[PATCH] ASoC: fsl_sai: Add asynchronous mode support

[Nicolin Chen]

From: Nicolin Chen <Guangy...@freescale.com>

SAI supports these operation modes:
1) asynchronous mode
Both Tx and Rx are set to be asynchronous.
2) synchronous mode (Rx sync with Tx)
Tx is set to be asynchronous, Rx is set to be synchronous.
3) synchronous mode (Tx sync with Rx)
Rx is set to be asynchronous, Tx is set to be synchronous.
4) synchronous mode (Tx/Rx sync with another SAI's Tx)
5) synchronous mode (Tx/Rx sync with another SAI's Rx)

* 4) and 5) are beyond this patch because they are related with another SAI.

As the initial version of this SAI driver, it supported 2) as default while
the others were totally missing.

So this patch just adds supports for 1) and 3).

Signed-off-by: Nicolin Chen <nicole...@gmail.com>
---
.../devicetree/bindings/sound/fsl-sai.txt | 16 ++++++++++++
sound/soc/fsl/fsl_sai.c | 30 +++++++++++++++++++---
sound/soc/fsl/fsl_sai.h | 4 +++
3 files changed, 46 insertions(+), 4 deletions(-)

diff --git a/Documentation/devicetree/bindings/sound/fsl-sai.txt b/Documentation/devicetree/bindings/sound/fsl-sai.txt
index 0f4e238..77864f4 100644
--- a/Documentation/devicetree/bindings/sound/fsl-sai.txt
+++ b/Documentation/devicetree/bindings/sound/fsl-sai.txt
@@ -24,6 +24,22 @@ Required properties:
- big-endian-data: If this property is absent, the little endian mode will
be in use as default, or the big endian mode will be in use for all the
fifo data.
+- fsl,sai-synchronous-rx: This is a boolean property. If present, indicating
+ that SAI will work in the synchronous mode (sync Tx with Rx) which means
+ both the transimitter and receiver will send and receive data by following
+ receiver's bit clocks and frame sync clocks.
+- fsl,sai-asynchronous: This is a boolean property. If present, indicating
+ that SAI will work in the asynchronous mode, which means both transimitter
+ and receiver will send and receive data by following their own bit clocks
+ and frame sync clocks separately.
+
+Note:
+- If both fsl,sai-asynchronous and fsl,sai-synchronous-rx are absent, the
+ default synchronous mode (sync Rx with Tx) will be used, which means both
+ transimitter and receiver will send and receive data by following clocks
+ of transimitter.
+- fsl,sai-asynchronous will be ignored if fsl,sai-synchronous-rx property is
+ already present.

Example:
sai2: sai@40031000 {
diff --git a/sound/soc/fsl/fsl_sai.c b/sound/soc/fsl/fsl_sai.c
index 9f10575..dc84f98 100644
--- a/sound/soc/fsl/fsl_sai.c
+++ b/sound/soc/fsl/fsl_sai.c
@@ -330,12 +330,14 @@ static int fsl_sai_trigger(struct snd_pcm_substream *substream, int cmd,
u32 xcsr, count = 100;

/*
- * The transmitter bit clock and frame sync are to be
- * used by both the transmitter and receiver.
+ * Asynchronous mode: Clear SYNC for both Tx and Rx.
+ * Rx sync with Tx clocks: Clear SYNC for Tx, set it for Rx.
+ * Tx sync with Rx clocks: Clear SYNC for Rx, set it for Tx.
*/
- regmap_update_bits(sai->regmap, FSL_SAI_TCR2, FSL_SAI_CR2_SYNC, 0);
+ regmap_update_bits(sai->regmap, FSL_SAI_TCR2, FSL_SAI_CR2_SYNC,
+ sai->synchronous[TX] ? FSL_SAI_CR2_SYNC : 0);
regmap_update_bits(sai->regmap, FSL_SAI_RCR2, FSL_SAI_CR2_SYNC,
- FSL_SAI_CR2_SYNC);
+ sai->synchronous[RX] ? FSL_SAI_CR2_SYNC : 0);

/*
* It is recommended that the transmitter is the last enabled
@@ -620,6 +622,26 @@ static int fsl_sai_probe(struct platform_device *pdev)
return ret;
}

+ /* Sync Tx with Rx as default by following old DT binding */
+ sai->synchronous[RX] = true;
+ sai->synchronous[TX] = false;
+ fsl_sai_dai.symmetric_rates = 1;
+ fsl_sai_dai.symmetric_channels = 1;
+ fsl_sai_dai.symmetric_samplebits = 1;
+
+ if (of_find_property(np, "fsl,sai-synchronous-rx", NULL)) {
+ /* Sync Rx with Tx */
+ sai->synchronous[RX] = false;
+ sai->synchronous[TX] = true;
+ } else if (of_find_property(np, "fsl,sai-asynchronous", NULL)) {
+ /* Discard all settings for asynchronous mode */
+ sai->synchronous[RX] = false;
+ sai->synchronous[TX] = false;
+ fsl_sai_dai.symmetric_rates = 0;
+ fsl_sai_dai.symmetric_channels = 0;
+ fsl_sai_dai.symmetric_samplebits = 0;
+ }
+
sai->dma_params_rx.addr = res->start + FSL_SAI_RDR;
sai->dma_params_tx.addr = res->start + FSL_SAI_TDR;
sai->dma_params_rx.maxburst = FSL_SAI_MAXBURST_RX;
diff --git a/sound/soc/fsl/fsl_sai.h b/sound/soc/fsl/fsl_sai.h
index 0e6c9f5..c24d3fd 100644
--- a/sound/soc/fsl/fsl_sai.h
+++ b/sound/soc/fsl/fsl_sai.h
@@ -135,9 +135,13 @@ struct fsl_sai {
bool big_endian_data;
bool is_dsp_mode;
bool sai_on_imx;
+ bool synchronous[2];

struct snd_dmaengine_dai_dma_data dma_params_rx;
struct snd_dmaengine_dai_dma_data dma_params_tx;
};

+#define TX 1
+#define RX 0
+
#endif /* __FSL_SAI_H */
--
1.8.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

[Varka Bhadram]

Would be readable if it like this...

fsl,sai-synchronous-rx: This is a boolean property. If present, indicating

that SAI will work in the synchronous mode (sync Tx with Rx) which means

both the transimitter and receiver will send and receive data by following

receiver's bit clocks and frame sync clocks.

- fsl,sai-asynchronous: This is a boolean property. If present, indicating

that SAI will work in the asynchronous mode, which means both transimitter

and receiver will send and receive data by following their own bit clocks

and frame sync clocks separately.

.....

--
Regards,
Varka Bhadram.

[Nicolin Chen]

Hi Varka,

I agree, however, the doc was initialized in that format. Adding
indentations for these two appended lines makes the whole text
look weird. :(

Thank you,
Nicolin

[Varka Bhadram]

Reading comfortably is important for us... :-)
see this:http://lxr.free-electrons.com/source/Documentation/devicetree/bindings/pinctrl/pinctrl-bindings.txt#L46

Thanks....

--
Regards,
Varka Bhadram.

[Nicolin Chen]

Hi Varka,

I know how to write a binding doc with indentations to make it clear.

What I'm talking about is that the doc was created without indentations,
so I don't feel comfortable to add indentations starting from this patch
while leaving the old lines in totally a different style because It's not
helpful to make people read comfortably at all but only forcing people to
think why these two lines are so special to take some steps back here.

So personally I agree with your idea to make doc more readable. But what
I prefer to do is not having a conversation on the style within this patch
but to create an extra style-refinement patch later after this one.

Let's just focus on the function itself right now. We can later send a
patch to refine this entire doc. Won't you think it's a better idea? :)

Thank you
Nicolin

[Mark Brown]

On Wed, Aug 06, 2014 at 10:42:37AM +0800, Nicolin Chen wrote:
> On Tue, Aug 05, 2014 at 04:44:29PM +0530, Varka Bhadram wrote:

> > Reading comfortably is important for us... :-)
> > see this:http://lxr.free-electrons.com/source/Documentation/devicetree/bindings/pinctrl/pinctrl-bindings.txt#L46

> I know how to write a binding doc with indentations to make it clear.

> What I'm talking about is that the doc was created without indentations,
> so I don't feel comfortable to add indentations starting from this patch
> while leaving the old lines in totally a different style because It's not
> helpful to make people read comfortably at all but only forcing people to
> think why these two lines are so special to take some steps back here.

> So personally I agree with your idea to make doc more readable. But what
> I prefer to do is not having a conversation on the style within this patch
> but to create an extra style-refinement patch later after this one.

This is the right approach.

[Mark Brown]

On Tue, Aug 05, 2014 at 03:32:05PM +0800, Nicolin Chen wrote:
> From: Nicolin Chen <Guangy...@freescale.com>
>
> SAI supports these operation modes:
> 1) asynchronous mode
> Both Tx and Rx are set to be asynchronous.
> 2) synchronous mode (Rx sync with Tx)
> Tx is set to be asynchronous, Rx is set to be synchronous.
> 3) synchronous mode (Tx sync with Rx)
> Rx is set to be asynchronous, Tx is set to be synchronous.
> 4) synchronous mode (Tx/Rx sync with another SAI's Tx)
> 5) synchronous mode (Tx/Rx sync with another SAI's Rx)

This seems to conflict with something else on my branch - not checked
what. Can you take a look please?

> +Note:
> +- If both fsl,sai-asynchronous and fsl,sai-synchronous-rx are absent, the
> + default synchronous mode (sync Rx with Tx) will be used, which means both
> + transimitter and receiver will send and receive data by following clocks
> + of transimitter.
> +- fsl,sai-asynchronous will be ignored if fsl,sai-synchronous-rx property is
> + already present.

Might be worth printing an error here.

[Nicolin Chen]

On Thu, Aug 07, 2014 at 10:45:27AM +0100, Mark Brown wrote:
> On Tue, Aug 05, 2014 at 03:32:05PM +0800, Nicolin Chen wrote:
> > From: Nicolin Chen <Guangy...@freescale.com>
> >
> > SAI supports these operation modes:
> > 1) asynchronous mode
> > Both Tx and Rx are set to be asynchronous.
> > 2) synchronous mode (Rx sync with Tx)
> > Tx is set to be asynchronous, Rx is set to be synchronous.
> > 3) synchronous mode (Tx sync with Rx)
> > Rx is set to be asynchronous, Tx is set to be synchronous.
> > 4) synchronous mode (Tx/Rx sync with another SAI's Tx)
> > 5) synchronous mode (Tx/Rx sync with another SAI's Rx)
>
> This seems to conflict with something else on my branch - not checked
> what. Can you take a look please?

Ah...there is another patch that should haven been merged into for-next
is currently missing on the remote branch.

This one:
http://mailman.alsa-project.org/pipermail/alsa-devel/2014-August/079689.html

> > +Note:
> > +- If both fsl,sai-asynchronous and fsl,sai-synchronous-rx are absent, the
> > + default synchronous mode (sync Rx with Tx) will be used, which means both
> > + transimitter and receiver will send and receive data by following clocks
> > + of transimitter.
> > +- fsl,sai-asynchronous will be ignored if fsl,sai-synchronous-rx property is
> > + already present.
>
> Might be worth printing an error here.

Hmm...Do I need to send a v2?

Thank you
Nicolin

[Mark Brown]

On Thu, Aug 07, 2014 at 05:44:05PM +0800, Nicolin Chen wrote:
> On Thu, Aug 07, 2014 at 10:45:27AM +0100, Mark Brown wrote:

> > Might be worth printing an error here.

> Hmm...Do I need to send a v2?

Send an incremental patch.

[Mark Brown]

On Tue, Aug 05, 2014 at 03:32:05PM +0800, Nicolin Chen wrote:
> From: Nicolin Chen <Guangy...@freescale.com>
>
> SAI supports these operation modes:
> 1) asynchronous mode
> Both Tx and Rx are set to be asynchronous.

Applied, thanks.