Re: [libvirt] [PATCH 2/4] qemu: Add architecture checks to qemuDomainMachineIsVirt()
On Thu, 2016-06-23 at 14:08 -0400, Laine Stump wrote:
On 06/23/2016 04:40 AM, Andrea Bolognani wrote:
> Remove all external architecture checks that have been
> made redundant by this change.
> ---
> src/qemu/qemu_capabilities.c | 12 +++++-------
> src/qemu/qemu_command.c | 4 +---
> src/qemu/qemu_domain.c | 12 +++++++++---
> 3 files changed, 15 insertions(+), 13 deletions(-)
>
> diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c
> index 4ed5b71..5fcd744 100644
> --- a/src/qemu/qemu_capabilities.c
> +++ b/src/qemu/qemu_capabilities.c
> @@ -2174,13 +2174,11 @@ bool virQEMUCapsHasPCIMultiBus(virQEMUCapsPtr qemuCaps,
> return false;
> }
>
> - if (ARCH_IS_ARM(def->os.arch)) {
> - /* If 'virt' supports PCI, it supports multibus.
> - * No extra conditions here for simplicity.
> - */
Just to be pedantic - here you're removing a check for ARMV6L or ARMV7L
or ARMV7B or AARCH64, and replacing it with a simpler check for ARMV7L
or AARCH64. But I guess the virt machinetype isn't available/possible on
those other two types?
As long as that's okay, ACK.
Yes, the virt machine type only supports ARMv7 and aarch64.
ARMv7, aarch64 and ppc64 can actually be used in either LE or BE mode
depending on the guest OS, so having separate PPC64LE/PPC64 and
ARMV7L/ARMV7B guest architectures is a questionable design decision:
the guest hardware is going to be exactly the same no matter what OS
you install on it. Note how, funnily enough, we don't have AARCH64BE.
Anyway, BE is basically unused on ARM: I haven't been able to find a
single BE guest OS for the architecture. And all other architecture
checks were already ignoring ARMV6L and ARMV7B, so it makes sense to
be consistent to avoid nasty surprises down the road.
--
Andrea Bolognani / Red Hat / Virtualization