[PATCH v2 0/4] Add misc support for QDU1000/QRU1000 SoCs

[PATCH v2 0/4] Add misc support for QDU1000/QRU1000 SoCs

[Melody Olvera]

This series firmware, SoC, rpmpd, tz-log, and mmc bindings as well as
pmic, rpmpd, and socinfo support for QDU1000 and QRU1000 SoCs.

The Qualcomm Technologies, Inc. Distributed Unit 1000 and Radio Unit
1000 are new SoCs meant for enabling Open RAN solutions. See more at
https://www.qualcomm.com/content/dam/qcomm-martech/dm-assets/documents/qualcomm_5g_ran_platforms_product_brief.pdf

Melody Olvera (4):
  dt-bindings: firmware: scm: Add QDU1000/QRU1000 compatibles
  dt-bindings: power: rpmpd: Add QDU1000/QRU1000 to rpmpd binding
  soc: qcom: rpmhpd: Add QDU1000/QRU1000 power domains
  soc: qcom: socinfo: Add QDU1000/QRU1000 and variant IDs

 .../devicetree/bindings/firmware/qcom,scm.yaml   | 16 ++++++++++++++++
 .../devicetree/bindings/power/qcom,rpmpd.yaml    |  2 ++
 drivers/soc/qcom/rpmhpd.c                        | 15 +++++++++++++++
 drivers/soc/qcom/socinfo.c                       |  6 ++++++
 include/dt-bindings/power/qcom-rpmpd.h           |  6 ++++++
 5 files changed, 45 insertions(+)


base-commit: dca0a0385a4963145593ba417e1417af88a7c18d
-- 
2.38.0

[PATCH v2 1/4] dt-bindings: firmware: scm: Add QDU1000/QRU1000 compatibles

[Melody Olvera]

Add compatibles for scm driver for QDU1000 and QRU1000 platforms.

Signed-off-by: Melody Olvera <quic_molvera@quicinc.com>
---
 .../devicetree/bindings/firmware/qcom,scm.yaml   | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
index c5b76c9f7ad0..47083f47f109 100644
--- a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
+++ b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
@@ -38,6 +38,8 @@ properties:
           - qcom,scm-msm8994
           - qcom,scm-msm8996
           - qcom,scm-msm8998
+          - qcom,scm-qdu1000
+          - qcom,scm-qru1000
           - qcom,scm-sc7180
           - qcom,scm-sc7280
           - qcom,scm-sc8280xp
@@ -80,6 +82,20 @@ properties:
     description: TCSR hardware block
 
 allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - qcom,scm-qdu1000
+              - qcom,scm-qru1000
+    then:
+      properties:
+        '#reset-cells':
+          maxItems: 1
+
+      required:
+        - '#reset-cells'
   - if:
       properties:
         compatible:
-- 
2.38.0

[PATCH v2 2/4] dt-bindings: power: rpmpd: Add QDU1000/QRU1000 to rpmpd binding

[Melody Olvera]

Add compatible and constants for the power domains exposed by the RPMH
in the Qualcomm QDU1000 and QRU1000 platforms.

Signed-off-by: Melody Olvera <quic_molvera@quicinc.com>
---
 Documentation/devicetree/bindings/power/qcom,rpmpd.yaml | 2 ++
 include/dt-bindings/power/qcom-rpmpd.h                  | 6 ++++++
 2 files changed, 8 insertions(+)

diff --git a/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml b/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml
index 5b4eda919911..ffac5fcf5f6f 100644
--- a/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml
+++ b/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml
@@ -28,6 +28,8 @@ properties:
       - qcom,msm8998-rpmpd
       - qcom,qcm2290-rpmpd
       - qcom,qcs404-rpmpd
+      - qcom,qdu1000-rpmhpd
+      - qcom,qru1000-rpmhpd
       - qcom,sa8540p-rpmhpd
       - qcom,sdm660-rpmpd
       - qcom,sc7180-rpmhpd
diff --git a/include/dt-bindings/power/qcom-rpmpd.h b/include/dt-bindings/power/qcom-rpmpd.h
index f5f82dde7399..965c065a9caa 100644
--- a/include/dt-bindings/power/qcom-rpmpd.h
+++ b/include/dt-bindings/power/qcom-rpmpd.h
@@ -103,6 +103,12 @@
 #define SM8450_MXC_AO	11
 #define SM8450_MSS	12
 
+/* QDU1000/QRU1000 Power Domain Indexes */
+#define QDRU1000_EBI	0
+#define QDRU1000_MSS	1
+#define QDRU1000_CX	2
+#define QDRU1000_MX	3
+
 /* SC7180 Power Domain Indexes */
 #define SC7180_CX	0
 #define SC7180_CX_AO	1
-- 
2.38.0

[PATCH v2 3/4] soc: qcom: rpmhpd: Add QDU1000/QRU1000 power domains

[Melody Olvera]

Add the power domains exposed by RPMH in the Qualcomm QDU1000
and QRU1000 platforms.

Signed-off-by: Melody Olvera <quic_molvera@quicinc.com>
---
 drivers/soc/qcom/rpmhpd.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/soc/qcom/rpmhpd.c b/drivers/soc/qcom/rpmhpd.c
index 092f6ab09acf..4b8198697245 100644
--- a/drivers/soc/qcom/rpmhpd.c
+++ b/drivers/soc/qcom/rpmhpd.c
@@ -353,6 +353,19 @@ static const struct rpmhpd_desc sm8450_desc = {
 	.num_pds = ARRAY_SIZE(sm8450_rpmhpds),
 };
 
+/* QDU1000/QRU1000 RPMH powerdomains */
+static struct rpmhpd *qdu1000_rpmhpds[] = {
+	[QDRU1000_CX] = &cx,
+	[QDRU1000_EBI] = &ebi,
+	[QDRU1000_MSS] = &mss,
+	[QDRU1000_MX] = &mx,
+};
+
+static const struct rpmhpd_desc qdu1000_desc = {
+	.rpmhpds = qdu1000_rpmhpds,
+	.num_pds = ARRAY_SIZE(qdu1000_rpmhpds),
+};
+
 /* SC7180 RPMH powerdomains */
 static struct rpmhpd *sc7180_rpmhpds[] = {
 	[SC7180_CX] = &cx_w_mx_parent,
@@ -430,6 +443,8 @@ static const struct rpmhpd_desc sc8280xp_desc = {
 };
 
 static const struct of_device_id rpmhpd_match_table[] = {
+	{ .compatible = "qcom,qdu1000-rpmhpd", .data = &qdu1000_desc },
+	{ .compatible = "qcom,qru1000-rpmhpd", .data = &qdu1000_desc },
 	{ .compatible = "qcom,sa8540p-rpmhpd", .data = &sa8540p_desc },
 	{ .compatible = "qcom,sc7180-rpmhpd", .data = &sc7180_desc },
 	{ .compatible = "qcom,sc7280-rpmhpd", .data = &sc7280_desc },
-- 
2.38.0

[PATCH v2 4/4] soc: qcom: socinfo: Add QDU1000/QRU1000 and variant IDs

[Melody Olvera]

Add soc IDs for QDU1000 and QRU1000 and their variants.

Signed-off-by: Melody Olvera <quic_molvera@quicinc.com>
---
 drivers/soc/qcom/socinfo.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/soc/qcom/socinfo.c b/drivers/soc/qcom/socinfo.c
index aa37e1bad095..5677b427c78d 100644
--- a/drivers/soc/qcom/socinfo.c
+++ b/drivers/soc/qcom/socinfo.c
@@ -336,6 +336,12 @@ static const struct soc_id soc_id[] = {
 	{ 487, "SC7280" },
 	{ 495, "SC7180P" },
 	{ 507, "SM6375" },
+	{ 539, "QRU1000" },
+	{ 545, "QDU1000" },
+	{ 587, "QDU1010" },
+	{ 588, "QRU1032" },
+	{ 589, "QRU1052" },
+	{ 590, "QRU1062" },
 };
 
 static const char *socinfo_machine(struct device *dev, unsigned int id)
-- 
2.38.0

Re: [PATCH v2 0/4] Add misc support for QDU1000/QRU1000 SoCs

[Melody Olvera]

On 10/14/2022 3:11 PM, Melody Olvera wrote:
> This series firmware, SoC, rpmpd, tz-log, and mmc bindings as well as
> pmic, rpmpd, and socinfo support for QDU1000 and QRU1000 SoCs.
>
> The Qualcomm Technologies, Inc. Distributed Unit 1000 and Radio Unit
> 1000 are new SoCs meant for enabling Open RAN solutions. See more at
> https://www.qualcomm.com/content/dam/qcomm-martech/dm-assets/documents/qualcomm_5g_ran_platforms_product_brief.pdf
Changes from v1:
- fixed ordering
- removed instances of qdru
- moved CPU documentation patch to dt patchset
- added compatible fields to rpmhpd match table
>
> Melody Olvera (4):
>   dt-bindings: firmware: scm: Add QDU1000/QRU1000 compatibles
>   dt-bindings: power: rpmpd: Add QDU1000/QRU1000 to rpmpd binding
>   soc: qcom: rpmhpd: Add QDU1000/QRU1000 power domains
>   soc: qcom: socinfo: Add QDU1000/QRU1000 and variant IDs
>
>  .../devicetree/bindings/firmware/qcom,scm.yaml   | 16 ++++++++++++++++
>  .../devicetree/bindings/power/qcom,rpmpd.yaml    |  2 ++
>  drivers/soc/qcom/rpmhpd.c                        | 15 +++++++++++++++
>  drivers/soc/qcom/socinfo.c                       |  6 ++++++
>  include/dt-bindings/power/qcom-rpmpd.h           |  6 ++++++
>  5 files changed, 45 insertions(+)
>
>
> base-commit: dca0a0385a4963145593ba417e1417af88a7c18d

Re: [PATCH v2 1/4] dt-bindings: firmware: scm: Add QDU1000/QRU1000 compatibles

[Krzysztof Kozlowski]

On 14/10/2022 18:11, Melody Olvera wrote:
> Add compatibles for scm driver for QDU1000 and QRU1000 platforms.
> 
> Signed-off-by: Melody Olvera <quic_molvera@quicinc.com>
> ---
>  .../devicetree/bindings/firmware/qcom,scm.yaml   | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
> index c5b76c9f7ad0..47083f47f109 100644
> --- a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
> +++ b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
> @@ -38,6 +38,8 @@ properties:
>            - qcom,scm-msm8994
>            - qcom,scm-msm8996
>            - qcom,scm-msm8998
> +          - qcom,scm-qdu1000
> +          - qcom,scm-qru1000

Why exactly we are no using qdu1000 as fallback? That was the
recommendation in previous discussion.

Patch is still incomplete - you still do no have proper changes in allOf
for the clocks. If you want to say that this SoC does not take any
clocks as input, then they should not be allowed.

Best regards,
Krzysztof

Re: [PATCH v2 2/4] dt-bindings: power: rpmpd: Add QDU1000/QRU1000 to rpmpd binding

[Krzysztof Kozlowski]

On 14/10/2022 18:11, Melody Olvera wrote:
> Add compatible and constants for the power domains exposed by the RPMH
> in the Qualcomm QDU1000 and QRU1000 platforms.
> 
> Signed-off-by: Melody Olvera <quic_molvera@quicinc.com>
> ---
>  Documentation/devicetree/bindings/power/qcom,rpmpd.yaml | 2 ++
>  include/dt-bindings/power/qcom-rpmpd.h                  | 6 ++++++
>  2 files changed, 8 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml b/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml
> index 5b4eda919911..ffac5fcf5f6f 100644
> --- a/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml
> +++ b/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml
> @@ -28,6 +28,8 @@ properties:
>        - qcom,msm8998-rpmpd
>        - qcom,qcm2290-rpmpd
>        - qcom,qcs404-rpmpd
> +      - qcom,qdu1000-rpmhpd
> +      - qcom,qru1000-rpmhpd

Same question about fallback.

Best regards,
Krzysztof

Re: [PATCH v2 1/4] dt-bindings: firmware: scm: Add QDU1000/QRU1000 compatibles

[Melody Olvera]

On 10/15/2022 6:34 AM, Krzysztof Kozlowski wrote:
> On 14/10/2022 18:11, Melody Olvera wrote:
>> Add compatibles for scm driver for QDU1000 and QRU1000 platforms.
>>
>> Signed-off-by: Melody Olvera <quic_molvera@quicinc.com>
>> ---
>>  .../devicetree/bindings/firmware/qcom,scm.yaml   | 16 ++++++++++++++++
>>  1 file changed, 16 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
>> index c5b76c9f7ad0..47083f47f109 100644
>> --- a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
>> +++ b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
>> @@ -38,6 +38,8 @@ properties:
>>            - qcom,scm-msm8994
>>            - qcom,scm-msm8996
>>            - qcom,scm-msm8998
>> +          - qcom,scm-qdu1000
>> +          - qcom,scm-qru1000
> Why exactly we are no using qdu1000 as fallback? That was the
> recommendation in previous discussion.
Will use only qdu; I think I misunderstood the outcome of that discussion.
>
> Patch is still incomplete - you still do no have proper changes in allOf
> for the clocks. If you want to say that this SoC does not take any
> clocks as input, then they should not be allowed.
That is what I'm trying to say; it seems most of our most recent SoCs (sm8*) don't have any
clocks associated with the scm. Does it make sense to remove the minItems earlier
in the binding, or is there something else that would communicate this in allOf better?

Thanks,
Melody

Re: [PATCH v2 1/4] dt-bindings: firmware: scm: Add QDU1000/QRU1000 compatibles

[Krzysztof Kozlowski]

On 19/10/2022 14:08, Melody Olvera wrote:
> 
> 
> On 10/15/2022 6:34 AM, Krzysztof Kozlowski wrote:
>> On 14/10/2022 18:11, Melody Olvera wrote:
>>> Add compatibles for scm driver for QDU1000 and QRU1000 platforms.
>>>
>>> Signed-off-by: Melody Olvera <quic_molvera@quicinc.com>
>>> ---
>>>  .../devicetree/bindings/firmware/qcom,scm.yaml   | 16 ++++++++++++++++
>>>  1 file changed, 16 insertions(+)
>>>
>>> diff --git a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
>>> index c5b76c9f7ad0..47083f47f109 100644
>>> --- a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
>>> +++ b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
>>> @@ -38,6 +38,8 @@ properties:
>>>            - qcom,scm-msm8994
>>>            - qcom,scm-msm8996
>>>            - qcom,scm-msm8998
>>> +          - qcom,scm-qdu1000
>>> +          - qcom,scm-qru1000
>> Why exactly we are no using qdu1000 as fallback? That was the
>> recommendation in previous discussion.
> Will use only qdu; I think I misunderstood the outcome of that discussion.

Actually, I think I commented about this in wrong patch. I think the
outcome was to use two compatibles for most of the cases, but as a
fallback, so:

QDU: "qcom,qdu1000-rpmhpd"
QRU: "qcom,qru1000-rpmhpd", "qcom,qdu1000-rpmhpd"
(or skip entirely second if you do not customize QRU in DTSI)

However here we already have a fallback, so these are fine:

"qcom,scm-qdu1000", "qcom,scm"
"qcom,scm-qru1000", "qcom,scm"

Still assuming you customize them in DTSI... which does not seem the
case, right?

>>
>> Patch is still incomplete - you still do no have proper changes in allOf
>> for the clocks. If you want to say that this SoC does not take any
>> clocks as input, then they should not be allowed.
> That is what I'm trying to say; it seems most of our most recent SoCs (sm8*) don't have any
> clocks associated with the scm. Does it make sense to remove the minItems earlier
> in the binding, or is there something else that would communicate this in allOf better?
> 


Then disallow clocks for your variant:

  - if:
     ....
    then:
     ...
      clocks: false
      clock-names: false

Best regards,
Krzysztof

Re: [PATCH v2 1/4] dt-bindings: firmware: scm: Add QDU1000/QRU1000 compatibles

[Melody Olvera]

On 10/20/2022 5:35 AM, Krzysztof Kozlowski wrote:
> On 19/10/2022 14:08, Melody Olvera wrote:
>>
>> On 10/15/2022 6:34 AM, Krzysztof Kozlowski wrote:
>>> On 14/10/2022 18:11, Melody Olvera wrote:
>>>> Add compatibles for scm driver for QDU1000 and QRU1000 platforms.
>>>>
>>>> Signed-off-by: Melody Olvera <quic_molvera@quicinc.com>
>>>> ---
>>>>  .../devicetree/bindings/firmware/qcom,scm.yaml   | 16 ++++++++++++++++
>>>>  1 file changed, 16 insertions(+)
>>>>
>>>> diff --git a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
>>>> index c5b76c9f7ad0..47083f47f109 100644
>>>> --- a/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
>>>> +++ b/Documentation/devicetree/bindings/firmware/qcom,scm.yaml
>>>> @@ -38,6 +38,8 @@ properties:
>>>>            - qcom,scm-msm8994
>>>>            - qcom,scm-msm8996
>>>>            - qcom,scm-msm8998
>>>> +          - qcom,scm-qdu1000
>>>> +          - qcom,scm-qru1000
>>> Why exactly we are no using qdu1000 as fallback? That was the
>>> recommendation in previous discussion.
>> Will use only qdu; I think I misunderstood the outcome of that discussion.
> Actually, I think I commented about this in wrong patch. I think the
> outcome was to use two compatibles for most of the cases, but as a
> fallback, so:
>
> QDU: "qcom,qdu1000-rpmhpd"
> QRU: "qcom,qru1000-rpmhpd", "qcom,qdu1000-rpmhpd"
> (or skip entirely second if you do not customize QRU in DTSI)
>
> However here we already have a fallback, so these are fine:
>
> "qcom,scm-qdu1000", "qcom,scm"
> "qcom,scm-qru1000", "qcom,scm"
>
> Still assuming you customize them in DTSI... which does not seem the
> case, right?
Yeah dtsi is largely shared between RU and DU. It probably makes more sense to
drop RU compat string all together unless there is a significant difference.
>>> Patch is still incomplete - you still do no have proper changes in allOf
>>> for the clocks. If you want to say that this SoC does not take any
>>> clocks as input, then they should not be allowed.
>> That is what I'm trying to say; it seems most of our most recent SoCs (sm8*) don't have any
>> clocks associated with the scm. Does it make sense to remove the minItems earlier
>> in the binding, or is there something else that would communicate this in allOf better?
>>
>
> Then disallow clocks for your variant:
>
>   - if:
>      ....
>     then:
>      ...
>       clocks: false
>       clock-names: false
Got it; thanks.

Thanks,
Melody