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

[PATCH 3/6] apr: make remove callback of apr driver void returned
by Dawei Li 05 Dec '22

05 Dec '22

Since commit fc7a6209d571 ("bus: Make remove callback return void") forces bus_type::remove be void-returned, it doesn't make much sense for any bus based driver implementing remove callbalk to return non-void to its caller. This change is for apr bus based drivers. Signed-off-by: Dawei Li <set_pte_at(a)outlook.com> --- include/linux/soc/qcom/apr.h | 2 +- sound/soc/qcom/qdsp6/q6core.c | 4 +--- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/include/linux/soc/qcom/apr.h b/include/linux/soc/qcom/apr.h index 23c5b30f3511..be98aebcb3e1 100644 --- a/include/linux/soc/qcom/apr.h +++ b/include/linux/soc/qcom/apr.h @@ -153,7 +153,7 @@ typedef struct apr_device gpr_device_t; struct apr_driver { int (*probe)(struct apr_device *sl); - int (*remove)(struct apr_device *sl); + void (*remove)(struct apr_device *sl); int (*callback)(struct apr_device *a, struct apr_resp_pkt *d); int (*gpr_callback)(struct gpr_resp_pkt *d, void *data, int op); diff --git a/sound/soc/qcom/qdsp6/q6core.c b/sound/soc/qcom/qdsp6/q6core.c index 5358fefd4210..49cfb32cd209 100644 --- a/sound/soc/qcom/qdsp6/q6core.c +++ b/sound/soc/qcom/qdsp6/q6core.c @@ -339,7 +339,7 @@ static int q6core_probe(struct apr_device *adev) return 0; } -static int q6core_exit(struct apr_device *adev) +static void q6core_exit(struct apr_device *adev) { struct q6core *core = dev_get_drvdata(&adev->dev); @@ -350,8 +350,6 @@ static int q6core_exit(struct apr_device *adev) g_core = NULL; kfree(core); - - return 0; } #ifdef CONFIG_OF -- 2.25.1

1 0

[PATCH] ASoC: soc-compress: Reposition and add pcm_mutex
by 강신형 05 Dec '22

05 Dec '22

>From 3f8fd1efedd77331d948b16a8e44ffa6cf0e811f Mon Sep 17 00:00:00 2001 From: "s47.kang" <s47.kang(a)samsung.com> Date: Fri, 2 Dec 2022 18:32:35 +0900 Subject: [PATCH] [COMMON] ASoC: soc-compress.c: Reposition and add pcm_mutex. If panic_on_warn is set and compress stream(DPCM) is started, then kernel panic will occurs because card->pcm_mutex isn't held appropriately in soc_compr_set_params_fe, soc_compr_open_fe and soc_compr_free_fe. So, I reposition and add pcm_mutex to resolve lockdep error. Change-Id: I15981b4f09b0bff37e7e98e6ee1f809339b2bcc4 --- sound/soc/soc-compress.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/sound/soc/soc-compress.c b/sound/soc/soc-compress.c index 870f13e1d389..7bce5088b455 100644 --- a/sound/soc/soc-compress.c +++ b/sound/soc/soc-compress.c @@ -149,6 +149,8 @@ static int soc_compr_open_fe(struct snd_compr_stream *cstream) if (ret < 0) goto be_err; + mutex_lock_nested(&fe->card->pcm_mutex, fe->card->pcm_subclass); + /* calculate valid and active FE <-> BE dpcms */ dpcm_process_paths(fe, stream, &list, 1); fe->dpcm[stream].runtime = fe_substream->runtime; @@ -184,7 +186,6 @@ static int soc_compr_open_fe(struct snd_compr_stream *cstream) fe->dpcm[stream].state = SND_SOC_DPCM_STATE_OPEN; fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO; - mutex_lock_nested(&fe->card->pcm_mutex, fe->card->pcm_subclass); snd_soc_runtime_activate(fe, stream); mutex_unlock(&fe->card->pcm_mutex); @@ -215,7 +216,6 @@ static int soc_compr_free_fe(struct snd_compr_stream *cstream) mutex_lock_nested(&fe->card->pcm_mutex, fe->card->pcm_subclass); snd_soc_runtime_deactivate(fe, stream); - mutex_unlock(&fe->card->pcm_mutex); fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_FE; @@ -234,6 +234,8 @@ static int soc_compr_free_fe(struct snd_compr_stream *cstream) dpcm_be_disconnect(fe, stream); + mutex_unlock(&fe->card->pcm_mutex); + fe->dpcm[stream].runtime = NULL; snd_soc_link_compr_shutdown(cstream, 0); @@ -409,8 +411,9 @@ static int soc_compr_set_params_fe(struct snd_compr_stream *cstream, ret = snd_soc_link_compr_set_params(cstream); if (ret < 0) goto out; - + mutex_lock_nested(&fe->card->pcm_mutex, fe->card->pcm_subclass); dpcm_dapm_stream_event(fe, stream, SND_SOC_DAPM_STREAM_START); + mutex_unlock(&fe->card->pcm_mutex); fe->dpcm[stream].state = SND_SOC_DPCM_STATE_PREPARE; out: -- 2.21.0

2 1

[PATCH 0/3] ALSA: Fix XRUN handling of USB-audio implicit fb
by Takashi Iwai 05 Dec '22

05 Dec '22

Hi, this is a patch set to cover a corner-case bug that happens with the USB-audio implicit feedback mode. When XRUN happens in the implicit fb full-duplex streams, the driver may get EPIPE error at prepare but re-issuing fails again. The essential fix is the 3rd patch but the former two are covering other corner cases I spotted (only theoretical). Takashi === Takashi Iwai (3): ALSA: pcm: Set missing stop_operating flag at undoing trigger start ALSA: pcm: Handle XRUN at trigger START ALSA: usb-audio: Workaround for XRUN at prepare sound/core/pcm_native.c | 12 ++++++++++-- sound/usb/pcm.c | 13 +++++++++++-- 2 files changed, 21 insertions(+), 4 deletions(-) -- 2.35.3

1 3

[PATCH v8 0/3] ASoC: SOF: Fix deadlock when shutdown a frozen userspace
by Ricardo Ribalda 05 Dec '22

05 Dec '22

Since: 83bfc7e793b5 ("ASoC: SOF: core: unregister clients and machine drivers in .shutdown") we wait for all the workloads to be completed during shutdown. This was done to avoid a stall once the device is started again. Unfortunately this has the side effect of stalling kexec(), if the userspace is frozen. Let's handle that case. To: Joel Fernandes <joel(a)joelfernandes.org> To: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> To: Liam Girdwood <lgirdwood(a)gmail.com> To: Peter Ujfalusi <peter.ujfalusi(a)linux.intel.com> To: Bard Liao <yung-chuan.liao(a)linux.intel.com> To: Ranjani Sridharan <ranjani.sridharan(a)linux.intel.com> To: Kai Vehmanen <kai.vehmanen(a)linux.intel.com> To: Daniel Baluta <daniel.baluta(a)nxp.com> To: Mark Brown <broonie(a)kernel.org> To: Jaroslav Kysela <perex(a)perex.cz> To: Takashi Iwai <tiwai(a)suse.com> To: Eric Biederman <ebiederm(a)xmission.com> To: Chromeos Kdump <chromeos-kdump(a)google.com> To: Steven Rostedt <rostedt(a)goodmis.org> To: Michael Ellerman <mpe(a)ellerman.id.au> To: Nicholas Piggin <npiggin(a)gmail.com> To: Christophe Leroy <christophe.leroy(a)csgroup.eu> To: "K. Y. Srinivasan" <kys(a)microsoft.com> To: Haiyang Zhang <haiyangz(a)microsoft.com> To: Wei Liu <wei.liu(a)kernel.org> To: Dexuan Cui <decui(a)microsoft.com> To: Thomas Gleixner <tglx(a)linutronix.de> To: Ingo Molnar <mingo(a)redhat.com> To: Borislav Petkov <bp(a)alien8.de> To: Dave Hansen <dave.hansen(a)linux.intel.com> To: x86(a)kernel.org To: "H. Peter Anvin" <hpa(a)zytor.com> To: Juergen Gross <jgross(a)suse.com> To: Boris Ostrovsky <boris.ostrovsky(a)oracle.com> To: Ard Biesheuvel <ardb(a)kernel.org> To: Bjorn Helgaas <bhelgaas(a)google.com> To: "Rafael J. Wysocki" <rafael(a)kernel.org> To: Pavel Machek <pavel(a)ucw.cz> To: Len Brown <len.brown(a)intel.com> Cc: stable(a)vger.kernel.org Cc: sound-open-firmware(a)alsa-project.org Cc: alsa-devel(a)alsa-project.org Cc: linux-kernel(a)vger.kernel.org Cc: kexec(a)lists.infradead.org Cc: linuxppc-dev(a)lists.ozlabs.org Cc: linux-hyperv(a)vger.kernel.org Cc: xen-devel(a)lists.xenproject.org Cc: linux-efi(a)vger.kernel.org Cc: linux-pci(a)vger.kernel.org Cc: linux-pm(a)vger.kernel.org Signed-off-by: Ricardo Ribalda <ribalda(a)chromium.org> --- Changes in v8: - Wrap pm_freezing and kexec_inprogress in functions. - Do not run snd_sof_machine_unregister(sdev, pdata) during kexec (Thanks Kai). - Link to v7: https://lore.kernel.org/r/20221127-snd-freeze-v7-0-127c582f1ca4@chromium.org Changes in v7: - Fix commit message (Thanks Pierre-Louis). - Link to v6: https://lore.kernel.org/r/20221127-snd-freeze-v6-0-3e90553f64a5@chromium.org Changes in v6: - Check if we are in kexec with the userspace frozen. - Link to v5: https://lore.kernel.org/r/20221127-snd-freeze-v5-0-4ededeb08ba0@chromium.org Changes in v5: - Edit subject prefix. - Link to v4: https://lore.kernel.org/r/20221127-snd-freeze-v4-0-51ca64b7f2ab@chromium.org Changes in v4: - Do not call snd_sof_machine_unregister from shutdown. - Link to v3: https://lore.kernel.org/r/20221127-snd-freeze-v3-0-a2eda731ca14@chromium.org Changes in v3: - Wrap pm_freezing in a function. - Link to v2: https://lore.kernel.org/r/20221127-snd-freeze-v2-0-d8a425ea9663@chromium.org Changes in v2: - Only use pm_freezing if CONFIG_FREEZER . - Link to v1: https://lore.kernel.org/r/20221127-snd-freeze-v1-0-57461a366ec2@chromium.org --- Ricardo Ribalda (3): kexec: Refactor kexec_in_progress into a function freezer: refactor pm_freezing into a function. ASoC: SOF: Fix deadlock when shutdown a frozen userspace arch/powerpc/platforms/pseries/vio.c | 2 +- arch/x86/kernel/cpu/mshyperv.c | 6 +++--- arch/x86/xen/enlighten_hvm.c | 2 +- drivers/firmware/efi/efi.c | 2 +- drivers/pci/pci-driver.c | 2 +- include/linux/freezer.h | 3 ++- include/linux/kexec.h | 5 ++--- kernel/freezer.c | 3 +-- kernel/kexec_core.c | 12 ++++++++++-- kernel/power/process.c | 24 ++++++++++++++++++++---- sound/soc/sof/core.c | 9 ++++++--- 11 files changed, 48 insertions(+), 22 deletions(-) --- base-commit: 4312098baf37ee17a8350725e6e0d0e8590252d4 change-id: 20221127-snd-freeze-1ee143228326 Best regards, -- Ricardo Ribalda <ribalda(a)chromium.org>

4 10

clangd friendly headers
by GitHub pull_request - opened 05 Dec '22

05 Dec '22

alsa-project/alsa-lib pull request #286 was opened from timblechmann: quality of life improvements when working with clangd enabled IDEs. these two patches fix issues when clangd parses the headers in isolation in a c++ compile context --- * make headers self-contained * avoid c++ keyword Request URL : https://github.com/alsa-project/alsa-lib/pull/286 Patch URL : https://github.com/alsa-project/alsa-lib/pull/286.patch Repository URL: https://github.com/alsa-project/alsa-lib

1 0

[PATCH] ASoC: soc-compress: Reposition and add pcm_mutex
by 강신형 05 Dec '22

05 Dec '22

>From a385cc2614c17ba3002b263c0ca47dfbf6395c78 Mon Sep 17 00:00:00 2001 From: "s47.kang" <s47.kang(a)samsung.com> Date: Fri, 2 Dec 2022 15:45:16 +0900 Subject: [PATCH] ASoC: soc-compress.c: Reposition and add pcm_mutex. If panic_on_warn is set and compress stream(DPCM) is started, then kernel panic will occurs because card->pcm_mutex isn't held appropriately in soc_compr_set_params_fe, soc_compr_open_fe and soc_compr_free_fe. So, I reposition and add pcm_mutex to resolve lockdep error. Signed-off-by: Shinhyung Kang <s47.kang(a)samsung.com> --- sound/soc/soc-compress.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/sound/soc/soc-compress.c b/sound/soc/soc-compress.c index 870f13e1d389..7bce5088b455 100644 --- a/sound/soc/soc-compress.c +++ b/sound/soc/soc-compress.c @@ -149,6 +149,8 @@ static int soc_compr_open_fe(struct snd_compr_stream *cstream) if (ret < 0) goto be_err; + mutex_lock_nested(&fe->card->pcm_mutex, fe->card->pcm_subclass); + /* calculate valid and active FE <-> BE dpcms */ dpcm_process_paths(fe, stream, &list, 1); fe->dpcm[stream].runtime = fe_substream->runtime; @@ -184,7 +186,6 @@ static int soc_compr_open_fe(struct snd_compr_stream *cstream) fe->dpcm[stream].state = SND_SOC_DPCM_STATE_OPEN; fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO; - mutex_lock_nested(&fe->card->pcm_mutex, fe->card->pcm_subclass); snd_soc_runtime_activate(fe, stream); mutex_unlock(&fe->card->pcm_mutex); @@ -215,7 +216,6 @@ static int soc_compr_free_fe(struct snd_compr_stream *cstream) mutex_lock_nested(&fe->card->pcm_mutex, fe->card->pcm_subclass); snd_soc_runtime_deactivate(fe, stream); - mutex_unlock(&fe->card->pcm_mutex); fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_FE; @@ -234,6 +234,8 @@ static int soc_compr_free_fe(struct snd_compr_stream *cstream) dpcm_be_disconnect(fe, stream); + mutex_unlock(&fe->card->pcm_mutex); + fe->dpcm[stream].runtime = NULL; snd_soc_link_compr_shutdown(cstream, 0); @@ -409,8 +411,9 @@ static int soc_compr_set_params_fe(struct snd_compr_stream *cstream, ret = snd_soc_link_compr_set_params(cstream); if (ret < 0) goto out; - + mutex_lock_nested(&fe->card->pcm_mutex, fe->card->pcm_subclass); dpcm_dapm_stream_event(fe, stream, SND_SOC_DAPM_STREAM_START); + mutex_unlock(&fe->card->pcm_mutex); fe->dpcm[stream].state = SND_SOC_DPCM_STATE_PREPARE; out: -- 2.21.0

2 1

[PATCH v1 0/2] ASoC/tda998x: Fix reporting of nonexistent capture streams
by Mark Brown 04 Dec '22

04 Dec '22

The recently added pcm-test selftest has pointed out that systems with the tda998x driver end up advertising that they support capture when in reality as far as I can see the tda998x devices are transmit only. The DAIs registered through hdmi-codec are bidirectional, meaning that for I2S systems when combined with a typical bidrectional CPU DAI the overall capability of the PCM is bidirectional. In most cases the I2S links will clock OK but no useful audio will be returned which isn't so bad but we should still not advertise the useless capability, and some systems may notice problems for example due to pinmux management. This is happening due to the hdmi-codec helpers not providing any mechanism for indicating unidirectional audio so add one and use it in the tda998x driver. It is likely other hdmi-codec users are also affected but I don't have those systems to hand. Mark Brown (2): ASoC: hdmi-codec: Allow playback and capture to be disabled drm: tda99x: Don't advertise non-existent capture support drivers/gpu/drm/i2c/tda998x_drv.c | 2 ++ include/sound/hdmi-codec.h | 4 ++++ sound/soc/codecs/hdmi-codec.c | 30 +++++++++++++++++++++++++----- 3 files changed, 31 insertions(+), 5 deletions(-) base-commit: f0c4d9fc9cc9462659728d168387191387e903cc -- 2.30.2

2 4

[PATCH v2 2/2] ASoC: tas5805m: add missing page switch.
by Daniel Beer 04 Dec '22

04 Dec '22

In tas5805m_refresh, we switch pages to update the DSP volume control, but we need to switch back to page 0 before trying to alter the soft-mute control. This latter page-switch was missing. Fixes: ec45268467f4 ("ASoC: add support for TAS5805M digital amplifier") Signed-off-by: Daniel Beer <daniel.beer(a)igorinstitute.com> --- sound/soc/codecs/tas5805m.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/sound/soc/codecs/tas5805m.c b/sound/soc/codecs/tas5805m.c index 6e2edf045446..4e38eb7acea1 100644 --- a/sound/soc/codecs/tas5805m.c +++ b/sound/soc/codecs/tas5805m.c @@ -203,6 +203,9 @@ static void tas5805m_refresh(struct tas5805m_priv *tas5805m) set_dsp_scale(rm, 0x24, tas5805m->vol[0]); set_dsp_scale(rm, 0x28, tas5805m->vol[1]); + regmap_write(rm, REG_PAGE, 0x00); + regmap_write(rm, REG_BOOK, 0x00); + /* Set/clear digital soft-mute */ regmap_write(rm, REG_DEVICE_CTRL_2, (tas5805m->is_muted ? DCTRL2_MUTE : 0) | -- 2.38.1

1 0

[PATCH v2 1/2] ASoC: tas5805m: rework to avoid scheduling while atomic.
by Daniel Beer 04 Dec '22

04 Dec '22

There's some setup we need to do in order to get the DSP initialized, and this can't be done until a bit-clock is ready. In an earlier version of this driver, this work was done in a DAPM callback. The DAPM callback doesn't guarantee that the bit-clock is running, so the work was moved instead to the trigger callback. Unfortunately this callback runs in atomic context, and the setup code needs to do I2C transactions. Here we use a work_struct to kick off the setup in a thread instead. Fixes: ec45268467f4 ("ASoC: add support for TAS5805M digital amplifier") Signed-off-by: Daniel Beer <daniel.beer(a)igorinstitute.com> --- sound/soc/codecs/tas5805m.c | 128 ++++++++++++++++++++++++------------ 1 file changed, 87 insertions(+), 41 deletions(-) diff --git a/sound/soc/codecs/tas5805m.c b/sound/soc/codecs/tas5805m.c index beb4ec629a03..6e2edf045446 100644 --- a/sound/soc/codecs/tas5805m.c +++ b/sound/soc/codecs/tas5805m.c @@ -154,6 +154,7 @@ static const uint32_t tas5805m_volume[] = { #define TAS5805M_VOLUME_MIN 0 struct tas5805m_priv { + struct i2c_client *i2c; struct regulator *pvdd; struct gpio_desc *gpio_pdn_n; @@ -165,6 +166,9 @@ struct tas5805m_priv { int vol[2]; bool is_powered; bool is_muted; + + struct work_struct work; + struct mutex lock; }; static void set_dsp_scale(struct regmap *rm, int offset, int vol) @@ -181,13 +185,11 @@ static void set_dsp_scale(struct regmap *rm, int offset, int vol) regmap_bulk_write(rm, offset, v, ARRAY_SIZE(v)); } -static void tas5805m_refresh(struct snd_soc_component *component) +static void tas5805m_refresh(struct tas5805m_priv *tas5805m) { - struct tas5805m_priv *tas5805m = - snd_soc_component_get_drvdata(component); struct regmap *rm = tas5805m->regmap; - dev_dbg(component->dev, "refresh: is_muted=%d, vol=%d/%d\n", + dev_dbg(&tas5805m->i2c->dev, "refresh: is_muted=%d, vol=%d/%d\n", tas5805m->is_muted, tas5805m->vol[0], tas5805m->vol[1]); regmap_write(rm, REG_PAGE, 0x00); @@ -226,8 +228,11 @@ static int tas5805m_vol_get(struct snd_kcontrol *kcontrol, struct tas5805m_priv *tas5805m = snd_soc_component_get_drvdata(component); + mutex_lock(&tas5805m->lock); ucontrol->value.integer.value[0] = tas5805m->vol[0]; ucontrol->value.integer.value[1] = tas5805m->vol[1]; + mutex_unlock(&tas5805m->lock); + return 0; } @@ -243,11 +248,13 @@ static int tas5805m_vol_put(struct snd_kcontrol *kcontrol, snd_soc_kcontrol_component(kcontrol); struct tas5805m_priv *tas5805m = snd_soc_component_get_drvdata(component); + int ret = 0; if (!(volume_is_valid(ucontrol->value.integer.value[0]) && volume_is_valid(ucontrol->value.integer.value[1]))) return -EINVAL; + mutex_lock(&tas5805m->lock); if (tas5805m->vol[0] != ucontrol->value.integer.value[0] || tas5805m->vol[1] != ucontrol->value.integer.value[1]) { tas5805m->vol[0] = ucontrol->value.integer.value[0]; @@ -256,11 +263,12 @@ static int tas5805m_vol_put(struct snd_kcontrol *kcontrol, tas5805m->vol[0], tas5805m->vol[1], tas5805m->is_powered); if (tas5805m->is_powered) - tas5805m_refresh(component); - return 1; + tas5805m_refresh(tas5805m); + ret = 1; } + mutex_unlock(&tas5805m->lock); - return 0; + return ret; } static const struct snd_kcontrol_new tas5805m_snd_controls[] = { @@ -294,54 +302,83 @@ static int tas5805m_trigger(struct snd_pcm_substream *substream, int cmd, struct snd_soc_component *component = dai->component; struct tas5805m_priv *tas5805m = snd_soc_component_get_drvdata(component); - struct regmap *rm = tas5805m->regmap; - unsigned int chan, global1, global2; switch (cmd) { case SNDRV_PCM_TRIGGER_START: case SNDRV_PCM_TRIGGER_RESUME: case SNDRV_PCM_TRIGGER_PAUSE_RELEASE: - dev_dbg(component->dev, "DSP startup\n"); - - /* We mustn't issue any I2C transactions until the I2S - * clock is stable. Furthermore, we must allow a 5ms - * delay after the first set of register writes to - * allow the DSP to boot before configuring it. - */ - usleep_range(5000, 10000); - send_cfg(rm, dsp_cfg_preboot, - ARRAY_SIZE(dsp_cfg_preboot)); - usleep_range(5000, 15000); - send_cfg(rm, tas5805m->dsp_cfg_data, - tas5805m->dsp_cfg_len); - - tas5805m->is_powered = true; - tas5805m_refresh(component); + dev_dbg(component->dev, "clock start\n"); + schedule_work(&tas5805m->work); break; case SNDRV_PCM_TRIGGER_STOP: case SNDRV_PCM_TRIGGER_SUSPEND: case SNDRV_PCM_TRIGGER_PAUSE_PUSH: - dev_dbg(component->dev, "DSP shutdown\n"); + break; - tas5805m->is_powered = false; + default: + return -EINVAL; + } - regmap_write(rm, REG_PAGE, 0x00); - regmap_write(rm, REG_BOOK, 0x00); + return 0; +} - regmap_read(rm, REG_CHAN_FAULT, &chan); - regmap_read(rm, REG_GLOBAL_FAULT1, &global1); - regmap_read(rm, REG_GLOBAL_FAULT2, &global2); +static void do_work(struct work_struct *work) +{ + struct tas5805m_priv *tas5805m = + container_of(work, struct tas5805m_priv, work); + struct regmap *rm = tas5805m->regmap; - dev_dbg(component->dev, - "fault regs: CHAN=%02x, GLOBAL1=%02x, GLOBAL2=%02x\n", - chan, global1, global2); + dev_dbg(&tas5805m->i2c->dev, "DSP startup\n"); - regmap_write(rm, REG_DEVICE_CTRL_2, DCTRL2_MODE_HIZ); - break; + mutex_lock(&tas5805m->lock); + /* We mustn't issue any I2C transactions until the I2S + * clock is stable. Furthermore, we must allow a 5ms + * delay after the first set of register writes to + * allow the DSP to boot before configuring it. + */ + usleep_range(5000, 10000); + send_cfg(rm, dsp_cfg_preboot, ARRAY_SIZE(dsp_cfg_preboot)); + usleep_range(5000, 15000); + send_cfg(rm, tas5805m->dsp_cfg_data, tas5805m->dsp_cfg_len); + + tas5805m->is_powered = true; + tas5805m_refresh(tas5805m); + mutex_unlock(&tas5805m->lock); +} - default: - return -EINVAL; +static int tas5805m_dac_event(struct snd_soc_dapm_widget *w, + struct snd_kcontrol *kcontrol, int event) +{ + struct snd_soc_component *component = snd_soc_dapm_to_component(w->dapm); + struct tas5805m_priv *tas5805m = + snd_soc_component_get_drvdata(component); + struct regmap *rm = tas5805m->regmap; + + if (event & SND_SOC_DAPM_PRE_PMD) { + unsigned int chan, global1, global2; + + dev_dbg(component->dev, "DSP shutdown\n"); + cancel_work_sync(&tas5805m->work); + + mutex_lock(&tas5805m->lock); + if (tas5805m->is_powered) { + tas5805m->is_powered = false; + + regmap_write(rm, REG_PAGE, 0x00); + regmap_write(rm, REG_BOOK, 0x00); + + regmap_read(rm, REG_CHAN_FAULT, &chan); + regmap_read(rm, REG_GLOBAL_FAULT1, &global1); + regmap_read(rm, REG_GLOBAL_FAULT2, &global2); + + dev_dbg(component->dev, "fault regs: CHAN=%02x, " + "GLOBAL1=%02x, GLOBAL2=%02x\n", + chan, global1, global2); + + regmap_write(rm, REG_DEVICE_CTRL_2, DCTRL2_MODE_HIZ); + } + mutex_unlock(&tas5805m->lock); } return 0; @@ -354,7 +391,8 @@ static const struct snd_soc_dapm_route tas5805m_audio_map[] = { static const struct snd_soc_dapm_widget tas5805m_dapm_widgets[] = { SND_SOC_DAPM_AIF_IN("DAC IN", "Playback", 0, SND_SOC_NOPM, 0, 0), - SND_SOC_DAPM_DAC("DAC", NULL, SND_SOC_NOPM, 0, 0), + SND_SOC_DAPM_DAC_E("DAC", NULL, SND_SOC_NOPM, 0, 0, + tas5805m_dac_event, SND_SOC_DAPM_PRE_PMD), SND_SOC_DAPM_OUTPUT("OUT") }; @@ -375,11 +413,14 @@ static int tas5805m_mute(struct snd_soc_dai *dai, int mute, int direction) struct tas5805m_priv *tas5805m = snd_soc_component_get_drvdata(component); + mutex_lock(&tas5805m->lock); dev_dbg(component->dev, "set mute=%d (is_powered=%d)\n", mute, tas5805m->is_powered); + tas5805m->is_muted = mute; if (tas5805m->is_powered) - tas5805m_refresh(component); + tas5805m_refresh(tas5805m); + mutex_unlock(&tas5805m->lock); return 0; } @@ -434,6 +475,7 @@ static int tas5805m_i2c_probe(struct i2c_client *i2c) if (!tas5805m) return -ENOMEM; + tas5805m->i2c = i2c; tas5805m->pvdd = devm_regulator_get(dev, "pvdd"); if (IS_ERR(tas5805m->pvdd)) { dev_err(dev, "failed to get pvdd supply: %ld\n", @@ -507,6 +549,9 @@ static int tas5805m_i2c_probe(struct i2c_client *i2c) gpiod_set_value(tas5805m->gpio_pdn_n, 1); usleep_range(10000, 15000); + INIT_WORK(&tas5805m->work, do_work); + mutex_init(&tas5805m->lock); + /* Don't register through devm. We need to be able to unregister * the component prior to deasserting PDN# */ @@ -527,6 +572,7 @@ static void tas5805m_i2c_remove(struct i2c_client *i2c) struct device *dev = &i2c->dev; struct tas5805m_priv *tas5805m = dev_get_drvdata(dev); + cancel_work_sync(&tas5805m->work); snd_soc_unregister_component(dev); gpiod_set_value(tas5805m->gpio_pdn_n, 0); usleep_range(10000, 15000); -- 2.38.1

1 0

[PATCH 0/2] soundwire: Remove redundant zeroing of page registers
by Richard Fitzgerald 02 Dec '22

02 Dec '22

Writing zero to the page registers after each message transaction can add up to a lot of overhead for codecs that need to transfer large amount of data - for example a firmware download. There's no spec reason I can see for this zeroing. The page registers are only used for a paged address. The bus code uses a non-paged address for registers in page 0. It always writes the page registers at the start of a paged transaction. If this zeroing was a workaround for anything, let me know and I will re-implement the zeroing as a quirk that can be enabled only when it is necessary. Richard Fitzgerald (2): soundwire: bus: Don't zero page registers after every transaction soundwire: bus: Remove unused reset_page_addr() callback drivers/soundwire/bus.c | 23 ----------------------- drivers/soundwire/cadence_master.c | 14 -------------- drivers/soundwire/cadence_master.h | 3 --- drivers/soundwire/intel_auxdevice.c | 1 - include/linux/soundwire/sdw.h | 3 --- 5 files changed, 44 deletions(-) -- 2.30.2

2 5