Re: [PATCH 4/5] ASoC: codecs: Add WCD939x Soundwire slave driver

[Neil Armstrong]

On 25/11/2023 12:55, Konrad Dybcio wrote:
> On 23.11.2023 15:49, Neil Armstrong wrote:
> > Add Soundwire Slave driver for the WCD9390/WCD9395 Audio Codec.
> >
> > The WCD9390/WCD9395 Soundwire Slaves will be used by the
> > main WCD9390/WCD9395 Audio Codec driver to access registers
> > and configure Soundwire RX and TX ports.
> >
> > Signed-off-by: Neil Armstrong <neil.armstrong@xxxxxxxxxx>
> > ---
> [...]
>
>
> > +static struct wcd939x_sdw_ch_info wcd939x_sdw_tx_ch_info[] = {
> > +	WCD_SDW_CH(WCD939X_ADC1, WCD939X_ADC_1_4_PORT, BIT(0)),
> > +	WCD_SDW_CH(WCD939X_ADC2, WCD939X_ADC_1_4_PORT, BIT(1)),
> > +	WCD_SDW_CH(WCD939X_ADC3, WCD939X_ADC_1_4_PORT, BIT(2)),
> > +	WCD_SDW_CH(WCD939X_ADC4, WCD939X_ADC_1_4_PORT, BIT(3)),
> > +	// TOFIX support ADC3/4 & DMIC0/1 on port 2
> Well, fix it or drop it :D
>
> > +	//WCD_SDW_CH(WCD939X_ADC3, WCD939X_ADC_DMIC_1_2_PORT, BIT(0)),
> > +	//WCD_SDW_CH(WCD939X_ADC4, WCD939X_ADC_DMIC_1_2_PORT, BIT(1)),
> > +	//WCD_SDW_CH(WCD939X_DMIC0, WCD939X_ADC_DMIC_1_2_PORT, BIT(2)),
> > +	//WCD_SDW_CH(WCD939X_DMIC1, WCD939X_ADC_DMIC_1_2_PORT, BIT(3)),
> > +	WCD_SDW_CH(WCD939X_DMIC0, WCD939X_DMIC_0_3_MBHC_PORT, BIT(0)),
> > +	WCD_SDW_CH(WCD939X_DMIC1, WCD939X_DMIC_0_3_MBHC_PORT, BIT(1)),
> > +	WCD_SDW_CH(WCD939X_MBHC, WCD939X_DMIC_0_3_MBHC_PORT, BIT(2)),
> > +	WCD_SDW_CH(WCD939X_DMIC2, WCD939X_DMIC_0_3_MBHC_PORT, BIT(2)),
> > +	WCD_SDW_CH(WCD939X_DMIC3, WCD939X_DMIC_0_3_MBHC_PORT, BIT(3)),
> > +	WCD_SDW_CH(WCD939X_DMIC4, WCD939X_DMIC_3_7_PORT, BIT(0)),
> > +	WCD_SDW_CH(WCD939X_DMIC5, WCD939X_DMIC_3_7_PORT, BIT(1)),
> > +	WCD_SDW_CH(WCD939X_DMIC6, WCD939X_DMIC_3_7_PORT, BIT(2)),
> > +	WCD_SDW_CH(WCD939X_DMIC7, WCD939X_DMIC_3_7_PORT, BIT(3)),
> > +};
> [...]
>
> > +
> > +int wcd939x_swr_get_current_bank(struct sdw_slave *sdev)
> > +{
> > +	int bank;
> > +
> > +	bank = sdw_read(sdev, SDW_SCP_CTRL);
> > +
> > +	return ((bank & 0x40) ? 1 : 0);
> bool conversion?
>
> Also, 0x40 == BIT(6), can you look up what it means and #define it?
Ack

> [...]
>
> > +
> > +static int wcd9390_bus_config(struct sdw_slave *slave,
> > +			      struct sdw_bus_params *params)
> > +{
> > +	sdw_write(slave, SWRS_SCP_HOST_CLK_DIV2_CTL_BANK(params->next_bank),
> > +		  0x01);
> similar, BIT(0)
Ack

> [...]
>
> > +	{ WCD939X_EAR_STATUS_REG_2, 0x08 },
> > +	{ WCD939X_FLYBACK_NEW_CTRL_2, 0x00 }, //??
> > +	{ WCD939X_FLYBACK_NEW_CTRL_3, 0x00 }, //??
> > +	{ WCD939X_FLYBACK_NEW_CTRL_4, 0x44 }, //??
> drop //s

Ack

> [...]
>
> > +static bool wcd939x_volatile_register(struct device *dev, unsigned int reg)
> > +{
> > +	if (reg <= WCD939X_BASE)
> > +		return false;
> Maybe move this check to readonly_register
> > +
> > +	if (wcd939x_readonly_register(dev, reg))
> > +		return true;
> and call readonly for .volatile_reg as well?
> [...]

Hmm, let me check

> > +	/**
> > +	 * Port map index starts with 0, however the data port for this codec
> > +	 * are from index 1
> > +	 */
> This is not kerneldoc

Ack

> > +	if (of_property_read_bool(dev->of_node, "qcom,tx-port-mapping")) {
> > +		wcd->is_tx = true;
> > +		ret = of_property_read_u32_array(dev->of_node,
> > +						 "qcom,tx-port-mapping",
> > +						 &pdev->m_port_map[1],
> > +						 WCD939X_MAX_TX_SWR_PORTS);
> > +	} else {
> > +		ret = of_property_read_u32_array(dev->of_node,
> > +						 "qcom,rx-port-mapping",
> > +						 &pdev->m_port_map[1],
> > +						 WCD939X_MAX_RX_SWR_PORTS);
> > +	}
> This is used in wcd9380 and will be used in wcd9370 when that happens some
> day, maybe it'd be worth to commonize it as qcom_{rx/tx}_portmap_get?
> [...]

OK but where ?

> > +static const struct sdw_device_id wcd9390_slave_id[] = {
> > +	SDW_SLAVE_ENTRY(0x0217, 0x10e, 0),
> 0x10e - WCD9380 or 9385 slave? an inline comment at the end of the line
> would be cool!

Ack

> Konrad