Merge tag 'kvm-riscv-6.19-1' of https://github.com/kvm-riscv/linux into HEAD

bonzini · bonzini · commit 63a9b0bc65d5 · 2025-12-02T18:35:25.000+01:00
KVM/riscv changes for 6.19

- SBI MPXY support for KVM guest
- New KVM_EXIT_FAIL_ENTRY_NO_VSFILE for the case when in-kernel
  AIA virtualization fails to allocate IMSIC VS-file
- Support enabling dirty log gradually in small chunks
- Fix guest page fault within HLV* instructions
- Flush VS-stage TLB after VCPU migration for Andes cores
diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst
@@ -8028,7 +8028,7 @@ will be initialized to 1 when created.  This also improves performance because
 dirty logging can be enabled gradually in small chunks on the first call
 to KVM_CLEAR_DIRTY_LOG.  KVM_DIRTY_LOG_INITIALLY_SET depends on
 KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is also only available on
-x86 and arm64 for now).
+x86, arm64 and riscv for now).
 
 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previously available under the name
 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the implementation had bugs that make
diff --git a/arch/riscv/include/asm/kvm_host.h b/arch/riscv/include/asm/kvm_host.h
@@ -59,6 +59,9 @@
 					 BIT(IRQ_VS_TIMER) | \
 					 BIT(IRQ_VS_EXT))
 
+#define KVM_DIRTY_LOG_MANUAL_CAPS	(KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE | \
+					 KVM_DIRTY_LOG_INITIALLY_SET)
+
 struct kvm_vm_stat {
 	struct kvm_vm_stat_generic generic;
 };
@@ -327,4 +330,7 @@ bool kvm_riscv_vcpu_stopped(struct kvm_vcpu *vcpu);
 
 void kvm_riscv_vcpu_record_steal_time(struct kvm_vcpu *vcpu);
 
+/* Flags representing implementation specific details */
+DECLARE_STATIC_KEY_FALSE(kvm_riscv_vsstage_tlb_no_gpa);
+
 #endif /* __RISCV_KVM_HOST_H__ */
diff --git a/arch/riscv/include/asm/kvm_tlb.h b/arch/riscv/include/asm/kvm_tlb.h
@@ -49,6 +49,7 @@ void kvm_riscv_local_hfence_vvma_gva(unsigned long vmid,
 				     unsigned long gva, unsigned long gvsz,
 				     unsigned long order);
 void kvm_riscv_local_hfence_vvma_all(unsigned long vmid);
+void kvm_riscv_local_tlb_sanitize(struct kvm_vcpu *vcpu);
 
 void kvm_riscv_tlb_flush_process(struct kvm_vcpu *vcpu);
 
diff --git a/arch/riscv/include/asm/kvm_vcpu_sbi.h b/arch/riscv/include/asm/kvm_vcpu_sbi.h
@@ -69,7 +69,9 @@ struct kvm_vcpu_sbi_extension {
 			     unsigned long reg_size, const void *reg_val);
 };
 
-void kvm_riscv_vcpu_sbi_forward(struct kvm_vcpu *vcpu, struct kvm_run *run);
+int kvm_riscv_vcpu_sbi_forward_handler(struct kvm_vcpu *vcpu,
+				       struct kvm_run *run,
+				       struct kvm_vcpu_sbi_return *retdata);
 void kvm_riscv_vcpu_sbi_system_reset(struct kvm_vcpu *vcpu,
 				     struct kvm_run *run,
 				     u32 type, u64 flags);
@@ -105,6 +107,7 @@ extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_dbcn;
 extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_susp;
 extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_sta;
 extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_fwft;
+extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_mpxy;
 extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_experimental;
 extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_vendor;
 
diff --git a/arch/riscv/include/asm/kvm_vmid.h b/arch/riscv/include/asm/kvm_vmid.h
@@ -22,6 +22,5 @@ unsigned long kvm_riscv_gstage_vmid_bits(void);
 int kvm_riscv_gstage_vmid_init(struct kvm *kvm);
 bool kvm_riscv_gstage_vmid_ver_changed(struct kvm_vmid *vmid);
 void kvm_riscv_gstage_vmid_update(struct kvm_vcpu *vcpu);
-void kvm_riscv_gstage_vmid_sanitize(struct kvm_vcpu *vcpu);
 
 #endif
diff --git a/arch/riscv/include/uapi/asm/kvm.h b/arch/riscv/include/uapi/asm/kvm.h
@@ -23,6 +23,8 @@
 #define KVM_INTERRUPT_SET	-1U
 #define KVM_INTERRUPT_UNSET	-2U
 
+#define KVM_EXIT_FAIL_ENTRY_NO_VSFILE	(1ULL << 0)
+
 /* for KVM_GET_REGS and KVM_SET_REGS */
 struct kvm_regs {
 };
@@ -211,6 +213,7 @@ enum KVM_RISCV_SBI_EXT_ID {
 	KVM_RISCV_SBI_EXT_STA,
 	KVM_RISCV_SBI_EXT_SUSP,
 	KVM_RISCV_SBI_EXT_FWFT,
+	KVM_RISCV_SBI_EXT_MPXY,
 	KVM_RISCV_SBI_EXT_MAX,
 };
 
diff --git a/arch/riscv/kvm/Makefile b/arch/riscv/kvm/Makefile
@@ -27,6 +27,7 @@ kvm-y += vcpu_onereg.o
 kvm-$(CONFIG_RISCV_PMU_SBI) += vcpu_pmu.o
 kvm-y += vcpu_sbi.o
 kvm-y += vcpu_sbi_base.o
+kvm-y += vcpu_sbi_forward.o
 kvm-y += vcpu_sbi_fwft.o
 kvm-y += vcpu_sbi_hsm.o
 kvm-$(CONFIG_RISCV_PMU_SBI) += vcpu_sbi_pmu.o
diff --git a/arch/riscv/kvm/aia_imsic.c b/arch/riscv/kvm/aia_imsic.c
@@ -814,7 +814,7 @@ int kvm_riscv_vcpu_aia_imsic_update(struct kvm_vcpu *vcpu)
 		/* For HW acceleration mode, we can't continue */
 		if (kvm->arch.aia.mode == KVM_DEV_RISCV_AIA_MODE_HWACCEL) {
 			run->fail_entry.hardware_entry_failure_reason =
-								CSR_HSTATUS;
+								KVM_EXIT_FAIL_ENTRY_NO_VSFILE;
 			run->fail_entry.cpu = vcpu->cpu;
 			run->exit_reason = KVM_EXIT_FAIL_ENTRY;
 			return 0;
diff --git a/arch/riscv/kvm/main.c b/arch/riscv/kvm/main.c
@@ -15,6 +15,18 @@
 #include <asm/kvm_nacl.h>
 #include <asm/sbi.h>
 
+DEFINE_STATIC_KEY_FALSE(kvm_riscv_vsstage_tlb_no_gpa);
+
+static void kvm_riscv_setup_vendor_features(void)
+{
+	/* Andes AX66: split two-stage TLBs */
+	if (riscv_cached_mvendorid(0) == ANDES_VENDOR_ID &&
+	    (riscv_cached_marchid(0) & 0xFFFF) == 0x8A66) {
+		static_branch_enable(&kvm_riscv_vsstage_tlb_no_gpa);
+		kvm_info("VS-stage TLB does not cache guest physical address and VMID\n");
+	}
+}
+
 long kvm_arch_dev_ioctl(struct file *filp,
 			unsigned int ioctl, unsigned long arg)
 {
@@ -160,6 +172,8 @@ static int __init riscv_kvm_init(void)
 		kvm_info("AIA available with %d guest external interrupts\n",
 			 kvm_riscv_aia_nr_hgei);
 
+	kvm_riscv_setup_vendor_features();
+
 	kvm_register_perf_callbacks(NULL);
 
 	rc = kvm_init(sizeof(struct kvm_vcpu), 0, THIS_MODULE);
diff --git a/arch/riscv/kvm/mmu.c b/arch/riscv/kvm/mmu.c
@@ -161,8 +161,11 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
 	 * allocated dirty_bitmap[], dirty pages will be tracked while
 	 * the memory slot is write protected.
 	 */
-	if (change != KVM_MR_DELETE && new->flags & KVM_MEM_LOG_DIRTY_PAGES)
+	if (change != KVM_MR_DELETE && new->flags & KVM_MEM_LOG_DIRTY_PAGES) {
+		if (kvm_dirty_log_manual_protect_and_init_set(kvm))
+			return;
 		mmu_wp_memory_region(kvm, new->id);
+	}
 }
 
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
diff --git a/arch/riscv/kvm/tlb.c b/arch/riscv/kvm/tlb.c
@@ -158,6 +158,36 @@ void kvm_riscv_local_hfence_vvma_all(unsigned long vmid)
 	csr_write(CSR_HGATP, hgatp);
 }
 
+void kvm_riscv_local_tlb_sanitize(struct kvm_vcpu *vcpu)
+{
+	unsigned long vmid;
+
+	if (!kvm_riscv_gstage_vmid_bits() ||
+	    vcpu->arch.last_exit_cpu == vcpu->cpu)
+		return;
+
+	/*
+	 * On RISC-V platforms with hardware VMID support, we share same
+	 * VMID for all VCPUs of a particular Guest/VM. This means we might
+	 * have stale G-stage TLB entries on the current Host CPU due to
+	 * some other VCPU of the same Guest which ran previously on the
+	 * current Host CPU.
+	 *
+	 * To cleanup stale TLB entries, we simply flush all G-stage TLB
+	 * entries by VMID whenever underlying Host CPU changes for a VCPU.
+	 */
+
+	vmid = READ_ONCE(vcpu->kvm->arch.vmid.vmid);
+	kvm_riscv_local_hfence_gvma_vmid_all(vmid);
+
+	/*
+	 * Flush VS-stage TLB entries for implementation where VS-stage
+	 * TLB does not cahce guest physical address and VMID.
+	 */
+	if (static_branch_unlikely(&kvm_riscv_vsstage_tlb_no_gpa))
+		kvm_riscv_local_hfence_vvma_all(vmid);
+}
+
 void kvm_riscv_fence_i_process(struct kvm_vcpu *vcpu)
 {
 	kvm_riscv_vcpu_pmu_incr_fw(vcpu, SBI_PMU_FW_FENCE_I_RCVD);
diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c
@@ -968,7 +968,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu)
 		 * Note: This should be done after G-stage VMID has been
 		 * updated using kvm_riscv_gstage_vmid_ver_changed()
 		 */
-		kvm_riscv_gstage_vmid_sanitize(vcpu);
+		kvm_riscv_local_tlb_sanitize(vcpu);
 
 		trace_kvm_entry(vcpu);
 
diff --git a/arch/riscv/kvm/vcpu_insn.c b/arch/riscv/kvm/vcpu_insn.c
@@ -298,6 +298,22 @@ static int system_opcode_insn(struct kvm_vcpu *vcpu, struct kvm_run *run,
 	return (rc <= 0) ? rc : 1;
 }
 
+static bool is_load_guest_page_fault(unsigned long scause)
+{
+	/**
+	 * If a g-stage page fault occurs, the direct approach
+	 * is to let the g-stage page fault handler handle it
+	 * naturally, however, calling the g-stage page fault
+	 * handler here seems rather strange.
+	 * Considering this is a corner case, we can directly
+	 * return to the guest and re-execute the same PC, this
+	 * will trigger a g-stage page fault again and then the
+	 * regular g-stage page fault handler will populate
+	 * g-stage page table.
+	 */
+	return (scause == EXC_LOAD_GUEST_PAGE_FAULT);
+}
+
 /**
  * kvm_riscv_vcpu_virtual_insn -- Handle virtual instruction trap
  *
@@ -323,6 +339,8 @@ int kvm_riscv_vcpu_virtual_insn(struct kvm_vcpu *vcpu, struct kvm_run *run,
 							  ct->sepc,
 							  &utrap);
 			if (utrap.scause) {
+				if (is_load_guest_page_fault(utrap.scause))
+					return 1;
 				utrap.sepc = ct->sepc;
 				kvm_riscv_vcpu_trap_redirect(vcpu, &utrap);
 				return 1;
@@ -378,6 +396,8 @@ int kvm_riscv_vcpu_mmio_load(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		insn = kvm_riscv_vcpu_unpriv_read(vcpu, true, ct->sepc,
 						  &utrap);
 		if (utrap.scause) {
+			if (is_load_guest_page_fault(utrap.scause))
+				return 1;
 			/* Redirect trap if we failed to read instruction */
 			utrap.sepc = ct->sepc;
 			kvm_riscv_vcpu_trap_redirect(vcpu, &utrap);
@@ -504,6 +524,8 @@ int kvm_riscv_vcpu_mmio_store(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		insn = kvm_riscv_vcpu_unpriv_read(vcpu, true, ct->sepc,
 						  &utrap);
 		if (utrap.scause) {
+			if (is_load_guest_page_fault(utrap.scause))
+				return 1;
 			/* Redirect trap if we failed to read instruction */
 			utrap.sepc = ct->sepc;
 			kvm_riscv_vcpu_trap_redirect(vcpu, &utrap);
diff --git a/arch/riscv/kvm/vcpu_sbi.c b/arch/riscv/kvm/vcpu_sbi.c
@@ -82,6 +82,10 @@ static const struct kvm_riscv_sbi_extension_entry sbi_ext[] = {
 		.ext_idx = KVM_RISCV_SBI_EXT_FWFT,
 		.ext_ptr = &vcpu_sbi_ext_fwft,
 	},
+	{
+		.ext_idx = KVM_RISCV_SBI_EXT_MPXY,
+		.ext_ptr = &vcpu_sbi_ext_mpxy,
+	},
 	{
 		.ext_idx = KVM_RISCV_SBI_EXT_EXPERIMENTAL,
 		.ext_ptr = &vcpu_sbi_ext_experimental,
@@ -120,7 +124,9 @@ static bool riscv_vcpu_supports_sbi_ext(struct kvm_vcpu *vcpu, int idx)
 	return sext && scontext->ext_status[sext->ext_idx] != KVM_RISCV_SBI_EXT_STATUS_UNAVAILABLE;
 }
 
-void kvm_riscv_vcpu_sbi_forward(struct kvm_vcpu *vcpu, struct kvm_run *run)
+int kvm_riscv_vcpu_sbi_forward_handler(struct kvm_vcpu *vcpu,
+				       struct kvm_run *run,
+				       struct kvm_vcpu_sbi_return *retdata)
 {
 	struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
 
@@ -137,6 +143,8 @@ void kvm_riscv_vcpu_sbi_forward(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	run->riscv_sbi.args[5] = cp->a5;
 	run->riscv_sbi.ret[0] = SBI_ERR_NOT_SUPPORTED;
 	run->riscv_sbi.ret[1] = 0;
+	retdata->uexit = true;
+	return 0;
 }
 
 void kvm_riscv_vcpu_sbi_system_reset(struct kvm_vcpu *vcpu,
diff --git a/arch/riscv/kvm/vcpu_sbi_base.c b/arch/riscv/kvm/vcpu_sbi_base.c
@@ -41,8 +41,7 @@ static int kvm_sbi_ext_base_handler(struct kvm_vcpu *vcpu, struct kvm_run *run,
 			 * For experimental/vendor extensions
 			 * forward it to the userspace
 			 */
-			kvm_riscv_vcpu_sbi_forward(vcpu, run);
-			retdata->uexit = true;
+			return kvm_riscv_vcpu_sbi_forward_handler(vcpu, run, retdata);
 		} else {
 			sbi_ext = kvm_vcpu_sbi_find_ext(vcpu, cp->a0);
 			*out_val = sbi_ext && sbi_ext->probe ?
@@ -71,28 +70,3 @@ const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_base = {
 	.extid_end = SBI_EXT_BASE,
 	.handler = kvm_sbi_ext_base_handler,
 };
-
-static int kvm_sbi_ext_forward_handler(struct kvm_vcpu *vcpu,
-				       struct kvm_run *run,
-				       struct kvm_vcpu_sbi_return *retdata)
-{
-	/*
-	 * Both SBI experimental and vendor extensions are
-	 * unconditionally forwarded to userspace.
-	 */
-	kvm_riscv_vcpu_sbi_forward(vcpu, run);
-	retdata->uexit = true;
-	return 0;
-}
-
-const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_experimental = {
-	.extid_start = SBI_EXT_EXPERIMENTAL_START,
-	.extid_end = SBI_EXT_EXPERIMENTAL_END,
-	.handler = kvm_sbi_ext_forward_handler,
-};
-
-const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_vendor = {
-	.extid_start = SBI_EXT_VENDOR_START,
-	.extid_end = SBI_EXT_VENDOR_END,
-	.handler = kvm_sbi_ext_forward_handler,
-};
diff --git a/arch/riscv/kvm/vcpu_sbi_forward.c b/arch/riscv/kvm/vcpu_sbi_forward.c
@@ -0,0 +1,34 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2025 Ventana Micro Systems Inc.
+ */
+
+#include <linux/kvm_host.h>
+#include <asm/kvm_vcpu_sbi.h>
+#include <asm/sbi.h>
+
+const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_experimental = {
+	.extid_start = SBI_EXT_EXPERIMENTAL_START,
+	.extid_end = SBI_EXT_EXPERIMENTAL_END,
+	.handler = kvm_riscv_vcpu_sbi_forward_handler,
+};
+
+const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_vendor = {
+	.extid_start = SBI_EXT_VENDOR_START,
+	.extid_end = SBI_EXT_VENDOR_END,
+	.handler = kvm_riscv_vcpu_sbi_forward_handler,
+};
+
+const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_dbcn = {
+	.extid_start = SBI_EXT_DBCN,
+	.extid_end = SBI_EXT_DBCN,
+	.default_disabled = true,
+	.handler = kvm_riscv_vcpu_sbi_forward_handler,
+};
+
+const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_mpxy = {
+	.extid_start = SBI_EXT_MPXY,
+	.extid_end = SBI_EXT_MPXY,
+	.default_disabled = true,
+	.handler = kvm_riscv_vcpu_sbi_forward_handler,
+};
diff --git a/arch/riscv/kvm/vcpu_sbi_replace.c b/arch/riscv/kvm/vcpu_sbi_replace.c
@@ -185,35 +185,3 @@ const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_srst = {
 	.extid_end = SBI_EXT_SRST,
 	.handler = kvm_sbi_ext_srst_handler,
 };
-
-static int kvm_sbi_ext_dbcn_handler(struct kvm_vcpu *vcpu,
-				    struct kvm_run *run,
-				    struct kvm_vcpu_sbi_return *retdata)
-{
-	struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
-	unsigned long funcid = cp->a6;
-
-	switch (funcid) {
-	case SBI_EXT_DBCN_CONSOLE_WRITE:
-	case SBI_EXT_DBCN_CONSOLE_READ:
-	case SBI_EXT_DBCN_CONSOLE_WRITE_BYTE:
-		/*
-		 * The SBI debug console functions are unconditionally
-		 * forwarded to the userspace.
-		 */
-		kvm_riscv_vcpu_sbi_forward(vcpu, run);
-		retdata->uexit = true;
-		break;
-	default:
-		retdata->err_val = SBI_ERR_NOT_SUPPORTED;
-	}
-
-	return 0;
-}
-
-const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_dbcn = {
-	.extid_start = SBI_EXT_DBCN,
-	.extid_end = SBI_EXT_DBCN,
-	.default_disabled = true,
-	.handler = kvm_sbi_ext_dbcn_handler,
-};
diff --git a/arch/riscv/kvm/vcpu_sbi_system.c b/arch/riscv/kvm/vcpu_sbi_system.c
@@ -47,9 +47,7 @@ static int kvm_sbi_ext_susp_handler(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		kvm_riscv_vcpu_sbi_request_reset(vcpu, cp->a1, cp->a2);
 
 		/* userspace provides the suspend implementation */
-		kvm_riscv_vcpu_sbi_forward(vcpu, run);
-		retdata->uexit = true;
-		break;
+		return kvm_riscv_vcpu_sbi_forward_handler(vcpu, run, retdata);
 	default:
 		retdata->err_val = SBI_ERR_NOT_SUPPORTED;
 		break;
diff --git a/arch/riscv/kvm/vcpu_sbi_v01.c b/arch/riscv/kvm/vcpu_sbi_v01.c
@@ -32,8 +32,7 @@ static int kvm_sbi_ext_v01_handler(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		 * The CONSOLE_GETCHAR/CONSOLE_PUTCHAR SBI calls cannot be
 		 * handled in kernel so we forward these to user-space
 		 */
-		kvm_riscv_vcpu_sbi_forward(vcpu, run);
-		retdata->uexit = true;
+		ret = kvm_riscv_vcpu_sbi_forward_handler(vcpu, run, retdata);
 		break;
 	case SBI_EXT_0_1_SET_TIMER:
 #if __riscv_xlen == 32
diff --git a/arch/riscv/kvm/vmid.c b/arch/riscv/kvm/vmid.c
diff --git a/tools/testing/selftests/kvm/riscv/get-reg-list.c b/tools/testing/selftests/kvm/riscv/get-reg-list.c
