[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230927-uncorrupt-bronco-72e35232b1cc@spud>
Date: Wed, 27 Sep 2023 17:01:03 +0100
From: Conor Dooley <conor@...nel.org>
To: Chen Wang <unicornxw@...il.com>
Cc: aou@...s.berkeley.edu, chao.wei@...hgo.com,
devicetree@...r.kernel.org, guoren@...nel.org, jszhang@...nel.org,
krzysztof.kozlowski+dt@...aro.org, linux-kernel@...r.kernel.org,
linux-riscv@...ts.infradead.org, palmer@...belt.com,
paul.walmsley@...ive.com, robh+dt@...nel.org,
xiaoguang.xing@...hgo.com, apatel@...tanamicro.com,
Inochi Amaoto <inochiama@...look.com>,
Chen Wang <wangchen20@...as.ac.cn>
Subject: Re: [PATCH v3 06/11] dt-bindings: timer: Add Sophgo sg2042 CLINT
timer
Hey,
On Wed, Sep 27, 2023 at 05:01:37PM +0800, Chen Wang wrote:
> From: Inochi Amaoto <inochiama@...look.com>
>
> The clint of Sophgo sg2042 is incompatible with the standard sifive
> clint, as the timer and ipi device on the different address, and can
> not be handled by the sifive,clint DT.
>
> In addition, the timers of sg2042 are mapped by per cluster, which is
> hard to merge with its ipi device.
I think the description here is kinda poor, it needs to be explained
that this is an implementation of the not frozen & likely abandoned
aclint spec.
> To avoid conficts caused by using the same clint compatible string when
> this device is parsed by SBI, add a new vendor specific compatible string
> to identify the timer of sg2042 soc.
And this whole section about avoiding conflicts is not relevant, since
the binding is specifically for the timer. It'd be better to mention why
a single compatible cannot work for all elements, than bring up a
situation that does not exist and would be a misuse of the binding in
the first place.
> Signed-off-by: Inochi Amaoto <inochiama@...look.com>
> Signed-off-by: Chen Wang <wangchen20@...as.ac.cn>
> Signed-off-by: Chen Wang <unicornxw@...il.com>
You only need to sign this off once. The iscas one looks like it
probably is the relevant signoff?
> ---
> .../timer/sophgo,sg2042-clint-mtimer.yaml | 42 +++++++++++++++++++
> 1 file changed, 42 insertions(+)
> create mode 100644 Documentation/devicetree/bindings/timer/sophgo,sg2042-clint-mtimer.yaml
>
> diff --git a/Documentation/devicetree/bindings/timer/sophgo,sg2042-clint-mtimer.yaml b/Documentation/devicetree/bindings/timer/sophgo,sg2042-clint-mtimer.yaml
> new file mode 100644
> index 000000000000..5da0947d048a
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/timer/sophgo,sg2042-clint-mtimer.yaml
> @@ -0,0 +1,42 @@
> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/timer/sophgo,sg2042-clint-mtimer.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Sophgo CLINT Timer
> +
> +maintainers:
> + - Inochi Amaoto <inochiama@...look.com>
> +
> +properties:
> + compatible:
> + oneOf:
> + - items:
> + - const: sophgo,sg2042-clint-mtimer
There's only one of these, so you don't need the oneOf.
Also, is the clint here not a thead IP? In which case, you need to add a
second compatible IMO. That second compatible then would be the one that
appears in opensbi etc.
Otherwise, this looks fine.
Thanks,
Conor.
> +
> + reg:
> + maxItems: 1
> +
> + interrupts-extended:
> + minItems: 1
> + maxItems: 4095
> +
> +additionalProperties: false
> +
> +required:
> + - compatible
> + - reg
> + - interrupts-extended
> +
> +examples:
> + - |
> + timer@...00000 {
> + compatible = "sophgo,sg2042-clint-mtimer";
> + interrupts-extended = <&cpu1intc 7>,
> + <&cpu2intc 7>,
> + <&cpu3intc 7>,
> + <&cpu4intc 7>;
> + reg = <0xac000000 0x00010000>;
> + };
> +...
> --
> 2.25.1
>
Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)
Powered by blists - more mailing lists