From: Marcus Cooper <[email protected]>
The i2s block supports multi-lane i2s output however this functionality
is only possible in earlier SoCs where the pins are exposed and for
the i2s block used for HDMI audio on the later SoCs.
To enable this functionality, an optional property has been added to
the bindings.
Signed-off-by: Marcus Cooper <[email protected]>
---
sound/soc/sunxi/sun4i-i2s.c | 44 ++++++++++++++++++++++++++++++++-----
1 file changed, 39 insertions(+), 5 deletions(-)
diff --git a/sound/soc/sunxi/sun4i-i2s.c b/sound/soc/sunxi/sun4i-i2s.c
index bca73b3c0d74..75217fb52bfa 100644
--- a/sound/soc/sunxi/sun4i-i2s.c
+++ b/sound/soc/sunxi/sun4i-i2s.c
@@ -23,7 +23,7 @@
#define SUN4I_I2S_CTRL_REG 0x00
#define SUN4I_I2S_CTRL_SDO_EN_MASK GENMASK(11, 8)
-#define SUN4I_I2S_CTRL_SDO_EN(sdo) BIT(8 + (sdo))
+#define SUN4I_I2S_CTRL_SDO_EN(lines) (((1 << lines) - 1) << 8)
#define SUN4I_I2S_CTRL_MODE_MASK BIT(5)
#define SUN4I_I2S_CTRL_MODE_SLAVE (1 << 5)
#define SUN4I_I2S_CTRL_MODE_MASTER (0 << 5)
@@ -355,14 +355,23 @@ static int sun4i_i2s_hw_params(struct snd_pcm_substream *substream,
struct sun4i_i2s *i2s = snd_soc_dai_get_drvdata(dai);
int sr, wss, channels;
u32 width;
+ int lines;
channels = params_channels(params);
- if (channels != 2) {
+ if ((channels > dai->driver->playback.channels_max) ||
+ (channels < dai->driver->playback.channels_min)) {
dev_err(dai->dev, "Unsupported number of channels: %d\n",
channels);
return -EINVAL;
}
+ lines = (channels + 1) / 2;
+
+ /* Enable the required output lines */
+ regmap_update_bits(i2s->regmap, SUN4I_I2S_CTRL_REG,
+ SUN4I_I2S_CTRL_SDO_EN_MASK,
+ SUN4I_I2S_CTRL_SDO_EN(lines));
+
if (i2s->variant->is_h3_i2s_based) {
regmap_update_bits(i2s->regmap, SUN8I_I2S_CHAN_CFG_REG,
SUN8I_I2S_CHAN_CFG_TX_SLOT_NUM_MASK,
@@ -373,8 +382,19 @@ static int sun4i_i2s_hw_params(struct snd_pcm_substream *substream,
}
/* Map the channels for playback and capture */
- regmap_field_write(i2s->field_txchanmap, 0x76543210);
regmap_field_write(i2s->field_rxchanmap, 0x00003210);
+ regmap_field_write(i2s->field_txchanmap, 0x10);
+ if (i2s->variant->is_h3_i2s_based) {
+ if (channels > 2)
+ regmap_write(i2s->regmap,
+ SUN8I_I2S_TX_CHAN_MAP_REG+4, 0x32);
+ if (channels > 4)
+ regmap_write(i2s->regmap,
+ SUN8I_I2S_TX_CHAN_MAP_REG+8, 0x54);
+ if (channels > 6)
+ regmap_write(i2s->regmap,
+ SUN8I_I2S_TX_CHAN_MAP_REG+12, 0x76);
+ }
/* Configure the channels */
regmap_field_write(i2s->field_txchansel,
@@ -1057,9 +1077,10 @@ static int sun4i_i2s_init_regmap_fields(struct device *dev,
static int sun4i_i2s_probe(struct platform_device *pdev)
{
struct sun4i_i2s *i2s;
+ struct snd_soc_dai_driver *soc_dai;
struct resource *res;
void __iomem *regs;
- int irq, ret;
+ int irq, ret, val;
i2s = devm_kzalloc(&pdev->dev, sizeof(*i2s), GFP_KERNEL);
if (!i2s)
@@ -1126,6 +1147,19 @@ static int sun4i_i2s_probe(struct platform_device *pdev)
i2s->capture_dma_data.addr = res->start + SUN4I_I2S_FIFO_RX_REG;
i2s->capture_dma_data.maxburst = 8;
+ soc_dai = devm_kmemdup(&pdev->dev, &sun4i_i2s_dai,
+ sizeof(*soc_dai), GFP_KERNEL);
+ if (!soc_dai) {
+ ret = -ENOMEM;
+ goto err_pm_disable;
+ }
+
+ if (!of_property_read_u32(pdev->dev.of_node,
+ "allwinner,playback-channels", &val)) {
+ if (val >= 2 && val <= 8)
+ soc_dai->playback.channels_max = val;
+ }
+
pm_runtime_enable(&pdev->dev);
if (!pm_runtime_enabled(&pdev->dev)) {
ret = sun4i_i2s_runtime_resume(&pdev->dev);
@@ -1135,7 +1169,7 @@ static int sun4i_i2s_probe(struct platform_device *pdev)
ret = devm_snd_soc_register_component(&pdev->dev,
&sun4i_i2s_component,
- &sun4i_i2s_dai, 1);
+ soc_dai, 1);
if (ret) {
dev_err(&pdev->dev, "Could not register DAI\n");
goto err_suspend;
--
2.21.0
On Mon, Jun 03, 2019 at 07:47:32PM +0200, [email protected] wrote:
> From: Marcus Cooper <[email protected]>
>
> The i2s block supports multi-lane i2s output however this functionality
> is only possible in earlier SoCs where the pins are exposed and for
> the i2s block used for HDMI audio on the later SoCs.
>
> To enable this functionality, an optional property has been added to
> the bindings.
>
> Signed-off-by: Marcus Cooper <[email protected]>
I'd like to have Mark's input on this, but I'm really worried about
the interaction with the proper TDM support.
Our fundamental issue is that the controller can have up to 8
channels, but either on 4 lines (instead of 1), or 8 channels on 1
(like proper TDM) (or any combination between the two, but that should
be pretty rare).
You're trying to do the first one, and I'm trying to do the second one.
There's a number of assumptions later on that will break the TDM case,
see below for examples
> ---
> sound/soc/sunxi/sun4i-i2s.c | 44 ++++++++++++++++++++++++++++++++-----
> 1 file changed, 39 insertions(+), 5 deletions(-)
>
> diff --git a/sound/soc/sunxi/sun4i-i2s.c b/sound/soc/sunxi/sun4i-i2s.c
> index bca73b3c0d74..75217fb52bfa 100644
> --- a/sound/soc/sunxi/sun4i-i2s.c
> +++ b/sound/soc/sunxi/sun4i-i2s.c
> @@ -23,7 +23,7 @@
>
> #define SUN4I_I2S_CTRL_REG 0x00
> #define SUN4I_I2S_CTRL_SDO_EN_MASK GENMASK(11, 8)
> -#define SUN4I_I2S_CTRL_SDO_EN(sdo) BIT(8 + (sdo))
> +#define SUN4I_I2S_CTRL_SDO_EN(lines) (((1 << lines) - 1) << 8)
> #define SUN4I_I2S_CTRL_MODE_MASK BIT(5)
> #define SUN4I_I2S_CTRL_MODE_SLAVE (1 << 5)
> #define SUN4I_I2S_CTRL_MODE_MASTER (0 << 5)
> @@ -355,14 +355,23 @@ static int sun4i_i2s_hw_params(struct snd_pcm_substream *substream,
> struct sun4i_i2s *i2s = snd_soc_dai_get_drvdata(dai);
> int sr, wss, channels;
> u32 width;
> + int lines;
>
> channels = params_channels(params);
> - if (channels != 2) {
> + if ((channels > dai->driver->playback.channels_max) ||
> + (channels < dai->driver->playback.channels_min)) {
> dev_err(dai->dev, "Unsupported number of channels: %d\n",
> channels);
> return -EINVAL;
> }
>
> + lines = (channels + 1) / 2;
> +
> + /* Enable the required output lines */
> + regmap_update_bits(i2s->regmap, SUN4I_I2S_CTRL_REG,
> + SUN4I_I2S_CTRL_SDO_EN_MASK,
> + SUN4I_I2S_CTRL_SDO_EN(lines));
> +
This has the assumption that each line will have 2 channels, which is wrong.
> if (i2s->variant->is_h3_i2s_based) {
> regmap_update_bits(i2s->regmap, SUN8I_I2S_CHAN_CFG_REG,
> SUN8I_I2S_CHAN_CFG_TX_SLOT_NUM_MASK,
> @@ -373,8 +382,19 @@ static int sun4i_i2s_hw_params(struct snd_pcm_substream *substream,
> }
>
> /* Map the channels for playback and capture */
> - regmap_field_write(i2s->field_txchanmap, 0x76543210);
> regmap_field_write(i2s->field_rxchanmap, 0x00003210);
> + regmap_field_write(i2s->field_txchanmap, 0x10);
> + if (i2s->variant->is_h3_i2s_based) {
> + if (channels > 2)
> + regmap_write(i2s->regmap,
> + SUN8I_I2S_TX_CHAN_MAP_REG+4, 0x32);
> + if (channels > 4)
> + regmap_write(i2s->regmap,
> + SUN8I_I2S_TX_CHAN_MAP_REG+8, 0x54);
> + if (channels > 6)
> + regmap_write(i2s->regmap,
> + SUN8I_I2S_TX_CHAN_MAP_REG+12, 0x76);
> + }
And this creates a mapping matching that.
> /* Configure the channels */
> regmap_field_write(i2s->field_txchansel,
> @@ -1057,9 +1077,10 @@ static int sun4i_i2s_init_regmap_fields(struct device *dev,
> static int sun4i_i2s_probe(struct platform_device *pdev)
> {
> struct sun4i_i2s *i2s;
> + struct snd_soc_dai_driver *soc_dai;
> struct resource *res;
> void __iomem *regs;
> - int irq, ret;
> + int irq, ret, val;
>
> i2s = devm_kzalloc(&pdev->dev, sizeof(*i2s), GFP_KERNEL);
> if (!i2s)
> @@ -1126,6 +1147,19 @@ static int sun4i_i2s_probe(struct platform_device *pdev)
> i2s->capture_dma_data.addr = res->start + SUN4I_I2S_FIFO_RX_REG;
> i2s->capture_dma_data.maxburst = 8;
>
> + soc_dai = devm_kmemdup(&pdev->dev, &sun4i_i2s_dai,
> + sizeof(*soc_dai), GFP_KERNEL);
> + if (!soc_dai) {
> + ret = -ENOMEM;
> + goto err_pm_disable;
> + }
> +
> + if (!of_property_read_u32(pdev->dev.of_node,
> + "allwinner,playback-channels", &val)) {
> + if (val >= 2 && val <= 8)
> + soc_dai->playback.channels_max = val;
> + }
> +
I'm not quite sure how this works.
of_property_read_u32 will return 0, so you will enter in the
condition. But what happens if the property is missing?
Maxime
--
Maxime Ripard, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
On Tue, 4 Jun 2019 at 09:58, Maxime Ripard <[email protected]> wrote:
>
> On Mon, Jun 03, 2019 at 07:47:32PM +0200, [email protected] wrote:
> > From: Marcus Cooper <[email protected]>
> >
> > The i2s block supports multi-lane i2s output however this functionality
> > is only possible in earlier SoCs where the pins are exposed and for
> > the i2s block used for HDMI audio on the later SoCs.
> >
> > To enable this functionality, an optional property has been added to
> > the bindings.
> >
> > Signed-off-by: Marcus Cooper <[email protected]>
>
> I'd like to have Mark's input on this, but I'm really worried about
> the interaction with the proper TDM support.
>
> Our fundamental issue is that the controller can have up to 8
> channels, but either on 4 lines (instead of 1), or 8 channels on 1
> (like proper TDM) (or any combination between the two, but that should
> be pretty rare).
I understand...maybe the TDM needs to be extended to support this to consider
channel mapping and multiple transfer lines. I was thinking about the later when
someone was requesting support on IIRC a while ago, I thought masking might
of been a solution. These can wait as the only consumer at the moment is
LibreELEC and we can patch it there.
Do you have any ideas Master?
CK
>
> You're trying to do the first one, and I'm trying to do the second one.
>
> There's a number of assumptions later on that will break the TDM case,
> see below for examples
>
> > ---
> > sound/soc/sunxi/sun4i-i2s.c | 44 ++++++++++++++++++++++++++++++++-----
> > 1 file changed, 39 insertions(+), 5 deletions(-)
> >
> > diff --git a/sound/soc/sunxi/sun4i-i2s.c b/sound/soc/sunxi/sun4i-i2s.c
> > index bca73b3c0d74..75217fb52bfa 100644
> > --- a/sound/soc/sunxi/sun4i-i2s.c
> > +++ b/sound/soc/sunxi/sun4i-i2s.c
> > @@ -23,7 +23,7 @@
> >
> > #define SUN4I_I2S_CTRL_REG 0x00
> > #define SUN4I_I2S_CTRL_SDO_EN_MASK GENMASK(11, 8)
> > -#define SUN4I_I2S_CTRL_SDO_EN(sdo) BIT(8 + (sdo))
> > +#define SUN4I_I2S_CTRL_SDO_EN(lines) (((1 << lines) - 1) << 8)
> > #define SUN4I_I2S_CTRL_MODE_MASK BIT(5)
> > #define SUN4I_I2S_CTRL_MODE_SLAVE (1 << 5)
> > #define SUN4I_I2S_CTRL_MODE_MASTER (0 << 5)
> > @@ -355,14 +355,23 @@ static int sun4i_i2s_hw_params(struct snd_pcm_substream *substream,
> > struct sun4i_i2s *i2s = snd_soc_dai_get_drvdata(dai);
> > int sr, wss, channels;
> > u32 width;
> > + int lines;
> >
> > channels = params_channels(params);
> > - if (channels != 2) {
> > + if ((channels > dai->driver->playback.channels_max) ||
> > + (channels < dai->driver->playback.channels_min)) {
> > dev_err(dai->dev, "Unsupported number of channels: %d\n",
> > channels);
> > return -EINVAL;
> > }
> >
> > + lines = (channels + 1) / 2;
> > +
> > + /* Enable the required output lines */
> > + regmap_update_bits(i2s->regmap, SUN4I_I2S_CTRL_REG,
> > + SUN4I_I2S_CTRL_SDO_EN_MASK,
> > + SUN4I_I2S_CTRL_SDO_EN(lines));
> > +
>
> This has the assumption that each line will have 2 channels, which is wrong.
>
> > if (i2s->variant->is_h3_i2s_based) {
> > regmap_update_bits(i2s->regmap, SUN8I_I2S_CHAN_CFG_REG,
> > SUN8I_I2S_CHAN_CFG_TX_SLOT_NUM_MASK,
> > @@ -373,8 +382,19 @@ static int sun4i_i2s_hw_params(struct snd_pcm_substream *substream,
> > }
> >
> > /* Map the channels for playback and capture */
> > - regmap_field_write(i2s->field_txchanmap, 0x76543210);
> > regmap_field_write(i2s->field_rxchanmap, 0x00003210);
> > + regmap_field_write(i2s->field_txchanmap, 0x10);
> > + if (i2s->variant->is_h3_i2s_based) {
> > + if (channels > 2)
> > + regmap_write(i2s->regmap,
> > + SUN8I_I2S_TX_CHAN_MAP_REG+4, 0x32);
> > + if (channels > 4)
> > + regmap_write(i2s->regmap,
> > + SUN8I_I2S_TX_CHAN_MAP_REG+8, 0x54);
> > + if (channels > 6)
> > + regmap_write(i2s->regmap,
> > + SUN8I_I2S_TX_CHAN_MAP_REG+12, 0x76);
> > + }
>
> And this creates a mapping matching that.
>
> > /* Configure the channels */
> > regmap_field_write(i2s->field_txchansel,
> > @@ -1057,9 +1077,10 @@ static int sun4i_i2s_init_regmap_fields(struct device *dev,
> > static int sun4i_i2s_probe(struct platform_device *pdev)
> > {
> > struct sun4i_i2s *i2s;
> > + struct snd_soc_dai_driver *soc_dai;
> > struct resource *res;
> > void __iomem *regs;
> > - int irq, ret;
> > + int irq, ret, val;
> >
> > i2s = devm_kzalloc(&pdev->dev, sizeof(*i2s), GFP_KERNEL);
> > if (!i2s)
> > @@ -1126,6 +1147,19 @@ static int sun4i_i2s_probe(struct platform_device *pdev)
> > i2s->capture_dma_data.addr = res->start + SUN4I_I2S_FIFO_RX_REG;
> > i2s->capture_dma_data.maxburst = 8;
> >
> > + soc_dai = devm_kmemdup(&pdev->dev, &sun4i_i2s_dai,
> > + sizeof(*soc_dai), GFP_KERNEL);
> > + if (!soc_dai) {
> > + ret = -ENOMEM;
> > + goto err_pm_disable;
> > + }
> > +
> > + if (!of_property_read_u32(pdev->dev.of_node,
> > + "allwinner,playback-channels", &val)) {
> > + if (val >= 2 && val <= 8)
> > + soc_dai->playback.channels_max = val;
> > + }
> > +
>
> I'm not quite sure how this works.
>
> of_property_read_u32 will return 0, so you will enter in the
> condition. But what happens if the property is missing?
>
> Maxime
>
> --
> Maxime Ripard, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com
On Tue, 4 Jun 2019 at 09:43, Code Kipper <[email protected]> wrote:
>
> On Tue, 4 Jun 2019 at 09:58, Maxime Ripard <[email protected]> wrote:
> >
> > On Mon, Jun 03, 2019 at 07:47:32PM +0200, [email protected] wrote:
> > > From: Marcus Cooper <[email protected]>
> > >
> > > The i2s block supports multi-lane i2s output however this functionality
> > > is only possible in earlier SoCs where the pins are exposed and for
> > > the i2s block used for HDMI audio on the later SoCs.
> > >
> > > To enable this functionality, an optional property has been added to
> > > the bindings.
> > >
> > > Signed-off-by: Marcus Cooper <[email protected]>
> >
> > I'd like to have Mark's input on this, but I'm really worried about
> > the interaction with the proper TDM support.
> >
> > Our fundamental issue is that the controller can have up to 8
> > channels, but either on 4 lines (instead of 1), or 8 channels on 1
> > (like proper TDM) (or any combination between the two, but that should
> > be pretty rare).
>
> I understand...maybe the TDM needs to be extended to support this to consider
> channel mapping and multiple transfer lines. I was thinking about the later when
> someone was requesting support on IIRC a while ago, I thought masking might
> of been a solution. These can wait as the only consumer at the moment is
> LibreELEC and we can patch it there.
Hi Marcus,
FWIW, the TI McASP driver has support for TDM & (i think?) multiple
transfer lines which are called serializers.
Maybe this can help with inspiration?
see https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/sound/soc/ti/davinci-mcasp.c
sample DTS:
&mcasp0 {
#sound-dai-cells = <0>;
status = "okay";
pinctrl-names = "default";
pinctrl-0 = <&mcasp0_pins>;
op-mode = <0>;
tdm-slots = <8>;
serial-dir = <
2 0 1 0
0 0 0 0
0 0 0 0
0 0 0 0
>;
tx-num-evt = <1>;
rx-num-evt = <1>;
};
Cheers!
> Do you have any ideas Master?
> CK
> >
> > You're trying to do the first one, and I'm trying to do the second one.
> >
> > There's a number of assumptions later on that will break the TDM case,
> > see below for examples
> >
> > > ---
> > > sound/soc/sunxi/sun4i-i2s.c | 44 ++++++++++++++++++++++++++++++++-----
> > > 1 file changed, 39 insertions(+), 5 deletions(-)
> > >
> > > diff --git a/sound/soc/sunxi/sun4i-i2s.c b/sound/soc/sunxi/sun4i-i2s.c
> > > index bca73b3c0d74..75217fb52bfa 100644
> > > --- a/sound/soc/sunxi/sun4i-i2s.c
> > > +++ b/sound/soc/sunxi/sun4i-i2s.c
> > > @@ -23,7 +23,7 @@
> > >
> > > #define SUN4I_I2S_CTRL_REG 0x00
> > > #define SUN4I_I2S_CTRL_SDO_EN_MASK GENMASK(11, 8)
> > > -#define SUN4I_I2S_CTRL_SDO_EN(sdo) BIT(8 + (sdo))
> > > +#define SUN4I_I2S_CTRL_SDO_EN(lines) (((1 << lines) - 1) << 8)
> > > #define SUN4I_I2S_CTRL_MODE_MASK BIT(5)
> > > #define SUN4I_I2S_CTRL_MODE_SLAVE (1 << 5)
> > > #define SUN4I_I2S_CTRL_MODE_MASTER (0 << 5)
> > > @@ -355,14 +355,23 @@ static int sun4i_i2s_hw_params(struct snd_pcm_substream *substream,
> > > struct sun4i_i2s *i2s = snd_soc_dai_get_drvdata(dai);
> > > int sr, wss, channels;
> > > u32 width;
> > > + int lines;
> > >
> > > channels = params_channels(params);
> > > - if (channels != 2) {
> > > + if ((channels > dai->driver->playback.channels_max) ||
> > > + (channels < dai->driver->playback.channels_min)) {
> > > dev_err(dai->dev, "Unsupported number of channels: %d\n",
> > > channels);
> > > return -EINVAL;
> > > }
> > >
> > > + lines = (channels + 1) / 2;
> > > +
> > > + /* Enable the required output lines */
> > > + regmap_update_bits(i2s->regmap, SUN4I_I2S_CTRL_REG,
> > > + SUN4I_I2S_CTRL_SDO_EN_MASK,
> > > + SUN4I_I2S_CTRL_SDO_EN(lines));
> > > +
> >
> > This has the assumption that each line will have 2 channels, which is wrong.
> >
> > > if (i2s->variant->is_h3_i2s_based) {
> > > regmap_update_bits(i2s->regmap, SUN8I_I2S_CHAN_CFG_REG,
> > > SUN8I_I2S_CHAN_CFG_TX_SLOT_NUM_MASK,
> > > @@ -373,8 +382,19 @@ static int sun4i_i2s_hw_params(struct snd_pcm_substream *substream,
> > > }
> > >
> > > /* Map the channels for playback and capture */
> > > - regmap_field_write(i2s->field_txchanmap, 0x76543210);
> > > regmap_field_write(i2s->field_rxchanmap, 0x00003210);
> > > + regmap_field_write(i2s->field_txchanmap, 0x10);
> > > + if (i2s->variant->is_h3_i2s_based) {
> > > + if (channels > 2)
> > > + regmap_write(i2s->regmap,
> > > + SUN8I_I2S_TX_CHAN_MAP_REG+4, 0x32);
> > > + if (channels > 4)
> > > + regmap_write(i2s->regmap,
> > > + SUN8I_I2S_TX_CHAN_MAP_REG+8, 0x54);
> > > + if (channels > 6)
> > > + regmap_write(i2s->regmap,
> > > + SUN8I_I2S_TX_CHAN_MAP_REG+12, 0x76);
> > > + }
> >
> > And this creates a mapping matching that.
> >
> > > /* Configure the channels */
> > > regmap_field_write(i2s->field_txchansel,
> > > @@ -1057,9 +1077,10 @@ static int sun4i_i2s_init_regmap_fields(struct device *dev,
> > > static int sun4i_i2s_probe(struct platform_device *pdev)
> > > {
> > > struct sun4i_i2s *i2s;
> > > + struct snd_soc_dai_driver *soc_dai;
> > > struct resource *res;
> > > void __iomem *regs;
> > > - int irq, ret;
> > > + int irq, ret, val;
> > >
> > > i2s = devm_kzalloc(&pdev->dev, sizeof(*i2s), GFP_KERNEL);
> > > if (!i2s)
> > > @@ -1126,6 +1147,19 @@ static int sun4i_i2s_probe(struct platform_device *pdev)
> > > i2s->capture_dma_data.addr = res->start + SUN4I_I2S_FIFO_RX_REG;
> > > i2s->capture_dma_data.maxburst = 8;
> > >
> > > + soc_dai = devm_kmemdup(&pdev->dev, &sun4i_i2s_dai,
> > > + sizeof(*soc_dai), GFP_KERNEL);
> > > + if (!soc_dai) {
> > > + ret = -ENOMEM;
> > > + goto err_pm_disable;
> > > + }
> > > +
> > > + if (!of_property_read_u32(pdev->dev.of_node,
> > > + "allwinner,playback-channels", &val)) {
> > > + if (val >= 2 && val <= 8)
> > > + soc_dai->playback.channels_max = val;
> > > + }
> > > +
> >
> > I'm not quite sure how this works.
> >
> > of_property_read_u32 will return 0, so you will enter in the
> > condition. But what happens if the property is missing?
> >
> > Maxime
> >
> > --
> > Maxime Ripard, Bootlin
> > Embedded Linux and Kernel engineering
> > https://bootlin.com
>
> --
> You received this message because you are subscribed to the Google Groups "linux-sunxi" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
> To view this discussion on the web, visit https://groups.google.com/d/msgid/linux-sunxi/CAEKpxB%3DRdYF9eEvAJ%2BR7sT6OtdtBWjhMM1am%2BEhaN%3D9ZO9Gd2A%40mail.gmail.com.
> For more options, visit https://groups.google.com/d/optout.