Merge branch 'linux-4.16' of git://github.com/skeggsb/linux into drm-next

- initial kepler clock gating support
- atomic gamma handling fixes
- support for gp108 "secure boot" (enables acceleration, finally)

* 'linux-4.16' of git://github.com/skeggsb/linux:
  drm/nouveau/clk: fix gcc-7 -Wint-in-bool-context warning
  drm/nouveau/mmu: Fix trailing semicolon
  drm/nouveau: Introduce NvPmEnableGating option
  drm/nouveau: Add support for SLCG for Kepler2
  drm/nouveau: Add support for BLCG on Kepler2
  drm/nouveau: Add support for BLCG on Kepler1
  drm/nouveau: Add support for basic clockgating on Kepler1
  drm/nouveau/kms/nv50: fix handling of gamma since atomic conversion
  drm/nouveau/kms/nv50: use INTERPOLATE_257_UNITY_RANGE LUT on newer chipsets
  drm/nouveau/kms/nv50: use "low res" lut for indexed mode
  drm/nouveau/kms/nv50: prepare for double-buffered LUTs
  drm/nouveau/bo: add helper functions for handling pinned+mapped buffers
  drm/nouveau/fbcon: add module parameter to select bits-per-pixel
  drm/nouveau/secboot/gp108: implement on top of acr_r370
  drm/nouveau/secboot/r370: implement support for booting LS SEC2 ucode
  drm/nouveau/secboot/r370: move a bunch of r375 stuff to a new implementation
  drm/nouveau: nouveau: use correct string length
  drm/nouveau/drm/nouveau/mmu: fix odd_ptr_err.cocci warnings
  drm/nouveau/pmu/fuc: don't use movw directly anymore

44 files changed, 2723 insertions, 1491 deletions

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h
index adb78f7d083a..92be0e5269c6 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h
@@ -75,6 +75,7 @@ int mcp89_fb_new(struct nvkm_device *, int, struct nvkm_fb **);
 int gf100_fb_new(struct nvkm_device *, int, struct nvkm_fb **);
 int gf108_fb_new(struct nvkm_device *, int, struct nvkm_fb **);
 int gk104_fb_new(struct nvkm_device *, int, struct nvkm_fb **);
+int gk110_fb_new(struct nvkm_device *, int, struct nvkm_fb **);
 int gk20a_fb_new(struct nvkm_device *, int, struct nvkm_fb **);
 int gm107_fb_new(struct nvkm_device *, int, struct nvkm_fb **);
 int gm200_fb_new(struct nvkm_device *, int, struct nvkm_fb **);
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/secboot.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/secboot.h
index 59f3ba551681..b57fe4ae93ba 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/secboot.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/secboot.h
@@ -60,6 +60,7 @@ int nvkm_secboot_reset(struct nvkm_secboot *, unsigned long);
 int gm200_secboot_new(struct nvkm_device *, int, struct nvkm_secboot **);
 int gm20b_secboot_new(struct nvkm_device *, int, struct nvkm_secboot **);
 int gp102_secboot_new(struct nvkm_device *, int, struct nvkm_secboot **);
+int gp108_secboot_new(struct nvkm_device *, int, struct nvkm_secboot **);
 int gp10b_secboot_new(struct nvkm_device *, int, struct nvkm_secboot **);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/therm.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/therm.h
index b1ac47eb786e..9398d9f09339 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/therm.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/therm.h
@@ -46,6 +46,16 @@ enum nvkm_therm_attr_type {
 	NVKM_THERM_ATTR_THRS_SHUTDOWN_HYST = 17,
 };
 
+struct nvkm_therm_clkgate_init {
+	u32 addr;
+	u8  count;
+	u32 data;
+};
+
+struct nvkm_therm_clkgate_pack {
+	const struct nvkm_therm_clkgate_init *init;
+};
+
 struct nvkm_therm {
 	const struct nvkm_therm_func *func;
 	struct nvkm_subdev subdev;
@@ -85,17 +95,24 @@ struct nvkm_therm {
 
 	int (*attr_get)(struct nvkm_therm *, enum nvkm_therm_attr_type);
 	int (*attr_set)(struct nvkm_therm *, enum nvkm_therm_attr_type, int);
+
+	bool clkgating_enabled;
 };
 
 int nvkm_therm_temp_get(struct nvkm_therm *);
 int nvkm_therm_fan_sense(struct nvkm_therm *);
 int nvkm_therm_cstate(struct nvkm_therm *, int, int);
+void nvkm_therm_clkgate_init(struct nvkm_therm *,
+			     const struct nvkm_therm_clkgate_pack *);
+void nvkm_therm_clkgate_enable(struct nvkm_therm *);
+void nvkm_therm_clkgate_fini(struct nvkm_therm *, bool);
 
 int nv40_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int nv50_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int g84_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int gt215_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int gf119_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
+int gk104_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int gm107_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int gm200_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
 int gp100_therm_new(struct nvkm_device *, int, struct nvkm_therm **);
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.h b/drivers/gpu/drm/nouveau/nouveau_bo.h
index 7b5cc5c73d20..be8e00b49cde 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.h
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.h
@@ -105,4 +105,32 @@ nvbo_kmap_obj_iovirtual(struct nouveau_bo *nvbo)
 	return ioptr;
 }
 
+static inline void
+nouveau_bo_unmap_unpin_unref(struct nouveau_bo **pnvbo)
+{
+	if (*pnvbo) {
+		nouveau_bo_unmap(*pnvbo);
+		nouveau_bo_unpin(*pnvbo);
+		nouveau_bo_ref(NULL, pnvbo);
+	}
+}
+
+static inline int
+nouveau_bo_new_pin_map(struct nouveau_cli *cli, u64 size, int align, u32 flags,
+		       struct nouveau_bo **pnvbo)
+{
+	int ret = nouveau_bo_new(cli, size, align, flags,
+				 0, 0, NULL, NULL, pnvbo);
+	if (ret == 0) {
+		ret = nouveau_bo_pin(*pnvbo, flags, true);
+		if (ret == 0) {
+			ret = nouveau_bo_map(*pnvbo);
+			if (ret == 0)
+				return ret;
+			nouveau_bo_unpin(*pnvbo);
+		}
+		nouveau_bo_ref(NULL, pnvbo);
+	}
+	return ret;
+}
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_crtc.h b/drivers/gpu/drm/nouveau/nouveau_crtc.h
index b7a18fbee6dc..366acb928f57 100644
--- a/drivers/gpu/drm/nouveau/nouveau_crtc.h
+++ b/drivers/gpu/drm/nouveau/nouveau_crtc.h
@@ -60,7 +60,6 @@ struct nouveau_crtc {
 	} cursor;
 
 	struct {
-		struct nouveau_bo *nvbo;
 		int depth;
 	} lut;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index ee5d1dc2eaf5..85c1f10bc2b6 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -56,6 +56,10 @@ MODULE_PARM_DESC(nofbaccel, "Disable fbcon acceleration");
 int nouveau_nofbaccel = 0;
 module_param_named(nofbaccel, nouveau_nofbaccel, int, 0400);
 
+MODULE_PARM_DESC(fbcon_bpp, "fbcon bits-per-pixel (default: auto)");
+static int nouveau_fbcon_bpp;
+module_param_named(fbcon_bpp, nouveau_fbcon_bpp, int, 0400);
+
 static void
 nouveau_fbcon_fillrect(struct fb_info *info, const struct fb_fillrect *rect)
 {
@@ -488,7 +492,7 @@ nouveau_fbcon_init(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_fbdev *fbcon;
-	int preferred_bpp;
+	int preferred_bpp = nouveau_fbcon_bpp;
 	int ret;
 
 	if (!dev->mode_config.num_crtc ||
@@ -512,13 +516,15 @@ nouveau_fbcon_init(struct drm_device *dev)
 	if (ret)
 		goto fini;
 
-	if (drm->client.device.info.ram_size <= 32 * 1024 * 1024)
-		preferred_bpp = 8;
-	else
-	if (drm->client.device.info.ram_size <= 64 * 1024 * 1024)
-		preferred_bpp = 16;
-	else
-		preferred_bpp = 32;
+	if (preferred_bpp != 8 && preferred_bpp != 16 && preferred_bpp != 32) {
+		if (drm->client.device.info.ram_size <= 32 * 1024 * 1024)
+			preferred_bpp = 8;
+		else
+		if (drm->client.device.info.ram_size <= 64 * 1024 * 1024)
+			preferred_bpp = 16;
+		else
+			preferred_bpp = 32;
+	}
 
 	/* disable all the possible outputs/crtcs before entering KMS mode */
 	if (!drm_drv_uses_atomic_modeset(dev))
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index b22c37bde13f..dd8d4352ed99 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -137,8 +137,10 @@ struct nv50_head_atom {
 	} mode;
 
 	struct {
+		bool visible;
 		u32 handle;
 		u64 offset:40;
+		u8  mode:4;
 	} lut;
 
 	struct {
@@ -192,6 +194,7 @@ struct nv50_head_atom {
 
 	union {
 		struct {
+			bool ilut:1;
 			bool core:1;
 			bool curs:1;
 		};
@@ -200,6 +203,7 @@ struct nv50_head_atom {
 
 	union {
 		struct {
+			bool ilut:1;
 			bool core:1;
 			bool curs:1;
 			bool view:1;
@@ -660,6 +664,10 @@ nv50_ovly_create(struct nvif_device *device, struct nvif_object *disp,
 
 struct nv50_head {
 	struct nouveau_crtc base;
+	struct {
+		struct nouveau_bo *nvbo[2];
+		int next;
+	} lut;
 	struct nv50_ovly ovly;
 	struct nv50_oimm oimm;
 };
@@ -1795,6 +1803,54 @@ nv50_head_lut_clr(struct nv50_head *head)
 }
 
 static void
+nv50_head_lut_load(struct drm_property_blob *blob, int mode,
+		   struct nouveau_bo *nvbo)
+{
+	struct drm_color_lut *in = (struct drm_color_lut *)blob->data;
+	void __iomem *lut = (u8 *)nvbo_kmap_obj_iovirtual(nvbo);
+	const int size = blob->length / sizeof(*in);
+	int bits, shift, i;
+	u16 zero, r, g, b;
+
+	/* This can't happen.. But it shuts the compiler up. */
+	if (WARN_ON(size != 256))
+		return;
+
+	switch (mode) {
+	case 0: /* LORES. */
+	case 1: /* HIRES. */
+		bits = 11;
+		shift = 3;
+		zero = 0x0000;
+		break;
+	case 7: /* INTERPOLATE_257_UNITY_RANGE. */
+		bits = 14;
+		shift = 0;
+		zero = 0x6000;
+		break;
+	default:
+		WARN_ON(1);
+		return;
+	}
+
+	for (i = 0; i < size; i++) {
+		r = (drm_color_lut_extract(in[i].  red, bits) + zero) << shift;
+		g = (drm_color_lut_extract(in[i].green, bits) + zero) << shift;
+		b = (drm_color_lut_extract(in[i]. blue, bits) + zero) << shift;
+		writew(r, lut + (i * 0x08) + 0);
+		writew(g, lut + (i * 0x08) + 2);
+		writew(b, lut + (i * 0x08) + 4);
+	}
+
+	/* INTERPOLATE modes require a "next" entry to interpolate with,
+	 * so we replicate the last entry to deal with this for now.
+	 */
+	writew(r, lut + (i * 0x08) + 0);
+	writew(g, lut + (i * 0x08) + 2);
+	writew(b, lut + (i * 0x08) + 4);
+}
+
+static void
 nv50_head_lut_set(struct nv50_head *head, struct nv50_head_atom *asyh)
 {
 	struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->mast.base;
@@ -1802,18 +1858,18 @@ nv50_head_lut_set(struct nv50_head *head, struct nv50_head_atom *asyh)
 	if ((push = evo_wait(core, 7))) {
 		if (core->base.user.oclass < G82_DISP_CORE_CHANNEL_DMA) {
 			evo_mthd(push, 0x0840 + (head->base.index * 0x400), 2);
-			evo_data(push, 0xc0000000);
+			evo_data(push, 0x80000000 | asyh->lut.mode << 30);
 			evo_data(push, asyh->lut.offset >> 8);
 		} else
 		if (core->base.user.oclass < GF110_DISP_CORE_CHANNEL_DMA) {
 			evo_mthd(push, 0x0840 + (head->base.index * 0x400), 2);
-			evo_data(push, 0xc0000000);
+			evo_data(push, 0x80000000 | asyh->lut.mode << 30);
 			evo_data(push, asyh->lut.offset >> 8);
 			evo_mthd(push, 0x085c + (head->base.index * 0x400), 1);
 			evo_data(push, asyh->lut.handle);
 		} else {
 			evo_mthd(push, 0x0440 + (head->base.index * 0x300), 4);
-			evo_data(push, 0x83000000);
+			evo_data(push, 0x80000000 | asyh->lut.mode << 24);
 			evo_data(push, asyh->lut.offset >> 8);
 			evo_data(push, 0x00000000);
 			evo_data(push, 0x00000000);
@@ -1896,7 +1952,7 @@ nv50_head_view(struct nv50_head *head, struct nv50_head_atom *asyh)
 static void
 nv50_head_flush_clr(struct nv50_head *head, struct nv50_head_atom *asyh, bool y)
 {
-	if (asyh->clr.core && (!asyh->set.core || y))
+	if (asyh->clr.ilut && (!asyh->set.ilut || y))
 		nv50_head_lut_clr(head);
 	if (asyh->clr.core && (!asyh->set.core || y))
 		nv50_head_core_clr(head);
@@ -1909,7 +1965,15 @@ nv50_head_flush_set(struct nv50_head *head, struct nv50_head_atom *asyh)
 {
 	if (asyh->set.view   ) nv50_head_view    (head, asyh);
 	if (asyh->set.mode   ) nv50_head_mode    (head, asyh);
-	if (asyh->set.core   ) nv50_head_lut_set (head, asyh);
+	if (asyh->set.ilut   ) {
+		struct nouveau_bo *nvbo = head->lut.nvbo[head->lut.next];
+		struct drm_property_blob *blob = asyh->state.gamma_lut;
+		if (blob)
+			nv50_head_lut_load(blob, asyh->lut.mode, nvbo);
+		asyh->lut.offset = nvbo->bo.offset;
+		head->lut.next ^= 1;
+		nv50_head_lut_set(head, asyh);
+	}
 	if (asyh->set.core   ) nv50_head_core_set(head, asyh);
 	if (asyh->set.curs   ) nv50_head_curs_set(head, asyh);
 	if (asyh->set.base   ) nv50_head_base    (head, asyh);
@@ -2044,6 +2108,37 @@ nv50_head_atomic_check_view(struct nv50_head_atom *armh,
 }
 
 static void
+nv50_head_atomic_check_lut(struct nv50_head *head,
+			   struct nv50_head_atom *armh,
+			   struct nv50_head_atom *asyh)
+{
+	struct nv50_disp *disp = nv50_disp(head->base.base.dev);
+
+	/* An I8 surface without an input LUT makes no sense, and
+	 * EVO will throw an error if you try.
+	 *
+	 * Legacy clients actually cause this due to the order in
+	 * which they call ioctls, so we will enable the LUT with
+	 * whatever contents the buffer already contains to avoid
+	 * triggering the error check.
+	 */
+	if (!asyh->state.gamma_lut && asyh->base.cpp != 1) {
+		asyh->lut.handle = 0;
+		asyh->clr.ilut = armh->lut.visible;
+		return;
+	}
+
+	if (disp->disp->oclass < GF110_DISP) {
+		asyh->lut.mode = (asyh->base.cpp == 1) ? 0 : 1;
+		asyh->set.ilut = true;
+	} else {
+		asyh->lut.mode = 7;
+		asyh->set.ilut = asyh->state.color_mgmt_changed;
+	}
+	asyh->lut.handle = disp->mast.base.vram.handle;
+}
+
+static void
 nv50_head_atomic_check_mode(struct nv50_head *head, struct nv50_head_atom *asyh)
 {
 	struct drm_display_mode *mode = &asyh->state.adjusted_mode;
@@ -2128,6 +2223,11 @@ nv50_head_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state)
 		if (asyh->state.mode_changed)
 			nv50_head_atomic_check_mode(head, asyh);
 
+		if (asyh->state.color_mgmt_changed ||
+		    asyh->base.cpp != armh->base.cpp)
+			nv50_head_atomic_check_lut(head, armh, asyh);
+		asyh->lut.visible = asyh->lut.handle != 0;
+
 		if (asyc) {
 			if (asyc->set.scaler)
 				nv50_head_atomic_check_view(armh, asyh, asyc);
@@ -2143,7 +2243,8 @@ nv50_head_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state)
 			asyh->core.w = asyh->base.w;
 			asyh->core.h = asyh->base.h;
 		} else
-		if ((asyh->core.visible = asyh->curs.visible)) {
+		if ((asyh->core.visible = asyh->curs.visible) ||
+		    (asyh->core.visible = asyh->lut.visible)) {
 			/*XXX: We need to either find some way of having the
 			 *     primary base layer appear black, while still
 			 *     being able to display the other layers, or we
@@ -2161,11 +2262,10 @@ nv50_head_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state)
 		asyh->core.layout = 1;
 		asyh->core.block = 0;
 		asyh->core.pitch = ALIGN(asyh->core.w, 64) * 4;
-		asyh->lut.handle = disp->mast.base.vram.handle;
-		asyh->lut.offset = head->base.lut.nvbo->bo.offset;
 		asyh->set.base = armh->base.cpp != asyh->base.cpp;
 		asyh->set.ovly = armh->ovly.cpp != asyh->ovly.cpp;
 	} else {
+		asyh->lut.visible = false;
 		asyh->core.visible = false;
 		asyh->curs.visible = false;
 		asyh->base.cpp = 0;
@@ -2189,8 +2289,10 @@ nv50_head_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state)
 			asyh->clr.curs = true;
 		}
 	} else {
+		asyh->clr.ilut = armh->lut.visible;
 		asyh->clr.core = armh->core.visible;
 		asyh->clr.curs = armh->curs.visible;
+		asyh->set.ilut = asyh->lut.visible;
 		asyh->set.core = asyh->core.visible;
 		asyh->set.curs = asyh->curs.visible;
 	}
@@ -2200,47 +2302,11 @@ nv50_head_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state)
 	return 0;
 }
 
-static void
-nv50_head_lut_load(struct drm_crtc *crtc)
-{
-	struct nv50_disp *disp = nv50_disp(crtc->dev);
-	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-	void __iomem *lut = nvbo_kmap_obj_iovirtual(nv_crtc->lut.nvbo);
-	u16 *r, *g, *b;
-	int i;
-
-	r = crtc->gamma_store;
-	g = r + crtc->gamma_size;
-	b = g + crtc->gamma_size;
-
-	for (i = 0; i < 256; i++) {
-		if (disp->disp->oclass < GF110_DISP) {
-			writew((*r++ >> 2) + 0x0000, lut + (i * 0x08) + 0);
-			writew((*g++ >> 2) + 0x0000, lut + (i * 0x08) + 2);
-			writew((*b++ >> 2) + 0x0000, lut + (i * 0x08) + 4);
-		} else {
-			/* 0x6000 interferes with the 14-bit color??? */
-			writew((*r++ >> 2) + 0x6000, lut + (i * 0x20) + 0);
-			writew((*g++ >> 2) + 0x6000, lut + (i * 0x20) + 2);
-			writew((*b++ >> 2) + 0x6000, lut + (i * 0x20) + 4);
-		}
-	}
-}
-
 static const struct drm_crtc_helper_funcs
 nv50_head_help = {
 	.atomic_check = nv50_head_atomic_check,
 };
 
-static int
-nv50_head_gamma_set(struct drm_crtc *crtc, u16 *r, u16 *g, u16 *b,
-		    uint32_t size,
-		    struct drm_modeset_acquire_ctx *ctx)
-{
-	nv50_head_lut_load(crtc);
-	return 0;
-}
-
 static void
 nv50_head_atomic_destroy_state(struct drm_crtc *crtc,
 			       struct drm_crtc_state *state)
@@ -2296,17 +2362,15 @@ nv50_head_reset(struct drm_crtc *crtc)
 static void
 nv50_head_destroy(struct drm_crtc *crtc)
 {
-	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
 	struct nv50_disp *disp = nv50_disp(crtc->dev);
 	struct nv50_head *head = nv50_head(crtc);
+	int i;
 
 	nv50_dmac_destroy(&head->ovly.base, disp->disp);
 	nv50_pioc_destroy(&head->oimm.base);
 
-	nouveau_bo_unmap(nv_crtc->lut.nvbo);
-	if (nv_crtc->lut.nvbo)
-		nouveau_bo_unpin(nv_crtc->lut.nvbo);
-	nouveau_bo_ref(NULL, &nv_crtc->lut.nvbo);
+	for (i = 0; i < ARRAY_SIZE(head->lut.nvbo); i++)
+		nouveau_bo_unmap_unpin_unref(&head->lut.nvbo[i]);
 
 	drm_crtc_cleanup(crtc);
 	kfree(crtc);
@@ -2315,7 +2379,7 @@ nv50_head_destroy(struct drm_crtc *crtc)
 static const struct drm_crtc_funcs
 nv50_head_func = {
 	.reset = nv50_head_reset,
-	.gamma_set = nv50_head_gamma_set,
+	.gamma_set = drm_atomic_helper_legacy_gamma_set,
 	.destroy = nv50_head_destroy,
 	.set_config = drm_atomic_helper_set_config,
 	.page_flip = drm_atomic_helper_page_flip,
@@ -2333,7 +2397,7 @@ nv50_head_create(struct drm_device *dev, int index)
 	struct nv50_base *base;
 	struct nv50_curs *curs;
 	struct drm_crtc *crtc;
-	int ret;
+	int ret, i;
 
 	head = kzalloc(sizeof(*head), GFP_KERNEL);
 	if (!head)
@@ -2355,22 +2419,14 @@ nv50_head_create(struct drm_device *dev, int index)
 	drm_crtc_helper_add(crtc, &nv50_head_help);
 	drm_mode_crtc_set_gamma_size(crtc, 256);
 
-	ret = nouveau_bo_new(&drm->client, 8192, 0x100, TTM_PL_FLAG_VRAM,
-			     0, 0x0000, NULL, NULL, &head->base.lut.nvbo);
-	if (!ret) {
-		ret = nouveau_bo_pin(head->base.lut.nvbo, TTM_PL_FLAG_VRAM, true);
-		if (!ret) {
-			ret = nouveau_bo_map(head->base.lut.nvbo);
-			if (ret)
-				nouveau_bo_unpin(head->base.lut.nvbo);
-		}
+	for (i = 0; i < ARRAY_SIZE(head->lut.nvbo); i++) {
+		ret = nouveau_bo_new_pin_map(&drm->client, 1025 * 8, 0x100,
+					     TTM_PL_FLAG_VRAM,
+					     &head->lut.nvbo[i]);
 		if (ret)
-			nouveau_bo_ref(NULL, &head->base.lut.nvbo);
+			goto out;
 	}
 
-	if (ret)
-		goto out;
-
 	/* allocate overlay resources */
 	ret = nv50_oimm_create(device, disp->disp, index, &head->oimm);
 	if (ret)
@@ -4350,7 +4406,6 @@ nv50_display_init(struct drm_device *dev)
 {
 	struct drm_encoder *encoder;
 	struct drm_plane *plane;
-	struct drm_crtc *crtc;
 	u32 *push;
 
 	push = evo_wait(nv50_mast(dev), 32);
@@ -4369,10 +4424,6 @@ nv50_display_init(struct drm_device *dev)
 		}
 	}
 
-	drm_for_each_crtc(crtc, dev) {
-		nv50_head_lut_load(crtc);
-	}
-
 	drm_for_each_plane(plane, dev) {
 		struct nv50_wndw *wndw = nv50_wndw(plane);
 		if (plane->funcs != &nv50_wndw)
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 00eeaaffeae5..073534ab6911 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -28,6 +28,7 @@
 #include <core/option.h>
 
 #include <subdev/bios.h>
+#include <subdev/therm.h>
 
 static DEFINE_MUTEX(nv_devices_mutex);
 static LIST_HEAD(nv_devices);
@@ -1682,7 +1683,7 @@ nve4_chipset = {
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gk104_pmu_new,
-	.therm = gf119_therm_new,
+	.therm = gk104_therm_new,
 	.timer = nv41_timer_new,
 	.top = gk104_top_new,
 	.volt = gk104_volt_new,
@@ -1721,7 +1722,7 @@ nve6_chipset = {
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gk104_pmu_new,
-	.therm = gf119_therm_new,
+	.therm = gk104_therm_new,
 	.timer = nv41_timer_new,
 	.top = gk104_top_new,
 	.volt = gk104_volt_new,
@@ -1760,7 +1761,7 @@ nve7_chipset = {
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gk104_pmu_new,
-	.therm = gf119_therm_new,
+	.therm = gk104_therm_new,
 	.timer = nv41_timer_new,
 	.top = gk104_top_new,
 	.volt = gk104_volt_new,
@@ -1811,7 +1812,7 @@ nvf0_chipset = {
 	.bus = gf100_bus_new,
 	.clk = gk104_clk_new,
 	.devinit = gf100_devinit_new,
-	.fb = gk104_fb_new,
+	.fb = gk110_fb_new,
 	.fuse = gf100_fuse_new,
 	.gpio = gk104_gpio_new,
 	.i2c = gk104_i2c_new,
@@ -1824,7 +1825,7 @@ nvf0_chipset = {
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gk110_pmu_new,
-	.therm = gf119_therm_new,
+	.therm = gk104_therm_new,
 	.timer = nv41_timer_new,
 	.top = gk104_top_new,
 	.volt = gk104_volt_new,
@@ -1849,7 +1850,7 @@ nvf1_chipset = {
 	.bus = gf100_bus_new,
 	.clk = gk104_clk_new,
 	.devinit = gf100_devinit_new,
-	.fb = gk104_fb_new,
+	.fb = gk110_fb_new,
 	.fuse = gf100_fuse_new,
 	.gpio = gk104_gpio_new,
 	.i2c = gk104_i2c_new,
@@ -1862,7 +1863,7 @@ nvf1_chipset = {
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gk110_pmu_new,
-	.therm = gf119_therm_new,
+	.therm = gk104_therm_new,
 	.timer = nv41_timer_new,
 	.top = gk104_top_new,
 	.volt = gk104_volt_new,
@@ -1887,7 +1888,7 @@ nv106_chipset = {
 	.bus = gf100_bus_new,
 	.clk = gk104_clk_new,
 	.devinit = gf100_devinit_new,
-	.fb = gk104_fb_new,
+	.fb = gk110_fb_new,
 	.fuse = gf100_fuse_new,
 	.gpio = gk104_gpio_new,
 	.i2c = gk104_i2c_new,
@@ -1900,7 +1901,7 @@ nv106_chipset = {
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gk208_pmu_new,
-	.therm = gf119_therm_new,
+	.therm = gk104_therm_new,
 	.timer = nv41_timer_new,
 	.top = gk