[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <aIHj27uJG18afnLc@linaro.org>
Date: Thu, 24 Jul 2025 09:42:19 +0200
From: Stephan Gerhold <stephan.gerhold@...aro.org>
To: srinivas.kandagatla@....qualcomm.com
Cc: andersson@...nel.org, konradybcio@...nel.org, robh@...nel.org,
krzk+dt@...nel.org, conor+dt@...nel.org,
cros-qcom-dts-watchers@...omium.org, linux-arm-msm@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 15/23] arm64: dts: qcom: msm8916: use dedicated
elite-audio dtsi
On Wed, Jul 23, 2025 at 11:27:29PM +0100, srinivas.kandagatla@....qualcomm.com wrote:
> From: Srinivas Kandagatla <srinivas.kandagatla@....qualcomm.com>
>
> Make use of elite-audio.dtsi and remove the gpr nodes from SoC dtsi.
> This move removes duplication.
>
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@....qualcomm.com>
> ---
> arch/arm64/boot/dts/qcom/msm8916-audio.dtsi | 5 +++
> arch/arm64/boot/dts/qcom/msm8916-pm8916.dtsi | 1 +
> arch/arm64/boot/dts/qcom/msm8916.dtsi | 44 --------------------
> 3 files changed, 6 insertions(+), 44 deletions(-)
> create mode 100644 arch/arm64/boot/dts/qcom/msm8916-audio.dtsi
>
> diff --git a/arch/arm64/boot/dts/qcom/msm8916-audio.dtsi b/arch/arm64/boot/dts/qcom/msm8916-audio.dtsi
> new file mode 100644
> index 000000000000..fe17bc28e825
> --- /dev/null
> +++ b/arch/arm64/boot/dts/qcom/msm8916-audio.dtsi
> @@ -0,0 +1,5 @@
> +// SPDX-License-Identifier: BSD-3-Clause
> +/*
> + * Copyright (c) 2025 Qualcomm Innovation Center, Inc. All rights reserved.
> + */
> +#include "elite-audio.dtsi"
I don't think this will work. elite-audio.dtsi contains
qcom,protection-domain = "avs/audio", "msm/adsp/audio_pd";
for all the APR service nodes, but the ADSP firmware on MSM8916 doesn't
have support for this as far as I know. This will likely prevent all the
drivers from probing.
The definition below doesn't use "qcom,protection-domain".
> diff --git a/arch/arm64/boot/dts/qcom/msm8916-pm8916.dtsi b/arch/arm64/boot/dts/qcom/msm8916-pm8916.dtsi
> index b1a7eafbee31..c0a48692c100 100644
> --- a/arch/arm64/boot/dts/qcom/msm8916-pm8916.dtsi
> +++ b/arch/arm64/boot/dts/qcom/msm8916-pm8916.dtsi
> @@ -9,6 +9,7 @@
> */
>
> #include "msm8916.dtsi"
> +#include "msm8916-audio.dtsi"
> #include "pm8916.dtsi"
>
> &camss {
> diff --git a/arch/arm64/boot/dts/qcom/msm8916.dtsi b/arch/arm64/boot/dts/qcom/msm8916.dtsi
> index b50c7e6e0bfc..6d9807adc618 100644
> --- a/arch/arm64/boot/dts/qcom/msm8916.dtsi
> +++ b/arch/arm64/boot/dts/qcom/msm8916.dtsi
> @@ -1997,51 +1997,7 @@ smd-edge {
> label = "hexagon";
>
> apr: apr {
> - compatible = "qcom,apr-v2";
> qcom,smd-channels = "apr_audio_svc";
> - qcom,domain = <APR_DOMAIN_ADSP>;
> - #address-cells = <1>;
> - #size-cells = <0>;
> - status = "disabled";
status = "disabled" got lost in the conversion. This is needed on DB410c
to prevent crashes when probing the broken APR services in the GPS-only
ADSP firmware.
Instead of disabling APR by default, you could move the status =
"disabled" to apq8016-sbc. I think that would also be fine.
Thanks,
Stephan
Powered by blists - more mailing lists