mbox series

[0/6] media: dt-bindings: media: camss: Fix interrupt types

Message ID 20240905164142.3475873-1-vladimir.zapolskiy@linaro.org
Headers show
Series media: dt-bindings: media: camss: Fix interrupt types | expand

Message

Vladimir Zapolskiy Sept. 5, 2024, 4:41 p.m. UTC 
It was discovered that on a few Qualcomm platforms types of interrupts
do not match both downstream code and a type requested by the CAMSS driver.

The mismatched interrupt type between firmware and the correspondent
CAMSS driver leads to known problems, which were discussed previously:

  https://lore.kernel.org/lkml/20220530080842.37024-4-manivannan.sadhasivam@linaro.org/

Here the situation is right the same, namely a repeated bind of camss
device is not possible due to a wrongly specified interrupt type, and
it may lead to an issue in runtime manifested like this:

  irq: type mismatch, failed to map hwirq-509 for interrupt-controller@17a00000!

Vladimir Zapolskiy (6):
  media: dt-bindings: media: qcom,sc8280xp-camss: Fix interrupt types
  media: dt-bindings: media: qcom,sdm845-camss: Fix interrupt types
  media: dt-bindings: media: qcom,sm8250-camss: Fix interrupt types
  arm64: dts: qcom: sc8280xp: Fix interrupt type of camss interrupts
  arm64: dts: qcom: sdm845: Fix interrupt types of camss interrupts
  arm64: dts: qcom: sm8250: Fix interrupt types of camss interrupts

 .../bindings/media/qcom,sc8280xp-camss.yaml   | 40 +++++++++----------
 .../bindings/media/qcom,sdm845-camss.yaml     | 20 +++++-----
 .../bindings/media/qcom,sm8250-camss.yaml     | 28 ++++++-------
 arch/arm64/boot/dts/qcom/sc8280xp.dtsi        | 40 +++++++++----------
 arch/arm64/boot/dts/qcom/sdm845.dtsi          | 20 +++++-----
 arch/arm64/boot/dts/qcom/sm8250.dtsi          | 28 ++++++-------
 6 files changed, 88 insertions(+), 88 deletions(-)

Comments

Krzysztof Kozlowski Sept. 6, 2024, 10:31 a.m. UTC | #1 
On Thu, Sep 05, 2024 at 07:41:37PM +0300, Vladimir Zapolskiy wrote:
> The expected type of all CAMSS interrupts is edge rising, fix it in
> the documented example from CAMSS device tree bindings for sc8280xp.

Subject: drop duplicated media. One media is enough (the first).

Best regards,
Krzysztof
Vladimir Zapolskiy Sept. 6, 2024, 11:06 a.m. UTC | #2 
Hi Krzysztof,

On 9/6/24 13:31, Krzysztof Kozlowski wrote:
> On Thu, Sep 05, 2024 at 07:41:37PM +0300, Vladimir Zapolskiy wrote:
>> The expected type of all CAMSS interrupts is edge rising, fix it in
>> the documented example from CAMSS device tree bindings for sc8280xp.
> 
> Subject: drop duplicated media. One media is enough (the first).
> 

thank you for review, here I've attemted to follow the established practice.

% git log -n 500 --oneline Documentation/devicetree/bindings/media/ | grep "media:.*media:" | wc -l
166

So 1/3 of relevant commits to the folder have the duplicated "media:"
in the subject, the reason is not totally clear to me, thus I've just
inflowed into the same.

--
Best wishes,
Vladimir
Krzysztof Kozlowski Sept. 6, 2024, 12:43 p.m. UTC | #3 
On 06/09/2024 13:06, Vladimir Zapolskiy wrote:
> Hi Krzysztof,
> 
> On 9/6/24 13:31, Krzysztof Kozlowski wrote:
>> On Thu, Sep 05, 2024 at 07:41:37PM +0300, Vladimir Zapolskiy wrote:
>>> The expected type of all CAMSS interrupts is edge rising, fix it in
>>> the documented example from CAMSS device tree bindings for sc8280xp.
>>
>> Subject: drop duplicated media. One media is enough (the first).
>>
> 
> thank you for review, here I've attemted to follow the established practice.
> 
> % git log -n 500 --oneline Documentation/devicetree/bindings/media/ | grep "media:.*media:" | wc -l
> 166
> 
> So 1/3 of relevant commits to the folder have the duplicated "media:"
> in the subject, the reason is not totally clear to me, thus I've just
> inflowed into the same.

People like to repeat same mistakes, but just because 166 times someone
did a mistake, is not a reason to do the same.

Best regards,
Krzysztof