RISC-V: KVM: Factor-out g-stage page table management

The upcoming nested virtualization can share g-stage page table
management with the current host g-stage implementation hence
factor-out g-stage page table management as separate sources
and also use "kvm_riscv_mmu_" prefix for host g-stage functions.

Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com>
Link: https://lore.kernel.org/r/20250618113532.471448-12-apatel@ventanamicro.com
Signed-off-by: Anup Patel <anup@brainfault.org>

arch/riscv/include/asm/kvm_gstage.h

@@ -0,0 +1,72 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ * Copyright (c) 2025 Ventana Micro Systems Inc.
+ */
+
+#ifndef __RISCV_KVM_GSTAGE_H_
+#define __RISCV_KVM_GSTAGE_H_
+
+#include <linux/kvm_types.h>
+
+struct kvm_gstage {
+	struct kvm *kvm;
+	unsigned long flags;
+#define KVM_GSTAGE_FLAGS_LOCAL		BIT(0)
+	unsigned long vmid;
+	pgd_t *pgd;
+};
+
+struct kvm_gstage_mapping {
+	gpa_t addr;
+	pte_t pte;
+	u32 level;
+};
+
+#ifdef CONFIG_64BIT
+#define kvm_riscv_gstage_index_bits	9
+#else
+#define kvm_riscv_gstage_index_bits	10
+#endif
+
+extern unsigned long kvm_riscv_gstage_mode;
+extern unsigned long kvm_riscv_gstage_pgd_levels;
+
+#define kvm_riscv_gstage_pgd_xbits	2
+#define kvm_riscv_gstage_pgd_size	(1UL << (HGATP_PAGE_SHIFT + kvm_riscv_gstage_pgd_xbits))
+#define kvm_riscv_gstage_gpa_bits	(HGATP_PAGE_SHIFT + \
+					 (kvm_riscv_gstage_pgd_levels * \
+					  kvm_riscv_gstage_index_bits) + \
+					 kvm_riscv_gstage_pgd_xbits)
+#define kvm_riscv_gstage_gpa_size	((gpa_t)(1ULL << kvm_riscv_gstage_gpa_bits))
+
+bool kvm_riscv_gstage_get_leaf(struct kvm_gstage *gstage, gpa_t addr,
+			       pte_t **ptepp, u32 *ptep_level);
+
+int kvm_riscv_gstage_set_pte(struct kvm_gstage *gstage,
+			     struct kvm_mmu_memory_cache *pcache,
+			     const struct kvm_gstage_mapping *map);
+
+int kvm_riscv_gstage_map_page(struct kvm_gstage *gstage,
+			      struct kvm_mmu_memory_cache *pcache,
+			      gpa_t gpa, phys_addr_t hpa, unsigned long page_size,
+			      bool page_rdonly, bool page_exec,
+			      struct kvm_gstage_mapping *out_map);
+
+enum kvm_riscv_gstage_op {
+	GSTAGE_OP_NOP = 0,	/* Nothing */
+	GSTAGE_OP_CLEAR,	/* Clear/Unmap */
+	GSTAGE_OP_WP,		/* Write-protect */
+};
+
+void kvm_riscv_gstage_op_pte(struct kvm_gstage *gstage, gpa_t addr,
+			     pte_t *ptep, u32 ptep_level, enum kvm_riscv_gstage_op op);
+
+void kvm_riscv_gstage_unmap_range(struct kvm_gstage *gstage,
+				  gpa_t start, gpa_t size, bool may_block);
+
+void kvm_riscv_gstage_wp_range(struct kvm_gstage *gstage, gpa_t start, gpa_t end);
+
+void kvm_riscv_gstage_mode_detect(void);
+
+#endif

arch/riscv/include/asm/kvm_mmu.h

@@ -6,28 +6,16 @@
 #ifndef __RISCV_KVM_MMU_H_
 #define __RISCV_KVM_MMU_H_
 
-#include <linux/kvm_types.h>
+#include <asm/kvm_gstage.h>
 
-struct kvm_gstage_mapping {
-	gpa_t addr;
-	pte_t pte;
-	u32 level;
-};
-
-int kvm_riscv_gstage_ioremap(struct kvm *kvm, gpa_t gpa,
-			     phys_addr_t hpa, unsigned long size,
-			     bool writable, bool in_atomic);
-void kvm_riscv_gstage_iounmap(struct kvm *kvm, gpa_t gpa,
-			      unsigned long size);
-int kvm_riscv_gstage_map(struct kvm_vcpu *vcpu,
-			 struct kvm_memory_slot *memslot,
-			 gpa_t gpa, unsigned long hva, bool is_write,
-			 struct kvm_gstage_mapping *out_map);
-int kvm_riscv_gstage_alloc_pgd(struct kvm *kvm);
-void kvm_riscv_gstage_free_pgd(struct kvm *kvm);
-void kvm_riscv_gstage_update_hgatp(struct kvm_vcpu *vcpu);
-void kvm_riscv_gstage_mode_detect(void);
-unsigned long kvm_riscv_gstage_mode(void);
-int kvm_riscv_gstage_gpa_bits(void);
+int kvm_riscv_mmu_ioremap(struct kvm *kvm, gpa_t gpa, phys_addr_t hpa,
+			  unsigned long size, bool writable, bool in_atomic);
+void kvm_riscv_mmu_iounmap(struct kvm *kvm, gpa_t gpa, unsigned long size);
+int kvm_riscv_mmu_map(struct kvm_vcpu *vcpu, struct kvm_memory_slot *memslot,
+		      gpa_t gpa, unsigned long hva, bool is_write,
+		      struct kvm_gstage_mapping *out_map);
+int kvm_riscv_mmu_alloc_pgd(struct kvm *kvm);
+void kvm_riscv_mmu_free_pgd(struct kvm *kvm);
+void kvm_riscv_mmu_update_hgatp(struct kvm_vcpu *vcpu);
 
 #endif

arch/riscv/kvm/Makefile

@@ -14,6 +14,7 @@ kvm-y += aia.o
 kvm-y += aia_aplic.o
 kvm-y += aia_device.o
 kvm-y += aia_imsic.o
+kvm-y += gstage.o
 kvm-y += main.o
 kvm-y += mmu.o
 kvm-y += nacl.o

arch/riscv/kvm/aia_imsic.c

@@ -746,9 +746,8 @@ void kvm_riscv_vcpu_aia_imsic_release(struct kvm_vcpu *vcpu)
 	 */
 
 	/* Purge the G-stage mapping */
-	kvm_riscv_gstage_iounmap(vcpu->kvm,
-				 vcpu->arch.aia_context.imsic_addr,
-				 IMSIC_MMIO_PAGE_SZ);
+	kvm_riscv_mmu_iounmap(vcpu->kvm, vcpu->arch.aia_context.imsic_addr,
+			      IMSIC_MMIO_PAGE_SZ);
 
 	/* TODO: Purge the IOMMU mapping ??? */
 
@@ -831,9 +830,9 @@ int kvm_riscv_vcpu_aia_imsic_update(struct kvm_vcpu *vcpu)
 	imsic_vsfile_local_clear(new_vsfile_hgei, imsic->nr_hw_eix);
 
 	/* Update G-stage mapping for the new IMSIC VS-file */
-	ret = kvm_riscv_gstage_ioremap(kvm, vcpu->arch.aia_context.imsic_addr,
-				       new_vsfile_pa, IMSIC_MMIO_PAGE_SZ,
-				       true, true);
+	ret = kvm_riscv_mmu_ioremap(kvm, vcpu->arch.aia_context.imsic_addr,
+				    new_vsfile_pa, IMSIC_MMIO_PAGE_SZ,
+				    true, true);
 	if (ret)
 		goto fail_free_vsfile_hgei;
 

arch/riscv/kvm/gstage.c

@@ -0,0 +1,337 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ * Copyright (c) 2025 Ventana Micro Systems Inc.
+ */
+
+#include <linux/bitops.h>
+#include <linux/errno.h>
+#include <linux/kvm_host.h>
+#include <linux/module.h>
+#include <linux/pgtable.h>
+#include <asm/kvm_gstage.h>
+
+#ifdef CONFIG_64BIT
+unsigned long kvm_riscv_gstage_mode __ro_after_init = HGATP_MODE_SV39X4;
+unsigned long kvm_riscv_gstage_pgd_levels __ro_after_init = 3;
+#else
+unsigned long kvm_riscv_gstage_mode __ro_after_init = HGATP_MODE_SV32X4;
+unsigned long kvm_riscv_gstage_pgd_levels __ro_after_init = 2;
+#endif
+
+#define gstage_pte_leaf(__ptep)	\
+	(pte_val(*(__ptep)) & (_PAGE_READ | _PAGE_WRITE | _PAGE_EXEC))
+
+static inline unsigned long gstage_pte_index(gpa_t addr, u32 level)
+{
+	unsigned long mask;
+	unsigned long shift = HGATP_PAGE_SHIFT + (kvm_riscv_gstage_index_bits * level);
+
+	if (level == (kvm_riscv_gstage_pgd_levels - 1))
+		mask = (PTRS_PER_PTE * (1UL << kvm_riscv_gstage_pgd_xbits)) - 1;
+	else
+		mask = PTRS_PER_PTE - 1;
+
+	return (addr >> shift) & mask;
+}
+
+static inline unsigned long gstage_pte_page_vaddr(pte_t pte)
+{
+	return (unsigned long)pfn_to_virt(__page_val_to_pfn(pte_val(pte)));
+}
+
+static int gstage_page_size_to_level(unsigned long page_size, u32 *out_level)
+{
+	u32 i;
+	unsigned long psz = 1UL << 12;
+
+	for (i = 0; i < kvm_riscv_gstage_pgd_levels; i++) {
+		if (page_size == (psz << (i * kvm_riscv_gstage_index_bits))) {
+			*out_level = i;
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
+static int gstage_level_to_page_order(u32 level, unsigned long *out_pgorder)
+{
+	if (kvm_riscv_gstage_pgd_levels < level)
+		return -EINVAL;
+
+	*out_pgorder = 12 + (level * kvm_riscv_gstage_index_bits);
+	return 0;
+}
+
+static int gstage_level_to_page_size(u32 level, unsigned long *out_pgsize)
+{
+	int rc;
+	unsigned long page_order = PAGE_SHIFT;
+
+	rc = gstage_level_to_page_order(level, &page_order);
+	if (rc)
+		return rc;
+
+	*out_pgsize = BIT(page_order);
+	return 0;
+}
+
+bool kvm_riscv_gstage_get_leaf(struct kvm_gstage *gstage, gpa_t addr,
+			       pte_t **ptepp, u32 *ptep_level)
+{
+	pte_t *ptep;
+	u32 current_level = kvm_riscv_gstage_pgd_levels - 1;
+
+	*ptep_level = current_level;
+	ptep = (pte_t *)gstage->pgd;
+	ptep = &ptep[gstage_pte_index(addr, current_level)];
+	while (ptep && pte_val(ptep_get(ptep))) {
+		if (gstage_pte_leaf(ptep)) {
+			*ptep_level = current_level;
+			*ptepp = ptep;
+			return true;
+		}
+
+		if (current_level) {
+			current_level--;
+			*ptep_level = current_level;
+			ptep = (pte_t *)gstage_pte_page_vaddr(ptep_get(ptep));
+			ptep = &ptep[gstage_pte_index(addr, current_level)];
+		} else {
+			ptep = NULL;
+		}
+	}
+
+	return false;
+}
+
+static void gstage_tlb_flush(struct kvm_gstage *gstage, u32 level, gpa_t addr)
+{
+	unsigned long order = PAGE_SHIFT;
+
+	if (gstage_level_to_page_order(level, &order))
+		return;
+	addr &= ~(BIT(order) - 1);
+
+	if (gstage->flags & KVM_GSTAGE_FLAGS_LOCAL)
+		kvm_riscv_local_hfence_gvma_vmid_gpa(gstage->vmid, addr, BIT(order), order);
+	else
+		kvm_riscv_hfence_gvma_vmid_gpa(gstage->kvm, -1UL, 0, addr, BIT(order), order);
+}
+
+int kvm_riscv_gstage_set_pte(struct kvm_gstage *gstage,
+			     struct kvm_mmu_memory_cache *pcache,
+			     const struct kvm_gstage_mapping *map)
+{
+	u32 current_level = kvm_riscv_gstage_pgd_levels - 1;
+	pte_t *next_ptep = (pte_t *)gstage->pgd;
+	pte_t *ptep = &next_ptep[gstage_pte_index(map->addr, current_level)];
+
+	if (current_level < map->level)
+		return -EINVAL;
+
+	while (current_level != map->level) {
+		if (gstage_pte_leaf(ptep))
+			return -EEXIST;
+
+		if (!pte_val(ptep_get(ptep))) {
+			if (!pcache)
+				return -ENOMEM;
+			next_ptep = kvm_mmu_memory_cache_alloc(pcache);
+			if (!next_ptep)
+				return -ENOMEM;
+			set_pte(ptep, pfn_pte(PFN_DOWN(__pa(next_ptep)),
+					      __pgprot(_PAGE_TABLE)));
+		} else {
+			if (gstage_pte_leaf(ptep))
+				return -EEXIST;
+			next_ptep = (pte_t *)gstage_pte_page_vaddr(ptep_get(ptep));
+		}
+
+		current_level--;
+		ptep = &next_ptep[gstage_pte_index(map->addr, current_level)];
+	}
+
+	if (pte_val(*ptep) != pte_val(map->pte)) {
+		set_pte(ptep, map->pte);
+		if (gstage_pte_leaf(ptep))
+			gstage_tlb_flush(gstage, current_level, map->addr);
+	}
+
+	return 0;
+}
+
+int kvm_riscv_gstage_map_page(struct kvm_gstage *gstage,
+			      struct kvm_mmu_memory_cache *pcache,
+			      gpa_t gpa, phys_addr_t hpa, unsigned long page_size,
+			      bool page_rdonly, bool page_exec,
+			      struct kvm_gstage_mapping *out_map)
+{
+	pgprot_t prot;
+	int ret;
+
+	out_map->addr = gpa;
+	out_map->level = 0;
+
+	ret = gstage_page_size_to_level(page_size, &out_map->level);
+	if (ret)
+		return ret;
+
+	/*
+	 * A RISC-V implementation can choose to either:
+	 * 1) Update 'A' and 'D' PTE bits in hardware
+	 * 2) Generate page fault when 'A' and/or 'D' bits are not set
+	 *    PTE so that software can update these bits.
+	 *
+	 * We support both options mentioned above. To achieve this, we
+	 * always set 'A' and 'D' PTE bits at time of creating G-stage
+	 * mapping. To support KVM dirty page logging with both options
+	 * mentioned above, we will write-protect G-stage PTEs to track
+	 * dirty pages.
+	 */
+
+	if (page_exec) {
+		if (page_rdonly)
+			prot = PAGE_READ_EXEC;
+		else
+			prot = PAGE_WRITE_EXEC;
+	} else {
+		if (page_rdonly)
+			prot = PAGE_READ;
+		else
+			prot = PAGE_WRITE;
+	}
+	out_map->pte = pfn_pte(PFN_DOWN(hpa), prot);
+	out_map->pte = pte_mkdirty(out_map->pte);
+
+	return kvm_riscv_gstage_set_pte(gstage, pcache, out_map);
+}
+
+void kvm_riscv_gstage_op_pte(struct kvm_gstage *gstage, gpa_t addr,
+			     pte_t *ptep, u32 ptep_level, enum kvm_riscv_gstage_op op)
+{
+	int i, ret;
+	pte_t old_pte, *next_ptep;
+	u32 next_ptep_level;
+	unsigned long next_page_size, page_size;
+
+	ret = gstage_level_to_page_size(ptep_level, &page_size);
+	if (ret)
+		return;
+
+	WARN_ON(addr & (page_size - 1));
+
+	if (!pte_val(ptep_get(ptep)))
+		return;
+
+	if (ptep_level && !gstage_pte_leaf(ptep)) {
+		next_ptep = (pte_t *)gstage_pte_page_vaddr(ptep_get(ptep));
+		next_ptep_level = ptep_level - 1;
+		ret = gstage_level_to_page_size(next_ptep_level, &next_page_size);
+		if (ret)
+			return;
+
+		if (op == GSTAGE_OP_CLEAR)
+			set_pte(ptep, __pte(0));
+		for (i = 0; i < PTRS_PER_PTE; i++)
+			kvm_riscv_gstage_op_pte(gstage, addr + i * next_page_size,
+						&next_ptep[i], next_ptep_level, op);
+		if (op == GSTAGE_OP_CLEAR)
+			put_page(virt_to_page(next_ptep));
+	} else {
+		old_pte = *ptep;
+		if (op == GSTAGE_OP_CLEAR)
+			set_pte(ptep, __pte(0));
+		else if (op == GSTAGE_OP_WP)
+			set_pte(ptep, __pte(pte_val(ptep_get(ptep)) & ~_PAGE_WRITE));
+		if (pte_val(*ptep) != pte_val(old_pte))
+			gstage_tlb_flush(gstage, ptep_level, addr);
+	}
+}
+
+void kvm_riscv_gstage_unmap_range(struct kvm_gstage *gstage,
+				  gpa_t start, gpa_t size, bool may_block)
+{
+	int ret;
+	pte_t *ptep;
+	u32 ptep_level;
+	bool found_leaf;
+	unsigned long page_size;
+	gpa_t addr = start, end = start + size;
+
+	while (addr < end) {
+		found_leaf = kvm_riscv_gstage_get_leaf(gstage, addr, &ptep, &ptep_level);
+		ret = gstage_level_to_page_size(ptep_level, &page_size);
+		if (ret)
+			break;
+
+		if (!found_leaf)
+			goto next;
+
+		if (!(addr & (page_size - 1)) && ((end - addr) >= page_size))
+			kvm_riscv_gstage_op_pte(gstage, addr, ptep,
+						ptep_level, GSTAGE_OP_CLEAR);
+
+next:
+		addr += page_size;
+
+		/*
+		 * If the range is too large, release the kvm->mmu_lock
+		 * to prevent starvation and lockup detector warnings.
+		 */
+		if (!(gstage->flags & KVM_GSTAGE_FLAGS_LOCAL) && may_block && addr < end)
+			cond_resched_lock(&gstage->kvm->mmu_lock);
+	}
+}
+
+void kvm_riscv_gstage_wp_range(struct kvm_gstage *gstage, gpa_t start, gpa_t end)
+{
+	int ret;
+	pte_t *ptep;
+	u32 ptep_level;
+	bool found_leaf;
+	gpa_t addr = start;
+	unsigned long page_size;
+
+	while (addr < end) {
+		found_leaf = kvm_riscv_gstage_get_leaf(gstage, addr, &ptep, &ptep_level);
+		ret = gstage_level_to_page_size(ptep_level, &page_size);
+		if (ret)
+			break;
+
+		if (!found_leaf)
+			goto next;
+
+		if (!(addr & (page_size - 1)) && ((end - addr) >= page_size))
+			kvm_riscv_gstage_op_pte(gstage, addr, ptep,
+						ptep_level, GSTAGE_OP_WP);
+
+next:
+		addr += page_size;
+	}
+}
+
+void __init kvm_riscv_gstage_mode_detect(void)
+{
+#ifdef CONFIG_64BIT
+	/* Try Sv57x4 G-stage mode */
+	csr_write(CSR_HGATP, HGATP_MODE_SV57X4 << HGATP_MODE_SHIFT);
+	if ((csr_read(CSR_HGATP) >> HGATP_MODE_SHIFT) == HGATP_MODE_SV57X4) {
+		kvm_riscv_gstage_mode = HGATP_MODE_SV57X4;
+		kvm_riscv_gstage_pgd_levels = 5;
+		goto skip_sv48x4_test;
+	}
+
+	/* Try Sv48x4 G-stage mode */
+	csr_write(CSR_HGATP, HGATP_MODE_SV48X4 << HGATP_MODE_SHIFT);
+	if ((csr_read(CSR_HGATP) >> HGATP_MODE_SHIFT) == HGATP_MODE_SV48X4) {
+		kvm_riscv_gstage_mode = HGATP_MODE_SV48X4;
+		kvm_riscv_gstage_pgd_levels = 4;
+	}
+skip_sv48x4_test:
+
+	csr_write(CSR_HGATP, 0);
+	kvm_riscv_local_hfence_gvma_all();
+#endif
+}

arch/riscv/kvm/main.c

@@ -135,7 +135,7 @@ static int __init riscv_kvm_init(void)
 			 (rc) ? slist : "no features");
 	}
 
-	switch (kvm_riscv_gstage_mode()) {
+	switch (kvm_riscv_gstage_mode) {
 	case HGATP_MODE_SV32X4:
 		str = "Sv32x4";
 		break;

arch/riscv/kvm/vcpu.c

@@ -622,7 +622,7 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 		}
 	}
 
-	kvm_riscv_gstage_update_hgatp(vcpu);
+	kvm_riscv_mmu_update_hgatp(vcpu);
 
 	kvm_riscv_vcpu_timer_restore(vcpu);
 
@@ -707,7 +707,7 @@ static void kvm_riscv_check_vcpu_requests(struct kvm_vcpu *vcpu)
 			kvm_riscv_reset_vcpu(vcpu, true);
 
 		if (kvm_check_request(KVM_REQ_UPDATE_HGATP, vcpu))
-			kvm_riscv_gstage_update_hgatp(vcpu);
+			kvm_riscv_mmu_update_hgatp(vcpu);
 
 		if (kvm_check_request(KVM_REQ_FENCE_I, vcpu))
 			kvm_riscv_fence_i_process(vcpu);

arch/riscv/kvm/vcpu_exit.c

@@ -43,8 +43,9 @@ static int gstage_page_fault(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		};
 	}
 
-	ret = kvm_riscv_gstage_map(vcpu, memslot, fault_addr, hva,
-		(trap->scause == EXC_STORE_GUEST_PAGE_FAULT) ? true : false, &host_map);
+	ret = kvm_riscv_mmu_map(vcpu, memslot, fault_addr, hva,
+				(trap->scause == EXC_STORE_GUEST_PAGE_FAULT) ? true : false,
+				&host_map);
 	if (ret < 0)
 		return ret;
 

arch/riscv/kvm/vm.c

@@ -32,13 +32,13 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 {
 	int r;
 
-	r = kvm_riscv_gstage_alloc_pgd(kvm);
+	r = kvm_riscv_mmu_alloc_pgd(kvm);
 	if (r)
 		return r;
 
 	r = kvm_riscv_gstage_vmid_init(kvm);
 	if (r) {
-		kvm_riscv_gstage_free_pgd(kvm);
+		kvm_riscv_mmu_free_pgd(kvm);
 		return r;
 	}
 
@@ -200,7 +200,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 		r = KVM_USER_MEM_SLOTS;
 		break;
 	case KVM_CAP_VM_GPA_BITS:
-		r = kvm_riscv_gstage_gpa_bits();
+		r = kvm_riscv_gstage_gpa_bits;
 		break;
 	default:
 		r = 0;