void vbo_exec_vtx_init( struct vbo_exec_context *exec )
{
struct gl_context *ctx = exec->ctx;
struct vbo_context *vbo = vbo_context(ctx);
GLuint i;
/* Allocate a buffer object. Will just reuse this object
* continuously, unless vbo_use_buffer_objects() is called to enable
* use of real VBOs.
*/
_mesa_reference_buffer_object(ctx,
&exec->vtx.bufferobj,
ctx->Shared->NullBufferObj);
ASSERT(!exec->vtx.buffer_map);
exec->vtx.buffer_map = _mesa_align_malloc(VBO_VERT_BUFFER_SIZE, 64);
exec->vtx.buffer_ptr = exec->vtx.buffer_map;
vbo_exec_vtxfmt_init( exec );
_mesa_noop_vtxfmt_init(&exec->vtxfmt_noop);
for (i = 0 ; i < VBO_ATTRIB_MAX ; i++) {
ASSERT(i < Elements(exec->vtx.attrsz));
exec->vtx.attrsz[i] = 0;
ASSERT(i < Elements(exec->vtx.attrtype));
exec->vtx.attrtype[i] = GL_FLOAT;
ASSERT(i < Elements(exec->vtx.active_sz));
exec->vtx.active_sz[i] = 0;
}
for (i = 0 ; i < VERT_ATTRIB_MAX; i++) {
ASSERT(i < Elements(exec->vtx.inputs));
ASSERT(i < Elements(exec->vtx.arrays));
exec->vtx.inputs[i] = &exec->vtx.arrays[i];
}
{
struct gl_client_array *arrays = exec->vtx.arrays;
unsigned i;
memcpy(arrays, &vbo->currval[VBO_ATTRIB_POS],
VERT_ATTRIB_FF_MAX * sizeof(arrays[0]));
for (i = 0; i < VERT_ATTRIB_FF_MAX; ++i) {
struct gl_client_array *array;
array = &arrays[VERT_ATTRIB_FF(i)];
array->BufferObj = NULL;
_mesa_reference_buffer_object(ctx, &array->BufferObj,
vbo->currval[VBO_ATTRIB_POS+i].BufferObj);
}
memcpy(arrays + VERT_ATTRIB_GENERIC(0),
&vbo->currval[VBO_ATTRIB_GENERIC0],
VERT_ATTRIB_GENERIC_MAX * sizeof(arrays[0]));
for (i = 0; i < VERT_ATTRIB_GENERIC_MAX; ++i) {
struct gl_client_array *array;
array = &arrays[VERT_ATTRIB_GENERIC(i)];
array->BufferObj = NULL;
_mesa_reference_buffer_object(ctx, &array->BufferObj,
vbo->currval[VBO_ATTRIB_GENERIC0+i].BufferObj);
}
}
exec->vtx.vertex_size = 0;
exec->begin_vertices_flags = FLUSH_UPDATE_CURRENT;
}
static HRESULT WINAPI
drm_create_adapter( int fd,
ID3DAdapter9 **ppAdapter )
{
struct d3dadapter9drm_context *ctx = CALLOC_STRUCT(d3dadapter9drm_context);
HRESULT hr;
int i, different_device;
const char *paths[] = {
getenv("D3D9_DRIVERS_PATH"),
getenv("D3D9_DRIVERS_DIR"),
PIPE_SEARCH_DIR
};
if (!ctx) { return E_OUTOFMEMORY; }
ctx->base.destroy = drm_destroy;
fd = loader_get_user_preferred_fd(fd, &different_device);
ctx->base.linear_framebuffer = !!different_device;
/* use pipe-loader to dlopen appropriate drm driver */
if (!pipe_loader_drm_probe_fd(&ctx->dev, fd, FALSE)) {
DBG("Failed to probe drm fd %d.\n", fd);
FREE(ctx);
close(fd);
return D3DERR_DRIVERINTERNALERROR;
}
/* use pipe-loader to create a drm screen (hal) */
ctx->base.hal = NULL;
for (i = 0; !ctx->base.hal && i < Elements(paths); ++i) {
if (!paths[i]) { continue; }
ctx->base.hal = pipe_loader_create_screen(ctx->dev, paths[i]);
}
if (!ctx->base.hal) {
DBG("Unable to load requested driver.\n");
pipe_loader_release(&ctx->dev, 1);
FREE(ctx);
return D3DERR_DRIVERINTERNALERROR;
}
/* wrap it to create a software screen that can share resources */
if (pipe_loader_sw_probe_wrapped(&ctx->swdev, ctx->base.hal)) {
ctx->base.ref = NULL;
for (i = 0; !ctx->base.ref && i < Elements(paths); ++i) {
if (!paths[i]) { continue; }
ctx->base.ref = pipe_loader_create_screen(ctx->swdev, paths[i]);
}
}
if (!ctx->base.ref) {
DBG("Couldn't wrap drm screen to swrast screen. Software devices "
"will be unavailable.\n");
}
/* read out PCI info */
read_descriptor(&ctx->base, fd);
/* create and return new ID3DAdapter9 */
hr = NineAdapter9_new(&ctx->base, (struct NineAdapter9 **)ppAdapter);
if (FAILED(hr)) {
if (ctx->swdev) { pipe_loader_release(&ctx->swdev, 1); }
pipe_loader_release(&ctx->dev, 1);
FREE(ctx);
return hr;
}
return D3D_OK;
}
/**
* Return ARB_v/f_prog-style input attrib string.
*/
static const char *
arb_input_attrib_string(GLint index, GLenum progType)
{
/*
* These strings should match the VERT_ATTRIB_x and VARYING_SLOT_x tokens.
*/
static const char *const vertAttribs[] = {
"vertex.position",
"vertex.weight",
"vertex.normal",
"vertex.color.primary",
"vertex.color.secondary",
"vertex.fogcoord",
"vertex.(six)", /* VERT_ATTRIB_COLOR_INDEX */
"vertex.(seven)", /* VERT_ATTRIB_EDGEFLAG */
"vertex.texcoord[0]",
"vertex.texcoord[1]",
"vertex.texcoord[2]",
"vertex.texcoord[3]",
"vertex.texcoord[4]",
"vertex.texcoord[5]",
"vertex.texcoord[6]",
"vertex.texcoord[7]",
"vertex.(sixteen)", /* VERT_ATTRIB_POINT_SIZE */
"vertex.attrib[0]",
"vertex.attrib[1]",
"vertex.attrib[2]",
"vertex.attrib[3]",
"vertex.attrib[4]",
"vertex.attrib[5]",
"vertex.attrib[6]",
"vertex.attrib[7]",
"vertex.attrib[8]",
"vertex.attrib[9]",
"vertex.attrib[10]",
"vertex.attrib[11]",
"vertex.attrib[12]",
"vertex.attrib[13]",
"vertex.attrib[14]",
"vertex.attrib[15]" /* MAX_VARYING = 16 */
};
static const char *const fragAttribs[] = {
"fragment.position",
"fragment.color.primary",
"fragment.color.secondary",
"fragment.fogcoord",
"fragment.texcoord[0]",
"fragment.texcoord[1]",
"fragment.texcoord[2]",
"fragment.texcoord[3]",
"fragment.texcoord[4]",
"fragment.texcoord[5]",
"fragment.texcoord[6]",
"fragment.texcoord[7]",
"fragment.(twelve)", /* VARYING_SLOT_PSIZ */
"fragment.(thirteen)", /* VARYING_SLOT_BFC0 */
"fragment.(fourteen)", /* VARYING_SLOT_BFC1 */
"fragment.(fifteen)", /* VARYING_SLOT_EDGE */
"fragment.(sixteen)", /* VARYING_SLOT_CLIP_VERTEX */
"fragment.(seventeen)", /* VARYING_SLOT_CLIP_DIST0 */
"fragment.(eighteen)", /* VARYING_SLOT_CLIP_DIST1 */
"fragment.(nineteen)", /* VARYING_SLOT_PRIMITIVE_ID */
"fragment.(twenty)", /* VARYING_SLOT_LAYER */
"fragment.(twenty-one)", /* VARYING_SLOT_FACE */
"fragment.(twenty-two)", /* VARYING_SLOT_PNTC */
"fragment.varying[0]",
"fragment.varying[1]",
"fragment.varying[2]",
"fragment.varying[3]",
"fragment.varying[4]",
"fragment.varying[5]",
"fragment.varying[6]",
"fragment.varying[7]",
"fragment.varying[8]",
"fragment.varying[9]",
"fragment.varying[10]",
"fragment.varying[11]",
"fragment.varying[12]",
"fragment.varying[13]",
"fragment.varying[14]",
"fragment.varying[15]",
"fragment.varying[16]",
"fragment.varying[17]",
"fragment.varying[18]",
"fragment.varying[19]",
"fragment.varying[20]",
"fragment.varying[21]",
"fragment.varying[22]",
"fragment.varying[23]",
"fragment.varying[24]",
"fragment.varying[25]",
"fragment.varying[26]",
"fragment.varying[27]",
"fragment.varying[28]",
"fragment.varying[29]",
"fragment.varying[30]",
"fragment.varying[31]", /* MAX_VARYING = 32 */
};
/* sanity checks */
STATIC_ASSERT(Elements(vertAttribs) == VERT_ATTRIB_MAX);
STATIC_ASSERT(Elements(fragAttribs) == VARYING_SLOT_MAX);
assert(strcmp(vertAttribs[VERT_ATTRIB_TEX0], "vertex.texcoord[0]") == 0);
assert(strcmp(vertAttribs[VERT_ATTRIB_GENERIC15], "vertex.attrib[15]") == 0);
assert(strcmp(fragAttribs[VARYING_SLOT_TEX0], "fragment.texcoord[0]") == 0);
assert(strcmp(fragAttribs[VARYING_SLOT_VAR0+15], "fragment.varying[15]") == 0);
if (progType == GL_VERTEX_PROGRAM_ARB) {
assert(index < Elements(vertAttribs));
return vertAttribs[index];
}
else {
assert(progType == GL_FRAGMENT_PROGRAM_ARB);
assert(index < Elements(fragAttribs));
return fragAttribs[index];
}
}
static struct st_context *
st_create_context_priv( struct gl_context *ctx, struct pipe_context *pipe,
const struct st_config_options *options)
{
struct pipe_screen *screen = pipe->screen;
uint i;
struct st_context *st = ST_CALLOC_STRUCT( st_context );
st->options = *options;
ctx->st = st;
st->ctx = ctx;
st->pipe = pipe;
/* XXX: this is one-off, per-screen init: */
st_debug_init();
/* state tracker needs the VBO module */
_vbo_CreateContext(ctx);
st->dirty.mesa = ~0;
st->dirty.st = ~0;
/* Create upload manager for vertex data for glBitmap, glDrawPixels,
* glClear, etc.
*/
st->uploader = u_upload_create(st->pipe, 65536, 4, PIPE_BIND_VERTEX_BUFFER);
if (!screen->get_param(screen, PIPE_CAP_USER_INDEX_BUFFERS)) {
st->indexbuf_uploader = u_upload_create(st->pipe, 128 * 1024, 4,
PIPE_BIND_INDEX_BUFFER);
}
if (!screen->get_param(screen, PIPE_CAP_USER_CONSTANT_BUFFERS)) {
unsigned alignment =
screen->get_param(screen, PIPE_CAP_CONSTANT_BUFFER_OFFSET_ALIGNMENT);
st->constbuf_uploader = u_upload_create(pipe, 128 * 1024, alignment,
PIPE_BIND_CONSTANT_BUFFER);
}
st->cso_context = cso_create_context(pipe);
st_init_atoms( st );
st_init_bitmap(st);
st_init_clear(st);
st_init_draw( st );
/* Choose texture target for glDrawPixels, glBitmap, renderbuffers */
if (pipe->screen->get_param(pipe->screen, PIPE_CAP_NPOT_TEXTURES))
st->internal_target = PIPE_TEXTURE_2D;
else
st->internal_target = PIPE_TEXTURE_RECT;
/* Vertex element objects used for drawing rectangles for glBitmap,
* glDrawPixels, glClear, etc.
*/
for (i = 0; i < Elements(st->velems_util_draw); i++) {
memset(&st->velems_util_draw[i], 0, sizeof(struct pipe_vertex_element));
st->velems_util_draw[i].src_offset = i * 4 * sizeof(float);
st->velems_util_draw[i].instance_divisor = 0;
st->velems_util_draw[i].vertex_buffer_index =
cso_get_aux_vertex_buffer_slot(st->cso_context);
st->velems_util_draw[i].src_format = PIPE_FORMAT_R32G32B32A32_FLOAT;
}
/* we want all vertex data to be placed in buffer objects */
vbo_use_buffer_objects(ctx);
/* make sure that no VBOs are left mapped when we're drawing. */
vbo_always_unmap_buffers(ctx);
/* Need these flags:
*/
st->ctx->FragmentProgram._MaintainTexEnvProgram = GL_TRUE;
st->ctx->VertexProgram._MaintainTnlProgram = GL_TRUE;
st->pixel_xfer.cache = _mesa_new_program_cache();
st->has_stencil_export =
screen->get_param(screen, PIPE_CAP_SHADER_STENCIL_EXPORT);
st->has_shader_model3 = screen->get_param(screen, PIPE_CAP_SM3);
st->has_etc1 = screen->is_format_supported(screen, PIPE_FORMAT_ETC1_RGB8,
PIPE_TEXTURE_2D, 0,
PIPE_BIND_SAMPLER_VIEW);
st->prefer_blit_based_texture_transfer = screen->get_param(screen,
PIPE_CAP_PREFER_BLIT_BASED_TEXTURE_TRANSFER);
st->needs_texcoord_semantic =
screen->get_param(screen, PIPE_CAP_TGSI_TEXCOORD);
st->apply_texture_swizzle_to_border_color =
!!(screen->get_param(screen, PIPE_CAP_TEXTURE_BORDER_COLOR_QUIRK) &
(PIPE_QUIRK_TEXTURE_BORDER_COLOR_SWIZZLE_NV50 |
PIPE_QUIRK_TEXTURE_BORDER_COLOR_SWIZZLE_R600));
st->has_time_elapsed =
screen->get_param(screen, PIPE_CAP_QUERY_TIME_ELAPSED);
/* GL limits and extensions */
st_init_limits(st->pipe->screen, &ctx->Const, &ctx->Extensions);
st_init_extensions(st->pipe->screen, &ctx->Const,
&ctx->Extensions, &st->options, ctx->Mesa_DXTn);
/* Enable shader-based fallbacks for ARB_color_buffer_float if needed. */
if (screen->get_param(screen, PIPE_CAP_VERTEX_COLOR_UNCLAMPED)) {
if (!screen->get_param(screen, PIPE_CAP_VERTEX_COLOR_CLAMPED)) {
st->clamp_vert_color_in_shader = GL_TRUE;
}
if (!screen->get_param(screen, PIPE_CAP_FRAGMENT_COLOR_CLAMPED)) {
st->clamp_frag_color_in_shader = GL_TRUE;
}
/* For drivers which cannot do color clamping, it's better to just
* disable ARB_color_buffer_float in the core profile, because
* the clamping is deprecated there anyway. */
if (ctx->API == API_OPENGL_CORE &&
(st->clamp_frag_color_in_shader || st->clamp_vert_color_in_shader)) {
st->clamp_vert_color_in_shader = GL_FALSE;
st->clamp_frag_color_in_shader = GL_FALSE;
ctx->Extensions.ARB_color_buffer_float = GL_FALSE;
}
}
/* called after _mesa_create_context/_mesa_init_point, fix default user
* settable max point size up
*/
st->ctx->Point.MaxSize = MAX2(ctx->Const.MaxPointSize,
ctx->Const.MaxPointSizeAA);
_mesa_compute_version(ctx);
if (ctx->Version == 0) {
/* This can happen when a core profile was requested, but the driver
* does not support some features of GL 3.1 or later.
*/
st_destroy_context_priv(st);
return NULL;
}
_mesa_initialize_dispatch_tables(ctx);
_mesa_initialize_vbo_vtxfmt(ctx);
return st;
}
NS_IMETHODIMP
HTMLFieldSetElement::GetElements(nsIDOMHTMLCollection** aElements)
{
NS_ADDREF(*aElements = Elements());
return NS_OK;
}
/**
* Called via glUniform*() functions.
*/
extern "C" void
_mesa_uniform(struct gl_context *ctx, struct gl_shader_program *shProg,
GLint location, GLsizei count,
const GLvoid *values, GLenum type)
{
unsigned loc, offset;
unsigned components;
unsigned src_components;
enum glsl_base_type basicType;
struct gl_uniform_storage *uni;
if (!validate_uniform_parameters(ctx, shProg, location, count,
&loc, &offset, "glUniform", false))
return;
uni = &shProg->UniformStorage[loc];
/* Verify that the types are compatible.
*/
switch (type) {
case GL_FLOAT:
basicType = GLSL_TYPE_FLOAT;
src_components = 1;
break;
case GL_FLOAT_VEC2:
basicType = GLSL_TYPE_FLOAT;
src_components = 2;
break;
case GL_FLOAT_VEC3:
basicType = GLSL_TYPE_FLOAT;
src_components = 3;
break;
case GL_FLOAT_VEC4:
basicType = GLSL_TYPE_FLOAT;
src_components = 4;
break;
case GL_UNSIGNED_INT:
basicType = GLSL_TYPE_UINT;
src_components = 1;
break;
case GL_UNSIGNED_INT_VEC2:
basicType = GLSL_TYPE_UINT;
src_components = 2;
break;
case GL_UNSIGNED_INT_VEC3:
basicType = GLSL_TYPE_UINT;
src_components = 3;
break;
case GL_UNSIGNED_INT_VEC4:
basicType = GLSL_TYPE_UINT;
src_components = 4;
break;
case GL_INT:
basicType = GLSL_TYPE_INT;
src_components = 1;
break;
case GL_INT_VEC2:
basicType = GLSL_TYPE_INT;
src_components = 2;
break;
case GL_INT_VEC3:
basicType = GLSL_TYPE_INT;
src_components = 3;
break;
case GL_INT_VEC4:
basicType = GLSL_TYPE_INT;
src_components = 4;
break;
case GL_BOOL:
case GL_BOOL_VEC2:
case GL_BOOL_VEC3:
case GL_BOOL_VEC4:
case GL_FLOAT_MAT2:
case GL_FLOAT_MAT2x3:
case GL_FLOAT_MAT2x4:
case GL_FLOAT_MAT3x2:
case GL_FLOAT_MAT3:
case GL_FLOAT_MAT3x4:
case GL_FLOAT_MAT4x2:
case GL_FLOAT_MAT4x3:
case GL_FLOAT_MAT4:
default:
_mesa_problem(NULL, "Invalid type in %s", __func__);
return;
}
if (uni->type->is_sampler()) {
components = 1;
} else {
components = uni->type->vector_elements;
}
bool match;
switch (uni->type->base_type) {
case GLSL_TYPE_BOOL:
match = true;
break;
case GLSL_TYPE_SAMPLER:
match = (basicType == GLSL_TYPE_INT);
break;
default:
match = (basicType == uni->type->base_type);
break;
}
if (uni->type->is_matrix() || components != src_components || !match) {
_mesa_error(ctx, GL_INVALID_OPERATION, "glUniform(type mismatch)");
return;
}
if (ctx->Shader.Flags & GLSL_UNIFORMS) {
log_uniform(values, basicType, components, 1, count,
false, shProg, location, uni);
}
/* Page 100 (page 116 of the PDF) of the OpenGL 3.0 spec says:
*
* "Setting a sampler's value to i selects texture image unit number
* i. The values of i range from zero to the implementation- dependent
* maximum supported number of texture image units."
*
* In addition, table 2.3, "Summary of GL errors," on page 17 (page 33 of
* the PDF) says:
*
* "Error Description Offending command
* ignored?
* ...
* INVALID_VALUE Numeric argument out of range Yes"
*
* Based on that, when an invalid sampler is specified, we generate a
* GL_INVALID_VALUE error and ignore the command.
*/
if (uni->type->is_sampler()) {
int i;
for (i = 0; i < count; i++) {
const unsigned texUnit = ((unsigned *) values)[i];
/* check that the sampler (tex unit index) is legal */
if (texUnit >= ctx->Const.MaxCombinedTextureImageUnits) {
_mesa_error(ctx, GL_INVALID_VALUE,
"glUniform1i(invalid sampler/tex unit index for "
"uniform %d)",
location);
return;
}
}
}
/* Page 82 (page 96 of the PDF) of the OpenGL 2.1 spec says:
*
* "When loading N elements starting at an arbitrary position k in a
* uniform declared as an array, elements k through k + N - 1 in the
* array will be replaced with the new values. Values for any array
* element that exceeds the highest array element index used, as
* reported by GetActiveUniform, will be ignored by the GL."
*
* Clamp 'count' to a valid value. Note that for non-arrays a count > 1
* will have already generated an error.
*/
if (uni->array_elements != 0) {
count = MIN2(count, (int) (uni->array_elements - offset));
}
FLUSH_VERTICES(ctx, _NEW_PROGRAM_CONSTANTS);
/* Store the data in the "actual type" backing storage for the uniform.
*/
if (!uni->type->is_boolean()) {
memcpy(&uni->storage[components * offset], values,
sizeof(uni->storage[0]) * components * count);
} else {
const union gl_constant_value *src =
(const union gl_constant_value *) values;
union gl_constant_value *dst = &uni->storage[components * offset];
const unsigned elems = components * count;
unsigned i;
for (i = 0; i < elems; i++) {
if (basicType == GLSL_TYPE_FLOAT) {
dst[i].i = src[i].f != 0.0f ? 1 : 0;
} else {
dst[i].i = src[i].i != 0 ? 1 : 0;
}
}
}
uni->initialized = true;
_mesa_propagate_uniforms_to_driver_storage(uni, offset, count);
/* If the uniform is a sampler, do the extra magic necessary to propagate
* the changes through.
*/
if (uni->type->is_sampler()) {
int i;
bool flushed = false;
for (i = 0; i < MESA_SHADER_TYPES; i++) {
struct gl_shader *const sh = shProg->_LinkedShaders[i];
int j;
/* If the shader stage doesn't use the sampler uniform, skip this.
*/
if (sh == NULL || !uni->sampler[i].active)
continue;
for (j = 0; j < count; j++) {
sh->SamplerUnits[uni->sampler[i].index + offset + j] =
((unsigned *) values)[j];
}
struct gl_program *const prog = sh->Program;
assert(sizeof(prog->SamplerUnits) == sizeof(sh->SamplerUnits));
/* Determine if any of the samplers used by this shader stage have
* been modified.
*/
bool changed = false;
for (unsigned j = 0; j < Elements(prog->SamplerUnits); j++) {
if ((sh->active_samplers & (1U << j)) != 0
&& (prog->SamplerUnits[j] != sh->SamplerUnits[j])) {
changed = true;
break;
}
}
if (changed) {
if (!flushed) {
FLUSH_VERTICES(ctx, _NEW_TEXTURE | _NEW_PROGRAM);
flushed = true;
}
memcpy(prog->SamplerUnits,
sh->SamplerUnits,
sizeof(sh->SamplerUnits));
_mesa_update_shader_textures_used(shProg, prog);
if (ctx->Driver.SamplerUniformChange)
ctx->Driver.SamplerUniformChange(ctx, prog->Target, prog);
}
}
}
}
static boolean parse_declaration( struct translate_ctx *ctx )
{
struct tgsi_full_declaration decl;
uint file;
struct parsed_dcl_bracket brackets[2];
int num_brackets;
uint writemask;
const char *cur;
uint advance;
boolean is_vs_input;
boolean is_imm_array;
assert(Elements(semantic_names) == TGSI_SEMANTIC_COUNT);
assert(Elements(interpolate_names) == TGSI_INTERPOLATE_COUNT);
if (!eat_white( &ctx->cur )) {
report_error( ctx, "Syntax error" );
return FALSE;
}
if (!parse_register_dcl( ctx, &file, brackets, &num_brackets))
return FALSE;
if (!parse_opt_writemask( ctx, &writemask ))
return FALSE;
decl = tgsi_default_full_declaration();
decl.Declaration.File = file;
decl.Declaration.UsageMask = writemask;
if (num_brackets == 1) {
decl.Range.First = brackets[0].first;
decl.Range.Last = brackets[0].last;
} else {
decl.Range.First = brackets[1].first;
decl.Range.Last = brackets[1].last;
decl.Declaration.Dimension = 1;
decl.Dim.Index2D = brackets[0].first;
}
is_vs_input = (file == TGSI_FILE_INPUT &&
ctx->processor == TGSI_PROCESSOR_VERTEX);
is_imm_array = (file == TGSI_FILE_IMMEDIATE_ARRAY);
cur = ctx->cur;
eat_opt_white( &cur );
if (*cur == ',' && !is_vs_input) {
uint i, j;
cur++;
eat_opt_white( &cur );
if (file == TGSI_FILE_RESOURCE) {
for (i = 0; i < TGSI_TEXTURE_COUNT; i++) {
if (str_match_no_case(&cur, texture_names[i])) {
if (!is_digit_alpha_underscore(cur)) {
decl.Resource.Resource = i;
break;
}
}
}
if (i == TGSI_TEXTURE_COUNT) {
report_error(ctx, "Expected texture target");
return FALSE;
}
eat_opt_white( &cur );
if (*cur != ',') {
report_error( ctx, "Expected `,'" );
return FALSE;
}
++cur;
eat_opt_white( &cur );
for (j = 0; j < 4; ++j) {
for (i = 0; i < PIPE_TYPE_COUNT; ++i) {
if (str_match_no_case(&cur, type_names[i])) {
if (!is_digit_alpha_underscore(cur)) {
switch (j) {
case 0:
decl.Resource.ReturnTypeX = i;
break;
case 1:
decl.Resource.ReturnTypeY = i;
break;
case 2:
decl.Resource.ReturnTypeZ = i;
break;
case 3:
decl.Resource.ReturnTypeW = i;
break;
default:
assert(0);
}
break;
}
}
}
if (i == PIPE_TYPE_COUNT) {
if (j == 0 || j > 2) {
report_error(ctx, "Expected type name");
return FALSE;
}
break;
} else {
const char *cur2 = cur;
eat_opt_white( &cur2 );
if (*cur2 == ',') {
cur2++;
eat_opt_white( &cur2 );
cur = cur2;
continue;
} else
break;
}
}
if (j < 4) {
decl.Resource.ReturnTypeY =
decl.Resource.ReturnTypeZ =
decl.Resource.ReturnTypeW =
decl.Resource.ReturnTypeX;
}
ctx->cur = cur;
} else {
for (i = 0; i < TGSI_SEMANTIC_COUNT; i++) {
if (str_match_no_case( &cur, semantic_names[i] )) {
const char *cur2 = cur;
uint index;
if (is_digit_alpha_underscore( cur ))
continue;
eat_opt_white( &cur2 );
if (*cur2 == '[') {
cur2++;
eat_opt_white( &cur2 );
if (!parse_uint( &cur2, &index )) {
report_error( ctx, "Expected literal integer" );
return FALSE;
}
eat_opt_white( &cur2 );
if (*cur2 != ']') {
report_error( ctx, "Expected `]'" );
return FALSE;
}
cur2++;
decl.Semantic.Index = index;
cur = cur2;
}
decl.Declaration.Semantic = 1;
decl.Semantic.Name = i;
ctx->cur = cur;
break;
}
}
}
} else if (is_imm_array) {
unsigned i;
float *vals_itr;
/* we have our immediate data */
if (*cur != '{') {
report_error( ctx, "Immediate array without data" );
return FALSE;
}
++cur;
ctx->cur = cur;
decl.ImmediateData.u =
MALLOC(sizeof(union tgsi_immediate_data) * 4 *
(decl.Range.Last + 1));
vals_itr = (float*)decl.ImmediateData.u;
for (i = 0; i <= decl.Range.Last; ++i) {
if (!parse_immediate_data(ctx, vals_itr)) {
FREE(decl.ImmediateData.u);
return FALSE;
}
vals_itr += 4;
eat_opt_white( &ctx->cur );
if (*ctx->cur != ',') {
if (i != decl.Range.Last) {
report_error( ctx, "Not enough data in immediate array!" );
FREE(decl.ImmediateData.u);
return FALSE;
}
} else
++ctx->cur;
}
eat_opt_white( &ctx->cur );
if (*ctx->cur != '}') {
FREE(decl.ImmediateData.u);
report_error( ctx, "Immediate array data missing closing '}'" );
return FALSE;
}
++ctx->cur;
}
cur = ctx->cur;
eat_opt_white( &cur );
if (*cur == ',' && !is_vs_input) {
uint i;
cur++;
eat_opt_white( &cur );
for (i = 0; i < TGSI_INTERPOLATE_COUNT; i++) {
if (str_match_no_case( &cur, interpolate_names[i] )) {
if (is_digit_alpha_underscore( cur ))
continue;
decl.Declaration.Interpolate = i;
ctx->cur = cur;
break;
}
}
if (i == TGSI_INTERPOLATE_COUNT) {
report_error( ctx, "Expected semantic or interpolate attribute" );
return FALSE;
}
}
advance = tgsi_build_full_declaration(
&decl,
ctx->tokens_cur,
ctx->header,
(uint) (ctx->tokens_end - ctx->tokens_cur) );
if (is_imm_array)
FREE(decl.ImmediateData.u);
if (advance == 0)
return FALSE;
ctx->tokens_cur += advance;
return TRUE;
}
static int Integrate(creal epsrel, creal epsabs,
cint flags, number mineval, cnumber maxeval, ccount key,
real *integral, real *error, real *prob)
{
TYPEDEFREGION;
count dim, comp;
int fail = 1;
Rule rule;
Totals totals[NCOMP];
Region *anchor = NULL, *region = NULL;
if( VERBOSE > 1 ) {
char s[256];
sprintf(s, "Cuhre input parameters:\n"
" ndim " COUNT "\n ncomp " COUNT "\n"
" epsrel " REAL "\n epsabs " REAL "\n"
" flags %d\n mineval " NUMBER "\n maxeval " NUMBER "\n"
" key " COUNT,
ndim_, ncomp_,
epsrel, epsabs,
flags, mineval, maxeval,
key);
Print(s);
}
#ifdef MLVERSION
if( setjmp(abort_) ) goto abort;
#endif
if( key == 13 && ndim_ == 2 ) Rule13Alloc(&rule);
else if( key == 11 && ndim_ == 3 ) Rule11Alloc(&rule);
else if( key == 9 ) Rule9Alloc(&rule);
else if( key == 7 ) Rule7Alloc(&rule);
else {
if( ndim_ == 2 ) Rule13Alloc(&rule);
else if( ndim_ == 3 ) Rule11Alloc(&rule);
else Rule9Alloc(&rule);
}
Alloc(rule.x, rule.n*(ndim_ + ncomp_));
rule.f = rule.x + rule.n*ndim_;
mineval = IMax(mineval, rule.n + 1);
Alloc(anchor, 1);
anchor->next = NULL;
anchor->div = 0;
for( dim = 0; dim < ndim_; ++dim ) {
Bounds *b = &anchor->bounds[dim];
b->lower = 0;
b->upper = 1;
}
Sample(&rule, anchor, flags);
for( comp = 0; comp < ncomp_; ++comp ) {
Totals *tot = &totals[comp];
Result *r = &anchor->result[comp];
tot->avg = tot->lastavg = tot->guess = r->avg;
tot->err = tot->lasterr = r->err;
tot->weightsum = 1/Max(Sq(r->err), NOTZERO);
tot->avgsum = tot->weightsum*r->avg;
tot->chisq = tot->chisqsum = tot->chisum = 0;
}
for( nregions_ = 1; ; ++nregions_ ) {
count maxcomp, bisectdim;
real maxratio, maxerr;
Region *regionL, *regionR, *reg, **parent, **par;
Bounds *bL, *bR;
if( VERBOSE ) {
char s[128 + 128*NCOMP], *p = s;
p += sprintf(p, "\n"
"Iteration " COUNT ": " NUMBER " integrand evaluations so far",
nregions_, neval_);
for( comp = 0; comp < ncomp_; ++comp ) {
cTotals *tot = &totals[comp];
p += sprintf(p, "\n[" COUNT "] "
REAL " +- " REAL " \tchisq " REAL " (" COUNT " df)",
comp + 1, tot->avg, tot->err, tot->chisq, nregions_ - 1);
}
Print(s);
}
maxratio = -INFTY;
maxcomp = 0;
for( comp = 0; comp < ncomp_; ++comp ) {
creal ratio = totals[comp].err/MaxErr(totals[comp].avg);
if( ratio > maxratio ) {
maxratio = ratio;
maxcomp = comp;
}
}
if( maxratio <= 1 && neval_ >= mineval ) {
fail = 0;
break;
}
if( neval_ >= maxeval ) break;
maxerr = -INFTY;
parent = &anchor;
region = anchor;
for( par = &anchor; (reg = *par); par = ®->next ) {
creal err = reg->result[maxcomp].err;
if( err > maxerr ) {
maxerr = err;
parent = par;
region = reg;
}
}
Alloc(regionL, 1);
Alloc(regionR, 1);
*parent = regionL;
regionL->next = regionR;
regionR->next = region->next;
regionL->div = regionR->div = region->div + 1;
VecCopy(regionL->bounds, region->bounds);
VecCopy(regionR->bounds, region->bounds);
bisectdim = region->result[maxcomp].bisectdim;
bL = ®ionL->bounds[bisectdim];
bR = ®ionR->bounds[bisectdim];
bL->upper = bR->lower = .5*(bL->upper + bL->lower);
Sample(&rule, regionL, flags);
Sample(&rule, regionR, flags);
for( comp = 0; comp < ncomp_; ++comp ) {
cResult *r = ®ion->result[comp];
Result *rL = ®ionL->result[comp];
Result *rR = ®ionR->result[comp];
Totals *tot = &totals[comp];
real diff, err, w, avg, sigsq;
tot->lastavg += diff = rL->avg + rR->avg - r->avg;
diff = fabs(.25*diff);
err = rL->err + rR->err;
if( err > 0 ) {
creal c = 1 + 2*diff/err;
rL->err *= c;
rR->err *= c;
}
rL->err += diff;
rR->err += diff;
tot->lasterr += rL->err + rR->err - r->err;
tot->weightsum += w = 1/Max(Sq(tot->lasterr), NOTZERO);
sigsq = 1/tot->weightsum;
tot->avgsum += w*tot->lastavg;
avg = sigsq*tot->avgsum;
tot->chisum += w *= tot->lastavg - tot->guess;
tot->chisqsum += w*tot->lastavg;
tot->chisq = tot->chisqsum - avg*tot->chisum;
if( LAST ) {
tot->avg = tot->lastavg;
tot->err = tot->lasterr;
}
else {
tot->avg = avg;
tot->err = sqrt(sigsq);
}
}
free(region);
region = NULL;
}
for( comp = 0; comp < ncomp_; ++comp ) {
cTotals *tot = &totals[comp];
integral[comp] = tot->avg;
error[comp] = tot->err;
prob[comp] = ChiSquare(tot->chisq, nregions_ - 1);
}
#ifdef MLVERSION
if( REGIONS ) {
MLPutFunction(stdlink, "List", 2);
MLPutFunction(stdlink, "List", nregions_);
for( region = anchor; region; region = region->next ) {
real lower[NDIM], upper[NDIM];
for( dim = 0; dim < ndim_; ++dim ) {
cBounds *b = ®ion->bounds[dim];
lower[dim] = b->lower;
upper[dim] = b->upper;
}
MLPutFunction(stdlink, "Cuba`Cuhre`region", 3);
MLPutRealList(stdlink, lower, ndim_);
MLPutRealList(stdlink, upper, ndim_);
MLPutFunction(stdlink, "List", ncomp_);
for( comp = 0; comp < ncomp_; ++comp ) {
cResult *r = ®ion->result[comp];
real res[] = {r->avg, r->err};
MLPutRealList(stdlink, res, Elements(res));
}
}
}
#endif
#ifdef MLVERSION
abort:
#endif
if( region ) free(region);
while( (region = anchor) ) {
anchor = anchor->next;
free(region);
}
free(rule.x);
RuleFree(&rule);
return fail;
}
/**
* Sample the texture/mipmap using given image filter and mip filter.
* data0_ptr and data1_ptr point to the two mipmap levels to sample
* from. width0/1_vec, height0/1_vec, depth0/1_vec indicate their sizes.
* If we're using nearest miplevel sampling the '1' values will be null/unused.
*/
static void
lp_build_sample_mipmap(struct lp_build_sample_context *bld,
unsigned img_filter,
unsigned mip_filter,
LLVMValueRef s,
LLVMValueRef t,
LLVMValueRef r,
LLVMValueRef ilevel0,
LLVMValueRef ilevel1,
LLVMValueRef lod_fpart,
LLVMValueRef colors_lo_var,
LLVMValueRef colors_hi_var)
{
LLVMBuilderRef builder = bld->gallivm->builder;
LLVMValueRef size0;
LLVMValueRef size1;
LLVMValueRef row_stride0_vec;
LLVMValueRef row_stride1_vec;
LLVMValueRef img_stride0_vec;
LLVMValueRef img_stride1_vec;
LLVMValueRef data_ptr0;
LLVMValueRef data_ptr1;
LLVMValueRef colors0_lo, colors0_hi;
LLVMValueRef colors1_lo, colors1_hi;
/* sample the first mipmap level */
lp_build_mipmap_level_sizes(bld, ilevel0,
&size0,
&row_stride0_vec, &img_stride0_vec);
data_ptr0 = lp_build_get_mipmap_level(bld, ilevel0);
if (img_filter == PIPE_TEX_FILTER_NEAREST) {
lp_build_sample_image_nearest(bld,
size0,
row_stride0_vec, img_stride0_vec,
data_ptr0, s, t, r,
&colors0_lo, &colors0_hi);
}
else {
assert(img_filter == PIPE_TEX_FILTER_LINEAR);
lp_build_sample_image_linear(bld,
size0,
row_stride0_vec, img_stride0_vec,
data_ptr0, s, t, r,
&colors0_lo, &colors0_hi);
}
/* Store the first level's colors in the output variables */
LLVMBuildStore(builder, colors0_lo, colors_lo_var);
LLVMBuildStore(builder, colors0_hi, colors_hi_var);
if (mip_filter == PIPE_TEX_MIPFILTER_LINEAR) {
LLVMValueRef h16_scale = lp_build_const_float(bld->gallivm, 256.0);
LLVMTypeRef i32_type = LLVMIntTypeInContext(bld->gallivm->context, 32);
struct lp_build_if_state if_ctx;
LLVMValueRef need_lerp;
lod_fpart = LLVMBuildFMul(builder, lod_fpart, h16_scale, "");
lod_fpart = LLVMBuildFPToSI(builder, lod_fpart, i32_type, "lod_fpart.fixed16");
/* need_lerp = lod_fpart > 0 */
need_lerp = LLVMBuildICmp(builder, LLVMIntSGT,
lod_fpart, LLVMConstNull(i32_type),
"need_lerp");
lp_build_if(&if_ctx, bld->gallivm, need_lerp);
{
struct lp_build_context h16_bld;
lp_build_context_init(&h16_bld, bld->gallivm, lp_type_ufixed(16));
/* sample the second mipmap level */
lp_build_mipmap_level_sizes(bld, ilevel1,
&size1,
&row_stride1_vec, &img_stride1_vec);
data_ptr1 = lp_build_get_mipmap_level(bld, ilevel1);
if (img_filter == PIPE_TEX_FILTER_NEAREST) {
lp_build_sample_image_nearest(bld,
size1,
row_stride1_vec, img_stride1_vec,
data_ptr1, s, t, r,
&colors1_lo, &colors1_hi);
}
else {
lp_build_sample_image_linear(bld,
size1,
row_stride1_vec, img_stride1_vec,
data_ptr1, s, t, r,
&colors1_lo, &colors1_hi);
}
/* interpolate samples from the two mipmap levels */
lod_fpart = LLVMBuildTrunc(builder, lod_fpart, h16_bld.elem_type, "");
lod_fpart = lp_build_broadcast_scalar(&h16_bld, lod_fpart);
#if HAVE_LLVM == 0x208
/* This is a work-around for a bug in LLVM 2.8.
* Evidently, something goes wrong in the construction of the
* lod_fpart short[8] vector. Adding this no-effect shuffle seems
* to force the vector to be properly constructed.
* Tested with mesa-demos/src/tests/mipmap_limits.c (press t, f).
*/
{
LLVMValueRef shuffles[8], shuffle;
int i;
assert(h16_bld.type.length <= Elements(shuffles));
for (i = 0; i < h16_bld.type.length; i++)
shuffles[i] = lp_build_const_int32(bld->gallivm, 2 * (i & 1));
shuffle = LLVMConstVector(shuffles, h16_bld.type.length);
lod_fpart = LLVMBuildShuffleVector(builder,
lod_fpart, lod_fpart,
shuffle, "");
}
#endif
colors0_lo = lp_build_lerp(&h16_bld, lod_fpart,
colors0_lo, colors1_lo);
colors0_hi = lp_build_lerp(&h16_bld, lod_fpart,
colors0_hi, colors1_hi);
LLVMBuildStore(builder, colors0_lo, colors_lo_var);
LLVMBuildStore(builder, colors0_hi, colors_hi_var);
}
lp_build_endif(&if_ctx);
}
}
JSObject*
CryptoBuffer::ToUint8Array(JSContext* aCx) const
{
return Uint8Array::Create(aCx, Length(), Elements());
}
JSObject*
CryptoBuffer::ToArrayBuffer(JSContext* aCx) const
{
return ArrayBuffer::Create(aCx, Length(), Elements());
}
/**
* Handle state changes.
* Called just prior to drawing anything (pipe::draw_arrays(), etc).
*
* Hopefully this will remain quite simple, otherwise need to pull in
* something like the state tracker mechanism.
*/
void llvmpipe_update_derived( struct llvmpipe_context *llvmpipe )
{
struct llvmpipe_screen *lp_screen = llvmpipe_screen(llvmpipe->pipe.screen);
/* Check for updated textures.
*/
if (llvmpipe->tex_timestamp != lp_screen->timestamp) {
llvmpipe->tex_timestamp = lp_screen->timestamp;
llvmpipe->dirty |= LP_NEW_SAMPLER_VIEW;
}
if (llvmpipe->dirty & (LP_NEW_RASTERIZER |
LP_NEW_FS |
LP_NEW_VS))
compute_vertex_info( llvmpipe );
if (llvmpipe->dirty & (LP_NEW_FS |
LP_NEW_FRAMEBUFFER |
LP_NEW_BLEND |
LP_NEW_SCISSOR |
LP_NEW_DEPTH_STENCIL_ALPHA |
LP_NEW_RASTERIZER |
LP_NEW_SAMPLER |
LP_NEW_SAMPLER_VIEW |
LP_NEW_OCCLUSION_QUERY))
llvmpipe_update_fs( llvmpipe );
if (llvmpipe->dirty & (LP_NEW_FS |
LP_NEW_RASTERIZER))
llvmpipe_update_setup( llvmpipe );
if (llvmpipe->dirty & LP_NEW_BLEND_COLOR)
lp_setup_set_blend_color(llvmpipe->setup,
&llvmpipe->blend_color);
if (llvmpipe->dirty & LP_NEW_SCISSOR)
lp_setup_set_scissors(llvmpipe->setup, llvmpipe->scissors);
if (llvmpipe->dirty & LP_NEW_DEPTH_STENCIL_ALPHA) {
lp_setup_set_alpha_ref_value(llvmpipe->setup,
llvmpipe->depth_stencil->alpha.ref_value);
lp_setup_set_stencil_ref_values(llvmpipe->setup,
llvmpipe->stencil_ref.ref_value);
}
if (llvmpipe->dirty & LP_NEW_CONSTANTS)
lp_setup_set_fs_constants(llvmpipe->setup,
Elements(llvmpipe->constants[PIPE_SHADER_FRAGMENT]),
llvmpipe->constants[PIPE_SHADER_FRAGMENT]);
if (llvmpipe->dirty & (LP_NEW_SAMPLER_VIEW))
lp_setup_set_fragment_sampler_views(llvmpipe->setup,
llvmpipe->num_sampler_views[PIPE_SHADER_FRAGMENT],
llvmpipe->sampler_views[PIPE_SHADER_FRAGMENT]);
if (llvmpipe->dirty & (LP_NEW_SAMPLER))
lp_setup_set_fragment_sampler_state(llvmpipe->setup,
llvmpipe->num_samplers[PIPE_SHADER_FRAGMENT],
llvmpipe->samplers[PIPE_SHADER_FRAGMENT]);
llvmpipe->dirty = 0;
}
/**
* Translate a geometry program to create a new variant.
*/
static struct st_gp_variant *
st_translate_geometry_program(struct st_context *st,
struct st_geometry_program *stgp,
const struct st_gp_variant_key *key)
{
GLuint inputMapping[VARYING_SLOT_MAX];
GLuint outputMapping[VARYING_SLOT_MAX];
struct pipe_context *pipe = st->pipe;
GLuint attr;
GLbitfield64 inputsRead;
GLuint vslot = 0;
uint gs_num_inputs = 0;
uint gs_builtin_inputs = 0;
uint gs_array_offset = 0;
ubyte gs_output_semantic_name[PIPE_MAX_SHADER_OUTPUTS];
ubyte gs_output_semantic_index[PIPE_MAX_SHADER_OUTPUTS];
uint gs_num_outputs = 0;
GLint i;
GLuint maxSlot = 0;
struct ureg_program *ureg;
struct st_gp_variant *gpv;
gpv = CALLOC_STRUCT(st_gp_variant);
if (!gpv)
return NULL;
if (!stgp->glsl_to_tgsi) {
_mesa_remove_output_reads(&stgp->Base.Base, PROGRAM_OUTPUT);
}
ureg = ureg_create( TGSI_PROCESSOR_GEOMETRY );
if (ureg == NULL) {
free(gpv);
return NULL;
}
/* which vertex output goes to the first geometry input */
vslot = 0;
memset(inputMapping, 0, sizeof(inputMapping));
memset(outputMapping, 0, sizeof(outputMapping));
/*
* Convert Mesa program inputs to TGSI input register semantics.
*/
inputsRead = stgp->Base.Base.InputsRead;
for (attr = 0; attr < VARYING_SLOT_MAX; attr++) {
if ((inputsRead & BITFIELD64_BIT(attr)) != 0) {
const GLuint slot = gs_num_inputs;
gs_num_inputs++;
inputMapping[attr] = slot;
stgp->input_map[slot + gs_array_offset] = vslot - gs_builtin_inputs;
stgp->input_to_index[attr] = vslot;
stgp->index_to_input[vslot] = attr;
++vslot;
if (attr != VARYING_SLOT_PRIMITIVE_ID) {
gs_array_offset += 2;
} else
++gs_builtin_inputs;
#if 0
debug_printf("input map at %d = %d\n",
slot + gs_array_offset, stgp->input_map[slot + gs_array_offset]);
#endif
switch (attr) {
case VARYING_SLOT_PRIMITIVE_ID:
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_PRIMID;
stgp->input_semantic_index[slot] = 0;
break;
case VARYING_SLOT_POS:
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_POSITION;
stgp->input_semantic_index[slot] = 0;
break;
case VARYING_SLOT_COL0:
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_COLOR;
stgp->input_semantic_index[slot] = 0;
break;
case VARYING_SLOT_COL1:
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_COLOR;
stgp->input_semantic_index[slot] = 1;
break;
case VARYING_SLOT_FOGC:
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_FOG;
stgp->input_semantic_index[slot] = 0;
break;
case VARYING_SLOT_CLIP_VERTEX:
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_CLIPVERTEX;
stgp->input_semantic_index[slot] = 0;
break;
case VARYING_SLOT_CLIP_DIST0:
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_CLIPDIST;
stgp->input_semantic_index[slot] = 0;
break;
case VARYING_SLOT_CLIP_DIST1:
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_CLIPDIST;
stgp->input_semantic_index[slot] = 1;
break;
case VARYING_SLOT_PSIZ:
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_PSIZE;
stgp->input_semantic_index[slot] = 0;
break;
case VARYING_SLOT_TEX0:
case VARYING_SLOT_TEX1:
case VARYING_SLOT_TEX2:
case VARYING_SLOT_TEX3:
case VARYING_SLOT_TEX4:
case VARYING_SLOT_TEX5:
case VARYING_SLOT_TEX6:
case VARYING_SLOT_TEX7:
stgp->input_semantic_name[slot] = st->needs_texcoord_semantic ?
TGSI_SEMANTIC_TEXCOORD : TGSI_SEMANTIC_GENERIC;
stgp->input_semantic_index[slot] = (attr - VARYING_SLOT_TEX0);
break;
case VARYING_SLOT_VAR0:
default:
assert(attr >= VARYING_SLOT_VAR0 && attr < VARYING_SLOT_MAX);
stgp->input_semantic_name[slot] = TGSI_SEMANTIC_GENERIC;
stgp->input_semantic_index[slot] = st->needs_texcoord_semantic ?
(attr - VARYING_SLOT_VAR0) : (attr - VARYING_SLOT_TEX0);
break;
}
}
}
/* initialize output semantics to defaults */
for (i = 0; i < PIPE_MAX_SHADER_OUTPUTS; i++) {
gs_output_semantic_name[i] = TGSI_SEMANTIC_GENERIC;
gs_output_semantic_index[i] = 0;
}
/*
* Determine number of outputs, the (default) output register
* mapping and the semantic information for each output.
*/
for (attr = 0; attr < VARYING_SLOT_MAX; attr++) {
if (stgp->Base.Base.OutputsWritten & BITFIELD64_BIT(attr)) {
GLuint slot;
slot = gs_num_outputs;
gs_num_outputs++;
outputMapping[attr] = slot;
switch (attr) {
case VARYING_SLOT_POS:
assert(slot == 0);
gs_output_semantic_name[slot] = TGSI_SEMANTIC_POSITION;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_COL0:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_COLOR;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_COL1:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_COLOR;
gs_output_semantic_index[slot] = 1;
break;
case VARYING_SLOT_BFC0:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_BCOLOR;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_BFC1:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_BCOLOR;
gs_output_semantic_index[slot] = 1;
break;
case VARYING_SLOT_FOGC:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_FOG;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_PSIZ:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_PSIZE;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_CLIP_VERTEX:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_CLIPVERTEX;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_CLIP_DIST0:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_CLIPDIST;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_CLIP_DIST1:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_CLIPDIST;
gs_output_semantic_index[slot] = 1;
break;
case VARYING_SLOT_LAYER:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_LAYER;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_PRIMITIVE_ID:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_PRIMID;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_VIEWPORT:
gs_output_semantic_name[slot] = TGSI_SEMANTIC_VIEWPORT_INDEX;
gs_output_semantic_index[slot] = 0;
break;
case VARYING_SLOT_TEX0:
case VARYING_SLOT_TEX1:
case VARYING_SLOT_TEX2:
case VARYING_SLOT_TEX3:
case VARYING_SLOT_TEX4:
case VARYING_SLOT_TEX5:
case VARYING_SLOT_TEX6:
case VARYING_SLOT_TEX7:
gs_output_semantic_name[slot] = st->needs_texcoord_semantic ?
TGSI_SEMANTIC_TEXCOORD : TGSI_SEMANTIC_GENERIC;
gs_output_semantic_index[slot] = (attr - VARYING_SLOT_TEX0);
break;
case VARYING_SLOT_VAR0:
default:
assert(slot < Elements(gs_output_semantic_name));
assert(attr >= VARYING_SLOT_VAR0);
gs_output_semantic_name[slot] = TGSI_SEMANTIC_GENERIC;
gs_output_semantic_index[slot] = st->needs_texcoord_semantic ?
(attr - VARYING_SLOT_VAR0) : (attr - VARYING_SLOT_TEX0);
break;
}
}
}
/* find max output slot referenced to compute gs_num_outputs */
for (attr = 0; attr < VARYING_SLOT_MAX; attr++) {
if (outputMapping[attr] != ~0 && outputMapping[attr] > maxSlot)
maxSlot = outputMapping[attr];
}
gs_num_outputs = maxSlot + 1;
#if 0 /* debug */
{
GLuint i;
printf("outputMapping? %d\n", outputMapping ? 1 : 0);
if (outputMapping) {
printf("attr -> slot\n");
for (i = 0; i < 16; i++) {
printf(" %2d %3d\n", i, outputMapping[i]);
}
}
printf("slot sem_name sem_index\n");
for (i = 0; i < gs_num_outputs; i++) {
printf(" %2d %d %d\n",
i,
gs_output_semantic_name[i],
gs_output_semantic_index[i]);
}
}
#endif
/* free old shader state, if any */
if (stgp->tgsi.tokens) {
st_free_tokens(stgp->tgsi.tokens);
stgp->tgsi.tokens = NULL;
}
ureg_property_gs_input_prim(ureg, stgp->Base.InputType);
ureg_property_gs_output_prim(ureg, stgp->Base.OutputType);
ureg_property_gs_max_vertices(ureg, stgp->Base.VerticesOut);
ureg_property_gs_invocations(ureg, stgp->Base.Invocations);
if (stgp->glsl_to_tgsi)
st_translate_program(st->ctx,
TGSI_PROCESSOR_GEOMETRY,
ureg,
stgp->glsl_to_tgsi,
&stgp->Base.Base,
/* inputs */
gs_num_inputs,
inputMapping,
stgp->input_semantic_name,
stgp->input_semantic_index,
NULL,
NULL,
/* outputs */
gs_num_outputs,
outputMapping,
gs_output_semantic_name,
gs_output_semantic_index,
FALSE,
FALSE);
else
st_translate_mesa_program(st->ctx,
TGSI_PROCESSOR_GEOMETRY,
ureg,
&stgp->Base.Base,
/* inputs */
gs_num_inputs,
inputMapping,
stgp->input_semantic_name,
stgp->input_semantic_index,
NULL,
/* outputs */
gs_num_outputs,
outputMapping,
gs_output_semantic_name,
gs_output_semantic_index,
FALSE,
FALSE);
stgp->num_inputs = gs_num_inputs;
stgp->tgsi.tokens = ureg_get_tokens( ureg, NULL );
ureg_destroy( ureg );
if (stgp->glsl_to_tgsi) {
st_translate_stream_output_info(stgp->glsl_to_tgsi,
outputMapping,
&stgp->tgsi.stream_output);
}
/* fill in new variant */
gpv->driver_shader = pipe->create_gs_state(pipe, &stgp->tgsi);
gpv->key = *key;
if ((ST_DEBUG & DEBUG_TGSI) && (ST_DEBUG & DEBUG_MESA)) {
_mesa_print_program(&stgp->Base.Base);
debug_printf("\n");
}
if (ST_DEBUG & DEBUG_TGSI) {
tgsi_dump(stgp->tgsi.tokens, 0);
debug_printf("\n");
}
return gpv;
}
#include "piglit-framework.h"
#include "piglit-util.h"
#include "rg-teximage-common.h"
#define WIDTH 256
#define HEIGHT 256
static float rgba_image[4 * WIDTH * HEIGHT];
static const GLenum internal_formats[] = {
GL_COMPRESSED_RG_RGTC2,
GL_COMPRESSED_SIGNED_RG_RGTC2,
GL_RG,
GL_RGBA,
};
GLuint tex[Elements(internal_formats)];
const unsigned num_tex = Elements(tex);
GLboolean pass = GL_TRUE;
int piglit_width = WIDTH * Elements(tex);
int piglit_height = HEIGHT;
int piglit_window_mode = GLUT_RGB | GLUT_DOUBLE;
void
piglit_init(int argc, char **argv)
{
unsigned i;
GLfloat result[4 * WIDTH * HEIGHT];
piglit_require_extension("GL_ARB_texture_compression_rgtc");
/**
* Do additional "completeness" testing of a framebuffer object.
*/
static void
intel_validate_framebuffer(struct gl_context *ctx, struct gl_framebuffer *fb)
{
struct intel_context *intel = intel_context(ctx);
const struct intel_renderbuffer *depthRb =
intel_get_renderbuffer(fb, BUFFER_DEPTH);
const struct intel_renderbuffer *stencilRb =
intel_get_renderbuffer(fb, BUFFER_STENCIL);
struct intel_mipmap_tree *depth_mt = NULL, *stencil_mt = NULL;
int i;
DBG("%s() on fb %p (%s)\n", __FUNCTION__,
fb, (fb == ctx->DrawBuffer ? "drawbuffer" :
(fb == ctx->ReadBuffer ? "readbuffer" : "other buffer")));
if (depthRb)
depth_mt = depthRb->mt;
if (stencilRb) {
stencil_mt = stencilRb->mt;
if (stencil_mt->stencil_mt)
stencil_mt = stencil_mt->stencil_mt;
}
if (depth_mt && stencil_mt) {
if (depth_mt == stencil_mt) {
/* For true packed depth/stencil (not faked on prefers-separate-stencil
* hardware) we need to be sure they're the same level/layer, since
* we'll be emitting a single packet describing the packed setup.
*/
if (depthRb->mt_level != stencilRb->mt_level ||
depthRb->mt_layer != stencilRb->mt_layer) {
DBG("depth image level/layer %d/%d != stencil image %d/%d\n",
depthRb->mt_level,
depthRb->mt_layer,
stencilRb->mt_level,
stencilRb->mt_layer);
fb->_Status = GL_FRAMEBUFFER_UNSUPPORTED_EXT;
}
} else {
if (!intel->has_separate_stencil) {
DBG("separate stencil unsupported\n");
fb->_Status = GL_FRAMEBUFFER_UNSUPPORTED_EXT;
}
if (stencil_mt->format != MESA_FORMAT_S8) {
DBG("separate stencil is %s instead of S8\n",
_mesa_get_format_name(stencil_mt->format));
fb->_Status = GL_FRAMEBUFFER_UNSUPPORTED_EXT;
}
if (intel->gen < 7 && depth_mt->hiz_mt == NULL) {
/* Before Gen7, separate depth and stencil buffers can be used
* only if HiZ is enabled. From the Sandybridge PRM, Volume 2,
* Part 1, Bit 3DSTATE_DEPTH_BUFFER.SeparateStencilBufferEnable:
* [DevSNB]: This field must be set to the same value (enabled
* or disabled) as Hierarchical Depth Buffer Enable.
*/
DBG("separate stencil without HiZ\n");
fb->_Status = GL_FRAMEBUFFER_UNSUPPORTED;
}
}
}
for (i = 0; i < Elements(fb->Attachment); i++) {
struct gl_renderbuffer *rb;
struct intel_renderbuffer *irb;
if (fb->Attachment[i].Type == GL_NONE)
continue;
/* A supported attachment will have a Renderbuffer set either
* from being a Renderbuffer or being a texture that got the
* intel_wrap_texture() treatment.
*/
rb = fb->Attachment[i].Renderbuffer;
if (rb == NULL) {
DBG("attachment without renderbuffer\n");
fb->_Status = GL_FRAMEBUFFER_UNSUPPORTED_EXT;
continue;
}
if (fb->Attachment[i].Type == GL_TEXTURE) {
const struct gl_texture_image *img =
_mesa_get_attachment_teximage_const(&fb->Attachment[i]);
if (img->Border) {
DBG("texture with border\n");
fb->_Status = GL_FRAMEBUFFER_UNSUPPORTED_EXT;
continue;
}
}
irb = intel_renderbuffer(rb);
if (irb == NULL) {
DBG("software rendering renderbuffer\n");
fb->_Status = GL_FRAMEBUFFER_UNSUPPORTED_EXT;
continue;
}
if (!intel->vtbl.render_target_supported(intel, rb)) {
DBG("Unsupported HW texture/renderbuffer format attached: %s\n",
_mesa_get_format_name(intel_rb_format(irb)));
fb->_Status = GL_FRAMEBUFFER_UNSUPPORTED_EXT;
}
}
}
/**
* Emit instructions to move sampling parameters to the message registers.
*/
static int
vs_add_sampler_params(struct toy_compiler *tc, int msg_type, int base_mrf,
struct toy_src coords, int num_coords,
struct toy_src bias_or_lod, struct toy_src ref_or_si,
struct toy_src ddx, struct toy_src ddy, int num_derivs)
{
const unsigned coords_writemask = (1 << num_coords) - 1;
struct toy_dst m[3];
int num_params, i;
assert(num_coords <= 4);
assert(num_derivs <= 3 && num_derivs <= num_coords);
for (i = 0; i < Elements(m); i++)
m[i] = tdst(TOY_FILE_MRF, base_mrf + i, 0);
switch (msg_type) {
case GEN6_MSG_SAMPLER_SAMPLE_L:
tc_MOV(tc, tdst_writemask(m[0], coords_writemask), coords);
tc_MOV(tc, tdst_writemask(m[1], TOY_WRITEMASK_X), bias_or_lod);
num_params = 5;
break;
case GEN6_MSG_SAMPLER_SAMPLE_D:
tc_MOV(tc, tdst_writemask(m[0], coords_writemask), coords);
tc_MOV(tc, tdst_writemask(m[1], TOY_WRITEMASK_XZ),
tsrc_swizzle(ddx, 0, 0, 1, 1));
tc_MOV(tc, tdst_writemask(m[1], TOY_WRITEMASK_YW),
tsrc_swizzle(ddy, 0, 0, 1, 1));
if (num_derivs > 2) {
tc_MOV(tc, tdst_writemask(m[2], TOY_WRITEMASK_X),
tsrc_swizzle1(ddx, 2));
tc_MOV(tc, tdst_writemask(m[2], TOY_WRITEMASK_Y),
tsrc_swizzle1(ddy, 2));
}
num_params = 4 + num_derivs * 2;
break;
case GEN6_MSG_SAMPLER_SAMPLE_L_C:
tc_MOV(tc, tdst_writemask(m[0], coords_writemask), coords);
tc_MOV(tc, tdst_writemask(m[1], TOY_WRITEMASK_X), ref_or_si);
tc_MOV(tc, tdst_writemask(m[1], TOY_WRITEMASK_Y), bias_or_lod);
num_params = 6;
break;
case GEN6_MSG_SAMPLER_LD:
assert(num_coords <= 3);
tc_MOV(tc, tdst_writemask(tdst_d(m[0]), coords_writemask), coords);
tc_MOV(tc, tdst_writemask(tdst_d(m[0]), TOY_WRITEMASK_W), bias_or_lod);
if (tc->dev->gen >= ILO_GEN(7)) {
num_params = 4;
}
else {
tc_MOV(tc, tdst_writemask(tdst_d(m[1]), TOY_WRITEMASK_X), ref_or_si);
num_params = 5;
}
break;
case GEN6_MSG_SAMPLER_RESINFO:
tc_MOV(tc, tdst_writemask(tdst_d(m[0]), TOY_WRITEMASK_X), bias_or_lod);
num_params = 1;
break;
default:
tc_fail(tc, "unknown sampler opcode");
num_params = 0;
break;
}
return (num_params + 3) / 4;
}
static boolean
drm_display_program(struct native_display *ndpy, int crtc_idx,
struct native_surface *nsurf, uint x, uint y,
const struct native_connector **nconns, int num_nconns,
const struct native_mode *nmode)
{
struct drm_display *drmdpy = drm_display(ndpy);
struct drm_surface *drmsurf = drm_surface(nsurf);
const struct drm_mode *drmmode = drm_mode(nmode);
uint32_t connector_ids[32];
uint32_t buffer_id;
drmModeModeInfo mode_tmp, *mode;
int i;
if (num_nconns > Elements(connector_ids)) {
_eglLog(_EGL_WARNING, "too many connectors (%d)", num_nconns);
num_nconns = Elements(connector_ids);
}
if (drmsurf) {
if (!drm_surface_init_framebuffers(&drmsurf->base, FALSE))
return FALSE;
buffer_id = drmsurf->front_fb.buffer_id;
/* the mode argument of drmModeSetCrtc is not constified */
mode_tmp = drmmode->mode;
mode = &mode_tmp;
}
else {
/* disable the CRTC */
buffer_id = 0;
mode = NULL;
num_nconns = 0;
}
for (i = 0; i < num_nconns; i++) {
struct drm_connector *drmconn = drm_connector(nconns[i]);
connector_ids[i] = drmconn->connector->connector_id;
}
if (!drm_display_set_crtc(&drmdpy->base, crtc_idx, buffer_id, x, y,
connector_ids, num_nconns, mode)) {
_eglLog(_EGL_WARNING, "failed to set CRTC %d", crtc_idx);
return FALSE;
}
if (drmdpy->shown_surfaces[crtc_idx])
drmdpy->shown_surfaces[crtc_idx]->is_shown = FALSE;
drmdpy->shown_surfaces[crtc_idx] = drmsurf;
/* remember the settings for buffer swapping */
if (drmsurf) {
uint32_t crtc_id = drmdpy->saved_crtcs[crtc_idx].crtc->crtc_id;
struct drm_crtc *drmcrtc = &drmsurf->current_crtc;
if (drmcrtc->crtc)
drmModeFreeCrtc(drmcrtc->crtc);
drmcrtc->crtc = drmModeGetCrtc(drmdpy->fd, crtc_id);
assert(num_nconns < Elements(drmcrtc->connectors));
memcpy(drmcrtc->connectors, connector_ids,
sizeof(*connector_ids) * num_nconns);
drmcrtc->num_connectors = num_nconns;
drmsurf->is_shown = TRUE;
}
return TRUE;
}
static void transform_inst(struct tgsi_transform_context *ctx,
struct tgsi_full_instruction *inst)
{
struct vs_transform_context *vsctx = (struct vs_transform_context *) ctx;
struct tgsi_full_instruction new_inst;
unsigned i;
if (!vsctx->first_instruction) {
vsctx->first_instruction = TRUE;
/* Insert the generic output for WPOS. */
emit_output(ctx, TGSI_SEMANTIC_GENERIC, vsctx->last_generic + 1,
TGSI_INTERPOLATE_PERSPECTIVE, vsctx->num_outputs);
/* Find a free temp for POSITION. */
for (i = 0; i < Elements(vsctx->temp_used); i++) {
if (!vsctx->temp_used[i]) {
emit_temp(ctx, i);
vsctx->pos_temp = i;
break;
}
}
}
if (inst->Instruction.Opcode == TGSI_OPCODE_END) {
/* MOV OUT[pos_output], TEMP[pos_temp]; */
new_inst = tgsi_default_full_instruction();
new_inst.Instruction.Opcode = TGSI_OPCODE_MOV;
new_inst.Instruction.NumDstRegs = 1;
new_inst.Dst[0].Register.File = TGSI_FILE_OUTPUT;
new_inst.Dst[0].Register.Index = vsctx->pos_output;
new_inst.Dst[0].Register.WriteMask = TGSI_WRITEMASK_XYZW;
new_inst.Instruction.NumSrcRegs = 1;
new_inst.Src[0].Register.File = TGSI_FILE_TEMPORARY;
new_inst.Src[0].Register.Index = vsctx->pos_temp;
ctx->emit_instruction(ctx, &new_inst);
/* MOV OUT[n-1], TEMP[pos_temp]; */
new_inst = tgsi_default_full_instruction();
new_inst.Instruction.Opcode = TGSI_OPCODE_MOV;
new_inst.Instruction.NumDstRegs = 1;
new_inst.Dst[0].Register.File = TGSI_FILE_OUTPUT;
new_inst.Dst[0].Register.Index = vsctx->num_outputs - 1;
new_inst.Dst[0].Register.WriteMask = TGSI_WRITEMASK_XYZW;
new_inst.Instruction.NumSrcRegs = 1;
new_inst.Src[0].Register.File = TGSI_FILE_TEMPORARY;
new_inst.Src[0].Register.Index = vsctx->pos_temp;
ctx->emit_instruction(ctx, &new_inst);
vsctx->end_instruction = TRUE;
} else {
/* Not an END instruction. */
/* Fix writes to outputs. */
for (i = 0; i < inst->Instruction.NumDstRegs; i++) {
struct tgsi_full_dst_register *dst = &inst->Dst[i];
if (dst->Register.File == TGSI_FILE_OUTPUT) {
if (dst->Register.Index == vsctx->pos_output) {
/* Replace writes to OUT[pos_output] with TEMP[pos_temp]. */
dst->Register.File = TGSI_FILE_TEMPORARY;
dst->Register.Index = vsctx->pos_temp;
} else {
/* Not a position, good...
* Since we were changing the indices of output decls,
* we must redirect writes into them too. */
dst->Register.Index = vsctx->out_remap[dst->Register.Index];
}
}
}
/* Inserting 2 instructions before the END opcode moves all following
* labels by 2. Subroutines are always after the END opcode so
* they're always moved. */
if (inst->Instruction.Opcode == TGSI_OPCODE_CAL) {
inst->Label.Label += 2;
}
/* The labels of the following opcodes are moved only after
* the END opcode. */
if (vsctx->end_instruction &&
(inst->Instruction.Opcode == TGSI_OPCODE_IF ||
inst->Instruction.Opcode == TGSI_OPCODE_ELSE ||
inst->Instruction.Opcode == TGSI_OPCODE_BGNLOOP ||
inst->Instruction.Opcode == TGSI_OPCODE_ENDLOOP)) {
inst->Label.Label += 2;
}
}
ctx->emit_instruction(ctx, inst);
}
/**
* Use pipe_screen::get_param() to query PIPE_CAP_ values to determine
* which GL extensions are supported.
* Quite a few extensions are always supported because they are standard
* features or can be built on top of other gallium features.
* Some fine tuning may still be needed.
*/
void st_init_extensions(struct st_context *st)
{
struct pipe_screen *screen = st->pipe->screen;
struct gl_context *ctx = st->ctx;
int i, glsl_feature_level;
GLboolean *extensions = (GLboolean *) &ctx->Extensions;
static const struct st_extension_cap_mapping cap_mapping[] = {
{ o(ARB_base_instance), PIPE_CAP_START_INSTANCE },
{ o(ARB_depth_clamp), PIPE_CAP_DEPTH_CLIP_DISABLE },
{ o(ARB_depth_texture), PIPE_CAP_TEXTURE_SHADOW_MAP },
{ o(ARB_draw_buffers_blend), PIPE_CAP_INDEP_BLEND_FUNC },
{ o(ARB_draw_instanced), PIPE_CAP_TGSI_INSTANCEID },
{ o(ARB_fragment_program_shadow), PIPE_CAP_TEXTURE_SHADOW_MAP },
{ o(ARB_instanced_arrays), PIPE_CAP_VERTEX_ELEMENT_INSTANCE_DIVISOR },
{ o(ARB_occlusion_query), PIPE_CAP_OCCLUSION_QUERY },
{ o(ARB_occlusion_query2), PIPE_CAP_OCCLUSION_QUERY },
{ o(ARB_point_sprite), PIPE_CAP_POINT_SPRITE },
{ o(ARB_seamless_cube_map), PIPE_CAP_SEAMLESS_CUBE_MAP },
{ o(ARB_shader_stencil_export), PIPE_CAP_SHADER_STENCIL_EXPORT },
{ o(ARB_shader_texture_lod), PIPE_CAP_SM3 },
{ o(ARB_shadow), PIPE_CAP_TEXTURE_SHADOW_MAP },
{ o(ARB_texture_mirror_clamp_to_edge), PIPE_CAP_TEXTURE_MIRROR_CLAMP },
{ o(ARB_texture_non_power_of_two), PIPE_CAP_NPOT_TEXTURES },
{ o(ARB_timer_query), PIPE_CAP_QUERY_TIMESTAMP },
{ o(ARB_transform_feedback2), PIPE_CAP_STREAM_OUTPUT_PAUSE_RESUME },
{ o(ARB_transform_feedback3), PIPE_CAP_STREAM_OUTPUT_PAUSE_RESUME },
{ o(EXT_blend_equation_separate), PIPE_CAP_BLEND_EQUATION_SEPARATE },
{ o(EXT_draw_buffers2), PIPE_CAP_INDEP_BLEND_ENABLE },
{ o(EXT_stencil_two_side), PIPE_CAP_TWO_SIDED_STENCIL },
{ o(EXT_texture_array), PIPE_CAP_MAX_TEXTURE_ARRAY_LAYERS },
{ o(EXT_texture_filter_anisotropic), PIPE_CAP_ANISOTROPIC_FILTER },
{ o(EXT_texture_mirror_clamp), PIPE_CAP_TEXTURE_MIRROR_CLAMP },
{ o(EXT_texture_swizzle), PIPE_CAP_TEXTURE_SWIZZLE },
{ o(EXT_transform_feedback), PIPE_CAP_MAX_STREAM_OUTPUT_BUFFERS },
{ o(AMD_seamless_cubemap_per_texture), PIPE_CAP_SEAMLESS_CUBE_MAP_PER_TEXTURE },
{ o(ATI_separate_stencil), PIPE_CAP_TWO_SIDED_STENCIL },
{ o(ATI_texture_mirror_once), PIPE_CAP_TEXTURE_MIRROR_CLAMP },
{ o(NV_conditional_render), PIPE_CAP_CONDITIONAL_RENDER },
{ o(NV_texture_barrier), PIPE_CAP_TEXTURE_BARRIER },
/* GL_NV_point_sprite is not supported by gallium because we don't
* support the GL_POINT_SPRITE_R_MODE_NV option. */
{ o(MESA_texture_array), PIPE_CAP_MAX_TEXTURE_ARRAY_LAYERS },
{ o(OES_standard_derivatives), PIPE_CAP_SM3 },
{ o(ARB_texture_cube_map_array), PIPE_CAP_CUBE_MAP_ARRAY },
{ o(ARB_texture_multisample), PIPE_CAP_TEXTURE_MULTISAMPLE }
};
/* Required: render target and sampler support */
static const struct st_extension_format_mapping rendertarget_mapping[] = {
{ { o(ARB_texture_float) },
{ PIPE_FORMAT_R32G32B32A32_FLOAT,
PIPE_FORMAT_R16G16B16A16_FLOAT } },
{ { o(ARB_texture_rgb10_a2ui) },
{ PIPE_FORMAT_B10G10R10A2_UINT } },
{ { o(EXT_framebuffer_sRGB) },
{ PIPE_FORMAT_A8B8G8R8_SRGB,
PIPE_FORMAT_B8G8R8A8_SRGB },
GL_TRUE }, /* at least one format must be supported */
{ { o(EXT_packed_float) },
{ PIPE_FORMAT_R11G11B10_FLOAT } },
{ { o(EXT_texture_integer) },
{ PIPE_FORMAT_R32G32B32A32_UINT,
PIPE_FORMAT_R32G32B32A32_SINT } },
{ { o(ARB_texture_rg) },
{ PIPE_FORMAT_R8_UNORM,
PIPE_FORMAT_R8G8_UNORM } },
};
/* Required: depth stencil and sampler support */
static const struct st_extension_format_mapping depthstencil_mapping[] = {
{ { o(ARB_depth_buffer_float) },
{ PIPE_FORMAT_Z32_FLOAT,
PIPE_FORMAT_Z32_FLOAT_S8X24_UINT } },
{ { o(EXT_packed_depth_stencil) },
{ PIPE_FORMAT_S8_UINT_Z24_UNORM,
PIPE_FORMAT_Z24_UNORM_S8_UINT },
GL_TRUE }, /* at least one format must be supported */
};
/* Required: sampler support */
static const struct st_extension_format_mapping texture_mapping[] = {
{ { o(ARB_texture_compression_rgtc) },
{ PIPE_FORMAT_RGTC1_UNORM,
PIPE_FORMAT_RGTC1_SNORM,
PIPE_FORMAT_RGTC2_UNORM,
PIPE_FORMAT_RGTC2_SNORM } },
{ { o(EXT_texture_compression_latc) },
{ PIPE_FORMAT_LATC1_UNORM,
PIPE_FORMAT_LATC1_SNORM,
PIPE_FORMAT_LATC2_UNORM,
PIPE_FORMAT_LATC2_SNORM } },
{ { o(EXT_texture_compression_s3tc),
o(ANGLE_texture_compression_dxt) },
{ PIPE_FORMAT_DXT1_RGB,
PIPE_FORMAT_DXT1_RGBA,
PIPE_FORMAT_DXT3_RGBA,
PIPE_FORMAT_DXT5_RGBA } },
{ { o(EXT_texture_shared_exponent) },
{ PIPE_FORMAT_R9G9B9E5_FLOAT } },
{ { o(EXT_texture_snorm) },
{ PIPE_FORMAT_R8G8B8A8_SNORM } },
{ { o(EXT_texture_sRGB),
o(EXT_texture_sRGB_decode) },
{ PIPE_FORMAT_A8B8G8R8_SRGB,
PIPE_FORMAT_B8G8R8A8_SRGB },
GL_TRUE }, /* at least one format must be supported */
{ { o(ATI_texture_compression_3dc) },
{ PIPE_FORMAT_LATC2_UNORM } },
{ { o(MESA_ycbcr_texture) },
{ PIPE_FORMAT_UYVY,
PIPE_FORMAT_YUYV },
GL_TRUE }, /* at least one format must be supported */
{ { o(OES_compressed_ETC1_RGB8_texture) },
{ PIPE_FORMAT_ETC1_RGB8 } },
};
/* Required: vertex fetch support. */
static const struct st_extension_format_mapping vertex_mapping[] = {
{ { o(ARB_vertex_type_2_10_10_10_rev) },
{ PIPE_FORMAT_R10G10B10A2_UNORM,
PIPE_FORMAT_B10G10R10A2_UNORM,
PIPE_FORMAT_R10G10B10A2_SNORM,
PIPE_FORMAT_B10G10R10A2_SNORM,
PIPE_FORMAT_R10G10B10A2_USCALED,
PIPE_FORMAT_B10G10R10A2_USCALED,
PIPE_FORMAT_R10G10B10A2_SSCALED,
PIPE_FORMAT_B10G10R10A2_SSCALED } },
};
static const struct st_extension_format_mapping tbo_rgb32[] = {
{ {o(ARB_texture_buffer_object_rgb32) },
{ PIPE_FORMAT_R32G32B32_FLOAT,
PIPE_FORMAT_R32G32B32_UINT,
PIPE_FORMAT_R32G32B32_SINT,
} },
};
/*
* Extensions that are supported by all Gallium drivers:
*/
ctx->Extensions.ARB_ES2_compatibility = GL_TRUE;
ctx->Extensions.ARB_draw_elements_base_vertex = GL_TRUE;
ctx->Extensions.ARB_explicit_attrib_location = GL_TRUE;
ctx->Extensions.ARB_fragment_coord_conventions = GL_TRUE;
ctx->Extensions.ARB_fragment_program = GL_TRUE;
ctx->Extensions.ARB_fragment_shader = GL_TRUE;
ctx->Extensions.ARB_half_float_pixel = GL_TRUE;
ctx->Extensions.ARB_half_float_vertex = GL_TRUE;
ctx->Extensions.ARB_internalformat_query = GL_TRUE;
ctx->Extensions.ARB_map_buffer_range = GL_TRUE;
ctx->Extensions.ARB_texture_border_clamp = GL_TRUE; /* XXX temp */
ctx->Extensions.ARB_texture_cube_map = GL_TRUE;
ctx->Extensions.ARB_texture_env_combine = GL_TRUE;
ctx->Extensions.ARB_texture_env_crossbar = GL_TRUE;
ctx->Extensions.ARB_texture_env_dot3 = GL_TRUE;
ctx->Extensions.ARB_vertex_program = GL_TRUE;
ctx->Extensions.ARB_vertex_shader = GL_TRUE;
ctx->Extensions.EXT_blend_color = GL_TRUE;
ctx->Extensions.EXT_blend_func_separate = GL_TRUE;
ctx->Extensions.EXT_blend_minmax = GL_TRUE;
ctx->Extensions.EXT_framebuffer_blit = GL_TRUE;
ctx->Extensions.EXT_gpu_program_parameters = GL_TRUE;
ctx->Extensions.EXT_pixel_buffer_object = GL_TRUE;
ctx->Extensions.EXT_point_parameters = GL_TRUE;
ctx->Extensions.EXT_provoking_vertex = GL_TRUE;
/* IMPORTANT:
* Don't enable EXT_separate_shader_objects. It disallows a certain
* optimization in the GLSL compiler and therefore is considered
* harmful.
*/
ctx->Extensions.EXT_separate_shader_objects = GL_FALSE;
ctx->Extensions.EXT_texture_env_dot3 = GL_TRUE;
ctx->Extensions.EXT_vertex_array_bgra = GL_TRUE;
ctx->Extensions.ATI_texture_env_combine3 = GL_TRUE;
ctx->Extensions.MESA_pack_invert = GL_TRUE;
ctx->Extensions.NV_fog_distance = GL_TRUE;
ctx->Extensions.NV_texture_env_combine4 = GL_TRUE;
ctx->Extensions.NV_texture_rectangle = GL_TRUE;
ctx->Extensions.NV_vdpau_interop = GL_TRUE;
ctx->Extensions.OES_EGL_image = GL_TRUE;
ctx->Extensions.OES_EGL_image_external = GL_TRUE;
ctx->Extensions.OES_draw_texture = GL_TRUE;
/* Expose the extensions which directly correspond to gallium caps. */
for (i = 0; i < Elements(cap_mapping); i++) {
if (screen->get_param(screen, cap_mapping[i].cap)) {
extensions[cap_mapping[i].extension_offset] = GL_TRUE;
}
}
/* Expose the extensions which directly correspond to gallium formats. */
init_format_extensions(st, rendertarget_mapping,
Elements(rendertarget_mapping), PIPE_TEXTURE_2D,
PIPE_BIND_RENDER_TARGET | PIPE_BIND_SAMPLER_VIEW);
init_format_extensions(st, depthstencil_mapping,
Elements(depthstencil_mapping), PIPE_TEXTURE_2D,
PIPE_BIND_DEPTH_STENCIL | PIPE_BIND_SAMPLER_VIEW);
init_format_extensions(st, texture_mapping, Elements(texture_mapping),
PIPE_TEXTURE_2D, PIPE_BIND_SAMPLER_VIEW);
init_format_extensions(st, vertex_mapping, Elements(vertex_mapping),
PIPE_BUFFER, PIPE_BIND_VERTEX_BUFFER);
/* Figure out GLSL support. */
glsl_feature_level = screen->get_param(screen, PIPE_CAP_GLSL_FEATURE_LEVEL);
if (glsl_feature_level >= 140) {
ctx->Const.GLSLVersion = 140;
} else if (glsl_feature_level >= 130) {
ctx->Const.GLSLVersion = 130;
} else {
ctx->Const.GLSLVersion = 120;
}
_mesa_override_glsl_version(st->ctx);
if (st->options.force_glsl_version > 0 &&
st->options.force_glsl_version <= ctx->Const.GLSLVersion) {
ctx->Const.ForceGLSLVersion = st->options.force_glsl_version;
}
if (ctx->Const.GLSLVersion >= 130) {
ctx->Const.NativeIntegers = GL_TRUE;
ctx->Const.MaxClipPlanes = 8;
/* Extensions that either depend on GLSL 1.30 or are a subset thereof. */
ctx->Extensions.ARB_conservative_depth = GL_TRUE;
ctx->Extensions.ARB_shading_language_packing = GL_TRUE;
ctx->Extensions.OES_depth_texture_cube_map = GL_TRUE;
ctx->Extensions.ARB_shading_language_420pack = GL_TRUE;
if (!st->options.disable_shader_bit_encoding) {
ctx->Extensions.ARB_shader_bit_encoding = GL_TRUE;
}
} else {
/* Optional integer support for GLSL 1.2. */
if (screen->get_shader_param(screen, PIPE_SHADER_VERTEX,
PIPE_SHADER_CAP_INTEGERS) &&
screen->get_shader_param(screen, PIPE_SHADER_FRAGMENT,
PIPE_SHADER_CAP_INTEGERS)) {
ctx->Const.NativeIntegers = GL_TRUE;
}
}
/* Below are the cases which cannot be moved into tables easily. */
if (!ctx->Mesa_DXTn && !st->options.force_s3tc_enable) {
ctx->Extensions.EXT_texture_compression_s3tc = GL_FALSE;
ctx->Extensions.ANGLE_texture_compression_dxt = GL_FALSE;
}
if (screen->get_shader_param(screen, PIPE_SHADER_GEOMETRY,
PIPE_SHADER_CAP_MAX_INSTRUCTIONS) > 0) {
#if 0 /* XXX re-enable when GLSL compiler again supports geometry shaders */
ctx->Extensions.ARB_geometry_shader4 = GL_TRUE;
#endif
}
ctx->Extensions.NV_primitive_restart = GL_TRUE;
if (!screen->get_param(screen, PIPE_CAP_PRIMITIVE_RESTART)) {
ctx->Const.PrimitiveRestartInSoftware = GL_TRUE;
}
/* ARB_color_buffer_float. */
if (screen->get_param(screen, PIPE_CAP_VERTEX_COLOR_UNCLAMPED)) {
ctx->Extensions.ARB_color_buffer_float = GL_TRUE;
if (!screen->get_param(screen, PIPE_CAP_VERTEX_COLOR_CLAMPED)) {
st->clamp_vert_color_in_shader = TRUE;
}
if (!screen->get_param(screen, PIPE_CAP_FRAGMENT_COLOR_CLAMPED)) {
st->clamp_frag_color_in_shader = TRUE;
}
/* For drivers which cannot do color clamping, it's better to just
* disable ARB_color_buffer_float in the core profile, because
* the clamping is deprecated there anyway. */
if (ctx->API == API_OPENGL_CORE &&
(st->clamp_frag_color_in_shader || st->clamp_vert_color_in_shader)) {
st->clamp_vert_color_in_shader = GL_FALSE;
st->clamp_frag_color_in_shader = GL_FALSE;
ctx->Extensions.ARB_color_buffer_float = GL_FALSE;
}
}
if (screen->fence_finish) {
ctx->Extensions.ARB_sync = GL_TRUE;
}
/* Maximum sample count. */
for (i = 16; i > 0; --i) {
enum pipe_format pformat = st_choose_format(st, GL_RGBA,
GL_NONE, GL_NONE,
PIPE_TEXTURE_2D, i,
PIPE_BIND_RENDER_TARGET, FALSE);
if (pformat != PIPE_FORMAT_NONE) {
ctx->Const.MaxSamples = i;
ctx->Const.MaxColorTextureSamples = i;
break;
}
}
for (i = ctx->Const.MaxSamples; i > 0; --i) {
enum pipe_format pformat = st_choose_format(st, GL_DEPTH_STENCIL,
GL_NONE, GL_NONE,
PIPE_TEXTURE_2D, i,
PIPE_BIND_DEPTH_STENCIL, FALSE);
if (pformat != PIPE_FORMAT_NONE) {
ctx->Const.MaxDepthTextureSamples = i;
break;
}
}
for (i = ctx->Const.MaxSamples; i > 0; --i) {
enum pipe_format pformat = st_choose_format(st, GL_RGBA_INTEGER,
GL_NONE, GL_NONE,
PIPE_TEXTURE_2D, i,
PIPE_BIND_RENDER_TARGET, FALSE);
if (pformat != PIPE_FORMAT_NONE) {
ctx->Const.MaxIntegerSamples = i;
break;
}
}
if (ctx->Const.MaxSamples == 1) {
/* one sample doesn't really make sense */
ctx->Const.MaxSamples = 0;
}
else if (ctx->Const.MaxSamples >= 2) {
ctx->Extensions.EXT_framebuffer_multisample = GL_TRUE;
ctx->Extensions.EXT_framebuffer_multisample_blit_scaled = GL_TRUE;
}
if (ctx->Const.MaxDualSourceDrawBuffers > 0 &&
!st->options.disable_blend_func_extended)
ctx->Extensions.ARB_blend_func_extended = GL_TRUE;
st->has_time_elapsed =
screen->get_param(screen, PIPE_CAP_QUERY_TIME_ELAPSED);
if (st->has_time_elapsed ||
ctx->Extensions.ARB_timer_query) {
ctx->Extensions.EXT_timer_query = GL_TRUE;
}
if (ctx->Extensions.ARB_transform_feedback2 &&
ctx->Extensions.ARB_draw_instanced) {
ctx->Extensions.ARB_transform_feedback_instanced = GL_TRUE;
}
if (st->options.force_glsl_extensions_warn)
ctx->Const.ForceGLSLExtensionsWarn = 1;
if (st->options.disable_glsl_line_continuations)
ctx->Const.DisableGLSLLineContinuations = 1;
ctx->Const.MinMapBufferAlignment =
screen->get_param(screen, PIPE_CAP_MIN_MAP_BUFFER_ALIGNMENT);
if (ctx->Const.MinMapBufferAlignment >= 64) {
ctx->Extensions.ARB_map_buffer_alignment = GL_TRUE;
}
if (screen->get_param(screen, PIPE_CAP_TEXTURE_BUFFER_OBJECTS)) {
ctx->Extensions.ARB_texture_buffer_object = GL_TRUE;
ctx->Const.MaxTextureBufferSize =
_min(screen->get_param(screen, PIPE_CAP_MAX_TEXTURE_BUFFER_SIZE),
(1u << 31) - 1);
ctx->Const.TextureBufferOffsetAlignment =
screen->get_param(screen, PIPE_CAP_TEXTURE_BUFFER_OFFSET_ALIGNMENT);
if (ctx->Const.TextureBufferOffsetAlignment)
ctx->Extensions.ARB_texture_buffer_range = GL_TRUE;
init_format_extensions(st, tbo_rgb32, Elements(tbo_rgb32),
PIPE_BUFFER, PIPE_BIND_SAMPLER_VIEW);
}
if (screen->get_param(screen, PIPE_CAP_MIXED_FRAMEBUFFER_SIZES) &&
ctx->Extensions.EXT_packed_depth_stencil) {
ctx->Extensions.ARB_framebuffer_object = GL_TRUE;
}
/* Unpacking a varying in the fragment shader costs 1 texture indirection.
* If the number of available texture indirections is very limited, then we
* prefer to disable varying packing rather than run the risk of varying
* packing preventing a shader from running.
*/
if (screen->get_shader_param(screen, PIPE_SHADER_FRAGMENT,
PIPE_SHADER_CAP_MAX_TEX_INDIRECTIONS) <= 8) {
/* We can't disable varying packing if transform feedback is available,
* because transform feedback code assumes a packed varying layout.
*/
if (!ctx->Extensions.EXT_transform_feedback)
ctx->Const.DisableVaryingPacking = GL_TRUE;
}
}
void r300_draw_init_vertex_shader(struct r300_context *r300,
struct r300_vertex_shader *vs)
{
struct draw_context *draw = r300->draw;
struct pipe_shader_state new_vs;
struct tgsi_shader_info info;
struct vs_transform_context transform;
const uint newLen = tgsi_num_tokens(vs->state.tokens) + 100 /* XXX */;
unsigned i;
tgsi_scan_shader(vs->state.tokens, &info);
new_vs.tokens = tgsi_alloc_tokens(newLen);
if (new_vs.tokens == NULL)
return;
memset(&transform, 0, sizeof(transform));
for (i = 0; i < Elements(transform.out_remap); i++) {
transform.out_remap[i] = i;
}
transform.last_generic = -1;
transform.base.transform_instruction = transform_inst;
transform.base.transform_declaration = transform_decl;
for (i = 0; i < info.num_outputs; i++) {
unsigned index = info.output_semantic_index[i];
switch (info.output_semantic_name[i]) {
case TGSI_SEMANTIC_COLOR:
assert(index < 2);
transform.color_used[index] = TRUE;
break;
case TGSI_SEMANTIC_BCOLOR:
assert(index < 2);
transform.bcolor_used[index] = TRUE;
break;
}
}
tgsi_transform_shader(vs->state.tokens,
(struct tgsi_token*)new_vs.tokens,
newLen, &transform.base);
#if 0
printf("----------------------------------------------\norig shader:\n");
tgsi_dump(vs->state.tokens, 0);
printf("----------------------------------------------\nnew shader:\n");
tgsi_dump(new_vs.tokens, 0);
printf("----------------------------------------------\n");
#endif
/* Free old tokens. */
FREE((void*)vs->state.tokens);
vs->draw_vs = draw_create_vertex_shader(draw, &new_vs);
/* Instead of duplicating and freeing the tokens, copy the pointer directly. */
vs->state.tokens = new_vs.tokens;
/* Init the VS output table for the rasterizer. */
r300_init_vs_outputs(r300, vs);
/* Make the last generic be WPOS. */
vs->outputs.wpos = vs->outputs.generic[transform.last_generic + 1];
vs->outputs.generic[transform.last_generic + 1] = ATTR_UNUSED;
}
const char *ir_expression::operator_string(ir_expression_operation op)
{
assert((unsigned int) op < Elements(operator_strs));
assert(Elements(operator_strs) == (ir_quadop_vector + 1));
return operator_strs[op];
}
/**
* Use pipe_screen::get_param() to query PIPE_CAP_ values to determine
* which GL extensions are supported.
* Quite a few extensions are always supported because they are standard
* features or can be built on top of other gallium features.
* Some fine tuning may still be needed.
*/
void st_init_extensions(struct st_context *st)
{
struct pipe_screen *screen = st->pipe->screen;
struct gl_context *ctx = st->ctx;
int i, glsl_feature_level;
GLboolean *extensions = (GLboolean *) &ctx->Extensions;
static const struct st_extension_cap_mapping cap_mapping[] = {
{ o(ARB_base_instance), PIPE_CAP_START_INSTANCE },
{ o(ARB_depth_clamp), PIPE_CAP_DEPTH_CLIP_DISABLE },
{ o(ARB_depth_texture), PIPE_CAP_TEXTURE_SHADOW_MAP },
{ o(ARB_draw_buffers_blend), PIPE_CAP_INDEP_BLEND_FUNC },
{ o(ARB_draw_instanced), PIPE_CAP_TGSI_INSTANCEID },
{ o(ARB_fragment_program_shadow), PIPE_CAP_TEXTURE_SHADOW_MAP },
{ o(ARB_instanced_arrays), PIPE_CAP_VERTEX_ELEMENT_INSTANCE_DIVISOR },
{ o(ARB_occlusion_query), PIPE_CAP_OCCLUSION_QUERY },
{ o(ARB_occlusion_query2), PIPE_CAP_OCCLUSION_QUERY },
{ o(ARB_point_sprite), PIPE_CAP_POINT_SPRITE },
{ o(ARB_seamless_cube_map), PIPE_CAP_SEAMLESS_CUBE_MAP },
{ o(ARB_shader_stencil_export), PIPE_CAP_SHADER_STENCIL_EXPORT },
{ o(ARB_shader_texture_lod), PIPE_CAP_SM3 },
{ o(ARB_shadow), PIPE_CAP_TEXTURE_SHADOW_MAP },
{ o(ARB_texture_non_power_of_two), PIPE_CAP_NPOT_TEXTURES },
{ o(ARB_transform_feedback2), PIPE_CAP_STREAM_OUTPUT_PAUSE_RESUME },
{ o(EXT_blend_equation_separate), PIPE_CAP_BLEND_EQUATION_SEPARATE },
{ o(EXT_draw_buffers2), PIPE_CAP_INDEP_BLEND_ENABLE },
{ o(EXT_shadow_funcs), PIPE_CAP_TEXTURE_SHADOW_MAP },
{ o(EXT_stencil_two_side), PIPE_CAP_TWO_SIDED_STENCIL },
{ o(EXT_texture_array), PIPE_CAP_MAX_TEXTURE_ARRAY_LAYERS },
{ o(EXT_texture_filter_anisotropic), PIPE_CAP_ANISOTROPIC_FILTER },
{ o(EXT_texture_mirror_clamp), PIPE_CAP_TEXTURE_MIRROR_CLAMP },
{ o(EXT_texture_swizzle), PIPE_CAP_TEXTURE_SWIZZLE },
{ o(EXT_timer_query), PIPE_CAP_TIMER_QUERY },
{ o(EXT_transform_feedback), PIPE_CAP_MAX_STREAM_OUTPUT_BUFFERS },
{ o(AMD_seamless_cubemap_per_texture), PIPE_CAP_SEAMLESS_CUBE_MAP_PER_TEXTURE },
{ o(ATI_separate_stencil), PIPE_CAP_TWO_SIDED_STENCIL },
{ o(ATI_texture_mirror_once), PIPE_CAP_TEXTURE_MIRROR_CLAMP },
{ o(NV_conditional_render), PIPE_CAP_CONDITIONAL_RENDER },
{ o(NV_texture_barrier), PIPE_CAP_TEXTURE_BARRIER },
/* GL_NV_point_sprite is not supported by gallium because we don't
* support the GL_POINT_SPRITE_R_MODE_NV option. */
{ o(MESA_texture_array), PIPE_CAP_MAX_TEXTURE_ARRAY_LAYERS },
};
/* Required: render target and sampler support */
static const struct st_extension_format_mapping rendertarget_mapping[] = {
{ { o(ARB_texture_float) },
{ PIPE_FORMAT_R32G32B32A32_FLOAT,
PIPE_FORMAT_R16G16B16A16_FLOAT } },
{ { o(ARB_texture_rgb10_a2ui) },
{ PIPE_FORMAT_B10G10R10A2_UINT } },
{ { o(EXT_framebuffer_sRGB) },
{ PIPE_FORMAT_A8B8G8R8_SRGB,
PIPE_FORMAT_B8G8R8A8_SRGB },
GL_TRUE }, /* at least one format must be supported */
{ { o(EXT_packed_float) },
{ PIPE_FORMAT_R11G11B10_FLOAT } },
{ { o(EXT_texture_integer) },
{ PIPE_FORMAT_R32G32B32A32_UINT,
PIPE_FORMAT_R32G32B32A32_SINT } },
};
/* Required: depth stencil and sampler support */
static const struct st_extension_format_mapping depthstencil_mapping[] = {
{ { o(ARB_depth_buffer_float) },
{ PIPE_FORMAT_Z32_FLOAT,
PIPE_FORMAT_Z32_FLOAT_S8X24_UINT } },
{ { o(ARB_framebuffer_object),
o(EXT_packed_depth_stencil) },
{ PIPE_FORMAT_S8_UINT_Z24_UNORM,
PIPE_FORMAT_Z24_UNORM_S8_UINT },
GL_TRUE }, /* at least one format must be supported */
};
/* Required: sampler support */
static const struct st_extension_format_mapping texture_mapping[] = {
{ { o(ARB_texture_compression_rgtc) },
{ PIPE_FORMAT_RGTC1_UNORM,
PIPE_FORMAT_RGTC1_SNORM,
PIPE_FORMAT_RGTC2_UNORM,
PIPE_FORMAT_RGTC2_SNORM } },
{ { o(ARB_texture_rg) },
{ PIPE_FORMAT_R8G8_UNORM } },
{ { o(EXT_texture_compression_latc) },
{ PIPE_FORMAT_LATC1_UNORM,
PIPE_FORMAT_LATC1_SNORM,
PIPE_FORMAT_LATC2_UNORM,
PIPE_FORMAT_LATC2_SNORM } },
{ { o(EXT_texture_compression_s3tc),
o(S3_s3tc) },
{ PIPE_FORMAT_DXT1_RGB,
PIPE_FORMAT_DXT1_RGBA,
PIPE_FORMAT_DXT3_RGBA,
PIPE_FORMAT_DXT5_RGBA } },
{ { o(EXT_texture_shared_exponent) },
{ PIPE_FORMAT_R9G9B9E5_FLOAT } },
{ { o(EXT_texture_snorm) },
{ PIPE_FORMAT_R8G8B8A8_SNORM } },
{ { o(EXT_texture_sRGB),
o(EXT_texture_sRGB_decode) },
{ PIPE_FORMAT_A8B8G8R8_SRGB,
PIPE_FORMAT_B8G8R8A8_SRGB },
GL_TRUE }, /* at least one format must be supported */
{ { o(ATI_texture_compression_3dc) },
{ PIPE_FORMAT_LATC2_UNORM } },
{ { o(MESA_ycbcr_texture) },
{ PIPE_FORMAT_UYVY,
PIPE_FORMAT_YUYV },
GL_TRUE }, /* at least one format must be supported */
{ { o(OES_compressed_ETC1_RGB8_texture) },
{ PIPE_FORMAT_ETC1_RGB8 } },
};
/* Required: vertex fetch support. */
static const struct st_extension_format_mapping vertex_mapping[] = {
{ { o(ARB_vertex_type_2_10_10_10_rev) },
{ PIPE_FORMAT_R10G10B10A2_UNORM,
PIPE_FORMAT_B10G10R10A2_UNORM,
PIPE_FORMAT_R10G10B10A2_SNORM,
PIPE_FORMAT_B10G10R10A2_SNORM,
PIPE_FORMAT_R10G10B10A2_USCALED,
PIPE_FORMAT_B10G10R10A2_USCALED,
PIPE_FORMAT_R10G10B10A2_SSCALED,
PIPE_FORMAT_B10G10R10A2_SSCALED } },
};
/*
* Extensions that are supported by all Gallium drivers:
*/
ctx->Extensions.ARB_ES2_compatibility = GL_TRUE;
ctx->Extensions.ARB_copy_buffer = GL_TRUE;
ctx->Extensions.ARB_draw_elements_base_vertex = GL_TRUE;
ctx->Extensions.ARB_explicit_attrib_location = GL_TRUE;
ctx->Extensions.ARB_fragment_coord_conventions = GL_TRUE;
ctx->Extensions.ARB_fragment_program = GL_TRUE;
ctx->Extensions.ARB_fragment_shader = GL_TRUE;
ctx->Extensions.ARB_half_float_pixel = GL_TRUE;
ctx->Extensions.ARB_half_float_vertex = GL_TRUE;
ctx->Extensions.ARB_map_buffer_range = GL_TRUE;
ctx->Extensions.ARB_sampler_objects = GL_TRUE;
ctx->Extensions.ARB_shader_objects = GL_TRUE;
ctx->Extensions.ARB_shading_language_100 = GL_TRUE;
ctx->Extensions.ARB_texture_border_clamp = GL_TRUE; /* XXX temp */
ctx->Extensions.ARB_texture_cube_map = GL_TRUE;
ctx->Extensions.ARB_texture_env_combine = GL_TRUE;
ctx->Extensions.ARB_texture_env_crossbar = GL_TRUE;
ctx->Extensions.ARB_texture_env_dot3 = GL_TRUE;
ctx->Extensions.ARB_texture_storage = GL_TRUE;
ctx->Extensions.ARB_vertex_array_object = GL_TRUE;
ctx->Extensions.ARB_vertex_program = GL_TRUE;
ctx->Extensions.ARB_vertex_shader = GL_TRUE;
ctx->Extensions.ARB_window_pos = GL_TRUE;
ctx->Extensions.EXT_blend_color = GL_TRUE;
ctx->Extensions.EXT_blend_func_separate = GL_TRUE;
ctx->Extensions.EXT_blend_minmax = GL_TRUE;
ctx->Extensions.EXT_framebuffer_blit = GL_TRUE;
ctx->Extensions.EXT_framebuffer_object = GL_TRUE;
ctx->Extensions.EXT_framebuffer_multisample = GL_TRUE;
ctx->Extensions.EXT_fog_coord = GL_TRUE;
ctx->Extensions.EXT_gpu_program_parameters = GL_TRUE;
ctx->Extensions.EXT_pixel_buffer_object = GL_TRUE;
ctx->Extensions.EXT_point_parameters = GL_TRUE;
ctx->Extensions.EXT_provoking_vertex = GL_TRUE;
ctx->Extensions.EXT_secondary_color = GL_TRUE;
ctx->Extensions.EXT_separate_shader_objects = GL_TRUE;
ctx->Extensions.EXT_texture_env_dot3 = GL_TRUE;
ctx->Extensions.EXT_vertex_array_bgra = GL_TRUE;
ctx->Extensions.APPLE_vertex_array_object = GL_TRUE;
ctx->Extensions.ATI_texture_env_combine3 = GL_TRUE;
ctx->Extensions.MESA_pack_invert = GL_TRUE;
ctx->Extensions.NV_blend_square = GL_TRUE;
ctx->Extensions.NV_fog_distance = GL_TRUE;
ctx->Extensions.NV_texgen_reflection = GL_TRUE;
ctx->Extensions.NV_texture_env_combine4 = GL_TRUE;
ctx->Extensions.NV_texture_rectangle = GL_TRUE;
#if 0
/* possibly could support the following two */
ctx->Extensions.NV_vertex_program = GL_TRUE;
ctx->Extensions.NV_vertex_program1_1 = GL_TRUE;
#endif
#if FEATURE_OES_EGL_image
ctx->Extensions.OES_EGL_image = GL_TRUE;
if (ctx->API != API_OPENGL)
ctx->Extensions.OES_EGL_image_external = GL_TRUE;
#endif
#if FEATURE_OES_draw_texture
ctx->Extensions.OES_draw_texture = GL_TRUE;
#endif
/* Expose the extensions which directly correspond to gallium caps. */
for (i = 0; i < Elements(cap_mapping); i++) {
if (screen->get_param(screen, cap_mapping[i].cap)) {
extensions[cap_mapping[i].extension_offset] = GL_TRUE;
}
}
/* Expose the extensions which directly correspond to gallium formats. */
init_format_extensions(st, rendertarget_mapping,
Elements(rendertarget_mapping), PIPE_TEXTURE_2D,
PIPE_BIND_RENDER_TARGET | PIPE_BIND_SAMPLER_VIEW);
init_format_extensions(st, depthstencil_mapping,
Elements(depthstencil_mapping), PIPE_TEXTURE_2D,
PIPE_BIND_DEPTH_STENCIL | PIPE_BIND_SAMPLER_VIEW);
init_format_extensions(st, texture_mapping, Elements(texture_mapping),
PIPE_TEXTURE_2D, PIPE_BIND_SAMPLER_VIEW);
init_format_extensions(st, vertex_mapping, Elements(vertex_mapping),
PIPE_BUFFER, PIPE_BIND_VERTEX_BUFFER);
/* Figure out GLSL support. */
glsl_feature_level = screen->get_param(screen, PIPE_CAP_GLSL_FEATURE_LEVEL);
if (glsl_feature_level >= 130) {
ctx->Const.GLSLVersion = 130;
} else {
ctx->Const.GLSLVersion = 120;
}
_mesa_override_glsl_version(st->ctx);
if (ctx->Const.GLSLVersion >= 130) {
ctx->Const.NativeIntegers = GL_TRUE;
ctx->Const.MaxClipPlanes = 8;
/* Extensions that only depend on GLSL 1.3. */
ctx->Extensions.ARB_conservative_depth = GL_TRUE;
} else {
/* Optional integer support for GLSL 1.2. */
if (screen->get_shader_param(screen, PIPE_SHADER_VERTEX,
PIPE_SHADER_CAP_INTEGERS) &&
screen->get_shader_param(screen, PIPE_SHADER_FRAGMENT,
PIPE_SHADER_CAP_INTEGERS)) {
ctx->Const.NativeIntegers = GL_TRUE;
}
}
/* Below are the cases which cannot be moved into tables easily. */
if (!ctx->Mesa_DXTn && !st_get_s3tc_override()) {
ctx->Extensions.EXT_texture_compression_s3tc = GL_FALSE;
ctx->Extensions.S3_s3tc = GL_FALSE;
}
if (ctx->Const.NativeIntegers) {
ctx->Extensions.ARB_shader_bit_encoding = GL_TRUE;
}
if (screen->get_shader_param(screen, PIPE_SHADER_GEOMETRY,
PIPE_SHADER_CAP_MAX_INSTRUCTIONS) > 0) {
#if 0 /* XXX re-enable when GLSL compiler again supports geometry shaders */
ctx->Extensions.ARB_geometry_shader4 = GL_TRUE;
#endif
}
ctx->Extensions.NV_primitive_restart = GL_TRUE;
if (!screen->get_param(screen, PIPE_CAP_PRIMITIVE_RESTART)) {
ctx->Const.PrimitiveRestartInSoftware = GL_TRUE;
}
if (screen->get_param(screen, PIPE_CAP_VERTEX_COLOR_UNCLAMPED)) {
ctx->Extensions.ARB_color_buffer_float = GL_TRUE;
if (!screen->get_param(screen, PIPE_CAP_VERTEX_COLOR_CLAMPED)) {
st->clamp_vert_color_in_shader = TRUE;
}
if (!screen->get_param(screen, PIPE_CAP_FRAGMENT_COLOR_CLAMPED)) {
st->clamp_frag_color_in_shader = TRUE;
}
}
if (screen->fence_finish) {
ctx->Extensions.ARB_sync = GL_TRUE;
}
/* Maximum sample count. */
for (i = 16; i > 0; --i) {
if (screen->is_format_supported(screen, PIPE_FORMAT_B8G8R8A8_UNORM,
PIPE_TEXTURE_2D, i,
PIPE_BIND_RENDER_TARGET)) {
ctx->Const.MaxSamples = i;
break;
}
}
if (ctx->Const.MaxDualSourceDrawBuffers > 0)
ctx->Extensions.ARB_blend_func_extended = GL_TRUE;
}
static int Integrate(This *t, real *integral, real *error, real *prob)
{
TYPEDEFREGION;
Totals totals[NCOMP];
real nneed, weight;
count dim, comp, iter, pass = 0, err, iregion;
number nwant, nmin = INT_MAX;
int fail;
if( VERBOSE > 1 ) {
char s[512];
sprintf(s, "Divonne input parameters:\n"
" ndim " COUNT "\n ncomp " COUNT "\n"
" epsrel " REAL "\n epsabs " REAL "\n"
" flags %d\n seed %d\n"
" mineval " NUMBER "\n maxeval " NUMBER "\n"
" key1 %d\n key2 %d\n key3 %d\n maxpass " COUNT "\n"
" border " REAL "\n maxchisq " REAL "\n mindeviation " REAL "\n"
" ngiven " NUMBER "\n nextra " NUMBER,
t->ndim, t->ncomp,
t->epsrel, t->epsabs,
t->flags, t->seed,
t->mineval, t->maxeval,
t->key1, t->key2, t->key3, t->maxpass,
t->border.lower, t->maxchisq, t->mindeviation,
t->ngiven, t->nextra);
Print(s);
}
if( BadComponent(t) ) return -2;
if( BadDimension(t, t->key1) ||
BadDimension(t, t->key2) ||
((t->key3 & -2) && BadDimension(t, t->key3)) ) return -1;
t->neval_opt = t->neval_cut = 0;
t->size = CHUNKSIZE;
MemAlloc(t->voidregion, t->size*sizeof(Region));
for( dim = 0; dim < t->ndim; ++dim ) {
Bounds *b = &RegionPtr(0)->bounds[dim];
b->lower = 0;
b->upper = 1;
}
RuleIni(&t->rule7);
RuleIni(&t->rule9);
RuleIni(&t->rule11);
RuleIni(&t->rule13);
SamplesIni(&t->samples[0]);
SamplesIni(&t->samples[1]);
SamplesIni(&t->samples[2]);
if( (fail = setjmp(t->abort)) ) goto abort;
t->epsabs = Max(t->epsabs, NOTZERO);
/* Step 1: partition the integration region */
if( VERBOSE ) Print("Partitioning phase:");
if( IsSobol(t->key1) || IsSobol(t->key2) || IsSobol(t->key3) )
IniRandom(t);
SamplesLookup(t, &t->samples[0], t->key1,
(number)47, (number)INT_MAX, (number)0);
SamplesAlloc(t, &t->samples[0]);
t->totals = totals;
Zap(totals);
t->phase = 1;
Explore(t, 0, &t->samples[0], INIDEPTH, 1);
for( iter = 1; ; ++iter ) {
Totals *maxtot;
count valid;
for( comp = 0; comp < t->ncomp; ++comp ) {
Totals *tot = &totals[comp];
tot->avg = tot->spreadsq = 0;
tot->spread = tot->secondspread = -INFTY;
}
for( iregion = 0; iregion < t->nregions; ++iregion ) {
Region *region = RegionPtr(iregion);
for( comp = 0; comp < t->ncomp; ++comp ) {
cResult *r = ®ion->result[comp];
Totals *tot = &totals[comp];
tot->avg += r->avg;
tot->spreadsq += Sq(r->spread);
if( r->spread > tot->spread ) {
tot->secondspread = tot->spread;
tot->spread = r->spread;
tot->iregion = iregion;
}
else if( r->spread > tot->secondspread )
tot->secondspread = r->spread;
}
}
maxtot = totals;
valid = 0;
for( comp = 0; comp < t->ncomp; ++comp ) {
Totals *tot = &totals[comp];
integral[comp] = tot->avg;
valid += tot->avg == tot->avg;
if( tot->spreadsq > maxtot->spreadsq ) maxtot = tot;
tot->spread = sqrt(tot->spreadsq);
error[comp] = tot->spread*t->samples[0].weight;
}
if( VERBOSE ) {
char s[128 + 64*NCOMP], *p = s;
p += sprintf(p, "\n"
"Iteration " COUNT " (pass " COUNT "): " COUNT " regions\n"
NUMBER7 " integrand evaluations so far,\n"
NUMBER7 " in optimizing regions,\n"
NUMBER7 " in finding cuts",
iter, pass, t->nregions, t->neval, t->neval_opt, t->neval_cut);
for( comp = 0; comp < t->ncomp; ++comp )
p += sprintf(p, "\n[" COUNT "] "
REAL " +- " REAL,
comp + 1, integral[comp], error[comp]);
Print(s);
}
if( valid == 0 ) goto abort; /* all NaNs */
if( t->neval > t->maxeval ) break;
nneed = maxtot->spread/MaxErr(maxtot->avg);
if( nneed < MAXPRIME ) {
cnumber n = t->neval + t->nregions*(number)ceil(nneed);
if( n < nmin ) {
nmin = n;
pass = 0;
}
else if( ++pass > t->maxpass && n >= t->mineval ) break;
}
Split(t, maxtot->iregion, DEPTH);
}
/* Step 2: do a "full" integration on each region */
/* nneed = t->samples[0].neff + 1; */
nneed = 2*t->samples[0].neff;
for( comp = 0; comp < t->ncomp; ++comp ) {
Totals *tot = &totals[comp];
creal maxerr = MaxErr(tot->avg);
tot->nneed = tot->spread/maxerr;
nneed = Max(nneed, tot->nneed);
tot->maxerrsq = Sq(maxerr);
tot->mindevsq = tot->maxerrsq*Sq(t->mindeviation);
}
nwant = (number)Min(ceil(nneed), MARKMASK/40.);
err = SamplesLookup(t, &t->samples[1], t->key2, nwant,
(t->maxeval - t->neval)/t->nregions + 1, t->samples[0].n + 1);
/* the number of points needed to reach the desired accuracy */
fail = Unmark(err)*t->nregions;
if( Marked(err) ) {
if( VERBOSE ) Print("\nNot enough samples left for main integration.");
for( comp = 0; comp < t->ncomp; ++comp )
prob[comp] = -999;
weight = t->samples[0].weight;
}
else {
bool can_adjust = (t->key3 == 1 && t->samples[1].sampler != SampleRule &&
(t->key2 < 0 || t->samples[1].neff < MAXPRIME));
count df, nlimit;
SamplesAlloc(t, &t->samples[1]);
if( VERBOSE ) {
char s[128];
sprintf(s, "\nMain integration on " COUNT
" regions with " NUMBER " samples per region.",
t->nregions, t->samples[1].neff);
Print(s);
}
ResClear(integral);
ResClear(error);
ResClear(prob);
nlimit = t->maxeval - t->nregions*t->samples[1].n;
df = 0;
for( iregion = 0; iregion < t->nregions; ++iregion ) {
Region *region = RegionPtr(iregion);
char s[64*NDIM + 256*NCOMP], *p = s;
int todo;
refine:
t->phase = 2;
t->samples[1].sampler(t, &t->samples[1], region->bounds, region->vol);
if( can_adjust )
for( comp = 0; comp < t->ncomp; ++comp )
totals[comp].spreadsq -= Sq(region->result[comp].spread);
nlimit += t->samples[1].n;
todo = 0;
for( comp = 0; comp < t->ncomp; ++comp ) {
cResult *r = ®ion->result[comp];
Totals *tot = &totals[comp];
t->samples[0].avg[comp] = r->avg;
t->samples[0].err[comp] = r->err;
if( t->neval < nlimit ) {
creal avg2 = t->samples[1].avg[comp];
creal err2 = t->samples[1].err[comp];
creal diffsq = Sq(avg2 - r->avg);
#define Var(s) Sq((s.err[comp] == 0) ? r->spread*s.weight : s.err[comp])
if( err2*tot->nneed > r->spread ||
diffsq > Max(t->maxchisq*(Var(t->samples[0]) + Var(t->samples[1])),
EPS*Sq(avg2)) ) {
if( t->key3 && diffsq > tot->mindevsq ) {
if( t->key3 == 1 ) {
ccount xregion = t->nregions;
if( VERBOSE > 2 ) Print("\nSplit");
t->phase = 1;
Explore(t, iregion, &t->samples[1], POSTDEPTH, 2);
if( can_adjust ) {
number nnew;
count ireg, xreg;
for( ireg = iregion, xreg = xregion;
ireg < t->nregions; ireg = xreg++ ) {
cResult *result = RegionPtr(ireg)->result;
count c;
for( c = 0; c < t->ncomp; ++c )
totals[c].spreadsq += Sq(result[c].spread);
}
nnew = (tot->spreadsq/Sq(MARKMASK) > tot->maxerrsq) ?
MARKMASK :
(number)ceil(sqrt(tot->spreadsq/tot->maxerrsq));
if( nnew > nwant + nwant/64 ) {
ccount err = SamplesLookup(t, &t->samples[1], t->key2, nnew,
(t->maxeval - t->neval)/t->nregions + 1, t->samples[1].n);
fail += Unmark(err)*t->nregions;
nwant = nnew;
SamplesFree(&t->samples[1]);
SamplesAlloc(t, &t->samples[1]);
if( t->key2 > 0 && t->samples[1].neff >= MAXPRIME )
can_adjust = false;
if( VERBOSE > 2 ) {
char s[128];
sprintf(s, "Sampling remaining " COUNT
" regions with " NUMBER " points per region.",
t->nregions, t->samples[1].neff);
Print(s);
}
}
}
goto refine;
}
todo |= 3;
}
todo |= 1;
}
}
}
if( can_adjust ) {
for( comp = 0; comp < t->ncomp; ++comp )
totals[comp].maxerrsq -=
Sq(region->result[comp].spread*t->samples[1].weight);
}
switch( todo ) {
case 1: /* get spread right */
Explore(t, iregion, &t->samples[1], 0, 2);
break;
case 3: /* sample region again with more points */
if( SamplesIniQ(&t->samples[2]) ) {
SamplesLookup(t, &t->samples[2], t->key3,
nwant, (number)INT_MAX, (number)0);
SamplesAlloc(t, &t->samples[2]);
}
t->phase = 3;
t->samples[2].sampler(t, &t->samples[2], region->bounds, region->vol);
Explore(t, iregion, &t->samples[2], 0, 2);
++region->depth; /* misused for df here */
++df;
}
++region->depth; /* misused for df here */
if( VERBOSE > 2 ) {
for( dim = 0; dim < t->ndim; ++dim ) {
cBounds *b = ®ion->bounds[dim];
p += sprintf(p,
(dim == 0) ? "\nRegion (" REALF ") - (" REALF ")" :
"\n (" REALF ") - (" REALF ")",
b->lower, b->upper);
}
}
for( comp = 0; comp < t->ncomp; ++comp ) {
Result *r = ®ion->result[comp];
creal x1 = t->samples[0].avg[comp];
creal s1 = Var(t->samples[0]);
creal x2 = t->samples[1].avg[comp];
creal s2 = Var(t->samples[1]);
creal r2 = (s1 == 0) ? Sq(t->samples[1].neff*t->samples[0].weight) : s2/s1;
real norm = 1 + r2;
real avg = x2 + r2*x1;
real sigsq = s2;
real chisq = Sq(x2 - x1);
real chiden = s1 + s2;
if( todo == 3 ) {
creal x3 = t->samples[2].avg[comp];
creal s3 = Var(t->samples[2]);
creal r3 = (s2 == 0) ? Sq(t->samples[2].neff*t->samples[1].weight) : s3/s2;
norm = 1 + r3*norm;
avg = x3 + r3*avg;
sigsq = s3;
chisq = s1*Sq(x3 - x2) + s2*Sq(x3 - x1) + s3*chisq;
chiden = s1*s2 + s3*chiden;
}
avg = LAST ? r->avg : (sigsq *= norm = 1/norm, avg*norm);
if( chisq > EPS ) chisq /= Max(chiden, NOTZERO);
#define Out(s) s.avg[comp], r->spread*s.weight, s.err[comp]
if( VERBOSE > 2 ) {
p += sprintf(p, "\n[" COUNT "] "
REAL " +- " REAL "(" REAL ")\n "
REAL " +- " REAL "(" REAL ")",
comp + 1, Out(t->samples[0]), Out(t->samples[1]));
if( todo == 3 ) p += sprintf(p, "\n "
REAL " +- " REAL "(" REAL ")",
Out(t->samples[2]));
p += sprintf(p, " \tchisq " REAL, chisq);
}
integral[comp] += avg;
error[comp] += sigsq;
prob[comp] += chisq;
r->avg = avg;
r->spread = sqrt(sigsq);
r->chisq = chisq;
}
if( VERBOSE > 2 ) Print(s);
}
for( comp = 0; comp < t->ncomp; ++comp )
error[comp] = sqrt(error[comp]);
df += t->nregions;
if( VERBOSE > 2 ) {
char s[16 + 128*NCOMP], *p = s;
p += sprintf(p, "\nTotals:");
for( comp = 0; comp < t->ncomp; ++comp )
p += sprintf(p, "\n[" COUNT "] "
REAL " +- " REAL " \tchisq " REAL " (" COUNT " df)",
comp + 1, integral[comp], error[comp], prob[comp], df);
Print(s);
}
for( comp = 0; comp < t->ncomp; ++comp )
prob[comp] = ChiSquare(prob[comp], df);
weight = 1;
}
#ifdef MLVERSION
if( REGIONS ) {
MLPutFunction(stdlink, "List", 2);
MLPutFunction(stdlink, "List", t->nregions);
for( iregion = 0; iregion < t->nregions; ++iregion ) {
Region *region = RegionPtr(iregion);
cBounds *b = region->bounds;
real lower[NDIM], upper[NDIM];
for( dim = 0; dim < t->ndim; ++dim ) {
lower[dim] = b[dim].lower;
upper[dim] = b[dim].upper;
}
MLPutFunction(stdlink, "Cuba`Divonne`region", 4);
MLPutRealList(stdlink, lower, t->ndim);
MLPutRealList(stdlink, upper, t->ndim);
MLPutFunction(stdlink, "List", t->ncomp);
for( comp = 0; comp < t->ncomp; ++comp ) {
cResult *r = ®ion->result[comp];
real res[] = {r->avg, r->spread*weight, r->chisq};
MLPutRealList(stdlink, res, Elements(res));
}
MLPutInteger(stdlink, region->depth); /* misused for df */
}
}
#endif
abort:
SamplesFree(&t->samples[2]);
SamplesFree(&t->samples[1]);
SamplesFree(&t->samples[0]);
RuleFree(&t->rule13);
RuleFree(&t->rule11);
RuleFree(&t->rule9);
RuleFree(&t->rule7);
free(t->voidregion);
return fail;
}
/**
* Return mask ? a : b;
*
* mask is a bitwise mask, composed of 0 or ~0 for each element. Any other value
* will yield unpredictable results.
*/
LLVMValueRef
lp_build_select(struct lp_build_context *bld,
LLVMValueRef mask,
LLVMValueRef a,
LLVMValueRef b)
{
LLVMBuilderRef builder = bld->gallivm->builder;
LLVMContextRef lc = bld->gallivm->context;
struct lp_type type = bld->type;
LLVMValueRef res;
assert(lp_check_value(type, a));
assert(lp_check_value(type, b));
if(a == b)
return a;
if (type.length == 1) {
mask = LLVMBuildTrunc(builder, mask, LLVMInt1TypeInContext(lc), "");
res = LLVMBuildSelect(builder, mask, a, b, "");
}
else if (0) {
/* Generate a vector select.
*
* XXX: Using vector selects would avoid emitting intrinsics, but they aren't
* properly supported yet.
*
* LLVM 3.0 includes experimental support provided the -promote-elements
* options is passed to LLVM's command line (e.g., via
* llvm::cl::ParseCommandLineOptions), but resulting code quality is much
* worse, probably because some optimization passes don't know how to
* handle vector selects.
*
* See also:
* - http://lists.cs.uiuc.edu/pipermail/llvmdev/2011-October/043659.html
*/
/* Convert the mask to a vector of booleans.
* XXX: There are two ways to do this. Decide what's best.
*/
if (1) {
LLVMTypeRef bool_vec_type = LLVMVectorType(LLVMInt1TypeInContext(lc), type.length);
mask = LLVMBuildTrunc(builder, mask, bool_vec_type, "");
} else {
mask = LLVMBuildICmp(builder, LLVMIntNE, mask, LLVMConstNull(bld->int_vec_type), "");
}
res = LLVMBuildSelect(builder, mask, a, b, "");
}
else if (((util_cpu_caps.has_sse4_1 &&
type.width * type.length == 128) ||
(util_cpu_caps.has_avx &&
type.width * type.length == 256 && type.width >= 32)) &&
!LLVMIsConstant(a) &&
!LLVMIsConstant(b) &&
!LLVMIsConstant(mask)) {
const char *intrinsic;
LLVMTypeRef arg_type;
LLVMValueRef args[3];
/*
* There's only float blend in AVX but can just cast i32/i64
* to float.
*/
if (type.width * type.length == 256) {
if (type.width == 64) {
intrinsic = "llvm.x86.avx.blendv.pd.256";
arg_type = LLVMVectorType(LLVMDoubleTypeInContext(lc), 4);
}
else {
intrinsic = "llvm.x86.avx.blendv.ps.256";
arg_type = LLVMVectorType(LLVMFloatTypeInContext(lc), 8);
}
}
else if (type.floating &&
type.width == 64) {
intrinsic = "llvm.x86.sse41.blendvpd";
arg_type = LLVMVectorType(LLVMDoubleTypeInContext(lc), 2);
} else if (type.floating &&
type.width == 32) {
intrinsic = "llvm.x86.sse41.blendvps";
arg_type = LLVMVectorType(LLVMFloatTypeInContext(lc), 4);
} else {
intrinsic = "llvm.x86.sse41.pblendvb";
arg_type = LLVMVectorType(LLVMInt8TypeInContext(lc), 16);
}
if (arg_type != bld->int_vec_type) {
mask = LLVMBuildBitCast(builder, mask, arg_type, "");
}
if (arg_type != bld->vec_type) {
a = LLVMBuildBitCast(builder, a, arg_type, "");
b = LLVMBuildBitCast(builder, b, arg_type, "");
}
args[0] = b;
args[1] = a;
args[2] = mask;
res = lp_build_intrinsic(builder, intrinsic,
arg_type, args, Elements(args));
if (arg_type != bld->vec_type) {
res = LLVMBuildBitCast(builder, res, bld->vec_type, "");
}
}
else {
res = lp_build_select_bitwise(bld, mask, a, b);
}
return res;
}
/**
* Make a list of per-vertex functions to call for each glArrayElement call.
* These functions access the array data (i.e. glVertex, glColor, glNormal,
* etc).
* Note: this may be called during display list construction.
*/
static void _ae_update_state( struct gl_context *ctx )
{
AEcontext *actx = AE_CONTEXT(ctx);
AEarray *aa = actx->arrays;
AEattrib *at = actx->attribs;
GLuint i;
struct gl_array_object *arrayObj = ctx->Array.ArrayObj;
actx->nr_vbos = 0;
/* conventional vertex arrays */
if (arrayObj->Index.Enabled) {
aa->array = &arrayObj->Index;
aa->offset = IndexFuncs[TYPE_IDX(aa->array->Type)];
check_vbo(actx, aa->array->BufferObj);
aa++;
}
if (arrayObj->EdgeFlag.Enabled) {
aa->array = &arrayObj->EdgeFlag;
aa->offset = _gloffset_EdgeFlagv;
check_vbo(actx, aa->array->BufferObj);
aa++;
}
if (arrayObj->Normal.Enabled) {
aa->array = &arrayObj->Normal;
aa->offset = NormalFuncs[TYPE_IDX(aa->array->Type)];
check_vbo(actx, aa->array->BufferObj);
aa++;
}
if (arrayObj->Color.Enabled) {
aa->array = &arrayObj->Color;
aa->offset = ColorFuncs[aa->array->Size-3][TYPE_IDX(aa->array->Type)];
check_vbo(actx, aa->array->BufferObj);
aa++;
}
if (arrayObj->SecondaryColor.Enabled) {
aa->array = &arrayObj->SecondaryColor;
aa->offset = SecondaryColorFuncs[TYPE_IDX(aa->array->Type)];
check_vbo(actx, aa->array->BufferObj);
aa++;
}
if (arrayObj->FogCoord.Enabled) {
aa->array = &arrayObj->FogCoord;
aa->offset = FogCoordFuncs[TYPE_IDX(aa->array->Type)];
check_vbo(actx, aa->array->BufferObj);
aa++;
}
for (i = 0; i < ctx->Const.MaxTextureCoordUnits; i++) {
struct gl_client_array *attribArray = &arrayObj->TexCoord[i];
if (attribArray->Enabled) {
/* NOTE: we use generic glVertexAttribNV functions here.
* If we ever remove GL_NV_vertex_program this will have to change.
*/
at->array = attribArray;
ASSERT(!at->array->Normalized);
at->func = AttribFuncsNV[at->array->Normalized]
[at->array->Size-1]
[TYPE_IDX(at->array->Type)];
at->index = VERT_ATTRIB_TEX0 + i;
check_vbo(actx, at->array->BufferObj);
at++;
}
}
/* generic vertex attribute arrays */
for (i = 1; i < Elements(arrayObj->VertexAttrib); i++) { /* skip zero! */
struct gl_client_array *attribArray = &arrayObj->VertexAttrib[i];
if (attribArray->Enabled) {
at->array = attribArray;
/* Note: we can't grab the _glapi_Dispatch->VertexAttrib1fvNV
* function pointer here (for float arrays) since the pointer may
* change from one execution of _ae_ArrayElement() to
* the next. Doing so caused UT to break.
*/
if (ctx->VertexProgram._Enabled
&& ctx->VertexProgram.Current->IsNVProgram) {
at->func = AttribFuncsNV[at->array->Normalized]
[at->array->Size-1]
[TYPE_IDX(at->array->Type)];
}
else {
GLint intOrNorm;
if (at->array->Integer)
intOrNorm = 2;
else if (at->array->Normalized)
intOrNorm = 1;
else
intOrNorm = 0;
at->func = AttribFuncsARB[intOrNorm]
[at->array->Size-1]
[TYPE_IDX(at->array->Type)];
}
at->index = i;
check_vbo(actx, at->array->BufferObj);
at++;
}
}
/* finally, vertex position */
if (arrayObj->VertexAttrib[0].Enabled) {
/* Use glVertex(v) instead of glVertexAttrib(0, v) to be sure it's
* issued as the last (provoking) attribute).
*/
aa->array = &arrayObj->VertexAttrib[0];
assert(aa->array->Size >= 2); /* XXX fix someday? */
aa->offset = VertexFuncs[aa->array->Size-2][TYPE_IDX(aa->array->Type)];
check_vbo(actx, aa->array->BufferObj);
aa++;
}
else if (arrayObj->Vertex.Enabled) {
aa->array = &arrayObj->Vertex;
aa->offset = VertexFuncs[aa->array->Size-2][TYPE_IDX(aa->array->Type)];
check_vbo(actx, aa->array->BufferObj);
aa++;
}
check_vbo(actx, ctx->Array.ElementArrayBufferObj);
ASSERT(at - actx->attribs <= VERT_ATTRIB_MAX);
ASSERT(aa - actx->arrays < 32);
at->func = NULL; /* terminate the list */
aa->offset = -1; /* terminate the list */
actx->NewState = 0;
}
[PIPE_PRIM_POINTS] = V_028A6C_OUTPRIM_TYPE_POINTLIST,
[PIPE_PRIM_LINES] = V_028A6C_OUTPRIM_TYPE_LINESTRIP,
[PIPE_PRIM_LINE_LOOP] = V_028A6C_OUTPRIM_TYPE_LINESTRIP,
[PIPE_PRIM_LINE_STRIP] = V_028A6C_OUTPRIM_TYPE_LINESTRIP,
[PIPE_PRIM_TRIANGLES] = V_028A6C_OUTPRIM_TYPE_TRISTRIP,
[PIPE_PRIM_TRIANGLE_STRIP] = V_028A6C_OUTPRIM_TYPE_TRISTRIP,
[PIPE_PRIM_TRIANGLE_FAN] = V_028A6C_OUTPRIM_TYPE_TRISTRIP,
[PIPE_PRIM_QUADS] = V_028A6C_OUTPRIM_TYPE_TRISTRIP,
[PIPE_PRIM_QUAD_STRIP] = V_028A6C_OUTPRIM_TYPE_TRISTRIP,
[PIPE_PRIM_POLYGON] = V_028A6C_OUTPRIM_TYPE_TRISTRIP,
[PIPE_PRIM_LINES_ADJACENCY] = V_028A6C_OUTPRIM_TYPE_LINESTRIP,
[PIPE_PRIM_LINE_STRIP_ADJACENCY] = V_028A6C_OUTPRIM_TYPE_LINESTRIP,
[PIPE_PRIM_TRIANGLES_ADJACENCY] = V_028A6C_OUTPRIM_TYPE_TRISTRIP,
[PIPE_PRIM_TRIANGLE_STRIP_ADJACENCY] = V_028A6C_OUTPRIM_TYPE_TRISTRIP
};
assert(mode < Elements(prim_conv));
return prim_conv[mode];
}
static bool si_update_draw_info_state(struct r600_context *rctx,
const struct pipe_draw_info *info)
{
struct si_pm4_state *pm4 = si_pm4_alloc_state(rctx);
struct si_shader *vs = &rctx->vs_shader->current->shader;
unsigned prim = si_conv_pipe_prim(info->mode);
unsigned gs_out_prim = r600_conv_prim_to_gs_out(info->mode);
unsigned ls_mask = 0;
if (pm4 == NULL)
return false;
/**
* Return ARB_v/f_prog-style output attrib string.
*/
static const char *
arb_output_attrib_string(GLint index, GLenum progType)
{
/*
* These strings should match the VARYING_SLOT_x and FRAG_RESULT_x tokens.
*/
static const char *const vertResults[] = {
"result.position",
"result.color.primary",
"result.color.secondary",
"result.fogcoord",
"result.texcoord[0]",
"result.texcoord[1]",
"result.texcoord[2]",
"result.texcoord[3]",
"result.texcoord[4]",
"result.texcoord[5]",
"result.texcoord[6]",
"result.texcoord[7]",
"result.pointsize", /* VARYING_SLOT_PSIZ */
"result.(thirteen)", /* VARYING_SLOT_BFC0 */
"result.(fourteen)", /* VARYING_SLOT_BFC1 */
"result.(fifteen)", /* VARYING_SLOT_EDGE */
"result.(sixteen)", /* VARYING_SLOT_CLIP_VERTEX */
"result.(seventeen)", /* VARYING_SLOT_CLIP_DIST0 */
"result.(eighteen)", /* VARYING_SLOT_CLIP_DIST1 */
"result.(nineteen)", /* VARYING_SLOT_PRIMITIVE_ID */
"result.(twenty)", /* VARYING_SLOT_LAYER */
"result.(twenty-one)", /* VARYING_SLOT_FACE */
"result.(twenty-two)", /* VARYING_SLOT_PNTC */
"result.varying[0]",
"result.varying[1]",
"result.varying[2]",
"result.varying[3]",
"result.varying[4]",
"result.varying[5]",
"result.varying[6]",
"result.varying[7]",
"result.varying[8]",
"result.varying[9]",
"result.varying[10]",
"result.varying[11]",
"result.varying[12]",
"result.varying[13]",
"result.varying[14]",
"result.varying[15]",
"result.varying[16]",
"result.varying[17]",
"result.varying[18]",
"result.varying[19]",
"result.varying[20]",
"result.varying[21]",
"result.varying[22]",
"result.varying[23]",
"result.varying[24]",
"result.varying[25]",
"result.varying[26]",
"result.varying[27]",
"result.varying[28]",
"result.varying[29]",
"result.varying[30]",
"result.varying[31]", /* MAX_VARYING = 32 */
};
static const char *const fragResults[] = {
"result.depth", /* FRAG_RESULT_DEPTH */
"result.(one)", /* FRAG_RESULT_STENCIL */
"result.color", /* FRAG_RESULT_COLOR */
"result.samplemask", /* FRAG_RESULT_SAMPLE_MASK */
"result.color[0]", /* FRAG_RESULT_DATA0 (named for GLSL's gl_FragData) */
"result.color[1]",
"result.color[2]",
"result.color[3]",
"result.color[4]",
"result.color[5]",
"result.color[6]",
"result.color[7]" /* MAX_DRAW_BUFFERS = 8 */
};
/* sanity checks */
STATIC_ASSERT(Elements(vertResults) == VARYING_SLOT_MAX);
STATIC_ASSERT(Elements(fragResults) == FRAG_RESULT_MAX);
assert(strcmp(vertResults[VARYING_SLOT_POS], "result.position") == 0);
assert(strcmp(vertResults[VARYING_SLOT_VAR0], "result.varying[0]") == 0);
assert(strcmp(fragResults[FRAG_RESULT_DATA0], "result.color[0]") == 0);
if (progType == GL_VERTEX_PROGRAM_ARB) {
assert(index < Elements(vertResults));
return vertResults[index];
}
else {
assert(progType == GL_FRAGMENT_PROGRAM_ARB);
assert(index < Elements(fragResults));
return fragResults[index];
}
}
static void
generate_110_uniforms(exec_list *instructions,
struct _mesa_glsl_parse_state *state,
bool add_deprecated)
{
glsl_symbol_table *const symtab = state->symbols;
if (add_deprecated) {
for (unsigned i = 0
; i < Elements(builtin_110_deprecated_uniforms)
; i++) {
add_builtin_variable(instructions, symtab,
& builtin_110_deprecated_uniforms[i]);
}
}
if (add_deprecated) {
add_builtin_constant(instructions, symtab, "gl_MaxLights",
state->Const.MaxLights);
add_builtin_constant(instructions, symtab, "gl_MaxClipPlanes",
state->Const.MaxClipPlanes);
add_builtin_constant(instructions, symtab, "gl_MaxTextureUnits",
state->Const.MaxTextureUnits);
add_builtin_constant(instructions, symtab, "gl_MaxTextureCoords",
state->Const.MaxTextureCoords);
}
add_builtin_constant(instructions, symtab, "gl_MaxVertexAttribs",
state->Const.MaxVertexAttribs);
add_builtin_constant(instructions, symtab, "gl_MaxVertexUniformComponents",
state->Const.MaxVertexUniformComponents);
add_builtin_constant(instructions, symtab, "gl_MaxVaryingFloats",
state->Const.MaxVaryingFloats);
add_builtin_constant(instructions, symtab, "gl_MaxVertexTextureImageUnits",
state->Const.MaxVertexTextureImageUnits);
add_builtin_constant(instructions, symtab, "gl_MaxCombinedTextureImageUnits",
state->Const.MaxCombinedTextureImageUnits);
add_builtin_constant(instructions, symtab, "gl_MaxTextureImageUnits",
state->Const.MaxTextureImageUnits);
add_builtin_constant(instructions, symtab, "gl_MaxFragmentUniformComponents",
state->Const.MaxFragmentUniformComponents);
if (add_deprecated) {
const glsl_type *const mat4_array_type =
glsl_type::get_array_instance(glsl_type::mat4_type,
state->Const.MaxTextureCoords);
add_uniform(instructions, symtab, "gl_TextureMatrix", mat4_array_type);
add_uniform(instructions, symtab, "gl_TextureMatrixInverse", mat4_array_type);
add_uniform(instructions, symtab, "gl_TextureMatrixTranspose", mat4_array_type);
add_uniform(instructions, symtab, "gl_TextureMatrixInverseTranspose", mat4_array_type);
}
add_uniform(instructions, symtab, "gl_DepthRange",
symtab->get_type("gl_DepthRangeParameters"));
if (add_deprecated) {
add_uniform(instructions, symtab, "gl_ClipPlane",
glsl_type::get_array_instance(glsl_type::vec4_type,
state->Const.MaxClipPlanes));
add_uniform(instructions, symtab, "gl_Point",
symtab->get_type("gl_PointParameters"));
const glsl_type *const material_parameters_type =
symtab->get_type("gl_MaterialParameters");
add_uniform(instructions, symtab, "gl_FrontMaterial", material_parameters_type);
add_uniform(instructions, symtab, "gl_BackMaterial", material_parameters_type);
const glsl_type *const light_source_array_type =
glsl_type::get_array_instance(symtab->get_type("gl_LightSourceParameters"), state->Const.MaxLights);
add_uniform(instructions, symtab, "gl_LightSource", light_source_array_type);
const glsl_type *const light_model_products_type =
symtab->get_type("gl_LightModelProducts");
add_uniform(instructions, symtab, "gl_FrontLightModelProduct",
light_model_products_type);
add_uniform(instructions, symtab, "gl_BackLightModelProduct",
light_model_products_type);
const glsl_type *const light_products_type =
glsl_type::get_array_instance(symtab->get_type("gl_LightProducts"),
state->Const.MaxLights);
add_uniform(instructions, symtab, "gl_FrontLightProduct", light_products_type);
add_uniform(instructions, symtab, "gl_BackLightProduct", light_products_type);
add_uniform(instructions, symtab, "gl_TextureEnvColor",
glsl_type::get_array_instance(glsl_type::vec4_type,
state->Const.MaxTextureUnits));
const glsl_type *const texcoords_vec4 =
glsl_type::get_array_instance(glsl_type::vec4_type,
state->Const.MaxTextureCoords);
add_uniform(instructions, symtab, "gl_EyePlaneS", texcoords_vec4);
add_uniform(instructions, symtab, "gl_EyePlaneT", texcoords_vec4);
add_uniform(instructions, symtab, "gl_EyePlaneR", texcoords_vec4);
add_uniform(instructions, symtab, "gl_EyePlaneQ", texcoords_vec4);
add_uniform(instructions, symtab, "gl_ObjectPlaneS", texcoords_vec4);
add_uniform(instructions, symtab, "gl_ObjectPlaneT", texcoords_vec4);
add_uniform(instructions, symtab, "gl_ObjectPlaneR", texcoords_vec4);
add_uniform(instructions, symtab, "gl_ObjectPlaneQ", texcoords_vec4);
add_uniform(instructions, symtab, "gl_Fog",
symtab->get_type("gl_FogParameters"));
}
/* Mesa-internal current attrib state */
const glsl_type *const vert_attribs =
glsl_type::get_array_instance(glsl_type::vec4_type, VERT_ATTRIB_MAX);
add_uniform(instructions, symtab, "gl_CurrentAttribVertMESA", vert_attribs);
const glsl_type *const frag_attribs =
glsl_type::get_array_instance(glsl_type::vec4_type, FRAG_ATTRIB_MAX);
add_uniform(instructions, symtab, "gl_CurrentAttribFragMESA", frag_attribs);
}
static void
gen6_emit_draw_dynamic_samplers(struct ilo_render *r,
const struct ilo_state_vector *vec,
int shader_type,
struct ilo_render_draw_session *session)
{
const struct ilo_sampler_cso * const *samplers =
vec->sampler[shader_type].cso;
const struct pipe_sampler_view * const *views =
(const struct pipe_sampler_view **) vec->view[shader_type].states;
uint32_t *sampler_state, *border_color_state;
int sampler_count;
bool emit_border_color = false;
bool skip = false;
ILO_DEV_ASSERT(r->dev, 6, 7.5);
/* SAMPLER_BORDER_COLOR_STATE and SAMPLER_STATE */
switch (shader_type) {
case PIPE_SHADER_VERTEX:
if (DIRTY(VS) || DIRTY(SAMPLER_VS) || DIRTY(VIEW_VS)) {
sampler_state = &r->state.vs.SAMPLER_STATE;
border_color_state = r->state.vs.SAMPLER_BORDER_COLOR_STATE;
if (DIRTY(VS) || DIRTY(SAMPLER_VS))
emit_border_color = true;
sampler_count = (vec->vs) ? ilo_shader_get_kernel_param(vec->vs,
ILO_KERNEL_SAMPLER_COUNT) : 0;
session->sampler_vs_changed = true;
} else {
skip = true;
}
break;
case PIPE_SHADER_FRAGMENT:
if (DIRTY(FS) || DIRTY(SAMPLER_FS) || DIRTY(VIEW_FS)) {
sampler_state = &r->state.wm.SAMPLER_STATE;
border_color_state = r->state.wm.SAMPLER_BORDER_COLOR_STATE;
if (DIRTY(VS) || DIRTY(SAMPLER_FS))
emit_border_color = true;
sampler_count = (vec->fs) ? ilo_shader_get_kernel_param(vec->fs,
ILO_KERNEL_SAMPLER_COUNT) : 0;
session->sampler_fs_changed = true;
} else {
skip = true;
}
break;
default:
skip = true;
break;
}
if (skip)
return;
assert(sampler_count <= Elements(vec->view[shader_type].states) &&
sampler_count <= Elements(vec->sampler[shader_type].cso));
if (emit_border_color) {
int i;
for (i = 0; i < sampler_count; i++) {
border_color_state[i] = (samplers[i]) ?
gen6_SAMPLER_BORDER_COLOR_STATE(r->builder, samplers[i]) : 0;
}
}
*sampler_state = gen6_SAMPLER_STATE(r->builder,
samplers, views, border_color_state, sampler_count);
}
int main(int argc, char **argv)
{
int ret = 0, n = 1;
const char *filename;
struct tgsi_token toks[65536];
struct tgsi_parse_context parse;
struct ir3_shader_variant v;
struct ir3_shader_key key = {};
const char *info;
void *ptr;
size_t size;
fd_mesa_debug |= FD_DBG_DISASM;
/* cmdline args which impact shader variant get spit out in a
* comment on the first line.. a quick/dirty way to preserve
* that info so when ir3test recompiles the shader with a new
* compiler version, we use the same shader-key settings:
*/
debug_printf("; options:");
while (n < argc) {
if (!strcmp(argv[n], "--verbose")) {
fd_mesa_debug |= FD_DBG_OPTDUMP | FD_DBG_MSGS | FD_DBG_OPTMSGS;
n++;
continue;
}
if (!strcmp(argv[n], "--binning-pass")) {
debug_printf(" %s", argv[n]);
key.binning_pass = true;
n++;
continue;
}
if (!strcmp(argv[n], "--color-two-side")) {
debug_printf(" %s", argv[n]);
key.color_two_side = true;
n++;
continue;
}
if (!strcmp(argv[n], "--half-precision")) {
debug_printf(" %s", argv[n]);
key.half_precision = true;
n++;
continue;
}
if (!strcmp(argv[n], "--alpha")) {
debug_printf(" %s", argv[n]);
key.alpha = true;
n++;
continue;
}
if (!strcmp(argv[n], "--saturate-s")) {
debug_printf(" %s %s", argv[n], argv[n+1]);
key.vsaturate_s = key.fsaturate_s = strtol(argv[n+1], NULL, 0);
n += 2;
continue;
}
if (!strcmp(argv[n], "--saturate-t")) {
debug_printf(" %s %s", argv[n], argv[n+1]);
key.vsaturate_t = key.fsaturate_t = strtol(argv[n+1], NULL, 0);
n += 2;
continue;
}
if (!strcmp(argv[n], "--saturate-r")) {
debug_printf(" %s %s", argv[n], argv[n+1]);
key.vsaturate_r = key.fsaturate_r = strtol(argv[n+1], NULL, 0);
n += 2;
continue;
}
if (!strcmp(argv[n], "--nocp")) {
fd_mesa_debug |= FD_DBG_NOCP;
n++;
continue;
}
if (!strcmp(argv[n], "--help")) {
print_usage();
return 0;
}
break;
}
debug_printf("\n");
filename = argv[n];
memset(&v, 0, sizeof(v));
v.key = key;
ret = read_file(filename, &ptr, &size);
if (ret) {
print_usage();
return ret;
}
if (!tgsi_text_translate(ptr, toks, Elements(toks)))
errx(1, "could not parse `%s'", filename);
tgsi_parse_init(&parse, toks);
switch (parse.FullHeader.Processor.Processor) {
case TGSI_PROCESSOR_FRAGMENT:
v.type = SHADER_FRAGMENT;
break;
case TGSI_PROCESSOR_VERTEX:
v.type = SHADER_VERTEX;
break;
case TGSI_PROCESSOR_COMPUTE:
v.type = SHADER_COMPUTE;
break;
}
if (!(fd_mesa_debug & FD_DBG_NOOPT)) {
/* with new compiler: */
info = "new compiler";
ret = ir3_compile_shader(&v, toks, key, true);
if (ret) {
reset_variant(&v, "new compiler failed, trying without copy propagation!");
info = "new compiler (no copy propagation)";
ret = ir3_compile_shader(&v, toks, key, false);
if (ret)
reset_variant(&v, "new compiler failed, trying fallback!\n");
}
}
if (ret) {
info = "old compiler";
ret = ir3_compile_shader_old(&v, toks, key);
}
if (ret) {
fprintf(stderr, "old compiler failed!\n");
return ret;
}
dump_info(&v, info);
}