uniform-u8vec4-srgb-taylor.frag

#version 430 core

vec3 taylorSrgbToRgb(in vec3 ColorSRGB)
{
	vec3 A = (ColorSRGB * 0.305306011 + 0.682171111);
	vec3 B = (ColorSRGB * A + 0.012522878);
	return ColorSRGB * B;
}

vec4 taylorSrgbToRgb(in vec4 ColorSRGB)
{
	return vec4(taylorSrgbToRgb(ColorSRGB.rgb), ColorSRGB.a);
}

layout(binding = 0) uniform bufferFetch
{
	uint Color;
} Buffer;

layout(location = 0, index = 0) out vec4 Color;

void main()
{
	Color = taylorSrgbToRgb(unpackUnorm4x8(Buffer.Color));
}

    1 |   0 |          | label_basic_block_1: s_mov_b32 s0, s1
    2 |   0 |          | s_movk_i32 s1, 0x0000
    3 |   0 |          | s_load_dwordx4 s[4:7], s[0:1], 0x140
    4 |   0 |          | s_waitcnt lgkmcnt(0)
    5 |   0 |          | s_andn2_b32 s5, s5, 0x3fff0000
    6 |   0 |          | s_buffer_load_dword s2, s[4:7], 0x00
    7 |   0 |          | s_waitcnt lgkmcnt(0)
    8 |   0 |          | s_lshl_b32 s2, s2, 4
    9 |   0 |          | s_load_dwordx4 s[0:3], s[0:1], s2
   10 |   0 |          | s_waitcnt lgkmcnt(0)
   11 |   0 |          | s_andn2_b32 s1, s1, 0x3fff0000
   12 |   0 |          | s_buffer_load_dword s0, s[0:3], 0x00
   13 |   0 |          | s_waitcnt lgkmcnt(0)
   14 |   1 | ^        | v_cvt_f32_ubyte0 v0, s0
   15 |   2 | :^       | v_cvt_f32_ubyte1 v1, s0
   16 |   3 | ::^      | v_cvt_f32_ubyte2 v2, s0
   17 |   4 | :::^     | v_cvt_f32_ubyte3 v3, s0
   18 |   5 | ::::^    | v_mov_b32 v4, 0x3b808081
   19 |   6 | v:::v^   | v_mul_f32 v5, v4, v0
   20 |   7 | :v::v:^  | v_mul_f32 v6, v1, v4
   21 |   8 | ::v:v::^ | v_mul_f32 v7, v2, v4
   22 |   8 | :::xv::: | v_mul_f32 v3, v3, v4
   23 |   8 | ::::^::: | v_mov_b32 v4, 0x3f2ea2c4
   24 |   8 | :::::::: | s_mov_b32 s0, 0x3a9cee01
   25 |   8 | x:::v::: | v_mad_f32 v0, v0, s0, v4
   26 |   8 | :x::v::: | v_mad_f32 v1, v1, s0, v4
   27 |   8 | ::v:x::: | v_mac_f32 v4, s0, v2
   28 |   7 | x: ::v:: | v_madak_f32 v0, v5, v0, 0x3c4d2cc2
   29 |   7 | :x :::v: | v_madak_f32 v1, v6, v1, 0x3c4d2cc2
   30 |   8 | ::^:v::v | v_madak_f32 v2, v7, v4, 0x3c4d2cc2
   31 |   7 | x::: v:: | v_mul_f32 v0, v5, v0
   32 |   6 | :x::  v: | v_mul_f32 v1, v6, v1
   33 |   5 | ::x:   v | v_mul_f32 v2, v7, v2
   34 |   4 | xv::     | v_cvt_pkrtz_f16_f32 v0, v0, v1
   35 |   4 | :^vv     | v_cvt_pkrtz_f16_f32 v1, v2, v3
   36 |   2 | vv       | exp mrt0, v0, v0, v1, v1
   37 |   0 |          | s_endpgm 

Maximum # VGPR used   8, # VGPR allocated:   8