[PATCH] PCI/TPH: Skip Root Port completer check for RC_END devices

[PATCH] PCI/TPH: Skip Root Port completer check for RC_END devices

[George Abraham P]

Root Complex Integrated Endpoint devices (PCI_EXP_TYPE_RC_END) are
directly integrated into the root complex and do not have an
associated Root Port in the traditional PCIe hierarchy. The current
TPH implementation incorrectly attempts to find and check a Root Port's
TPH completer capability for these devices.

Add a check to skip Root Port completer type verification for RC_END
devices, allowing them to use their full TPH requester capability
without being limited by a non-existent Root Port's completer support.

For RC_END devices, the root complex itself acts as the TPH completer,
and this relationship is handled differently than the standard
endpoint-to-Root-Port model.

Signed-off-by: George Abraham P <george.abraham.p@intel.com>
---
 drivers/pci/tph.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/pci/tph.c b/drivers/pci/tph.c
index cc64f93709a4..c61456d24f61 100644
--- a/drivers/pci/tph.c
+++ b/drivers/pci/tph.c
@@ -397,10 +397,13 @@ int pcie_enable_tph(struct pci_dev *pdev, int mode)
 	else
 		pdev->tph_req_type = PCI_TPH_REQ_TPH_ONLY;
 
-	rp_req_type = get_rp_completer_type(pdev);
+	/* Check if the device is behind a Root Port */
+	if (pci_pcie_type(pdev) != PCI_EXP_TYPE_RC_END) {
+		rp_req_type = get_rp_completer_type(pdev);
 
-	/* Final req_type is the smallest value of two */
-	pdev->tph_req_type = min(pdev->tph_req_type, rp_req_type);
+		/* Final req_type is the smallest value of two */
+		pdev->tph_req_type = min(pdev->tph_req_type, rp_req_type);
+	}
 
 	if (pdev->tph_req_type == PCI_TPH_REQ_DISABLE)
 		return -EINVAL;

base-commit: c29008e61d8e75ac7da3efd5310e253c035e0458
-- 
2.40.1

Re: [PATCH] PCI/TPH: Skip Root Port completer check for RC_END devices

[Bjorn Helgaas]

[+cc authors of TPH support]

On Thu, Sep 18, 2025 at 02:19:40PM +0530, George Abraham P wrote:
> Root Complex Integrated Endpoint devices (PCI_EXP_TYPE_RC_END) are
> directly integrated into the root complex and do not have an
> associated Root Port in the traditional PCIe hierarchy. The current
> TPH implementation incorrectly attempts to find and check a Root Port's
> TPH completer capability for these devices.
> 
> Add a check to skip Root Port completer type verification for RC_END
> devices, allowing them to use their full TPH requester capability
> without being limited by a non-existent Root Port's completer support.
> 
> For RC_END devices, the root complex itself acts as the TPH completer,
> and this relationship is handled differently than the standard
> endpoint-to-Root-Port model.

I thought maybe the spec would mention TPH Completer Supported for a
Root Complex in an RCRB, but I looked through PCIe r7.0 and didn't see
anything in RCRB related to the Root Port TPH Completer Supported
field in Device Capabilities 2.

It seems sort of surprising that Root Ports have to advertise what
kinds of TPH Completers they support, but we can assume that Root
Complexes support both TPH and Extended TPH Completers.  Do you have
any insight into that?

But I certainly agree that as-is, TPH is useless for RCiEPs since
there's no Root Port, so we assume the completer has no TPH Completer
support at all.

Do you think we should add a Fixes: tag for f69767a1ada3 ("PCI: Add
TLP Processing Hints (TPH) support"), where the TPH support was added?

> Signed-off-by: George Abraham P <george.abraham.p@intel.com>
> ---
>  drivers/pci/tph.c | 9 ++++++---
>  1 file changed, 6 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/pci/tph.c b/drivers/pci/tph.c
> index cc64f93709a4..c61456d24f61 100644
> --- a/drivers/pci/tph.c
> +++ b/drivers/pci/tph.c
> @@ -397,10 +397,13 @@ int pcie_enable_tph(struct pci_dev *pdev, int mode)
>  	else
>  		pdev->tph_req_type = PCI_TPH_REQ_TPH_ONLY;
>  
> -	rp_req_type = get_rp_completer_type(pdev);
> +	/* Check if the device is behind a Root Port */
> +	if (pci_pcie_type(pdev) != PCI_EXP_TYPE_RC_END) {
> +		rp_req_type = get_rp_completer_type(pdev);
>  
> -	/* Final req_type is the smallest value of two */
> -	pdev->tph_req_type = min(pdev->tph_req_type, rp_req_type);
> +		/* Final req_type is the smallest value of two */
> +		pdev->tph_req_type = min(pdev->tph_req_type, rp_req_type);
> +	}
>  
>  	if (pdev->tph_req_type == PCI_TPH_REQ_DISABLE)
>  		return -EINVAL;
> 
> base-commit: c29008e61d8e75ac7da3efd5310e253c035e0458
> -- 
> 2.40.1
> 

Re: [PATCH] PCI/TPH: Skip Root Port completer check for RC_END devices

[George Abraham P]

Hi Bjorn,

On 19-Sep-25 12:58 AM, Bjorn Helgaas wrote:
> [+cc authors of TPH support]
> 
> On Thu, Sep 18, 2025 at 02:19:40PM +0530, George Abraham P wrote:
>> Root Complex Integrated Endpoint devices (PCI_EXP_TYPE_RC_END) are
>> directly integrated into the root complex and do not have an
>> associated Root Port in the traditional PCIe hierarchy. The current
>> TPH implementation incorrectly attempts to find and check a Root Port's
>> TPH completer capability for these devices.
>>
>> Add a check to skip Root Port completer type verification for RC_END
>> devices, allowing them to use their full TPH requester capability
>> without being limited by a non-existent Root Port's completer support.
>>
>> For RC_END devices, the root complex itself acts as the TPH completer,
>> and this relationship is handled differently than the standard
>> endpoint-to-Root-Port model.
> 
> I thought maybe the spec would mention TPH Completer Supported for a
> Root Complex in an RCRB, but I looked through PCIe r7.0 and didn't see
> anything in RCRB related to the Root Port TPH Completer Supported
> field in Device Capabilities 2.
> 
> It seems sort of surprising that Root Ports have to advertise what
> kinds of TPH Completers they support, but we can assume that Root
> Complexes support both TPH and Extended TPH Completers.  Do you have
> any insight into that?
> 

I too was was scouring the PCI spec for this same information. Unfortunately
couldn't find the solution to the same.
The only statement in the spec that supports this patch would be that the TPH
Completer Supported field is only applicable to Root Ports and Endpoints.
Hence, for RCiEP, this check should be bypassed.
If there is a better way to handle this, please suggest.

> But I certainly agree that as-is, TPH is useless for RCiEPs since
> there's no Root Port, so we assume the completer has no TPH Completer
> support at all.
> 
> Do you think we should add a Fixes: tag for f69767a1ada3 ("PCI: Add
> TLP Processing Hints (TPH) support"), where the TPH support was added?
> 

Sure, will modify as suggested. Thanks.

>> Signed-off-by: George Abraham P <george.abraham.p@intel.com>
>> ---
>>  drivers/pci/tph.c | 9 ++++++---
>>  1 file changed, 6 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/pci/tph.c b/drivers/pci/tph.c
>> index cc64f93709a4..c61456d24f61 100644
>> --- a/drivers/pci/tph.c
>> +++ b/drivers/pci/tph.c
>> @@ -397,10 +397,13 @@ int pcie_enable_tph(struct pci_dev *pdev, int mode)
>>  	else
>>  		pdev->tph_req_type = PCI_TPH_REQ_TPH_ONLY;
>>  
>> -	rp_req_type = get_rp_completer_type(pdev);
>> +	/* Check if the device is behind a Root Port */
>> +	if (pci_pcie_type(pdev) != PCI_EXP_TYPE_RC_END) {
>> +		rp_req_type = get_rp_completer_type(pdev);
>>  
>> -	/* Final req_type is the smallest value of two */
>> -	pdev->tph_req_type = min(pdev->tph_req_type, rp_req_type);
>> +		/* Final req_type is the smallest value of two */
>> +		pdev->tph_req_type = min(pdev->tph_req_type, rp_req_type);
>> +	}
>>  
>>  	if (pdev->tph_req_type == PCI_TPH_REQ_DISABLE)
>>  		return -EINVAL;
>>
>> base-commit: c29008e61d8e75ac7da3efd5310e253c035e0458
>> -- 
>> 2.40.1
>>

Regards,
George