[PATCH 0/2] PCI: qcom: Add basic interconnect support

[PATCH 0/2] PCI: qcom: Add basic interconnect support

[Johan Hovold]

On Qualcomm platforms like SC8280XP and SA8540P interconnect bandwidth
must be requested before enabling interconnect clocks.

Add basic support for managing an optional "pcie-mem" interconnect path
by setting a low constraint before enabling clocks and updating it after
the link is up.

This is specifically needed to prevent a crash on SC8280XP/SA8540P when
the interconnect constraints are enforced during boot.

As support for these platforms was added in 6.1-rc1 it would be nice to
have this merged as a fix for 6.1, but deferring for 6.2 works as well.

Johan


Johan Hovold (2):
  dt-bindings: PCI: qcom: Add SC8280XP/SA8540P interconnects
  PCI: qcom: Add basic interconnect support

 .../devicetree/bindings/pci/qcom,pcie.yaml    | 25 ++++++
 drivers/pci/controller/dwc/pcie-qcom.c        | 76 +++++++++++++++++++
 2 files changed, 101 insertions(+)

-- 
2.37.3

[PATCH 1/2] dt-bindings: PCI: qcom: Add SC8280XP/SA8540P interconnects

[Johan Hovold]

Add the missing SC8280XP/SA8540P "pcie-mem" and "cpu-pcie" interconnect
paths to the bindings.

Fixes: 76d777ae045e ("dt-bindings: PCI: qcom: Add SC8280XP to binding")
Fixes: 76c4207f4085 ("dt-bindings: PCI: qcom: Add SA8540P to binding")
Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
---
 .../devicetree/bindings/pci/qcom,pcie.yaml    | 25 +++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
index 22a2aac4c23f..a55434f95edd 100644
--- a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
+++ b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
@@ -62,6 +62,12 @@ properties:
     minItems: 3
     maxItems: 12
 
+  interconnects:
+    maxItems: 2
+
+  interconnect-names:
+    maxItems: 2
+
   resets:
     minItems: 1
     maxItems: 12
@@ -629,6 +635,25 @@ allOf:
           items:
             - const: pci # PCIe core reset
 
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - qcom,pcie-sa8540p
+              - qcom,pcie-sc8280xp
+    then:
+      properties:
+        interconnects:
+          maxItems: 2
+        interconnect-names:
+          items:
+            - const: pcie-mem
+            - const: cpu-pcie
+      required:
+        - interconnects
+        - interconnect-names
+
   - if:
       not:
         properties:
-- 
2.37.3

Re: [PATCH 1/2] dt-bindings: PCI: qcom: Add SC8280XP/SA8540P interconnects

[Krzysztof Kozlowski]

On 17/10/2022 07:24, Johan Hovold wrote:
> Add the missing SC8280XP/SA8540P "pcie-mem" and "cpu-pcie" interconnect
> paths to the bindings.
> 
> Fixes: 76d777ae045e ("dt-bindings: PCI: qcom: Add SC8280XP to binding")
> Fixes: 76c4207f4085 ("dt-bindings: PCI: qcom: Add SA8540P to binding")
> Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
> ---
>  .../devicetree/bindings/pci/qcom,pcie.yaml    | 25 +++++++++++++++++++
>  1 file changed, 25 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
> index 22a2aac4c23f..a55434f95edd 100644
> --- a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
> +++ b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
> @@ -62,6 +62,12 @@ properties:
>      minItems: 3
>      maxItems: 12
>  
> +  interconnects:
> +    maxItems: 2
> +
> +  interconnect-names:
> +    maxItems: 2
> +
>    resets:
>      minItems: 1
>      maxItems: 12
> @@ -629,6 +635,25 @@ allOf:
>            items:
>              - const: pci # PCIe core reset
>  
> +  - if:
> +      properties:
> +        compatible:
> +          contains:
> +            enum:
> +              - qcom,pcie-sa8540p
> +              - qcom,pcie-sc8280xp
> +    then:
> +      properties:
> +        interconnects:
> +          maxItems: 2

No need for this.

> +        interconnect-names:
> +          items:
> +            - const: pcie-mem
> +            - const: cpu-pcie
> +      required:
> +        - interconnects
> +        - interconnect-names

else:
  ??

Otherwise, you allow any names for other variants.

Best regards,
Krzysztof

Re: [PATCH 1/2] dt-bindings: PCI: qcom: Add SC8280XP/SA8540P interconnects

[Johan Hovold]

On Wed, Oct 19, 2022 at 10:37:31AM -0400, Krzysztof Kozlowski wrote:
> On 17/10/2022 07:24, Johan Hovold wrote:
> > Add the missing SC8280XP/SA8540P "pcie-mem" and "cpu-pcie" interconnect
> > paths to the bindings.
> > 
> > Fixes: 76d777ae045e ("dt-bindings: PCI: qcom: Add SC8280XP to binding")
> > Fixes: 76c4207f4085 ("dt-bindings: PCI: qcom: Add SA8540P to binding")
> > Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
> > ---
> >  .../devicetree/bindings/pci/qcom,pcie.yaml    | 25 +++++++++++++++++++
> >  1 file changed, 25 insertions(+)
> > 
> > diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
> > index 22a2aac4c23f..a55434f95edd 100644
> > --- a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
> > +++ b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
> > @@ -62,6 +62,12 @@ properties:
> >      minItems: 3
> >      maxItems: 12
> >  
> > +  interconnects:
> > +    maxItems: 2
> > +
> > +  interconnect-names:
> > +    maxItems: 2
> > +
> >    resets:
> >      minItems: 1
> >      maxItems: 12
> > @@ -629,6 +635,25 @@ allOf:
> >            items:
> >              - const: pci # PCIe core reset
> >  
> > +  - if:
> > +      properties:
> > +        compatible:
> > +          contains:
> > +            enum:
> > +              - qcom,pcie-sa8540p
> > +              - qcom,pcie-sc8280xp
> > +    then:
> > +      properties:
> > +        interconnects:
> > +          maxItems: 2
> 
> No need for this.
> 
> > +        interconnect-names:
> > +          items:
> > +            - const: pcie-mem
> > +            - const: cpu-pcie
> > +      required:
> > +        - interconnects
> > +        - interconnect-names
> 
> else:
>   ??
> 
> Otherwise, you allow any names for other variants.

Are you suggesting something like moving the names to the common
constraints for now:

  interconnects:
    maxItems: 2

  interconnect-names:
    items:
      - const: pcie-mem
      - const: cpu-pcie

and then in the allOf:

  - if:
      properties:
        compatible:
          contains:
            enum:
              - qcom,pcie-sa8540p
              - qcom,pcie-sc8280xp
    then:
      required:
        - interconnects
        - interconnect-names
    else:
      properties:
        interconnects: false
        interconnect-names: false

This way we'd catch anyone adding interconnects to a DTS without first
updating the bindings, but it also seems to go against the idea of
bindings fully describing the hardware by saying that no other platforms
have interconnects (when they actually do even if we don't describe it
just yet).

Or should we do the above but without the else clause to have some
constraints in place on the names at least?

Johan

Re: [PATCH 1/2] dt-bindings: PCI: qcom: Add SC8280XP/SA8540P interconnects

[Krzysztof Kozlowski]

On 20/10/2022 03:57, Johan Hovold wrote:
> On Wed, Oct 19, 2022 at 10:37:31AM -0400, Krzysztof Kozlowski wrote:
>> On 17/10/2022 07:24, Johan Hovold wrote:
>>> Add the missing SC8280XP/SA8540P "pcie-mem" and "cpu-pcie" interconnect
>>> paths to the bindings.
>>>
>>> Fixes: 76d777ae045e ("dt-bindings: PCI: qcom: Add SC8280XP to binding")
>>> Fixes: 76c4207f4085 ("dt-bindings: PCI: qcom: Add SA8540P to binding")
>>> Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
>>> ---
>>>  .../devicetree/bindings/pci/qcom,pcie.yaml    | 25 +++++++++++++++++++
>>>  1 file changed, 25 insertions(+)
>>>
>>> diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
>>> index 22a2aac4c23f..a55434f95edd 100644
>>> --- a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
>>> +++ b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
>>> @@ -62,6 +62,12 @@ properties:
>>>      minItems: 3
>>>      maxItems: 12
>>>  
>>> +  interconnects:
>>> +    maxItems: 2
>>> +
>>> +  interconnect-names:
>>> +    maxItems: 2
>>> +
>>>    resets:
>>>      minItems: 1
>>>      maxItems: 12
>>> @@ -629,6 +635,25 @@ allOf:
>>>            items:
>>>              - const: pci # PCIe core reset
>>>  
>>> +  - if:
>>> +      properties:
>>> +        compatible:
>>> +          contains:
>>> +            enum:
>>> +              - qcom,pcie-sa8540p
>>> +              - qcom,pcie-sc8280xp
>>> +    then:
>>> +      properties:
>>> +        interconnects:
>>> +          maxItems: 2
>>
>> No need for this.
>>
>>> +        interconnect-names:
>>> +          items:
>>> +            - const: pcie-mem
>>> +            - const: cpu-pcie
>>> +      required:
>>> +        - interconnects
>>> +        - interconnect-names
>>
>> else:
>>   ??
>>
>> Otherwise, you allow any names for other variants.
> 
> Are you suggesting something like moving the names to the common
> constraints for now:
> 
>   interconnects:
>     maxItems: 2
> 
>   interconnect-names:
>     items:
>       - const: pcie-mem
>       - const: cpu-pcie
> 
> and then in the allOf:
> 
>   - if:
>       properties:
>         compatible:
>           contains:
>             enum:
>               - qcom,pcie-sa8540p
>               - qcom,pcie-sc8280xp
>     then:
>       required:
>         - interconnects
>         - interconnect-names
>     else:
>       properties:
>         interconnects: false
>         interconnect-names: false
> 
> This way we'd catch anyone adding interconnects to a DTS without first
> updating the bindings, but it also seems to go against the idea of
> bindings fully describing the hardware by saying that no other platforms
> have interconnects (when they actually do even if we don't describe it
> just yet).

You can add a comment to the else like "TODO: Not described yet". I
would prefer to have specific but incomplete bindings, instead of loose
one which later might cause people adding whatever names they like.

> Or should we do the above but without the else clause to have some
> constraints in place on the names at least?

This would work as well if you think the names are applicable for other
devices.

Best regards,
Krzysztof

Re: [PATCH 1/2] dt-bindings: PCI: qcom: Add SC8280XP/SA8540P interconnects

[Johan Hovold]

On Thu, Oct 20, 2022 at 08:29:02AM -0400, Krzysztof Kozlowski wrote:
> On 20/10/2022 03:57, Johan Hovold wrote:
> > On Wed, Oct 19, 2022 at 10:37:31AM -0400, Krzysztof Kozlowski wrote:
> >> On 17/10/2022 07:24, Johan Hovold wrote:
> >>> Add the missing SC8280XP/SA8540P "pcie-mem" and "cpu-pcie" interconnect
> >>> paths to the bindings.
> >>>
> >>> Fixes: 76d777ae045e ("dt-bindings: PCI: qcom: Add SC8280XP to binding")
> >>> Fixes: 76c4207f4085 ("dt-bindings: PCI: qcom: Add SA8540P to binding")
> >>> Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
> >>> ---
> >>>  .../devicetree/bindings/pci/qcom,pcie.yaml    | 25 +++++++++++++++++++
> >>>  1 file changed, 25 insertions(+)
> >>>
> >>> diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
> >>> index 22a2aac4c23f..a55434f95edd 100644
> >>> --- a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml
> >>> +++ b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml

> > Are you suggesting something like moving the names to the common
> > constraints for now:
> > 
> >   interconnects:
> >     maxItems: 2
> > 
> >   interconnect-names:
> >     items:
> >       - const: pcie-mem
> >       - const: cpu-pcie
> > 
> > and then in the allOf:
> > 
> >   - if:
> >       properties:
> >         compatible:
> >           contains:
> >             enum:
> >               - qcom,pcie-sa8540p
> >               - qcom,pcie-sc8280xp
> >     then:
> >       required:
> >         - interconnects
> >         - interconnect-names
> >     else:
> >       properties:
> >         interconnects: false
> >         interconnect-names: false
> > 
> > This way we'd catch anyone adding interconnects to a DTS without first
> > updating the bindings, but it also seems to go against the idea of
> > bindings fully describing the hardware by saying that no other platforms
> > have interconnects (when they actually do even if we don't describe it
> > just yet).
> 
> You can add a comment to the else like "TODO: Not described yet". I
> would prefer to have specific but incomplete bindings, instead of loose
> one which later might cause people adding whatever names they like.
> 
> > Or should we do the above but without the else clause to have some
> > constraints in place on the names at least?
> 
> This would work as well if you think the names are applicable for other
> devices.

I think that's a reasonable assumption so I'll go with this alternative.

Thanks!

Johan

[PATCH 2/2] PCI: qcom: Add basic interconnect support

[Johan Hovold]

On Qualcomm platforms like SC8280XP and SA8540P interconnect bandwidth
must be requested before enabling interconnect clocks.

Add basic support for managing an optional "pcie-mem" interconnect path
by setting a low constraint before enabling clocks and updating it after
the link is up.

Note that it is not possible for a controller driver to set anything but
a maximum peak bandwidth as expected average bandwidth will vary with
use case and actual use (and power policy?). This very much remains an
unresolved problem with the interconnect framework.

Also note that no constraint is set for the SC8280XP/SA8540P "cpu-pcie"
path for now as it is not clear what an appropriate constraint would be
(and the system does not crash when left unspecified currently).

Fixes: 70574511f3fc ("PCI: qcom: Add support for SC8280XP")
Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
---
 drivers/pci/controller/dwc/pcie-qcom.c | 76 ++++++++++++++++++++++++++
 1 file changed, 76 insertions(+)

diff --git a/drivers/pci/controller/dwc/pcie-qcom.c b/drivers/pci/controller/dwc/pcie-qcom.c
index 7db94a22238d..0c13f976626f 100644
--- a/drivers/pci/controller/dwc/pcie-qcom.c
+++ b/drivers/pci/controller/dwc/pcie-qcom.c
@@ -12,6 +12,7 @@
 #include <linux/crc8.h>
 #include <linux/delay.h>
 #include <linux/gpio/consumer.h>
+#include <linux/interconnect.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/iopoll.h>
@@ -224,6 +225,7 @@ struct qcom_pcie {
 	union qcom_pcie_resources res;
 	struct phy *phy;
 	struct gpio_desc *reset;
+	struct icc_path *icc_mem;
 	const struct qcom_pcie_cfg *cfg;
 };
 
@@ -1644,6 +1646,74 @@ static const struct dw_pcie_ops dw_pcie_ops = {
 	.start_link = qcom_pcie_start_link,
 };
 
+static int qcom_pcie_icc_init(struct qcom_pcie *pcie)
+{
+	struct dw_pcie *pci = pcie->pci;
+	int ret;
+
+	pcie->icc_mem = devm_of_icc_get(pci->dev, "pcie-mem");
+	if (IS_ERR(pcie->icc_mem)) {
+		ret = PTR_ERR(pcie->icc_mem);
+		return ret;
+	}
+
+	/*
+	 * Some Qualcomm platforms require interconnect bandwidth constraints
+	 * to be set before enabling interconnect clocks.
+	 *
+	 * Set an initial peak bandwidth corresponding to single-lane Gen 1
+	 * for the pcie-mem path.
+	 */
+	ret = icc_set_bw(pcie->icc_mem, 0, MBps_to_icc(250));
+	if (ret) {
+		dev_err(pci->dev, "failed to set interconnect bandwidth: %d\n",
+			ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void qcom_pcie_icc_update(struct qcom_pcie *pcie)
+{
+	struct dw_pcie *pci = pcie->pci;
+	u32 offset, status, bw;
+	int speed, width;
+	int ret;
+
+	if (!pcie->icc_mem)
+		return;
+
+	offset = dw_pcie_find_capability(pci, PCI_CAP_ID_EXP);
+	status = readw(pci->dbi_base + offset + PCI_EXP_LNKSTA);
+
+	/* Only update constraints if link is up. */
+	if (!(status & PCI_EXP_LNKSTA_DLLLA))
+		return;
+
+	speed = FIELD_GET(PCI_EXP_LNKSTA_CLS, status);
+	width = FIELD_GET(PCI_EXP_LNKSTA_NLW, status);
+
+	switch (speed) {
+	case 1:
+		bw = MBps_to_icc(250);
+		break;
+	case 2:
+		bw = MBps_to_icc(500);
+		break;
+	default:
+	case 3:
+		bw = MBps_to_icc(985);
+		break;
+	}
+
+	ret = icc_set_bw(pcie->icc_mem, 0, width * bw);
+	if (ret) {
+		dev_err(pci->dev, "failed to set interconnect bandwidth: %d\n",
+			ret);
+	}
+}
+
 static int qcom_pcie_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
@@ -1704,6 +1774,10 @@ static int qcom_pcie_probe(struct platform_device *pdev)
 		goto err_pm_runtime_put;
 	}
 
+	ret = qcom_pcie_icc_init(pcie);
+	if (ret)
+		goto err_pm_runtime_put;
+
 	ret = pcie->cfg->ops->get_resources(pcie);
 	if (ret)
 		goto err_pm_runtime_put;
@@ -1722,6 +1796,8 @@ static int qcom_pcie_probe(struct platform_device *pdev)
 		goto err_phy_exit;
 	}
 
+	qcom_pcie_icc_update(pcie);
+
 	return 0;
 
 err_phy_exit:
-- 
2.37.3

Re: [PATCH 2/2] PCI: qcom: Add basic interconnect support

[Brian Masney]

On Mon, Oct 17, 2022 at 01:24:49PM +0200, Johan Hovold wrote:
> +	/*
> +	 * Some Qualcomm platforms require interconnect bandwidth constraints
> +	 * to be set before enabling interconnect clocks.
> +	 *
> +	 * Set an initial peak bandwidth corresponding to single-lane Gen 1
> +	 * for the pcie-mem path.
> +	 */
> +	ret = icc_set_bw(pcie->icc_mem, 0, MBps_to_icc(250));

[snip]

> +	speed = FIELD_GET(PCI_EXP_LNKSTA_CLS, status);
> +	width = FIELD_GET(PCI_EXP_LNKSTA_NLW, status);
> +
> +	switch (speed) {
> +	case 1:
> +		bw = MBps_to_icc(250);
> +		break;
> +	case 2:
> +		bw = MBps_to_icc(500);
> +		break;
> +	default:
> +	case 3:
> +		bw = MBps_to_icc(985);
> +		break;
> +	}

Just curious: These platforms have a 4 lane PCIe bus. Why use 985
instead of 1000 for the maximum?

Brian

Re: [PATCH 2/2] PCI: qcom: Add basic interconnect support

[Johan Hovold]

On Wed, Oct 19, 2022 at 10:32:41AM -0400, Brian Masney wrote:
> On Mon, Oct 17, 2022 at 01:24:49PM +0200, Johan Hovold wrote:
> > +	/*
> > +	 * Some Qualcomm platforms require interconnect bandwidth constraints
> > +	 * to be set before enabling interconnect clocks.
> > +	 *
> > +	 * Set an initial peak bandwidth corresponding to single-lane Gen 1
> > +	 * for the pcie-mem path.
> > +	 */
> > +	ret = icc_set_bw(pcie->icc_mem, 0, MBps_to_icc(250));
> 
> [snip]
> 
> > +	speed = FIELD_GET(PCI_EXP_LNKSTA_CLS, status);
> > +	width = FIELD_GET(PCI_EXP_LNKSTA_NLW, status);
> > +
> > +	switch (speed) {
> > +	case 1:
> > +		bw = MBps_to_icc(250);
> > +		break;
> > +	case 2:
> > +		bw = MBps_to_icc(500);
> > +		break;
> > +	default:
> > +	case 3:
> > +		bw = MBps_to_icc(985);
> > +		break;
> > +	}
> 
> Just curious: These platforms have a 4 lane PCIe bus. Why use 985
> instead of 1000 for the maximum?

This is the per-lane bandwidth that depends on encoding. The four-lane
peak throughput would be about 3.94 GB/s.

Johan

Re: [PATCH 2/2] PCI: qcom: Add basic interconnect support

[Brian Masney]

On Mon, Oct 17, 2022 at 01:24:49PM +0200, Johan Hovold wrote:
> On Qualcomm platforms like SC8280XP and SA8540P interconnect bandwidth
> must be requested before enabling interconnect clocks.
> 
> Add basic support for managing an optional "pcie-mem" interconnect path
> by setting a low constraint before enabling clocks and updating it after
> the link is up.
> 
> Note that it is not possible for a controller driver to set anything but
> a maximum peak bandwidth as expected average bandwidth will vary with
> use case and actual use (and power policy?). This very much remains an
> unresolved problem with the interconnect framework.
> 
> Also note that no constraint is set for the SC8280XP/SA8540P "cpu-pcie"
> path for now as it is not clear what an appropriate constraint would be
> (and the system does not crash when left unspecified currently).
> 
> Fixes: 70574511f3fc ("PCI: qcom: Add support for SC8280XP")
> Signed-off-by: Johan Hovold <johan+linaro@kernel.org>

Reviewed-by: Brian Masney <bmasney@redhat.com>