[PATCH 0/7]KVM: selftests: Clean up x86 page walker utilities

[PATCH 0/7]KVM: selftests: Clean up x86 page walker utilities

[Sean Christopherson]

Clean up and dedup x86 utilities that walk page tables, e.g. to insert new
mappings or to retrieve existing mappings.  There's no end user that
benefits from the refactoring; I did all the work thinking that I needed
some super fancy test to hit an NX huge page bug...

Patch 1 has been posted by Vitaly (at my request)[*], but should be
identical, i.e. the order shouldn't matter.

[*] https://lore.kernel.org/all/20221004123956.188909-37-vkuznets@redhat.com

Sean Christopherson (7):
  KVM: selftests: Drop helpers to read/write page table entries
  KVM: selftests: Drop reserved bit checks from PTE accessor
  KVM: selftests: Remove useless shifts when creating guest page tables
  KVM: selftests: Verify parent PTE is PRESENT when getting child PTE
  KVM: selftests: Use virt_get_pte() when getting PTE pointer
  KVM: selftests: Use vm_get_page_table_entry() in addr_arch_gva2gpa()
  KVM: selftests: Play nice with huge pages when getting PTEs/GPAs

 .../selftests/kvm/include/x86_64/processor.h  |  19 ++-
 .../selftests/kvm/lib/x86_64/processor.c      | 159 ++++++------------
 .../kvm/x86_64/emulator_error_test.c          |   8 +-
 3 files changed, 70 insertions(+), 116 deletions(-)


base-commit: e18d6152ff0f41b7f01f9817372022df04e0d354
-- 
2.38.0.rc1.362.ged0d419d3c-goog

[PATCH 1/7] KVM: selftests: Drop helpers to read/write page table entries

[Sean Christopherson]

Drop vm_{g,s}et_page_table_entry() and instead expose the "inner"
helper (was _vm_get_page_table_entry()) that returns a _pointer_ to the
PTE, i.e. let tests directly modify PTEs instead of bouncing through
helpers that just make life difficult.

Opportunsitically use BIT_ULL() in emulator_error_test, and use the
MAXPHYADDR define to set the "rogue" GPA bit instead of open coding the
same value.

No functional change intended.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 .../selftests/kvm/include/x86_64/processor.h  |  6 ++----
 .../selftests/kvm/lib/x86_64/processor.c      | 21 ++-----------------
 .../kvm/x86_64/emulator_error_test.c          |  6 ++++--
 3 files changed, 8 insertions(+), 25 deletions(-)

diff --git a/tools/testing/selftests/kvm/include/x86_64/processor.h b/tools/testing/selftests/kvm/include/x86_64/processor.h
index e8ca0d8a6a7e..30d5df1ebaad 100644
--- a/tools/testing/selftests/kvm/include/x86_64/processor.h
+++ b/tools/testing/selftests/kvm/include/x86_64/processor.h
@@ -827,10 +827,8 @@ static inline uint8_t wrmsr_safe(uint32_t msr, uint64_t val)
 
 bool kvm_is_tdp_enabled(void);
 
-uint64_t vm_get_page_table_entry(struct kvm_vm *vm, struct kvm_vcpu *vcpu,
-				 uint64_t vaddr);
-void vm_set_page_table_entry(struct kvm_vm *vm, struct kvm_vcpu *vcpu,
-			     uint64_t vaddr, uint64_t pte);
+uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, struct kvm_vcpu *vcpu,
+				  uint64_t vaddr);
 
 uint64_t kvm_hypercall(uint64_t nr, uint64_t a0, uint64_t a1, uint64_t a2,
 		       uint64_t a3);
diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c
index 39c4409ef56a..90b35998b0f3 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c
@@ -241,9 +241,8 @@ void virt_map_level(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr,
 	}
 }
 
-static uint64_t *_vm_get_page_table_entry(struct kvm_vm *vm,
-					  struct kvm_vcpu *vcpu,
-					  uint64_t vaddr)
+uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, struct kvm_vcpu *vcpu,
+				  uint64_t vaddr)
 {
 	uint16_t index[4];
 	uint64_t *pml4e, *pdpe, *pde;
@@ -313,22 +312,6 @@ static uint64_t *_vm_get_page_table_entry(struct kvm_vm *vm,
 	return &pte[index[0]];
 }
 
-uint64_t vm_get_page_table_entry(struct kvm_vm *vm, struct kvm_vcpu *vcpu,
-				 uint64_t vaddr)
-{
-	uint64_t *pte = _vm_get_page_table_entry(vm, vcpu, vaddr);
-
-	return *(uint64_t *)pte;
-}
-
-void vm_set_page_table_entry(struct kvm_vm *vm, struct kvm_vcpu *vcpu,
-			     uint64_t vaddr, uint64_t pte)
-{
-	uint64_t *new_pte = _vm_get_page_table_entry(vm, vcpu, vaddr);
-
-	*(uint64_t *)new_pte = pte;
-}
-
 void virt_arch_dump(FILE *stream, struct kvm_vm *vm, uint8_t indent)
 {
 	uint64_t *pml4e, *pml4e_start;
diff --git a/tools/testing/selftests/kvm/x86_64/emulator_error_test.c b/tools/testing/selftests/kvm/x86_64/emulator_error_test.c
index 236e11755ba6..bde247f3c8a1 100644
--- a/tools/testing/selftests/kvm/x86_64/emulator_error_test.c
+++ b/tools/testing/selftests/kvm/x86_64/emulator_error_test.c
@@ -152,8 +152,9 @@ int main(int argc, char *argv[])
 {
 	struct kvm_vcpu *vcpu;
 	struct kvm_vm *vm;
-	uint64_t gpa, pte;
+	uint64_t *pte;
 	uint64_t *hva;
+	uint64_t gpa;
 	int rc;
 
 	/* Tell stdout not to buffer its content */
@@ -178,8 +179,9 @@ int main(int argc, char *argv[])
 	virt_map(vm, MEM_REGION_GVA, MEM_REGION_GPA, 1);
 	hva = addr_gpa2hva(vm, MEM_REGION_GPA);
 	memset(hva, 0, PAGE_SIZE);
+
 	pte = vm_get_page_table_entry(vm, vcpu, MEM_REGION_GVA);
-	vm_set_page_table_entry(vm, vcpu, MEM_REGION_GVA, pte | (1ull << 36));
+	*pte |= BIT_ULL(MAXPHYADDR);
 
 	vcpu_run(vcpu);
 	process_exit_on_emulation_error(vcpu);
-- 
2.38.0.rc1.362.ged0d419d3c-goog

[PATCH 2/7] KVM: selftests: Drop reserved bit checks from PTE accessor

[Sean Christopherson]

Drop the reserved bit checks from the helper to retrieve a PTE, there's
very little value in sanity checking the constructed page tables as any
will quickly be noticed in the form of an unexpected #PF.  The checks
also place unnecessary restrictions on the usage of the helpers, e.g. if
a test _wanted_ to set reserved bits for whatever reason.

Removing the NX check in particular allows for the removal of the @vcpu
param, which will in turn allow the helper to be reused nearly verbatim
for addr_gva2gpa().

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 .../selftests/kvm/include/x86_64/processor.h  |  3 +--
 .../selftests/kvm/lib/x86_64/processor.c      | 26 +------------------
 .../kvm/x86_64/emulator_error_test.c          |  2 +-
 3 files changed, 3 insertions(+), 28 deletions(-)

diff --git a/tools/testing/selftests/kvm/include/x86_64/processor.h b/tools/testing/selftests/kvm/include/x86_64/processor.h
index 30d5df1ebaad..53d52a5ace48 100644
--- a/tools/testing/selftests/kvm/include/x86_64/processor.h
+++ b/tools/testing/selftests/kvm/include/x86_64/processor.h
@@ -827,8 +827,7 @@ static inline uint8_t wrmsr_safe(uint32_t msr, uint64_t val)
 
 bool kvm_is_tdp_enabled(void);
 
-uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, struct kvm_vcpu *vcpu,
-				  uint64_t vaddr);
+uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr);
 
 uint64_t kvm_hypercall(uint64_t nr, uint64_t a0, uint64_t a1, uint64_t a2,
 		       uint64_t a3);
diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c
index 90b35998b0f3..9e196837a794 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c
@@ -241,29 +241,11 @@ void virt_map_level(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr,
 	}
 }
 
-uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, struct kvm_vcpu *vcpu,
-				  uint64_t vaddr)
+uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr)
 {
 	uint16_t index[4];
 	uint64_t *pml4e, *pdpe, *pde;
 	uint64_t *pte;
-	struct kvm_sregs sregs;
-	uint64_t rsvd_mask = 0;
-
-	/* Set the high bits in the reserved mask. */
-	if (vm->pa_bits < 52)
-		rsvd_mask = GENMASK_ULL(51, vm->pa_bits);
-
-	/*
-	 * SDM vol 3, fig 4-11 "Formats of CR3 and Paging-Structure Entries
-	 * with 4-Level Paging and 5-Level Paging".
-	 * If IA32_EFER.NXE = 0 and the P flag of a paging-structure entry is 1,
-	 * the XD flag (bit 63) is reserved.
-	 */
-	vcpu_sregs_get(vcpu, &sregs);
-	if ((sregs.efer & EFER_NX) == 0) {
-		rsvd_mask |= PTE_NX_MASK;
-	}
 
 	TEST_ASSERT(vm->mode == VM_MODE_PXXV48_4K, "Attempt to use "
 		"unknown or unsupported guest mode, mode: 0x%x", vm->mode);
@@ -286,24 +268,18 @@ uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, struct kvm_vcpu *vcpu,
 	pml4e = addr_gpa2hva(vm, vm->pgd);
 	TEST_ASSERT(pml4e[index[3]] & PTE_PRESENT_MASK,
 		"Expected pml4e to be present for gva: 0x%08lx", vaddr);
-	TEST_ASSERT((pml4e[index[3]] & (rsvd_mask | PTE_LARGE_MASK)) == 0,
-		"Unexpected reserved bits set.");
 
 	pdpe = addr_gpa2hva(vm, PTE_GET_PFN(pml4e[index[3]]) * vm->page_size);
 	TEST_ASSERT(pdpe[index[2]] & PTE_PRESENT_MASK,
 		"Expected pdpe to be present for gva: 0x%08lx", vaddr);
 	TEST_ASSERT(!(pdpe[index[2]] & PTE_LARGE_MASK),
 		"Expected pdpe to map a pde not a 1-GByte page.");
-	TEST_ASSERT((pdpe[index[2]] & rsvd_mask) == 0,
-		"Unexpected reserved bits set.");
 
 	pde = addr_gpa2hva(vm, PTE_GET_PFN(pdpe[index[2]]) * vm->page_size);
 	TEST_ASSERT(pde[index[1]] & PTE_PRESENT_MASK,
 		"Expected pde to be present for gva: 0x%08lx", vaddr);
 	TEST_ASSERT(!(pde[index[1]] & PTE_LARGE_MASK),
 		"Expected pde to map a pte not a 2-MByte page.");
-	TEST_ASSERT((pde[index[1]] & rsvd_mask) == 0,
-		"Unexpected reserved bits set.");
 
 	pte = addr_gpa2hva(vm, PTE_GET_PFN(pde[index[1]]) * vm->page_size);
 	TEST_ASSERT(pte[index[0]] & PTE_PRESENT_MASK,
diff --git a/tools/testing/selftests/kvm/x86_64/emulator_error_test.c b/tools/testing/selftests/kvm/x86_64/emulator_error_test.c
index bde247f3c8a1..1abb34735754 100644
--- a/tools/testing/selftests/kvm/x86_64/emulator_error_test.c
+++ b/tools/testing/selftests/kvm/x86_64/emulator_error_test.c
@@ -180,7 +180,7 @@ int main(int argc, char *argv[])
 	hva = addr_gpa2hva(vm, MEM_REGION_GPA);
 	memset(hva, 0, PAGE_SIZE);
 
-	pte = vm_get_page_table_entry(vm, vcpu, MEM_REGION_GVA);
+	pte = vm_get_page_table_entry(vm, MEM_REGION_GVA);
 	*pte |= BIT_ULL(MAXPHYADDR);
 
 	vcpu_run(vcpu);
-- 
2.38.0.rc1.362.ged0d419d3c-goog

[PATCH 3/7] KVM: selftests: Remove useless shifts when creating guest page tables

[Sean Christopherson]

Remove the pointless shift from GPA=>GFN and immediately back to
GFN=>GPA when creating guest page tables.  Ignore the other walkers
that have a similar pattern for the moment, they will be converted
to use virt_get_pte() in the near future.

No functional change intended.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 .../selftests/kvm/include/x86_64/processor.h    |  3 ++-
 .../selftests/kvm/lib/x86_64/processor.c        | 17 ++++++++---------
 2 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/tools/testing/selftests/kvm/include/x86_64/processor.h b/tools/testing/selftests/kvm/include/x86_64/processor.h
index 53d52a5ace48..9676a3464758 100644
--- a/tools/testing/selftests/kvm/include/x86_64/processor.h
+++ b/tools/testing/selftests/kvm/include/x86_64/processor.h
@@ -177,7 +177,8 @@ struct kvm_x86_cpu_feature {
 #define PAGE_MASK		(~(PAGE_SIZE-1))
 
 #define PHYSICAL_PAGE_MASK      GENMASK_ULL(51, 12)
-#define PTE_GET_PFN(pte)        (((pte) & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT)
+#define PTE_GET_PA(pte)		((pte) & PHYSICAL_PAGE_MASK)
+#define PTE_GET_PFN(pte)        (PTE_GET_PA(pte) >> PAGE_SHIFT)
 
 /* General Registers in 64-Bit Mode */
 struct gpr64_regs {
diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c
index 9e196837a794..324bf24564a1 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c
@@ -131,23 +131,23 @@ void virt_arch_pgd_alloc(struct kvm_vm *vm)
 	}
 }
 
-static void *virt_get_pte(struct kvm_vm *vm, uint64_t pt_pfn, uint64_t vaddr,
+static void *virt_get_pte(struct kvm_vm *vm, uint64_t pt_gpa, uint64_t vaddr,
 			  int level)
 {
-	uint64_t *page_table = addr_gpa2hva(vm, pt_pfn << vm->page_shift);
+	uint64_t *page_table = addr_gpa2hva(vm, pt_gpa);
 	int index = (vaddr >> PG_LEVEL_SHIFT(level)) & 0x1ffu;
 
 	return &page_table[index];
 }
 
 static uint64_t *virt_create_upper_pte(struct kvm_vm *vm,
-				       uint64_t pt_pfn,
+				       uint64_t pt_gpa,
 				       uint64_t vaddr,
 				       uint64_t paddr,
 				       int current_level,
 				       int target_level)
 {
-	uint64_t *pte = virt_get_pte(vm, pt_pfn, vaddr, current_level);
+	uint64_t *pte = virt_get_pte(vm, pt_gpa, vaddr, current_level);
 
 	if (!(*pte & PTE_PRESENT_MASK)) {
 		*pte = PTE_PRESENT_MASK | PTE_WRITABLE_MASK;
@@ -197,21 +197,20 @@ void __virt_pg_map(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr, int level)
 	 * Allocate upper level page tables, if not already present.  Return
 	 * early if a hugepage was created.
 	 */
-	pml4e = virt_create_upper_pte(vm, vm->pgd >> vm->page_shift,
-				      vaddr, paddr, PG_LEVEL_512G, level);
+	pml4e = virt_create_upper_pte(vm, vm->pgd, vaddr, paddr, PG_LEVEL_512G, level);
 	if (*pml4e & PTE_LARGE_MASK)
 		return;
 
-	pdpe = virt_create_upper_pte(vm, PTE_GET_PFN(*pml4e), vaddr, paddr, PG_LEVEL_1G, level);
+	pdpe = virt_create_upper_pte(vm, PTE_GET_PA(*pml4e), vaddr, paddr, PG_LEVEL_1G, level);
 	if (*pdpe & PTE_LARGE_MASK)
 		return;
 
-	pde = virt_create_upper_pte(vm, PTE_GET_PFN(*pdpe), vaddr, paddr, PG_LEVEL_2M, level);
+	pde = virt_create_upper_pte(vm, PTE_GET_PA(*pdpe), vaddr, paddr, PG_LEVEL_2M, level);
 	if (*pde & PTE_LARGE_MASK)
 		return;
 
 	/* Fill in page table entry. */
-	pte = virt_get_pte(vm, PTE_GET_PFN(*pde), vaddr, PG_LEVEL_4K);
+	pte = virt_get_pte(vm, PTE_GET_PA(*pde), vaddr, PG_LEVEL_4K);
 	TEST_ASSERT(!(*pte & PTE_PRESENT_MASK),
 		    "PTE already present for 4k page at vaddr: 0x%lx\n", vaddr);
 	*pte = PTE_PRESENT_MASK | PTE_WRITABLE_MASK | (paddr & PHYSICAL_PAGE_MASK);
-- 
2.38.0.rc1.362.ged0d419d3c-goog

[PATCH 4/7] KVM: selftests: Verify parent PTE is PRESENT when getting child PTE

[Sean Christopherson]

Verify the parent PTE is PRESENT when getting a child via virt_get_pte()
so that the helper can be used for getting PTEs/GPAs without losing
sanity checks that the walker isn't wandering into the weeds.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 .../selftests/kvm/lib/x86_64/processor.c      | 21 ++++++++++++-------
 1 file changed, 13 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c
index 324bf24564a1..c9649f19aca1 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c
@@ -131,23 +131,28 @@ void virt_arch_pgd_alloc(struct kvm_vm *vm)
 	}
 }
 
-static void *virt_get_pte(struct kvm_vm *vm, uint64_t pt_gpa, uint64_t vaddr,
-			  int level)
+static void *virt_get_pte(struct kvm_vm *vm, uint64_t *parent_pte,
+			  uint64_t vaddr, int level)
 {
+	uint64_t pt_gpa = PTE_GET_PA(*parent_pte);
 	uint64_t *page_table = addr_gpa2hva(vm, pt_gpa);
 	int index = (vaddr >> PG_LEVEL_SHIFT(level)) & 0x1ffu;
 
+	TEST_ASSERT((*parent_pte & PTE_PRESENT_MASK) || parent_pte == &vm->pgd,
+		    "Parent PTE (level %d) not PRESENT for gva: 0x%08lx",
+		    level + 1, vaddr);
+
 	return &page_table[index];
 }
 
 static uint64_t *virt_create_upper_pte(struct kvm_vm *vm,
-				       uint64_t pt_gpa,
+				       uint64_t *parent_pte,
 				       uint64_t vaddr,
 				       uint64_t paddr,
 				       int current_level,
 				       int target_level)
 {
-	uint64_t *pte = virt_get_pte(vm, pt_gpa, vaddr, current_level);
+	uint64_t *pte = virt_get_pte(vm, parent_pte, vaddr, current_level);
 
 	if (!(*pte & PTE_PRESENT_MASK)) {
 		*pte = PTE_PRESENT_MASK | PTE_WRITABLE_MASK;
@@ -197,20 +202,20 @@ void __virt_pg_map(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr, int level)
 	 * Allocate upper level page tables, if not already present.  Return
 	 * early if a hugepage was created.
 	 */
-	pml4e = virt_create_upper_pte(vm, vm->pgd, vaddr, paddr, PG_LEVEL_512G, level);
+	pml4e = virt_create_upper_pte(vm, &vm->pgd, vaddr, paddr, PG_LEVEL_512G, level);
 	if (*pml4e & PTE_LARGE_MASK)
 		return;
 
-	pdpe = virt_create_upper_pte(vm, PTE_GET_PA(*pml4e), vaddr, paddr, PG_LEVEL_1G, level);
+	pdpe = virt_create_upper_pte(vm, pml4e, vaddr, paddr, PG_LEVEL_1G, level);
 	if (*pdpe & PTE_LARGE_MASK)
 		return;
 
-	pde = virt_create_upper_pte(vm, PTE_GET_PA(*pdpe), vaddr, paddr, PG_LEVEL_2M, level);
+	pde = virt_create_upper_pte(vm, pdpe, vaddr, paddr, PG_LEVEL_2M, level);
 	if (*pde & PTE_LARGE_MASK)
 		return;
 
 	/* Fill in page table entry. */
-	pte = virt_get_pte(vm, PTE_GET_PA(*pde), vaddr, PG_LEVEL_4K);
+	pte = virt_get_pte(vm, pde, vaddr, PG_LEVEL_4K);
 	TEST_ASSERT(!(*pte & PTE_PRESENT_MASK),
 		    "PTE already present for 4k page at vaddr: 0x%lx\n", vaddr);
 	*pte = PTE_PRESENT_MASK | PTE_WRITABLE_MASK | (paddr & PHYSICAL_PAGE_MASK);
-- 
2.38.0.rc1.362.ged0d419d3c-goog

[PATCH 5/7] KVM: selftests: Use virt_get_pte() when getting PTE pointer

[Sean Christopherson]

Use virt_get_pte() in vm_get_page_table_entry() instead of open coding
equivalent code.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 .../selftests/kvm/lib/x86_64/processor.c      | 29 ++++---------------
 1 file changed, 6 insertions(+), 23 deletions(-)

diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c
index c9649f19aca1..09b550fd8815 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c
@@ -247,9 +247,7 @@ void virt_map_level(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr,
 
 uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr)
 {
-	uint16_t index[4];
 	uint64_t *pml4e, *pdpe, *pde;
-	uint64_t *pte;
 
 	TEST_ASSERT(vm->mode == VM_MODE_PXXV48_4K, "Attempt to use "
 		"unknown or unsupported guest mode, mode: 0x%x", vm->mode);
@@ -264,32 +262,17 @@ uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr)
 	TEST_ASSERT(vaddr == (((int64_t)vaddr << 16) >> 16),
 		"Canonical check failed.  The virtual address is invalid.");
 
-	index[0] = (vaddr >> 12) & 0x1ffu;
-	index[1] = (vaddr >> 21) & 0x1ffu;
-	index[2] = (vaddr >> 30) & 0x1ffu;
-	index[3] = (vaddr >> 39) & 0x1ffu;
+	pml4e = virt_get_pte(vm, &vm->pgd, vaddr, PG_LEVEL_512G);
 
-	pml4e = addr_gpa2hva(vm, vm->pgd);
-	TEST_ASSERT(pml4e[index[3]] & PTE_PRESENT_MASK,
-		"Expected pml4e to be present for gva: 0x%08lx", vaddr);
-
-	pdpe = addr_gpa2hva(vm, PTE_GET_PFN(pml4e[index[3]]) * vm->page_size);
-	TEST_ASSERT(pdpe[index[2]] & PTE_PRESENT_MASK,
-		"Expected pdpe to be present for gva: 0x%08lx", vaddr);
-	TEST_ASSERT(!(pdpe[index[2]] & PTE_LARGE_MASK),
+	pdpe = virt_get_pte(vm, pml4e, vaddr, PG_LEVEL_1G);
+	TEST_ASSERT(!(*pdpe & PTE_LARGE_MASK),
 		"Expected pdpe to map a pde not a 1-GByte page.");
 
-	pde = addr_gpa2hva(vm, PTE_GET_PFN(pdpe[index[2]]) * vm->page_size);
-	TEST_ASSERT(pde[index[1]] & PTE_PRESENT_MASK,
-		"Expected pde to be present for gva: 0x%08lx", vaddr);
-	TEST_ASSERT(!(pde[index[1]] & PTE_LARGE_MASK),
+	pde = virt_get_pte(vm, pdpe, vaddr, PG_LEVEL_2M);
+	TEST_ASSERT(!(*pde & PTE_LARGE_MASK),
 		"Expected pde to map a pte not a 2-MByte page.");
 
-	pte = addr_gpa2hva(vm, PTE_GET_PFN(pde[index[1]]) * vm->page_size);
-	TEST_ASSERT(pte[index[0]] & PTE_PRESENT_MASK,
-		"Expected pte to be present for gva: 0x%08lx", vaddr);
-
-	return &pte[index[0]];
+	return virt_get_pte(vm, pde, vaddr, PG_LEVEL_4K);
 }
 
 void virt_arch_dump(FILE *stream, struct kvm_vm *vm, uint8_t indent)
-- 
2.38.0.rc1.362.ged0d419d3c-goog

[PATCH 6/7] KVM: selftests: Use vm_get_page_table_entry() in addr_arch_gva2gpa()

[Sean Christopherson]

Use vm_get_page_table_entry() in addr_arch_gva2gpa() to get the leaf PTE
instead of manually walking page tables.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 .../selftests/kvm/lib/x86_64/processor.c      | 38 ++-----------------
 1 file changed, 4 insertions(+), 34 deletions(-)

diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c
index 09b550fd8815..053f64191122 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c
@@ -458,41 +458,11 @@ static void kvm_seg_set_kernel_data_64bit(struct kvm_vm *vm, uint16_t selector,
 
 vm_paddr_t addr_arch_gva2gpa(struct kvm_vm *vm, vm_vaddr_t gva)
 {
-	uint16_t index[4];
-	uint64_t *pml4e, *pdpe, *pde;
-	uint64_t *pte;
+	uint64_t *pte = vm_get_page_table_entry(vm, gva);
 
-	TEST_ASSERT(vm->mode == VM_MODE_PXXV48_4K, "Attempt to use "
-		"unknown or unsupported guest mode, mode: 0x%x", vm->mode);
-
-	index[0] = (gva >> 12) & 0x1ffu;
-	index[1] = (gva >> 21) & 0x1ffu;
-	index[2] = (gva >> 30) & 0x1ffu;
-	index[3] = (gva >> 39) & 0x1ffu;
-
-	if (!vm->pgd_created)
-		goto unmapped_gva;
-	pml4e = addr_gpa2hva(vm, vm->pgd);
-	if (!(pml4e[index[3]] & PTE_PRESENT_MASK))
-		goto unmapped_gva;
-
-	pdpe = addr_gpa2hva(vm, PTE_GET_PFN(pml4e[index[3]]) * vm->page_size);
-	if (!(pdpe[index[2]] & PTE_PRESENT_MASK))
-		goto unmapped_gva;
-
-	pde = addr_gpa2hva(vm, PTE_GET_PFN(pdpe[index[2]]) * vm->page_size);
-	if (!(pde[index[1]] & PTE_PRESENT_MASK))
-		goto unmapped_gva;
-
-	pte = addr_gpa2hva(vm, PTE_GET_PFN(pde[index[1]]) * vm->page_size);
-	if (!(pte[index[0]] & PTE_PRESENT_MASK))
-		goto unmapped_gva;
-
-	return (PTE_GET_PFN(pte[index[0]]) * vm->page_size) + (gva & ~PAGE_MASK);
-
-unmapped_gva:
-	TEST_FAIL("No mapping for vm virtual address, gva: 0x%lx", gva);
-	exit(EXIT_FAILURE);
+	TEST_ASSERT(*pte & PTE_PRESENT_MASK,
+		    "Leaf PTE not PRESENT for gva: 0x%08lx", gva);
+	return PTE_GET_PA(*pte) | (gva & ~PAGE_MASK);
 }
 
 static void kvm_setup_gdt(struct kvm_vm *vm, struct kvm_dtable *dt)
-- 
2.38.0.rc1.362.ged0d419d3c-goog

[PATCH 7/7] KVM: selftests: Play nice with huge pages when getting PTEs/GPAs

[Sean Christopherson]

Play nice with huge pages when getting PTEs and translating GVAs to GPAs,
there's no reason to disallow using huge pages in selftests.  Use
PG_LEVEL_NONE to indicate that the caller doesn't care about the mapping
level and just wants to get the pte+level.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 .../selftests/kvm/include/x86_64/processor.h  | 11 ++++-
 .../selftests/kvm/lib/x86_64/processor.c      | 45 ++++++++++++++++---
 2 files changed, 47 insertions(+), 9 deletions(-)

diff --git a/tools/testing/selftests/kvm/include/x86_64/processor.h b/tools/testing/selftests/kvm/include/x86_64/processor.h
index 9676a3464758..e000e35c948f 100644
--- a/tools/testing/selftests/kvm/include/x86_64/processor.h
+++ b/tools/testing/selftests/kvm/include/x86_64/processor.h
@@ -172,11 +172,16 @@ struct kvm_x86_cpu_feature {
 #define PTE_GLOBAL_MASK         BIT_ULL(8)
 #define PTE_NX_MASK             BIT_ULL(63)
 
+#define PHYSICAL_PAGE_MASK      GENMASK_ULL(51, 12)
+
 #define PAGE_SHIFT		12
 #define PAGE_SIZE		(1ULL << PAGE_SHIFT)
-#define PAGE_MASK		(~(PAGE_SIZE-1))
+#define PAGE_MASK		(~(PAGE_SIZE-1) & PHYSICAL_PAGE_MASK)
+
+#define HUGEPAGE_SHIFT(x)	(PAGE_SHIFT + (((x) - 1) * 9))
+#define HUGEPAGE_SIZE(x)	(1UL << HUGEPAGE_SHIFT(x))
+#define HUGEPAGE_MASK(x)	(~(HUGEPAGE_SIZE(x) - 1) & PHYSICAL_PAGE_MASK)
 
-#define PHYSICAL_PAGE_MASK      GENMASK_ULL(51, 12)
 #define PTE_GET_PA(pte)		((pte) & PHYSICAL_PAGE_MASK)
 #define PTE_GET_PFN(pte)        (PTE_GET_PA(pte) >> PAGE_SHIFT)
 
@@ -828,6 +833,8 @@ static inline uint8_t wrmsr_safe(uint32_t msr, uint64_t val)
 
 bool kvm_is_tdp_enabled(void);
 
+uint64_t *__vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr,
+				    int *level);
 uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr);
 
 uint64_t kvm_hypercall(uint64_t nr, uint64_t a0, uint64_t a1, uint64_t a2,
diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c
index 053f64191122..efa20d0f9927 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c
@@ -245,10 +245,26 @@ void virt_map_level(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr,
 	}
 }
 
-uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr)
+static bool vm_is_target_pte(uint64_t *pte, int *level, int current_level)
+{
+	if (*pte & PTE_LARGE_MASK) {
+		TEST_ASSERT(*level == PG_LEVEL_NONE ||
+			    *level == current_level,
+			    "Unexpected hugepage at level %d\n", current_level);
+		*level = current_level;
+	}
+
+	return *level == current_level;
+}
+
+uint64_t *__vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr,
+				    int *level)
 {
 	uint64_t *pml4e, *pdpe, *pde;
 
+	TEST_ASSERT(*level >= PG_LEVEL_NONE && *level < PG_LEVEL_NUM,
+		    "Invalid PG_LEVEL_* '%d'", *level);
+
 	TEST_ASSERT(vm->mode == VM_MODE_PXXV48_4K, "Attempt to use "
 		"unknown or unsupported guest mode, mode: 0x%x", vm->mode);
 	TEST_ASSERT(sparsebit_is_set(vm->vpages_valid,
@@ -263,18 +279,27 @@ uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr)
 		"Canonical check failed.  The virtual address is invalid.");
 
 	pml4e = virt_get_pte(vm, &vm->pgd, vaddr, PG_LEVEL_512G);
+	if (vm_is_target_pte(pml4e, level, PG_LEVEL_512G))
+		return pml4e;
 
 	pdpe = virt_get_pte(vm, pml4e, vaddr, PG_LEVEL_1G);
-	TEST_ASSERT(!(*pdpe & PTE_LARGE_MASK),
-		"Expected pdpe to map a pde not a 1-GByte page.");
+	if (vm_is_target_pte(pdpe, level, PG_LEVEL_1G))
+		return pdpe;
 
 	pde = virt_get_pte(vm, pdpe, vaddr, PG_LEVEL_2M);
-	TEST_ASSERT(!(*pde & PTE_LARGE_MASK),
-		"Expected pde to map a pte not a 2-MByte page.");
+	if (vm_is_target_pte(pde, level, PG_LEVEL_2M))
+		return pde;
 
 	return virt_get_pte(vm, pde, vaddr, PG_LEVEL_4K);
 }
 
+uint64_t *vm_get_page_table_entry(struct kvm_vm *vm, uint64_t vaddr)
+{
+	int level = PG_LEVEL_4K;
+
+	return __vm_get_page_table_entry(vm, vaddr, &level);
+}
+
 void virt_arch_dump(FILE *stream, struct kvm_vm *vm, uint8_t indent)
 {
 	uint64_t *pml4e, *pml4e_start;
@@ -458,11 +483,17 @@ static void kvm_seg_set_kernel_data_64bit(struct kvm_vm *vm, uint16_t selector,
 
 vm_paddr_t addr_arch_gva2gpa(struct kvm_vm *vm, vm_vaddr_t gva)
 {
-	uint64_t *pte = vm_get_page_table_entry(vm, gva);
+	int level = PG_LEVEL_NONE;
+	uint64_t *pte = __vm_get_page_table_entry(vm, gva, &level);
 
 	TEST_ASSERT(*pte & PTE_PRESENT_MASK,
 		    "Leaf PTE not PRESENT for gva: 0x%08lx", gva);
-	return PTE_GET_PA(*pte) | (gva & ~PAGE_MASK);
+
+	/*
+	 * No need for a hugepage mask on the PTE, x86-64 requires the "unused"
+	 * address bits to be zero.
+	 */
+	return PTE_GET_PA(*pte) | (gva & ~HUGEPAGE_MASK(level));
 }
 
 static void kvm_setup_gdt(struct kvm_vm *vm, struct kvm_dtable *dt)
-- 
2.38.0.rc1.362.ged0d419d3c-goog