Skip to contentSkip to conversation

ALSA/SOF Intel: Enforce stricter period size for NVL

PATCH Applied Peter Ujfalusi 4 participants 9 messages Apr 8

Subsystems (2)

v1 of 2 Applied Latest is v2 →

lore
Share & tools
lore
#0 lore raw

cover ALSA/SOF Intel: Enforce stricter period size for NVL

Hi,

NVL and NVL-S (ACE4) needs to use stricter period size constraint to meet the address alignment for each BDLE buffer (start of each period in the continuous ALSA buffer) set in the HDA specification.

It would be great if these can be sent for 7.0 as last minute if it is doable, I left out the Fixes tag from the first patch as that is introduced in 7.0.

Regards
Peter
---
Kai Vehmanen (2):
ALSA: hda/intel: enforce stricter period-size alignment for Intel NVL
ASoC: SOF: Intel: hda: modify period size constraints for ACE4

 sound/hda/controllers/intel.c |  7 +++++--
 sound/soc/sof/intel/hda-pcm.c | 14 ++++++++++++--
 2 files changed, 17 insertions(+), 4 deletions(-)
Signature · 1 line
2.53.0
#1 lore raw

PATCH 1/2 ALSA: hda/intel: enforce stricter period-size alignment for Intel NVL

From: Kai Vehmanen <[email protected]>

Intel ACE4 based products set more strict constraints on HDA BDLE start address and length alignment. Modify capability flags to drop AZX_DCAPS_NO_ALIGN_BUFSIZE for Intel Nova Lake platforms.

Fixes 7f428282fde3 ("ALSA: hda: controllers: intel: add support for Nova Lake")
---
 sound/hda/controllers/intel.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
sound/hda/controllers/intel.c +5 -2
index 1b22dbf7a719..257c498c3260 100644
@@ -295,6 +295,9 @@ enum {
 #define AZX_DCAPS_INTEL_LNL \
 	(AZX_DCAPS_INTEL_SKYLAKE | AZX_DCAPS_PIO_COMMANDS)
 
+#define AZX_DCAPS_INTEL_NVL \
+	(AZX_DCAPS_INTEL_LNL & ~AZX_DCAPS_NO_ALIGN_BUFSIZE)
+
 /* quirks for ATI SB / AMD Hudson */
 #define AZX_DCAPS_PRESET_ATI_SB \
 	(AZX_DCAPS_NO_TCSEL | AZX_DCAPS_POSFIX_LPIB |\
@@ -2552,8 +2555,8 @@ static const struct pci_device_id azx_ids[] = {
 	/* Wildcat Lake */
 	{ PCI_DEVICE_DATA(INTEL, HDA_WCL, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_LNL) },
 	/* Nova Lake */
-	{ PCI_DEVICE_DATA(INTEL, HDA_NVL, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_LNL) },
-	{ PCI_DEVICE_DATA(INTEL, HDA_NVL_S, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_LNL) },
+	{ PCI_DEVICE_DATA(INTEL, HDA_NVL, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_NVL) },
+	{ PCI_DEVICE_DATA(INTEL, HDA_NVL_S, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_NVL) },
 	/* Apollolake (Broxton-P) */
 	{ PCI_DEVICE_DATA(INTEL, HDA_APL, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_BROXTON) },
 	/* Gemini-Lake */
Signature · 1 line
2.53.0
maintainer
#2 lore raw

Re: [PATCH for 7.0 1/2] ALSA: hda/intel: enforce stricter period-size alignment for Intel NVL

Quoted 10 lines · Peter Ujfalusi

From: Kai Vehmanen <[email protected]>

Intel ACE4 based products set more strict constraints on HDA BDLE start address and length alignment. Modify capability flags to drop AZX_DCAPS_NO_ALIGN_BUFSIZE for Intel Nova Lake platforms.

Fixes 7f428282fde3 ("ALSA: hda: controllers: intel: add support for Nova Lake")

This also needs a:

Cc stable <[email protected]>

As the commit you reference here is in the stable 6.19.y tree.

thanks,

greg k-h

Applied
#3 lore raw

Re: [PATCH for 7.0 1/2] ALSA: hda/intel: enforce stricter period-size alignment for Intel NVL

In reply to Peter Ujfalusi
Quoted 11 lines · Peter Ujfalusi

From: Kai Vehmanen <[email protected]>

Intel ACE4 based products set more strict constraints on HDA BDLE start address and length alignment. Modify capability flags to drop AZX_DCAPS_NO_ALIGN_BUFSIZE for Intel Nova Lake platforms.

Fixes 7f428282fde3 ("ALSA: hda: controllers: intel: add support for Nova Lake")

Applied this one now (with Cc to stable).

thanks,

Takashi

#4 lore raw

PATCH 2/2 ASoC: SOF: Intel: hda: modify period size constraints for ACE4

In reply to Peter Ujfalusi

From: Kai Vehmanen <[email protected]>

Intel ACE4 based products set more strict constraints on HDA BDLE start address and length alignment. Add a constraint to align period size to 128 bytes.

The commit removes the "minimum as per HDA spec" comment. This comment was misleading as spec actually does allow a 2 byte BDLE length, and more importantly, period size also directly impacts how the BDLE start addresses are aligned, so it is not sufficient just to consider allowed buffer length.

Fixes d3df422f66e8 ("ASoC: SOF: Intel: add initial support for NVL-S")
---
 sound/soc/sof/intel/hda-pcm.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)
sound/soc/sof/intel/hda-pcm.c +12 -2
index da6c1e7263cd..16a364072821 100644
@@ -219,6 +219,7 @@ EXPORT_SYMBOL_NS(hda_dsp_pcm_pointer, "SND_SOC_SOF_INTEL_HDA_COMMON");
 int hda_dsp_pcm_open(struct snd_sof_dev *sdev,
 		     struct snd_pcm_substream *substream)
 {
+	const struct sof_intel_dsp_desc *chip_info = get_chip_info(sdev->pdata);
 	struct snd_soc_pcm_runtime *rtd = snd_soc_substream_to_rtd(substream);
 	struct snd_pcm_runtime *runtime = substream->runtime;
 	struct snd_soc_component *scomp = sdev->component;
@@ -268,8 +269,17 @@ int hda_dsp_pcm_open(struct snd_sof_dev *sdev,
 		return -ENODEV;
 	}
 
-	/* minimum as per HDA spec */
-	snd_pcm_hw_constraint_step(substream->runtime, 0, SNDRV_PCM_HW_PARAM_PERIOD_BYTES, 4);
+	/*
+	 * Set period size constraint to ensure BDLE buffer length and
+	 * start address alignment requirements are met. Align to 128
+	 * bytes for newer Intel platforms, with older ones using 4 byte alignment.
+	 */
+	if (chip_info->hw_ip_version >= SOF_INTEL_ACE_4_0)
+		snd_pcm_hw_constraint_step(substream->runtime, 0,
+					   SNDRV_PCM_HW_PARAM_PERIOD_BYTES, 128);
+	else
+		snd_pcm_hw_constraint_step(substream->runtime, 0,
+					   SNDRV_PCM_HW_PARAM_PERIOD_BYTES, 4);
 
 	/* avoid circular buffer wrap in middle of period */
 	snd_pcm_hw_constraint_integer(substream->runtime,
Signature · 1 line
2.53.0
#5 lore raw
In reply to Peter Ujfalusi
Quoted 10 lines · Peter Ujfalusi

Hi,

NVL and NVL-S (ACE4) needs to use stricter period size constraint to meet the address alignment for each BDLE buffer (start of each period in the continuous ALSA buffer) set in the HDA specification.

It would be great if these can be sent for 7.0 as last minute if it is doable, I left out the Fixes tag from the first patch as that is introduced in 7.0.

Mark, shall I pick up both patches to my tree for the next PR (planned for tomorrow or on Friday)?
Basically both patches are independent, and I can apply the first one in anyway.

thanks,

Takashi

Quoted 15 lines

Regards
Peter
---
Kai Vehmanen (2):
ALSA: hda/intel: enforce stricter period-size alignment for Intel NVL
ASoC: SOF: Intel: hda: modify period size constraints for ACE4

 sound/hda/controllers/intel.c |  7 +++++--
 sound/soc/sof/intel/hda-pcm.c | 14 ++++++++++++--
 2 files changed, 17 insertions(+), 4 deletions(-)
Signature · 1 line
2.53.0
#6 lore raw
Quoted 6 lines · Takashi Iwai
Quoted 3 lines · Peter Ujfalusi

NVL and NVL-S (ACE4) needs to use stricter period size constraint to meet the address alignment for each BDLE buffer (start of each period in the continuous ALSA buffer) set in the HDA specification.

Quoted 8 lines
Quoted 3 lines

It would be great if these can be sent for 7.0 as last minute if it is doable, I left out the Fixes tag from the first patch as that is introduced in 7.0.

Mark, shall I pick up both patches to my tree for the next PR (planned for tomorrow or on Friday)?
Basically both patches are independent, and I can apply the first one in anyway.

I've already got the ASoC one in process for a final fixes PR today and was going to complain at Peter about combining patches for multiple trees into a single series when there's no dependencies.

#7 lore raw
Quoted 12 lines · Mark Brown
Quoted 9 lines · Takashi Iwai
Quoted 7 lines · Peter Ujfalusi

NVL and NVL-S (ACE4) needs to use stricter period size constraint to meet the address alignment for each BDLE buffer (start of each period in the continuous ALSA buffer) set in the HDA specification.

It would be great if these can be sent for 7.0 as last minute if it is doable, I left out the Fixes tag from the first patch as that is introduced in 7.0.

Quoted 8 lines
Quoted 4 lines

Mark, shall I pick up both patches to my tree for the next PR (planned for tomorrow or on Friday)?
Basically both patches are independent, and I can apply the first one in anyway.

I've already got the ASoC one in process for a final fixes PR today and was going to complain at Peter about combining patches for multiple trees into a single series when there's no dependencies.

OK, then I'll wait for your PR for the second patch, while I take the first one on mine.

thanks,

Takashi

#8 lore raw

Re: (subset) [PATCH for 7.0 0/2] ALSA/SOF Intel: Enforce stricter period size for NVL

In reply to Peter Ujfalusi
Quoted 9 lines · Peter Ujfalusi

ALSA/SOF Intel: Enforce stricter period size for NVL

Hi,

NVL and NVL-S (ACE4) needs to use stricter period size constraint to meet the address alignment for each BDLE buffer (start of each period in the continuous ALSA buffer) set in the HDA specification.

[...]

Applied to

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-7.0

Thanks!

[2/2] ASoC: SOF: Intel: hda: modify period size constraints for ACE4

      https://git.kernel.org/broonie/sound/c/0f7186605726

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