[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <78e60d723c27b7fa0f03bc6a74f6ad37d6508734.camel@pengutronix.de>
Date: Tue, 25 Feb 2025 14:12:03 +0100
From: Philipp Zabel <p.zabel@...gutronix.de>
To: Daniel Baluta <daniel.baluta@....com>, robh@...nel.org,
krzk+dt@...nel.org, shawnguo@...nel.org, mathieu.poirier@...aro.org
Cc: conor+dt@...nel.org, s.hauer@...gutronix.de, kernel@...gutronix.de,
festevam@...il.com, linux-kernel@...r.kernel.org,
devicetree@...r.kernel.org, imx@...ts.linux.dev,
linux-arm-kernel@...ts.infradead.org, linux-remoteproc@...r.kernel.org,
andersson@...nel.org, Frank.Li@....com, peng.fan@....com,
laurentiu.mihalcea@....com, iuliana.prodan@....com
Subject: Re: [PATCH v3 2/8] dt-bindings: dsp: fsl,dsp: Add resets property
On Di, 2025-02-25 at 12:19 +0200, Daniel Baluta wrote:
> Assert and deassert functionality of the DSP found on i.MX8MP is
> realized by combining control bits from two modules: Audio Block
> Control and Debug Access Port.
>
> Audio block control bits are used to for Run/Stall the DSP core
> while the DAP bits are used for software reset the core.
>
> The original plan was to use fsl,dsp-ctrl property and to refer the
> audiomix bits via syscon interface. This proposal received NACK from
> community we shouldn't abuse the syscon interface [1].
>
> So remove fsl,dsp-ctrl property for i.MX8MP and use reset control
> interface instead.
>
> [1] https://patchwork.kernel.org/project/imx/patch/20250212085222.107102-6-daniel.baluta@nxp.com/
>
> Signed-off-by: Daniel Baluta <daniel.baluta@....com>
> Reviewed-by: Frank Li <Frank.Li@....com>
> ---
> .../devicetree/bindings/dsp/fsl,dsp.yaml | 16 +++++++++++++++-
> 1 file changed, 15 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/devicetree/bindings/dsp/fsl,dsp.yaml b/Documentation/devicetree/bindings/dsp/fsl,dsp.yaml
> index ab93ffd3d2e5..b3550c9d12e7 100644
> --- a/Documentation/devicetree/bindings/dsp/fsl,dsp.yaml
> +++ b/Documentation/devicetree/bindings/dsp/fsl,dsp.yaml
> @@ -82,6 +82,9 @@ properties:
> description:
> Phandle to syscon block which provide access for processor enablement
>
> + resets:
> + maxItems: 1
The DAP core reset is mentioned in the commit message. Why is it
missing here? After reading the discussion in [1], I'd expect both the
stall and the (core) reset signal to be documented, something like:
resets:
maxItems: 2
reset-names:
items:
- const: core
- const: stall
> +
> required:
> - compatible
> - reg
> @@ -164,6 +167,16 @@ allOf:
> - const: txdb1
> - const: rxdb0
> - const: rxdb1
> + - if:
> + properties:
> + compatible:
> + contains:
> + enum:
> + - fsl,imx8mp-dsp
> + - fsl,imx8mp-hifi4
> + then:
> + required:
> + - resets
>
> additionalProperties: false
>
> @@ -186,6 +199,7 @@ examples:
> };
> - |
> #include <dt-bindings/clock/imx8mp-clock.h>
> + #include <dt-bindings/reset/imx8mp-reset-audiomix.h>
> dsp_reserved: dsp@...00000 {
> reg = <0x92400000 0x1000000>;
> no-map;
> @@ -220,5 +234,5 @@ examples:
> <&mu2 3 0>;
> memory-region = <&dsp_vdev0buffer>, <&dsp_vdev0vring0>,
> <&dsp_vdev0vring1>, <&dsp_reserved>;
> - fsl,dsp-ctrl = <&audio_blk_ctrl>;
Is there nothing else in this range that will have to be controlled by
the DSP driver in the future, such as the IMPWIRE register or the
XOCDMODE[OCDHALTONRESET] bit?
regards
Philipp
Powered by blists - more mailing lists