On 4 December 2017 at 08:34, Vasily Khoruzhick anarsoul@gmail.com wrote:
I'd keep it sun50i_a64_acodec_i2s, since other 3 I2S modules aren't compatible with this one, but similar to H3.
Well I wouldn't and the description of it is usage is clearly stated in the devicetree for the dai.
On Sun, Dec 3, 2017 at 10:42 PM, Code Kipper codekipper@gmail.com wrote:
On 3 December 2017 at 21:41, Vasily Khoruzhick anarsoul@gmail.com wrote:
From: Marcus Cooper codekipper@gmail.com
The I2S block used for the audio codec in the A64 is very similar to what is used by the A10(sun4i) devices. However, its TX FIFO is located at a different address.
[vasilykh: - added fixed_wss and wss_value to A64 quirks, - changed compatible to 'allwinner,sun50i-a64-acodec-i2s, since A64 has 3 more I2S blocks that are not compatible with audio codec I2S]
This needs to be investigated more...I'm suspicious of the SUN8I_I2S_FMT0_LRCK_PERIOD value which is hardcoded to 0x1f. I've made this following change https://github.com/codekipper/linux-sunxi/commit/4c5fe5f5576cfbb2e1e94a99cd1... which seems to work for me but I would like to look at it with a logic analyser and test other sample depths. CK
Signed-off-by: Marcus Cooper codekipper@gmail.com Signed-off-by: Vasily Khoruzhick anarsoul@gmail.com
.../devicetree/bindings/sound/sun4i-i2s.txt | 2 ++ sound/soc/sunxi/sun4i-i2s.c | 23 ++++++++++++++++++++++ 2 files changed, 25 insertions(+)
diff --git a/Documentation/devicetree/bindings/sound/sun4i-i2s.txt b/Documentation/devicetree/bindings/sound/sun4i-i2s.txt index 05d7135a8d2f..ab86f266962a 100644 --- a/Documentation/devicetree/bindings/sound/sun4i-i2s.txt +++ b/Documentation/devicetree/bindings/sound/sun4i-i2s.txt @@ -9,6 +9,7 @@ Required properties: - "allwinner,sun4i-a10-i2s" - "allwinner,sun6i-a31-i2s" - "allwinner,sun8i-h3-i2s"
- "allwinner,sun50i-a64-acodec-i2s"
- reg: physical base address of the controller and length of memory mapped region.
- interrupts: should contain the I2S interrupt.
@@ -24,6 +25,7 @@ Required properties: Required properties for the following compatibles: - "allwinner,sun6i-a31-i2s" - "allwinner,sun8i-h3-i2s"
- "allwinner,sun50i-a64-acodec-i2s"I would keep the compatible format the same. ie. allwinner,sun50i-a64-i2s. The A33 for example uses the sun6i-a31-i2s for it's dai.
- resets: phandle to the reset line for this codec
Example: diff --git a/sound/soc/sunxi/sun4i-i2s.c b/sound/soc/sunxi/sun4i-i2s.c index 54c16eb64713..2c060e015725 100644 --- a/sound/soc/sunxi/sun4i-i2s.c +++ b/sound/soc/sunxi/sun4i-i2s.c @@ -942,6 +942,25 @@ static const struct sun4i_i2s_quirks sun8i_h3_i2s_quirks = { .field_rxchansel = REG_FIELD(SUN8I_I2S_RX_CHAN_SEL_REG, 0, 2), };
+static const struct sun4i_i2s_quirks sun50i_a64_acodec_i2s_quirks = {
ditto
.has_reset = true,.reg_offset_txdata = SUN8I_I2S_FIFO_TX_REG,.sun4i_i2s_regmap = &sun4i_i2s_regmap_config,.has_slave_select_bit = true,.fixed_wss = true,.wss_value = 3,.field_clkdiv_mclk_en = REG_FIELD(SUN4I_I2S_CLK_DIV_REG, 7, 7),.field_fmt_wss = REG_FIELD(SUN4I_I2S_FMT0_REG, 2, 3),.field_fmt_sr = REG_FIELD(SUN4I_I2S_FMT0_REG, 4, 5),.field_fmt_bclk = REG_FIELD(SUN4I_I2S_FMT0_REG, 6, 6),.field_fmt_lrclk = REG_FIELD(SUN4I_I2S_FMT0_REG, 7, 7),.field_fmt_mode = REG_FIELD(SUN4I_I2S_FMT0_REG, 0, 1),.field_txchanmap = REG_FIELD(SUN4I_I2S_TX_CHAN_MAP_REG, 0, 31),.field_rxchanmap = REG_FIELD(SUN4I_I2S_RX_CHAN_MAP_REG, 0, 31),.field_txchansel = REG_FIELD(SUN4I_I2S_TX_CHAN_SEL_REG, 0, 2),.field_rxchansel = REG_FIELD(SUN4I_I2S_RX_CHAN_SEL_REG, 0, 2),+};
static int sun4i_i2s_init_regmap_fields(struct device *dev, struct sun4i_i2s *i2s) { @@ -1146,6 +1165,10 @@ static const struct of_device_id sun4i_i2s_match[] = { .compatible = "allwinner,sun8i-h3-i2s", .data = &sun8i_h3_i2s_quirks, },
{.compatible = "allwinner,sun50i-a64-acodec-i2s",.data = &sun50i_a64_acodec_i2s_quirks,ditto BR, CK
}, {}}; MODULE_DEVICE_TABLE(of, sun4i_i2s_match); -- 2.15.0