The patch

ASoC: SOF: add a field to store the current D0 substate of DSP

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 4c19030c511fd6eab029bae838f736256d2f43cd Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:40:57 -0500 Subject: [PATCH] ASoC: SOF: add a field to store the current D0 substate of DSP

Add field d0_substate to struct snd_sof_dev to store the current DSP D0 sub-state(only meaningful when DSP in D0), which could be D0I0 or D0I3.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-2-pierre-louis.bossart@linux.i... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/sof-priv.h | 9 +++++++++ 1 file changed, 9 insertions(+)

diff --git a/sound/soc/sof/sof-priv.h b/sound/soc/sof/sof-priv.h index 2d40de5ee285..481dfe4ee2d0 100644 --- a/sound/soc/sof/sof-priv.h +++ b/sound/soc/sof/sof-priv.h @@ -68,6 +68,12 @@ extern int sof_core_debug;

#define DMA_CHAN_INVALID 0xFFFFFFFF

+/* DSP D0ix sub-state */ +enum sof_d0_substate { + SOF_DSP_D0I0 = 0, /* DSP default D0 substate */ + SOF_DSP_D0I3, /* DSP D0i3(low power) substate*/ +}; + struct snd_sof_dev; struct snd_sof_ipc_msg; struct snd_sof_ipc; @@ -387,6 +393,9 @@ struct snd_sof_dev { */ struct snd_soc_component_driver plat_drv;

+ /* power states related */ + enum sof_d0_substate d0_substate; + /* DSP firmware boot */ wait_queue_head_t boot_wait; u32 boot_complete;

The patch

ASoC: SOF: reset default d0_substate at probe() and resume()

has been applied to the asoc tree at

https://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

From 09fe6b528886c0d07ce539b837749edcc46618d0 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:40:58 -0500 Subject: [PATCH] ASoC: SOF: reset default d0_substate at probe() and resume()

We initialize/reset d0_substate to default d0i0 value when doing transition D3-->D0, e.g. at success of probing and resuming.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-3-pierre-louis.bossart@linux.i... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/core.c | 3 +++ sound/soc/sof/pm.c | 3 +++ 2 files changed, 6 insertions(+)

diff --git a/sound/soc/sof/core.c b/sound/soc/sof/core.c index 5998861a9002..8661c2cca76b 100644 --- a/sound/soc/sof/core.c +++ b/sound/soc/sof/core.c @@ -458,6 +458,9 @@ int snd_sof_device_probe(struct device *dev, struct snd_sof_pdata *plat_data) /* initialize sof device */ sdev->dev = dev;

+ /* initialize default D0 sub-state */ + sdev->d0_substate = SOF_DSP_D0I0; + sdev->pdata = plat_data; sdev->first_boot = true; dev_set_drvdata(dev, sdev); diff --git a/sound/soc/sof/pm.c b/sound/soc/sof/pm.c index e23beaeefe00..81e623dfc7e5 100644 --- a/sound/soc/sof/pm.c +++ b/sound/soc/sof/pm.c @@ -326,6 +326,9 @@ static int sof_resume(struct device *dev, bool runtime_resume) "error: ctx_restore ipc error during resume %d\n", ret);

+ /* initialize default D0 sub-state */ + sdev->d0_substate = SOF_DSP_D0I0; + return ret; }

The patch

ASoC: SOF: add set_power_state() to dsp_ops for power state update

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From e8f112d8c29f44ded83f97828c104bf0904871ec Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:40:59 -0500 Subject: [PATCH] ASoC: SOF: add set_power_state() to dsp_ops for power state update

D0i3 is a platform-defined substate of D0, so we need a platform-specific callback in dsp_ops to handle the relevant configurations.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-4-pierre-louis.bossart@linux.i... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/ops.h | 9 +++++++++ sound/soc/sof/sof-priv.h | 2 ++ 2 files changed, 11 insertions(+)

diff --git a/sound/soc/sof/ops.h b/sound/soc/sof/ops.h index 824d36fe59fd..d7c8fc06f961 100644 --- a/sound/soc/sof/ops.h +++ b/sound/soc/sof/ops.h @@ -193,6 +193,15 @@ static inline int snd_sof_dsp_set_clk(struct snd_sof_dev *sdev, u32 freq) return 0; }

+static inline int snd_sof_dsp_set_power_state(struct snd_sof_dev *sdev, + enum sof_d0_substate substate) +{ + if (sof_ops(sdev)->set_power_state) + return sof_ops(sdev)->set_power_state(sdev, substate); + + return 0; +} + /* debug */ static inline void snd_sof_dsp_dbg_dump(struct snd_sof_dev *sdev, u32 flags) { diff --git a/sound/soc/sof/sof-priv.h b/sound/soc/sof/sof-priv.h index 481dfe4ee2d0..e9902e4e8e55 100644 --- a/sound/soc/sof/sof-priv.h +++ b/sound/soc/sof/sof-priv.h @@ -189,6 +189,8 @@ struct snd_sof_dsp_ops { int (*runtime_resume)(struct snd_sof_dev *sof_dev); /* optional */ int (*runtime_idle)(struct snd_sof_dev *sof_dev); /* optional */ int (*set_hw_params_upon_resume)(struct snd_sof_dev *sdev); /* optional */ + int (*set_power_state)(struct snd_sof_dev *sdev, + enum sof_d0_substate d0_substate); /* optional */

/* DSP clocking */ int (*set_clk)(struct snd_sof_dev *sof_dev, u32 freq); /* optional */

The patch

ASoC: SOF: Intel: hda-dsp: Add helper for setting DSP D0ix substate

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 62f8f76604623980d41cf73691ca45288871efd9 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:00 -0500 Subject: [PATCH] ASoC: SOF: Intel: hda-dsp: Add helper for setting DSP D0ix substate

Adding helper to implement setting dsp to d0i3 or d0i0 status, this will be needed for driver D0ix support.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-5-pierre-louis.bossart@linux.i... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/hda-dsp.c | 46 +++++++++++++++++++++++++++++++++++ sound/soc/sof/intel/hda.h | 10 ++++++++ 2 files changed, 56 insertions(+)

diff --git a/sound/soc/sof/intel/hda-dsp.c b/sound/soc/sof/intel/hda-dsp.c index 3ea401646e0c..fa2f1f66c72c 100644 --- a/sound/soc/sof/intel/hda-dsp.c +++ b/sound/soc/sof/intel/hda-dsp.c @@ -306,6 +306,52 @@ void hda_dsp_ipc_int_disable(struct snd_sof_dev *sdev) HDA_DSP_REG_HIPCCTL_BUSY | HDA_DSP_REG_HIPCCTL_DONE, 0); }

+static int hda_dsp_wait_d0i3c_done(struct snd_sof_dev *sdev, int retry) +{ + struct hdac_bus *bus = sof_to_bus(sdev); + + while (snd_hdac_chip_readb(bus, VS_D0I3C) & SOF_HDA_VS_D0I3C_CIP) { + if (!retry--) + return -ETIMEDOUT; + usleep_range(10, 15); + } + + return 0; +} + +int hda_dsp_set_power_state(struct snd_sof_dev *sdev, + enum sof_d0_substate d0_substate) +{ + struct hdac_bus *bus = sof_to_bus(sdev); + int retry = 50; + int ret; + u8 value; + + /* Write to D0I3C after Command-In-Progress bit is cleared */ + ret = hda_dsp_wait_d0i3c_done(sdev, retry); + if (ret < 0) { + dev_err(bus->dev, "CIP timeout before update D0I3C!\n"); + return ret; + } + + /* Update D0I3C register */ + value = d0_substate == SOF_DSP_D0I3 ? SOF_HDA_VS_D0I3C_I3 : 0; + snd_hdac_chip_updateb(bus, VS_D0I3C, SOF_HDA_VS_D0I3C_I3, value); + + /* Wait for cmd in progress to be cleared before exiting the function */ + retry = 50; + ret = hda_dsp_wait_d0i3c_done(sdev, retry); + if (ret < 0) { + dev_err(bus->dev, "CIP timeout after D0I3C updated!\n"); + return ret; + } + + dev_vdbg(bus->dev, "D0I3C updated, register = 0x%x\n", + snd_hdac_chip_readb(bus, VS_D0I3C)); + + return 0; +} + static int hda_suspend(struct snd_sof_dev *sdev, bool runtime_suspend) { struct sof_intel_hda_dev *hda = sdev->pdata->hw_pdata; diff --git a/sound/soc/sof/intel/hda.h b/sound/soc/sof/intel/hda.h index ea02bf40cb25..0e7c366b8f71 100644 --- a/sound/soc/sof/intel/hda.h +++ b/sound/soc/sof/intel/hda.h @@ -64,6 +64,13 @@ #define SOF_HDA_PPCTL_PIE BIT(31) #define SOF_HDA_PPCTL_GPROCEN BIT(30)

+/*Vendor Specific Registers*/ +#define SOF_HDA_VS_D0I3C 0x104A + +/* D0I3C Register fields */ +#define SOF_HDA_VS_D0I3C_CIP BIT(0) /* Command-In-Progress */ +#define SOF_HDA_VS_D0I3C_I3 BIT(2) /* D0i3 enable bit */ + /* DPIB entry size: 8 Bytes = 2 DWords */ #define SOF_HDA_DPIB_ENTRY_SIZE 0x8

@@ -455,6 +462,9 @@ int hda_dsp_core_reset_power_down(struct snd_sof_dev *sdev, void hda_dsp_ipc_int_enable(struct snd_sof_dev *sdev); void hda_dsp_ipc_int_disable(struct snd_sof_dev *sdev);

+int hda_dsp_set_power_state(struct snd_sof_dev *sdev, + enum sof_d0_substate d0_substate); + int hda_dsp_suspend(struct snd_sof_dev *sdev); int hda_dsp_resume(struct snd_sof_dev *sdev); int hda_dsp_runtime_suspend(struct snd_sof_dev *sdev);

On Tue, Oct 29, 2019 at 3:10 AM Cezary Rojewski cezary.rojewski@intel.com wrote:

On 2019-10-26 00:41, Pierre-Louis Bossart wrote:

...

From: Keyon Jie yang.jie@linux.intel.com

Adding helper to implement setting dsp to d0i3 or d0i0 status, this will be needed for driver D0ix support.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart <

pierre-louis.bossart@linux.intel.com>

...

+static int hda_dsp_wait_d0i3c_done(struct snd_sof_dev *sdev, int retry) +{

• struct hdac_bus *bus = sof_to_bus(sdev);
  

• while (snd_hdac_chip_readb(bus, VS_D0I3C) & SOF_HDA_VS_D0I3C_CIP) {
  

•         if (!retry--)
  

•                 return -ETIMEDOUT;
  

•         usleep_range(10, 15);
  

• }
  

• return 0;
  

+}

+int hda_dsp_set_power_state(struct snd_sof_dev *sdev,

•                     enum sof_d0_substate d0_substate)
  

+{

• struct hdac_bus *bus = sof_to_bus(sdev);
  

• int retry = 50;
  

• int ret;
  

• u8 value;
  

• /* Write to D0I3C after Command-In-Progress bit is cleared */
  

• ret = hda_dsp_wait_d0i3c_done(sdev, retry);
  

• if (ret < 0) {
  

•         dev_err(bus->dev, "CIP timeout before update D0I3C!\n");
  

•         return ret;
  

• }
  

• /* Update D0I3C register */
  

• value = d0_substate == SOF_DSP_D0I3 ? SOF_HDA_VS_D0I3C_I3 : 0;
  

Some indentation or parenthesis would make this cleaner.

...

• snd_hdac_chip_updateb(bus, VS_D0I3C, SOF_HDA_VS_D0I3C_I3, value);
  

• /* Wait for cmd in progress to be cleared before exiting the
  

function */

...

• retry = 50;
  

• ret = hda_dsp_wait_d0i3c_done(sdev, retry);
  

• if (ret < 0) {
  

•         dev_err(bus->dev, "CIP timeout after D0I3C updated!\n");
  

•         return ret;
  

• }
  

• dev_vdbg(bus->dev, "D0I3C updated, register = 0x%x\n",
  

•          snd_hdac_chip_readb(bus, VS_D0I3C));
  

• return 0;
  

+}

I believe hda_dsp_wait_d0i3c_done function could have had its argument list simplified. "retry" is passed externally and it is always set to 50. One could put the "retry" right inside _done function. This or allow the caller to modify the sleep period. Another option is to replace "retry" with "timeout period" (total timeout, that is) entirely.

Cezary,

This has been fixed later in the series to use the HDA_DSP_REG_POLL_RETRY_COUNT but I agree that this can further be simplified to just use the macro inside the hda_dsp_wait_d0i3c_done() instead of passing the argument.

In regard to maintenance, both ret checks (resulting in dev_errs) assume -ETIMEOUT outcome on _done failure. If said function gets updated in the future, these implicit checks might cause problems.

Possibly, but at the moment -ETIMEOUT looks like the correct error to be reported.

Thanks, Ranjani

...

static int hda_suspend(struct snd_sof_dev *sdev, bool runtime_suspend) { struct sof_intel_hda_dev *hda = sdev->pdata->hw_pdata; diff --git a/sound/soc/sof/intel/hda.h b/sound/soc/sof/intel/hda.h index ea02bf40cb25..0e7c366b8f71 100644 --- a/sound/soc/sof/intel/hda.h +++ b/sound/soc/sof/intel/hda.h @@ -64,6 +64,13 @@ #define SOF_HDA_PPCTL_PIE BIT(31) #define SOF_HDA_PPCTL_GPROCEN BIT(30)

+/*Vendor Specific Registers*/

Missing spaces.

...

+#define SOF_HDA_VS_D0I3C 0x104A

---

Alsa-devel mailing list Alsa-devel@alsa-project.org https://mailman.alsa-project.org/mailman/listinfo/alsa-devel

The patch

ASoC: SOF: Intel: CNL: add set_power_state() ops

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 73b51957122dc47a40325a5359b44ef337cbf217 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:01 -0500 Subject: [PATCH] ASoC: SOF: Intel: CNL: add set_power_state() ops

Using hda_dsp_set_power_state() as set_power_state() ops for cnl to do d0ix platform configuration updates.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-6-pierre-louis.bossart@linux.i... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/cnl.c | 1 + 1 file changed, 1 insertion(+)

diff --git a/sound/soc/sof/intel/cnl.c b/sound/soc/sof/intel/cnl.c index e9cdcc040fb5..982b81a0b13a 100644 --- a/sound/soc/sof/intel/cnl.c +++ b/sound/soc/sof/intel/cnl.c @@ -255,6 +255,7 @@ const struct snd_sof_dsp_ops sof_cnl_ops = { .runtime_resume = hda_dsp_runtime_resume, .runtime_idle = hda_dsp_runtime_idle, .set_hw_params_upon_resume = hda_dsp_set_hw_params_upon_resume, + .set_power_state = hda_dsp_set_power_state,

/* ALSA HW info flags */ .hw_info = SNDRV_PCM_INFO_MMAP |

The patch

ASoC: SOF: Intel: APL: add set_power_state() ops

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 35c930ba6025964ac71f041065ae212a0fcc7cca Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:02 -0500 Subject: [PATCH] ASoC: SOF: Intel: APL: add set_power_state() ops

Using hda_dsp_set_power_state() as set_power_state() ops for apl to do d0ix platform configuration updates.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-7-pierre-louis.bossart@linux.i... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/apl.c | 1 + 1 file changed, 1 insertion(+)

diff --git a/sound/soc/sof/intel/apl.c b/sound/soc/sof/intel/apl.c index 15d26e8d90a9..7daa8eb456c8 100644 --- a/sound/soc/sof/intel/apl.c +++ b/sound/soc/sof/intel/apl.c @@ -97,6 +97,7 @@ const struct snd_sof_dsp_ops sof_apl_ops = { .runtime_resume = hda_dsp_runtime_resume, .runtime_idle = hda_dsp_runtime_idle, .set_hw_params_upon_resume = hda_dsp_set_hw_params_upon_resume, + .set_power_state = hda_dsp_set_power_state,

/* ALSA HW info flags */ .hw_info = SNDRV_PCM_INFO_MMAP |

The patch

ASoC: SOF: add flag to snd_sof_pcm_stream for D0i3 compatible stream

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From e5c97e88084b8dca6850eba5d2937716b88306c1 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:03 -0500 Subject: [PATCH] ASoC: SOF: add flag to snd_sof_pcm_stream for D0i3 compatible stream

Add flag d0i3_compatible to struct snd_sof_pcm_stream to denote if the stream can tolerate a transition to the D0i3 substate while opened (thus seen as 'active' by pm_runtime).

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-8-pierre-louis.bossart@linux.i... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/sof-priv.h | 1 + 1 file changed, 1 insertion(+)

diff --git a/sound/soc/sof/sof-priv.h b/sound/soc/sof/sof-priv.h index e9902e4e8e55..7a21a45d3635 100644 --- a/sound/soc/sof/sof-priv.h +++ b/sound/soc/sof/sof-priv.h @@ -310,6 +310,7 @@ struct snd_sof_pcm_stream { struct sof_ipc_stream_posn posn; struct snd_pcm_substream *substream; struct work_struct period_elapsed_work; + bool d0i3_compatible; /* DSP can be in D0I3 when this pcm is opened */ };

/* ALSA SOF PCM device */

The patch

ASoC: SOF: token: add tokens for PCM compatible with D0i3 substate

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 4a94940988cc44adeb383401dea0beeac4abbe63 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:04 -0500 Subject: [PATCH] ASoC: SOF: token: add tokens for PCM compatible with D0i3 substate

Add stream token SOF_TKN_STREAM_PLAYBACK_COMPATIBLE_D0I3 and SOF_TKN_STREAM_CAPTURE_COMPATIBLE_D0I3 to denote if the stream can be opened at low power d0i3 status or not.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-9-pierre-louis.bossart@linux.i... Signed-off-by: Mark Brown broonie@kernel.org --- include/uapi/sound/sof/tokens.h | 4 ++++ sound/soc/sof/topology.c | 10 ++++++++++ 2 files changed, 14 insertions(+)

diff --git a/include/uapi/sound/sof/tokens.h b/include/uapi/sound/sof/tokens.h index d65406f34361..76883e6fb750 100644 --- a/include/uapi/sound/sof/tokens.h +++ b/include/uapi/sound/sof/tokens.h @@ -113,6 +113,10 @@ /* ESAI */ #define SOF_TKN_IMX_ESAI_MCLK_ID 1100

+/* Stream */ +#define SOF_TKN_STREAM_PLAYBACK_COMPATIBLE_D0I3 1200 +#define SOF_TKN_STREAM_CAPTURE_COMPATIBLE_D0I3 1201 + /* Led control for mute switches */ #define SOF_TKN_MUTE_LED_USE 1300 #define SOF_TKN_MUTE_LED_DIRECTION 1301 diff --git a/sound/soc/sof/topology.c b/sound/soc/sof/topology.c index b3ab6f23fc93..121e5d6f8477 100644 --- a/sound/soc/sof/topology.c +++ b/sound/soc/sof/topology.c @@ -567,6 +567,16 @@ static const struct sof_topology_token pcm_tokens[] = { offsetof(struct sof_ipc_comp_host, dmac_config), 0}, };

+/* PCM */ +static const struct sof_topology_token stream_tokens[] = { + {SOF_TKN_STREAM_PLAYBACK_COMPATIBLE_D0I3, + SND_SOC_TPLG_TUPLE_TYPE_BOOL, get_token_u16, + offsetof(struct snd_sof_pcm, stream[0].d0i3_compatible), 0}, + {SOF_TKN_STREAM_CAPTURE_COMPATIBLE_D0I3, + SND_SOC_TPLG_TUPLE_TYPE_BOOL, get_token_u16, + offsetof(struct snd_sof_pcm, stream[1].d0i3_compatible), 0}, +}; + /* Generic components */ static const struct sof_topology_token comp_tokens[] = { {SOF_TKN_COMP_PERIOD_SINK_COUNT,

The patch

ASoC: SOF: topology: parse and store d0i3_compatible flag

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From c5232c0171428f005a3204e1c264231fb5999b28 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:05 -0500 Subject: [PATCH] ASoC: SOF: topology: parse and store d0i3_compatible flag

Parses the token from tplg file and store it to snd_sof_pcm_stream d0i3_compatible flag, which can be used later for d0ix transition management.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-10-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/topology.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)

diff --git a/sound/soc/sof/topology.c b/sound/soc/sof/topology.c index 121e5d6f8477..cc138ac6a8eb 100644 --- a/sound/soc/sof/topology.c +++ b/sound/soc/sof/topology.c @@ -2307,6 +2307,7 @@ static int sof_dai_load(struct snd_soc_component *scomp, int index, { struct snd_sof_dev *sdev = snd_soc_component_get_drvdata(scomp); struct snd_soc_tplg_stream_caps *caps; + struct snd_soc_tplg_private *private = &pcm->priv; struct snd_sof_pcm *spcm; int stream = SNDRV_PCM_STREAM_PLAYBACK; int ret = 0; @@ -2329,10 +2330,22 @@ static int sof_dai_load(struct snd_soc_component *scomp, int index, dai_drv->dobj.private = spcm; list_add(&spcm->list, &sdev->pcm_list);

+ ret = sof_parse_tokens(scomp, spcm, stream_tokens, + ARRAY_SIZE(stream_tokens), private->array, + le32_to_cpu(private->size)); + if (ret) { + dev_err(sdev->dev, "error: parse stream tokens failed %d\n", + le32_to_cpu(private->size)); + return ret; + } + /* do we need to allocate playback PCM DMA pages */ if (!spcm->pcm.playback) goto capture;

+ dev_vdbg(sdev->dev, "tplg: pcm %s stream tokens: playback d0i3:%d\n", + spcm->pcm.pcm_name, spcm->stream[0].d0i3_compatible); + caps = &spcm->pcm.caps[stream];

/* allocate playback page table buffer */ @@ -2360,6 +2373,9 @@ static int sof_dai_load(struct snd_soc_component *scomp, int index, if (!spcm->pcm.capture) return ret;

+ dev_vdbg(sdev->dev, "tplg: pcm %s stream tokens: capture d0i3:%d\n", + spcm->pcm.pcm_name, spcm->stream[1].d0i3_compatible); + caps = &spcm->pcm.caps[stream];

/* allocate capture page table buffer */

The patch

ASoC: SOF: Intel: hda-dsp: align the comments for D0I3C update

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From aae7c82d01219bf568a7b12f5839d7ca52b17d59 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:06 -0500 Subject: [PATCH] ASoC: SOF: Intel: hda-dsp: align the comments for D0I3C update

Align the logs for CIP timeout at D0I3C.I3 updating.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-11-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/hda-dsp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/sound/soc/sof/intel/hda-dsp.c b/sound/soc/sof/intel/hda-dsp.c index fa2f1f66c72c..74805a066183 100644 --- a/sound/soc/sof/intel/hda-dsp.c +++ b/sound/soc/sof/intel/hda-dsp.c @@ -330,7 +330,7 @@ int hda_dsp_set_power_state(struct snd_sof_dev *sdev, /* Write to D0I3C after Command-In-Progress bit is cleared */ ret = hda_dsp_wait_d0i3c_done(sdev, retry); if (ret < 0) { - dev_err(bus->dev, "CIP timeout before update D0I3C!\n"); + dev_err(bus->dev, "CIP timeout before D0I3C update!\n"); return ret; }

@@ -342,7 +342,7 @@ int hda_dsp_set_power_state(struct snd_sof_dev *sdev, retry = 50; ret = hda_dsp_wait_d0i3c_done(sdev, retry); if (ret < 0) { - dev_err(bus->dev, "CIP timeout after D0I3C updated!\n"); + dev_err(bus->dev, "CIP timeout after D0I3C update!\n"); return ret; }

The patch

ASoC: SOF: Intel: HDA: use macro for register polling retry count

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 92f4beb718d76e93b76343a3ba872df6cb210672 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:07 -0500 Subject: [PATCH] ASoC: SOF: Intel: HDA: use macro for register polling retry count

Define macro and use it for the register polling retry count.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-12-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/hda-dsp.c | 6 ++---- sound/soc/sof/intel/hda.h | 1 + 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/sound/soc/sof/intel/hda-dsp.c b/sound/soc/sof/intel/hda-dsp.c index 74805a066183..936361bd25e9 100644 --- a/sound/soc/sof/intel/hda-dsp.c +++ b/sound/soc/sof/intel/hda-dsp.c @@ -323,12 +323,11 @@ int hda_dsp_set_power_state(struct snd_sof_dev *sdev, enum sof_d0_substate d0_substate) { struct hdac_bus *bus = sof_to_bus(sdev); - int retry = 50; int ret; u8 value;

/* Write to D0I3C after Command-In-Progress bit is cleared */ - ret = hda_dsp_wait_d0i3c_done(sdev, retry); + ret = hda_dsp_wait_d0i3c_done(sdev, HDA_DSP_REG_POLL_RETRY_COUNT); if (ret < 0) { dev_err(bus->dev, "CIP timeout before D0I3C update!\n"); return ret; @@ -339,8 +338,7 @@ int hda_dsp_set_power_state(struct snd_sof_dev *sdev, snd_hdac_chip_updateb(bus, VS_D0I3C, SOF_HDA_VS_D0I3C_I3, value);

/* Wait for cmd in progress to be cleared before exiting the function */ - retry = 50; - ret = hda_dsp_wait_d0i3c_done(sdev, retry); + ret = hda_dsp_wait_d0i3c_done(sdev, HDA_DSP_REG_POLL_RETRY_COUNT); if (ret < 0) { dev_err(bus->dev, "CIP timeout after D0I3C update!\n"); return ret; diff --git a/sound/soc/sof/intel/hda.h b/sound/soc/sof/intel/hda.h index 0e7c366b8f71..99ec60218c16 100644 --- a/sound/soc/sof/intel/hda.h +++ b/sound/soc/sof/intel/hda.h @@ -214,6 +214,7 @@ #define HDA_DSP_CTRL_RESET_TIMEOUT 100 #define HDA_DSP_WAIT_TIMEOUT 500 /* 500 msec */ #define HDA_DSP_REG_POLL_INTERVAL_US 500 /* 0.5 msec */ +#define HDA_DSP_REG_POLL_RETRY_COUNT 50

#define HDA_DSP_ADSPIC_IPC 1 #define HDA_DSP_ADSPIS_IPC 1

The patch

ASoC: SOF: PM: rename sof_send_pm_ipc to sof_send_pm_ctx_ipc

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 7c7eba2402c11137d1fa1d3fb964a2c6fc1ded3e Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:08 -0500 Subject: [PATCH] ASoC: SOF: PM: rename sof_send_pm_ipc to sof_send_pm_ctx_ipc

The helper sof_send_pm_ipc() is only suitable for context save/restore IPCs' sending, so rename it to sof_send_pm_ctx_ipc here.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-13-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/pm.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/sound/soc/sof/pm.c b/sound/soc/sof/pm.c index 81e623dfc7e5..ac900fb3379e 100644 --- a/sound/soc/sof/pm.c +++ b/sound/soc/sof/pm.c @@ -197,7 +197,7 @@ static int sof_restore_pipelines(struct snd_sof_dev *sdev) return ret; }

-static int sof_send_pm_ipc(struct snd_sof_dev *sdev, int cmd) +static int sof_send_pm_ctx_ipc(struct snd_sof_dev *sdev, int cmd) { struct sof_ipc_pm_ctx pm_ctx; struct sof_ipc_reply reply; @@ -320,7 +320,7 @@ static int sof_resume(struct device *dev, bool runtime_resume) }

/* notify DSP of system resume */ - ret = sof_send_pm_ipc(sdev, SOF_IPC_PM_CTX_RESTORE); + ret = sof_send_pm_ctx_ipc(sdev, SOF_IPC_PM_CTX_RESTORE); if (ret < 0) dev_err(sdev->dev, "error: ctx_restore ipc error during resume %d\n", @@ -361,7 +361,7 @@ static int sof_suspend(struct device *dev, bool runtime_suspend) sof_cache_debugfs(sdev); #endif /* notify DSP of upcoming power down */ - ret = sof_send_pm_ipc(sdev, SOF_IPC_PM_CTX_SAVE); + ret = sof_send_pm_ctx_ipc(sdev, SOF_IPC_PM_CTX_SAVE); if (ret == -EBUSY || ret == -EAGAIN) { /* * runtime PM has logic to handle -EBUSY/-EAGAIN so

The patch

ASoC: SOF: ipc: introduce message for DSP power gating

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 10992004e1e960c7e256ec70c7ab22895a7003d7 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:09 -0500 Subject: [PATCH] ASoC: SOF: ipc: introduce message for DSP power gating

Add new ipc messages which will be sent from driver to FW, to ask FW to enter specific power saving state.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-14-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- include/sound/sof/header.h | 1 + include/sound/sof/pm.h | 8 ++++++++ 2 files changed, 9 insertions(+)

diff --git a/include/sound/sof/header.h b/include/sound/sof/header.h index 332143ff7278..bf3edd9c08b4 100644 --- a/include/sound/sof/header.h +++ b/include/sound/sof/header.h @@ -75,6 +75,7 @@ #define SOF_IPC_PM_CLK_GET SOF_CMD_TYPE(0x005) #define SOF_IPC_PM_CLK_REQ SOF_CMD_TYPE(0x006) #define SOF_IPC_PM_CORE_ENABLE SOF_CMD_TYPE(0x007) +#define SOF_IPC_PM_GATE SOF_CMD_TYPE(0x008)

/* component runtime config - multiple different types */ #define SOF_IPC_COMP_SET_VALUE SOF_CMD_TYPE(0x001) diff --git a/include/sound/sof/pm.h b/include/sound/sof/pm.h index 003879401d63..3cf2e0f39d94 100644 --- a/include/sound/sof/pm.h +++ b/include/sound/sof/pm.h @@ -45,4 +45,12 @@ struct sof_ipc_pm_core_config { uint32_t enable_mask; } __packed;

+struct sof_ipc_pm_gate { + struct sof_ipc_cmd_hdr hdr; + uint32_t flags; /* platform specific */ + + /* reserved for future use */ + uint32_t reserved[5]; +} __packed; + #endif

The patch

ASoC: SOF: Intel: hda-ipc: Don't read mailbox for PM_GATE reply

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 463fbf6dc03aff41e973602bbad6cb135fa6d48a Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:10 -0500 Subject: [PATCH] ASoC: SOF: Intel: hda-ipc: Don't read mailbox for PM_GATE reply

Memory windows could be powered off before receiving PM_GATE IPC reply from FW, we can't read the mailbox to get reply.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-15-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/hda-ipc.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/sound/soc/sof/intel/hda-ipc.c b/sound/soc/sof/intel/hda-ipc.c index 6aae6f18b3dc..0fd2153c1769 100644 --- a/sound/soc/sof/intel/hda-ipc.c +++ b/sound/soc/sof/intel/hda-ipc.c @@ -83,10 +83,12 @@ void hda_dsp_ipc_get_reply(struct snd_sof_dev *sdev) }

hdr = msg->msg_data; - if (hdr->cmd == (SOF_IPC_GLB_PM_MSG | SOF_IPC_PM_CTX_SAVE)) { + if (hdr->cmd == (SOF_IPC_GLB_PM_MSG | SOF_IPC_PM_CTX_SAVE) || + hdr->cmd == (SOF_IPC_GLB_PM_MSG | SOF_IPC_PM_GATE)) { /* * memory windows are powered off before sending IPC reply, - * so we can't read the mailbox for CTX_SAVE reply. + * so we can't read the mailbox for CTX_SAVE and PM_GATE + * replies. */ reply.error = 0; reply.hdr.cmd = SOF_IPC_GLB_REPLY;

The patch

ASoC: SOF: Intel: HDA: add cAVS specific compact IPC header file

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 5056193d4d1a8b73087145add21141c46027d6a2 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:11 -0500 Subject: [PATCH] ASoC: SOF: Intel: HDA: add cAVS specific compact IPC header file

On cAVS platforms, some IPCs are required to be sent via IPC registers only(e.g. when in D0i3, mailbox is unaccessible), add hda-ipc.h to hold definition of those compact IPCs.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-16-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/hda-ipc.h | 51 +++++++++++++++++++++++++++++++++++ 1 file changed, 51 insertions(+) create mode 100644 sound/soc/sof/intel/hda-ipc.h

diff --git a/sound/soc/sof/intel/hda-ipc.h b/sound/soc/sof/intel/hda-ipc.h new file mode 100644 index 000000000000..aef0ceac9803 --- /dev/null +++ b/sound/soc/sof/intel/hda-ipc.h @@ -0,0 +1,51 @@ +/* SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause) */ +/* + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * Copyright(c) 2019 Intel Corporation. All rights reserved. + * + * Author: Keyon Jie yang.jie@linux.intel.com + */ + +#ifndef __SOF_INTEL_HDA_IPC_H +#define __SOF_INTEL_HDA_IPC_H + +/* + * Primary register, mapped to + * - DIPCTDR (HIPCIDR) in sideband IPC (cAVS 1.8+) + * - DIPCT in cAVS 1.5 IPC + * + * Secondary register, mapped to: + * - DIPCTDD (HIPCIDD) in sideband IPC (cAVS 1.8+) + * - DIPCTE in cAVS 1.5 IPC + */ + +/* Common bits in primary register */ + +/* Reserved for doorbell */ +#define HDA_IPC_RSVD_31 BIT(31) +/* Target, 0 - normal message, 1 - compact message(cAVS compatible) */ +#define HDA_IPC_MSG_COMPACT BIT(30) +/* Direction, 0 - request, 1 - response */ +#define HDA_IPC_RSP BIT(29) + +#define HDA_IPC_TYPE_SHIFT 24 +#define HDA_IPC_TYPE_MASK GENMASK(28, 24) +#define HDA_IPC_TYPE(x) ((x) << HDA_IPC_TYPE_SHIFT) + +#define HDA_IPC_PM_GATE HDA_IPC_TYPE(0x8U) + +/* Command specific payload bits in secondary register */ + +/* Disable DMA tracing (0 - keep tracing, 1 - to disable DMA trace) */ +#define HDA_PM_NO_DMA_TRACE BIT(4) +/* Prevent clock gating (0 - cg allowed, 1 - DSP clock always on) */ +#define HDA_PM_PCG BIT(3) +/* Prevent power gating (0 - deep power state transitions allowed) */ +#define HDA_PM_PPG BIT(2) +/* Indicates whether streaming is active */ +#define HDA_PM_PG_STREAMING BIT(1) +#define HDA_PM_PG_RSVD BIT(0) + +#endif

The patch

ASoC: SOF: configure D0ix IPC flags in set_power_state

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 534037fddd34b58be86a826d449a5a6635ecdbf5 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:12 -0500 Subject: [PATCH] ASoC: SOF: configure D0ix IPC flags in set_power_state

The configuration for D0ix in FW is platform specific, let's do this and send IPC in the platform set_power_state() ops.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-17-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/hda-dsp.c | 32 +++++++++++++++++++++++++++++++- 1 file changed, 31 insertions(+), 1 deletion(-)

diff --git a/sound/soc/sof/intel/hda-dsp.c b/sound/soc/sof/intel/hda-dsp.c index 936361bd25e9..b5070409a5e3 100644 --- a/sound/soc/sof/intel/hda-dsp.c +++ b/sound/soc/sof/intel/hda-dsp.c @@ -19,6 +19,7 @@ #include <sound/hda_register.h> #include "../ops.h" #include "hda.h" +#include "hda-ipc.h"

/* * DSP Core control. @@ -319,10 +320,28 @@ static int hda_dsp_wait_d0i3c_done(struct snd_sof_dev *sdev, int retry) return 0; }

+static int hda_dsp_send_pm_gate_ipc(struct snd_sof_dev *sdev, u32 flags) +{ + struct sof_ipc_pm_gate pm_gate; + struct sof_ipc_reply reply; + + memset(&pm_gate, 0, sizeof(pm_gate)); + + /* configure pm_gate ipc message */ + pm_gate.hdr.size = sizeof(pm_gate); + pm_gate.hdr.cmd = SOF_IPC_GLB_PM_MSG | SOF_IPC_PM_GATE; + pm_gate.flags = flags; + + /* send pm_gate ipc to dsp */ + return sof_ipc_tx_message(sdev->ipc, pm_gate.hdr.cmd, &pm_gate, + sizeof(pm_gate), &reply, sizeof(reply)); +} + int hda_dsp_set_power_state(struct snd_sof_dev *sdev, enum sof_d0_substate d0_substate) { struct hdac_bus *bus = sof_to_bus(sdev); + u32 flags; int ret; u8 value;

@@ -347,7 +366,18 @@ int hda_dsp_set_power_state(struct snd_sof_dev *sdev, dev_vdbg(bus->dev, "D0I3C updated, register = 0x%x\n", snd_hdac_chip_readb(bus, VS_D0I3C));

- return 0; + if (d0_substate == SOF_DSP_D0I0) + flags = HDA_PM_PPG;/* prevent power gating in D0 */ + else + flags = HDA_PM_NO_DMA_TRACE;/* disable DMA trace in D0I3*/ + + /* sending pm_gate IPC */ + ret = hda_dsp_send_pm_gate_ipc(sdev, flags); + if (ret < 0) + dev_err(sdev->dev, + "error: PM_GATE ipc error %d\n", ret); + + return ret; }

static int hda_suspend(struct snd_sof_dev *sdev, bool runtime_suspend)

The patch

ASoC: SOF: Intel: CNL: add support for sending compact IPC

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 0267de58acfe5059ace739741f1533dd605ed22f Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:14 -0500 Subject: [PATCH] ASoC: SOF: Intel: CNL: add support for sending compact IPC

For compact IPCs, we will send the IPC header/command via the HIPCIDR register and the first 32bit payload via the HIPCIDD register, no mailbox will be used.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-19-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/cnl.c | 42 ++++++++++++++++++++++++++++++++++----- sound/soc/sof/intel/hda.h | 1 + 2 files changed, 38 insertions(+), 5 deletions(-)

diff --git a/sound/soc/sof/intel/cnl.c b/sound/soc/sof/intel/cnl.c index 982b81a0b13a..0e1e265f3f3b 100644 --- a/sound/soc/sof/intel/cnl.c +++ b/sound/soc/sof/intel/cnl.c @@ -17,6 +17,7 @@

#include "../ops.h" #include "hda.h" +#include "hda-ipc.h"

static const struct snd_sof_debugfs_map cnl_dsp_debugfs[] = { {"hda", HDA_DSP_HDA_BAR, 0, 0x4000, SOF_DEBUGFS_ACCESS_ALWAYS}, @@ -150,14 +151,45 @@ static void cnl_ipc_dsp_done(struct snd_sof_dev *sdev) CNL_DSP_REG_HIPCCTL_DONE); }

+static bool cnl_compact_ipc_compress(struct snd_sof_ipc_msg *msg, + u32 *dr, u32 *dd) +{ + struct sof_ipc_pm_gate *pm_gate; + + if (msg->header == (SOF_IPC_GLB_PM_MSG | SOF_IPC_PM_GATE)) { + pm_gate = msg->msg_data; + + /* send the compact message via the primary register */ + *dr = HDA_IPC_MSG_COMPACT | HDA_IPC_PM_GATE; + + /* send payload via the extended data register */ + *dd = pm_gate->flags; + + return true; + } + + return false; +} + static int cnl_ipc_send_msg(struct snd_sof_dev *sdev, struct snd_sof_ipc_msg *msg) { - /* send the message */ - sof_mailbox_write(sdev, sdev->host_box.offset, msg->msg_data, - msg->msg_size); - snd_sof_dsp_write(sdev, HDA_DSP_BAR, CNL_DSP_REG_HIPCIDR, - CNL_DSP_REG_HIPCIDR_BUSY); + u32 dr = 0; + u32 dd = 0; + + if (cnl_compact_ipc_compress(msg, &dr, &dd)) { + /* send the message via IPC registers */ + snd_sof_dsp_write(sdev, HDA_DSP_BAR, CNL_DSP_REG_HIPCIDD, + dd); + snd_sof_dsp_write(sdev, HDA_DSP_BAR, CNL_DSP_REG_HIPCIDR, + CNL_DSP_REG_HIPCIDR_BUSY | dr); + } else { + /* send the message via mailbox */ + sof_mailbox_write(sdev, sdev->host_box.offset, msg->msg_data, + msg->msg_size); + snd_sof_dsp_write(sdev, HDA_DSP_BAR, CNL_DSP_REG_HIPCIDR, + CNL_DSP_REG_HIPCIDR_BUSY); + }

return 0; } diff --git a/sound/soc/sof/intel/hda.h b/sound/soc/sof/intel/hda.h index 99ec60218c16..52a87a47029d 100644 --- a/sound/soc/sof/intel/hda.h +++ b/sound/soc/sof/intel/hda.h @@ -312,6 +312,7 @@ #define CNL_DSP_REG_HIPCTDD (CNL_DSP_IPC_BASE + 0x08) #define CNL_DSP_REG_HIPCIDR (CNL_DSP_IPC_BASE + 0x10) #define CNL_DSP_REG_HIPCIDA (CNL_DSP_IPC_BASE + 0x14) +#define CNL_DSP_REG_HIPCIDD (CNL_DSP_IPC_BASE + 0x18) #define CNL_DSP_REG_HIPCCTL (CNL_DSP_IPC_BASE + 0x28)

/* HIPCI */

The patch

ASoC: SOF: add a flag to indicate the system suspend target

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 7367d3096b1443e55a2c730d1966f423b15d5cad Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:15 -0500 Subject: [PATCH] ASoC: SOF: add a flag to indicate the system suspend target

Add flag 's0_suspend' to indicate if the system is entering S0ix or not.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-20-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/sof-priv.h | 2 ++ 1 file changed, 2 insertions(+)

diff --git a/sound/soc/sof/sof-priv.h b/sound/soc/sof/sof-priv.h index 2231c673c678..e715673f20e5 100644 --- a/sound/soc/sof/sof-priv.h +++ b/sound/soc/sof/sof-priv.h @@ -398,6 +398,8 @@ struct snd_sof_dev {

/* power states related */ enum sof_d0_substate d0_substate; + /* flag to track if the intended power target of suspend is S0ix */ + bool s0_suspend;

/* DSP firmware boot */ wait_queue_head_t boot_wait;

The patch

ASoC: SOF: add a flag suspend_ignored for sof stream

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 4cd933abd4820da7eefe672ff439b32d199a07be Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:16 -0500 Subject: [PATCH] ASoC: SOF: add a flag suspend_ignored for sof stream

Add a suspend_ignored flag to snd_sof_pcm_stream that will be used to decide if the corresponding FW pipeline should be kept active to perform always on tasks when the system is entering the S0ix state.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-21-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/sof-priv.h | 5 +++++ 1 file changed, 5 insertions(+)

diff --git a/sound/soc/sof/sof-priv.h b/sound/soc/sof/sof-priv.h index e715673f20e5..c2541d020bde 100644 --- a/sound/soc/sof/sof-priv.h +++ b/sound/soc/sof/sof-priv.h @@ -311,6 +311,11 @@ struct snd_sof_pcm_stream { struct snd_pcm_substream *substream; struct work_struct period_elapsed_work; bool d0i3_compatible; /* DSP can be in D0I3 when this pcm is opened */ + /* + * flag to indicate that the DSP pipelines should be kept + * active or not while suspending the stream + */ + bool suspend_ignored; };

/* ALSA SOF PCM device */

The patch

ASoC: SOF: PM: implement prepare/complete callbacks

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 0b50b3b1c3bc2a2c9eeab418b3de3e60e0530cf4 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:17 -0500 Subject: [PATCH] ASoC: SOF: PM: implement prepare/complete callbacks

Implement the prepare() and complete() callbacks for power management, initialize s0_suspend flag at prepare(), and reset it at complete().

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-22-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/pm.c | 23 +++++++++++++++++++++++ sound/soc/sof/sof-priv.h | 2 ++ 2 files changed, 25 insertions(+)

diff --git a/sound/soc/sof/pm.c b/sound/soc/sof/pm.c index 584241e9734a..99e4e6ffff74 100644 --- a/sound/soc/sof/pm.c +++ b/sound/soc/sof/pm.c @@ -439,3 +439,26 @@ int snd_sof_suspend(struct device *dev) return sof_suspend(dev, false); } EXPORT_SYMBOL(snd_sof_suspend); + +int snd_sof_prepare(struct device *dev) +{ + struct snd_sof_dev *sdev = dev_get_drvdata(dev); + +#if defined(CONFIG_ACPI) + sdev->s0_suspend = acpi_target_system_state() == ACPI_STATE_S0; +#else + /* will suspend to S3 by default */ + sdev->s0_suspend = false; +#endif + + return 0; +} +EXPORT_SYMBOL(snd_sof_prepare); + +void snd_sof_complete(struct device *dev) +{ + struct snd_sof_dev *sdev = dev_get_drvdata(dev); + + sdev->s0_suspend = false; +} +EXPORT_SYMBOL(snd_sof_complete); diff --git a/sound/soc/sof/sof-priv.h b/sound/soc/sof/sof-priv.h index c2541d020bde..6408ac88a3e5 100644 --- a/sound/soc/sof/sof-priv.h +++ b/sound/soc/sof/sof-priv.h @@ -492,6 +492,8 @@ int snd_sof_runtime_resume(struct device *dev); int snd_sof_runtime_idle(struct device *dev); int snd_sof_resume(struct device *dev); int snd_sof_suspend(struct device *dev); +int snd_sof_prepare(struct device *dev); +void snd_sof_complete(struct device *dev); int snd_sof_set_d0_substate(struct snd_sof_dev *sdev, enum sof_d0_substate d0_substate);

The patch

ASoC: SOF: ignore suspend/resume for D0ix compatible streams

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From ac8c046f19f94ec419c60d7a073af75a71386e97 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:18 -0500 Subject: [PATCH] ASoC: SOF: ignore suspend/resume for D0ix compatible streams

During system suspend, the PM framework will freeze all applications and the ALSA/ASoC core will suspend all RUNNING PCM streams.

However, D0ix-compatible PCM streams should keep the related pipelines active in the DSP when the system is entering S0ix. The TRIGGER_SUSPEND event is trapped in such cases to prevent the pipelines from being stopped. Likewise, the TRIGGER_RESUME/START events should not affect the pipeline state.

The SOF driver also triggers some DSP Firmware pipelines based on the DAPM widgets power events. In such cases, we also ignore PRE_PMU and POST_PMD events to keep the pipelines active.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-23-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/pcm.c | 31 +++++++++++++++++++++++++++++++ sound/soc/sof/topology.c | 17 ++++++++++++++++- 2 files changed, 47 insertions(+), 1 deletion(-)

diff --git a/sound/soc/sof/pcm.c b/sound/soc/sof/pcm.c index 9ba505cf5966..3d5cd1b445ba 100644 --- a/sound/soc/sof/pcm.c +++ b/sound/soc/sof/pcm.c @@ -346,6 +346,16 @@ static int sof_pcm_trigger(struct snd_soc_component *component, stream.hdr.cmd |= SOF_IPC_STREAM_TRIG_RELEASE; break; case SNDRV_PCM_TRIGGER_RESUME: + if (spcm->stream[substream->stream].suspend_ignored) { + /* + * this case will be triggered when INFO_RESUME is + * supported, no need to resume streams that remained + * enabled in D0ix. + */ + spcm->stream[substream->stream].suspend_ignored = false; + return 0; + } + /* set up hw_params */ ret = sof_pcm_prepare(component, substream); if (ret < 0) { @@ -356,9 +366,30 @@ static int sof_pcm_trigger(struct snd_soc_component *component,

/* fallthrough */ case SNDRV_PCM_TRIGGER_START: + if (spcm->stream[substream->stream].suspend_ignored) { + /* + * This case will be triggered when INFO_RESUME is + * not supported, no need to re-start streams that + * remained enabled in D0ix. + */ + spcm->stream[substream->stream].suspend_ignored = false; + return 0; + } stream.hdr.cmd |= SOF_IPC_STREAM_TRIG_START; break; case SNDRV_PCM_TRIGGER_SUSPEND: + if (sdev->s0_suspend && + spcm->stream[substream->stream].d0i3_compatible) { + /* + * trap the event, not sending trigger stop to + * prevent the FW pipelines from being stopped, + * and mark the flag to ignore the upcoming DAPM + * PM events. + */ + spcm->stream[substream->stream].suspend_ignored = true; + return 0; + } + /* fallthrough */ case SNDRV_PCM_TRIGGER_STOP: stream.hdr.cmd |= SOF_IPC_STREAM_TRIG_STOP; ipc_first = true; diff --git a/sound/soc/sof/topology.c b/sound/soc/sof/topology.c index cc138ac6a8eb..e0e2ae734632 100644 --- a/sound/soc/sof/topology.c +++ b/sound/soc/sof/topology.c @@ -135,7 +135,9 @@ static int sof_keyword_dapm_event(struct snd_soc_dapm_widget *w, struct snd_kcontrol *k, int event) { struct snd_sof_widget *swidget = w->dobj.private; + int stream = SNDRV_PCM_STREAM_CAPTURE; struct snd_sof_dev *sdev; + struct snd_sof_pcm *spcm; int ret = 0;

if (!swidget) @@ -146,11 +148,19 @@ static int sof_keyword_dapm_event(struct snd_soc_dapm_widget *w, dev_dbg(sdev->dev, "received event %d for widget %s\n", event, w->name);

+ /* get runtime PCM params using widget's stream name */ + spcm = snd_sof_find_spcm_name(sdev, swidget->widget->sname); + /* process events */ switch (event) { case SND_SOC_DAPM_PRE_PMU: + if (spcm->stream[stream].suspend_ignored) { + dev_dbg(sdev->dev, "PRE_PMU event ignored, KWD pipeline is already RUNNING\n"); + return 0; + } + /* set pcm params */ - ret = ipc_pcm_params(swidget, SOF_IPC_STREAM_CAPTURE); + ret = ipc_pcm_params(swidget, stream); if (ret < 0) { dev_err(sdev->dev, "error: failed to set pcm params for widget %s\n", @@ -166,6 +176,11 @@ static int sof_keyword_dapm_event(struct snd_soc_dapm_widget *w, swidget->widget->name); break; case SND_SOC_DAPM_POST_PMD: + if (spcm->stream[stream].suspend_ignored) { + dev_dbg(sdev->dev, "POST_PMD even ignored, KWD pipeline will remain RUNNING\n"); + return 0; + } + /* stop trigger */ ret = ipc_trigger(swidget, SOF_IPC_STREAM_TRIG_STOP); if (ret < 0)

The patch

ASoC: SOF: Intel: hda-dsp: implement suspend/resume for S0ix<->S0 transition

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 66e40876ddc325d892c493a6d83574bbba5770ce Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:19 -0500 Subject: [PATCH] ASoC: SOF: Intel: hda-dsp: implement suspend/resume for S0ix<->S0 transition

Enable system wake up via IPC interrupt from DSP when the system is suspending to the S0ix state, and disable it in the corresponding resuming.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-24-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/intel/hda-dsp.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+)

diff --git a/sound/soc/sof/intel/hda-dsp.c b/sound/soc/sof/intel/hda-dsp.c index b5070409a5e3..d23573d9e9c4 100644 --- a/sound/soc/sof/intel/hda-dsp.c +++ b/sound/soc/sof/intel/hda-dsp.c @@ -477,6 +477,15 @@ static int hda_resume(struct snd_sof_dev *sdev, bool runtime_resume)

int hda_dsp_resume(struct snd_sof_dev *sdev) { + struct pci_dev *pci = to_pci_dev(sdev->dev); + + if (sdev->s0_suspend) { + /* restore and disable the system wakeup */ + pci_restore_state(pci); + disable_irq_wake(pci->irq); + return 0; + } + /* init hda controller. DSP cores will be powered up during fw boot */ return hda_resume(sdev, false); } @@ -509,8 +518,16 @@ int hda_dsp_runtime_suspend(struct snd_sof_dev *sdev) int hda_dsp_suspend(struct snd_sof_dev *sdev) { struct hdac_bus *bus = sof_to_bus(sdev); + struct pci_dev *pci = to_pci_dev(sdev->dev); int ret;

+ if (sdev->s0_suspend) { + /* enable the system waking up via IPC IRQ */ + enable_irq_wake(pci->irq); + pci_save_state(pci); + return 0; + } + /* stop hda controller and power dsp off */ ret = hda_suspend(sdev, false); if (ret < 0) {

The patch

ASoC: SOF: return -ENOTSUPP if D0I3 is not supported

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 5655ce660a6d3eec4a3ad37feb99fb228540d1a9 Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:20 -0500 Subject: [PATCH] ASoC: SOF: return -ENOTSUPP if D0I3 is not supported

No set_power_state ops means that the platform doesn't support D0i3, return -ENOTSUPP for the case.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-25-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/ops.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/sound/soc/sof/ops.h b/sound/soc/sof/ops.h index d7c8fc06f961..93512dcbaacd 100644 --- a/sound/soc/sof/ops.h +++ b/sound/soc/sof/ops.h @@ -199,7 +199,8 @@ static inline int snd_sof_dsp_set_power_state(struct snd_sof_dev *sdev, if (sof_ops(sdev)->set_power_state) return sof_ops(sdev)->set_power_state(sdev, substate);

- return 0; + /* D0 substate is not supported */ + return -ENOTSUPP; }

/* debug */

The patch

ASoC: SOF: PM: Add support for DSP D0i3 state when entering S0ix

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From c470fc3f61b25e955f8ba90fc9dc554881e4e72c Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:21 -0500 Subject: [PATCH] ASoC: SOF: PM: Add support for DSP D0i3 state when entering S0ix

When system is entering into S0ix, the PCI device may transition to the D0i3 substate instead of D3. In D0i3, some always-on functionality can be enabled, such as acoustic event detection, voice activity detection or hotwording. When an event is detected, the DSP firmware can wake-up the device for a transition to D0 with an interrupt.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-26-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/pm.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+)

diff --git a/sound/soc/sof/pm.c b/sound/soc/sof/pm.c index 99e4e6ffff74..560a937e0484 100644 --- a/sound/soc/sof/pm.c +++ b/sound/soc/sof/pm.c @@ -430,12 +430,58 @@ EXPORT_SYMBOL(snd_sof_set_d0_substate);

int snd_sof_resume(struct device *dev) { + struct snd_sof_dev *sdev = dev_get_drvdata(dev); + int ret; + + if (sdev->s0_suspend) { + /* resume from D0I3 */ + dev_dbg(sdev->dev, "DSP will exit from D0i3...\n"); + ret = snd_sof_set_d0_substate(sdev, SOF_DSP_D0I0); + if (ret == -ENOTSUPP) { + /* fallback to resume from D3 */ + dev_dbg(sdev->dev, "D0i3 not supported, fall back to resume from D3...\n"); + goto d3_resume; + } else if (ret < 0) { + dev_err(sdev->dev, "error: failed to exit from D0I3 %d\n", + ret); + return ret; + } + + /* platform-specific resume from D0i3 */ + return snd_sof_dsp_resume(sdev); + } + +d3_resume: + /* resume from D3 */ return sof_resume(dev, false); } EXPORT_SYMBOL(snd_sof_resume);

int snd_sof_suspend(struct device *dev) { + struct snd_sof_dev *sdev = dev_get_drvdata(dev); + int ret; + + if (sdev->s0_suspend) { + /* suspend to D0i3 */ + dev_dbg(sdev->dev, "DSP is trying to enter D0i3...\n"); + ret = snd_sof_set_d0_substate(sdev, SOF_DSP_D0I3); + if (ret == -ENOTSUPP) { + /* fallback to D3 suspend */ + dev_dbg(sdev->dev, "D0i3 not supported, fall back to D3...\n"); + goto d3_suspend; + } else if (ret < 0) { + dev_err(sdev->dev, "error: failed to enter D0I3, %d\n", + ret); + return ret; + } + + /* platform-specific suspend to D0i3 */ + return snd_sof_dsp_suspend(sdev); + } + +d3_suspend: + /* suspend to D3 */ return sof_suspend(dev, false); } EXPORT_SYMBOL(snd_sof_suspend);

The patch

ASoC: SOF: pci: Add prepare/complete PM callbacks

has been applied to the asoc tree at

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-5.5

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

From 6d407a39cc9619d0088b44333c6526924be00f8d Mon Sep 17 00:00:00 2001

From: Keyon Jie yang.jie@linux.intel.com Date: Fri, 25 Oct 2019 17:41:22 -0500 Subject: [PATCH] ASoC: SOF: pci: Add prepare/complete PM callbacks

Use the new implemented snd_sof_prepare() and snd_sof_complete() as the power management callbacks for pci probing platforms.

Signed-off-by: Keyon Jie yang.jie@linux.intel.com Signed-off-by: Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com Link: https://lore.kernel.org/r/20191025224122.7718-27-pierre-louis.bossart@linux.... Signed-off-by: Mark Brown broonie@kernel.org --- sound/soc/sof/sof-pci-dev.c | 2 ++ 1 file changed, 2 insertions(+)

diff --git a/sound/soc/sof/sof-pci-dev.c b/sound/soc/sof/sof-pci-dev.c index 4adbb27c76c5..487b1f29f84d 100644 --- a/sound/soc/sof/sof-pci-dev.c +++ b/sound/soc/sof/sof-pci-dev.c @@ -264,6 +264,8 @@ static const struct sof_dev_desc jsl_desc = { #endif

static const struct dev_pm_ops sof_pci_pm = { + .prepare = snd_sof_prepare, + .complete = snd_sof_complete, SET_SYSTEM_SLEEP_PM_OPS(snd_sof_suspend, snd_sof_resume) SET_RUNTIME_PM_OPS(snd_sof_runtime_suspend, snd_sof_runtime_resume, snd_sof_runtime_idle)