drm/amdgpu: implement 2+1 PD support for Raven v3

Instead of falling back to 2 level and very limited address space use
2+1 PD support and 128TB + 512GB of virtual address space.

v2: cleanup defines, rebase on top of level enum
v3: fix inverted check in hardware setup
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-and-Tested-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -540,6 +540,7 @@ struct amdgpu_mc {
 	u64					private_aperture_end;
 	/* protects concurrent invalidation */
 	spinlock_t		invalidate_lock;
+	bool			translate_further;
 };
 
 /*

drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h

@@ -70,6 +70,12 @@ struct amdgpu_bo_list_entry;
 /* PDE is handled as PTE for VEGA10 */
 #define AMDGPU_PDE_PTE		(1ULL << 54)
 
+/* PTE is handled as PDE for VEGA10 (Translate Further) */
+#define AMDGPU_PTE_TF		(1ULL << 56)
+
+/* PDE Block Fragment Size for VEGA10 */
+#define AMDGPU_PDE_BFS(a)	((uint64_t)a << 59)
+
 /* VEGA10 only */
 #define AMDGPU_PTE_MTYPE(a)    ((uint64_t)a << 57)
 #define AMDGPU_PTE_MTYPE_MASK	AMDGPU_PTE_MTYPE(3ULL)

drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c

@@ -143,8 +143,15 @@ static void gfxhub_v1_0_init_cache_regs(struct amdgpu_device *adev)
 	WREG32_SOC15(GC, 0, mmVM_L2_CNTL2, tmp);
 
 	tmp = mmVM_L2_CNTL3_DEFAULT;
+	if (adev->mc.translate_further) {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+				    L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+	} else {
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+				    L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
+	}
 	WREG32_SOC15(GC, 0, mmVM_L2_CNTL3, tmp);
 
 	tmp = mmVM_L2_CNTL4_DEFAULT;
@@ -182,18 +189,27 @@ static void gfxhub_v1_0_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 {
-	int i;
+	unsigned num_level, block_size;
 	uint32_t tmp;
+	int i;
+
+	num_level = adev->vm_manager.num_level;
+	block_size = adev->vm_manager.block_size;
+	if (adev->mc.translate_further)
+		num_level -= 1;
+	else
+		block_size -= 9;
 
 	for (i = 0; i <= 14; i++) {
 		tmp = RREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT1_CNTL, i);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
-				    adev->vm_manager.num_level);
+				    num_level);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+				    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT,
+				    1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
@@ -206,7 +222,7 @@ static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 				    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    PAGE_TABLE_BLOCK_SIZE,
-				adev->vm_manager.block_size - 9);
+				    block_size);
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);

drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c

@@ -476,6 +476,21 @@ static void gmc_v9_0_get_vm_pde(struct amdgpu_device *adev, int level,
 		*addr = adev->vm_manager.vram_base_offset + *addr -
 			adev->mc.vram_start;
 	BUG_ON(*addr & 0xFFFF00000000003FULL);
+
+	if (!adev->mc.translate_further)
+		return;
+
+	if (level == AMDGPU_VM_PDB1) {
+		/* Set the block fragment size */
+		if (!(*flags & AMDGPU_PDE_PTE))
+			*flags |= AMDGPU_PDE_BFS(0x9);
+
+	} else if (level == AMDGPU_VM_PDB0) {
+		if (*flags & AMDGPU_PDE_PTE)
+			*flags &= ~AMDGPU_PDE_PTE;
+		else
+			*flags |= AMDGPU_PTE_TF;
+	}
 }
 
 static const struct amdgpu_gart_funcs gmc_v9_0_gart_funcs = {
@@ -773,11 +788,14 @@ static int gmc_v9_0_sw_init(void *handle)
 	switch (adev->asic_type) {
 	case CHIP_RAVEN:
 		adev->mc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN;
-		if (adev->rev_id == 0x0 || adev->rev_id == 0x1)
+		if (adev->rev_id == 0x0 || adev->rev_id == 0x1) {
 			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
-		else
-			/* vm_size is 64GB for legacy 2-level page support */
-			amdgpu_vm_adjust_size(adev, 64, 9, 1, 48);
+		} else {
+			/* vm_size is 128TB + 512GB for legacy 3-level page support */
+			amdgpu_vm_adjust_size(adev, 128 * 1024 + 512, 9, 2, 48);
+			adev->mc.translate_further =
+				adev->vm_manager.num_level > 1;
+		}
 		break;
 	case CHIP_VEGA10:
 		/* XXX Don't know how to get VRAM type yet. */

drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c

@@ -155,10 +155,15 @@ static void mmhub_v1_0_init_cache_regs(struct amdgpu_device *adev)
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
 	WREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL2, tmp);
 
-	tmp = mmVM_L2_CNTL3_DEFAULT;
+	if (adev->mc.translate_further) {
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+				    L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+	} else {
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
-	WREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL3, tmp);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+				    L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
+	}
 
 	tmp = mmVM_L2_CNTL4_DEFAULT;
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
@@ -196,19 +201,27 @@ static void mmhub_v1_0_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 {
-	int i;
+	unsigned num_level, block_size;
 	uint32_t tmp;
+	int i;
+
+	num_level = adev->vm_manager.num_level;
+	block_size = adev->vm_manager.block_size;
+	if (adev->mc.translate_further)
+		num_level -= 1;
+	else
+		block_size -= 9;
 
 	for (i = 0; i <= 14; i++) {
 		tmp = RREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT1_CNTL, i);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				ENABLE_CONTEXT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				PAGE_TABLE_DEPTH, adev->vm_manager.num_level);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
+				    num_level);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+				    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT,
+				    1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
@@ -221,7 +234,7 @@ static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 				    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    PAGE_TABLE_BLOCK_SIZE,
-				adev->vm_manager.block_size - 9);
+				    block_size);
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);