Re: [PATCH v5 13/16] cxl/pci: Add error handler for CXL PCIe Port RAS errors

[Bowman, Terry]

On 1/14/2025 4:51 PM, Ira Weiny wrote:
> Terry Bowman wrote:
>> Introduce correctable and uncorrectable CXL PCIe Port Protocol Error
>> handlers.
>>
>> The handlers will be called with a 'struct pci_dev' parameter
>> indicating the CXL Port device requiring handling. The CXL PCIe Port
>> device's underlying 'struct device' will match the port device in the
>> CXL topology.
>>
>> Use the PCIe Port's device object to find the matching CXL Upstream Switch
>> Port, CXL Downstream Switch Port, or CXL Root Port in the CXL topology. The
>> matching CXL Port device should contain a cached reference to the RAS
>> register block. The cached RAS block will be used handling the error.
>>
>> Invoke the existing __cxl_handle_ras() or __cxl_handle_cor_ras() using
>> a reference to the RAS registers as a parameter. These functions will use
>> the RAS register reference to indicate an error and clear the device's RAS
>> status.
>>
>> Future patches will assign the error handlers and add trace logging.
>>
>> Signed-off-by: Terry Bowman <terry.bowman@xxxxxxx>
>> ---
>>  drivers/cxl/core/pci.c | 63 ++++++++++++++++++++++++++++++++++++++++++
>>  1 file changed, 63 insertions(+)
>>
>> diff --git a/drivers/cxl/core/pci.c b/drivers/cxl/core/pci.c
>> index 8275b3dc3589..411834f7efe0 100644
>> --- a/drivers/cxl/core/pci.c
>> +++ b/drivers/cxl/core/pci.c
>> @@ -776,6 +776,69 @@ static void cxl_disable_rch_root_ints(struct cxl_dport *dport)
>>  	writel(aer_cmd, aer_base + PCI_ERR_ROOT_COMMAND);
>>  }
>>  
>> +static int match_uport(struct device *dev, const void *data)
>> +{
>> +	struct device *uport_dev = (struct device *)data;
>> +	struct cxl_port *port;
>> +
>> +	if (!is_cxl_port(dev))
>> +		return 0;
>> +
>> +	port = to_cxl_port(dev);
>> +
>> +	return port->uport_dev == uport_dev;
>> +}
>> +
>> +static void __iomem *cxl_pci_port_ras(struct pci_dev *pdev)
>> +{
>> +	struct cxl_port *port;
>> +
>> +	if (!pdev)
>> +		return NULL;
>> +
>> +	if ((pci_pcie_type(pdev) == PCI_EXP_TYPE_ROOT_PORT) ||
>> +	    (pci_pcie_type(pdev) == PCI_EXP_TYPE_DOWNSTREAM)) {
>> +		struct cxl_dport *dport;
>> +		void __iomem *ras_base;
>> +
>> +		port = find_cxl_port(&pdev->dev, &dport);
>> +		ras_base = dport ? dport->regs.ras : NULL;
>> +		if (port)
>> +			put_device(&port->dev);
>> +		return ras_base;
>> +	} else if (pci_pcie_type(pdev) == PCI_EXP_TYPE_UPSTREAM) {
>> +		struct device *port_dev;
>> +
>> +		port_dev = bus_find_device(&cxl_bus_type, NULL, &pdev->dev,
>> +					   match_uport);
>> +		if (!port_dev)
>> +			return NULL;
>> +
>> +		port = to_cxl_port(port_dev);
>> +		if (!port)
>> +			return NULL;
>> +
>> +		put_device(port_dev);
> Is there any chance the cxl_port (and subsequently the mapping of the ras
> registers) could go away between here and their use in
> __cxl_handle_*_ras()?
>
> Ira
Yes, this could happen.

>> +		return port->uport_regs.ras;
>> +	}
>> +
>> +	return NULL;
>> +}
>> +
>> +static void cxl_port_cor_error_detected(struct pci_dev *pdev)
>> +{
>> +	void __iomem *ras_base = cxl_pci_port_ras(pdev);
>> +
>> +	__cxl_handle_cor_ras(&pdev->dev, ras_base);
>> +}
>> +
>> +static bool cxl_port_error_detected(struct pci_dev *pdev)
>> +{
>> +	void __iomem *ras_base = cxl_pci_port_ras(pdev);
>> +
>> +	return __cxl_handle_ras(&pdev->dev, ras_base);
>> +}
>> +
>>  void cxl_uport_init_ras_reporting(struct cxl_port *port)
>>  {
>>  	/* uport may have more than 1 downstream EP. Check if already mapped. */
>> -- 
>> 2.34.1
>>
>