KVM: PPC: Book3S 64: Move GUEST_MODE_SKIP test into KVM

Move the GUEST_MODE_SKIP logic into KVM code. This is quite a KVM
internal detail that has no real need to be in common handlers.

Add a comment explaining the what and why of KVM "skip" interrupts.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210528090752.3542186-3-npiggin@gmail.com

1 files changed, 58 insertions, 1 deletions

diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index 7a039ea78f15..bf927e7a06af 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -1,6 +1,7 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 #include <asm/asm-offsets.h>
 #include <asm/cache.h>
+#include <asm/exception-64s.h>
 #include <asm/kvm_asm.h>
 #include <asm/kvm_book3s_asm.h>
 #include <asm/ppc_asm.h>
@@ -20,9 +21,12 @@ kvmppc_interrupt:
 	 * guest R12 saved in shadow VCPU SCRATCH0
 	 * guest R13 saved in SPRN_SCRATCH0
 	 */
-#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	std	r9,HSTATE_SCRATCH2(r13)
 	lbz	r9,HSTATE_IN_GUEST(r13)
+	cmpwi	r9,KVM_GUEST_MODE_SKIP
+	beq-	.Lmaybe_skip
+.Lno_skip:
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	cmpwi	r9,KVM_GUEST_MODE_HOST_HV
 	beq	kvmppc_bad_host_intr
 #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
@@ -32,5 +36,58 @@ kvmppc_interrupt:
 #endif
 	b	kvmppc_interrupt_hv
 #else
+	ld	r9,HSTATE_SCRATCH2(r13)
 	b	kvmppc_interrupt_pr
 #endif
+
+/*
+ * "Skip" interrupts are part of a trick KVM uses a with hash guests to load
+ * the faulting instruction in guest memory from the the hypervisor without
+ * walking page tables.
+ *
+ * When the guest takes a fault that requires the hypervisor to load the
+ * instruction (e.g., MMIO emulation), KVM is running in real-mode with HV=1
+ * and the guest MMU context loaded. It sets KVM_GUEST_MODE_SKIP, and sets
+ * MSR[DR]=1 while leaving MSR[IR]=0, so it continues to fetch HV instructions
+ * but loads and stores will access the guest context. This is used to load
+ * the faulting instruction using the faulting guest effective address.
+ *
+ * However the guest context may not be able to translate, or it may cause a
+ * machine check or other issue, which results in a fault in the host
+ * (even with KVM-HV).
+ *
+ * These faults come here because KVM_GUEST_MODE_SKIP was set, so if they
+ * are (or are likely) caused by that load, the instruction is skipped by
+ * just returning with the PC advanced +4, where it is noticed the load did
+ * not execute and it goes to the slow path which walks the page tables to
+ * read guest memory.
+ */
+.Lmaybe_skip:
+	cmpwi	r12,BOOK3S_INTERRUPT_MACHINE_CHECK
+	beq	1f
+	cmpwi	r12,BOOK3S_INTERRUPT_DATA_STORAGE
+	beq	1f
+	cmpwi	r12,BOOK3S_INTERRUPT_DATA_SEGMENT
+	beq	1f
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+	/* HSRR interrupts get 2 added to interrupt number */
+	cmpwi	r12,BOOK3S_INTERRUPT_H_DATA_STORAGE | 0x2
+	beq	2f
+#endif
+	b	.Lno_skip
+1:	mfspr	r9,SPRN_SRR0
+	addi	r9,r9,4
+	mtspr	SPRN_SRR0,r9
+	ld	r12,HSTATE_SCRATCH0(r13)
+	ld	r9,HSTATE_SCRATCH2(r13)
+	GET_SCRATCH0(r13)
+	RFI_TO_KERNEL
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+2:	mfspr	r9,SPRN_HSRR0
+	addi	r9,r9,4
+	mtspr	SPRN_HSRR0,r9
+	ld	r12,HSTATE_SCRATCH0(r13)
+	ld	r9,HSTATE_SCRATCH2(r13)
+	GET_SCRATCH0(r13)
+	HRFI_TO_KERNEL
+#endif