The series of patches fix couple of issues as below:

PATCH[1]: Don't enable EEH functionality on the emulated PCI devices,
          which shouldn't be affected during EEH recovery
PATCH[2]: Clear PE's frozen state on reboot. Some PEs in the guest
          might be put into offline because excessive EEH errors and
          those PEs are still in frozen state. Users want them to work
          again after reboot
PATCH[3]: Flush stable MSIx vectors during PE reset.

Changelog:
=========
v3 -> v4:
   * Added PATCH[1]
   * Separate function to reenable EEH functionality in PATCH[2]
   * Move the whole logic into sPAPR platform in PATCH[3]

Gavin Shan (3):
  sPAPR: Don't enable EEH on emulated PCI devices
  sPAPR: Reenable EEH functionality on reboot
  sPAPR: Clear stale MSIx table during EEH reset

 hw/ppc/spapr_pci.c      |  7 ++++++
 hw/ppc/spapr_pci_vfio.c | 65 ++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 71 insertions(+), 1 deletion(-)

-- 
2.1.0


Reply via email to