Merge tag 'drm-next-2019-09-18' of git://anongit.freedesktop.org/drm/drm

Pull drm updates from Dave Airlie:
 "This is the main pull request for 5.4-rc1 merge window. I don't think
  there is anything outstanding so next week should just be fixes, but
  we'll see if I missed anything. I landed some fixes earlier in the
  week but got delayed writing summary and sending it out, due to a mix
  of sick kid and jetlag!

  There are some fixes pending, but I'd rather get the main merge out of
  the way instead of delaying it longer.

  It's also pretty large in commit count and new amd header file size.
  The largest thing is four new amdgpu products (navi12/14, arcturus and
  renoir APU support).

  Otherwise it's pretty much lots of work across the board, i915 has
  started landing tigerlake support, lots of icelake fixes and lots of
  locking reworking for future gpu support, lots of header file rework
  (drmP.h is nearly gone), some old legacy hacks (DRM_WAIT_ON) have been
  put into the places they are needed.

  uapi:
   - content protection type property for HDCP

  core:
   - rework include dependencies
   - lots of drmP.h removals
   - link rate calculation robustness fix
   - make fb helper map only when required
   - add connector->DDC adapter link
   - DRM_WAIT_ON removed
   - drop DRM_AUTH usage from drivers

  dma-buf:
   - reservation object fence helper

  dma-fence:
   - shrink dma_fence struct
   - merge signal functions
   - store timestamps in dma_fence
   - selftests

  ttm:
   - embed drm_get_object struct into ttm_buffer_object
   - release_notify callback

  bridges:
   - sii902x - audio graph card support
   - tc358767 - aux data handling rework
   - ti-snd64dsi86 - debugfs support, DSI mode flags support

  panels:
   - Support for GiantPlus GPM940B0, Sharp LQ070Y3DG3B, Ortustech
     COM37H3M, Novatek NT39016, Sharp LS020B1DD01D, Raydium RM67191, Boe
     Himax8279d, Sharp LD-D5116Z01B
   - TI nspire, NEC NL8048HL11, LG Philips LB035Q02, Sharp LS037V7DW01,
     Sony ACX565AKM, Toppoly TD028TTEC1 Toppoly TD043MTEA1

  i915:
   - Initial tigerlake platform support
   - Locking simplification work, general all over refactoring.
   - Selftests
   - HDCP debug info improvements
   - DSI properties
   - Icelake display PLL fixes, colorspace fixes, bandwidth fixes, DSI
     suspend/resume
   - GuC fixes
   - Perf fixes
   - ElkhartLake enablement
   - DP MST fixes
   - GVT - command parser enhancements

  amdgpu:
   - add wipe memory on release flag for buffer creation
   - Navi12/14 support (may be marked experimental)
   - Arcturus support
   - Renoir APU support
   - mclk DPM for Navi
   - DC display fixes
   - Raven scatter/gather support
   - RAS support for GFX
   - Navi12 + Arcturus power features
   - GPU reset for Picasso
   - smu11 i2c controller support

  amdkfd:
   - navi12/14 support
   - Arcturus support

  radeon:
   - kexec fix

  nouveau:
   - improved display color management
   - detect lack of GPU power cables

  vmwgfx:
   - evicition priority support
   - remove unused security feature

  msm:
   - msm8998 display support
   - better async commit support for cursor updates

  etnaviv:
   - per-process address space support
   - performance counter fixes
   - softpin support

  mcde:
   - DCS transfers fix

  exynos:
   - drmP.h cleanup

  lima:
   - reduce logging

  kirin:
   - misc clenaups

  komeda:
   - dual-link support
   - DT memory regions

  hisilicon:
   - misc fixes

  imx:
   - IPUv3 image converter fixes
   - 32-bit RGB V4L2 pixel format support

  ingenic:
   - more support for panel related cases

  mgag200:
   - cursor support fix

  panfrost:
   - export GPU features register to userspace
   - gpu heap allocations
   - per-fd address space support

  pl111:
   - CLD pads wiring support removed from DT

  rockchip:
   - rework to use DRM PSR helpers
   - fix bug in VOP_WIN_GET macro
   - DSI DT binding rework

  sun4i:
   - improve support for color encoding and range
   - DDC enabled GPIO

  tinydrm:
   - rework SPI support
   - improve MIPI-DBI support
   - moved to drm/tiny

  vkms:
   - rework CRC tracking

  dw-hdmi:
   - get_eld and i2s improvements

  gm12u320:
   - misc fixes

  meson:
   - global code cleanup
   - vpu feature detect

  omap:
   - alpha/pixel blend mode properties

  rcar-du:
   - misc fixes"

* tag 'drm-next-2019-09-18' of git://anongit.freedesktop.org/drm/drm: (2112 commits)
  drm/nouveau/bar/gm20b: Avoid BAR1 teardown during init
  drm/nouveau: Fix ordering between TTM and GEM release
  drm/nouveau/prime: Extend DMA reservation object lock
  drm/nouveau: Fix fallout from reservation object rework
  drm/nouveau/kms/nv50-: Don't create MSTMs for eDP connectors
  drm/i915: Use NOEVICT for first pass on attemping to pin a GGTT mmap
  drm/i915: to make vgpu ppgtt notificaiton as atomic operation
  drm/i915: Flush the existing fence before GGTT read/write
  drm/i915: Hold irq-off for the entire fake lock period
  drm/i915/gvt: update RING_START reg of vGPU when the context is submitted to i915
  drm/i915/gvt: update vgpu workload head pointer correctly
  drm/mcde: Fix DSI transfers
  drm/msm: Use the correct dma_sync calls harder
  drm/msm: remove unlikely() from WARN_ON() conditions
  drm/msm/dsi: Fix return value check for clk_get_parent
  drm/msm: add atomic traces
  drm/msm/dpu: async commit support
  drm/msm: async commit support
  drm/msm: split power control from prepare/complete_commit
  drm/msm: add kms->flush_commit()
  ...

16 files changed, 383 insertions, 79 deletions

diff --git a/drivers/gpu/drm/nouveau/dispnv50/atom.h b/drivers/gpu/drm/nouveau/dispnv50/atom.h
index b5fae5ab3fa8..43df86c38f58 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/atom.h
+++ b/drivers/gpu/drm/nouveau/dispnv50/atom.h
@@ -185,6 +185,11 @@ struct nv50_wndw_atom {
 	} xlut;
 
 	struct {
+		u32 matrix[12];
+		bool valid;
+	} csc;
+
+	struct {
 		u8  mode:2;
 		u8  interval:4;
 
@@ -216,14 +221,23 @@ struct nv50_wndw_atom {
 		u16 y;
 	} point;
 
+	struct {
+		u8 depth;
+		u8 k1;
+		u8 src_color:4;
+		u8 dst_color:4;
+	} blend;
+
 	union nv50_wndw_atom_mask {
 		struct {
 			bool ntfy:1;
 			bool sema:1;
 			bool xlut:1;
+			bool csc:1;
 			bool image:1;
 			bool scale:1;
 			bool point:1;
+			bool blend:1;
 		};
 		u8 mask;
 	} set, clr;
diff --git a/drivers/gpu/drm/nouveau/dispnv50/base507c.c b/drivers/gpu/drm/nouveau/dispnv50/base507c.c
index d5e295ca2caa..00a85f1e1a4a 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/base507c.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/base507c.c
@@ -25,7 +25,9 @@
 #include <nvif/event.h>
 
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_plane_helper.h>
+
 #include "nouveau_bo.h"
 
 void
@@ -56,12 +58,21 @@ static void
 base507c_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
 {
 	u32 *push;
-	if ((push = evo_wait(&wndw->wndw, 10))) {
+	if ((push = evo_wait(&wndw->wndw, 13))) {
 		evo_mthd(push, 0x0084, 1);
 		evo_data(push, asyw->image.mode << 8 |
 			       asyw->image.interval << 4);
 		evo_mthd(push, 0x00c0, 1);
 		evo_data(push, asyw->image.handle[0]);
+		if (asyw->image.format == 0xca) {
+			evo_mthd(push, 0x0110, 2);
+			evo_data(push, 1);
+			evo_data(push, 0x6400);
+		} else {
+			evo_mthd(push, 0x0110, 2);
+			evo_data(push, 0);
+			evo_data(push, 0);
+		}
 		evo_mthd(push, 0x0800, 5);
 		evo_data(push, asyw->image.offset[0] >> 8);
 		evo_data(push, 0x00000000);
@@ -179,9 +190,6 @@ base507c_acquire(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw,
 	const struct drm_framebuffer *fb = asyw->state.fb;
 	int ret;
 
-	if (!fb->format->depth)
-		return -EINVAL;
-
 	ret = drm_atomic_helper_check_plane_state(&asyw->state, &asyh->state,
 						  DRM_PLANE_HELPER_NO_SCALING,
 						  DRM_PLANE_HELPER_NO_SCALING,
@@ -200,6 +208,14 @@ base507c_acquire(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw,
 	asyh->base.y = asyw->state.src.y1 >> 16;
 	asyh->base.w = asyw->state.fb->width;
 	asyh->base.h = asyw->state.fb->height;
+
+	/* Some newer formats, esp FP16 ones, don't have a
+	 * "depth". There's nothing that really makes sense there
+	 * either, so just set it to the implicit bit count.
+	 */
+	if (!asyh->base.depth)
+		asyh->base.depth = asyh->base.cpp * 8;
+
 	return 0;
 }
 
@@ -215,6 +231,8 @@ base507c_format[] = {
 	DRM_FORMAT_ABGR2101010,
 	DRM_FORMAT_XBGR8888,
 	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XBGR16161616F,
+	DRM_FORMAT_ABGR16161616F,
 	0
 };
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/base827c.c b/drivers/gpu/drm/nouveau/dispnv50/base827c.c
index 73646819a0d6..f4c05949dd62 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/base827c.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/base827c.c
@@ -25,12 +25,21 @@ static void
 base827c_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
 {
 	u32 *push;
-	if ((push = evo_wait(&wndw->wndw, 10))) {
+	if ((push = evo_wait(&wndw->wndw, 13))) {
 		evo_mthd(push, 0x0084, 1);
 		evo_data(push, asyw->image.mode << 8 |
 			       asyw->image.interval << 4);
 		evo_mthd(push, 0x00c0, 1);
 		evo_data(push, asyw->image.handle[0]);
+		if (asyw->image.format == 0xca) {
+			evo_mthd(push, 0x0110, 2);
+			evo_data(push, 1);
+			evo_data(push, 0x6400);
+		} else {
+			evo_mthd(push, 0x0110, 2);
+			evo_data(push, 0);
+			evo_data(push, 0);
+		}
 		evo_mthd(push, 0x0800, 5);
 		evo_data(push, asyw->image.offset[0] >> 8);
 		evo_data(push, 0x00000000);
diff --git a/drivers/gpu/drm/nouveau/dispnv50/base907c.c b/drivers/gpu/drm/nouveau/dispnv50/base907c.c
index 049ce6da321c..5f2de77e0f32 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/base907c.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/base907c.c
@@ -83,6 +83,68 @@ base907c_ilut(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
 	asyw->xlut.i.load = head907d_olut_load;
 }
 
+static inline u32
+csc_drm_to_base(u64 in)
+{
+	/* base takes a 19-bit 2's complement value in S3.16 format */
+	bool sign = in & BIT_ULL(63);
+	u32 integer = (in >> 32) & 0x7fffffff;
+	u32 fraction = in & 0xffffffff;
+
+	if (integer >= 4) {
+		return (1 << 18) - (sign ? 0 : 1);
+	} else {
+		u32 ret = (integer << 16) | (fraction >> 16);
+		if (sign)
+			ret = -ret;
+		return ret & GENMASK(18, 0);
+	}
+}
+
+void
+base907c_csc(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw,
+	     const struct drm_color_ctm *ctm)
+{
+	int i, j;
+
+	for (j = 0; j < 3; j++) {
+		for (i = 0; i < 4; i++) {
+			u32 *val = &asyw->csc.matrix[j * 4 + i];
+			/* DRM does not support constant offset, while
+			 * HW CSC does. Skip it. */
+			if (i == 3) {
+				*val = 0;
+			} else {
+				*val = csc_drm_to_base(ctm->matrix[j * 3 + i]);
+			}
+		}
+	}
+}
+
+static void
+base907c_csc_clr(struct nv50_wndw *wndw)
+{
+	u32 *push;
+	if ((push = evo_wait(&wndw->wndw, 2))) {
+		evo_mthd(push, 0x0140, 1);
+		evo_data(push, 0x00000000);
+		evo_kick(push, &wndw->wndw);
+	}
+}
+
+static void
+base907c_csc_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
+{
+	u32 *push, i;
+	if ((push = evo_wait(&wndw->wndw, 13))) {
+		evo_mthd(push, 0x0140, 12);
+		evo_data(push, asyw->csc.matrix[0] | 0x80000000);
+		for (i = 1; i < 12; i++)
+			evo_data(push, asyw->csc.matrix[i]);
+		evo_kick(push, &wndw->wndw);
+	}
+}
+
 const struct nv50_wndw_func
 base907c = {
 	.acquire = base507c_acquire,
@@ -94,6 +156,9 @@ base907c = {
 	.ntfy_clr = base507c_ntfy_clr,
 	.ntfy_wait_begun = base507c_ntfy_wait_begun,
 	.ilut = base907c_ilut,
+	.csc = base907c_csc,
+	.csc_set = base907c_csc_set,
+	.csc_clr = base907c_csc_clr,
 	.olut_core = true,
 	.xlut_set = base907c_xlut_set,
 	.xlut_clr = base907c_xlut_clr,
diff --git a/drivers/gpu/drm/nouveau/dispnv50/base917c.c b/drivers/gpu/drm/nouveau/dispnv50/base917c.c
index 54d705bb81a5..a1baed4fe0e9 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/base917c.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/base917c.c
@@ -36,6 +36,8 @@ base917c_format[] = {
 	DRM_FORMAT_ABGR8888,
 	DRM_FORMAT_XRGB2101010,
 	DRM_FORMAT_ARGB2101010,
+	DRM_FORMAT_XBGR16161616F,
+	DRM_FORMAT_ABGR16161616F,
 	0
 };
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/corec37d.c b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c
index 7860774b65bc..40d9b654ab8c 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/corec37d.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c
@@ -82,7 +82,7 @@ corec37d_init(struct nv50_core *core)
 		for (i = 0; i < windows; i++) {
 			evo_mthd(push, 0x1000 + (i * 0x080), 3);
 			evo_data(push, i >> 1);
-			evo_data(push, 0x00000017);
+			evo_data(push, 0x0000001f);
 			evo_data(push, 0x00000000);
 			evo_mthd(push, 0x1010 + (i * 0x080), 1);
 			evo_data(push, 0x00127fff);
diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c
index 5c36c75232e6..b46be8a091e9 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c
@@ -30,14 +30,14 @@
 #include <linux/dma-mapping.h>
 #include <linux/hdmi.h>
 
-#include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_dp_helper.h>
+#include <drm/drm_edid.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_plane_helper.h>
 #include <drm/drm_probe_helper.h>
 #include <drm/drm_scdc_helper.h>
-#include <drm/drm_edid.h>
+#include <drm/drm_vblank.h>
 
 #include <nvif/class.h>
 #include <nvif/cl0002.h>
@@ -1603,7 +1603,8 @@ nv50_sor_create(struct drm_connector *connector, struct dcb_output *dcbe)
 			nv_encoder->aux = aux;
 		}
 
-		if ((data = nvbios_dp_table(bios, &ver, &hdr, &cnt, &len)) &&
+		if (nv_connector->type != DCB_CONNECTOR_eDP &&
+		    (data = nvbios_dp_table(bios, &ver, &hdr, &cnt, &len)) &&
 		    ver >= 0x40 && (nvbios_rd08(bios, data + 0x08) & 0x04)) {
 			ret = nv50_mstm_new(nv_encoder, &nv_connector->aux, 16,
 					    nv_connector->base.base.id,
@@ -1830,8 +1831,11 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 
 		NV_ATOMIC(drm, "%s: clr %04x (set %04x)\n", crtc->name,
 			  asyh->clr.mask, asyh->set.mask);
-		if (old_crtc_state->active && !new_crtc_state->active)
+
+		if (old_crtc_state->active && !new_crtc_state->active) {
+			pm_runtime_put_noidle(dev->dev);
 			drm_crtc_vblank_off(crtc);
+		}
 
 		if (asyh->clr.mask) {
 			nv50_head_flush_clr(head, asyh, atom->flush_disable);
@@ -1917,8 +1921,10 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 		}
 
 		if (new_crtc_state->active) {
-			if (!old_crtc_state->active)
+			if (!old_crtc_state->active) {
 				drm_crtc_vblank_on(crtc);
+				pm_runtime_get_noresume(dev->dev);
+			}
 			if (new_crtc_state->event)
 				drm_crtc_vblank_get(crtc);
 		}
@@ -1983,6 +1989,10 @@ nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
 	drm_atomic_helper_cleanup_planes(dev, state);
 	drm_atomic_helper_commit_cleanup_done(state);
 	drm_atomic_state_put(state);
+
+	/* Drop the RPM ref we got from nv50_disp_atomic_commit() */
+	pm_runtime_mark_last_busy(dev->dev);
+	pm_runtime_put_autosuspend(dev->dev);
 }
 
 static void
@@ -1997,11 +2007,8 @@ static int
 nv50_disp_atomic_commit(struct drm_device *dev,
 			struct drm_atomic_state *state, bool nonblock)
 {
-	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct drm_plane_state *new_plane_state;
 	struct drm_plane *plane;
-	struct drm_crtc *crtc;
-	bool active = false;
 	int ret, i;
 
 	ret = pm_runtime_get_sync(dev->dev);
@@ -2038,27 +2045,17 @@ nv50_disp_atomic_commit(struct drm_device *dev,
 
 	drm_atomic_state_get(state);
 
+	/*
+	 * Grab another RPM ref for the commit tail, which will release the
+	 * ref when it's finished
+	 */
+	pm_runtime_get_noresume(dev->dev);
+
 	if (nonblock)
 		queue_work(system_unbound_wq, &state->commit_work);
 	else
 		nv50_disp_atomic_commit_tail(state);
 
-	drm_for_each_crtc(crtc, dev) {
-		if (crtc->state->active) {
-			if (!drm->have_disp_power_ref) {
-				drm->have_disp_power_ref = true;
-				return 0;
-			}
-			active = true;
-			break;
-		}
-	}
-
-	if (!active && drm->have_disp_power_ref) {
-		pm_runtime_put_autosuspend(dev->dev);
-		drm->have_disp_power_ref = false;
-	}
-
 err_cleanup:
 	if (ret)
 		drm_atomic_helper_cleanup_planes(dev, state);
@@ -2320,6 +2317,7 @@ nv50_display_create(struct drm_device *dev)
 	disp->disp = &nouveau_display(dev)->disp;
 	dev->mode_config.funcs = &nv50_disp_func;
 	dev->mode_config.quirk_addfb_prefer_xbgr_30bpp = true;
+	dev->mode_config.normalize_zpos = true;
 
 	/* small shared memory area we use for notifiers and semaphores */
 	ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM,
diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.c b/drivers/gpu/drm/nouveau/dispnv50/head.c
index 929d93b1677e..71c23bf1fe25 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/head.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/head.c
@@ -480,7 +480,7 @@ nv50_head_create(struct drm_device *dev, int index)
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nv50_disp *disp = nv50_disp(dev);
 	struct nv50_head *head;
-	struct nv50_wndw *curs, *wndw;
+	struct nv50_wndw *base, *ovly, *curs;
 	struct drm_crtc *crtc;
 	int ret;
 
@@ -492,13 +492,13 @@ nv50_head_create(struct drm_device *dev, int index)
 	head->base.index = index;
 
 	if (disp->disp->object.oclass < GV100_DISP) {
-		ret = nv50_ovly_new(drm, head->base.index, &wndw);
-		ret = nv50_base_new(drm, head->base.index, &wndw);
+		ret = nv50_base_new(drm, head->base.index, &base);
+		ret = nv50_ovly_new(drm, head->base.index, &ovly);
 	} else {
-		ret = nv50_wndw_new(drm, DRM_PLANE_TYPE_OVERLAY,
-				    head->base.index * 2 + 1, &wndw);
 		ret = nv50_wndw_new(drm, DRM_PLANE_TYPE_PRIMARY,
-				    head->base.index * 2 + 0, &wndw);
+				    head->base.index * 2 + 0, &base);
+		ret = nv50_wndw_new(drm, DRM_PLANE_TYPE_OVERLAY,
+				    head->base.index * 2 + 1, &ovly);
 	}
 	if (ret == 0)
 		ret = nv50_curs_new(drm, head->base.index, &curs);
@@ -508,10 +508,14 @@ nv50_head_create(struct drm_device *dev, int index)
 	}
 
 	crtc = &head->base.base;
-	drm_crtc_init_with_planes(dev, crtc, &wndw->plane, &curs->plane,
+	drm_crtc_init_with_planes(dev, crtc, &base->plane, &curs->plane,
 				  &nv50_head_func, "head-%d", head->base.index);
 	drm_crtc_helper_add(crtc, &nv50_head_help);
 	drm_mode_crtc_set_gamma_size(crtc, 256);
+	if (disp->disp->object.oclass >= GF110_DISP)
+		drm_crtc_enable_color_mgmt(crtc, 256, true, 256);
+	else
+		drm_crtc_enable_color_mgmt(crtc, 0, false, 256);
 
 	if (head->func->olut_set) {
 		ret = nv50_lut_init(disp, &drm->client.mmu, &head->olut);
diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly507e.c b/drivers/gpu/drm/nouveau/dispnv50/ovly507e.c
index cc417664f823..8ccd96113bad 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/ovly507e.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/ovly507e.c
@@ -23,6 +23,7 @@
 #include "atom.h"
 
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_plane_helper.h>
 
 #include <nvif/cl507e.h>
@@ -160,9 +161,7 @@ ovly507e_format[] = {
 	DRM_FORMAT_YUYV,
 	DRM_FORMAT_UYVY,
 	DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_ARGB8888,
 	DRM_FORMAT_XRGB1555,
-	DRM_FORMAT_ARGB1555,
 	0
 };
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c b/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c
index aaa9fe5a4fc8..2e68fc736fe1 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/ovly827e.c
@@ -90,11 +90,8 @@ ovly827e_format[] = {
 	DRM_FORMAT_YUYV,
 	DRM_FORMAT_UYVY,
 	DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_ARGB8888,
 	DRM_FORMAT_XRGB1555,
-	DRM_FORMAT_ARGB1555,
 	DRM_FORMAT_XBGR2101010,
-	DRM_FORMAT_ABGR2101010,
 	0
 };
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly907e.c b/drivers/gpu/drm/nouveau/dispnv50/ovly907e.c
index a3ce53046015..9efe5e9d5ce4 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/ovly907e.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/ovly907e.c
@@ -61,10 +61,21 @@ ovly907e = {
 	.update = ovly507e_update,
 };
 
+static const u32
+ovly907e_format[] = {
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_XBGR2101010,
+	DRM_FORMAT_XBGR16161616F,
+	0
+};
+
 int
 ovly907e_new(struct nouveau_drm *drm, int head, s32 oclass,
 	     struct nv50_wndw **pwndw)
 {
-	return ovly507e_new_(&ovly907e, ovly827e_format, drm, head, oclass,
+	return ovly507e_new_(&ovly907e, ovly907e_format, drm, head, oclass,
 			     0x00000004 << (head * 4), pwndw);
 }
diff --git a/drivers/gpu/drm/nouveau/dispnv50/ovly917e.c b/drivers/gpu/drm/nouveau/dispnv50/ovly917e.c
index 505fa7e78523..e24d6fd23450 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/ovly917e.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/ovly917e.c
@@ -26,13 +26,10 @@ ovly917e_format[] = {
 	DRM_FORMAT_YUYV,
 	DRM_FORMAT_UYVY,
 	DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_ARGB8888,
 	DRM_FORMAT_XRGB1555,
-	DRM_FORMAT_ARGB1555,
 	DRM_FORMAT_XBGR2101010,
-	DRM_FORMAT_ABGR2101010,
 	DRM_FORMAT_XRGB2101010,
-	DRM_FORMAT_ARGB2101010,
+	DRM_FORMAT_XBGR16161616F,
 	0
 };
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.c b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
index 283ff690350e..2db029371c91 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndw.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
@@ -26,6 +26,8 @@
 #include <nvif/cl0002.h>
 
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_fourcc.h>
+
 #include "nouveau_bo.h"
 
 static void
@@ -118,6 +120,7 @@ nv50_wndw_flush_clr(struct nv50_wndw *wndw, u32 *interlock, bool flush,
 	if (clr.sema ) wndw->func-> sema_clr(wndw);
 	if (clr.ntfy ) wndw->func-> ntfy_clr(wndw);
 	if (clr.xlut ) wndw->func-> xlut_clr(wndw);
+	if (clr.csc  ) wndw->func->  csc_clr(wndw);
 	if (clr.image) wndw->func->image_clr(wndw);
 
 	interlock[wndw->interlock.type] |= wndw->interlock.data;
@@ -145,7 +148,9 @@ nv50_wndw_flush_set(struct nv50_wndw *wndw, u32 *interlock,
 		wndw->func->xlut_set(wndw, asyw);
 	}
 
+	if (asyw->set.csc  ) wndw->func->csc_set  (wndw, asyw);
 	if (asyw->set.scale) wndw->func->scale_set(wndw, asyw);
+	if (asyw->set.blend) wndw->func->blend_set(wndw, asyw);
 	if (asyw->set.point) {
 		if (asyw->set.point = false, asyw->set.mask)
 			interlock[wndw->interlock.type] |= wndw->interlock.data;
@@ -202,18 +207,20 @@ static int
 nv50_wndw_atomic_check_acquire_rgb(struct nv50_wndw_atom *asyw)
 {
 	switch (asyw->state.fb->format->format) {
-	case DRM_FORMAT_C8         : asyw->image.format = 0x1e; break;
-	case DRM_FORMAT_XRGB8888   :
-	case DRM_FORMAT_ARGB8888   : asyw->image.format = 0xcf; break;
-	case DRM_FORMAT_RGB565     : asyw->image.format = 0xe8; break;
-	case DRM_FORMAT_XRGB1555   :
-	case DRM_FORMAT_ARGB1555   : asyw->image.format = 0xe9; break;
-	case DRM_FORMAT_XBGR2101010:
-	case DRM_FORMAT_ABGR2101010: asyw->image.format = 0xd1; break;
-	case DRM_FORMAT_XBGR8888   :
-	case DRM_FORMAT_ABGR8888   : asyw->image.format = 0xd5; break;
-	case DRM_FORMAT_XRGB2101010:
-	case DRM_FORMAT_ARGB2101010: asyw->image.format = 0xdf; break;
+	case DRM_FORMAT_C8           : asyw->image.format = 0x1e; break;
+	case DRM_FORMAT_XRGB8888     :
+	case DRM_FORMAT_ARGB8888     : asyw->image.format = 0xcf; break;
+	case DRM_FORMAT_RGB565       : asyw->image.format = 0xe8; break;
+	case DRM_FORMAT_XRGB1555     :
+	case DRM_FORMAT_ARGB1555     : asyw->image.format = 0xe9; break;
+	case DRM_FORMAT_XBGR2101010  :
+	case DRM_FORMAT_ABGR2101010  : asyw->image.format = 0xd1; break;
+	case DRM_FORMAT_XBGR8888     :
+	case DRM_FORMAT_ABGR8888     : asyw->image.format = 0xd5; break;
+	case DRM_FORMAT_XRGB2101010  :
+	case DRM_FORMAT_ARGB2101010  : asyw->image.format = 0xdf; break;
+	case DRM_FORMAT_XBGR16161616F:
+	case DRM_FORMAT_ABGR16161616F: asyw->image.format = 0xca; break;
 	default:
 		return -EINVAL;
 	}
@@ -279,6 +286,28 @@ nv50_wndw_atomic_check_acquire(struct nv50_wndw *wndw, bool modeset,
 			asyw->set.scale = true;
 	}
 
+	if (wndw->func->blend_set) {
+		asyw->blend.depth = 255 - asyw->state.normalized_zpos;
+		asyw->blend.k1 = asyw->state.alpha >> 8;
+		switch (asyw->state.pixel_blend_mode) {
+		case DRM_MODE_BLEND_PREMULTI:
+			asyw->blend.src_color = 2; /* K1 */
+			asyw->blend.dst_color = 7; /* NEG_K1_TIMES_SRC */
+			break;
+		case DRM_MODE_BLEND_COVERAGE:
+			asyw->blend.src_color = 5; /* K1_TIMES_SRC */
+			asyw->blend.dst_color = 7; /* NEG_K1_TIMES_SRC */
+			break;
+		case DRM_MODE_BLEND_PIXEL_NONE:
+		default:
+			asyw->blend.src_color = 2; /* K1 */
+			asyw->blend.dst_color = 4; /* NEG_K1 */
+			break;
+		}
+		if (memcmp(&armw->blend, &asyw->blend, sizeof(asyw->blend)))
+			asyw->set.blend = true;
+	}
+
 	if (wndw->immd) {
 		asyw->point.x = asyw->state.crtc_x;
 		asyw->point.y = asyw->state.crtc_y;
@@ -320,7 +349,9 @@ nv50_wndw_atomic_check_lut(struct nv50_wndw *wndw,
 		asyh->wndw.olut &= ~BIT(wndw->id);
 	}
 
-	if (!ilut && wndw->func->ilut_identity) {
+	if (!ilut && wndw->func->ilut_identity &&
+	    asyw->state.fb->format->format != DRM_FORMAT_XBGR16161616F &&
+	    asyw->state.fb->format->format != DRM_FORMAT_ABGR16161616F) {
 		static struct drm_property_blob dummy = {};
 		ilut = &dummy;
 	}
@@ -332,6 +363,8 @@ nv50_wndw_atomic_check_lut(struct nv50_wndw *wndw,
 		asyw->xlut.handle = wndw->wndw.vram.handle;
 		asyw->xlut.i.buffer = !asyw->xlut.i.buffer;
 		asyw->set.xlut = true;
+	} else {
+		asyw->clr.xlut = armw->xlut.handle != 0;
 	}
 
 	/* Handle setting base SET_OUTPUT_LUT_LO_ENABLE_USE_CORE_LUT. */
@@ -339,6 +372,16 @@ nv50_wndw_atomic_check_lut(struct nv50_wndw *wndw,
 	    (!armw->visible || (armw->xlut.handle && !asyw->xlut.handle)))
 		asyw->set.xlut = true;
 
+	if (wndw->func->csc && asyh->state.ctm) {
+		const struct drm_color_ctm *ctm = asyh->state.ctm->data;
+		wndw->func->csc(wndw, asyw, ctm);
+		asyw->csc.valid = true;
+		asyw->set.csc = true;
+	} else {
+		asyw->csc.valid = false;
+		asyw->clr.csc = armw->csc.valid;
+	}
+
 	/* Can't do an immediate flip while changing the LUT. */
 	asyh->state.pageflip_flags &= ~DRM_MODE_PAGE_FLIP_ASYNC;
 }
@@ -408,6 +451,7 @@ nv50_wndw_atomic_check(struct drm_plane *plane, struct drm_plane_state *state)
 		asyw->clr.ntfy = armw->ntfy.handle != 0;
 		asyw->clr.sema = armw->sema.handle != 0;
 		asyw->clr.xlut = armw->xlut.handle != 0;
+		asyw->clr.csc  = armw->csc.valid;
 		if (wndw->func->image_clr)
 			asyw->clr.image = armw->image.handle[0] != 0;
 	}
@@ -457,7 +501,7 @@ nv50_wndw_prepare_fb(struct drm_plane *plane, struct drm_plane_state *state)
 		asyw->image.handle[0] = ctxdma->object.handle;
 	}
 
-	asyw->state.fence = reservation_object_get_excl_rcu(fb->nvbo->bo.resv);
+	asyw->state.fence = dma_resv_get_excl_rcu(fb->nvbo->bo.base.resv);
 	asyw->image.offset[0] = fb->nvbo->bo.offset;
 
 	if (wndw->func->prepare) {
@@ -499,6 +543,7 @@ nv50_wndw_atomic_duplicate_state(struct drm_plane *plane)
 	asyw->ntfy = armw->ntfy;
 	asyw->ilut = NULL;
 	asyw->xlut = armw->xlut;
+	asyw->csc  = armw->csc;
 	asyw->image = armw->image;
 	asyw->point = armw->point;
 	asyw->clr.mask = 0;
@@ -506,6 +551,13 @@ nv50_wndw_atomic_duplicate_state(struct drm_plane *plane)
 	return &asyw->state;
 }
 
+static int
+nv50_wndw_zpos_default(struct drm_plane *plane)
+{
+	return (plane->type == DRM_PLANE_TYPE_PRIMARY) ? 0 :
+	       (plane->type == DRM_PLANE_TYPE_OVERLAY) ? 1 : 255;
+}
+
 static void
 nv50_wndw_reset(struct drm_plane *plane)
 {
@@ -516,9 +568,10 @@ nv50_wndw_reset(struct drm_plane *plane)
 
 	if (plane->state)
 		plane->funcs->atomic_destroy_state(plane, plane->state);
-	plane->state = &asyw->state;
-	plane->state->plane = plane;
-	plane->state->rotation = DRM_MODE_ROTATE_0;
+
+	__drm_atomic_helper_plane_reset(plane, &asyw->state);
+	plane->state->zpos = nv50_wndw_zpos_default(plane);
+	plane->state->normalized_zpos = nv50_wndw_zpos_default(plane);
 }
 
 static void
@@ -613,6 +666,30 @@ nv50_wndw_new_(const struct nv50_wndw_func *func, struct drm_device *dev,
 	}
 
 	wndw->notify.func = nv50_wndw_notify;
+
+	if (wndw->func->blend_set) {
+		ret = drm_plane_create_zpos_property(&wndw->plane,
+				nv50_wndw_zpos_default(&wndw->plane), 0, 254);
+		if (ret)
+			return ret;
+
+		ret = drm_plane_create_alpha_property(&wndw->plane);
+		if (ret)
+			return ret;
+
+		ret = drm_plane_create_blend_mode_property(&wndw->plane,
+				BIT(DRM_MODE_BLEND_PIXEL_NONE) |
+				BIT(DRM_MODE_BLEND_PREMULTI) |
+				BIT(DRM_MODE_BLEND_COVERAGE));
+		if (ret)
+			return ret;
+	} else {
+		ret = drm_plane_create_zpos_immutable_property(&wndw->plane,
+				nv50_wndw_zpos_default(&wndw->plane));
+		if (ret)
+			return ret;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.h b/drivers/gpu/drm/nouveau/dispnv50/wndw.h
index 03f3d8dc235a..c63bd3bdaf06 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndw.h
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.h
@@ -65,6 +65,10 @@ struct nv50_wndw_func {
 	int (*ntfy_wait_begun)(struct nouveau_bo *, u32 offset,
 			       struct nvif_device *);
 	void (*ilut)(struct nv50_wndw *, struct nv50_wndw_atom *);
+	void (*csc)(struct nv50_wndw *, struct nv50_wndw_atom *,
+		    const struct drm_color_ctm *);
+	void (*csc_set)(struct nv50_wndw *, struct nv50_wndw_atom *);
+	void (*csc_clr)(struct nv50_wndw *);
 	bool ilut_identity;
 	bool olut_core;
 	void (*xlut_set)(struct nv50_wndw *, struct nv50_wndw_atom *);
@@ -72,6 +76,7 @@ struct nv50_wndw_func {
 	void (*image_set)(struct nv50_wndw *, struct nv50_wndw_atom *);
 	void (*image_clr)(struct nv50_wndw *);
 	void (*scale_set)(struct nv50_wndw *, struct nv50_wndw_atom *);
+	void (*blend_set)(struct nv50_wndw *, struct nv50_wndw_atom *);
 
 	void (*update)(struct nv50_wndw *, u32 *interlock);
 };
@@ -81,6 +86,9 @@ extern const struct drm_plane_funcs nv50_wndw;
 void base507c_ntfy_reset(struct nouveau_bo *, u32);
 int base507c_ntfy_wait_begun(struct nouveau_bo *, u32, struct nvif_device *);
 
+void base907c_csc(struct nv50_wndw *, struct nv50_wndw_atom *,
+		  const struct drm_color_ctm *);
+
 struct nv50_wimm_func {
 	void (*point)(struct nv50_wndw *, struct nv50_wndw_atom *);
 
@@ -102,8 +110,8 @@ void wndwc37e_sema_set(struct nv50_wndw *, struct nv50_wndw_atom *);
 void wndwc37e_sema_clr(struct nv50_wndw *);
 void wndwc37e_ntfy_set(struct nv50_wndw *, struct nv50_wndw_atom *);
 void wndwc37e_ntfy_clr(struct nv50_wndw *);
-void wndwc37e_image_set(struct nv50_wndw *, struct nv50_wndw_atom *);
 void wndwc37e_image_clr(struct nv50_wndw *);
+void wndwc37e_blend_set(struct nv50_wndw *, struct nv50_wndw_atom *);
 void wndwc37e_update(struct nv50_wndw *, u32 *);
 
 int wndwc57e_new(struct nouveau_drm *, enum drm_plane_type, int, s32,
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c b/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c
index e52a85c83f7a..0f9402162bde 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c
@@ -29,6 +29,23 @@
 #include <nvif/clc37e.h>
 
 static void
+wndwc37e_csc_clr(struct nv50_wndw *wndw)
+{
+}
+
+static void
+wndwc37e_csc_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
+{
+	u32 *push, i;
+	if ((push = evo_wait(&wndw->wndw, 13))) {
+		 evo_mthd(push, 0x02bc, 12);
+		 for (i = 0; i < 12; i++)
+			  evo_data(push, asyw->csc.matrix[i]);
+		 evo_kick(push, &wndw->wndw);
+	}
+}
+
+static void
 wndwc37e_ilut_clr(struct nv50_wndw *wndw)
 {
 	u32 *push;
@@ -65,6 +82,26 @@ wndwc37e_ilut(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
 }
 
 void
+wndwc37e_blend_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
+{
+	u32 *push;
+	if ((push = evo_wait(&wndw->wndw, 8))) {
+		evo_mthd(push, 0x02ec, 7);
+		evo_data(push, asyw->blend.depth << 4);
+		evo_data(push, asyw->blend.k1);
+		evo_data(push, asyw->blend.dst_color << 12 |
+			       asyw->blend.dst_color << 8 |
+			       asyw->blend.src_color << 4 |
+			       asyw->blend.src_color);
+		evo_data(push, 0xffff0000);
+		evo_data(push, 0xffff0000);
+		evo_data(push, 0xffff0000);
+		evo_data(push, 0xffff0000);
+		evo_kick(push, &wndw->wndw);
+	}
+}
+
+void
 wndwc37e_image_clr(struct nv50_wndw *wndw)
 {
 	u32 *push;
@@ -77,12 +114,12 @@ wndwc37e_image_clr(struct nv50_wndw *wndw)
 	}
 }
 
-void
+static void
 wndwc37e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
 {
 	u32 *push;
 
-	if (!(push = evo_wait(&wndw->wndw, 25)))
+	if (!(push = evo_wait(&wndw->wndw, 17)))
 		return;
 
 	evo_mthd(push, 0x0308, 1);
@@ -90,7 +127,9 @@ wndwc37e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
 	evo_mthd(push, 0x0224, 4);
 	evo_data(push, asyw->image.h << 16 | asyw->image.w);
 	evo_data(push, asyw->image.layout << 4 | asyw->image.blockh);
-	evo_data(push, asyw->image.colorspace << 8 | asyw->image.format);
+	evo_data(push, asyw->csc.valid << 17 |
+		       asyw->image.colorspace << 8 |
+		       asyw->image.format);
 	evo_data(push, asyw->image.blocks[0] | (asyw->image.pitch[0] >> 6));
 	evo_mthd(push, 0x0240, 1);
 	evo_data(push, asyw->image.handle[0]);
@@ -105,16 +144,6 @@ wndwc37e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
 	evo_mthd(push, 0x02a4, 1);
 	evo_data(push, asyw->state.crtc_h << 16 |
 		       asyw->state.crtc_w);
-
-	/*XXX: Composition-related stuff.  Need to implement properly. */
-	evo_mthd(push, 0x02ec, 1);
-	evo_data(push, (2 - (wndw->id & 1)) << 4);
-	evo_mthd(push, 0x02f4, 5);
-	evo_data(push, 0x00000011);
-	evo_data(push, 0xffff0000);
-	evo_data(push, 0xffff0000);
-	evo_data(push, 0xffff0000);
-	evo_data(push, 0xffff0000);
 	evo_kick(push, &wndw->wndw);
 }
 
@@ -216,6 +245,8 @@ wndwc37e_format[] = {
 	DRM_FORMAT_ABGR8888,
 	DRM_FORMAT_XRGB2101010,
 	DRM_FORMAT_ARGB2101010,
+	DRM_FORMAT_XBGR16161616F,
+	DRM_FORMAT_ABGR16161616F,
 	0
 };
 
@@ -232,8 +263,12 @@ wndwc37e = {
 	.ilut = wndwc37e_ilut,
 	.xlut_set = wndwc37e_ilut_set,
 	.xlut_clr = wndwc37e_ilut_clr,
+	.csc = base907c_csc,
+	.csc_set = wndwc37e_csc_set,
+	.csc_clr = wndwc37e_csc_clr,
 	.image_set = wndwc37e_image_set,
 	.image_clr = wndwc37e_image_clr,
+	.blend_set = wndwc37e_blend_set,
 	.update = wndwc37e_update,
 };
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndwc57e.c b/drivers/gpu/drm/nouveau/dispnv50/wndwc57e.c
index ba89f1a5fcfa..a311c79e5295 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndwc57e.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndwc57e.c
@@ -29,6 +29,72 @@
 #include <nvif/clc37e.h>
 
 static void
+wndwc57e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
+{
+	u32 *push;
+
+	if (!(push = evo_wait(&wndw->wndw, 17)))
+		return;
+
+	evo_mthd(push, 0x0308, 1);
+	evo_data(push, asyw->image.mode << 4 | asyw->image.interval);
+	evo_mthd(push, 0x0224, 4);
+	evo_data(push, asyw->image.h << 16 | asyw->image.w);
+	evo_data(push, asyw->image.layout << 4 | asyw->image.blockh);
+	evo_data(push, asyw->image.colorspace << 8 |
+		       asyw->image.format);
+	evo_data(push, asyw->image.blocks[0] | (asyw->image.pitch[0] >> 6));
+	evo_mthd(push, 0x0240, 1);
+	evo_data(push, asyw->image.handle[0]);
+	evo_mthd(push, 0x0260, 1);
+	evo_data(push, asyw->image.offset[0] >> 8);
+	evo_mthd(push, 0x0290, 1);
+	evo_data(push, (asyw->state.src_y >> 16) << 16 |
+		       (asyw->state.src_x >> 16));
+	evo_mthd(push, 0x0298, 1);
+	evo_data(push, (asyw->state.src_h >> 16) << 16 |
+		       (asyw->state.src_w >> 16));
+	evo_mthd(push, 0x02a4, 1);
+	evo_data(push, asyw->state.crtc_h << 16 |
+		       asyw->state.crtc_w);
+	evo_kick(push, &wndw->wndw);
+}
+
+static void
+wndwc57e_csc_clr(struct nv50_wndw *wndw)
+{
+	u32 *push;
+	if ((push = evo_wait(&wndw->wndw, 13))) {
+		 evo_mthd(push, 0x0400, 12);
+		 evo_data(push, 0x00010000);
+		 evo_data(push, 0x00000000);
+		 evo_data(push, 0x00000000);
+		 evo_data(push, 0x00000000);
+		 evo_data(push, 0x00000000);
+		 evo_data(push, 0x00010000);
+		 evo_data(push, 0x00000000);
+		 evo_data(push, 0x00000000);
+		 evo_data(push, 0x00000000);
+		 evo_data(push, 0x00000000);
+		 evo_data(push, 0x00010000);
+		 evo_data(push, 0x00000000);
+		 evo_kick(push, &wndw->wndw);
+	}
+}
+
+static void
+wndwc57e_csc_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
+{
+	u32 *push, i;
+	if ((push = evo_wait(&wndw->wndw, 13))) {
+		 evo_mthd(push, 0x0400, 12);
+		 for (i = 0; i < 12; i++)
+			  evo_data(push, asyw->csc.matrix[i]);
+		 evo_kick(push, &wndw->wndw);
+	}
+}
+
+static void
 wndwc57e_ilut_clr(struct nv50_wndw *wndw)
 {
 	u32 *push;
@@ -119,8 +185,12 @@ wndwc57e = {
 	.ilut_identity = true,
 	.xlut_set = wndwc57e_ilut_set,
 	.xlut_clr = wndwc57e_ilut_clr,
-	.image_set = wndwc37e_image_set,
+	.csc = base907c_csc,
+	.csc_set = wndwc57e_csc_set,
+	.csc_clr = wndwc57e_csc_clr,
+	.image_set = wndwc57e_image_set,
 	.image_clr = wndwc37e_image_clr,
+	.blend_set = wndwc37e_blend_set,
 	.update = wndwc37e_update,
 };