x86,kvm,vmx: Preserve CR4 across VM entry
authorAndy Lutomirski <luto@amacapital.net>
Wed, 8 Oct 2014 16:02:13 +0000 (09:02 -0700)
committerBen Hutchings <ben@decadent.org.uk>
Wed, 5 Nov 2014 20:27:48 +0000 (20:27 +0000)
commit9e793c5ed9204271ecc2cb7c899010e70561a452
treecedfe5d3c6d03d5de9c561b02749246cadef416c
parent3a8c709ba4cf6fe86f5069c71325029d412bcf1e
x86,kvm,vmx: Preserve CR4 across VM entry

commit d974baa398f34393db76be45f7d4d04fbdbb4a0a upstream.

CR4 isn't constant; at least the TSD and PCE bits can vary.

TBH, treating CR0 and CR3 as constant scares me a bit, too, but it looks
like it's correct.

This adds a branch and a read from cr4 to each vm entry.  Because it is
extremely likely that consecutive entries into the same vcpu will have
the same host cr4 value, this fixes up the vmcs instead of restoring cr4
after the fact.  A subsequent patch will add a kernel-wide cr4 shadow,
reducing the overhead in the common case to just two memory reads and a
branch.

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Cc: stable@vger.kernel.org
Cc: Petr Matousek <pmatouse@redhat.com>
Cc: Gleb Natapov <gleb@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
[bwh: Backported to 3.2:
 - Adjust context
 - Add struct vcpu_vmx *vmx parameter to vmx_set_constant_host_state(), done
   upstream in commit a547c6db4d2f ("KVM: VMX: Enable acknowledge interupt
   on vmexit")]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
arch/x86/kvm/vmx.c