From: Yangtao Li <frank@allwinnertech.com>

A100 perf1 hava MicroSD slot and on-board eMMC module, add support for them.

Signed-off-by: Yangtao Li <frank@allwinnertech.com>
Signed-off-by: Cody Eksal <masterr3c0rd@epochal.quest>
---
 .../allwinner/sun50i-a100-allwinner-perf1.dts    | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a100-allwinner-perf1.dts b/arch/arm64/boot/dts/allwinner/sun50i-a100-allwinner-perf1.dts
index 2f8c7ee60283..d418fc272b3c 100644
--- a/arch/arm64/boot/dts/allwinner/sun50i-a100-allwinner-perf1.dts
+++ b/arch/arm64/boot/dts/allwinner/sun50i-a100-allwinner-perf1.dts
@@ -39,6 +39,22 @@ &ehci1 {
 	status = "okay";
 };
 
+&mmc0 {
+	vmmc-supply = <&reg_dcdc1>;
+	cd-gpios = <&pio 5 6 GPIO_ACTIVE_LOW>; /* PF6 */
+	bus-width = <4>;
+	status = "okay";
+};
+
+&mmc2 {
+	vmmc-supply = <&reg_dcdc1>;
+	vqmmc-supply = <&reg_aldo1>;
+	cap-mmc-hw-reset;
+	non-removable;
+	bus-width = <8>;
+	status = "okay";
+};
+
 &ohci0 {
 	status = "okay";
 };
-- 
2.47.0

On Thu, Oct 31, 2024 at 3:03 PM Cody Eksal <masterr3c0rd@epochal.quest> wrote:
>
> From: Yangtao Li <frank@allwinnertech.com>
>
> A100 perf1 hava MicroSD slot and on-board eMMC module, add support for them.
>
> Signed-off-by: Yangtao Li <frank@allwinnertech.com>
> Signed-off-by: Cody Eksal <masterr3c0rd@epochal.quest>

So I skipped this one since Andre had some questions about the previous
mmc clock patch. And I believe this one won't work without it?

ChenYu

> ---
>  .../allwinner/sun50i-a100-allwinner-perf1.dts    | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a100-allwinner-perf1.dts b/arch/arm64/boot/dts/allwinner/sun50i-a100-allwinner-perf1.dts
> index 2f8c7ee60283..d418fc272b3c 100644
> --- a/arch/arm64/boot/dts/allwinner/sun50i-a100-allwinner-perf1.dts
> +++ b/arch/arm64/boot/dts/allwinner/sun50i-a100-allwinner-perf1.dts
> @@ -39,6 +39,22 @@ &ehci1 {
>         status = "okay";
>  };
>
> +&mmc0 {
> +       vmmc-supply = <&reg_dcdc1>;
> +       cd-gpios = <&pio 5 6 GPIO_ACTIVE_LOW>; /* PF6 */
> +       bus-width = <4>;
> +       status = "okay";
> +};
> +
> +&mmc2 {
> +       vmmc-supply = <&reg_dcdc1>;
> +       vqmmc-supply = <&reg_aldo1>;
> +       cap-mmc-hw-reset;
> +       non-removable;
> +       bus-width = <8>;
> +       status = "okay";
> +};
> +
>  &ohci0 {
>         status = "okay";
>  };
> --
> 2.47.0
>

On 2024/11/02 8:48 am, Chen-Yu Tsai wrote:
> On Thu, Oct 31, 2024 at 3:03 PM Cody Eksal <masterr3c0rd@epochal.quest> wrote:
>>
>> From: Yangtao Li <frank@allwinnertech.com>
>>
>> A100 perf1 hava MicroSD slot and on-board eMMC module, add support for them.
>>
>> Signed-off-by: Yangtao Li <frank@allwinnertech.com>
>> Signed-off-by: Cody Eksal <masterr3c0rd@epochal.quest>
> 
> So I skipped this one since Andre had some questions about the previous
> mmc clock patch. And I believe this one won't work without it?
eMMC (mmc2) works fine after more testing, even without the clock
reparenting patch (in fact, it appears to work more consistently), but
mmc0 and mmc1 require more testing and debugging. I'll deal with
the remaining MMC issues in a future patch.

- Cody

> ChenYu
> 
>> ---
>>  .../allwinner/sun50i-a100-allwinner-perf1.dts    | 16 ++++++++++++++++
>>  1 file changed, 16 insertions(+)

On Sun, Nov 3, 2024 at 5:23 AM Cody Eksal <masterr3c0rd@epochal.quest> wrote:
>
> On 2024/11/02 8:48 am, Chen-Yu Tsai wrote:
> > On Thu, Oct 31, 2024 at 3:03 PM Cody Eksal <masterr3c0rd@epochal.quest> wrote:
> >>
> >> From: Yangtao Li <frank@allwinnertech.com>
> >>
> >> A100 perf1 hava MicroSD slot and on-board eMMC module, add support for them.
> >>
> >> Signed-off-by: Yangtao Li <frank@allwinnertech.com>
> >> Signed-off-by: Cody Eksal <masterr3c0rd@epochal.quest>
> >
> > So I skipped this one since Andre had some questions about the previous
> > mmc clock patch. And I believe this one won't work without it?
> eMMC (mmc2) works fine after more testing, even without the clock
> reparenting patch (in fact, it appears to work more consistently), but
> mmc0 and mmc1 require more testing and debugging. I'll deal with
> the remaining MMC issues in a future patch.

I think it would be better if only the eMMC was enabled. Enabling something
that is almost guaranteed to cause data corruption isn't going to be fun
for users.

ChenYu