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

[PATCH 0/7] ALSA: add virtio sound driver
by Anton Yakovlev 24 Jan '21

24 Jan '21

This series implements a driver part of the virtio sound device specification v8 [1]. The driver supports PCM playback and capture substreams, jack and channel map controls. A message-based transport is used to write/read PCM frames to/from a device. The series is based (and was actually tested) on Linus's master branch [2], on top of commit 1e2a199f6ccd ("Merge tag 'spi-fix-v5.11-rc4' of ...") As a device part was used OpenSynergy proprietary implementation. Any comments are very welcome. [1] https://lists.oasis-open.org/archives/virtio-dev/202003/msg00185.html [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git Anton Yakovlev (7): uapi: virtio_ids: add a sound device type ID from OASIS spec uapi: virtio_snd: add the sound device header file ALSA: virtio: add virtio sound driver ALSA: virtio: handling control and I/O messages for the PCM device ALSA: virtio: PCM substream operators ALSA: virtio: introduce jack support ALSA: virtio: introduce device suspend/resume support MAINTAINERS | 8 + include/uapi/linux/virtio_ids.h | 1 + include/uapi/linux/virtio_snd.h | 361 +++++++++++++++++++ sound/Kconfig | 2 + sound/Makefile | 3 +- sound/virtio/Kconfig | 10 + sound/virtio/Makefile | 13 + sound/virtio/virtio_card.c | 593 ++++++++++++++++++++++++++++++++ sound/virtio/virtio_card.h | 121 +++++++ sound/virtio/virtio_chmap.c | 237 +++++++++++++ sound/virtio/virtio_ctl_msg.c | 293 ++++++++++++++++ sound/virtio/virtio_ctl_msg.h | 122 +++++++ sound/virtio/virtio_jack.c | 255 ++++++++++++++ sound/virtio/virtio_pcm.c | 582 +++++++++++++++++++++++++++++++ sound/virtio/virtio_pcm.h | 132 +++++++ sound/virtio/virtio_pcm_msg.c | 317 +++++++++++++++++ sound/virtio/virtio_pcm_ops.c | 524 ++++++++++++++++++++++++++++ 17 files changed, 3573 insertions(+), 1 deletion(-) create mode 100644 include/uapi/linux/virtio_snd.h create mode 100644 sound/virtio/Kconfig create mode 100644 sound/virtio/Makefile create mode 100644 sound/virtio/virtio_card.c create mode 100644 sound/virtio/virtio_card.h create mode 100644 sound/virtio/virtio_chmap.c create mode 100644 sound/virtio/virtio_ctl_msg.c create mode 100644 sound/virtio/virtio_ctl_msg.h create mode 100644 sound/virtio/virtio_jack.c create mode 100644 sound/virtio/virtio_pcm.c create mode 100644 sound/virtio/virtio_pcm.h create mode 100644 sound/virtio/virtio_pcm_msg.c create mode 100644 sound/virtio/virtio_pcm_ops.c -- 2.30.0

3 19

jack: use program_invocation_short_name
by GitHub pull_request - opened 24 Jan '21

24 Jan '21

alsa-project/alsa-plugins pull request #18 was opened from ArsenArsen: this results in more identifiable names in jack patchbays and control programs Request URL : https://github.com/alsa-project/alsa-plugins/pull/18 Patch URL : https://github.com/alsa-project/alsa-plugins/pull/18.patch Repository URL: https://github.com/alsa-project/alsa-plugins

1 0

[PATCH] ALSA: usb-audio: Re-add sample rate quirk for Pioneer DJM-750
by Olivia Mackintosh 24 Jan '21

24 Jan '21

Re-add the sample-rate quirk originally implemented by Dmitry Panchenko for the DJM-900NXS2 but for the DJM-750. This commit only adds it for the DJM-750 since I can only verify that it is absolutely required for this device. Other models may need this patch but I'm hesitant to add them as I cannot test. The 'wIndex' is taken from `fmt->sync_ep` as different devices have a different sync endpoint which is specified in 'quirks-table.h'. Adding other devices should only require adding an additional case. Signed-off-by: Olivia Mackintosh <livvy(a)base.nu> --- sound/usb/quirks.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index e196e364cef1..517346054fb7 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -1470,6 +1470,22 @@ static void set_format_emu_quirk(struct snd_usb_substream *subs, subs->pkt_offset_adj = (emu_samplerate_id >= EMU_QUIRK_SR_176400HZ) ? 4 : 0; } +static int pioneer_djm_set_format_quirk(struct snd_usb_substream *subs, + const struct audioformat *fmt) +{ + unsigned int cur_rate = subs->data_endpoint->cur_rate; + u8 sr[3]; + // Convert to little endian + sr[0] = cur_rate&0xff; + sr[1] = (cur_rate>>8)&0xff; + sr[2] = (cur_rate>>16)&0xff; + usb_set_interface(subs->dev, 0, 1); + snd_usb_ctl_msg(subs->stream->chip->dev, + usb_rcvctrlpipe(subs->stream->chip->dev, 0), + 0x01, 0x22, 0x0100, fmt->sync_ep, &sr, 0x0003); + return 0; +} + void snd_usb_set_format_quirk(struct snd_usb_substream *subs, const struct audioformat *fmt) { @@ -1483,6 +1499,9 @@ void snd_usb_set_format_quirk(struct snd_usb_substream *subs, case USB_ID(0x534d, 0x2109): /* MacroSilicon MS2109 */ subs->stream_offset_adj = 2; break; + case USB_ID(0x08e4, 0x017f): /* Pioneer DJM-750 */ + pioneer_djm_set_format_quirk(subs, fmt); + break; } } -- 2.30.0

1 0

[RFC][PATCH v6 0/1] audio jack software injection
by Hui Wang 24 Jan '21

24 Jan '21

the changes in the v6: - use the sizeof(buf) to replace the digits in the scnprintf - squash the document patch into the 0001-xxx.patch. the changes in the v5: - add a Kconfig to enable/disable the jack injection feature - replace all kzalloc with local char[] in the debugfs fops - replace the chars of !isalnum() to '_' for the jack folder's name - squash all .c files patches into one patch - add a document to explain jack injection, for easier review, put the document in a standalone patch. If needed, will squash this patch to the previous one. the changes in the v4: - change the sound-core to sound and change the sound_core_debugfs_root to sound_debugfs_root in the 0001-xxx.patch - change kzalloc/kfree to kvzalloc/kvfree in the debugfs fops for 0001-xxx.patch and 0003-xxx.patch - And if needed, I could squash 4 patches into 1 patch before merging. the changes in the v3 (for easy to review, divide change into 4 patches): - address the comment about the snd_jack_report() and _snd_jack_report(), the v2 design is hard to understand and is hard to review, in the v3, separate the jack_report to snd_jack_report() and snd_jack_inject_report(), hw jack events call snd_jack_report() as before, if a jack contains multi jack_kctl and the jack_kctl's sw_inject is enabled, the status and the related input-dev's events will not be reproted. The injection events call snd_jack_inject_report(). This change is squashed to 0001-xxx.patch, the rest part of 0001-xxx.patch is same as v2. - address the comment about folders'name in the 0002-xxx.patch, so far, drop the '/', ',', '=' and ' ' from the folders' name. - address the comment about adding more debugfs nodes in the 0003-xxx.patch, it adds kctl_id, mask_bits, status and type. - address the comment about save-n-restore jack's hw status in the 0004-xxx.patch, adding a hw_status_cache and save the last reported jack hw event, once the sw_inject is disabled, will restore all jack_kctl's state under the same snd_jack with hw_status_cache. [snip] the changes in the V2: - using debugfs instead of sysfs - using jack_ctrl to create a folder instead of snd_jack, since ASoC drivers could create multi jack_ctrls within a snd_jack - create a folder for each jack_ctrl instead for all jack_ctrls [ snip ] Hui Wang (1): alsa: jack: implement software jack injection via debugfs Documentation/sound/designs/index.rst | 1 + .../sound/designs/jack-injection.rst | 124 ++++++++ include/sound/core.h | 2 + include/sound/jack.h | 1 + sound/core/Kconfig | 9 + sound/core/init.c | 10 + sound/core/jack.c | 298 +++++++++++++++++- sound/core/sound.c | 8 + 8 files changed, 449 insertions(+), 4 deletions(-) create mode 100644 Documentation/sound/designs/jack-injection.rst -- 2.25.1

2 3

[PATCH] sound/pci/hda: boolean values to a bool variable
by Jiapeng Zhong 24 Jan '21

24 Jan '21

Fix the following coccicheck warnings: ./sound/pci/hda/patch_conexant.c:570:2-20: WARNING: Assignment of 0/1 to bool variable. Reported-by: Abaci Robot <abaci(a)linux.alibaba.com> Signed-off-by: Jiapeng Zhong <abaci-bugfix(a)linux.alibaba.com> --- sound/pci/hda/patch_conexant.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c index d49cc44..f2aa226 100644 --- a/sound/pci/hda/patch_conexant.c +++ b/sound/pci/hda/patch_conexant.c @@ -567,7 +567,7 @@ static void cxt_fixup_mute_led_eapd(struct hda_codec *codec, if (action == HDA_FIXUP_ACT_PRE_PROBE) { spec->mute_led_eapd = 0x1b; - spec->dynamic_eapd = 1; + spec->dynamic_eapd = true; snd_hda_gen_add_mute_led_cdev(codec, cx_auto_vmaster_mute_led); } } -- 1.8.3.1

2 1

ASoC: cpcap: Implement set_tdm_slot for voice call support
by Pavel Machek 23 Jan '21

23 Jan '21

From: Tony Lindgren <tony(a)atomide.com> ASoC: cpcap: Implement set_tdm_slot for voice call support For using cpcap for voice calls, we need to route audio directly from the modem to cpcap for TDM (Time Division Multiplexing). The voice call is direct data between the modem and cpcap with no CPU involvment. In this mode, the cpcap related audio mixer controls work for the speaker selection and volume though. To do this, we need to implement standard snd_soc_dai_set_tdm_slot() for cpcap. Then the modem codec driver can use snd_soc_dai_set_sysclk(), snd_soc_dai_set_fmt(), and snd_soc_dai_set_tdm_slot() to configure a voice call. Let's add cpcap_voice_set_tdm_slot() for this, and cpcap_voice_call() helper to configure the additional registers needed for voice call. Let's also clear CPCAP_REG_VAUDIOC on init in case we have the bit for CPCAP_BIT_VAUDIO_MODE0 set on init. Signed-off-by: Tony Lindgren <tony(a)atomide.com> Signed-off-by: Pavel Machek <pavel(a)ucw.cz> diff --git a/sound/soc/codecs/cpcap.c b/sound/soc/codecs/cpcap.c index f046987ee4cd..e266d993ab2a 100644 --- a/sound/soc/codecs/cpcap.c +++ b/sound/soc/codecs/cpcap.c @@ -16,6 +16,14 @@ #include <sound/soc.h> #include <sound/tlv.h> +/* Register 512 CPCAP_REG_VAUDIOC --- Audio Regulator and Bias Voltage */ +#define CPCAP_BIT_AUDIO_LOW_PWR 6 +#define CPCAP_BIT_AUD_LOWPWR_SPEED 5 +#define CPCAP_BIT_VAUDIOPRISTBY 4 +#define CPCAP_BIT_VAUDIO_MODE1 2 +#define CPCAP_BIT_VAUDIO_MODE0 1 +#define CPCAP_BIT_V_AUDIO_EN 0 + /* Register 513 CPCAP_REG_CC --- CODEC */ #define CPCAP_BIT_CDC_CLK2 15 #define CPCAP_BIT_CDC_CLK1 14 @@ -221,6 +229,7 @@ struct cpcap_reg_info { }; static const struct cpcap_reg_info cpcap_default_regs[] = { + { CPCAP_REG_VAUDIOC, 0x003F, 0x0000 }, { CPCAP_REG_CC, 0xFFFF, 0x0000 }, { CPCAP_REG_CC, 0xFFFF, 0x0000 }, { CPCAP_REG_CDI, 0xBFFF, 0x0000 }, @@ -1371,8 +1380,121 @@ static int cpcap_voice_set_dai_fmt(struct snd_soc_dai *codec_dai, return 0; } -static int cpcap_voice_set_mute(struct snd_soc_dai *dai, - int mute, int direction) + +/* + * Configure codec for voice call if requested. + * + * We can configure most with snd_soc_dai_set_sysclk(), snd_soc_dai_set_fmt() + * and snd_soc_dai_set_tdm_slot(). This function configures the rest of the + * cpcap related hardware as CPU is not involved in the voice call. + */ +static int cpcap_voice_call(struct cpcap_audio *cpcap, struct snd_soc_dai *dai, + bool voice_call) +{ + int mask, err; + + /* Modem to codec VAUDIO_MODE1 */ + mask = BIT(CPCAP_BIT_VAUDIO_MODE1); + err = regmap_update_bits(cpcap->regmap, CPCAP_REG_VAUDIOC, + mask, voice_call ? mask : 0); + if (err) + return err; + + /* Clear MIC1_MUX for call */ + mask = BIT(CPCAP_BIT_MIC1_MUX); + err = regmap_update_bits(cpcap->regmap, CPCAP_REG_TXI, + mask, voice_call ? 0 : mask); + if (err) + return err; + + /* Set MIC2_MUX for call */ + mask = BIT(CPCAP_BIT_MB_ON1L) | BIT(CPCAP_BIT_MB_ON1R) | + BIT(CPCAP_BIT_MIC2_MUX) | BIT(CPCAP_BIT_MIC2_PGA_EN); + err = regmap_update_bits(cpcap->regmap, CPCAP_REG_TXI, + mask, voice_call ? mask : 0); + if (err) + return err; + + /* Enable LDSP for call */ + mask = BIT(CPCAP_BIT_A2_LDSP_L_EN) | BIT(CPCAP_BIT_A2_LDSP_R_EN); + err = regmap_update_bits(cpcap->regmap, CPCAP_REG_RXOA, + mask, voice_call ? mask : 0); + if (err) + return err; + + /* Enable CPCAP_BIT_PGA_CDC_EN for call */ + mask = BIT(CPCAP_BIT_PGA_CDC_EN); + err = regmap_update_bits(cpcap->regmap, CPCAP_REG_RXCOA, + mask, voice_call ? mask : 0); + if (err) + return err; + + /* Unmute voice for call */ + if (dai) { + err = snd_soc_dai_digital_mute(dai, !voice_call, + SNDRV_PCM_STREAM_PLAYBACK); + if (err) + return err; + } + + /* Set modem to codec mic CDC and HPF for call */ + mask = BIT(CPCAP_BIT_MIC2_CDC_EN) | BIT(CPCAP_BIT_CDC_EN_RX) | + BIT(CPCAP_BIT_AUDOHPF_1) | BIT(CPCAP_BIT_AUDOHPF_0) | + BIT(CPCAP_BIT_AUDIHPF_1) | BIT(CPCAP_BIT_AUDIHPF_0); + err = regmap_update_bits(cpcap->regmap, CPCAP_REG_CC, + mask, voice_call ? mask : 0); + if (err) + return err; + + /* Enable modem to codec CDC for call*/ + mask = BIT(CPCAP_BIT_CDC_CLK_EN); + err = regmap_update_bits(cpcap->regmap, CPCAP_REG_CDI, + mask, voice_call ? mask : 0); + + return err; +} + +static int cpcap_voice_set_tdm_slot(struct snd_soc_dai *dai, + unsigned int tx_mask, unsigned int rx_mask, + int slots, int slot_width) +{ + struct snd_soc_component *component = dai->component; + struct cpcap_audio *cpcap = snd_soc_component_get_drvdata(component); + int err, ts_mask, mask; + bool voice_call; + + /* + * Primitive test for voice call, probably needs more checks + * later on for 16-bit calls detected, Bluetooth headset etc. + */ + if (tx_mask == 0 && rx_mask == 1 && slot_width == 8) + voice_call = true; + else + voice_call = false; + + ts_mask = 0x7 << CPCAP_BIT_MIC2_TIMESLOT0; + ts_mask |= 0x7 << CPCAP_BIT_MIC1_RX_TIMESLOT0; + + mask = (tx_mask & 0x7) << CPCAP_BIT_MIC2_TIMESLOT0; + mask |= (rx_mask & 0x7) << CPCAP_BIT_MIC1_RX_TIMESLOT0; + + err = regmap_update_bits(cpcap->regmap, CPCAP_REG_CDI, + ts_mask, mask); + if (err) + return err; + + err = cpcap_set_samprate(cpcap, CPCAP_DAI_VOICE, slot_width * 1000); + if (err) + return err; + + err = cpcap_voice_call(cpcap, dai, voice_call); + if (err) + return err; + + return 0; +} + +static int cpcap_voice_set_mute(struct snd_soc_dai *dai, int mute, int direction) { struct snd_soc_component *component = dai->component; struct cpcap_audio *cpcap = snd_soc_component_get_drvdata(component); @@ -1393,6 +1515,7 @@ static const struct snd_soc_dai_ops cpcap_dai_voice_ops = { .hw_params = cpcap_voice_hw_params, .set_sysclk = cpcap_voice_set_dai_sysclk, .set_fmt = cpcap_voice_set_dai_fmt, + .set_tdm_slot = cpcap_voice_set_tdm_slot, .mute_stream = cpcap_voice_set_mute, .no_capture_mute = 1, }; -- http://www.livejournal.com/~pavelmachek

2 2

[PATCH alsa-ucm-conf 1/3] chtnau8824: Move DAC Channel Source selection to Speaker/Headphones EnableSeq
by Hans de Goede 23 Jan '21

23 Jan '21

Move DAC Channel Source selection to Speaker/Headphones EnableSeq. The main reason for doing this is to make it easier to allow using variables, as the Speaker/Headphones conf files are evaluated after the main HiFi.conf had a chance to set them. Signed-off-by: Hans de Goede <hdegoede(a)redhat.com> --- ucm2/codecs/nau8824/EnableSeq.conf | 3 --- ucm2/codecs/nau8824/HeadPhones.conf | 2 ++ ucm2/codecs/nau8824/MonoSpeaker.conf | 2 ++ ucm2/codecs/nau8824/Speaker.conf | 2 ++ 4 files changed, 6 insertions(+), 3 deletions(-) diff --git a/ucm2/codecs/nau8824/EnableSeq.conf b/ucm2/codecs/nau8824/EnableSeq.conf index 3650d06..2595e5f 100644 --- a/ucm2/codecs/nau8824/EnableSeq.conf +++ b/ucm2/codecs/nau8824/EnableSeq.conf @@ -1,7 +1,4 @@ EnableSequence [ - # Playback TDM configuration - cset "name='DACL Channel Source' 0" - cset "name='DACR Channel Source' 1" # Input Configuration cset "name='DMIC1 Enable Switch' off" cset "name='DMIC2 Enable Switch' off" diff --git a/ucm2/codecs/nau8824/HeadPhones.conf b/ucm2/codecs/nau8824/HeadPhones.conf index a807f84..46ec5d7 100644 --- a/ucm2/codecs/nau8824/HeadPhones.conf +++ b/ucm2/codecs/nau8824/HeadPhones.conf @@ -12,6 +12,8 @@ SectionDevice."Headphones" { ] EnableSequence [ + cset "name='DACL Channel Source' 0" + cset "name='DACR Channel Source' 1" cset "name='Headphone Switch' on" ] diff --git a/ucm2/codecs/nau8824/MonoSpeaker.conf b/ucm2/codecs/nau8824/MonoSpeaker.conf index 6b4ef8a..09edea5 100644 --- a/ucm2/codecs/nau8824/MonoSpeaker.conf +++ b/ucm2/codecs/nau8824/MonoSpeaker.conf @@ -11,6 +11,8 @@ SectionDevice."Speaker" { ] EnableSequence [ + cset "name='DACL Channel Source' 0" + cset "name='DACR Channel Source' 1" # nau8824 mono speaker boards have the speaker on the right chan cset "name='Speaker Left DACL Volume' 0" cset "name='Speaker Left DACR Volume' 0" diff --git a/ucm2/codecs/nau8824/Speaker.conf b/ucm2/codecs/nau8824/Speaker.conf index a995873..64d7eee 100644 --- a/ucm2/codecs/nau8824/Speaker.conf +++ b/ucm2/codecs/nau8824/Speaker.conf @@ -11,6 +11,8 @@ SectionDevice."Speaker" { ] EnableSequence [ + cset "name='DACL Channel Source' 0" + cset "name='DACR Channel Source' 1" cset "name='Speaker Left DACL Volume' 1" cset "name='Speaker Left DACR Volume' 0" cset "name='Speaker Right DACL Volume' 0" -- 2.28.0

2 6

[PATCH] ALSA: usb-audio: workaround for iface reset issue
by Takashi Iwai 23 Jan '21

23 Jan '21

The recently introduced sample rate validation code seems causing a problem on some devices; namely, after performing this, the bus gets screwed and it influences even on other USB devices. As a quick workaround, perform it only for the necessary devices; currently MOTU devices are known to need the valid altset checks, so filter out other devices. Fixes: 93db51d06b32 ("ALSA: usb-audio: Check valid altsetting at parsing rates for UAC2/3") Reported-by: Jamie Heilman <jamie(a)audible.transient.net> BugLink: https://bugzilla.suse.com/show_bug.cgi?id=1178203 Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/usb/format.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/sound/usb/format.c b/sound/usb/format.c index 9ebc5d202c87..e6ff317a6785 100644 --- a/sound/usb/format.c +++ b/sound/usb/format.c @@ -466,6 +466,17 @@ static int validate_sample_rate_table_v2v3(struct snd_usb_audio *chip, unsigned int nr_rates; int i, err; + /* performing the rate verification may lead to unexpected USB bus + * behavior afterwards by some unknown reason. Do this only for the + * known devices. + */ + switch (USB_ID_VENDOR(chip->usb_id)) { + case 0x07fd: /* MOTU */ + break; + default: + return 0; /* don't perform the validation as default */ + } + table = kcalloc(fp->nr_rates, sizeof(*table), GFP_KERNEL); if (!table) return -ENOMEM; -- 2.26.2

1 0

[PATCH] ALSA: pcm: One more dependency for hw constraints
by Takashi Iwai 23 Jan '21

23 Jan '21

The fix for a long-standing USB-audio bug required one more dependency variable to be added to the hw constraints. Unfortunately I didn't realize at debugging that the new addition may result in the overflow of the dependency array of each snd_pcm_hw_rule (up to three plus a sentinel), because USB-audio driver adds one more dependency only for a certain device and bus, hence it works as is for many devices. But in a bad case, a simple open always results in -EINVAL (with kernel WARNING if CONFIG_SND_DEBUG is set) no matter what is passed. Since the dependencies are real and unavoidable (USB-audio restricts the hw_params per looping over the format/rate/channels combos), the only good solution seems to raise the bar for one more dependency for snd_pcm_hw_rule -- so does this patch: now the hw constraint dependencies can be up to four. Fixes: 506c203cc3de ("ALSA: usb-audio: Fix hw constraints dependencies") Reported-by: Jamie Heilman <jamie(a)audible.transient.net> Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- include/sound/pcm.h | 2 +- sound/core/pcm_native.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/include/sound/pcm.h b/include/sound/pcm.h index 2336bf9243e1..2e1200d17d0c 100644 --- a/include/sound/pcm.h +++ b/include/sound/pcm.h @@ -229,7 +229,7 @@ typedef int (*snd_pcm_hw_rule_func_t)(struct snd_pcm_hw_params *params, struct snd_pcm_hw_rule { unsigned int cond; int var; - int deps[4]; + int deps[5]; snd_pcm_hw_rule_func_t func; void *private; diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c index 9f3f8e953ff0..c4aac703dc22 100644 --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -382,8 +382,8 @@ static int constrain_params_by_rules(struct snd_pcm_substream *substream, continue; /* - * The 'deps' array includes maximum three dependencies - * to SNDRV_PCM_HW_PARAM_XXXs for this rule. The fourth + * The 'deps' array includes maximum four dependencies + * to SNDRV_PCM_HW_PARAM_XXXs for this rule. The fifth * member of this array is a sentinel and should be * negative value. * -- 2.26.2

1 0

Question about hdac_ext_link ref_count management
by Ranjani Sridharan 23 Jan '21

23 Jan '21

Hi Takashi, While exploring some power optimizations on Intel platforms, I noticed that the hdac_ext_link ref_count is incremented during codec probe in hdac_hda_codec_probe() and the ref_count is held until the codec device is removed. I was wondering if it would be possible to call the get/put for the hdac_ext_link in the codec runtime suspend/resume callbacks so that the link is powered up only when the link is in use. Are there any downsides to doing this? Thanks, Ranjani

2 5