On Wed, Feb 21, 2018 at 07:18:32PM +0100, Andrew Jones wrote:

On Wed, Feb 21, 2018 at 06:43:00PM +0100, Andrew Jones wrote:

quoted

On Thu, Feb 15, 2018 at 10:03:05PM +0100, Christoffer Dall wrote:

quoted

So far this is mostly (see below) a copy of the legacy non-VHE switch
function, but we will start reworking these functions in separate
directions to work on VHE and non-VHE in the most optimal way in later
patches.

The only difference after this patch between the VHE and non-VHE run
functions is that we omit the branch-predictor variant-2 hardening for
QC Falkor CPUs, because this workaround is specific to a series of
non-VHE ARMv8.0 CPUs.

Reviewed-by: Marc Zyngier <redacted>
Signed-off-by: Christoffer Dall <redacted>
---

Notes:
    Changes since v3:
     - Added BUG() to 32-bit ARM VHE run function
     - Omitted QC Falkor BP Hardening functionality from VHE-specific
       function
    
    Changes since v2:
     - Reworded commit message
    
    Changes since v1:
     - Rename kvm_vcpu_run to kvm_vcpu_run_vhe and rename __kvm_vcpu_run to
       __kvm_vcpu_run_nvhe
     - Removed stray whitespace line

 arch/arm/include/asm/kvm_asm.h   |  5 ++-
 arch/arm/kvm/hyp/switch.c        |  2 +-
 arch/arm64/include/asm/kvm_asm.h |  4 ++-
 arch/arm64/kvm/hyp/switch.c      | 66 +++++++++++++++++++++++++++++++++++++++-
 virt/kvm/arm/arm.c               |  5 ++-
 5 files changed, 77 insertions(+), 5 deletions(-)

...

quoted

diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 2062d9357971..5bd879c78951 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c

@@ -736,7 +736,10 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
 		if (has_vhe())
 			kvm_arm_vhe_guest_enter();
 
-		ret = kvm_call_hyp(__kvm_vcpu_run, vcpu);
+		if (has_vhe())
+			ret = kvm_vcpu_run_vhe(vcpu);
+		else
+			ret = kvm_call_hyp(__kvm_vcpu_run_nvhe, vcpu);
 
 		if (has_vhe())
 			kvm_arm_vhe_guest_exit();

We can combine these has_vhe()'s

 if (has_vhe()) {
    kvm_arm_vhe_guest_enter();
    ret = kvm_vcpu_run_vhe(vcpu);
    kvm_arm_vhe_guest_exit();
 } else
    ret = kvm_call_hyp(__kvm_vcpu_run_nvhe, vcpu);

Maybe even do a cleanup patch that removes
kvm_arm_vhe_guest_enter/exit by putting the daif
masking/restoring directly into kvm_vcpu_run_vhe()?

Yes, indeed.  This is a blind rebasing result on my part.

Thanks,
-Christoffer