"Correctable error" or "Uncorrectable error" from PCIe AER is logged

Issue

• After installing a new PCI Express (PCIe) card or during normal system operation, error messages related to Advanced Error Reporting (AER) are recorded in the kernel log (dmesg or journalctl -k).

• Example of a Correctable Error:

  kernel: pcieport 0000:00:1b.2: AER: Correctable error received from 0000:03:04.0
  kernel: pcieport 0000:03:04.0: PCIe Bus Error: severity=Correctable, type=Physical Layer, (Receiver ID)
  kernel: pcieport 0000:03:04.0:   device [111d:8063] error status/mask=00000001/00002000
  kernel: pcieport 0000:03:04.0:    [ 0] RxErr
  

• In some cases, more severe Uncorrectable Errors are also logged:

  kernel: pcieport 0000:00:1b.2: AER: Uncorrectable (Fatal) error message received from 0000:03:04.0
  kernel: pcieport 0000:03:04.0: PCIe Bus Error: severity=Uncorrectable (Fatal), type=Transaction Layer, (Receiver ID)
  kernel: pcieport 0000:03:04.0:   device [111d:8063] error status/mask=00000020/00000000
  kernel: pcieport 0000:03:04.0:    [ 5] SDES                   (First)
  kernel: pci 0000:05:00.0: AER: can't recover (no error_detected callback)
  kernel: pcieport 0000:03:04.0: AER: Downstream Port link has been reset (0)
  kernel: pcieport 0000:03:04.0: AER: device recovery failed
  

• What do these messages mean, and should they be addressed with OS-level configuration changes?