mrq/GLdc
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

More performance work and fix bugs introduced in last commit

Browse Source
This commit is contained in:
Luke Benstead 2018-10-09 09:27:53 +01:00
parent 18759422ea
commit df44c0ea73
5 changed files with 363 additions and 81 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

320
GL/draw.c
View File

@ -460,6 +460,180 @@ static inline PolyBuildFunc _calcBuildFunc(const GLenum type) {
return &_buildStrip;
}
static inline void genElementsCommon(
ClipVertex* output,
const GLubyte* iptr, GLuint istride, GLenum type,
GLsizei count,
const GLubyte* vptr, GLuint vstride,
const GLubyte* cptr, GLuint cstride,
const GLubyte* uvptr, GLuint uvstride,
const GLubyte* stptr, GLuint ststride,
const GLubyte* nptr, GLuint nstride,
GLboolean doTexture, GLboolean doMultitexture, GLboolean doLighting
) {
const FloatParseFunc vertexFunc = _calcVertexParseFunc();
const ByteParseFunc diffuseFunc = _calcDiffuseParseFunc();
const FloatParseFunc uvFunc = _calcUVParseFunc();
const FloatParseFunc stFunc = _calcSTParseFunc();
const FloatParseFunc normalFunc = _calcNormalParseFunc();
const IndexParseFunc indexFunc = _calcParseIndexFunc(type);
GLsizei i = 0;
const GLubyte* idx = iptr;
ClipVertex* vertex = output;
for(; i < count; ++i, idx += istride, ++vertex) {
GLuint j = indexFunc(idx);
vertex->flags = PVR_CMD_VERTEX;
vertexFunc(vertex->xyz, vptr + (j * vstride));
}
idx = iptr;
vertex = output;
for(i = 0; i < count; ++i, idx += istride, ++vertex) {
GLuint j = indexFunc(idx);
diffuseFunc(vertex->bgra, cptr + (j * cstride));
}
if(doTexture) {
idx = iptr;
vertex = output;
for(i = 0; i < count; ++i, idx += istride, ++vertex) {
GLuint j = indexFunc(idx);
uvFunc(vertex->uv, uvptr + (j * uvstride));
}
}
if(doMultitexture) {
idx = iptr;
vertex = output;
for(i = 0; i < count; ++i, idx += istride, ++vertex) {
GLuint j = indexFunc(idx);
stFunc(vertex->st, stptr + (j * ststride));
}
}
if(doLighting) {
idx = iptr;
vertex = output;
for(i = 0; i < count; ++i, idx += istride, ++vertex) {
GLuint j = indexFunc(idx);
normalFunc(vertex->nxyz, nptr + (j * nstride));
}
}
}
static inline void genElementsTriangles(
ClipVertex* output,
GLsizei count,
const GLubyte* iptr, GLuint istride, GLenum type,
const GLubyte* vptr, GLuint vstride,
const GLubyte* cptr, GLuint cstride,
const GLubyte* uvptr, GLuint uvstride,
const GLubyte* stptr, GLuint ststride,
const GLubyte* nptr, GLuint nstride,
GLboolean doTexture, GLboolean doMultitexture, GLboolean doLighting) {
genElementsCommon(
output,
iptr, istride, type, count,
vptr, vstride, cptr, cstride, uvptr, uvstride, stptr, ststride, nptr, nstride,
doTexture, doMultitexture, doLighting
);
GLsizei i = 2;
for(; i < count; i += 3) {
output[i].flags = PVR_CMD_VERTEX_EOL;
}
}
static inline void genElementsQuads(
ClipVertex* output,
GLsizei count,
const GLubyte* iptr, GLuint istride, GLenum type,
const GLubyte* vptr, GLuint vstride,
const GLubyte* cptr, GLuint cstride,
const GLubyte* uvptr, GLuint uvstride,
const GLubyte* stptr, GLuint ststride,
const GLubyte* nptr, GLuint nstride,
GLboolean doTexture, GLboolean doMultitexture, GLboolean doLighting) {
genElementsCommon(
output,
iptr, istride, type, count,
vptr, vstride, cptr, cstride, uvptr, uvstride, stptr, ststride, nptr, nstride,
doTexture, doMultitexture, doLighting
);
GLsizei i = 3;
for(; i < count; i += 4) {
swapVertex(&output[i], &output[i - 1]);
output[i].flags = PVR_CMD_VERTEX_EOL;
}
}
static inline void genElementsTriangleFan(
ClipVertex* output,
GLsizei count,
const GLubyte* iptr, GLuint istride, GLenum type,
const GLubyte* vptr, GLuint vstride,
const GLubyte* cptr, GLuint cstride,
const GLubyte* uvptr, GLuint uvstride,
const GLubyte* stptr, GLuint ststride,
const GLubyte* nptr, GLuint nstride,
GLboolean doTexture, GLboolean doMultitexture, GLboolean doLighting) {
genElementsCommon(
output,
iptr, istride, type, count,
vptr, vstride, cptr, cstride, uvptr, uvstride, stptr, ststride, nptr, nstride,
doTexture, doMultitexture, doLighting
);
swapVertex(&output[1], &output[2]);
output[2].flags = PVR_CMD_VERTEX_EOL;
GLsizei i = 3;
ClipVertex* first = &output[0];
for(; i < count - 1; ++i) {
ClipVertex* next = &output[i + 1];
ClipVertex* previous = &output[i - 1];
ClipVertex* vertex = &output[i];
*next = *first;
swapVertex(next, vertex);
vertex = next + 1;
*vertex = *previous;
vertex->flags = PVR_CMD_VERTEX_EOL;
}
}
static inline void genElementsTriangleStrip(
ClipVertex* output,
GLsizei count,
const GLubyte* iptr, GLuint istride, GLenum type,
const GLubyte* vptr, GLuint vstride,
const GLubyte* cptr, GLuint cstride,
const GLubyte* uvptr, GLuint uvstride,
const GLubyte* stptr, GLuint ststride,
const GLubyte* nptr, GLuint nstride,
GLboolean doTexture, GLboolean doMultitexture, GLboolean doLighting) {
genElementsCommon(
output,
iptr, istride, type, count,
vptr, vstride, cptr, cstride, uvptr, uvstride, stptr, ststride, nptr, nstride,
doTexture, doMultitexture, doLighting
);
output[count - 1].flags = PVR_CMD_VERTEX_EOL;
}
static inline void genArraysCommon(
ClipVertex* output,
GLsizei count,
@ -566,12 +740,14 @@ static void genArraysQuads(
doTexture, doMultitexture, doLighting
);
GLsizei i = count;
ClipVertex* vertex = output;
GLsizei i = 3;
for(i = 3; i < count; i += 4) {
swapVertex(&vertex[i], &vertex[i - 1]);
vertex[i].flags = PVR_CMD_VERTEX_EOL;
for(; i < count; i += 4) {
ClipVertex* this = output + i;
ClipVertex* previous = output + (i - 1);
swapVertex(previous, this);
this->flags = PVR_CMD_VERTEX_EOL;
}
}
@ -642,13 +818,14 @@ static void generate(ClipVertex* output, const GLenum mode, const GLsizei first,
const GLuint ststride = (ST_POINTER.stride) ? ST_POINTER.stride : ST_POINTER.size * byte_size(ST_POINTER.type);
const GLuint nstride = (NORMAL_POINTER.stride) ? NORMAL_POINTER.stride : NORMAL_POINTER.size * byte_size(NORMAL_POINTER.type);
if(!indices) {
const GLubyte* vptr = VERTEX_POINTER.ptr + (first * vstride);
const GLubyte* cptr = DIFFUSE_POINTER.ptr + (first * cstride);
const GLubyte* uvptr = UV_POINTER.ptr + (first * uvstride);
const GLubyte* stptr = ST_POINTER.ptr + (first * ststride);
const GLubyte* nptr = NORMAL_POINTER.ptr + (first * nstride);
const GLubyte* vptr = VERTEX_POINTER.ptr + (first * vstride);
const GLubyte* cptr = DIFFUSE_POINTER.ptr + (first * cstride);
const GLubyte* uvptr = UV_POINTER.ptr + (first * uvstride);
const GLubyte* stptr = ST_POINTER.ptr + (first * ststride);
const GLubyte* nptr = NORMAL_POINTER.ptr + (first * nstride);
const GLsizei istride = byte_size(type);
if(!indices) {
// Drawing arrays
switch(mode) {
case GL_TRIANGLES:
@ -662,6 +839,7 @@ static void generate(ClipVertex* output, const GLenum mode, const GLsizei first,
nptr, nstride,
doTexture, doMultitexture, doLighting
);
break;
case GL_QUADS:
genArraysQuads(
output,
@ -673,6 +851,7 @@ static void generate(ClipVertex* output, const GLenum mode, const GLsizei first,
nptr, nstride,
doTexture, doMultitexture, doLighting
);
break;
case GL_TRIANGLE_FAN:
genArraysTriangleFan(
output,
@ -684,6 +863,7 @@ static void generate(ClipVertex* output, const GLenum mode, const GLsizei first,
nptr, nstride,
doTexture, doMultitexture, doLighting
);
break;
case GL_TRIANGLE_STRIP:
default:
genArraysTriangleStrip(
@ -697,76 +877,54 @@ static void generate(ClipVertex* output, const GLenum mode, const GLsizei first,
doTexture, doMultitexture, doLighting
);
}
return;
}
const GLsizei max = first + count;
ClipVertex* vertex = output;
const FloatParseFunc vertexFunc = _calcVertexParseFunc();
const ByteParseFunc diffuseFunc = _calcDiffuseParseFunc();
const FloatParseFunc uvFunc = _calcUVParseFunc();
const FloatParseFunc stFunc = _calcSTParseFunc();
const FloatParseFunc normalFunc = _calcNormalParseFunc();
const PolyBuildFunc buildFunc = _calcBuildFunc(mode);
const IndexParseFunc indexFunc = _calcParseIndexFunc(type);
const GLsizei type_byte_size = byte_size(type);
ClipVertex* previous = NULL;
ClipVertex* firstV = vertex;
ClipVertex* next = NULL;
ClipVertex* target = NULL;
GLsizei i, j = 0;
GLuint idx;
for(i = first; i < max; ++i, ++j, ++vertex) {
if(mode == GL_QUADS) {
/* Performance optimisation to prevent copying to a temporary */
GLsizei mod = (j + 1) % 4;
if(mod == 0) {
target = vertex - 1;
target->flags = PVR_CMD_VERTEX;
} else if(mod == 3) {
target = vertex + 1;
target->flags = PVR_CMD_VERTEX_EOL;
} else {
target = vertex;
target->flags = PVR_CMD_VERTEX;
}
} else {
target = vertex;
target->flags = PVR_CMD_VERTEX;
}
idx = (indices) ?
indexFunc(&indices[type_byte_size * i]) : i;
vertexFunc(target->xyz, VERTEX_POINTER.ptr + (idx * vstride));
diffuseFunc(target->bgra, DIFFUSE_POINTER.ptr + (idx * cstride));
if(doTexture) {
uvFunc(target->uv, UV_POINTER.ptr + (idx * uvstride));
}
if(doMultitexture) {
stFunc(target->st, ST_POINTER.ptr + (idx * ststride));
}
if(doLighting) {
normalFunc(target->nxyz, NORMAL_POINTER.ptr + (idx * nstride));
}
if(mode != GL_QUADS) {
next = (j < count - 1) ? vertex + 1 : NULL;
previous = (j > 0) ? vertex - 1 : NULL;
buildFunc(firstV, previous, vertex, next, j);
}
} else if(mode == GL_TRIANGLES) {
genElementsTriangles(
output,
count,
indices, istride, type,
vptr, vstride,
cptr, cstride,
uvptr, uvstride,
stptr, ststride,
nptr, nstride,
doTexture, doMultitexture, doLighting
);
} else if(mode == GL_QUADS) {
genElementsQuads(
output,
count,
indices, istride, type,
vptr, vstride,
cptr, cstride,
uvptr, uvstride,
stptr, ststride,
nptr, nstride,
doTexture, doMultitexture, doLighting
);
} else if(mode == GL_TRIANGLE_FAN) {
genElementsTriangleFan(
output,
count,
indices, istride, type,
vptr, vstride,
cptr, cstride,
uvptr, uvstride,
stptr, ststride,
nptr, nstride,
doTexture, doMultitexture, doLighting
);
} else {
genElementsTriangleStrip(
output,
count,
indices, istride, type,
vptr, vstride,
cptr, cstride,
uvptr, uvstride,
stptr, ststride,
nptr, nstride,
doTexture, doMultitexture, doLighting
);
}
}

1
samples/Makefile
View File

@ -9,6 +9,7 @@ export OBJEXTRA := $(LIB_DIR)/libGLdc.a
all:
$(KOS_MAKE) -C nehe02 all
$(KOS_MAKE) -C nehe02va all
$(KOS_MAKE) -C nehe02de all
$(KOS_MAKE) -C nehe03 all
$(KOS_MAKE) -C nehe06 all
$(KOS_MAKE) -C nehe06_vq all

29
samples/nehe02de/Makefile Normal file
View File

@ -0,0 +1,29 @@
TARGET = nehe02de.elf
OBJS = main.o
all: rm-elf $(TARGET)
include $(KOS_BASE)/Makefile.rules
clean:
-rm -f $(TARGET) $(OBJS) romdisk.*
rm-elf:
-rm -f $(TARGET) romdisk.*
$(TARGET): $(OBJS) romdisk.o
$(KOS_CC) $(KOS_CFLAGS) $(KOS_LDFLAGS) -o $(TARGET) $(KOS_START) \
$(OBJS) romdisk.o $(OBJEXTRA) -lm -lkosutils $(KOS_LIBS)
romdisk.img:
$(KOS_GENROMFS) -f romdisk.img -d romdisk -v
romdisk.o: romdisk.img
$(KOS_BASE)/utils/bin2o/bin2o romdisk.img romdisk romdisk.o
run: $(TARGET)
$(KOS_LOADER) $(TARGET)
dist:
rm -f $(OBJS) romdisk.o romdisk.img
$(KOS_STRIP) $(TARGET)

94
samples/nehe02de/main.c Normal file
View File

@ -0,0 +1,94 @@
#include "gl.h"
#include "glu.h"
#include "glkos.h"
/* A general OpenGL initialization function. Sets all of the initial parameters. */
void InitGL(int Width, int Height) // We call this right after our OpenGL window is created.
{
glClearColor(0.0f, 0.0f, 0.0f, 0.0f); // This Will Clear The Background Color To Black
glClearDepth(1.0); // Enables Clearing Of The Depth Buffer
glDepthFunc(GL_LESS); // The Type Of Depth Test To Do
glEnable(GL_DEPTH_TEST); // Enables Depth Testing
glShadeModel(GL_SMOOTH); // Enables Smooth Color Shading
glEnableClientState(GL_VERTEX_ARRAY); // Enable vertex arrays
glMatrixMode(GL_PROJECTION);
glLoadIdentity(); // Reset The Projection Matrix
gluPerspective(45.0f,(GLfloat)Width/(GLfloat)Height,0.1f,100.0f); // Calculate The Aspect Ratio Of The Window
glMatrixMode(GL_MODELVIEW);
}
/* The function called when our window is resized (which shouldn't happen, because we're fullscreen) */
void ReSizeGLScene(int Width, int Height)
{
if (Height == 0) // Prevent A Divide By Zero If The Window Is Too Small
Height = 1;
glViewport(0, 0, Width, Height); // Reset The Current Viewport And Perspective Transformation
glMatrixMode(GL_PROJECTION);
glLoadIdentity();
gluPerspective(45.0f,(GLfloat)Width/(GLfloat)Height,0.1f,100.0f);
glMatrixMode(GL_MODELVIEW);
}
/* The main drawing function. */
void DrawGLScene()
{
const GLfloat triangle [] = {
0.0f, 1.0f, 0.0f,
1.0f, -1.0f, 0.0f,
-1.0f, -1.0f, 0.0f
};
const GLfloat square [] = {
-1.0f, 1.0f, 0.0f,
1.0f, 1.0f, 0.0f,
1.0f, -1.0f, 0.0f,
-1.0f, -1.0f, 0.0f
};
const GLuint triangleIdx [] = {
0, 1, 2
};
const GLuint squareIdx [] = {
0, 1, 2, 3
};
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT); // Clear The Screen And The Depth Buffer
glLoadIdentity(); // Reset The View
glTranslatef(-1.5f,0.0f,-6.0f); // Move Left 1.5 Units And Into The Screen 6.0
glVertexPointer(3, GL_FLOAT, 0, triangle);
glDrawElements(GL_TRIANGLES, 3, GL_UNSIGNED_INT, triangleIdx);
glTranslatef(3.0f,0.0f,0.0f); // Move Right 3 Units
glVertexPointer(3, GL_FLOAT, 0, square);
glDrawElements(GL_QUADS, 4, GL_UNSIGNED_INT, squareIdx);
// swap buffers to display, since we're double buffered.
glKosSwapBuffers();
}
int main(int argc, char **argv)
{
glKosInit();
InitGL(640, 480);
ReSizeGLScene(640, 480);
while(1) {
DrawGLScene();
}
return 0;
}

0
samples/nehe02de/romdisk/PLACEHOLDER Normal file
View File