OrangePi 3 can optionally have 8 GiB eMMC (soldered on board). Because
those pins are dedicated to eMMC exclusively, node can be added for both
variants (with and without eMMC). Kernel will then scan bus for presence
of eMMC and act accordingly.
Signed-off-by: Jernej Skrabec <[email protected]>
---
Changes since v1:
- don't make separate DT just for -emmc variant - add node to existing
orangepi 3 DT
arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
index c311eee52a35..1e0abd9d047f 100644
--- a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
+++ b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
@@ -144,6 +144,15 @@ brcm: sdio-wifi@1 {
};
};
+&mmc2 {
+ vmmc-supply = <®_cldo1>;
+ vqmmc-supply = <®_bldo2>;
+ cap-mmc-hw-reset;
+ non-removable;
+ bus-width = <8>;
+ status = "okay";
+};
+
&ohci0 {
status = "okay";
};
--
2.25.0
On Mon, Feb 10, 2020 at 06:40:07PM +0100, Jernej Skrabec wrote:
> OrangePi 3 can optionally have 8 GiB eMMC (soldered on board). Because
> those pins are dedicated to eMMC exclusively, node can be added for both
> variants (with and without eMMC). Kernel will then scan bus for presence
> of eMMC and act accordingly.
>
> Signed-off-by: Jernej Skrabec <[email protected]>
> ---
> Changes since v1:
> - don't make separate DT just for -emmc variant - add node to existing
> orangepi 3 DT
>
> arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> index c311eee52a35..1e0abd9d047f 100644
> --- a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> +++ b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> @@ -144,6 +144,15 @@ brcm: sdio-wifi@1 {
> };
> };
>
> +&mmc2 {
> + vmmc-supply = <®_cldo1>;
> + vqmmc-supply = <®_bldo2>;
> + cap-mmc-hw-reset;
> + non-removable;
Given that non-removable is documented as "Non-removable slot (like
eMMC); assume always present.", we should probably get rid of that
property?
Maxime
Hi!
Dne torek, 11. februar 2020 ob 07:51:41 CET je Maxime Ripard napisal(a):
> On Mon, Feb 10, 2020 at 06:40:07PM +0100, Jernej Skrabec wrote:
> > OrangePi 3 can optionally have 8 GiB eMMC (soldered on board). Because
> > those pins are dedicated to eMMC exclusively, node can be added for both
> > variants (with and without eMMC). Kernel will then scan bus for presence
> > of eMMC and act accordingly.
> >
> > Signed-off-by: Jernej Skrabec <[email protected]>
> > ---
> > Changes since v1:
> > - don't make separate DT just for -emmc variant - add node to existing
> >
> > orangepi 3 DT
> >
> > arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts | 9 +++++++++
> > 1 file changed, 9 insertions(+)
> >
> > diff --git a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> > b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts index
> > c311eee52a35..1e0abd9d047f 100644
> > --- a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> > +++ b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> > @@ -144,6 +144,15 @@ brcm: sdio-wifi@1 {
> >
> > };
> >
> > };
> >
> > +&mmc2 {
> > + vmmc-supply = <®_cldo1>;
> > + vqmmc-supply = <®_bldo2>;
> > + cap-mmc-hw-reset;
> > + non-removable;
>
> Given that non-removable is documented as "Non-removable slot (like
> eMMC); assume always present.", we should probably get rid of that
> property?
I checked mmc core code and this property means that bus will be scanned only
once. In this form, node doesn't tell what kind of device is connected, so
core has to scan it no matter if "non-removable" property is present or not. I
maybe missed something though, so it would be great if someone can check it
again.
Best regards,
Jernej
On Tue, Feb 11, 2020 at 08:09:57AM +0100, Jernej Škrabec wrote:
> Hi!
>
> Dne torek, 11. februar 2020 ob 07:51:41 CET je Maxime Ripard napisal(a):
> > On Mon, Feb 10, 2020 at 06:40:07PM +0100, Jernej Skrabec wrote:
> > > OrangePi 3 can optionally have 8 GiB eMMC (soldered on board). Because
> > > those pins are dedicated to eMMC exclusively, node can be added for both
> > > variants (with and without eMMC). Kernel will then scan bus for presence
> > > of eMMC and act accordingly.
> > >
> > > Signed-off-by: Jernej Skrabec <[email protected]>
> > > ---
> > > Changes since v1:
> > > - don't make separate DT just for -emmc variant - add node to existing
> > >
> > > orangepi 3 DT
> > >
> > > arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts | 9 +++++++++
> > > 1 file changed, 9 insertions(+)
> > >
> > > diff --git a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> > > b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts index
> > > c311eee52a35..1e0abd9d047f 100644
> > > --- a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> > > +++ b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> > > @@ -144,6 +144,15 @@ brcm: sdio-wifi@1 {
> > >
> > > };
> > >
> > > };
> > >
> > > +&mmc2 {
> > > + vmmc-supply = <®_cldo1>;
> > > + vqmmc-supply = <®_bldo2>;
> > > + cap-mmc-hw-reset;
> > > + non-removable;
> >
> > Given that non-removable is documented as "Non-removable slot (like
> > eMMC); assume always present.", we should probably get rid of that
> > property?
>
> I checked mmc core code and this property means that bus will be scanned only
> once. In this form, node doesn't tell what kind of device is connected, so
> core has to scan it no matter if "non-removable" property is present or not. I
> maybe missed something though, so it would be great if someone can check it
> again.
It looks like it does indeed :)
I've applied that patch, thanks!
Maxime
Hi,
On Mon, Feb 10, 2020 at 06:40:07PM +0100, Jernej Skrabec wrote:
> OrangePi 3 can optionally have 8 GiB eMMC (soldered on board). Because
> those pins are dedicated to eMMC exclusively, node can be added for both
> variants (with and without eMMC). Kernel will then scan bus for presence
> of eMMC and act accordingly.
Tested-by: Ondrej Jirman <[email protected]> (on a variant without eMMC)
It didn't magically add extra 8GiB of storage, but it didn't break anything
either. :)
regards,
o.
> Signed-off-by: Jernej Skrabec <[email protected]>
> ---
> Changes since v1:
> - don't make separate DT just for -emmc variant - add node to existing
> orangepi 3 DT
>
> arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> index c311eee52a35..1e0abd9d047f 100644
> --- a/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> +++ b/arch/arm64/boot/dts/allwinner/sun50i-h6-orangepi-3.dts
> @@ -144,6 +144,15 @@ brcm: sdio-wifi@1 {
> };
> };
>
> +&mmc2 {
> + vmmc-supply = <®_cldo1>;
> + vqmmc-supply = <®_bldo2>;
> + cap-mmc-hw-reset;
> + non-removable;
> + bus-width = <8>;
> + status = "okay";
> +};
> +
> &ohci0 {
> status = "okay";
> };
> --
> 2.25.0
>
> --
> 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/20200210174007.118575-1-jernej.skrabec%40siol.net.