amd/vpelib: update 3dlut and shaper FL

[WHY]
Fast load support is required for 3DLUT and Shaper features.
The calculation logic needs to be modularized and exposed via
the resource interface to support this.

[HOW]
1. Add `calculate_shaper` and `program_fastload` function pointers to the `resource` struct.
2. Move shaper normalization, HDR multiplier update, and 3DLUT update logic from
   `vpe_color_update_movable_cm` into a new core function `vpe_calculate_shaper`.
3. Implement `vpe10_calculate_shaper` and assign it to the resource function table for VPE10 and VPE11.
4. Update `vpe_create_engine` return signature to remove `const` qualifier.

Acked-by: Chuanyu Tseng <Chuanyu.Tseng@amd.com>
Signed-off-by: Nawwar Ali <Nawwar.Ali@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39848>

src/amd/vpelib/inc/vpelib.h

@@ -158,7 +158,7 @@ enum vpe_status vpe_build_resolve_query(
  * @param[in] param  provide the asic version.
  * @return           vpe engine instance if valid. NULL otherwise
  */
-const struct vpe_engine *vpe_create_engine(struct vpe_init_data *param);
+struct vpe_engine *vpe_create_engine(struct vpe_init_data *param);
 
 /**
  * destroy the vpe engine instance.

src/amd/vpelib/src/chip/vpe10/inc/vpe10_resource.h

@@ -106,6 +106,9 @@ bool vpe10_validate_cached_param(struct vpe_priv *vpe_priv, const struct vpe_bui
 void vpe10_setup_check_funcs(struct vpe_check_support_funcs *funcs);
 
 const struct vpe_caps *vpe10_get_capability(void);
+
+enum vpe_status vpe10_calculate_shaper(struct vpe_priv *vpe_priv, struct stream_ctx *stream_ctx);
+
 #ifdef __cplusplus
 }
 #endif

src/amd/vpelib/src/chip/vpe10/vpe10_resource.c

@@ -438,6 +438,7 @@ enum vpe_status vpe10_construct_resource(struct vpe_priv *vpe_priv, struct resou
     res->update_blnd_gamma                 = vpe10_update_blnd_gamma;
     res->update_output_gamma               = vpe10_update_output_gamma;
     res->validate_cached_param             = vpe10_validate_cached_param;
+    res->calculate_shaper                  = vpe10_calculate_shaper;
 
     return VPE_STATUS_OK;
 err:
@@ -1535,3 +1536,8 @@ void vpe10_setup_check_funcs(struct vpe_check_support_funcs *funcs)
     funcs->get_dcc_compression_input_cap  = vpe10_get_dcc_compression_input_cap;
     funcs->get_dcc_compression_output_cap = vpe10_get_dcc_compression_output_cap;
 }
+
+enum vpe_status vpe10_calculate_shaper(struct vpe_priv *vpe_priv, struct stream_ctx *stream_ctx)
+{
+    return vpe_calculate_shaper(vpe_priv, stream_ctx); // calculate shaper data and config
+}

src/amd/vpelib/src/chip/vpe11/vpe11_resource.c

@@ -210,6 +210,7 @@ enum vpe_status vpe11_construct_resource(struct vpe_priv *vpe_priv, struct resou
     res->update_blnd_gamma                 = vpe10_update_blnd_gamma;
     res->update_output_gamma               = vpe10_update_output_gamma;
     res->validate_cached_param             = vpe11_validate_cached_param;
+    res->calculate_shaper                  = vpe10_calculate_shaper;
 
     return VPE_STATUS_OK;
 err:

src/amd/vpelib/src/core/color.c

@@ -777,6 +777,40 @@ enum vpe_status vpe_color_update_shaper(const struct vpe_priv *vpe_priv, uint16_
     return ret;
 }
 
+enum vpe_status vpe_calculate_shaper(struct vpe_priv *vpe_priv, struct stream_ctx *stream_ctx)
+{
+    uint32_t                 shaper_norm_factor;
+    struct vpe_color_space   cs;
+    enum color_space         out_lut_cs;
+    enum color_transfer_func lut_in_tf;
+    bool                     need_update_3dlut = true;
+    bool                     enable_3dlut =
+        stream_ctx->stream.tm_params.UID != 0 || stream_ctx->stream.tm_params.enable_3dlut;
+
+    // Get the normalization factor for the shaper based on tone mapping parameters.
+    get_shaper_norm_factor(&stream_ctx->stream.tm_params, stream_ctx, &shaper_norm_factor);
+
+    // Update the HDR multiplier based on the shaper normalization factor and other parameters.
+    vpe_color_tm_update_hdr_mult(SHAPER_EXP_MAX_IN, shaper_norm_factor,
+        &stream_ctx->lut3d_func->hdr_multiplier, enable_3dlut,
+        vpe_is_fp16(stream_ctx->stream.surface_info.format));
+
+    if (need_update_3dlut) {
+        /* DirectConfig loading case */
+        vpe_color_update_3dlut(vpe_priv, stream_ctx, enable_3dlut);
+    }
+
+    // Build the shaper color space based on tone mapping parameters and output surface.
+    vpe_color_build_shaper_cs(&stream_ctx->stream.tm_params, &vpe_priv->output_ctx.surface, &cs);
+
+    // Extract the color space and transfer function from the shaper color space.
+    vpe_color_get_color_space_and_tf(&cs, &out_lut_cs, &lut_in_tf);
+
+    // Update the shaper transfer function for the current stream.
+    return vpe_color_update_shaper(
+        vpe_priv, SHAPER_EXP_MAX_IN, stream_ctx, lut_in_tf, enable_3dlut);
+}
+
 enum vpe_status vpe_color_update_movable_cm(
     struct vpe_priv *vpe_priv, const struct vpe_build_param *param)
 {
@@ -793,12 +827,9 @@ enum vpe_status vpe_color_update_movable_cm(
 
         if (stream_ctx->uid_3dlut != stream_ctx->stream.tm_params.UID) {
 
-            uint32_t                 shaper_norm_factor;
             struct vpe_color_space   tm_out_cs;
-            struct vpe_color_space   cs;
             enum color_space         out_lut_cs;
-            enum color_transfer_func tf, lut_in_tf;
-
+            enum color_transfer_func tf;
             if (!stream_ctx->in_shaper_func) {
                 stream_ctx->in_shaper_func =
                     (struct transfer_func *)vpe_zalloc(sizeof(struct transfer_func));
@@ -838,17 +869,7 @@ enum vpe_status vpe_color_update_movable_cm(
                 }
             }
 
-            get_shaper_norm_factor(&stream_ctx->stream.tm_params, stream_ctx, &shaper_norm_factor);
-
-            vpe_color_tm_update_hdr_mult(SHAPER_EXP_MAX_IN, shaper_norm_factor,
-                &stream_ctx->lut3d_func->hdr_multiplier, enable_3dlut,
-                vpe_is_fp16(stream_ctx->stream.surface_info.format));
-
-            vpe_color_build_shaper_cs(
-                &stream_ctx->stream.tm_params, &vpe_priv->output_ctx.surface, &cs);
-            vpe_color_get_color_space_and_tf(&cs, &out_lut_cs, &lut_in_tf);
-            vpe_color_update_shaper(
-                vpe_priv, SHAPER_EXP_MAX_IN, stream_ctx, lut_in_tf, enable_3dlut);
+            vpe_priv->resource.calculate_shaper(vpe_priv, stream_ctx);
 
             vpe_color_build_tm_cs(
                 &stream_ctx->stream.tm_params, &vpe_priv->output_ctx.surface, &tm_out_cs);

src/amd/vpelib/src/core/inc/cmd_builder.h

@@ -26,6 +26,7 @@
 
 #include "vpe_types.h"
 #include "plane_desc_writer.h"
+#include "hw_shared.h"
 
 #ifdef __cplusplus
 extern "C" {
@@ -46,6 +47,7 @@ struct cmd_builder {
 
     enum vpe_status (*build_collaborate_sync_cmd)(
         struct vpe_priv *vpe_priv, struct vpe_build_bufs *cur_bufs);
+
 };
 
 #ifdef __cplusplus

src/amd/vpelib/src/core/inc/color.h

@@ -264,6 +264,8 @@ enum vpe_status vpe_color_update_movable_cm(
 void vpe_color_get_color_space_and_tf(
     const struct vpe_color_space *vcs, enum color_space *cs, enum color_transfer_func *tf);
 
+enum vpe_status vpe_shaper_generator(struct vpe_priv *vpe_priv);
+
 bool vpe_use_csc_adjust(const struct vpe_color_adjust *adjustments);
 
 bool vpe_is_rgb_equal(const struct pwl_result_data *rgb, uint32_t num);
@@ -302,6 +304,8 @@ bool vpe_color_update_degamma_tf(struct vpe_priv *vpe_priv, enum color_transfer_
 enum color_range_type vpe_get_range_type(
     enum color_space color_space, enum vpe_surface_pixel_format format);
 
+enum vpe_status vpe_calculate_shaper(struct vpe_priv *vpe_priv, struct stream_ctx *stream_ctx);
+
 #ifdef __cplusplus
 }
 #endif

src/amd/vpelib/src/core/inc/resource.h

@@ -114,6 +114,9 @@ struct resource {
         bool geometric_scaling);
 
     bool (*validate_cached_param)(struct vpe_priv *vpe_priv, const struct vpe_build_param *param);
+
+    enum vpe_status (*calculate_shaper)(struct vpe_priv *vpe_priv, struct stream_ctx *stream_ctx);
+
     // Indicates the nominal range hdr input content should be in during processing.
     int internal_hdr_normalization;
 

src/amd/vpelib/src/core/vpelib.c

@@ -1011,7 +1011,7 @@ enum vpe_status vpe_build_resolve_query(
     return result;
 }
 
-const struct vpe_engine *vpe_create_engine(struct vpe_init_data *params)
+struct vpe_engine *vpe_create_engine(struct vpe_init_data *params)
 {
     struct vpe_engine_priv *engine_priv;
     struct vpe_engine      *engine_handle;