Re: [PATCH 1/3] pinctrl: rockchip: update rk3308 iomux routes

[Heiko Stübner <heiko@xxxxxxxxx>]

Am Mittwoch, 15. Mai 2024, 14:16:32 CEST schrieb Dmitry Yashin:
> Some of the rk3308 iomux routes in rk3308_mux_route_data belong to
> the rk3308b SoC. Remove them and correct i2c3 routes.
> 
> Fixes: 7825aeb7b208 ("pinctrl: rockchip: add rk3308 SoC support")
> Signed-off-by: Dmitry Yashin <dmt.yashin@xxxxxxxxx>
> ---
>  drivers/pinctrl/pinctrl-rockchip.c | 17 ++---------------
>  1 file changed, 2 insertions(+), 15 deletions(-)
> 
> diff --git a/drivers/pinctrl/pinctrl-rockchip.c b/drivers/pinctrl/pinctrl-rockchip.c
> index 3bedf36a0019..cc647db76927 100644
> --- a/drivers/pinctrl/pinctrl-rockchip.c
> +++ b/drivers/pinctrl/pinctrl-rockchip.c
> @@ -870,9 +870,8 @@ static struct rockchip_mux_route_data rk3308_mux_route_data[] = {
>  	RK_MUXROUTE_SAME(0, RK_PC3, 1, 0x314, BIT(16 + 0) | BIT(0)), /* rtc_clk */
>  	RK_MUXROUTE_SAME(1, RK_PC6, 2, 0x314, BIT(16 + 2) | BIT(16 + 3)), /* uart2_rxm0 */
>  	RK_MUXROUTE_SAME(4, RK_PD2, 2, 0x314, BIT(16 + 2) | BIT(16 + 3) | BIT(2)), /* uart2_rxm1 */
> -	RK_MUXROUTE_SAME(0, RK_PB7, 2, 0x608, BIT(16 + 8) | BIT(16 + 9)), /* i2c3_sdam0 */
> -	RK_MUXROUTE_SAME(3, RK_PB4, 2, 0x608, BIT(16 + 8) | BIT(16 + 9) | BIT(8)), /* i2c3_sdam1 */
> -	RK_MUXROUTE_SAME(2, RK_PA0, 3, 0x608, BIT(16 + 8) | BIT(16 + 9) | BIT(9)), /* i2c3_sdam2 */
> +	RK_MUXROUTE_SAME(0, RK_PB7, 2, 0x314, BIT(16 + 4)), /* i2c3_sdam0 */
> +	RK_MUXROUTE_SAME(3, RK_PB4, 2, 0x314, BIT(16 + 4) | BIT(4)), /* i2c3_sdam1 */

checked against TRM and no devicetree in the kernel currently enables i2c3 at all.

>  	RK_MUXROUTE_SAME(1, RK_PA3, 2, 0x308, BIT(16 + 3)), /* i2s-8ch-1-sclktxm0 */
>  	RK_MUXROUTE_SAME(1, RK_PA4, 2, 0x308, BIT(16 + 3)), /* i2s-8ch-1-sclkrxm0 */
>  	RK_MUXROUTE_SAME(1, RK_PB5, 2, 0x308, BIT(16 + 3) | BIT(3)), /* i2s-8ch-1-sclktxm1 */
> @@ -881,18 +880,6 @@ static struct rockchip_mux_route_data rk3308_mux_route_data[] = {
>  	RK_MUXROUTE_SAME(1, RK_PB6, 4, 0x308, BIT(16 + 12) | BIT(16 + 13) | BIT(12)), /* pdm-clkm1 */
>  	RK_MUXROUTE_SAME(2, RK_PA6, 2, 0x308, BIT(16 + 12) | BIT(16 + 13) | BIT(13)), /* pdm-clkm2 */
>  	RK_MUXROUTE_SAME(2, RK_PA4, 3, 0x600, BIT(16 + 2) | BIT(2)), /* pdm-clkm-m2 */
> -	RK_MUXROUTE_SAME(3, RK_PB2, 3, 0x314, BIT(16 + 9)), /* spi1_miso */
> -	RK_MUXROUTE_SAME(2, RK_PA4, 2, 0x314, BIT(16 + 9) | BIT(9)), /* spi1_miso_m1 */
> -	RK_MUXROUTE_SAME(0, RK_PB3, 3, 0x314, BIT(16 + 10) | BIT(16 + 11)), /* owire_m0 */
> -	RK_MUXROUTE_SAME(1, RK_PC6, 7, 0x314, BIT(16 + 10) | BIT(16 + 11) | BIT(10)), /* owire_m1 */
> -	RK_MUXROUTE_SAME(2, RK_PA2, 5, 0x314, BIT(16 + 10) | BIT(16 + 11) | BIT(11)), /* owire_m2 */
> -	RK_MUXROUTE_SAME(0, RK_PB3, 2, 0x314, BIT(16 + 12) | BIT(16 + 13)), /* can_rxd_m0 */
> -	RK_MUXROUTE_SAME(1, RK_PC6, 5, 0x314, BIT(16 + 12) | BIT(16 + 13) | BIT(12)), /* can_rxd_m1 */
> -	RK_MUXROUTE_SAME(2, RK_PA2, 4, 0x314, BIT(16 + 12) | BIT(16 + 13) | BIT(13)), /* can_rxd_m2 */

> -	RK_MUXROUTE_SAME(1, RK_PC4, 3, 0x314, BIT(16 + 14)), /* mac_rxd0_m0 */
> -	RK_MUXROUTE_SAME(4, RK_PA2, 2, 0x314, BIT(16 + 14) | BIT(14)), /* mac_rxd0_m1 */

> -	RK_MUXROUTE_SAME(3, RK_PB4, 4, 0x314, BIT(16 + 15)), /* uart3_rx */
> -	RK_MUXROUTE_SAME(0, RK_PC1, 3, 0x314, BIT(16 + 15) | BIT(15)), /* uart3_rx_m1 */

checked against the TRM too.
And as far as usage:
- spi1: not used at all
- owire: not used at all
- can: not used at all
- mac_rxd0: all boards use the (default) m0 variant
- uart3: not currently enabled anywhere


And I guess the rk3308 is niche enough that people either use the vendor-kernel
or have the dts in mainline, so we should have a ok grasp on not causing breakage
by some accidental rk3308b somewhere.

Reviewed-by: Heiko Stuebner <heiko@xxxxxxxxx>