[GIT PULL] Please pull my kvm-ppc-next-4.16-2 tag

[GIT PULL] Please pull my kvm-ppc-next-4.16-2 tag

[Paul Mackerras]

Paolo or Radim,

Please do a pull from my kvm-ppc-next-4.16-2 tag to get some PPC KVM
fixes that I would like to have go into 4.16.  Most of them are quite
self-contained; there are two that are larger (and arguably don't
quite count as bug fixes), but I have checked them carefully and
convinced myself that they have a low risk of causing regressions, and
they do address issues that have been found in testing, so I put them
in.

With things like this I would normally put them in my kvm-ppc-fixes
branch and base it on current upstream.  However, as far as I can see
the kvm next branch hasn't been merged upstream yet.  Therefore I made
this series a continuation of my kvm-ppc-next branch.

When do you expect that kvm/next will get merged upstream?

Thanks,
Paul.

The following changes since commit 9b9b13a6d1537ddc4caccd6f1c41b78edbc08437:

  KVM: PPC: Book3S HV: Keep XIVE escalation interrupt masked unless ceded (2018-01-19 12:10:21 +1100)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc tags/kvm-ppc-next-4.16-2

for you to fetch changes up to 09f984961c137c4b252c368adab7e1c9f035fa59:

  KVM: PPC: Book3S: Add MMIO emulation for VMX instructions (2018-02-09 16:51:51 +1100)

----------------------------------------------------------------
Second PPC KVM update for 4.16

Seven fixes that are either trivial or that address bugs that people
are actually hitting.  The main ones are:

- Drop spinlocks before reading guest memory

- Fix a bug causing corruption of VCPU state in PR KVM with preemption
  enabled

- Make HPT resizing work on POWER9

- Add MMIO emulation for vector loads and stores, because guests now
  use these instructions in memcpy and similar routines.

----------------------------------------------------------------
Alexander Graf (2):
      KVM: PPC: Book3S PR: Fix svcpu copying with preemption enabled
      KVM: PPC: Book3S HV: Branch inside feature section

David Gibson (1):
      KVM: PPC: Book3S HV: Make HPT resizing work on POWER9

Jose Ricardo Ziviani (1):
      KVM: PPC: Book3S: Add MMIO emulation for VMX instructions

Paul Mackerras (2):
      KVM: PPC: Book3S HV: Drop locks before reading guest memory
      KVM: PPC: Book3S HV: Fix handling of secondary HPTEG in HPT resizing code

Ulf Magnusson (1):
      KVM: PPC: Book3S PR: Fix broken select due to misspelling

 arch/powerpc/include/asm/kvm_book3s.h   |   6 +-
 arch/powerpc/include/asm/kvm_host.h     |   2 +
 arch/powerpc/include/asm/kvm_ppc.h      |   4 +
 arch/powerpc/include/asm/ppc-opcode.h   |   6 ++
 arch/powerpc/kvm/Kconfig                |   2 +-
 arch/powerpc/kvm/book3s_64_mmu_hv.c     |  38 +++++---
 arch/powerpc/kvm/book3s_hv.c            |  16 +++-
 arch/powerpc/kvm/book3s_hv_rmhandlers.S |   3 +-
 arch/powerpc/kvm/book3s_interrupts.S    |   4 +-
 arch/powerpc/kvm/book3s_pr.c            |  20 ++---
 arch/powerpc/kvm/emulate_loadstore.c    |  36 ++++++++
 arch/powerpc/kvm/powerpc.c              | 153 +++++++++++++++++++++++++++++++-
 12 files changed, 251 insertions(+), 39 deletions(-)

Re: [GIT PULL] Please pull my kvm-ppc-next-4.16-2 tag

[Radim Krčmář]

2018-02-09 20:41+1100, Paul Mackerras:
> Paolo or Radim,
> 
> Please do a pull from my kvm-ppc-next-4.16-2 tag to get some PPC KVM
> fixes that I would like to have go into 4.16.  Most of them are quite
> self-contained; there are two that are larger (and arguably don't
> quite count as bug fixes), but I have checked them carefully and
> convinced myself that they have a low risk of causing regressions, and
> they do address issues that have been found in testing, so I put them
> in.
> 
> With things like this I would normally put them in my kvm-ppc-fixes
> branch and base it on current upstream.  However, as far as I can see
> the kvm next branch hasn't been merged upstream yet.  Therefore I made
> this series a continuation of my kvm-ppc-next branch.

I have resolved the resulting conflict and pulled, thanks.

> When do you expect that kvm/next will get merged upstream?

Hopefully on Sunday, sorry for the delay.