On 2 January 2017 at 20:01, Andrew Jones <drjo...@redhat.com> wrote: > v2: > - all Igor's suggestions > - added four mach-virt acpi cleanup patches > - added the remaining two acpi patches needed to bring equivalence > to Peter's DT changes > > This series is based on Peter's qemu-arm gicv3-virt branch. It's > main goal (patches 07-13), which was suggested by Eduardo, is to > remove an unnecessary structure, VirtGuestInfo, which is a > maintenance burden, as it requires duplicating Virt machine state. > Additionally patches 05-06 do some mach-virt cleanups and patches > 14-16 add the ACPI equivalents of the DT patches in Peter's > gicv3-virt branch. The first four patches are cleanups to mach-virt's > acpi code. > > Patches available here > https://github.com/rhdrjones/qemu/commits/virt/remove-guest-info-v2
Thanks. Apart from the last 2 patches which depend on the gicv3-virt changes, I think this is all OK to go in now. Since the gicv3-virt code still needs a respin (I haven't looked at the failure to boot smp secondaries into EL2 yet), I have taken the first 3 patches from gicv3-virt and put them into target-arm.next: hw/arm/virt: Merge VirtBoardInfo and VirtMachineState hw/arm/virt: Rename 'vbi' variables to 'vms' hw/arm/virt: Don't incorrectly claim architectural timer to be edge-triggered and then rebased patches 1-14 from this series on top of that, and applied all that to target-arm.next. I'll put patches 15 and 16 into my gicv3-virt branch for when I send out v2 of that. thanks -- PMM