Merge branch 'drm-next-4.15' of git://people.freedesktop.org/~agd5f/linux into drm-next

Some amdgpu/ttm fixes.

* 'drm-next-4.15' of git://people.freedesktop.org/~agd5f/linux:
  drm/amd/powerplay: wrong control mode cause the fan spins faster unnecessarily
  drm/amd/powerplay: fix memory leak of hardcoded pptable
  drm/amdgpu:add fw-vram-usage for atomfirmware
  drm/radeon: fix atombios on big endian
  drm/ttm:fix memory leak due to individualize
  drm/amdgpu: fix error handling in amdgpu_bo_do_create
  drm/ttm: once more fix ttm_buffer_object_transfer
  drm/amd/powerplay: change ASIC temperature reading on Vega10

9 files changed, 68 insertions, 51 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
index f66d33e4baca..f450b69323fa 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
@@ -1766,34 +1766,32 @@ bool amdgpu_atombios_scratch_need_asic_init(struct amdgpu_device *adev)
 		return true;
 }
 
-/* Atom needs data in little endian format
- * so swap as appropriate when copying data to
- * or from atom. Note that atom operates on
- * dw units.
+/* Atom needs data in little endian format so swap as appropriate when copying
+ * data to or from atom. Note that atom operates on dw units.
+ *
+ * Use to_le=true when sending data to atom and provide at least
+ * ALIGN(num_bytes,4) bytes in the dst buffer.
+ *
+ * Use to_le=false when receiving data from atom and provide ALIGN(num_bytes,4)
+ * byes in the src buffer.
  */
 void amdgpu_atombios_copy_swap(u8 *dst, u8 *src, u8 num_bytes, bool to_le)
 {
 #ifdef __BIG_ENDIAN
-	u8 src_tmp[20], dst_tmp[20]; /* used for byteswapping */
-	u32 *dst32, *src32;
+	u32 src_tmp[5], dst_tmp[5];
 	int i;
+	u8 align_num_bytes = ALIGN(num_bytes, 4);
 
-	memcpy(src_tmp, src, num_bytes);
-	src32 = (u32 *)src_tmp;
-	dst32 = (u32 *)dst_tmp;
 	if (to_le) {
-		for (i = 0; i < ((num_bytes + 3) / 4); i++)
-			dst32[i] = cpu_to_le32(src32[i]);
-		memcpy(dst, dst_tmp, num_bytes);
+		memcpy(src_tmp, src, num_bytes);
+		for (i = 0; i < align_num_bytes / 4; i++)
+			dst_tmp[i] = cpu_to_le32(src_tmp[i]);
+		memcpy(dst, dst_tmp, align_num_bytes);
 	} else {
-		u8 dws = num_bytes & ~3;
-		for (i = 0; i < ((num_bytes + 3) / 4); i++)
-			dst32[i] = le32_to_cpu(src32[i]);
-		memcpy(dst, dst_tmp, dws);
-		if (num_bytes % 4) {
-			for (i = 0; i < (num_bytes % 4); i++)
-				dst[dws+i] = dst_tmp[dws+i];
-		}
+		memcpy(src_tmp, src, align_num_bytes);
+		for (i = 0; i < align_num_bytes / 4; i++)
+			dst_tmp[i] = le32_to_cpu(src_tmp[i]);
+		memcpy(dst, dst_tmp, num_bytes);
 	}
 #else
 	memcpy(dst, src, num_bytes);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
index f9ffe8ef0cd6..ff8efd0f8fd5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
@@ -71,19 +71,33 @@ int amdgpu_atomfirmware_allocate_fb_scratch(struct amdgpu_device *adev)
 	struct atom_context *ctx = adev->mode_info.atom_context;
 	int index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
 						vram_usagebyfirmware);
+	struct vram_usagebyfirmware_v2_1 *	firmware_usage;
+	uint32_t start_addr, size;
 	uint16_t data_offset;
 	int usage_bytes = 0;
 
 	if (amdgpu_atom_parse_data_header(ctx, index, NULL, NULL, NULL, &data_offset)) {
-		struct vram_usagebyfirmware_v2_1 *firmware_usage =
-			(struct vram_usagebyfirmware_v2_1 *)(ctx->bios + data_offset);
-
+		firmware_usage = (struct vram_usagebyfirmware_v2_1 *)(ctx->bios + data_offset);
 		DRM_DEBUG("atom firmware requested %08x %dkb fw %dkb drv\n",
 			  le32_to_cpu(firmware_usage->start_address_in_kb),
 			  le16_to_cpu(firmware_usage->used_by_firmware_in_kb),
 			  le16_to_cpu(firmware_usage->used_by_driver_in_kb));
 
-		usage_bytes = le16_to_cpu(firmware_usage->used_by_driver_in_kb) * 1024;
+		start_addr = le32_to_cpu(firmware_usage->start_address_in_kb);
+		size = le16_to_cpu(firmware_usage->used_by_firmware_in_kb);
+
+		if ((uint32_t)(start_addr & ATOM_VRAM_OPERATION_FLAGS_MASK) ==
+			(uint32_t)(ATOM_VRAM_BLOCK_SRIOV_MSG_SHARE_RESERVATION <<
+			ATOM_VRAM_OPERATION_FLAGS_SHIFT)) {
+			/* Firmware request VRAM reservation for SR-IOV */
+			adev->fw_vram_usage.start_offset = (start_addr &
+				(~ATOM_VRAM_OPERATION_FLAGS_MASK)) << 10;
+			adev->fw_vram_usage.size = size << 10;
+			/* Use the default scratch size */
+			usage_bytes = 0;
+		} else {
+			usage_bytes = le16_to_cpu(firmware_usage->used_by_driver_in_kb) << 10;
+		}
 	}
 	ctx->scratch_size_bytes = 0;
 	if (usage_bytes == 0)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 8b4ed8a98a18..ea25164e7f4b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -369,6 +369,9 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
 	r = ttm_bo_init_reserved(&adev->mman.bdev, &bo->tbo, size, type,
 				 &bo->placement, page_align, !kernel, NULL,
 				 acc_size, sg, resv, &amdgpu_ttm_bo_destroy);
+	if (unlikely(r != 0))
+		return r;
+
 	bytes_moved = atomic64_read(&adev->num_bytes_moved) -
 		      initial_bytes_moved;
 	if (adev->mc.visible_vram_size < adev->mc.real_vram_size &&
@@ -378,9 +381,6 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
 	else
 		amdgpu_cs_report_moved_bytes(adev, bytes_moved, 0);
 
-	if (unlikely(r != 0))
-		return r;
-
 	if (kernel)
 		bo->tbo.priority = 1;
 
diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
index 0f8b6dc853a5..c7e34128cbde 100644
--- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
@@ -78,6 +78,9 @@ static int amd_powerplay_destroy(void *handle)
 {
 	struct pp_instance *instance = (struct pp_instance *)handle;
 
+	kfree(instance->hwmgr->hardcode_pp_table);
+	instance->hwmgr->hardcode_pp_table = NULL;
+
 	kfree(instance->hwmgr);
 	instance->hwmgr = NULL;
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
index 203ef10cafdc..4239b98cf6db 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
@@ -4234,7 +4234,7 @@ static void vega10_set_fan_control_mode(struct pp_hwmgr *hwmgr, uint32_t mode)
 			vega10_fan_ctrl_stop_smc_fan_control(hwmgr);
 		break;
 	case AMD_FAN_CTRL_AUTO:
-		if (!vega10_fan_ctrl_set_static_mode(hwmgr, mode))
+		if (PP_CAP(PHM_PlatformCaps_MicrocodeFanControl))
 			vega10_fan_ctrl_start_smc_fan_control(hwmgr);
 		break;
 	default:
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
index 1feefac49ea9..dc3761bcb9b6 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c
@@ -365,8 +365,8 @@ int vega10_thermal_get_temperature(struct pp_hwmgr *hwmgr)
 
 	temp = cgs_read_register(hwmgr->device, reg);
 
-	temp = (temp & CG_MULT_THERMAL_STATUS__ASIC_MAX_TEMP_MASK) >>
-			CG_MULT_THERMAL_STATUS__ASIC_MAX_TEMP__SHIFT;
+	temp = (temp & CG_MULT_THERMAL_STATUS__CTF_TEMP_MASK) >>
+			CG_MULT_THERMAL_STATUS__CTF_TEMP__SHIFT;
 
 	temp = temp & 0x1ff;
 
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index a904c80c30e6..3e798593e042 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -45,34 +45,32 @@ static char *pre_emph_names[] = {
 
 /***** radeon AUX functions *****/
 
-/* Atom needs data in little endian format
- * so swap as appropriate when copying data to
- * or from atom. Note that atom operates on
- * dw units.
+/* Atom needs data in little endian format so swap as appropriate when copying
+ * data to or from atom. Note that atom operates on dw units.
+ *
+ * Use to_le=true when sending data to atom and provide at least
+ * ALIGN(num_bytes,4) bytes in the dst buffer.
+ *
+ * Use to_le=false when receiving data from atom and provide ALIGN(num_bytes,4)
+ * byes in the src buffer.
  */
 void radeon_atom_copy_swap(u8 *dst, u8 *src, u8 num_bytes, bool to_le)
 {
 #ifdef __BIG_ENDIAN
-	u8 src_tmp[20], dst_tmp[20]; /* used for byteswapping */
-	u32 *dst32, *src32;
+	u32 src_tmp[5], dst_tmp[5];
 	int i;
+	u8 align_num_bytes = ALIGN(num_bytes, 4);
 
-	memcpy(src_tmp, src, num_bytes);
-	src32 = (u32 *)src_tmp;
-	dst32 = (u32 *)dst_tmp;
 	if (to_le) {
-		for (i = 0; i < ((num_bytes + 3) / 4); i++)
-			dst32[i] = cpu_to_le32(src32[i]);
-		memcpy(dst, dst_tmp, num_bytes);
+		memcpy(src_tmp, src, num_bytes);
+		for (i = 0; i < align_num_bytes / 4; i++)
+			dst_tmp[i] = cpu_to_le32(src_tmp[i]);
+		memcpy(dst, dst_tmp, align_num_bytes);
 	} else {
-		u8 dws = num_bytes & ~3;
-		for (i = 0; i < ((num_bytes + 3) / 4); i++)
-			dst32[i] = le32_to_cpu(src32[i]);
-		memcpy(dst, dst_tmp, dws);
-		if (num_bytes % 4) {
-			for (i = 0; i < (num_bytes % 4); i++)
-				dst[dws+i] = dst_tmp[dws+i];
-		}
+		memcpy(src_tmp, src, align_num_bytes);
+		for (i = 0; i < align_num_bytes / 4; i++)
+			dst_tmp[i] = le32_to_cpu(src_tmp[i]);
+		memcpy(dst, dst_tmp, num_bytes);
 	}
 #else
 	memcpy(dst, src, num_bytes);
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index d79607a1187c..379ec41d2c69 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -457,8 +457,11 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
 		if (reservation_object_test_signaled_rcu(&bo->ttm_resv, true)) {
 			ttm_bo_del_from_lru(bo);
 			spin_unlock(&glob->lru_lock);
-			if (bo->resv != &bo->ttm_resv)
+			if (bo->resv != &bo->ttm_resv) {
 				reservation_object_unlock(&bo->ttm_resv);
+				reservation_object_fini(&bo->ttm_resv);
+			}
+
 			ttm_bo_cleanup_memtype_use(bo);
 			return;
 		}
diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index 78cb99be7146..e7a519f1849b 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -474,6 +474,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo,
 	INIT_LIST_HEAD(&fbo->lru);
 	INIT_LIST_HEAD(&fbo->swap);
 	INIT_LIST_HEAD(&fbo->io_reserve_lru);
+	mutex_init(&fbo->wu_mutex);
 	fbo->moving = NULL;
 	drm_vma_node_reset(&fbo->vma_node);
 	atomic_set(&fbo->cpu_writers, 0);