[PATCH v2] ASoC: simple-card: add comment to indicate don't remove platforms

[PATCH v2] ASoC: simple-card: add comment to indicate don't remove platforms

[Kuninori Morimoto]

Basically CPU and Platform are different Component, but if CPU is using
soc-generic-dmaengine-pcm, same dev will be shared between CPU and
Platform, and Simple Card had been supporting it.

When we focus to clean up Simple Card driver, we tend to remove platforms
if no Platform was selected, but it is wrong because of above reasons.

This patch adds comment why we shouldn't remove platforms.

In case of CPU is not using soc-generic-dmaengine-pcm, CPU and Platform
will be duplicated, but it will be ignored by snd_soc_rtd_add_component().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
---
v1 -> v2

	- add +1 comment on asoc_simple_canonicalize_platform()

 sound/soc/generic/audio-graph-card.c  | 13 ++++++++++++-
 sound/soc/generic/audio-graph-card2.c | 17 +++++++++++++++++
 sound/soc/generic/simple-card-utils.c | 11 ++++++++++-
 sound/soc/generic/simple-card.c       | 22 +++++++++++++++++++++-
 4 files changed, 60 insertions(+), 3 deletions(-)

diff --git a/sound/soc/generic/audio-graph-card.c b/sound/soc/generic/audio-graph-card.c
index d788f5f23a8a..4e85536a1b26 100644
--- a/sound/soc/generic/audio-graph-card.c
+++ b/sound/soc/generic/audio-graph-card.c
@@ -613,10 +613,16 @@ static int graph_count_noml(struct asoc_simple_priv *priv,
 		return -EINVAL;
 	}
 
+	/*
+	 * DON'T REMOVE platforms
+	 * see
+	 *	simple-card.c :: simple_count_noml()
+	 */
 	li->num[li->link].cpus		= 1;
-	li->num[li->link].codecs	= 1;
 	li->num[li->link].platforms     = 1;
 
+	li->num[li->link].codecs	= 1;
+
 	li->link += 1; /* 1xCPU-Codec */
 
 	dev_dbg(dev, "Count As Normal\n");
@@ -637,6 +643,11 @@ static int graph_count_dpcm(struct asoc_simple_priv *priv,
 	}
 
 	if (li->cpu) {
+		/*
+		 * DON'T REMOVE platforms
+		 * see
+		 *	simple-card.c :: simple_count_noml()
+		 */
 		li->num[li->link].cpus		= 1;
 		li->num[li->link].platforms     = 1;
 
diff --git a/sound/soc/generic/audio-graph-card2.c b/sound/soc/generic/audio-graph-card2.c
index 259544f64df9..d145b74d41e4 100644
--- a/sound/soc/generic/audio-graph-card2.c
+++ b/sound/soc/generic/audio-graph-card2.c
@@ -1046,8 +1046,14 @@ static int graph_count_normal(struct asoc_simple_priv *priv,
 	 * =>		lnk: port { endpoint { .. }; };
 	 *	};
 	 */
+	/*
+	 * DON'T REMOVE platforms
+	 * see
+	 *	simple-card.c :: simple_count_noml()
+	 */
 	li->num[li->link].cpus		=
 	li->num[li->link].platforms	= graph_counter(cpu_port);
+
 	li->num[li->link].codecs	= graph_counter(codec_port);
 
 	of_node_put(cpu_ep);
@@ -1079,6 +1085,11 @@ static int graph_count_dpcm(struct asoc_simple_priv *priv,
 	 */
 
 	if (asoc_graph_is_ports0(lnk)) {
+		/*
+		 * DON'T REMOVE platforms
+		 * see
+		 *	simple-card.c :: simple_count_noml()
+		 */
 		li->num[li->link].cpus		= graph_counter(rport); /* FE */
 		li->num[li->link].platforms	= graph_counter(rport);
 	} else {
@@ -1113,8 +1124,14 @@ static int graph_count_c2c(struct asoc_simple_priv *priv,
 	 *	};
 	 * };
 	 */
+	/*
+	 * DON'T REMOVE platforms
+	 * see
+	 *	simple-card.c :: simple_count_noml()
+	 */
 	li->num[li->link].cpus		=
 	li->num[li->link].platforms	= graph_counter(codec0);
+
 	li->num[li->link].codecs	= graph_counter(codec1);
 
 	of_node_put(ports);
diff --git a/sound/soc/generic/simple-card-utils.c b/sound/soc/generic/simple-card-utils.c
index 56552a616f21..34a9b91e676f 100644
--- a/sound/soc/generic/simple-card-utils.c
+++ b/sound/soc/generic/simple-card-utils.c
@@ -638,7 +638,16 @@ EXPORT_SYMBOL_GPL(asoc_simple_dai_init);
 void asoc_simple_canonicalize_platform(struct snd_soc_dai_link_component *platforms,
 				       struct snd_soc_dai_link_component *cpus)
 {
-	/* Assumes platform == cpu */
+	/*
+	 * Assumes Platform == CPU
+	 *
+	 * Some CPU might be using soc-generic-dmaengine-pcm. This means CPU and Platform
+	 * are different Component, but are sharing same component->dev.
+	 *
+	 * Let's assume Platform is same as CPU if it doesn't identify Platform on DT.
+	 * see
+	 *	simple-card.c :: simple_count_noml()
+	 */
 	if (!platforms->of_node)
 		platforms->of_node = cpus->of_node;
 }
diff --git a/sound/soc/generic/simple-card.c b/sound/soc/generic/simple-card.c
index e98932c16754..fcd6c44f89f6 100644
--- a/sound/soc/generic/simple-card.c
+++ b/sound/soc/generic/simple-card.c
@@ -509,10 +509,25 @@ static int simple_count_noml(struct asoc_simple_priv *priv,
 		return -EINVAL;
 	}
 
+	/*
+	 * DON'T REMOVE platforms
+	 *
+	 * Some CPU might be using soc-generic-dmaengine-pcm. This means CPU and Platform
+	 * are different Component, but are sharing same component->dev.
+	 * Simple Card had been supported it without special Platform selection.
+	 * We need platforms here.
+	 *
+	 * In case of no Platform, it will be Platform == CPU, but Platform will be
+	 * ignored by snd_soc_rtd_add_component().
+	 *
+	 * see
+	 *	simple-card-utils.c :: asoc_simple_canonicalize_platform()
+	 */
 	li->num[li->link].cpus		= 1;
-	li->num[li->link].codecs	= 1;
 	li->num[li->link].platforms	= 1;
 
+	li->num[li->link].codecs	= 1;
+
 	li->link += 1;
 
 	return 0;
@@ -531,6 +546,11 @@ static int simple_count_dpcm(struct asoc_simple_priv *priv,
 	}
 
 	if (li->cpu) {
+		/*
+		 * DON'T REMOVE platforms
+		 * see
+		 *	simple_count_noml()
+		 */
 		li->num[li->link].cpus		= 1;
 		li->num[li->link].platforms	= 1;
 
-- 
2.25.1

Re: [PATCH v2] ASoC: simple-card: add comment to indicate don't remove platforms

[Mark Brown]

On Wed, 22 Mar 2023 01:12:24 +0000, Kuninori Morimoto wrote:
> Basically CPU and Platform are different Component, but if CPU is using
> soc-generic-dmaengine-pcm, same dev will be shared between CPU and
> Platform, and Simple Card had been supporting it.
> 
> When we focus to clean up Simple Card driver, we tend to remove platforms
> if no Platform was selected, but it is wrong because of above reasons.
> 
> [...]

Applied to

   broonie/sound.git for-next

Thanks!

[1/1] ASoC: simple-card: add comment to indicate don't remove platforms
      commit: dcf084247e376b35c139c922d30917f4ff716a4c

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