* name and parameters.
*/
+
+/**
+ * Unfortunately, some versions of MinGW produce bad code if this file
+ * is compiled with -O2 or -O3. The resulting driver will crash in random
+ * places if the app uses GLSL.
+ * The work-around is to disable optimizations for just this file. Luckily,
+ * this code is basically just executed once.
+ *
+ * MinGW 4.6.3 (in Ubuntu 13.10) does not have this bug.
+ * MinGW 5.3.1 (in Ubuntu 16.04) definitely has this bug.
+ * MinGW 6.2.0 (in Ubuntu 16.10) definitely has this bug.
+ * MinGW x.y.z - don't know. Assume versions after 4.6.x are buggy
+ */
+
+#if defined(__MINGW32__) && ((__GNUC__ * 100) + __GNUC_MINOR >= 407)
+#warning "disabling optimizations for this file to work around compiler bug"
+#pragma GCC optimize("O1")
+#endif
+
+
#include <stdarg.h>
#include <stdio.h>
-#include "main/core.h" /* for struct gl_shader */
+#include "main/mtypes.h"
#include "main/shaderobj.h"
#include "ir_builder.h"
#include "glsl_parser_extras.h"
#include "program/prog_instruction.h"
#include <math.h>
+#include "builtin_functions.h"
+#include "util/hash_table.h"
#define M_PIf ((float) M_PI)
#define M_PI_2f ((float) M_PI_2)
compatibility_vs_only(const _mesa_glsl_parse_state *state)
{
return state->stage == MESA_SHADER_VERTEX &&
- state->language_version <= 130 &&
+ (state->compat_shader || state->ARB_compatibility_enable) &&
!state->es_shader;
}
}
static bool
-v130_fs_only(const _mesa_glsl_parse_state *state)
+v130_desktop(const _mesa_glsl_parse_state *state)
{
- return state->is_version(130, 300) &&
- state->stage == MESA_SHADER_FRAGMENT;
+ return state->is_version(130, 0);
}
static bool
-v140(const _mesa_glsl_parse_state *state)
+v460_desktop(const _mesa_glsl_parse_state *state)
+{
+ return state->is_version(460, 0);
+}
+
+static bool
+v130_fs_only(const _mesa_glsl_parse_state *state)
{
- return state->is_version(140, 0);
+ return state->is_version(130, 300) &&
+ state->stage == MESA_SHADER_FRAGMENT;
}
static bool
state->stage == MESA_SHADER_FRAGMENT;
}
-static bool
-es31(const _mesa_glsl_parse_state *state)
-{
- return state->is_version(0, 310);
-}
-
static bool
texture_rectangle(const _mesa_glsl_parse_state *state)
{
return state->OES_EGL_image_external_enable;
}
+static bool
+texture_external_es3(const _mesa_glsl_parse_state *state)
+{
+ return state->OES_EGL_image_external_essl3_enable &&
+ state->es_shader &&
+ state->is_version(0, 300);
+}
+
/** True if texturing functions with explicit LOD are allowed. */
static bool
lod_exists_in_stage(const _mesa_glsl_parse_state *state)
return !state->es_shader && lod_exists_in_stage(state);
}
+static bool
+texture_buffer(const _mesa_glsl_parse_state *state)
+{
+ return state->is_version(140, 320) ||
+ state->EXT_texture_buffer_enable ||
+ state->OES_texture_buffer_enable;
+}
+
static bool
shader_texture_lod(const _mesa_glsl_parse_state *state)
{
shader_integer_mix(const _mesa_glsl_parse_state *state)
{
return state->is_version(450, 310) ||
+ state->ARB_ES3_1_compatibility_enable ||
(v130(state) && state->EXT_shader_integer_mix_enable);
}
return state->is_version(400, 0) || state->ARB_gpu_shader5_enable;
}
+static bool
+gpu_shader5_es(const _mesa_glsl_parse_state *state)
+{
+ return state->is_version(400, 320) ||
+ state->ARB_gpu_shader5_enable ||
+ state->EXT_gpu_shader5_enable ||
+ state->OES_gpu_shader5_enable;
+}
+
+static bool
+gpu_shader5_or_OES_texture_cube_map_array(const _mesa_glsl_parse_state *state)
+{
+ return state->is_version(400, 320) ||
+ state->ARB_gpu_shader5_enable ||
+ state->EXT_texture_cube_map_array_enable ||
+ state->OES_texture_cube_map_array_enable;
+}
+
+static bool
+es31_not_gs5(const _mesa_glsl_parse_state *state)
+{
+ return state->is_version(0, 310) && !gpu_shader5_es(state);
+}
+
static bool
gpu_shader5_or_es31(const _mesa_glsl_parse_state *state)
{
}
static bool
-fs_gpu_shader5(const _mesa_glsl_parse_state *state)
+gpu_shader5_or_es31_or_integer_functions(const _mesa_glsl_parse_state *state)
+{
+ return gpu_shader5_or_es31(state) ||
+ state->MESA_shader_integer_functions_enable;
+}
+
+static bool
+fs_interpolate_at(const _mesa_glsl_parse_state *state)
{
return state->stage == MESA_SHADER_FRAGMENT &&
- (state->is_version(400, 0) || state->ARB_gpu_shader5_enable);
+ (state->is_version(400, 320) ||
+ state->ARB_gpu_shader5_enable ||
+ state->OES_shader_multisample_interpolation_enable);
}
fs_texture_cube_map_array(const _mesa_glsl_parse_state *state)
{
return state->stage == MESA_SHADER_FRAGMENT &&
- (state->is_version(400, 0) ||
- state->ARB_texture_cube_map_array_enable);
+ state->has_texture_cube_map_array();
}
static bool
texture_cube_map_array(const _mesa_glsl_parse_state *state)
{
- return state->is_version(400, 0) ||
- state->ARB_texture_cube_map_array_enable;
+ return state->has_texture_cube_map_array();
}
static bool
}
static bool
-texture_gather(const _mesa_glsl_parse_state *state)
+texture_gather_cube_map_array(const _mesa_glsl_parse_state *state)
{
- return state->is_version(400, 0) ||
+ return state->is_version(400, 320) ||
state->ARB_texture_gather_enable ||
- state->ARB_gpu_shader5_enable;
+ state->ARB_gpu_shader5_enable ||
+ state->EXT_texture_cube_map_array_enable ||
+ state->OES_texture_cube_map_array_enable;
}
static bool
static bool
texture_gather_only_or_es31(const _mesa_glsl_parse_state *state)
{
- return !state->is_version(400, 0) &&
+ return !state->is_version(400, 320) &&
!state->ARB_gpu_shader5_enable &&
+ !state->EXT_gpu_shader5_enable &&
+ !state->OES_gpu_shader5_enable &&
(state->ARB_texture_gather_enable ||
state->is_version(0, 310));
}
{
return state->stage == MESA_SHADER_FRAGMENT &&
(state->is_version(110, 300) ||
- state->OES_standard_derivatives_enable);
+ state->OES_standard_derivatives_enable ||
+ state->ctx->Const.AllowGLSLRelaxedES);
}
static bool
return state->has_atomic_counters();
}
+static bool
+shader_atomic_counter_ops(const _mesa_glsl_parse_state *state)
+{
+ return state->ARB_shader_atomic_counter_ops_enable;
+}
+
+static bool
+shader_atomic_counter_ops_or_v460_desktop(const _mesa_glsl_parse_state *state)
+{
+ return state->ARB_shader_atomic_counter_ops_enable || v460_desktop(state);
+}
+
+static bool
+shader_ballot(const _mesa_glsl_parse_state *state)
+{
+ return state->ARB_shader_ballot_enable;
+}
+
+static bool
+supports_arb_fragment_shader_interlock(const _mesa_glsl_parse_state *state)
+{
+ return state->ARB_fragment_shader_interlock_enable;
+}
+
+static bool
+supports_nv_fragment_shader_interlock(const _mesa_glsl_parse_state *state)
+{
+ return state->NV_fragment_shader_interlock_enable;
+}
+
+static bool
+supports_intel_fragment_shader_ordering(const _mesa_glsl_parse_state *state)
+{
+ return state->INTEL_fragment_shader_ordering_enable;
+}
+
static bool
shader_clock(const _mesa_glsl_parse_state *state)
{
return state->ARB_shader_clock_enable;
}
+static bool
+shader_clock_int64(const _mesa_glsl_parse_state *state)
+{
+ return state->ARB_shader_clock_enable &&
+ (state->ARB_gpu_shader_int64_enable ||
+ state->AMD_gpu_shader_int64_enable);
+}
+
static bool
shader_storage_buffer_object(const _mesa_glsl_parse_state *state)
{
static bool
shader_image_atomic(const _mesa_glsl_parse_state *state)
{
- return (state->is_version(420, 0) ||
- state->ARB_shader_image_load_store_enable);
+ return (state->is_version(420, 320) ||
+ state->ARB_shader_image_load_store_enable ||
+ state->OES_shader_image_atomic_enable);
+}
+
+static bool
+shader_image_atomic_exchange_float(const _mesa_glsl_parse_state *state)
+{
+ return (state->is_version(450, 320) ||
+ state->ARB_ES3_1_compatibility_enable ||
+ state->OES_shader_image_atomic_enable ||
+ state->NV_shader_atomic_float_enable);
+}
+
+static bool
+shader_image_atomic_add_float(const _mesa_glsl_parse_state *state)
+{
+ return state->NV_shader_atomic_float_enable;
}
static bool
return state->has_double();
}
+static bool
+int64(const _mesa_glsl_parse_state *state)
+{
+ return state->has_int64();
+}
+
+static bool
+int64_fp64(const _mesa_glsl_parse_state *state)
+{
+ return state->has_int64() && state->has_double();
+}
+
static bool
compute_shader(const _mesa_glsl_parse_state *state)
{
return state->stage == MESA_SHADER_COMPUTE;
}
+static bool
+compute_shader_supported(const _mesa_glsl_parse_state *state)
+{
+ return state->has_compute_shader();
+}
+
static bool
buffer_atomics_supported(const _mesa_glsl_parse_state *state)
{
state->stage == MESA_SHADER_TESS_CTRL;
}
+static bool
+vote(const _mesa_glsl_parse_state *state)
+{
+ return state->ARB_shader_group_vote_enable;
+}
+
+static bool
+vote_or_v460_desktop(const _mesa_glsl_parse_state *state)
+{
+ return state->ARB_shader_group_vote_enable || v460_desktop(state);
+}
+
+static bool
+integer_functions_supported(const _mesa_glsl_parse_state *state)
+{
+ return state->extensions->MESA_shader_integer_functions;
+}
+
+static bool
+NV_shader_atomic_float_supported(const _mesa_glsl_parse_state *state)
+{
+ return state->extensions->NV_shader_atomic_float;
+}
+
+static bool
+shader_atomic_float_add(const _mesa_glsl_parse_state *state)
+{
+ return state->NV_shader_atomic_float_enable;
+}
+
+static bool
+shader_atomic_float_exchange(const _mesa_glsl_parse_state *state)
+{
+ return state->NV_shader_atomic_float_enable ||
+ state->INTEL_shader_atomic_float_minmax_enable;
+}
+
+static bool
+INTEL_shader_atomic_float_minmax_supported(const _mesa_glsl_parse_state *state)
+{
+ return state->extensions->INTEL_shader_atomic_float_minmax;
+}
+
+static bool
+shader_atomic_float_minmax(const _mesa_glsl_parse_state *state)
+{
+ return state->INTEL_shader_atomic_float_minmax_enable;
+}
/** @} */
/******************************************************************************/
private:
void *mem_ctx;
- /** Global variables used by built-in functions. */
- ir_variable *gl_ModelViewProjectionMatrix;
- ir_variable *gl_Vertex;
-
void create_shader();
void create_intrinsics();
void create_builtins();
ir_dereference_array *array_ref(ir_variable *var, int i);
ir_swizzle *matrix_elt(ir_variable *var, int col, int row);
- ir_expression *asin_expr(ir_variable *x);
+ ir_expression *asin_expr(ir_variable *x, float p0, float p1);
void do_atan(ir_factory &body, const glsl_type *type, ir_variable *res, operand y_over_x);
/**
unsigned num_arguments,
unsigned flags);
- enum image_function_flags {
- IMAGE_FUNCTION_EMIT_STUB = (1 << 0),
- IMAGE_FUNCTION_RETURNS_VOID = (1 << 1),
- IMAGE_FUNCTION_HAS_VECTOR_DATA_TYPE = (1 << 2),
- IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE = (1 << 3),
- IMAGE_FUNCTION_READ_ONLY = (1 << 4),
- IMAGE_FUNCTION_WRITE_ONLY = (1 << 5),
- IMAGE_FUNCTION_AVAIL_ATOMIC = (1 << 6),
- IMAGE_FUNCTION_MS_ONLY = (1 << 7),
- };
-
/**
* Create a new image built-in function for all known image types.
* \p flags is a bitfield of \c image_function_flags flags.
const char *intrinsic_name,
image_prototype_ctr prototype,
unsigned num_arguments,
- unsigned flags);
+ unsigned flags,
+ enum ir_intrinsic_id id);
/**
* Create new functions for all known image built-ins and types.
ir_expression_operation opcode,
const glsl_type *return_type,
const glsl_type *param0_type,
- const glsl_type *param1_type);
+ const glsl_type *param1_type,
+ bool swap_operands = false);
#define B0(X) ir_function_signature *_##X();
#define B1(X) ir_function_signature *_##X(const glsl_type *);
B1(floatBitsToUint)
B1(intBitsToFloat)
B1(uintBitsToFloat)
+
+ BA1(doubleBitsToInt64)
+ BA1(doubleBitsToUint64)
+ BA1(int64BitsToDouble)
+ BA1(uint64BitsToDouble)
+
ir_function_signature *_packUnorm2x16(builtin_available_predicate avail);
ir_function_signature *_packSnorm2x16(builtin_available_predicate avail);
ir_function_signature *_packUnorm4x8(builtin_available_predicate avail);
ir_function_signature *_unpackHalf2x16(builtin_available_predicate avail);
ir_function_signature *_packDouble2x32(builtin_available_predicate avail);
ir_function_signature *_unpackDouble2x32(builtin_available_predicate avail);
+ ir_function_signature *_packInt2x32(builtin_available_predicate avail);
+ ir_function_signature *_unpackInt2x32(builtin_available_predicate avail);
+ ir_function_signature *_packUint2x32(builtin_available_predicate avail);
+ ir_function_signature *_unpackUint2x32(builtin_available_predicate avail);
BA1(length)
BA1(distance);
B1(all);
B1(not);
BA2(textureSize);
- B1(textureSamples);
+ BA1(textureSamples);
/** Flags to _texture() */
#define TEX_PROJECT 1
const glsl_type *sampler_type,
const glsl_type *coord_type,
int flags = 0);
- B0(textureCubeArrayShadow);
+ BA1(textureCubeArrayShadow);
ir_function_signature *_texelFetch(builtin_available_predicate avail,
const glsl_type *return_type,
const glsl_type *sampler_type,
B0(barrier)
BA2(textureQueryLod);
- B1(textureQueryLevels);
+ BA1(textureQueryLevels);
BA2(textureSamplesIdentical);
B1(dFdx);
B1(dFdy);
B1(interpolateAtOffset)
B1(interpolateAtSample)
- ir_function_signature *_atomic_counter_intrinsic(builtin_available_predicate avail);
+ ir_function_signature *_atomic_counter_intrinsic(builtin_available_predicate avail,
+ enum ir_intrinsic_id id);
+ ir_function_signature *_atomic_counter_intrinsic1(builtin_available_predicate avail,
+ enum ir_intrinsic_id id);
+ ir_function_signature *_atomic_counter_intrinsic2(builtin_available_predicate avail,
+ enum ir_intrinsic_id id);
ir_function_signature *_atomic_counter_op(const char *intrinsic,
builtin_available_predicate avail);
+ ir_function_signature *_atomic_counter_op1(const char *intrinsic,
+ builtin_available_predicate avail);
+ ir_function_signature *_atomic_counter_op2(const char *intrinsic,
+ builtin_available_predicate avail);
ir_function_signature *_atomic_intrinsic2(builtin_available_predicate avail,
- const glsl_type *type);
+ const glsl_type *type,
+ enum ir_intrinsic_id id);
ir_function_signature *_atomic_op2(const char *intrinsic,
builtin_available_predicate avail,
const glsl_type *type);
ir_function_signature *_atomic_intrinsic3(builtin_available_predicate avail,
- const glsl_type *type);
+ const glsl_type *type,
+ enum ir_intrinsic_id id);
ir_function_signature *_atomic_op3(const char *intrinsic,
builtin_available_predicate avail,
const glsl_type *type);
const glsl_type *image_type,
const char *intrinsic_name,
unsigned num_arguments,
- unsigned flags);
+ unsigned flags,
+ enum ir_intrinsic_id id);
ir_function_signature *_memory_barrier_intrinsic(
- builtin_available_predicate avail);
+ builtin_available_predicate avail,
+ enum ir_intrinsic_id id);
ir_function_signature *_memory_barrier(const char *intrinsic_name,
builtin_available_predicate avail);
+ ir_function_signature *_ballot_intrinsic();
+ ir_function_signature *_ballot();
+ ir_function_signature *_read_first_invocation_intrinsic(const glsl_type *type);
+ ir_function_signature *_read_first_invocation(const glsl_type *type);
+ ir_function_signature *_read_invocation_intrinsic(const glsl_type *type);
+ ir_function_signature *_read_invocation(const glsl_type *type);
+
+
+ ir_function_signature *_invocation_interlock_intrinsic(
+ builtin_available_predicate avail,
+ enum ir_intrinsic_id id);
+ ir_function_signature *_invocation_interlock(
+ const char *intrinsic_name,
+ builtin_available_predicate avail);
+
ir_function_signature *_shader_clock_intrinsic(builtin_available_predicate avail,
const glsl_type *type);
ir_function_signature *_shader_clock(builtin_available_predicate avail,
const glsl_type *type);
+ ir_function_signature *_vote_intrinsic(builtin_available_predicate avail,
+ enum ir_intrinsic_id id);
+ ir_function_signature *_vote(const char *intrinsic_name,
+ builtin_available_predicate avail);
+
#undef B0
#undef B1
#undef B2
/** @} */
};
+enum image_function_flags {
+ IMAGE_FUNCTION_EMIT_STUB = (1 << 0),
+ IMAGE_FUNCTION_RETURNS_VOID = (1 << 1),
+ IMAGE_FUNCTION_HAS_VECTOR_DATA_TYPE = (1 << 2),
+ IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE = (1 << 3),
+ IMAGE_FUNCTION_READ_ONLY = (1 << 4),
+ IMAGE_FUNCTION_WRITE_ONLY = (1 << 5),
+ IMAGE_FUNCTION_AVAIL_ATOMIC = (1 << 6),
+ IMAGE_FUNCTION_MS_ONLY = (1 << 7),
+ IMAGE_FUNCTION_AVAIL_ATOMIC_EXCHANGE = (1 << 8),
+ IMAGE_FUNCTION_AVAIL_ATOMIC_ADD = (1 << 9),
+};
+
} /* anonymous namespace */
/**
* @{
*/
builtin_builder::builtin_builder()
- : shader(NULL),
- gl_ModelViewProjectionMatrix(NULL),
- gl_Vertex(NULL)
+ : shader(NULL)
{
mem_ctx = NULL;
}
* GLSL utility code that could be linked against any stage, so just
* arbitrarily pick GL_VERTEX_SHADER.
*/
- shader = _mesa_new_shader(NULL, 0, GL_VERTEX_SHADER);
+ shader = _mesa_new_shader(0, MESA_SHADER_VERTEX);
shader->symbols = new(mem_ctx) glsl_symbol_table;
-
- gl_ModelViewProjectionMatrix =
- new(mem_ctx) ir_variable(glsl_type::mat4_type,
- "gl_ModelViewProjectionMatrix",
- ir_var_uniform);
-
- shader->symbols->add_variable(gl_ModelViewProjectionMatrix);
-
- gl_Vertex = in_var(glsl_type::vec4_type, "gl_Vertex");
- shader->symbols->add_variable(gl_Vertex);
}
/** @} */
builtin_builder::create_intrinsics()
{
add_function("__intrinsic_atomic_read",
- _atomic_counter_intrinsic(shader_atomic_counters),
+ _atomic_counter_intrinsic(shader_atomic_counters,
+ ir_intrinsic_atomic_counter_read),
NULL);
add_function("__intrinsic_atomic_increment",
- _atomic_counter_intrinsic(shader_atomic_counters),
+ _atomic_counter_intrinsic(shader_atomic_counters,
+ ir_intrinsic_atomic_counter_increment),
NULL);
add_function("__intrinsic_atomic_predecrement",
- _atomic_counter_intrinsic(shader_atomic_counters),
+ _atomic_counter_intrinsic(shader_atomic_counters,
+ ir_intrinsic_atomic_counter_predecrement),
NULL);
add_function("__intrinsic_atomic_add",
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::uint_type),
+ glsl_type::uint_type,
+ ir_intrinsic_generic_atomic_add),
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::int_type),
+ glsl_type::int_type,
+ ir_intrinsic_generic_atomic_add),
+ _atomic_intrinsic2(NV_shader_atomic_float_supported,
+ glsl_type::float_type,
+ ir_intrinsic_generic_atomic_add),
+ _atomic_counter_intrinsic1(shader_atomic_counter_ops_or_v460_desktop,
+ ir_intrinsic_atomic_counter_add),
NULL);
add_function("__intrinsic_atomic_min",
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::uint_type),
+ glsl_type::uint_type,
+ ir_intrinsic_generic_atomic_min),
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::int_type),
+ glsl_type::int_type,
+ ir_intrinsic_generic_atomic_min),
+ _atomic_intrinsic2(INTEL_shader_atomic_float_minmax_supported,
+ glsl_type::float_type,
+ ir_intrinsic_generic_atomic_min),
+ _atomic_counter_intrinsic1(shader_atomic_counter_ops_or_v460_desktop,
+ ir_intrinsic_atomic_counter_min),
NULL);
add_function("__intrinsic_atomic_max",
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::uint_type),
+ glsl_type::uint_type,
+ ir_intrinsic_generic_atomic_max),
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::int_type),
+ glsl_type::int_type,
+ ir_intrinsic_generic_atomic_max),
+ _atomic_intrinsic2(INTEL_shader_atomic_float_minmax_supported,
+ glsl_type::float_type,
+ ir_intrinsic_generic_atomic_max),
+ _atomic_counter_intrinsic1(shader_atomic_counter_ops_or_v460_desktop,
+ ir_intrinsic_atomic_counter_max),
NULL);
add_function("__intrinsic_atomic_and",
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::uint_type),
+ glsl_type::uint_type,
+ ir_intrinsic_generic_atomic_and),
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::int_type),
+ glsl_type::int_type,
+ ir_intrinsic_generic_atomic_and),
+ _atomic_counter_intrinsic1(shader_atomic_counter_ops_or_v460_desktop,
+ ir_intrinsic_atomic_counter_and),
NULL);
add_function("__intrinsic_atomic_or",
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::uint_type),
+ glsl_type::uint_type,
+ ir_intrinsic_generic_atomic_or),
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::int_type),
+ glsl_type::int_type,
+ ir_intrinsic_generic_atomic_or),
+ _atomic_counter_intrinsic1(shader_atomic_counter_ops_or_v460_desktop,
+ ir_intrinsic_atomic_counter_or),
NULL);
add_function("__intrinsic_atomic_xor",
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::uint_type),
+ glsl_type::uint_type,
+ ir_intrinsic_generic_atomic_xor),
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::int_type),
+ glsl_type::int_type,
+ ir_intrinsic_generic_atomic_xor),
+ _atomic_counter_intrinsic1(shader_atomic_counter_ops_or_v460_desktop,
+ ir_intrinsic_atomic_counter_xor),
NULL);
add_function("__intrinsic_atomic_exchange",
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::uint_type),
+ glsl_type::uint_type,
+ ir_intrinsic_generic_atomic_exchange),
_atomic_intrinsic2(buffer_atomics_supported,
- glsl_type::int_type),
+ glsl_type::int_type,
+ ir_intrinsic_generic_atomic_exchange),
+ _atomic_intrinsic2(NV_shader_atomic_float_supported,
+ glsl_type::float_type,
+ ir_intrinsic_generic_atomic_exchange),
+ _atomic_counter_intrinsic1(shader_atomic_counter_ops_or_v460_desktop,
+ ir_intrinsic_atomic_counter_exchange),
NULL);
add_function("__intrinsic_atomic_comp_swap",
_atomic_intrinsic3(buffer_atomics_supported,
- glsl_type::uint_type),
+ glsl_type::uint_type,
+ ir_intrinsic_generic_atomic_comp_swap),
_atomic_intrinsic3(buffer_atomics_supported,
- glsl_type::int_type),
+ glsl_type::int_type,
+ ir_intrinsic_generic_atomic_comp_swap),
+ _atomic_intrinsic3(INTEL_shader_atomic_float_minmax_supported,
+ glsl_type::float_type,
+ ir_intrinsic_generic_atomic_comp_swap),
+ _atomic_counter_intrinsic2(shader_atomic_counter_ops_or_v460_desktop,
+ ir_intrinsic_atomic_counter_comp_swap),
NULL);
add_image_functions(false);
add_function("__intrinsic_memory_barrier",
- _memory_barrier_intrinsic(shader_image_load_store),
+ _memory_barrier_intrinsic(shader_image_load_store,
+ ir_intrinsic_memory_barrier),
NULL);
add_function("__intrinsic_group_memory_barrier",
- _memory_barrier_intrinsic(compute_shader),
+ _memory_barrier_intrinsic(compute_shader,
+ ir_intrinsic_group_memory_barrier),
NULL);
add_function("__intrinsic_memory_barrier_atomic_counter",
- _memory_barrier_intrinsic(compute_shader),
+ _memory_barrier_intrinsic(compute_shader_supported,
+ ir_intrinsic_memory_barrier_atomic_counter),
NULL);
add_function("__intrinsic_memory_barrier_buffer",
- _memory_barrier_intrinsic(compute_shader),
+ _memory_barrier_intrinsic(compute_shader_supported,
+ ir_intrinsic_memory_barrier_buffer),
NULL);
add_function("__intrinsic_memory_barrier_image",
- _memory_barrier_intrinsic(compute_shader),
+ _memory_barrier_intrinsic(compute_shader_supported,
+ ir_intrinsic_memory_barrier_image),
NULL);
add_function("__intrinsic_memory_barrier_shared",
- _memory_barrier_intrinsic(compute_shader),
+ _memory_barrier_intrinsic(compute_shader,
+ ir_intrinsic_memory_barrier_shared),
NULL);
+ add_function("__intrinsic_begin_invocation_interlock",
+ _invocation_interlock_intrinsic(
+ supports_arb_fragment_shader_interlock,
+ ir_intrinsic_begin_invocation_interlock), NULL);
+
+ add_function("__intrinsic_end_invocation_interlock",
+ _invocation_interlock_intrinsic(
+ supports_arb_fragment_shader_interlock,
+ ir_intrinsic_end_invocation_interlock), NULL);
+
+ add_function("__intrinsic_begin_fragment_shader_ordering",
+ _invocation_interlock_intrinsic(
+ supports_intel_fragment_shader_ordering,
+ ir_intrinsic_begin_fragment_shader_ordering), NULL);
+
add_function("__intrinsic_shader_clock",
_shader_clock_intrinsic(shader_clock,
glsl_type::uvec2_type),
NULL);
+
+ add_function("__intrinsic_vote_all",
+ _vote_intrinsic(vote_or_v460_desktop, ir_intrinsic_vote_all),
+ NULL);
+ add_function("__intrinsic_vote_any",
+ _vote_intrinsic(vote_or_v460_desktop, ir_intrinsic_vote_any),
+ NULL);
+ add_function("__intrinsic_vote_eq",
+ _vote_intrinsic(vote_or_v460_desktop, ir_intrinsic_vote_eq),
+ NULL);
+
+ add_function("__intrinsic_ballot", _ballot_intrinsic(), NULL);
+
+ add_function("__intrinsic_read_invocation",
+ _read_invocation_intrinsic(glsl_type::float_type),
+ _read_invocation_intrinsic(glsl_type::vec2_type),
+ _read_invocation_intrinsic(glsl_type::vec3_type),
+ _read_invocation_intrinsic(glsl_type::vec4_type),
+
+ _read_invocation_intrinsic(glsl_type::int_type),
+ _read_invocation_intrinsic(glsl_type::ivec2_type),
+ _read_invocation_intrinsic(glsl_type::ivec3_type),
+ _read_invocation_intrinsic(glsl_type::ivec4_type),
+
+ _read_invocation_intrinsic(glsl_type::uint_type),
+ _read_invocation_intrinsic(glsl_type::uvec2_type),
+ _read_invocation_intrinsic(glsl_type::uvec3_type),
+ _read_invocation_intrinsic(glsl_type::uvec4_type),
+ NULL);
+
+ add_function("__intrinsic_read_first_invocation",
+ _read_first_invocation_intrinsic(glsl_type::float_type),
+ _read_first_invocation_intrinsic(glsl_type::vec2_type),
+ _read_first_invocation_intrinsic(glsl_type::vec3_type),
+ _read_first_invocation_intrinsic(glsl_type::vec4_type),
+
+ _read_first_invocation_intrinsic(glsl_type::int_type),
+ _read_first_invocation_intrinsic(glsl_type::ivec2_type),
+ _read_first_invocation_intrinsic(glsl_type::ivec3_type),
+ _read_first_invocation_intrinsic(glsl_type::ivec4_type),
+
+ _read_first_invocation_intrinsic(glsl_type::uint_type),
+ _read_first_invocation_intrinsic(glsl_type::uvec2_type),
+ _read_first_invocation_intrinsic(glsl_type::uvec3_type),
+ _read_first_invocation_intrinsic(glsl_type::uvec4_type),
+ NULL);
+
}
/**
#define FDGS5(NAME) \
add_function(#NAME, \
- _##NAME(gpu_shader5, glsl_type::float_type), \
- _##NAME(gpu_shader5, glsl_type::vec2_type), \
- _##NAME(gpu_shader5, glsl_type::vec3_type), \
- _##NAME(gpu_shader5, glsl_type::vec4_type), \
+ _##NAME(gpu_shader5_es, glsl_type::float_type), \
+ _##NAME(gpu_shader5_es, glsl_type::vec2_type), \
+ _##NAME(gpu_shader5_es, glsl_type::vec3_type), \
+ _##NAME(gpu_shader5_es, glsl_type::vec4_type), \
_##NAME(fp64, glsl_type::double_type), \
_##NAME(fp64, glsl_type::dvec2_type), \
_##NAME(fp64, glsl_type::dvec3_type), \
_##NAME(glsl_type::ivec4_type), \
NULL);
-#define FID(NAME) \
+#define FI64(NAME) \
add_function(#NAME, \
_##NAME(always_available, glsl_type::float_type), \
_##NAME(always_available, glsl_type::vec2_type), \
_##NAME(fp64, glsl_type::dvec2_type), \
_##NAME(fp64, glsl_type::dvec3_type), \
_##NAME(fp64, glsl_type::dvec4_type), \
+ _##NAME(int64, glsl_type::int64_t_type), \
+ _##NAME(int64, glsl_type::i64vec2_type), \
+ _##NAME(int64, glsl_type::i64vec3_type), \
+ _##NAME(int64, glsl_type::i64vec4_type), \
NULL);
-#define FIUD(NAME) \
+#define FIUD_VEC(NAME) \
add_function(#NAME, \
- _##NAME(always_available, glsl_type::float_type), \
_##NAME(always_available, glsl_type::vec2_type), \
_##NAME(always_available, glsl_type::vec3_type), \
_##NAME(always_available, glsl_type::vec4_type), \
\
- _##NAME(always_available, glsl_type::int_type), \
_##NAME(always_available, glsl_type::ivec2_type), \
_##NAME(always_available, glsl_type::ivec3_type), \
_##NAME(always_available, glsl_type::ivec4_type), \
\
- _##NAME(v130, glsl_type::uint_type), \
_##NAME(v130, glsl_type::uvec2_type), \
_##NAME(v130, glsl_type::uvec3_type), \
_##NAME(v130, glsl_type::uvec4_type), \
- _##NAME(fp64, glsl_type::double_type), \
_##NAME(fp64, glsl_type::dvec2_type), \
_##NAME(fp64, glsl_type::dvec3_type), \
_##NAME(fp64, glsl_type::dvec4_type), \
+ _##NAME(int64, glsl_type::int64_t_type), \
+ _##NAME(int64, glsl_type::i64vec2_type), \
+ _##NAME(int64, glsl_type::i64vec3_type), \
+ _##NAME(int64, glsl_type::i64vec4_type), \
+ _##NAME(int64, glsl_type::uint64_t_type), \
+ _##NAME(int64, glsl_type::u64vec2_type), \
+ _##NAME(int64, glsl_type::u64vec3_type), \
+ _##NAME(int64, glsl_type::u64vec4_type), \
NULL);
#define IU(NAME) \
_##NAME(glsl_type::uvec4_type), \
NULL);
-#define FIUBD(NAME) \
+#define FIUBD_VEC(NAME) \
add_function(#NAME, \
- _##NAME(always_available, glsl_type::float_type), \
_##NAME(always_available, glsl_type::vec2_type), \
_##NAME(always_available, glsl_type::vec3_type), \
_##NAME(always_available, glsl_type::vec4_type), \
\
- _##NAME(always_available, glsl_type::int_type), \
_##NAME(always_available, glsl_type::ivec2_type), \
_##NAME(always_available, glsl_type::ivec3_type), \
_##NAME(always_available, glsl_type::ivec4_type), \
\
- _##NAME(v130, glsl_type::uint_type), \
_##NAME(v130, glsl_type::uvec2_type), \
_##NAME(v130, glsl_type::uvec3_type), \
_##NAME(v130, glsl_type::uvec4_type), \
\
- _##NAME(always_available, glsl_type::bool_type), \
_##NAME(always_available, glsl_type::bvec2_type), \
_##NAME(always_available, glsl_type::bvec3_type), \
_##NAME(always_available, glsl_type::bvec4_type), \
\
- _##NAME(fp64, glsl_type::double_type), \
_##NAME(fp64, glsl_type::dvec2_type), \
_##NAME(fp64, glsl_type::dvec3_type), \
_##NAME(fp64, glsl_type::dvec4_type), \
+ _##NAME(int64, glsl_type::int64_t_type), \
+ _##NAME(int64, glsl_type::i64vec2_type), \
+ _##NAME(int64, glsl_type::i64vec3_type), \
+ _##NAME(int64, glsl_type::i64vec4_type), \
+ _##NAME(int64, glsl_type::uint64_t_type), \
+ _##NAME(int64, glsl_type::u64vec2_type), \
+ _##NAME(int64, glsl_type::u64vec3_type), \
+ _##NAME(int64, glsl_type::u64vec4_type), \
NULL);
#define FIUD2_MIXED(NAME) \
_##NAME(fp64, glsl_type::dvec2_type, glsl_type::dvec2_type), \
_##NAME(fp64, glsl_type::dvec3_type, glsl_type::dvec3_type), \
_##NAME(fp64, glsl_type::dvec4_type, glsl_type::dvec4_type), \
+ \
+ _##NAME(int64, glsl_type::int64_t_type, glsl_type::int64_t_type), \
+ _##NAME(int64, glsl_type::i64vec2_type, glsl_type::int64_t_type), \
+ _##NAME(int64, glsl_type::i64vec3_type, glsl_type::int64_t_type), \
+ _##NAME(int64, glsl_type::i64vec4_type, glsl_type::int64_t_type), \
+ _##NAME(int64, glsl_type::i64vec2_type, glsl_type::i64vec2_type), \
+ _##NAME(int64, glsl_type::i64vec3_type, glsl_type::i64vec3_type), \
+ _##NAME(int64, glsl_type::i64vec4_type, glsl_type::i64vec4_type), \
+ _##NAME(int64, glsl_type::uint64_t_type, glsl_type::uint64_t_type), \
+ _##NAME(int64, glsl_type::u64vec2_type, glsl_type::uint64_t_type), \
+ _##NAME(int64, glsl_type::u64vec3_type, glsl_type::uint64_t_type), \
+ _##NAME(int64, glsl_type::u64vec4_type, glsl_type::uint64_t_type), \
+ _##NAME(int64, glsl_type::u64vec2_type, glsl_type::u64vec2_type), \
+ _##NAME(int64, glsl_type::u64vec3_type, glsl_type::u64vec3_type), \
+ _##NAME(int64, glsl_type::u64vec4_type, glsl_type::u64vec4_type), \
NULL);
F(radians)
F(log2)
FD(sqrt)
FD(inversesqrt)
- FID(abs)
- FID(sign)
+ FI64(abs)
+ FI64(sign)
FD(floor)
FD(trunc)
FD(round)
_mix_sel(shader_integer_mix, glsl_type::bvec2_type, glsl_type::bvec2_type),
_mix_sel(shader_integer_mix, glsl_type::bvec3_type, glsl_type::bvec3_type),
_mix_sel(shader_integer_mix, glsl_type::bvec4_type, glsl_type::bvec4_type),
+
+ _mix_sel(int64, glsl_type::int64_t_type, glsl_type::bool_type),
+ _mix_sel(int64, glsl_type::i64vec2_type, glsl_type::bvec2_type),
+ _mix_sel(int64, glsl_type::i64vec3_type, glsl_type::bvec3_type),
+ _mix_sel(int64, glsl_type::i64vec4_type, glsl_type::bvec4_type),
+
+ _mix_sel(int64, glsl_type::uint64_t_type, glsl_type::bool_type),
+ _mix_sel(int64, glsl_type::u64vec2_type, glsl_type::bvec2_type),
+ _mix_sel(int64, glsl_type::u64vec3_type, glsl_type::bvec3_type),
+ _mix_sel(int64, glsl_type::u64vec4_type, glsl_type::bvec4_type),
NULL);
add_function("step",
_uintBitsToFloat(glsl_type::uvec4_type),
NULL);
+ add_function("doubleBitsToInt64",
+ _doubleBitsToInt64(int64_fp64, glsl_type::double_type),
+ _doubleBitsToInt64(int64_fp64, glsl_type::dvec2_type),
+ _doubleBitsToInt64(int64_fp64, glsl_type::dvec3_type),
+ _doubleBitsToInt64(int64_fp64, glsl_type::dvec4_type),
+ NULL);
+
+ add_function("doubleBitsToUint64",
+ _doubleBitsToUint64(int64_fp64, glsl_type::double_type),
+ _doubleBitsToUint64(int64_fp64, glsl_type::dvec2_type),
+ _doubleBitsToUint64(int64_fp64, glsl_type::dvec3_type),
+ _doubleBitsToUint64(int64_fp64, glsl_type::dvec4_type),
+ NULL);
+
+ add_function("int64BitsToDouble",
+ _int64BitsToDouble(int64_fp64, glsl_type::int64_t_type),
+ _int64BitsToDouble(int64_fp64, glsl_type::i64vec2_type),
+ _int64BitsToDouble(int64_fp64, glsl_type::i64vec3_type),
+ _int64BitsToDouble(int64_fp64, glsl_type::i64vec4_type),
+ NULL);
+
+ add_function("uint64BitsToDouble",
+ _uint64BitsToDouble(int64_fp64, glsl_type::uint64_t_type),
+ _uint64BitsToDouble(int64_fp64, glsl_type::u64vec2_type),
+ _uint64BitsToDouble(int64_fp64, glsl_type::u64vec3_type),
+ _uint64BitsToDouble(int64_fp64, glsl_type::u64vec4_type),
+ NULL);
+
add_function("packUnorm2x16", _packUnorm2x16(shader_packing_or_es3_or_gpu_shader5), NULL);
add_function("packSnorm2x16", _packSnorm2x16(shader_packing_or_es3), NULL);
add_function("packUnorm4x8", _packUnorm4x8(shader_packing_or_es31_or_gpu_shader5), NULL);
add_function("packDouble2x32", _packDouble2x32(fp64), NULL);
add_function("unpackDouble2x32", _unpackDouble2x32(fp64), NULL);
+ add_function("packInt2x32", _packInt2x32(int64), NULL);
+ add_function("unpackInt2x32", _unpackInt2x32(int64), NULL);
+ add_function("packUint2x32", _packUint2x32(int64), NULL);
+ add_function("unpackUint2x32", _unpackUint2x32(int64), NULL);
FD(length)
FD(distance)
_transpose(fp64, glsl_type::dmat4x2_type),
_transpose(fp64, glsl_type::dmat4x3_type),
NULL);
- FIUD(lessThan)
- FIUD(lessThanEqual)
- FIUD(greaterThan)
- FIUD(greaterThanEqual)
- FIUBD(notEqual)
- FIUBD(equal)
+ FIUD_VEC(lessThan)
+ FIUD_VEC(lessThanEqual)
+ FIUD_VEC(greaterThan)
+ FIUD_VEC(greaterThanEqual)
+ FIUBD_VEC(notEqual)
+ FIUBD_VEC(equal)
add_function("any",
_any(glsl_type::bvec2_type),
_textureSize(v130, glsl_type::ivec2_type, glsl_type::usampler2DRect_type),
_textureSize(v130, glsl_type::ivec2_type, glsl_type::sampler2DRectShadow_type),
- _textureSize(v140, glsl_type::int_type, glsl_type::samplerBuffer_type),
- _textureSize(v140, glsl_type::int_type, glsl_type::isamplerBuffer_type),
- _textureSize(v140, glsl_type::int_type, glsl_type::usamplerBuffer_type),
+ _textureSize(texture_buffer, glsl_type::int_type, glsl_type::samplerBuffer_type),
+ _textureSize(texture_buffer, glsl_type::int_type, glsl_type::isamplerBuffer_type),
+ _textureSize(texture_buffer, glsl_type::int_type, glsl_type::usamplerBuffer_type),
_textureSize(texture_multisample, glsl_type::ivec2_type, glsl_type::sampler2DMS_type),
_textureSize(texture_multisample, glsl_type::ivec2_type, glsl_type::isampler2DMS_type),
_textureSize(texture_multisample, glsl_type::ivec2_type, glsl_type::usampler2DMS_type),
NULL);
add_function("textureSamples",
- _textureSamples(glsl_type::sampler2DMS_type),
- _textureSamples(glsl_type::isampler2DMS_type),
- _textureSamples(glsl_type::usampler2DMS_type),
+ _textureSamples(shader_samples, glsl_type::sampler2DMS_type),
+ _textureSamples(shader_samples, glsl_type::isampler2DMS_type),
+ _textureSamples(shader_samples, glsl_type::usampler2DMS_type),
- _textureSamples(glsl_type::sampler2DMSArray_type),
- _textureSamples(glsl_type::isampler2DMSArray_type),
- _textureSamples(glsl_type::usampler2DMSArray_type),
+ _textureSamples(shader_samples, glsl_type::sampler2DMSArray_type),
+ _textureSamples(shader_samples, glsl_type::isampler2DMSArray_type),
+ _textureSamples(shader_samples, glsl_type::usampler2DMSArray_type),
NULL);
add_function("texture",
_texture(ir_tex, v130, glsl_type::float_type, glsl_type::sampler1DArrayShadow_type, glsl_type::vec3_type),
_texture(ir_tex, v130, glsl_type::float_type, glsl_type::sampler2DArrayShadow_type, glsl_type::vec4_type),
/* samplerCubeArrayShadow is special; it has an extra parameter
- * for the shadow comparitor since there is no vec5 type.
+ * for the shadow comparator since there is no vec5 type.
*/
- _textureCubeArrayShadow(),
+ _textureCubeArrayShadow(texture_cube_map_array, glsl_type::samplerCubeArrayShadow_type),
_texture(ir_tex, v130, glsl_type::vec4_type, glsl_type::sampler2DRect_type, glsl_type::vec2_type),
_texture(ir_tex, v130, glsl_type::ivec4_type, glsl_type::isampler2DRect_type, glsl_type::vec2_type),
_texture(ir_tex, v130, glsl_type::float_type, glsl_type::sampler2DRectShadow_type, glsl_type::vec3_type),
+ _texture(ir_tex, texture_external_es3, glsl_type::vec4_type, glsl_type::samplerExternalOES_type, glsl_type::vec2_type),
+
_texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::float_type),
_texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::float_type),
_texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler1D_type, glsl_type::float_type),
_texture(ir_tex, v130, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET),
_texture(ir_tex, v130, glsl_type::float_type, glsl_type::sampler1DArrayShadow_type, glsl_type::vec3_type, TEX_OFFSET),
+ /* The next one was forgotten in GLSL 1.30 spec. It's from
+ * EXT_gpu_shader4 originally. It was added in 4.30 with the
+ * wrong syntax. This was corrected in 4.40. 4.30 indicates
+ * that it was intended to be included previously, so allow it
+ * in 1.30.
+ */
+ _texture(ir_tex, v130_desktop, glsl_type::float_type, glsl_type::sampler2DArrayShadow_type, glsl_type::vec4_type, TEX_OFFSET),
_texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::float_type, TEX_OFFSET),
_texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::float_type, TEX_OFFSET),
_texture(ir_tex, v130, glsl_type::vec4_type, glsl_type::sampler2DRect_type, glsl_type::vec3_type, TEX_PROJECT),
_texture(ir_tex, v130, glsl_type::ivec4_type, glsl_type::isampler2DRect_type, glsl_type::vec3_type, TEX_PROJECT),
+ _texture(ir_tex, texture_external_es3, glsl_type::vec4_type, glsl_type::samplerExternalOES_type, glsl_type::vec3_type, TEX_PROJECT),
+ _texture(ir_tex, texture_external_es3, glsl_type::vec4_type, glsl_type::samplerExternalOES_type, glsl_type::vec4_type, TEX_PROJECT),
+
_texture(ir_tex, v130, glsl_type::uvec4_type, glsl_type::usampler2DRect_type, glsl_type::vec3_type, TEX_PROJECT),
_texture(ir_tex, v130, glsl_type::vec4_type, glsl_type::sampler2DRect_type, glsl_type::vec4_type, TEX_PROJECT),
_texture(ir_tex, v130, glsl_type::ivec4_type, glsl_type::isampler2DRect_type, glsl_type::vec4_type, TEX_PROJECT),
_texture(ir_tex, v130, glsl_type::float_type, glsl_type::sampler2DRectShadow_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::vec2_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::vec2_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler1D_type, glsl_type::vec2_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler1D_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::vec2_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::vec2_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler1D_type, glsl_type::vec2_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler1D_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec3_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec3_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec3_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec3_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec3_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec3_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler3D_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler3D_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler3D_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler3D_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler3D_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler3D_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::float_type, glsl_type::sampler1DShadow_type, glsl_type::vec4_type, TEX_PROJECT),
- _texture(ir_txb, v130, glsl_type::float_type, glsl_type::sampler2DShadow_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::float_type, glsl_type::sampler1DShadow_type, glsl_type::vec4_type, TEX_PROJECT),
+ _texture(ir_txb, v130_fs_only, glsl_type::float_type, glsl_type::sampler2DShadow_type, glsl_type::vec4_type, TEX_PROJECT),
NULL);
add_function("texelFetch",
_texelFetch(v130, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::ivec3_type),
_texelFetch(v130, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::ivec3_type),
- _texelFetch(v140, glsl_type::vec4_type, glsl_type::samplerBuffer_type, glsl_type::int_type),
- _texelFetch(v140, glsl_type::ivec4_type, glsl_type::isamplerBuffer_type, glsl_type::int_type),
- _texelFetch(v140, glsl_type::uvec4_type, glsl_type::usamplerBuffer_type, glsl_type::int_type),
+ _texelFetch(texture_buffer, glsl_type::vec4_type, glsl_type::samplerBuffer_type, glsl_type::int_type),
+ _texelFetch(texture_buffer, glsl_type::ivec4_type, glsl_type::isamplerBuffer_type, glsl_type::int_type),
+ _texelFetch(texture_buffer, glsl_type::uvec4_type, glsl_type::usamplerBuffer_type, glsl_type::int_type),
_texelFetch(texture_multisample, glsl_type::vec4_type, glsl_type::sampler2DMS_type, glsl_type::ivec2_type),
_texelFetch(texture_multisample, glsl_type::ivec4_type, glsl_type::isampler2DMS_type, glsl_type::ivec2_type),
_texelFetch(texture_multisample_array, glsl_type::vec4_type, glsl_type::sampler2DMSArray_type, glsl_type::ivec3_type),
_texelFetch(texture_multisample_array, glsl_type::ivec4_type, glsl_type::isampler2DMSArray_type, glsl_type::ivec3_type),
_texelFetch(texture_multisample_array, glsl_type::uvec4_type, glsl_type::usampler2DMSArray_type, glsl_type::ivec3_type),
+
+ _texelFetch(texture_external_es3, glsl_type::vec4_type, glsl_type::samplerExternalOES_type, glsl_type::ivec2_type),
+
NULL);
+
add_function("texelFetchOffset",
_texelFetch(v130, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::int_type, glsl_type::int_type),
_texelFetch(v130, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::int_type, glsl_type::int_type),
_texture(ir_tex, v130, glsl_type::float_type, glsl_type::sampler2DRectShadow_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::vec2_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::vec2_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler1D_type, glsl_type::vec2_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler1D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::vec2_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::vec2_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler1D_type, glsl_type::vec2_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler1D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler1D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler1D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec3_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec3_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec3_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec3_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec3_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec3_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::vec4_type, glsl_type::sampler3D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::ivec4_type, glsl_type::isampler3D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::uvec4_type, glsl_type::usampler3D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::vec4_type, glsl_type::sampler3D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::ivec4_type, glsl_type::isampler3D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::uvec4_type, glsl_type::usampler3D_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::float_type, glsl_type::sampler1DShadow_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
- _texture(ir_txb, v130, glsl_type::float_type, glsl_type::sampler2DShadow_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::float_type, glsl_type::sampler1DShadow_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
+ _texture(ir_txb, v130_fs_only, glsl_type::float_type, glsl_type::sampler2DShadow_type, glsl_type::vec4_type, TEX_PROJECT | TEX_OFFSET),
NULL);
add_function("textureLodOffset",
NULL);
add_function("textureQueryLevels",
- _textureQueryLevels(glsl_type::sampler1D_type),
- _textureQueryLevels(glsl_type::sampler2D_type),
- _textureQueryLevels(glsl_type::sampler3D_type),
- _textureQueryLevels(glsl_type::samplerCube_type),
- _textureQueryLevels(glsl_type::sampler1DArray_type),
- _textureQueryLevels(glsl_type::sampler2DArray_type),
- _textureQueryLevels(glsl_type::samplerCubeArray_type),
- _textureQueryLevels(glsl_type::sampler1DShadow_type),
- _textureQueryLevels(glsl_type::sampler2DShadow_type),
- _textureQueryLevels(glsl_type::samplerCubeShadow_type),
- _textureQueryLevels(glsl_type::sampler1DArrayShadow_type),
- _textureQueryLevels(glsl_type::sampler2DArrayShadow_type),
- _textureQueryLevels(glsl_type::samplerCubeArrayShadow_type),
-
- _textureQueryLevels(glsl_type::isampler1D_type),
- _textureQueryLevels(glsl_type::isampler2D_type),
- _textureQueryLevels(glsl_type::isampler3D_type),
- _textureQueryLevels(glsl_type::isamplerCube_type),
- _textureQueryLevels(glsl_type::isampler1DArray_type),
- _textureQueryLevels(glsl_type::isampler2DArray_type),
- _textureQueryLevels(glsl_type::isamplerCubeArray_type),
-
- _textureQueryLevels(glsl_type::usampler1D_type),
- _textureQueryLevels(glsl_type::usampler2D_type),
- _textureQueryLevels(glsl_type::usampler3D_type),
- _textureQueryLevels(glsl_type::usamplerCube_type),
- _textureQueryLevels(glsl_type::usampler1DArray_type),
- _textureQueryLevels(glsl_type::usampler2DArray_type),
- _textureQueryLevels(glsl_type::usamplerCubeArray_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::sampler1D_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::sampler2D_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::sampler3D_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::samplerCube_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::sampler1DArray_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::sampler2DArray_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::samplerCubeArray_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::sampler1DShadow_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::sampler2DShadow_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::samplerCubeShadow_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::sampler1DArrayShadow_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::sampler2DArrayShadow_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::samplerCubeArrayShadow_type),
+
+ _textureQueryLevels(texture_query_levels, glsl_type::isampler1D_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::isampler2D_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::isampler3D_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::isamplerCube_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::isampler1DArray_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::isampler2DArray_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::isamplerCubeArray_type),
+
+ _textureQueryLevels(texture_query_levels, glsl_type::usampler1D_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::usampler2D_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::usampler3D_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::usamplerCube_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::usampler1DArray_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::usampler2DArray_type),
+ _textureQueryLevels(texture_query_levels, glsl_type::usamplerCubeArray_type),
NULL);
_texture(ir_tg4, texture_gather_or_es31, glsl_type::ivec4_type, glsl_type::isamplerCube_type, glsl_type::vec3_type),
_texture(ir_tg4, texture_gather_or_es31, glsl_type::uvec4_type, glsl_type::usamplerCube_type, glsl_type::vec3_type),
- _texture(ir_tg4, texture_gather, glsl_type::vec4_type, glsl_type::samplerCubeArray_type, glsl_type::vec4_type),
- _texture(ir_tg4, texture_gather, glsl_type::ivec4_type, glsl_type::isamplerCubeArray_type, glsl_type::vec4_type),
- _texture(ir_tg4, texture_gather, glsl_type::uvec4_type, glsl_type::usamplerCubeArray_type, glsl_type::vec4_type),
+ _texture(ir_tg4, texture_gather_cube_map_array, glsl_type::vec4_type, glsl_type::samplerCubeArray_type, glsl_type::vec4_type),
+ _texture(ir_tg4, texture_gather_cube_map_array, glsl_type::ivec4_type, glsl_type::isamplerCubeArray_type, glsl_type::vec4_type),
+ _texture(ir_tg4, texture_gather_cube_map_array, glsl_type::uvec4_type, glsl_type::usamplerCubeArray_type, glsl_type::vec4_type),
_texture(ir_tg4, gpu_shader5_or_es31, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_COMPONENT),
_texture(ir_tg4, gpu_shader5_or_es31, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_COMPONENT),
_texture(ir_tg4, gpu_shader5_or_es31, glsl_type::ivec4_type, glsl_type::isamplerCube_type, glsl_type::vec3_type, TEX_COMPONENT),
_texture(ir_tg4, gpu_shader5_or_es31, glsl_type::uvec4_type, glsl_type::usamplerCube_type, glsl_type::vec3_type, TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::samplerCubeArray_type, glsl_type::vec4_type, TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isamplerCubeArray_type, glsl_type::vec4_type, TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usamplerCubeArray_type, glsl_type::vec4_type, TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_or_OES_texture_cube_map_array, glsl_type::vec4_type, glsl_type::samplerCubeArray_type, glsl_type::vec4_type, TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_or_OES_texture_cube_map_array, glsl_type::ivec4_type, glsl_type::isamplerCubeArray_type, glsl_type::vec4_type, TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_or_OES_texture_cube_map_array, glsl_type::uvec4_type, glsl_type::usamplerCubeArray_type, glsl_type::vec4_type, TEX_COMPONENT),
_texture(ir_tg4, gpu_shader5_or_es31, glsl_type::vec4_type, glsl_type::sampler2DShadow_type, glsl_type::vec2_type),
_texture(ir_tg4, gpu_shader5_or_es31, glsl_type::vec4_type, glsl_type::sampler2DArrayShadow_type, glsl_type::vec3_type),
_texture(ir_tg4, gpu_shader5_or_es31, glsl_type::vec4_type, glsl_type::samplerCubeShadow_type, glsl_type::vec3_type),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::samplerCubeArrayShadow_type, glsl_type::vec4_type),
+ _texture(ir_tg4, gpu_shader5_or_OES_texture_cube_map_array, glsl_type::vec4_type, glsl_type::samplerCubeArrayShadow_type, glsl_type::vec4_type),
_texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DRectShadow_type, glsl_type::vec2_type),
NULL);
_texture(ir_tg4, texture_gather_only_or_es31, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET),
_texture(ir_tg4, texture_gather_only_or_es31, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET),
- _texture(ir_tg4, es31, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET | TEX_COMPONENT),
- _texture(ir_tg4, es31, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET | TEX_COMPONENT),
- _texture(ir_tg4, es31, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET | TEX_COMPONENT),
+ _texture(ir_tg4, es31_not_gs5, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET | TEX_COMPONENT),
+ _texture(ir_tg4, es31_not_gs5, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET | TEX_COMPONENT),
+ _texture(ir_tg4, es31_not_gs5, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET | TEX_COMPONENT),
- _texture(ir_tg4, es31, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET | TEX_COMPONENT),
- _texture(ir_tg4, es31, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET | TEX_COMPONENT),
- _texture(ir_tg4, es31, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET | TEX_COMPONENT),
+ _texture(ir_tg4, es31_not_gs5, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET | TEX_COMPONENT),
+ _texture(ir_tg4, es31_not_gs5, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET | TEX_COMPONENT),
+ _texture(ir_tg4, es31_not_gs5, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
- _texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
- _texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST),
- _texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST),
- _texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST),
_texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
_texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
_texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
_texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
_texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
_texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DShadow_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DArrayShadow_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2DShadow_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2DArrayShadow_type, glsl_type::vec3_type, TEX_OFFSET_NONCONST),
_texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DRectShadow_type, glsl_type::vec2_type, TEX_OFFSET_NONCONST),
- _texture(ir_tg4, es31, glsl_type::vec4_type, glsl_type::sampler2DShadow_type, glsl_type::vec2_type, TEX_OFFSET),
- _texture(ir_tg4, es31, glsl_type::vec4_type, glsl_type::sampler2DArrayShadow_type, glsl_type::vec3_type, TEX_OFFSET),
+ _texture(ir_tg4, es31_not_gs5, glsl_type::vec4_type, glsl_type::sampler2DShadow_type, glsl_type::vec2_type, TEX_OFFSET),
+ _texture(ir_tg4, es31_not_gs5, glsl_type::vec4_type, glsl_type::sampler2DArrayShadow_type, glsl_type::vec3_type, TEX_OFFSET),
NULL);
add_function("textureGatherOffsets",
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
- _texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
- _texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::ivec4_type, glsl_type::isampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::uvec4_type, glsl_type::usampler2D_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY),
- _texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY),
- _texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::ivec4_type, glsl_type::isampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::uvec4_type, glsl_type::usampler2DArray_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
_texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
_texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
_texture(ir_tg4, gpu_shader5, glsl_type::ivec4_type, glsl_type::isampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
_texture(ir_tg4, gpu_shader5, glsl_type::uvec4_type, glsl_type::usampler2DRect_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY | TEX_COMPONENT),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DShadow_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
- _texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DArrayShadow_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2DShadow_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
+ _texture(ir_tg4, gpu_shader5_es, glsl_type::vec4_type, glsl_type::sampler2DArrayShadow_type, glsl_type::vec3_type, TEX_OFFSET_ARRAY),
_texture(ir_tg4, gpu_shader5, glsl_type::vec4_type, glsl_type::sampler2DRectShadow_type, glsl_type::vec2_type, TEX_OFFSET_ARRAY),
NULL);
shader_atomic_counters),
NULL);
+ add_function("atomicCounterAddARB",
+ _atomic_counter_op1("__intrinsic_atomic_add",
+ shader_atomic_counter_ops),
+ NULL);
+ add_function("atomicCounterSubtractARB",
+ _atomic_counter_op1("__intrinsic_atomic_sub",
+ shader_atomic_counter_ops),
+ NULL);
+ add_function("atomicCounterMinARB",
+ _atomic_counter_op1("__intrinsic_atomic_min",
+ shader_atomic_counter_ops),
+ NULL);
+ add_function("atomicCounterMaxARB",
+ _atomic_counter_op1("__intrinsic_atomic_max",
+ shader_atomic_counter_ops),
+ NULL);
+ add_function("atomicCounterAndARB",
+ _atomic_counter_op1("__intrinsic_atomic_and",
+ shader_atomic_counter_ops),
+ NULL);
+ add_function("atomicCounterOrARB",
+ _atomic_counter_op1("__intrinsic_atomic_or",
+ shader_atomic_counter_ops),
+ NULL);
+ add_function("atomicCounterXorARB",
+ _atomic_counter_op1("__intrinsic_atomic_xor",
+ shader_atomic_counter_ops),
+ NULL);
+ add_function("atomicCounterExchangeARB",
+ _atomic_counter_op1("__intrinsic_atomic_exchange",
+ shader_atomic_counter_ops),
+ NULL);
+ add_function("atomicCounterCompSwapARB",
+ _atomic_counter_op2("__intrinsic_atomic_comp_swap",
+ shader_atomic_counter_ops),
+ NULL);
+
+ add_function("atomicCounterAdd",
+ _atomic_counter_op1("__intrinsic_atomic_add",
+ v460_desktop),
+ NULL);
+ add_function("atomicCounterSubtract",
+ _atomic_counter_op1("__intrinsic_atomic_sub",
+ v460_desktop),
+ NULL);
+ add_function("atomicCounterMin",
+ _atomic_counter_op1("__intrinsic_atomic_min",
+ v460_desktop),
+ NULL);
+ add_function("atomicCounterMax",
+ _atomic_counter_op1("__intrinsic_atomic_max",
+ v460_desktop),
+ NULL);
+ add_function("atomicCounterAnd",
+ _atomic_counter_op1("__intrinsic_atomic_and",
+ v460_desktop),
+ NULL);
+ add_function("atomicCounterOr",
+ _atomic_counter_op1("__intrinsic_atomic_or",
+ v460_desktop),
+ NULL);
+ add_function("atomicCounterXor",
+ _atomic_counter_op1("__intrinsic_atomic_xor",
+ v460_desktop),
+ NULL);
+ add_function("atomicCounterExchange",
+ _atomic_counter_op1("__intrinsic_atomic_exchange",
+ v460_desktop),
+ NULL);
+ add_function("atomicCounterCompSwap",
+ _atomic_counter_op2("__intrinsic_atomic_comp_swap",
+ v460_desktop),
+ NULL);
+
add_function("atomicAdd",
_atomic_op2("__intrinsic_atomic_add",
buffer_atomics_supported,
_atomic_op2("__intrinsic_atomic_add",
buffer_atomics_supported,
glsl_type::int_type),
+ _atomic_op2("__intrinsic_atomic_add",
+ shader_atomic_float_add,
+ glsl_type::float_type),
NULL);
add_function("atomicMin",
_atomic_op2("__intrinsic_atomic_min",
_atomic_op2("__intrinsic_atomic_min",
buffer_atomics_supported,
glsl_type::int_type),
+ _atomic_op2("__intrinsic_atomic_min",
+ shader_atomic_float_minmax,
+ glsl_type::float_type),
NULL);
add_function("atomicMax",
_atomic_op2("__intrinsic_atomic_max",
_atomic_op2("__intrinsic_atomic_max",
buffer_atomics_supported,
glsl_type::int_type),
+ _atomic_op2("__intrinsic_atomic_max",
+ shader_atomic_float_minmax,
+ glsl_type::float_type),
NULL);
add_function("atomicAnd",
_atomic_op2("__intrinsic_atomic_and",
_atomic_op2("__intrinsic_atomic_exchange",
buffer_atomics_supported,
glsl_type::int_type),
+ _atomic_op2("__intrinsic_atomic_exchange",
+ shader_atomic_float_exchange,
+ glsl_type::float_type),
NULL);
add_function("atomicCompSwap",
_atomic_op3("__intrinsic_atomic_comp_swap",
_atomic_op3("__intrinsic_atomic_comp_swap",
buffer_atomics_supported,
glsl_type::int_type),
+ _atomic_op3("__intrinsic_atomic_comp_swap",
+ shader_atomic_float_minmax,
+ glsl_type::float_type),
NULL);
add_function("min3",
NULL);
add_function("memoryBarrierAtomicCounter",
_memory_barrier("__intrinsic_memory_barrier_atomic_counter",
- compute_shader),
+ compute_shader_supported),
NULL);
add_function("memoryBarrierBuffer",
_memory_barrier("__intrinsic_memory_barrier_buffer",
- compute_shader),
+ compute_shader_supported),
NULL);
add_function("memoryBarrierImage",
_memory_barrier("__intrinsic_memory_barrier_image",
- compute_shader),
+ compute_shader_supported),
NULL);
add_function("memoryBarrierShared",
_memory_barrier("__intrinsic_memory_barrier_shared",
compute_shader),
NULL);
+ add_function("ballotARB", _ballot(), NULL);
+
+ add_function("readInvocationARB",
+ _read_invocation(glsl_type::float_type),
+ _read_invocation(glsl_type::vec2_type),
+ _read_invocation(glsl_type::vec3_type),
+ _read_invocation(glsl_type::vec4_type),
+
+ _read_invocation(glsl_type::int_type),
+ _read_invocation(glsl_type::ivec2_type),
+ _read_invocation(glsl_type::ivec3_type),
+ _read_invocation(glsl_type::ivec4_type),
+
+ _read_invocation(glsl_type::uint_type),
+ _read_invocation(glsl_type::uvec2_type),
+ _read_invocation(glsl_type::uvec3_type),
+ _read_invocation(glsl_type::uvec4_type),
+ NULL);
+
+ add_function("readFirstInvocationARB",
+ _read_first_invocation(glsl_type::float_type),
+ _read_first_invocation(glsl_type::vec2_type),
+ _read_first_invocation(glsl_type::vec3_type),
+ _read_first_invocation(glsl_type::vec4_type),
+
+ _read_first_invocation(glsl_type::int_type),
+ _read_first_invocation(glsl_type::ivec2_type),
+ _read_first_invocation(glsl_type::ivec3_type),
+ _read_first_invocation(glsl_type::ivec4_type),
+
+ _read_first_invocation(glsl_type::uint_type),
+ _read_first_invocation(glsl_type::uvec2_type),
+ _read_first_invocation(glsl_type::uvec3_type),
+ _read_first_invocation(glsl_type::uvec4_type),
+ NULL);
+
add_function("clock2x32ARB",
_shader_clock(shader_clock,
glsl_type::uvec2_type),
NULL);
+ add_function("clockARB",
+ _shader_clock(shader_clock_int64,
+ glsl_type::uint64_t_type),
+ NULL);
+
+ add_function("beginInvocationInterlockARB",
+ _invocation_interlock(
+ "__intrinsic_begin_invocation_interlock",
+ supports_arb_fragment_shader_interlock),
+ NULL);
+
+ add_function("endInvocationInterlockARB",
+ _invocation_interlock(
+ "__intrinsic_end_invocation_interlock",
+ supports_arb_fragment_shader_interlock),
+ NULL);
+
+ add_function("beginInvocationInterlockNV",
+ _invocation_interlock(
+ "__intrinsic_begin_invocation_interlock",
+ supports_nv_fragment_shader_interlock),
+ NULL);
+
+ add_function("endInvocationInterlockNV",
+ _invocation_interlock(
+ "__intrinsic_end_invocation_interlock",
+ supports_nv_fragment_shader_interlock),
+ NULL);
+
+ add_function("beginFragmentShaderOrderingINTEL",
+ _invocation_interlock(
+ "__intrinsic_begin_fragment_shader_ordering",
+ supports_intel_fragment_shader_ordering),
+ NULL);
+
+ add_function("anyInvocationARB",
+ _vote("__intrinsic_vote_any", vote),
+ NULL);
+
+ add_function("allInvocationsARB",
+ _vote("__intrinsic_vote_all", vote),
+ NULL);
+
+ add_function("allInvocationsEqualARB",
+ _vote("__intrinsic_vote_eq", vote),
+ NULL);
+
+ add_function("anyInvocation",
+ _vote("__intrinsic_vote_any", v460_desktop),
+ NULL);
+
+ add_function("allInvocations",
+ _vote("__intrinsic_vote_all", v460_desktop),
+ NULL);
+
+ add_function("allInvocationsEqual",
+ _vote("__intrinsic_vote_eq", v460_desktop),
+ NULL);
+
+ add_function("__builtin_idiv64",
+ generate_ir::idiv64(mem_ctx, integer_functions_supported),
+ NULL);
+
+ add_function("__builtin_imod64",
+ generate_ir::imod64(mem_ctx, integer_functions_supported),
+ NULL);
+
+ add_function("__builtin_sign64",
+ generate_ir::sign64(mem_ctx, integer_functions_supported),
+ NULL);
+
+ add_function("__builtin_udiv64",
+ generate_ir::udiv64(mem_ctx, integer_functions_supported),
+ NULL);
+
+ add_function("__builtin_umod64",
+ generate_ir::umod64(mem_ctx, integer_functions_supported),
+ NULL);
+
+ add_function("__builtin_umul64",
+ generate_ir::umul64(mem_ctx, integer_functions_supported),
+ NULL);
+
#undef F
#undef FI
-#undef FIUD
-#undef FIUBD
+#undef FIUD_VEC
+#undef FIUBD_VEC
#undef FIU2_MIXED
}
const char *intrinsic_name,
image_prototype_ctr prototype,
unsigned num_arguments,
- unsigned flags)
+ unsigned flags,
+ enum ir_intrinsic_id intrinsic_id)
{
static const glsl_type *const types[] = {
glsl_type::image1D_type,
ir_function *f = new(mem_ctx) ir_function(name);
for (unsigned i = 0; i < ARRAY_SIZE(types); ++i) {
- if ((types[i]->sampler_type != GLSL_TYPE_FLOAT ||
+ if ((types[i]->sampled_type != GLSL_TYPE_FLOAT ||
(flags & IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE)) &&
(types[i]->sampler_dimensionality == GLSL_SAMPLER_DIM_MS ||
!(flags & IMAGE_FUNCTION_MS_ONLY)))
f->add_signature(_image(prototype, types[i], intrinsic_name,
- num_arguments, flags));
+ num_arguments, flags, intrinsic_id));
}
shader->symbols->add_function(f);
&builtin_builder::_image_prototype, 0,
(flags | IMAGE_FUNCTION_HAS_VECTOR_DATA_TYPE |
IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE |
- IMAGE_FUNCTION_READ_ONLY));
+ IMAGE_FUNCTION_READ_ONLY),
+ ir_intrinsic_image_load);
add_image_function(glsl ? "imageStore" : "__intrinsic_image_store",
"__intrinsic_image_store",
(flags | IMAGE_FUNCTION_RETURNS_VOID |
IMAGE_FUNCTION_HAS_VECTOR_DATA_TYPE |
IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE |
- IMAGE_FUNCTION_WRITE_ONLY));
+ IMAGE_FUNCTION_WRITE_ONLY),
+ ir_intrinsic_image_store);
const unsigned atom_flags = flags | IMAGE_FUNCTION_AVAIL_ATOMIC;
add_image_function(glsl ? "imageAtomicAdd" : "__intrinsic_image_atomic_add",
"__intrinsic_image_atomic_add",
- &builtin_builder::_image_prototype, 1, atom_flags);
+ &builtin_builder::_image_prototype, 1,
+ (flags | IMAGE_FUNCTION_AVAIL_ATOMIC_ADD |
+ IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE),
+ ir_intrinsic_image_atomic_add);
add_image_function(glsl ? "imageAtomicMin" : "__intrinsic_image_atomic_min",
"__intrinsic_image_atomic_min",
- &builtin_builder::_image_prototype, 1, atom_flags);
+ &builtin_builder::_image_prototype, 1, atom_flags,
+ ir_intrinsic_image_atomic_min);
add_image_function(glsl ? "imageAtomicMax" : "__intrinsic_image_atomic_max",
"__intrinsic_image_atomic_max",
- &builtin_builder::_image_prototype, 1, atom_flags);
+ &builtin_builder::_image_prototype, 1, atom_flags,
+ ir_intrinsic_image_atomic_max);
add_image_function(glsl ? "imageAtomicAnd" : "__intrinsic_image_atomic_and",
"__intrinsic_image_atomic_and",
- &builtin_builder::_image_prototype, 1, atom_flags);
+ &builtin_builder::_image_prototype, 1, atom_flags,
+ ir_intrinsic_image_atomic_and);
add_image_function(glsl ? "imageAtomicOr" : "__intrinsic_image_atomic_or",
"__intrinsic_image_atomic_or",
- &builtin_builder::_image_prototype, 1, atom_flags);
+ &builtin_builder::_image_prototype, 1, atom_flags,
+ ir_intrinsic_image_atomic_or);
add_image_function(glsl ? "imageAtomicXor" : "__intrinsic_image_atomic_xor",
"__intrinsic_image_atomic_xor",
- &builtin_builder::_image_prototype, 1, atom_flags);
+ &builtin_builder::_image_prototype, 1, atom_flags,
+ ir_intrinsic_image_atomic_xor);
add_image_function((glsl ? "imageAtomicExchange" :
"__intrinsic_image_atomic_exchange"),
"__intrinsic_image_atomic_exchange",
- &builtin_builder::_image_prototype, 1, atom_flags);
+ &builtin_builder::_image_prototype, 1,
+ (flags | IMAGE_FUNCTION_AVAIL_ATOMIC_EXCHANGE |
+ IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE),
+ ir_intrinsic_image_atomic_exchange);
add_image_function((glsl ? "imageAtomicCompSwap" :
"__intrinsic_image_atomic_comp_swap"),
"__intrinsic_image_atomic_comp_swap",
- &builtin_builder::_image_prototype, 2, atom_flags);
+ &builtin_builder::_image_prototype, 2, atom_flags,
+ ir_intrinsic_image_atomic_comp_swap);
add_image_function(glsl ? "imageSize" : "__intrinsic_image_size",
"__intrinsic_image_size",
&builtin_builder::_image_size_prototype, 1,
- flags | IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE);
+ flags | IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE,
+ ir_intrinsic_image_size);
add_image_function(glsl ? "imageSamples" : "__intrinsic_image_samples",
"__intrinsic_image_samples",
&builtin_builder::_image_samples_prototype, 1,
flags | IMAGE_FUNCTION_SUPPORTS_FLOAT_DATA_TYPE |
- IMAGE_FUNCTION_MS_ONLY);
+ IMAGE_FUNCTION_MS_ONLY,
+ ir_intrinsic_image_samples);
}
ir_variable *
return new(mem_ctx) ir_constant(type, &data);
}
-#define IMM_FP(type, val) (type->base_type == GLSL_TYPE_DOUBLE) ? imm(val) : imm((float)val)
+#define IMM_FP(type, val) (type->is_double()) ? imm(val) : imm((float)val)
ir_dereference_variable *
builtin_builder::var_ref(ir_variable *var)
ir_factory body(&sig->body, mem_ctx); \
sig->is_defined = true;
-#define MAKE_INTRINSIC(return_type, avail, ...) \
+#define MAKE_INTRINSIC(return_type, id, avail, ...) \
ir_function_signature *sig = \
new_sig(return_type, avail, __VA_ARGS__); \
- sig->is_intrinsic = true;
+ sig->intrinsic_id = id;
ir_function_signature *
builtin_builder::unop(builtin_available_predicate avail,
ir_expression_operation opcode,
const glsl_type *return_type,
const glsl_type *param0_type,
- const glsl_type *param1_type)
+ const glsl_type *param1_type,
+ bool swap_operands)
{
ir_variable *x = in_var(param0_type, "x");
ir_variable *y = in_var(param1_type, "y");
MAKE_SIG(return_type, avail, 2, x, y);
- body.emit(ret(expr(opcode, x, y)));
+
+ if (swap_operands)
+ body.emit(ret(expr(opcode, y, x)));
+ else
+ body.emit(ret(expr(opcode, x, y)));
+
return sig;
}
}
ir_expression *
-builtin_builder::asin_expr(ir_variable *x)
+builtin_builder::asin_expr(ir_variable *x, float p0, float p1)
{
return mul(sign(x),
sub(imm(M_PI_2f),
mul(abs(x),
add(imm(M_PI_4f - 1.0f),
mul(abs(x),
- add(imm(0.086566724f),
- mul(abs(x), imm(-0.03102955f))))))))));
+ add(imm(p0),
+ mul(abs(x), imm(p1))))))))));
}
+/**
+ * Generate a ir_call to a function with a set of parameters
+ *
+ * The input \c params can either be a list of \c ir_variable or a list of
+ * \c ir_dereference_variable. In the latter case, all nodes will be removed
+ * from \c params and used directly as the parameters to the generated
+ * \c ir_call.
+ */
ir_call *
builtin_builder::call(ir_function *f, ir_variable *ret, exec_list params)
{
exec_list actual_params;
- foreach_in_list(ir_variable, var, ¶ms) {
- actual_params.push_tail(var_ref(var));
+ foreach_in_list_safe(ir_instruction, ir, ¶ms) {
+ ir_dereference_variable *d = ir->as_dereference_variable();
+ if (d != NULL) {
+ d->remove();
+ actual_params.push_tail(d);
+ } else {
+ ir_variable *var = ir->as_variable();
+ assert(var != NULL);
+ actual_params.push_tail(var_ref(var));
+ }
}
ir_function_signature *sig =
ir_variable *x = in_var(type, "x");
MAKE_SIG(type, always_available, 1, x);
- body.emit(ret(asin_expr(x)));
+ body.emit(ret(asin_expr(x, 0.086566724f, -0.03102955f)));
return sig;
}
ir_variable *x = in_var(type, "x");
MAKE_SIG(type, always_available, 1, x);
- body.emit(ret(sub(imm(M_PI_2f), asin_expr(x))));
+ body.emit(ret(sub(imm(M_PI_2f), asin_expr(x, 0.08132463f, -0.02363318f))));
return sig;
}
ir_function_signature *
builtin_builder::_atan2(const glsl_type *type)
{
- ir_variable *vec_y = in_var(type, "vec_y");
- ir_variable *vec_x = in_var(type, "vec_x");
- MAKE_SIG(type, always_available, 2, vec_y, vec_x);
-
- ir_variable *vec_result = body.make_temp(type, "vec_result");
- ir_variable *r = body.make_temp(glsl_type::float_type, "r");
- for (int i = 0; i < type->vector_elements; i++) {
- ir_variable *y = body.make_temp(glsl_type::float_type, "y");
- ir_variable *x = body.make_temp(glsl_type::float_type, "x");
- body.emit(assign(y, swizzle(vec_y, i, 1)));
- body.emit(assign(x, swizzle(vec_x, i, 1)));
-
- /* If |x| >= 1.0e-8 * |y|: */
- ir_if *outer_if =
- new(mem_ctx) ir_if(greater(abs(x), mul(imm(1.0e-8f), abs(y))));
-
- ir_factory outer_then(&outer_if->then_instructions, mem_ctx);
-
- /* Then...call atan(y/x) */
- do_atan(outer_then, glsl_type::float_type, r, div(y, x));
-
- /* ...and fix it up: */
- ir_if *inner_if = new(mem_ctx) ir_if(less(x, imm(0.0f)));
- inner_if->then_instructions.push_tail(
- if_tree(gequal(y, imm(0.0f)),
- assign(r, add(r, imm(M_PIf))),
- assign(r, sub(r, imm(M_PIf)))));
- outer_then.emit(inner_if);
+ const unsigned n = type->vector_elements;
+ ir_variable *y = in_var(type, "y");
+ ir_variable *x = in_var(type, "x");
+ MAKE_SIG(type, always_available, 2, y, x);
- /* Else... */
- outer_if->else_instructions.push_tail(
- assign(r, mul(sign(y), imm(M_PI_2f))));
+ /* If we're on the left half-plane rotate the coordinates π/2 clock-wise
+ * for the y=0 discontinuity to end up aligned with the vertical
+ * discontinuity of atan(s/t) along t=0. This also makes sure that we
+ * don't attempt to divide by zero along the vertical line, which may give
+ * unspecified results on non-GLSL 4.1-capable hardware.
+ */
+ ir_variable *flip = body.make_temp(glsl_type::bvec(n), "flip");
+ body.emit(assign(flip, gequal(imm(0.0f, n), x)));
+ ir_variable *s = body.make_temp(type, "s");
+ body.emit(assign(s, csel(flip, abs(x), y)));
+ ir_variable *t = body.make_temp(type, "t");
+ body.emit(assign(t, csel(flip, y, abs(x))));
- body.emit(outer_if);
+ /* If the magnitude of the denominator exceeds some huge value, scale down
+ * the arguments in order to prevent the reciprocal operation from flushing
+ * its result to zero, which would cause precision problems, and for s
+ * infinite would cause us to return a NaN instead of the correct finite
+ * value.
+ *
+ * If fmin and fmax are respectively the smallest and largest positive
+ * normalized floating point values representable by the implementation,
+ * the constants below should be in agreement with:
+ *
+ * huge <= 1 / fmin
+ * scale <= 1 / fmin / fmax (for |t| >= huge)
+ *
+ * In addition scale should be a negative power of two in order to avoid
+ * loss of precision. The values chosen below should work for most usual
+ * floating point representations with at least the dynamic range of ATI's
+ * 24-bit representation.
+ */
+ ir_constant *huge = imm(1e18f, n);
+ ir_variable *scale = body.make_temp(type, "scale");
+ body.emit(assign(scale, csel(gequal(abs(t), huge),
+ imm(0.25f, n), imm(1.0f, n))));
+ ir_variable *rcp_scaled_t = body.make_temp(type, "rcp_scaled_t");
+ body.emit(assign(rcp_scaled_t, rcp(mul(t, scale))));
+ ir_expression *s_over_t = mul(mul(s, scale), rcp_scaled_t);
+
+ /* For |x| = |y| assume tan = 1 even if infinite (i.e. pretend momentarily
+ * that ∞/∞ = 1) in order to comply with the rather artificial rules
+ * inherited from IEEE 754-2008, namely:
+ *
+ * "atan2(±∞, −∞) is ±3π/4
+ * atan2(±∞, +∞) is ±π/4"
+ *
+ * Note that this is inconsistent with the rules for the neighborhood of
+ * zero that are based on iterated limits:
+ *
+ * "atan2(±0, −0) is ±π
+ * atan2(±0, +0) is ±0"
+ *
+ * but GLSL specifically allows implementations to deviate from IEEE rules
+ * at (0,0), so we take that license (i.e. pretend that 0/0 = 1 here as
+ * well).
+ */
+ ir_expression *tan = csel(equal(abs(x), abs(y)),
+ imm(1.0f, n), abs(s_over_t));
- body.emit(assign(vec_result, r, 1 << i));
- }
- body.emit(ret(vec_result));
+ /* Calculate the arctangent and fix up the result if we had flipped the
+ * coordinate system.
+ */
+ ir_variable *arc = body.make_temp(type, "arc");
+ do_atan(body, type, arc, tan);
+ body.emit(assign(arc, add(arc, mul(b2f(flip), imm(M_PI_2f)))));
+
+ /* Rather convoluted calculation of the sign of the result. When x < 0 we
+ * cannot use fsign because we need to be able to distinguish between
+ * negative and positive zero. Unfortunately we cannot use bitwise
+ * arithmetic tricks either because of back-ends without integer support.
+ * When x >= 0 rcp_scaled_t will always be non-negative so this won't be
+ * able to distinguish between negative and positive zero, but we don't
+ * care because atan2 is continuous along the whole positive y = 0
+ * half-line, so it won't affect the result significantly.
+ */
+ body.emit(ret(csel(less(min2(y, rcp_scaled_t), imm(0.0f, n)),
+ neg(arc), arc)));
return sig;
}
ir_variable *x = in_var(type, "x");
MAKE_SIG(type, v130, 1, x);
- /* (e^x - e^(-x)) / (e^x + e^(-x)) */
- body.emit(ret(div(sub(exp(x), exp(neg(x))),
- add(exp(x), exp(neg(x))))));
+ /* tanh(x) := (0.5 * (e^x - e^(-x))) / (0.5 * (e^x + e^(-x)))
+ *
+ * With a little algebra this reduces to (e^2x - 1) / (e^2x + 1)
+ *
+ * Clamp x to (-inf, +10] to avoid precision problems. When x > 10, e^2x
+ * is so much larger than 1.0 that 1.0 gets flushed to zero in the
+ * computation e^2x +/- 1 so it can be ignored.
+ */
+ ir_variable *t = body.make_temp(type, "tmp");
+ body.emit(assign(t, min2(x, imm(10.0f))));
+
+ body.emit(ret(div(sub(exp(mul(t, imm(2.0f))), imm(1.0f)),
+ add(exp(mul(t, imm(2.0f))), imm(1.0f)))));
return sig;
}
ir_variable *t = body.make_temp(x_type, "t");
if (x_type->vector_elements == 1) {
/* Both are floats */
- if (edge_type->base_type == GLSL_TYPE_DOUBLE)
+ if (edge_type->is_double())
body.emit(assign(t, f2d(b2f(gequal(x, edge)))));
else
body.emit(assign(t, b2f(gequal(x, edge))));
} else if (edge_type->vector_elements == 1) {
/* x is a vector but edge is a float */
for (int i = 0; i < x_type->vector_elements; i++) {
- if (edge_type->base_type == GLSL_TYPE_DOUBLE)
+ if (edge_type->is_double())
body.emit(assign(t, f2d(b2f(gequal(swizzle(x, i, 1), edge))), 1 << i));
else
body.emit(assign(t, b2f(gequal(swizzle(x, i, 1), edge)), 1 << i));
} else {
/* Both are vectors */
for (int i = 0; i < x_type->vector_elements; i++) {
- if (edge_type->base_type == GLSL_TYPE_DOUBLE)
+ if (edge_type->is_double())
body.emit(assign(t, f2d(b2f(gequal(swizzle(x, i, 1), swizzle(edge, i, 1)))),
1 << i));
else
return sig;
}
+ir_function_signature *
+builtin_builder::_doubleBitsToInt64(builtin_available_predicate avail, const glsl_type *type)
+{
+ ir_variable *x = in_var(type, "x");
+ MAKE_SIG(glsl_type::i64vec(type->vector_elements), avail, 1, x);
+ body.emit(ret(bitcast_d2i64(x)));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_doubleBitsToUint64(builtin_available_predicate avail, const glsl_type *type)
+{
+ ir_variable *x = in_var(type, "x");
+ MAKE_SIG(glsl_type::u64vec(type->vector_elements), avail, 1, x);
+ body.emit(ret(bitcast_d2u64(x)));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_int64BitsToDouble(builtin_available_predicate avail, const glsl_type *type)
+{
+ ir_variable *x = in_var(type, "x");
+ MAKE_SIG(glsl_type::dvec(type->vector_elements), avail, 1, x);
+ body.emit(ret(bitcast_i642d(x)));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_uint64BitsToDouble(builtin_available_predicate avail, const glsl_type *type)
+{
+ ir_variable *x = in_var(type, "x");
+ MAKE_SIG(glsl_type::dvec(type->vector_elements), avail, 1, x);
+ body.emit(ret(bitcast_u642d(x)));
+ return sig;
+}
+
ir_function_signature *
builtin_builder::_packUnorm2x16(builtin_available_predicate avail)
{
return sig;
}
+ir_function_signature *
+builtin_builder::_packInt2x32(builtin_available_predicate avail)
+{
+ ir_variable *v = in_var(glsl_type::ivec2_type, "v");
+ MAKE_SIG(glsl_type::int64_t_type, avail, 1, v);
+ body.emit(ret(expr(ir_unop_pack_int_2x32, v)));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_unpackInt2x32(builtin_available_predicate avail)
+{
+ ir_variable *p = in_var(glsl_type::int64_t_type, "p");
+ MAKE_SIG(glsl_type::ivec2_type, avail, 1, p);
+ body.emit(ret(expr(ir_unop_unpack_int_2x32, p)));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_packUint2x32(builtin_available_predicate avail)
+{
+ ir_variable *v = in_var(glsl_type::uvec2_type, "v");
+ MAKE_SIG(glsl_type::uint64_t_type, avail, 1, v);
+ body.emit(ret(expr(ir_unop_pack_uint_2x32, v)));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_unpackUint2x32(builtin_available_predicate avail)
+{
+ ir_variable *p = in_var(glsl_type::uint64_t_type, "p");
+ MAKE_SIG(glsl_type::uvec2_type, avail, 1, p);
+ body.emit(ret(expr(ir_unop_unpack_uint_2x32, p)));
+ return sig;
+}
+
ir_function_signature *
builtin_builder::_length(builtin_available_predicate avail, const glsl_type *type)
{
{
MAKE_SIG(glsl_type::vec4_type, compatibility_vs_only, 0);
- body.emit(ret(new(mem_ctx) ir_expression(ir_binop_mul,
- glsl_type::vec4_type,
- var_ref(gl_ModelViewProjectionMatrix),
- var_ref(gl_Vertex))));
-
- /* FINISHME: Once the ir_expression() constructor handles type inference
- * for matrix operations, we can simplify this to:
- *
- * body.emit(ret(mul(gl_ModelViewProjectionMatrix, gl_Vertex)));
+ /* ftransform() refers to global variables, and is always emitted
+ * directly by ast_function.cpp. Just emit a prototype here so we
+ * can recognize calls to it.
*/
return sig;
}
ir_variable *c;
ir_variable *r;
- if (type->base_type == GLSL_TYPE_DOUBLE) {
+ if (type->is_double()) {
r = in_var(glsl_type::dvec(type->matrix_columns), "r");
c = in_var(glsl_type::dvec(type->vector_elements), "c");
} else {
builtin_builder::_lessThanEqual(builtin_available_predicate avail,
const glsl_type *type)
{
- return binop(avail, ir_binop_lequal,
- glsl_type::bvec(type->vector_elements), type, type);
+ return binop(avail, ir_binop_gequal,
+ glsl_type::bvec(type->vector_elements), type, type,
+ true);
}
ir_function_signature *
builtin_builder::_greaterThan(builtin_available_predicate avail,
const glsl_type *type)
{
- return binop(avail, ir_binop_greater,
- glsl_type::bvec(type->vector_elements), type, type);
+ return binop(avail, ir_binop_less,
+ glsl_type::bvec(type->vector_elements), type, type,
+ true);
}
ir_function_signature *
}
ir_function_signature *
-builtin_builder::_textureSamples(const glsl_type *sampler_type)
+builtin_builder::_textureSamples(builtin_available_predicate avail,
+ const glsl_type *sampler_type)
{
ir_variable *s = in_var(sampler_type, "sampler");
- MAKE_SIG(glsl_type::int_type, shader_samples, 1, s);
+ MAKE_SIG(glsl_type::int_type, avail, 1, s);
ir_texture *tex = new(mem_ctx) ir_texture(ir_texture_samples);
tex->set_sampler(new(mem_ctx) ir_dereference_variable(s), glsl_type::int_type);
if (coord_size == coord_type->vector_elements) {
tex->coordinate = var_ref(P);
} else {
- /* The incoming coordinate also has the projector or shadow comparitor,
+ /* The incoming coordinate also has the projector or shadow comparator,
* so we need to swizzle those away.
*/
tex->coordinate = swizzle_for_size(P, coord_size);
*/
ir_variable *refz = in_var(glsl_type::float_type, "refz");
sig->parameters.push_tail(refz);
- tex->shadow_comparitor = var_ref(refz);
+ tex->shadow_comparator = var_ref(refz);
} else {
- /* The shadow comparitor is normally in the Z component, but a few types
+ /* The shadow comparator is normally in the Z component, but a few types
* have sufficiently large coordinates that it's in W.
*/
- tex->shadow_comparitor = swizzle(P, MAX2(coord_size, SWIZZLE_Z), 1);
+ tex->shadow_comparator = swizzle(P, MAX2(coord_size, SWIZZLE_Z), 1);
}
}
}
ir_function_signature *
-builtin_builder::_textureCubeArrayShadow()
+builtin_builder::_textureCubeArrayShadow(builtin_available_predicate avail,
+ const glsl_type *sampler_type)
{
- ir_variable *s = in_var(glsl_type::samplerCubeArrayShadow_type, "sampler");
+ ir_variable *s = in_var(sampler_type, "sampler");
ir_variable *P = in_var(glsl_type::vec4_type, "P");
ir_variable *compare = in_var(glsl_type::float_type, "compare");
- MAKE_SIG(glsl_type::float_type, texture_cube_map_array, 3, s, P, compare);
+ MAKE_SIG(glsl_type::float_type, avail, 3, s, P, compare);
ir_texture *tex = new(mem_ctx) ir_texture(ir_tex);
tex->set_sampler(var_ref(s), glsl_type::float_type);
tex->coordinate = var_ref(P);
- tex->shadow_comparitor = var_ref(compare);
+ tex->shadow_comparator = var_ref(compare);
body.emit(ret(tex));
}
ir_function_signature *
-builtin_builder::_textureQueryLevels(const glsl_type *sampler_type)
+builtin_builder::_textureQueryLevels(builtin_available_predicate avail,
+ const glsl_type *sampler_type)
{
ir_variable *s = in_var(sampler_type, "sampler");
const glsl_type *return_type = glsl_type::int_type;
- MAKE_SIG(return_type, texture_query_levels, 1, s);
+ MAKE_SIG(return_type, avail, 1, s);
ir_texture *tex = new(mem_ctx) ir_texture(ir_query_levels);
tex->set_sampler(var_ref(s), return_type);
ir_variable *value = in_var(type, "value");
ir_variable *offset = in_var(glsl_type::int_type, "offset");
ir_variable *bits = in_var(glsl_type::int_type, "bits");
- MAKE_SIG(type, gpu_shader5_or_es31, 3, value, offset, bits);
+ MAKE_SIG(type, gpu_shader5_or_es31_or_integer_functions, 3, value, offset,
+ bits);
operand cast_offset = is_uint ? i2u(offset) : operand(offset);
operand cast_bits = is_uint ? i2u(bits) : operand(bits);
ir_variable *insert = in_var(type, "insert");
ir_variable *offset = in_var(glsl_type::int_type, "offset");
ir_variable *bits = in_var(glsl_type::int_type, "bits");
- MAKE_SIG(type, gpu_shader5_or_es31, 4, base, insert, offset, bits);
+ MAKE_SIG(type, gpu_shader5_or_es31_or_integer_functions, 4, base, insert,
+ offset, bits);
operand cast_offset = is_uint ? i2u(offset) : operand(offset);
operand cast_bits = is_uint ? i2u(bits) : operand(bits);
return sig;
}
-UNOP(bitfieldReverse, ir_unop_bitfield_reverse, gpu_shader5_or_es31)
+UNOP(bitfieldReverse, ir_unop_bitfield_reverse, gpu_shader5_or_es31_or_integer_functions)
ir_function_signature *
builtin_builder::_bitCount(const glsl_type *type)
{
- return unop(gpu_shader5_or_es31, ir_unop_bit_count,
+ return unop(gpu_shader5_or_es31_or_integer_functions, ir_unop_bit_count,
glsl_type::ivec(type->vector_elements), type);
}
ir_function_signature *
builtin_builder::_findLSB(const glsl_type *type)
{
- return unop(gpu_shader5_or_es31, ir_unop_find_lsb,
+ return unop(gpu_shader5_or_es31_or_integer_functions, ir_unop_find_lsb,
glsl_type::ivec(type->vector_elements), type);
}
ir_function_signature *
builtin_builder::_findMSB(const glsl_type *type)
{
- return unop(gpu_shader5_or_es31, ir_unop_find_msb,
+ return unop(gpu_shader5_or_es31_or_integer_functions, ir_unop_find_msb,
glsl_type::ivec(type->vector_elements), type);
}
ir_function_signature *
builtin_builder::_ldexp(const glsl_type *x_type, const glsl_type *exp_type)
{
- return binop(x_type->base_type == GLSL_TYPE_DOUBLE ? fp64 : gpu_shader5_or_es31,
+ return binop(x_type->is_double() ? fp64 : gpu_shader5_or_es31_or_integer_functions,
ir_binop_ldexp, x_type, x_type, exp_type);
}
{
ir_variable *x = in_var(x_type, "x");
ir_variable *exponent = out_var(exp_type, "exp");
- MAKE_SIG(x_type, gpu_shader5_or_es31, 2, x, exponent);
+ MAKE_SIG(x_type, gpu_shader5_or_es31_or_integer_functions, 2, x, exponent);
const unsigned vec_elem = x_type->vector_elements;
const glsl_type *bvec = glsl_type::get_instance(GLSL_TYPE_BOOL, vec_elem, 1);
ir_variable *x = in_var(type, "x");
ir_variable *y = in_var(type, "y");
ir_variable *carry = out_var(type, "carry");
- MAKE_SIG(type, gpu_shader5_or_es31, 3, x, y, carry);
+ MAKE_SIG(type, gpu_shader5_or_es31_or_integer_functions, 3, x, y, carry);
body.emit(assign(carry, ir_builder::carry(x, y)));
body.emit(ret(add(x, y)));
ir_variable *x = in_var(type, "x");
ir_variable *y = in_var(type, "y");
ir_variable *borrow = out_var(type, "borrow");
- MAKE_SIG(type, gpu_shader5_or_es31, 3, x, y, borrow);
+ MAKE_SIG(type, gpu_shader5_or_es31_or_integer_functions, 3, x, y, borrow);
body.emit(assign(borrow, ir_builder::borrow(x, y)));
body.emit(ret(sub(x, y)));
ir_variable *y = in_var(type, "y");
ir_variable *msb = out_var(type, "msb");
ir_variable *lsb = out_var(type, "lsb");
- MAKE_SIG(glsl_type::void_type, gpu_shader5_or_es31, 4, x, y, msb, lsb);
+ MAKE_SIG(glsl_type::void_type, gpu_shader5_or_es31_or_integer_functions, 4, x, y, msb, lsb);
body.emit(assign(msb, imul_high(x, y)));
body.emit(assign(lsb, mul(x, y)));
{
ir_variable *interpolant = in_var(type, "interpolant");
interpolant->data.must_be_shader_input = 1;
- MAKE_SIG(type, fs_gpu_shader5, 1, interpolant);
+ MAKE_SIG(type, fs_interpolate_at, 1, interpolant);
body.emit(ret(interpolate_at_centroid(interpolant)));
ir_variable *interpolant = in_var(type, "interpolant");
interpolant->data.must_be_shader_input = 1;
ir_variable *offset = in_var(glsl_type::vec2_type, "offset");
- MAKE_SIG(type, fs_gpu_shader5, 2, interpolant, offset);
+ MAKE_SIG(type, fs_interpolate_at, 2, interpolant, offset);
body.emit(ret(interpolate_at_offset(interpolant, offset)));
ir_variable *interpolant = in_var(type, "interpolant");
interpolant->data.must_be_shader_input = 1;
ir_variable *sample_num = in_var(glsl_type::int_type, "sample_num");
- MAKE_SIG(type, fs_gpu_shader5, 2, interpolant, sample_num);
+ MAKE_SIG(type, fs_interpolate_at, 2, interpolant, sample_num);
body.emit(ret(interpolate_at_sample(interpolant, sample_num)));
}
ir_function_signature *
-builtin_builder::_atomic_counter_intrinsic(builtin_available_predicate avail)
+builtin_builder::_atomic_counter_intrinsic(builtin_available_predicate avail,
+ enum ir_intrinsic_id id)
+{
+ ir_variable *counter = in_var(glsl_type::atomic_uint_type, "counter");
+ MAKE_INTRINSIC(glsl_type::uint_type, id, avail, 1, counter);
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_atomic_counter_intrinsic1(builtin_available_predicate avail,
+ enum ir_intrinsic_id id)
{
ir_variable *counter = in_var(glsl_type::atomic_uint_type, "counter");
- MAKE_INTRINSIC(glsl_type::uint_type, avail, 1, counter);
+ ir_variable *data = in_var(glsl_type::uint_type, "data");
+ MAKE_INTRINSIC(glsl_type::uint_type, id, avail, 2, counter, data);
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_atomic_counter_intrinsic2(builtin_available_predicate avail,
+ enum ir_intrinsic_id id)
+{
+ ir_variable *counter = in_var(glsl_type::atomic_uint_type, "counter");
+ ir_variable *compare = in_var(glsl_type::uint_type, "compare");
+ ir_variable *data = in_var(glsl_type::uint_type, "data");
+ MAKE_INTRINSIC(glsl_type::uint_type, id, avail, 3, counter, compare, data);
return sig;
}
ir_function_signature *
builtin_builder::_atomic_intrinsic2(builtin_available_predicate avail,
- const glsl_type *type)
+ const glsl_type *type,
+ enum ir_intrinsic_id id)
{
ir_variable *atomic = in_var(type, "atomic");
ir_variable *data = in_var(type, "data");
- MAKE_INTRINSIC(type, avail, 2, atomic, data);
+ MAKE_INTRINSIC(type, id, avail, 2, atomic, data);
return sig;
}
ir_function_signature *
builtin_builder::_atomic_intrinsic3(builtin_available_predicate avail,
- const glsl_type *type)
+ const glsl_type *type,
+ enum ir_intrinsic_id id)
{
ir_variable *atomic = in_var(type, "atomic");
ir_variable *data1 = in_var(type, "data1");
ir_variable *data2 = in_var(type, "data2");
- MAKE_INTRINSIC(type, avail, 3, atomic, data1, data2);
+ MAKE_INTRINSIC(type, id, avail, 3, atomic, data1, data2);
return sig;
}
return sig;
}
+ir_function_signature *
+builtin_builder::_atomic_counter_op1(const char *intrinsic,
+ builtin_available_predicate avail)
+{
+ ir_variable *counter = in_var(glsl_type::atomic_uint_type, "atomic_counter");
+ ir_variable *data = in_var(glsl_type::uint_type, "data");
+ MAKE_SIG(glsl_type::uint_type, avail, 2, counter, data);
+
+ ir_variable *retval = body.make_temp(glsl_type::uint_type, "atomic_retval");
+
+ /* Instead of generating an __intrinsic_atomic_sub, generate an
+ * __intrinsic_atomic_add with the data parameter negated.
+ */
+ if (strcmp("__intrinsic_atomic_sub", intrinsic) == 0) {
+ ir_variable *const neg_data =
+ body.make_temp(glsl_type::uint_type, "neg_data");
+
+ body.emit(assign(neg_data, neg(data)));
+
+ exec_list parameters;
+
+ parameters.push_tail(new(mem_ctx) ir_dereference_variable(counter));
+ parameters.push_tail(new(mem_ctx) ir_dereference_variable(neg_data));
+
+ ir_function *const func =
+ shader->symbols->get_function("__intrinsic_atomic_add");
+ ir_instruction *const c = call(func, retval, parameters);
+
+ assert(c != NULL);
+ assert(parameters.is_empty());
+
+ body.emit(c);
+ } else {
+ body.emit(call(shader->symbols->get_function(intrinsic), retval,
+ sig->parameters));
+ }
+
+ body.emit(ret(retval));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_atomic_counter_op2(const char *intrinsic,
+ builtin_available_predicate avail)
+{
+ ir_variable *counter = in_var(glsl_type::atomic_uint_type, "atomic_counter");
+ ir_variable *compare = in_var(glsl_type::uint_type, "compare");
+ ir_variable *data = in_var(glsl_type::uint_type, "data");
+ MAKE_SIG(glsl_type::uint_type, avail, 3, counter, compare, data);
+
+ ir_variable *retval = body.make_temp(glsl_type::uint_type, "atomic_retval");
+ body.emit(call(shader->symbols->get_function(intrinsic), retval,
+ sig->parameters));
+ body.emit(ret(retval));
+ return sig;
+}
+
ir_function_signature *
builtin_builder::_atomic_op2(const char *intrinsic,
builtin_available_predicate avail,
return sig;
}
+static builtin_available_predicate
+get_image_available_predicate(const glsl_type *type, unsigned flags)
+{
+ if ((flags & IMAGE_FUNCTION_AVAIL_ATOMIC_EXCHANGE) &&
+ type->sampled_type == GLSL_TYPE_FLOAT)
+ return shader_image_atomic_exchange_float;
+
+ if ((flags & IMAGE_FUNCTION_AVAIL_ATOMIC_ADD) &&
+ type->sampled_type == GLSL_TYPE_FLOAT)
+ return shader_image_atomic_add_float;
+
+ else if (flags & (IMAGE_FUNCTION_AVAIL_ATOMIC_EXCHANGE |
+ IMAGE_FUNCTION_AVAIL_ATOMIC_ADD |
+ IMAGE_FUNCTION_AVAIL_ATOMIC))
+ return shader_image_atomic;
+
+ else
+ return shader_image_load_store;
+}
+
ir_function_signature *
builtin_builder::_image_prototype(const glsl_type *image_type,
unsigned num_arguments,
unsigned flags)
{
const glsl_type *data_type = glsl_type::get_instance(
- image_type->sampler_type,
+ image_type->sampled_type,
(flags & IMAGE_FUNCTION_HAS_VECTOR_DATA_TYPE ? 4 : 1),
1);
const glsl_type *ret_type = (flags & IMAGE_FUNCTION_RETURNS_VOID ?
ir_variable *coord = in_var(
glsl_type::ivec(image_type->coordinate_components()), "coord");
- const builtin_available_predicate avail =
- (flags & IMAGE_FUNCTION_AVAIL_ATOMIC ? shader_image_atomic :
- shader_image_load_store);
- ir_function_signature *sig = new_sig(ret_type, avail, 2, image, coord);
+ ir_function_signature *sig = new_sig(
+ ret_type, get_image_available_predicate(image_type, flags),
+ 2, image, coord);
/* Sample index for multisample images. */
if (image_type->sampler_dimensionality == GLSL_SAMPLER_DIM_MS)
* accept everything that needs to be accepted, and reject cases
* like loads from write-only or stores to read-only images.
*/
- image->data.image_read_only = (flags & IMAGE_FUNCTION_READ_ONLY) != 0;
- image->data.image_write_only = (flags & IMAGE_FUNCTION_WRITE_ONLY) != 0;
- image->data.image_coherent = true;
- image->data.image_volatile = true;
- image->data.image_restrict = true;
+ image->data.memory_read_only = (flags & IMAGE_FUNCTION_READ_ONLY) != 0;
+ image->data.memory_write_only = (flags & IMAGE_FUNCTION_WRITE_ONLY) != 0;
+ image->data.memory_coherent = true;
+ image->data.memory_volatile = true;
+ image->data.memory_restrict = true;
return sig;
}
* accept everything that needs to be accepted, and reject cases
* like loads from write-only or stores to read-only images.
*/
- image->data.image_read_only = true;
- image->data.image_write_only = true;
- image->data.image_coherent = true;
- image->data.image_volatile = true;
- image->data.image_restrict = true;
+ image->data.memory_read_only = true;
+ image->data.memory_write_only = true;
+ image->data.memory_coherent = true;
+ image->data.memory_volatile = true;
+ image->data.memory_restrict = true;
return sig;
}
* accept everything that needs to be accepted, and reject cases
* like loads from write-only or stores to read-only images.
*/
- image->data.image_read_only = true;
- image->data.image_write_only = true;
- image->data.image_coherent = true;
- image->data.image_volatile = true;
- image->data.image_restrict = true;
+ image->data.memory_read_only = true;
+ image->data.memory_write_only = true;
+ image->data.memory_coherent = true;
+ image->data.memory_volatile = true;
+ image->data.memory_restrict = true;
return sig;
}
const glsl_type *image_type,
const char *intrinsic_name,
unsigned num_arguments,
- unsigned flags)
+ unsigned flags,
+ enum ir_intrinsic_id id)
{
ir_function_signature *sig = (this->*prototype)(image_type,
num_arguments, flags);
sig->is_defined = true;
} else {
- sig->is_intrinsic = true;
+ sig->intrinsic_id = id;
}
return sig;
}
ir_function_signature *
-builtin_builder::_memory_barrier_intrinsic(builtin_available_predicate avail)
+builtin_builder::_memory_barrier_intrinsic(builtin_available_predicate avail,
+ enum ir_intrinsic_id id)
{
- MAKE_INTRINSIC(glsl_type::void_type, avail, 0);
+ MAKE_INTRINSIC(glsl_type::void_type, id, avail, 0);
return sig;
}
return sig;
}
+ir_function_signature *
+builtin_builder::_ballot_intrinsic()
+{
+ ir_variable *value = in_var(glsl_type::bool_type, "value");
+ MAKE_INTRINSIC(glsl_type::uint64_t_type, ir_intrinsic_ballot, shader_ballot,
+ 1, value);
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_ballot()
+{
+ ir_variable *value = in_var(glsl_type::bool_type, "value");
+
+ MAKE_SIG(glsl_type::uint64_t_type, shader_ballot, 1, value);
+ ir_variable *retval = body.make_temp(glsl_type::uint64_t_type, "retval");
+
+ body.emit(call(shader->symbols->get_function("__intrinsic_ballot"),
+ retval, sig->parameters));
+ body.emit(ret(retval));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_read_first_invocation_intrinsic(const glsl_type *type)
+{
+ ir_variable *value = in_var(type, "value");
+ MAKE_INTRINSIC(type, ir_intrinsic_read_first_invocation, shader_ballot,
+ 1, value);
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_read_first_invocation(const glsl_type *type)
+{
+ ir_variable *value = in_var(type, "value");
+
+ MAKE_SIG(type, shader_ballot, 1, value);
+ ir_variable *retval = body.make_temp(type, "retval");
+
+ body.emit(call(shader->symbols->get_function("__intrinsic_read_first_invocation"),
+ retval, sig->parameters));
+ body.emit(ret(retval));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_read_invocation_intrinsic(const glsl_type *type)
+{
+ ir_variable *value = in_var(type, "value");
+ ir_variable *invocation = in_var(glsl_type::uint_type, "invocation");
+ MAKE_INTRINSIC(type, ir_intrinsic_read_invocation, shader_ballot,
+ 2, value, invocation);
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_read_invocation(const glsl_type *type)
+{
+ ir_variable *value = in_var(type, "value");
+ ir_variable *invocation = in_var(glsl_type::uint_type, "invocation");
+
+ MAKE_SIG(type, shader_ballot, 2, value, invocation);
+ ir_variable *retval = body.make_temp(type, "retval");
+
+ body.emit(call(shader->symbols->get_function("__intrinsic_read_invocation"),
+ retval, sig->parameters));
+ body.emit(ret(retval));
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_invocation_interlock_intrinsic(builtin_available_predicate avail,
+ enum ir_intrinsic_id id)
+{
+ MAKE_INTRINSIC(glsl_type::void_type, id, avail, 0);
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_invocation_interlock(const char *intrinsic_name,
+ builtin_available_predicate avail)
+{
+ MAKE_SIG(glsl_type::void_type, avail, 0);
+ body.emit(call(shader->symbols->get_function(intrinsic_name),
+ NULL, sig->parameters));
+ return sig;
+}
+
ir_function_signature *
builtin_builder::_shader_clock_intrinsic(builtin_available_predicate avail,
const glsl_type *type)
{
- MAKE_INTRINSIC(type, avail, 0);
+ MAKE_INTRINSIC(type, ir_intrinsic_shader_clock, avail, 0);
return sig;
}
{
MAKE_SIG(type, avail, 0);
- ir_variable *retval = body.make_temp(type, "clock_retval");
+ ir_variable *retval = body.make_temp(glsl_type::uvec2_type, "clock_retval");
body.emit(call(shader->symbols->get_function("__intrinsic_shader_clock"),
retval, sig->parameters));
+
+ if (type == glsl_type::uint64_t_type) {
+ body.emit(ret(expr(ir_unop_pack_uint_2x32, retval)));
+ } else {
+ body.emit(ret(retval));
+ }
+
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_vote_intrinsic(builtin_available_predicate avail,
+ enum ir_intrinsic_id id)
+{
+ ir_variable *value = in_var(glsl_type::bool_type, "value");
+ MAKE_INTRINSIC(glsl_type::bool_type, id, avail, 1, value);
+ return sig;
+}
+
+ir_function_signature *
+builtin_builder::_vote(const char *intrinsic_name,
+ builtin_available_predicate avail)
+{
+ ir_variable *value = in_var(glsl_type::bool_type, "value");
+
+ MAKE_SIG(glsl_type::bool_type, avail, 1, value);
+
+ ir_variable *retval = body.make_temp(glsl_type::bool_type, "retval");
+
+ body.emit(call(shader->symbols->get_function(intrinsic_name),
+ retval, sig->parameters));
body.emit(ret(retval));
return sig;
}
_mesa_glsl_find_builtin_function(_mesa_glsl_parse_state *state,
const char *name, exec_list *actual_parameters)
{
- ir_function_signature * s;
+ ir_function_signature *s;
mtx_lock(&builtins_lock);
s = builtins.find(state, name, actual_parameters);
mtx_unlock(&builtins_lock);
+
return s;
}
-ir_function *
-_mesa_glsl_find_builtin_function_by_name(const char *name)
+bool
+_mesa_glsl_has_builtin_function(_mesa_glsl_parse_state *state, const char *name)
{
ir_function *f;
+ bool ret = false;
mtx_lock(&builtins_lock);
f = builtins.shader->symbols->get_function(name);
+ if (f != NULL) {
+ foreach_in_list(ir_function_signature, sig, &f->signatures) {
+ if (sig->is_builtin_available(state)) {
+ ret = true;
+ break;
+ }
+ }
+ }
mtx_unlock(&builtins_lock);
- return f;
+
+ return ret;
}
gl_shader *
* Get the function signature for main from a shader
*/
ir_function_signature *
-_mesa_get_main_function_signature(gl_shader *sh)
+_mesa_get_main_function_signature(glsl_symbol_table *symbols)
{
- ir_function *const f = sh->symbols->get_function("main");
+ ir_function *const f = symbols->get_function("main");
if (f != NULL) {
exec_list void_parameters;
projects / mesa.git / blobdiff