Re: [RFC v5 6/7] PCI: Add a new bit to pci_bus_flags to indicate interrupt remapping

2016-04-05 Thread Yongji Xie
On 2016/4/6 8:11, Gavin Shan wrote: On Tue, Apr 05, 2016 at 09:46:43PM +0800, Yongji Xie wrote: I'm trying to find a proper way to indicate the capability of interrupt remapping on PPC64 because we need this to determine whether it is safe to mmap MSI-X table in VFIO driver. There is a existing

Re: [RFC v5 6/7] PCI: Add a new bit to pci_bus_flags to indicate interrupt remapping

2016-04-05 Thread Gavin Shan
On Tue, Apr 05, 2016 at 09:46:43PM +0800, Yongji Xie wrote: >I'm trying to find a proper way to indicate >the capability of interrupt remapping on PPC64 >because we need this to determine whether it is >safe to mmap MSI-X table in VFIO driver. > >There is a existing flag for this in the IOMMU >spac

[RFC v5 6/7] PCI: Add a new bit to pci_bus_flags to indicate interrupt remapping

2016-04-05 Thread Yongji Xie
I'm trying to find a proper way to indicate the capability of interrupt remapping on PPC64 because we need this to determine whether it is safe to mmap MSI-X table in VFIO driver. There is a existing flag for this in the IOMMU space: enum iommu_cap { IOMMU_CAP_CACHE_COHERENCY, --->IOM