On Mon, Nov 07, 2016 at 09:52:35AM +0100, Andrew Jones wrote: > On Mon, Nov 07, 2016 at 08:56:39AM +0100, Alexander Gordeev wrote: > > + > > + cmd = PCI_COMMAND_SERR | PCI_COMMAND_PARITY; > > + > > I don't think the parity error response bit is necessary for > virtual (qemu emulated) devices. I'd also rather minimize > the amount of setup we do for devices here, leaving most of > the setup to the unit tests that want to drive the devices. > IOW, I wouldn't bother with the parity thing, and, IMO, if > we can't allocate resources here, for whatever reason, then > we shouldn't touch the device at all; we should just leave > it to the unit tests. > OK, you've convinced me (off-list) that we might as well keep this. It doesn't hurt and, anyway, rolling things back takes even more code. Let's just leave it alone. You'll send v11 with the writeb/w updates though. Thanks, drew -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
