Re: [GIT PULL] KVM changes for 4.8 merge window

[Paolo Bonzini <pbonzini@redhat.com>]

On 03/08/2016 05:21, Michael Ellerman wrote:
> Paolo Bonzini <pbonzini@redhat.com> writes:
> ...
>> - arch/powerpc: what a mess.  For the idle_book3s.S conflict, the KVM
>> tree is the right one; everything else is trivial.  In this case I am
>> not quite sure what went wrong.  The commit that is causing the mess
>> (fd7bacbca47a, "KVM: PPC: Book3S HV: Fix TB corruption in guest exit
>> path on HMI interrupt", 2016-05-15) touches both arch/powerpc/kernel/
>> and arch/powerpc/kvm/.  It's large, but at 396 insertions/5 deletions
>> I guessed that it wasn't really possible to split it and that the 5
>> deletions wouldn't conflict.  That wasn't the case.
> 
> In fact I think the problem is that this patch shouldn't have gone via the KVM
> tree at all.
> 
> If you look at the diffstat, it doesn't touch anything in generic KVM, but lots
> of arch code:

The KVM tree merges all arch/*/kvm code from submaintainers.  Only Radim
and I send patches directly to Linus.

Considering the h in "hmi" is for hypervisor, actual non-virt code in
that patch was this:

  arch/powerpc/include/asm/paca.h         |   6 +++
  arch/powerpc/kernel/Makefile            |   2 +-
  arch/powerpc/kernel/exceptions-64s.S    |   4 +-
  arch/powerpc/kernel/idle_power7.S       |   5 ++-
  arch/powerpc/kernel/traps.c             |   5 +++

So the changes are pretty small, yet apart from paca.h every file ended
up having a conflict with the PPC tree.  So I think it's just very bad
luck in this case.  Having this patch in a topic branch merged by both
PPC and KVM maintainers would have still been a good idea, because I
guess Paul knew of Ben's idle_power7.S cleanup.

Paolo