The controllers present in the D1 are extremely similar to the R40
and require the same reset quirks, but An extra quirk is needed to support
receiving packets.

Signed-off-by: John Watts <contact@jookia.org>
---
 drivers/net/can/Kconfig     |  4 ++--
 drivers/net/can/sun4i_can.c | 12 +++++++++++-
 2 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig
index a5c5036dfb94..e626de33e735 100644
--- a/drivers/net/can/Kconfig
+++ b/drivers/net/can/Kconfig
@@ -185,10 +185,10 @@ config CAN_SLCAN
 
 config CAN_SUN4I
 	tristate "Allwinner A10 CAN controller"
-	depends on MACH_SUN4I || MACH_SUN7I || COMPILE_TEST
+	depends on MACH_SUN4I || MACH_SUN7I || RISCV || COMPILE_TEST
 	help
 	  Say Y here if you want to use CAN controller found on Allwinner
-	  A10/A20 SoCs.
+	  A10/A20/D1 SoCs.
 
 	  To compile this driver as a module, choose M here: the module will
 	  be called sun4i_can.
diff --git a/drivers/net/can/sun4i_can.c b/drivers/net/can/sun4i_can.c
index 1f90fe6dbb8b..c508a328e38d 100644
--- a/drivers/net/can/sun4i_can.c
+++ b/drivers/net/can/sun4i_can.c
@@ -91,6 +91,8 @@
 #define SUN4I_REG_BUF12_ADDR	0x0070	/* CAN Tx/Rx Buffer 12 */
 #define SUN4I_REG_ACPC_ADDR	0x0040	/* CAN Acceptance Code 0 */
 #define SUN4I_REG_ACPM_ADDR	0x0044	/* CAN Acceptance Mask 0 */
+#define SUN4I_REG_ACPC_ADDR_D1	0x0028	/* CAN Acceptance Code 0 on the D1 */
+#define SUN4I_REG_ACPM_ADDR_D1	0x002C	/* CAN Acceptance Mask 0 on the D1 */
 #define SUN4I_REG_RBUF_RBACK_START_ADDR	0x0180	/* CAN transmit buffer start */
 #define SUN4I_REG_RBUF_RBACK_END_ADDR	0x01b0	/* CAN transmit buffer end */
 
@@ -779,6 +781,11 @@ static const struct sun4ican_quirks sun4ican_quirks_r40 = {
 	.acp_offset = 0,
 };
 
+static const struct sun4ican_quirks sun4ican_quirks_d1 = {
+	.has_reset = true,
+	.acp_offset = (SUN4I_REG_ACPC_ADDR_D1 - SUN4I_REG_ACPC_ADDR),
+};
+
 static const struct of_device_id sun4ican_of_match[] = {
 	{
 		.compatible = "allwinner,sun4i-a10-can",
@@ -789,6 +796,9 @@ static const struct of_device_id sun4ican_of_match[] = {
 	}, {
 		.compatible = "allwinner,sun8i-r40-can",
 		.data = &sun4ican_quirks_r40
+	}, {
+		.compatible = "allwinner,sun20i-d1-can",
+		.data = &sun4ican_quirks_d1
 	}, {
 		/* sentinel */
 	},
@@ -913,4 +923,4 @@ module_platform_driver(sun4i_can_driver);
 MODULE_AUTHOR("Peter Chen <xingkongcp@gmail.com>");
 MODULE_AUTHOR("Gerhard Bertelsmann <info@gerhard-bertelsmann.de>");
 MODULE_LICENSE("Dual BSD/GPL");
-MODULE_DESCRIPTION("CAN driver for Allwinner SoCs (A10/A20)");
+MODULE_DESCRIPTION("CAN driver for Allwinner SoCs (A10/A20/D1)");
-- 
2.41.0

Hi John,

On Sat, Jul 22, 2023 at 12:18 AM John Watts <contact@jookia.org> wrote:
> The controllers present in the D1 are extremely similar to the R40
> and require the same reset quirks, but An extra quirk is needed to support
> receiving packets.
>
> Signed-off-by: John Watts <contact@jookia.org>

Thanks for your patch, which is now commit 8abb95250ae6af2d ("can:
sun4i_can: Add support for the Allwinner D1") in linux-can-next/master.

> --- a/drivers/net/can/Kconfig
> +++ b/drivers/net/can/Kconfig
> @@ -185,10 +185,10 @@ config CAN_SLCAN
>
>  config CAN_SUN4I
>         tristate "Allwinner A10 CAN controller"
> -       depends on MACH_SUN4I || MACH_SUN7I || COMPILE_TEST
> +       depends on MACH_SUN4I || MACH_SUN7I || RISCV || COMPILE_TEST

This makes this question pop up when configuring a kernel for any RISC-V
platform, not just for Allwinner RISC-V platforms.

In comparison, drivers/clk/sunxi-ng/Kconfig does have some

    depends on MACH_SUN<foo>I || RISCV || COMPILE_TEST

but these are gated by ARCH_SUNXI at the top of the file.

I'm not sure what's the best way to fix this:
  - Replace RISCV by ARCH_SUNXI?
    This would expose it on more ARM sun<foo>i platforms, making the
    MACH_SUN4I || MACH_SUN7I superfluous?
  - Replace RISCV by RISCV && ARCH_SUNXI?

Thanks for your comments!

>         help
>           Say Y here if you want to use CAN controller found on Allwinner
> -         A10/A20 SoCs.
> +         A10/A20/D1 SoCs.
>
>           To compile this driver as a module, choose M here: the module will
>           be called sun4i_can.

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

On Tue, Aug 22, 2023 at 02:30:16PM +0200, Geert Uytterhoeven wrote:
> Hi John,
> 
> This makes this question pop up when configuring a kernel for any RISC-V
> platform, not just for Allwinner RISC-V platforms.

Oh dear.

> In comparison, drivers/clk/sunxi-ng/Kconfig does have some
> 
>     depends on MACH_SUN<foo>I || RISCV || COMPILE_TEST
> 
> but these are gated by ARCH_SUNXI at the top of the file.

Ah, that is what I copied.

> I'm not sure what's the best way to fix this:
>   - Replace RISCV by ARCH_SUNXI?
>     This would expose it on more ARM sun<foo>i platforms, making the
>     MACH_SUN4I || MACH_SUN7I superfluous?
>   - Replace RISCV by RISCV && ARCH_SUNXI?

I'm not sure what the best approach here is. Just having it require ARCH_SUNXI
would make sense to me but I'm not too sure why where's so many different MACH
here in the first place.

> Thanks for your comments!
> 
> >         help
> >           Say Y here if you want to use CAN controller found on Allwinner
> > -         A10/A20 SoCs.
> > +         A10/A20/D1 SoCs.
> >
> >           To compile this driver as a module, choose M here: the module will
> >           be called sun4i_can.
> 
> Gr{oetje,eeting}s,
> 
>                         Geert

John.