Skip to contentSkip to conversation

RISC-V: KVM: Redirect illegal instruction traps to guest

PATCH In review Anup Patel 2 participants 2 messages Jan 27, 2023

Subsystems (3)
lore
Share & tools
lore

Anup Patel

author
#0 lore raw

PATCH v2 1/1 RISC-V: KVM: Redirect illegal instruction traps to guest

From: Andy Chiu <[email protected]>

The M-mode redirects an unhandled illegal instruction trap back to S-mode. However, KVM running in HS-mode terminates the VS-mode software when it receives illegal instruction trap. Instead, KVM should redirect the illegal instruction trap back to VS-mode, and let VS-mode trap handler decide the next step. This futher allows guest kernel to implement on-demand enabling of vector extension for a guest user space process upon first-use.

---
 arch/riscv/kvm/vcpu_exit.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/riscv/kvm/vcpu_exit.c +9 -0
index c9f741ab26f5..4ea101a73d8b 100644
@@ -160,6 +160,9 @@ void kvm_riscv_vcpu_trap_redirect(struct kvm_vcpu *vcpu,
 
 	/* Set Guest PC to Guest exception vector */
 	vcpu->arch.guest_context.sepc = csr_read(CSR_VSTVEC);
+
+	/* Set Guest privilege mode to supervisor */
+	vcpu->arch.guest_context.sstatus |= SR_SPP;
 }
 
 /*
@@ -179,6 +182,12 @@ int kvm_riscv_vcpu_exit(struct kvm_vcpu *vcpu, struct kvm_run *run,
 	ret = -EFAULT;
 	run->exit_reason = KVM_EXIT_UNKNOWN;
 	switch (trap->scause) {
+	case EXC_INST_ILLEGAL:
+		if (vcpu->arch.guest_context.hstatus & HSTATUS_SPV) {
+			kvm_riscv_vcpu_trap_redirect(vcpu, trap);
+			ret = 1;
+		}
+		break;
 	case EXC_VIRTUAL_INST_FAULT:
 		if (vcpu->arch.guest_context.hstatus & HSTATUS_SPV)
 			ret = kvm_riscv_vcpu_virtual_insn(vcpu, run, trap);
Signature · 1 line
2.34.1
Reviewed-by on v2 1/1
#1 lore raw
Quoted 21 lines · Anup Patel

From: Andy Chiu <[email protected]>

The M-mode redirects an unhandled illegal instruction trap back to S-mode. However, KVM running in HS-mode terminates the VS-mode software when it receives illegal instruction trap. Instead, KVM should redirect the illegal instruction trap back to VS-mode, and let VS-mode trap handler decide the next step. This futher allows guest kernel to implement on-demand enabling of vector extension for a guest user space process upon first-use.

---
 arch/riscv/kvm/vcpu_exit.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/riscv/kvm/vcpu_exit.c +0 -0
index c9f741ab26f5..4ea101a73d8b 100644

arch/riscv/kvm/vcpu_exit.c void kvm_riscv_vcpu_trap_redirect(struct kvm_vcpu *vcpu, · line 165

Quoted 6 lines
arch/riscv/kvm/vcpu_exit.c +3 -0
@@ -160,6 +160,9 @@ void kvm_riscv_vcpu_trap_redirect(struct kvm_vcpu *vcpu,
 
 	/* Set Guest PC to Guest exception vector */
 	vcpu->arch.guest_context.sepc = csr_read(CSR_VSTVEC);
+
+	/* Set Guest privilege mode to supervisor */
+	vcpu->arch.guest_context.sstatus |= SR_SPP;

This could/should be a separate fix patch, right?

Quoted 19 lines
arch/riscv/kvm/vcpu_exit.c +6 -0
 }
 
 /*
@@ -179,6 +182,12 @@ int kvm_riscv_vcpu_exit(struct kvm_vcpu *vcpu, struct kvm_run *run,
 	ret = -EFAULT;
 	run->exit_reason = KVM_EXIT_UNKNOWN;
 	switch (trap->scause) {
+	case EXC_INST_ILLEGAL:
+		if (vcpu->arch.guest_context.hstatus & HSTATUS_SPV) {
+			kvm_riscv_vcpu_trap_redirect(vcpu, trap);
+			ret = 1;
+		}
+		break;
 	case EXC_VIRTUAL_INST_FAULT:
 		if (vcpu->arch.guest_context.hstatus & HSTATUS_SPV)
 			ret = kvm_riscv_vcpu_virtual_insn(vcpu, run, trap);
Signature · 1 line
2.34.1

Otherwise,

Thanks,
drew