arm: KVM: force execution of HCPTR access on VM exit

commit 85e84ba3 upstream.

On VM entry, we disable access to the VFP registers in order to
perform a lazy save/restore of these registers.

On VM exit, we restore access, test if we did enable them before,
and save/restore the guest/host registers if necessary. In this
sequence, the FPEXC register is always accessed, irrespective
of the trapping configuration.

If the guest didn't touch the VFP registers, then the HCPTR access
has now enabled such access, but we're missing a barrier to ensure
architectural execution of the new HCPTR configuration. If the HCPTR
access has been delayed/reordered, the subsequent access to FPEXC
will cause a trap, which we aren't prepared to handle at all.

The same condition exists when trapping to enable VFP for the guest.

The fix is to introduce a barrier after enabling VFP access. In the
vmexit case, it can be relaxed to only takes place if the guest hasn't
accessed its view of the VFP registers, making the access to FPEXC safe.

The set_hcptr macro is modified to deal with both vmenter/vmexit and
vmtrap operations, and now takes an optional label that is branched to
when the guest hasn't touched the VFP registers.
Reported-by: Vikram Sethi <vikrams@codeaurora.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Luis Henriques <luis.henriques@canonical.com>

arch/arm/kvm/interrupts.S

@@ -159,13 +159,9 @@ __kvm_vcpu_return:
 	@ Don't trap coprocessor accesses for host kernel
 	set_hstr vmexit
 	set_hdcr vmexit
-	set_hcptr vmexit, (HCPTR_TTA | HCPTR_TCP(10) | HCPTR_TCP(11))
+	set_hcptr vmexit, (HCPTR_TTA | HCPTR_TCP(10) | HCPTR_TCP(11)), after_vfp_restore
 
 #ifdef CONFIG_VFPv3
-	@ Save floating point registers we if let guest use them.
-	tst	r2, #(HCPTR_TCP(10) | HCPTR_TCP(11))
-	bne	after_vfp_restore
-
 	@ Switch VFP/NEON hardware state to the host's
 	add	r7, vcpu, #VCPU_VFP_GUEST
 	store_vfp_state r7
@@ -177,6 +173,8 @@ after_vfp_restore:
 	@ Restore FPEXC_EN which we clobbered on entry
 	pop	{r2}
 	VFPFMXR FPEXC, r2
+#else
+after_vfp_restore:
 #endif
 
 	@ Reset Hyp-role
@@ -467,7 +465,7 @@ switch_to_guest_vfp:
 	push	{r3-r7}
 
 	@ NEON/VFP used.  Turn on VFP access.
-	set_hcptr vmexit, (HCPTR_TCP(10) | HCPTR_TCP(11))
+	set_hcptr vmtrap, (HCPTR_TCP(10) | HCPTR_TCP(11))
 
 	@ Switch VFP/NEON hardware state to the guest's
 	add	r7, r0, #VCPU_VFP_HOST

arch/arm/kvm/interrupts_head.S

@@ -578,8 +578,13 @@ vcpu	.req	r0		@ vcpu pointer always in r0
 .endm
 
 /* Configures the HCPTR (Hyp Coprocessor Trap Register) on entry/return
- * (hardware reset value is 0). Keep previous value in r2. */
-.macro set_hcptr operation, mask
+ * (hardware reset value is 0). Keep previous value in r2.
+ * An ISB is emited on vmexit/vmtrap, but executed on vmexit only if
+ * VFP wasn't already enabled (always executed on vmtrap).
+ * If a label is specified with vmexit, it is branched to if VFP wasn't
+ * enabled.
+ */
+.macro set_hcptr operation, mask, label = none
 	mrc	p15, 4, r2, c1, c1, 2
 	ldr	r3, =\mask
 	.if \operation == vmentry
@@ -588,6 +593,17 @@ vcpu	.req	r0		@ vcpu pointer always in r0
 	bic	r3, r2, r3		@ Don't trap defined coproc-accesses
 	.endif
 	mcr	p15, 4, r3, c1, c1, 2
+	.if \operation != vmentry
+	.if \operation == vmexit
+	tst	r2, #(HCPTR_TCP(10) | HCPTR_TCP(11))
+	beq	1f
+	.endif
+	isb
+	.if \label != none
+	b	\label
+	.endif
+1:
+	.endif
 .endm
 
 /* Configures the HDCR (Hyp Debug Configuration Register) on entry/return