[PATCH 0/2] arm64: dts: renesas: Add SD/OE pin properties

[PATCH 0/2] arm64: dts: renesas: Add SD/OE pin properties

[Sean Anderson]

Linux can configure almost every part of this clock generator without
relying on the OTP settings. However, the correct configuration for the
SD/OE pin cannot be determined automatically. Therefore, add the
appropriate properties to configure this pin.

I have CC'd some people who appear to have access to the following
boards which use the versaclock5:

- Salvator-X
- HiHope RZ/G2[MN] Main Board
- Beacon Embedded Works RZ/G2N Development Kit

as I could not find schematics for these boards. You can help me out by
(pick one):

- Run the following command and send me the output:

    $ grep 10: /sys/kernel/debug/regmap/*-006a/registers

- Measure the voltage on the SD/OE pin on a functioning board.
- Inspect (or send me) the schematic to determine the state of the SD/OE
  pin during normal operation.

My suspicion is that all of these boards use the same configuration
(SD/OE pin tied high) owing to their common design heritage. Thanks in
advance.


Sean Anderson (2):
  arm64: dts: renesas: salvator-xs: Add SD/OE pin properties
  arm64: dts: renesas: ulcb: Add SD/OE pin properties

 arch/arm64/boot/dts/renesas/salvator-xs.dtsi | 2 ++
 arch/arm64/boot/dts/renesas/ulcb.dtsi        | 2 ++
 2 files changed, 4 insertions(+)

-- 
2.35.1.1320.gc452695387.dirty

[PATCH 1/2] arm64: dts: renesas: salvator-xs: Add SD/OE pin properties

[Sean Anderson]

Add SD/OE pin properties to the devicetree so that Linux can configure
the pin without relying on the OTP. This matches the register
configuration reported by Geert [1], as well as my inspection of the
schematic (which shows the SD/OE pin permanently tied high).

[1] https://lore.kernel.org/linux-clk/CAMuHMdW9LMuQLuPEF-Fcs1E6Q7dDzY17VZqu4awKDj5WSTRt=A@mail.gmail.com/

Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
---

 arch/arm64/boot/dts/renesas/salvator-xs.dtsi | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/arm64/boot/dts/renesas/salvator-xs.dtsi b/arch/arm64/boot/dts/renesas/salvator-xs.dtsi
index 08b925624e12..1d18dedb1ff0 100644
--- a/arch/arm64/boot/dts/renesas/salvator-xs.dtsi
+++ b/arch/arm64/boot/dts/renesas/salvator-xs.dtsi
@@ -25,6 +25,8 @@ versaclock6: clock-generator@6a {
 		#clock-cells = <1>;
 		clocks = <&x23_clk>;
 		clock-names = "xin";
+		idt,shutdown = <0>;
+		idt,output-enable-active = <1>;
 	};
 };
 
-- 
2.35.1.1320.gc452695387.dirty

Re: [PATCH 1/2] arm64: dts: renesas: salvator-xs: Add SD/OE pin properties

[Geert Uytterhoeven]

Hi Sean,

On Thu, Sep 26, 2024 at 8:09 PM Sean Anderson <sean.anderson@linux.dev> wrote:
> Add SD/OE pin properties to the devicetree so that Linux can configure
> the pin without relying on the OTP. This matches the register
> configuration reported by Geert [1], as well as my inspection of the
> schematic (which shows the SD/OE pin permanently tied high).
>
> [1] https://lore.kernel.org/linux-clk/CAMuHMdW9LMuQLuPEF-Fcs1E6Q7dDzY17VZqu4awKDj5WSTRt=A@mail.gmail.com/
>
> Signed-off-by: Sean Anderson <sean.anderson@linux.dev>

Thanks for your patch!

Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>

There is no change in the output of

    grep 10: /sys/kernel/debug/regmap/*-006a/registers

before/after this patch, so
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>

BTW, I applied the same patch to salvator-x.dtsi, and that seems to be
fine, too.

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH 1/2] arm64: dts: renesas: salvator-xs: Add SD/OE pin properties

[Geert Uytterhoeven]

On Fri, Sep 27, 2024 at 12:34 PM Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
> On Thu, Sep 26, 2024 at 8:09 PM Sean Anderson <sean.anderson@linux.dev> wrote:
> > Add SD/OE pin properties to the devicetree so that Linux can configure
> > the pin without relying on the OTP. This matches the register
> > configuration reported by Geert [1], as well as my inspection of the
> > schematic (which shows the SD/OE pin permanently tied high).
> >
> > [1] https://lore.kernel.org/linux-clk/CAMuHMdW9LMuQLuPEF-Fcs1E6Q7dDzY17VZqu4awKDj5WSTRt=A@mail.gmail.com/
> >
> > Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
>
> Thanks for your patch!
>
> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
>
> There is no change in the output of
>
>     grep 10: /sys/kernel/debug/regmap/*-006a/registers
>
> before/after this patch, so
> Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
>
> BTW, I applied the same patch to salvator-x.dtsi, and that seems to be
> fine, too.

Thanks, will queue in renesas-devel for v6.13.

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

[PATCH 2/2] arm64: dts: renesas: ulcb: Add SD/OE pin properties

[Sean Anderson]

Add SD/OE pin properties to the devicetree so that Linux can configure
the pin without relying on the OTP. This configuration is based on
inspection of the schematic (which shows the SD/OE pin permanently tied
high).

Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
---

 arch/arm64/boot/dts/renesas/ulcb.dtsi | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/arm64/boot/dts/renesas/ulcb.dtsi b/arch/arm64/boot/dts/renesas/ulcb.dtsi
index a2f66f916048..2a2cd788ffec 100644
--- a/arch/arm64/boot/dts/renesas/ulcb.dtsi
+++ b/arch/arm64/boot/dts/renesas/ulcb.dtsi
@@ -234,6 +234,8 @@ versaclock5: clock-generator@6a {
 		#clock-cells = <1>;
 		clocks = <&x23_clk>;
 		clock-names = "xin";
+		idt,shutdown = <0>;
+		idt,output-enable-active = <1>;
 	};
 };
 
-- 
2.35.1.1320.gc452695387.dirty

Re: [PATCH 2/2] arm64: dts: renesas: ulcb: Add SD/OE pin properties

[Geert Uytterhoeven]

Hi Sean,

On Thu, Sep 26, 2024 at 8:09 PM Sean Anderson <sean.anderson@linux.dev> wrote:
> Add SD/OE pin properties to the devicetree so that Linux can configure
> the pin without relying on the OTP. This configuration is based on
> inspection of the schematic (which shows the SD/OE pin permanently tied
> high).
>
> Signed-off-by: Sean Anderson <sean.anderson@linux.dev>

Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>

There is no change in the output of

    grep 10: /sys/kernel/debug/regmap/*-006a/registers

before/after this patch, so
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH 2/2] arm64: dts: renesas: ulcb: Add SD/OE pin properties

[Geert Uytterhoeven]

On Fri, Sep 27, 2024 at 12:34 PM Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
> On Thu, Sep 26, 2024 at 8:09 PM Sean Anderson <sean.anderson@linux.dev> wrote:
> > Add SD/OE pin properties to the devicetree so that Linux can configure
> > the pin without relying on the OTP. This configuration is based on
> > inspection of the schematic (which shows the SD/OE pin permanently tied
> > high).
> >
> > Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
>
> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
>
> There is no change in the output of
>
>     grep 10: /sys/kernel/debug/regmap/*-006a/registers
>
> before/after this patch, so
> Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>

Thanks, will queue in renesas-devel for v6.13.

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH 0/2] arm64: dts: renesas: Add SD/OE pin properties

[Sean Anderson]

On 9/26/24 14:09, Sean Anderson wrote:
> Linux can configure almost every part of this clock generator without
> relying on the OTP settings. However, the correct configuration for the
> SD/OE pin cannot be determined automatically. Therefore, add the
> appropriate properties to configure this pin.
> 
> I have CC'd some people who appear to have access to the following
> boards which use the versaclock5:
> 
> - Salvator-X
> - HiHope RZ/G2[MN] Main Board
> - Beacon Embedded Works RZ/G2N Development Kit
> 
> as I could not find schematics for these boards. You can help me out by
> (pick one):
> 
> - Run the following command and send me the output:
> 
>     $ grep 10: /sys/kernel/debug/regmap/*-006a/registers
> 
> - Measure the voltage on the SD/OE pin on a functioning board.
> - Inspect (or send me) the schematic to determine the state of the SD/OE
>   pin during normal operation.
> 
> My suspicion is that all of these boards use the same configuration
> (SD/OE pin tied high) owing to their common design heritage. Thanks in
> advance.

+CC the people I meant to CC in the first place

Re: [PATCH 0/2] arm64: dts: renesas: Add SD/OE pin properties

[Geert Uytterhoeven]

Hi Sean,

On Thu, Sep 26, 2024 at 8:24 PM Sean Anderson <sean.anderson@linux.dev> wrote:
> On 9/26/24 14:09, Sean Anderson wrote:
> > Linux can configure almost every part of this clock generator without
> > relying on the OTP settings. However, the correct configuration for the
> > SD/OE pin cannot be determined automatically. Therefore, add the
> > appropriate properties to configure this pin.
> >
> > I have CC'd some people who appear to have access to the following
> > boards which use the versaclock5:
> >
> > - Salvator-X
> > - HiHope RZ/G2[MN] Main Board
> > - Beacon Embedded Works RZ/G2N Development Kit
> >
> > as I could not find schematics for these boards. You can help me out by
> > (pick one):
> >
> > - Run the following command and send me the output:
> >
> >     $ grep 10: /sys/kernel/debug/regmap/*-006a/registers

On Salvator-X (5p49v5923) and ULCB (5p49v5925): 82
On Salvator-XS (5p49v6901): 8a

> > - Inspect (or send me) the schematic to determine the state of the SD/OE
> >   pin during normal operation.
> >
> > My suspicion is that all of these boards use the same configuration
> > (SD/OE pin tied high) owing to their common design heritage. Thanks in
> > advance.

According to the schematics, the SD/OE pin is indeed strapped high on
all of the Salvator-X(S), ULCB, HiHope, and Renesom base and SoM boards.

Gr{oetje,eeting}s,

                        Geert


--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH 0/2] arm64: dts: renesas: Add SD/OE pin properties

[Adam Ford]

On Thu, Sep 26, 2024 at 1:24 PM Sean Anderson <sean.anderson@linux.dev> wrote:
>
> On 9/26/24 14:09, Sean Anderson wrote:
> > Linux can configure almost every part of this clock generator without
> > relying on the OTP settings. However, the correct configuration for the
> > SD/OE pin cannot be determined automatically. Therefore, add the
> > appropriate properties to configure this pin.
> >
> > I have CC'd some people who appear to have access to the following
> > boards which use the versaclock5:
> >
> > - Salvator-X
> > - HiHope RZ/G2[MN] Main Board
> > - Beacon Embedded Works RZ/G2N Development Kit
> >
> > as I could not find schematics for these boards. You can help me out by
> > (pick one):
> >
> > - Run the following command and send me the output:
> >
> >     $ grep 10: /sys/kernel/debug/regmap/*-006a/registers

Without any of the patches from this series, for the Beacon RZ/G2 Boards:

grep 10: /sys/kernel/debug/regmap/*-006a/registers
/sys/kernel/debug/regmap/2-006a/registers:10: a0
/sys/kernel/debug/regmap/4-006a/registers:10: a0

> >
> > - Measure the voltage on the SD/OE pin on a functioning board.
> > - Inspect (or send me) the schematic to determine the state of the SD/OE
> >   pin during normal operation.

The SD/OE pins on the Beacon boards are not tied high by default.
They have an optional pull-up resistor, but it is not populated by
default.

adam
> >
> > My suspicion is that all of these boards use the same configuration
> > (SD/OE pin tied high) owing to their common design heritage. Thanks in
> > advance.
>
> +CC the people I meant to CC in the first place

Re: [PATCH 0/2] arm64: dts: renesas: Add SD/OE pin properties

[Geert Uytterhoeven]

Hi Adam,

On Fri, Sep 27, 2024 at 2:20 PM Adam Ford <aford173@gmail.com> wrote:
> On Thu, Sep 26, 2024 at 1:24 PM Sean Anderson <sean.anderson@linux.dev> wrote:
> > > - Inspect (or send me) the schematic to determine the state of the SD/OE
> > >   pin during normal operation.
>
> The SD/OE pins on the Beacon boards are not tied high by default.
> They have an optional pull-up resistor, but it is not populated by
> default.

Whoops, this "np" suffix is really tiny! I had completely missed it :-(

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH 0/2] arm64: dts: renesas: Add SD/OE pin properties

[Adam Ford]

On Fri, Sep 27, 2024 at 7:38 AM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>
> Hi Adam,
>
> On Fri, Sep 27, 2024 at 2:20 PM Adam Ford <aford173@gmail.com> wrote:
> > On Thu, Sep 26, 2024 at 1:24 PM Sean Anderson <sean.anderson@linux.dev> wrote:
> > > > - Inspect (or send me) the schematic to determine the state of the SD/OE
> > > >   pin during normal operation.
> >
> > The SD/OE pins on the Beacon boards are not tied high by default.
> > They have an optional pull-up resistor, but it is not populated by
> > default.
>
> Whoops, this "np" suffix is really tiny! I had completely missed it :-(

I have worked at Beacon/Logic PD for 10 years and I still miss the
'np' at times.  It's done that way, so when the bill-of-material is
generated from the schematic, the system throws away anything with an
'np' suffix, but it makes it really easy to enable it and know what
the part should be.

adam
>
> Gr{oetje,eeting}s,
>
>                         Geert
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
>
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
>                                 -- Linus Torvalds