KVM: arm64: Move the handling of !FP outside of the fast path

We currently start by assuming that the host owns the FP unit
at load time, then check again whether this is the case as
we are about to run. Only at this point do we account for the
fact that there is a (vanishingly small) chance that we're running
on a system without a FPSIMD unit (yes, this is madness).

We can actually move this FPSIMD check as early as load-time,
and drop the check at run time.

No intended change in behaviour.

Suggested-by: Reiji Watanabe <[email protected]>
Reviewed-by: Reiji Watanabe <[email protected]>
Signed-off-by: Marc Zyngier <[email protected]>

Author: Marc Zyngier

arch/arm64/kvm/arm.c

@@ -328,6 +328,12 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
 
 	vcpu->arch.mmu_page_cache.gfp_zero = __GFP_ZERO;
 
+	/*
+	 * Default value for the FP state, will be overloaded at load
+	 * time if we support FP (pretty likely)
+	 */
+	vcpu->arch.fp_state = FP_STATE_FREE;
+
 	/* Set up the timer */
 	kvm_timer_vcpu_init(vcpu);
 

arch/arm64/kvm/fpsimd.c

@@ -77,6 +77,9 @@ void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu)
 	BUG_ON(!current->mm);
 	BUG_ON(test_thread_flag(TIF_SVE));
 
+	if (!system_supports_fpsimd())
+		return;
+
 	vcpu->arch.fp_state = FP_STATE_HOST_OWNED;
 
 	vcpu_clear_flag(vcpu, HOST_SVE_ENABLED);
@@ -110,13 +113,10 @@ void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu)
  * FP while we were preemptible (such as off the back of an interrupt),
  * then neither the host nor the guest own the FP hardware (and it was the
  * responsibility of the code that used FP to save the existing state).
- *
- * Note that not supporting FP is basically the same thing as far as the
- * hypervisor is concerned (nothing to save).
  */
 void kvm_arch_vcpu_ctxflush_fp(struct kvm_vcpu *vcpu)
 {
-	if (!system_supports_fpsimd() || test_thread_flag(TIF_FOREIGN_FPSTATE))
+	if (test_thread_flag(TIF_FOREIGN_FPSTATE))
 		vcpu->arch.fp_state = FP_STATE_FREE;
 }