Hi Bjorn, On 02/22/20 at 10:56am, Bjorn Helgaas wrote: > [+cc Khalid, Deepa, Randy, Dave, Myron] > > On Thu, Dec 26, 2019 at 03:21:18AM +0800, Kairui Song wrote: > > There are reports about kdump hang upon reboot on some HPE machines, > > kernel hanged when trying to shutdown a PCIe port, an uncorrectable > > error occurred and crashed the system. > > Did we ever make progress on this? This definitely sounds like a > problem that needs to be fixed, but I don't see a resolution here. > I'm not familar with the PCI details, but if this only adds a quirk for kdump use and no other risks added then it should be good to have. Or we can provide a kernel parameter for the quirk? Then it is even limited to only be effective when in-kdump && the-param-is-used Anyway still prefer to people who know more about this area to evaluate the risk. Thanks Dave
