ASoC: SOF: Intel: fix LNL DMA handling and enable chain_dma support #4533

plbossart · 2023-08-18T20:28:58Z

The DMA handling is not correct on capture when we have 4 channels aggregated from 2 DAIs on 2 different links. We currently setup the DMA for 2 channels, which causes the data to be recorded at half-speed.

Also this enables chain-DMA support for all interfaces, only tested on ALH playback for now.

Still draft since the capture w/ chain DMA does not work - xrun on capture.

plbossart · 2023-08-29T16:40:21Z

Moving to ready. This was tested in hardware, and the only problematic case is the capture which is also broken for MTL, so it's a firmware issue clearly.

plbossart · 2023-08-29T16:41:28Z

actually, need to fix checkpatch issues.

plbossart · 2023-08-29T18:16:06Z

wow, clang found an uninitialized variable....Amazing that other compilers don't detect this!

sound/soc/sof/intel/hda-dai.c:473:4: error: variable 'ch_mask' is uninitialized when used here [-Werror,-Wuninitialized]
                        ch_mask |= rtd->dai_link->codec_ch_maps[j].ch_mask;
                        ^~~~~~~
sound/soc/sof/intel/hda-dai.c:443:13: note: initialize the variable 'ch_mask' to silence this warning
        int ch_mask;
                   ^
                    = 0

bardliao

LGTM

ujfalusi

@plbossart, I'm not sure about the tying of the ChainDMA to DSPless mode, is it future proof?

sound/soc/sof/intel/hda-dai.c

sound/soc/sof/intel/hda-dai-ops.c

sound/soc/sof/ops.h

sound/soc/sof/intel/hda.c

sound/soc/sof/ipc4-topology.c

RanderWang

LGTM

sound/soc/sof/ops.h

sound/soc/sof/intel/hda-dai-ops.c

plbossart · 2023-09-12T21:13:38Z

@RanderWang @ujfalusi v2 updated, see diff
v2.diff.txt

sound/soc/sof/intel/hda-dai.c

sound/soc/sof/intel/hda-dai-ops.c

ujfalusi

I'm not sure about the definition of the snd_sof_is_dai_supported and especially the SOF_DAI_FEATURE_ definition.

Also, I think there is an issue with the dai validity check for Intel platforms.

sound/soc/sof/intel/hda-dai-ops.c

sound/soc/sof/sof-priv.h

sound/soc/sof/intel/hda.c

…ed dailink The existing code derives the channel map used to program the HDaudio link DMA from the hw_params, but that is not quite right in the case of aggregation. The code in soc-pcm.c splits the hw_params depending on the codec_ch_map, and we need to reconstruct the channel-map to insert the data in the right places. This issue is seen only on amplifier feedback capture where the data from the second amplifier was replaced by that of the first amplifier. Note that the loop iterator of the macro for_each_rtd_cpu_dais() is reused in a following loop. This is different to all existing usages of that macro, hence the use of a boolean flag to avoid an access to an uninitialized variable. Fixes: 2960ee5 ("ASoC: SOF: Intel: hda-dai: add helpers for SoundWire callbacks") Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

The generic format calculation is shared between SSP, DMIC and SoundWire. In the latter case, the dailink may handle more channels than what each DAI can handle. This is specifically the case for aggregated configurations, specifically for capture where we may want to handle a 4ch stream using two DAIs. The issue is that soc-pcm.c already splits the channels on a per-dai basis, so we need to count the total number of channels handled by the dailink. This isn't very elegant but the information on the dailink configuration is not available at the DAI level. The net effect of this confusion between DAI and dailink channels was that the data was recorded 2x slower than normal, and when played at the normal rate a 2x frequency shift can be heard. Fixes: 1254773 ("ASoC: SOF: Intel: hda-dai-ops: add/select DMA ops for SSP") Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

The chain_dma mode is currently only handled for HDaudio, but can be used for orther DAIs starting with LunarLake. Move the chain_dma handling earlier. Error detection for the chain_dma case for older platforms is handled at a different level. Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

IPC4 introduced a 'chain-dma' mode when host and link DMA are connected by firmware without using a regular pipeline or the ability to add intermediate connections. This mode is not available on all platforms and all links, so add a platform-specific callback to help the SOF ipc4-topology core handle different hardware+firmware configurations. Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

Reuse existing function to get the interface mask and expose it to the SOF core with a callback - the main user is the IPC4 topology so only HDaudio platforms provide this callback. Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

plbossart · 2023-09-25T23:26:48Z

new version with simpler 'is_chain_dma_supported' callback and all comments addressed. Feedback welcome.

RanderWang

LGTM

sound/soc/sof/intel/hda-dai-ops.c

Use the existing callbacks and mix/match of HDaudio and SoundWire support. Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

Now that we have a 'is_chain_dma_supported' callback we can use it to double-check possible disconnects between a topology file enabling chain-dma for a DAI and the hardware/firmware capabilities. Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

RanderWang

LGTM

plbossart marked this pull request as ready for review August 29, 2023 16:39

plbossart requested review from RanderWang, bardliao, dbaluta, juimonen, kuanhsuncheng, kv2019i, lgirdwood, libinyang, lyakh, ranj063, ujfalusi, vijendarmukunda and yaochunhung as code owners August 29, 2023 16:39

plbossart force-pushed the fix/lnl-alh-chain-dma branch from f5a0eb6 to aee60c1 Compare August 29, 2023 16:46

plbossart force-pushed the fix/lnl-alh-chain-dma branch from aee60c1 to d690df9 Compare August 29, 2023 18:16

bardliao previously approved these changes Aug 30, 2023

View reviewed changes

ujfalusi reviewed Aug 30, 2023

View reviewed changes

RanderWang previously approved these changes Aug 31, 2023

View reviewed changes

sound/soc/sof/ops.h Show resolved Hide resolved

sound/soc/sof/intel/hda-dai-ops.c Outdated Show resolved Hide resolved

plbossart dismissed stale reviews from RanderWang and bardliao via 97a0775 September 12, 2023 21:12

plbossart force-pushed the fix/lnl-alh-chain-dma branch from d690df9 to 97a0775 Compare September 12, 2023 21:12

RanderWang reviewed Sep 13, 2023

View reviewed changes

sound/soc/sof/intel/hda-dai.c Outdated Show resolved Hide resolved

lyakh reviewed Sep 13, 2023

View reviewed changes

sound/soc/sof/intel/hda-dai-ops.c Show resolved Hide resolved

ujfalusi reviewed Sep 13, 2023

View reviewed changes

sound/soc/sof/intel/hda-dai-ops.c Show resolved Hide resolved

sound/soc/sof/sof-priv.h Outdated Show resolved Hide resolved

sound/soc/sof/intel/hda.c Outdated Show resolved Hide resolved

plbossart added 4 commits September 25, 2023 19:10

plbossart requested review from lyakh and ujfalusi September 25, 2023 23:26

plbossart force-pushed the fix/lnl-alh-chain-dma branch from 97a0775 to da910d3 Compare September 25, 2023 23:26

plbossart requested review from RanderWang and bardliao September 25, 2023 23:26

bardliao previously approved these changes Sep 26, 2023

View reviewed changes

RanderWang previously approved these changes Sep 26, 2023

View reviewed changes

sound/soc/sof/intel/hda-dai-ops.c Show resolved Hide resolved

plbossart added 2 commits October 3, 2023 12:07

ASoC: SOF: Intel: hda-dai-ops: enable chain_dma for ALH

931e1e4

Use the existing callbacks and mix/match of HDaudio and SoundWire support. Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

plbossart dismissed stale reviews from RanderWang and bardliao via ad29504 October 3, 2023 16:08

plbossart force-pushed the fix/lnl-alh-chain-dma branch from da910d3 to ad29504 Compare October 3, 2023 16:08

plbossart requested review from RanderWang and bardliao October 3, 2023 16:09

bardliao approved these changes Oct 4, 2023

View reviewed changes

RanderWang approved these changes Oct 10, 2023

View reviewed changes

plbossart merged commit 42730b1 into thesofproject:topic/sof-dev Oct 10, 2023

plbossart mentioned this pull request Oct 10, 2023

Fix bad merge #4628

Merged

keqiaozhang mentioned this pull request Oct 25, 2023

[BUG][LNL] call trace error happens when doing audio playback on LNL-NOCODEC platform #4655

Closed

ASoC: SOF: Intel: fix LNL DMA handling and enable chain_dma support #4533

ASoC: SOF: Intel: fix LNL DMA handling and enable chain_dma support #4533

Uh oh!

Conversation

plbossart commented Aug 18, 2023

Uh oh!

plbossart commented Aug 29, 2023

Uh oh!

plbossart commented Aug 29, 2023

Uh oh!

plbossart commented Aug 29, 2023

Uh oh!

bardliao left a comment

Choose a reason for hiding this comment

Uh oh!

ujfalusi left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

RanderWang left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

plbossart commented Sep 12, 2023

Uh oh!

Uh oh!

Uh oh!

ujfalusi left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

plbossart commented Sep 25, 2023

Uh oh!

RanderWang left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

RanderWang left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants