renderer: optimise shader usage further, split shaders and add more caching (#12992)

* shader: split CM rgba/rgbx into discard ones

make it branchless if we have no discards.

* shader: ensure we dont stall on vbo uv buffer

if we render a new texture before the previous was done gpu wise its
going to stall until done, call glBufferData to orphan the data.
this allows the driver to return a new memory block immediately
if the GPU is still reading from the previous one

* protocols: ensure we reset GL_PACK_ALIGNMENT

reset GL_PACK_ALIGNMENT back to the default initial value of 4

* shader: use unsigned short in VAO

loose a tiny bit of precision but gain massive bandwidth reductions.
use GL_UNSIGNED_SHORT and set it as normalized. clamp and round the UV
for uint16_t in customUv.

* shader: interleave vertex buffers

use std::array for fullverts, use a single interleaved buffer for
position and uv, should in theory improve cache locality. and also remove
the need to have two buffers around.

* shader: revert precision drop

we need the float precision because we might have 1.01 or similiar
floats entering CM shader maths, and rounding/clamping those means the
maths turns out wrong. so revert back to float, sadly higher bandwidth
usage.

* update doColorManagement api

* convert primaries to XYZ on cpu

* remove unused primaries uniform

---------

Co-authored-by: UjinT34 <ujint34@mail.ru>

src/render/shaders/glsl/CM.glsl

@@ -401,13 +401,12 @@ vec4 tonemap(vec4 color, mat3 dstXYZ) {
     return vec4(fromLMS * toLinear(vec4(ICtCpPQInv * ICtCp, 1.0), CM_TRANSFER_FUNCTION_ST2084_PQ).rgb * HDR_MAX_LUMINANCE * refScale, color[3]); 
 }
 
-vec4 doColorManagement(vec4 pixColor, int srcTF, int dstTF, mat4x2 dstPrimaries) {
+vec4 doColorManagement(vec4 pixColor, int srcTF, int dstTF, mat3 dstxyz) {
     pixColor.rgb /= max(pixColor.a, 0.001);
     pixColor.rgb = toLinearRGB(pixColor.rgb, srcTF);
     pixColor.rgb = convertMatrix * pixColor.rgb;
     pixColor = toNit(pixColor, srcTFRange);
     pixColor.rgb *= pixColor.a;
-    mat3 dstxyz = primaries2xyz(dstPrimaries);
     pixColor = tonemap(pixColor, dstxyz);
     pixColor = fromLinearNit(pixColor, dstTF, dstTFRange);
     if ((srcTF == CM_TRANSFER_FUNCTION_SRGB || srcTF == CM_TRANSFER_FUNCTION_GAMMA22) && dstTF == CM_TRANSFER_FUNCTION_ST2084_PQ) {

src/render/shaders/glsl/CMborder.frag

@@ -6,7 +6,7 @@ in vec2 v_texcoord;
 
 uniform int sourceTF; // eTransferFunction
 uniform int targetTF; // eTransferFunction
-uniform mat4x2 targetPrimaries;
+uniform mat3 targetPrimariesXYZ;
 
 uniform vec2 fullSizeUntransformed;
 uniform float radiusOuter;
@@ -90,7 +90,7 @@ void main() {
     pixColor = getColorForCoord(v_texcoord);
     pixColor.rgb *= pixColor[3];
 
-    pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimaries);
+    pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimariesXYZ);
 
     pixColor *= alpha * additionalAlpha;
 

src/render/shaders/glsl/CMrgba.frag

@@ -7,14 +7,9 @@ uniform sampler2D tex;
 
 uniform int sourceTF; // eTransferFunction
 uniform int targetTF; // eTransferFunction
-uniform mat4x2 targetPrimaries;
+uniform mat3 targetPrimariesXYZ;
 
 uniform float alpha;
-
-uniform bool discardOpaque;
-uniform bool discardAlpha;
-uniform float discardAlphaValue;
-
 uniform bool applyTint;
 uniform vec3 tint;
 
@@ -25,14 +20,8 @@ layout(location = 0) out vec4 fragColor;
 void main() {
     vec4 pixColor = texture(tex, v_texcoord);
 
-    if (discardOpaque && pixColor.a * alpha == 1.0)
-        discard;
-
-    if (discardAlpha && pixColor.a <= discardAlphaValue)
-        discard;
-
     // this shader shouldn't be used when skipCM == 1
-    pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimaries);
+    pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimariesXYZ);
 
     if (applyTint)
         pixColor.rgb *= tint;

src/render/shaders/glsl/CMrgbadiscard.frag

@@ -0,0 +1,44 @@
+#version 300 es
+#extension GL_ARB_shading_language_include : enable
+
+precision highp float;
+in vec2 v_texcoord;
+uniform sampler2D tex;
+
+uniform int sourceTF; // eTransferFunction
+uniform int targetTF; // eTransferFunction
+uniform mat3 targetPrimariesXYZ;
+
+uniform float alpha;
+
+uniform bool discardOpaque;
+uniform bool discardAlpha;
+uniform float discardAlphaValue;
+
+uniform bool applyTint;
+uniform vec3 tint;
+
+#include "rounding.glsl"
+#include "CM.glsl"
+
+layout(location = 0) out vec4 fragColor;
+void main() {
+    vec4 pixColor = texture(tex, v_texcoord);
+
+    if (discardOpaque && pixColor.a * alpha == 1.0)
+        discard;
+
+    if (discardAlpha && pixColor.a <= discardAlphaValue)
+        discard;
+
+    // this shader shouldn't be used when skipCM == 1
+    pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimariesXYZ);
+
+    if (applyTint)
+        pixColor.rgb *= tint;
+
+    if (radius > 0.0)
+        pixColor = rounding(pixColor);
+
+    fragColor = pixColor * alpha;
+}

src/render/shaders/glsl/CMrgbx.frag

@@ -7,14 +7,9 @@ uniform sampler2D tex;
 
 uniform int sourceTF; // eTransferFunction
 uniform int targetTF; // eTransferFunction
-uniform mat4x2 targetPrimaries;
+uniform mat3 targetPrimariesXYZ;
 
 uniform float alpha;
-
-uniform bool discardOpaque;
-uniform bool discardAlpha;
-uniform float discardAlphaValue;
-
 uniform bool applyTint;
 uniform vec3 tint;
 
@@ -25,14 +20,8 @@ layout(location = 0) out vec4 fragColor;
 void main() {
     vec4 pixColor = vec4(texture(tex, v_texcoord).rgb, 1.0);
 
-    if (discardOpaque && pixColor.a * alpha == 1.0)
-        discard;
-
-    if (discardAlpha && pixColor.a <= discardAlphaValue)
-        discard;
-
     // this shader shouldn't be used when skipCM == 1
-    pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimaries);
+    pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimariesXYZ);
 
     if (applyTint)
         pixColor.rgb *= tint;

src/render/shaders/glsl/CMrgbxdiscard.frag

@@ -0,0 +1,44 @@
+#version 300 es
+#extension GL_ARB_shading_language_include : enable
+
+precision highp float;
+in vec2 v_texcoord;
+uniform sampler2D tex;
+
+uniform int sourceTF; // eTransferFunction
+uniform int targetTF; // eTransferFunction
+uniform mat3 targetPrimariesXYZ;
+
+uniform float alpha;
+
+uniform bool discardOpaque;
+uniform bool discardAlpha;
+uniform float discardAlphaValue;
+
+uniform bool applyTint;
+uniform vec3 tint;
+
+#include "rounding.glsl"
+#include "CM.glsl"
+
+layout(location = 0) out vec4 fragColor;
+void main() {
+    vec4 pixColor = vec4(texture(tex, v_texcoord).rgb, 1.0);
+
+    if (discardOpaque && pixColor.a * alpha == 1.0)
+        discard;
+
+    if (discardAlpha && pixColor.a <= discardAlphaValue)
+        discard;
+
+    // this shader shouldn't be used when skipCM == 1
+    pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimariesXYZ);
+
+    if (applyTint)
+        pixColor.rgb *= tint;
+
+    if (radius > 0.0)
+        pixColor = rounding(pixColor);
+
+    fragColor = pixColor * alpha;
+}

src/render/shaders/glsl/shadow.frag

@@ -8,7 +8,7 @@ in vec2 v_texcoord;
 uniform int skipCM;
 uniform int sourceTF; // eTransferFunction
 uniform int targetTF; // eTransferFunction
-uniform mat4x2 targetPrimaries;
+uniform mat3 targetPrimariesXYZ;
 
 uniform vec2 topLeft;
 uniform vec2 bottomRight;
@@ -93,7 +93,7 @@ void main() {
     pixColor.rgb *= pixColor[3];
 
 	if (skipCM == 0)
-        pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimaries);
+        pixColor = doColorManagement(pixColor, sourceTF, targetTF, targetPrimariesXYZ);
 
 	fragColor = pixColor;
 }