On Wed, Jan 22, 2014 at 11:29:50AM -0500, Anthony DeRobertis wrote: > On 01/20/2014 07:26 AM, Neil Horman wrote: > >Right, the X58 variant of the chipset should be covered by > >803075dba31c17af110e1d9a915fe7262165b213. And it is, because we still get > >the > >remapping warning when you boot up. Whats odd is that you're still seeing > >the > >issue, so there must be something else about your chipset that acting > >differently than other X58's > > Indeed. X58 isn't exactly a weird chipset, I can't be the only > person with it... > You're by no means the only person with it, but you are the only person that I know of thats complaining of the non-draining irq issue after the quirk disables interrupt remapping, which leads me to think that there is something about your x58 chipset that differs from everyone elses.
> I've gone back to booting with my patch (the one which > removes/disables the quirk). and it continues to work fine. As long > as I ignore Intel's erratum. > which makes no sense, save for the possibility that you have encountered a different problem that exhibits the same symptom as the irq remapping issue. > I'm not sure where to look next. Going to throw some random printk's > in, but I stress random—I'm not familiar with the code or the > hardware. Suggestions of what to try next appreciated. > _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
