diff mbox series

[4/6] arm64: dts: qcom: sm8250: move sound node out of soc

Message ID 20220328143035.519909-5-vkoul@kernel.org (mailing list archive)
State Superseded
Headers show
Series arm64: dts: qcom: Get rid of some warnings | expand

Commit Message

Vinod Koul March 28, 2022, 2:30 p.m. UTC
The soc node expects all the nodes to have unit addresses. The sound
node does not have that which causes warnings:

arch/arm64/boot/dts/qcom/sm8250.dtsi:2806.16-2807.5:
Warning (simple_bus_reg): /soc@0/sound: missing or empty reg/ranges property

Move sound node out of soc to fix this

Signed-off-by: Vinod Koul <vkoul@kernel.org>
---
 arch/arm64/boot/dts/qcom/sm8250.dtsi | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

Comments

Krzysztof Kozlowski March 28, 2022, 3:14 p.m. UTC | #1
On 28/03/2022 16:30, Vinod Koul wrote:
> The soc node expects all the nodes to have unit addresses. The sound
> node does not have that which causes warnings:
> 
> arch/arm64/boot/dts/qcom/sm8250.dtsi:2806.16-2807.5:
> Warning (simple_bus_reg): /soc@0/sound: missing or empty reg/ranges property
> 
> Move sound node out of soc to fix this
> 
> Signed-off-by: Vinod Koul <vkoul@kernel.org>
> ---
>  arch/arm64/boot/dts/qcom/sm8250.dtsi | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 

I don't know the SM8250, but usually the sound node (e.g. containing
audio-codec) is not part of SoC. I propose to remove it entirely from
DTSI and define in same place in each DTS. It makes more sense logically
in such case - one clearly see which board defines the sounds, which
does not.


Best regards,
Krzysztof
Vinod Koul March 28, 2022, 5:14 p.m. UTC | #2
On 28-03-22, 17:14, Krzysztof Kozlowski wrote:
> On 28/03/2022 16:30, Vinod Koul wrote:
> > The soc node expects all the nodes to have unit addresses. The sound
> > node does not have that which causes warnings:
> > 
> > arch/arm64/boot/dts/qcom/sm8250.dtsi:2806.16-2807.5:
> > Warning (simple_bus_reg): /soc@0/sound: missing or empty reg/ranges property
> > 
> > Move sound node out of soc to fix this
> > 
> > Signed-off-by: Vinod Koul <vkoul@kernel.org>
> > ---
> >  arch/arm64/boot/dts/qcom/sm8250.dtsi | 6 +++---
> >  1 file changed, 3 insertions(+), 3 deletions(-)
> > 
> 
> I don't know the SM8250, but usually the sound node (e.g. containing
> audio-codec) is not part of SoC. I propose to remove it entirely from
> DTSI and define in same place in each DTS. It makes more sense logically
> in such case - one clearly see which board defines the sounds, which
> does not.

Most of our boards have sound, should we duplicate it in all the
boards..? Bjorn..?
Krzysztof Kozlowski March 28, 2022, 5:16 p.m. UTC | #3
On 28/03/2022 19:14, Vinod Koul wrote:
> On 28-03-22, 17:14, Krzysztof Kozlowski wrote:
>> On 28/03/2022 16:30, Vinod Koul wrote:
>>> The soc node expects all the nodes to have unit addresses. The sound
>>> node does not have that which causes warnings:
>>>
>>> arch/arm64/boot/dts/qcom/sm8250.dtsi:2806.16-2807.5:
>>> Warning (simple_bus_reg): /soc@0/sound: missing or empty reg/ranges property
>>>
>>> Move sound node out of soc to fix this
>>>
>>> Signed-off-by: Vinod Koul <vkoul@kernel.org>
>>> ---
>>>  arch/arm64/boot/dts/qcom/sm8250.dtsi | 6 +++---
>>>  1 file changed, 3 insertions(+), 3 deletions(-)
>>>
>>
>> I don't know the SM8250, but usually the sound node (e.g. containing
>> audio-codec) is not part of SoC. I propose to remove it entirely from
>> DTSI and define in same place in each DTS. It makes more sense logically
>> in such case - one clearly see which board defines the sounds, which
>> does not.
> 
> Most of our boards have sound, should we duplicate it in all the
> boards..? Bjorn..?

It's anyway empty here, so not much of saving lines f code...


Best regards,
Krzysztof
Bjorn Andersson April 9, 2022, 12:25 a.m. UTC | #4
On Mon 28 Mar 12:14 CDT 2022, Vinod Koul wrote:

> On 28-03-22, 17:14, Krzysztof Kozlowski wrote:
> > On 28/03/2022 16:30, Vinod Koul wrote:
> > > The soc node expects all the nodes to have unit addresses. The sound
> > > node does not have that which causes warnings:
> > > 
> > > arch/arm64/boot/dts/qcom/sm8250.dtsi:2806.16-2807.5:
> > > Warning (simple_bus_reg): /soc@0/sound: missing or empty reg/ranges property
> > > 
> > > Move sound node out of soc to fix this
> > > 
> > > Signed-off-by: Vinod Koul <vkoul@kernel.org>
> > > ---
> > >  arch/arm64/boot/dts/qcom/sm8250.dtsi | 6 +++---
> > >  1 file changed, 3 insertions(+), 3 deletions(-)
> > > 
> > 
> > I don't know the SM8250, but usually the sound node (e.g. containing
> > audio-codec) is not part of SoC. I propose to remove it entirely from
> > DTSI and define in same place in each DTS. It makes more sense logically
> > in such case - one clearly see which board defines the sounds, which
> > does not.
> 
> Most of our boards have sound, should we duplicate it in all the
> boards..? Bjorn..?
> 

But is the sndcard platform or device specific? E.g. all the SDM845
boards seems to use either qcom,sdm845-sndcard or qcom,db845c-sndcard.

Are there room for some common properties in this node?


Otherwise it seems reasonable to skip it in the platform dtsi.

Regards,
Bjorn
diff mbox series

Patch

diff --git a/arch/arm64/boot/dts/qcom/sm8250.dtsi b/arch/arm64/boot/dts/qcom/sm8250.dtsi
index b120babdb26c..fb94b342f2d1 100644
--- a/arch/arm64/boot/dts/qcom/sm8250.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8250.dtsi
@@ -2803,9 +2803,6 @@  compute-cb@8 {
 			};
 		};
 
-		sound: sound {
-		};
-
 		usb_1_hsphy: phy@88e3000 {
 			compatible = "qcom,sm8250-usb-hs-phy",
 				     "qcom,usb-snps-hs-7nm-phy";
@@ -4684,6 +4681,9 @@  cpufreq_hw: cpufreq@18591000 {
 		};
 	};
 
+	sound: sound {
+	};
+
 	timer {
 		compatible = "arm,armv8-timer";
 		interrupts = <GIC_PPI 13