[PATCH 0/4] ASoC: Intel: machine driver corrections

[PATCH 0/4] ASoC: Intel: machine driver corrections

[Pierre-Louis Bossart]

The first fix solves an underflow in SoundWire platforms using the
max98373 amplifier, the rest of the patches are minor corrections in
machine drivers.

The fix should be queued for the 5.14 cycle, the rest should be
harmless but can be deferred for 5.15 if it's too late already.

Brent Lu (2):
  ASoC: SOF: add a helper to get topology configured bclk
  ASoC: Intel: sof_cs42l42: use helper function to get bclk frequency

Gongjun Song (1):
  ASoC: Intel: soc-acpi: add support for SoundWire of TGL-H-RVP

Rander Wang (1):
  ASoC: Intel: boards: fix xrun issue on platform with max98373

 include/sound/sof.h                           |  1 +
 sound/soc/intel/boards/sof_cs42l42.c          |  8 +-
 sound/soc/intel/boards/sof_sdw_max98373.c     | 81 ++++++++++++-------
 .../intel/common/soc-acpi-intel-tgl-match.c   | 15 ++++
 sound/soc/sof/sof-audio.c                     | 42 ++++++++--
 5 files changed, 111 insertions(+), 36 deletions(-)

-- 
2.25.1

[PATCH 1/4] ASoC: Intel: boards: fix xrun issue on platform with max98373

[Pierre-Louis Bossart]

From: Rander Wang <rander.wang@intel.com>

On TGL platform with max98373 codec the trigger start sequence is
fe first, then codec component and sdw link is the last. Recently
a delay was introduced in max98373 codec driver and this resulted
to the start of sdw stream transmission was delayed and the data
transmitted by fw can't be consumed by sdw controller, so xrun happened.

Adding delay in trigger function is a bad idea. This patch enable spk
pin in prepare function and disable it in hw_free to avoid xrun issue
caused by delay in trigger.

Fixes: 3a27875e91fb ("ASoC: max98373: Added 30ms turn on/off time delay")
BugLink: https://github.com/thesofproject/sof/issues/4066
Reviewed-by: Bard Liao <bard.liao@intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Signed-off-by: Rander Wang <rander.wang@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
---
 sound/soc/intel/boards/sof_sdw_max98373.c | 81 +++++++++++++++--------
 1 file changed, 53 insertions(+), 28 deletions(-)

diff --git a/sound/soc/intel/boards/sof_sdw_max98373.c b/sound/soc/intel/boards/sof_sdw_max98373.c
index 0e7ed906b341..25daef910aee 100644
--- a/sound/soc/intel/boards/sof_sdw_max98373.c
+++ b/sound/soc/intel/boards/sof_sdw_max98373.c
@@ -55,43 +55,68 @@ static int spk_init(struct snd_soc_pcm_runtime *rtd)
 	return ret;
 }
 
-static int max98373_sdw_trigger(struct snd_pcm_substream *substream, int cmd)
+static int mx8373_enable_spk_pin(struct snd_pcm_substream *substream, bool enable)
 {
+	struct snd_soc_pcm_runtime *rtd = asoc_substream_to_rtd(substream);
+	struct snd_soc_dai *codec_dai;
+	struct snd_soc_dai *cpu_dai;
 	int ret;
+	int j;
 
-	switch (cmd) {
-	case SNDRV_PCM_TRIGGER_START:
-	case SNDRV_PCM_TRIGGER_RESUME:
-	case SNDRV_PCM_TRIGGER_PAUSE_RELEASE:
-		/* enable max98373 first */
-		ret = max_98373_trigger(substream, cmd);
-		if (ret < 0)
-			break;
-
-		ret = sdw_trigger(substream, cmd);
-		break;
-	case SNDRV_PCM_TRIGGER_STOP:
-	case SNDRV_PCM_TRIGGER_SUSPEND:
-	case SNDRV_PCM_TRIGGER_PAUSE_PUSH:
-		ret = sdw_trigger(substream, cmd);
-		if (ret < 0)
-			break;
-
-		ret = max_98373_trigger(substream, cmd);
-		break;
-	default:
-		ret = -EINVAL;
-		break;
+	/* set spk pin by playback only */
+	if (substream->stream == SNDRV_PCM_STREAM_CAPTURE)
+		return 0;
+
+	cpu_dai = asoc_rtd_to_cpu(rtd, 0);
+	for_each_rtd_codec_dais(rtd, j, codec_dai) {
+		struct snd_soc_dapm_context *dapm =
+				snd_soc_component_get_dapm(cpu_dai->component);
+		char pin_name[16];
+
+		snprintf(pin_name, ARRAY_SIZE(pin_name), "%s Spk",
+			 codec_dai->component->name_prefix);
+
+		if (enable)
+			ret = snd_soc_dapm_enable_pin(dapm, pin_name);
+		else
+			ret = snd_soc_dapm_disable_pin(dapm, pin_name);
+
+		if (!ret)
+			snd_soc_dapm_sync(dapm);
 	}
 
-	return ret;
+	return 0;
+}
+
+static int mx8373_sdw_prepare(struct snd_pcm_substream *substream)
+{
+	int ret = 0;
+
+	/* according to soc_pcm_prepare dai link prepare is called first */
+	ret = sdw_prepare(substream);
+	if (ret < 0)
+		return ret;
+
+	return mx8373_enable_spk_pin(substream, true);
+}
+
+static int mx8373_sdw_hw_free(struct snd_pcm_substream *substream)
+{
+	int ret = 0;
+
+	/* according to soc_pcm_hw_free dai link free is called first */
+	ret = sdw_hw_free(substream);
+	if (ret < 0)
+		return ret;
+
+	return mx8373_enable_spk_pin(substream, false);
 }
 
 static const struct snd_soc_ops max_98373_sdw_ops = {
 	.startup = sdw_startup,
-	.prepare = sdw_prepare,
-	.trigger = max98373_sdw_trigger,
-	.hw_free = sdw_hw_free,
+	.prepare = mx8373_sdw_prepare,
+	.trigger = sdw_trigger,
+	.hw_free = mx8373_sdw_hw_free,
 	.shutdown = sdw_shutdown,
 };
 
-- 
2.25.1

[PATCH 2/4] ASoC: Intel: soc-acpi: add support for SoundWire of TGL-H-RVP

[Pierre-Louis Bossart]

From: Gongjun Song <gongjun.song@intel.com>

snd_soc_acpi_intel_tgl_sdw_machines structure array does not contain
the configuration information of TGL-H-RVP, which results in the
inability to enable soundwire on the TGL-H-RVP platform.

Add the corresponding configuration information to enable soundwire
on the TGL-H-RVP platform.

Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Gongjun Song <gongjun.song@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
---
 sound/soc/intel/common/soc-acpi-intel-tgl-match.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/sound/soc/intel/common/soc-acpi-intel-tgl-match.c b/sound/soc/intel/common/soc-acpi-intel-tgl-match.c
index 66595e3ab13f..d9b8902658c7 100644
--- a/sound/soc/intel/common/soc-acpi-intel-tgl-match.c
+++ b/sound/soc/intel/common/soc-acpi-intel-tgl-match.c
@@ -196,6 +196,15 @@ static const struct snd_soc_acpi_link_adr tgl_rvp[] = {
 	{}
 };
 
+static const struct snd_soc_acpi_link_adr tgl_rvp_headset_only[] = {
+	{
+		.mask = BIT(0),
+		.num_adr = ARRAY_SIZE(rt711_0_adr),
+		.adr_d = rt711_0_adr,
+	},
+	{}
+};
+
 static const struct snd_soc_acpi_link_adr tgl_hp[] = {
 	{
 		.mask = BIT(0),
@@ -398,6 +407,12 @@ struct snd_soc_acpi_mach snd_soc_acpi_intel_tgl_sdw_machines[] = {
 		.drv_name = "sof_sdw",
 		.sof_tplg_filename = "sof-tgl-sdw-max98373-rt5682.tplg",
 	},
+	{
+		.link_mask = 0x1, /* rt711 on link 0 */
+		.links = tgl_rvp_headset_only,
+		.drv_name = "sof_sdw",
+		.sof_tplg_filename = "sof-tgl-rt711.tplg",
+	},
 	{},
 };
 EXPORT_SYMBOL_GPL(snd_soc_acpi_intel_tgl_sdw_machines);
-- 
2.25.1

[PATCH 3/4] ASoC: SOF: add a helper to get topology configured bclk

[Pierre-Louis Bossart]

From: Brent Lu <brent.lu@intel.com>

Add helper function sof_dai_ssp_bclk() to get the BCLK frequency
configured by topology.

Reviewed-by: Bard Liao <bard.liao@intel.com>
Signed-off-by: Brent Lu <brent.lu@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
---
 include/sound/sof.h       |  1 +
 sound/soc/sof/sof-audio.c | 42 ++++++++++++++++++++++++++++++++-------
 2 files changed, 36 insertions(+), 7 deletions(-)

diff --git a/include/sound/sof.h b/include/sound/sof.h
index 502ed9b8d6a1..6a1cd8e783d8 100644
--- a/include/sound/sof.h
+++ b/include/sound/sof.h
@@ -101,5 +101,6 @@ struct sof_dev_desc {
 };
 
 int sof_dai_get_mclk(struct snd_soc_pcm_runtime *rtd);
+int sof_dai_get_bclk(struct snd_soc_pcm_runtime *rtd);
 
 #endif
diff --git a/sound/soc/sof/sof-audio.c b/sound/soc/sof/sof-audio.c
index 510883cd9107..989912f2b739 100644
--- a/sound/soc/sof/sof-audio.c
+++ b/sound/soc/sof/sof-audio.c
@@ -433,11 +433,10 @@ struct snd_sof_dai *snd_sof_find_dai(struct snd_soc_component *scomp,
 	return NULL;
 }
 
-/*
- * Helper to get SSP MCLK from a pcm_runtime.
- * Return 0 if not exist.
- */
-int sof_dai_get_mclk(struct snd_soc_pcm_runtime *rtd)
+#define SOF_DAI_CLK_INTEL_SSP_MCLK	0
+#define SOF_DAI_CLK_INTEL_SSP_BCLK	1
+
+static int sof_dai_get_clk(struct snd_soc_pcm_runtime *rtd, int clk_type)
 {
 	struct snd_soc_component *component =
 		snd_soc_rtdcom_lookup(rtd, SOF_AUDIO_PCM_DRV_NAME);
@@ -450,16 +449,45 @@ int sof_dai_get_mclk(struct snd_soc_pcm_runtime *rtd)
 
 	switch (dai->dai_config->type) {
 	case SOF_DAI_INTEL_SSP:
-		return dai->dai_config->ssp.mclk_rate;
+		switch (clk_type) {
+		case SOF_DAI_CLK_INTEL_SSP_MCLK:
+			return dai->dai_config->ssp.mclk_rate;
+		case SOF_DAI_CLK_INTEL_SSP_BCLK:
+			return dai->dai_config->ssp.bclk_rate;
+		default:
+			dev_err(rtd->dev, "fail to get SSP clk %d rate\n",
+				clk_type);
+			return -EINVAL;
+		}
+		break;
 	default:
 		/* not yet implemented for platforms other than the above */
-		dev_err(rtd->dev, "mclk for dai_config->type %d not supported yet!\n",
+		dev_err(rtd->dev, "DAI type %d not supported yet!\n",
 			dai->dai_config->type);
 		return -EINVAL;
 	}
 }
+
+/*
+ * Helper to get SSP MCLK from a pcm_runtime.
+ * Return 0 if not exist.
+ */
+int sof_dai_get_mclk(struct snd_soc_pcm_runtime *rtd)
+{
+	return sof_dai_get_clk(rtd, SOF_DAI_CLK_INTEL_SSP_MCLK);
+}
 EXPORT_SYMBOL(sof_dai_get_mclk);
 
+/*
+ * Helper to get SSP BCLK from a pcm_runtime.
+ * Return 0 if not exist.
+ */
+int sof_dai_get_bclk(struct snd_soc_pcm_runtime *rtd)
+{
+	return sof_dai_get_clk(rtd, SOF_DAI_CLK_INTEL_SSP_BCLK);
+}
+EXPORT_SYMBOL(sof_dai_get_bclk);
+
 /*
  * SOF Driver enumeration.
  */
-- 
2.25.1

[PATCH 4/4] ASoC: Intel: sof_cs42l42: use helper function to get bclk frequency

[Pierre-Louis Bossart]

From: Brent Lu <brent.lu@intel.com>

Use sof_dai_get_bclk() function to get bclk frequency instead of a
hard-coded magic number depending on configuration in topology.

Reviewed-by: Bard Liao <bard.liao@intel.com>
Signed-off-by: Brent Lu <brent.lu@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
---
 sound/soc/intel/boards/sof_cs42l42.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/sound/soc/intel/boards/sof_cs42l42.c b/sound/soc/intel/boards/sof_cs42l42.c
index 42aadf801f72..2b16011b7a06 100644
--- a/sound/soc/intel/boards/sof_cs42l42.c
+++ b/sound/soc/intel/boards/sof_cs42l42.c
@@ -16,6 +16,7 @@
 #include <sound/pcm.h>
 #include <sound/pcm_params.h>
 #include <sound/soc.h>
+#include <sound/sof.h>
 #include <sound/soc-acpi.h>
 #include <dt-bindings/sound/cs42l42.h>
 #include "../../codecs/hdac_hdmi.h"
@@ -122,7 +123,12 @@ static int sof_cs42l42_hw_params(struct snd_pcm_substream *substream,
 	struct snd_soc_dai *codec_dai = asoc_rtd_to_codec(rtd, 0);
 	int clk_freq, ret;
 
-	clk_freq = 3072000; /* BCLK freq */
+	clk_freq = sof_dai_get_bclk(rtd); /* BCLK freq */
+
+	if (clk_freq <= 0) {
+		dev_err(rtd->dev, "get bclk freq failed: %d\n", clk_freq);
+		return -EINVAL;
+	}
 
 	/* Configure sysclk for codec */
 	ret = snd_soc_dai_set_sysclk(codec_dai, 0,
-- 
2.25.1

Re: (subset) [PATCH 0/4] ASoC: Intel: machine driver corrections

[Mark Brown]

On Fri, 25 Jun 2021 15:50:38 -0500, Pierre-Louis Bossart wrote:
> The first fix solves an underflow in SoundWire platforms using the
> max98373 amplifier, the rest of the patches are minor corrections in
> machine drivers.
> 
> The fix should be queued for the 5.14 cycle, the rest should be
> harmless but can be deferred for 5.15 if it's too late already.
> 
> [...]

Applied to

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-next

Thanks!

[1/4] ASoC: Intel: boards: fix xrun issue on platform with max98373
      commit: 33c8516841ea4fa12fdb8961711bf95095c607ee

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark

Re: (subset) [PATCH 0/4] ASoC: Intel: machine driver corrections

[Mark Brown]

On Fri, 25 Jun 2021 15:50:38 -0500, Pierre-Louis Bossart wrote:
> The first fix solves an underflow in SoundWire platforms using the
> max98373 amplifier, the rest of the patches are minor corrections in
> machine drivers.
> 
> The fix should be queued for the 5.14 cycle, the rest should be
> harmless but can be deferred for 5.15 if it's too late already.
> 
> [...]

Applied to

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-next

Thanks!

[2/4] ASoC: Intel: soc-acpi: add support for SoundWire of TGL-H-RVP
      commit: f99acc259f621ae6667782778b2065c15e109693
[3/4] ASoC: SOF: add a helper to get topology configured bclk
      commit: bc619cfc6278c87b4e310f9db9f45abc263220e8
[4/4] ASoC: Intel: sof_cs42l42: use helper function to get bclk frequency
      commit: 837ad6da36ba765d9ff8120c93dd243b9200957e

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark