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/codecs/wcd9335.c b/sound/soc/codecs/wcd9335.c index 85737fe54474..a04a7cedd99d 100644 --- a/sound/soc/codecs/wcd9335.c +++ b/sound/soc/codecs/wcd9335.c @@ -2734,7 +2734,7 @@ static int wcd9335_codec_enable_dec(struct snd_soc_dapm_widget *w, char *dec; u8 hpf_coff_freq; - widget_name = kmemdup_nul(w->name, 15, GFP_KERNEL); + widget_name = kstrndup(w->name, 15, GFP_KERNEL); if (!widget_name) return -ENOMEM; -- Ben Hutchings, Software Developer Codethink Ltd https://www.codethink.co.uk/ Dale House, 35 Dale Street Manchester, M1 2HF, United Kingdom

2 2

[alsa-devel] [PATCH 00/19] ALSA: bebob/fireworks/oxfw: code refactoring toward rework for reservation of isochronous resources
by Takashi Sakamoto 12 Jun '19

12 Jun '19

Hi, This patchset is a part of patches to reserve/release isochronous resources in pcm.hw_params/hw_free callbacks, like posted patchsets below: https://mailman.alsa-project.org/pipermail/alsa-devel/2019-June/150118.html https://mailman.alsa-project.org/pipermail/alsa-devel/2019-June/150863.html However, in this patchset, I focus on code refactoring, due to kernel API implementation in firewire-lib. The target devices supported by ALSA bebob, fireworks and oxfw drivers implements Connection Management Procedure (CMP) in IEC 61883-1. For CMP, the drivers use kernel API in firewire-lib. This API has private data including the data for isochronous resources, and its implementation is tightly-coupled to maintenance of isochronous resources. This patchset is a preparation for future work to change the CMP implementation. Takashi Sakamoto (19): ALSA: bebob: configure sampling transfer frequency in pcm.hw_params callback ALSA: bebob: don't set XRUN in stop streaming ALSA: bebob: obsolete useless member of private structure ALSA: bebob: code refactoring to initialize/destroy stream data ALSA: fireworks: unify substream counter ALSA: fireworks: code refactoring for rawmidi.open/close ALSA: fireworks: code refactoring for pcm.hw_params/hw_free ALSA: fireworks: configure sampling transfer frequency in pcm.hw_params callback ALSA: fireworks: configure stream parameters in pcm.hw_params callback ALSA: fireworks: don't set XRUN in stop streaming ALSA: oxfw: code refactoring for stop condition of packet streaming ALSA: oxfw: set packet parameter according to current configuration ALSA: oxfw: start duplex streams if supported ALSA: oxfw: break packet streaming at bus-reset handler ALSA: oxfw: expand stop procedure for packet streaming ALSA: oxfw: rename helper functions for duplex streams ALSA: oxfw: unify substreams counter ALSA: oxfw: configure packet format in pcm.hw_params callback ALSA: oxfw: configure stream parameter in pcm.hw_params callback sound/firewire/bebob/bebob.h | 5 +- sound/firewire/bebob/bebob_midi.c | 22 +- sound/firewire/bebob/bebob_pcm.c | 22 +- sound/firewire/bebob/bebob_stream.c | 339 ++++++++++---------- sound/firewire/fireworks/fireworks.h | 6 +- sound/firewire/fireworks/fireworks_midi.c | 54 +--- sound/firewire/fireworks/fireworks_pcm.c | 59 +--- sound/firewire/fireworks/fireworks_stream.c | 159 +++++---- sound/firewire/oxfw/oxfw-midi.c | 22 +- sound/firewire/oxfw/oxfw-pcm.c | 34 +- sound/firewire/oxfw/oxfw-stream.c | 335 ++++++++++--------- sound/firewire/oxfw/oxfw.c | 17 +- sound/firewire/oxfw/oxfw.h | 22 +- 13 files changed, 553 insertions(+), 543 deletions(-) -- 2.20.1

2 20

[alsa-devel] [PATCH] ASoC: codec: hdac_hdmi: fix pin connections at cvt enable
by Kai Vehmanen 12 Jun '19

12 Jun '19

In display codecs supported by hdac_hdmi, the connection indices are shared by all converters. At boot and resume from suspend, the connection state may be reset to default values. In case of multiple connected pins (multiple monitors connected with audio capability), routing and mute status of pins that are not connected to any PCM, may interfere with other pins. E.g. after resume from S3 with multiple monitors, unless all converters are in active use, playback to some PCMs may be muted due to the default settings of unrelated converters. Avoid this by ensuring all pin:cvt selections are correct in codec whenever a converter is enabled for playback. Signed-off-by: Kai Vehmanen <kai.vehmanen(a)linux.intel.com> --- sound/soc/codecs/hdac_hdmi.c | 41 ++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) diff --git a/sound/soc/codecs/hdac_hdmi.c b/sound/soc/codecs/hdac_hdmi.c index 660e0587f3999..bb87240d2fb22 100644 --- a/sound/soc/codecs/hdac_hdmi.c +++ b/sound/soc/codecs/hdac_hdmi.c @@ -546,6 +546,39 @@ static struct hdac_hdmi_port *hdac_hdmi_get_port_from_cvt( return NULL; } +/** + * Go through all converters and ensure connection is set to + * the correct pin as set via kcontrols. + */ +static void hdac_hdmi_verify_connect_sel_all_pins(struct hdac_device *hdev) +{ + struct hdac_hdmi_port *port; + struct hdac_hdmi_priv *hdmi = hdev_to_hdmi_priv(hdev); + struct hdac_hdmi_cvt *cvt; + int cvt_idx = 0; + int curr; + + list_for_each_entry(cvt, &hdmi->cvt_list, head) { + port = hdac_hdmi_get_port_from_cvt(hdev, hdmi, cvt); + if (port && port->pin) { + curr = snd_hdac_codec_read(hdev, port->pin->nid, + 0, AC_VERB_GET_CONNECT_SEL, + 0); + if (curr != cvt_idx) { + snd_hdac_codec_write(hdev, + port->pin->nid, 0, + AC_VERB_SET_CONNECT_SEL, + cvt_idx); + dev_dbg(&hdev->dev, + "%s: %s set connect %d -> %d\n", + __func__, cvt->name, port->pin->nid, + cvt_idx); + } + } + ++cvt_idx; + } +} + /* * This tries to get a valid pin and set the HW constraints based on the * ELD. Even if a valid pin is not found return success so that device open @@ -806,6 +839,14 @@ static int hdac_hdmi_cvt_output_widget_event(struct snd_soc_dapm_widget *w, AC_VERB_SET_CHANNEL_STREAMID, pcm->stream_tag); snd_hdac_codec_write(hdev, cvt->nid, 0, AC_VERB_SET_STREAM_FORMAT, pcm->format); + + /** + * The connection indices are shared by all converters and + * may interfere with each other. Ensure correct + * routing for all converters at stream start. + */ + hdac_hdmi_verify_connect_sel_all_pins(hdev); + break; case SND_SOC_DAPM_POST_PMD: -- 2.17.1

2 2

[alsa-devel] [PATCH 0/6] alsa-lib: Various portability fixes
by Michael Forney 12 Jun '19

12 Jun '19

Even though alsa-lib doesn't aim for ISO C conformance, hopefully these changes are uncontroversial enough to be considered anyway. Michael Forney (6): Use __func__ instead of __FUNCTION__ List cases in range explicitly Don't return in a void function Make sure parameter qualifiers match between declaration and definition Remove unused empty struct Avoid pointer arithmetic on `void *` aserver/aserver.c | 4 ++-- include/error.h | 8 ++++---- include/local.h | 4 ++-- src/conf.c | 22 ++++++++++++---------- src/pcm/pcm.c | 16 ++++++++-------- src/pcm/pcm_direct.h | 2 -- src/pcm/pcm_hw.c | 2 +- src/pcm/pcm_rate.c | 8 ++++---- src/pcm/pcm_share.c | 4 ++-- 9 files changed, 35 insertions(+), 35 deletions(-) -- 2.20.1

2 7

[alsa-devel] [PATCH 00/10] Move part of cros-ec out of MFD subsystem
by Enric Balletbo i Serra 11 Jun '19

11 Jun '19

Hi, This is the first attempt to clean up a bit more the cros-ec drivers to have a better separation on what is part of the MFD subsystem and what is part of platform/chrome. It'd be really nice have some reviews, acks and tested on different platforms from the chromiumos people before merge all this patchset, as this moves a lot of code. The major changes introduced by this patchset are: 1. Move the core driver to platform/chrome, as is not really related to an MFD device driver. 2. Create a new misc chardev driver to replace the chardev bits from cros-ec-dev (MFD) 3. Added some convenience structs in cros-ec-dev (MFD) to easy add more subdrivers and avoid to add more boiler plate. Once applied we have moved all the code to platform/chrome except the cros-ec-dev driver, which is the one that instantiates the different subdrivers as cells of the MFD device. I tested the following patches on Veyron, Kevin, Samus, Peach Pi and Peach Pit without noticing any problem, but they would need a lot of more tests. I'll continue testing while the reviewing process of this patchset. Waiting for your feedback, Enric Enric Balletbo i Serra (10): mfd / platform: cros_ec: Handle chained ECs as platform devices mfd / platform: cros_ec: Move cros-ec core driver out from MFD mfd / platform: cros_ec: Miscellaneous character device to talk with the EC mfd: cros_ec: Switch to use the new cros-ec-chardev driver mfd / platform: cros_ec: Rename config to a better name mfd / platform: cros_ec: Reorganize platform and mfd includes mfd: cros_ec: Update with SPDX Licence identifier and fix description mfd: cros_ec: Use kzalloc and cros_ec_cmd_xfer_status helper mfd: cros_ec: Add convenience struct to define dedicated CrOS EC MCUs mfd: cros_ec: Add convenience struct to define autodetectable CrOS EC subdevices Documentation/ioctl/ioctl-number.txt | 2 +- drivers/extcon/Kconfig | 2 +- drivers/extcon/extcon-usbc-cros-ec.c | 3 +- drivers/hid/Kconfig | 2 +- drivers/hid/hid-google-hammer.c | 4 +- drivers/i2c/busses/Kconfig | 2 +- drivers/i2c/busses/i2c-cros-ec-tunnel.c | 4 +- drivers/iio/accel/cros_ec_accel_legacy.c | 3 +- drivers/iio/common/cros_ec_sensors/Kconfig | 2 +- .../common/cros_ec_sensors/cros_ec_sensors.c | 3 +- .../cros_ec_sensors/cros_ec_sensors_core.c | 3 +- drivers/iio/light/cros_ec_light_prox.c | 3 +- drivers/iio/pressure/cros_ec_baro.c | 3 +- drivers/input/keyboard/Kconfig | 2 +- drivers/input/keyboard/cros_ec_keyb.c | 4 +- drivers/media/platform/Kconfig | 3 +- .../media/platform/cros-ec-cec/cros-ec-cec.c | 4 +- drivers/mfd/Kconfig | 26 +- drivers/mfd/Makefile | 4 +- drivers/mfd/cros_ec_dev.c | 433 +++++------------- drivers/platform/chrome/Kconfig | 48 +- drivers/platform/chrome/Makefile | 2 + drivers/{mfd => platform/chrome}/cros_ec.c | 64 +-- drivers/platform/chrome/cros_ec_chardev.c | 279 +++++++++++ drivers/platform/chrome/cros_ec_debugfs.c | 3 +- drivers/platform/chrome/cros_ec_i2c.c | 12 +- drivers/platform/chrome/cros_ec_lightbar.c | 3 +- drivers/platform/chrome/cros_ec_lpc.c | 7 +- drivers/platform/chrome/cros_ec_lpc_reg.c | 4 +- drivers/platform/chrome/cros_ec_proto.c | 3 +- drivers/platform/chrome/cros_ec_rpmsg.c | 6 +- drivers/platform/chrome/cros_ec_spi.c | 12 +- drivers/platform/chrome/cros_ec_sysfs.c | 3 +- drivers/platform/chrome/cros_ec_trace.c | 2 +- drivers/platform/chrome/cros_ec_trace.h | 4 +- drivers/platform/chrome/cros_ec_vbc.c | 3 +- drivers/platform/chrome/cros_usbpd_logger.c | 5 +- drivers/power/supply/Kconfig | 2 +- drivers/power/supply/cros_usbpd-charger.c | 5 +- drivers/pwm/Kconfig | 2 +- drivers/pwm/pwm-cros-ec.c | 4 +- drivers/rtc/Kconfig | 2 +- drivers/rtc/rtc-cros-ec.c | 3 +- .../linux/iio/common/cros_ec_sensors_core.h | 3 +- include/linux/mfd/cros_ec.h | 302 +----------- .../{mfd => platform_data}/cros_ec_commands.h | 0 include/linux/platform_data/cros_ec_proto.h | 315 +++++++++++++ .../uapi/linux/cros_ec_chardev.h | 18 +- sound/soc/codecs/cros_ec_codec.c | 4 +- sound/soc/qcom/Kconfig | 2 +- 50 files changed, 902 insertions(+), 732 deletions(-) rename drivers/{mfd => platform/chrome}/cros_ec.c (85%) create mode 100644 drivers/platform/chrome/cros_ec_chardev.c rename include/linux/{mfd => platform_data}/cros_ec_commands.h (100%) create mode 100644 include/linux/platform_data/cros_ec_proto.h rename drivers/mfd/cros_ec_dev.h => include/uapi/linux/cros_ec_chardev.h (70%) -- 2.20.1

13 22

[alsa-devel] [PATCH] ALSA: HDA: fix a mask
by Guennadi Liakhovetski 11 Jun '19

11 Jun '19

Response tag only occupies 6 bits, not 7. This bug is harmless, since the value has just been shifted to the right by 26 bits, so this is only a cosmetic fix. Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski(a)linux.intel.com> --- sound/pci/hda/hda_jack.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/hda/hda_jack.c b/sound/pci/hda/hda_jack.c index 74b4695..60a5485 100644 --- a/sound/pci/hda/hda_jack.c +++ b/sound/pci/hda/hda_jack.c @@ -563,7 +563,7 @@ static void call_jack_callback(struct hda_codec *codec, unsigned int res, void snd_hda_jack_unsol_event(struct hda_codec *codec, unsigned int res) { struct hda_jack_tbl *event; - int tag = (res >> AC_UNSOL_RES_TAG_SHIFT) & 0x7f; + int tag = (res & AC_UNSOL_RES_TAG) >> AC_UNSOL_RES_TAG_SHIFT; event = snd_hda_jack_tbl_get_from_tag(codec, tag); if (!event) -- 1.9.3

2 1

[alsa-devel] [PATCH 0/2] pcm_file write file assert case handling
by Adam Miartus 11 Jun '19

11 Jun '19

patched possible case where failed file write can cause assert in pcm_file plugin, by reporting error EPIPE and printing debug message. reporting an error with debug hint message is preferred to assert, user may have a way to recover in upper API level. providing a debug message makes it easier to debug the issue from logs. previously, reaching assert did not provide useful message to API level user. Adam Miartus (2): pcm_file: do not disrupt playback on output file write fail pcm_file: report write output file error to api user src/pcm/pcm_file.c | 63 +++++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 46 insertions(+), 17 deletions(-) -- 2.7.4

4 5

[alsa-devel] [PATCH 00/12] ALSA: firewire-digi00x/dice: reserve/release isochronous resources in pcm.hw_params/hw_free callbacks
by Takashi Sakamoto 11 Jun '19

11 Jun '19

Hi, This patchset is a part of series of patches for all of drivers in ALSA firewire stack to reserve/release isochronous resources in pcm.hw_params/hw_free callbacks. In current implementation, the resources are reserved at the same time to start packet streaming, and released at the same time to stop packet streaming. However, once allocated, the resources are available independent of lifetime of each of packet streaming. The isochronous resources are the resources of IEEE 1394 bus. On the other side of view, it's a kind of resources of hardware to maintain the bus (isochronous resource manager). For this kind of reservation and release, hw_params and hw_free operations are suitable in ALSA PCM interface. Ideally, the operation to reserve/release isochronous resource should be separated from the operation to start/stop packet streaming. However, IEEE 1394 bus has reset event. Once reset occurs, isochronous resource manager releases allocated resources. The resources should be reallocated by requesters themselves. For this reason, in this patchset, bus generation is checked before starting packet streaming. If generation is updated, reallocation is requested to isochronous resource manager, then packet streaming starts. Takashi Sakamoto (12): ALSA: firewire-digi00x: refactoring to move timing of registration for isochronous channel ALSA: firewire-digi00x: code refactoring to finish streaming session ALSA: firewire-digi00x: simplify error path to begin streaming session ALSA: firewire-digi00x: code refactoring to keep isochronous resources ALSA: firewire-digi00x: reserve/release isochronous resources in pcm.hw_params/hw_free callbacks ALSA: firewire-digi00x: update isochronous resources when starting packet streaming after bus-reset ALSA: firewire-digi00x: code refactoring for pcm.hw_params/hw_free callbacks ALSA: dice: code refactoring to stop packet streaming ALSA: dice: code refactoring to keep isochronous resources ALSA: dice: reserve/release isochronous resources in pcm.hw_params/hw_free callbacks ALSA: dice: update isochronous resources when starting packet streaming after bus-reset ALSA: dice: code refactoring for pcm.hw_params/hw_free callbacks sound/firewire/dice/dice-midi.c | 10 +- sound/firewire/dice/dice-pcm.c | 62 ++--- sound/firewire/dice/dice-stream.c | 350 +++++++++++++----------- sound/firewire/dice/dice.h | 4 +- sound/firewire/digi00x/digi00x-midi.c | 10 +- sound/firewire/digi00x/digi00x-pcm.c | 65 ++--- sound/firewire/digi00x/digi00x-stream.c | 189 ++++++------- sound/firewire/digi00x/digi00x.h | 4 +- 8 files changed, 342 insertions(+), 352 deletions(-) -- 2.20.1

2 13

[alsa-devel] [PATCH] ALSA: usb-audio: Enable .product_name override for Emagic, Unitor 8.
by Stefan Sauer 11 Jun '19

11 Jun '19

>From a41303935608e9a98653d4dff1e10baf1ce07e05 Mon Sep 17 00:00:00 2001 From: Stefan Sauer <ensonic(a)google.com> Date: Mon, 10 Jun 2019 11:26:53 +0200 Subject: [PATCH] ALSA: usb-audio: Enable .product_name override for Emagic Unitor 8. The Emagic Unitor 8 does not provide iManufacturer and iProduct fields in its device descriptor. These fields are used by alsa to make build the device name. Thus uncomment the .product-name in the quirks-table. Without this change the device shows up as 'USB Device 0x86a:0x01'. Output of lsusb and amidi: https://gist.github.com/ensonic/7820a102e91f31575be355da2b6b33bc Signed-off-by: Stefan Sauer <ensonic(a)google.com> --- sound/usb/quirks-table.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/usb/quirks-table.h b/sound/usb/quirks-table.h index 9e049f60e80e..e918ce346027 100644 --- a/sound/usb/quirks-table.h +++ b/sound/usb/quirks-table.h @@ -2408,7 +2408,7 @@ YAMAHA_DEVICE(0x7010, "UB99"), USB_DEVICE(0x086a, 0x0001), .driver_info = (unsigned long) & (const struct snd_usb_audio_quirk) { .vendor_name = "Emagic", - /* .product_name = "Unitor8", */ + .product_name = "Unitor8", .ifnum = 2, .type = QUIRK_MIDI_EMAGIC, .data = & (const struct snd_usb_midi_endpoint_info) { -- 2.21.0

2 1