Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "This is mostly amdgpu/radeon fixes, and imx related fixes.

  There is also one one TTM fix, one nouveau fix, and one hdlcd fix.

  The AMD ones are some fixes for power management after suspend/resume
  one some GPUs, and some vblank fixes.

  The IMX ones are for more stricter plane checks and some cleanups.

  I'm off until Monday, so therre might be some fixes early next week if
  anyone missed me"

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (34 commits)
  drm/nouveau/tegra: acquire and enable reference clock if needed
  drm/amdgpu: total vram size also reduces pin size
  drm/amd/powerplay: add uvd/vce dpm enabling flag default.
  drm/amd/powerplay: fix issue that resume back, dpm can't work on FIJI.
  drm/amdgpu: save and restore the firwmware cache part when suspend resume
  drm/amdgpu: save and restore UVD context with suspend and resume
  drm/ttm: use phys_addr_t for ttm_bus_placement
  drm: ARM HDLCD - fix an error code
  drm: ARM HDLCD - get rid of devm_clk_put()
  drm/radeon: Only call drm_vblank_on/off between drm_vblank_init/cleanup
  drm/amdgpu: fence wait old rcu slot
  drm/amdgpu: fix leaking fence in the pageflip code
  drm/amdgpu: print vram type rather than just DDR
  drm/amdgpu/gmc: use proper register for vram type on Fiji
  drm/amdgpu/gmc: move vram type fetching into sw_init
  drm/amdgpu: Set vblank_disable_allowed = true
  drm/radeon: Set vblank_disable_allowed = true
  drm/amd/powerplay: Need to change boot to performance state in resume.
  drm/amd/powerplay: add new Fiji function for not setting same ps.
  drm/amdgpu: check dpm state before pm system fs initialized.
  ...

31 files changed, 400 insertions, 166 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index c4a21c6428f5..62a778012fe0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1591,6 +1591,7 @@ struct amdgpu_uvd {
 	struct amdgpu_bo	*vcpu_bo;
 	void			*cpu_addr;
 	uint64_t		gpu_addr;
+	void			*saved_bo;
 	atomic_t		handles[AMDGPU_MAX_UVD_HANDLES];
 	struct drm_file		*filp[AMDGPU_MAX_UVD_HANDLES];
 	struct delayed_work	idle_work;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
index 7a4b101e10c6..6043dc7c3a94 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
@@ -816,10 +816,13 @@ static int amdgpu_cgs_get_active_displays_info(void *cgs_device,
 	struct drm_device *ddev = adev->ddev;
 	struct drm_crtc *crtc;
 	uint32_t line_time_us, vblank_lines;
+	struct cgs_mode_info *mode_info;
 
 	if (info == NULL)
 		return -EINVAL;
 
+	mode_info = info->mode_info;
+
 	if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) {
 		list_for_each_entry(crtc,
 				&ddev->mode_config.crtc_list, head) {
@@ -828,7 +831,7 @@ static int amdgpu_cgs_get_active_displays_info(void *cgs_device,
 				info->active_display_mask |= (1 << amdgpu_crtc->crtc_id);
 				info->display_count++;
 			}
-			if (info->mode_info != NULL &&
+			if (mode_info != NULL &&
 				crtc->enabled && amdgpu_crtc->enabled &&
 				amdgpu_crtc->hw_mode.clock) {
 				line_time_us = (amdgpu_crtc->hw_mode.crtc_htotal * 1000) /
@@ -836,10 +839,10 @@ static int amdgpu_cgs_get_active_displays_info(void *cgs_device,
 				vblank_lines = amdgpu_crtc->hw_mode.crtc_vblank_end -
 							amdgpu_crtc->hw_mode.crtc_vdisplay +
 							(amdgpu_crtc->v_border * 2);
-				info->mode_info->vblank_time_us = vblank_lines * line_time_us;
-				info->mode_info->refresh_rate = drm_mode_vrefresh(&amdgpu_crtc->hw_mode);
-				info->mode_info->ref_clock = adev->clock.spll.reference_freq;
-				info->mode_info++;
+				mode_info->vblank_time_us = vblank_lines * line_time_us;
+				mode_info->refresh_rate = drm_mode_vrefresh(&amdgpu_crtc->hw_mode);
+				mode_info->ref_clock = adev->clock.spll.reference_freq;
+				mode_info = NULL;
 			}
 		}
 	}
@@ -847,6 +850,16 @@ static int amdgpu_cgs_get_active_displays_info(void *cgs_device,
 	return 0;
 }
 
+
+static int amdgpu_cgs_notify_dpm_enabled(void *cgs_device, bool enabled)
+{
+	CGS_FUNC_ADEV;
+
+	adev->pm.dpm_enabled = enabled;
+
+	return 0;
+}
+
 /** \brief evaluate acpi namespace object, handle or pathname must be valid
  *  \param cgs_device
  *  \param info input/output arguments for the control method
@@ -1097,6 +1110,7 @@ static const struct cgs_ops amdgpu_cgs_ops = {
 	amdgpu_cgs_set_powergating_state,
 	amdgpu_cgs_set_clockgating_state,
 	amdgpu_cgs_get_active_displays_info,
+	amdgpu_cgs_notify_dpm_enabled,
 	amdgpu_cgs_call_acpi_method,
 	amdgpu_cgs_query_system_info,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index f0ed974bd4e0..3fb405b3a614 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -57,7 +57,7 @@ static bool amdgpu_flip_handle_fence(struct amdgpu_flip_work *work,
 	if (!fence_add_callback(fence, &work->cb, amdgpu_flip_callback))
 		return true;
 
-	fence_put(*f);
+	fence_put(fence);
 	return false;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 4303b447efe8..d81f1f4883a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -121,7 +121,7 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct fence **f)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_fence *fence;
-	struct fence **ptr;
+	struct fence *old, **ptr;
 	uint32_t seq;
 
 	fence = kmem_cache_alloc(amdgpu_fence_slab, GFP_KERNEL);
@@ -141,7 +141,11 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct fence **f)
 	/* This function can't be called concurrently anyway, otherwise
 	 * emitting the fence would mess up the hardware ring buffer.
 	 */
-	BUG_ON(rcu_dereference_protected(*ptr, 1));
+	old = rcu_dereference_protected(*ptr, 1);
+	if (old && !fence_is_signaled(old)) {
+		DRM_INFO("rcu slot is busy\n");
+		fence_wait(old, false);
+	}
 
 	rcu_assign_pointer(*ptr, fence_get(&fence->base));
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index f594cfaa97e5..762cfdb85147 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -219,6 +219,8 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
 	if (r) {
 		return r;
 	}
+	adev->ddev->vblank_disable_allowed = true;
+
 	/* enable msi */
 	adev->irq.msi_enabled = false;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 7805a8706af7..598eb0cd5aab 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -382,6 +382,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		struct drm_amdgpu_info_vram_gtt vram_gtt;
 
 		vram_gtt.vram_size = adev->mc.real_vram_size;
+		vram_gtt.vram_size -= adev->vram_pin_size;
 		vram_gtt.vram_cpu_accessible_size = adev->mc.visible_vram_size;
 		vram_gtt.vram_cpu_accessible_size -= adev->vram_pin_size;
 		vram_gtt.gtt_size  = adev->mc.gtt_size;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 56d1458393cc..5b6639faa731 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -476,6 +476,17 @@ int amdgpu_bo_evict_vram(struct amdgpu_device *adev)
 	return ttm_bo_evict_mm(&adev->mman.bdev, TTM_PL_VRAM);
 }
 
+static const char *amdgpu_vram_names[] = {
+	"UNKNOWN",
+	"GDDR1",
+	"DDR2",
+	"GDDR3",
+	"GDDR4",
+	"GDDR5",
+	"HBM",
+	"DDR3"
+};
+
 int amdgpu_bo_init(struct amdgpu_device *adev)
 {
 	/* Add an MTRR for the VRAM */
@@ -484,8 +495,8 @@ int amdgpu_bo_init(struct amdgpu_device *adev)
 	DRM_INFO("Detected VRAM RAM=%lluM, BAR=%lluM\n",
 		adev->mc.mc_vram_size >> 20,
 		(unsigned long long)adev->mc.aper_size >> 20);
-	DRM_INFO("RAM width %dbits DDR\n",
-			adev->mc.vram_width);
+	DRM_INFO("RAM width %dbits %s\n",
+		 adev->mc.vram_width, amdgpu_vram_names[adev->mc.vram_type]);
 	return amdgpu_ttm_init(adev);
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c
index 3cb6d6c413c7..e9c6ae6ed2f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c
@@ -143,7 +143,7 @@ static int amdgpu_pp_late_init(void *handle)
 					adev->powerplay.pp_handle);
 
 #ifdef CONFIG_DRM_AMD_POWERPLAY
-	if (adev->pp_enabled) {
+	if (adev->pp_enabled && adev->pm.dpm_enabled) {
 		amdgpu_pm_sysfs_init(adev);
 		amdgpu_dpm_dispatch_task(adev, AMD_PP_EVENT_COMPLETE_INIT, NULL, NULL);
 	}
@@ -161,12 +161,8 @@ static int amdgpu_pp_sw_init(void *handle)
 					adev->powerplay.pp_handle);
 
 #ifdef CONFIG_DRM_AMD_POWERPLAY
-	if (adev->pp_enabled) {
-		if (amdgpu_dpm == 0)
-			adev->pm.dpm_enabled = false;
-		else
-			adev->pm.dpm_enabled = true;
-	}
+	if (adev->pp_enabled)
+		adev->pm.dpm_enabled = true;
 #endif
 
 	return ret;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
index c1a581044417..338da80006b6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
@@ -241,32 +241,28 @@ int amdgpu_uvd_sw_fini(struct amdgpu_device *adev)
 
 int amdgpu_uvd_suspend(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *ring = &adev->uvd.ring;
-	int i, r;
+	unsigned size;
+	void *ptr;
+	int i;
 
 	if (adev->uvd.vcpu_bo == NULL)
 		return 0;
 
-	for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) {
-		uint32_t handle = atomic_read(&adev->uvd.handles[i]);
-		if (handle != 0) {
-			struct fence *fence;
+	for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i)
+		if (atomic_read(&adev->uvd.handles[i]))
+			break;
 
-			amdgpu_uvd_note_usage(adev);
+	if (i == AMDGPU_MAX_UVD_HANDLES)
+		return 0;
 
-			r = amdgpu_uvd_get_destroy_msg(ring, handle, false, &fence);
-			if (r) {
-				DRM_ERROR("Error destroying UVD (%d)!\n", r);
-				continue;
-			}
+	size = amdgpu_bo_size(adev->uvd.vcpu_bo);
+	ptr = adev->uvd.cpu_addr;
 
-			fence_wait(fence, false);
-			fence_put(fence);
+	adev->uvd.saved_bo = kmalloc(size, GFP_KERNEL);
+	if (!adev->uvd.saved_bo)
+		return -ENOMEM;
 
-			adev->uvd.filp[i] = NULL;
-			atomic_set(&adev->uvd.handles[i], 0);
-		}
-	}
+	memcpy(adev->uvd.saved_bo, ptr, size);
 
 	return 0;
 }
@@ -275,23 +271,29 @@ int amdgpu_uvd_resume(struct amdgpu_device *adev)
 {
 	unsigned size;
 	void *ptr;
-	const struct common_firmware_header *hdr;
-	unsigned offset;
 
 	if (adev->uvd.vcpu_bo == NULL)
 		return -EINVAL;
 
-	hdr = (const struct common_firmware_header *)adev->uvd.fw->data;
-	offset = le32_to_cpu(hdr->ucode_array_offset_bytes);
-	memcpy(adev->uvd.cpu_addr, (adev->uvd.fw->data) + offset,
-		(adev->uvd.fw->size) - offset);
-
 	size = amdgpu_bo_size(adev->uvd.vcpu_bo);
-	size -= le32_to_cpu(hdr->ucode_size_bytes);
 	ptr = adev->uvd.cpu_addr;
-	ptr += le32_to_cpu(hdr->ucode_size_bytes);
 
-	memset(ptr, 0, size);
+	if (adev->uvd.saved_bo != NULL) {
+		memcpy(ptr, adev->uvd.saved_bo, size);
+		kfree(adev->uvd.saved_bo);
+		adev->uvd.saved_bo = NULL;
+	} else {
+		const struct common_firmware_header *hdr;
+		unsigned offset;
+
+		hdr = (const struct common_firmware_header *)adev->uvd.fw->data;
+		offset = le32_to_cpu(hdr->ucode_array_offset_bytes);
+		memcpy(adev->uvd.cpu_addr, (adev->uvd.fw->data) + offset,
+			(adev->uvd.fw->size) - offset);
+		size -= le32_to_cpu(hdr->ucode_size_bytes);
+		ptr += le32_to_cpu(hdr->ucode_size_bytes);
+		memset(ptr, 0, size);
+	}
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 82ce7d943884..05b0353d3880 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -903,14 +903,6 @@ static int gmc_v7_0_early_init(void *handle)
 	gmc_v7_0_set_gart_funcs(adev);
 	gmc_v7_0_set_irq_funcs(adev);
 
-	if (adev->flags & AMD_IS_APU) {
-		adev->mc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN;
-	} else {
-		u32 tmp = RREG32(mmMC_SEQ_MISC0);
-		tmp &= MC_SEQ_MISC0__MT__MASK;
-		adev->mc.vram_type = gmc_v7_0_convert_vram_type(tmp);
-	}
-
 	return 0;
 }
 
@@ -927,6 +919,14 @@ static int gmc_v7_0_sw_init(void *handle)
 	int dma_bits;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (adev->flags & AMD_IS_APU) {
+		adev->mc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN;
+	} else {
+		u32 tmp = RREG32(mmMC_SEQ_MISC0);
+		tmp &= MC_SEQ_MISC0__MT__MASK;
+		adev->mc.vram_type = gmc_v7_0_convert_vram_type(tmp);
+	}
+
 	r = amdgpu_irq_add_id(adev, 146, &adev->mc.vm_fault);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 29bd7b57dc91..02deb3229405 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -863,14 +863,6 @@ static int gmc_v8_0_early_init(void *handle)
 	gmc_v8_0_set_gart_funcs(adev);
 	gmc_v8_0_set_irq_funcs(adev);
 
-	if (adev->flags & AMD_IS_APU) {
-		adev->mc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN;
-	} else {
-		u32 tmp = RREG32(mmMC_SEQ_MISC0);
-		tmp &= MC_SEQ_MISC0__MT__MASK;
-		adev->mc.vram_type = gmc_v8_0_convert_vram_type(tmp);
-	}
-
 	return 0;
 }
 
@@ -881,12 +873,27 @@ static int gmc_v8_0_late_init(void *handle)
 	return amdgpu_irq_get(adev, &adev->mc.vm_fault, 0);
 }
 
+#define mmMC_SEQ_MISC0_FIJI 0xA71
+
 static int gmc_v8_0_sw_init(void *handle)
 {
 	int r;
 	int dma_bits;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (adev->flags & AMD_IS_APU) {
+		adev->mc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN;
+	} else {
+		u32 tmp;
+
+		if (adev->asic_type == CHIP_FIJI)
+			tmp = RREG32(mmMC_SEQ_MISC0_FIJI);
+		else
+			tmp = RREG32(mmMC_SEQ_MISC0);
+		tmp &= MC_SEQ_MISC0__MT__MASK;
+		adev->mc.vram_type = gmc_v8_0_convert_vram_type(tmp);
+	}
+
 	r = amdgpu_irq_add_id(adev, 146, &adev->mc.vm_fault);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
index c606ccb38d8b..cb463753115b 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
@@ -224,11 +224,11 @@ static int uvd_v4_2_suspend(void *handle)
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	r = amdgpu_uvd_suspend(adev);
+	r = uvd_v4_2_hw_fini(adev);
 	if (r)
 		return r;
 
-	r = uvd_v4_2_hw_fini(adev);
+	r = amdgpu_uvd_suspend(adev);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
index e3c852d9d79a..16476d80f475 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
@@ -220,11 +220,11 @@ static int uvd_v5_0_suspend(void *handle)
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	r = amdgpu_uvd_suspend(adev);
+	r = uvd_v5_0_hw_fini(adev);
 	if (r)
 		return r;
 
-	r = uvd_v5_0_hw_fini(adev);
+	r = amdgpu_uvd_suspend(adev);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
index 3375e614ac67..d49379145ef2 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
@@ -214,15 +214,16 @@ static int uvd_v6_0_suspend(void *handle)
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	r = uvd_v6_0_hw_fini(adev);
+	if (r)
+		return r;
+
 	/* Skip this for APU for now */
 	if (!(adev->flags & AMD_IS_APU)) {
 		r = amdgpu_uvd_suspend(adev);
 		if (r)
 			return r;
 	}
-	r = uvd_v6_0_hw_fini(adev);
-	if (r)
-		return r;
 
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/include/cgs_common.h b/drivers/gpu/drm/amd/include/cgs_common.h
index aec38fc3834f..ab84d4947247 100644
--- a/drivers/gpu/drm/amd/include/cgs_common.h
+++ b/drivers/gpu/drm/amd/include/cgs_common.h
@@ -589,6 +589,8 @@ typedef int(*cgs_get_active_displays_info)(
 					void *cgs_device,
 					struct cgs_display_info *info);
 
+typedef int (*cgs_notify_dpm_enabled)(void *cgs_device, bool enabled);
+
 typedef int (*cgs_call_acpi_method)(void *cgs_device,
 					uint32_t acpi_method,
 					uint32_t acpi_function,
@@ -644,6 +646,8 @@ struct cgs_ops {
 	cgs_set_clockgating_state set_clockgating_state;
 	/* display manager */
 	cgs_get_active_displays_info get_active_displays_info;
+	/* notify dpm enabled */
+	cgs_notify_dpm_enabled notify_dpm_enabled;
 	/* ACPI */
 	cgs_call_acpi_method call_acpi_method;
 	/* get system info */
@@ -734,8 +738,12 @@ struct cgs_device
 	CGS_CALL(set_powergating_state, dev, block_type, state)
 #define cgs_set_clockgating_state(dev, block_type, state)	\
 	CGS_CALL(set_clockgating_state, dev, block_type, state)
+#define cgs_notify_dpm_enabled(dev, enabled)	\
+	CGS_CALL(notify_dpm_enabled, dev, enabled)
+
 #define cgs_get_active_displays_info(dev, info)	\
 	CGS_CALL(get_active_displays_info, dev, info)
+
 #define cgs_call_acpi_method(dev, acpi_method, acpi_function, pintput, poutput, output_count, input_size, output_size)	\
 	CGS_CALL(call_acpi_method, dev, acpi_method, acpi_function, pintput, poutput, output_count, input_size, output_size)
 #define cgs_query_system_info(dev, sys_info)	\
diff --git a/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c b/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c
index 6b52c78cb404..56856a2864d1 100644
--- a/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c
+++ b/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c
@@ -137,14 +137,14 @@ static const pem_event_action *resume_event[] = {
 	reset_display_configCounter_tasks,
 	update_dal_configuration_tasks,
 	vari_bright_resume_tasks,
-	block_adjust_power_state_tasks,
 	setup_asic_tasks,
 	enable_stutter_mode_tasks, /*must do this in boot state and before SMC is started */
 	enable_dynamic_state_management_tasks,
 	enable_clock_power_gatings_tasks,
 	enable_disable_bapm_tasks,
 	initialize_thermal_controller_tasks,
-	reset_boot_state_tasks,
+	get_2d_performance_state_tasks,
+	set_performance_state_tasks,
 	adjust_power_state_tasks,
 	enable_disable_fps_tasks,
 	notify_hw_power_source_tasks,
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c
index 51dedf84623c..89f31bc5b68b 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c
@@ -2389,6 +2389,7 @@ static int fiji_populate_smc_vce_level(struct pp_hwmgr *hwmgr,
 
 	for(count = 0; count < table->VceLevelCount; count++) {
 		table->VceLevel[count].Frequency = mm_table->entries[count].eclk;
+		table->VceLevel[count].MinVoltage = 0;
 		table->VceLevel[count].MinVoltage |=
 				(mm_table->entries[count].vddc * VOLTAGE_SCALE) << VDDC_SHIFT;
 		table->VceLevel[count].MinVoltage |=
@@ -2465,6 +2466,7 @@ static int fiji_populate_smc_samu_level(struct pp_hwmgr *hwmgr,
 
 	for (count = 0; count < table->SamuLevelCount; count++) {
 		/* not sure whether we need evclk or not */
+		table->SamuLevel[count].MinVoltage = 0;
 		table->SamuLevel[count].Frequency = mm_table->entries[count].samclock;
 		table->SamuLevel[count].MinVoltage |= (mm_table->entries[count].vddc *
 				VOLTAGE_SCALE) << VDDC_SHIFT;
@@ -2562,6 +2564,7 @@ static int fiji_populate_smc_uvd_level(struct pp_hwmgr *hwmgr,
 	table->UvdBootLevel = 0;
 
 	for (count = 0; count < table->UvdLevelCount; count++) {
+		table->UvdLevel[count].MinVoltage = 0;
 		table->UvdLevel[count].VclkFrequency = mm_table->entries[count].vclk;
 		table->UvdLevel[count].DclkFrequency = mm_table->entries[count].dclk;
 		table->UvdLevel[count].MinVoltage |= (mm_table->entries[count].vddc *
@@ -2900,6 +2903,8 @@ static int fiji_init_smc_table(struct pp_hwmgr *hwmgr)
 	if(FIJI_VOLTAGE_CONTROL_NONE != data->voltage_control)
 		fiji_populate_smc_voltage_tables(hwmgr, table);
 
+	table->SystemFlags = 0;
+
 	if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps,
 			PHM_PlatformCaps_AutomaticDCTransition))
 		table->SystemFlags |= PPSMC_SYSTEMFLAG_GPIO_DC;
@@ -2997,6 +3002,7 @@ static int fiji_init_smc_table(struct pp_hwmgr *hwmgr)
 	table->MemoryThermThrottleEnable = 1;
 	table->PCIeBootLinkLevel = 0;      /* 0:Gen1 1:Gen2 2:Gen3*/
 	table->PCIeGenInterval = 1;
+	table->VRConfig = 0;
 
 	result = fiji_populate_vr_config(hwmgr, table);
 	PP_ASSERT_WITH_CODE(0 == result,
@@ -5195,6 +5201,67 @@ static int fiji_print_clock_levels(struct pp_hwmgr *hwmgr,
 	return size;
 }
 
+static inline bool fiji_are_power_levels_equal(const struct fiji_performance_level *pl1,
+							   const struct fiji_performance_level *pl2)
+{
+	return ((pl1->memory_clock == pl2->memory_clock) &&
+		  (pl1->engine_clock == pl2->engine_clock) &&
+		  (pl1->pcie_gen == pl2->pcie_gen) &&
+		  (pl1->pcie_lane == pl2->pcie_lane));
+}
+
+int fiji_check_states_equal(struct pp_hwmgr *hwmgr, const struct pp_hw_power_state *pstate1, const struct pp_hw_power_state *pstate2, bool *equal)
+{
+	const struct fiji_power_state *psa = cast_const_phw_fiji_power_state(pstate1);
+	const struct fiji_power_state *psb = cast_const_phw_fiji_power_state(pstate2);
+	int i;
+
+	if (equal == NULL || psa == NULL || psb == NULL)
+		return -EINVAL;
+
+	/* If the two states don't even have the same number of performance levels they cannot be the same state. */
+	if (psa->performance_level_count != psb->performance_level_count) {
+		*equal = false;
+		return 0;
+	}
+
+	for (i = 0; i < psa->performance_level_count; i++) {
+		if (!fiji_are_power_levels_equal(&(psa->performance_levels[i]), &(psb->performance_levels[i]))) {
+			/* If we have found even one performance level pair that is different the states are different. */
+			*equal = false;
+			return 0;
+		}
+	}
+
+	/* If all performance levels are the same try to use the UVD clocks to break the tie.*/
+	*equal = ((psa->uvd_clks.vclk == psb->uvd_clks.vclk) && (psa->uvd_clks.dclk == psb->uvd_clks.dclk));
+	*equal &= ((psa->vce_clks.evclk == psb->vce_clks.evclk) && (psa->vce_clks.ecclk == psb->vce_clks.ecclk));
+	*equal &= (psa->sclk_threshold == psb->sclk_threshold);
+	*equal &= (psa->acp_clk == psb->acp_clk);
+
+	return 0;
+}
+
+bool fiji_check_smc_update_required_for_display_configuration(struct pp_hwmgr *hwmgr)
+{
+	struct fiji_hwmgr *data = (struct fiji_hwmgr *)(hwmgr->backend);
+	bool is_update_required = false;
+	struct cgs_display_info info = {0,0,NULL};
+
+	cgs_get_active_displays_info(hwmgr->device, &info);
+
+	if (data->display_timing.num_existing_displays != info.display_count)
+		is_update_required = true;
+/* TO DO NEED TO GET DEEP SLEEP CLOCK FROM DAL
+	if (phm_cap_enabled(hwmgr->hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_SclkDeepSleep)) {
+		cgs_get_min_clock_settings(hwmgr->device, &min_clocks);
+		if(min_clocks.engineClockInSR != data->display_timing.minClockInSR)
+			is_update_required = true;
+*/
+	return is_update_required;
+}
+
+
 static const struct pp_hwmgr_func fiji_hwmgr_funcs = {
 	.backend_init = &fiji_hwmgr_backend_init,
 	.backend_fini = &tonga_hwmgr_backend_fini,
@@ -5230,6 +5297,8 @@ static const struct pp_hwmgr_func fiji_hwmgr_funcs = {
 	.register_internal_thermal_interrupt = fiji_register_internal_thermal_interrupt,
 	.set_fan_control_mode = fiji_set_fan_control_mode,
 	.get_fan_control_mode = fiji_get_fan_control_mode,
+	.check_states_equal = fiji_check_states_equal,
+	.check_smc_update_required_for_display_configuration = fiji_check_smc_update_required_for_display_configuration,
 	.get_pp_table = fiji_get_pp_table,
 	.set_pp_table = fiji_set_pp_table,
 	.force_clock_level = fiji_force_clock_level,
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
index be31bed2538a..fa208ada6892 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
@@ -58,6 +58,9 @@ void phm_init_dynamic_caps(struct pp_hwmgr *hwmgr)
 
 	phm_cap_unset(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_VpuRecoveryInProgress);
 
+	phm_cap_set(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_UVDDPM);
+	phm_cap_set(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_VCEDPM);
+
 	if (acpi_atcs_functions_supported(hwmgr->device, ATCS_FUNCTION_PCIE_PERFORMANCE_REQUEST) &&
 		acpi_atcs_functions_supported(hwmgr->device, ATCS_FUNCTION_PCIE_DEVICE_READY_NOTIFICATION))
 		phm_cap_set(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_PCIEPerformanceRequest);
@@ -130,18 +133,25 @@ int phm_set_power_state(struct pp_hwmgr *hwmgr,
 
 int phm_enable_dynamic_state_management(struct pp_hwmgr *hwmgr)
 {
+	int ret = 1;
+	bool enabled;
 	PHM_FUNC_CHECK(hwmgr);
 
 	if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps,
 		PHM_PlatformCaps_TablelessHardwareInterface)) {
 		if (NULL != hwmgr->hwmgr_func->dynamic_state_management_enable)
-			return hwmgr->hwmgr_func->dynamic_state_management_enable(hwmgr);
+			ret = hwmgr->hwmgr_func->dynamic_state_management_enable(hwmgr);
 	} else {
-		return phm_dispatch_table(hwmgr,
+		ret = phm_dispatch_table(hwmgr,
 				&(hwmgr->enable_dynamic_state_management),
 				NULL, NULL);
 	}
-	return 0;
+
+	enabled = ret == 0 ? true : false;
+
+	cgs_notify_dpm_enabled(hwmgr->device, enabled);
+
+	return ret;
 }
 
 int phm_force_dpm_levels(struct pp_hwmgr *hwmgr, enum amd_dpm_forced_level level)
diff --git a/drivers/gpu/drm/arm/hdlcd_drv.c b/drivers/gpu/drm/arm/hdlcd_drv.c
index 56b829f97699..3ac1ae4d8caf 100644
--- a/drivers/gpu/drm/arm/hdlcd_drv.c
+++ b/drivers/gpu/drm/arm/hdlcd_drv.c
@@ -57,14 +57,13 @@ static int hdlcd_load(struct drm_device *drm, unsigned long flags)
 		DRM_ERROR("failed to map control registers area\n");
 		ret = PTR_ERR(hdlcd->mmio);
 		hdlcd->mmio = NULL;
-		goto fail;
+		return ret;
 	}
 
 	version = hdlcd_read(hdlcd, HDLCD_REG_VERSION);
 	if ((version & HDLCD_PRODUCT_MASK) != HDLCD_PRODUCT_ID) {
 		DRM_ERROR("unknown product id: 0x%x\n", version);
-		ret = -EINVAL;
-		goto fail;
+		return -EINVAL;
 	}
 	DRM_INFO("found ARM HDLCD version r%dp%d\n",
 		(version & HDLCD_VERSION_MAJOR_MASK) >> 8,
@@ -73,7 +72,7 @@ static int hdlcd_load(struct drm_device *drm, unsigned long flags)
 	/* Get the optional framebuffer memory resource */
 	ret = of_reserved_mem_device_init(drm->dev);
 	if (ret && ret != -ENODEV)
-		goto fail;
+		return ret;
 
 	ret = dma_set_mask_and_coherent(drm->dev, DMA_BIT_MASK(32));
 	if (ret)
@@ -101,8 +100,6 @@ irq_fail:
 	drm_crtc_cleanup(&hdlcd->crtc);
 setup_fail:
 	of_reserved_mem_device_release(drm->dev);
-fail:
-	devm_clk_put(drm->dev, hdlcd->clk);
 
 	return ret;
 }
@@ -412,7 +409,6 @@ err_unload:
 	pm_runtime_put_sync(drm->dev);
 	pm_runtime_disable(drm->dev);
 	of_reserved_mem_device_release(drm->dev);
-	devm_clk_put(dev, hdlcd->clk);
 err_free:
 	drm_dev_unref(drm);
 
@@ -436,10 +432,6 @@ static void hdlcd_drm_unbind(struct device *dev)
 	pm_runtime_put_sync(drm->dev);
 	pm_runtime_disable(drm->dev);
 	of_reserved_mem_device_release(drm->dev);
-	if (!IS_ERR(hdlcd->clk)) {
-		devm_clk_put(drm->dev, hdlcd->clk);
-		hdlcd->clk = NULL;
-	}
 	drm_mode_config_cleanup(drm);
 	drm_dev_unregister(drm);
 	drm_dev_unref(drm);
diff --git a/drivers/gpu/drm/imx/dw_hdmi-imx.c b/drivers/gpu/drm/imx/dw_hdmi-imx.c
index 2a95d10e9d92..a24631fdf4ad 100644
--- a/drivers/gpu/drm/imx/dw_hdmi-imx.c
+++ b/drivers/gpu/drm/imx/dw_hdmi-imx.c
@@ -225,8 +225,6 @@ static int dw_hdmi_imx_bind(struct device *dev, struct device *master,
 	if (!iores)
 		return -ENXIO;
 
-	platform_set_drvdata(pdev, hdmi);
-
 	encoder->possible_crtcs = drm_of_find_possible_crtcs(drm, dev->of_node);
 	/*
 	 * If we failed to find the CRTC(s) which this encoder is
@@ -245,7 +243,16 @@ static int dw_hdmi_imx_bind(struct device *dev, struct device *master,
 	drm_encoder_init(drm, encoder, &dw_hdmi_imx_encoder_funcs,
 			 DRM_MODE_ENCODER_TMDS, NULL);
 
-	return dw_hdmi_bind(dev, master, data, encoder, iores, irq, plat_data);
+	ret = dw_hdmi_bind(dev, master, data, encoder, iores, irq, plat_data);
+
+	/*
+	 * If dw_hdmi_bind() fails we'll never call dw_hdmi_unbind(),
+	 * which would have called the encoder cleanup.  Do it manually.
+	 */
+	if (ret)
+		drm_encoder_cleanup(encoder);
+
+	return ret;
 }
 
 static void dw_hdmi_imx_unbind(struct device *dev, struct device *master,
diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c
index 9876e0f0c3e1..e26dcdec2aba 100644
--- a/drivers/gpu/drm/imx/imx-drm-core.c
+++ b/drivers/gpu/drm/imx/imx-drm-core.c
@@ -326,7 +326,6 @@ int imx_drm_add_crtc(struct drm_device *drm, struct drm_crtc *crtc,
 {
 	struct imx_drm_device *imxdrm = drm->dev_private;
 	struct imx_drm_crtc *imx_drm_crtc;
-	int ret;
 
 	/*
 	 * The vblank arrays are dimensioned by MAX_CRTC - we can't
@@ -351,10 +350,6 @@ int imx_drm_add_crtc(struct drm_device *drm, struct drm_crtc *crtc,
 
 	*new_crtc = imx_drm_crtc;
 
-	ret = drm_mode_crtc_set_gamma_size(imx_drm_crtc->crtc, 256);
-	if (ret)
-		goto err_register;
-
 	drm_crtc_helper_add(crtc,
 			imx_drm_crtc->imx_drm_helper_funcs.crtc_helper_funcs);
 
@@ -362,11 +357,6 @@ int imx_drm_add_crtc(struct drm_device *drm, struct drm_crtc *crtc,
 			imx_drm_crtc->imx_drm_helper_funcs.crtc_funcs, NULL);
 
 	return 0;
-
-err_register:
-	imxdrm->crtc[--imxdrm->pipes] = NULL;
-	kfree(imx_drm_crtc);
-	return ret;
 }
 EXPORT_SYMBOL_GPL(imx_drm_add_crtc);
 
diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c
index 588827844f30..681ec6eb77d9 100644
--- a/drivers/gpu/drm/imx/ipuv3-plane.c
+++ b/drivers/gpu/drm/imx/ipuv3-plane.c
@@ -72,22 +72,101 @@ static inline int calc_bandwidth(int width, int height, unsigned int vref)
 int ipu_plane_set_base(struct ipu_plane *ipu_plane, struct drm_framebuffer *fb,
 		       int x, int y)
 {
-	struct drm_gem_cma_object *cma_obj;
-	unsigned long eba;
-	int active;
-
-	cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
-	if (!cma_obj) {
-		DRM_DEBUG_KMS("entry is null.\n");
-		return -EFAULT;
+	struct drm_gem_cma_object *cma_obj[3];
+	unsigned long eba, ubo, vbo;
+	int active, i;
+
+	for (i = 0; i < drm_format_num_planes(fb->pixel_format); i++) {
+		cma_obj[i] = drm_fb_cma_get_gem_obj(fb, i);
+		if (!cma_obj[i]) {
+			DRM_DEBUG_KMS("plane %d entry is null.\n", i);
+			return -EFAULT;
+		}
 	}
 
-	dev_dbg(ipu_plane->base.dev->dev, "phys = %pad, x = %d, y = %d",
-		&cma_obj->paddr, x, y);
-
-	eba = cma_obj->paddr + fb->offsets[0] +
+	eba = cma_obj[0]->paddr + fb->offsets[0] +
 	      fb->pitches[0] * y + (fb->bits_per_pixel >> 3) * x;
 
+	if (eba & 0x7) {
+		DRM_DEBUG_KMS("base address must be a multiple of 8.\n");
+		return -EINVAL;
+	}
+
+	if (fb->pitches[0] < 1 || fb->pitches[0] > 16384) {
+		DRM_DEBUG_KMS("pitches out of range.\n");
+		return -EINVAL;
+	}
+
+	if (ipu_plane->enabled && fb->pitches[0] != ipu_plane->stride[0]) {
+		DRM_DEBUG_KMS("pitches must not change while plane is enabled.\n");
+		return -EINVAL;
+	}
+
+	ipu_plane->stride[0] = fb->pitches[0];
+
+	switch (fb->pixel_format) {
+	case DRM_FORMAT_YUV420:
+	case DRM_FORMAT_YVU420:
+		/*
+		 * Multiplanar formats have to meet the following restrictions:
+		 * - The (up to) three plane addresses are EBA, EBA+UBO, EBA+VBO
+		 * - EBA, UBO and VBO are a multiple of 8
+		 * - UBO and VBO are unsigned and not larger than 0xfffff8
+		 * - Only EBA may be changed while scanout is active
+		 * - The strides of U and V planes must be identical.
+		 */
+		ubo = cma_obj[1]->paddr + fb->offsets[1] +
+		      fb->pitches[1] * y / 2 + x / 2 - eba;
+		vbo = cma_obj[2]->paddr + fb->offsets[2] +
+		      fb->pitches[2] * y / 2 + x / 2 - eba;
+
+		if ((ubo & 0x7) || (vbo & 0x7)) {
+			DRM_DEBUG_KMS("U/V buffer offsets must be a multiple of 8.\n");
+			return -EINVAL;
+		}
+
+		if ((ubo > 0xfffff8) || (vbo > 0xfffff8)) {
+			DRM_DEBUG_KMS("U/V buffer offsets must be positive and not larger than 0xfffff8.\n");
+			return -EINVAL;
+		}
+
+		if (ipu_plane->enabled && ((ipu_plane->u_offset != ubo) ||
+					   (ipu_plane->v_offset != vbo))) {
+			DRM_DEBUG_KMS("U/V buffer offsets must not change while plane is enabled.\n");
+			return -EINVAL;
+		}
+
+		if (fb->pitches[1] != fb->pitches[2]) {
+			DRM_DEBUG_KMS("U/V pitches must be identical.\n");
+			return -EINVAL;
+		}
+
+		if (fb->pitches[1] < 1 || fb->pitches[1] > 16384) {
+			DRM_DEBUG_KMS("U/V pitches out of range.\n");
+			return -EINVAL;
+		}
+
+		if (ipu_plane->enabled &&
+		    (ipu_plane->stride[1] != fb->pitches[1])) {
+			DRM_DEBUG_KMS("U/V pitches must not change while plane is enabled.\n");
+			return -EINVAL;
+		}
+
+		ipu_plane->u_offset = ubo;
+		ipu_plane->v_offset = vbo;
+		ipu_plane->stride[1] = fb->pitches[1];
+
+		dev_dbg(ipu_plane->base.dev->dev,
+			"phys = %pad %pad %pad, x = %d, y = %d",
+			&cma_obj[0]->paddr, &cma_obj[1]->paddr,
+			&cma_obj[2]->paddr, x, y);
+		break;
+	default:
+		dev_dbg(ipu_plane->base.dev->dev, "phys = %pad, x = %d, y = %d",
+			&cma_obj[0]->paddr, x, y);
+		break;
+	}
+
 	if (ipu_plane->enabled) {
 		active = ipu_idmac_get_current_buffer(ipu_plane->ipu_ch);
 		ipu_cpmem_set_buffer(ipu_plane->ipu_ch, !active, eba);
@@ -201,12 +280,6 @@ int ipu_plane_mode_set(struct ipu_plane *ipu_plane, struct drm_crtc *crtc,
 		}
 	}
 
-	ret = ipu_dmfc_init_channel(ipu_plane->dmfc, crtc_w);
-	if (ret) {
-		dev_err(dev, "initializing dmfc channel failed with %d\n", ret);
-		return ret;
-	}
-
 	ret = ipu_dmfc_alloc_bandwidth(ipu_plane->dmfc,
 			calc_bandwidth(crtc_w, crtc_h,
 				       calc_vref(mode)), 64);
@@ -215,6 +288,8 @@ int ipu_plane_mode_set(struct ipu_plane *ipu_plane, struct drm_crtc *crtc,
 		return ret;
 	}
 
+	ipu_dmfc_config_wait4eot(ipu_plane->dmfc, crtc_w);
+
 	ipu_cpmem_zero(ipu_plane->ipu_ch);
 	ipu_cpmem_set_resolution(ipu_plane->ipu_ch, src_w, src_h);
 	ret = ipu_cpmem_set_fmt(ipu_plane->ipu_ch, fb->pixel_format);
@@ -233,6 +308,18 @@ int ipu_plane_mode_set(struct ipu_plane *ipu_plane, struct drm_crtc *crtc,
 	if (interlaced)
 		ipu_cpmem_interlaced_scan(ipu_plane->ipu_ch, fb->pitches[0]);
 
+	if (fb->pixel_format == DRM_FORMAT_YUV420) {
+		ipu_cpmem_set_yuv_planar_full(ipu_plane->ipu_ch,
+					      ipu_plane->stride[1],
+					      ipu_plane->u_offset,
+					      ipu_plane->v_offset);
+	} else if (fb->pixel_format == DRM_FORMAT_YVU420) {
+		ipu_cpmem_set_yuv_planar_full(ipu_plane->ipu_ch,
+					      ipu_plane->stride[1],
+					      ipu_plane->v_offset,
+					      ipu_plane->u_offset);
+	}
+
 	ipu_plane->w = src_w;
 	ipu_plane->h = src_h;
 
diff --git a/drivers/gpu/drm/imx/ipuv3-plane.h b/drivers/gpu/drm/imx/ipuv3-plane.h
index 3a443b413c60..4448fd4ad4eb 100644
--- a/drivers/gpu/drm/imx/ipuv3-plane.h
+++ b/drivers/gpu/drm/imx/ipuv3-plane.h
@@ -29,6 +29,10 @@ struct ipu_plane {
 	int			w;
 	int			h;
 
+	unsigned int		u_offset;
+	unsigned int		v_offset;
+	unsigned int		stride[2];
+
 	bool			enabled;
 };
 
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/tegra.h b/drivers/gpu/drm/nouveau/include/nvkm/core/tegra.h
index 16641cec18a2..b5370cb56e3c 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/core/tegra.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/core/tegra.h
@@ -11,6 +11,7 @@ struct nvkm_device_tegra {
 
 	struct reset_control *rst;
 	struct clk *clk;
+	struct clk *clk_ref;
 	struct clk *clk_pwr;
 
 	struct regulator *vdd;
@@ -36,6 +37,10 @@ struct nvkm_device_tegra_func {
 	 * bypassed). A value of 0 means an IOMMU is never used.
 	 */
 	u8 iommu_bit;
+	/*
+	 * Whether the chip requires a reference clock
+	 */
+	bool require_ref_clk;
 };
 
 int nvkm_device_tegra_new(const struct nvkm_device_tegra_func *,
diff --git a/drivers/gpu/drm/nouveau/nouveau_platform.c b/drivers/gpu/drm/nouveau/nouveau_platform.c
index 2dfe58af12e4..4c4cc2260257 100644
--- a/drivers/gpu/drm/nouveau/nouveau_platform.c
+++ b/drivers/gpu/drm/nouveau/nouveau_platform.c
@@ -55,6 +55,11 @@ static const struct nvkm_device_tegra_func gk20a_platform_data = {
 	.iommu_bit = 34,
 };
 
+static const struct nvkm_device_tegra_func gm20b_platform_data = {
+	.iommu_bit = 34,
+	.require_ref_clk = true,
+};
+
 static const struct of_device_id nouveau_platform_match[] = {
 	{
 		.compatible = "nvidia,gk20a",
@@ -62,7 +67,7 @@ static const struct of_device_id nouveau_platform_match[] = {
 	},
 	{
 		.compatible = "nvidia,gm20b",
-		.data = &gk20a_platform_data,
+		.data = &gm20b_platform_data,
 	},
 	{ }
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
index 9afa5f3e3c1c..ec12efb4689a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
@@ -35,6 +35,11 @@ nvkm_device_tegra_power_up(struct nvkm_device_tegra *tdev)
 	ret = clk_prepare_enable(tdev->clk);
 	if (ret)
 		goto err_clk;
+	if (tdev->clk_ref) {
+		ret = clk_prepare_enable(tdev->clk_ref);
+		if (ret)
+			goto err_clk_ref;
+	}
 	ret = clk_prepare_enable(tdev->clk_pwr);
 	if (ret)
 		goto err_clk_pwr;
@@ -57,6 +62,9 @@ nvkm_device_tegra_power_up(struct nvkm_device_tegra *tdev)
 err_clamp:
 	clk_disable_unprepare(tdev->clk_pwr);
 err_clk_pwr:
+	if (tdev->clk_ref)
+		clk_disable_unprepare(tdev->clk_ref);
+err_clk_ref:
 	clk_disable_unprepare(tdev->clk);
 err_clk:
 	regulator_disable(tdev->vdd);
@@ -71,6 +79,8 @@ nvkm_device_tegra_power_down(struct nvkm_device_tegra *tdev)
 	udelay(10);
 
 	clk_disable_unprepare(tdev->clk_pwr);
+	if (tdev->clk_ref)
+		clk_disable_unprepare(tdev->clk_ref);
 	clk_disable_unprepare(tdev->clk);
 	udelay(10);
 
@@ -274,6 +284,13 @@ nvkm_device_tegra_new(const struct nvkm_device_tegra_func *func,
 		goto free;
 	}
 
+	if (func->require_ref_clk)
+		tdev->clk_ref = devm_clk_get(&pdev->dev, "ref");
+	if (IS_ERR(tdev->clk_ref)) {
+		ret = PTR_ERR(tdev->clk_ref);
+		goto free;
+	}
+
 	tdev->clk_pwr = devm_clk_get(&pdev->dev, "pwr");
 	if (IS_ERR(tdev->clk_pwr)) {
 		ret = PTR_ERR(tdev->clk_pwr);
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index cf61e0856f4a..b80b08f71cb4 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -275,13 +275,15 @@ void atombios_crtc_dpms(struct drm_crtc *crtc, int mode)
 		if (ASIC_IS_DCE3(rdev) && !ASIC_IS_DCE6(rdev))
 			atombios_enable_crtc_memreq(crtc, ATOM_ENABLE);
 		atombios_blank_crtc(crtc, ATOM_DISABLE);
-		drm_vblank_on(dev, radeon_crtc->crtc_id);
+		if (dev->num_crtcs > radeon_crtc->crtc_id)
+			drm_vblank_on(dev, radeon_crtc->crtc_id);
 		radeon_crtc_load_lut(crtc);
 		break;
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 	case DRM_MODE_DPMS_OFF:
-		drm_vblank_off(dev, radeon_crtc->crtc_id);
+		if (dev->num_crtcs > radeon_crtc->crtc_id)
+			drm_vblank_off(dev, radeon_crtc->crtc_id);
 		if (radeon_crtc->enabled)
 			atombios_blank_crtc(crtc, ATOM_ENABLE);
 		if (ASIC_IS_DCE3(rdev) && !ASIC_IS_DCE6(rdev))
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index 979f3bf65f2c..1e9304d1c88f 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -291,6 +291,8 @@ int radeon_irq_kms_init(struct radeon_device *rdev)
 	if (r) {
 		return r;
 	}
+	rdev->ddev->vblank_disable_allowed = true;
+
 	/* enable msi */
 	rdev->msi_enabled = 0;
 
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
index 24152dfef199..478d4099b0d0 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
@@ -331,13 +331,15 @@ static void radeon_crtc_dpms(struct drm_crtc *crtc, int mode)
 									 RADEON_CRTC_DISP_REQ_EN_B));
 			WREG32_P(RADEON_CRTC_EXT_CNTL, crtc_ext_cntl, ~(mask | crtc_ext_cntl));
 		}
-		drm_vblank_on(dev, radeon_crtc->crtc_id);
+		if (dev->num_crtcs > radeon_crtc->crtc_id)
+			drm_vblank_on(dev, radeon_crtc->crtc_id);
 		radeon_crtc_load_lut(crtc);
 		break;
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 	case DRM_MODE_DPMS_OFF:
-		drm_vblank_off(dev, radeon_crtc->crtc_id);
+		if (dev->num_crtcs > radeon_crtc->crtc_id)
+			drm_vblank_off(dev, radeon_crtc->crtc_id);
 		if (radeon_crtc->crtc_id)
 			WREG32_P(RADEON_CRTC2_GEN_CNTL, mask, ~(RADEON_CRTC2_EN | mask));
 		else {
diff --git a/drivers/gpu/ipu-v3/ipu-cpmem.c b/drivers/gpu/ipu-v3/ipu-cpmem.c
index 883a314cd83a..6494a4d28171 100644
--- a/drivers/gpu/ipu-v3/ipu-cpmem.c
+++ b/drivers/gpu/ipu-v3/ipu-cpmem.c
@@ -395,60 +395,48 @@ void ipu_cpmem_set_yuv_interleaved(struct ipuv3_channel *ch, u32 pixel_format)
 EXPORT_SYMBOL_GPL(ipu_cpmem_set_yuv_interleaved);
 
 void ipu_cpmem_set_yuv_planar_full(struct ipuv3_channel *ch,
-				   u32 pixel_format, int stride,
-				   int u_offset, int v_offset)
+				   unsigned int uv_stride,
+				   unsigned int u_offset, unsigned int v_offset)
 {
-	switch (pixel_format) {
-	case V4L2_PIX_FMT_YUV420:
-	case V4L2_PIX_FMT_YUV422P:
-		ipu_ch_param_write_field(ch, IPU_FIELD_SLUV, (stride / 2) - 1);
-		ipu_ch_param_write_field(ch, IPU_FIELD_UBO, u_offset / 8);
-		ipu_ch_param_write_field(ch, IPU_FIELD_VBO, v_offset / 8);
-		break;
-	case V4L2_PIX_FMT_YVU420:
-		ipu_ch_param_write_field(ch, IPU_FIELD_SLUV, (stride / 2) - 1);
-		ipu_ch_param_write_field(ch, IPU_FIELD_UBO, v_offset / 8);
-		ipu_ch_param_write_field(ch, IPU_FIELD_VBO, u_offset / 8);
-		break;
-	case V4L2_PIX_FMT_NV12:
-	case V4L2_PIX_FMT_NV16:
-		ipu_ch_param_write_field(ch, IPU_FIELD_SLUV, stride - 1);
-		ipu_ch_param_write_field(ch, IPU_FIELD_UBO, u_offset / 8);
-		ipu_ch_param_write_field(ch, IPU_FIELD_VBO, u_offset / 8);
-		break;
-	}
+	ipu_ch_param_write_field(ch, IPU_FIELD_SLUV, uv_stride - 1);
+	ipu_ch_param_write_field(ch, IPU_FIELD_UBO, u_offset / 8);
+	ipu_ch_param_write_field(ch, IPU_FIELD_VBO, v_offset / 8);
 }
 EXPORT_SYMBOL_GPL(ipu_cpmem_set_yuv_planar_full);
 
 void ipu_cpmem_set_yuv_planar(struct ipuv3_channel *ch,
 			      u32 pixel_format, int stride, int height)
 {
-	int u_offset, v_offset;
+	int fourcc, u_offset, v_offset;
 	int uv_stride = 0;
 
-	switch (pixel_format) {
-	case V4L2_PIX_FMT_YUV420:
-	case V4L2_PIX_FMT_YVU420:
+	fourcc = v4l2_pix_fmt_to_drm_fourcc(pixel_format);
+	switch (fourcc) {
+	case DRM_FORMAT_YUV420:
 		uv_stride = stride / 2;
 		u_offset = stride * height;
 		v_offset = u_offset + (uv_stride * height / 2);
-		ipu_cpmem_set_yuv_planar_full(ch, pixel_format, stride,
-					      u_offset, v_offset);
 		break;
-	case V4L2_PIX_FMT_YUV422P:
+	case DRM_FORMAT_YVU420:
+		uv_stride = stride / 2;
+		v_offset = stride * height;
+		u_offset = v_offset + (uv_stride * height / 2);
+		break;
+	case DRM_FORMAT_YUV422:
 		uv_stride = stride / 2;
 		u_offset = stride * height;
 		v_offset = u_offset + (uv_stride * height);
-		ipu_cpmem_set_yuv_planar_full(ch, pixel_format, stride,
-					      u_offset, v_offset);
 		break;
-	case V4L2_PIX_FMT_NV12:
-	case V4L2_PIX_FMT_NV16:
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_NV16:
+		uv_stride = stride;
 		u_offset = stride * height;
-		ipu_cpmem_set_yuv_planar_full(ch, pixel_format, stride,
-					      u_offset, 0);
+		v_offset = 0;
 		break;
+	default:
+		return;
 	}
+	ipu_cpmem_set_yuv_planar_full(ch, uv_stride, u_offset, v_offset);
 }
 EXPORT_SYMBOL_GPL(ipu_cpmem_set_yuv_planar);
 
@@ -684,17 +672,25 @@ int ipu_cpmem_set_image(struct ipuv3_channel *ch, struct ipu_image *image)
 
 	switch (pix->pixelformat) {
 	case V4L2_PIX_FMT_YUV420:
-	case V4L2_PIX_FMT_YVU420:
 		offset = Y_OFFSET(pix, image->rect.left, image->rect.top);
 		u_offset = U_OFFSET(pix, image->rect.left,
 				    image->rect.top) - offset;
 		v_offset = V_OFFSET(pix, image->rect.left,
 				    image->rect.top) - offset;
 
-		ipu_cpmem_set_yuv_planar_full(ch, pix->pixelformat,
-					      pix->bytesperline,
+		ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline / 2,
 					      u_offset, v_offset);
 		break;
+	case V4L2_PIX_FMT_YVU420:
+		offset = Y_OFFSET(pix, image->rect.left, image->rect.top);
+		u_offset = U_OFFSET(pix, image->rect.left,
+				    image->rect.top) - offset;
+		v_offset = V_OFFSET(pix, image->rect.left,
+				    image->rect.top) - offset;
+
+		ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline / 2,
+					      v_offset, u_offset);
+		break;
 	case V4L2_PIX_FMT_YUV422P:
 		offset = Y_OFFSET(pix, image->rect.left, image->rect.top);
 		u_offset = U2_OFFSET(pix, image->rect.left,
@@ -702,8 +698,7 @@ int ipu_cpmem_set_image(struct ipuv3_channel *ch, struct ipu_image *image)
 		v_offset = V2_OFFSET(pix, image->rect.left,
 				     image->rect.top) - offset;
 
-		ipu_cpmem_set_yuv_planar_full(ch, pix->pixelformat,
-					      pix->bytesperline,
+		ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline / 2,
 					      u_offset, v_offset);
 		break;
 	case V4L2_PIX_FMT_NV12:
@@ -712,8 +707,7 @@ int ipu_cpmem_set_image(struct ipuv3_channel *ch, struct ipu_image *image)
 				     image->rect.top) - offset;
 		v_offset = 0;
 
-		ipu_cpmem_set_yuv_planar_full(ch, pix->pixelformat,
-					      pix->bytesperline,
+		ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline,
 					      u_offset, v_offset);
 		break;
 	case V4L2_PIX_FMT_NV16:
@@ -722,8 +716,7 @@ int ipu_cpmem_set_image(struct ipuv3_channel *ch, struct ipu_image *image)
 				      image->rect.top) - offset;
 		v_offset = 0;
 
-		ipu_cpmem_set_yuv_planar_full(ch, pix->pixelformat,
-					      pix->bytesperline,
+		ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline,
 					      u_offset, v_offset);
 		break;
 	case V4L2_PIX_FMT_UYVY:
diff --git a/drivers/gpu/ipu-v3/ipu-dmfc.c b/drivers/gpu/ipu-v3/ipu-dmfc.c
index 042c3958e2a0..837b1ec22800 100644
--- a/drivers/gpu/ipu-v3/ipu-dmfc.c
+++ b/drivers/gpu/ipu-v3/ipu-dmfc.c
@@ -350,11 +350,13 @@ out:
 }
 EXPORT_SYMBOL_GPL(ipu_dmfc_alloc_bandwidth);
 
-int ipu_dmfc_init_channel(struct dmfc_channel *dmfc, int width)
+void ipu_dmfc_config_wait4eot(struct dmfc_channel *dmfc, int width)
 {
 	struct ipu_dmfc_priv *priv = dmfc->priv;
 	u32 dmfc_gen1;
 
+	mutex_lock(&priv->mutex);
+
 	dmfc_gen1 = readl(priv->base + DMFC_GENERAL1);
 
 	if ((dmfc->slots * 64 * 4) / width > dmfc->data->max_fifo_lines)
@@ -364,9 +366,9 @@ int ipu_dmfc_init_channel(struct dmfc_channel *dmfc, int width)
 
 	writel(dmfc_gen1, priv->base + DMFC_GENERAL1);
 
-	return 0;
+	mutex_unlock(&priv->mutex);
 }
-EXPORT_SYMBOL_GPL(ipu_dmfc_init_channel);
+EXPORT_SYMBOL_GPL(ipu_dmfc_config_wait4eot);
 
 struct dmfc_channel *ipu_dmfc_get(struct ipu_soc *ipu, int ipu_channel)
 {