[PATCH 0/3] Fix tcsr_mutex register for IPQ6018

[PATCH 0/3] Fix tcsr_mutex register for IPQ6018

[Vignesh Viswanathan]

IPQ6018 has 32 tcsr_mutext hwlock registers of 0x1000 size each.
The compatible string qcom,ipq6018-tcsr-mutex is mapped to
of_msm8226_tcsr_mutex which has 32 locks configured with stride of 0x80
and doesn't match the HW present in IPQ6018.

This series fixes the following:
1. Fix the tcsr_mutex hwlock register size to 0x20000 in ipq6018.dtsi.
2. Remove qcom,ipq6018-tcsr-mutex compatible string for tcsr_mutex in
ipq6018.dtsi.
3. Drop unused qcom,ipq6018-tcsr-mutex compatible string from
qcom_hwspinlock driver and dt-bindings.

Vignesh Viswanathan (3):
  arm64: dts: qcom: ipq6018: Fix tcsr_mutex register size
  hwspinlock: qcom: Drop unused qcom,ipq6018-tcsr-mutex
  dt-bindings: hwlock: Drop unused qcom,ipq6018-tcsr-mutex

 Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml | 1 -
 arch/arm64/boot/dts/qcom/ipq6018.dtsi                         | 4 ++--
 drivers/hwspinlock/qcom_hwspinlock.c                          | 1 -
 3 files changed, 2 insertions(+), 4 deletions(-)

-- 
2.41.0

[PATCH 1/3] arm64: dts: qcom: ipq6018: Fix tcsr_mutex register size

[Vignesh Viswanathan]

IPQ6018 has 32 tcsr_mutex hwlock registers and size of each is 0x1000.

Fix size of the tcsr_mutex hwlock register to 0x20000.
Also, remove the qcom,ipq6018-tcsr-mutex compatible as this maps to
incorrect stride and mutex config for IPQ6018 in hwspinlock driver.

Signed-off-by: Vignesh Viswanathan <quic_viswanat@quicinc.com>
---
 arch/arm64/boot/dts/qcom/ipq6018.dtsi | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/boot/dts/qcom/ipq6018.dtsi b/arch/arm64/boot/dts/qcom/ipq6018.dtsi
index 47b8b1d6730a..56537cf8243f 100644
--- a/arch/arm64/boot/dts/qcom/ipq6018.dtsi
+++ b/arch/arm64/boot/dts/qcom/ipq6018.dtsi
@@ -392,8 +392,8 @@ gcc: gcc@1800000 {
 		};
 
 		tcsr_mutex: hwlock@1905000 {
-			compatible = "qcom,ipq6018-tcsr-mutex", "qcom,tcsr-mutex";
-			reg = <0x0 0x01905000 0x0 0x1000>;
+			compatible = "qcom,tcsr-mutex";
+			reg = <0x0 0x01905000 0x0 0x20000>;
 			#hwlock-cells = <1>;
 		};
 
-- 
2.41.0

[PATCH 2/3] hwspinlock: qcom: Drop unused qcom,ipq6018-tcsr-mutex

[Vignesh Viswanathan]

qcom,ipq6018-tcsr-mutex maps to incorrect config of IPQ6018 and is
dropped from the devictree. IPQ6018 will use qcom,tcsr-mutex compatible
string.

Drop qcom,ipq6018-tcsr-mutex compatible string from
qcom_hwspinlock_of_match table.

Signed-off-by: Vignesh Viswanathan <quic_viswanat@quicinc.com>
---
 drivers/hwspinlock/qcom_hwspinlock.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/hwspinlock/qcom_hwspinlock.c b/drivers/hwspinlock/qcom_hwspinlock.c
index a0fd67fd2934..814dfe8697bf 100644
--- a/drivers/hwspinlock/qcom_hwspinlock.c
+++ b/drivers/hwspinlock/qcom_hwspinlock.c
@@ -115,7 +115,6 @@ static const struct of_device_id qcom_hwspinlock_of_match[] = {
 	{ .compatible = "qcom,sfpb-mutex", .data = &of_sfpb_mutex },
 	{ .compatible = "qcom,tcsr-mutex", .data = &of_tcsr_mutex },
 	{ .compatible = "qcom,apq8084-tcsr-mutex", .data = &of_msm8226_tcsr_mutex },
-	{ .compatible = "qcom,ipq6018-tcsr-mutex", .data = &of_msm8226_tcsr_mutex },
 	{ .compatible = "qcom,msm8226-tcsr-mutex", .data = &of_msm8226_tcsr_mutex },
 	{ .compatible = "qcom,msm8974-tcsr-mutex", .data = &of_msm8226_tcsr_mutex },
 	{ .compatible = "qcom,msm8994-tcsr-mutex", .data = &of_msm8226_tcsr_mutex },
-- 
2.41.0

[PATCH 3/3] dt-bindings: hwlock: Drop unused qcom,ipq6018-tcsr-mutex

[Vignesh Viswanathan]

Drop qcom,ipq6018-tcsr-mutex enum as this is does not map to the correct
config for IPQ6018. IPQ6018 will use qcom,tcsr-mutex compatible string.

Signed-off-by: Vignesh Viswanathan <quic_viswanat@quicinc.com>
---
 Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml | 1 -
 1 file changed, 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml
index ee2726149cf3..a595396fd3f9 100644
--- a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml
+++ b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml
@@ -22,7 +22,6 @@ properties:
       - items:
           - enum:
               - qcom,apq8084-tcsr-mutex
-              - qcom,ipq6018-tcsr-mutex
               - qcom,msm8226-tcsr-mutex
               - qcom,msm8994-tcsr-mutex
           - const: qcom,tcsr-mutex
-- 
2.41.0

Re: [PATCH 1/3] arm64: dts: qcom: ipq6018: Fix tcsr_mutex register size

[Krzysztof Kozlowski]

On 04/09/2023 07:50, Vignesh Viswanathan wrote:
> IPQ6018 has 32 tcsr_mutex hwlock registers and size of each is 0x1000.
> 
> Fix size of the tcsr_mutex hwlock register to 0x20000.
> Also, remove the qcom,ipq6018-tcsr-mutex compatible as this maps to
> incorrect stride and mutex config for IPQ6018 in hwspinlock driver.

This is 6018, so if 6018 compatible is incorrect... this does not make
any sense.

NAK


Best regards,
Krzysztof

Re: [PATCH 2/3] hwspinlock: qcom: Drop unused qcom,ipq6018-tcsr-mutex

[Krzysztof Kozlowski]

On 04/09/2023 07:50, Vignesh Viswanathan wrote:
> qcom,ipq6018-tcsr-mutex maps to incorrect config of IPQ6018 and is
> dropped from the devictree. 

No, it is not dropped.


> IPQ6018 will use qcom,tcsr-mutex compatible
> string.

No, it will not.

> 
> Drop qcom,ipq6018-tcsr-mutex compatible string from
> qcom_hwspinlock_of_match table.

Why? Do not write what you are doing here, but why you are doing it.


Best regards,
Krzysztof

Re: [PATCH 3/3] dt-bindings: hwlock: Drop unused qcom,ipq6018-tcsr-mutex

[Krzysztof Kozlowski]

On 04/09/2023 07:50, Vignesh Viswanathan wrote:
> Drop qcom,ipq6018-tcsr-mutex enum as this is does not map to the correct
> config for IPQ6018. IPQ6018 will use qcom,tcsr-mutex compatible string.
> 
> Signed-off-by: Vignesh Viswanathan <quic_viswanat@quicinc.com>
> ---
>  Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml
> index ee2726149cf3..a595396fd3f9 100644
> --- a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml
> +++ b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml
> @@ -22,7 +22,6 @@ properties:
>        - items:
>            - enum:
>                - qcom,apq8084-tcsr-mutex
> -              - qcom,ipq6018-tcsr-mutex

NAK, this is a correct compatible for IPQ6018.

Best regards,
Krzysztof

Re: [PATCH 1/3] arm64: dts: qcom: ipq6018: Fix tcsr_mutex register size

[Vignesh Viswanathan]

On 9/4/2023 12:10 PM, Krzysztof Kozlowski wrote:
> On 04/09/2023 07:50, Vignesh Viswanathan wrote:
>> IPQ6018 has 32 tcsr_mutex hwlock registers and size of each is 0x1000.
>>
>> Fix size of the tcsr_mutex hwlock register to 0x20000.
>> Also, remove the qcom,ipq6018-tcsr-mutex compatible as this maps to
>> incorrect stride and mutex config for IPQ6018 in hwspinlock driver.
> 
> This is 6018, so if 6018 compatible is incorrect... this does not make
> any sense.
> 
> NAK
> 

Understood, Instead of removing the 6018 compatible, I'll use the same
and fix the config mapped to 6018 in the hwspinlock driver.

Thanks,
Vignesh
> 
> Best regards,
> Krzysztof
> 

Re: [PATCH 0/3] Fix tcsr_mutex register for IPQ6018

[Konrad Dybcio]

On 4.09.2023 07:50, Vignesh Viswanathan wrote:
> IPQ6018 has 32 tcsr_mutext hwlock registers of 0x1000 size each.
> The compatible string qcom,ipq6018-tcsr-mutex is mapped to
> of_msm8226_tcsr_mutex which has 32 locks configured with stride of 0x80
> and doesn't match the HW present in IPQ6018.
> 
> This series fixes the following:
> 1. Fix the tcsr_mutex hwlock register size to 0x20000 in ipq6018.dtsi.
> 2. Remove qcom,ipq6018-tcsr-mutex compatible string for tcsr_mutex in
> ipq6018.dtsi.
> 3. Drop unused qcom,ipq6018-tcsr-mutex compatible string from
> qcom_hwspinlock driver and dt-bindings.
> 
Sounds like 20230904063344.4144086-1-quic_viswanat@quicinc.com can't
go in without this then?

Konrad

Re: [PATCH 2/3] hwspinlock: qcom: Drop unused qcom,ipq6018-tcsr-mutex

[Konrad Dybcio]

On 4.09.2023 08:42, Krzysztof Kozlowski wrote:
> On 04/09/2023 07:50, Vignesh Viswanathan wrote:
>> qcom,ipq6018-tcsr-mutex maps to incorrect config of IPQ6018 and is
>> dropped from the devictree. 
> 
> No, it is not dropped.
> 
> 
>> IPQ6018 will use qcom,tcsr-mutex compatible
>> string.
> 
> No, it will not.
> 
>>
>> Drop qcom,ipq6018-tcsr-mutex compatible string from
>> qcom_hwspinlock_of_match table.
> 
> Why? Do not write what you are doing here, but why you are doing it.
More importantly, looks like the ipq6018 compatible was added after
support for this SoC was introduced (see f5e303aefc06 and 5bf635621245a),
so if it's going to use of_tcsr_mutex data with the fallback compat, the
SoC-specific compatible can be removed from the driver.

Konrad

Re: [PATCH 2/3] hwspinlock: qcom: Drop unused qcom,ipq6018-tcsr-mutex

[Vignesh Viswanathan]

On 9/4/2023 9:31 PM, Konrad Dybcio wrote:
> On 4.09.2023 08:42, Krzysztof Kozlowski wrote:
>> On 04/09/2023 07:50, Vignesh Viswanathan wrote:
>>> qcom,ipq6018-tcsr-mutex maps to incorrect config of IPQ6018 and is
>>> dropped from the devictree.
>>
>> No, it is not dropped.
>>
>>
>>> IPQ6018 will use qcom,tcsr-mutex compatible
>>> string.
>>
>> No, it will not.
>>
>>>
>>> Drop qcom,ipq6018-tcsr-mutex compatible string from
>>> qcom_hwspinlock_of_match table.
>>
>> Why? Do not write what you are doing here, but why you are doing it.
> More importantly, looks like the ipq6018 compatible was added after
> support for this SoC was introduced (see f5e303aefc06 and 5bf635621245a),
> so if it's going to use of_tcsr_mutex data with the fallback compat, the
> SoC-specific compatible can be removed from the driver.
> 
Hi Konrad, Krzysztof,

I was planning to update the SOC-specific compatible for IPQ6018
qcom,ipq6018-tcsr-mutex to point to of_tcsr_mutex data in the of_match
table in the hwspinlock driver in V2.

Do you think this would be okay? or should I go ahead with removal of
IPQ6018 specific compatible so that it falls back to of_tcsr_mutex?

Thanks,
Vignesh

> Konrad

Re: [PATCH 0/3] Fix tcsr_mutex register for IPQ6018

[Vignesh Viswanathan]

On 9/4/2023 9:28 PM, Konrad Dybcio wrote:
> On 4.09.2023 07:50, Vignesh Viswanathan wrote:
>> IPQ6018 has 32 tcsr_mutext hwlock registers of 0x1000 size each.
>> The compatible string qcom,ipq6018-tcsr-mutex is mapped to
>> of_msm8226_tcsr_mutex which has 32 locks configured with stride of 0x80
>> and doesn't match the HW present in IPQ6018.
>>
>> This series fixes the following:
>> 1. Fix the tcsr_mutex hwlock register size to 0x20000 in ipq6018.dtsi.
>> 2. Remove qcom,ipq6018-tcsr-mutex compatible string for tcsr_mutex in
>> ipq6018.dtsi.
>> 3. Drop unused qcom,ipq6018-tcsr-mutex compatible string from
>> qcom_hwspinlock driver and dt-bindings.
>>
> Sounds like 20230904063344.4144086-1-quic_viswanat@quicinc.com can't
> go in without this then?
> 

They can independently go in as the current max size and stride set for
IPQ6018 still has 32 locks and moving the index 3 shouldn't cause any 
issues.

However, functionally IPQ6018's HW lock is broken and will be fixed once 
this series goes in.

Thanks,
Vignesh
> Konrad

Re: [PATCH 2/3] hwspinlock: qcom: Drop unused qcom,ipq6018-tcsr-mutex

[Krzysztof Kozlowski]

On 04/09/2023 18:09, Vignesh Viswanathan wrote:
> 
> 
> On 9/4/2023 9:31 PM, Konrad Dybcio wrote:
>> On 4.09.2023 08:42, Krzysztof Kozlowski wrote:
>>> On 04/09/2023 07:50, Vignesh Viswanathan wrote:
>>>> qcom,ipq6018-tcsr-mutex maps to incorrect config of IPQ6018 and is
>>>> dropped from the devictree.
>>>
>>> No, it is not dropped.
>>>
>>>
>>>> IPQ6018 will use qcom,tcsr-mutex compatible
>>>> string.
>>>
>>> No, it will not.
>>>
>>>>
>>>> Drop qcom,ipq6018-tcsr-mutex compatible string from
>>>> qcom_hwspinlock_of_match table.
>>>
>>> Why? Do not write what you are doing here, but why you are doing it.
>> More importantly, looks like the ipq6018 compatible was added after
>> support for this SoC was introduced (see f5e303aefc06 and 5bf635621245a),
>> so if it's going to use of_tcsr_mutex data with the fallback compat, the
>> SoC-specific compatible can be removed from the driver.
>>
> Hi Konrad, Krzysztof,
> 
> I was planning to update the SOC-specific compatible for IPQ6018
> qcom,ipq6018-tcsr-mutex to point to of_tcsr_mutex data in the of_match
> table in the hwspinlock driver in V2.
> 
> Do you think this would be okay? or should I go ahead with removal of
> IPQ6018 specific compatible so that it falls back to of_tcsr_mutex?

Remove, it's not needed in the driver.

Best regards,
Krzysztof

Re: [PATCH 2/3] hwspinlock: qcom: Drop unused qcom,ipq6018-tcsr-mutex

[Vignesh Viswanathan]

On 9/4/2023 11:46 PM, Krzysztof Kozlowski wrote:
> On 04/09/2023 18:09, Vignesh Viswanathan wrote:
>>
>>
>> On 9/4/2023 9:31 PM, Konrad Dybcio wrote:
>>> On 4.09.2023 08:42, Krzysztof Kozlowski wrote:
>>>> On 04/09/2023 07:50, Vignesh Viswanathan wrote:
>>>>> qcom,ipq6018-tcsr-mutex maps to incorrect config of IPQ6018 and is
>>>>> dropped from the devictree.
>>>>
>>>> No, it is not dropped.
>>>>
>>>>
>>>>> IPQ6018 will use qcom,tcsr-mutex compatible
>>>>> string.
>>>>
>>>> No, it will not.
>>>>
>>>>>
>>>>> Drop qcom,ipq6018-tcsr-mutex compatible string from
>>>>> qcom_hwspinlock_of_match table.
>>>>
>>>> Why? Do not write what you are doing here, but why you are doing it.
>>> More importantly, looks like the ipq6018 compatible was added after
>>> support for this SoC was introduced (see f5e303aefc06 and 5bf635621245a),
>>> so if it's going to use of_tcsr_mutex data with the fallback compat, the
>>> SoC-specific compatible can be removed from the driver.
>>>
>> Hi Konrad, Krzysztof,
>>
>> I was planning to update the SOC-specific compatible for IPQ6018
>> qcom,ipq6018-tcsr-mutex to point to of_tcsr_mutex data in the of_match
>> table in the hwspinlock driver in V2.
>>
>> Do you think this would be okay? or should I go ahead with removal of
>> IPQ6018 specific compatible so that it falls back to of_tcsr_mutex?
> 
> Remove, it's not needed in the driver.

Ack, will remove in V2.
> 
> Best regards,
> Krzysztof
>