Merge v4.18-rc3 into drm-next

Two requests have come in for a backmerge,
and I've got some pull reqs on rc2, so this
just makes sense.

Signed-off-by: Dave Airlie <airlied@redhat.com>

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

@@ -2157,10 +2157,18 @@ bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type)
 	switch (asic_type) {
 #if defined(CONFIG_DRM_AMD_DC)
 	case CHIP_BONAIRE:
-	case CHIP_HAWAII:
 	case CHIP_KAVERI:
 	case CHIP_KABINI:
 	case CHIP_MULLINS:
+		/*
+		 * We have systems in the wild with these ASICs that require
+		 * LVDS and VGA support which is not supported with DC.
+		 *
+		 * Fallback to the non-DC driver here by default so as not to
+		 * cause regressions.
+		 */
+		return amdgpu_dc > 0;
+	case CHIP_HAWAII:
 	case CHIP_CARRIZO:
 	case CHIP_STONEY:
 	case CHIP_POLARIS10:

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

@@ -376,7 +376,7 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
 	struct amdgpu_device *adev = ring->adev;
 	uint64_t index;
 
-	if (ring != &adev->uvd.inst[ring->me].ring) {
+	if (ring->funcs->type != AMDGPU_RING_TYPE_UVD) {
 		ring->fence_drv.cpu_addr = &adev->wb.wb[ring->fence_offs];
 		ring->fence_drv.gpu_addr = adev->wb.gpu_addr + (ring->fence_offs * 4);
 	} else {

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

@@ -918,8 +918,7 @@ int amdgpu_bo_pin_restricted(struct amdgpu_bo *bo, u32 domain,
 	domain = amdgpu_mem_type_to_domain(bo->tbo.mem.mem_type);
 	if (domain == AMDGPU_GEM_DOMAIN_VRAM) {
 		adev->vram_pin_size += amdgpu_bo_size(bo);
-		if (bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)
-			adev->invisible_pin_size += amdgpu_bo_size(bo);
+		adev->invisible_pin_size += amdgpu_vram_mgr_bo_invisible_size(bo);
 	} else if (domain == AMDGPU_GEM_DOMAIN_GTT) {
 		adev->gart_pin_size += amdgpu_bo_size(bo);
 	}
@@ -969,25 +968,22 @@ int amdgpu_bo_unpin(struct amdgpu_bo *bo)
 	bo->pin_count--;
 	if (bo->pin_count)
 		return 0;
+
+	if (bo->tbo.mem.mem_type == TTM_PL_VRAM) {
+		adev->vram_pin_size -= amdgpu_bo_size(bo);
+		adev->invisible_pin_size -= amdgpu_vram_mgr_bo_invisible_size(bo);
+	} else if (bo->tbo.mem.mem_type == TTM_PL_TT) {
+		adev->gart_pin_size -= amdgpu_bo_size(bo);
+	}
+
 	for (i = 0; i < bo->placement.num_placement; i++) {
 		bo->placements[i].lpfn = 0;
 		bo->placements[i].flags &= ~TTM_PL_FLAG_NO_EVICT;
 	}
 	r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
-	if (unlikely(r)) {
+	if (unlikely(r))
 		dev_err(adev->dev, "%p validate failed for unpin\n", bo);
-		goto error;
-	}
 
-	if (bo->tbo.mem.mem_type == TTM_PL_VRAM) {
-		adev->vram_pin_size -= amdgpu_bo_size(bo);
-		if (bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)
-			adev->invisible_pin_size -= amdgpu_bo_size(bo);
-	} else if (bo->tbo.mem.mem_type == TTM_PL_TT) {
-		adev->gart_pin_size -= amdgpu_bo_size(bo);
-	}
-
-error:
 	return r;
 }
 

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

@@ -73,6 +73,7 @@ bool amdgpu_gtt_mgr_has_gart_addr(struct ttm_mem_reg *mem);
 uint64_t amdgpu_gtt_mgr_usage(struct ttm_mem_type_manager *man);
 int amdgpu_gtt_mgr_recover(struct ttm_mem_type_manager *man);
 
+u64 amdgpu_vram_mgr_bo_invisible_size(struct amdgpu_bo *bo);
 uint64_t amdgpu_vram_mgr_usage(struct ttm_mem_type_manager *man);
 uint64_t amdgpu_vram_mgr_vis_usage(struct ttm_mem_type_manager *man);
 

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

@@ -130,7 +130,7 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev)
 	unsigned family_id;
 	int i, j, r;
 
-	INIT_DELAYED_WORK(&adev->uvd.inst->idle_work, amdgpu_uvd_idle_work_handler);
+	INIT_DELAYED_WORK(&adev->uvd.idle_work, amdgpu_uvd_idle_work_handler);
 
 	switch (adev->asic_type) {
 #ifdef CONFIG_DRM_AMDGPU_CIK
@@ -331,12 +331,12 @@ int amdgpu_uvd_suspend(struct amdgpu_device *adev)
 	void *ptr;
 	int i, j;
 
+	cancel_delayed_work_sync(&adev->uvd.idle_work);
+
 	for (j = 0; j < adev->uvd.num_uvd_inst; ++j) {
 		if (adev->uvd.inst[j].vcpu_bo == NULL)
 			continue;
 
-		cancel_delayed_work_sync(&adev->uvd.inst[j].idle_work);
-
 		/* only valid for physical mode */
 		if (adev->asic_type < CHIP_POLARIS10) {
 			for (i = 0; i < adev->uvd.max_handles; ++i)
@@ -1162,7 +1162,7 @@ int amdgpu_uvd_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
 static void amdgpu_uvd_idle_work_handler(struct work_struct *work)
 {
 	struct amdgpu_device *adev =
-		container_of(work, struct amdgpu_device, uvd.inst->idle_work.work);
+		container_of(work, struct amdgpu_device, uvd.idle_work.work);
 	unsigned fences = 0, i, j;
 
 	for (i = 0; i < adev->uvd.num_uvd_inst; ++i) {
@@ -1184,7 +1184,7 @@ static void amdgpu_uvd_idle_work_handler(struct work_struct *work)
 							       AMD_CG_STATE_GATE);
 		}
 	} else {
-		schedule_delayed_work(&adev->uvd.inst->idle_work, UVD_IDLE_TIMEOUT);
+		schedule_delayed_work(&adev->uvd.idle_work, UVD_IDLE_TIMEOUT);
 	}
 }
 
@@ -1196,7 +1196,7 @@ void amdgpu_uvd_ring_begin_use(struct amdgpu_ring *ring)
 	if (amdgpu_sriov_vf(adev))
 		return;
 
-	set_clocks = !cancel_delayed_work_sync(&adev->uvd.inst->idle_work);
+	set_clocks = !cancel_delayed_work_sync(&adev->uvd.idle_work);
 	if (set_clocks) {
 		if (adev->pm.dpm_enabled) {
 			amdgpu_dpm_enable_uvd(adev, true);
@@ -1213,7 +1213,7 @@ void amdgpu_uvd_ring_begin_use(struct amdgpu_ring *ring)
 void amdgpu_uvd_ring_end_use(struct amdgpu_ring *ring)
 {
 	if (!amdgpu_sriov_vf(ring->adev))
-		schedule_delayed_work(&ring->adev->uvd.inst->idle_work, UVD_IDLE_TIMEOUT);
+		schedule_delayed_work(&ring->adev->uvd.idle_work, UVD_IDLE_TIMEOUT);
 }
 
 /**

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

@@ -44,7 +44,6 @@ struct amdgpu_uvd_inst {
 	void			*saved_bo;
 	atomic_t		handles[AMDGPU_MAX_UVD_HANDLES];
 	struct drm_file		*filp[AMDGPU_MAX_UVD_HANDLES];
-	struct delayed_work	idle_work;
 	struct amdgpu_ring	ring;
 	struct amdgpu_ring	ring_enc[AMDGPU_MAX_UVD_ENC_RINGS];
 	struct amdgpu_irq_src	irq;
@@ -62,6 +61,7 @@ struct amdgpu_uvd {
 	bool			address_64_bit;
 	bool			use_ctx_buf;
 	struct amdgpu_uvd_inst		inst[AMDGPU_MAX_UVD_INSTANCES];
+	struct delayed_work	idle_work;
 };
 
 int amdgpu_uvd_sw_init(struct amdgpu_device *adev);

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

@@ -52,7 +52,7 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 	unsigned long bo_size;
 	const char *fw_name;
 	const struct common_firmware_header *hdr;
-	unsigned version_major, version_minor, family_id;
+	unsigned char fw_check;
 	int r;
 
 	INIT_DELAYED_WORK(&adev->vcn.idle_work, amdgpu_vcn_idle_work_handler);
@@ -83,12 +83,33 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 
 	hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
 	adev->vcn.fw_version = le32_to_cpu(hdr->ucode_version);
-	family_id = le32_to_cpu(hdr->ucode_version) & 0xff;
-	version_major = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xff;
-	version_minor = (le32_to_cpu(hdr->ucode_version) >> 8) & 0xff;
-	DRM_INFO("Found VCN firmware Version: %hu.%hu Family ID: %hu\n",
-		version_major, version_minor, family_id);
 
+	/* Bit 20-23, it is encode major and non-zero for new naming convention.
+	 * This field is part of version minor and DRM_DISABLED_FLAG in old naming
+	 * convention. Since the l:wq!atest version minor is 0x5B and DRM_DISABLED_FLAG
+	 * is zero in old naming convention, this field is always zero so far.
+	 * These four bits are used to tell which naming convention is present.
+	 */
+	fw_check = (le32_to_cpu(hdr->ucode_version) >> 20) & 0xf;
+	if (fw_check) {
+		unsigned int dec_ver, enc_major, enc_minor, vep, fw_rev;
+
+		fw_rev = le32_to_cpu(hdr->ucode_version) & 0xfff;
+		enc_minor = (le32_to_cpu(hdr->ucode_version) >> 12) & 0xff;
+		enc_major = fw_check;
+		dec_ver = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xf;
+		vep = (le32_to_cpu(hdr->ucode_version) >> 28) & 0xf;
+		DRM_INFO("Found VCN firmware Version ENC: %hu.%hu DEC: %hu VEP: %hu Revision: %hu\n",
+			enc_major, enc_minor, dec_ver, vep, fw_rev);
+	} else {
+		unsigned int version_major, version_minor, family_id;
+
+		family_id = le32_to_cpu(hdr->ucode_version) & 0xff;
+		version_major = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xff;
+		version_minor = (le32_to_cpu(hdr->ucode_version) >> 8) & 0xff;
+		DRM_INFO("Found VCN firmware Version: %hu.%hu Family ID: %hu\n",
+			version_major, version_minor, family_id);
+	}
 
 	bo_size = AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8)
 		  +  AMDGPU_VCN_STACK_SIZE + AMDGPU_VCN_HEAP_SIZE

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

@@ -1577,7 +1577,9 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 			uint64_t count;
 
 			max_entries = min(max_entries, 16ull * 1024ull);
-			for (count = 1; count < max_entries; ++count) {
+			for (count = 1;
+			     count < max_entries / (PAGE_SIZE / AMDGPU_GPU_PAGE_SIZE);
+			     ++count) {
 				uint64_t idx = pfn + count;
 
 				if (pages_addr[idx] !=
@@ -1590,7 +1592,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 				dma_addr = pages_addr;
 			} else {
 				addr = pages_addr[pfn];
-				max_entries = count;
+				max_entries = count * (PAGE_SIZE / AMDGPU_GPU_PAGE_SIZE);
 			}
 
 		} else if (flags & AMDGPU_PTE_VALID) {
@@ -1605,7 +1607,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 		if (r)
 			return r;
 
-		pfn += last - start + 1;
+		pfn += (last - start + 1) / (PAGE_SIZE / AMDGPU_GPU_PAGE_SIZE);
 		if (nodes && nodes->size == pfn) {
 			pfn = 0;
 			++nodes;

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

@@ -96,6 +96,38 @@ static u64 amdgpu_vram_mgr_vis_size(struct amdgpu_device *adev,
 		adev->gmc.visible_vram_size : end) - start;
 }
 
+/**
+ * amdgpu_vram_mgr_bo_invisible_size - CPU invisible BO size
+ *
+ * @bo: &amdgpu_bo buffer object (must be in VRAM)
+ *
+ * Returns:
+ * How much of the given &amdgpu_bo buffer object lies in CPU invisible VRAM.
+ */
+u64 amdgpu_vram_mgr_bo_invisible_size(struct amdgpu_bo *bo)
+{
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+	struct ttm_mem_reg *mem = &bo->tbo.mem;
+	struct drm_mm_node *nodes = mem->mm_node;
+	unsigned pages = mem->num_pages;
+	u64 usage = 0;
+
+	if (adev->gmc.visible_vram_size == adev->gmc.real_vram_size)
+		return 0;
+
+	if (mem->start >= adev->gmc.visible_vram_size >> PAGE_SHIFT)
+		return amdgpu_bo_size(bo);
+
+	while (nodes && pages) {
+		usage += nodes->size << PAGE_SHIFT;
+		usage -= amdgpu_vram_mgr_vis_size(adev, nodes);
+		pages -= nodes->size;
+		++nodes;
+	}
+
+	return usage;
+}
+
 /**
  * amdgpu_vram_mgr_new - allocate new ranges
  *
@@ -135,7 +167,8 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 		num_nodes = DIV_ROUND_UP(mem->num_pages, pages_per_node);
 	}
 
-	nodes = kcalloc(num_nodes, sizeof(*nodes), GFP_KERNEL);
+	nodes = kvmalloc_array(num_nodes, sizeof(*nodes),
+			       GFP_KERNEL | __GFP_ZERO);
 	if (!nodes)
 		return -ENOMEM;
 
@@ -190,7 +223,7 @@ error:
 		drm_mm_remove_node(&nodes[i]);
 	spin_unlock(&mgr->lock);
 
-	kfree(nodes);
+	kvfree(nodes);
 	return r == -ENOSPC ? 0 : r;
 }
 
@@ -229,7 +262,7 @@ static void amdgpu_vram_mgr_del(struct ttm_mem_type_manager *man,
 	atomic64_sub(usage, &mgr->usage);
 	atomic64_sub(vis_usage, &mgr->vis_usage);
 
-	kfree(mem->mm_node);
+	kvfree(mem->mm_node);
 	mem->mm_node = NULL;
 }