Merge tag 'drm-next-2021-02-19' of git://anongit.freedesktop.org/drm/drm

Pull drm updates from Dave Airlie:
 "A pretty normal tree, lots of refactoring across the board, ttm, i915,
  nouveau, and bunch of features in various drivers.

  docs:
   - lots of updated docs

  core:
   - require crtc to have unique primary plane
   - fourcc macro fix
   - PCI bar quirk for bar resizing
   - don't sent hotplug on error
   - move vm code to legacy
   - nuke hose only used on old oboslete alpha

  dma-buf:
   - kernel doc updates
   - improved lock tracking

  dp/hdmi:
   - DP-HDMI2.1 protocol converter support

  ttm:
   - bo size handling cleanup
   - release a pinned bo warning
   - cleanup lru handler
   - avoid using pages with drm_prime_sg_to_page_addr_arrays

  cma-helper:
   - prime/mmap fixes

  bridge:
   - add DP support

  gma500:
   - remove gma3600 support

  i915:
   - try eDP fast/narrow link again with fallback
   - Intel eDP backlight control
   - replace display register read/write macros
   - refactor intel_display.c
   - display power improvements
   - HPD code cleanup
   - Rocketlake display fixes
   - Power/backlight/RPM fixes
   - DG1 display fix
   - IVB/BYT clear residuals security fix again
   - make i915 mitigations options via parameter
   - HSW GT1 GPU hangs fixes
   - DG1 workaround hang fixes
   - TGL DMAR hang avoidance
   - Lots of GT fixes
   - follow on fixes for residuals clear
   - gen7 per-engine-reset support
   - HDCP2.2 + HDCP1.4 GEN12 DP MST support
   - TGL clear color support
   - backlight refactoring
   - VRR/Adaptive sync enabling on DP/EDP for TGL+
   - async flips for all ilk+

  amdgpu:
   - rework IH ring handling (Vega/Navi)
   - rework HDP handling (Vega/Navi)
   - swSMU updates for renoir/vangogh
   - Sienna Cichild overdrive support
   - FP16 on DCE8-11 support
   - GPU reset on navy flounder/vangogh
   - SMU profile fixes for APU
   - SR-IOV fixes
   - Vangogh SMU fixes
   - fan speed control fixes

  amdkfd:
   - config handling fix
   - buffer free fix
   - recursive lock warnings fix

  nouveau:
   - Turing MMU fault recovery fixes
   - mDP connectors reporting fix
   - audio locking fixes
   - rework engines/instances code to support new scheme

  tegra:
   - VIC newer firmware support
   - display/gr2d fixes for older tegra
   - pm reference leak fix

  mediatek:
   - SOC MT8183 support
   - decouple sub driver + share mtk mutex driver

  radeon:
   - PCI resource fix for some platforms

  ingenic:
   - pm support
   - 8-bit delta RGB panels

  vmwgfx:
   - managed driver helpers

  vc4:
   - BCM2711 DSI1 support
   - converted to atomic helpers
   - enable 10/12 bpc outputs
   - gem prime mmap helpers
   - CEC fix

  omap:
   - use degamma table
   - CTM support
   - rework DSI support

  imx:
   - stack usage fixes
   - drm managed support
   - imx-tve clock provider leak fix
-

  rcar-du:
   - default mode fixes
   - conversion to managed API

  hisilicon:
   - use simple encoder

  vkms:
   - writeback connector support

  d3:
   - BT2020 support"

* tag 'drm-next-2021-02-19' of git://anongit.freedesktop.org/drm/drm: (1459 commits)
  drm/amdgpu: Set reference clock to 100Mhz on Renoir (v2)
  drm/radeon: OLAND boards don't have VCE
  drm/amdkfd: Fix recursive lock warnings
  drm/amd/display: Add FPU wrappers to dcn21_validate_bandwidth()
  drm/amd/display: Fix potential integer overflow
  drm/amdgpu/display: remove hdcp_srm sysfs on device removal
  drm/amdgpu: fix CGTS_TCC_DISABLE register offset on gfx10.3
  drm/i915/gt: Correct surface base address for renderclear
  drm/i915: Disallow plane x+w>stride on ilk+ with X-tiling
  drm/nouveau/top/ga100: initial support
  drm/nouveau/top: add ioctrl/nvjpg
  drm/nouveau/privring: rename from ibus
  drm/nouveau/nvkm: remove nvkm_subdev.index
  drm/nouveau/nvkm: determine subdev id/order from layout
  drm/nouveau/vic: switch to instanced constructor
  drm/nouveau/sw: switch to instanced constructor
  drm/nouveau/sec2: switch to instanced constructor
  drm/nouveau/sec: switch to instanced constructor
  drm/nouveau/pm: switch to instanced constructor
  drm/nouveau/nvenc: switch to instanced constructor
  ...

14 files changed, 608 insertions, 431 deletions

diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c
index 469d1b4f2643..fddaeb0b09c1 100644
--- a/drivers/gpu/drm/vc4/vc4_bo.c
+++ b/drivers/gpu/drm/vc4/vc4_bo.c
@@ -21,7 +21,7 @@
 #include "vc4_drv.h"
 #include "uapi/drm/vc4_drm.h"
 
-static vm_fault_t vc4_fault(struct vm_fault *vmf);
+static const struct drm_gem_object_funcs vc4_gem_object_funcs;
 
 static const char * const bo_type_names[] = {
 	"kernel",
@@ -376,20 +376,6 @@ out:
 	return bo;
 }
 
-static const struct vm_operations_struct vc4_vm_ops = {
-	.fault = vc4_fault,
-	.open = drm_gem_vm_open,
-	.close = drm_gem_vm_close,
-};
-
-static const struct drm_gem_object_funcs vc4_gem_object_funcs = {
-	.free = vc4_free_object,
-	.export = vc4_prime_export,
-	.get_sg_table = drm_gem_cma_prime_get_sg_table,
-	.vmap = vc4_prime_vmap,
-	.vm_ops = &vc4_vm_ops,
-};
-
 /**
  * vc4_create_object - Implementation of driver->gem_create_object.
  * @dev: DRM device
@@ -538,7 +524,7 @@ static void vc4_bo_cache_free_old(struct drm_device *dev)
 /* Called on the last userspace/kernel unreference of the BO.  Returns
  * it to the BO cache if possible, otherwise frees it.
  */
-void vc4_free_object(struct drm_gem_object *gem_bo)
+static void vc4_free_object(struct drm_gem_object *gem_bo)
 {
 	struct drm_device *dev = gem_bo->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
@@ -673,7 +659,7 @@ static void vc4_bo_cache_time_timer(struct timer_list *t)
 	schedule_work(&vc4->bo_cache.time_work);
 }
 
-struct dma_buf * vc4_prime_export(struct drm_gem_object *obj, int flags)
+static struct dma_buf *vc4_prime_export(struct drm_gem_object *obj, int flags)
 {
 	struct vc4_bo *bo = to_vc4_bo(obj);
 	struct dma_buf *dmabuf;
@@ -718,19 +704,9 @@ static vm_fault_t vc4_fault(struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
-int vc4_mmap(struct file *filp, struct vm_area_struct *vma)
+static int vc4_gem_object_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma)
 {
-	struct drm_gem_object *gem_obj;
-	unsigned long vm_pgoff;
-	struct vc4_bo *bo;
-	int ret;
-
-	ret = drm_gem_mmap(filp, vma);
-	if (ret)
-		return ret;
-
-	gem_obj = vma->vm_private_data;
-	bo = to_vc4_bo(gem_obj);
+	struct vc4_bo *bo = to_vc4_bo(obj);
 
 	if (bo->validated_shader && (vma->vm_flags & VM_WRITE)) {
 		DRM_DEBUG("mmaping of shader BOs for writing not allowed.\n");
@@ -744,72 +720,23 @@ int vc4_mmap(struct file *filp, struct vm_area_struct *vma)
 		return -EINVAL;
 	}
 
-	/*
-	 * Clear the VM_PFNMAP flag that was set by drm_gem_mmap(), and set the
-	 * vm_pgoff (used as a fake buffer offset by DRM) to 0 as we want to map
-	 * the whole buffer.
-	 */
-	vma->vm_flags &= ~VM_PFNMAP;
-
-	/* This ->vm_pgoff dance is needed to make all parties happy:
-	 * - dma_mmap_wc() uses ->vm_pgoff as an offset within the allocated
-	 *   mem-region, hence the need to set it to zero (the value set by
-	 *   the DRM core is a virtual offset encoding the GEM object-id)
-	 * - the mmap() core logic needs ->vm_pgoff to be restored to its
-	 *   initial value before returning from this function because it
-	 *   encodes the  offset of this GEM in the dev->anon_inode pseudo-file
-	 *   and this information will be used when we invalidate userspace
-	 *   mappings  with drm_vma_node_unmap() (called from vc4_gem_purge()).
-	 */
-	vm_pgoff = vma->vm_pgoff;
-	vma->vm_pgoff = 0;
-	ret = dma_mmap_wc(bo->base.base.dev->dev, vma, bo->base.vaddr,
-			  bo->base.paddr, vma->vm_end - vma->vm_start);
-	vma->vm_pgoff = vm_pgoff;
-
-	if (ret)
-		drm_gem_vm_close(vma);
-
-	return ret;
-}
-
-int vc4_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma)
-{
-	struct vc4_bo *bo = to_vc4_bo(obj);
-
-	if (bo->validated_shader && (vma->vm_flags & VM_WRITE)) {
-		DRM_DEBUG("mmaping of shader BOs for writing not allowed.\n");
-		return -EINVAL;
-	}
-
-	return drm_gem_cma_prime_mmap(obj, vma);
-}
-
-int vc4_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map)
-{
-	struct vc4_bo *bo = to_vc4_bo(obj);
-
-	if (bo->validated_shader) {
-		DRM_DEBUG("mmaping of shader BOs not allowed.\n");
-		return -EINVAL;
-	}
-
-	return drm_gem_cma_prime_vmap(obj, map);
+	return drm_gem_cma_mmap(obj, vma);
 }
 
-struct drm_gem_object *
-vc4_prime_import_sg_table(struct drm_device *dev,
-			  struct dma_buf_attachment *attach,
-			  struct sg_table *sgt)
-{
-	struct drm_gem_object *obj;
-
-	obj = drm_gem_cma_prime_import_sg_table(dev, attach, sgt);
-	if (IS_ERR(obj))
-		return obj;
+static const struct vm_operations_struct vc4_vm_ops = {
+	.fault = vc4_fault,
+	.open = drm_gem_vm_open,
+	.close = drm_gem_vm_close,
+};
 
-	return obj;
-}
+static const struct drm_gem_object_funcs vc4_gem_object_funcs = {
+	.free = vc4_free_object,
+	.export = vc4_prime_export,
+	.get_sg_table = drm_gem_cma_get_sg_table,
+	.vmap = drm_gem_cma_vmap,
+	.mmap = vc4_gem_object_mmap,
+	.vm_ops = &vc4_vm_ops,
+};
 
 static int vc4_grab_bin_bo(struct vc4_dev *vc4, struct vc4_file *vc4file)
 {
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c
index ea710beb8e00..269390bc586e 100644
--- a/drivers/gpu/drm/vc4/vc4_crtc.c
+++ b/drivers/gpu/drm/vc4/vc4_crtc.c
@@ -403,7 +403,9 @@ static void require_hvs_enabled(struct drm_device *dev)
 		     SCALER_DISPCTRL_ENABLE);
 }
 
-static int vc4_crtc_disable(struct drm_crtc *crtc, unsigned int channel)
+static int vc4_crtc_disable(struct drm_crtc *crtc,
+			    struct drm_atomic_state *state,
+			    unsigned int channel)
 {
 	struct drm_encoder *encoder = vc4_get_crtc_encoder(crtc);
 	struct vc4_encoder *vc4_encoder = to_vc4_encoder(encoder);
@@ -435,13 +437,13 @@ static int vc4_crtc_disable(struct drm_crtc *crtc, unsigned int channel)
 	mdelay(20);
 
 	if (vc4_encoder && vc4_encoder->post_crtc_disable)
-		vc4_encoder->post_crtc_disable(encoder);
+		vc4_encoder->post_crtc_disable(encoder, state);
 
 	vc4_crtc_pixelvalve_reset(crtc);
 	vc4_hvs_stop_channel(dev, channel);
 
 	if (vc4_encoder && vc4_encoder->post_crtc_powerdown)
-		vc4_encoder->post_crtc_powerdown(encoder);
+		vc4_encoder->post_crtc_powerdown(encoder, state);
 
 	return 0;
 }
@@ -468,7 +470,7 @@ int vc4_crtc_disable_at_boot(struct drm_crtc *crtc)
 	if (channel < 0)
 		return 0;
 
-	return vc4_crtc_disable(crtc, channel);
+	return vc4_crtc_disable(crtc, NULL, channel);
 }
 
 static void vc4_crtc_atomic_disable(struct drm_crtc *crtc,
@@ -484,7 +486,7 @@ static void vc4_crtc_atomic_disable(struct drm_crtc *crtc,
 	/* Disable vblank irq handling before crtc is disabled. */
 	drm_crtc_vblank_off(crtc);
 
-	vc4_crtc_disable(crtc, old_vc4_state->assigned_channel);
+	vc4_crtc_disable(crtc, state, old_vc4_state->assigned_channel);
 
 	/*
 	 * Make sure we issue a vblank event after disabling the CRTC if
@@ -503,8 +505,6 @@ static void vc4_crtc_atomic_disable(struct drm_crtc *crtc,
 static void vc4_crtc_atomic_enable(struct drm_crtc *crtc,
 				   struct drm_atomic_state *state)
 {
-	struct drm_crtc_state *old_state = drm_atomic_get_old_crtc_state(state,
-									 crtc);
 	struct drm_device *dev = crtc->dev;
 	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
 	struct drm_encoder *encoder = vc4_get_crtc_encoder(crtc);
@@ -517,17 +517,17 @@ static void vc4_crtc_atomic_enable(struct drm_crtc *crtc,
 	 */
 	drm_crtc_vblank_on(crtc);
 
-	vc4_hvs_atomic_enable(crtc, old_state);
+	vc4_hvs_atomic_enable(crtc, state);
 
 	if (vc4_encoder->pre_crtc_configure)
-		vc4_encoder->pre_crtc_configure(encoder);
+		vc4_encoder->pre_crtc_configure(encoder, state);
 
 	vc4_crtc_config_pv(crtc);
 
 	CRTC_WRITE(PV_CONTROL, CRTC_READ(PV_CONTROL) | PV_CONTROL_EN);
 
 	if (vc4_encoder->pre_crtc_enable)
-		vc4_encoder->pre_crtc_enable(encoder);
+		vc4_encoder->pre_crtc_enable(encoder, state);
 
 	/* When feeding the transposer block the pixelvalve is unneeded and
 	 * should not be enabled.
@@ -536,7 +536,7 @@ static void vc4_crtc_atomic_enable(struct drm_crtc *crtc,
 		   CRTC_READ(PV_V_CONTROL) | PV_VCONTROL_VIDEN);
 
 	if (vc4_encoder->post_crtc_enable)
-		vc4_encoder->post_crtc_enable(encoder);
+		vc4_encoder->post_crtc_enable(encoder, state);
 }
 
 static enum drm_mode_status vc4_crtc_mode_valid(struct drm_crtc *crtc,
@@ -593,7 +593,7 @@ static int vc4_crtc_atomic_check(struct drm_crtc *crtc,
 	struct drm_connector_state *conn_state;
 	int ret, i;
 
-	ret = vc4_hvs_atomic_check(crtc, crtc_state);
+	ret = vc4_hvs_atomic_check(crtc, state);
 	if (ret)
 		return ret;
 
@@ -697,7 +697,6 @@ vc4_async_page_flip_complete(struct vc4_seqno_cb *cb)
 		container_of(cb, struct vc4_async_flip_state, cb);
 	struct drm_crtc *crtc = flip_state->crtc;
 	struct drm_device *dev = crtc->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct drm_plane *plane = crtc->primary;
 
 	vc4_plane_async_set_fb(plane, flip_state->fb);
@@ -729,8 +728,6 @@ vc4_async_page_flip_complete(struct vc4_seqno_cb *cb)
 	}
 
 	kfree(flip_state);
-
-	up(&vc4->async_modeset);
 }
 
 /* Implements async (non-vblank-synced) page flips.
@@ -745,7 +742,6 @@ static int vc4_async_page_flip(struct drm_crtc *crtc,
 			       uint32_t flags)
 {
 	struct drm_device *dev = crtc->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct drm_plane *plane = crtc->primary;
 	int ret = 0;
 	struct vc4_async_flip_state *flip_state;
@@ -774,15 +770,6 @@ static int vc4_async_page_flip(struct drm_crtc *crtc,
 	flip_state->crtc = crtc;
 	flip_state->event = event;
 
-	/* Make sure all other async modesetes have landed. */
-	ret = down_interruptible(&vc4->async_modeset);
-	if (ret) {
-		drm_framebuffer_put(fb);
-		vc4_bo_dec_usecnt(bo);
-		kfree(flip_state);
-		return ret;
-	}
-
 	/* Save the current FB before it's replaced by the new one in
 	 * drm_atomic_set_fb_for_plane(). We'll need the old FB in
 	 * vc4_async_page_flip_complete() to decrement the BO usecnt and keep
@@ -884,7 +871,6 @@ static const struct drm_crtc_funcs vc4_crtc_funcs = {
 	.reset = vc4_crtc_reset,
 	.atomic_duplicate_state = vc4_crtc_duplicate_state,
 	.atomic_destroy_state = vc4_crtc_destroy_state,
-	.gamma_set = drm_atomic_helper_legacy_gamma_set,
 	.enable_vblank = vc4_enable_vblank,
 	.disable_vblank = vc4_disable_vblank,
 	.get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp,
diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c
index 2cd97a39c286..556ad0f02a0d 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.c
+++ b/drivers/gpu/drm/vc4/vc4_drv.c
@@ -140,17 +140,7 @@ static void vc4_close(struct drm_device *dev, struct drm_file *file)
 	kfree(vc4file);
 }
 
-static const struct file_operations vc4_drm_fops = {
-	.owner = THIS_MODULE,
-	.open = drm_open,
-	.release = drm_release,
-	.unlocked_ioctl = drm_ioctl,
-	.mmap = vc4_mmap,
-	.poll = drm_poll,
-	.read = drm_read,
-	.compat_ioctl = drm_compat_ioctl,
-	.llseek = noop_llseek,
-};
+DEFINE_DRM_GEM_FOPS(vc4_drm_fops);
 
 static const struct drm_ioctl_desc vc4_drm_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(VC4_SUBMIT_CL, vc4_submit_cl_ioctl, DRM_RENDER_ALLOW),
@@ -190,12 +180,7 @@ static struct drm_driver vc4_drm_driver = {
 
 	.gem_create_object = vc4_create_object,
 
-	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
-	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
-	.gem_prime_import_sg_table = vc4_prime_import_sg_table,
-	.gem_prime_mmap = vc4_prime_mmap,
-
-	.dumb_create = vc4_dumb_create,
+	DRM_GEM_CMA_DRIVER_OPS_WITH_DUMB_CREATE(vc4_dumb_create),
 
 	.ioctls = vc4_drm_ioctls,
 	.num_ioctls = ARRAY_SIZE(vc4_drm_ioctls),
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h
index 43a1af110b3e..a7500716cf3f 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.h
+++ b/drivers/gpu/drm/vc4/vc4_drv.h
@@ -77,7 +77,6 @@ struct vc4_dev {
 	struct vc4_hvs *hvs;
 	struct vc4_v3d *v3d;
 	struct vc4_dpi *dpi;
-	struct vc4_dsi *dsi1;
 	struct vc4_vec *vec;
 	struct vc4_txp *txp;
 
@@ -215,8 +214,6 @@ struct vc4_dev {
 		struct work_struct reset_work;
 	} hangcheck;
 
-	struct semaphore async_modeset;
-
 	struct drm_modeset_lock ctm_state_lock;
 	struct drm_private_obj ctm_manager;
 	struct drm_private_obj hvs_channels;
@@ -444,12 +441,12 @@ struct vc4_encoder {
 	enum vc4_encoder_type type;
 	u32 clock_select;
 
-	void (*pre_crtc_configure)(struct drm_encoder *encoder);
-	void (*pre_crtc_enable)(struct drm_encoder *encoder);
-	void (*post_crtc_enable)(struct drm_encoder *encoder);
+	void (*pre_crtc_configure)(struct drm_encoder *encoder, struct drm_atomic_state *state);
+	void (*pre_crtc_enable)(struct drm_encoder *encoder, struct drm_atomic_state *state);
+	void (*post_crtc_enable)(struct drm_encoder *encoder, struct drm_atomic_state *state);
 
-	void (*post_crtc_disable)(struct drm_encoder *encoder);
-	void (*post_crtc_powerdown)(struct drm_encoder *encoder);
+	void (*post_crtc_disable)(struct drm_encoder *encoder, struct drm_atomic_state *state);
+	void (*post_crtc_powerdown)(struct drm_encoder *encoder, struct drm_atomic_state *state);
 };
 
 static inline struct vc4_encoder *
@@ -785,13 +782,11 @@ struct vc4_validated_shader_info {
 
 /* vc4_bo.c */
 struct drm_gem_object *vc4_create_object(struct drm_device *dev, size_t size);
-void vc4_free_object(struct drm_gem_object *gem_obj);
 struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t size,
 			     bool from_cache, enum vc4_kernel_bo_type type);
 int vc4_dumb_create(struct drm_file *file_priv,
 		    struct drm_device *dev,
 		    struct drm_mode_create_dumb *args);
-struct dma_buf *vc4_prime_export(struct drm_gem_object *obj, int flags);
 int vc4_create_bo_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
 int vc4_create_shader_bo_ioctl(struct drm_device *dev, void *data,
@@ -806,12 +801,6 @@ int vc4_get_hang_state_ioctl(struct drm_device *dev, void *data,
 			     struct drm_file *file_priv);
 int vc4_label_bo_ioctl(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
-int vc4_mmap(struct file *filp, struct vm_area_struct *vma);
-int vc4_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma);
-struct drm_gem_object *vc4_prime_import_sg_table(struct drm_device *dev,
-						 struct dma_buf_attachment *attach,
-						 struct sg_table *sgt);
-int vc4_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map);
 int vc4_bo_cache_init(struct drm_device *dev);
 int vc4_bo_inc_usecnt(struct vc4_bo *bo);
 void vc4_bo_dec_usecnt(struct vc4_bo *bo);
@@ -916,11 +905,10 @@ void vc4_irq_reset(struct drm_device *dev);
 extern struct platform_driver vc4_hvs_driver;
 void vc4_hvs_stop_channel(struct drm_device *dev, unsigned int output);
 int vc4_hvs_get_fifo_from_output(struct drm_device *dev, unsigned int output);
-int vc4_hvs_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state);
-void vc4_hvs_atomic_enable(struct drm_crtc *crtc, struct drm_crtc_state *old_state);
-void vc4_hvs_atomic_disable(struct drm_crtc *crtc, struct drm_crtc_state *old_state);
-void vc4_hvs_atomic_flush(struct drm_crtc *crtc,
-			  struct drm_atomic_state *state);
+int vc4_hvs_atomic_check(struct drm_crtc *crtc, struct drm_atomic_state *state);
+void vc4_hvs_atomic_enable(struct drm_crtc *crtc, struct drm_atomic_state *state);
+void vc4_hvs_atomic_disable(struct drm_crtc *crtc, struct drm_atomic_state *state);
+void vc4_hvs_atomic_flush(struct drm_crtc *crtc, struct drm_atomic_state *state);
 void vc4_hvs_dump_state(struct drm_device *dev);
 void vc4_hvs_unmask_underrun(struct drm_device *dev, int channel);
 void vc4_hvs_mask_underrun(struct drm_device *dev, int channel);
diff --git a/drivers/gpu/drm/vc4/vc4_dsi.c b/drivers/gpu/drm/vc4/vc4_dsi.c
index 19aab4e7e209..a55256ed0955 100644
--- a/drivers/gpu/drm/vc4/vc4_dsi.c
+++ b/drivers/gpu/drm/vc4/vc4_dsi.c
@@ -306,11 +306,11 @@
 # define DSI0_PHY_AFEC0_RESET			BIT(11)
 # define DSI1_PHY_AFEC0_PD_BG			BIT(11)
 # define DSI0_PHY_AFEC0_PD			BIT(10)
-# define DSI1_PHY_AFEC0_PD_DLANE3		BIT(10)
+# define DSI1_PHY_AFEC0_PD_DLANE1		BIT(10)
 # define DSI0_PHY_AFEC0_PD_BG			BIT(9)
 # define DSI1_PHY_AFEC0_PD_DLANE2		BIT(9)
 # define DSI0_PHY_AFEC0_PD_DLANE1		BIT(8)
-# define DSI1_PHY_AFEC0_PD_DLANE1		BIT(8)
+# define DSI1_PHY_AFEC0_PD_DLANE3		BIT(8)
 # define DSI_PHY_AFEC0_PTATADJ_MASK		VC4_MASK(7, 4)
 # define DSI_PHY_AFEC0_PTATADJ_SHIFT		4
 # define DSI_PHY_AFEC0_CTATADJ_MASK		VC4_MASK(3, 0)
@@ -493,6 +493,18 @@
  */
 #define DSI1_ID			0x8c
 
+struct vc4_dsi_variant {
+	/* Whether we're on bcm2835's DSI0 or DSI1. */
+	unsigned int port;
+
+	bool broken_axi_workaround;
+
+	const char *debugfs_name;
+	const struct debugfs_reg32 *regs;
+	size_t nregs;
+
+};
+
 /* General DSI hardware state. */
 struct vc4_dsi {
 	struct platform_device *pdev;
@@ -509,8 +521,7 @@ struct vc4_dsi {
 	u32 *reg_dma_mem;
 	dma_addr_t reg_paddr;
 
-	/* Whether we're on bcm2835's DSI0 or DSI1. */
-	int port;
+	const struct vc4_dsi_variant *variant;
 
 	/* DSI channel for the panel we're connected to. */
 	u32 channel;
@@ -586,10 +597,10 @@ dsi_dma_workaround_write(struct vc4_dsi *dsi, u32 offset, u32 val)
 #define DSI_READ(offset) readl(dsi->regs + (offset))
 #define DSI_WRITE(offset, val) dsi_dma_workaround_write(dsi, offset, val)
 #define DSI_PORT_READ(offset) \
-	DSI_READ(dsi->port ? DSI1_##offset : DSI0_##offset)
+	DSI_READ(dsi->variant->port ? DSI1_##offset : DSI0_##offset)
 #define DSI_PORT_WRITE(offset, val) \
-	DSI_WRITE(dsi->port ? DSI1_##offset : DSI0_##offset, val)
-#define DSI_PORT_BIT(bit) (dsi->port ? DSI1_##bit : DSI0_##bit)
+	DSI_WRITE(dsi->variant->port ? DSI1_##offset : DSI0_##offset, val)
+#define DSI_PORT_BIT(bit) (dsi->variant->port ? DSI1_##bit : DSI0_##bit)
 
 /* VC4 DSI encoder KMS struct */
 struct vc4_dsi_encoder {
@@ -837,7 +848,7 @@ static void vc4_dsi_encoder_enable(struct drm_encoder *encoder)
 
 	ret = pm_runtime_get_sync(dev);
 	if (ret) {
-		DRM_ERROR("Failed to runtime PM enable on DSI%d\n", dsi->port);
+		DRM_ERROR("Failed to runtime PM enable on DSI%d\n", dsi->variant->port);
 		return;
 	}
 
@@ -871,7 +882,7 @@ static void vc4_dsi_encoder_enable(struct drm_encoder *encoder)
 	DSI_PORT_WRITE(STAT, DSI_PORT_READ(STAT));
 
 	/* Set AFE CTR00/CTR1 to release powerdown of analog. */
-	if (dsi->port == 0) {
+	if (dsi->variant->port == 0) {
 		u32 afec0 = (VC4_SET_FIELD(7, DSI_PHY_AFEC0_PTATADJ) |
 			     VC4_SET_FIELD(7, DSI_PHY_AFEC0_CTATADJ));
 
@@ -1017,7 +1028,7 @@ static void vc4_dsi_encoder_enable(struct drm_encoder *encoder)
 		       DSI_PORT_BIT(PHYC_CLANE_ENABLE) |
 		       ((dsi->mode_flags & MIPI_DSI_CLOCK_NON_CONTINUOUS) ?
 			0 : DSI_PORT_BIT(PHYC_HS_CLK_CONTINUOUS)) |
-		       (dsi->port == 0 ?
+		       (dsi->variant->port == 0 ?
 			VC4_SET_FIELD(lpx - 1, DSI0_PHYC_ESC_CLK_LPDT) :
 			VC4_SET_FIELD(lpx - 1, DSI1_PHYC_ESC_CLK_LPDT)));
 
@@ -1043,13 +1054,13 @@ static void vc4_dsi_encoder_enable(struct drm_encoder *encoder)
 		       DSI_DISP1_ENABLE);
 
 	/* Ungate the block. */
-	if (dsi->port == 0)
+	if (dsi->variant->port == 0)
 		DSI_PORT_WRITE(CTRL, DSI_PORT_READ(CTRL) | DSI0_CTRL_CTRL0);
 	else
 		DSI_PORT_WRITE(CTRL, DSI_PORT_READ(CTRL) | DSI1_CTRL_EN);
 
 	/* Bring AFE out of reset. */
-	if (dsi->port == 0) {
+	if (dsi->variant->port == 0) {
 	} else {
 		DSI_PORT_WRITE(PHY_AFEC0,
 			       DSI_PORT_READ(PHY_AFEC0) &
@@ -1313,8 +1324,32 @@ static const struct drm_encoder_helper_funcs vc4_dsi_encoder_helper_funcs = {
 	.mode_fixup = vc4_dsi_encoder_mode_fixup,
 };
 
+static const struct vc4_dsi_variant bcm2711_dsi1_variant = {
+	.port			= 1,
+	.debugfs_name		= "dsi1_regs",
+	.regs			= dsi1_regs,
+	.nregs			= ARRAY_SIZE(dsi1_regs),
+};
+
+static const struct vc4_dsi_variant bcm2835_dsi0_variant = {
+	.port			= 0,
+	.debugfs_name		= "dsi0_regs",
+	.regs			= dsi0_regs,
+	.nregs			= ARRAY_SIZE(dsi0_regs),
+};
+
+static const struct vc4_dsi_variant bcm2835_dsi1_variant = {
+	.port			= 1,
+	.broken_axi_workaround	= true,
+	.debugfs_name		= "dsi1_regs",
+	.regs			= dsi1_regs,
+	.nregs			= ARRAY_SIZE(dsi1_regs),
+};
+
 static const struct of_device_id vc4_dsi_dt_match[] = {
-	{ .compatible = "brcm,bcm2835-dsi1", (void *)(uintptr_t)1 },
+	{ .compatible = "brcm,bcm2711-dsi1", &bcm2711_dsi1_variant },
+	{ .compatible = "brcm,bcm2835-dsi0", &bcm2835_dsi0_variant },
+	{ .compatible = "brcm,bcm2835-dsi1", &bcm2835_dsi1_variant },
 	{}
 };
 
@@ -1325,7 +1360,7 @@ static void dsi_handle_error(struct vc4_dsi *dsi,
 	if (!(stat & bit))
 		return;
 
-	DRM_ERROR("DSI%d: %s error\n", dsi->port, type);
+	DRM_ERROR("DSI%d: %s error\n", dsi->variant->port, type);
 	*ret = IRQ_HANDLED;
 }
 
@@ -1398,12 +1433,12 @@ vc4_dsi_init_phy_clocks(struct vc4_dsi *dsi)
 	struct device *dev = &dsi->pdev->dev;
 	const char *parent_name = __clk_get_name(dsi->pll_phy_clock);
 	static const struct {
-		const char *dsi0_name, *dsi1_name;
+		const char *name;
 		int div;
 	} phy_clocks[] = {
-		{ "dsi0_byte", "dsi1_byte", 8 },
-		{ "dsi0_ddr2", "dsi1_ddr2", 4 },
-		{ "dsi0_ddr", "dsi1_ddr", 2 },
+		{ "byte", 8 },
+		{ "ddr2", 4 },
+		{ "ddr", 2 },
 	};
 	int i;
 
@@ -1419,8 +1454,12 @@ vc4_dsi_init_phy_clocks(struct vc4_dsi *dsi)
 	for (i = 0; i < ARRAY_SIZE(phy_clocks); i++) {
 		struct clk_fixed_factor *fix = &dsi->phy_clocks[i];
 		struct clk_init_data init;
+		char clk_name[16];
 		int ret;
 
+		snprintf(clk_name, sizeof(clk_name),
+			 "dsi%u_%s", dsi->variant->port, phy_clocks[i].name);
+
 		/* We just use core fixed factor clock ops for the PHY
 		 * clocks.  The clocks are actually gated by the
 		 * PHY_AFEC0_DDRCLK_EN bits, which we should be
@@ -1437,10 +1476,7 @@ vc4_dsi_init_phy_clocks(struct vc4_dsi *dsi)
 		memset(&init, 0, sizeof(init));
 		init.parent_names = &parent_name;
 		init.num_parents = 1;
-		if (dsi->port == 1)
-			init.name = phy_clocks[i].dsi1_name;
-		else
-			init.name = phy_clocks[i].dsi0_name;
+		init.name = clk_name;
 		init.ops = &clk_fixed_factor_ops;
 
 		ret = devm_clk_hw_register(dev, &fix->hw);
@@ -1459,7 +1495,6 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct drm_device *drm = dev_get_drvdata(master);
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
 	struct vc4_dsi *dsi = dev_get_drvdata(dev);
 	struct vc4_dsi_encoder *vc4_dsi_encoder;
 	struct drm_panel *panel;
@@ -1471,7 +1506,7 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 	if (!match)
 		return -ENODEV;
 
-	dsi->port = (uintptr_t)match->data;
+	dsi->variant = match->data;
 
 	vc4_dsi_encoder = devm_kzalloc(dev, sizeof(*vc4_dsi_encoder),
 				       GFP_KERNEL);
@@ -1488,13 +1523,8 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 		return PTR_ERR(dsi->regs);
 
 	dsi->regset.base = dsi->regs;
-	if (dsi->port == 0) {
-		dsi->regset.regs = dsi0_regs;
-		dsi->regset.nregs = ARRAY_SIZE(dsi0_regs);
-	} else {
-		dsi->regset.regs = dsi1_regs;
-		dsi->regset.nregs = ARRAY_SIZE(dsi1_regs);
-	}
+	dsi->regset.regs = dsi->variant->regs;
+	dsi->regset.nregs = dsi->variant->nregs;
 
 	if (DSI_PORT_READ(ID) != DSI_ID_VALUE) {
 		dev_err(dev, "Port returned 0x%08x for ID instead of 0x%08x\n",
@@ -1502,11 +1532,11 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 		return -ENODEV;
 	}
 
-	/* DSI1 has a broken AXI slave that doesn't respond to writes
-	 * from the ARM.  It does handle writes from the DMA engine,
+	/* DSI1 on BCM2835/6/7 has a broken AXI slave that doesn't respond to
+	 * writes from the ARM.  It does handle writes from the DMA engine,
 	 * so set up a channel for talking to it.
 	 */
-	if (dsi->port == 1) {
+	if (dsi->variant->broken_axi_workaround) {
 		dsi->reg_dma_mem = dma_alloc_coherent(dev, 4,
 						      &dsi->reg_dma_paddr,
 						      GFP_KERNEL);
@@ -1612,9 +1642,6 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 	if (ret)
 		return ret;
 
-	if (dsi->port == 1)
-		vc4->dsi1 = dsi;
-
 	drm_simple_encoder_init(drm, dsi->encoder, DRM_MODE_ENCODER_DSI);
 	drm_encoder_helper_add(dsi->encoder, &vc4_dsi_encoder_helper_funcs);
 
@@ -1630,10 +1657,7 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 	 */
 	list_splice_init(&dsi->encoder->bridge_chain, &dsi->bridge_chain);
 
-	if (dsi->port == 0)
-		vc4_debugfs_add_regset32(drm, "dsi0_regs", &dsi->regset);
-	else
-		vc4_debugfs_add_regset32(drm, "dsi1_regs", &dsi->regset);
+	vc4_debugfs_add_regset32(drm, dsi->variant->debugfs_name, &dsi->regset);
 
 	pm_runtime_enable(dev);
 
@@ -1643,8 +1667,6 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 static void vc4_dsi_unbind(struct device *dev, struct device *master,
 			   void *data)
 {
-	struct drm_device *drm = dev_get_drvdata(master);
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
 	struct vc4_dsi *dsi = dev_get_drvdata(dev);
 
 	if (dsi->bridge)
@@ -1656,9 +1678,6 @@ static void vc4_dsi_unbind(struct device *dev, struct device *master,
 	 */
 	list_splice_init(&dsi->bridge_chain, &dsi->encoder->bridge_chain);
 	drm_encoder_cleanup(dsi->encoder);
-
-	if (dsi->port == 1)
-		vc4->dsi1 = NULL;
 }
 
 static const struct component_ops vc4_dsi_ops = {
diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c
index b641252939d8..445d3bab89e0 100644
--- a/drivers/gpu/drm/vc4/vc4_gem.c
+++ b/drivers/gpu/drm/vc4/vc4_gem.c
@@ -1026,7 +1026,6 @@ int vc4_queue_seqno_cb(struct drm_device *dev,
 		       void (*func)(struct vc4_seqno_cb *cb))
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int ret = 0;
 	unsigned long irqflags;
 
 	cb->func = func;
@@ -1041,7 +1040,7 @@ int vc4_queue_seqno_cb(struct drm_device *dev,
 	}
 	spin_unlock_irqrestore(&vc4->job_lock, irqflags);
 
-	return ret;
+	return 0;
 }
 
 /* Scheduled when any job has been completed, this walks the list of
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 98cab0bbe92d..1fda574579af 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -76,12 +76,25 @@
 #define VC5_HDMI_VERTB_VSPO_SHIFT		16
 #define VC5_HDMI_VERTB_VSPO_MASK		VC4_MASK(29, 16)
 
+#define VC5_HDMI_DEEP_COLOR_CONFIG_1_INIT_PACK_PHASE_SHIFT	8
+#define VC5_HDMI_DEEP_COLOR_CONFIG_1_INIT_PACK_PHASE_MASK	VC4_MASK(10, 8)
+
+#define VC5_HDMI_DEEP_COLOR_CONFIG_1_COLOR_DEPTH_SHIFT		0
+#define VC5_HDMI_DEEP_COLOR_CONFIG_1_COLOR_DEPTH_MASK		VC4_MASK(3, 0)
+
+#define VC5_HDMI_GCP_CONFIG_GCP_ENABLE		BIT(31)
+
+#define VC5_HDMI_GCP_WORD_1_GCP_SUBPACKET_BYTE_1_SHIFT	8
+#define VC5_HDMI_GCP_WORD_1_GCP_SUBPACKET_BYTE_1_MASK	VC4_MASK(15, 8)
+
 # define VC4_HD_M_SW_RST			BIT(2)
 # define VC4_HD_M_ENABLE			BIT(0)
 
 #define CEC_CLOCK_FREQ 40000
 #define VC4_HSM_MID_CLOCK 149985000
 
+#define HDMI_14_MAX_TMDS_CLK   (340 * 1000 * 1000)
+
 static int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
@@ -119,24 +132,57 @@ static void vc5_hdmi_reset(struct vc4_hdmi *vc4_hdmi)
 		   HDMI_READ(HDMI_CLOCK_STOP) | VC4_DVP_HT_CLOCK_STOP_PIXEL);
 }
 
+#ifdef CONFIG_DRM_VC4_HDMI_CEC
+static void vc4_hdmi_cec_update_clk_div(struct vc4_hdmi *vc4_hdmi)
+{
+	u16 clk_cnt;
+	u32 value;
+
+	value = HDMI_READ(HDMI_CEC_CNTRL_1);
+	value &= ~VC4_HDMI_CEC_DIV_CLK_CNT_MASK;
+
+	/*
+	 * Set the clock divider: the hsm_clock rate and this divider
+	 * setting will give a 40 kHz CEC clock.
+	 */
+	clk_cnt = clk_get_rate(vc4_hdmi->cec_clock) / CEC_CLOCK_FREQ;
+	value |= clk_cnt << VC4_HDMI_CEC_DIV_CLK_CNT_SHIFT;
+	HDMI_WRITE(HDMI_CEC_CNTRL_1, value);
+}
+#else
+static void vc4_hdmi_cec_update_clk_div(struct vc4_hdmi *vc4_hdmi) {}
+#endif
+
 static enum drm_connector_status
 vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
 {
 	struct vc4_hdmi *vc4_hdmi = connector_to_vc4_hdmi(connector);
+	bool connected = false;
 
 	if (vc4_hdmi->hpd_gpio) {
 		if (gpio_get_value_cansleep(vc4_hdmi->hpd_gpio) ^
 		    vc4_hdmi->hpd_active_low)
-			return connector_status_connected;
-		cec_phys_addr_invalidate(vc4_hdmi->cec_adap);
-		return connector_status_disconnected;
+			connected = true;
+	} else if (drm_probe_ddc(vc4_hdmi->ddc)) {
+		connected = true;
+	} else if (HDMI_READ(HDMI_HOTPLUG) & VC4_HDMI_HOTPLUG_CONNECTED) {
+		connected = true;
 	}
 
-	if (drm_probe_ddc(vc4_hdmi->ddc))
-		return connector_status_connected;
+	if (connected) {
+		if (connector->status != connector_status_connected) {
+			struct edid *edid = drm_get_edid(connector, vc4_hdmi->ddc);
+
+			if (edid) {
+				cec_s_phys_addr_from_edid(vc4_hdmi->cec_adap, edid);
+				vc4_hdmi->encoder.hdmi_monitor = drm_detect_hdmi_monitor(edid);
+				kfree(edid);
+			}
+		}
 
-	if (HDMI_READ(HDMI_HOTPLUG) & VC4_HDMI_HOTPLUG_CONNECTED)
 		return connector_status_connected;
+	}
+
 	cec_phys_addr_invalidate(vc4_hdmi->cec_adap);
 	return connector_status_disconnected;
 }
@@ -170,16 +216,48 @@ static int vc4_hdmi_connector_get_modes(struct drm_connector *connector)
 
 static void vc4_hdmi_connector_reset(struct drm_connector *connector)
 {
-	drm_atomic_helper_connector_reset(connector);
+	struct vc4_hdmi_connector_state *old_state =
+		conn_state_to_vc4_hdmi_conn_state(connector->state);
+	struct vc4_hdmi_connector_state *new_state =
+		kzalloc(sizeof(*new_state), GFP_KERNEL);
+
+	if (connector->state)
+		__drm_atomic_helper_connector_destroy_state(connector->state);
+
+	kfree(old_state);
+	__drm_atomic_helper_connector_reset(connector, &new_state->base);
+
+	if (!new_state)
+		return;
+
+	new_state->base.max_bpc = 8;
+	new_state->base.max_requested_bpc = 8;
 	drm_atomic_helper_connector_tv_reset(connector);
 }
 
+static struct drm_connector_state *
+vc4_hdmi_connector_duplicate_state(struct drm_connector *connector)
+{
+	struct drm_connector_state *conn_state = connector->state;
+	struct vc4_hdmi_connector_state *vc4_state = conn_state_to_vc4_hdmi_conn_state(conn_state);
+	struct vc4_hdmi_connector_state *new_state;
+
+	new_state = kzalloc(sizeof(*new_state), GFP_KERNEL);
+	if (!new_state)
+		return NULL;
+
+	new_state->pixel_rate = vc4_state->pixel_rate;
+	__drm_atomic_helper_connector_duplicate_state(connector, &new_state->base);
+
+	return &new_state->base;
+}
+
 static const struct drm_connector_funcs vc4_hdmi_connector_funcs = {
 	.detect = vc4_hdmi_connector_detect,
 	.fill_modes = drm_helper_probe_single_connector_modes,
 	.destroy = vc4_hdmi_connector_destroy,
 	.reset = vc4_hdmi_connector_reset,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_duplicate_state = vc4_hdmi_connector_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
 };
 
@@ -200,12 +278,20 @@ static int vc4_hdmi_connector_init(struct drm_device *dev,
 				    vc4_hdmi->ddc);
 	drm_connector_helper_add(connector, &vc4_hdmi_connector_helper_funcs);
 
+	/*
+	 * Some of the properties below require access to state, like bpc.
+	 * Allocate some default initial connector state with our reset helper.
+	 */
+	if (connector->funcs->reset)
+		connector->funcs->reset(connector);
+
 	/* Create and attach TV margin props to this connector. */
 	ret = drm_mode_create_tv_margin_properties(dev);
 	if (ret)
 		return ret;
 
 	drm_connector_attach_tv_margin_properties(connector);
+	drm_connector_attach_max_bpc_property(connector, 8, 12);
 
 	connector->polled = (DRM_CONNECTOR_POLL_CONNECT |
 			     DRM_CONNECTOR_POLL_DISCONNECT);
@@ -219,7 +305,8 @@ static int vc4_hdmi_connector_init(struct drm_device *dev,
 }
 
 static int vc4_hdmi_stop_packet(struct drm_encoder *encoder,
-				enum hdmi_infoframe_type type)
+				enum hdmi_infoframe_type type,
+				bool poll)
 {
 	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 	u32 packet_id = type - 0x80;
@@ -227,6 +314,9 @@ static int vc4_hdmi_stop_packet(struct drm_encoder *encoder,
 	HDMI_WRITE(HDMI_RAM_PACKET_CONFIG,
 		   HDMI_READ(HDMI_RAM_PACKET_CONFIG) & ~BIT(packet_id));
 
+	if (!poll)
+		return 0;
+
 	return wait_for(!(HDMI_READ(HDMI_RAM_PACKET_STATUS) &
 			  BIT(packet_id)), 100);
 }
@@ -253,7 +343,7 @@ static void vc4_hdmi_write_infoframe(struct drm_encoder *encoder,
 	if (len < 0)
 		return;
 
-	ret = vc4_hdmi_stop_packet(encoder, frame->any.type);
+	ret = vc4_hdmi_stop_packet(encoder, frame->any.type, true);
 	if (ret) {
 		DRM_ERROR("Failed to wait for infoframe to go idle: %d\n", ret);
 		return;
@@ -356,7 +446,8 @@ static void vc4_hdmi_set_infoframes(struct drm_encoder *encoder)
 		vc4_hdmi_set_audio_infoframe(encoder);
 }
 
-static void vc4_hdmi_encoder_post_crtc_disable(struct drm_encoder *encoder)
+static void vc4_hdmi_encoder_post_crtc_disable(struct drm_encoder *encoder,
+					       struct drm_atomic_state *state)
 {
 	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 
@@ -369,7 +460,8 @@ static void vc4_hdmi_encoder_post_crtc_disable(struct drm_encoder *encoder)
 		   HDMI_READ(HDMI_VID_CTL) | VC4_HD_VID_CTL_BLANKPIX);
 }
 
-static void vc4_hdmi_encoder_post_crtc_powerdown(struct drm_encoder *encoder)
+static void vc4_hdmi_encoder_post_crtc_powerdown(struct drm_encoder *encoder,
+						 struct drm_atomic_state *state)
 {
 	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 	int ret;
@@ -468,6 +560,7 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi, bool enable)
 }
 
 static void vc4_hdmi_set_timings(struct vc4_hdmi *vc4_hdmi,
+				 struct drm_connector_state *state,
 				 struct drm_display_mode *mode)
 {
 	bool hsync_pos = mode->flags & DRM_MODE_FLAG_PHSYNC;
@@ -511,7 +604,9 @@ static void vc4_hdmi_set_timings(struct vc4_hdmi *vc4_hdmi,
 	HDMI_WRITE(HDMI_VERTB0, vertb_even);
 	HDMI_WRITE(HDMI_VERTB1, vertb);
 }
+
 static void vc5_hdmi_set_timings(struct vc4_hdmi *vc4_hdmi,
+				 struct drm_connector_state *state,
 				 struct drm_display_mode *mode)
 {
 	bool hsync_pos = mode->flags & DRM_MODE_FLAG_PHSYNC;
@@ -531,6 +626,9 @@ static void vc5_hdmi_set_timings(struct vc4_hdmi *vc4_hdmi,
 					mode->crtc_vsync_end -
 					interlaced,
 					VC4_HDMI_VERTB_VBP));
+	unsigned char gcp;
+	bool gcp_en;
+	u32 reg;
 
 	HDMI_WRITE(HDMI_VEC_INTERFACE_XBAR, 0x354021);
 	HDMI_WRITE(HDMI_HORZA,
@@ -556,6 +654,39 @@ static void vc5_hdmi_set_timings(struct vc4_hdmi *vc4_hdmi,
 	HDMI_WRITE(HDMI_VERTB0, vertb_even);
 	HDMI_WRITE(HDMI_VERTB1, vertb);
 
+	switch (state->max_bpc) {
+	case 12:
+		gcp = 6;
+		gcp_en = true;
+		break;
+	case 10:
+		gcp = 5;
+		gcp_en = true;
+		break;
+	case 8:
+	default:
+		gcp = 4;
+		gcp_en = false;
+		break;
+	}
+
+	reg = HDMI_READ(HDMI_DEEP_COLOR_CONFIG_1);
+	reg &= ~(VC5_HDMI_DEEP_COLOR_CONFIG_1_INIT_PACK_PHASE_MASK |
+		 VC5_HDMI_DEEP_COLOR_CONFIG_1_COLOR_DEPTH_MASK);
+	reg |= VC4_SET_FIELD(2, VC5_HDMI_DEEP_COLOR_CONFIG_1_INIT_PACK_PHASE) |
+	       VC4_SET_FIELD(gcp, VC5_HDMI_DEEP_COLOR_CONFIG_1_COLOR_DEPTH);
+	HDMI_WRITE(HDMI_DEEP_COLOR_CONFIG_1, reg);
+
+	reg = HDMI_READ(HDMI_GCP_WORD_1);
+	reg &= ~VC5_HDMI_GCP_WORD_1_GCP_SUBPACKET_BYTE_1_MASK;
+	reg |= VC4_SET_FIELD(gcp, VC5_HDMI_GCP_WORD_1_GCP_SUBPACKET_BYTE_1);
+	HDMI_WRITE(HDMI_GCP_WORD_1, reg);
+
+	reg = HDMI_READ(HDMI_GCP_CONFIG);
+	reg &= ~VC5_HDMI_GCP_CONFIG_GCP_ENABLE;
+	reg |= gcp_en ? VC5_HDMI_GCP_CONFIG_GCP_ENABLE : 0;
+	HDMI_WRITE(HDMI_GCP_CONFIG, reg);
+
 	HDMI_WRITE(HDMI_CLOCK_STOP, 0);
 }
 
@@ -583,8 +714,29 @@ static void vc4_hdmi_recenter_fifo(struct vc4_hdmi *vc4_hdmi)
 		  "VC4_HDMI_FIFO_CTL_RECENTER_DONE");
 }
 
-static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder)
+static struct drm_connector_state *
+vc4_hdmi_encoder_get_connector_state(struct drm_encoder *encoder,
+				     struct drm_atomic_state *state)
 {
+	struct drm_connector_state *conn_state;
+	struct drm_connector *connector;
+	unsigned int i;
+
+	for_each_new_connector_in_state(state, connector, conn_state, i) {
+		if (conn_state->best_encoder == encoder)
+			return conn_state;
+	}
+
+	return NULL;
+}
+
+static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder,
+						struct drm_atomic_state *state)
+{
+	struct drm_connector_state *conn_state =
+		vc4_hdmi_encoder_get_connector_state(encoder, state);
+	struct vc4_hdmi_connector_state *vc4_conn_state =
+		conn_state_to_vc4_hdmi_conn_state(conn_state);
 	struct drm_display_mode *mode = &encoder->crtc->state->adjusted_mode;
 	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 	unsigned long pixel_rate, hsm_rate;
@@ -596,7 +748,7 @@ static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder)
 		return;
 	}
 
-	pixel_rate = mode->clock * 1000 * ((mode->flags & DRM_MODE_FLAG_DBLCLK) ? 2 : 1);
+	pixel_rate = vc4_conn_state->pixel_rate;
 	ret = clk_set_rate(vc4_hdmi->pixel_clock, pixel_rate);
 	if (ret) {
 		DRM_ERROR("Failed to set pixel clock rate: %d\n", ret);
@@ -639,6 +791,8 @@ static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder)
 		return;
 	}
 
+	vc4_hdmi_cec_update_clk_div(vc4_hdmi);
+
 	/*
 	 * FIXME: When the pixel freq is 594MHz (4k60), this needs to be setup
 	 * at 300MHz.
@@ -660,11 +814,8 @@ static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder)
 		return;
 	}
 
-	if (vc4_hdmi->variant->reset)
-		vc4_hdmi->variant->reset(vc4_hdmi);
-
 	if (vc4_hdmi->variant->phy_init)
-		vc4_hdmi->variant->phy_init(vc4_hdmi, mode);
+		vc4_hdmi->variant->phy_init(vc4_hdmi, vc4_conn_state);
 
 	HDMI_WRITE(HDMI_SCHEDULER_CONTROL,
 		   HDMI_READ(HDMI_SCHEDULER_CONTROL) |
@@ -672,10 +823,11 @@ static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder)
 		   VC4_HDMI_SCHEDULER_CONTROL_IGNORE_VSYNC_PREDICTS);
 
 	if (vc4_hdmi->variant->set_timings)
-		vc4_hdmi->variant->set_timings(vc4_hdmi, mode);
+		vc4_hdmi->variant->set_timings(vc4_hdmi, conn_state, mode);
 }
 
-static void vc4_hdmi_encoder_pre_crtc_enable(struct drm_encoder *encoder)
+static void vc4_hdmi_encoder_pre_crtc_enable(struct drm_encoder *encoder,
+					     struct drm_atomic_state *state)
 {
 	struct drm_display_mode *mode = &encoder->crtc->state->adjusted_mode;
 	struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder);
@@ -697,7 +849,8 @@ static void vc4_hdmi_encoder_pre_crtc_enable(struct drm_encoder *encoder)
 	HDMI_WRITE(HDMI_FIFO_CTL, VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N);
 }
 
-static void vc4_hdmi_encoder_post_crtc_enable(struct drm_encoder *encoder)
+static void vc4_hdmi_encoder_post_crtc_enable(struct drm_encoder *encoder,
+					      struct drm_atomic_state *state)
 {
 	struct drm_display_mode *mode = &encoder->crtc->state->adjusted_mode;
 	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
@@ -766,6 +919,7 @@ static int vc4_hdmi_encoder_atomic_check(struct drm_encoder *encoder,
 					 struct drm_crtc_state *crtc_state,
 					 struct drm_connector_state *conn_state)
 {
+	struct vc4_hdmi_connector_state *vc4_state = conn_state_to_vc4_hdmi_conn_state(conn_state);
 	struct drm_display_mode *mode = &crtc_state->adjusted_mode;
 	struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder);
 	unsigned long long pixel_rate = mode->clock * 1000;
@@ -790,9 +944,22 @@ static int vc4_hdmi_encoder_atomic_check(struct drm_encoder *encoder,
 		pixel_rate = mode->clock * 1000;
 	}
 
+	if (conn_state->max_bpc == 12) {
+		pixel_rate = pixel_rate * 150;
+		do_div(pixel_rate, 100);
+	} else if (conn_state->max_bpc == 10) {
+		pixel_rate = pixel_rate * 125;
+		do_div(pixel_rate, 100);
+	}
+
+	if (mode->flags & DRM_MODE_FLAG_DBLCLK)
+		pixel_rate = pixel_rate * 2;
+
 	if (pixel_rate > vc4_hdmi->variant->max_pixel_clock)
 		return -EINVAL;
 
+	vc4_state->pixel_rate = pixel_rate;
+
 	return 0;
 }
 
@@ -936,7 +1103,7 @@ static void vc4_hdmi_audio_reset(struct vc4_hdmi *vc4_hdmi)
 	int ret;
 
 	vc4_hdmi->audio.streaming = false;
-	ret = vc4_hdmi_stop_packet(encoder, HDMI_INFOFRAME_TYPE_AUDIO);
+	ret = vc4_hdmi_stop_packet(encoder, HDMI_INFOFRAME_TYPE_AUDIO, false);
 	if (ret)
 		dev_err(dev, "Failed to stop audio infoframe: %d\n", ret);
 
@@ -1288,15 +1455,22 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *vc4_hdmi)
 }
 
 #ifdef CONFIG_DRM_VC4_HDMI_CEC
-static irqreturn_t vc4_cec_irq_handler_thread(int irq, void *priv)
+static irqreturn_t vc4_cec_irq_handler_rx_thread(int irq, void *priv)
+{
+	struct vc4_hdmi *vc4_hdmi = priv;
+
+	if (vc4_hdmi->cec_rx_msg.len)
+		cec_received_msg(vc4_hdmi->cec_adap,
+				 &vc4_hdmi->cec_rx_msg);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t vc4_cec_irq_handler_tx_thread(int irq, void *priv)
 {
 	struct vc4_hdmi *vc4_hdmi = priv;
 
-	if (vc4_hdmi->cec_irq_was_rx) {
-		if (vc4_hdmi->cec_rx_msg.len)
-			cec_received_msg(vc4_hdmi->cec_adap,
-					 &vc4_hdmi->cec_rx_msg);
-	} else if (vc4_hdmi->cec_tx_ok) {
+	if (vc4_hdmi->cec_tx_ok) {
 		cec_transmit_done(vc4_hdmi->cec_adap, CEC_TX_STATUS_OK,
 				  0, 0, 0, 0);
 	} else {
@@ -1310,15 +1484,35 @@ static irqreturn_t vc4_cec_irq_handler_thread(int irq, void *priv)
 	return IRQ_HANDLED;
 }
 
+static irqreturn_t vc4_cec_irq_handler_thread(int irq, void *priv)
+{
+	struct vc4_hdmi *vc4_hdmi = priv;
+	irqreturn_t ret;
+
+	if (vc4_hdmi->cec_irq_was_rx)
+		ret = vc4_cec_irq_handler_rx_thread(irq, priv);
+	else
+		ret = vc4_cec_irq_handler_tx_thread(irq, priv);
+
+	return ret;
+}
+
 static void vc4_cec_read_msg(struct vc4_hdmi *vc4_hdmi, u32 cntrl1)
 {
+	struct drm_device *dev = vc4_hdmi->connector.dev;
 	struct cec_msg *msg = &vc4_hdmi->cec_rx_msg;
 	unsigned int i;
 
 	msg->len = 1 + ((cntrl1 & VC4_HDMI_CEC_REC_WRD_CNT_MASK) >>
 					VC4_HDMI_CEC_REC_WRD_CNT_SHIFT);
+
+	if (msg->len > 16) {
+		drm_err(dev, "Attempting to read too much data (%d)\n", msg->len);
+		return;
+	}
+
 	for (i = 0; i < msg->len; i += 4) {
-		u32 val = HDMI_READ(HDMI_CEC_RX_DATA_1 + i);
+		u32 val = HDMI_READ(HDMI_CEC_RX_DATA_1 + (i >> 2));
 
 		msg->msg[i] = val & 0xff;
 		msg->msg[i + 1] = (val >> 8) & 0xff;
@@ -1327,31 +1521,55 @@ static void vc4_cec_read_msg(struct vc4_hdmi *vc4_hdmi, u32 cntrl1)
 	}
 }
 
+static irqreturn_t vc4_cec_irq_handler_tx_bare(int irq, void *priv)
+{
+	struct vc4_hdmi *vc4_hdmi = priv;
+	u32 cntrl1;
+
+	cntrl1 = HDMI_READ(HDMI_CEC_CNTRL_1);
+	vc4_hdmi->cec_tx_ok = cntrl1 & VC4_HDMI_CEC_TX_STATUS_GOOD;
+	cntrl1 &= ~VC4_HDMI_CEC_START_XMIT_BEGIN;
+	HDMI_WRITE(HDMI_CEC_CNTRL_1, cntrl1);
+
+	return IRQ_WAKE_THREAD;
+}
+
+static irqreturn_t vc4_cec_irq_handler_rx_bare(int irq, void *priv)
+{
+	struct vc4_hdmi *vc4_hdmi = priv;
+	u32 cntrl1;
+
+	vc4_hdmi->cec_rx_msg.len = 0;
+	cntrl1 = HDMI_READ(HDMI_CEC_CNTRL_1);
+	vc4_cec_read_msg(vc4_hdmi, cntrl1);
+	cntrl1 |= VC4_HDMI_CEC_CLEAR_RECEIVE_OFF;
+	HDMI_WRITE(HDMI_CEC_CNTRL_1, cntrl1);
+	cntrl1 &= ~VC4_HDMI_CEC_CLEAR_RECEIVE_OFF;
+
+	HDMI_WRITE(HDMI_CEC_CNTRL_1, cntrl1);
+
+	return IRQ_WAKE_THREAD;
+}
+
 static irqreturn_t vc4_cec_irq_handler(int irq, void *priv)
 {
 	struct vc4_hdmi *vc4_hdmi = priv;
 	u32 stat = HDMI_READ(HDMI_CEC_CPU_STATUS);
-	u32 cntrl1, cntrl5;
+	irqreturn_t ret;
+	u32 cntrl5;
 
 	if (!(stat & VC4_HDMI_CPU_CEC))
 		return IRQ_NONE;
-	vc4_hdmi->cec_rx_msg.len = 0;
-	cntrl1 = HDMI_READ(HDMI_CEC_CNTRL_1);
+
 	cntrl5 = HDMI_READ(HDMI_CEC_CNTRL_5);
 	vc4_hdmi->cec_irq_was_rx = cntrl5 & VC4_HDMI_CEC_RX_CEC_INT;
-	if (vc4_hdmi->cec_irq_was_rx) {
-		vc4_cec_read_msg(vc4_hdmi, cntrl1);
-		cntrl1 |= VC4_HDMI_CEC_CLEAR_RECEIVE_OFF;
-		HDMI_WRITE(HDMI_CEC_CNTRL_1, cntrl1);
-		cntrl1 &= ~VC4_HDMI_CEC_CLEAR_RECEIVE_OFF;
-	} else {
-		vc4_hdmi->cec_tx_ok = cntrl1 & VC4_HDMI_CEC_TX_STATUS_GOOD;
-		cntrl1 &= ~VC4_HDMI_CEC_START_XMIT_BEGIN;
-	}
-	HDMI_WRITE(HDMI_CEC_CNTRL_1, cntrl1);
-	HDMI_WRITE(HDMI_CEC_CPU_CLEAR, VC4_HDMI_CPU_CEC);
+	if (vc4_hdmi->cec_irq_was_rx)
+		ret = vc4_cec_irq_handler_rx_bare(irq, priv);
+	else
+		ret = vc4_cec_irq_handler_tx_bare(irq, priv);
 
-	return IRQ_WAKE_THREAD;
+	HDMI_WRITE(HDMI_CEC_CPU_CLEAR, VC4_HDMI_CPU_CEC);
+	return ret;
 }
 
 static int vc4_hdmi_cec_adap_enable(struct cec_adapter *adap, bool enable)
@@ -1388,9 +1606,11 @@ static int vc4_hdmi_cec_adap_enable(struct cec_adapter *adap, bool enable)
 			   ((3600 / usecs) << VC4_HDMI_CEC_CNT_TO_3600_US_SHIFT) |
 			   ((3500 / usecs) << VC4_HDMI_CEC_CNT_TO_3500_US_SHIFT));
 
-		HDMI_WRITE(HDMI_CEC_CPU_MASK_CLEAR, VC4_HDMI_CPU_CEC);
+		if (!vc4_hdmi->variant->external_irq_controller)
+			HDMI_WRITE(HDMI_CEC_CPU_MASK_CLEAR, VC4_HDMI_CPU_CEC);
 	} else {
-		HDMI_WRITE(HDMI_CEC_CPU_MASK_SET, VC4_HDMI_CPU_CEC);
+		if (!vc4_hdmi->variant->external_irq_controller)
+			HDMI_WRITE(HDMI_CEC_CPU_MASK_SET, VC4_HDMI_CPU_CEC);
 		HDMI_WRITE(HDMI_CEC_CNTRL_5, val |
 			   VC4_HDMI_CEC_TX_SW_RESET | VC4_HDMI_CEC_RX_SW_RESET);
 	}
@@ -1411,11 +1631,17 @@ static int vc4_hdmi_cec_adap_transmit(struct cec_adapter *adap, u8 attempts,
 				      u32 signal_free_time, struct cec_msg *msg)
 {
 	struct vc4_hdmi *vc4_hdmi = cec_get_drvdata(adap);
+	struct drm_device *dev = vc4_hdmi->connector.dev;
 	u32 val;
 	unsigned int i;
 
+	if (msg->len > 16) {
+		drm_err(dev, "Attempting to transmit too much data (%d)\n", msg->len);
+		return -ENOMEM;
+	}
+
 	for (i = 0; i < msg->len; i += 4)
-		HDMI_WRITE(HDMI_CEC_TX_DATA_1 + i,
+		HDMI_WRITE(HDMI_CEC_TX_DATA_1 + (i >> 2),
 			   (msg->msg[i]) |
 			   (msg->msg[i + 1] << 8) |
 			   (msg->msg[i + 2] << 16) |
@@ -1442,11 +1668,14 @@ static int vc4_hdmi_cec_init(struct vc4_hdmi *vc4_hdmi)
 {
 	struct cec_connector_info conn_info;
 	struct platform_device *pdev = vc4_hdmi->pdev;
+	struct device *dev = &pdev->dev;
 	u32 value;
 	int ret;
 
-	if (!vc4_hdmi->variant->cec_available)
+	if (!of_find_property(dev->of_node, "interrupts", NULL)) {
+		dev_warn(dev, "'interrupts' DT property is missing, no CEC\n");
 		return 0;
+	}
 
 	vc4_hdmi->cec_adap = cec_allocate_adapter(&vc4_hdmi_cec_adap_ops,
 						  vc4_hdmi, "vc4",
@@ -1459,23 +1688,39 @@ static int vc4_hdmi_cec_init(struct vc4_hdmi *vc4_hdmi)
 	cec_fill_conn_info_from_drm(&conn_info, &vc4_hdmi->connector);
 	cec_s_conn_info(vc4_hdmi->cec_adap, &conn_info);
 
-	HDMI_WRITE(HDMI_CEC_CPU_MASK_SET, 0xffffffff);
 	value = HDMI_READ(HDMI_CEC_CNTRL_1);
-	value &= ~VC4_HDMI_CEC_DIV_CLK_CNT_MASK;
-	/*
-	 * Set the logical address to Unregistered and set the clock
-	 * divider: the hsm_clock rate and this divider setting will
-	 * give a 40 kHz CEC clock.
-	 */
-	value |= VC4_HDMI_CEC_ADDR_MASK |
-		 (4091 << VC4_HDMI_CEC_DIV_CLK_CNT_SHIFT);
+	/* Set the logical address to Unregistered */
+	value |= VC4_HDMI_CEC_ADDR_MASK;
 	HDMI_WRITE(HDMI_CEC_CNTRL_1, value);
-	ret = devm_request_threaded_irq(&pdev->dev, platform_get_irq(pdev, 0),
-					vc4_cec_irq_handler,
-					vc4_cec_irq_handler_thread, 0,
-					"vc4 hdmi cec", vc4_hdmi);
-	if (ret)
-		goto err_delete_cec_adap;
+
+	vc4_hdmi_cec_update_clk_div(vc4_hdmi);
+
+	if (vc4_hdmi->variant->external_irq_controller) {
+		ret = devm_request_threaded_irq(&pdev->dev,
+						platform_get_irq_byname(pdev, "cec-rx"),
+						vc4_cec_irq_handler_rx_bare,
+						vc4_cec_irq_handler_rx_thread, 0,
+						"vc4 hdmi cec rx", vc4_hdmi);
+		if (ret)
+			goto err_delete_cec_adap;
+
+		ret = devm_request_threaded_irq(&pdev->dev,
+						platform_get_irq_byname(pdev, "cec-tx"),
+						vc4_cec_irq_handler_tx_bare,
+						vc4_cec_irq_handler_tx_thread, 0,
+						"vc4 hdmi cec tx", vc4_hdmi);
+		if (ret)
+			goto err_delete_cec_adap;
+	} else {
+		HDMI_WRITE(HDMI_CEC_CPU_MASK_SET, 0xffffffff);
+
+		ret = devm_request_threaded_irq(&pdev->dev, platform_get_irq(pdev, 0),
+						vc4_cec_irq_handler,
+						vc4_cec_irq_handler_thread, 0,
+						"vc4 hdmi cec", vc4_hdmi);
+		if (ret)
+			goto err_delete_cec_adap;
+	}
 
 	ret = cec_register_adapter(vc4_hdmi->cec_adap, &pdev->dev);
 	if (ret < 0)
@@ -1575,6 +1820,7 @@ static int vc4_hdmi_init_resources(struct vc4_hdmi *vc4_hdmi)
 		return PTR_ERR(vc4_hdmi->hsm_clock);
 	}
 	vc4_hdmi->audio_clock = vc4_hdmi->hsm_clock;
+	vc4_hdmi->cec_clock = vc4_hdmi->hsm_clock;
 
 	return 0;
 }
@@ -1668,6 +1914,12 @@ static int vc5_hdmi_init_resources(struct vc4_hdmi *vc4_hdmi)
 		return PTR_ERR(vc4_hdmi->audio_clock);
 	}
 
+	vc4_hdmi->cec_clock = devm_clk_get(dev, "cec");
+	if (IS_ERR(vc4_hdmi->cec_clock)) {
+		DRM_ERROR("Failed to get CEC clock\n");
+		return PTR_ERR(vc4_hdmi->cec_clock);
+	}
+
 	vc4_hdmi->reset = devm_reset_control_get(dev, NULL);
 	if (IS_ERR(vc4_hdmi->reset)) {
 		DRM_ERROR("Failed to get HDMI reset line\n");
@@ -1740,6 +1992,9 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
 	vc4_hdmi->disable_wifi_frequencies =
 		of_property_read_bool(dev->of_node, "wifi-2.4ghz-coexistence");
 
+	if (vc4_hdmi->variant->reset)
+		vc4_hdmi->variant->reset(vc4_hdmi);
+
 	pm_runtime_enable(dev);
 
 	drm_simple_encoder_init(drm, encoder, DRM_MODE_ENCODER_TMDS);
@@ -1835,7 +2090,6 @@ static const struct vc4_hdmi_variant bcm2835_variant = {
 	.debugfs_name		= "hdmi_regs",
 	.card_name		= "vc4-hdmi",
 	.max_pixel_clock	= 162000000,
-	.cec_available		= true,
 	.registers		= vc4_hdmi_fields,
 	.num_registers		= ARRAY_SIZE(vc4_hdmi_fields),
 
@@ -1854,7 +2108,7 @@ static const struct vc4_hdmi_variant bcm2711_hdmi0_variant = {
 	.encoder_type		= VC4_ENCODER_TYPE_HDMI0,
 	.debugfs_name		= "hdmi0_regs",
 	.card_name		= "vc4-hdmi-0",
-	.max_pixel_clock	= 297000000,
+	.max_pixel_clock	= HDMI_14_MAX_TMDS_CLK,
 	.registers		= vc5_hdmi_hdmi0_fields,
 	.num_registers		= ARRAY_SIZE(vc5_hdmi_hdmi0_fields),
 	.phy_lane_mapping	= {
@@ -1864,6 +2118,7 @@ static const struct vc4_hdmi_variant bcm2711_hdmi0_variant = {
 		PHY_LANE_CK,
 	},
 	.unsupported_odd_h_timings	= true,
+	.external_irq_controller	= true,
 
 	.init_resources		= vc5_hdmi_init_resources,
 	.csc_setup		= vc5_hdmi_csc_setup,
@@ -1880,7 +2135,7 @@ static const struct vc4_hdmi_variant bcm2711_hdmi1_variant = {
 	.encoder_type		= VC4_ENCODER_TYPE_HDMI1,
 	.debugfs_name		= "hdmi1_regs",
 	.card_name		= "vc4-hdmi-1",
-	.max_pixel_clock	= 297000000,
+	.max_pixel_clock	= HDMI_14_MAX_TMDS_CLK,
 	.registers		= vc5_hdmi_hdmi1_fields,
 	.num_registers		= ARRAY_SIZE(vc5_hdmi_hdmi1_fields),
 	.phy_lane_mapping	= {
@@ -1890,6 +2145,7 @@ static const struct vc4_hdmi_variant bcm2711_hdmi1_variant = {
 		PHY_LANE_2,
 	},
 	.unsupported_odd_h_timings	= true,
+	.external_irq_controller	= true,
 
 	.init_resources		= vc5_hdmi_init_resources,
 	.csc_setup		= vc5_hdmi_csc_setup,
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.h b/drivers/gpu/drm/vc4/vc4_hdmi.h
index 0526a9cf608a..3cebd1fd00fc 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.h
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.h
@@ -21,10 +21,9 @@ to_vc4_hdmi_encoder(struct drm_encoder *encoder)
 	return container_of(encoder, struct vc4_hdmi_encoder, base.base);
 }
 
-struct drm_display_mode;
-
 struct vc4_hdmi;
 struct vc4_hdmi_register;
+struct vc4_hdmi_connector_state;
 
 enum vc4_hdmi_phy_channel {
 	PHY_LANE_0 = 0,
@@ -43,9 +42,6 @@ struct vc4_hdmi_variant {
 	/* Filename to expose the registers in debugfs */
 	const char *debugfs_name;
 
-	/* Set to true when the CEC support is available */
-	bool cec_available;
-
 	/* Maximum pixel clock supported by the controller (in Hz) */
 	unsigned long long max_pixel_clock;
 
@@ -65,6 +61,13 @@ struct vc4_hdmi_variant {
 	/* The BCM2711 cannot deal with odd horizontal pixel timings */
 	bool unsupported_odd_h_timings;
 
+	/*
+	 * The BCM2711 CEC/hotplug IRQ controller is shared between the
+	 * two HDMI controllers, and we have a proper irqchip driver for
+	 * it.
+	 */
+	bool external_irq_controller;
+
 	/* Callback to get the resources (memory region, interrupts,
 	 * clocks, etc) for that variant.
 	 */
@@ -78,11 +81,12 @@ struct vc4_hdmi_variant {
 
 	/* Callback to configure the video timings in the HDMI block */
 	void (*set_timings)(struct vc4_hdmi *vc4_hdmi,
+			    struct drm_connector_state *state,
 			    struct drm_display_mode *mode);
 
-	/* Callback to initialize the PHY according to the mode */
+	/* Callback to initialize the PHY according to the connector state */
 	void (*phy_init)(struct vc4_hdmi *vc4_hdmi,
-			 struct drm_display_mode *mode);
+			 struct vc4_hdmi_connector_state *vc4_conn_state);
 
 	/* Callback to disable the PHY */
 	void (*phy_disable)(struct vc4_hdmi *vc4_hdmi);
@@ -155,6 +159,7 @@ struct vc4_hdmi {
 	bool cec_tx_ok;
 	bool cec_irq_was_rx;
 
+	struct clk *cec_clock;
 	struct clk *pixel_clock;
 	struct clk *hsm_clock;
 	struct clk *audio_clock;
@@ -180,14 +185,25 @@ encoder_to_vc4_hdmi(struct drm_encoder *encoder)
 	return container_of(_encoder, struct vc4_hdmi, encoder);
 }
 
+struct vc4_hdmi_connector_state {
+	struct drm_connector_state	base;
+	unsigned long long		pixel_rate;
+};
+
+static inline struct vc4_hdmi_connector_state *
+conn_state_to_vc4_hdmi_conn_state(struct drm_connector_state *conn_state)
+{
+	return container_of(conn_state, struct vc4_hdmi_connector_state, base);
+}
+
 void vc4_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi,
-		       struct drm_display_mode *mode);
+		       struct vc4_hdmi_connector_state *vc4_conn_state);
 void vc4_hdmi_phy_disable(struct vc4_hdmi *vc4_hdmi);
 void vc4_hdmi_phy_rng_enable(struct vc4_hdmi *vc4_hdmi);
 void vc4_hdmi_phy_rng_disable(struct vc4_hdmi *vc4_hdmi);
 
 void vc5_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi,
-		       struct drm_display_mode *mode);
+		       struct vc4_hdmi_connector_state *vc4_conn_state);
 void vc5_hdmi_phy_disable(struct vc4_hdmi *vc4_hdmi);
 void vc5_hdmi_phy_rng_enable(struct vc4_hdmi *vc4_hdmi);
 void vc5_hdmi_phy_rng_disable(struct vc4_hdmi *vc4_hdmi);
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi_phy.c b/drivers/gpu/drm/vc4/vc4_hdmi_phy.c
index 057796b54c51..36535480f8e2 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi_phy.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi_phy.c
@@ -127,7 +127,8 @@
 
 #define OSCILLATOR_FREQUENCY	54000000
 
-void vc4_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi, struct drm_display_mode *mode)
+void vc4_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi,
+		       struct vc4_hdmi_connector_state *conn_state)
 {
 	/* PHY should be in reset, like
 	 * vc4_hdmi_encoder_disable() does.
@@ -339,11 +340,12 @@ static void vc5_hdmi_reset_phy(struct vc4_hdmi *vc4_hdmi)
 	HDMI_WRITE(HDMI_TX_PHY_POWERDOWN_CTL, BIT(10));
 }
 
-void vc5_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi, struct drm_display_mode *mode)
+void vc5_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi,
+		       struct vc4_hdmi_connector_state *conn_state)
 {
 	const struct phy_lane_settings *chan0_settings, *chan1_settings, *chan2_settings, *clock_settings;
 	const struct vc4_hdmi_variant *variant = vc4_hdmi->variant;
-	unsigned long long pixel_freq = mode->clock * 1000;
+	unsigned long long pixel_freq = conn_state->pixel_rate;
 	unsigned long long vco_freq;
 	unsigned char word_sel;
 	u8 vco_sel, vco_div;
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi_regs.h b/drivers/gpu/drm/vc4/vc4_hdmi_regs.h
index 96d764ebfe67..e1b58eac766f 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi_regs.h
+++ b/drivers/gpu/drm/vc4/vc4_hdmi_regs.h
@@ -29,6 +29,7 @@ enum vc4_hdmi_field {
 	HDMI_CEC_CPU_MASK_SET,
 	HDMI_CEC_CPU_MASK_STATUS,
 	HDMI_CEC_CPU_STATUS,
+	HDMI_CEC_CPU_SET,
 
 	/*
 	 * Transmit data, first byte is low byte of the 32-bit reg.
@@ -59,9 +60,12 @@ enum vc4_hdmi_field {
 	 */
 	HDMI_CTS_0,
 	HDMI_CTS_1,
+	HDMI_DEEP_COLOR_CONFIG_1,
 	HDMI_DVP_CTL,
 	HDMI_FIFO_CTL,
 	HDMI_FRAME_COUNT,
+	HDMI_GCP_CONFIG,
+	HDMI_GCP_WORD_1,
 	HDMI_HORZA,
 	HDMI_HORZB,
 	HDMI_HOTPLUG,
@@ -196,9 +200,10 @@ static const struct vc4_hdmi_register __maybe_unused vc4_hdmi_fields[] = {
 	VC4_HDMI_REG(HDMI_TX_PHY_RESET_CTL, 0x02c0),
 	VC4_HDMI_REG(HDMI_TX_PHY_CTL_0, 0x02c4),
 	VC4_HDMI_REG(HDMI_CEC_CPU_STATUS, 0x0340),
+	VC4_HDMI_REG(HDMI_CEC_CPU_SET, 0x0344),
 	VC4_HDMI_REG(HDMI_CEC_CPU_CLEAR, 0x0348),
 	VC4_HDMI_REG(HDMI_CEC_CPU_MASK_STATUS, 0x034c),
-	VC4_HDMI_REG(HDMI_CEC_CPU_MASK_SET, 0x034c),
+	VC4_HDMI_REG(HDMI_CEC_CPU_MASK_SET, 0x0350),
 	VC4_HDMI_REG(HDMI_CEC_CPU_MASK_CLEAR, 0x0354),
 	VC4_HDMI_REG(HDMI_RAM_PACKET_START, 0x0400),
 };
@@ -229,6 +234,9 @@ static const struct vc4_hdmi_register __maybe_unused vc5_hdmi_hdmi0_fields[] = {
 	VC4_HDMI_REG(HDMI_VERTB1, 0x0f8),
 	VC4_HDMI_REG(HDMI_MAI_CHANNEL_MAP, 0x09c),
 	VC4_HDMI_REG(HDMI_MAI_CONFIG, 0x0a0),
+	VC4_HDMI_REG(HDMI_DEEP_COLOR_CONFIG_1, 0x170),
+	VC4_HDMI_REG(HDMI_GCP_CONFIG, 0x178),
+	VC4_HDMI_REG(HDMI_GCP_WORD_1, 0x17c),
 	VC4_HDMI_REG(HDMI_HOTPLUG, 0x1a8),
 
 	VC5_DVP_REG(HDMI_CLOCK_STOP, 0x0bc),
@@ -305,6 +313,9 @@ static const struct vc4_hdmi_register __maybe_unused vc5_hdmi_hdmi1_fields[] = {
 	VC4_HDMI_REG(HDMI_VERTB1, 0x0f8),
 	VC4_HDMI_REG(HDMI_MAI_CHANNEL_MAP, 0x09c),
 	VC4_HDMI_REG(HDMI_MAI_CONFIG, 0x0a0),
+	VC4_HDMI_REG(HDMI_DEEP_COLOR_CONFIG_1, 0x170),
+	VC4_HDMI_REG(HDMI_GCP_CONFIG, 0x178),
+	VC4_HDMI_REG(HDMI_GCP_WORD_1, 0x17c),
 	VC4_HDMI_REG(HDMI_HOTPLUG, 0x1a8),
 
 	VC5_DVP_REG(HDMI_CLOCK_STOP, 0x0bc),
diff --git a/drivers/gpu/drm/vc4/vc4_hvs.c b/drivers/gpu/drm/vc4/vc4_hvs.c
index 3b722252d1fb..c239045e05d6 100644
--- a/drivers/gpu/drm/vc4/vc4_hvs.c
+++ b/drivers/gpu/drm/vc4/vc4_hvs.c
@@ -326,10 +326,10 @@ void vc4_hvs_stop_channel(struct drm_device *dev, unsigned int chan)
 		     SCALER_DISPSTATX_EMPTY);
 }
 
-int vc4_hvs_atomic_check(struct drm_crtc *crtc,
-			 struct drm_crtc_state *state)
+int vc4_hvs_atomic_check(struct drm_crtc *crtc, struct drm_atomic_state *state)
 {
-	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(state);
+	struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
+	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc_state);
 	struct drm_device *dev = crtc->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct drm_plane *plane;
@@ -341,10 +341,10 @@ int vc4_hvs_atomic_check(struct drm_crtc *crtc,
 	/* The pixelvalve can only feed one encoder (and encoders are
 	 * 1:1 with connectors.)
 	 */
-	if (hweight32(state->connector_mask) > 1)
+	if (hweight32(crtc_state->connector_mask) > 1)
 		return -EINVAL;
 
-	drm_atomic_crtc_state_for_each_plane_state(plane, plane_state, state)
+	drm_atomic_crtc_state_for_each_plane_state(plane, plane_state, crtc_state)
 		dlist_count += vc4_plane_dlist_size(plane_state);
 
 	dlist_count++; /* Account for SCALER_CTL0_END. */
@@ -391,11 +391,12 @@ static void vc4_hvs_update_dlist(struct drm_crtc *crtc)
 }
 
 void vc4_hvs_atomic_enable(struct drm_crtc *crtc,
-			   struct drm_crtc_state *old_state)
+			   struct drm_atomic_state *state)
 {
 	struct drm_device *dev = crtc->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc->state);
+	struct drm_crtc_state *new_crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
+	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(new_crtc_state);
 	struct drm_display_mode *mode = &crtc->state->adjusted_mode;
 	bool oneshot = vc4_state->feed_txp;
 
@@ -404,9 +405,10 @@ void vc4_hvs_atomic_enable(struct drm_crtc *crtc,
 }
 
 void vc4_hvs_atomic_disable(struct drm_crtc *crtc,
-			    struct drm_crtc_state *old_state)
+			    struct drm_atomic_state *state)
 {
 	struct drm_device *dev = crtc->dev;
+	struct drm_crtc_state *old_state = drm_atomic_get_old_crtc_state(state, crtc);
 	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(old_state);
 	unsigned int chan = vc4_state->assigned_channel;
 
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
index ba310c0ab5f6..f09254c2497d 100644
--- a/drivers/gpu/drm/vc4/vc4_kms.c
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -39,7 +39,11 @@ static struct vc4_ctm_state *to_vc4_ctm_state(struct drm_private_state *priv)
 
 struct vc4_hvs_state {
 	struct drm_private_state base;
-	unsigned int unassigned_channels;
+
+	struct {
+		unsigned in_use: 1;
+		struct drm_crtc_commit *pending_commit;
+	} fifo_state[HVS_NUM_CHANNELS];
 };
 
 static struct vc4_hvs_state *
@@ -183,6 +187,32 @@ vc4_ctm_commit(struct vc4_dev *vc4, struct drm_atomic_state *state)
 }
 
 static struct vc4_hvs_state *
+vc4_hvs_get_new_global_state(struct drm_atomic_state *state)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(state->dev);
+	struct drm_private_state *priv_state;
+
+	priv_state = drm_atomic_get_new_private_obj_state(state, &vc4->hvs_channels);
+	if (IS_ERR(priv_state))
+		return ERR_CAST(priv_state);
+
+	return to_vc4_hvs_state(priv_state);
+}
+
+static struct vc4_hvs_state *
+vc4_hvs_get_old_global_state(struct drm_atomic_state *state)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(state->dev);
+	struct drm_private_state *priv_state;
+
+	priv_state = drm_atomic_get_old_private_obj_state(state, &vc4->hvs_channels);
+	if (IS_ERR(priv_state))
+		return ERR_CAST(priv_state);
+
+	return to_vc4_hvs_state(priv_state);
+}
+
+static struct vc4_hvs_state *
 vc4_hvs_get_global_state(struct drm_atomic_state *state)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(state->dev);
@@ -302,14 +332,15 @@ static void vc5_hvs_pv_muxing_commit(struct vc4_dev *vc4,
 	}
 }
 
-static void
-vc4_atomic_complete_commit(struct drm_atomic_state *state)
+static void vc4_atomic_commit_tail(struct drm_atomic_state *state)
 {
 	struct drm_device *dev = state->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_hvs *hvs = vc4->hvs;
+	struct drm_crtc_state *old_crtc_state;
 	struct drm_crtc_state *new_crtc_state;
 	struct drm_crtc *crtc;
+	struct vc4_hvs_state *old_hvs_state;
 	int i;
 
 	for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) {
@@ -325,9 +356,35 @@ vc4_atomic_complete_commit(struct drm_atomic_state *state)
 	if (vc4->hvs->hvs5)
 		clk_set_min_rate(hvs->core_clk, 500000000);
 
-	drm_atomic_helper_wait_for_fences(dev, state, false);
+	old_hvs_state = vc4_hvs_get_old_global_state(state);
+	if (!old_hvs_state)
+		return;
+
+	for_each_old_crtc_in_state(state, crtc, old_crtc_state, i) {
+		struct vc4_crtc_state *vc4_crtc_state =
+			to_vc4_crtc_state(old_crtc_state);
+		struct drm_crtc_commit *commit;
+		unsigned int channel = vc4_crtc_state->assigned_channel;
+		unsigned long done;
+
+		if (channel == VC4_HVS_CHANNEL_DISABLED)
+			continue;
+
+		if (!old_hvs_state->fifo_state[channel].in_use)
+			continue;
+
+		commit = old_hvs_state->fifo_state[i].pending_commit;
+		if (!commit)
+			continue;
 
-	drm_atomic_helper_wait_for_dependencies(state);
+		done = wait_for_completion_timeout(&commit->hw_done, 10 * HZ);
+		if (!done)
+			drm_err(dev, "Timed out waiting for hw_done\n");
+
+		done = wait_for_completion_timeout(&commit->flip_done, 10 * HZ);
+		if (!done)
+			drm_err(dev, "Timed out waiting for flip_done\n");
+	}
 
 	drm_atomic_helper_commit_modeset_disables(dev, state);
 
@@ -350,125 +407,37 @@ vc4_atomic_complete_commit(struct drm_atomic_state *state)
 
 	drm_atomic_helper_cleanup_planes(dev, state);
 
-	drm_atomic_helper_commit_cleanup_done(state);
-
 	if (vc4->hvs->hvs5)
 		clk_set_min_rate(hvs->core_clk, 0);
-
-	drm_atomic_state_put(state);
-
-	up(&vc4->async_modeset);
-}
-
-static void commit_work(struct work_struct *work)
-{
-	struct drm_atomic_state *state = container_of(work,
-						      struct drm_atomic_state,
-						      commit_work);
-	vc4_atomic_complete_commit(state);
 }
 
-/**
- * vc4_atomic_commit - commit validated state object
- * @dev: DRM device
- * @state: the driver state object
- * @nonblock: nonblocking commit
- *
- * This function commits a with drm_atomic_helper_check() pre-validated state
- * object. This can still fail when e.g. the framebuffer reservation fails. For
- * now this doesn't implement asynchronous commits.
- *
- * RETURNS
- * Zero for success or -errno.
- */
-static int vc4_atomic_commit(struct drm_device *dev,
-			     struct drm_atomic_state *state,
-			     bool nonblock)
+static int vc4_atomic_commit_setup(struct drm_atomic_state *state)
 {
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int ret;
-
-	if (state->async_update) {
-		ret = down_interruptible(&vc4->async_modeset);
-		if (ret)
-			return ret;
-
-		ret = drm_atomic_helper_prepare_planes(dev, state);
-		if (ret) {
-			up(&vc4->async_modeset);
-			return ret;
-		}
-
-		drm_atomic_helper_async_commit(dev, state);
-
-		drm_atomic_helper_cleanup_planes(dev, state);
-
-		up(&vc4->async_modeset);
-
-		return 0;
-	}
+	struct drm_crtc_state *crtc_state;
+	struct vc4_hvs_state *hvs_state;
+	struct drm_crtc *crtc;
+	unsigned int i;
 
-	/* We know for sure we don't want an async update here. Set
-	 * state->legacy_cursor_update to false to prevent
-	 * drm_atomic_helper_setup_commit() from auto-completing
-	 * commit->flip_done.
-	 */
-	state->legacy_cursor_update = false;
-	ret = drm_atomic_helper_setup_commit(state, nonblock);
-	if (ret)
-		return ret;
+	hvs_state = vc4_hvs_get_new_global_state(state);
+	if (!hvs_state)
+		return -EINVAL;
 
-	INIT_WORK(&state->commit_work, commit_work);
+	for_each_new_crtc_in_state(state, crtc, crtc_state, i) {
+		struct vc4_crtc_state *vc4_crtc_state =
+			to_vc4_crtc_state(crtc_state);
+		unsigned int channel =
+			vc4_crtc_state->assigned_channel;
 
-	ret = down_interruptible(&vc4->async_modeset);
-	if (ret)
-		return ret;
+		if (channel == VC4_HVS_CHANNEL_DISABLED)
+			continue;
 
-	ret = drm_atomic_helper_prepare_planes(dev, state);
-	if (ret) {
-		up(&vc4->async_modeset);
-		return ret;
-	}
+		if (!hvs_state->fifo_state[channel].in_use)
+			continue;
 
-	if (!nonblock) {
-		ret = drm_atomic_helper_wait_for_fences(dev, state, true);
-		if (ret) {
-			drm_atomic_helper_cleanup_planes(dev, state);
-			up(&vc4->async_modeset);
-			return ret;
-		}
+		hvs_state->fifo_state[channel].pending_commit =
+			drm_crtc_commit_get(crtc_state->commit);
 	}
 
-	/*
-	 * This is the point of no return - everything below never fails except
-	 * when the hw goes bonghits. Which means we can commit the new state on
-	 * the software side now.
-	 */
-
-	BUG_ON(drm_atomic_helper_swap_state(state, false) < 0);
-
-	/*
-	 * Everything below can be run asynchronously without the need to grab
-	 * any modeset locks at all under one condition: It must be guaranteed
-	 * that the asynchronous work has either been cancelled (if the driver
-	 * supports it, which at least requires that the framebuffers get
-	 * cleaned up with drm_atomic_helper_cleanup_planes()) or completed
-	 * before the new state gets committed on the software side with
-	 * drm_atomic_helper_swap_state().
-	 *
-	 * This scheme allows new atomic state updates to be prepared and
-	 * checked in parallel to the asynchronous completion of the previous
-	 * update. Which is important since compositors need to figure out the
-	 * composition of the next frame right after having submitted the
-	 * current layout.
-	 */
-
-	drm_atomic_state_get(state);
-	if (nonblock)
-		queue_work(system_unbound_wq, &state->commit_work);
-	else
-		vc4_atomic_complete_commit(state);
-
 	return 0;
 }
 
@@ -697,6 +666,7 @@ vc4_hvs_channels_duplicate_state(struct drm_private_obj *obj)
 {
 	struct vc4_hvs_state *old_state = to_vc4_hvs_state(obj->state);
 	struct vc4_hvs_state *state;
+	unsigned int i;
 
 	state = kzalloc(sizeof(*state), GFP_KERNEL);
 	if (!state)
@@ -704,7 +674,16 @@ vc4_hvs_channels_duplicate_state(struct drm_private_obj *obj)
 
 	__drm_atomic_helper_private_obj_duplicate_state(obj, &state->base);
 
-	state->unassigned_channels = old_state->unassigned_channels;
+
+	for (i = 0; i < HVS_NUM_CHANNELS; i++) {
+		state->fifo_state[i].in_use = old_state->fifo_state[i].in_use;
+
+		if (!old_state->fifo_state[i].pending_commit)
+			continue;
+
+		state->fifo_state[i].pending_commit =
+			drm_crtc_commit_get(old_state->fifo_state[i].pending_commit);
+	}
 
 	return &state->base;
 }
@@ -713,6 +692,14 @@ static void vc4_hvs_channels_destroy_state(struct drm_private_obj *obj,
 					   struct drm_private_state *state)
 {
 	struct vc4_hvs_state *hvs_state = to_vc4_hvs_state(state);
+	unsigned int i;
+
+	for (i = 0; i < HVS_NUM_CHANNELS; i++) {
+		if (!hvs_state->fifo_state[i].pending_commit)
+			continue;
+
+		drm_crtc_commit_put(hvs_state->fifo_state[i].pending_commit);
+	}
 
 	kfree(hvs_state);
 }
@@ -737,7 +724,6 @@ static int vc4_hvs_channels_obj_init(struct vc4_dev *vc4)
 	if (!state)
 		return -ENOMEM;
 
-	state->unassigned_channels = GENMASK(HVS_NUM_CHANNELS - 1, 0);
 	drm_atomic_private_obj_init(&vc4->base, &vc4->hvs_channels,
 				    &state->base,
 				    &vc4_hvs_state_funcs);
@@ -781,12 +767,17 @@ static int vc4_pv_muxing_atomic_check(struct drm_device *dev,
 	struct vc4_hvs_state *hvs_new_state;
 	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
 	struct drm_crtc *crtc;
+	unsigned int unassigned_channels = 0;
 	unsigned int i;
 
 	hvs_new_state = vc4_hvs_get_global_state(state);
 	if (!hvs_new_state)
 		return -EINVAL;
 
+	for (i = 0; i < ARRAY_SIZE(hvs_new_state->fifo_state); i++)
+		if (!hvs_new_state->fifo_state[i].in_use)
+			unassigned_channels |= BIT(i);
+
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		struct vc4_crtc_state *old_vc4_crtc_state =
 			to_vc4_crtc_state(old_crtc_state);
@@ -794,6 +785,7 @@ static int vc4_pv_muxing_atomic_check(struct drm_device *dev,
 			to_vc4_crtc_state(new_crtc_state);
 		struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
 		unsigned int matching_channels;
+		unsigned int channel;
 
 		/* Nothing to do here, let's skip it */
 		if (old_crtc_state->enable == new_crtc_state->enable)
@@ -804,7 +796,8 @@ static int vc4_pv_muxing_atomic_check(struct drm_device *dev,
 
 		/* If we're disabling our CRTC, we put back our channel */
 		if (!new_crtc_state->enable) {
-			hvs_new_state->unassigned_channels |= BIT(old_vc4_crtc_state->assigned_channel);
+			channel = old_vc4_crtc_state->assigned_channel;
+			hvs_new_state->fifo_state[channel].in_use = false;
 			new_vc4_crtc_state->assigned_channel = VC4_HVS_CHANNEL_DISABLED;
 			continue;
 		}
@@ -833,15 +826,14 @@ static int vc4_pv_muxing_atomic_check(struct drm_device *dev,
 		 * the future, we will need to have something smarter,
 		 * but it works so far.
 		 */
-		matching_channels = hvs_new_state->unassigned_channels & vc4_crtc->data->hvs_available_channels;
-		if (matching_channels) {
-			unsigned int channel = ffs(matching_channels) - 1;
-
-			new_vc4_crtc_state->assigned_channel = channel;
-			hvs_new_state->unassigned_channels &= ~BIT(channel);
-		} else {
+		matching_channels = unassigned_channels & vc4_crtc->data->hvs_available_channels;
+		if (!matching_channels)
 			return -EINVAL;
-		}
+
+		channel = ffs(matching_channels) - 1;
+		new_vc4_crtc_state->assigned_channel = channel;
+		unassigned_channels &= ~BIT(channel);
+		hvs_new_state->fifo_state[channel].in_use = true;
 	}
 
 	return 0;
@@ -867,9 +859,14 @@ vc4_atomic_check(struct drm_device *dev, struct drm_atomic_state *state)
 	return vc4_load_tracker_atomic_check(state);
 }
 
+static struct drm_mode_config_helper_funcs vc4_mode_config_helpers = {
+	.atomic_commit_setup	= vc4_atomic_commit_setup,
+	.atomic_commit_tail	= vc4_atomic_commit_tail,
+};
+
 static const struct drm_mode_config_funcs vc4_mode_funcs = {
 	.atomic_check = vc4_atomic_check,
-	.atomic_commit = vc4_atomic_commit,
+	.atomic_commit = drm_atomic_helper_commit,
 	.fb_create = vc4_fb_create,
 };
 
@@ -889,8 +886,6 @@ int vc4_kms_load(struct drm_device *dev)
 		vc4->load_tracker_enabled = true;
 	}
 
-	sema_init(&vc4->async_modeset, 1);
-
 	/* Set support for vblank irq fast disable, before drm_vblank_init() */
 	dev->vblank_disable_immediate = true;
 
@@ -910,6 +905,7 @@ int vc4_kms_load(struct drm_device *dev)
 	}
 
 	dev->mode_config.funcs = &vc4_mode_funcs;
+	dev->mode_config.helper_private = &vc4_mode_config_helpers;
 	dev->mode_config.preferred_depth = 24;
 	dev->mode_config.async_page_flip = true;
 	dev->mode_config.allow_fb_modifiers = true;
diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c
index af4b8944a603..7322169c0682 100644
--- a/drivers/gpu/drm/vc4/vc4_plane.c
+++ b/drivers/gpu/drm/vc4/vc4_plane.c
@@ -1283,11 +1283,6 @@ static const struct drm_plane_helper_funcs vc4_plane_helper_funcs = {
 	.atomic_async_update = vc4_plane_atomic_async_update,
 };
 
-static void vc4_plane_destroy(struct drm_plane *plane)
-{
-	drm_plane_cleanup(plane);
-}
-
 static bool vc4_format_mod_supported(struct drm_plane *plane,
 				     uint32_t format,
 				     uint64_t modifier)
@@ -1338,7 +1333,7 @@ static bool vc4_format_mod_supported(struct drm_plane *plane,
 static const struct drm_plane_funcs vc4_plane_funcs = {
 	.update_plane = drm_atomic_helper_update_plane,
 	.disable_plane = drm_atomic_helper_disable_plane,
-	.destroy = vc4_plane_destroy,
+	.destroy = drm_plane_cleanup,
 	.set_property = NULL,
 	.reset = vc4_plane_reset,
 	.atomic_duplicate_state = vc4_plane_duplicate_state,
diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c
index 8aa5220885f4..c0122d83b651 100644
--- a/drivers/gpu/drm/vc4/vc4_txp.c
+++ b/drivers/gpu/drm/vc4/vc4_txp.c
@@ -382,7 +382,6 @@ static const struct drm_crtc_funcs vc4_txp_crtc_funcs = {
 	.reset			= vc4_crtc_reset,
 	.atomic_duplicate_state	= vc4_crtc_duplicate_state,
 	.atomic_destroy_state	= vc4_crtc_destroy_state,
-	.gamma_set		= drm_atomic_helper_legacy_gamma_set,
 	.enable_vblank		= vc4_txp_enable_vblank,
 	.disable_vblank		= vc4_txp_disable_vblank,
 };
@@ -395,7 +394,7 @@ static int vc4_txp_atomic_check(struct drm_crtc *crtc,
 	struct vc4_crtc_state *vc4_state = to_vc4_crtc_state(crtc_state);
 	int ret;
 
-	ret = vc4_hvs_atomic_check(crtc, crtc_state);
+	ret = vc4_hvs_atomic_check(crtc, state);
 	if (ret)
 		return ret;
 
@@ -408,23 +407,19 @@ static int vc4_txp_atomic_check(struct drm_crtc *crtc,
 static void vc4_txp_atomic_enable(struct drm_crtc *crtc,
 				  struct drm_atomic_state *state)
 {
-	struct drm_crtc_state *old_state = drm_atomic_get_old_crtc_state(state,
-									 crtc);
 	drm_crtc_vblank_on(crtc);
-	vc4_hvs_atomic_enable(crtc, old_state);
+	vc4_hvs_atomic_enable(crtc, state);
 }
 
 static void vc4_txp_atomic_disable(struct drm_crtc *crtc,
 				   struct drm_atomic_state *state)
 {
-	struct drm_crtc_state *old_state = drm_atomic_get_old_crtc_state(state,
-									 crtc);
 	struct drm_device *dev = crtc->dev;
 
 	/* Disable vblank irq handling before crtc is disabled. */
 	drm_crtc_vblank_off(crtc);
 
-	vc4_hvs_atomic_disable(crtc, old_state);
+	vc4_hvs_atomic_disable(crtc, state);
 
 	/*
 	 * Make sure we issue a vblank event after disabling the CRTC if