From: Krzysztof Kozlowski <krzk@kernel.org>
To: Dhanunjanrao Katta <katta.dhanunjanrao@amd.com>,
Mark Brown <broonie@kernel.org>,
Liam Girdwood <lgirdwood@gmail.com>,
linux-sound@vger.kernel.org
Cc: Vincenzo Frascino <vincenzo.frascino@arm.com>,
Rob Herring <robh@kernel.org>,
Krzysztof Kozlowski <krzk+dt@kernel.org>,
Conor Dooley <conor+dt@kernel.org>,
Jaroslav Kysela <perex@perex.cz>, Takashi Iwai <tiwai@suse.com>,
Michal Simek <michal.simek@amd.com>,
devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-arm-kernel@lists.infradead.org, git@amd.com
Subject: Re: [PATCH 1/3] ASoC: dt-bindings: xlnx,spdif: describe AXI4-Stream and audio clocks
Date: Tue, 17 Nov 2026 11:06:19 +0100 [thread overview]
Message-ID: <5be284b8-f321-4a9f-be59-7c85f5805fe9@kernel.org> (raw)
In-Reply-To: <20261117085739.3807943-2-katta.dhanunjanrao@amd.com>
On 17/11/2026 09:57, Dhanunjanrao Katta wrote:
> The SPDIF IP is fed by three clocks: the audio clock used at the
> SPDIF/AES3 interface, the AXI4-Lite control clock and the AXI4-Stream
> data clock. So far the binding described only two of them and carried
> the audio clock frequency in the custom "xlnx,aud_clk_i" property.
>
> Describe the AXI4-Stream clock and pick its name from the IP mode,
> s_axis_aclk for the transmitter and m_axis_aclk for the receiver. The
> minimum clock count stays at two, so that device trees written against
> the earlier binding keep validating and the third clock is effectively
> optional.
>
> Mark "xlnx,aud_clk_i" deprecated instead of removing it. The frequency
> it carries is available from the rate of the audio clock, but the
> property is part of the published binding and device trees in the field
> still pass it. As the binding sets "additionalProperties: false",
> removing the property outright would make those device trees fail
> validation.
>
> Make "xlnx,spdif-mode" required. The conditional that selects the
That's ABI break.
> AXI4-Stream clock name keys on that property, and a conditional with no
> "required" entry of its own also matches when the property is absent.
> A device tree omitting the mode would therefore have been held to the
> transmitter clock name by default. Requiring the property turns the
> omission into a clear validation failure instead.
You need to explain why, not what. Entire commit msg looks LLM written,
honestly difficult to parse. LLM is not supposed to be used to write
commit msg, because it means you did not understand what you did here.
>
> Signed-off-by: Dhanunjanrao Katta <katta.dhanunjanrao@amd.com>
> ---
> .../devicetree/bindings/sound/xlnx,spdif.yaml | 54 +++++++++++++++----
> 1 file changed, 44 insertions(+), 10 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/sound/xlnx,spdif.yaml b/Documentation/devicetree/bindings/sound/xlnx,spdif.yaml
> index a45d8a0755fe..a4abca2027d3 100644
> --- a/Documentation/devicetree/bindings/sound/xlnx,spdif.yaml
> +++ b/Documentation/devicetree/bindings/sound/xlnx,spdif.yaml
> @@ -14,6 +14,28 @@ maintainers:
>
> allOf:
> - $ref: dai-common.yaml#
> + - if:
Move allOf to the bottom, before additional/unevaluatedProps
> + properties:
> + xlnx,spdif-mode:
> + const: 1
> + required:
> + - xlnx,spdif-mode
> + then:
> + properties:
> + clock-names:
> + minItems: 2
> + items:
> + - const: aud_clk_i
> + - const: s_axi_aclk
> + - const: s_axis_aclk
> + else:
> + properties:
> + clock-names:
> + minItems: 2
> + items:
> + - const: aud_clk_i
> + - const: s_axi_aclk
> + - const: m_axis_aclk
That's the same clock. Just call it "axis"
>
> properties:
> compatible:
> @@ -28,15 +50,15 @@ properties:
> - description: SPDIF audio interrupt
>
> clock-names:
> - items:
> - - const: aud_clk_i
> - - const: s_axi_aclk
> + minItems: 2
> + maxItems: 3
List stays here with minItems
>
> clocks:
> - minItems: 1
> + minItems: 2
> items:
> - - description: input audio clock
> - - description: clock for the AXI data stream
> + - description: Audio clock input used at the SPDIF/AES3 interface
> + - description: AXI4-Lite clock
> + - description: AXI4-Stream clock
>
> xlnx,spdif-mode:
> $ref: /schemas/types.yaml#/definitions/uint32
> @@ -49,8 +71,10 @@ properties:
>
> xlnx,aud_clk_i:
> $ref: /schemas/types.yaml#/definitions/uint32
> + deprecated: true
> description:
> - Input audio clock frequency. It affects the sampling rate.
> + Frequency of the audio clock input. Superseded by the rate of the
> + clock named aud_clk_i.
>
> required:
> - compatible
> @@ -58,6 +82,7 @@ required:
> - interrupts
> - clock-names
> - clocks
> + - xlnx,spdif-mode
ABI break
Best regards,
Krzysztof
next prev parent reply other threads:[~2026-11-17 10:06 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-17 8:57 [PATCH 0/3] ASoC: xlnx: spdif: switch to the common clock framework Dhanunjanrao Katta
2026-11-17 8:57 ` [PATCH 1/3] ASoC: dt-bindings: xlnx,spdif: describe AXI4-Stream and audio clocks Dhanunjanrao Katta
2026-11-17 10:06 ` Krzysztof Kozlowski [this message]
2026-11-17 8:57 ` [PATCH 2/3] ASoC: xlnx: spdif: namespace register macros and use a divider enum Dhanunjanrao Katta
2026-11-17 8:57 ` [PATCH 3/3] ASoC: xlnx: spdif: get clocks through the common clock framework Dhanunjanrao Katta
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5be284b8-f321-4a9f-be59-7c85f5805fe9@kernel.org \
--to=krzk@kernel.org \
--cc=broonie@kernel.org \
--cc=conor+dt@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=git@amd.com \
--cc=katta.dhanunjanrao@amd.com \
--cc=krzk+dt@kernel.org \
--cc=lgirdwood@gmail.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-sound@vger.kernel.org \
--cc=michal.simek@amd.com \
--cc=perex@perex.cz \
--cc=robh@kernel.org \
--cc=tiwai@suse.com \
--cc=vincenzo.frascino@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®