Skip to contentSkip to conversation

ASoC: Intel: Skylake: More driver updates

PATCH In review [email protected] 5 participants 22 messages Mar 24, 2017 – Mar 29, 2017

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

v2 cover ASoC: Intel: Skylake: More driver updates

From: Jeeja KP <[email protected]>

This patch series provides updates on the following:
o Fix LLCH register read.
o Add 16-bit constraint to FE bxt_rt298 machine.
o Fix DMA position reporting for capture stream.
o Use the sig_bits to define dai bps caps.
o Update sig_bits based on converter capability.
o Rearrangement of code to cleanup SKL SST library.
o Add support for deferred DSP module bind.

Changes in v2:
- Addressed Pierre's comments for "Fix DMA position reporting for capture stream" patch.

B, Jayachandran (1):
ALSA: hda: Fix LLCH register read

G Kranthi (1):
ASoC: Intel: Skylake: Add 16-bit constraint to FE bxt_rt298 machine

Hardik T Shah (1):
ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Jeeja KP (5):
ASoC: Intel: Skylake: Use the sig_bits to define dai bps capability
ASoC: hdac_hdmi: Update sig_bits based on converter capability
ASoC: Intel: Skylake: Rearrangement of code to cleanup SKL SST library
ASoC: Intel: Skylake: Fix module state after unbind and delete
ASoC: Intel: Skylake: Add support for deferred DSP module bind

Vinod Koul (3):
ASoC: Intel: Skylake: Don't unload module when in use
ASoC: Intel: Skylake: Remove redundant vmixer handler
ASoC: Intel: Skylake: remove hard coded ACPI path

 sound/hda/hdac_controller.c             |   2 +-
 sound/soc/codecs/hdac_hdmi.c            |   4 +-
 sound/soc/intel/boards/bxt_rt298.c      |   3 +
 sound/soc/intel/skylake/skl-messages.c  |   2 +-
 sound/soc/intel/skylake/skl-nhlt.c      |   7 +-
 sound/soc/intel/skylake/skl-pcm.c       |  95 ++++++++++++++++---
 sound/soc/intel/skylake/skl-sst-dsp.h   |  24 +++--
 sound/soc/intel/skylake/skl-sst-ipc.h   |   8 ++
 sound/soc/intel/skylake/skl-sst-utils.c |  60 ++----------
 sound/soc/intel/skylake/skl-sst.c       |   5 +
 sound/soc/intel/skylake/skl-topology.c  | 159 +++++++++++++++++++++++++-------
 sound/soc/intel/skylake/skl-topology.h  |  17 ++--
 sound/soc/intel/skylake/skl.h           |   1 +
 13 files changed, 263 insertions(+), 124 deletions(-)
Signature · 1 line
2.5.0
#1 lore raw

PATCH v2 1/11 ALSA: hda: Fix LLCH register read

From: "B, Jayachandran" <[email protected]>

LLCH is a 16 bit register. Use readw instead of readl API.

---
 sound/hda/hdac_controller.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/hda/hdac_controller.c +1 -1
index 0430658..6f1e99c 100644
@@ -268,7 +268,7 @@ int snd_hdac_bus_parse_capabilities(struct hdac_bus *bus)
 	unsigned int offset;
 	unsigned int counter = 0;
 
-	offset = snd_hdac_chip_readl(bus, LLCH);
+	offset = snd_hdac_chip_readw(bus, LLCH);
 
 	/* Lets walk the linked capabilities list */
 	do {
Signature · 1 line
2.5.0
Acked-by on v2 1/11
#2 lore raw

Re: [PATCH v2 01/11] ALSA: hda: Fix LLCH register read

Quoted 7 lines · On Fri, 24 Mar 2017 18:40:24 +0100, [email protected] wrote:

From: "B, Jayachandran" <[email protected]>

LLCH is a 16 bit register. Use readw instead of readl API.

Takashi

Quoted 25 lines
---
 sound/hda/hdac_controller.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/hda/hdac_controller.c +1 -1
index 0430658..6f1e99c 100644
@@ -268,7 +268,7 @@ int snd_hdac_bus_parse_capabilities(struct hdac_bus *bus)
 	unsigned int offset;
 	unsigned int counter = 0;
 
-	offset = snd_hdac_chip_readl(bus, LLCH);
+	offset = snd_hdac_chip_readw(bus, LLCH);
 
 	/* Lets walk the linked capabilities list */
 	do {
Signature · 6 lines
2.5.0

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
#3 lore raw

PATCH v2 2/11 ASoC: Intel: Skylake: Use the sig_bits to define dai bps capability

In reply to [email protected]

From: Jeeja KP <[email protected]>

For calculating the HDA DMA format, use the max_bps supported by the DAI caps instead of fixing it to 32/24. For host DMA the Max bps support is 32, but in case of link DMA, this depends on the codec capability. So use the sig_bits to define the bps supported by dai.

---
 sound/soc/intel/skylake/skl-pcm.c      | 25 ++++++++++++++++++++++---
 sound/soc/intel/skylake/skl-topology.c |  2 ++
 sound/soc/intel/skylake/skl-topology.h |  2 ++
 3 files changed, 26 insertions(+), 3 deletions(-)
sound/soc/intel/skylake/skl-pcm.c +22 -3
index 2f90bc4..3c61dba 100644
@@ -155,7 +155,7 @@ int skl_pcm_host_dma_prepare(struct device *dev, struct skl_pipe_params *params)
 	snd_hdac_ext_stream_decouple(ebus, stream, true);
 
 	format_val = snd_hdac_calc_stream_format(params->s_freq,
-				params->ch, params->format, 32, 0);
+			params->ch, params->format, params->host_bps, 0);
 
 	dev_dbg(dev, "format_val=%d, rate=%d, ch=%d, format=%d\n",
 		format_val, params->s_freq, params->ch, params->format);
@@ -190,8 +190,8 @@ int skl_pcm_link_dma_prepare(struct device *dev, struct skl_pipe_params *params)
 
 	stream = stream_to_hdac_ext_stream(hstream);
 	snd_hdac_ext_stream_decouple(ebus, stream, true);
-	format_val = snd_hdac_calc_stream_format(params->s_freq,
-				params->ch, params->format, 24, 0);
+	format_val = snd_hdac_calc_stream_format(params->s_freq, params->ch,
+					params->format, params->link_bps, 0);
 
 	dev_dbg(dev, "format_val=%d, rate=%d, ch=%d, format=%d\n",
 		format_val, params->s_freq, params->ch, params->format);
@@ -309,6 +309,11 @@ static int skl_pcm_hw_params(struct snd_pcm_substream *substream,
 	p_params.host_dma_id = dma_id;
 	p_params.stream = substream->stream;
 	p_params.format = params_format(params);
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+		p_params.host_bps = dai->driver->playback.sig_bits;
+	else
+		p_params.host_bps = dai->driver->capture.sig_bits;
+
 
 	m_cfg = skl_tplg_fe_get_cpr_module(dai, p_params.stream);
 	if (m_cfg)
@@ -547,6 +552,11 @@ static int skl_link_hw_params(struct snd_pcm_substream *substream,
 	p_params.link_index = link->index;
 	p_params.format = params_format(params);
 
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+		p_params.link_bps = codec_dai->driver->playback.sig_bits;
+	else
+		p_params.link_bps = codec_dai->driver->capture.sig_bits;
+
 	return skl_tplg_be_update_params(dai, &p_params);
 }
 
@@ -652,6 +662,7 @@ static struct snd_soc_dai_driver skl_platform_dai[] = {
 		.rates = SNDRV_PCM_RATE_48000 | SNDRV_PCM_RATE_16000 | SNDRV_PCM_RATE_8000,
 		.formats = SNDRV_PCM_FMTBIT_S16_LE |
 			SNDRV_PCM_FMTBIT_S24_LE | SNDRV_PCM_FMTBIT_S32_LE,
+		.sig_bits = 32,
 	},
 	.capture = {
 		.stream_name = "System Capture",
@@ -659,6 +670,7 @@ static struct snd_soc_dai_driver skl_platform_dai[] = {
 		.channels_max = HDA_STEREO,
 		.rates = SNDRV_PCM_RATE_48000 | SNDRV_PCM_RATE_16000,
 		.formats = SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S24_LE,
+		.sig_bits = 32,
 	},
 },
 {
@@ -670,6 +682,7 @@ static struct snd_soc_dai_driver skl_platform_dai[] = {
 		.channels_max = HDA_QUAD,
 		.rates = SNDRV_PCM_RATE_48000 | SNDRV_PCM_RATE_16000,
 		.formats = SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S24_LE,
+		.sig_bits = 32,
 	},
 },
 {
@@ -681,6 +694,7 @@ static struct snd_soc_dai_driver skl_platform_dai[] = {
 		.channels_max = HDA_STEREO,
 		.rates = SNDRV_PCM_RATE_48000,
 		.formats = SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S24_LE,
+		.sig_bits = 32,
 	},
 },
 {
@@ -692,6 +706,7 @@ static struct snd_soc_dai_driver skl_platform_dai[] = {
 		.channels_max = HDA_STEREO,
 		.rates = SNDRV_PCM_RATE_48000,
 		.formats = SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S24_LE,
+		.sig_bits = 32,
 	},
 },
 {
@@ -703,6 +718,7 @@ static struct snd_soc_dai_driver skl_platform_dai[] = {
 		.channels_max = HDA_QUAD,
 		.rates = SNDRV_PCM_RATE_48000 | SNDRV_PCM_RATE_16000,
 		.formats = SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S24_LE,
+		.sig_bits = 32,
 	},
 },
 {
@@ -718,6 +734,7 @@ static struct snd_soc_dai_driver skl_platform_dai[] = {
 			SNDRV_PCM_RATE_192000,
 		.formats = SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S24_LE |
 			SNDRV_PCM_FMTBIT_S32_LE,
+		.sig_bits = 32,
 	},
 },
 {
@@ -733,6 +750,7 @@ static struct snd_soc_dai_driver skl_platform_dai[] = {
 			SNDRV_PCM_RATE_192000,
 		.formats = SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S24_LE |
 			SNDRV_PCM_FMTBIT_S32_LE,
+		.sig_bits = 32,
 	},
 },
 {
@@ -748,6 +766,7 @@ static struct snd_soc_dai_driver skl_platform_dai[] = {
 			SNDRV_PCM_RATE_192000,
 		.formats = SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S24_LE |
 			SNDRV_PCM_FMTBIT_S32_LE,
+		.sig_bits = 32,
 	},
 },
 
sound/soc/intel/skylake/skl-topology.c +2 -0
index c6bd4bb..43f9cb3 100644
@@ -1242,10 +1242,12 @@ static void skl_tplg_fill_dma_id(struct skl_module_cfg *mcfg,
 		case SKL_DEVICE_HDALINK:
 			pipe->p_params->link_dma_id = params->link_dma_id;
 			pipe->p_params->link_index = params->link_index;
+			pipe->p_params->link_bps = params->link_bps;
 			break;
 
 		case SKL_DEVICE_HDAHOST:
 			pipe->p_params->host_dma_id = params->host_dma_id;
+			pipe->p_params->host_bps = params->host_bps;
 			break;
 
 		default:
sound/soc/intel/skylake/skl-topology.h +2 -0
index fefab0e..bf2c63b 100644
@@ -257,6 +257,8 @@ struct skl_pipe_params {
 	snd_pcm_format_t format;
 	int link_index;
 	int stream;
+	unsigned int host_bps;
+	unsigned int link_bps;
 };
 
 struct skl_pipe {
Signature · 1 line
2.5.0
#4 lore raw

PATCH v2 3/11 ASoC: hdac_hdmi: Update sig_bits based on converter capability

In reply to [email protected]

From: Jeeja KP <[email protected]>

When creating the codec dai, use sig_bits to update the max bps based on the codec capability. So both the link DMA and codec format will be calculated based on DAI sig_bits.

So update the sig_bits with converter capability and use the sig_bits for HDA format calculation.

---
 sound/soc/codecs/hdac_hdmi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
sound/soc/codecs/hdac_hdmi.c +2 -2
index 78fca8a..5b76947 100644
@@ -469,7 +469,7 @@ static int hdac_hdmi_set_hw_params(struct snd_pcm_substream *substream,
 
 	format = snd_hdac_calc_stream_format(params_rate(hparams),
 			params_channels(hparams), params_format(hparams),
-			24, 0);
+			dai->driver->playback.sig_bits, 0);
 
 	pcm = hdac_hdmi_get_pcm_from_cvt(hdmi, dai_map->cvt);
 	if (!pcm)
@@ -1419,8 +1419,8 @@ static int hdac_hdmi_create_dais(struct hdac_device *hdac,
 		hdmi_dais[i].playback.rate_min = rate_min;
 		hdmi_dais[i].playback.channels_min = 2;
 		hdmi_dais[i].playback.channels_max = 2;
+		hdmi_dais[i].playback.sig_bits = bps;
 		hdmi_dais[i].ops = &hdmi_dai_ops;
-
 		i++;
 	}
 
Signature · 1 line
2.5.0
#5 lore raw

PATCH v2 4/11 ASoC: Intel: Skylake: Add 16-bit constraint to FE bxt_rt298 machine

In reply to [email protected]

From: G Kranthi <[email protected]>

Add constraint to FE to restrict sample format to 16-bit for bxt_rt298 machine

---
 sound/soc/intel/boards/bxt_rt298.c | 3 +++
 1 file changed, 3 insertions(+)
sound/soc/intel/boards/bxt_rt298.c +3 -0
index 176c080..1a68d04 100644
@@ -274,12 +274,15 @@ static int bxt_fe_startup(struct snd_pcm_substream *substream)
 	 * on this platform for PCM device we support:
 	 *      48Khz
 	 *      stereo
+	 *	16-bit audio
 	 */
 
 	runtime->hw.channels_max = 2;
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_CHANNELS,
 				&constraints_channels);
 
+	runtime->hw.formats = SNDRV_PCM_FMTBIT_S16_LE;
+	snd_pcm_hw_constraint_msbits(runtime, 0, 16, 16);
 	snd_pcm_hw_constraint_list(runtime, 0,
 				SNDRV_PCM_HW_PARAM_RATE, &constraints_rates);
 
Signature · 1 line
2.5.0
#6 lore raw

PATCH v2 5/11 ASoC: Intel: Skylake: Don't unload module when in use

In reply to [email protected]

From: Vinod Koul <[email protected]>

A module may have multiple instances in DSP, so unload only when usage count is zero.

---
 sound/soc/intel/skylake/skl-sst.c | 5 +++++
 1 file changed, 5 insertions(+)
sound/soc/intel/skylake/skl-sst.c +5 -0
index 39d4aaa..5395297 100644
@@ -417,6 +417,11 @@ static int skl_unload_module(struct sst_dsp *ctx, u16 mod_id)
 		dev_err(ctx->dev, "Module bad usage cnt!:%d\n", usage_cnt);
 		return -EIO;
 	}
+
+	/* if module is used by others return, no need to unload */
+	if (usage_cnt > 0)
+		return 0;
+
 	ret = skl_ipc_unload_modules(&skl->ipc,
 			SKL_NUM_MODULES, &mod_id);
 	if (ret < 0) {
Signature · 1 line
2.5.0
#7 lore raw

PATCH v2 6/11 ASoC: Intel: Skylake: Remove redundant vmixer handler

In reply to [email protected]

From: Vinod Koul <[email protected]>

Initially vmixer and mixer widget handlers were bit different, but over time they became same so remove the duplicate code.

---
 sound/soc/intel/skylake/skl-topology.c | 32 +-------------------------------
 1 file changed, 1 insertion(+), 31 deletions(-)
sound/soc/intel/skylake/skl-topology.c +1 -31
index 43f9cb3..35d9d7b 100644
@@ -1073,36 +1073,6 @@ static int skl_tplg_pga_dapm_post_pmd_event(struct snd_soc_dapm_widget *w,
 }
 
 /*
- * In modelling, we assume there will be ONLY one mixer in a pipeline.  If
- * mixer is not required then it is treated as static mixer aka vmixer with
- * a hard path to source module
- * So we don't need to check if source is started or not as hard path puts
- * dependency on each other
- */
-static int skl_tplg_vmixer_event(struct snd_soc_dapm_widget *w,
-				struct snd_kcontrol *k, int event)
-{
-	struct snd_soc_dapm_context *dapm = w->dapm;
-	struct skl *skl = get_skl_ctx(dapm->dev);
-
-	switch (event) {
-	case SND_SOC_DAPM_PRE_PMU:
-		return skl_tplg_mixer_dapm_pre_pmu_event(w, skl);
-
-	case SND_SOC_DAPM_POST_PMU:
-		return skl_tplg_mixer_dapm_post_pmu_event(w, skl);
-
-	case SND_SOC_DAPM_PRE_PMD:
-		return skl_tplg_mixer_dapm_pre_pmd_event(w, skl);
-
-	case SND_SOC_DAPM_POST_PMD:
-		return skl_tplg_mixer_dapm_post_pmd_event(w, skl);
-	}
-
-	return 0;
-}
-
-/*
  * In modelling, we assume there will be ONLY one mixer in a pipeline. If a
  * second one is required that is created as another pipe entity.
  * The mixer is responsible for pipe management and represent a pipeline
@@ -1570,7 +1540,7 @@ int skl_tplg_be_update_params(struct snd_soc_dai *dai,
 
 static const struct snd_soc_tplg_widget_events skl_tplg_widget_ops[] = {
 	{SKL_MIXER_EVENT, skl_tplg_mixer_event},
-	{SKL_VMIXER_EVENT, skl_tplg_vmixer_event},
+	{SKL_VMIXER_EVENT, skl_tplg_mixer_event},
 	{SKL_PGA_EVENT, skl_tplg_pga_event},
 };
 
Signature · 1 line
2.5.0
#8 lore raw

PATCH v2 7/11 ASoC: Intel: Skylake: remove hard coded ACPI path

In reply to [email protected]

From: Vinod Koul <[email protected]>

We should not hard code the ACPI path to get acpi_handle. Instead use ACPI_HANDLE macro to do the job.

---
 sound/soc/intel/skylake/skl-nhlt.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
sound/soc/intel/skylake/skl-nhlt.c +3 -4
index 7eb9c41..e3f0667 100644
@@ -24,8 +24,6 @@
 static u8 OSC_UUID[16] = {0x6E, 0x88, 0x9F, 0xA6, 0xEB, 0x6C, 0x94, 0x45,
 				0xA4, 0x1F, 0x7B, 0x5D, 0xCE, 0x24, 0xC5, 0x53};
 
-#define DSDT_NHLT_PATH "\\_SB.PCI0.HDAS"
-
 struct nhlt_acpi_table *skl_nhlt_init(struct device *dev)
 {
 	acpi_handle handle;
@@ -33,8 +31,9 @@ struct nhlt_acpi_table *skl_nhlt_init(struct device *dev)
 	struct nhlt_resource_desc  *nhlt_ptr = NULL;
 	struct nhlt_acpi_table *nhlt_table = NULL;
 
-	if (ACPI_FAILURE(acpi_get_handle(NULL, DSDT_NHLT_PATH, &handle))) {
-		dev_err(dev, "Requested NHLT device not found\n");
+	handle = ACPI_HANDLE(dev);
+	if (!handle) {
+		dev_err(dev, "Didn't find ACPI_HANDLE\n");
 		return NULL;
 	}
 
Signature · 1 line
2.5.0
#9 lore raw

PATCH v2 8/11 ASoC: Intel: Skylake: Rearrangement of code to cleanup SKL SST library

In reply to [email protected]

From: Jeeja KP <[email protected]>

Skylake driver topology header/driver structure is referenced and used in SST library which creates circular dependency. Hence the rearrangement.

---
 sound/soc/intel/skylake/skl-pcm.c       | 35 +++++++++++++++----
 sound/soc/intel/skylake/skl-sst-dsp.h   | 24 +++++++++----
 sound/soc/intel/skylake/skl-sst-ipc.h   |  8 +++++
 sound/soc/intel/skylake/skl-sst-utils.c | 60 +++++----------------------------
 sound/soc/intel/skylake/skl-topology.c  | 11 ++++--
 sound/soc/intel/skylake/skl-topology.h  | 13 -------
 6 files changed, 70 insertions(+), 81 deletions(-)
sound/soc/intel/skylake/skl-pcm.c +29 -6
index 3c61dba..ef440d8 100644
@@ -1174,29 +1174,52 @@ static int skl_pcm_new(struct snd_soc_pcm_runtime *rtd)
 	return retval;
 }
 
+static int skl_get_module_info(struct skl *skl, struct skl_module_cfg *mconfig)
+{
+	struct skl_sst *ctx = skl->skl_sst;
+	struct uuid_module *module;
+	uuid_le *uuid_mod;
+
+	uuid_mod = (uuid_le *)mconfig->guid;
+
+	if (list_empty(&ctx->uuid_list)) {
+		dev_err(ctx->dev, "Module list is empty\n");
+		return -EIO;
+	}
+
+	list_for_each_entry(module, &ctx->uuid_list, list) {
+		if (uuid_le_cmp(*uuid_mod, module->uuid) == 0) {
+			mconfig->id.module_id = module->id;
+			mconfig->is_loadable = module->is_loadable;
+			return 0;
+		}
+	}
+
+	return -EIO;
+}
+
 static int skl_populate_modules(struct skl *skl)
 {
 	struct skl_pipeline *p;
 	struct skl_pipe_module *m;
 	struct snd_soc_dapm_widget *w;
 	struct skl_module_cfg *mconfig;
-	int ret;
+	int ret = 0;
 
 	list_for_each_entry(p, &skl->ppl_list, node) {
 		list_for_each_entry(m, &p->pipe->w_list, node) {
-
 			w = m->w;
 			mconfig = w->priv;
 
-			ret = snd_skl_get_module_info(skl->skl_sst, mconfig);
+			ret = skl_get_module_info(skl, mconfig);
 			if (ret < 0) {
 				dev_err(skl->skl_sst->dev,
-					"query module info failed:%d\n", ret);
-				goto err;
+					"query module info failed\n");
+				return ret;
 			}
 		}
 	}
-err:
+
 	return ret;
 }
 
sound/soc/intel/skylake/skl-sst-dsp.h +17 -7
index 5d7a93a..7229a12 100644
@@ -17,13 +17,14 @@
 #define __SKL_SST_DSP_H__
 
 #include <linux/interrupt.h>
+#include <linux/uuid.h>
 #include <sound/memalloc.h>
 #include "skl-sst-cldma.h"
-#include "skl-topology.h"
 
 struct sst_dsp;
 struct skl_sst;
 struct sst_dsp_device;
+struct skl_lib_info;
 
 /* Intel HD Audio General DSP Registers */
 #define SKL_ADSP_GEN_BASE		0x0
@@ -172,6 +173,19 @@ struct skl_dsp_loader_ops {
 				 int stream_tag);
 };
 
+#define MAX_INSTANCE_BUFF 2
+
+struct uuid_module {
+	uuid_le uuid;
+	int id;
+	int is_loadable;
+	int max_instance;
+	u64 pvt_id[MAX_INSTANCE_BUFF];
+	int *instance_id;
+
+	struct list_head list;
+};
+
 struct skl_load_module_info {
 	u16 mod_id;
 	const struct firmware *fw;
@@ -223,14 +237,10 @@ int bxt_sst_init_fw(struct device *dev, struct skl_sst *ctx);
 void skl_sst_dsp_cleanup(struct device *dev, struct skl_sst *ctx);
 void bxt_sst_dsp_cleanup(struct device *dev, struct skl_sst *ctx);
 
-int snd_skl_get_module_info(struct skl_sst *ctx,
-				struct skl_module_cfg *mconfig);
 int snd_skl_parse_uuids(struct sst_dsp *ctx, const struct firmware *fw,
 				unsigned int offset, int index);
-int skl_get_pvt_id(struct skl_sst *ctx,
-				struct skl_module_cfg *mconfig);
-int skl_put_pvt_id(struct skl_sst *ctx,
-				struct skl_module_cfg *mconfig);
+int skl_get_pvt_id(struct skl_sst *ctx, uuid_le *uuid_mod, int instance_id);
+int skl_put_pvt_id(struct skl_sst *ctx, uuid_le *uuid_mod, int *pvt_id);
 int skl_get_pvt_instance_id_map(struct skl_sst *ctx,
 				int module_id, int instance_id);
 void skl_freeup_uuid_list(struct skl_sst *ctx);
sound/soc/intel/skylake/skl-sst-ipc.h +8 -0
index fc07c39..4abf98c 100644
@@ -69,6 +69,14 @@ struct skl_d0i3_data {
 	struct delayed_work work;
 };
 
+#define SKL_LIB_NAME_LENGTH 128
+#define SKL_MAX_LIB 16
+
+struct skl_lib_info {
+	char name[SKL_LIB_NAME_LENGTH];
+	const struct firmware *fw;
+};
+
 struct skl_sst {
 	struct device *dev;
 	struct sst_dsp *dsp;
sound/soc/intel/skylake/skl-sst-utils.c +8 -52

103 lines not on this page · raw message

sound/soc/intel/skylake/skl-topology.c +8 -3
index 35d9d7b..8bd5ded 100644
@@ -539,6 +539,7 @@ skl_tplg_init_pipe_modules(struct skl *skl, struct skl_pipe *pipe)
 	int ret = 0;
 
 	list_for_each_entry(w_module, &pipe->w_list, node) {
+		uuid_le *uuid_mod;
 		w = w_module->w;
 		mconfig = w->priv;
 
@@ -576,13 +577,15 @@ skl_tplg_init_pipe_modules(struct skl *skl, struct skl_pipe *pipe)
 		 * FE/BE params
 		 */
 		skl_tplg_update_module_params(w, ctx);
-		mconfig->id.pvt_id = skl_get_pvt_id(ctx, mconfig);
+		uuid_mod = (uuid_le *)mconfig->guid;
+		mconfig->id.pvt_id = skl_get_pvt_id(ctx, uuid_mod,
+						mconfig->id.instance_id);
 		if (mconfig->id.pvt_id < 0)
 			return ret;
 		skl_tplg_set_module_init_data(w);
 		ret = skl_init_module(ctx, mconfig);
 		if (ret < 0) {
-			skl_put_pvt_id(ctx, mconfig);
+			skl_put_pvt_id(ctx, uuid_mod, &mconfig->id.pvt_id);
 			return ret;
 		}
 		skl_tplg_alloc_pipe_mcps(skl, mconfig);
@@ -602,7 +605,9 @@ static int skl_tplg_unload_pipe_modules(struct skl_sst *ctx,
 	struct skl_module_cfg *mconfig = NULL;
 
 	list_for_each_entry(w_module, &pipe->w_list, node) {
+		uuid_le *uuid_mod;
 		mconfig  = w_module->w->priv;
+		uuid_mod = (uuid_le *)mconfig->guid;
 
 		if (mconfig->is_loadable && ctx->dsp->fw_ops.unload_mod &&
 			mconfig->m_state > SKL_MODULE_UNINIT) {
@@ -611,7 +616,7 @@ static int skl_tplg_unload_pipe_modules(struct skl_sst *ctx,
 			if (ret < 0)
 				return -EIO;
 		}
-		skl_put_pvt_id(ctx, mconfig);
+		skl_put_pvt_id(ctx, uuid_mod, &mconfig->id.pvt_id);
 	}
 
 	/* no modules to unload in this path, so return */
sound/soc/intel/skylake/skl-topology.h +0 -13
index bf2c63b..8536d03 100644
@@ -336,19 +336,6 @@ struct skl_pipeline {
 	struct list_head node;
 };
 
-#define SKL_LIB_NAME_LENGTH 128
-#define SKL_MAX_LIB 16
-
-struct skl_lib_info {
-	char name[SKL_LIB_NAME_LENGTH];
-	const struct firmware *fw;
-};
-
-struct skl_manifest {
-	u32 lib_count;
-	struct skl_lib_info lib[SKL_MAX_LIB];
-};
-
 static inline struct skl *get_skl_ctx(struct device *dev)
 {
 	struct hdac_ext_bus *ebus = dev_get_drvdata(dev);
Signature · 1 line
2.5.0
#10 lore raw

PATCH v2 9/11 ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

In reply to [email protected]

From: Hardik T Shah <[email protected]>

As per hardware recommendation, for every capture stream completion following operations need to be done in order to reflect the actual data that is received in position buffer.

1. Wait for 20us before reading the DMA position in buffer once the interrupt is generated for stream completion.
2. Read any of the register to flush the DMA position value. This is dummy read operation.

---
 sound/soc/intel/skylake/skl-pcm.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)
sound/soc/intel/skylake/skl-pcm.c +21 -2
index ef440d8..1823197 100644
@@ -21,6 +21,7 @@
 
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
+#include <linux/delay.h>
 #include <sound/pcm_params.h>
 #include <sound/soc.h>
 #include "skl.h"
@@ -1063,13 +1064,31 @@ static snd_pcm_uframes_t skl_platform_pcm_pointer
 	 * HAD space reflects the actual data that is transferred.
 	 * Use the position buffer for capture, as DPIB write gets
 	 * completed earlier than the actual data written to the DDR.
+	 *
+	 * For capture stream following workaround is required to fix the
+	 * incorrect position reporting.
+	 *
+	 * 1. Wait for 20us before reading the DMA position in buffer once
+	 * the interrupt is generated for stream completion as update happens
+	 * on the HDA frame boundary i.e. 20.833uSec.
+	 * 2. Read DPIB register to flush the DMA position value. This dummy
+	 * read is required to flush DMA position value.
+	 * 3. Read the DMA Position-in-Buffer. This value now will be equal to
+	 * or greater than period boundary.
 	 */
-	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
 		pos = readl(ebus->bus.remap_addr + AZX_REG_VS_SDXDPIB_XBASE +
 				(AZX_REG_VS_SDXDPIB_XINTERVAL *
 				hdac_stream(hstream)->index));
-	else
+	} else {
+		udelay(20);
+		readl(ebus->bus.remap_addr +
+				AZX_REG_VS_SDXDPIB_XBASE +
+				(AZX_REG_VS_SDXDPIB_XINTERVAL *
+				 hdac_stream(hstream)->index));
 		pos = snd_hdac_stream_get_pos_posbuf(hdac_stream(hstream));
+	}
 
 	if (pos >= hdac_stream(hstream)->bufsize)
 		pos = 0;
Signature · 1 line
2.5.0
#11 lore raw

Re: [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 11 lines · On Fri, 24 Mar 2017 18:40:32 +0100, [email protected] wrote:

From: Hardik T Shah <[email protected]>

As per hardware recommendation, for every capture stream completion following operations need to be done in order to reflect the actual data that is received in position buffer.

1. Wait for 20us before reading the DMA position in buffer once the interrupt is generated for stream completion.
2. Read any of the register to flush the DMA position value. This is dummy read operation.

Are these workarounds needed for the legacy driver? If yes, which chips require it?

thanks,

Takashi

Quoted 62 lines
---
 sound/soc/intel/skylake/skl-pcm.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)
sound/soc/intel/skylake/skl-pcm.c +21 -2
index ef440d8..1823197 100644
@@ -21,6 +21,7 @@
 
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
+#include <linux/delay.h>
 #include <sound/pcm_params.h>
 #include <sound/soc.h>
 #include "skl.h"
@@ -1063,13 +1064,31 @@ static snd_pcm_uframes_t skl_platform_pcm_pointer
 	 * HAD space reflects the actual data that is transferred.
 	 * Use the position buffer for capture, as DPIB write gets
 	 * completed earlier than the actual data written to the DDR.
+	 *
+	 * For capture stream following workaround is required to fix the
+	 * incorrect position reporting.
+	 *
+	 * 1. Wait for 20us before reading the DMA position in buffer once
+	 * the interrupt is generated for stream completion as update happens
+	 * on the HDA frame boundary i.e. 20.833uSec.
+	 * 2. Read DPIB register to flush the DMA position value. This dummy
+	 * read is required to flush DMA position value.
+	 * 3. Read the DMA Position-in-Buffer. This value now will be equal to
+	 * or greater than period boundary.
 	 */
-	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
 		pos = readl(ebus->bus.remap_addr + AZX_REG_VS_SDXDPIB_XBASE +
 				(AZX_REG_VS_SDXDPIB_XINTERVAL *
 				hdac_stream(hstream)->index));
-	else
+	} else {
+		udelay(20);
+		readl(ebus->bus.remap_addr +
+				AZX_REG_VS_SDXDPIB_XBASE +
+				(AZX_REG_VS_SDXDPIB_XINTERVAL *
+				 hdac_stream(hstream)->index));
 		pos = snd_hdac_stream_get_pos_posbuf(hdac_stream(hstream));
+	}
 
 	if (pos >= hdac_stream(hstream)->bufsize)
 		pos = 0;
Signature · 6 lines
2.5.0

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
#12 lore raw

Re: [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 23 lines

-----Original Message-----
From: [email protected] [mailto:alsa-devel-bounces@alsa- project.org] On Behalf Of Takashi Iwai
Sent: Friday, March 24, 2017 11:38 AM
To: Kp, Jeeja <[email protected]>
Cc: [email protected]; R, Dharageswari <[email protected]>; Patches Audio <[email protected]>; Shah, Hardik T <[email protected]>; [email protected]; Girdwood, Liam R <[email protected]>
Subject: Re: [alsa-devel] [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 7 lines · On Fri, 24 Mar 2017 18:40:32 +0100, [email protected] wrote:

From: Hardik T Shah <[email protected]>

As per hardware recommendation, for every capture stream completion following operations need to be done in order to reflect the actual data that is received in position buffer.

Quoted 8 lines
Quoted 4 lines

1. Wait for 20us before reading the DMA position in buffer once the interrupt is generated for stream completion.
2. Read any of the register to flush the DMA position value. This is dummy read operation.

Are these workarounds needed for the legacy driver? If yes, which chips require it?

Yes, these are needed in legacy driver as well.

Quoted 77 lines

From SKL and BXT onwards, it is needed.

thanks,

Takashi

Review

Quoted 9 lines
---
 sound/soc/intel/skylake/skl-pcm.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)
(quoted diff) +0 -0

b/sound/soc/intel/skylake/skl-pcm.c

sound/soc/intel/skylake/skl-pcm.c static snd_pcm_uframes_t

Quoted 12 lines

index ef440d8..1823197 100644

sound/soc/intel/skylake/skl-pcm.c +1 -0
@@ -21,6 +21,7 @@
 
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
+#include <linux/delay.h>
 #include <sound/pcm_params.h>
 #include <sound/soc.h>
 #include "skl.h"
@@ -1063,13 +1064,31 @@ static snd_pcm_uframes_t

skl_platform_pcm_pointer

Quoted 9 lines
sound/soc/intel/skylake/skl-pcm.c +6 -0
 	 * HAD space reflects the actual data that is transferred.
 	 * Use the position buffer for capture, as DPIB write gets
 	 * completed earlier than the actual data written to the DDR.
+	 *
+	 * For capture stream following workaround is required to fix the
+	 * incorrect position reporting.
+	 *
+	 * 1. Wait for 20us before reading the DMA position in buffer once
+	 * the interrupt is generated for stream completion as update happens

sound/soc/intel/skylake/skl-pcm.c static snd_pcm_uframes_t skl_platform_pcm_pointer

Quoted 10 lines
sound/soc/intel/skylake/skl-pcm.c +7 -1
+	 * on the HDA frame boundary i.e. 20.833uSec.
+	 * 2. Read DPIB register to flush the DMA position value. This dummy
+	 * read is required to flush DMA position value.
+	 * 3. Read the DMA Position-in-Buffer. This value now will be equal to
+	 * or greater than period boundary.
 	 */
-	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
 		pos = readl(ebus->bus.remap_addr +

AZX_REG_VS_SDXDPIB_XBASE +

sound/soc/intel/skylake/skl-pcm.c static snd_pcm_uframes_t skl_platform_pcm_pointer

Quoted 10 lines
sound/soc/intel/skylake/skl-pcm.c +6 -1
 				(AZX_REG_VS_SDXDPIB_XINTERVAL *
 				hdac_stream(hstream)->index));
-	else
+	} else {
+		udelay(20);
+		readl(ebus->bus.remap_addr +
+				AZX_REG_VS_SDXDPIB_XBASE +
+				(AZX_REG_VS_SDXDPIB_XINTERVAL *
+				 hdac_stream(hstream)->index));
 		pos =

snd_hdac_stream_get_pos_posbuf(hdac_stream(hstream));

Quoted 7 lines
sound/soc/intel/skylake/skl-pcm.c +1 -1
+	}
 
 	if (pos >= hdac_stream(hstream)->bufsize)
 		pos = 0;
--

2.5.0

Quoted 5 lines

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

#13 lore raw

Re: [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 29 lines · Rakesh A
Quoted 29 lines

-----Original Message-----
From: [email protected] [mailto:alsa-devel-bounces@alsa- project.org] On Behalf Of Takashi Iwai
Sent: Friday, March 24, 2017 11:38 AM
To: Kp, Jeeja <[email protected]>
Cc: [email protected]; R, Dharageswari <[email protected]>; Patches Audio <[email protected]>; Shah, Hardik T <[email protected]>; [email protected]; Girdwood, Liam R <[email protected]>
Subject: Re: [alsa-devel] [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 11 lines · On Fri, 24 Mar 2017 18:40:32 +0100, [email protected] wrote:

From: Hardik T Shah <[email protected]>

As per hardware recommendation, for every capture stream completion following operations need to be done in order to reflect the actual data that is received in position buffer.

1. Wait for 20us before reading the DMA position in buffer once the interrupt is generated for stream completion.
2. Read any of the register to flush the DMA position value. This is dummy read operation.

Quoted 6 lines
Quoted 3 lines

Are these workarounds needed for the legacy driver? If yes, which chips require it?

Yes, these are needed in legacy driver as well.
From SKL and BXT onwards, it is needed.

OK, thanks for confirmation.

Now, from what I read in the above, is the workaround required *only* after the interrupt is generated? 20us delay isn't so cheap, and we tend to inquire PCM positions often. If the workaround is needed only after the PCM period elapse, we can set some flag in the irq handler and apply the workaround only when necessary.

Takashi

Quoted 77 lines
Quoted 76 lines

thanks,

Takashi

Review

Quoted 9 lines
---
 sound/soc/intel/skylake/skl-pcm.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)
(quoted diff) +0 -0

b/sound/soc/intel/skylake/skl-pcm.c

sound/soc/intel/skylake/skl-pcm.c static snd_pcm_uframes_t

Quoted 12 lines

index ef440d8..1823197 100644

sound/soc/intel/skylake/skl-pcm.c +1 -0
@@ -21,6 +21,7 @@
 
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
+#include <linux/delay.h>
 #include <sound/pcm_params.h>
 #include <sound/soc.h>
 #include "skl.h"
@@ -1063,13 +1064,31 @@ static snd_pcm_uframes_t

skl_platform_pcm_pointer

Quoted 9 lines
sound/soc/intel/skylake/skl-pcm.c +6 -0
 	 * HAD space reflects the actual data that is transferred.
 	 * Use the position buffer for capture, as DPIB write gets
 	 * completed earlier than the actual data written to the DDR.
+	 *
+	 * For capture stream following workaround is required to fix the
+	 * incorrect position reporting.
+	 *
+	 * 1. Wait for 20us before reading the DMA position in buffer once
+	 * the interrupt is generated for stream completion as update happens

sound/soc/intel/skylake/skl-pcm.c static snd_pcm_uframes_t skl_platform_pcm_pointer

Quoted 10 lines
sound/soc/intel/skylake/skl-pcm.c +7 -1
+	 * on the HDA frame boundary i.e. 20.833uSec.
+	 * 2. Read DPIB register to flush the DMA position value. This dummy
+	 * read is required to flush DMA position value.
+	 * 3. Read the DMA Position-in-Buffer. This value now will be equal to
+	 * or greater than period boundary.
 	 */
-	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
 		pos = readl(ebus->bus.remap_addr +

AZX_REG_VS_SDXDPIB_XBASE +

sound/soc/intel/skylake/skl-pcm.c static snd_pcm_uframes_t skl_platform_pcm_pointer

Quoted 10 lines
sound/soc/intel/skylake/skl-pcm.c +6 -1
 				(AZX_REG_VS_SDXDPIB_XINTERVAL *
 				hdac_stream(hstream)->index));
-	else
+	} else {
+		udelay(20);
+		readl(ebus->bus.remap_addr +
+				AZX_REG_VS_SDXDPIB_XBASE +
+				(AZX_REG_VS_SDXDPIB_XINTERVAL *
+				 hdac_stream(hstream)->index));
 		pos =

snd_hdac_stream_get_pos_posbuf(hdac_stream(hstream));

Quoted 7 lines
sound/soc/intel/skylake/skl-pcm.c +1 -1
+	}
 
 	if (pos >= hdac_stream(hstream)->bufsize)
 		pos = 0;
--

2.5.0

Quoted 5 lines

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

#14 lore raw

Re: [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 54 lines

-----Original Message-----
From: [email protected] [mailto:alsa-devel-bounces@alsa- project.org] On Behalf Of Takashi Iwai
Sent: Friday, March 24, 2017 11:50 AM
To: Ughreja, Rakesh A <[email protected]>
Cc: [email protected]; R, Dharageswari <[email protected]>; Patches Audio <[email protected]>; Shah, Hardik T <[email protected]>; [email protected]; Girdwood, Liam R <[email protected]>; Kp, Jeeja <[email protected]>
Subject: Re: [alsa-devel] [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 5 lines · Rakesh A
Quoted 5 lines

-----Original Message-----
From: [email protected] [mailto:alsa-devel-

bounces@alsa-

Quoted 4 lines
Quoted 4 lines

project.org] On Behalf Of Takashi Iwai
Sent: Friday, March 24, 2017 11:38 AM
To: Kp, Jeeja <[email protected]>
Cc: [email protected]; R, Dharageswari

<[email protected]>;

Quoted 20 lines
Quoted 20 lines

Patches Audio <[email protected]>; Shah, Hardik T <[email protected]>; [email protected]; Girdwood, Liam R <[email protected]>
Subject: Re: [alsa-devel] [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 11 lines · On Fri, 24 Mar 2017 18:40:32 +0100, [email protected] wrote:

From: Hardik T Shah <[email protected]>

As per hardware recommendation, for every capture stream completion following operations need to be done in order to reflect the actual data that is received in position buffer.

1. Wait for 20us before reading the DMA position in buffer once the interrupt is generated for stream completion.
2. Read any of the register to flush the DMA position value. This is dummy read operation.

Quoted 6 lines
Quoted 3 lines

Are these workarounds needed for the legacy driver? If yes, which chips require it?

Yes, these are needed in legacy driver as well.
From SKL and BXT onwards, it is needed.

Quoted 8 lines

OK, thanks for confirmation.

Now, from what I read in the above, is the workaround required *only* after the interrupt is generated? 20us delay isn't so cheap, and we tend to inquire PCM positions often. If the workaround is needed only after the PCM period elapse, we can set some flag in the irq handler and apply the workaround only when necessary.

Yes, Takashi the workaround is required only in the period elapsed interrupt. In some cases the DMA Position updates are delayed and so when the period elapsed interrupt occurs the wait_for_avail thinks that one period worth of data is not available and so returns only on the next period elapsed interrupt. This creates problem for 2 periods playback/capture streams.

So even in the period elapsed interrupt the wait is required only if the position is less than the period boundary.

Quoted 84 lines

Takashi

Quoted 67 lines
Quoted 67 lines

thanks,

Takashi

Review

Quoted 9 lines
---
 sound/soc/intel/skylake/skl-pcm.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)
(quoted diff) +0 -0

b/sound/soc/intel/skylake/skl-pcm.c

sound/soc/intel/skylake/skl-pcm.c static snd_pcm_uframes_t

Quoted 12 lines

index ef440d8..1823197 100644

sound/soc/intel/skylake/skl-pcm.c +1 -0
@@ -21,6 +21,7 @@
 
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
+#include <linux/delay.h>
 #include <sound/pcm_params.h>
 #include <sound/soc.h>
 #include "skl.h"
@@ -1063,13 +1064,31 @@ static snd_pcm_uframes_t

skl_platform_pcm_pointer

Quoted 9 lines
sound/soc/intel/skylake/skl-pcm.c +6 -0
 	 * HAD space reflects the actual data that is transferred.
 	 * Use the position buffer for capture, as DPIB write gets
 	 * completed earlier than the actual data written to the DDR.
+	 *
+	 * For capture stream following workaround is required to fix the
+	 * incorrect position reporting.
+	 *
+	 * 1. Wait for 20us before reading the DMA position in buffer once
+	 * the interrupt is generated for stream completion as update happens

sound/soc/intel/skylake/skl-pcm.c static snd_pcm_uframes_t skl_platform_pcm_pointer

Quoted 10 lines
sound/soc/intel/skylake/skl-pcm.c +7 -1
+	 * on the HDA frame boundary i.e. 20.833uSec.
+	 * 2. Read DPIB register to flush the DMA position value. This dummy
+	 * read is required to flush DMA position value.
+	 * 3. Read the DMA Position-in-Buffer. This value now will be equal to
+	 * or greater than period boundary.
 	 */
-	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
 		pos = readl(ebus->bus.remap_addr +

AZX_REG_VS_SDXDPIB_XBASE +

sound/soc/intel/skylake/skl-pcm.c static snd_pcm_uframes_t skl_platform_pcm_pointer

Quoted 10 lines
sound/soc/intel/skylake/skl-pcm.c +6 -1
 				(AZX_REG_VS_SDXDPIB_XINTERVAL *
 				hdac_stream(hstream)->index));
-	else
+	} else {
+		udelay(20);
+		readl(ebus->bus.remap_addr +
+				AZX_REG_VS_SDXDPIB_XBASE +
+				(AZX_REG_VS_SDXDPIB_XINTERVAL *
+				 hdac_stream(hstream)->index));
 		pos =

snd_hdac_stream_get_pos_posbuf(hdac_stream(hstream));

Quoted 7 lines
sound/soc/intel/skylake/skl-pcm.c +1 -1
+	}
 
 	if (pos >= hdac_stream(hstream)->bufsize)
 		pos = 0;
--

2.5.0

Quoted 10 lines
Quoted 9 lines
Quoted 5 lines

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

#15 lore raw

Re: [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 23 lines · Rakesh A
Quoted 7 lines
Quoted 6 lines
Quoted 3 lines

Are these workarounds needed for the legacy driver? If yes, which chips require it?

Yes, these are needed in legacy driver as well.
From SKL and BXT onwards, it is needed.

Quoted 8 lines

OK, thanks for confirmation.

Now, from what I read in the above, is the workaround required *only* after the interrupt is generated? 20us delay isn't so cheap, and we tend to inquire PCM positions often. If the workaround is needed only after the PCM period elapse, we can set some flag in the irq handler and apply the workaround only when necessary.

Yes, Takashi the workaround is required only in the period elapsed interrupt. In some cases the DMA Position updates are delayed and so when the period elapsed interrupt occurs the wait_for_avail thinks that one period worth of data is not available and so returns only on the next period elapsed interrupt. This creates problem for 2 periods playback/capture streams.

Quoted 2 lines

So even in the period elapsed interrupt the wait is required only if the position is less than the period boundary.

Hi Takashi,

So we need this for 2 periods when the device is in irq mode. Not for other cases. ie SKL_PLUS..

But have you seen any user reports on this till now.

Signature · 1 line
~Vinod
#16 lore raw

Re: [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 25 lines · Vinod Koul
Quoted 23 lines · Rakesh A
Quoted 7 lines
Quoted 6 lines
Quoted 3 lines

Are these workarounds needed for the legacy driver? If yes, which chips require it?

Yes, these are needed in legacy driver as well.
From SKL and BXT onwards, it is needed.

Quoted 8 lines

OK, thanks for confirmation.

Now, from what I read in the above, is the workaround required *only* after the interrupt is generated? 20us delay isn't so cheap, and we tend to inquire PCM positions often. If the workaround is needed only after the PCM period elapse, we can set some flag in the irq handler and apply the workaround only when necessary.

Yes, Takashi the workaround is required only in the period elapsed interrupt. In some cases the DMA Position updates are delayed and so when the period elapsed interrupt occurs the wait_for_avail thinks that one period worth of data is not available and so returns only on the next period elapsed interrupt. This creates problem for 2 periods playback/capture streams.

Quoted 7 lines
Quoted 2 lines

So even in the period elapsed interrupt the wait is required only if the position is less than the period boundary.

Hi Takashi,

So we need this for 2 periods when the device is in irq mode. Not for other cases. ie SKL_PLUS..

Yeah, thanks. I'd cook a couple of patches to do that for the legacy driver. But I still wonder whether the wait is always needed.

Judging from the description, does the discrepancy of posbuf read happen *only* when the DMA position goes across the BD boundary? Or does it happen at any time?

When you trace, you can see that the apps frequently inquires the position. So, an unconditional wait should be really avoided.

Quoted 1 line

But have you seen any user reports on this till now.

I've seen some bug reports mentioning about crackling audio capture on SKL (I forgot URLs). It might be triggered by that.

thanks,

Takashi

#17 lore raw

Re: [PATCH v2 09/11] ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 35 lines · Takashi Iwai
Quoted 25 lines · Vinod Koul
Quoted 23 lines · Rakesh A
Quoted 7 lines
Quoted 6 lines
Quoted 3 lines

Are these workarounds needed for the legacy driver? If yes, which chips require it?

Yes, these are needed in legacy driver as well.
From SKL and BXT onwards, it is needed.

Quoted 8 lines

OK, thanks for confirmation.

Now, from what I read in the above, is the workaround required *only* after the interrupt is generated? 20us delay isn't so cheap, and we tend to inquire PCM positions often. If the workaround is needed only after the PCM period elapse, we can set some flag in the irq handler and apply the workaround only when necessary.

Yes, Takashi the workaround is required only in the period elapsed interrupt. In some cases the DMA Position updates are delayed and so when the period elapsed interrupt occurs the wait_for_avail thinks that one period worth of data is not available and so returns only on the next period elapsed interrupt. This creates problem for 2 periods playback/capture streams.

Quoted 7 lines
Quoted 2 lines

So even in the period elapsed interrupt the wait is required only if the position is less than the period boundary.

Hi Takashi,

So we need this for 2 periods when the device is in irq mode. Not for other cases. ie SKL_PLUS..

Quoted 6 lines

Yeah, thanks. I'd cook a couple of patches to do that for the legacy driver. But I still wonder whether the wait is always needed.

Judging from the description, does the discrepancy of posbuf read happen *only* when the DMA position goes across the BD boundary? Or does it happen at any time?

I think it can happen anytime, but the impact is not felt unless we have a 2 period case. The update is in-flight, so read returns a value lesser than period boundary. We will sleep till next period ie next wake, which results in overrun. For more than 2 periods it doesn't impact much as the overrun case should not happen

Quoted 2 lines

When you trace, you can see that the apps frequently inquires the position. So, an unconditional wait should be really avoided.

If they enquire after a while then we should be okay, but if they are written nicely with good power behaviour then we may have issue, as writes are typically done after period boundaries.

Quoted 5 lines
Quoted 2 lines

But have you seen any user reports on this till now.

I've seen some bug reports mentioning about crackling audio capture on SKL (I forgot URLs). It might be triggered by that.

Quiet possible..

Signature · 1 line
~Vinod
maintainer
#18 lore raw

Applied "ASoC: Intel: Skylake: Fix DMA position reporting for capture stream" to the asoc tree

In reply to [email protected]

The patch

ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

has been applied to the asoc tree at

git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git

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

Quoted 1 line

From fdd85a054b850db43c6abe39c1da28b581be5e93 Mon Sep 17 00:00:00 2001

From: Hardik T Shah <[email protected]>
Date: Fri, 24 Mar 2017 23:10:32 +0530
Subject: [PATCH] ASoC: Intel: Skylake: Fix DMA position reporting for capture
stream

As per hardware recommendation, for every capture stream completion following operations need to be done in order to reflect the actual data that is received in position buffer.

1. Wait for 20us before reading the DMA position in buffer once the interrupt is generated for stream completion.
2. Read any of the register to flush the DMA position value. This is dummy read operation.

---
 sound/soc/intel/skylake/skl-pcm.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)
sound/soc/intel/skylake/skl-pcm.c +21 -2
index ef440d8629e8..1823197c15c8 100644
@@ -21,6 +21,7 @@
 
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
+#include <linux/delay.h>
 #include <sound/pcm_params.h>
 #include <sound/soc.h>
 #include "skl.h"
@@ -1063,13 +1064,31 @@ static snd_pcm_uframes_t skl_platform_pcm_pointer
 	 * HAD space reflects the actual data that is transferred.
 	 * Use the position buffer for capture, as DPIB write gets
 	 * completed earlier than the actual data written to the DDR.
+	 *
+	 * For capture stream following workaround is required to fix the
+	 * incorrect position reporting.
+	 *
+	 * 1. Wait for 20us before reading the DMA position in buffer once
+	 * the interrupt is generated for stream completion as update happens
+	 * on the HDA frame boundary i.e. 20.833uSec.
+	 * 2. Read DPIB register to flush the DMA position value. This dummy
+	 * read is required to flush DMA position value.
+	 * 3. Read the DMA Position-in-Buffer. This value now will be equal to
+	 * or greater than period boundary.
 	 */
-	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
 		pos = readl(ebus->bus.remap_addr + AZX_REG_VS_SDXDPIB_XBASE +
 				(AZX_REG_VS_SDXDPIB_XINTERVAL *
 				hdac_stream(hstream)->index));
-	else
+	} else {
+		udelay(20);
+		readl(ebus->bus.remap_addr +
+				AZX_REG_VS_SDXDPIB_XBASE +
+				(AZX_REG_VS_SDXDPIB_XINTERVAL *
+				 hdac_stream(hstream)->index));
 		pos = snd_hdac_stream_get_pos_posbuf(hdac_stream(hstream));
+	}
 
 	if (pos >= hdac_stream(hstream)->bufsize)
 		pos = 0;
Signature · 1 line
2.11.0
#19 lore raw

PATCH v2 10/11 ASoC: Intel: Skylake: Fix module state after unbind and delete

In reply to [email protected]

From: Jeeja KP <[email protected]>

When DSP module is unbound, the module state needs to be in INIT_DONE state instead of UNINT. Also the state needs to be set to UNINIT after module is deleted from DSP pipeline.

So, set the module state to INIT_DONE after unbind and then UNINIT after module is deleted.

---
 sound/soc/intel/skylake/skl-messages.c | 2 +-
 sound/soc/intel/skylake/skl-topology.c | 5 +++++
 2 files changed, 6 insertions(+), 1 deletion(-)
sound/soc/intel/skylake/skl-messages.c +1 -1
index ba1ec97..09730dd 100644
@@ -862,7 +862,7 @@ static void skl_clear_module_state(struct skl_module_pin *mpin, int max,
 	}
 
 	if (!found)
-		mcfg->m_state = SKL_MODULE_UNINIT;
+		mcfg->m_state = SKL_MODULE_INIT_DONE;
 	return;
 }
 
sound/soc/intel/skylake/skl-topology.c +5 -0
index 8bd5ded..e960d9f 100644
@@ -1037,6 +1037,11 @@ static int skl_tplg_mixer_dapm_post_pmd_event(struct snd_soc_dapm_widget *w,
 
 	skl_delete_pipe(ctx, mconfig->pipe);
 
+	list_for_each_entry(w_module, &s_pipe->w_list, node) {
+		src_module = w_module->w->priv;
+		src_module->m_state = SKL_MODULE_UNINIT;
+	}
+
 	return skl_tplg_unload_pipe_modules(ctx, s_pipe);
 }
 
Signature · 1 line
2.5.0
#20 lore raw

PATCH v2 11/11 ASoC: Intel: Skylake: Add support for deferred DSP module bind

In reply to [email protected]

From: Jeeja KP <[email protected]>

Module at the end of DSP pipeline that needs to be connected to a module in another pipeline are represented as a PGA(leaf node) and in PGA event handler these modules are bound/unbounded. Modules other than PGA leaf can be connected directly or via switch to a module in another pipeline.
Example: reference path.

To support the deferred DSP module bind, following changes are done:
o When the path is enabled, the destination module that needs to be bound may not be initialized. If the module is not initialized, add these modules in a deferred bind list.
o When the destination module is initialized, check for these modules in deferred bind list. If found, bind them.
o When the destination module is deleted, Unbind the modules.
o When the source module is deleted, remove the entry from the deferred bind list.

---
 sound/soc/intel/skylake/skl-pcm.c      |  12 ++++
 sound/soc/intel/skylake/skl-topology.c | 109 ++++++++++++++++++++++++++++++++-
 sound/soc/intel/skylake/skl-topology.h |   6 ++
 sound/soc/intel/skylake/skl.h          |   1 +
 4 files changed, 127 insertions(+), 1 deletion(-)
sound/soc/intel/skylake/skl-pcm.c +12 -0
index 1823197..600faad 100644
@@ -1300,6 +1300,7 @@ int skl_platform_register(struct device *dev)
 	struct skl *skl = ebus_to_skl(ebus);
 
 	INIT_LIST_HEAD(&skl->ppl_list);
+	INIT_LIST_HEAD(&skl->bind_list);
 
 	ret = snd_soc_register_platform(dev, &skl_platform_drv);
 	if (ret) {
@@ -1320,6 +1321,17 @@ int skl_platform_register(struct device *dev)
 
 int skl_platform_unregister(struct device *dev)
 {
+	struct hdac_ext_bus *ebus = dev_get_drvdata(dev);
+	struct skl *skl = ebus_to_skl(ebus);
+	struct skl_module_deferred_bind *modules;
+
+	if (!list_empty(&skl->bind_list)) {
+		list_for_each_entry(modules, &skl->bind_list, node) {
+			list_del(&modules->node);
+			kfree(modules);
+		}
+	}
+
 	snd_soc_unregister_component(dev);
 	snd_soc_unregister_platform(dev);
 	return 0;
sound/soc/intel/skylake/skl-topology.c +108 -1

146 lines not on this page · raw message

sound/soc/intel/skylake/skl-topology.h +6 -0
index 8536d03..cc64d6b 100644
@@ -336,6 +336,12 @@ struct skl_pipeline {
 	struct list_head node;
 };
 
+struct skl_module_deferred_bind {
+	struct skl_module_cfg *src;
+	struct skl_module_cfg *dst;
+	struct list_head node;
+};
+
 static inline struct skl *get_skl_ctx(struct device *dev)
 {
 	struct hdac_ext_bus *ebus = dev_get_drvdata(dev);
sound/soc/intel/skylake/skl.h +1 -0
index bbef77d2b..5b3fa4b 100644
@@ -77,6 +77,7 @@ struct skl {
 
 	struct skl_dsp_resource resource;
 	struct list_head ppl_list;
+	struct list_head bind_list;
 
 	const char *fw_name;
 	char tplg_name[64];
Signature · 1 line
2.5.0
Acked-by on the v2 series
#21 lore raw
In reply to [email protected]
Quoted 24 lines · On Fri, Mar 24, 2017 at 11:10:23PM +0530, [email protected] wrote:

From: Jeeja KP <[email protected]>

This patch series provides updates on the following:
o Fix LLCH register read.
o Add 16-bit constraint to FE bxt_rt298 machine.
o Fix DMA position reporting for capture stream.
o Use the sig_bits to define dai bps caps.
o Update sig_bits based on converter capability.
o Rearrangement of code to cleanup SKL SST library.
o Add support for deferred DSP module bind.

Changes in v2:
- Addressed Pierre's comments for "Fix DMA position reporting for capture stream" patch.

B, Jayachandran (1):
ALSA: hda: Fix LLCH register read

G Kranthi (1):
ASoC: Intel: Skylake: Add 16-bit constraint to FE bxt_rt298 machine

Hardik T Shah (1):
ASoC: Intel: Skylake: Fix DMA position reporting for capture stream

Quoted 6 lines

Jeeja KP (5):
ASoC: Intel: Skylake: Use the sig_bits to define dai bps capability
ASoC: hdac_hdmi: Update sig_bits based on converter capability
ASoC: Intel: Skylake: Rearrangement of code to cleanup SKL SST library
ASoC: Intel: Skylake: Fix module state after unbind and delete
ASoC: Intel: Skylake: Add support for deferred DSP module bind

All these:

Signature · 1 line
~Vinod