Re: [PATCH v4 2/2] PCI/ERR: Split the fatal and non-fatal error recovery handling

[Kuppuswamy, Sathyanarayanan]

On 10/13/20 4:56 AM, Christoph Hellwig wrote:
> You might want to split out pcie_do_fatal_recovery and get rid of the
> state argument:
This is how it was before Keith merged fatal and non-fatal error recovery
paths. When the comparison is between additional-parameter vs new-interface
, I choose the former. But I can merge your change in next version.

> diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
> index fa12f7cbc1a095..eec0d3fe9fd967 100644
> --- a/drivers/pci/pci.h
> +++ b/drivers/pci/pci.h
> @@ -556,7 +556,8 @@ static inline int pci_dev_specific_disable_acs_redir(struct pci_dev *dev)
>   
>   /* PCI error reporting and recovery */
>   pci_ers_result_t pcie_do_recovery(struct pci_dev *dev,
> -			pci_channel_state_t state,
> +			pci_ers_result_t (*reset_link)(struct pci_dev *pdev));
> +pci_ers_result_t pcie_do_fatal_recovery(struct pci_dev *dev,
>   			pci_ers_result_t (*reset_link)(struct pci_dev *pdev));
>   
>   bool pcie_wait_for_link(struct pci_dev *pdev, bool active);
> diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c
> index 65dff5f3457ac0..4bf7ebb34cf854 100644
> --- a/drivers/pci/pcie/aer.c
> +++ b/drivers/pci/pcie/aer.c
> @@ -947,9 +947,9 @@ static void handle_error_source(struct pci_dev *dev, struct aer_err_info *info)
>   		if (pcie_aer_is_native(dev))
>   			pcie_clear_device_status(dev);
>   	} else if (info->severity == AER_NONFATAL)
> -		pcie_do_recovery(dev, pci_channel_io_normal, aer_root_reset);
> +		pcie_do_recovery(dev, aer_root_reset);
>   	else if (info->severity == AER_FATAL)
> -		pcie_do_recovery(dev, pci_channel_io_frozen, aer_root_reset);
> +		pcie_do_fatal_recovery(dev, aer_root_reset);
>   	pci_dev_put(dev);
>   }
>   
> @@ -985,11 +985,9 @@ static void aer_recover_work_func(struct work_struct *work)
>   		}
>   		cper_print_aer(pdev, entry.severity, entry.regs);
>   		if (entry.severity == AER_NONFATAL)
> -			pcie_do_recovery(pdev, pci_channel_io_normal,
> -					 aer_root_reset);
> +			pcie_do_recovery(pdev, aer_root_reset);
>   		else if (entry.severity == AER_FATAL)
> -			pcie_do_recovery(pdev, pci_channel_io_frozen,
> -					 aer_root_reset);
> +			pcie_do_fatal_recovery(pdev, aer_root_reset);
>   		pci_dev_put(pdev);
>   	}
>   }
> diff --git a/drivers/pci/pcie/dpc.c b/drivers/pci/pcie/dpc.c
> index daa9a4153776ce..74e7d1da3cf054 100644
> --- a/drivers/pci/pcie/dpc.c
> +++ b/drivers/pci/pcie/dpc.c
> @@ -233,7 +233,7 @@ static irqreturn_t dpc_handler(int irq, void *context)
>   	dpc_process_error(pdev);
>   
>   	/* We configure DPC so it only triggers on ERR_FATAL */
> -	pcie_do_recovery(pdev, pci_channel_io_frozen, dpc_reset_link);
> +	pcie_do_fatal_recovery(pdev, dpc_reset_link);
>   
>   	return IRQ_HANDLED;
>   }
> diff --git a/drivers/pci/pcie/edr.c b/drivers/pci/pcie/edr.c
> index a6b9b479b97ad0..87379bc566f691 100644
> --- a/drivers/pci/pcie/edr.c
> +++ b/drivers/pci/pcie/edr.c
> @@ -183,7 +183,7 @@ static void edr_handle_event(acpi_handle handle, u32 event, void *data)
>   	 * or ERR_NONFATAL, since the link is already down, use the FATAL
>   	 * error recovery path for both cases.
>   	 */
> -	estate = pcie_do_recovery(edev, pci_channel_io_frozen, dpc_reset_link);
> +	estate = pcie_do_fatal_recovery(edev, dpc_reset_link);
>   
>   send_ost:
>   
> diff --git a/drivers/pci/pcie/err.c b/drivers/pci/pcie/err.c
> index c2ae4d08801a4d..11fcff16b17303 100644
> --- a/drivers/pci/pcie/err.c
> +++ b/drivers/pci/pcie/err.c
> @@ -141,7 +141,7 @@ static int report_resume(struct pci_dev *dev, void *data)
>   	return 0;
>   }
>   
> -static pci_ers_result_t pcie_do_fatal_recovery(struct pci_dev *dev,
> +pci_ers_result_t pcie_do_fatal_recovery(struct pci_dev *dev,
>   			pci_ers_result_t (*reset_link)(struct pci_dev *pdev))
>   {
>   	struct pci_dev *udev;
> @@ -194,15 +194,11 @@ static pci_ers_result_t pcie_do_fatal_recovery(struct pci_dev *dev,
>   }
>   
>   pci_ers_result_t pcie_do_recovery(struct pci_dev *dev,
> -			pci_channel_state_t state,
>   			pci_ers_result_t (*reset_link)(struct pci_dev *pdev))
>   {
>   	pci_ers_result_t status = PCI_ERS_RESULT_CAN_RECOVER;
>   	struct pci_bus *bus;
>   
> -	if (state == pci_channel_io_frozen)
> -		return pcie_do_fatal_recovery(dev, reset_link);
> -
>   	/*
>   	 * Error recovery runs on all subordinates of the first downstream port.
>   	 * If the downstream port detected the error, it is cleared at the end.

--
Sathyanarayanan Kuppuswamy
Linux Kernel Developer