Patchwork [3/3] KVM: PPC: BookE: Support FPU on non-hv systems

login
register
mail settings
Submitter Alexander Graf
Date Aug. 16, 2012, 11:17 a.m.
Message ID <1345115854-3816-4-git-send-email-agraf@suse.de>
Download mbox | patch
Permalink /patch/177954/
State New
Headers show

Comments

Alexander Graf - Aug. 16, 2012, 11:17 a.m.
When running on HV aware hosts, we can not trap when the guest sets the FP
bit, so we just let it do so when it wants to, because it has full access to
MSR.

For non-HV aware hosts with an FPU (like 440), we need to also adjust the
shadow MSR though. Otherwise the guest gets an FP unavailable trap even when
it really enabled the FP bit in MSR.

Signed-off-by: Alexander Graf <agraf@suse.de>
---
 arch/powerpc/kvm/booke.c |   11 +++++++++++
 1 files changed, 11 insertions(+), 0 deletions(-)

Patch

diff --git a/arch/powerpc/kvm/booke.c b/arch/powerpc/kvm/booke.c
index 959aae9..5f0476a 100644
--- a/arch/powerpc/kvm/booke.c
+++ b/arch/powerpc/kvm/booke.c
@@ -122,6 +122,16 @@  static void kvmppc_vcpu_sync_spe(struct kvm_vcpu *vcpu)
 }
 #endif
 
+static void kvmppc_vcpu_sync_fpu(struct kvm_vcpu *vcpu)
+{
+#if defined(CONFIG_PPC_FPU) && !defined(CONFIG_KVM_BOOKE_HV)
+	/* We always treat the FP bit as enabled from the host
+	   perspective, so only need to adjust the shadow MSR */
+	vcpu->arch.shadow_msr &= ~MSR_FP;
+	vcpu->arch.shadow_msr |= vcpu->arch.shared->msr & MSR_FP;
+#endif
+}
+
 /*
  * Helper function for "full" MSR writes.  No need to call this if only
  * EE/CE/ME/DE/RI are changing.
@@ -138,6 +148,7 @@  void kvmppc_set_msr(struct kvm_vcpu *vcpu, u32 new_msr)
 
 	kvmppc_mmu_msr_notify(vcpu, old_msr);
 	kvmppc_vcpu_sync_spe(vcpu);
+	kvmppc_vcpu_sync_fpu(vcpu);
 }
 
 static void kvmppc_booke_queue_irqprio(struct kvm_vcpu *vcpu,