powerpc/xive: Remove incorrect debug code

index a3b8d7d1316e..838ebdbfe4c5 100644
@@ -1269,11 +1269,6 @@ static void xive_setup_cpu(void)
 {
 	struct xive_cpu *xc = __this_cpu_read(xive_cpu);
 
-	/* Debug: Dump the TM state */
-	pr_devel("CPU %d [HW 0x%02x] VT=%02x\n",
-	    smp_processor_id(), hard_smp_processor_id(),
-	    in_8(xive_tima + xive_tima_offset + TM_WORD2));
-
 	/* The backend might have additional things to do */
 	if (xive_ops->setup_cpu)
 		xive_ops->setup_cpu(smp_processor_id(), xc);

index 8956a9856604..d3124c302146 100644
@@ -10,8 +10,6 @@
  *      2 of the License, or (at your option) any later version.
  */
 
-#define DEBUG
-
 #include <linux/export.h>
 #include <linux/string.h>
 #include <linux/sched.h>

index 041ddbd1fc57..2479e3396ed8 100644
@@ -127,7 +127,7 @@ int __init early_init_dt_scan_opal(unsigned long node,
 
 	if (of_flat_dt_is_compatible(node, "ibm,opal-v3")) {
 		powerpc_firmware_features |= FW_FEATURE_OPAL;
-		pr_info("OPAL detected !\n");
+		pr_debug("OPAL detected !\n");
 	} else {
 		panic("OPAL != V3 detected, no longer supported.\n");
 	}

index 0d750d274c4e..cd21c891be9e 100644
@@ -1795,6 +1795,7 @@ static int xive_debug_show(struct seq_file *m, void *private)
 
 	kvm_for_each_vcpu(i, vcpu, kvm) {
 		struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu;
+		unsigned int i;
 
 		if (!xc)
 			continue;

@@ -1804,6 +1805,33 @@ static int xive_debug_show(struct seq_file *m, void *private)
 			   xc->server_num, xc->cppr, xc->hw_cppr,
 			   xc->mfrr, xc->pending,
 			   xc->stat_rm_h_xirr, xc->stat_vm_h_xirr);
+		for (i = 0; i < KVMPPC_XIVE_Q_COUNT; i++) {
+			struct xive_q *q = &xc->queues[i];
+			u32 i0, i1, idx;
+
+			if (!q->qpage && !xc->esc_virq[i])
+				continue;
+
+			seq_printf(m, " [q%d]: ", i);
+
+			if (q->qpage) {
+				idx = q->idx;
+				i0 = be32_to_cpup(q->qpage + idx);
+				idx = (idx + 1) & q->msk;
+				i1 = be32_to_cpup(q->qpage + idx);
+				seq_printf(m, "T=%d %08x %08x... \n", q->toggle, i0, i1);
+			}
+			if (xc->esc_virq[i]) {
+				struct irq_data *d = irq_get_irq_data(xc->esc_virq[i]);
+				struct xive_irq_data *xd = irq_data_get_irq_handler_data(d);
+				u64 pq = xive_vm_esb_load(xd, XIVE_ESB_GET);
+				seq_printf(m, "E:%c%c I(%d:%llx:%llx)",
+					   (pq & XIVE_ESB_VAL_P) ? 'P' : 'p',
+					   (pq & XIVE_ESB_VAL_Q) ? 'Q' : 'q',
+					   xc->esc_virq[i], pq, xd->eoi_page);
+				seq_printf(m, "\n");
+			}
+		}
 
 		t_rm_h_xirr += xc->stat_rm_h_xirr;
 		t_rm_h_ipoll += xc->stat_rm_h_ipoll;

index 233c7504b1f2..fc926743647e 100644
@@ -1073,6 +1073,7 @@ enum {
 /* Flags for OPAL_XIVE_GET/SET_VP_INFO */
 enum {
 	OPAL_XIVE_VP_ENABLED		= 0x00000001,
+	OPAL_XIVE_VP_SINGLE_ESCALATION	= 0x00000002,
 };
 
 /* "Any chip" replacement for chip ID for allocation functions */

index 371fbebf1ec9..11d5edeb5c22 100644
@@ -143,9 +143,10 @@ extern void xive_native_disable_queue(u32 vp_id, struct xive_q *q, u8 prio);
 
 extern void xive_native_sync_source(u32 hw_irq);
 extern bool is_xive_irq(struct irq_chip *chip);
-extern int xive_native_enable_vp(u32 vp_id);
+extern int xive_native_enable_vp(u32 vp_id, bool single_escalation);
 extern int xive_native_disable_vp(u32 vp_id);
 extern int xive_native_get_vp_info(u32 vp_id, u32 *out_cam_id, u32 *out_chip_id);
+extern bool xive_native_has_single_escalation(void);
 
 #else
 

111 lines not on this page · raw message

index 6ba63f8e8a61..a08ae6fd4c51 100644
@@ -120,6 +120,8 @@ struct kvmppc_xive {
 	u32	q_order;
 	u32	q_page_order;
 
+	/* Flags */
+	u8	single_escalation;
 };
 
 #define KVMPPC_XIVE_Q_COUNT	8

@@ -201,25 +203,20 @@ static inline struct kvmppc_xive_src_block *kvmppc_xive_find_source(struct kvmpp
  * is as follow.
  *
  * Guest request for 0...6 are honored. Guest request for anything
- * higher results in a priority of 7 being applied.
- *
- * However, when XIRR is returned via H_XIRR, 7 is translated to 0xb
- * in order to match AIX expectations
+ * higher results in a priority of 6 being applied.
  *
  * Similar mapping is done for CPPR values
  */
 static inline u8 xive_prio_from_guest(u8 prio)
 {
-	if (prio = 0xff || prio < 8)
+	if (prio = 0xff || prio < 6)
 		return prio;
-	return 7;
+	return 6;
 }
 
 static inline u8 xive_prio_to_guest(u8 prio)
 {
-	if (prio = 0xff || prio < 7)
-		return prio;
-	return 0xb;
+	return prio;
 }
 
 static inline u32 __xive_read_eq(__be32 *qpage, u32 msk, u32 *idx, u32 *toggle)

index ebc244b08d67..d22aeb0b69e1 100644
@@ -42,6 +42,7 @@ static u32 xive_provision_chip_count;
 static u32 xive_queue_shift;
 static u32 xive_pool_vps = XIVE_INVALID_VP;
 static struct kmem_cache *xive_provision_cache;
+static bool xive_has_single_esc;
 
 int xive_native_populate_irq_data(u32 hw_irq, struct xive_irq_data *data)
 {

@@ -571,6 +572,10 @@ bool __init xive_native_init(void)
 			break;
 	}
 
+	/* Do we support single escalation */
+	if (of_get_property(np, "single-escalation-support", NULL) != NULL)
+		xive_has_single_esc = true;
+
 	/* Configure Thread Management areas for KVM */
 	for_each_possible_cpu(cpu)
 		kvmppc_set_xive_tima(cpu, r.start, tima);

@@ -667,12 +672,15 @@ void xive_native_free_vp_block(u32 vp_base)
 }
 EXPORT_SYMBOL_GPL(xive_native_free_vp_block);
 
-int xive_native_enable_vp(u32 vp_id)
+int xive_native_enable_vp(u32 vp_id, bool single_escalation)
 {
 	s64 rc;
+	u64 flags = OPAL_XIVE_VP_ENABLED;
 
+	if (single_escalation)
+		flags |= OPAL_XIVE_VP_SINGLE_ESCALATION;
 	for (;;) {
-		rc = opal_xive_set_vp_info(vp_id, OPAL_XIVE_VP_ENABLED, 0);
+		rc = opal_xive_set_vp_info(vp_id, flags, 0);
 		if (rc != OPAL_BUSY)
 			break;
 		msleep(1);

@@ -710,3 +718,9 @@ int xive_native_get_vp_info(u32 vp_id, u32 *out_cam_id, u32 *out_chip_id)
 	return 0;
 }
 EXPORT_SYMBOL_GPL(xive_native_get_vp_info);
+
+bool xive_native_has_single_escalation(void)
+{
+	return xive_has_single_esc;
+}
+EXPORT_SYMBOL_GPL(xive_native_has_single_escalation);

index 3aa5b577cd60..bfe51356af5e 100644
@@ -709,6 +709,7 @@ struct kvm_vcpu_arch {
 	u8 ceded;
 	u8 prodded;
 	u8 doorbell_request;
+	u8 irq_pending; /* Used by XIVE to signal pending guest irqs */
 	u32 last_inst;
 
 	struct swait_queue_head *wqp;

index 6b958414b4e0..825089cf3e23 100644
@@ -514,6 +514,7 @@ int main(void)
 	OFFSET(VCPU_PENDING_EXC, kvm_vcpu, arch.pending_exceptions);
 	OFFSET(VCPU_CEDED, kvm_vcpu, arch.ceded);
 	OFFSET(VCPU_PRODDED, kvm_vcpu, arch.prodded);
+	OFFSET(VCPU_IRQ_PENDING, kvm_vcpu, arch.irq_pending);
 	OFFSET(VCPU_DBELL_REQ, kvm_vcpu, arch.doorbell_request);
 	OFFSET(VCPU_MMCR, kvm_vcpu, arch.mmcr);
 	OFFSET(VCPU_PMC, kvm_vcpu, arch.pmc);

index 2d46037ce936..eafd722dce56 100644
@@ -2987,7 +2987,7 @@ static inline bool xive_interrupt_pending(struct kvm_vcpu *vcpu)
 {
 	if (!xive_enabled())
 		return false;
-	return vcpu->arch.xive_saved_state.pipr <
+	return vcpu->arch.irq_pending || vcpu->arch.xive_saved_state.pipr <
 		vcpu->arch.xive_saved_state.cppr;
 }
 #else

index 87f814e81e7d..de78ac38f7ed 100644
@@ -84,8 +84,7 @@ static irqreturn_t xive_esc_irq(int irq, void *data)
 {
 	struct kvm_vcpu *vcpu = data;
 
-	/* We use the existing H_PROD mechanism to wake up the target */
-	vcpu->arch.prodded = 1;
+	vcpu->arch.irq_pending = 1;
 	smp_mb();
 	if (vcpu->arch.ceded)
 		kvmppc_fast_vcpu_kick(vcpu);

index bfe51356af5e..2cf2cc6b389a 100644
@@ -739,7 +739,8 @@ struct kvm_vcpu_arch {
 	struct kvmppc_icp *icp; /* XICS presentation controller */
 	struct kvmppc_xive_vcpu *xive_vcpu; /* XIVE virtual CPU data */
 	__be32 xive_cam_word;    /* Cooked W2 in proper endian with valid bit */
-	u32 xive_pushed;	 /* Is the VP pushed on the physical CPU ? */
+	u8 xive_pushed;		 /* Is the VP pushed on the physical CPU ? */
+	u8 xive_esc_on;		 /* Is the escalation irq enabled ? */
 	union xive_tma_w01 xive_saved_state; /* W0..1 of XIVE thread state */
 #endif
 

index 1d3f2be5ae39..fa4288822b68 100644
@@ -9,6 +9,41 @@
 #ifndef _ASM_POWERPC_XIVE_REGS_H
 #define _ASM_POWERPC_XIVE_REGS_H
 
+/*
+ * "magic" Event State Buffer (ESB) MMIO offsets.
+ *
+ * Each interrupt source has a 2-bit state machine called ESB
+ * which can be controlled by MMIO. It's made of 2 bits, P and
+ * Q. P indicates that an interrupt is pending (has been sent
+ * to a queue and is waiting for an EOI). Q indicates that the
+ * interrupt has been triggered while pending.
+ *
+ * This acts as a coalescing mechanism in order to guarantee
+ * that a given interrupt only occurs at most once in a queue.
+ *
+ * When doing an EOI, the Q bit will indicate if the interrupt
+ * needs to be re-triggered.
+ *
+ * The following offsets into the ESB MMIO allow to read or
+ * manipulate the PQ bits. They must be used with an 8-bytes
+ * load instruction. They all return the previous state of the
+ * interrupt (atomically).
+ *
+ * Additionally, some ESB pages support doing an EOI via a
+ * store at 0 and some ESBs support doing a trigger via a
+ * separate trigger page.
+ */
+#define XIVE_ESB_STORE_EOI	0x400 /* Store */
+#define XIVE_ESB_LOAD_EOI	0x000 /* Load */
+#define XIVE_ESB_GET		0x800 /* Load */
+#define XIVE_ESB_SET_PQ_00	0xc00 /* Load */
+#define XIVE_ESB_SET_PQ_01	0xd00 /* Load */
+#define XIVE_ESB_SET_PQ_10	0xe00 /* Load */
+#define XIVE_ESB_SET_PQ_11	0xf00 /* Load */
+
+#define XIVE_ESB_VAL_P		0x2
+#define XIVE_ESB_VAL_Q		0x1
+
 /*
  * Thread Management (aka "TM") registers
  */

index 11d5edeb5c22..f786617ea8b7 100644
@@ -72,41 +72,6 @@ struct xive_q {
 	atomic_t		pending_count;
 };
 
-/*
- * "magic" Event State Buffer (ESB) MMIO offsets.
- *
- * Each interrupt source has a 2-bit state machine called ESB
- * which can be controlled by MMIO. It's made of 2 bits, P and
- * Q. P indicates that an interrupt is pending (has been sent
- * to a queue and is waiting for an EOI). Q indicates that the
- * interrupt has been triggered while pending.
- *
- * This acts as a coalescing mechanism in order to guarantee
- * that a given interrupt only occurs at most once in a queue.
- *
- * When doing an EOI, the Q bit will indicate if the interrupt
- * needs to be re-triggered.
- *
- * The following offsets into the ESB MMIO allow to read or
- * manipulate the PQ bits. They must be used with an 8-bytes
- * load instruction. They all return the previous state of the
- * interrupt (atomically).
- *
- * Additionally, some ESB pages support doing an EOI via a
- * store at 0 and some ESBs support doing a trigger via a
- * separate trigger page.
- */
-#define XIVE_ESB_STORE_EOI	0x400 /* Store */
-#define XIVE_ESB_LOAD_EOI	0x000 /* Load */
-#define XIVE_ESB_GET		0x800 /* Load */
-#define XIVE_ESB_SET_PQ_00	0xc00 /* Load */
-#define XIVE_ESB_SET_PQ_01	0xd00 /* Load */
-#define XIVE_ESB_SET_PQ_10	0xe00 /* Load */
-#define XIVE_ESB_SET_PQ_11	0xf00 /* Load */
-
-#define XIVE_ESB_VAL_P		0x2
-#define XIVE_ESB_VAL_Q		0x1
-
 /* Global enable flags for the XIVE support */
 extern bool __xive_enabled;
 

index f786617ea8b7..e602903c3029 100644
@@ -58,6 +58,9 @@ struct xive_irq_data {
 #define XIVE_IRQ_FLAG_EOI_FW	0x10
 #define XIVE_IRQ_FLAG_H_INT_ESB	0x20
 
+/* Special flag set by KVM for excalation interrupts */
+#define XIVE_IRQ_NO_EOI		0x80
+
 #define XIVE_INVALID_CHIP_ID	-1
 
 /* A queue tracking structure in a CPU */

index 838ebdbfe4c5..40c06110821c 100644
@@ -367,7 +367,8 @@ static void xive_irq_eoi(struct irq_data *d)
 	 * EOI the source if it hasn't been disabled and hasn't
 	 * been passed-through to a KVM guest
 	 */
-	if (!irqd_irq_disabled(d) && !irqd_is_forwarded_to_vcpu(d))
+	if (!irqd_irq_disabled(d) && !irqd_is_forwarded_to_vcpu(d) &&
+	    !(xd->flags & XIVE_IRQ_NO_EOI))
 		xive_do_source_eoi(irqd_to_hwirq(d), xd);
 
 	/*

index 2cf2cc6b389a..fef8133becc8 100644
@@ -742,6 +742,8 @@ struct kvm_vcpu_arch {
 	u8 xive_pushed;		 /* Is the VP pushed on the physical CPU ? */
 	u8 xive_esc_on;		 /* Is the escalation irq enabled ? */
 	union xive_tma_w01 xive_saved_state; /* W0..1 of XIVE thread state */
+	u64 xive_esc_raddr;	 /* Escalation interrupt ESB real addr */
+	u64 xive_esc_vaddr;	 /* Escalation interrupt ESB virt addr */
 #endif
 
 #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE

index 825089cf3e23..1672dffd94e2 100644
@@ -734,6 +734,9 @@ int main(void)
 	DEFINE(VCPU_XIVE_CAM_WORD, offsetof(struct kvm_vcpu,
 					    arch.xive_cam_word));
 	DEFINE(VCPU_XIVE_PUSHED, offsetof(struct kvm_vcpu, arch.xive_pushed));
+	DEFINE(VCPU_XIVE_ESC_ON, offsetof(struct kvm_vcpu, arch.xive_esc_on));
+	DEFINE(VCPU_XIVE_ESC_RADDR, offsetof(struct kvm_vcpu, arch.xive_esc_raddr));
+	DEFINE(VCPU_XIVE_ESC_VADDR, offsetof(struct kvm_vcpu, arch.xive_esc_vaddr));
 #endif
 
 #ifdef CONFIG_KVM_EXIT_TIMING

index de78ac38f7ed..badfdbb857a2 100644
@@ -89,6 +89,17 @@ static irqreturn_t xive_esc_irq(int irq, void *data)
 	if (vcpu->arch.ceded)
 		kvmppc_fast_vcpu_kick(vcpu);
 
+	/* Since we have the no-EOI flag, the interrupt is effectively
+	 * disabled now. Clearing xive_esc_on means we won't bother
+	 * doing so on the next entry.
+	 *
+	 * This also allows the entry code to know that if a PQ combination
+	 * of 10 is observed while xive_esc_on is true, it means the queue
+	 * contains an unprocessed escalation interrupt. We don't make use of
+	 * that knowledge today but might (see comment in book3s_hv_rmhandler.S)
+	 */
+	vcpu->arch.xive_esc_on = false;
+
 	return IRQ_HANDLED;
 }
 

@@ -134,6 +145,25 @@ static int xive_attach_escalation(struct kvm_vcpu *vcpu, u8 prio)
 		goto error;
 	}
 	xc->esc_virq_names[prio] = name;
+
+	/* In single escalation mode, we grab the ESB MMIO of the
+	 * interrupt and mask it. Also populate the VCPU v/raddr
+	 * of the ESB page for use by asm entry/exit code. Finally
+	 * set the XIVE_IRQ_NO_EOI flag which will prevent the
+	 * core code from performing an EOI on the escalation
+	 * interrupt, thus leaving it effectively masked after
+	 * it fires once.
+	 */
+	if (xc->xive->single_escalation) {
+		struct irq_data *d = irq_get_irq_data(xc->esc_virq[prio]);
+		struct xive_irq_data *xd = irq_data_get_irq_handler_data(d);
+
+		xive_vm_esb_load(xd, XIVE_ESB_SET_PQ_01);
+		vcpu->arch.xive_esc_raddr = xd->eoi_page;
+		vcpu->arch.xive_esc_vaddr = (__force u64)xd->eoi_mmio;
+		xd->flags |= XIVE_IRQ_NO_EOI;
+	}
+
 	return 0;
 error:
 	irq_dispose_mapping(xc->esc_virq[prio]);

index 72be0c32e902..397d6e64e3df 100644
@@ -1409,7 +1409,7 @@ void show_regs(struct pt_regs * regs)
 	print_msr_bits(regs->msr);
 	pr_cont("  CR: %08lx  XER: %08lx\n", regs->ccr, regs->xer);
 	trap = TRAP(regs);
-	if ((regs->trap != 0xc00) && cpu_has_feature(CPU_FTR_CFAR))
+	if ((TRAP(regs) != 0xc00) && cpu_has_feature(CPU_FTR_CFAR))
 		pr_cont("CFAR: "REG" ", regs->orig_gpr3);
 	if (trap = 0x200 || trap = 0x300 || trap = 0x600)
 #if defined(CONFIG_4xx) || defined(CONFIG_BOOKE)

index f3eb61be0d30..d61989be28e1 100644
@@ -1564,7 +1564,7 @@ void facility_unavailable_exception(struct pt_regs *regs)
 	u8 status;
 	bool hv;
 
-	hv = (regs->trap = 0xf80);
+	hv = (TRAP(regs) = 0xf80);
 	if (hv)
 		value = mfspr(SPRN_HFSCR);
 	else

index 6e1e39035380..7b15fe2ac986 100644
@@ -576,7 +576,7 @@ void bad_page_fault(struct pt_regs *regs, unsigned long address, int sig)
 
 	/* kernel has accessed a bad area */
 
-	switch (regs->trap) {
+	switch (TRAP(regs)) {
 	case 0x300:
 	case 0x380:
 		printk(KERN_ALERT "Unable to handle kernel paging request for "

index fdcff76e9a25..0a663dfc28b5 100644
@@ -178,7 +178,7 @@ end_##sname:
  * TRAMP_REAL_*   - real, unrelocated helpers (virt can call these)
  * TRAMP_VIRT_*   - virt, unreloc helpers (in practice, real can use)
  * TRAMP_KVM      - KVM handlers that get put into real, unrelocated
- * EXC_COMMON_*   - virt, relocated common handlers
+ * EXC_COMMON     - virt, relocated common handlers
  *
  * The EXC handlers are given a name, and branch to name_common, or the
  * appropriate KVM or masking function. Vector handler verieties are as

@@ -211,7 +211,6 @@ end_##sname:
  * EXC_COMMON_BEGIN/END - used to open-code the handler
  * EXC_COMMON
  * EXC_COMMON_ASYNC
- * EXC_COMMON_HV
  *
  * TRAMP_REAL and TRAMP_VIRT can be used with BEGIN/END. KVM
  * and OOL handlers are implemented as types of TRAMP and TRAMP_VIRT handlers.

@@ -413,10 +412,6 @@ end_##sname:
 	EXC_COMMON_BEGIN(name);						\
 	STD_EXCEPTION_COMMON_ASYNC(realvec, name, hdlr);		\
 
-#define EXC_COMMON_HV(name, realvec, hdlr)				\
-	EXC_COMMON_BEGIN(name);						\
-	STD_EXCEPTION_COMMON(realvec + 0x2, name, hdlr);		\
-
 #endif /* __ASSEMBLY__ */
 
 #endif	/* _ASM_POWERPC_HEAD_64_H */

index b27205297e1d..64936a89c79c 100644
@@ -243,10 +243,10 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 	EXCEPTION_PROLOG_1(area, extra, vec);				\
 	EXCEPTION_PROLOG_PSERIES_1(label, h);
 
-#define __KVMTEST(h, n)							\
+#define __KVMTEST(h, n, v)						\
 	lbz	r10,HSTATE_IN_GUEST(r13);				\
 	cmpwi	r10,0;							\
-	bne	do_kvm_##h##n
+	bne	do_kvm_##h##n##v
 
 #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 /*

@@ -353,12 +353,12 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 	b	kvmppc_skip_##h##interrupt
 
 #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
-#define KVMTEST(h, n)			__KVMTEST(h, n)
+#define KVMTEST(h, n, v)		__KVMTEST(h, n, v)
 #define KVM_HANDLER(area, h, n)		__KVM_HANDLER(area, h, n)
 #define KVM_HANDLER_SKIP(area, h, n)	__KVM_HANDLER_SKIP(area, h, n)
 
 #else
-#define KVMTEST(h, n)
+#define KVMTEST(h, n, v)
 #define KVM_HANDLER(area, h, n)
 #define KVM_HANDLER_SKIP(area, h, n)
 #endif

@@ -479,13 +479,13 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 	EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, vec);		\
 	EXCEPTION_RELON_PROLOG_PSERIES_1(label, EXC_STD)
 
-#define STD_RELON_EXCEPTION_HV(loc, vec, label)		\
-	SET_SCRATCH0(r13);	/* save r13 */		\
+#define STD_RELON_EXCEPTION_HV(loc, vec, label)			\
+	SET_SCRATCH0(r13);	/* save r13 */			\
 	EXCEPTION_RELON_PROLOG_PSERIES(PACA_EXGEN, label,	\
-				       EXC_HV, KVMTEST_HV, vec);
+				       EXC_HV, KVMTEST_HV_VIRT, vec);
 
 #define STD_RELON_EXCEPTION_HV_OOL(vec, label)			\
-	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV, vec);	\
+	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV_VIRT, vec);	\
 	EXCEPTION_RELON_PROLOG_PSERIES_1(label, EXC_HV)
 
 /* This associate vector numbers with bits in paca->irq_happened */

@@ -506,18 +506,25 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 #define _SOFTEN_TEST(h, vec)	__SOFTEN_TEST(h, vec)
 
 #define SOFTEN_TEST_PR(vec)						\
-	KVMTEST(EXC_STD, vec);						\
+	KVMTEST(EXC_STD, vec,);						\
 	_SOFTEN_TEST(EXC_STD, vec)
 
 #define SOFTEN_TEST_HV(vec)						\
-	KVMTEST(EXC_HV, vec);						\
+	KVMTEST(EXC_HV, vec,);						\
+	_SOFTEN_TEST(EXC_HV, vec)
+
+#define SOFTEN_TEST_HV_VIRT(vec)					\
+	KVMTEST(EXC_HV, vec, _VIRT);					\
 	_SOFTEN_TEST(EXC_HV, vec)
 
 #define KVMTEST_PR(vec)							\
-	KVMTEST(EXC_STD, vec)
+	KVMTEST(EXC_STD, vec,)
 
 #define KVMTEST_HV(vec)							\
-	KVMTEST(EXC_HV, vec)
+	KVMTEST(EXC_HV, vec,)
+
+#define KVMTEST_HV_VIRT(vec)						\
+	KVMTEST(EXC_HV, vec, _VIRT)
 
 #define SOFTEN_NOTEST_PR(vec)		_SOFTEN_TEST(EXC_STD, vec)
 #define SOFTEN_NOTEST_HV(vec)		_SOFTEN_TEST(EXC_HV, vec)

@@ -562,10 +569,10 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 
 #define MASKABLE_RELON_EXCEPTION_HV(loc, vec, label)			\
 	_MASKABLE_RELON_EXCEPTION_PSERIES(vec, label,			\
-					  EXC_HV, SOFTEN_TEST_HV)
+					  EXC_HV, SOFTEN_TEST_HV_VIRT)
 
 #define MASKABLE_RELON_EXCEPTION_HV_OOL(vec, label)			\
-	EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_HV, vec);		\
+	EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_HV_VIRT, vec);	\
 	EXCEPTION_RELON_PROLOG_PSERIES_1(label, EXC_HV)
 
 /*

index 0a663dfc28b5..faf8d03f0797 100644
@@ -404,6 +404,14 @@ end_##sname:
 	TRAMP_KVM_BEGIN(do_kvm_H##n);					\
 	KVM_HANDLER_SKIP(area, EXC_HV, n + 0x2);			\
 
+#define TRAMP_KVM_HV_VIRT(area, n)					\
+	TRAMP_KVM_BEGIN(do_kvm_H##n##_VIRT);				\
+	KVM_HANDLER(area, EXC_HV, n + 0x4002);				\
+
+#define TRAMP_KVM_HV_VIRT_SKIP(area, n)					\
+	TRAMP_KVM_BEGIN(do_kvm_H##n##_VIRT);				\
+	KVM_HANDLER_SKIP(area, EXC_HV, n + 0x4002);			\
+
 #define EXC_COMMON(name, realvec, hdlr)					\
 	EXC_COMMON_BEGIN(name);						\
 	STD_EXCEPTION_COMMON(realvec, name, hdlr);			\

index ab386af2904f..c6afd99d91fb 100644
@@ -125,6 +125,7 @@ struct kvmppc_host_state {
 	void __iomem *xive_tima_phys;
 	void __iomem *xive_tima_virt;
 	u32 saved_xirr;
+	u32 exit_virt;
 	u64 dabr;
 	u64 host_mmcr[7];	/* MMCR 0,1,A, SIAR, SDAR, MMCR2, SIER */
 	u32 host_pmc[8];

index 1672dffd94e2..25ee619531c2 100644
@@ -641,6 +641,7 @@ int main(void)
 	HSTATE_FIELD(HSTATE_XIVE_TIMA_PHYS, xive_tima_phys);
 	HSTATE_FIELD(HSTATE_XIVE_TIMA_VIRT, xive_tima_virt);
 	HSTATE_FIELD(HSTATE_SAVED_XIRR, saved_xirr);
+	HSTATE_FIELD(HSTATE_EXIT_VIRT, exit_virt);
 	HSTATE_FIELD(HSTATE_HOST_IPI, host_ipi);
 	HSTATE_FIELD(HSTATE_PTID, ptid);
 	HSTATE_FIELD(HSTATE_TID, tid);

index 49a2c7825e04..5fbaa8710e0c 100644
@@ -209,8 +209,7 @@ long kvmppc_h_random(struct kvm_vcpu *vcpu)
 {
 	int r;
 
-	/* Only need to do the expensive mfmsr() on radix */
-	if (kvm_is_radix(vcpu->kvm) && (mfmsr() & MSR_IR))
+	if (local_paca->kvm_hstate.exit_virt)
 		r = powernv_get_random_long(&vcpu->arch.gpr[4]);
 	else
 		r = powernv_get_random_real_mode(&vcpu->arch.gpr[4]);

@@ -541,7 +540,7 @@ static long kvmppc_read_one_intr(bool *again)
 #ifdef CONFIG_KVM_XICS
 static inline bool is_rm(void)
 {
-	return !(mfmsr() & MSR_DR);
+	return !local_paca->kvm_hstate.exit_virt;
 }
 
 unsigned long kvmppc_rm_h_xirr(struct kvm_vcpu *vcpu)