On 10/29/19 4:03 PM, Bjorn Helgaas wrote:
On Tue, Oct 29, 2019 at 12:58:14PM -0700, Kuppuswamy Sathyanarayanan wrote:
On 10/28/19 4:22 PM, Bjorn Helgaas wrote:
On Thu, Oct 03, 2019 at 04:39:01PM -0700, sathyanarayanan.kuppuswamy@xxxxxxxxxxxxxxx wrote:
From: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@xxxxxxxxxxxxxxx>
@@ -430,9 +424,6 @@ int pci_cleanup_aer_error_status_regs(struct pci_dev *dev)
if (!pos)
return -EIO;
- if (pcie_aer_get_firmware_first(dev))
- return -EIO;
-
port_type = pci_pcie_type(dev);
if (port_type == PCI_EXP_TYPE_ROOT_PORT) {
pci_read_config_dword(dev, pos + PCI_ERR_ROOT_STATUS, &status);
@@ -455,7 +446,8 @@ void pci_aer_init(struct pci_dev *dev)
if (dev->aer_cap)
dev->aer_stats = kzalloc(sizeof(struct aer_stats), GFP_KERNEL);
- pci_cleanup_aer_error_status_regs(dev);
+ if (!pcie_aer_get_firmware_first(dev))
+ pci_cleanup_aer_error_status_regs(dev);
This effectively moves the "if (pcie_aer_get_firmware_first())" check
from pci_cleanup_aer_error_status_regs() into one of the callers. But
there are two other callers: pci_aer_init() and pci_restore_state().
Do they need the change, or do you want to cleanup the AER error
registers there, but not here?
Good catch. I have added this check to pci_aer_init(). But it needs
to be added to pci_restore_state() as well. Instead of moving the
checks to the caller, If you agree, I could change the API to
pci_cleanup_aer_error_status_regs(struct pci_dev *dev, bool
skip_ff_check) and let the caller decide whether they want skip the
check or not.
If all callers of pci_cleanup_aer_error_status_regs() would have to
check pcie_aer_get_firmware_first(), I don't understand why you're
moving the check at all.
We need exception for the call made from DPC driver. If
pcie_aer_get_firmware_first()
call is made from DPC driver during EDR mode execution, then FF mode
check needs
to be skipped.
--
Sathyanarayanan Kuppuswamy
Linux kernel developer
