drm/amd/display: add shaper and blend colorops for 1D Curve Custom LUT

This patch adds colorops for custom 1D LUTs in the SHAPER and
BLND HW blocks.

With this change the following IGT tests pass:
kms_colorop --run plane-XR30-XR30-srgb_inv_eotf_lut
kms_colorop --run plane-XR30-XR30-srgb_inv_eotf_lut-srgb_eotf_lut

The color pipeline now consists of the following colorops:
1. 1D curve colorop
2. 1D curve colorop
3. 1D LUT
4. 1D curve colorop
5. 1D LUT

The 1D curve colorops support sRGB, BT2020, and PQ scaled to 125.0.

Signed-off-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Simon Ser <contact@emersion.fr>
Link: https://patch.msgid.link/20251115000237.3561250-39-alex.hung@amd.com

2 files changed, 251 insertions, 77 deletions

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
index 0ad3d818189e..4d04da502e84 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
@@ -73,6 +73,7 @@
  */
 
 #define MAX_DRM_LUT_VALUE 0xFFFF
+#define MAX_DRM_LUT32_VALUE 0xFFFFFFFF
 #define SDR_WHITE_LEVEL_INIT_VALUE 80
 
 /**
@@ -343,6 +344,21 @@ __extract_blob_lut(const struct drm_property_blob *blob, uint32_t *size)
 }
 
 /**
+ * __extract_blob_lut32 - Extracts the DRM lut and lut size from a blob.
+ * @blob: DRM color mgmt property blob
+ * @size: lut size
+ *
+ * Returns:
+ * DRM LUT or NULL
+ */
+static const struct drm_color_lut32 *
+__extract_blob_lut32(const struct drm_property_blob *blob, uint32_t *size)
+{
+	*size = blob ? drm_color_lut32_size(blob) : 0;
+	return blob ? (struct drm_color_lut32 *)blob->data : NULL;
+}
+
+/**
  * __is_lut_linear - check if the given lut is a linear mapping of values
  * @lut: given lut to check values
  * @size: lut size
@@ -416,6 +432,24 @@ static void __drm_lut_to_dc_gamma(const struct drm_color_lut *lut,
 }
 
 /**
+ * __drm_lut32_to_dc_gamma - convert the drm_color_lut to dc_gamma.
+ * @lut: DRM lookup table for color conversion
+ * @gamma: DC gamma to set entries
+ *
+ * The conversion depends on the size of the lut - whether or not it's legacy.
+ */
+static void __drm_lut32_to_dc_gamma(const struct drm_color_lut32 *lut, struct dc_gamma *gamma)
+{
+	int i;
+
+	for (i = 0; i < MAX_COLOR_LUT_ENTRIES; i++) {
+		gamma->entries.red[i] = dc_fixpt_from_fraction(lut[i].red, MAX_DRM_LUT32_VALUE);
+		gamma->entries.green[i] = dc_fixpt_from_fraction(lut[i].green, MAX_DRM_LUT32_VALUE);
+		gamma->entries.blue[i] = dc_fixpt_from_fraction(lut[i].blue, MAX_DRM_LUT32_VALUE);
+	}
+}
+
+/**
  * __drm_ctm_to_dc_matrix - converts a DRM CTM to a DC CSC float matrix
  * @ctm: DRM color transformation matrix
  * @matrix: DC CSC float matrix
@@ -567,6 +601,63 @@ static int __set_output_tf(struct dc_transfer_func *func,
 	return res ? 0 : -ENOMEM;
 }
 
+/**
+ * __set_output_tf_32 - calculates the output transfer function based on expected input space.
+ * @func: transfer function
+ * @lut: lookup table that defines the color space
+ * @lut_size: size of respective lut
+ * @has_rom: if ROM can be used for hardcoded curve
+ *
+ * Returns:
+ * 0 in case of success. -ENOMEM if fails.
+ */
+static int __set_output_tf_32(struct dc_transfer_func *func,
+			      const struct drm_color_lut32 *lut, uint32_t lut_size,
+			      bool has_rom)
+{
+	struct dc_gamma *gamma = NULL;
+	struct calculate_buffer cal_buffer = {0};
+	bool res;
+
+	cal_buffer.buffer_index = -1;
+
+	if (lut_size) {
+		gamma = dc_create_gamma();
+		if (!gamma)
+			return -ENOMEM;
+
+		gamma->num_entries = lut_size;
+		__drm_lut32_to_dc_gamma(lut, gamma);
+	}
+
+	if (func->tf == TRANSFER_FUNCTION_LINEAR) {
+		/*
+		 * Color module doesn't like calculating regamma params
+		 * on top of a linear input. But degamma params can be used
+		 * instead to simulate this.
+		 */
+		if (gamma)
+			gamma->type = GAMMA_CUSTOM;
+		res = mod_color_calculate_degamma_params(NULL, func,
+							 gamma, gamma != NULL);
+	} else {
+		/*
+		 * Assume sRGB. The actual mapping will depend on whether the
+		 * input was legacy or not.
+		 */
+		if (gamma)
+			gamma->type = GAMMA_CS_TFM_1D;
+		res = mod_color_calculate_regamma_params(func, gamma, gamma != NULL,
+							 has_rom, NULL, &cal_buffer);
+	}
+
+	if (gamma)
+		dc_gamma_release(&gamma);
+
+	return res ? 0 : -ENOMEM;
+}
+
+
 static int amdgpu_dm_set_atomic_regamma(struct dc_transfer_func *out_tf,
 					const struct drm_color_lut *regamma_lut,
 					uint32_t regamma_size, bool has_rom,
@@ -639,6 +730,42 @@ static int __set_input_tf(struct dc_color_caps *caps, struct dc_transfer_func *f
 	return res ? 0 : -ENOMEM;
 }
 
+/**
+ * __set_input_tf_32 - calculates the input transfer function based on expected
+ * input space.
+ * @caps: dc color capabilities
+ * @func: transfer function
+ * @lut: lookup table that defines the color space
+ * @lut_size: size of respective lut.
+ *
+ * Returns:
+ * 0 in case of success. -ENOMEM if fails.
+ */
+static int __set_input_tf_32(struct dc_color_caps *caps, struct dc_transfer_func *func,
+			     const struct drm_color_lut32 *lut, uint32_t lut_size)
+{
+	struct dc_gamma *gamma = NULL;
+	bool res;
+
+	if (lut_size) {
+		gamma = dc_create_gamma();
+		if (!gamma)
+			return -ENOMEM;
+
+		gamma->type = GAMMA_CUSTOM;
+		gamma->num_entries = lut_size;
+
+		__drm_lut32_to_dc_gamma(lut, gamma);
+	}
+
+	res = mod_color_calculate_degamma_params(caps, func, gamma, gamma != NULL);
+
+	if (gamma)
+		dc_gamma_release(&gamma);
+
+	return res ? 0 : -ENOMEM;
+}
+
 static enum dc_transfer_func_predefined
 amdgpu_tf_to_dc_tf(enum amdgpu_transfer_function tf)
 {
@@ -1252,38 +1379,6 @@ __set_dm_plane_colorop_degamma(struct drm_plane_state *plane_state,
 }
 
 static int
-__set_colorop_in_shaper_1d_curve(struct dc_plane_state *dc_plane_state,
-				 struct drm_colorop_state *colorop_state)
-{
-	struct dc_transfer_func *tf = &dc_plane_state->in_shaper_func;
-	struct drm_colorop *colorop = colorop_state->colorop;
-	struct drm_device *drm = colorop->dev;
-
-	if (colorop->type != DRM_COLOROP_1D_CURVE)
-		return -EINVAL;
-
-	if (!(BIT(colorop_state->curve_1d_type) & amdgpu_dm_supported_shaper_tfs))
-		return -EINVAL;
-
-	if (colorop_state->bypass) {
-		tf->type = TF_TYPE_BYPASS;
-		tf->tf = TRANSFER_FUNCTION_LINEAR;
-		return 0;
-	}
-
-	drm_dbg(drm, "Shaper colorop with ID: %d\n", colorop->base.id);
-
-	if (colorop->type == DRM_COLOROP_1D_CURVE) {
-		tf->type = TF_TYPE_DISTRIBUTED_POINTS;
-		tf->tf = amdgpu_colorop_tf_to_dc_tf(colorop_state->curve_1d_type);
-		tf->sdr_ref_white_level = SDR_WHITE_LEVEL_INIT_VALUE;
-		return __set_output_tf(tf, 0, 0, false);
-	}
-
-	return -EINVAL;
-}
-
-static int
 __set_dm_plane_colorop_shaper(struct drm_plane_state *plane_state,
 			      struct dc_plane_state *dc_plane_state,
 			      struct drm_colorop *colorop)
@@ -1291,62 +1386,66 @@ __set_dm_plane_colorop_shaper(struct drm_plane_state *plane_state,
 	struct drm_colorop *old_colorop;
 	struct drm_colorop_state *colorop_state = NULL, *new_colorop_state;
 	struct drm_atomic_state *state = plane_state->state;
-	int i = 0;
+	enum dc_transfer_func_predefined default_tf = TRANSFER_FUNCTION_LINEAR;
+	struct dc_transfer_func *tf = &dc_plane_state->in_shaper_func;
+	const struct drm_color_lut32 *shaper_lut;
+	struct drm_device *dev = colorop->dev;
+	u32 shaper_size;
+	int i = 0, ret = 0;
 
+	/* 1D Curve - SHAPER TF */
 	old_colorop = colorop;
-
-	/* 2nd op: 1d curve - shaper */
 	for_each_new_colorop_in_state(state, colorop, new_colorop_state, i) {
 		if (new_colorop_state->colorop == old_colorop &&
 		    (BIT(new_colorop_state->curve_1d_type) & amdgpu_dm_supported_shaper_tfs)) {
 			colorop_state = new_colorop_state;
 			break;
 		}
-
-		if (new_colorop_state->colorop == old_colorop) {
-			colorop_state = new_colorop_state;
-			break;
-		}
 	}
 
-	if (!colorop_state)
-		return -EINVAL;
-
-	return __set_colorop_in_shaper_1d_curve(dc_plane_state, colorop_state);
-}
-
-static int
-__set_colorop_1d_curve_blend_tf_lut(struct dc_plane_state *dc_plane_state,
-				    struct drm_colorop_state *colorop_state)
-{
-	struct dc_transfer_func *tf = &dc_plane_state->blend_tf;
-	struct drm_colorop *colorop = colorop_state->colorop;
-	struct drm_device *drm = colorop->dev;
-	const struct drm_color_lut *blend_lut = NULL;
-	u32 blend_size = 0;
-
-	if (colorop->type != DRM_COLOROP_1D_CURVE)
-		return -EINVAL;
+	if (colorop_state && !colorop_state->bypass && colorop->type == DRM_COLOROP_1D_CURVE) {
+		drm_dbg(dev, "Shaper TF colorop with ID: %d\n", colorop->base.id);
+		tf->type = TF_TYPE_DISTRIBUTED_POINTS;
+		tf->tf = default_tf = amdgpu_colorop_tf_to_dc_tf(colorop_state->curve_1d_type);
+		tf->sdr_ref_white_level = SDR_WHITE_LEVEL_INIT_VALUE;
+		ret = __set_output_tf(tf, 0, 0, false);
+		if (ret)
+			return ret;
+	}
 
-	if (!(BIT(colorop_state->curve_1d_type) & amdgpu_dm_supported_blnd_tfs))
+	/* 1D LUT - SHAPER LUT */
+	colorop = old_colorop->next;
+	if (!colorop) {
+		drm_dbg(dev, "no Shaper LUT colorop found\n");
 		return -EINVAL;
-
-	if (colorop_state->bypass) {
-		tf->type = TF_TYPE_BYPASS;
-		tf->tf = TRANSFER_FUNCTION_LINEAR;
-		return 0;
 	}
 
-	drm_dbg(drm, "Blend colorop with ID: %d\n", colorop->base.id);
+	old_colorop = colorop;
+	for_each_new_colorop_in_state(state, colorop, new_colorop_state, i) {
+		if (new_colorop_state->colorop == old_colorop &&
+		    new_colorop_state->colorop->type == DRM_COLOROP_1D_LUT) {
+			colorop_state = new_colorop_state;
+			break;
+		}
+	}
 
-	if (colorop->type == DRM_COLOROP_1D_CURVE) {
+	if (colorop_state && !colorop_state->bypass && colorop->type == DRM_COLOROP_1D_LUT) {
+		drm_dbg(dev, "Shaper LUT colorop with ID: %d\n", colorop->base.id);
 		tf->type = TF_TYPE_DISTRIBUTED_POINTS;
-		tf->tf = amdgpu_colorop_tf_to_dc_tf(colorop_state->curve_1d_type);
+		tf->tf = default_tf;
 		tf->sdr_ref_white_level = SDR_WHITE_LEVEL_INIT_VALUE;
-		return __set_input_tf(NULL, tf, blend_lut, blend_size);
+		shaper_lut = __extract_blob_lut32(colorop_state->data, &shaper_size);
+		shaper_size = shaper_lut != NULL ? shaper_size : 0;
+
+		/* Custom LUT size must be the same as supported size */
+		if (shaper_size == colorop->size) {
+			ret = __set_output_tf_32(tf, shaper_lut, shaper_size, false);
+			if (ret)
+				return ret;
+		}
 	}
 
-	return -EINVAL;
+	return 0;
 }
 
 static int
@@ -1357,28 +1456,63 @@ __set_dm_plane_colorop_blend(struct drm_plane_state *plane_state,
 	struct drm_colorop *old_colorop;
 	struct drm_colorop_state *colorop_state = NULL, *new_colorop_state;
 	struct drm_atomic_state *state = plane_state->state;
+	enum dc_transfer_func_predefined default_tf = TRANSFER_FUNCTION_LINEAR;
+	struct dc_transfer_func *tf = &dc_plane_state->blend_tf;
+	const struct drm_color_lut32 *blend_lut = NULL;
+	struct drm_device *dev = colorop->dev;
+	uint32_t blend_size = 0;
 	int i = 0;
 
+	/* 1D Curve - BLND TF */
 	old_colorop = colorop;
-
-	/* 3nd op: 1d curve - blend */
 	for_each_new_colorop_in_state(state, colorop, new_colorop_state, i) {
 		if (new_colorop_state->colorop == old_colorop &&
 		    (BIT(new_colorop_state->curve_1d_type) & amdgpu_dm_supported_blnd_tfs)) {
 			colorop_state = new_colorop_state;
 			break;
 		}
+	}
 
-		if (new_colorop_state->colorop == old_colorop) {
+	if (colorop_state && !colorop_state->bypass && colorop->type == DRM_COLOROP_1D_CURVE &&
+	    (BIT(colorop_state->curve_1d_type) & amdgpu_dm_supported_blnd_tfs)) {
+		drm_dbg(dev, "Blend TF colorop with ID: %d\n", colorop->base.id);
+		tf->type = TF_TYPE_DISTRIBUTED_POINTS;
+		tf->tf = default_tf = amdgpu_colorop_tf_to_dc_tf(colorop_state->curve_1d_type);
+		tf->sdr_ref_white_level = SDR_WHITE_LEVEL_INIT_VALUE;
+		__set_input_tf_32(NULL, tf, blend_lut, blend_size);
+	}
+
+	/* 1D Curve - BLND LUT */
+	colorop = old_colorop->next;
+	if (!colorop) {
+		drm_dbg(dev, "no Blend LUT colorop found\n");
+		return -EINVAL;
+	}
+
+	old_colorop = colorop;
+	for_each_new_colorop_in_state(state, colorop, new_colorop_state, i) {
+		if (new_colorop_state->colorop == old_colorop &&
+		    new_colorop_state->colorop->type == DRM_COLOROP_1D_LUT) {
 			colorop_state = new_colorop_state;
 			break;
 		}
 	}
 
-	if (!colorop_state)
-		return -EINVAL;
+	if (colorop_state && !colorop_state->bypass && colorop->type == DRM_COLOROP_1D_LUT &&
+	    (BIT(colorop_state->curve_1d_type) & amdgpu_dm_supported_blnd_tfs)) {
+		drm_dbg(dev, "Blend LUT colorop with ID: %d\n", colorop->base.id);
+		tf->type = TF_TYPE_DISTRIBUTED_POINTS;
+		tf->tf = default_tf;
+		tf->sdr_ref_white_level = SDR_WHITE_LEVEL_INIT_VALUE;
+		blend_lut = __extract_blob_lut32(colorop_state->data, &blend_size);
+		blend_size = blend_lut != NULL ? blend_size : 0;
 
-	return __set_colorop_1d_curve_blend_tf_lut(dc_plane_state, colorop_state);
+		/* Custom LUT size must be the same as supported size */
+		if (blend_size == colorop->size)
+			__set_input_tf_32(NULL, tf, blend_lut, blend_size);
+	}
+
+	return 0;
 }
 
 static int
@@ -1447,7 +1581,7 @@ amdgpu_dm_plane_set_colorop_properties(struct drm_plane_state *plane_state,
 	if (ret)
 		return ret;
 
-	/* 1D Curve - SHAPER TF */
+	/* 1D Curve & LUT - SHAPER TF & LUT */
 	colorop = colorop->next;
 	if (!colorop) {
 		drm_dbg(dev, "no Shaper TF colorop found\n");
@@ -1458,7 +1592,12 @@ amdgpu_dm_plane_set_colorop_properties(struct drm_plane_state *plane_state,
 	if (ret)
 		return ret;
 
-	/* 1D Curve - BLND TF */
+	/* Shaper LUT colorop is already handled, just skip here */
+	colorop = colorop->next;
+	if (!colorop)
+		return -EINVAL;
+
+	/* 1D Curve & LUT - BLND TF & LUT */
 	colorop = colorop->next;
 	if (!colorop) {
 		drm_dbg(dev, "no Blend TF colorop found\n");
@@ -1469,6 +1608,11 @@ amdgpu_dm_plane_set_colorop_properties(struct drm_plane_state *plane_state,
 	if (ret)
 		return ret;
 
+	/* BLND LUT colorop is already handled, just skip here */
+	colorop = colorop->next;
+	if (!colorop)
+		return -EINVAL;
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c
index e72261a8beb9..532b0b8bf8c1 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c
@@ -29,6 +29,7 @@
 #include <drm/drm_property.h>
 #include <drm/drm_colorop.h>
 
+#include "amdgpu.h"
 #include "amdgpu_dm_colorop.h"
 
 const u64 amdgpu_dm_supported_degam_tfs =
@@ -88,6 +89,21 @@ int amdgpu_dm_initialize_default_pipeline(struct drm_plane *plane, struct drm_pr
 
 	i++;
 
+	/* 1D LUT - SHAPER LUT */
+	ops[i] = kzalloc(sizeof(struct drm_colorop), GFP_KERNEL);
+	if (!ops[i]) {
+		ret = -ENOMEM;
+		goto cleanup;
+	}
+
+	ret = drm_plane_colorop_curve_1d_lut_init(dev, ops[i], plane, MAX_COLOR_LUT_ENTRIES);
+	if (ret)
+		goto cleanup;
+
+	drm_colorop_set_next_property(ops[i-1], ops[i]);
+
+	i++;
+
 	/* 1D curve - BLND TF */
 	ops[i] = kzalloc(sizeof(*ops[0]), GFP_KERNEL);
 	if (!ops[i]) {
@@ -101,6 +117,20 @@ int amdgpu_dm_initialize_default_pipeline(struct drm_plane *plane, struct drm_pr
 
 	drm_colorop_set_next_property(ops[i - 1], ops[i]);
 
+	i++;
+
+	/* 1D LUT - BLND LUT */
+	ops[i] = kzalloc(sizeof(struct drm_colorop), GFP_KERNEL);
+	if (!ops[i]) {
+		ret = -ENOMEM;
+		goto cleanup;
+	}
+
+	ret = drm_plane_colorop_curve_1d_lut_init(dev, ops[i], plane, MAX_COLOR_LUT_ENTRIES);
+	if (ret)
+		goto cleanup;
+
+	drm_colorop_set_next_property(ops[i-1], ops[i]);
 	return 0;
 
 cleanup: