Subject: Re: [GIT PULL] KVM updates for v4.9-rc1
To: =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        Linus Torvalds <torvalds@linux-foundation.org>
References: <20161006111646.GA19924@potion>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
        Marc Zyngier <marc.zyngier@arm.com>
From: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <aeb67611-3771-ff0d-6be7-2fddeba1e3da@redhat.com>
Date: Thu, 6 Oct 2016 13:34:37 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101
 Thunderbird/45.3.0
MIME-Version: 1.0
In-Reply-To: <20161006111646.GA19924@potion>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1056
Lines: 29


On 06/10/2016 13:16, Radim Krčmář wrote:
> Linus,
> 
> this merge needs one manual conflict resolution that is foretold in
> 3d9cd95f90b2 ("ARM: gic-v3: Work around definition of gic_write_bpr1").
> The commit says that the conflict is "obvious to resolve", but
> linux-next got it wrong, so I have pasted the correct resolution at the
> bottom -- inline assembly and surrounding #if should be removed, because
> the #if would always select the write_sysreg() branch.

This is nowhere as bad as 4.8, but I'd like to point out the right way to
do it: prepare a topic branch with the arch/arm/include/asm patches, and merge
it into both the irqchip and KVM-ARM trees.  You can find three examples in this
merge window alone:

Paolo Bonzini (12):
      Merge branch 'x86/amd-avic' of git://git.kernel.org/.../joro/iommu into HEAD

Paul Mackerras (8):
      Merge branch 'kvm-ppc-infrastructure' into kvm-ppc-next

Christian Borntraeger (1):
      Merge remote-tracking branch 'kvms390/s390forkvm' into kvms390next

So we can do better in 4.10. :)

Paolo