[PATCH 73/89] KVM: arm64: Add HVC handling for protected guests at EL2

[Will Deacon]

From: Fuad Tabba <tabba at google.com>

Rather than forwarding guest hypercalls back to the host for handling,
implement some basic handling at EL2 which will later be extending to
provide additional functionality such as PSCI.

Signed-off-by: Fuad Tabba <tabba at google.com>
---
 arch/arm64/kvm/hyp/include/nvhe/pkvm.h |  2 ++
 arch/arm64/kvm/hyp/nvhe/hyp-main.c     | 24 ++++++++++++++++++++++++
 arch/arm64/kvm/hyp/nvhe/pkvm.c         | 22 ++++++++++++++++++++++
 arch/arm64/kvm/hyp/nvhe/switch.c       |  1 +
 4 files changed, 49 insertions(+)

diff --git a/arch/arm64/kvm/hyp/include/nvhe/pkvm.h b/arch/arm64/kvm/hyp/include/nvhe/pkvm.h
index e772f9835a86..33d34cc639ea 100644
--- a/arch/arm64/kvm/hyp/include/nvhe/pkvm.h
+++ b/arch/arm64/kvm/hyp/include/nvhe/pkvm.h
@@ -101,4 +101,6 @@ bool kvm_handle_pvm_restricted(struct kvm_vcpu *vcpu, u64 *exit_code);
 void kvm_reset_pvm_sys_regs(struct kvm_vcpu *vcpu);
 int kvm_check_pvm_sysreg_table(void);
 
+bool kvm_handle_pvm_hvc64(struct kvm_vcpu *vcpu, u64 *exit_code);
+
 #endif /* __ARM64_KVM_NVHE_PKVM_H__ */
diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-main.c b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
index 1e39dc7eab4d..26c8709f5494 100644
--- a/arch/arm64/kvm/hyp/nvhe/hyp-main.c
+++ b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
@@ -4,6 +4,8 @@
  * Author: Andrew Scull <ascull at google.com>
  */
 
+#include <kvm/arm_hypercalls.h>
+
 #include <hyp/adjust_pc.h>
 
 #include <asm/pgtable-types.h>
@@ -42,6 +44,13 @@ static void handle_pvm_entry_wfx(struct kvm_vcpu *host_vcpu, struct kvm_vcpu *sh
 				   KVM_ARM64_INCREMENT_PC;
 }
 
+static void handle_pvm_entry_hvc64(struct kvm_vcpu *host_vcpu, struct kvm_vcpu *shadow_vcpu)
+{
+	u64 ret = READ_ONCE(host_vcpu->arch.ctxt.regs.regs[0]);
+
+	vcpu_set_reg(shadow_vcpu, 0, ret);
+}
+
 static void handle_pvm_entry_sys64(struct kvm_vcpu *host_vcpu, struct kvm_vcpu *shadow_vcpu)
 {
 	unsigned long host_flags;
@@ -195,6 +204,19 @@ static void handle_pvm_exit_sys64(struct kvm_vcpu *host_vcpu, struct kvm_vcpu *s
 	}
 }
 
+static void handle_pvm_exit_hvc64(struct kvm_vcpu *host_vcpu, struct kvm_vcpu *shadow_vcpu)
+{
+	int i;
+
+	WRITE_ONCE(host_vcpu->arch.fault.esr_el2,
+		   shadow_vcpu->arch.fault.esr_el2);
+
+	/* Pass the hvc function id (r0) as well as any potential arguments. */
+	for (i = 0; i < 8; i++)
+		WRITE_ONCE(host_vcpu->arch.ctxt.regs.regs[i],
+			   vcpu_get_reg(shadow_vcpu, i));
+}
+
 static void handle_pvm_exit_iabt(struct kvm_vcpu *host_vcpu, struct kvm_vcpu *shadow_vcpu)
 {
 	WRITE_ONCE(host_vcpu->arch.fault.esr_el2,
@@ -273,6 +295,7 @@ static void handle_vm_exit_abt(struct kvm_vcpu *host_vcpu, struct kvm_vcpu *shad
 static const shadow_entry_exit_handler_fn entry_pvm_shadow_handlers[] = {
 	[0 ... ESR_ELx_EC_MAX]		= NULL,
 	[ESR_ELx_EC_WFx]		= handle_pvm_entry_wfx,
+	[ESR_ELx_EC_HVC64]		= handle_pvm_entry_hvc64,
 	[ESR_ELx_EC_SYS64]		= handle_pvm_entry_sys64,
 	[ESR_ELx_EC_IABT_LOW]		= handle_pvm_entry_iabt,
 	[ESR_ELx_EC_DABT_LOW]		= handle_pvm_entry_dabt,
@@ -281,6 +304,7 @@ static const shadow_entry_exit_handler_fn entry_pvm_shadow_handlers[] = {
 static const shadow_entry_exit_handler_fn exit_pvm_shadow_handlers[] = {
 	[0 ... ESR_ELx_EC_MAX]		= NULL,
 	[ESR_ELx_EC_WFx]		= handle_pvm_exit_wfx,
+	[ESR_ELx_EC_HVC64]		= handle_pvm_exit_hvc64,
 	[ESR_ELx_EC_SYS64]		= handle_pvm_exit_sys64,
 	[ESR_ELx_EC_IABT_LOW]		= handle_pvm_exit_iabt,
 	[ESR_ELx_EC_DABT_LOW]		= handle_pvm_exit_dabt,
diff --git a/arch/arm64/kvm/hyp/nvhe/pkvm.c b/arch/arm64/kvm/hyp/nvhe/pkvm.c
index 9feeb0b5433a..92e60ebeced5 100644
--- a/arch/arm64/kvm/hyp/nvhe/pkvm.c
+++ b/arch/arm64/kvm/hyp/nvhe/pkvm.c
@@ -7,6 +7,8 @@
 #include <linux/kvm_host.h>
 #include <linux/mm.h>
 
+#include <kvm/arm_hypercalls.h>
+
 #include <asm/kvm_emulate.h>
 
 #include <nvhe/mem_protect.h>
@@ -797,3 +799,23 @@ int __pkvm_teardown_shadow(unsigned int shadow_handle)
 	hyp_spin_unlock(&shadow_lock);
 	return err;
 }
+
+/*
+ * Handler for protected VM HVC calls.
+ *
+ * Returns true if the hypervisor has handled the exit, and control should go
+ * back to the guest, or false if it hasn't.
+ */
+bool kvm_handle_pvm_hvc64(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	u32 fn = smccc_get_function(vcpu);
+
+	switch (fn) {
+	case ARM_SMCCC_VERSION_FUNC_ID:
+		/* Nothing to be handled by the host. Go back to the guest. */
+		smccc_set_retval(vcpu, ARM_SMCCC_VERSION_1_1, 0, 0, 0);
+		return true;
+	default:
+		return false;
+	}
+}
diff --git a/arch/arm64/kvm/hyp/nvhe/switch.c b/arch/arm64/kvm/hyp/nvhe/switch.c
index 6bb979ee51cc..87338775288c 100644
--- a/arch/arm64/kvm/hyp/nvhe/switch.c
+++ b/arch/arm64/kvm/hyp/nvhe/switch.c
@@ -205,6 +205,7 @@ static const exit_handler_fn hyp_exit_handlers[] = {
 
 static const exit_handler_fn pvm_exit_handlers[] = {
 	[0 ... ESR_ELx_EC_MAX]		= NULL,
+	[ESR_ELx_EC_HVC64]		= kvm_handle_pvm_hvc64,
 	[ESR_ELx_EC_SYS64]		= kvm_handle_pvm_sys64,
 	[ESR_ELx_EC_SVE]		= kvm_handle_pvm_restricted,
 	[ESR_ELx_EC_FP_ASIMD]		= kvm_handle_pvm_fpsimd,
-- 
2.36.1.124.g0e6072fb45-goog