[PATCH 0/5] KVM: selftests: Reserve memslot 1 for ucall MMIO

[PATCH 0/5] KVM: selftests: Reserve memslot 1 for ucall MMIO

[Tharit Tangkijwanichakul]

This series addresses the TODO in __vm_create() to reserve memslot 1
for the ucall MMIO region. It introduces a shared base for test
memslots, migrates existing test users, and adds a read-only region to
prevent overlapping allocations.

A regression test verifies overlap rejection and continued ucall
operation.

Testing:
- Intel Core i5-10210U and Rockchip RK3588 (arm64).

These tests are changed but not tested: 
- arm64/sea_to_user.c: not enough memory allocation.
- x86/nested_vmsave_vmload_test.c: requires X86_FEATURE_SVM.
- riscv/sbi_pmu_test.c: riscv hardware unavailability.

The following test failures reproduce on both the unmodified base
<8edffbc1f6272db167037e1a8c49b99e9cd04803> with host kernel 
7.3.0-rc1-gd4b7fb647204 on RK3588 and with this series applied:
- arm64/vgic_its_save.c
- arm64/hypercalls.c

These failures are unchanged by this series.

Tharit Tangkijwanichakul (5):
  KVM: selftests: Introduce a shared base for test memslots
  KVM: selftests: Use the shared memslot base in memslot_perf_test
  KVM: selftests: Use the shared memslot base in memstress
  KVM: selftests: Reserve memslot 1 for the ucall MMIO address
  KVM: selftests: Test protection of the ucall MMIO region

 tools/testing/selftests/kvm/Makefile.kvm      |  1 +
 .../testing/selftests/kvm/arm64/hypercalls.c  |  3 +-
 .../testing/selftests/kvm/arm64/sea_to_user.c |  2 +-
 .../kvm/arm64/stage2_block_transitions.c      |  2 +-
 .../selftests/kvm/arm64/vgic_its_save.c       |  2 +-
 .../selftests/kvm/arm64/vgic_lpi_stress.c     |  2 +-
 tools/testing/selftests/kvm/dirty_log_test.c  |  2 +-
 .../testing/selftests/kvm/guest_memfd_test.c  |  2 +-
 .../testing/selftests/kvm/include/kvm_util.h  |  1 +
 .../testing/selftests/kvm/include/memstress.h |  2 +-
 .../selftests/kvm/kvm_page_table_test.c       |  2 +-
 tools/testing/selftests/kvm/lib/kvm_util.c    | 19 ++++--
 .../testing/selftests/kvm/memslot_perf_test.c | 62 +++++++++++--------
 tools/testing/selftests/kvm/mmu_stress_test.c |  2 +-
 .../selftests/kvm/reserved_slot_test.c        | 61 ++++++++++++++++++
 .../selftests/kvm/riscv/sbi_pmu_test.c        |  3 +-
 tools/testing/selftests/kvm/steal_time.c      |  3 +-
 .../selftests/kvm/x86/nested_dirty_log_test.c |  2 +-
 .../kvm/x86/nested_vmsave_vmload_test.c       |  2 +-
 19 files changed, 128 insertions(+), 47 deletions(-)
 create mode 100644 tools/testing/selftests/kvm/reserved_slot_test.c

-- 
2.53.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH 1/5] KVM: selftests: Introduce a shared base for test memslots

[Tharit Tangkijwanichakul]

Replace hardcoded slot 1 assignments with TEST_MEM_SLOT_BASE
This prepares tests for reserving slot 1 for library use.

Signed-off-by: Tharit Tangkijwanichakul <tharitt97@gmail.com>
---
 tools/testing/selftests/kvm/arm64/hypercalls.c               | 3 ++-
 tools/testing/selftests/kvm/arm64/sea_to_user.c              | 2 +-
 tools/testing/selftests/kvm/arm64/stage2_block_transitions.c | 2 +-
 tools/testing/selftests/kvm/arm64/vgic_its_save.c            | 2 +-
 tools/testing/selftests/kvm/arm64/vgic_lpi_stress.c          | 2 +-
 tools/testing/selftests/kvm/dirty_log_test.c                 | 2 +-
 tools/testing/selftests/kvm/guest_memfd_test.c               | 2 +-
 tools/testing/selftests/kvm/include/kvm_util.h               | 1 +
 tools/testing/selftests/kvm/kvm_page_table_test.c            | 2 +-
 tools/testing/selftests/kvm/mmu_stress_test.c                | 2 +-
 tools/testing/selftests/kvm/riscv/sbi_pmu_test.c             | 3 ++-
 tools/testing/selftests/kvm/steal_time.c                     | 3 ++-
 tools/testing/selftests/kvm/x86/nested_dirty_log_test.c      | 2 +-
 tools/testing/selftests/kvm/x86/nested_vmsave_vmload_test.c  | 2 +-
 14 files changed, 17 insertions(+), 13 deletions(-)

diff --git a/tools/testing/selftests/kvm/arm64/hypercalls.c b/tools/testing/selftests/kvm/arm64/hypercalls.c
index 391e2ba58b0b..ff7c21738a96 100644
--- a/tools/testing/selftests/kvm/arm64/hypercalls.c
+++ b/tools/testing/selftests/kvm/arm64/hypercalls.c
@@ -191,7 +191,8 @@ static void steal_time_init(struct kvm_vcpu *vcpu)
 	unsigned int gpages;
 
 	gpages = vm_calc_num_guest_pages(VM_MODE_DEFAULT, STEAL_TIME_SIZE);
-	vm_userspace_mem_region_add(vcpu->vm, VM_MEM_SRC_ANONYMOUS, ST_GPA_BASE, 1, gpages, 0);
+	vm_userspace_mem_region_add(vcpu->vm, VM_MEM_SRC_ANONYMOUS, ST_GPA_BASE,
+				    TEST_MEM_SLOT_BASE, gpages, 0);
 
 	vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PVTIME_CTRL,
 			     KVM_ARM_VCPU_PVTIME_IPA, &st_ipa);
diff --git a/tools/testing/selftests/kvm/arm64/sea_to_user.c b/tools/testing/selftests/kvm/arm64/sea_to_user.c
index d4af6e0eed28..bb552debc61c 100644
--- a/tools/testing/selftests/kvm/arm64/sea_to_user.c
+++ b/tools/testing/selftests/kvm/arm64/sea_to_user.c
@@ -323,7 +323,7 @@ static struct kvm_vm *vm_create_with_sea_handler(struct kvm_vcpu **vcpu)
 		/*vm=*/vm,
 		/*src_type=*/src_type,
 		/*gpa=*/start_gpa,
-		/*slot=*/1,
+		/*slot=*/TEST_MEM_SLOT_BASE,
 		/*npages=*/num_guest_pages,
 		/*flags=*/0);
 
diff --git a/tools/testing/selftests/kvm/arm64/stage2_block_transitions.c b/tools/testing/selftests/kvm/arm64/stage2_block_transitions.c
index 5fd47f4ada1f..0e23d7b55782 100644
--- a/tools/testing/selftests/kvm/arm64/stage2_block_transitions.c
+++ b/tools/testing/selftests/kvm/arm64/stage2_block_transitions.c
@@ -41,7 +41,7 @@
 #include "test_util.h"
 #include "ucall.h"
 
-#define DATA_SLOT		1
+#define DATA_SLOT		TEST_MEM_SLOT_BASE
 #define TEST_GVA		0xc0000000UL
 #define BLOCK_SIZE		SZ_2M
 
diff --git a/tools/testing/selftests/kvm/arm64/vgic_its_save.c b/tools/testing/selftests/kvm/arm64/vgic_its_save.c
index 864da01539f3..e858a4aa741d 100644
--- a/tools/testing/selftests/kvm/arm64/vgic_its_save.c
+++ b/tools/testing/selftests/kvm/arm64/vgic_its_save.c
@@ -29,7 +29,7 @@
 #include "ucall.h"
 #include "vgic.h"
 
-#define TEST_MEMSLOT_INDEX	1
+#define TEST_MEMSLOT_INDEX	TEST_MEM_SLOT_BASE
 
 /* All three ITS table entry sizes are 8 bytes in ABI 0. */
 #define ESZ			8
diff --git a/tools/testing/selftests/kvm/arm64/vgic_lpi_stress.c b/tools/testing/selftests/kvm/arm64/vgic_lpi_stress.c
index 9a69c439a0f8..1d5c56481053 100644
--- a/tools/testing/selftests/kvm/arm64/vgic_lpi_stress.c
+++ b/tools/testing/selftests/kvm/arm64/vgic_lpi_stress.c
@@ -18,7 +18,7 @@
 #include "ucall.h"
 #include "vgic.h"
 
-#define TEST_MEMSLOT_INDEX	1
+#define TEST_MEMSLOT_INDEX	TEST_MEM_SLOT_BASE
 
 #define GIC_LPI_OFFSET	8192
 
diff --git a/tools/testing/selftests/kvm/dirty_log_test.c b/tools/testing/selftests/kvm/dirty_log_test.c
index af5eb0334a74..75084fe3bec6 100644
--- a/tools/testing/selftests/kvm/dirty_log_test.c
+++ b/tools/testing/selftests/kvm/dirty_log_test.c
@@ -26,7 +26,7 @@
 #define PAGE_SHIFT_4K  12
 
 /* The memory slot index to track dirty pages */
-#define TEST_MEM_SLOT_INDEX		1
+#define TEST_MEM_SLOT_INDEX		TEST_MEM_SLOT_BASE
 
 /* Default guest test virtual memory offset */
 #define DEFAULT_GUEST_TEST_MEM		0xc0000000
diff --git a/tools/testing/selftests/kvm/guest_memfd_test.c b/tools/testing/selftests/kvm/guest_memfd_test.c
index 2233d871a38f..231bfe238e31 100644
--- a/tools/testing/selftests/kvm/guest_memfd_test.c
+++ b/tools/testing/selftests/kvm/guest_memfd_test.c
@@ -510,7 +510,7 @@ static void test_guest_memfd_guest(void)
 	 * the PCI hole and other MMIO regions that need to be avoided.
 	 */
 	const gpa_t gpa = SZ_4G;
-	const int slot = 1;
+	const int slot = TEST_MEM_SLOT_BASE;
 
 	struct kvm_vcpu *vcpu;
 	struct kvm_vm *vm;
diff --git a/tools/testing/selftests/kvm/include/kvm_util.h b/tools/testing/selftests/kvm/include/kvm_util.h
index 5a7a455b5387..759d604d44ce 100644
--- a/tools/testing/selftests/kvm/include/kvm_util.h
+++ b/tools/testing/selftests/kvm/include/kvm_util.h
@@ -30,6 +30,7 @@
 
 #define KVM_DEV_PATH "/dev/kvm"
 #define KVM_MAX_VCPUS 512
+#define TEST_MEM_SLOT_BASE 1
 
 #define NSEC_PER_SEC 1000000000L
 
diff --git a/tools/testing/selftests/kvm/kvm_page_table_test.c b/tools/testing/selftests/kvm/kvm_page_table_test.c
index 974813ea8953..f18720109319 100644
--- a/tools/testing/selftests/kvm/kvm_page_table_test.c
+++ b/tools/testing/selftests/kvm/kvm_page_table_test.c
@@ -20,7 +20,7 @@
 #include "guest_modes.h"
 #include "ucall_common.h"
 
-#define TEST_MEM_SLOT_INDEX             1
+#define TEST_MEM_SLOT_INDEX             TEST_MEM_SLOT_BASE
 
 /* Default size(1GB) of the memory for testing */
 #define DEFAULT_TEST_MEM_SIZE		(1 << 30)
diff --git a/tools/testing/selftests/kvm/mmu_stress_test.c b/tools/testing/selftests/kvm/mmu_stress_test.c
index 741e343962ed..b965a8b9d214 100644
--- a/tools/testing/selftests/kvm/mmu_stress_test.c
+++ b/tools/testing/selftests/kvm/mmu_stress_test.c
@@ -273,7 +273,7 @@ int main(int argc, char *argv[])
 	 * 1gb-3gb,but it's simpler to skip straight to 4gb.
 	 */
 	const u64 start_gpa = SZ_4G;
-	const int first_slot = 1;
+	const int first_slot = TEST_MEM_SLOT_BASE;
 
 	struct timespec time_start, time_run1, time_reset, time_run2, time_ro, time_rw;
 	u64 max_gpa, gpa, slot_size, max_mem, i;
diff --git a/tools/testing/selftests/kvm/riscv/sbi_pmu_test.c b/tools/testing/selftests/kvm/riscv/sbi_pmu_test.c
index 20388f0b959d..6abd74870e01 100644
--- a/tools/testing/selftests/kvm/riscv/sbi_pmu_test.c
+++ b/tools/testing/selftests/kvm/riscv/sbi_pmu_test.c
@@ -605,7 +605,8 @@ static void test_vm_events_test(void *guest_code)
 static void test_vm_setup_snapshot_mem(struct kvm_vm *vm, struct kvm_vcpu *vcpu)
 {
 	/* PMU Snapshot requires single page only */
-	vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS, PMU_SNAPSHOT_GPA_BASE, 1, 1, 0);
+	vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS, PMU_SNAPSHOT_GPA_BASE,
+				    TEST_MEM_SLOT_BASE, 1, 0);
 	/* PMU_SNAPSHOT_GPA_BASE is identity mapped */
 	virt_map(vm, PMU_SNAPSHOT_GPA_BASE, PMU_SNAPSHOT_GPA_BASE, 1);
 
diff --git a/tools/testing/selftests/kvm/steal_time.c b/tools/testing/selftests/kvm/steal_time.c
index 785d19f9ee8f..e386580ef540 100644
--- a/tools/testing/selftests/kvm/steal_time.c
+++ b/tools/testing/selftests/kvm/steal_time.c
@@ -509,7 +509,8 @@ static struct kvm_vm *vm_create_steal_time(u32 nr_vcpus, void *guest_code,
 	/* Create a VM and an identity mapped memslot for the steal time structure */
 	vm = vm_create_with_vcpus(nr_vcpus, guest_code, vcpus);
 	gpages = vm_calc_num_guest_pages(VM_MODE_DEFAULT, STEAL_TIME_SIZE * nr_vcpus);
-	vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS, ST_GPA_BASE, 1, gpages, 0);
+	vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS, ST_GPA_BASE,
+				    TEST_MEM_SLOT_BASE, gpages, 0);
 	virt_map(vm, ST_GPA_BASE, ST_GPA_BASE, gpages);
 
 	return vm;
diff --git a/tools/testing/selftests/kvm/x86/nested_dirty_log_test.c b/tools/testing/selftests/kvm/x86/nested_dirty_log_test.c
index 26b474bf1353..e86d625b3e89 100644
--- a/tools/testing/selftests/kvm/x86/nested_dirty_log_test.c
+++ b/tools/testing/selftests/kvm/x86/nested_dirty_log_test.c
@@ -16,7 +16,7 @@
 #include "vmx.h"
 
 /* The memory slot index to track dirty pages */
-#define TEST_MEM_SLOT_INDEX		1
+#define TEST_MEM_SLOT_INDEX		TEST_MEM_SLOT_BASE
 
 /*
  * Allocate four pages total.  Two pages are used to verify that the KVM marks
diff --git a/tools/testing/selftests/kvm/x86/nested_vmsave_vmload_test.c b/tools/testing/selftests/kvm/x86/nested_vmsave_vmload_test.c
index a130759f39a1..4bc57e5e40e3 100644
--- a/tools/testing/selftests/kvm/x86/nested_vmsave_vmload_test.c
+++ b/tools/testing/selftests/kvm/x86/nested_vmsave_vmload_test.c
@@ -17,7 +17,7 @@
  * interpreted as a direct L1 GPA or translated using NPT as an L2 GPA, depends
  * on which VMCB is accessed.
  */
-#define TEST_MEM_SLOT_INDEX		1
+#define TEST_MEM_SLOT_INDEX		TEST_MEM_SLOT_BASE
 #define TEST_MEM_PAGES			2
 #define TEST_MEM_BASE			0xc0000000
 
-- 
2.53.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH 2/5] KVM: selftests: Use the shared memslot base in memslot_perf_test

[Tharit Tangkijwanichakul]

Use TEST_MEM_SLOT_BASE for test memslot IDs and account for the
reserved slots when calculating the available slot count. This
prepares memslot_perf_test for reserving slot 1 for library use.

Signed-off-by: Tharit Tangkijwanichakul <tharitt97@gmail.com>
---
 .../testing/selftests/kvm/memslot_perf_test.c | 62 +++++++++++--------
 1 file changed, 35 insertions(+), 27 deletions(-)

diff --git a/tools/testing/selftests/kvm/memslot_perf_test.c b/tools/testing/selftests/kvm/memslot_perf_test.c
index e1299611be32..cbb900ab58af 100644
--- a/tools/testing/selftests/kvm/memslot_perf_test.c
+++ b/tools/testing/selftests/kvm/memslot_perf_test.c
@@ -60,7 +60,7 @@
  * other memslot(s).
  *
  * We have different number of memory slots, excluding the reserved
- * memory slot 0, on various architectures and configurations. The
+ * memory slots 0 and 1, on various architectures and configurations. The
  * memory size in this test is calculated by picking the maximal
  * last memory slot's memory size, with alignment to the largest
  * supported page size (64KB). In this way, the selected memory
@@ -68,9 +68,9 @@
  *
  * architecture   slots    memory-per-slot    memory-on-last-slot
  * --------------------------------------------------------------
- * x86-4KB        32763    16KB               160KB
- * arm64-4KB      32766    16KB               112KB
- * arm64-16KB     32766    16KB               112KB
+ * x86-4KB        32762    16KB               176KB
+ * arm64-4KB      32765    16KB               128KB
+ * arm64-16KB     32765    16KB               128KB
  * arm64-64KB     8192     64KB               128KB
  */
 #define MEM_TEST_MOVE_SIZE		(3 * SZ_64K)
@@ -276,7 +276,7 @@ static u64 get_max_slots(struct vm_data *data, u32 host_page_size)
 		rempages = mempages % pages_per_slot;
 		if (check_slot_pages(host_page_size, guest_page_size,
 				     pages_per_slot, rempages))
-			return slots + 1;	/* slot 0 is reserved */
+			return slots + TEST_MEM_SLOT_BASE;
 	}
 
 	return 0;
@@ -313,38 +313,42 @@ static bool prepare_vm(struct vm_data *data, int nslots, u64 *maxslots,
 	data->hva_slots = malloc(sizeof(*data->hva_slots) * data->nslots);
 	TEST_ASSERT(data->hva_slots, "malloc() fail");
 
-	pr_info_v("Adding slots 1..%i, each slot with %"PRIu64" pages + %"PRIu64" extra pages last\n",
-		data->nslots, data->pages_per_slot, rempages);
+	pr_info_v("Adding slots %u..%u, each slot with %" PRIu64
+		  " pages + %" PRIu64 " extra pages last\n",
+		  (u32)TEST_MEM_SLOT_BASE,
+		  TEST_MEM_SLOT_BASE + data->nslots - 1,
+		  data->pages_per_slot, rempages);
 
 	clock_gettime(CLOCK_MONOTONIC, &tstart);
-	for (slot = 1, guest_addr = MEM_GPA; slot <= data->nslots; slot++) {
+	for (slot = 0, guest_addr = MEM_GPA; slot < data->nslots; slot++) {
 		u64 npages;
 
 		npages = data->pages_per_slot;
-		if (slot == data->nslots)
+		if (slot == data->nslots - 1)
 			npages += rempages;
 
 		vm_userspace_mem_region_add(data->vm, VM_MEM_SRC_ANONYMOUS,
-					    guest_addr, slot, npages,
-					    0);
+					    guest_addr, TEST_MEM_SLOT_BASE + slot,
+					    npages, 0);
 		guest_addr += npages * guest_page_size;
 	}
 	*slot_runtime = timespec_elapsed(tstart);
 
-	for (slot = 1, guest_addr = MEM_GPA; slot <= data->nslots; slot++) {
+	for (slot = 0, guest_addr = MEM_GPA; slot < data->nslots; slot++) {
 		u64 npages;
 		gpa_t gpa;
 
 		npages = data->pages_per_slot;
-		if (slot == data->nslots)
+		if (slot == data->nslots - 1)
 			npages += rempages;
 
-		gpa = vm_phy_pages_alloc(data->vm, npages, guest_addr, slot);
+		gpa = vm_phy_pages_alloc(data->vm, npages, guest_addr,
+					 TEST_MEM_SLOT_BASE + slot);
 		TEST_ASSERT(gpa == guest_addr,
 			    "vm_phy_pages_alloc() failed");
 
-		data->hva_slots[slot - 1] = addr_gpa2hva(data->vm, guest_addr);
-		memset(data->hva_slots[slot - 1], 0, npages * guest_page_size);
+		data->hva_slots[slot] = addr_gpa2hva(data->vm, guest_addr);
+		memset(data->hva_slots[slot], 0, npages * guest_page_size);
 
 		guest_addr += npages * guest_page_size;
 	}
@@ -631,9 +635,10 @@ static void test_memslot_move_loop(struct vm_data *data, struct sync_area *sync)
 	u64 movesrcgpa;
 
 	movesrcgpa = vm_slot2gpa(data, data->nslots - 1);
-	vm_mem_region_move(data->vm, data->nslots - 1 + 1,
+	vm_mem_region_move(data->vm, TEST_MEM_SLOT_BASE + data->nslots - 1,
 			   MEM_TEST_MOVE_GPA_DEST);
-	vm_mem_region_move(data->vm, data->nslots - 1 + 1, movesrcgpa);
+	vm_mem_region_move(data->vm, TEST_MEM_SLOT_BASE + data->nslots - 1,
+			   movesrcgpa);
 }
 
 static void test_memslot_do_unmap(struct vm_data *data,
@@ -988,8 +993,10 @@ static bool parse_args(int argc, char *argv[],
 #endif
 		case 's':
 			targs->nslots = atoi_paranoid(optarg);
-			if (targs->nslots <= 1 && targs->nslots != -1) {
-				pr_info("Slot count cap must be larger than 1 or -1 for no cap\n");
+			if (targs->nslots <= TEST_MEM_SLOT_BASE &&
+			    targs->nslots != -1) {
+				pr_info("Slot count cap must be larger than %d or -1 for no cap\n",
+					TEST_MEM_SLOT_BASE);
 				return false;
 			}
 			break;
@@ -1027,20 +1034,21 @@ static bool parse_args(int argc, char *argv[],
 	}
 
 	max_mem_slots = kvm_check_cap(KVM_CAP_NR_MEMSLOTS);
-	if (max_mem_slots <= 1) {
-		pr_info("KVM_CAP_NR_MEMSLOTS should be greater than 1\n");
+	if (max_mem_slots <= TEST_MEM_SLOT_BASE) {
+		pr_info("KVM_CAP_NR_MEMSLOTS should be greater than %d\n",
+			TEST_MEM_SLOT_BASE);
 		return false;
 	}
 
-	/* Memory slot 0 is reserved */
+	/* Memory slots below TEST_MEM_SLOT_BASE are reserved. */
 	if (targs->nslots == -1)
-		targs->nslots = max_mem_slots - 1;
+		targs->nslots = max_mem_slots - TEST_MEM_SLOT_BASE;
 	else
-		targs->nslots = min_t(int, targs->nslots, max_mem_slots) - 1;
+		targs->nslots = min_t(int, targs->nslots, max_mem_slots)
+				- TEST_MEM_SLOT_BASE;
 
 	pr_info_v("Allowed Number of memory slots: %"PRIu32"\n",
-		  targs->nslots + 1);
-
+		 (u32)(targs->nslots + TEST_MEM_SLOT_BASE));
 	return true;
 }
 
-- 
2.53.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH 3/5] KVM: selftests: Use the shared memslot base in memstress

[Tharit Tangkijwanichakul]

Define MEMSTRESS_MEM_SLOT_INDEX using TEST_MEM_SLOT_BASE so
memstress and its users follow the library's reserved-slot
convention.

The tests affected but require no change are:
	dirty_log_perf_test.c
	demand_paging_test.c
	access_tracking_perf_test.c
	memslot_modification_stress_test.c
	x86/dirty_log_page_splitting_test.c

Signed-off-by: Tharit Tangkijwanichakul <tharitt97@gmail.com>
---
 tools/testing/selftests/kvm/include/memstress.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/kvm/include/memstress.h b/tools/testing/selftests/kvm/include/memstress.h
index 0d1d6230cc05..4da03541e805 100644
--- a/tools/testing/selftests/kvm/include/memstress.h
+++ b/tools/testing/selftests/kvm/include/memstress.h
@@ -17,7 +17,7 @@
 
 #define DEFAULT_PER_VCPU_MEM_SIZE	(1 << 30) /* 1G */
 
-#define MEMSTRESS_MEM_SLOT_INDEX	1
+#define MEMSTRESS_MEM_SLOT_INDEX	TEST_MEM_SLOT_BASE
 
 struct memstress_vcpu_args {
 	gpa_t gpa;
-- 
2.53.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH 4/5] KVM: selftests: Reserve memslot 1 for the ucall MMIO address

[Tharit Tangkijwanichakul]

Back the ucall MMIO address with a read-only region in slot 1
when KVM_CAP_READONLY_MEM is available. Writes still exit as
MMIO, while overlapping memslot creation is rejected.

Start test memslots at slot 2, leaving slot 1 reserved by
convention even when read-only memory is unsupported.

Signed-off-by: Tharit Tangkijwanichakul <tharitt97@gmail.com>
---
 .../testing/selftests/kvm/include/kvm_util.h  |  2 +-
 tools/testing/selftests/kvm/lib/kvm_util.c    | 19 +++++++++++++------
 2 files changed, 14 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/kvm/include/kvm_util.h b/tools/testing/selftests/kvm/include/kvm_util.h
index 759d604d44ce..4dbb2ef332fc 100644
--- a/tools/testing/selftests/kvm/include/kvm_util.h
+++ b/tools/testing/selftests/kvm/include/kvm_util.h
@@ -30,7 +30,7 @@
 
 #define KVM_DEV_PATH "/dev/kvm"
 #define KVM_MAX_VCPUS 512
-#define TEST_MEM_SLOT_BASE 1
+#define TEST_MEM_SLOT_BASE 2
 
 #define NSEC_PER_SEC 1000000000L
 
diff --git a/tools/testing/selftests/kvm/lib/kvm_util.c b/tools/testing/selftests/kvm/lib/kvm_util.c
index 9ddc047d5c27..73e28814a226 100644
--- a/tools/testing/selftests/kvm/lib/kvm_util.c
+++ b/tools/testing/selftests/kvm/lib/kvm_util.c
@@ -20,6 +20,7 @@
 #include <linux/kernel.h>
 
 #define KVM_UTIL_MIN_PFN	2
+#define KVM_UCALL_MEMSLOT	1
 
 u32 kvm_random_seed;
 struct kvm_random_state kvm_rng;
@@ -492,6 +493,7 @@ struct kvm_vm *__vm_create(struct vm_shape shape, u32 nr_runnable_vcpus,
 	struct userspace_mem_region *slot0;
 	struct kvm_vm *vm;
 	int i, flags;
+	gpa_t mmio_gpa;
 
 	kvm_set_files_rlimit(nr_runnable_vcpus);
 
@@ -514,14 +516,19 @@ struct kvm_vm *__vm_create(struct vm_shape shape, u32 nr_runnable_vcpus,
 
 	kvm_vm_elf_load(vm, program_invocation_name);
 
-	/*
-	 * TODO: Add proper defines to protect the library's memslots, and then
-	 * carve out memslot1 for the ucall MMIO address.  KVM treats writes to
-	 * read-only memslots as MMIO, and creating a read-only memslot for the
-	 * MMIO region would prevent silently clobbering the MMIO region.
+	/* KVM treats writes to read-only memslots as MMIO, and creating
+	 * a read-only memslot for the MMIO region would prevent silently
+	 * clobbering the MMIO region.
 	 */
 	slot0 = memslot2region(vm, 0);
-	ucall_init(vm, slot0->region.guest_phys_addr + slot0->region.memory_size);
+	mmio_gpa = slot0->region.guest_phys_addr + slot0->region.memory_size;
+	ucall_init(vm, mmio_gpa);
+
+	if (vm_check_cap(vm, KVM_CAP_READONLY_MEM))
+		vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS,
+					    mmio_gpa, KVM_UCALL_MEMSLOT,
+					    vm_adjust_num_guest_pages(vm->mode, 1),
+					    KVM_MEM_READONLY);
 
 	if (kvm_random_seed != last_kvm_seed)
 		kvm_seed_rng(kvm_random_seed);
-- 
2.53.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH 5/5] KVM: selftests: Test protection of the ucall MMIO region

[Tharit Tangkijwanichakul]

Verify that creating a memslot overlapping the ucall MMIO region
fails with EEXIST and that ucalls work before and after the
attempt. Skip when MMIO ucalls or read-only memory are unsupported.

Signed-off-by: Tharit Tangkijwanichakul <tharitt97@gmail.com>
---
 tools/testing/selftests/kvm/Makefile.kvm      |  1 +
 .../selftests/kvm/reserved_slot_test.c        | 61 +++++++++++++++++++
 2 files changed, 62 insertions(+)
 create mode 100644 tools/testing/selftests/kvm/reserved_slot_test.c

diff --git a/tools/testing/selftests/kvm/Makefile.kvm b/tools/testing/selftests/kvm/Makefile.kvm
index 1f4bc2a1322d..ff4a5e9733e9 100644
--- a/tools/testing/selftests/kvm/Makefile.kvm
+++ b/tools/testing/selftests/kvm/Makefile.kvm
@@ -68,6 +68,7 @@ TEST_GEN_PROGS_COMMON += set_memory_region_test
 TEST_GEN_PROGS_COMMON += memslot_modification_stress_test
 TEST_GEN_PROGS_COMMON += memslot_perf_test
 TEST_GEN_PROGS_COMMON += vm_types_test
+TEST_GEN_PROGS_COMMON += reserved_slot_test
 
 # Compiled test targets
 TEST_GEN_PROGS_x86 = $(TEST_GEN_PROGS_COMMON)
diff --git a/tools/testing/selftests/kvm/reserved_slot_test.c b/tools/testing/selftests/kvm/reserved_slot_test.c
new file mode 100644
index 000000000000..4a908017ca73
--- /dev/null
+++ b/tools/testing/selftests/kvm/reserved_slot_test.c
@@ -0,0 +1,61 @@
+// SPDX-License-Identifier: GPL-2.0
+#include "kvm_util.h"
+#include "processor.h"
+#include "test_util.h"
+
+#define CLOBBERED_SLOT		TEST_MEM_SLOT_BASE
+
+static void guest_code(void)
+{
+	GUEST_SYNC(1);
+	GUEST_SYNC(2);
+}
+
+static void test_slot1_protected(void)
+{
+	struct kvm_vcpu *vcpu;
+	struct kvm_run *run;
+	struct kvm_vm *vm;
+	struct ucall uc;
+	u32 nr_pages;
+	size_t slot_size;
+	void *slot_mem;
+	int ret;
+
+	vm = vm_create_with_one_vcpu(&vcpu, guest_code);
+	nr_pages = vm_adjust_num_guest_pages(vm->mode, 1);
+	slot_size = (size_t)nr_pages * vm->page_size;
+
+	run = vcpu->run;
+	vcpu_run(vcpu);
+	TEST_ASSERT(run->exit_reason == KVM_EXIT_MMIO,
+		    "Unexpected exit reason = %u", run->exit_reason);
+	TEST_ASSERT(get_ucall(vcpu, &uc) == UCALL_SYNC && uc.args[1] == 1,
+		    "Expected the first sync before clobbering the ucall GPA");
+
+	slot_mem = kvm_mmap(slot_size, PROT_READ | PROT_WRITE,
+			    MAP_PRIVATE | MAP_ANONYMOUS, -1);
+	ret = __vm_set_user_memory_region2(vm, CLOBBERED_SLOT, 0,
+					   vm->ucall_mmio_addr, slot_size, slot_mem, 0, 0);
+	TEST_ASSERT(ret == -1 && errno == EEXIST,
+		    "Expected the ucall GPA overlap to be rejected: ret=%d errno=%d",
+		    ret, errno);
+
+	vcpu_run(vcpu);
+	TEST_ASSERT(run->exit_reason == KVM_EXIT_MMIO,
+		    "Unexpected exit reason = %u", run->exit_reason);
+	TEST_ASSERT(get_ucall(vcpu, &uc) == UCALL_SYNC && uc.args[1] == 2,
+		    "Expected the second sync after attempting to clobber the ucall GPA");
+
+	kvm_vm_free(vm);
+	TEST_ASSERT(!munmap(slot_mem, slot_size), "munmap() failed, errno=%d", errno);
+}
+
+int main(void)
+{
+	TEST_REQUIRE(UCALL_EXIT_REASON == KVM_EXIT_MMIO);
+	TEST_REQUIRE(kvm_has_cap(KVM_CAP_READONLY_MEM));
+
+	test_slot1_protected();
+	return 0;
+}
-- 
2.53.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 5/5] KVM: selftests: Test protection of the ucall MMIO region

[Tharit Tangkijwanichakul]

> +#define CLOBBERED_SLOT         TEST_MEM_SLOT_BASE
Upon self-review, this should have been hardcoded to 1 (to clobber the
reserved ucal mmio).

> +       ret = __vm_set_user_memory_region2(vm, CLOBBERED_SLOT, 0,
> +                                          vm->ucall_mmio_addr, slot_size, slot_mem, 0, 0);
> +       TEST_ASSERT(ret == -1 && errno == EEXIST,
> +                   "Expected the ucall GPA overlap to be rejected: ret=%d errno=%d",
> +                   ret, errno);
This should have been errno == EINVAL check to catch writing to the
exiting read-only slot.

The earlier test accidentally passed because it used a different slot
ID (=2, not reserved) and received EEXIST for overlapping the reserved
ucall GPA. It therefore tested GPA overlap rejection, not the intended
rejection of overwriting slot 1.

I will fix it in v2

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv