On 01/20/16 at 08:03am, Mark Hounschell wrote: > On 01/20/2016 07:31 AM, Joerg Roedel wrote: > >Hi Baoquan, > > > >On Wed, Jan 13, 2016 at 11:09:50AM +0800, Baoquan He wrote: > >>Seems 0f:00.0 is not attached to a group so that it get a domain. Can > >>you also paste your "lspci -t" and "lspci -vvv" output? Here 0f:00.0 > >>should be a pci bridge which need set the same domain with its > >>peripheral 0000:0f:04.0. > >> > >>Please apply this patch and try again. > > > >Thanks for the fix, mind submitting it so that I can queue it and send > >it upstream after the merge window closes? > > > > > > > > Joerg > > I suspect the fix is just not complete yet as my box is still hosed.
No, the io_page_fault is caused by the not correct alias setting so that the scsi storage controller like 0f:04.0/0a:0a.0 can't be handled correclty. You can see that their alias are 0f:00.0 and 09:00.0 which are reported in io_page_fualt. But the other 2 disks still didn't come up, I suspect if the irq is not set right. This should be another different issue, no any idea yet. Thanks Baoquan _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
