Re: [Qemu-devel] [PULL 3/6] vfio-pci: Lazy PCI option ROM loading

2013-10-04 Thread Alex Williamson
On Fri, 2013-10-04 at 22:13 +1000, Alexey Kardashevskiy wrote: > On 10/04/2013 01:39 AM, Alex Williamson wrote: > > During vfio-pci initfn, the device is not always in a state where the > > option ROM can be read. In the case of graphics cards, there's often > > no per function reset, which means

Re: [Qemu-devel] [PULL 3/6] vfio-pci: Lazy PCI option ROM loading

2013-10-04 Thread Alexey Kardashevskiy
On 10/04/2013 01:39 AM, Alex Williamson wrote: > During vfio-pci initfn, the device is not always in a state where the > option ROM can be read. In the case of graphics cards, there's often > no per function reset, which means we have host driver state affecting > whether the option ROM is usable.

Re: [Qemu-devel] [PULL 3/6] vfio-pci: Lazy PCI option ROM loading

2013-10-03 Thread Alex Williamson
On Thu, 2013-10-03 at 20:46 +0200, Paolo Bonzini wrote: > Il 03/10/2013 17:39, Alex Williamson ha scritto: > > +static const MemoryRegionOps vfio_rom_ops = { > > +.read = vfio_rom_read, > > +.endianness = DEVICE_LITTLE_ENDIAN, > > +}; > > + > > I think you need to define a write callback t

Re: [Qemu-devel] [PULL 3/6] vfio-pci: Lazy PCI option ROM loading

2013-10-03 Thread Paolo Bonzini
Il 03/10/2013 17:39, Alex Williamson ha scritto: > +static const MemoryRegionOps vfio_rom_ops = { > +.read = vfio_rom_read, > +.endianness = DEVICE_LITTLE_ENDIAN, > +}; > + I think you need to define a write callback too (unless you're sure for some other reason that the area will never be