KVM: arm64: nvhe: Save the SPE context early

The nVHE KVM hyp drains and disables the SPE buffer, before
entering the guest, as the EL1&0 translation regime
is going to be loaded with that of the guest.

But this operation is performed way too late, because :
  - The owning translation regime of the SPE buffer
    is transferred to EL2. (MDCR_EL2_E2PB == 0)
  - The guest Stage1 is loaded.

Thus the flush could use the host EL1 virtual address,
but use the EL2 translations instead of host EL1, for writing
out any cached data.

Fix this by moving the SPE buffer handling early enough.
The restore path is doing the right thing.

Fixes: 014c4c77 ("KVM: arm64: Improve debug register save/restore flow")
Cc: stable@vger.kernel.org
Cc: Christoffer Dall <christoffer.dall@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Alexandru Elisei <alexandru.elisei@arm.com>
Reviewed-by: Alexandru Elisei <alexandru.elisei@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210302120345.3102874-1-suzuki.poulose@arm.com
Message-Id: <20210305185254.3730990-2-maz@kernel.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

arch/arm64/include/asm/kvm_hyp.h

@@ -83,6 +83,11 @@ void sysreg_restore_guest_state_vhe(struct kvm_cpu_context *ctxt);
 void __debug_switch_to_guest(struct kvm_vcpu *vcpu);
 void __debug_switch_to_host(struct kvm_vcpu *vcpu);
 
+#ifdef __KVM_NVHE_HYPERVISOR__
+void __debug_save_host_buffers_nvhe(struct kvm_vcpu *vcpu);
+void __debug_restore_host_buffers_nvhe(struct kvm_vcpu *vcpu);
+#endif
+
 void __fpsimd_save_state(struct user_fpsimd_state *fp_regs);
 void __fpsimd_restore_state(struct user_fpsimd_state *fp_regs);
 

arch/arm64/kvm/hyp/nvhe/debug-sr.c

@@ -58,16 +58,24 @@ static void __debug_restore_spe(u64 pmscr_el1)
 	write_sysreg_s(pmscr_el1, SYS_PMSCR_EL1);
 }
 
-void __debug_switch_to_guest(struct kvm_vcpu *vcpu)
+void __debug_save_host_buffers_nvhe(struct kvm_vcpu *vcpu)
 {
 	/* Disable and flush SPE data generation */
 	__debug_save_spe(&vcpu->arch.host_debug_state.pmscr_el1);
+}
+
+void __debug_switch_to_guest(struct kvm_vcpu *vcpu)
+{
 	__debug_switch_to_guest_common(vcpu);
 }
 
-void __debug_switch_to_host(struct kvm_vcpu *vcpu)
+void __debug_restore_host_buffers_nvhe(struct kvm_vcpu *vcpu)
 {
 	__debug_restore_spe(vcpu->arch.host_debug_state.pmscr_el1);
+}
+
+void __debug_switch_to_host(struct kvm_vcpu *vcpu)
+{
 	__debug_switch_to_host_common(vcpu);
 }
 

arch/arm64/kvm/hyp/nvhe/switch.c

@@ -192,6 +192,14 @@ int __kvm_vcpu_run(struct kvm_vcpu *vcpu)
 	pmu_switch_needed = __pmu_switch_to_guest(host_ctxt);
 
 	__sysreg_save_state_nvhe(host_ctxt);
+	/*
+	 * We must flush and disable the SPE buffer for nVHE, as
+	 * the translation regime(EL1&0) is going to be loaded with
+	 * that of the guest. And we must do this before we change the
+	 * translation regime to EL2 (via MDCR_EL2_E2PB == 0) and
+	 * before we load guest Stage1.
+	 */
+	__debug_save_host_buffers_nvhe(vcpu);
 
 	__adjust_pc(vcpu);
 
@@ -234,11 +242,12 @@ int __kvm_vcpu_run(struct kvm_vcpu *vcpu)
 	if (vcpu->arch.flags & KVM_ARM64_FP_ENABLED)
 		__fpsimd_save_fpexc32(vcpu);
 
+	__debug_switch_to_host(vcpu);
 	/*
 	 * This must come after restoring the host sysregs, since a non-VHE
 	 * system may enable SPE here and make use of the TTBRs.
 	 */
-	__debug_switch_to_host(vcpu);
+	__debug_restore_host_buffers_nvhe(vcpu);
 
 	if (pmu_switch_needed)
 		__pmu_switch_to_host(host_ctxt);