Reuh/ctruLua
1
0
Fork 0
mirror of https://github.com/ctruLua/ctruLua.git synced 2025-10-28 16:59:30 +00:00
Code Releases Activity

Updated all the libs, added citro3d, added ctr.swkbd (WIP, untested)

Browse source
This commit is contained in:
Firew0lf 2016-08-05 17:30:24 +02:00
parent 68a44645f7
commit 49c87e5526
97 changed files with 7341 additions and 944 deletions
Download patch file Download diff file Expand all files Collapse all files

267
libs/sf2dlib/libsf2d/source/sf2d.c
View file

@ -1,22 +1,13 @@
#include <string.h>
#include "sf2d.h"
#include "sf2d_private.h"
#include "shader_vsh_shbin.h"
#include "shader_shbin.h"
static int sf2d_initialized = 0;
static u32 clear_color = 0;
static u32 *gpu_cmd = NULL;
//GPU init variables
static int gpu_cmd_size = 0;
// Temporary memory pool
static void *pool_addr = NULL;
static u32 pool_index = 0;
static u32 pool_size = 0;
//GPU framebuffer address
static u32 *gpu_fb_addr = NULL;
//GPU depth buffer address
static u32 *gpu_depth_fb_addr = NULL;
//VBlank wait
static int vblank_wait = 1;
//FPS calculation
@ -29,19 +20,14 @@ static gfx3dSide_t cur_side = GFX_LEFT;
//Shader stuff
static DVLB_s *dvlb = NULL;
static shaderProgram_s shader;
static u32 projection_desc = -1;
//Matrix
static float ortho_matrix_top[4*4];
static float ortho_matrix_bot[4*4];
static int projection_desc = -1;
static int transform_desc = -1;
static int useTransform_desc = -1;
//Rendertarget things
static sf2d_rendertarget * currentRenderTarget = NULL;
static void * targetDepthBuffer;
static int targetDepthBufferLen = 0;
//Apt hook cookie
static aptHookCookie apt_hook_cookie;
//Functions
static void apt_hook_func(APT_HookType hook, void *param);
static void reset_gpu_apt_resume();
static sf2d_rendertarget * targetTopLeft;
static sf2d_rendertarget * targetTopRight;
static sf2d_rendertarget * targetBottom;
static int in_render;
int sf2d_init()
{
@ -54,47 +40,43 @@ int sf2d_init_advanced(int gpucmd_size, int temppool_size)
{
if (sf2d_initialized) return 0;
gpu_fb_addr = vramMemAlign(400*240*8, 0x100);
gpu_depth_fb_addr = vramMemAlign(400*240*8, 0x100);
gpu_cmd = linearAlloc(gpucmd_size * 4);
pool_addr = linearAlloc(temppool_size);
pool_size = temppool_size;
gpu_cmd_size = gpucmd_size;
gfxInitDefault();
GPU_Init(NULL);
gfxSet3D(false);
GPU_Reset(NULL, gpu_cmd, gpucmd_size);
C3D_Init(gpucmd_size*8);
//Setup rendertargets
targetTopLeft = sf2d_create_rendertarget(400, 240);
targetTopRight = sf2d_create_rendertarget(400, 240);
targetBottom = sf2d_create_rendertarget(320, 240);
sf2d_set_clear_color(0);
C3D_RenderTargetSetOutput(targetTopLeft->target, GFX_TOP, GFX_LEFT, 0x1000);
C3D_RenderTargetSetOutput(targetTopRight->target, GFX_TOP, GFX_RIGHT, 0x1000);
C3D_RenderTargetSetOutput(targetBottom->target, GFX_BOTTOM, GFX_LEFT, 0x1000);
//Setup temp pool
pool_addr = linearAlloc(temppool_size);
pool_size = temppool_size;
//Setup the shader
dvlb = DVLB_ParseFile((u32 *)shader_vsh_shbin, shader_vsh_shbin_size);
dvlb = DVLB_ParseFile((u32 *)shader_shbin, shader_shbin_size);
shaderProgramInit(&shader);
shaderProgramSetVsh(&shader, &dvlb->DVLE[0]);
//Get shader uniform descriptors
projection_desc = shaderInstanceGetUniformLocation(shader.vertexShader, "projection");
transform_desc = shaderInstanceGetUniformLocation(shader.vertexShader, "transform");
useTransform_desc = shaderInstanceGetUniformLocation(shader.vertexShader, "useTransform");
shaderProgramUse(&shader);
matrix_init_orthographic(ortho_matrix_top, 0.0f, 400.0f, 0.0f, 240.0f, 0.0f, 1.0f);
matrix_init_orthographic(ortho_matrix_bot, 0.0f, 320.0f, 0.0f, 240.0f, 0.0f, 1.0f);
matrix_gpu_set_uniform(ortho_matrix_top, projection_desc);
//Register the apt callback hook
aptHook(&apt_hook_cookie, apt_hook_func, NULL);
C3D_BindProgram(&shader);
C3D_CullFace(GPU_CULL_NONE);
C3D_DepthTest(true, GPU_GEQUAL, GPU_WRITE_ALL);
C3D_BoolUnifSet(GPU_VERTEX_SHADER, useTransform_desc, false);
vblank_wait = 1;
current_fps = 0.0f;
frames = 0;
last_time = osGetTime();
cur_screen = GFX_TOP;
cur_side = GFX_LEFT;
GPUCMD_Finalize();
GPUCMD_FlushAndRun();
gspWaitForP3D();
sf2d_pool_reset();
sf2d_initialized = 1;
@ -106,17 +88,12 @@ int sf2d_fini()
{
if (!sf2d_initialized) return 0;
aptUnhook(&apt_hook_cookie);
linearFree(pool_addr);
gfxExit();
shaderProgramFree(&shader);
DVLB_Free(dvlb);
linearFree(pool_addr);
linearFree(gpu_cmd);
vramFree(gpu_fb_addr);
vramFree(gpu_depth_fb_addr);
linearFree(targetDepthBuffer);
C3D_Fini();
gfxExit();
sf2d_initialized = 0;
@ -128,143 +105,54 @@ void sf2d_set_3D(int enable)
gfxSet3D(enable);
}
void sf2d_set_transform(C3D_Mtx* mtx)
{
C3D_BoolUnifSet(GPU_VERTEX_SHADER, useTransform_desc, mtx != NULL);
if (mtx) {
C3D_FVUnifMtx4x4(GPU_VERTEX_SHADER, transform_desc, mtx);
}
}
void sf2d_start_frame(gfxScreen_t screen, gfx3dSide_t side)
{
sf2d_pool_reset();
GPUCMD_SetBufferOffset(0);
cur_screen = screen;
cur_side = side;
// Only upload the uniform if the screen changes
if (screen != cur_screen) {
if (screen == GFX_TOP) {
matrix_gpu_set_uniform(ortho_matrix_top, projection_desc);
} else {
matrix_gpu_set_uniform(ortho_matrix_bot, projection_desc);
}
cur_screen = screen;
}
int screen_w;
if (screen == GFX_TOP) {
screen_w = 400;
cur_side = side;
if (side == GFX_LEFT) {
sf2d_start_frame_target(targetTopLeft);
} else {
sf2d_start_frame_target(targetTopRight);
}
} else {
screen_w = 320;
sf2d_start_frame_target(targetBottom);
}
GPU_SetViewport((u32 *)osConvertVirtToPhys(gpu_depth_fb_addr),
(u32 *)osConvertVirtToPhys(gpu_fb_addr),
0, 0, 240, screen_w);
GPU_DepthMap(-1.0f, 0.0f);
GPU_SetFaceCulling(GPU_CULL_NONE);
GPU_SetStencilTest(false, GPU_ALWAYS, 0x00, 0xFF, 0x00);
GPU_SetStencilOp(GPU_STENCIL_KEEP, GPU_STENCIL_KEEP, GPU_STENCIL_KEEP);
GPU_SetBlendingColor(0,0,0,0);
GPU_SetDepthTestAndWriteMask(true, GPU_GEQUAL, GPU_WRITE_ALL);
GPUCMD_AddMaskedWrite(GPUREG_EARLYDEPTH_TEST1, 0x1, 0);
GPUCMD_AddWrite(GPUREG_EARLYDEPTH_TEST2, 0);
GPU_SetAlphaBlending(
GPU_BLEND_ADD,
GPU_BLEND_ADD,
GPU_SRC_ALPHA, GPU_ONE_MINUS_SRC_ALPHA,
GPU_ONE, GPU_ZERO
);
GPU_SetAlphaTest(false, GPU_ALWAYS, 0x00);
GPU_SetDummyTexEnv(1);
GPU_SetDummyTexEnv(2);
GPU_SetDummyTexEnv(3);
GPU_SetDummyTexEnv(4);
GPU_SetDummyTexEnv(5);
}
void sf2d_start_frame_target(sf2d_rendertarget *target)
{
sf2d_pool_reset();
GPUCMD_SetBufferOffset(0);
// Upload saved uniform
matrix_gpu_set_uniform(target->projection, projection_desc);
int bufferLen = target->texture.width * target->texture.height * 4; // apparently depth buffer is (or can be) 32bit?
if (bufferLen > targetDepthBufferLen) { // expand depth buffer
if (targetDepthBufferLen > 0) linearFree(targetDepthBuffer);
targetDepthBuffer = linearAlloc(bufferLen);
memset(targetDepthBuffer, 0, bufferLen);
targetDepthBufferLen = bufferLen;
if (!in_render) {
sf2d_pool_reset();
C3D_FrameBegin(vblank_wait ? C3D_FRAME_SYNCDRAW : 0);
in_render = 1;
}
GPU_SetViewport((u32 *)osConvertVirtToPhys(targetDepthBuffer),
(u32 *)osConvertVirtToPhys(target->texture.data),
0, 0, target->texture.height, target->texture.width);
currentRenderTarget = target;
GPU_DepthMap(-1.0f, 0.0f);
GPU_SetFaceCulling(GPU_CULL_NONE);
GPU_SetStencilTest(false, GPU_ALWAYS, 0x00, 0xFF, 0x00);
GPU_SetStencilOp(GPU_STENCIL_KEEP, GPU_STENCIL_KEEP, GPU_STENCIL_KEEP);
GPU_SetBlendingColor(0,0,0,0);
GPU_SetDepthTestAndWriteMask(true, GPU_GEQUAL, GPU_WRITE_ALL);
GPUCMD_AddMaskedWrite(GPUREG_EARLYDEPTH_TEST1, 0x1, 0);
GPUCMD_AddWrite(GPUREG_EARLYDEPTH_TEST2, 0);
GPU_SetAlphaBlending(
GPU_BLEND_ADD,
GPU_BLEND_ADD,
GPU_SRC_ALPHA, GPU_ONE_MINUS_SRC_ALPHA,
GPU_ONE, GPU_ZERO
);
GPU_SetAlphaTest(false, GPU_ALWAYS, 0x00);
GPU_SetDummyTexEnv(1);
GPU_SetDummyTexEnv(2);
GPU_SetDummyTexEnv(3);
GPU_SetDummyTexEnv(4);
GPU_SetDummyTexEnv(5);
C3D_FrameDrawOn(target->target);
C3D_FVUnifMtx4x4(GPU_VERTEX_SHADER, projection_desc, &target->projection);
}
void sf2d_end_frame()
{
GPU_FinishDrawing();
GPUCMD_Finalize();
GPUCMD_FlushAndRun();
gspWaitForP3D();
if (!currentRenderTarget) {
//Copy the GPU rendered FB to the screen FB
if (cur_screen == GFX_TOP) {
GX_DisplayTransfer(gpu_fb_addr, GX_BUFFER_DIM(240, 400),
(u32 *)gfxGetFramebuffer(GFX_TOP, cur_side, NULL, NULL),
GX_BUFFER_DIM(240, 400), 0x1000);
} else {
GX_DisplayTransfer(gpu_fb_addr, GX_BUFFER_DIM(240, 320),
(u32 *)gfxGetFramebuffer(GFX_BOTTOM, GFX_LEFT, NULL, NULL),
GX_BUFFER_DIM(240, 320), 0x1000);
}
gspWaitForPPF();
//Clear the screen
GX_MemoryFill(
gpu_fb_addr, clear_color, &gpu_fb_addr[240*400], GX_FILL_TRIGGER | GX_FILL_32BIT_DEPTH,
gpu_depth_fb_addr, 0, &gpu_depth_fb_addr[240*400], GX_FILL_TRIGGER | GX_FILL_32BIT_DEPTH);
gspWaitForPSC0();
} else {
//gspWaitForPPF();
//gspWaitForPSC0();
sf2d_texture_tile32(&(currentRenderTarget->texture));
}
currentRenderTarget = NULL;
// Nothing
}
void sf2d_swapbuffers()
{
gfxSwapBuffersGpu();
if (vblank_wait) {
gspWaitForEvent(GSPGPU_EVENT_VBlank0, false);
}
if (!in_render) return;
in_render = 0;
C3D_FrameEnd(0);
//Calculate FPS
frames++;
u64 delta_time = osGetTime() - last_time;
@ -323,19 +211,17 @@ void sf2d_pool_reset()
void sf2d_set_clear_color(u32 color)
{
// GX_SetMemoryFill wants the color inverted?
clear_color = RGBA8_GET_R(color) << 24 |
RGBA8_GET_G(color) << 16 |
RGBA8_GET_B(color) << 8 |
RGBA8_GET_A(color) << 0;
sf2d_clear_target(targetTopLeft, color);
sf2d_clear_target(targetTopRight, color);
sf2d_clear_target(targetBottom, color);
}
void sf2d_set_scissor_test(GPU_SCISSORMODE mode, u32 x, u32 y, u32 w, u32 h)
{
if (cur_screen == GFX_TOP) {
GPU_SetScissorTest(mode, 240 - (y + h), 400 - (x + w), 240 - y, 400 - x);
C3D_SetScissor(mode, 240 - (y + h), 400 - (x + w), 240 - y, 400 - x);
} else {
GPU_SetScissorTest(mode, 240 - (y + h), 320 - (x + w), 240 - y, 320 - x);
C3D_SetScissor(mode, 240 - (y + h), 320 - (x + w), 240 - y, 320 - x);
}
}
@ -348,26 +234,3 @@ gfx3dSide_t sf2d_get_current_side()
{
return cur_side;
}
static void apt_hook_func(APT_HookType hook, void *param)
{
if (hook == APTHOOK_ONRESTORE) {
reset_gpu_apt_resume();
}
}
static void reset_gpu_apt_resume()
{
GPU_Reset(NULL, gpu_cmd, gpu_cmd_size); // Only required for custom GPU cmd sizes
shaderProgramUse(&shader);
if (cur_screen == GFX_TOP) {
matrix_gpu_set_uniform(ortho_matrix_top, projection_desc);
} else {
matrix_gpu_set_uniform(ortho_matrix_bot, projection_desc);
}
GPUCMD_Finalize();
GPUCMD_FlushAndRun();
gspWaitForP3D();
}

71
libs/sf2dlib/libsf2d/source/sf2d_draw.c
View file

@ -1,5 +1,4 @@
#include "sf2d.h"
#include "sf2d_private.h"
#include <math.h>
#ifndef M_PI
@ -7,27 +6,19 @@
#endif
void sf2d_setup_env_internal(const sf2d_vertex_pos_col* vertices) {
GPU_SetTexEnv(
0,
GPU_TEVSOURCES(GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR),
GPU_TEVSOURCES(GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR),
GPU_TEVOPERANDS(0, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_REPLACE, GPU_REPLACE,
0xFFFFFFFF
);
C3D_TexEnv* env = C3D_GetTexEnv(0);
C3D_TexEnvSrc(env, C3D_Both, GPU_PRIMARY_COLOR, 0, 0);
C3D_TexEnvOp(env, C3D_Both, 0, 0, 0);
C3D_TexEnvFunc(env, C3D_Both, GPU_REPLACE);
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 4, GPU_UNSIGNED_BYTE),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_UNSIGNED_BYTE, 4);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_col), 2, 0x10);
}
void sf2d_draw_line(float x0, float y0, float x1, float y1, float width, u32 color)
@ -64,21 +55,21 @@ void sf2d_draw_line(float x0, float y0, float x1, float y1, float width, u32 col
sf2d_setup_env_internal(vertices);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_rectangle_internal(const sf2d_vertex_pos_col *vertices)
{
sf2d_setup_env_internal(vertices);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_triangle_internal(const sf2d_vertex_pos_col *vertices)
{
sf2d_setup_env_internal(vertices);
GPU_DrawArray(GPU_TRIANGLES, 0, 3);
C3D_DrawArrays(GPU_TRIANGLES, 0, 3);
}
void sf2d_draw_rectangle(int x, int y, int w, int h, u32 color)
@ -133,17 +124,14 @@ void sf2d_draw_rectangle_rotate(int x, int y, int w, int h, u32 color, float rad
vertices[2].color = vertices[0].color;
vertices[3].color = vertices[0].color;
float m[4*4];
matrix_set_z_rotation(m, rad);
sf2d_vector_3f rot[4];
int i;
for (i = 0; i < 4; i++) {
vector_mult_matrix4x4(m, &vertices[i].position, &rot[i]);
vertices[i].position = (sf2d_vector_3f){rot[i].x + x + w2, rot[i].y + y + h2, rot[i].z};
}
C3D_Mtx m;
Mtx_Identity(&m);
Mtx_Translate(&m, x+w2, y+h2, 0, true);
Mtx_RotateZ(&m, rad, true);
sf2d_set_transform(&m);
sf2d_draw_rectangle_internal(vertices);
sf2d_set_transform(NULL);
}
void sf2d_draw_rectangle_gradient(int x, int y, int w, int h, u32 color1, u32 color2, sf2d_gradient_dir direction)
@ -182,17 +170,14 @@ void sf2d_draw_rectangle_gradient_rotate(int x, int y, int w, int h, u32 color1,
vertices[2].color = (direction == SF2D_LEFT_TO_RIGHT) ? color1 : color2;
vertices[3].color = color2;
float m[4*4];
matrix_set_z_rotation(m, rad);
sf2d_vector_3f rot[4];
int i;
for (i = 0; i < 4; i++) {
vector_mult_matrix4x4(m, &vertices[i].position, &rot[i]);
vertices[i].position = (sf2d_vector_3f){rot[i].x + x + w2, rot[i].y + y + h2, rot[i].z};
}
C3D_Mtx m;
Mtx_Identity(&m);
Mtx_Translate(&m, x+w2, y+h2, 0, true);
Mtx_RotateZ(&m, rad, true);
sf2d_set_transform(&m);
sf2d_draw_rectangle_internal(vertices);
sf2d_set_transform(NULL);
}
void sf2d_draw_fill_circle(int x, int y, int radius, u32 color)
@ -227,5 +212,5 @@ void sf2d_draw_fill_circle(int x, int y, int radius, u32 color)
sf2d_setup_env_internal(vertices);
GPU_DrawArray(GPU_TRIANGLE_FAN, 0, num_segments + 2);
C3D_DrawArrays(GPU_TRIANGLE_FAN, 0, num_segments + 2);
}

150
libs/sf2dlib/libsf2d/source/sf2d_private.c
View file

@ -1,150 +0,0 @@
#include <string.h>
#include <math.h>
#include "sf2d_private.h"
#ifndef M_PI
#define M_PI (3.14159265358979323846)
#endif
//stolen from staplebutt
void GPU_SetDummyTexEnv(u8 num)
{
GPU_SetTexEnv(num,
GPU_TEVSOURCES(GPU_PREVIOUS, 0, 0),
GPU_TEVSOURCES(GPU_PREVIOUS, 0, 0),
GPU_TEVOPERANDS(0,0,0),
GPU_TEVOPERANDS(0,0,0),
GPU_REPLACE,
GPU_REPLACE,
0xFFFFFFFF);
}
void vector_mult_matrix4x4(const float *msrc, const sf2d_vector_3f *vsrc, sf2d_vector_3f *vdst)
{
vdst->x = msrc[0*4 + 0]*vsrc->x + msrc[0*4 + 1]*vsrc->y + msrc[0*4 + 2]*vsrc->z + msrc[0*4 + 3];
vdst->y = msrc[1*4 + 0]*vsrc->x + msrc[1*4 + 1]*vsrc->y + msrc[1*4 + 2]*vsrc->z + msrc[1*4 + 3];
vdst->z = msrc[2*4 + 0]*vsrc->x + msrc[2*4 + 1]*vsrc->y + msrc[2*4 + 2]*vsrc->z + msrc[2*4 + 3];
}
void matrix_gpu_set_uniform(const float *m, u32 startreg)
{
GPU_SetFloatUniform(GPU_VERTEX_SHADER, startreg, (u32 *)m, 4);
}
void matrix_copy(float *dst, const float *src)
{
memcpy(dst, src, sizeof(float)*4*4);
}
void matrix_identity4x4(float *m)
{
m[0] = m[5] = m[10] = m[15] = 1.0f;
m[1] = m[2] = m[3] = 0.0f;
m[4] = m[6] = m[7] = 0.0f;
m[8] = m[9] = m[11] = 0.0f;
m[12] = m[13] = m[14] = 0.0f;
}
void matrix_mult4x4(const float *src1, const float *src2, float *dst)
{
int i, j, k;
for (i = 0; i < 4; i++) {
for (j = 0; j < 4; j++) {
dst[i*4 + j] = 0.0f;
for (k = 0; k < 4; k++) {
dst[i*4 + j] += src1[i*4 + k]*src2[k*4 + j];
}
}
}
}
void matrix_set_z_rotation(float *m, float rad)
{
float c = cosf(rad);
float s = sinf(rad);
matrix_identity4x4(m);
m[0] = c;
m[1] = -s;
m[4] = s;
m[5] = c;
}
void matrix_rotate_z(float *m, float rad)
{
float mr[4*4], mt[4*4];
matrix_set_z_rotation(mr, rad);
matrix_mult4x4(mr, m, mt);
matrix_copy(m, mt);
}
void matrix_set_scaling(float *m, float x_scale, float y_scale, float z_scale)
{
matrix_identity4x4(m);
m[0] = x_scale;
m[5] = y_scale;
m[10] = z_scale;
}
void matrix_swap_xy(float *m)
{
float ms[4*4], mt[4*4];
matrix_identity4x4(ms);
ms[0] = 0.0f;
ms[1] = 1.0f;
ms[4] = 1.0f;
ms[5] = 0.0f;
matrix_mult4x4(ms, m, mt);
matrix_copy(m, mt);
}
void matrix_init_orthographic(float *m, float left, float right, float bottom, float top, float near, float far)
{
float mo[4*4], mp[4*4];
mo[0x0] = 2.0f/(right-left);
mo[0x1] = 0.0f;
mo[0x2] = 0.0f;
mo[0x3] = -(right+left)/(right-left);
mo[0x4] = 0.0f;
mo[0x5] = 2.0f/(top-bottom);
mo[0x6] = 0.0f;
mo[0x7] = -(top+bottom)/(top-bottom);
mo[0x8] = 0.0f;
mo[0x9] = 0.0f;
mo[0xA] = -2.0f/(far-near);
mo[0xB] = (far+near)/(far-near);
mo[0xC] = 0.0f;
mo[0xD] = 0.0f;
mo[0xE] = 0.0f;
mo[0xF] = 1.0f;
matrix_identity4x4(mp);
mp[0xA] = 0.5;
mp[0xB] = -0.5;
//Convert Z [-1, 1] to [-1, 0] (PICA shiz)
matrix_mult4x4(mp, mo, m);
// Rotate 180 degrees
matrix_rotate_z(m, M_PI);
// Swap X and Y axis
matrix_swap_xy(m);
}
//Grabbed from: http://graphics.stanford.edu/~seander/bithacks.html#RoundUpPowerOf2
unsigned int next_pow2(unsigned int v)
{
v--;
v |= v >> 1;
v |= v >> 2;
v |= v >> 4;
v |= v >> 8;
v |= v >> 16;
return v+1;
}

559
libs/sf2dlib/libsf2d/source/sf2d_texture.c
View file

@ -2,7 +2,6 @@
#include <string.h>
#include <math.h>
#include "sf2d.h"
#include "sf2d_private.h"
#ifndef M_PI
#define M_PI (3.14159265358979323846)
@ -10,159 +9,98 @@
#define TEX_MIN_SIZE 32
static unsigned int nibbles_per_pixel(sf2d_texfmt format)
//Grabbed from: http://graphics.stanford.edu/~seander/bithacks.html#RoundUpPowerOf2
unsigned int next_pow2(unsigned int v)
{
switch (format) {
case TEXFMT_RGBA8:
return 8;
case TEXFMT_RGB8:
return 6;
case TEXFMT_RGB5A1:
case TEXFMT_RGB565:
case TEXFMT_RGBA4:
case TEXFMT_IA8:
return 4;
case TEXFMT_A4:
return 1;
case TEXFMT_I8:
case TEXFMT_A8:
case TEXFMT_IA4:
default:
return 2;
}
}
static int calc_buffer_size(sf2d_texfmt pixel_format, int width, int height)
{
return width * height * (nibbles_per_pixel(pixel_format)>>1);
v--;
v |= v >> 1;
v |= v >> 2;
v |= v >> 4;
v |= v >> 8;
v |= v >> 16;
v++;
return v >= TEX_MIN_SIZE ? v : TEX_MIN_SIZE;
}
sf2d_texture *sf2d_create_texture(int width, int height, sf2d_texfmt pixel_format, sf2d_place place)
{
int pow2_w = next_pow2(width);
int pow2_h = next_pow2(height);
if (pow2_w < TEX_MIN_SIZE) pow2_w = TEX_MIN_SIZE;
if (pow2_h < TEX_MIN_SIZE) pow2_h = TEX_MIN_SIZE;
int data_size = calc_buffer_size(pixel_format, pow2_w, pow2_h);
void *data;
sf2d_texture *texture = calloc(1, sizeof(*texture));
if (!texture) return NULL;
bool success = false;
texture->width = width;
texture->height = height;
if (place == SF2D_PLACE_RAM) {
// If there's not enough linear heap space, return
if (linearSpaceFree() < data_size) {
return NULL;
}
data = linearMemAlign(data_size, 0x80);
success = C3D_TexInit(&texture->tex, next_pow2(width), next_pow2(height), pixel_format);
} else if (place == SF2D_PLACE_VRAM) {
// If there's not enough VRAM heap space, return
if (vramSpaceFree() < data_size) {
return NULL;
}
data = vramMemAlign(data_size, 0x80);
} else if (place == SF2D_PLACE_TEMP) {
if (sf2d_pool_space_free() < data_size) {
return NULL;
}
data = sf2d_pool_memalign(data_size, 0x80);
} else {
//wot?
success = C3D_TexInitVRAM(&texture->tex, next_pow2(width), next_pow2(height), pixel_format);
}
if (!success) {
free(texture);
return NULL;
}
sf2d_texture *texture = malloc(sizeof(*texture));
texture->tiled = 0;
texture->place = place;
texture->pixel_format = pixel_format;
texture->params = GPU_TEXTURE_MAG_FILTER(GPU_NEAREST)
| GPU_TEXTURE_MIN_FILTER(GPU_NEAREST)
| GPU_TEXTURE_WRAP_S(GPU_CLAMP_TO_BORDER)
| GPU_TEXTURE_WRAP_T(GPU_CLAMP_TO_BORDER);
texture->width = width;
texture->height = height;
texture->pow2_w = pow2_w;
texture->pow2_h = pow2_h;
texture->data_size = data_size;
texture->data = data;
if (place == SF2D_PLACE_VRAM) {
GX_MemoryFill(texture->data, 0x00000000, (u32*)&((u8*)texture->data)[texture->data_size], GX_FILL_TRIGGER | GX_FILL_32BIT_DEPTH,
NULL, 0x00000000, NULL, 0);
gspWaitForPSC0();
} else {
memset(texture->data, 0, texture->data_size);
}
C3D_TexSetWrap(&texture->tex, GPU_CLAMP_TO_BORDER, GPU_CLAMP_TO_BORDER);
return texture;
}
sf2d_rendertarget *sf2d_create_rendertarget(int width, int height)
{
sf2d_texture *tx = sf2d_create_texture(width, height, TEXFMT_RGBA8, SF2D_PLACE_RAM);
sf2d_rendertarget *rt = malloc(sizeof(*rt));
//memcpy(rt, tx, sizeof(*tx));
rt->texture = *tx;
free(tx);
//tx = * rt->texture;
//rt->projection
sf2d_rendertarget *rt = calloc(1, sizeof(*rt));
if (!rt) return NULL;
matrix_init_orthographic(rt->projection, 0.0f, width, height, 0.0f, 0.0f, 1.0f);
matrix_rotate_z(rt->projection, M_PI / 2.0f);
rt->target = C3D_RenderTargetCreate(height, width, GPU_RB_RGBA8, GPU_RB_DEPTH24_STENCIL8);
if (!rt->target)
{
free(rt);
return NULL;
}
Mtx_OrthoTilt(&rt->projection, 0.0f, width, height, 0.0f, 0.0f, 1.0f, true);
return rt;
}
void sf2d_free_texture(sf2d_texture *texture)
{
if (texture) {
if (texture->place == SF2D_PLACE_RAM) {
linearFree(texture->data);
} else if (texture->place == SF2D_PLACE_VRAM) {
vramFree(texture->data);
}
C3D_TexDelete(&texture->tex);
free(texture);
}
}
void sf2d_free_target(sf2d_rendertarget *target)
{
sf2d_free_texture(&(target->texture));
//free(target); // unnecessary since the texture is the start of the target struct
C3D_RenderTargetDelete(target->target);
free(target);
}
void sf2d_clear_target(sf2d_rendertarget *target, u32 color) {
if (color == 0) { // if fully transparent, take a shortcut
memset(target->texture.data, 0, target->texture.width * target->texture.height * 4);
sf2d_texture_tile32(&(target->texture));
return;
}
void sf2d_clear_target(sf2d_rendertarget *target, u32 color)
{
color = ((color>>24)&0x000000FF) | ((color>>8)&0x0000FF00) | ((color<<8)&0x00FF0000) | ((color<<24)&0xFF000000); // reverse byte order
int itarget = target->texture.width * target->texture.height;
for (int i = 0; i < itarget; i++) { memcpy(target->texture.data + i*4, &color, 4); }
sf2d_texture_tile32(&(target->texture));
C3D_RenderTargetSetClear(target->target, C3D_CLEAR_ALL, color, 0);
}
void sf2d_texture_tile32_hardware(sf2d_texture *texture, const void *data, int w, int h)
{
if (texture->tiled) return;
const u32 flags = (GX_TRANSFER_FLIP_VERT(1) | GX_TRANSFER_OUT_TILED(1) | GX_TRANSFER_RAW_COPY(0) |
GX_TRANSFER_IN_FORMAT(GX_TRANSFER_FMT_RGBA8) | GX_TRANSFER_OUT_FORMAT(GX_TRANSFER_FMT_RGBA8) |
GX_TRANSFER_SCALING(GX_TRANSFER_SCALE_NO));
const u32 flags = (GX_TRANSFER_FLIP_VERT(1) | GX_TRANSFER_OUT_TILED(1) | GX_TRANSFER_RAW_COPY(0) |
GX_TRANSFER_IN_FORMAT(GX_TRANSFER_FMT_RGBA8) | GX_TRANSFER_OUT_FORMAT(GX_TRANSFER_FMT_RGBA8) |
GX_TRANSFER_SCALING(GX_TRANSFER_SCALE_NO));
GSPGPU_FlushDataCache(data, (w*h)<<2);
GX_DisplayTransfer(
(u32*)data,
GX_BUFFER_DIM(w, h),
(u32*)texture->data,
GX_BUFFER_DIM(texture->pow2_w, texture->pow2_h),
flags
);
gspWaitForPPF();
GSPGPU_InvalidateDataCache(texture->data, texture->data_size);
GSPGPU_FlushDataCache(texture->tex.data, texture->tex.size);
C3D_SafeDisplayTransfer(
(u32*)data,
GX_BUFFER_DIM(w, h),
(u32*)texture->tex.data,
GX_BUFFER_DIM(texture->tex.width, texture->tex.height),
flags
);
gspWaitForPPF();
texture->tiled = 1;
}
@ -170,14 +108,14 @@ void sf2d_fill_texture_from_RGBA8(sf2d_texture *dst, const void *rgba8, int sour
{
// TODO: add support for non-RGBA8 textures
u8 *tmp = linearAlloc((dst->pow2_w * dst->pow2_h)<<2);
u8 *tmp = linearAlloc((dst->tex.width * dst->tex.height)<<2);
int i, j;
for (i = 0; i < source_h; i++) {
for (j = 0; j < source_w; j++) {
((u32 *)tmp)[i*dst->pow2_w + j] = __builtin_bswap32(((u32 *)rgba8)[i*source_w + j]);
((u32 *)tmp)[i*dst->tex.width + j] = __builtin_bswap32(((u32 *)rgba8)[i*source_w + j]);
}
}
sf2d_texture_tile32_hardware(dst, tmp, dst->pow2_w, dst->pow2_h);
sf2d_texture_tile32_hardware(dst, tmp, dst->tex.width, dst->tex.height);
linearFree(tmp);
}
@ -190,86 +128,45 @@ sf2d_texture *sf2d_create_texture_mem_RGBA8(const void *src_buffer, int src_w, i
return tex;
}
static inline int sf2d_get_texunit(GPU_TEXUNIT unit)
{
switch (unit) {
case GPU_TEXUNIT0: return 0;
case GPU_TEXUNIT1: return 1;
case GPU_TEXUNIT2: return 2;
default: return -1;
}
}
void sf2d_bind_texture(const sf2d_texture *texture, GPU_TEXUNIT unit)
{
GPU_SetTextureEnable(unit);
C3D_TexBind(sf2d_get_texunit(unit), (C3D_Tex*)&texture->tex);
GPU_SetTexEnv(
0,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_TEXTURE0, GPU_TEXTURE0),
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_TEXTURE0, GPU_TEXTURE0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_REPLACE, GPU_REPLACE,
0xFFFFFFFF
);
GPU_SetTexture(
unit,
(u32 *)osConvertVirtToPhys(texture->data),
texture->pow2_w,
texture->pow2_h,
texture->params,
texture->pixel_format
);
C3D_TexEnv* env = C3D_GetTexEnv(0);
C3D_TexEnvSrc(env, C3D_Both, GPU_TEXTURE0, 0, 0);
C3D_TexEnvOp(env, C3D_Both, 0, 0, 0);
C3D_TexEnvFunc(env, C3D_Both, GPU_REPLACE);
}
void sf2d_bind_texture_color(const sf2d_texture *texture, GPU_TEXUNIT unit, u32 color)
{
GPU_SetTextureEnable(unit);
C3D_TexBind(sf2d_get_texunit(unit), (C3D_Tex*)&texture->tex);
GPU_SetTexEnv(
0,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, GPU_CONSTANT),
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, GPU_CONSTANT),
GPU_TEVOPERANDS(0, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_MODULATE, GPU_MODULATE,
color
);
GPU_SetTexture(
unit,
(u32 *)osConvertVirtToPhys(texture->data),
texture->pow2_w,
texture->pow2_h,
texture->params,
texture->pixel_format
);
}
void sf2d_bind_texture_parameters(const sf2d_texture *texture, GPU_TEXUNIT unit, unsigned int params)
{
GPU_SetTextureEnable(unit);
GPU_SetTexEnv(
0,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_TEXTURE0, GPU_TEXTURE0),
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_TEXTURE0, GPU_TEXTURE0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_REPLACE, GPU_REPLACE,
0xFFFFFFFF
);
GPU_SetTexture(
unit,
(u32 *)osConvertVirtToPhys(texture->data),
texture->pow2_w,
texture->pow2_h,
params,
texture->pixel_format
);
C3D_TexEnv* env = C3D_GetTexEnv(0);
C3D_TexEnvSrc(env, C3D_Both, GPU_TEXTURE0, GPU_CONSTANT, 0);
C3D_TexEnvOp(env, C3D_Both, 0, 0, 0);
C3D_TexEnvFunc(env, C3D_Both, GPU_MODULATE);
C3D_TexEnvColor(env, color);
}
void sf2d_texture_set_params(sf2d_texture *texture, u32 params)
{
texture->params = params;
texture->tex.param = params;
}
int sf2d_texture_get_params(const sf2d_texture *texture)
{
return texture->params;
return texture->tex.param;
}
static inline void sf2d_draw_texture_generic(const sf2d_texture *texture, int x, int y)
@ -285,27 +182,24 @@ static inline void sf2d_draw_texture_generic(const sf2d_texture *texture, int x,
vertices[2].position = (sf2d_vector_3f){(float)x, (float)y+h, SF2D_DEFAULT_DEPTH};
vertices[3].position = (sf2d_vector_3f){(float)x+w, (float)y+h, SF2D_DEFAULT_DEPTH};
float u = texture->width/(float)texture->pow2_w;
float v = texture->height/(float)texture->pow2_h;
float u = texture->width/(float)texture->tex.width;
float v = texture->height/(float)texture->tex.height;
vertices[0].texcoord = (sf2d_vector_2f){0.0f, 0.0f};
vertices[1].texcoord = (sf2d_vector_2f){u, 0.0f};
vertices[2].texcoord = (sf2d_vector_2f){0.0f, v};
vertices[3].texcoord = (sf2d_vector_2f){u, v};
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_FLOAT, 2);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_tex), 2, 0x10);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_texture(const sf2d_texture *texture, int x, int y)
@ -344,8 +238,8 @@ static inline void sf2d_draw_texture_rotate_hotspot_generic(const sf2d_texture *
vertices[3].position.y = h - center_y;
vertices[3].position.z = SF2D_DEFAULT_DEPTH;
float u = w/(float)texture->pow2_w;
float v = h/(float)texture->pow2_h;
float u = w/(float)texture->tex.width;
float v = h/(float)texture->tex.height;
vertices[0].texcoord = (sf2d_vector_2f){0.0f, 0.0f};
vertices[1].texcoord = (sf2d_vector_2f){u, 0.0f};
@ -362,19 +256,16 @@ static inline void sf2d_draw_texture_rotate_hotspot_generic(const sf2d_texture *
vertices[i].position.y = _x*s + _y*c + y;
}
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_FLOAT, 2);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_tex), 2, 0x10);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_texture_rotate_hotspot(const sf2d_texture *texture, int x, int y, float rad, float center_x, float center_y)
@ -428,8 +319,8 @@ static inline void sf2d_draw_texture_rotate_scale_hotspot_generic(const sf2d_tex
vertices[3].position.y = h - center_y * scale_y;
vertices[3].position.z = SF2D_DEFAULT_DEPTH;
float u = w/(float)texture->pow2_w;
float v = h/(float)texture->pow2_h;
float u = w/(float)texture->tex.width;
float v = h/(float)texture->tex.height;
vertices[0].texcoord = (sf2d_vector_2f){0.0f, 0.0f};
vertices[1].texcoord = (sf2d_vector_2f){u, 0.0f};
@ -446,19 +337,16 @@ static inline void sf2d_draw_texture_rotate_scale_hotspot_generic(const sf2d_tex
vertices[i].position.y = _x*s + _y*c + y;
}
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_FLOAT, 2);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_tex), 2, 0x10);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_texture_rotate_scale_hotspot(const sf2d_texture *texture, int x, int y, float rad, float scale_x, float scale_y, float center_x, float center_y)
@ -483,29 +371,26 @@ static inline void sf2d_draw_texture_part_generic(const sf2d_texture *texture, i
vertices[2].position = (sf2d_vector_3f){(float)x, (float)y+tex_h, SF2D_DEFAULT_DEPTH};
vertices[3].position = (sf2d_vector_3f){(float)x+tex_w, (float)y+tex_h, SF2D_DEFAULT_DEPTH};
float u0 = tex_x/(float)texture->pow2_w;
float v0 = tex_y/(float)texture->pow2_h;
float u1 = (tex_x+tex_w)/(float)texture->pow2_w;
float v1 = (tex_y+tex_h)/(float)texture->pow2_h;
float u0 = tex_x/(float)texture->tex.width;
float v0 = tex_y/(float)texture->tex.height;
float u1 = (tex_x+tex_w)/(float)texture->tex.width;
float v1 = (tex_y+tex_h)/(float)texture->tex.height;
vertices[0].texcoord = (sf2d_vector_2f){u0, v0};
vertices[1].texcoord = (sf2d_vector_2f){u1, v0};
vertices[2].texcoord = (sf2d_vector_2f){u0, v1};
vertices[3].texcoord = (sf2d_vector_2f){u1, v1};
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_FLOAT, 2);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_tex), 2, 0x10);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_texture_part(const sf2d_texture *texture, int x, int y, int tex_x, int tex_y, int tex_w, int tex_h)
@ -533,27 +418,24 @@ static inline void sf2d_draw_texture_scale_generic(const sf2d_texture *texture,
vertices[2].position = (sf2d_vector_3f){(float)x, (float)y+hs, SF2D_DEFAULT_DEPTH};
vertices[3].position = (sf2d_vector_3f){(float)x+ws, (float)y+hs, SF2D_DEFAULT_DEPTH};
float u = texture->width/(float)texture->pow2_w;
float v = texture->height/(float)texture->pow2_h;
float u = texture->width/(float)texture->tex.width;
float v = texture->height/(float)texture->tex.height;
vertices[0].texcoord = (sf2d_vector_2f){0.0f, 0.0f};
vertices[1].texcoord = (sf2d_vector_2f){u, 0.0f};
vertices[2].texcoord = (sf2d_vector_2f){0.0f, v};
vertices[3].texcoord = (sf2d_vector_2f){u, v};
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_FLOAT, 2);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_tex), 2, 0x10);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_texture_scale(const sf2d_texture *texture, int x, int y, float x_scale, float y_scale)
@ -573,10 +455,10 @@ static inline void sf2d_draw_texture_part_scale_generic(const sf2d_texture *text
sf2d_vertex_pos_tex *vertices = sf2d_pool_memalign(4 * sizeof(sf2d_vertex_pos_tex), 8);
if (!vertices) return;
float u0 = tex_x/(float)texture->pow2_w;
float v0 = tex_y/(float)texture->pow2_h;
float u1 = (tex_x+tex_w)/(float)texture->pow2_w;
float v1 = (tex_y+tex_h)/(float)texture->pow2_h;
float u0 = tex_x/(float)texture->tex.width;
float v0 = tex_y/(float)texture->tex.height;
float u1 = (tex_x+tex_w)/(float)texture->tex.width;
float v1 = (tex_y+tex_h)/(float)texture->tex.height;
vertices[0].texcoord = (sf2d_vector_2f){u0, v0};
vertices[1].texcoord = (sf2d_vector_2f){u1, v0};
@ -591,19 +473,16 @@ static inline void sf2d_draw_texture_part_scale_generic(const sf2d_texture *text
vertices[2].position = (sf2d_vector_3f){(float)x, (float)y+tex_h, SF2D_DEFAULT_DEPTH};
vertices[3].position = (sf2d_vector_3f){(float)x+tex_w, (float)y+tex_h, SF2D_DEFAULT_DEPTH};
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_FLOAT, 2);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_tex), 2, 0x10);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_texture_part_scale(const sf2d_texture *texture, float x, float y, float tex_x, float tex_y, float tex_w, float tex_h, float x_scale, float y_scale)
@ -618,23 +497,23 @@ void sf2d_draw_texture_part_scale_blend(const sf2d_texture *texture, float x, fl
sf2d_draw_texture_part_scale_generic(texture, x, y, tex_x, tex_y, tex_w, tex_h, x_scale, y_scale);
}
static inline void sf2d_draw_texture_part_rotate_scale_hotspot_generic(const sf2d_texture *texture, int x, int y, float rad, int tex_x, int tex_y, int tex_w, int tex_h, float x_scale, float y_scale, float center_x, float center_y)
static inline void sf2d_draw_texture_part_rotate_scale_generic(const sf2d_texture *texture, int x, int y, float rad, int tex_x, int tex_y, int tex_w, int tex_h, float x_scale, float y_scale)
{
sf2d_vertex_pos_tex *vertices = sf2d_pool_memalign(4 * sizeof(sf2d_vertex_pos_tex), 8);
if (!vertices) return;
int w = tex_w;
int h = tex_h;
int w2 = (tex_w * x_scale)/2.0f;
int h2 = (tex_h * y_scale)/2.0f;
vertices[0].position = (sf2d_vector_3f){(float)-center_x * x_scale, (float)-center_y * y_scale, SF2D_DEFAULT_DEPTH};
vertices[1].position = (sf2d_vector_3f){(float) (w - center_x) * x_scale, (float)-center_y * y_scale, SF2D_DEFAULT_DEPTH};
vertices[2].position = (sf2d_vector_3f){(float)-center_x * x_scale, (float) (h - center_y) * y_scale, SF2D_DEFAULT_DEPTH};
vertices[3].position = (sf2d_vector_3f){(float) (w - center_x) * x_scale, (float) h - center_y * y_scale, SF2D_DEFAULT_DEPTH};
vertices[0].position = (sf2d_vector_3f){(float)-w2, (float)-h2, SF2D_DEFAULT_DEPTH};
vertices[1].position = (sf2d_vector_3f){(float) w2, (float)-h2, SF2D_DEFAULT_DEPTH};
vertices[2].position = (sf2d_vector_3f){(float)-w2, (float) h2, SF2D_DEFAULT_DEPTH};
vertices[3].position = (sf2d_vector_3f){(float) w2, (float) h2, SF2D_DEFAULT_DEPTH};
float u0 = tex_x/(float)texture->pow2_w;
float v0 = tex_y/(float)texture->pow2_h;
float u1 = (tex_x+tex_w)/(float)texture->pow2_w;
float v1 = (tex_y+tex_h)/(float)texture->pow2_h;
float u0 = tex_x/(float)texture->tex.width;
float v0 = tex_y/(float)texture->tex.height;
float u1 = (tex_x+tex_w)/(float)texture->tex.width;
float v1 = (tex_y+tex_h)/(float)texture->tex.height;
vertices[0].texcoord = (sf2d_vector_2f){u0, v0};
vertices[1].texcoord = (sf2d_vector_2f){u1, v0};
@ -651,37 +530,28 @@ static inline void sf2d_draw_texture_part_rotate_scale_hotspot_generic(const sf2
vertices[i].position.y = _x*s + _y*c + y;
}
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_FLOAT, 2);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_tex), 2, 0x10);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_texture_part_rotate_scale(const sf2d_texture *texture, int x, int y, float rad, int tex_x, int tex_y, int tex_w, int tex_h, float x_scale, float y_scale)
{
sf2d_bind_texture(texture, GPU_TEXUNIT0);
sf2d_draw_texture_part_rotate_scale_hotspot_generic(texture, x, y, rad, tex_x, tex_y, tex_w, tex_h, x_scale, y_scale, tex_w/2.0f, tex_h/2.0f);
sf2d_draw_texture_part_rotate_scale_generic(texture, x, y, rad, tex_x, tex_y, tex_w, tex_h, x_scale, y_scale);
}
void sf2d_draw_texture_part_rotate_scale_blend(const sf2d_texture *texture, int x, int y, float rad, int tex_x, int tex_y, int tex_w, int tex_h, float x_scale, float y_scale, u32 color)
{
sf2d_bind_texture_color(texture, GPU_TEXUNIT0, color);
sf2d_draw_texture_part_rotate_scale_hotspot_generic(texture, x, y, rad, tex_x, tex_y, tex_w, tex_h, x_scale, y_scale, tex_w/2.0f, tex_h/2.0f);
}
void sf2d_draw_texture_part_rotate_scale_hotspot_blend(const sf2d_texture *texture, int x, int y, float rad, int tex_x, int tex_y, int tex_w, int tex_h, float x_scale, float y_scale, float center_x, float center_y, u32 color)
{
sf2d_bind_texture_color(texture, GPU_TEXUNIT0, color);
sf2d_draw_texture_part_rotate_scale_hotspot_generic(texture, x, y, rad, tex_x, tex_y, tex_w, tex_h, x_scale, y_scale, center_x, center_y);
sf2d_draw_texture_part_rotate_scale_generic(texture, x, y, rad, tex_x, tex_y, tex_w, tex_h, x_scale, y_scale);
}
static inline void sf2d_draw_texture_depth_generic(const sf2d_texture *texture, int x, int y, signed short z)
@ -698,27 +568,24 @@ static inline void sf2d_draw_texture_depth_generic(const sf2d_texture *texture,
vertices[2].position = (sf2d_vector_3f){(float)x, (float)y+h, depth};
vertices[3].position = (sf2d_vector_3f){(float)x+w, (float)y+h, depth};
float u = texture->width/(float)texture->pow2_w;
float v = texture->height/(float)texture->pow2_h;
float u = texture->width/(float)texture->tex.width;
float v = texture->height/(float)texture->tex.height;
vertices[0].texcoord = (sf2d_vector_2f){0.0f, 0.0f};
vertices[1].texcoord = (sf2d_vector_2f){u, 0.0f};
vertices[2].texcoord = (sf2d_vector_2f){0.0f, v};
vertices[3].texcoord = (sf2d_vector_2f){u, v};
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_FLOAT, 2);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_tex), 2, 0x10);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_texture_depth(const sf2d_texture *texture, int x, int y, signed short z)
@ -733,8 +600,7 @@ void sf2d_draw_texture_depth_blend(const sf2d_texture *texture, int x, int y, si
sf2d_draw_texture_depth_generic(texture, x, y, z);
}
void sf2d_draw_quad_uv(const sf2d_texture *texture, float left, float top, float right, float bottom, float u0, float v0, float u1, float v1, unsigned int params)
void sf2d_draw_quad_uv_current(float left, float top, float right, float bottom, float u0, float v0, float u1, float v1)
{
sf2d_vertex_pos_tex *vertices = sf2d_pool_memalign(4 * sizeof(sf2d_vertex_pos_tex), 8);
if (!vertices) return;
@ -749,21 +615,28 @@ void sf2d_draw_quad_uv(const sf2d_texture *texture, float left, float top, float
vertices[2].texcoord = (sf2d_vector_2f){u0, v1};
vertices[3].texcoord = (sf2d_vector_2f){u1, v1};
sf2d_bind_texture_parameters(texture, GPU_TEXUNIT0, params);
C3D_AttrInfo* attrInfo = C3D_GetAttrInfo();
AttrInfo_Init(attrInfo);
AttrInfo_AddLoader(attrInfo, 0, GPU_FLOAT, 3);
AttrInfo_AddLoader(attrInfo, 1, GPU_FLOAT, 2);
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(vertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT) | GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC, //0b1100
0x10,
1, //number of buffers
(u32[]){0x0}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
C3D_BufInfo* bufInfo = C3D_GetBufInfo();
BufInfo_Init(bufInfo);
BufInfo_Add(bufInfo, vertices, sizeof(sf2d_vertex_pos_tex), 2, 0x10);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
C3D_DrawArrays(GPU_TRIANGLE_STRIP, 0, 4);
}
void sf2d_draw_quad_uv(const sf2d_texture *texture, float left, float top, float right, float bottom, float u0, float v0, float u1, float v1)
{
sf2d_bind_texture(texture, GPU_TEXUNIT0);
sf2d_draw_quad_uv_current(left, top, right, bottom, u0, v0, u1, v1);
}
void sf2d_draw_quad_uv_blend(const sf2d_texture *texture, float left, float top, float right, float bottom, float u0, float v0, float u1, float v1, u32 color)
{
sf2d_bind_texture_color(texture, GPU_TEXUNIT0, color);
sf2d_draw_quad_uv_current(left, top, right, bottom, u0, v0, u1, v1);
}
// Grabbed from Citra Emulator (citra/src/video_core/utils.h)
@ -786,25 +659,25 @@ static inline u32 get_morton_offset(u32 x, u32 y, u32 bytes_per_pixel)
void sf2d_set_pixel(sf2d_texture *texture, int x, int y, u32 new_color)
{
y = (texture->pow2_h - 1 - y);
y = (texture->tex.height - 1 - y);
if (texture->tiled) {
u32 coarse_y = y & ~7;
u32 offset = get_morton_offset(x, y, 4) + coarse_y * texture->pow2_w * 4;
*(u32 *)(texture->data + offset) = new_color;
u32 offset = get_morton_offset(x, y, 4) + coarse_y * texture->tex.width * 4;
*(u32 *)(texture->tex.data + offset) = new_color;
} else {
((u32 *)texture->data)[x + y * texture->pow2_w] = new_color;
((u32 *)texture->tex.data)[x + y * texture->tex.width] = new_color;
}
}
u32 sf2d_get_pixel(sf2d_texture *texture, int x, int y)
{
y = (texture->pow2_h - 1 - y);
y = (texture->tex.height - 1 - y);
if (texture->tiled) {
u32 coarse_y = y & ~7;
u32 offset = get_morton_offset(x, y, 4) + coarse_y * texture->pow2_w * 4;
return *(u32 *)(texture->data + offset);
u32 offset = get_morton_offset(x, y, 4) + coarse_y * texture->tex.width * 4;
return *(u32 *)(texture->tex.data + offset);
} else {
return ((u32 *)texture->data)[x + y * texture->pow2_w];
return ((u32 *)texture->tex.data)[x + y * texture->tex.width];
}
}
@ -814,21 +687,21 @@ void sf2d_texture_tile32(sf2d_texture *texture)
if (texture->tiled) return;
// TODO: add support for non-RGBA8 textures
u8 *tmp = linearAlloc(texture->pow2_w * texture->pow2_h * 4);
u8 *tmp = linearAlloc(texture->tex.width * texture->tex.height * 4);
int i, j;
for (j = 0; j < texture->pow2_h; j++) {
for (i = 0; i < texture->pow2_w; i++) {
for (j = 0; j < texture->tex.height; j++) {
for (i = 0; i < texture->tex.width; i++) {
u32 coarse_y = j & ~7;
u32 dst_offset = get_morton_offset(i, j, 4) + coarse_y * texture->pow2_w * 4;
u32 dst_offset = get_morton_offset(i, j, 4) + coarse_y * texture->tex.width * 4;
u32 v = ((u32 *)texture->data)[i + (texture->pow2_h - 1 - j)*texture->pow2_w];
u32 v = ((u32 *)texture->tex.data)[i + (texture->tex.height - 1 - j)*texture->tex.width];
*(u32 *)(tmp + dst_offset) = __builtin_bswap32(v); /* RGBA8 -> ABGR8 */
}
}
memcpy(texture->data, tmp, texture->pow2_w*texture->pow2_h*4);
memcpy(texture->tex.data, tmp, texture->tex.width*texture->tex.height*4);
linearFree(tmp);
texture->tiled = 1;