void drawPlayerGun(player_s* p)
{
if(!p)return;
gsPushMatrix();
useCamera(&p->camera);
gsLoadIdentity();
GPU_SetDepthTestAndWriteMask(true, GPU_ALWAYS, GPU_WRITE_DEPTH);
gsSwitchRenderMode(-1);
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys(rectangleBaseAddr), // we use the start of linear heap as base since that's where all our buffers are located
GPU_ATTRIBFMT(0, 3, GPU_FLOAT)|GPU_ATTRIBFMT(1, 2, GPU_FLOAT), // we want v0 (vertex position) and v1 (texcoord)
0xFFC, // mask : we want v0 and v1
0x10, // permutation : we use identity
1, // number of buffers : we have one attribute per buffer
(u32[]){(u32)rectangleVertexData-rectangleBaseAddr}, // buffer offsets (placeholders)
(u64[]){0x10}, // attribute permutations for each buffer
(u8[]){2} // number of attributes for each buffer
);
gsSetShader(&passthroughProgram);
GPU_DrawArray(GPU_TRIANGLES, 6);
GPU_SetDepthTestAndWriteMask(true, GPU_ALWAYS, GPU_WRITE_ALL);
GPUCMD_AddWrite(GPUREG_ATTRIBBUFFER0_CONFIG0, (u32)crosshairVertexData-rectangleBaseAddr);
textureBind(&crosshairTexture, GPU_TEXUNIT0);
GPU_DrawArray(GPU_TRIANGLES, 6);
GPU_SetDepthTestAndWriteMask(true, GPU_GREATER, GPU_WRITE_ALL);
gsSwitchRenderMode(md2GsMode);
gsTranslate(1.3, -1.65, -3.1);
gsRotateY(p->tempAngle.y);
gsRotateX(-p->tempAngle.x);
gsRotateX(-0.1);
gsRotateZ(M_PI/2);
gsRotateY(-M_PI/2);
gsRotateX(-M_PI/2);
gsScale(3.0f/8, 3.0f/8, 3.0f/8);
md2InstanceDraw(&p->gunInstance);
gsPopMatrix();
}
static void sceneRender(void)
{
// Bind the shader program
shaderProgramUse(&program);
// Configure the first fragment shading substage to just pass through the vertex color
// See https://www.opengl.org/sdk/docs/man2/xhtml/glTexEnv.xml for more insight
GPU_SetTexEnv(0,
GPU_TEVSOURCES(GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR), // RGB channels
GPU_TEVSOURCES(GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR), // Alpha
GPU_TEVOPERANDS(0, 0, 0), // RGB
GPU_TEVOPERANDS(0, 0, 0), // Alpha
GPU_REPLACE, GPU_REPLACE, // RGB, Alpha
0xFFFFFFFF);
// Configure the "attribute buffers" (that is, the vertex input buffers)
GPU_SetAttributeBuffers(
1, // Number of inputs per vertex
(u32*)osConvertVirtToPhys((u32)vbo_data), // Location of the VBO
GPU_ATTRIBFMT(0, 3, GPU_FLOAT), // Format of the inputs (in this case the only input is a 3-element float vector)
0xFFE, // Unused attribute mask, in our case bit 0 is cleared since it is used
0x0, // Attribute permutations (here it is the identity)
1, // Number of buffers
(u32[]) { 0x0 }, // Buffer offsets (placeholders)
(u64[]) { 0x0 }, // Attribute permutations for each buffer (identity again)
(u8[]) { 1 }); // Number of attributes for each buffer
// Upload the projection matrix
GPU_SetFloatUniformMatrix(GPU_VERTEX_SHADER, uLoc_projection, &projection);
// Draw the VBO
GPU_DrawArray(GPU_TRIANGLES, vertex_list_count);
}
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)
{
sf2d_vertex_pos_tex *vertices = sf2d_pool_malloc(4 * sizeof(sf2d_vertex_pos_tex));
vertices[0].position = (sf2d_vector_3f){(float)x, (float)y, 0.5f};
vertices[1].position = (sf2d_vector_3f){(float)x+tex_w, (float)y, 0.5f};
vertices[2].position = (sf2d_vector_3f){(float)x, (float)y+tex_h, 0.5f};
vertices[3].position = (sf2d_vector_3f){(float)x+tex_w, (float)y+tex_h, 0.5f};
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;
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};
sf2d_bind_texture(texture, GPU_TEXUNIT0);
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys((u32)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
);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 4);
}
void sf2d_draw_texture_rotate_cut_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_vertex_pos_tex *vertices = sf2d_pool_malloc(4 * sizeof(sf2d_vertex_pos_tex));
//Don't even try to understand what I'm doing here (because I don't even understand it).
//Matrices are boring.
int w2 = (texture->width * x_scale)/2.0f;
int h2 = (texture->height * y_scale)/2.0f;
vertices[0].position = (sf2d_vector_3f){(float)-w2, (float)-h2, 0.5f};
vertices[1].position = (sf2d_vector_3f){(float) w2, (float)-h2, 0.5f};
vertices[2].position = (sf2d_vector_3f){(float)-w2, (float) h2, 0.5f};
vertices[3].position = (sf2d_vector_3f){(float) w2, (float) h2, 0.5f};
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;
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};
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[0].position = (sf2d_vector_3f){rot[0].x + x + w2, rot[0].y + y + h2, rot[0].z};
vertices[1].position = (sf2d_vector_3f){rot[1].x + x * x_scale + w2, rot[1].y + y + h2, rot[1].z};
vertices[2].position = (sf2d_vector_3f){rot[2].x + x + w2, rot[2].y + y * y_scale + h2, rot[2].z};
vertices[3].position = (sf2d_vector_3f){rot[3].x + x * x_scale + w2, rot[3].y + y * y_scale + h2, rot[3].z};
sf2d_bind_texture(texture, GPU_TEXUNIT0);
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys((u32)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
);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 4);
}
void drawPortals(portal_s* portals[], int n, renderSceneCallback_t callback, camera_s* c, int depth, u8 stencil)
{
if(!portals || !portalVertexData || !callback || !c || !depth)return;
int i;
gsSwitchRenderMode(-1);
GPU_SetAttributeBuffers(
1, // number of attributes
(u32*)osConvertVirtToPhys(portalBaseAddr), // we use the start of linear heap as base since that's where all our buffers are located
GPU_ATTRIBFMT(0, 3, GPU_FLOAT), // we want v0 (vertex position)
0xFFE, // mask : we want v0
0x0, // permutation : we use identity
1, // number of buffers : we have one attribute per buffer
(u32[]){(u32)portalOutlineVertexData-portalBaseAddr}, // buffer offsets (placeholders)
(u64[]){0x0}, // attribute permutations for each buffer
(u8[]){1} // number of attributes for each buffer
);
gsPushMatrix();
shaderInstanceSetBool(portalProgram.vertexShader, 0, true);
gsSetShader(&portalProgram);
//TEMP
int colorUniformLoc = shaderInstanceGetUniformLocation(portalProgram.vertexShader, "color");
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_MODULATE, GPU_MODULATE,
0xFFFFFFFF);
for(i=0; i<n; i++)
{
portal_s* p = portals[i];
if(!p->draw)continue;
GPU_SetFloatUniform(GPU_VERTEX_SHADER, colorUniformLoc, (u32*)(float[]){1.0f, p->color.z, p->color.y, p->color.x}, 1);
gsPushMatrix();
gsTranslate(p->position.x, p->position.y, p->position.z);
gsMultMatrix(p->matrix);
gsUpdateTransformation();
GPU_DrawArray(GPU_TRIANGLE_STRIP, portalNumVertices);
gsPopMatrix();
}
void sf2d_draw_texture_rotate(const sf2d_texture *texture, int x, int y, float rad)
{
sf2d_vertex_pos_tex *vertices = sf2d_pool_malloc(4 * sizeof(sf2d_vertex_pos_tex));
int w2 = texture->width/2.0f;
int h2 = texture->height/2.0f;
vertices[0].position = (sf2d_vector_3f){(float)-w2, (float)-h2, 0.5f};
vertices[1].position = (sf2d_vector_3f){(float) w2, (float)-h2, 0.5f};
vertices[2].position = (sf2d_vector_3f){(float)-w2, (float) h2, 0.5f};
vertices[3].position = (sf2d_vector_3f){(float) w2, (float) h2, 0.5f};
float u = texture->width/(float)texture->pow2_w;
float v = texture->height/(float)texture->pow2_h;
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};
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]);
}
for (i = 0; i < 4; i++) {
vertices[i].position = (sf2d_vector_3f){rot[i].x + x + w2, rot[i].y + y + h2, rot[i].z};
}
sf2d_bind_texture(texture, GPU_TEXUNIT0);
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys((u32)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
);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 4);
}
void gpuDrawVbo(u32 vbo) {
VboData* vboData = (VboData*) vbo;
if(vboData == NULL || vboData->data == NULL) {
return;
}
gpuUpdateState();
static u32 attributeBufferOffset = 0;
GPU_SetAttributeBuffers(vboData->attributeCount, (u32*) osConvertVirtToPhys((u32) vboData->data), vboData->attributes, vboData->attributeMask, vboData->attributePermutations, 1, &attributeBufferOffset, &vboData->attributePermutations, &vboData->attributeCount);
if(vboData->indices != NULL) {
GPU_DrawElements((GPU_Primitive_t) vboData->primitive, (u32*) vboData->indices, vboData->numVertices);
} else {
GPU_DrawArray((GPU_Primitive_t) vboData->primitive, vboData->numVertices);
}
}
void sf2d_draw_texture_scale(const sf2d_texture *texture, int x, int y, float x_scale, float y_scale)
{
sf2d_vertex_pos_tex *vertices = sf2d_pool_malloc(4 * sizeof(sf2d_vertex_pos_tex));
int ws = texture->width * x_scale;
int hs = texture->height * y_scale;
vertices[0].position = (sf2d_vector_3f){(float)x, (float)y, 0.5f};
vertices[1].position = (sf2d_vector_3f){(float)x+ws, (float)y, 0.5f};
vertices[2].position = (sf2d_vector_3f){(float)x, (float)y+hs, 0.5f};
vertices[3].position = (sf2d_vector_3f){(float)x+ws, (float)y+hs, 0.5f};
float u = texture->width/(float)texture->pow2_w;
float v = texture->height/(float)texture->pow2_h;
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};
sf2d_bind_texture(texture, GPU_TEXUNIT0);
GPU_SetAttributeBuffers(
2, // number of attributes
(u32*)osConvertVirtToPhys((u32)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
);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 4);
}
void RectangleShape::draw()
{
if (!_device->isInitialized() || !_visible)
return;
updateTransform();
if (_color.a > 0)
{
ShaderProgram::setUniform(UNIFORM_MODELVIEW_NAME, getCombinedTransform());
#ifdef _3DS
auto vertices = VertexPool::getInstance()->
pushVertices<VertexPositionTexture>(4);
vertices[0].position = Vec2(0, 0);
vertices[1].position = Vec2(_size.x, 0);
vertices[2].position = Vec2(0, _size.y);
vertices[3].position = Vec2(_size.x, _size.y);
vertices[0].texcoord = Vec2::Zero;
vertices[1].texcoord = Vec2::Zero;
vertices[2].texcoord = Vec2::Zero;
vertices[3].texcoord = Vec2::Zero;
//=========================================================================
GPU_SetTexEnv(0,
GPU_TEVSOURCES(GPU_CONSTANT, GPU_CONSTANT, GPU_CONSTANT),
GPU_TEVSOURCES(GPU_CONSTANT, GPU_CONSTANT, GPU_CONSTANT),
GPU_TEVOPERANDS(0, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_REPLACE, GPU_REPLACE,
_color.r | _color.g << 8 | _color.b << 16 | _color.a << 24);
u32 bufferOffsets[] = { 0x00 };
u64 bufferPermutations[] = { 0x10 };
u8 bufferNumAttributes[] = { 2 };
GPU_SetAttributeBuffers(
2,
(u32*)osConvertVirtToPhys((u32)vertices),
GPU_ATTRIBFMT(0, 2, GPU_FLOAT) |
GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFFC,
0x10,
1,
bufferOffsets,
bufferPermutations,
bufferNumAttributes);
GPU_DrawArray(GPU_TRIANGLE_STRIP, 0, 4);
#else
Vec4 c = _color.toVector();
float vertices[] =
{
0, 0, -0.5f, c.x, c.y, c.z, c.w, 0, 0,
_size.x, 0, -0.5f, c.x, c.y, c.z, c.w, 0, 0,
0, _size.y, -0.5f, c.x, c.y, c.z, c.w, 0, 0,
_size.x, _size.y, -0.5f, c.x, c.y, c.z, c.w, 0, 0
};
glUniform1i(ShaderProgram::getCurrentProgram()->
getUniformLocation("textureEnabled"), false);
glBindBuffer(GL_ARRAY_BUFFER, _vbo);
glBufferSubData(GL_ARRAY_BUFFER, 0, sizeof(float) * 36, vertices);
glBindBuffer(GL_ARRAY_BUFFER, 0);
glBindVertexArray(_vao);
glDrawArrays(GL_TRIANGLE_STRIP, 0, 4);
glBindVertexArray(0);
#endif // _3DS
}
if (getOutlineColor().a > 0)
{
_outline.setTransform(getCombinedTransform());
_outline.setLine(0, 0, _size.x, 0);
_outline.draw();
_outline.setLine(_size.x, 0, _size.x, _size.y);
_outline.draw();
_outline.setLine(0, _size.y, _size.x, _size.y);
_outline.draw();
_outline.setLine(0, 0, 0, _size.y);
_outline.draw();
}
Node::draw();
}
static void menu_display_ctr_draw(void *data)
{
struct ctr_texture *texture = NULL;
const float *color = NULL;
ctr_video_t *ctr = (ctr_video_t*)video_driver_get_ptr(false);
menu_display_ctx_draw_t *draw = (menu_display_ctx_draw_t*)data;
if (!ctr || !draw)
return;
texture = (struct ctr_texture*)draw->texture;
color = draw->coords->color;
if (!texture)
return;
ctr_scale_vector_t scale_vector;
ctr_set_scale_vector(&scale_vector,
CTR_TOP_FRAMEBUFFER_WIDTH, CTR_TOP_FRAMEBUFFER_HEIGHT,
texture->width, texture->height);
ctrGuSetVertexShaderFloatUniform(0, (float*)&scale_vector, 1);
if ((ctr->vertex_cache.size - (ctr->vertex_cache.current - ctr->vertex_cache.buffer)) < 1)
ctr->vertex_cache.current = ctr->vertex_cache.buffer;
ctr_vertex_t* v = ctr->vertex_cache.current++;
v->x0 = draw->x;
v->y0 = 240 - draw->height - draw->y;
v->x1 = v->x0 + draw->width;
v->y1 = v->y0 + draw->height;
v->u0 = 0;
v->v0 = 0;
v->u1 = texture->active_width;
v->v1 = texture->active_height;
ctrGuSetAttributeBuffers(2,
VIRT_TO_PHYS(v),
CTRGU_ATTRIBFMT(GPU_SHORT, 4) << 0 |
CTRGU_ATTRIBFMT(GPU_SHORT, 4) << 4,
sizeof(ctr_vertex_t));
color = draw->coords->color;
int colorR = (int)((*color++)*255.f);
int colorG = (int)((*color++)*255.f);
int colorB = (int)((*color++)*255.f);
int colorA = (int)((*color++)*255.f);
GPU_SetTexEnv(0,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, 0),
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, 0),
0,
0,
GPU_MODULATE, GPU_MODULATE,
COLOR_ABGR(colorR,colorG,colorB,colorA)
);
// GPU_SetTexEnv(0,
// GPU_TEVSOURCES(GPU_CONSTANT, GPU_CONSTANT, 0),
// GPU_TEVSOURCES(GPU_CONSTANT, GPU_CONSTANT, 0),
// 0,
// GPU_TEVOPERANDS(GPU_TEVOP_RGB_SRC_COLOR, GPU_TEVOP_RGB_SRC_COLOR, 0),
// GPU_REPLACE, GPU_REPLACE,
// 0x3FFFFFFF);
ctrGuSetTexture(GPU_TEXUNIT0, VIRT_TO_PHYS(texture->data), texture->width, texture->height,
GPU_TEXTURE_MAG_FILTER(GPU_LINEAR) | GPU_TEXTURE_MIN_FILTER(GPU_LINEAR) |
GPU_TEXTURE_WRAP_S(GPU_CLAMP_TO_EDGE) | GPU_TEXTURE_WRAP_T(GPU_CLAMP_TO_EDGE),
GPU_RGBA8);
GPU_SetViewport(NULL,
VIRT_TO_PHYS(ctr->drawbuffers.top.left),
0, 0, CTR_TOP_FRAMEBUFFER_HEIGHT,
ctr->video_mode == CTR_VIDEO_MODE_800x240 ? CTR_TOP_FRAMEBUFFER_WIDTH * 2 : CTR_TOP_FRAMEBUFFER_WIDTH);
GPU_DrawArray(GPU_GEOMETRY_PRIM, 0, 1);
if (ctr->video_mode == CTR_VIDEO_MODE_3D)
{
GPU_SetViewport(NULL,
VIRT_TO_PHYS(ctr->drawbuffers.top.right),
0, 0, CTR_TOP_FRAMEBUFFER_HEIGHT,
CTR_TOP_FRAMEBUFFER_WIDTH);
GPU_DrawArray(GPU_GEOMETRY_PRIM, 0, 1);
}
GPU_SetTexEnv(0, GPU_TEXTURE0, GPU_TEXTURE0, 0, 0, GPU_REPLACE, GPU_REPLACE, 0);
// printf("(%i,%i,%i,%i) , (%i,%i)\n", (int)draw->x, (int)draw->y, (int)draw->width, (int)draw->height, texture->width, texture->height);
}
static bool ctr_frame(void* data, const void* frame,
unsigned width, unsigned height, unsigned pitch, const char* msg)
{
ctr_video_t* ctr = (ctr_video_t*)data;
settings_t* settings = config_get_ptr();
static uint64_t currentTick,lastTick;
static float fps = 0.0;
static int total_frames = 0;
static int frames = 0;
extern bool select_pressed;
if (!width || !height)
{
gspWaitForEvent(GSPEVENT_VBlank0, true);
return true;
}
if(!aptMainLoop())
{
event_command(EVENT_CMD_QUIT);
return true;
}
if (select_pressed)
{
event_command(EVENT_CMD_QUIT);
return true;
}
svcWaitSynchronization(gspEvents[GSPEVENT_P3D], 20000000);
svcClearEvent(gspEvents[GSPEVENT_P3D]);
svcWaitSynchronization(gspEvents[GSPEVENT_PPF], 20000000);
svcClearEvent(gspEvents[GSPEVENT_PPF]);
gfxSwapBuffersGpu();
frames++;
if (ctr->vsync)
svcWaitSynchronization(gspEvents[GSPEVENT_VBlank0], U64_MAX);
svcClearEvent(gspEvents[GSPEVENT_VBlank0]);
currentTick = svcGetSystemTick();
uint32_t diff = currentTick - lastTick;
if(diff > CTR_CPU_TICKS_PER_SECOND)
{
fps = (float)frames * ((float) CTR_CPU_TICKS_PER_SECOND / (float) diff);
lastTick = currentTick;
frames = 0;
}
printf("fps: %8.4f frames: %i\r", fps, total_frames++);
fflush(stdout);
ctrGuSetMemoryFill(true, (u32*)CTR_GPU_FRAMEBUFFER, 0x00000000,
(u32*)(CTR_GPU_FRAMEBUFFER + CTR_TOP_FRAMEBUFFER_WIDTH * CTR_TOP_FRAMEBUFFER_HEIGHT * sizeof(uint32_t)),
0x201, (u32*)CTR_GPU_DEPTHBUFFER, 0x00000000,
(u32*)(CTR_GPU_DEPTHBUFFER + CTR_TOP_FRAMEBUFFER_WIDTH * CTR_TOP_FRAMEBUFFER_HEIGHT * sizeof(uint32_t)),
0x201);
GPUCMD_SetBufferOffset(0);
if (width > ctr->texture_width)
width = ctr->texture_width;
if (height > ctr->texture_height)
height = ctr->texture_height;
if(frame)
{
if(((((u32)(frame)) >= 0x14000000 && ((u32)(frame)) < 0x1c000000)) /* frame in linear memory */
&& !((u32)frame & 0x7F) /* 128-byte aligned */
&& !((pitch) & 0xF)) /* 16-byte aligned */
{
/* can copy the buffer directly with the GPU */
ctrGuCopyImage(false, frame, pitch / 2, height, CTRGU_RGB565, false,
ctr->texture_swizzled, ctr->texture_width, CTRGU_RGB565, true);
}
else
{
int i;
uint16_t* dst = (uint16_t*)ctr->texture_linear;
const uint8_t* src = frame;
for (i = 0; i < height; i++)
{
memcpy(dst, src, width * sizeof(uint16_t));
dst += ctr->texture_width;
src += pitch;
}
GSPGPU_FlushDataCache(NULL, ctr->texture_linear,
ctr->texture_width * ctr->texture_height * sizeof(uint16_t));
ctrGuCopyImage(false, ctr->texture_linear, ctr->texture_width, ctr->menu.texture_height, CTRGU_RGB565, false,
ctr->texture_swizzled, ctr->texture_width, CTRGU_RGB565, true);
}
}
ctrGuSetTexture(GPU_TEXUNIT0, VIRT_TO_PHYS(ctr->texture_swizzled), ctr->texture_width, ctr->texture_height,
GPU_TEXTURE_MAG_FILTER(GPU_LINEAR) | GPU_TEXTURE_MIN_FILTER(GPU_LINEAR) |
GPU_TEXTURE_WRAP_S(GPU_CLAMP_TO_EDGE) | GPU_TEXTURE_WRAP_T(GPU_CLAMP_TO_EDGE),
GPU_RGB565);
ctr->frame_coords->u = width;
ctr->frame_coords->v = height;
GSPGPU_FlushDataCache(NULL, (u8*)ctr->frame_coords, sizeof(ctr_vertex_t));
ctrGuSetAttributeBuffersAddress(VIRT_TO_PHYS(ctr->frame_coords));
ctrGuSetVertexShaderFloatUniform(0, (float*)&ctr->scale_vector, 1);
GPU_DrawArray(GPU_UNKPRIM, 1);
if (ctr->menu_texture_enable)
{
GSPGPU_FlushDataCache(NULL, ctr->menu.texture_linear,
ctr->menu.texture_width * ctr->menu.texture_height * sizeof(uint16_t));
ctrGuCopyImage(false, ctr->menu.texture_linear, ctr->menu.texture_width, ctr->menu.texture_height, CTRGU_RGBA4444,false,
ctr->menu.texture_swizzled, ctr->menu.texture_width, CTRGU_RGBA4444, true);
ctrGuSetTexture(GPU_TEXUNIT0, VIRT_TO_PHYS(ctr->menu.texture_swizzled), ctr->menu.texture_width, ctr->menu.texture_height,
GPU_TEXTURE_MAG_FILTER(GPU_LINEAR) | GPU_TEXTURE_MIN_FILTER(GPU_LINEAR) |
GPU_TEXTURE_WRAP_S(GPU_CLAMP_TO_EDGE) | GPU_TEXTURE_WRAP_T(GPU_CLAMP_TO_EDGE),
GPU_RGBA4);
ctrGuSetAttributeBuffersAddress(VIRT_TO_PHYS(ctr->menu.frame_coords));
ctrGuSetVertexShaderFloatUniform(1, (float*)&ctr->menu.scale_vector, 1);
GPU_DrawArray(GPU_UNKPRIM, 1);
}
GPU_FinishDrawing();
GPUCMD_Finalize();
ctrGuFlushAndRun(true);
ctrGuDisplayTransfer(true, CTR_GPU_FRAMEBUFFER, 240,400, CTRGU_RGBA8,
gfxGetFramebuffer(GFX_TOP, GFX_LEFT, NULL, NULL), 240,400,CTRGU_RGB8, CTRGU_MULTISAMPLE_NONE);
ctr->frame_count++;
return true;
}
static bool ctr_frame(void* data, const void* frame,
unsigned width, unsigned height,
uint64_t frame_count,
unsigned pitch, const char* msg)
{
uint32_t diff;
static uint64_t currentTick,lastTick;
ctr_video_t *ctr = (ctr_video_t*)data;
settings_t *settings = config_get_ptr();
static float fps = 0.0;
static int total_frames = 0;
static int frames = 0;
static struct retro_perf_counter ctrframe_f = {0};
uint32_t state_tmp;
touchPosition state_tmp_touch;
extern bool select_pressed;
if (!width || !height)
{
gspWaitForEvent(GSPEVENT_VBlank0, true);
return true;
}
if(!aptMainLoop())
{
event_command(EVENT_CMD_QUIT);
return true;
}
if (select_pressed)
{
event_command(EVENT_CMD_QUIT);
return true;
}
state_tmp = hidKeysDown();
hidTouchRead(&state_tmp_touch);
if((state_tmp & KEY_TOUCH) && (state_tmp_touch.py < 120))
{
Handle lcd_handle;
u8 not_2DS;
extern PrintConsole* currentConsole;
gfxBottomFramebuffers[0] = ctr->lcd_buttom_on ? (u8*)ctr->empty_framebuffer:
(u8*)currentConsole->frameBuffer;
CFGU_GetModelNintendo2DS(¬_2DS);
if(not_2DS && srvGetServiceHandle(&lcd_handle, "gsp::Lcd") >= 0)
{
u32 *cmdbuf = getThreadCommandBuffer();
cmdbuf[0] = ctr->lcd_buttom_on? 0x00120040: 0x00110040;
cmdbuf[1] = 2;
svcSendSyncRequest(lcd_handle);
svcCloseHandle(lcd_handle);
}
ctr->lcd_buttom_on = !ctr->lcd_buttom_on;
}
svcWaitSynchronization(gspEvents[GSPEVENT_P3D], 20000000);
svcClearEvent(gspEvents[GSPEVENT_P3D]);
svcWaitSynchronization(gspEvents[GSPEVENT_PPF], 20000000);
svcClearEvent(gspEvents[GSPEVENT_PPF]);
frames++;
if (ctr->vsync)
svcWaitSynchronization(gspEvents[GSPEVENT_VBlank0], U64_MAX);
svcClearEvent(gspEvents[GSPEVENT_VBlank0]);
currentTick = svcGetSystemTick();
diff = currentTick - lastTick;
if(diff > CTR_CPU_TICKS_PER_SECOND)
{
fps = (float)frames * ((float) CTR_CPU_TICKS_PER_SECOND / (float) diff);
lastTick = currentTick;
frames = 0;
}
// extern u32 __linear_heap_size;
// extern u32 gpuCmdBufOffset;
// extern u32 __heap_size;
// printf("0x%08X 0x%08X 0x%08X\r", __heap_size, gpuCmdBufOffset, (__linear_heap_size - linearSpaceFree() +0x3FF) & ~0x3FF);
// printf("fps: %8.4f frames: %i (%X)\r", fps, total_frames++, (__linear_heap_size - linearSpaceFree()));
printf("fps: %8.4f frames: %i\r", fps, total_frames++);
fflush(stdout);
rarch_perf_init(&ctrframe_f, "ctrframe_f");
retro_perf_start(&ctrframe_f);
if (ctr->should_resize)
ctr_update_viewport(ctr);
ctrGuSetMemoryFill(true, (u32*)CTR_GPU_FRAMEBUFFER, 0x00000000,
(u32*)(CTR_GPU_FRAMEBUFFER + CTR_TOP_FRAMEBUFFER_WIDTH * CTR_TOP_FRAMEBUFFER_HEIGHT * sizeof(uint32_t)),
0x201, (u32*)CTR_GPU_DEPTHBUFFER, 0x00000000,
(u32*)(CTR_GPU_DEPTHBUFFER + CTR_TOP_FRAMEBUFFER_WIDTH * CTR_TOP_FRAMEBUFFER_HEIGHT * sizeof(uint32_t)),
0x201);
GPUCMD_SetBufferOffset(0);
if (width > ctr->texture_width)
width = ctr->texture_width;
if (height > ctr->texture_height)
height = ctr->texture_height;
if(frame)
{
if(((((u32)(frame)) >= 0x14000000 && ((u32)(frame)) < 0x40000000)) /* frame in linear memory */
&& !((u32)frame & 0x7F) /* 128-byte aligned */
&& !((pitch) & 0xF)) /* 16-byte aligned */
{
/* can copy the buffer directly with the GPU */
ctrGuCopyImage(false, frame, pitch / (ctr->rgb32? 4: 2), height, ctr->rgb32 ? CTRGU_RGBA8: CTRGU_RGB565, false,
ctr->texture_swizzled, ctr->texture_width, ctr->rgb32 ? CTRGU_RGBA8: CTRGU_RGB565, true);
}
else
{
int i;
uint8_t *dst = (uint8_t*)ctr->texture_linear;
const uint8_t *src = frame;
for (i = 0; i < height; i++)
{
memcpy(dst, src, width * (ctr->rgb32? 4: 2));
dst += ctr->texture_width * (ctr->rgb32? 4: 2);
src += pitch;
}
GSPGPU_FlushDataCache(ctr->texture_linear,
ctr->texture_width * ctr->texture_height * (ctr->rgb32? 4: 2));
ctrGuCopyImage(false, ctr->texture_linear, ctr->texture_width, ctr->texture_height, ctr->rgb32 ? CTRGU_RGBA8: CTRGU_RGB565, false,
ctr->texture_swizzled, ctr->texture_width, ctr->rgb32 ? CTRGU_RGBA8: CTRGU_RGB565, true);
}
}
ctrGuSetTexture(GPU_TEXUNIT0, VIRT_TO_PHYS(ctr->texture_swizzled), ctr->texture_width, ctr->texture_height,
(ctr->smooth? GPU_TEXTURE_MAG_FILTER(GPU_LINEAR) | GPU_TEXTURE_MIN_FILTER(GPU_LINEAR)
: GPU_TEXTURE_MAG_FILTER(GPU_NEAREST) | GPU_TEXTURE_MIN_FILTER(GPU_NEAREST)) |
GPU_TEXTURE_WRAP_S(GPU_CLAMP_TO_EDGE) | GPU_TEXTURE_WRAP_T(GPU_CLAMP_TO_EDGE),
ctr->rgb32 ? GPU_RGBA8: GPU_RGB565);
ctr->frame_coords->u = width;
ctr->frame_coords->v = height;
GSPGPU_FlushDataCache(ctr->frame_coords, sizeof(ctr_vertex_t));
ctrGuSetAttributeBuffersAddress(VIRT_TO_PHYS(ctr->frame_coords));
ctrGuSetVertexShaderFloatUniform(0, (float*)&ctr->scale_vector, 1);
/* ARGB --> RGBA */
if (ctr->rgb32)
{
GPU_SetTexEnv(0,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, 0),
GPU_TEVSOURCES(GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR, 0),
GPU_TEVOPERANDS(GPU_TEVOP_RGB_SRC_G, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_MODULATE, GPU_MODULATE,
0x0000FF);
GPU_SetTexEnv(1,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, GPU_PREVIOUS),
GPU_TEVSOURCES(GPU_PREVIOUS, GPU_PREVIOUS, 0),
GPU_TEVOPERANDS(GPU_TEVOP_RGB_SRC_B, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_MULTIPLY_ADD, GPU_MODULATE,
0x00FF00);
GPU_SetTexEnv(2,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, GPU_PREVIOUS),
GPU_TEVSOURCES(GPU_PREVIOUS, GPU_PREVIOUS, 0),
GPU_TEVOPERANDS(GPU_TEVOP_RGB_SRC_ALPHA, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_MULTIPLY_ADD, GPU_MODULATE,
0xFF0000);
}
GPU_DrawArray(GPU_GEOMETRY_PRIM, 0, 1);
/* restore */
if (ctr->rgb32)
{
GPU_SetTexEnv(0,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_PRIMARY_COLOR, 0),
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_PRIMARY_COLOR, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_MODULATE, GPU_MODULATE,
0xFFFFFFFF);
GPU_SetTexEnv(1, GPU_PREVIOUS,GPU_PREVIOUS, 0, 0, 0, 0, 0);
GPU_SetTexEnv(2, GPU_PREVIOUS,GPU_PREVIOUS, 0, 0, 0, 0, 0);
}
if (ctr->menu_texture_enable)
{
GSPGPU_FlushDataCache(ctr->menu.texture_linear,
ctr->menu.texture_width * ctr->menu.texture_height * sizeof(uint16_t));
ctrGuCopyImage(false, ctr->menu.texture_linear, ctr->menu.texture_width, ctr->menu.texture_height, CTRGU_RGBA4444,false,
ctr->menu.texture_swizzled, ctr->menu.texture_width, CTRGU_RGBA4444, true);
ctrGuSetTexture(GPU_TEXUNIT0, VIRT_TO_PHYS(ctr->menu.texture_swizzled), ctr->menu.texture_width, ctr->menu.texture_height,
GPU_TEXTURE_MAG_FILTER(GPU_LINEAR) | GPU_TEXTURE_MIN_FILTER(GPU_LINEAR) |
GPU_TEXTURE_WRAP_S(GPU_CLAMP_TO_EDGE) | GPU_TEXTURE_WRAP_T(GPU_CLAMP_TO_EDGE),
GPU_RGBA4);
ctrGuSetAttributeBuffersAddress(VIRT_TO_PHYS(ctr->menu.frame_coords));
ctrGuSetVertexShaderFloatUniform(0, (float*)&ctr->menu.scale_vector, 1);
GPU_DrawArray(GPU_GEOMETRY_PRIM, 0, 1);
}
GPU_FinishDrawing();
GPUCMD_Finalize();
ctrGuFlushAndRun(true);
ctrGuDisplayTransfer(true, CTR_GPU_FRAMEBUFFER, 240,400, CTRGU_RGBA8,
gfxGetFramebuffer(GFX_TOP, GFX_LEFT, NULL, NULL), 240,400,CTRGU_RGB8, CTRGU_MULTISAMPLE_NONE);
gfxSwapBuffersGpu();
retro_perf_stop(&ctrframe_f);
return true;
}
static bool ctr_frame(void* data, const void* frame,
unsigned width, unsigned height,
uint64_t frame_count,
unsigned pitch, const char* msg)
{
uint32_t diff;
static uint64_t currentTick,lastTick;
ctr_video_t *ctr = (ctr_video_t*)data;
settings_t *settings = config_get_ptr();
static float fps = 0.0;
static int total_frames = 0;
static int frames = 0;
static struct retro_perf_counter ctrframe_f = {0};
uint32_t state_tmp;
touchPosition state_tmp_touch;
extern bool select_pressed;
if (!width || !height)
{
gspWaitForEvent(GSPGPU_EVENT_VBlank0, true);
return true;
}
if(!aptMainLoop())
{
command_event(CMD_EVENT_QUIT, NULL);
return true;
}
if (select_pressed)
{
command_event(CMD_EVENT_QUIT, NULL);
return true;
}
state_tmp = hidKeysDown();
hidTouchRead(&state_tmp_touch);
if((state_tmp & KEY_TOUCH) && (state_tmp_touch.py < 120))
{
Handle lcd_handle;
u8 not_2DS;
extern PrintConsole* currentConsole;
gfxBottomFramebuffers[0] = ctr->lcd_buttom_on ? (u8*)ctr->empty_framebuffer:
(u8*)currentConsole->frameBuffer;
CFGU_GetModelNintendo2DS(¬_2DS);
if(not_2DS && srvGetServiceHandle(&lcd_handle, "gsp::Lcd") >= 0)
{
u32 *cmdbuf = getThreadCommandBuffer();
cmdbuf[0] = ctr->lcd_buttom_on? 0x00120040: 0x00110040;
cmdbuf[1] = 2;
svcSendSyncRequest(lcd_handle);
svcCloseHandle(lcd_handle);
}
ctr->lcd_buttom_on = !ctr->lcd_buttom_on;
}
svcWaitSynchronization(gspEvents[GSPGPU_EVENT_P3D], 20000000);
svcClearEvent(gspEvents[GSPGPU_EVENT_P3D]);
svcWaitSynchronization(gspEvents[GSPGPU_EVENT_PPF], 20000000);
svcClearEvent(gspEvents[GSPGPU_EVENT_PPF]);
frames++;
if (ctr->vsync)
svcWaitSynchronization(gspEvents[GSPGPU_EVENT_VBlank0], U64_MAX);
svcClearEvent(gspEvents[GSPGPU_EVENT_VBlank0]);
currentTick = svcGetSystemTick();
diff = currentTick - lastTick;
if(diff > CTR_CPU_TICKS_PER_SECOND)
{
fps = (float)frames * ((float) CTR_CPU_TICKS_PER_SECOND / (float) diff);
lastTick = currentTick;
frames = 0;
}
//#define CTR_INSPECT_MEMORY_USAGE
#ifdef CTR_INSPECT_MEMORY_USAGE
uint32_t ctr_get_stack_usage(void);
void ctr_linear_get_stats(void);
extern u32 __linear_heap_size;
extern u32 __heap_size;
MemInfo mem_info;
PageInfo page_info;
u32 query_addr = 0x08000000;
printf(PRINTFPOS(0,0));
while (query_addr < 0x40000000)
{
svcQueryMemory(&mem_info, &page_info, query_addr);
printf("0x%08X --> 0x%08X (0x%08X) \n", mem_info.base_addr, mem_info.base_addr + mem_info.size, mem_info.size);
query_addr = mem_info.base_addr + mem_info.size;
if(query_addr == 0x1F000000)
query_addr = 0x30000000;
}
// static u32* dummy_pointer;
// if(total_frames == 500)
// dummy_pointer = malloc(0x2000000);
// if(total_frames == 1000)
// free(dummy_pointer);
printf("========================================");
printf("0x%08X 0x%08X 0x%08X\n", __heap_size, gpuCmdBufOffset, (__linear_heap_size - linearSpaceFree()));
printf("fps: %8.4f frames: %i (%X)\n", fps, total_frames++, (__linear_heap_size - linearSpaceFree()));
printf("========================================");
u32 app_memory = *((u32*)0x1FF80040);
u64 mem_used;
svcGetSystemInfo(&mem_used, 0, 1);
printf("total mem : 0x%08X \n", app_memory);
printf("used: 0x%08X free: 0x%08X \n", (u32)mem_used, app_memory - (u32)mem_used);
static u32 stack_usage = 0;
extern u32 __stack_bottom;
if(!(total_frames & 0x3F))
stack_usage = ctr_get_stack_usage();
printf("stack total:0x%08X used: 0x%08X\n", 0x10000000 - __stack_bottom, stack_usage);
printf("========================================");
ctr_linear_get_stats();
printf("========================================");
#else
printf(PRINTFPOS(29,0)"fps: %8.4f frames: %i\r", fps, total_frames++);
#endif
fflush(stdout);
performance_counter_init(&ctrframe_f, "ctrframe_f");
performance_counter_start(&ctrframe_f);
if (ctr->should_resize)
ctr_update_viewport(ctr);
ctrGuSetMemoryFill(true, (u32*)CTR_TOP_FRAMEBUFFER, 0x00000000,
(u32*)(CTR_TOP_FRAMEBUFFER + 2 * CTR_TOP_FRAMEBUFFER_WIDTH * CTR_TOP_FRAMEBUFFER_HEIGHT * sizeof(uint32_t)),
0x201, (u32*)CTR_GPU_DEPTHBUFFER, 0x00000000,
(u32*)(CTR_GPU_DEPTHBUFFER + CTR_TOP_FRAMEBUFFER_WIDTH * CTR_TOP_FRAMEBUFFER_HEIGHT * sizeof(uint32_t)),
0x201);
GPUCMD_SetBufferOffset(0);
if (width > ctr->texture_width)
width = ctr->texture_width;
if (height > ctr->texture_height)
height = ctr->texture_height;
if(frame)
{
if(((((u32)(frame)) >= 0x14000000 && ((u32)(frame)) < 0x40000000)) /* frame in linear memory */
&& !((u32)frame & 0x7F) /* 128-byte aligned */
&& !(pitch & 0xF) /* 16-byte aligned */
&& (pitch > 0x40))
{
/* can copy the buffer directly with the GPU */
// GSPGPU_FlushDataCache(frame, pitch * height);
ctrGuSetCommandList_First(true,(void*)frame, pitch * height,0,0,0,0);
ctrGuCopyImage(true, frame, pitch / (ctr->rgb32? 4: 2), height, ctr->rgb32 ? CTRGU_RGBA8: CTRGU_RGB565, false,
ctr->texture_swizzled, ctr->texture_width, ctr->rgb32 ? CTRGU_RGBA8: CTRGU_RGB565, true);
}
else
{
int i;
uint8_t *dst = (uint8_t*)ctr->texture_linear;
const uint8_t *src = frame;
for (i = 0; i < height; i++)
{
memcpy(dst, src, width * (ctr->rgb32? 4: 2));
dst += ctr->texture_width * (ctr->rgb32? 4: 2);
src += pitch;
}
GSPGPU_FlushDataCache(ctr->texture_linear,
ctr->texture_width * ctr->texture_height * (ctr->rgb32? 4: 2));
ctrGuCopyImage(false, ctr->texture_linear, ctr->texture_width, ctr->texture_height, ctr->rgb32 ? CTRGU_RGBA8: CTRGU_RGB565, false,
ctr->texture_swizzled, ctr->texture_width, ctr->rgb32 ? CTRGU_RGBA8: CTRGU_RGB565, true);
}
ctr->frame_coords->u0 = 0;
ctr->frame_coords->v0 = 0;
ctr->frame_coords->u1 = width;
ctr->frame_coords->v1 = height;
GSPGPU_FlushDataCache(ctr->frame_coords, sizeof(ctr_vertex_t));
ctrGuSetVertexShaderFloatUniform(0, (float*)&ctr->scale_vector, 1);
}
ctrGuSetTexture(GPU_TEXUNIT0, VIRT_TO_PHYS(ctr->texture_swizzled), ctr->texture_width, ctr->texture_height,
(ctr->smooth? GPU_TEXTURE_MAG_FILTER(GPU_LINEAR) | GPU_TEXTURE_MIN_FILTER(GPU_LINEAR)
: GPU_TEXTURE_MAG_FILTER(GPU_NEAREST) | GPU_TEXTURE_MIN_FILTER(GPU_NEAREST)) |
GPU_TEXTURE_WRAP_S(GPU_CLAMP_TO_EDGE) | GPU_TEXTURE_WRAP_T(GPU_CLAMP_TO_EDGE),
ctr->rgb32 ? GPU_RGBA8: GPU_RGB565);
ctr_check_3D_slider(ctr);
/* ARGB --> RGBA */
if (ctr->rgb32)
{
GPU_SetTexEnv(0,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, 0),
GPU_TEVSOURCES(GPU_PRIMARY_COLOR, GPU_PRIMARY_COLOR, 0),
GPU_TEVOPERANDS(GPU_TEVOP_RGB_SRC_G, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_MODULATE, GPU_MODULATE,
0x0000FF);
GPU_SetTexEnv(1,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, GPU_PREVIOUS),
GPU_TEVSOURCES(GPU_PREVIOUS, GPU_PREVIOUS, 0),
GPU_TEVOPERANDS(GPU_TEVOP_RGB_SRC_B, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_MULTIPLY_ADD, GPU_MODULATE,
0x00FF00);
GPU_SetTexEnv(2,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_CONSTANT, GPU_PREVIOUS),
GPU_TEVSOURCES(GPU_PREVIOUS, GPU_PREVIOUS, 0),
GPU_TEVOPERANDS(GPU_TEVOP_RGB_SRC_ALPHA, 0, 0),
GPU_TEVOPERANDS(0, 0, 0),
GPU_MULTIPLY_ADD, GPU_MODULATE,
0xFF0000);
}
GPU_SetViewport(VIRT_TO_PHYS(CTR_GPU_DEPTHBUFFER),
VIRT_TO_PHYS(CTR_TOP_FRAMEBUFFER),
0, 0, CTR_TOP_FRAMEBUFFER_HEIGHT,
ctr->video_mode == CTR_VIDEO_MODE_800x240 ? CTR_TOP_FRAMEBUFFER_WIDTH * 2 : CTR_TOP_FRAMEBUFFER_WIDTH);
if (ctr->video_mode == CTR_VIDEO_MODE_3D)
{
if (ctr->menu_texture_enable)
{
ctrGuSetAttributeBuffersAddress(VIRT_TO_PHYS(&ctr->frame_coords[1]));
GPU_DrawArray(GPU_GEOMETRY_PRIM, 0, 1);
ctrGuSetAttributeBuffersAddress(VIRT_TO_PHYS(&ctr->frame_coords[2]));
}
else
{
ctrGuSetAttributeBuffersAddress(VIRT_TO_PHYS(ctr->frame_coords));
GPU_DrawArray(GPU_GEOMETRY_PRIM, 0, 1);
}
GPU_SetViewport(VIRT_TO_PHYS(CTR_GPU_DEPTHBUFFER),
VIRT_TO_PHYS(CTR_TOP_FRAMEBUFFER_RIGHT),
0, 0, CTR_TOP_FRAMEBUFFER_HEIGHT,
CTR_TOP_FRAMEBUFFER_WIDTH);
}
else
ctrGuSetAttributeBuffersAddress(VIRT_TO_PHYS(ctr->frame_coords));
GPU_DrawArray(GPU_GEOMETRY_PRIM, 0, 1);
/* restore */
if (ctr->rgb32)
{
GPU_SetTexEnv(0, GPU_TEXTURE0, GPU_TEXTURE0, 0, 0, GPU_REPLACE, GPU_REPLACE, 0);
GPU_SetTexEnv(1, GPU_PREVIOUS, GPU_PREVIOUS, 0, 0, 0, 0, 0);
GPU_SetTexEnv(2, GPU_PREVIOUS, GPU_PREVIOUS, 0, 0, 0, 0, 0);
}
if (ctr->menu_texture_enable)
{
GSPGPU_FlushDataCache(ctr->menu.texture_linear,
ctr->menu.texture_width * ctr->menu.texture_height * sizeof(uint16_t));
ctrGuCopyImage(false, ctr->menu.texture_linear, ctr->menu.texture_width, ctr->menu.texture_height, CTRGU_RGBA4444,false,
ctr->menu.texture_swizzled, ctr->menu.texture_width, CTRGU_RGBA4444, true);
ctrGuSetTexture(GPU_TEXUNIT0, VIRT_TO_PHYS(ctr->menu.texture_swizzled), ctr->menu.texture_width, ctr->menu.texture_height,
GPU_TEXTURE_MAG_FILTER(GPU_LINEAR) | GPU_TEXTURE_MIN_FILTER(GPU_LINEAR) |
GPU_TEXTURE_WRAP_S(GPU_CLAMP_TO_EDGE) | GPU_TEXTURE_WRAP_T(GPU_CLAMP_TO_EDGE),
GPU_RGBA4);
ctrGuSetVertexShaderFloatUniform(0, (float*)&ctr->menu.scale_vector, 1);
GPU_SetViewport(VIRT_TO_PHYS(CTR_GPU_DEPTHBUFFER),
VIRT_TO_PHYS(CTR_TOP_FRAMEBUFFER),
0, 0, CTR_TOP_FRAMEBUFFER_HEIGHT,
ctr->video_mode == CTR_VIDEO_MODE_800x240 ? CTR_TOP_FRAMEBUFFER_WIDTH * 2 : CTR_TOP_FRAMEBUFFER_WIDTH);
ctrGuSetAttributeBuffersAddress(VIRT_TO_PHYS(ctr->menu.frame_coords));
GPU_DrawArray(GPU_GEOMETRY_PRIM, 0, 1);
if (ctr->video_mode == CTR_VIDEO_MODE_3D)
{
GPU_SetViewport(VIRT_TO_PHYS(CTR_GPU_DEPTHBUFFER),
VIRT_TO_PHYS(CTR_TOP_FRAMEBUFFER_RIGHT),
0, 0, CTR_TOP_FRAMEBUFFER_HEIGHT,
CTR_TOP_FRAMEBUFFER_WIDTH);
GPU_DrawArray(GPU_GEOMETRY_PRIM, 0, 1);
}
}
GPU_FinishDrawing();
GPUCMD_Finalize();
ctrGuFlushAndRun(true);
ctrGuDisplayTransfer(true, CTR_TOP_FRAMEBUFFER,
240,
ctr->video_mode == CTR_VIDEO_MODE_800x240 ? 800 : 400,
CTRGU_RGBA8,
gfxTopLeftFramebuffers[ctr->current_buffer_top], 240,CTRGU_RGB8, CTRGU_MULTISAMPLE_NONE);
if ((ctr->video_mode == CTR_VIDEO_MODE_400x240) || (ctr->video_mode == CTR_VIDEO_MODE_3D))
ctrGuDisplayTransfer(true, CTR_TOP_FRAMEBUFFER_RIGHT,
240,
400,
CTRGU_RGBA8,
gfxTopRightFramebuffers[ctr->current_buffer_top], 240,CTRGU_RGB8, CTRGU_MULTISAMPLE_NONE);
// Swap buffers :
ctr->current_buffer_top ^= 1;
extern GSPGPU_FramebufferInfo topFramebufferInfo;
extern u8* gfxSharedMemory;
extern u8 gfxThreadID;
topFramebufferInfo.active_framebuf=ctr->current_buffer_top;
topFramebufferInfo.framebuf0_vaddr=(u32*)gfxTopLeftFramebuffers[ctr->current_buffer_top];
if(ctr->video_mode == CTR_VIDEO_MODE_800x240)
{
topFramebufferInfo.framebuf1_vaddr=(u32*)(gfxTopLeftFramebuffers[ctr->current_buffer_top] + 240 * 3);
topFramebufferInfo.framebuf_widthbytesize = 240 * 3 * 2;
}
else
{
topFramebufferInfo.framebuf1_vaddr=(u32*)gfxTopRightFramebuffers[ctr->current_buffer_top];
topFramebufferInfo.framebuf_widthbytesize = 240 * 3;
}
topFramebufferInfo.format=(1<<8)|(1<<5)|GSP_BGR8_OES;
topFramebufferInfo.framebuf_dispselect=ctr->current_buffer_top;
topFramebufferInfo.unk=0x00000000;
u8* framebufferInfoHeader=gfxSharedMemory+0x200+gfxThreadID*0x80;
GSPGPU_FramebufferInfo* framebufferInfo=(GSPGPU_FramebufferInfo*)&framebufferInfoHeader[0x4];
framebufferInfoHeader[0x0] ^= 1;
framebufferInfo[framebufferInfoHeader[0x0]] = topFramebufferInfo;
framebufferInfoHeader[0x1]=1;
performance_counter_stop(&ctrframe_f);
return true;
}
void RenderTopScreen()
{
// notes on the drawing process
// GPU hangs if we attempt to draw an even number of arrays :/ which is why we draw the border 'twice'
// textures used here are actually 512x256. TODO: investigate if GPU_SetTexture() really has the params in the wrong order
// or if we did something wrong.
//general setup
GPU_SetViewport((u32*)osConvertVirtToPhys((u32)gpuDOut),(u32*)osConvertVirtToPhys((u32)gpuOut),0,0,240*2,400);
GPU_DepthRange(-1.0f, 0.0f);
GPU_SetFaceCulling(GPU_CULL_BACK_CCW);
GPU_SetStencilTest(false, GPU_ALWAYS, 0x00);
GPU_SetDepthTest(false, GPU_ALWAYS, 0x1F);
// ?
GPUCMD_AddSingleParam(0x00010062, 0x00000000); //param always 0x0 according to code
GPUCMD_AddSingleParam(0x000F0118, 0x00000000);
//setup shader
SHDR_UseProgram(shader, 0);
//?
GPUCMD_AddSingleParam(0x000F0100, 0x00E40100);
GPUCMD_AddSingleParam(0x000F0101, 0x01010000);
GPUCMD_AddSingleParam(0x000F0104, 0x00000010);
//texturing stuff
GPUCMD_AddSingleParam(0x0002006F, 0x00000100);
GPUCMD_AddSingleParam(0x000F0080, 0x00011001); //enables/disables texturing
//texenv
GPU_SetTexEnv(0,
GPU_TEVSOURCES(GPU_TEXTURE0, 0, 0),
GPU_TEVSOURCES(GPU_CONSTANT, 0, 0),
GPU_TEVOPERANDS(0,0,0),
GPU_TEVOPERANDS(0,0,0),
GPU_REPLACE, GPU_REPLACE,
0xFFFFFFFF);
GPU_SetDummyTexEnv(1);
GPU_SetDummyTexEnv(2);
GPU_SetDummyTexEnv(3);
GPU_SetDummyTexEnv(4);
GPU_SetDummyTexEnv(5);
//texturing stuff
GPU_SetTexture((u32*)osConvertVirtToPhys((u32)BorderTex),256,512,0,GPU_RGBA8); // texture is actually 512x256
//setup matrices
setUniformMatrix(0x24, mvMatrix);
setUniformMatrix(0x20, projMatrix);
// border
GPU_SetAttributeBuffers(2, (u32*)osConvertVirtToPhys((u32)borderVertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT)|GPU_ATTRIBFMT(1, 2, GPU_FLOAT),
0xFFC, 0x10, 1, (u32[]){0x00000000}, (u64[]){0x10}, (u8[]){2});
GPU_DrawArray(GPU_TRIANGLES, 3);
GPU_DrawArray(GPU_TRIANGLES, 2*3);
GPU_DepthRange(-1.0f, 0.0f);
GPU_SetFaceCulling(GPU_CULL_BACK_CCW);
GPU_SetStencilTest(false, GPU_ALWAYS, 0x00);
GPU_SetDepthTest(false, GPU_ALWAYS, 0x1F);
GPUCMD_AddSingleParam(0x00010062, 0x00000000);
GPUCMD_AddSingleParam(0x000F0118, 0x00000000);
GPUCMD_AddSingleParam(0x000F0100, 0x00000100);
GPUCMD_AddSingleParam(0x000F0101, 0x01010000);
GPUCMD_AddSingleParam(0x000F0104, 0x00000010);
//texturing stuff
GPUCMD_AddSingleParam(0x0002006F, 0x00000700); // enables/disables texcoord output
GPUCMD_AddSingleParam(0x000F0080, 0x00011007); //enables/disables texturing
// TEXTURE ENV STAGES
// ---
// blending operation: (Main.Color +- (Sub.Color * Main.Alpha)) * Sub.Alpha
// Main.Alpha = 0/255 depending on color math
// Sub.Alpha = 128/255 depending on color div2
// note: the main/sub intensities are halved to prevent overflow during the operations.
// (each TEV stage output is clamped to [0,255])
// stage 4 makes up for this
// ---
// STAGE 1: Out.Color = Sub.Color * Main.Alpha, Out.Alpha = Sub.Alpha + (1-Main.Alpha) (cancel out div2 when color math doesn't happen)
GPU_SetTexEnv(0,
GPU_TEVSOURCES(GPU_TEXTURE1, GPU_TEXTURE0, 0),
GPU_TEVSOURCES(GPU_TEXTURE1, GPU_TEXTURE0, 0),
GPU_TEVOPERANDS(0,2,0),
GPU_TEVOPERANDS(0,1,0),
GPU_MODULATE,
GPU_ADD,
0xFFFFFFFF);
// STAGE 2: Out.Color = Main.Color +- Prev.Color, Out.Alpha = Prev.Alpha
GPU_SetTexEnv(1,
GPU_TEVSOURCES(GPU_TEXTURE0, GPU_PREVIOUS, 0),
GPU_TEVSOURCES(GPU_PREVIOUS, 0, 0),
GPU_TEVOPERANDS(0,0,0),
GPU_TEVOPERANDS(0,0,0),
(PPU_ColorMath & 0x80) ? GPU_SUBTRACT:GPU_ADD,
GPU_REPLACE,
0xFFFFFFFF);
// STAGE 3: Out.Color = Prev.Color * Prev.Alpha, Out.Alpha = Prev.Alpha
GPU_SetTexEnv(2,
GPU_TEVSOURCES(GPU_PREVIOUS, GPU_PREVIOUS, 0),
GPU_TEVSOURCES(GPU_PREVIOUS, 0, 0),
GPU_TEVOPERANDS(0,2,0),
GPU_TEVOPERANDS(0,0,0),
GPU_MODULATE,
GPU_REPLACE,
0xFFFFFFFF);
// STAGE 4: Out.Color = Prev.Color + Prev.Color (doubling color intensity), Out.Alpha = Const.Alpha
GPU_SetTexEnv(3,
GPU_TEVSOURCES(GPU_PREVIOUS, GPU_PREVIOUS, 0),
GPU_TEVSOURCES(GPU_CONSTANT, 0, 0),
GPU_TEVOPERANDS(0,0,0),
GPU_TEVOPERANDS(0,0,0),
GPU_ADD,
GPU_REPLACE,
0xFFFFFFFF);
// STAGE 5: master brightness - Out.Color = Prev.Color * Bright.Alpha, Out.Alpha = Const.Alpha
GPU_SetTexEnv(4,
GPU_TEVSOURCES(GPU_PREVIOUS, GPU_TEXTURE2, 0),
GPU_TEVSOURCES(GPU_CONSTANT, 0, 0),
GPU_TEVOPERANDS(0,2,0),
GPU_TEVOPERANDS(0,0,0),
GPU_MODULATE,
GPU_REPLACE,
0xFFFFFFFF);
// STAGE 6: dummy
GPU_SetDummyTexEnv(5);
GPU_SetAttributeBuffers(3, (u32*)osConvertVirtToPhys((u32)screenVertices),
GPU_ATTRIBFMT(0, 3, GPU_FLOAT)|GPU_ATTRIBFMT(1, 2, GPU_FLOAT)|GPU_ATTRIBFMT(2, 2, GPU_FLOAT),
0xFFC, 0x210, 1, (u32[]){0x00000000}, (u64[]){0x210}, (u8[]){3});
GPU_SetTexture((u32*)osConvertVirtToPhys((u32)MainScreenTex),256,512,0,GPU_RGBA8);
GPU_SetTexture1((u32*)osConvertVirtToPhys((u32)SubScreenTex),256,512,0,GPU_RGBA8);
GPU_SetTexture2((u32*)osConvertVirtToPhys((u32)BrightnessTex),256,8,0x200,GPU_A8);
GPU_DrawArray(GPU_TRIANGLES, 2*3);
}
