fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-21 11:30:11 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1
mesa/src/compiler/glsl/glsl_parser_extras.cpp

1974 lines
64 KiB
C++
Raw Normal View History

Initial commit. lol
2010-02-22 13:19:34 -08:00
/*
* Copyright © 2008, 2009 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
#include <stdio.h>
#include <stdarg.h>
#include <string.h>
#include <assert.h>
Drop GLcontext typedef and use struct gl_context instead
2010-10-12 12:26:10 -04:00
#include "main/core.h" /* for struct gl_context */
glsl: add support for using API_OPENGL_CORE Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Eric Anholt <eric@anholt.net> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2012-07-24 14:48:57 -07:00
#include "main/context.h"
mesa: move GL_ARB_debug_output code into new debug_output.c file The errors.c file had grown quite large so split off this extension code into its own file. This involved making a handful of functions non-static. Acked-by: Timothy Arceri <timothy.arceri@collabora.com>
2016-02-08 09:29:38 -07:00
#include "main/debug_output.h"
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
#include "main/shaderobj.h"
glsl: Initialize static temporaries_allocate_names once per process. Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
2014-11-21 18:04:21 -08:00
#include "util/u_atomic.h" /* for p_atomic_cmpxchg */
util: Move ralloc to a new src/util directory. For a long time, we've wanted a place to put utility code which isn't directly tied to Mesa or Gallium internals. This patch creates a new src/util directory for exactly that purpose, and builds the contents as libmesautil.la. ralloc seemed like a good first candidate. These days, it's directly used by mesa/main, i965, i915, and r300g, so keeping it in src/glsl didn't make much sense. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> v2 (Jason Ekstrand): More realloc uses and some scons fixes Signed-off-by: Jason Ekstrand <jason.ekstrand@intel.com> Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2014-02-24 23:39:14 -08:00
#include "util/ralloc.h"
Initial commit. lol
2010-02-22 13:19:34 -08:00
#include "ast.h"
#include "glsl_parser_extras.h"
autoconf for the ...
2010-02-25 17:17:23 -08:00
#include "glsl_parser.h"
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
#include "ir_optimization.h"
glsl2: Perform initial bits of loop analysis during compilation
2010-08-26 16:45:22 -07:00
#include "loop_analysis.h"
Initial commit. lol
2010-02-22 13:19:34 -08:00
glsl: Make a function to express a GLSL version ir human-readable form. This will be useful in generating more helpful error messages, especially with the addition of GLSL 3.00 ES support. [v2, idr]: Rename ctx parameter to mem_ctx Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-01 18:36:57 -07:00
/**
* Format a short human-readable description of the given GLSL version.
*/
const char *
glsl_compute_version_string(void *mem_ctx, bool is_es, unsigned version)
{
return ralloc_asprintf(mem_ctx, "GLSL%s %d.%02d", is_es ? " ES" : "",
version / 100, version % 100);
}
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
glsl: make static constant variables "static const" This allows them to be moved to .rodata, and allow us to be sure that they will not be modified. Signed-off-by: Chia-I Wu <olv@lunarg.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Timothy Arceri <t_arceri@yahoo.com.au>
2014-04-22 16:58:16 +08:00
static const unsigned known_desktop_glsl_versions[] =
main, glsl: Bump max known desktop glsl version to 4.50 Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-11-24 13:56:29 -08:00
{ 110, 120, 130, 140, 150, 330, 400, 410, 420, 430, 440, 450 };
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
glsl: add gl_context member Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 16:14:31 +00:00
_mesa_glsl_parse_state::_mesa_glsl_parse_state(struct gl_context *_ctx,
glsl: Change _mesa_glsl_parse_state ctor to use gl_shader_stage enum. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> v2: Also rename "target" param to "stage". Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-07 09:46:10 -08:00
gl_shader_stage stage,
void *mem_ctx)
glsl/cs: Handle compute shader local_size_{x,y,z} declaration. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 09:09:31 -08:00
: ctx(_ctx), cs_input_local_size_specified(false), cs_input_local_size(),
switch_state()
glsl2: Add a constructor for _mesa_glsl_parse_state Coming changes to the handling of built-in functions necessitate this.
2010-07-18 15:59:43 -07:00
{
glsl: Change _mesa_glsl_parse_state ctor to use gl_shader_stage enum. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> v2: Also rename "target" param to "stage". Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-07 09:46:10 -08:00
assert(stage < MESA_SHADER_STAGES);
this->stage = stage;
glsl2: Add a constructor for _mesa_glsl_parse_state Coming changes to the handling of built-in functions necessitate this.
2010-07-18 15:59:43 -07:00
this->scanner = NULL;
this->translation_unit.make_empty();
this->symbols = new(mem_ctx) glsl_symbol_table;
glsl: Initialize all member variables of _mesa_glsl_parse_state on construction. The _mesa_glsl_parse_state object relies on the memory allocator zeroing out its contents before it's initialized, which is quite an unusual practice in the C++ world because it ties objects to some specific allocation scheme, and gives unpredictable results when an object is created with a different allocator -- Stack allocation, array allocation, or aggregation inside a different object are some of the useful possibilities that come to my mind. Initialize all fields from the constructor and stop using the zeroing allocator. Reviewed-by: Paul Berry <stereotype441@gmail.com> Reviewed-by: Chad Versace <chad.versace@linux.intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-09-11 20:07:53 -07:00
Convert everything from the talloc API to the ralloc API.
2011-01-21 14:32:31 -08:00
this->info_log = ralloc_strdup(mem_ctx, "");
glsl2: Add a constructor for _mesa_glsl_parse_state Coming changes to the handling of built-in functions necessitate this.
2010-07-18 15:59:43 -07:00
this->error = false;
glsl: Generate IR for switch statements Up until now modifying the GLSL compiler has been pretty straightforward. This is where things get interesting. But still pretty straightforward. Switch statements can be thought of a series of if/then/else statements. Case labels are compared with the value of a test expression and the case statements are executed if the comparison is true. There are a couple of aspects of switch statements that complicate this simple view of the world. The primary one is that cases can fall through sequentially to subsequent case, unless a break statement is encountered, in which case, the switch statement exits completely. But break handling is further complicated by the fact that a break statement can impact the exit of a loop. Thus, we need to coordinate break processing between switch statements and loop statements. The code generated by a switch statement maintains three temporary state variables: int test_value; bool is_fallthru; bool is_break; test_value is initialized to the value of the test expression at the head of the switch statement. This is the value that case labels are compared against. is_fallthru is used to sequentially fall through to subsequent cases and is initialized to false. When a case label matches the test expression, this state variable is set to true. It will also be forced to false if a break statement has been encountered. This forcing to false on break MUST be after every case test. In practice, we defer that forcing to immediately after the last case comparison prior to executing a case statement, but that is an optimization. is_break is used to indicate that a break statement has been executed and is initialized to false. When a break statement is encountered, it is set to true. This state variable is then used to conditionally force is_fallthru to to false to prevent subsequent case statements from executing. Code generation for break statements depends on whether the break statement is inside a switch statement or inside a loop statement. If it inside a loop statement is inside a break statement, the same code as before gets generated. But if a switch statement is inside a loop statement, code is emitted to set the is_break state to true. Just as ASTs for loop statements are managed in a stack-like manner to handle nesting, we also add a bool to capture the innermost switch or loop condition. Note that we still need to maintain a loop AST stack to properly handle for-loop code generation on a continue statement. Technically, we don't (yet) need a switch AST stack, but I am using one for orthogonality with loop statements, in anticipation of future use. Note that a simple boolean stack would have sufficed. We will illustrate a switch statement with its analogous conditional code that a switch statement corresponds to by examining an example. Consider the following switch statement: switch (42) { case 0: case 1: gl_FragColor = vec4(1.0, 2.0, 3.0, 4.0); case 2: case 3: gl_FragColor = vec4(4.0, 3.0, 2.0, 1.0); break; case 4: default: gl_FragColor = vec4(0.0, 0.0, 0.0, 0.0); } Note that case 0 and case 1 fall through to cases 2 and 3 if they occur. Note that case 4 and the default case must be reached explicitly, since cases 2 and 3 break at the end of their case. Finally, note that case 4 and the default case don't break but simply fall through to the end of the switch. For this code, the equivalent code can be expressed as: int test_val = 42; // capture value of test expression bool is_fallthru = false; // prevent initial fall through bool is_break = false; // capture the execution of a break stmt is_fallthru |= (test_val == 0); // enable fallthru on case 0 is_fallthru |= (test_val == 1); // enable fallthru on case 1 is_fallthru &= !is_break; // inhibit fallthru on previous break if (is_fallthru) { gl_FragColor = vec4(1.0, 2.0, 3.0, 4.0); } is_fallthru |= (test_val == 2); // enable fallthru on case 2 is_fallthru |= (test_val == 3); // enable fallthru on case 3 is_fallthru &= !is_break; // inhibit fallthru on previous break if (is_fallthru) { gl_FragColor = vec4(4.0, 3.0, 2.0, 1.0); is_break = true; // inhibit all subsequent fallthru for break } is_fallthru |= (test_val == 4); // enable fallthru on case 4 is_fallthru = true; // enable fallthru for default case is_fallthru &= !is_break; // inhibit fallthru on previous break if (is_fallthru) { gl_FragColor = vec4(0.0, 0.0, 0.0, 0.0); } The code generate for |= and &= uses the conditional assignment capabilities of the IR. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2011-11-07 16:17:58 -08:00
this->loop_nesting_ast = NULL;
glsl: Set default language version in mesa_glsl_parse_state constructor. This should make it easier to change the default version based on the API (say, version 1.00 for OpenGL ES). Also, synchronize the symbol table's version with the parse state's version just before doing AST-to-HIR. This way, it will be set when it matters, but the main initialization code doesn't have to care about the symbol table.
2010-09-05 00:31:28 -07:00
glsl: Simplify the built-in function linking code. Previously, we stored an array of up to 16 additional shaders to link, as well as a count of how many each shader actually needed. Since the built-in functions rewrite, all the built-ins are stored in a single shader. So all we need is a boolean indicating whether a shader needs to link against built-ins or not. During linking, we can avoid creating the temporary array if none of the shaders being linked need built-ins. Otherwise, it's simply a copy of the array that has one additional element. This is much simpler. This patch saves approximately 128 bytes of memory per gl_shader object. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-11-23 12:11:34 -08:00
this->uses_builtin_functions = false;
glsl: Defer initialization of built-in functions until they're needed. Very simple shaders don't actually use GLSL built-ins. For example: - gl_Position = gl_ModelViewProjectionMatrix * gl_Vertex; - gl_FragColor = vec4(0.0); Both of the shaders used by _mesa_meta_glsl_Clear() also qualify. By waiting to initialize the built-ins until the first time we need to look for a signature, we can avoid the overhead entirely in these cases. Makes piglit run roughly 18% faster (255 vs. 312 seconds). Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Eric Anholt <eric@anholt.net>
2011-09-19 18:30:15 -07:00
glsl: Set default language version in mesa_glsl_parse_state constructor. This should make it easier to change the default version based on the API (say, version 1.00 for OpenGL ES). Also, synchronize the symbol table's version with the parse state's version just before doing AST-to-HIR. This way, it will be set when it matters, but the main initialization code doesn't have to care about the symbol table.
2010-09-05 00:31:28 -07:00
/* Set default language version and extensions */
glsl: allow ForceGLSLVersion to override #version directives Previously, the ctx->Const.ForceGLSLVersion setting only worked if the shader lacked a #version directive. Now, the ForceGLSLVersion setting will override the #version directive too. This change should be safe since it should be rare to have an app that has a mix of shader versions and we only wanted to override the #version for shaders which lacked the #version directive. Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-03-27 10:54:10 -06:00
this->language_version = 110;
this->forced_language_version = ctx->Const.ForceGLSLVersion;
glsl: Accept language version 100 and make it the default on ES2.
2010-08-16 12:34:53 -07:00
this->es_shader = false;
glsl2: Add a constructor for _mesa_glsl_parse_state Coming changes to the handling of built-in functions necessitate this.
2010-07-18 15:59:43 -07:00
this->ARB_texture_rectangle_enable = true;
glsl: Require a context in _mesa_glsl_parse_state. Create a dummy context in the standalone compiler and pass it to _mesa_glsl_parse_state.
2010-09-08 18:48:12 +08:00
/* OpenGL ES 2.0 has different defaults from desktop GL. */
if (ctx->API == API_OPENGLES2) {
this->language_version = 100;
this->es_shader = true;
this->ARB_texture_rectangle_enable = false;
glsl2: Add a constructor for _mesa_glsl_parse_state Coming changes to the handling of built-in functions necessitate this.
2010-07-18 15:59:43 -07:00
}
glsl: Require a context in _mesa_glsl_parse_state. Create a dummy context in the standalone compiler and pass it to _mesa_glsl_parse_state.
2010-09-08 18:48:12 +08:00
this->extensions = &ctx->Extensions;
this->Const.MaxLights = ctx->Const.MaxLights;
this->Const.MaxClipPlanes = ctx->Const.MaxClipPlanes;
this->Const.MaxTextureUnits = ctx->Const.MaxTextureUnits;
this->Const.MaxTextureCoords = ctx->Const.MaxTextureCoordUnits;
mesa: replace ctx->Const.{Vertex,Fragment,Geomtery}Program with an array. These are replaced with ctx->Const.Program[MESA_SHADER_{VERTEX,FRAGMENT,GEOMETRY}]. In patches to follow, this will allow us to replace a lot of ad-hoc logic with a variable index into the array. With the exception of the changes to mtypes.h, this patch was generated entirely by the command: find src -type f '(' -iname '*.c' -o -iname '*.cpp' -o -iname '*.py' \ -o -iname '*.y' ')' -print0 | xargs -0 sed -i \ -e 's/Const\.VertexProgram/Const.Program[MESA_SHADER_VERTEX]/g' \ -e 's/Const\.GeometryProgram/Const.Program[MESA_SHADER_GEOMETRY]/g' \ -e 's/Const\.FragmentProgram/Const.Program[MESA_SHADER_FRAGMENT]/g' Suggested-by: Brian Paul <brianp@vmware.com> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-01-08 10:00:28 -08:00
this->Const.MaxVertexAttribs = ctx->Const.Program[MESA_SHADER_VERTEX].MaxAttribs;
this->Const.MaxVertexUniformComponents = ctx->Const.Program[MESA_SHADER_VERTEX].MaxUniformComponents;
this->Const.MaxVertexTextureImageUnits = ctx->Const.Program[MESA_SHADER_VERTEX].MaxTextureImageUnits;
glsl: Require a context in _mesa_glsl_parse_state. Create a dummy context in the standalone compiler and pass it to _mesa_glsl_parse_state.
2010-09-08 18:48:12 +08:00
this->Const.MaxCombinedTextureImageUnits = ctx->Const.MaxCombinedTextureImageUnits;
mesa: replace ctx->Const.{Vertex,Fragment,Geomtery}Program with an array. These are replaced with ctx->Const.Program[MESA_SHADER_{VERTEX,FRAGMENT,GEOMETRY}]. In patches to follow, this will allow us to replace a lot of ad-hoc logic with a variable index into the array. With the exception of the changes to mtypes.h, this patch was generated entirely by the command: find src -type f '(' -iname '*.c' -o -iname '*.cpp' -o -iname '*.py' \ -o -iname '*.y' ')' -print0 | xargs -0 sed -i \ -e 's/Const\.VertexProgram/Const.Program[MESA_SHADER_VERTEX]/g' \ -e 's/Const\.GeometryProgram/Const.Program[MESA_SHADER_GEOMETRY]/g' \ -e 's/Const\.FragmentProgram/Const.Program[MESA_SHADER_FRAGMENT]/g' Suggested-by: Brian Paul <brianp@vmware.com> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-01-08 10:00:28 -08:00
this->Const.MaxTextureImageUnits = ctx->Const.Program[MESA_SHADER_FRAGMENT].MaxTextureImageUnits;
this->Const.MaxFragmentUniformComponents = ctx->Const.Program[MESA_SHADER_FRAGMENT].MaxUniformComponents;
glsl: Make {Min,Max}ProgramTexelOffset available to compiler. These constants need to be made available to shaders in GLSL 3.00 ES. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Eric Anholt <eric@anholt.net> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-04 10:00:20 -07:00
this->Const.MinProgramTexelOffset = ctx->Const.MinProgramTexelOffset;
this->Const.MaxProgramTexelOffset = ctx->Const.MaxProgramTexelOffset;
glsl: Require a context in _mesa_glsl_parse_state. Create a dummy context in the standalone compiler and pass it to _mesa_glsl_parse_state.
2010-09-08 18:48:12 +08:00
this->Const.MaxDrawBuffers = ctx->Const.MaxDrawBuffers;
glsl: Reject shader versions not supported by the implementation Previously we'd happily compile GLSL 1.30 shaders on any driver. We'd also happily compile GLSL 1.10 and 1.20 shaders in an ES2 context. This has been a long standing FINISHME in the compiler. NOTE: This is a candidate for the 7.9 and 7.10 branches
2011-01-31 15:02:24 -08:00
glsl: add EXT_blend_func_extended parser enables This adds a state for the maximum dual source draw variables available and the variable for determining if the extension has been enabled in the program shaders. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-11-05 10:53:40 -06:00
this->Const.MaxDualSourceDrawBuffers = ctx->Const.MaxDualSourceDrawBuffers;
glsl: Add new GLSL 1.50 constants. This patch populates the following built-in GLSL 1.50 variables based on constants stored in ctx->Const: - gl_MaxVertexOutputComponents - gl_MaxGeometryInputComponents - gl_MaxGeometryOutputComponents - gl_MaxFragmentInputComponents - gl_MaxGeometryTextureImageUnits - gl_MaxGeometryOutputVertices - gl_MaxGeometryTotalOutputComponents - gl_MaxGeometryUniformComponents - gl_MaxGeometryVaryingComponents On i965/gen7, fixes all Piglit tests in "spec/glsl-1.50/built-in constants/*" except for gl_MaxCombinedTextureImageUnits and gl_MaxGeometryUniformComponents. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-26 11:55:52 -07:00
/* 1.50 constants */
mesa: replace ctx->Const.{Vertex,Fragment,Geomtery}Program with an array. These are replaced with ctx->Const.Program[MESA_SHADER_{VERTEX,FRAGMENT,GEOMETRY}]. In patches to follow, this will allow us to replace a lot of ad-hoc logic with a variable index into the array. With the exception of the changes to mtypes.h, this patch was generated entirely by the command: find src -type f '(' -iname '*.c' -o -iname '*.cpp' -o -iname '*.py' \ -o -iname '*.y' ')' -print0 | xargs -0 sed -i \ -e 's/Const\.VertexProgram/Const.Program[MESA_SHADER_VERTEX]/g' \ -e 's/Const\.GeometryProgram/Const.Program[MESA_SHADER_GEOMETRY]/g' \ -e 's/Const\.FragmentProgram/Const.Program[MESA_SHADER_FRAGMENT]/g' Suggested-by: Brian Paul <brianp@vmware.com> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-01-08 10:00:28 -08:00
this->Const.MaxVertexOutputComponents = ctx->Const.Program[MESA_SHADER_VERTEX].MaxOutputComponents;
this->Const.MaxGeometryInputComponents = ctx->Const.Program[MESA_SHADER_GEOMETRY].MaxInputComponents;
this->Const.MaxGeometryOutputComponents = ctx->Const.Program[MESA_SHADER_GEOMETRY].MaxOutputComponents;
this->Const.MaxFragmentInputComponents = ctx->Const.Program[MESA_SHADER_FRAGMENT].MaxInputComponents;
this->Const.MaxGeometryTextureImageUnits = ctx->Const.Program[MESA_SHADER_GEOMETRY].MaxTextureImageUnits;
glsl: Add new GLSL 1.50 constants. This patch populates the following built-in GLSL 1.50 variables based on constants stored in ctx->Const: - gl_MaxVertexOutputComponents - gl_MaxGeometryInputComponents - gl_MaxGeometryOutputComponents - gl_MaxFragmentInputComponents - gl_MaxGeometryTextureImageUnits - gl_MaxGeometryOutputVertices - gl_MaxGeometryTotalOutputComponents - gl_MaxGeometryUniformComponents - gl_MaxGeometryVaryingComponents On i965/gen7, fixes all Piglit tests in "spec/glsl-1.50/built-in constants/*" except for gl_MaxCombinedTextureImageUnits and gl_MaxGeometryUniformComponents. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-26 11:55:52 -07:00
this->Const.MaxGeometryOutputVertices = ctx->Const.MaxGeometryOutputVertices;
this->Const.MaxGeometryTotalOutputComponents = ctx->Const.MaxGeometryTotalOutputComponents;
mesa: replace ctx->Const.{Vertex,Fragment,Geomtery}Program with an array. These are replaced with ctx->Const.Program[MESA_SHADER_{VERTEX,FRAGMENT,GEOMETRY}]. In patches to follow, this will allow us to replace a lot of ad-hoc logic with a variable index into the array. With the exception of the changes to mtypes.h, this patch was generated entirely by the command: find src -type f '(' -iname '*.c' -o -iname '*.cpp' -o -iname '*.py' \ -o -iname '*.y' ')' -print0 | xargs -0 sed -i \ -e 's/Const\.VertexProgram/Const.Program[MESA_SHADER_VERTEX]/g' \ -e 's/Const\.GeometryProgram/Const.Program[MESA_SHADER_GEOMETRY]/g' \ -e 's/Const\.FragmentProgram/Const.Program[MESA_SHADER_FRAGMENT]/g' Suggested-by: Brian Paul <brianp@vmware.com> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-01-08 10:00:28 -08:00
this->Const.MaxGeometryUniformComponents = ctx->Const.Program[MESA_SHADER_GEOMETRY].MaxUniformComponents;
glsl: Add new GLSL 1.50 constants. This patch populates the following built-in GLSL 1.50 variables based on constants stored in ctx->Const: - gl_MaxVertexOutputComponents - gl_MaxGeometryInputComponents - gl_MaxGeometryOutputComponents - gl_MaxFragmentInputComponents - gl_MaxGeometryTextureImageUnits - gl_MaxGeometryOutputVertices - gl_MaxGeometryTotalOutputComponents - gl_MaxGeometryUniformComponents - gl_MaxGeometryVaryingComponents On i965/gen7, fixes all Piglit tests in "spec/glsl-1.50/built-in constants/*" except for gl_MaxCombinedTextureImageUnits and gl_MaxGeometryUniformComponents. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-26 11:55:52 -07:00
mesa: replace ctx->Const.{Vertex,Fragment,Geomtery}Program with an array. These are replaced with ctx->Const.Program[MESA_SHADER_{VERTEX,FRAGMENT,GEOMETRY}]. In patches to follow, this will allow us to replace a lot of ad-hoc logic with a variable index into the array. With the exception of the changes to mtypes.h, this patch was generated entirely by the command: find src -type f '(' -iname '*.c' -o -iname '*.cpp' -o -iname '*.py' \ -o -iname '*.y' ')' -print0 | xargs -0 sed -i \ -e 's/Const\.VertexProgram/Const.Program[MESA_SHADER_VERTEX]/g' \ -e 's/Const\.GeometryProgram/Const.Program[MESA_SHADER_GEOMETRY]/g' \ -e 's/Const\.FragmentProgram/Const.Program[MESA_SHADER_FRAGMENT]/g' Suggested-by: Brian Paul <brianp@vmware.com> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-01-08 10:00:28 -08:00
this->Const.MaxVertexAtomicCounters = ctx->Const.Program[MESA_SHADER_VERTEX].MaxAtomicCounters;
glsl: add builtin constants for ARB_tessellation_shader Limits from other extensions added by Marek. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-08-17 22:37:16 +12:00
this->Const.MaxTessControlAtomicCounters = ctx->Const.Program[MESA_SHADER_TESS_CTRL].MaxAtomicCounters;
this->Const.MaxTessEvaluationAtomicCounters = ctx->Const.Program[MESA_SHADER_TESS_EVAL].MaxAtomicCounters;
mesa: replace ctx->Const.{Vertex,Fragment,Geomtery}Program with an array. These are replaced with ctx->Const.Program[MESA_SHADER_{VERTEX,FRAGMENT,GEOMETRY}]. In patches to follow, this will allow us to replace a lot of ad-hoc logic with a variable index into the array. With the exception of the changes to mtypes.h, this patch was generated entirely by the command: find src -type f '(' -iname '*.c' -o -iname '*.cpp' -o -iname '*.py' \ -o -iname '*.y' ')' -print0 | xargs -0 sed -i \ -e 's/Const\.VertexProgram/Const.Program[MESA_SHADER_VERTEX]/g' \ -e 's/Const\.GeometryProgram/Const.Program[MESA_SHADER_GEOMETRY]/g' \ -e 's/Const\.FragmentProgram/Const.Program[MESA_SHADER_FRAGMENT]/g' Suggested-by: Brian Paul <brianp@vmware.com> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-01-08 10:00:28 -08:00
this->Const.MaxGeometryAtomicCounters = ctx->Const.Program[MESA_SHADER_GEOMETRY].MaxAtomicCounters;
this->Const.MaxFragmentAtomicCounters = ctx->Const.Program[MESA_SHADER_FRAGMENT].MaxAtomicCounters;
mesa: make compute maximums reflect driver-provided values Looks like the various max's were never plumbed through. Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2016-01-20 03:06:53 -05:00
this->Const.MaxComputeAtomicCounters = ctx->Const.Program[MESA_SHADER_COMPUTE].MaxAtomicCounters;
glsl: Add built-in functions and constants required for ARB_shader_atomic_counters. v2: Represent atomics as GLSL intrinsics. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-10-20 12:39:16 -07:00
this->Const.MaxCombinedAtomicCounters = ctx->Const.MaxCombinedAtomicCounters;
this->Const.MaxAtomicBufferBindings = ctx->Const.MaxAtomicBufferBindings;
glsl/es3.1: Allow interger mix built-ins in GLSL ES 3.10 v2: Add missing lexer support. Noticed by Tapani. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com> Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu> [v1]
2015-04-28 12:50:51 -07:00
this->Const.MaxVertexAtomicCounterBuffers =
ctx->Const.Program[MESA_SHADER_VERTEX].MaxAtomicBuffers;
glsl: add builtin constants for ARB_tessellation_shader Limits from other extensions added by Marek. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-08-17 22:37:16 +12:00
this->Const.MaxTessControlAtomicCounterBuffers =
ctx->Const.Program[MESA_SHADER_TESS_CTRL].MaxAtomicBuffers;
this->Const.MaxTessEvaluationAtomicCounterBuffers =
ctx->Const.Program[MESA_SHADER_TESS_EVAL].MaxAtomicBuffers;
glsl/es3.1: Allow interger mix built-ins in GLSL ES 3.10 v2: Add missing lexer support. Noticed by Tapani. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com> Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu> [v1]
2015-04-28 12:50:51 -07:00
this->Const.MaxGeometryAtomicCounterBuffers =
ctx->Const.Program[MESA_SHADER_GEOMETRY].MaxAtomicBuffers;
this->Const.MaxFragmentAtomicCounterBuffers =
ctx->Const.Program[MESA_SHADER_FRAGMENT].MaxAtomicBuffers;
mesa: make compute maximums reflect driver-provided values Looks like the various max's were never plumbed through. Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2016-01-20 03:06:53 -05:00
this->Const.MaxComputeAtomicCounterBuffers =
ctx->Const.Program[MESA_SHADER_COMPUTE].MaxAtomicBuffers;
glsl/es3.1: Allow interger mix built-ins in GLSL ES 3.10 v2: Add missing lexer support. Noticed by Tapani. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com> Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu> [v1]
2015-04-28 12:50:51 -07:00
this->Const.MaxCombinedAtomicCounterBuffers =
ctx->Const.MaxCombinedAtomicBuffers;
this->Const.MaxAtomicCounterBufferSize =
ctx->Const.MaxAtomicBufferSize;
glsl: Add built-in functions and constants required for ARB_shader_atomic_counters. v2: Represent atomics as GLSL intrinsics. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-10-20 12:39:16 -07:00
glsl: add transform feedback built-in constants These are new built-ins added by ARB_enhanced_layouts. Reviewed-by: Dave Airlie <airlied@redhat.com>
2016-01-13 14:40:05 +11:00
/* ARB_enhanced_layouts constants */
this->Const.MaxTransformFeedbackBuffers = ctx->Const.MaxTransformFeedbackBuffers;
this->Const.MaxTransformFeedbackInterleavedComponents = ctx->Const.MaxTransformFeedbackInterleavedComponents;
mesa/cs: Implement MAX_COMPUTE_WORK_GROUP_SIZE constant. v2: Document that the 3-element array MaxComputeWorkGroupSize is indexed by dimension. Reviewed-by: Matt Turner <mattst88@gmail.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 13:31:58 -08:00
/* Compute shader constants */
glsl: replace Elements() with ARRAY_SIZE() Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-02-28 09:11:23 -07:00
for (unsigned i = 0; i < ARRAY_SIZE(this->Const.MaxComputeWorkGroupCount); i++)
mesa/cs: Implement MAX_COMPUTE_WORK_GROUP_COUNT constant. v2: Document that the 3-element array MaxComputeWorkGroupCount is indexed by dimension. Reviewed-by: Matt Turner <mattst88@gmail.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-08 01:42:58 -08:00
this->Const.MaxComputeWorkGroupCount[i] = ctx->Const.MaxComputeWorkGroupCount[i];
glsl: replace Elements() with ARRAY_SIZE() Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-02-28 09:11:23 -07:00
for (unsigned i = 0; i < ARRAY_SIZE(this->Const.MaxComputeWorkGroupSize); i++)
mesa/cs: Implement MAX_COMPUTE_WORK_GROUP_SIZE constant. v2: Document that the 3-element array MaxComputeWorkGroupSize is indexed by dimension. Reviewed-by: Matt Turner <mattst88@gmail.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 13:31:58 -08:00
this->Const.MaxComputeWorkGroupSize[i] = ctx->Const.MaxComputeWorkGroupSize[i];
mesa: make compute maximums reflect driver-provided values Looks like the various max's were never plumbed through. Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2016-01-20 03:06:53 -05:00
this->Const.MaxComputeTextureImageUnits = ctx->Const.Program[MESA_SHADER_COMPUTE].MaxTextureImageUnits;
this->Const.MaxComputeUniformComponents = ctx->Const.Program[MESA_SHADER_COMPUTE].MaxUniformComponents;
glsl: Add built-in constants for ARB_shader_image_load_store. v2: Add them on GLSL version 4.20 too. Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-02-12 17:15:21 +01:00
this->Const.MaxImageUnits = ctx->Const.MaxImageUnits;
mesa: Rename MaxCombinedImageUnitsAndFragmentOutputs to MaxCombinedShaderOutputResources. The name of both the GLSL built-in variable and the glGetInteger param with the same value changed in GLSL ES 3.1 and GL 4.5. Its semantics also changed slightly, since the limit now also takes into account the number of SSBs in use. Switch our internal data structures to the up-to-date name. Reviewed-by: Timothy Arceri <t_arceri@yahoo.com.au> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2015-08-17 19:10:46 +03:00
this->Const.MaxCombinedShaderOutputResources = ctx->Const.MaxCombinedShaderOutputResources;
glsl: Add built-in constants for ARB_shader_image_load_store. v2: Add them on GLSL version 4.20 too. Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-02-12 17:15:21 +01:00
this->Const.MaxImageSamples = ctx->Const.MaxImageSamples;
this->Const.MaxVertexImageUniforms = ctx->Const.Program[MESA_SHADER_VERTEX].MaxImageUniforms;
glsl: add builtin constants for ARB_tessellation_shader Limits from other extensions added by Marek. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-08-17 22:37:16 +12:00
this->Const.MaxTessControlImageUniforms = ctx->Const.Program[MESA_SHADER_TESS_CTRL].MaxImageUniforms;
this->Const.MaxTessEvaluationImageUniforms = ctx->Const.Program[MESA_SHADER_TESS_EVAL].MaxImageUniforms;
glsl: Add built-in constants for ARB_shader_image_load_store. v2: Add them on GLSL version 4.20 too. Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-02-12 17:15:21 +01:00
this->Const.MaxGeometryImageUniforms = ctx->Const.Program[MESA_SHADER_GEOMETRY].MaxImageUniforms;
this->Const.MaxFragmentImageUniforms = ctx->Const.Program[MESA_SHADER_FRAGMENT].MaxImageUniforms;
mesa: make compute maximums reflect driver-provided values Looks like the various max's were never plumbed through. Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2016-01-20 03:06:53 -05:00
this->Const.MaxComputeImageUniforms = ctx->Const.Program[MESA_SHADER_COMPUTE].MaxImageUniforms;
glsl: Add built-in constants for ARB_shader_image_load_store. v2: Add them on GLSL version 4.20 too. Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-02-12 17:15:21 +01:00
this->Const.MaxCombinedImageUniforms = ctx->Const.MaxCombinedImageUniforms;
glsl: Add gl_MaxViewports to available builtin constants It seems to have been forgotten during viewports array implementation time. Cc: "10.4 10.3" <mesa-stable@lists.freedesktop.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-12-10 08:09:24 +01:00
/* ARB_viewport_array */
this->Const.MaxViewports = ctx->Const.MaxViewports;
glsl: add tessellation shader parsing support (v2) v2: Fixed things that Ken suggested. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-03-20 22:44:43 +01:00
/* tessellation shader constants */
this->Const.MaxPatchVertices = ctx->Const.MaxPatchVertices;
glsl: add builtin constants for ARB_tessellation_shader Limits from other extensions added by Marek. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-08-17 22:37:16 +12:00
this->Const.MaxTessGenLevel = ctx->Const.MaxTessGenLevel;
this->Const.MaxTessControlInputComponents = ctx->Const.Program[MESA_SHADER_TESS_CTRL].MaxInputComponents;
this->Const.MaxTessControlOutputComponents = ctx->Const.Program[MESA_SHADER_TESS_CTRL].MaxOutputComponents;
this->Const.MaxTessControlTextureImageUnits = ctx->Const.Program[MESA_SHADER_TESS_CTRL].MaxTextureImageUnits;
this->Const.MaxTessEvaluationInputComponents = ctx->Const.Program[MESA_SHADER_TESS_EVAL].MaxInputComponents;
this->Const.MaxTessEvaluationOutputComponents = ctx->Const.Program[MESA_SHADER_TESS_EVAL].MaxOutputComponents;
this->Const.MaxTessEvaluationTextureImageUnits = ctx->Const.Program[MESA_SHADER_TESS_EVAL].MaxTextureImageUnits;
this->Const.MaxTessPatchComponents = ctx->Const.MaxTessPatchComponents;
this->Const.MaxTessControlTotalOutputComponents = ctx->Const.MaxTessControlTotalOutputComponents;
this->Const.MaxTessControlUniformComponents = ctx->Const.Program[MESA_SHADER_TESS_CTRL].MaxUniformComponents;
this->Const.MaxTessEvaluationUniformComponents = ctx->Const.Program[MESA_SHADER_TESS_EVAL].MaxUniformComponents;
glsl: add tessellation shader parsing support (v2) v2: Fixed things that Ken suggested. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-03-20 22:44:43 +01:00
glsl: Initialize all member variables of _mesa_glsl_parse_state on construction. The _mesa_glsl_parse_state object relies on the memory allocator zeroing out its contents before it's initialized, which is quite an unusual practice in the C++ world because it ties objects to some specific allocation scheme, and gives unpredictable results when an object is created with a different allocator -- Stack allocation, array allocation, or aggregation inside a different object are some of the useful possibilities that come to my mind. Initialize all fields from the constructor and stop using the zeroing allocator. Reviewed-by: Paul Berry <stereotype441@gmail.com> Reviewed-by: Chad Versace <chad.versace@linux.intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-09-11 20:07:53 -07:00
this->current_function = NULL;
this->toplevel_ir = NULL;
this->found_return = false;
this->all_invariant = false;
this->user_structures = NULL;
this->num_user_structures = 0;
glsl: add ast/parser support for subroutine parsing storage (v3.2) This is the guts of the GLSL parser and AST support for shader subroutines. The code creates a subroutine type in the parser, and uses that there to validate the identifiers. The parser also distinguishes between subroutine types/function prototypes /uniforms and subroutine defintions for functions. Then in the AST conversion it recreates the types, and stores the subroutine definition info or subroutine info into the ir_function along with a side lookup table in the parser state. It also converts subroutine calls into the enhanced ir_call. v2: move to handling method calls in function handling not in field selection. v3: merge Chris's previous parser patches in here, to make it clearer what's changed in one place. v3.1: add more documentation, drop unused include v3.2: drop is_subroutine_def Reviewed-by: Chris Forbes <chrisf@ijw.co.nz> Signed-off-by: Dave Airlie <airlied@redhat.com>
2015-06-01 10:55:47 +10:00
this->num_subroutines = 0;
this->subroutines = NULL;
this->num_subroutine_types = 0;
this->subroutine_types = NULL;
glsl: Initialize all member variables of _mesa_glsl_parse_state on construction. The _mesa_glsl_parse_state object relies on the memory allocator zeroing out its contents before it's initialized, which is quite an unusual practice in the C++ world because it ties objects to some specific allocation scheme, and gives unpredictable results when an object is created with a different allocator -- Stack allocation, array allocation, or aggregation inside a different object are some of the useful possibilities that come to my mind. Initialize all fields from the constructor and stop using the zeroing allocator. Reviewed-by: Paul Berry <stereotype441@gmail.com> Reviewed-by: Chad Versace <chad.versace@linux.intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-09-11 20:07:53 -07:00
main, glsl: Bump max known desktop glsl version to 4.50 Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-11-24 13:56:29 -08:00
/* supported_versions should be large enough to support the known desktop
glsl/es3.1: Allow 3.10 ES shaders in a GLES 3.1 context Currently no 3.10 ES features (beyond 3.00 ES) are enabled. That will come later. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
2015-04-28 12:13:21 -07:00
* GLSL versions plus 3 GLES versions (ES 1.00, ES 3.00, and ES 3.10))
main, glsl: Bump max known desktop glsl version to 4.50 Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-11-24 13:56:29 -08:00
*/
glsl/es3.1: Allow 3.10 ES shaders in a GLES 3.1 context Currently no 3.10 ES features (beyond 3.00 ES) are enabled. That will come later. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
2015-04-28 12:13:21 -07:00
STATIC_ASSERT((ARRAY_SIZE(known_desktop_glsl_versions) + 3) ==
main, glsl: Bump max known desktop glsl version to 4.50 Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-11-24 13:56:29 -08:00
ARRAY_SIZE(this->supported_versions));
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
/* Populate the list of supported GLSL versions */
/* FINISHME: Once the OpenGL 3.0 'forward compatible' context or
* the OpenGL 3.2 Core context is supported, this logic will need
* change. Older versions of GLSL are no longer supported
* outside the compatibility contexts of 3.x.
*/
this->num_supported_versions = 0;
if (_mesa_is_desktop_gl(ctx)) {
for (unsigned i = 0; i < ARRAY_SIZE(known_desktop_glsl_versions); i++) {
if (known_desktop_glsl_versions[i] <= ctx->Const.GLSLVersion) {
this->supported_versions[this->num_supported_versions].ver
= known_desktop_glsl_versions[i];
this->supported_versions[this->num_supported_versions].es = false;
this->num_supported_versions++;
}
}
}
if (ctx->API == API_OPENGLES2 || ctx->Extensions.ARB_ES2_compatibility) {
this->supported_versions[this->num_supported_versions].ver = 100;
this->supported_versions[this->num_supported_versions].es = true;
this->num_supported_versions++;
}
if (_mesa_is_gles3(ctx) || ctx->Extensions.ARB_ES3_compatibility) {
this->supported_versions[this->num_supported_versions].ver = 300;
this->supported_versions[this->num_supported_versions].es = true;
this->num_supported_versions++;
}
glsl/es3.1: Allow 3.10 ES shaders in a GLES 3.1 context Currently no 3.10 ES features (beyond 3.00 ES) are enabled. That will come later. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
2015-04-28 12:13:21 -07:00
if (_mesa_is_gles31(ctx)) {
this->supported_versions[this->num_supported_versions].ver = 310;
this->supported_versions[this->num_supported_versions].es = true;
this->num_supported_versions++;
}
glsl: Reject shader versions not supported by the implementation Previously we'd happily compile GLSL 1.30 shaders on any driver. We'd also happily compile GLSL 1.10 and 1.20 shaders in an ES2 context. This has been a long standing FINISHME in the compiler. NOTE: This is a candidate for the 7.9 and 7.10 branches
2011-01-31 15:02:24 -08:00
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
/* Create a string for use in error messages to tell the user which GLSL
* versions are supported.
*/
char *supported = ralloc_strdup(this, "");
for (unsigned i = 0; i < this->num_supported_versions; i++) {
unsigned ver = this->supported_versions[i].ver;
const char *const prefix = (i == 0)
glsl: Reject shader versions not supported by the implementation Previously we'd happily compile GLSL 1.30 shaders on any driver. We'd also happily compile GLSL 1.10 and 1.20 shaders in an ES2 context. This has been a long standing FINISHME in the compiler. NOTE: This is a candidate for the 7.9 and 7.10 branches
2011-01-31 15:02:24 -08:00
? ""
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
: ((i == this->num_supported_versions - 1) ? ", and " : ", ");
const char *const suffix = (this->supported_versions[i].es) ? " ES" : "";
glsl: Reject shader versions not supported by the implementation Previously we'd happily compile GLSL 1.30 shaders on any driver. We'd also happily compile GLSL 1.10 and 1.20 shaders in an ES2 context. This has been a long standing FINISHME in the compiler. NOTE: This is a candidate for the 7.9 and 7.10 branches
2011-01-31 15:02:24 -08:00
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
ralloc_asprintf_append(& supported, "%s%u.%02u%s",
glsl: Reject shader versions not supported by the implementation Previously we'd happily compile GLSL 1.30 shaders on any driver. We'd also happily compile GLSL 1.10 and 1.20 shaders in an ES2 context. This has been a long standing FINISHME in the compiler. NOTE: This is a candidate for the 7.9 and 7.10 branches
2011-01-31 15:02:24 -08:00
prefix,
ver / 100, ver % 100,
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
suffix);
glsl: Reject shader versions not supported by the implementation Previously we'd happily compile GLSL 1.30 shaders on any driver. We'd also happily compile GLSL 1.10 and 1.20 shaders in an ES2 context. This has been a long standing FINISHME in the compiler. NOTE: This is a candidate for the 7.9 and 7.10 branches
2011-01-31 15:02:24 -08:00
}
this->supported_version_string = supported;
mesa: Add a flag for forcing all GLSL extensions to "warn". NOTE: This is a candidate for the 8.0 branch. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2012-01-18 12:14:09 -08:00
if (ctx->Const.ForceGLSLExtensionsWarn)
_mesa_glsl_process_extension("all", NULL, "warn", NULL, this);
glsl: Add support for default layout qualifiers for uniforms. I ended up having to add rallocing of the ast_type_qualifier in order to avoid pulling in ast.h for glsl_parser_extras.h, because I wanted to track an ast_type_qualifier in the state. Fixes piglit ARB_uniform_buffer_object/row-major. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2012-06-22 13:36:35 -07:00
this->default_uniform_qualifier = new(this) ast_type_qualifier();
this->default_uniform_qualifier->flags.q.shared = 1;
this->default_uniform_qualifier->flags.q.column_major = 1;
glsl: ignore default qualifier declarations when checking for duplicate layout qualifiers Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-03 09:47:56 +02:00
this->default_uniform_qualifier->is_default_qualifier = true;
glsl: Parse the GLSL 1.50 GS layout qualifiers. Limited semantic checking (compatibility between declarations, checking that they're in the right shader target, etc.) is done. v2: Remove stray debug printfs. v3 (Paul Berry <stereotype441@gmail.com>): Process input layout qualifiers at ast_to_hir time rather than at parse time, since certain error conditions depend on the relative ordering between input layout qualifiers, declarations, and calls to .length(). Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 14:03:49 -07:00
glsl: allow default qualifiers for shader storage block definitions This kind of definitions: layout(xxx) buffer; was not supported by commit 84fc5fece006. Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-03 12:00:16 +02:00
this->default_shader_storage_qualifier = new(this) ast_type_qualifier();
this->default_shader_storage_qualifier->flags.q.shared = 1;
this->default_shader_storage_qualifier->flags.q.column_major = 1;
this->default_shader_storage_qualifier->is_default_qualifier = true;
glsl: Link error if fs defines conflicting qualifiers for gl_FragCoord GLSL 1.50 spec says: "If gl_FragCoord is redeclared in any fragment shader in a program, it must be redeclared in all the fragment shaders in that program that have a static use gl_FragCoord. All redeclarations of gl_FragCoord in all fragment shaders in a single program must have the same set of qualifiers." This patch causes the shader link to fail if we have multiple fragment shaders with conflicting layout qualifiers for gl_FragCoord. V2: Restructure the code and add conditions to correctly handle the following case: fragment shader 1: layout(origin_upper_left) in vec4 gl_FragCoord; void main() { foo(); gl_FragColor = gl_FragData; } fragment shader 2: layout(pixel_center_integer) in vec4 gl_FragCoord; void foo() { } V3: Allow linking in the following case: fragment shader 1: void main() { foo(); gl_FragColor = gl_FragCoord; } fragment shader 2: in vec4 gl_FragCoord; void foo() { ... } Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com> Cc: <mesa-stable@lists.freedesktop.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-02-05 15:01:58 -08:00
this->fs_uses_gl_fragcoord = false;
this->fs_redeclares_gl_fragcoord = false;
this->fs_origin_upper_left = false;
this->fs_pixel_center_integer = false;
this->fs_redeclares_gl_fragcoord_with_no_layout_qualifiers = false;
glsl: Parse the GLSL 1.50 GS layout qualifiers. Limited semantic checking (compatibility between declarations, checking that they're in the right shader target, etc.) is done. v2: Remove stray debug printfs. v3 (Paul Berry <stereotype441@gmail.com>): Process input layout qualifiers at ast_to_hir time rather than at parse time, since certain error conditions depend on the relative ordering between input layout qualifiers, declarations, and calls to .length(). Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 14:03:49 -07:00
this->gs_input_prim_type_specified = false;
glsl: add tessellation shader parsing support (v2) v2: Fixed things that Ken suggested. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-03-20 22:44:43 +01:00
this->tcs_output_vertices_specified = false;
glsl: Initialize all member variables of _mesa_glsl_parse_state on construction. The _mesa_glsl_parse_state object relies on the memory allocator zeroing out its contents before it's initialized, which is quite an unusual practice in the C++ world because it ties objects to some specific allocation scheme, and gives unpredictable results when an object is created with a different allocator -- Stack allocation, array allocation, or aggregation inside a different object are some of the useful possibilities that come to my mind. Initialize all fields from the constructor and stop using the zeroing allocator. Reviewed-by: Paul Berry <stereotype441@gmail.com> Reviewed-by: Chad Versace <chad.versace@linux.intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-09-11 20:07:53 -07:00
this->gs_input_size = 0;
glsl: convert GS input primitive to use ast_type_qualifier We introduce a new merge_in_qualifier ast_type_qualifier which allows specialized handling of merging input layout qualifiers. By merging layout qualifiers into state->in_qualifier, we allow multiple input qualifiers. For example, the primitive type can be specified specified separately from the invocations count (ARB_gpu_shader5). state->gs_input_prim_type is moved into state->in_qualifier->prim_type state->gs_input_prim_type_specified is still processed separately so we can determine when the input primitive is specified. This is important since certain scenerios are not supported until after the primitive type has been specified in the shader code. v4: * Merge with compute shader input layout qualifiers Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
2014-02-02 17:49:15 -08:00
this->in_qualifier = new(this) ast_type_qualifier();
glsl: Parse the GLSL 1.50 GS layout qualifiers. Limited semantic checking (compatibility between declarations, checking that they're in the right shader target, etc.) is done. v2: Remove stray debug printfs. v3 (Paul Berry <stereotype441@gmail.com>): Process input layout qualifiers at ast_to_hir time rather than at parse time, since certain error conditions depend on the relative ordering between input layout qualifiers, declarations, and calls to .length(). Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 14:03:49 -07:00
this->out_qualifier = new(this) ast_type_qualifier();
glsl: Keep track of the early_fragment_tests flag in gl_shader. And rename _mesa_glsl_parse_state::early_fragment_tests to fs_early_fragment_tests for consistency with other FS-specific flags in the same struct. Reviewed-by: Matt Turner <mattst88@gmail.com>
2015-01-28 17:42:37 +02:00
this->fs_early_fragment_tests = false;
glsl: Initialize _mesa_glsl_parse_state::atomic_counter_offsets before using it. Cc: Ian Romanick <ian.d.romanick@intel.com> Cc: "10.0" <mesa-stable@lists.freedesktop.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-11-26 12:43:13 -08:00
memset(this->atomic_counter_offsets, 0,
sizeof(this->atomic_counter_offsets));
glsl: add a mechanism to allow #extension directives in the middle of shaders This is needed to make Unigine Heaven 4.0 and Unigine Valley 1.0 work with sample shading. Also, if this is disabled, the error message at least makes sense now. Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-07-08 20:20:22 +02:00
this->allow_extension_directive_midshader =
ctx->Const.AllowGLSLExtensionDirectiveMidShader;
glsl2: Add a constructor for _mesa_glsl_parse_state Coming changes to the handling of built-in functions necessitate this.
2010-07-18 15:59:43 -07:00
}
glsl: Add GLSL version query functions. With the advent of GLSL 3.00 ES, the version checks we perform in the GLSL compiler (to determine which language features are present) will become more complicated. To reduce the complexity, this patch adds functions check_version() and is_version() to _mesa_glsl_parse_state. These functions take two version numbers: a desktop GLSL version and a GLSL ES version, and return a boolean indicating whether the GLSL version being compiled is at least the required version. So, for example, is_version(130, 300) returns true if the GLSL version being compiled is at least desktop GLSL 1.30 or GLSL 3.00. The check_version() function additionally produces an error message if the version check fails, informing the user of which GLSL version(s) support the given feature. [v2, idr]: Add PRINTFLIKE annotation to the new method. The numbering of th parameters is correct because GCC is silly. [v3, idr]: Fix copy-and-paste error in the comment before _mesa_glsl_parse_state::is_version. Noticed by Ken. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-01 19:09:24 -07:00
/**
* Determine whether the current GLSL version is sufficiently high to support
* a certain feature, and generate an error message if it isn't.
*
* \param required_glsl_version and \c required_glsl_es_version are
* interpreted as they are in _mesa_glsl_parse_state::is_version().
*
* \param locp is the parser location where the error should be reported.
*
* \param fmt (and additional arguments) constitute a printf-style error
* message to report if the version check fails. Information about the
* current and required GLSL versions will be appended. So, for example, if
* the GLSL version being compiled is 1.20, and check_version(130, 300, locp,
* "foo unsupported") is called, the error message will be "foo unsupported in
* GLSL 1.20 (GLSL 1.30 or GLSL 3.00 ES required)".
*/
bool
_mesa_glsl_parse_state::check_version(unsigned required_glsl_version,
unsigned required_glsl_es_version,
YYLTYPE *locp, const char *fmt, ...)
{
if (this->is_version(required_glsl_version, required_glsl_es_version))
return true;
va_list args;
va_start(args, fmt);
glsl_parser_extras.cpp: fixup gl vs mem contexts again. This should fix: https://bugs.freedesktop.org/show_bug.cgi?id=58039 Tested-by: Darxus on bug 58039 Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-12-16 11:47:01 +10:00
char *problem = ralloc_vasprintf(this, fmt, args);
glsl: Add GLSL version query functions. With the advent of GLSL 3.00 ES, the version checks we perform in the GLSL compiler (to determine which language features are present) will become more complicated. To reduce the complexity, this patch adds functions check_version() and is_version() to _mesa_glsl_parse_state. These functions take two version numbers: a desktop GLSL version and a GLSL ES version, and return a boolean indicating whether the GLSL version being compiled is at least the required version. So, for example, is_version(130, 300) returns true if the GLSL version being compiled is at least desktop GLSL 1.30 or GLSL 3.00. The check_version() function additionally produces an error message if the version check fails, informing the user of which GLSL version(s) support the given feature. [v2, idr]: Add PRINTFLIKE annotation to the new method. The numbering of th parameters is correct because GCC is silly. [v3, idr]: Fix copy-and-paste error in the comment before _mesa_glsl_parse_state::is_version. Noticed by Ken. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-01 19:09:24 -07:00
va_end(args);
const char *glsl_version_string
glsl: Fix gl_context vs. ralloc context in check_version again, again. Dave found some, but there were more. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=58039
2012-12-17 11:20:53 -08:00
= glsl_compute_version_string(this, false, required_glsl_version);
glsl: Add GLSL version query functions. With the advent of GLSL 3.00 ES, the version checks we perform in the GLSL compiler (to determine which language features are present) will become more complicated. To reduce the complexity, this patch adds functions check_version() and is_version() to _mesa_glsl_parse_state. These functions take two version numbers: a desktop GLSL version and a GLSL ES version, and return a boolean indicating whether the GLSL version being compiled is at least the required version. So, for example, is_version(130, 300) returns true if the GLSL version being compiled is at least desktop GLSL 1.30 or GLSL 3.00. The check_version() function additionally produces an error message if the version check fails, informing the user of which GLSL version(s) support the given feature. [v2, idr]: Add PRINTFLIKE annotation to the new method. The numbering of th parameters is correct because GCC is silly. [v3, idr]: Fix copy-and-paste error in the comment before _mesa_glsl_parse_state::is_version. Noticed by Ken. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-01 19:09:24 -07:00
const char *glsl_es_version_string
glsl: Fix gl_context vs. ralloc context in check_version again, again. Dave found some, but there were more. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=58039
2012-12-17 11:20:53 -08:00
= glsl_compute_version_string(this, true, required_glsl_es_version);
glsl: Add GLSL version query functions. With the advent of GLSL 3.00 ES, the version checks we perform in the GLSL compiler (to determine which language features are present) will become more complicated. To reduce the complexity, this patch adds functions check_version() and is_version() to _mesa_glsl_parse_state. These functions take two version numbers: a desktop GLSL version and a GLSL ES version, and return a boolean indicating whether the GLSL version being compiled is at least the required version. So, for example, is_version(130, 300) returns true if the GLSL version being compiled is at least desktop GLSL 1.30 or GLSL 3.00. The check_version() function additionally produces an error message if the version check fails, informing the user of which GLSL version(s) support the given feature. [v2, idr]: Add PRINTFLIKE annotation to the new method. The numbering of th parameters is correct because GCC is silly. [v3, idr]: Fix copy-and-paste error in the comment before _mesa_glsl_parse_state::is_version. Noticed by Ken. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-01 19:09:24 -07:00
const char *requirement_string = "";
if (required_glsl_version && required_glsl_es_version) {
glsl_parser_extras.cpp: fixup gl vs mem contexts again. This should fix: https://bugs.freedesktop.org/show_bug.cgi?id=58039 Tested-by: Darxus on bug 58039 Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-12-16 11:47:01 +10:00
requirement_string = ralloc_asprintf(this, " (%s or %s required)",
glsl: Add GLSL version query functions. With the advent of GLSL 3.00 ES, the version checks we perform in the GLSL compiler (to determine which language features are present) will become more complicated. To reduce the complexity, this patch adds functions check_version() and is_version() to _mesa_glsl_parse_state. These functions take two version numbers: a desktop GLSL version and a GLSL ES version, and return a boolean indicating whether the GLSL version being compiled is at least the required version. So, for example, is_version(130, 300) returns true if the GLSL version being compiled is at least desktop GLSL 1.30 or GLSL 3.00. The check_version() function additionally produces an error message if the version check fails, informing the user of which GLSL version(s) support the given feature. [v2, idr]: Add PRINTFLIKE annotation to the new method. The numbering of th parameters is correct because GCC is silly. [v3, idr]: Fix copy-and-paste error in the comment before _mesa_glsl_parse_state::is_version. Noticed by Ken. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-01 19:09:24 -07:00
glsl_version_string,
glsl_es_version_string);
} else if (required_glsl_version) {
glsl_parser_extras.cpp: fixup gl vs mem contexts again. This should fix: https://bugs.freedesktop.org/show_bug.cgi?id=58039 Tested-by: Darxus on bug 58039 Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-12-16 11:47:01 +10:00
requirement_string = ralloc_asprintf(this, " (%s required)",
glsl: Add GLSL version query functions. With the advent of GLSL 3.00 ES, the version checks we perform in the GLSL compiler (to determine which language features are present) will become more complicated. To reduce the complexity, this patch adds functions check_version() and is_version() to _mesa_glsl_parse_state. These functions take two version numbers: a desktop GLSL version and a GLSL ES version, and return a boolean indicating whether the GLSL version being compiled is at least the required version. So, for example, is_version(130, 300) returns true if the GLSL version being compiled is at least desktop GLSL 1.30 or GLSL 3.00. The check_version() function additionally produces an error message if the version check fails, informing the user of which GLSL version(s) support the given feature. [v2, idr]: Add PRINTFLIKE annotation to the new method. The numbering of th parameters is correct because GCC is silly. [v3, idr]: Fix copy-and-paste error in the comment before _mesa_glsl_parse_state::is_version. Noticed by Ken. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-01 19:09:24 -07:00
glsl_version_string);
} else if (required_glsl_es_version) {
glsl_parser_extras.cpp: fixup gl vs mem contexts again. This should fix: https://bugs.freedesktop.org/show_bug.cgi?id=58039 Tested-by: Darxus on bug 58039 Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-12-16 11:47:01 +10:00
requirement_string = ralloc_asprintf(this, " (%s required)",
glsl: Add GLSL version query functions. With the advent of GLSL 3.00 ES, the version checks we perform in the GLSL compiler (to determine which language features are present) will become more complicated. To reduce the complexity, this patch adds functions check_version() and is_version() to _mesa_glsl_parse_state. These functions take two version numbers: a desktop GLSL version and a GLSL ES version, and return a boolean indicating whether the GLSL version being compiled is at least the required version. So, for example, is_version(130, 300) returns true if the GLSL version being compiled is at least desktop GLSL 1.30 or GLSL 3.00. The check_version() function additionally produces an error message if the version check fails, informing the user of which GLSL version(s) support the given feature. [v2, idr]: Add PRINTFLIKE annotation to the new method. The numbering of th parameters is correct because GCC is silly. [v3, idr]: Fix copy-and-paste error in the comment before _mesa_glsl_parse_state::is_version. Noticed by Ken. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-01 19:09:24 -07:00
glsl_es_version_string);
}
glsl: Be consistent about '\n', '.', and capitalization in errors/warnings. The majority of calls to _mesa_glsl_error(), _mesa_glsl_warning(), and _mesa_glsl_parse_state::check_version() use a message that begins with a lower case letter and ends without a period. This patch makes all messages follow that convention. Also, error/warning messages shouldn't end in '\n', since _mesa_glsl_msg() automatically adds '\n' at the end of the message. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-25 19:56:43 -07:00
_mesa_glsl_error(locp, this, "%s in %s%s",
glsl: Add GLSL version query functions. With the advent of GLSL 3.00 ES, the version checks we perform in the GLSL compiler (to determine which language features are present) will become more complicated. To reduce the complexity, this patch adds functions check_version() and is_version() to _mesa_glsl_parse_state. These functions take two version numbers: a desktop GLSL version and a GLSL ES version, and return a boolean indicating whether the GLSL version being compiled is at least the required version. So, for example, is_version(130, 300) returns true if the GLSL version being compiled is at least desktop GLSL 1.30 or GLSL 3.00. The check_version() function additionally produces an error message if the version check fails, informing the user of which GLSL version(s) support the given feature. [v2, idr]: Add PRINTFLIKE annotation to the new method. The numbering of th parameters is correct because GCC is silly. [v3, idr]: Fix copy-and-paste error in the comment before _mesa_glsl_parse_state::is_version. Noticed by Ken. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-01 19:09:24 -07:00
problem, this->get_version_string(),
requirement_string);
return false;
}
glsl/parser: Extract version directive processing into a function. Version directive handling is going to have to be used within two parser rules, one for desktop-style version directives (e.g. "#version 130") and one for the new ES-style version directive (e.g. "#version 300 es"), so this patch moves it to a function that can be called from both rules. No functional change. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:02:55 -07:00
/**
* Process a GLSL #version directive.
*
* \param version is the integer that follows the #version token.
glsl/parser: Handle "#version 300 es" directive. Note that GLSL 1.00 is selected using "#version 100", so "#version 100 es" is prohibited. v2: Check for GLES3 before allowing '#version 300 es' v3: Make sure a correct language_version is set in _mesa_glsl_parse_state::process_version_directive. Signed-off-by: Paul Berry <stereotype441@gmail.com> Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:17:30 -07:00
*
* \param ident is a string identifier that follows the integer, if any is
* present. Otherwise NULL.
glsl/parser: Extract version directive processing into a function. Version directive handling is going to have to be used within two parser rules, one for desktop-style version directives (e.g. "#version 130") and one for the new ES-style version directive (e.g. "#version 300 es"), so this patch moves it to a function that can be called from both rules. No functional change. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:02:55 -07:00
*/
void
glsl/parser: Handle "#version 300 es" directive. Note that GLSL 1.00 is selected using "#version 100", so "#version 100 es" is prohibited. v2: Check for GLES3 before allowing '#version 300 es' v3: Make sure a correct language_version is set in _mesa_glsl_parse_state::process_version_directive. Signed-off-by: Paul Berry <stereotype441@gmail.com> Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:17:30 -07:00
_mesa_glsl_parse_state::process_version_directive(YYLTYPE *locp, int version,
const char *ident)
glsl/parser: Extract version directive processing into a function. Version directive handling is going to have to be used within two parser rules, one for desktop-style version directives (e.g. "#version 130") and one for the new ES-style version directive (e.g. "#version 300 es"), so this patch moves it to a function that can be called from both rules. No functional change. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:02:55 -07:00
{
glsl/parser: Handle "#version 300 es" directive. Note that GLSL 1.00 is selected using "#version 100", so "#version 100 es" is prohibited. v2: Check for GLES3 before allowing '#version 300 es' v3: Make sure a correct language_version is set in _mesa_glsl_parse_state::process_version_directive. Signed-off-by: Paul Berry <stereotype441@gmail.com> Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:17:30 -07:00
bool es_token_present = false;
if (ident) {
if (strcmp(ident, "es") == 0) {
es_token_present = true;
glsl: Parse "#version 150 core" directives. Previously we only supported "#version 150". This patch recognizes "compatibility" to give the user a more descriptive error message. Fixes Piglit's version-150-core-profile test. Reviewed-by: Matt Turner <mattst88@gmail.com> Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-07 21:28:59 -07:00
} else if (version >= 150) {
if (strcmp(ident, "core") == 0) {
/* Accept the token. There's no need to record that this is
* a core profile shader since that's the only profile we support.
*/
} else if (strcmp(ident, "compatibility") == 0) {
_mesa_glsl_error(locp, this,
glsl: Be consistent about '\n', '.', and capitalization in errors/warnings. The majority of calls to _mesa_glsl_error(), _mesa_glsl_warning(), and _mesa_glsl_parse_state::check_version() use a message that begins with a lower case letter and ends without a period. This patch makes all messages follow that convention. Also, error/warning messages shouldn't end in '\n', since _mesa_glsl_msg() automatically adds '\n' at the end of the message. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-25 19:56:43 -07:00
"the compatibility profile is not supported");
glsl: Parse "#version 150 core" directives. Previously we only supported "#version 150". This patch recognizes "compatibility" to give the user a more descriptive error message. Fixes Piglit's version-150-core-profile test. Reviewed-by: Matt Turner <mattst88@gmail.com> Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-07 21:28:59 -07:00
} else {
_mesa_glsl_error(locp, this,
"\"%s\" is not a valid shading language profile; "
glsl: Be consistent about '\n', '.', and capitalization in errors/warnings. The majority of calls to _mesa_glsl_error(), _mesa_glsl_warning(), and _mesa_glsl_parse_state::check_version() use a message that begins with a lower case letter and ends without a period. This patch makes all messages follow that convention. Also, error/warning messages shouldn't end in '\n', since _mesa_glsl_msg() automatically adds '\n' at the end of the message. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-25 19:56:43 -07:00
"if present, it must be \"core\"", ident);
glsl: Parse "#version 150 core" directives. Previously we only supported "#version 150". This patch recognizes "compatibility" to give the user a more descriptive error message. Fixes Piglit's version-150-core-profile test. Reviewed-by: Matt Turner <mattst88@gmail.com> Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-07 21:28:59 -07:00
}
glsl/parser: Handle "#version 300 es" directive. Note that GLSL 1.00 is selected using "#version 100", so "#version 100 es" is prohibited. v2: Check for GLES3 before allowing '#version 300 es' v3: Make sure a correct language_version is set in _mesa_glsl_parse_state::process_version_directive. Signed-off-by: Paul Berry <stereotype441@gmail.com> Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:17:30 -07:00
} else {
_mesa_glsl_error(locp, this,
glsl: Be consistent about '\n', '.', and capitalization in errors/warnings. The majority of calls to _mesa_glsl_error(), _mesa_glsl_warning(), and _mesa_glsl_parse_state::check_version() use a message that begins with a lower case letter and ends without a period. This patch makes all messages follow that convention. Also, error/warning messages shouldn't end in '\n', since _mesa_glsl_msg() automatically adds '\n' at the end of the message. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-25 19:56:43 -07:00
"illegal text following version number");
glsl/parser: Handle "#version 300 es" directive. Note that GLSL 1.00 is selected using "#version 100", so "#version 100 es" is prohibited. v2: Check for GLES3 before allowing '#version 300 es' v3: Make sure a correct language_version is set in _mesa_glsl_parse_state::process_version_directive. Signed-off-by: Paul Berry <stereotype441@gmail.com> Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:17:30 -07:00
}
}
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
this->es_shader = es_token_present;
if (version == 100) {
if (es_token_present) {
glsl/parser: Handle "#version 300 es" directive. Note that GLSL 1.00 is selected using "#version 100", so "#version 100 es" is prohibited. v2: Check for GLES3 before allowing '#version 300 es' v3: Make sure a correct language_version is set in _mesa_glsl_parse_state::process_version_directive. Signed-off-by: Paul Berry <stereotype441@gmail.com> Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:17:30 -07:00
_mesa_glsl_error(locp, this,
"GLSL 1.00 ES should be selected using "
glsl: Be consistent about '\n', '.', and capitalization in errors/warnings. The majority of calls to _mesa_glsl_error(), _mesa_glsl_warning(), and _mesa_glsl_parse_state::check_version() use a message that begins with a lower case letter and ends without a period. This patch makes all messages follow that convention. Also, error/warning messages shouldn't end in '\n', since _mesa_glsl_msg() automatically adds '\n' at the end of the message. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-25 19:56:43 -07:00
"`#version 100'");
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
} else {
glsl/parser: Handle "#version 300 es" directive. Note that GLSL 1.00 is selected using "#version 100", so "#version 100 es" is prohibited. v2: Check for GLES3 before allowing '#version 300 es' v3: Make sure a correct language_version is set in _mesa_glsl_parse_state::process_version_directive. Signed-off-by: Paul Berry <stereotype441@gmail.com> Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:17:30 -07:00
this->es_shader = true;
}
glsl/parser: Extract version directive processing into a function. Version directive handling is going to have to be used within two parser rules, one for desktop-style version directives (e.g. "#version 130") and one for the new ES-style version directive (e.g. "#version 300 es"), so this patch moves it to a function that can be called from both rules. No functional change. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:02:55 -07:00
}
glsl: Disable ARB_texture_rectangle in shader version 100. OpenGL with ARB_ES2_compatibility allows shaders that specify #version 100. This fixes the Khronos OpenGL test(Texture_Rectangle_Samplers_frag.test) failure. Cc: mesa-stable@lists.freedesktop.org Reviewed-by: Matt Turner <mattst88@gmail.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
2014-01-17 12:23:05 -08:00
if (this->es_shader) {
this->ARB_texture_rectangle_enable = false;
}
glsl: allow ForceGLSLVersion to override #version directives Previously, the ctx->Const.ForceGLSLVersion setting only worked if the shader lacked a #version directive. Now, the ForceGLSLVersion setting will override the #version directive too. This change should be safe since it should be rare to have an app that has a mix of shader versions and we only wanted to override the #version for shaders which lacked the #version directive. Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-03-27 10:54:10 -06:00
if (this->forced_language_version)
this->language_version = this->forced_language_version;
else
this->language_version = version;
glsl/parser: Extract version directive processing into a function. Version directive handling is going to have to be used within two parser rules, one for desktop-style version directives (e.g. "#version 130") and one for the new ES-style version directive (e.g. "#version 300 es"), so this patch moves it to a function that can be called from both rules. No functional change. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:02:55 -07:00
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
bool supported = false;
for (unsigned i = 0; i < this->num_supported_versions; i++) {
glsl: allow ForceGLSLVersion to override #version directives Previously, the ctx->Const.ForceGLSLVersion setting only worked if the shader lacked a #version directive. Now, the ForceGLSLVersion setting will override the #version directive too. This change should be safe since it should be rare to have an app that has a mix of shader versions and we only wanted to override the #version for shaders which lacked the #version directive. Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-03-27 10:54:10 -06:00
if (this->supported_versions[i].ver == this->language_version
glsl: Fix unsupported version error for GLSL ES 3.00, future proof for 3.30. When the user specifies an unsupported GLSL version, _mesa_glsl_parse_state::process_version_directive() nicely gives them an error message telling them which GLSL versions are supported. Previous to this patch, the logic for determining whether a given language version was supported was independent from the logic to generate this error message string; as a result, we had a bug where GLSL 3.00 would never be listed in the error message as an available language version, even if it was really available. To make matters worse, the code for generating the error message string assumed that desktop GL versions were always separated by 0.10, an assumption that will be wrong as soon as we support GLSL 3.30. This patch fixes both problems by adding a table of supported GLSL versions to _mesa_glsl_parse_state; this table is used both to generate the error message and to check whether a given version is supported. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-02-05 15:07:26 -08:00
&& this->supported_versions[i].es == this->es_shader) {
supported = true;
break;
}
}
glsl/parser: Extract version directive processing into a function. Version directive handling is going to have to be used within two parser rules, one for desktop-style version directives (e.g. "#version 130") and one for the new ES-style version directive (e.g. "#version 300 es"), so this patch moves it to a function that can be called from both rules. No functional change. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:02:55 -07:00
if (!supported) {
_mesa_glsl_error(locp, this, "%s is not supported. "
glsl: Be consistent about '\n', '.', and capitalization in errors/warnings. The majority of calls to _mesa_glsl_error(), _mesa_glsl_warning(), and _mesa_glsl_parse_state::check_version() use a message that begins with a lower case letter and ends without a period. This patch makes all messages follow that convention. Also, error/warning messages shouldn't end in '\n', since _mesa_glsl_msg() automatically adds '\n' at the end of the message. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-25 19:56:43 -07:00
"Supported versions are: %s",
glsl/parser: Extract version directive processing into a function. Version directive handling is going to have to be used within two parser rules, one for desktop-style version directives (e.g. "#version 130") and one for the new ES-style version directive (e.g. "#version 300 es"), so this patch moves it to a function that can be called from both rules. No functional change. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:02:55 -07:00
this->get_version_string(),
this->supported_version_string);
glsl/parser: Handle "#version 300 es" directive. Note that GLSL 1.00 is selected using "#version 100", so "#version 100 es" is prohibited. v2: Check for GLES3 before allowing '#version 300 es' v3: Make sure a correct language_version is set in _mesa_glsl_parse_state::process_version_directive. Signed-off-by: Paul Berry <stereotype441@gmail.com> Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:17:30 -07:00
/* On exit, the language_version must be set to a valid value.
* Later calls to _mesa_glsl_initialize_types will misbehave if
* the version is invalid.
*/
switch (this->ctx->API) {
case API_OPENGL_COMPAT:
case API_OPENGL_CORE:
this->language_version = this->ctx->Const.GLSLVersion;
break;
case API_OPENGLES:
assert(!"Should not get here.");
/* FALLTHROUGH */
case API_OPENGLES2:
this->language_version = 100;
break;
}
glsl/parser: Extract version directive processing into a function. Version directive handling is going to have to be used within two parser rules, one for desktop-style version directives (e.g. "#version 130") and one for the new ES-style version directive (e.g. "#version 300 es"), so this patch moves it to a function that can be called from both rules. No functional change. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Acked-by: Carl Worth <cworth@cworth.org>
2012-08-02 11:02:55 -07:00
}
}
mesa: Use shared code for converting shader targets to short strings. We were duplicating this code all over the place, and they all would need updating for the next set of shader targets. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2013-06-12 16:57:11 -07:00
glsl: consolidate error/warning code And lay the groundwork for GL_ARB_debug_output. v2: Add descriptive comments. Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 15:39:38 +00:00
/* This helper function will append the given message to the shader's
info log and report it via GL_ARB_debug_output. Per that extension,
'type' is one of the enum values classifying the message, and
'id' is the implementation-defined ID of the given message. */
static void
_mesa_glsl_msg(const YYLTYPE *locp, _mesa_glsl_parse_state *state,
mesa: Report ARB_debug_output for both shader errors and warnings. This ends up reusing the dynamic ID support, so a silly enum gets to go away. We don't assign good IDs to different messages yet, but at least that's tractable now. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2013-02-22 15:57:25 -08:00
GLenum type, const char *fmt, va_list ap)
glsl: consolidate error/warning code And lay the groundwork for GL_ARB_debug_output. v2: Add descriptive comments. Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 15:39:38 +00:00
{
mesa: Report ARB_debug_output for both shader errors and warnings. This ends up reusing the dynamic ID support, so a silly enum gets to go away. We don't assign good IDs to different messages yet, but at least that's tractable now. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2013-02-22 15:57:25 -08:00
bool error = (type == MESA_DEBUG_TYPE_ERROR);
GLuint msg_id = 0;
glsl: consolidate error/warning code And lay the groundwork for GL_ARB_debug_output. v2: Add descriptive comments. Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 15:39:38 +00:00
assert(state->info_log != NULL);
glsl: report errors via GL_ARB_debug_output Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 16:26:05 +00:00
/* Get the offset that the new message will be written to. */
int msg_offset = strlen(state->info_log);
glsl: consolidate error/warning code And lay the groundwork for GL_ARB_debug_output. v2: Add descriptive comments. Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 15:39:38 +00:00
ralloc_asprintf_append(&state->info_log, "%u:%u(%u): %s: ",
locp->source,
locp->first_line,
locp->first_column,
error ? "error" : "warning");
ralloc_vasprintf_append(&state->info_log, fmt, ap);
glsl: report errors via GL_ARB_debug_output Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 16:26:05 +00:00
const char *const msg = &state->info_log[msg_offset];
struct gl_context *ctx = state->ctx;
mesa: Report ARB_debug_output for both shader errors and warnings. This ends up reusing the dynamic ID support, so a silly enum gets to go away. We don't assign good IDs to different messages yet, but at least that's tractable now. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2013-02-22 15:57:25 -08:00
glsl: report errors via GL_ARB_debug_output Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 16:26:05 +00:00
/* Report the error via GL_ARB_debug_output. */
mesa: remove len argument from _mesa_shader_debug() There was only a single user which was using strlen(buf). As this function is not user facing (i.e. we don't need to feed back original length via a callback), we can simplify things. Suggested-by: Timothy Arceri <timothy.arceri@collabora.com> Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
2015-11-27 13:12:59 +00:00
_mesa_shader_debug(ctx, type, &msg_id, msg);
glsl: report errors via GL_ARB_debug_output Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 16:26:05 +00:00
glsl: consolidate error/warning code And lay the groundwork for GL_ARB_debug_output. v2: Add descriptive comments. Reviewed-by: Brian Paul <brianp@vmware.com>
2012-04-02 15:39:38 +00:00
ralloc_strcat(&state->info_log, "\n");
}
Add utility function to get the name of a shader target
2010-04-07 16:44:30 -07:00
Initial commit. lol
2010-02-22 13:19:34 -08:00
void
Track generation of errors and halt compilation appropriately
2010-03-11 14:08:33 -08:00
_mesa_glsl_error(YYLTYPE *locp, _mesa_glsl_parse_state *state,
const char *fmt, ...)
Initial commit. lol
2010-02-22 13:19:34 -08:00
{
va_list ap;
Disallow passing NULL for state to _mesa_glsl_error The two places that were still passing NULL had a state pointer to pass. Not passing it in these places prevented termination of compilation of erroneous programs.
2010-03-23 13:21:19 -07:00
state->error = true;
Track generation of errors and halt compilation appropriately
2010-03-11 14:08:33 -08:00
Initial commit. lol
2010-02-22 13:19:34 -08:00
va_start(ap, fmt);
mesa: Report ARB_debug_output for both shader errors and warnings. This ends up reusing the dynamic ID support, so a silly enum gets to go away. We don't assign good IDs to different messages yet, but at least that's tractable now. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2013-02-22 15:57:25 -08:00
_mesa_glsl_msg(locp, state, MESA_DEBUG_TYPE_ERROR, fmt, ap);
Initial commit. lol
2010-02-22 13:19:34 -08:00
va_end(ap);
}
Add _mesa_glsl_warning to emit warnings to the shader log
2010-04-07 14:47:46 -07:00
void
Store warnings and errors in a parser state infolog. Cleans up compile warning about unused state in _mesa_glsl_warning. We would want infolog handling roughly like this anyway.
2010-04-29 18:00:33 -07:00
_mesa_glsl_warning(const YYLTYPE *locp, _mesa_glsl_parse_state *state,
Add _mesa_glsl_warning to emit warnings to the shader log
2010-04-07 14:47:46 -07:00
const char *fmt, ...)
{
va_list ap;
va_start(ap, fmt);
mesa: Report ARB_debug_output for both shader errors and warnings. This ends up reusing the dynamic ID support, so a silly enum gets to go away. We don't assign good IDs to different messages yet, but at least that's tractable now. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2013-02-22 15:57:25 -08:00
_mesa_glsl_msg(locp, state, MESA_DEBUG_TYPE_OTHER, fmt, ap);
Add _mesa_glsl_warning to emit warnings to the shader log
2010-04-07 14:47:46 -07:00
va_end(ap);
}
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
/**
* Enum representing the possible behaviors that can be specified in
* an #extension directive.
*/
enum ext_behavior {
extension_disable,
extension_enable,
extension_require,
extension_warn
};
/**
* Element type for _mesa_glsl_supported_extensions
*/
struct _mesa_glsl_extension {
/**
* Name of the extension when referred to in a GLSL extension
* statement
*/
const char *name;
/** True if this extension is available to desktop GL shaders */
bool avail_in_GL;
/** True if this extension is available to GLES shaders */
bool avail_in_ES;
/**
* Flag in the gl_extensions struct indicating whether this
* extension is supported by the driver, or
* &gl_extensions::dummy_true if supported by all drivers.
*
* Note: the type (GLboolean gl_extensions::*) is a "pointer to
* member" type, the type-safe alternative to the "offsetof" macro.
* In a nutshell:
*
* - foo bar::* p declares p to be an "offset" to a field of type
* foo that exists within struct bar
* - &bar::baz computes the "offset" of field baz within struct bar
* - x.*p accesses the field of x that exists at "offset" p
* - x->*p is equivalent to (*x).*p
*/
const GLboolean gl_extensions::* supported_flag;
/**
* Flag in the _mesa_glsl_parse_state struct that should be set
* when this extension is enabled.
*
* See note in _mesa_glsl_extension::supported_flag about "pointer
* to member" types.
*/
bool _mesa_glsl_parse_state::* enable_flag;
/**
* Flag in the _mesa_glsl_parse_state struct that should be set
* when the shader requests "warn" behavior for this extension.
*
* See note in _mesa_glsl_extension::supported_flag about "pointer
* to member" types.
*/
bool _mesa_glsl_parse_state::* warn_flag;
bool compatible_with_state(const _mesa_glsl_parse_state *state) const;
void set_flags(_mesa_glsl_parse_state *state, ext_behavior behavior) const;
};
glsl: Remove shader stage checking for extension handling. Certain extensions only add functionality to particular shader stages. (For example, ARB_draw_instanced only adds variables to the vertex shader stage.) Previously, we only allowed such extensions to be enabled in the shader stages where they're useful. However, I've never found any text which mandates that behavior; in my opinion, you should be able to turn on extensions in any shader stage, even if they have no effect. Fixes Piglit tests glslparsertest/glsl2/draw_buffers-05.vert and ARB_draw_instanced/preprocessor/feature-macro-enabled.frag. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29185 Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
2013-07-29 10:40:51 -07:00
#define EXT(NAME, GL, ES, SUPPORTED_FLAG) \
{ "GL_" #NAME, GL, ES, &gl_extensions::SUPPORTED_FLAG, \
&_mesa_glsl_parse_state::NAME##_enable, \
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
&_mesa_glsl_parse_state::NAME##_warn }
/**
* Table of extensions that can be enabled/disabled within a shader,
* and the conditions under which they are supported.
*/
static const _mesa_glsl_extension _mesa_glsl_supported_extensions[] = {
glsl: Remove shader stage checking for extension handling. Certain extensions only add functionality to particular shader stages. (For example, ARB_draw_instanced only adds variables to the vertex shader stage.) Previously, we only allowed such extensions to be enabled in the shader stages where they're useful. However, I've never found any text which mandates that behavior; in my opinion, you should be able to turn on extensions in any shader stage, even if they have no effect. Fixes Piglit tests glslparsertest/glsl2/draw_buffers-05.vert and ARB_draw_instanced/preprocessor/feature-macro-enabled.frag. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29185 Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
2013-07-29 10:40:51 -07:00
/* API availability */
/* name GL ES supported flag */
glsl: Sort the list of extensions ARB, OES, then everything else. If there's ever a KHR shading language extension, it should go between ARB and OES. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Acked-by: Eric Anholt <eric@anholt.net>
2014-03-26 11:43:32 -07:00
/* ARB extensions go here, sorted alphabetically.
*/
mesa: define ARB_shader_storage_buffer_object extension Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2015-03-17 12:17:27 +01:00
EXT(ARB_arrays_of_arrays, true, false, ARB_arrays_of_arrays),
EXT(ARB_compute_shader, true, false, ARB_compute_shader),
EXT(ARB_conservative_depth, true, false, ARB_conservative_depth),
EXT(ARB_derivative_control, true, false, ARB_derivative_control),
EXT(ARB_draw_buffers, true, false, dummy_true),
EXT(ARB_draw_instanced, true, false, ARB_draw_instanced),
mesa: add ARB_enhanced_layouts Reviewed-by: Emil Velikov <emil.velikov@collabora.co.uk>
2015-10-05 00:01:45 +11:00
EXT(ARB_enhanced_layouts, true, false, ARB_enhanced_layouts),
mesa: define ARB_shader_storage_buffer_object extension Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2015-03-17 12:17:27 +01:00
EXT(ARB_explicit_attrib_location, true, false, ARB_explicit_attrib_location),
EXT(ARB_explicit_uniform_location, true, false, ARB_explicit_uniform_location),
EXT(ARB_fragment_coord_conventions, true, false, ARB_fragment_coord_conventions),
EXT(ARB_fragment_layer_viewport, true, false, ARB_fragment_layer_viewport),
EXT(ARB_gpu_shader5, true, false, ARB_gpu_shader5),
EXT(ARB_gpu_shader_fp64, true, false, ARB_gpu_shader_fp64),
EXT(ARB_sample_shading, true, false, ARB_sample_shading),
EXT(ARB_separate_shader_objects, true, false, dummy_true),
mesa: add GL_ARB_shader_atomic_counter_ops support Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
2016-02-19 23:07:38 -05:00
EXT(ARB_shader_atomic_counter_ops, true, false, ARB_shader_atomic_counter_ops),
mesa: define ARB_shader_storage_buffer_object extension Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2015-03-17 12:17:27 +01:00
EXT(ARB_shader_atomic_counters, true, false, ARB_shader_atomic_counters),
EXT(ARB_shader_bit_encoding, true, false, ARB_shader_bit_encoding),
glsl: add ARB_shader_clock infrastructure Signed-off-by: Emil Velikov <emil.velikov@collabora.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2015-10-02 09:56:37 +01:00
EXT(ARB_shader_clock, true, false, ARB_shader_clock),
mesa: Add core mesa support for GL_ARB_shader_draw_parameters Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
2015-12-10 12:07:43 -08:00
EXT(ARB_shader_draw_parameters, true, false, ARB_shader_draw_parameters),
mesa: define ARB_shader_storage_buffer_object extension Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2015-03-17 12:17:27 +01:00
EXT(ARB_shader_image_load_store, true, false, ARB_shader_image_load_store),
main: add extension GL_ARB_shader_image_size Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com> Reviewed-by: Francisco Jerez <currojerez@riseup.net> Signed-off-by: Martin Peres <martin.peres@linux.intel.com>
2015-04-27 20:05:14 +03:00
EXT(ARB_shader_image_size, true, false, ARB_shader_image_size),
mesa: define ARB_shader_storage_buffer_object extension Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2015-03-17 12:17:27 +01:00
EXT(ARB_shader_precision, true, false, ARB_shader_precision),
EXT(ARB_shader_stencil_export, true, false, ARB_shader_stencil_export),
mesa: enable ARB_shader_storage_buffer_object extension for GLES 3.1 Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com> Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-07-14 12:23:42 +02:00
EXT(ARB_shader_storage_buffer_object, true, true, ARB_shader_storage_buffer_object),
glsl: Add extension plumbing and define for ARB_shader_subroutine Reviewed-by: Tapani Pälli <tapani.palli@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Signed-off-by: Chris Forbes <chrisf@ijw.co.nz> Signed-off-by: Dave Airlie <airlied@redhat.com>
2014-08-10 21:31:06 +12:00
EXT(ARB_shader_subroutine, true, false, ARB_shader_subroutine),
glsl: add ARB_shader_texture_image_samples infrastructure Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2015-08-27 23:06:29 -04:00
EXT(ARB_shader_texture_image_samples, true, false, ARB_shader_texture_image_samples),
mesa: define ARB_shader_storage_buffer_object extension Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2015-03-17 12:17:27 +01:00
EXT(ARB_shader_texture_lod, true, false, ARB_shader_texture_lod),
EXT(ARB_shading_language_420pack, true, false, ARB_shading_language_420pack),
EXT(ARB_shading_language_packing, true, false, ARB_shading_language_packing),
glsl: add tessellation shader parsing support (v2) v2: Fixed things that Ken suggested. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-03-20 22:44:43 +01:00
EXT(ARB_tessellation_shader, true, false, ARB_tessellation_shader),
mesa: define ARB_shader_storage_buffer_object extension Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2015-03-17 12:17:27 +01:00
EXT(ARB_texture_cube_map_array, true, false, ARB_texture_cube_map_array),
EXT(ARB_texture_gather, true, false, ARB_texture_gather),
EXT(ARB_texture_multisample, true, false, ARB_texture_multisample),
EXT(ARB_texture_query_levels, true, false, ARB_texture_query_levels),
EXT(ARB_texture_query_lod, true, false, ARB_texture_query_lod),
EXT(ARB_texture_rectangle, true, false, dummy_true),
EXT(ARB_uniform_buffer_object, true, false, ARB_uniform_buffer_object),
EXT(ARB_vertex_attrib_64bit, true, false, ARB_vertex_attrib_64bit),
EXT(ARB_viewport_array, true, false, ARB_viewport_array),
glsl: Sort the list of extensions ARB, OES, then everything else. If there's ever a KHR shading language extension, it should go between ARB and OES. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Acked-by: Eric Anholt <eric@anholt.net>
2014-03-26 11:43:32 -07:00
/* KHR extensions go here, sorted alphabetically.
*/
/* OES extensions go here, sorted alphabetically.
*/
EXT(OES_EGL_image_external, false, true, OES_EGL_image_external),
glsl: add GL_OES_geometry_point_size and conditionalize gl_PointSize For now this will be enabled in tandem with GL_OES_geometry_shader. Should a driver come along that wants to separate them out, another enable can be added. Also adds the missed GL_OES_geometry_shader define in glcpp. Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by: Marta Lofstedt <marta.lofstedt@intel.com>
2016-01-22 13:07:10 -05:00
EXT(OES_geometry_point_size, false, true, OES_geometry_shader),
glsl: add support for GL_OES_geometry_shader This adds glsl support of GL_OES_geometry_shader for OpenGL ES 3.1. Signed-off-by: Marta Lofstedt <marta.lofstedt@linux.intel.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2016-01-21 16:17:31 +01:00
EXT(OES_geometry_shader, false, true, OES_geometry_shader),
mesa: add GL_OES_gpu_shader5 and GL_EXT_gpu_shader5 support The two extensions are identical, and are largely taking bits of already existing desktop functionality. We continue to do a poor job of supporting the 'precise' keyword, just like we do on desktop. This passes the relevant dEQP tests that I could find. Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
2016-02-19 19:08:35 -05:00
EXT(OES_gpu_shader5, false, true, ARB_gpu_shader5),
glsl: Add usual extension boilerplate for OES_shader_image_atomic. v2: No need for extension enable bits (Ilia). Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2016-02-22 15:06:45 -08:00
EXT(OES_shader_image_atomic, false, true, ARB_shader_image_load_store),
glsl: Sort the list of extensions ARB, OES, then everything else. If there's ever a KHR shading language extension, it should go between ARB and OES. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Acked-by: Eric Anholt <eric@anholt.net>
2014-03-26 11:43:32 -07:00
EXT(OES_standard_derivatives, false, true, OES_standard_derivatives),
mesa: Replace gl_extensions::EXT_texture3D with ::dummy_true Mesa unconditionally sets this driver flag to true in _mesa_init_extensions(). There is therefore no need for the driver to communicate support for this extension. Replace the driver capability flag with ::dummy_true. Signed-off-by: Nanley Chery <nanley.g.chery@intel.com> Reviewed-by: Chad Versace <chad.versace@intel.com>
2015-10-16 10:14:39 -07:00
EXT(OES_texture_3D, false, true, dummy_true),
glsl: add OES_texture_buffer and EXT_texture_buffer support Expose the samplerBuffer/imageBuffer types, and allow the various functions to operate on them. Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2016-02-27 16:13:50 -05:00
EXT(OES_texture_buffer, false, true, OES_texture_buffer),
glsl: add support for OES_texture_storage_multisample_2d_array v2: use ARB_texture_multisample enable bit Patch adds extension enable bit and enables required keywords and builtin functions for the extension. Signed-off-by: Tapani Pälli <tapani.palli@intel.com> Reviewed-by: Marta Lofstedt <marta.lofstedt@intel.com> Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-08-21 09:42:10 +03:00
EXT(OES_texture_storage_multisample_2d_array, false, true, ARB_texture_multisample),
glsl: Sort the list of extensions ARB, OES, then everything else. If there's ever a KHR shading language extension, it should go between ARB and OES. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Acked-by: Eric Anholt <eric@anholt.net>
2014-03-26 11:43:32 -07:00
/* All other extensions go here, sorted alphabetically.
*/
EXT(AMD_conservative_depth, true, false, ARB_conservative_depth),
EXT(AMD_shader_stencil_export, true, false, ARB_shader_stencil_export),
EXT(AMD_shader_trinary_minmax, true, false, dummy_true),
glsl: Remove shader stage checking for extension handling. Certain extensions only add functionality to particular shader stages. (For example, ARB_draw_instanced only adds variables to the vertex shader stage.) Previously, we only allowed such extensions to be enabled in the shader stages where they're useful. However, I've never found any text which mandates that behavior; in my opinion, you should be able to turn on extensions in any shader stage, even if they have no effect. Fixes Piglit tests glslparsertest/glsl2/draw_buffers-05.vert and ARB_draw_instanced/preprocessor/feature-macro-enabled.frag. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29185 Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
2013-07-29 10:40:51 -07:00
EXT(AMD_vertex_shader_layer, true, false, AMD_vertex_shader_layer),
glsl: add support for AMD_vertex_shader_viewport_index Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Chris Forbes <chrisf@ijw.co.nz> Tested-by: Tobias Droste <tdroste@gmx.de>
2014-07-02 12:12:51 -04:00
EXT(AMD_vertex_shader_viewport_index, true, false, AMD_vertex_shader_viewport_index),
glsl: add EXT_blend_func_extended parser enables This adds a state for the maximum dual source draw variables available and the variable for determining if the extension has been enabled in the program shaders. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-11-05 10:53:40 -06:00
EXT(EXT_blend_func_extended, false, true, ARB_blend_func_extended),
mesa/glsl/glapi: enable GL_EXT_draw_buffers extension Patch enables ES2 extension that utilizes existing ES3 functionality. Changes make all the subtests to run and pass in WebGL conformance test 'webgl-draw-buffers' when running Chrome on OpenGL ES, also Piglit test 'draw_buffers_gles2' passes. v2: remove unused boolean (Ilia Mirkin) v3: proper error checking for invalid values (Chad Versace) v4: run error check explicitly for ES2 and ES3 (Kenneth Graunke) Signed-off-by: Tapani Pälli <tapani.palli@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com> Reviewed-by: Chad Versace <chad.versace@intel.com>
2014-11-25 06:10:30 -05:00
EXT(EXT_draw_buffers, false, true, dummy_true),
mesa: add GL_OES_gpu_shader5 and GL_EXT_gpu_shader5 support The two extensions are identical, and are largely taking bits of already existing desktop functionality. We continue to do a poor job of supporting the 'precise' keyword, just like we do on desktop. This passes the relevant dEQP tests that I could find. Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
2016-02-19 19:08:35 -05:00
EXT(EXT_gpu_shader5, false, true, ARB_gpu_shader5),
mesa: Enable GL_EXT_separate_shader_objects for OpenGL ES Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Eric Anholt <eric@anholt.net>
2014-03-26 13:11:44 -07:00
EXT(EXT_separate_shader_objects, false, true, dummy_true),
mesa: Rename MESA_shader_integer_mix to EXT_shader_integer_mix Everyone at the Khronos meeting was as surprised that GLSL didn't already support this as we were. Several vendors said they'd ship it, but there didn't seem to be enough interest to put in the effort to make it ARB or KHR. v2: Fix a couple typos and rename the spec file to EXT_shader_integer_mix.spec. Suggested by Roland. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com> Reviewed-by: Roland Scheidegger <sroland@vmware.com>
2013-09-12 11:40:00 -05:00
EXT(EXT_shader_integer_mix, true, true, EXT_shader_integer_mix),
glsl: Extension tracking for EXT_shader_samples_indentical Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
2015-11-17 15:36:15 -08:00
EXT(EXT_shader_samples_identical, true, true, EXT_shader_samples_identical),
glsl: Sort the list of extensions ARB, OES, then everything else. If there's ever a KHR shading language extension, it should go between ARB and OES. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Acked-by: Eric Anholt <eric@anholt.net>
2014-03-26 11:43:32 -07:00
EXT(EXT_texture_array, true, false, EXT_texture_array),
glsl: add OES_texture_buffer and EXT_texture_buffer support Expose the samplerBuffer/imageBuffer types, and allow the various functions to operate on them. Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2016-02-27 16:13:50 -05:00
EXT(EXT_texture_buffer, false, true, OES_texture_buffer),
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
};
#undef EXT
/**
* Determine whether a given extension is compatible with the target,
* API, and extension information in the current parser state.
*/
bool _mesa_glsl_extension::compatible_with_state(const _mesa_glsl_parse_state *
state) const
{
/* Check that this extension matches whether we are compiling
* for desktop GL or GLES.
*/
if (state->es_shader) {
if (!this->avail_in_ES) return false;
} else {
if (!this->avail_in_GL) return false;
}
/* Check that this extension is supported by the OpenGL
* implementation.
*
* Note: the ->* operator indexes into state->extensions by the
* offset this->supported_flag. See
* _mesa_glsl_extension::supported_flag for more info.
*/
return state->extensions->*(this->supported_flag);
}
/**
* Set the appropriate flags in the parser state to establish the
* given behavior for this extension.
*/
void _mesa_glsl_extension::set_flags(_mesa_glsl_parse_state *state,
ext_behavior behavior) const
{
/* Note: the ->* operator indexes into state by the
* offsets this->enable_flag and this->warn_flag. See
* _mesa_glsl_extension::supported_flag for more info.
*/
state->*(this->enable_flag) = (behavior != extension_disable);
state->*(this->warn_flag) = (behavior == extension_warn);
}
/**
* Find an extension by name in _mesa_glsl_supported_extensions. If
* the name is not found, return NULL.
*/
static const _mesa_glsl_extension *find_extension(const char *name)
{
glsl: replace Elements() with ARRAY_SIZE() Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-02-28 09:11:23 -07:00
for (unsigned i = 0; i < ARRAY_SIZE(_mesa_glsl_supported_extensions); ++i) {
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
if (strcmp(name, _mesa_glsl_supported_extensions[i].name) == 0) {
return &_mesa_glsl_supported_extensions[i];
}
}
return NULL;
}
Begin processing #extension directive Nowhere near complete. It just parses correctly at this point.
2010-04-07 16:46:25 -07:00
bool
_mesa_glsl_process_extension(const char *name, YYLTYPE *name_locp,
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
const char *behavior_string, YYLTYPE *behavior_locp,
Begin processing #extension directive Nowhere near complete. It just parses correctly at this point.
2010-04-07 16:46:25 -07:00
_mesa_glsl_parse_state *state)
{
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
ext_behavior behavior;
if (strcmp(behavior_string, "warn") == 0) {
behavior = extension_warn;
} else if (strcmp(behavior_string, "require") == 0) {
behavior = extension_require;
} else if (strcmp(behavior_string, "enable") == 0) {
behavior = extension_enable;
} else if (strcmp(behavior_string, "disable") == 0) {
behavior = extension_disable;
Begin processing #extension directive Nowhere near complete. It just parses correctly at this point.
2010-04-07 16:46:25 -07:00
} else {
_mesa_glsl_error(behavior_locp, state,
glsl: Be consistent about '\n', '.', and capitalization in errors/warnings. The majority of calls to _mesa_glsl_error(), _mesa_glsl_warning(), and _mesa_glsl_parse_state::check_version() use a message that begins with a lower case letter and ends without a period. This patch makes all messages follow that convention. Also, error/warning messages shouldn't end in '\n', since _mesa_glsl_msg() automatically adds '\n' at the end of the message. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-25 19:56:43 -07:00
"unknown extension behavior `%s'",
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
behavior_string);
Begin processing #extension directive Nowhere near complete. It just parses correctly at this point.
2010-04-07 16:46:25 -07:00
return false;
}
if (strcmp(name, "all") == 0) {
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
if ((behavior == extension_enable) || (behavior == extension_require)) {
glsl: Be consistent about '\n', '.', and capitalization in errors/warnings. The majority of calls to _mesa_glsl_error(), _mesa_glsl_warning(), and _mesa_glsl_parse_state::check_version() use a message that begins with a lower case letter and ends without a period. This patch makes all messages follow that convention. Also, error/warning messages shouldn't end in '\n', since _mesa_glsl_msg() automatically adds '\n' at the end of the message. Reviewed-by: Matt Turner <mattst88@gmail.com>
2013-07-25 19:56:43 -07:00
_mesa_glsl_error(name_locp, state, "cannot %s all extensions",
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
(behavior == extension_enable)
Begin processing #extension directive Nowhere near complete. It just parses correctly at this point.
2010-04-07 16:46:25 -07:00
? "enable" : "require");
return false;
Add support for GL_ARB_draw_buffers extension
2010-04-07 16:59:46 -07:00
} else {
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
for (unsigned i = 0;
glsl: replace Elements() with ARRAY_SIZE() Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-02-28 09:11:23 -07:00
i < ARRAY_SIZE(_mesa_glsl_supported_extensions); ++i) {
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
const _mesa_glsl_extension *extension
= &_mesa_glsl_supported_extensions[i];
if (extension->compatible_with_state(state)) {
_mesa_glsl_supported_extensions[i].set_flags(state, behavior);
}
}
Add support for GL_ARB_draw_buffers extension
2010-04-07 16:59:46 -07:00
}
Begin processing #extension directive Nowhere near complete. It just parses correctly at this point.
2010-04-07 16:46:25 -07:00
} else {
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
const _mesa_glsl_extension *extension = find_extension(name);
if (extension && extension->compatible_with_state(state)) {
extension->set_flags(state, behavior);
Emit a warning when an unknown extension is used with #extension
2010-04-07 14:50:36 -07:00
} else {
glsl: make static constant variables "static const" This allows them to be moved to .rodata, and allow us to be sure that they will not be modified. Signed-off-by: Chia-I Wu <olv@lunarg.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Timothy Arceri <t_arceri@yahoo.com.au>
2014-04-22 16:58:16 +08:00
static const char fmt[] = "extension `%s' unsupported in %s shader";
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
if (behavior == extension_require) {
_mesa_glsl_error(name_locp, state, fmt,
mesa: Clean up nomenclature for pipeline stages. Previously, we had an enum called gl_shader_type which represented pipeline stages in the order they occur in the pipeline (i.e. MESA_SHADER_VERTEX=0, MESA_SHADER_GEOMETRY=1, etc), and several inconsistently named functions for converting between it and other representations: - _mesa_shader_type_to_string: gl_shader_type -> string - _mesa_shader_type_to_index: GLenum (GL_*_SHADER) -> gl_shader_type - _mesa_program_target_to_index: GLenum (GL_*_PROGRAM) -> gl_shader_type - _mesa_shader_enum_to_string: GLenum (GL_*_{SHADER,PROGRAM}) -> string This patch tries to clean things up so that we use more consistent terminology: the enum is now called gl_shader_stage (to emphasize that it is in the order of pipeline stages), and the conversion functions are: - _mesa_shader_stage_to_string: gl_shader_stage -> string - _mesa_shader_enum_to_shader_stage: GLenum (GL_*_SHADER) -> gl_shader_stage - _mesa_program_enum_to_shader_stage: GLenum (GL_*_PROGRAM) -> gl_shader_stage - _mesa_progshader_enum_to_string: GLenum (GL_*_{SHADER,PROGRAM}) -> string In addition, MESA_SHADER_TYPES has been renamed to MESA_SHADER_STAGES, for consistency with the new name for the enum. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> v2: Also rename the "target" field of _mesa_glsl_parse_state and the "target" parameter of _mesa_shader_stage_to_string to "stage". Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-07 10:11:39 -08:00
name, _mesa_shader_stage_to_string(state->stage));
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
return false;
} else {
_mesa_glsl_warning(name_locp, state, fmt,
mesa: Clean up nomenclature for pipeline stages. Previously, we had an enum called gl_shader_type which represented pipeline stages in the order they occur in the pipeline (i.e. MESA_SHADER_VERTEX=0, MESA_SHADER_GEOMETRY=1, etc), and several inconsistently named functions for converting between it and other representations: - _mesa_shader_type_to_string: gl_shader_type -> string - _mesa_shader_type_to_index: GLenum (GL_*_SHADER) -> gl_shader_type - _mesa_program_target_to_index: GLenum (GL_*_PROGRAM) -> gl_shader_type - _mesa_shader_enum_to_string: GLenum (GL_*_{SHADER,PROGRAM}) -> string This patch tries to clean things up so that we use more consistent terminology: the enum is now called gl_shader_stage (to emphasize that it is in the order of pipeline stages), and the conversion functions are: - _mesa_shader_stage_to_string: gl_shader_stage -> string - _mesa_shader_enum_to_shader_stage: GLenum (GL_*_SHADER) -> gl_shader_stage - _mesa_program_enum_to_shader_stage: GLenum (GL_*_PROGRAM) -> gl_shader_stage - _mesa_progshader_enum_to_string: GLenum (GL_*_{SHADER,PROGRAM}) -> string In addition, MESA_SHADER_TYPES has been renamed to MESA_SHADER_STAGES, for consistency with the new name for the enum. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> v2: Also rename the "target" field of _mesa_glsl_parse_state and the "target" parameter of _mesa_shader_stage_to_string to "stage". Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-07 10:11:39 -08:00
name, _mesa_shader_stage_to_string(state->stage));
glsl: Rewrote _mesa_glsl_process_extension to use table-driven logic. Instead of using a chain of manually maintained if/else blocks to handle "#extension" directives, we now consult a table that specifies, for each extension, the circumstances under which it is available, and what flags in _mesa_glsl_parse_state need to be set in order to activate it. This makes it easier to add new GLSL extensions in the future, and fixes the following bugs: - Previously, _mesa_glsl_process_extension would sometimes set the "_enable" and "_warn" flags for an extension before checking whether the extension was supported by the driver; as a result, specifying "enable" behavior for an unsupported extension would sometimes cause front-end support for that extension to be switched on in spite of the fact that back-end support was not available, leading to strange failures, such as those in https://bugs.freedesktop.org/show_bug.cgi?id=38015. - "#extension all: warn" and "#extension all: disable" had no effect. Notes: - All extensions are currently marked as unavailable in geometry shaders. This should not have any adverse effects since geometry shaders aren't supported yet. When we return to working on geometry shader support, we'll need to update the table for those extensions that are available in geometry shaders. - Previous to this commit, if a shader mentioned ARB_shader_texture_lod, extension ARB_texture_rectangle would be automatically turned on in order to ensure that the types sampler2DRect and sampler2DRectShadow would be defined. This was unnecessary, because (a) ARB_shader_texture_lod works perfectly well without those types provided that the builtin functions that reference them are not called, and (b) ARB_texture_rectangle is enabled by default in non-ES contexts anyway. I eliminated this unnecessary behavior in order to make the behavior of all extensions consistent. NOTE: This is a candidate for the 7.10 and 7.11 branches. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2011-06-24 15:34:04 -07:00
}
Begin processing #extension directive Nowhere near complete. It just parses correctly at this point.
2010-04-07 16:46:25 -07:00
}
}
return true;
}
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
/**
* Recurses through <type> and <expr> if <expr> is an aggregate initializer
* and sets <expr>'s <constructor_type> field to <type>. Gives later functions
* (process_array_constructor, et al) sufficient information to do type
* checking.
*
* Operates on assignments involving an aggregate initializer. E.g.,
*
* vec4 pos = {1.0, -1.0, 0.0, 1.0};
*
* or more ridiculously,
*
* struct S {
* vec4 v[2];
* };
*
* struct {
* S a[2], b;
* int c;
* } aggregate = {
* {
* {
* {
* {1.0, 2.0, 3.0, 4.0}, // a[0].v[0]
* {5.0, 6.0, 7.0, 8.0} // a[0].v[1]
* } // a[0].v
* }, // a[0]
* {
* {
* {1.0, 2.0, 3.0, 4.0}, // a[1].v[0]
* {5.0, 6.0, 7.0, 8.0} // a[1].v[1]
* } // a[1].v
* } // a[1]
* }, // a
* {
* {
* {1.0, 2.0, 3.0, 4.0}, // b.v[0]
* {5.0, 6.0, 7.0, 8.0} // b.v[1]
* } // b.v
* }, // b
* 4 // c
* };
*
* This pass is necessary because the right-hand side of <type> e = { ... }
* doesn't contain sufficient information to determine if the types match.
*/
void
glsl: Simplify aggregate type inference to prepare for ARB_arrays_of_arrays. Most of the time it is not necessary to perform type inference to compile GLSL; the type of every expression can be inferred from the contents of the expression itself (and previous type declarations). The exception is aggregate initializers: their type is determined by the LHS of the variable being assigned to. For example, in the statement: mat2 foo = { { 1, 2 }, { 3, 4 } }; the type of { 1, 2 } is only known to be vec2 (as opposed to, say, ivec2, uvec2, int[2], or a struct) because of the fact that the result is being assigned to a mat2. Previous to this patch, we handled this situation by doing some type inference during parsing: when parsing a declaration like the one above, we would call _mesa_set_aggregate_type(), which would infer the type of each aggregate initializer and store it in the corresponding ast_aggregate_initializer::constructor_type field. Since this happened at parse time, we couldn't do the type inference using glsl_type objects; we had to use ast_type_specifiers, which are much more awkward to work with. Things are about to get more complicated when we add support for ARB_arrays_of_arrays. This patch simplifies things by postponing the call to _mesa_set_aggregate_type() until ast-to-hir time, when we have access to glsl_type objects. As a side benefit, we only need to have one call to _mesa_set_aggregate_type() now, instead of six. Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-01-21 15:41:26 -08:00
_mesa_ast_set_aggregate_type(const glsl_type *type,
ast_expression *expr)
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
{
ast_aggregate_initializer *ai = (ast_aggregate_initializer *)expr;
glsl: Simplify aggregate type inference to prepare for ARB_arrays_of_arrays. Most of the time it is not necessary to perform type inference to compile GLSL; the type of every expression can be inferred from the contents of the expression itself (and previous type declarations). The exception is aggregate initializers: their type is determined by the LHS of the variable being assigned to. For example, in the statement: mat2 foo = { { 1, 2 }, { 3, 4 } }; the type of { 1, 2 } is only known to be vec2 (as opposed to, say, ivec2, uvec2, int[2], or a struct) because of the fact that the result is being assigned to a mat2. Previous to this patch, we handled this situation by doing some type inference during parsing: when parsing a declaration like the one above, we would call _mesa_set_aggregate_type(), which would infer the type of each aggregate initializer and store it in the corresponding ast_aggregate_initializer::constructor_type field. Since this happened at parse time, we couldn't do the type inference using glsl_type objects; we had to use ast_type_specifiers, which are much more awkward to work with. Things are about to get more complicated when we add support for ARB_arrays_of_arrays. This patch simplifies things by postponing the call to _mesa_set_aggregate_type() until ast-to-hir time, when we have access to glsl_type objects. As a side benefit, we only need to have one call to _mesa_set_aggregate_type() now, instead of six. Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-01-21 15:41:26 -08:00
ai->constructor_type = type;
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
/* If the aggregate is an array, recursively set its elements' types. */
glsl: Simplify aggregate type inference to prepare for ARB_arrays_of_arrays. Most of the time it is not necessary to perform type inference to compile GLSL; the type of every expression can be inferred from the contents of the expression itself (and previous type declarations). The exception is aggregate initializers: their type is determined by the LHS of the variable being assigned to. For example, in the statement: mat2 foo = { { 1, 2 }, { 3, 4 } }; the type of { 1, 2 } is only known to be vec2 (as opposed to, say, ivec2, uvec2, int[2], or a struct) because of the fact that the result is being assigned to a mat2. Previous to this patch, we handled this situation by doing some type inference during parsing: when parsing a declaration like the one above, we would call _mesa_set_aggregate_type(), which would infer the type of each aggregate initializer and store it in the corresponding ast_aggregate_initializer::constructor_type field. Since this happened at parse time, we couldn't do the type inference using glsl_type objects; we had to use ast_type_specifiers, which are much more awkward to work with. Things are about to get more complicated when we add support for ARB_arrays_of_arrays. This patch simplifies things by postponing the call to _mesa_set_aggregate_type() until ast-to-hir time, when we have access to glsl_type objects. As a side benefit, we only need to have one call to _mesa_set_aggregate_type() now, instead of six. Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-01-21 15:41:26 -08:00
if (type->is_array()) {
glsl: remove element_type() helper We now have is_array() and without_array() that make the code much clearer and remove the need for this. For all remaining calls to this we already knew that the type was an array so returning a null wasn't adding any value. v2: use without_array() in _mesa_ast_array_index_to_hir() and don't use without_array() in lower_clip_distance_visitor() as we want to make sure the array is 2D. Reviewed-by: Matt Turner <mattst88@gmail.com>
2015-04-30 20:45:54 +10:00
/* Each array element has the type type->fields.array.
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
*
* E.g., if <type> if struct S[2] we want to set each element's type to
* struct S.
*/
for (exec_node *expr_node = ai->expressions.head;
!expr_node->is_tail_sentinel();
expr_node = expr_node->next) {
ast_expression *expr = exec_node_data(ast_expression, expr_node,
link);
if (expr->oper == ast_aggregate)
glsl: remove element_type() helper We now have is_array() and without_array() that make the code much clearer and remove the need for this. For all remaining calls to this we already knew that the type was an array so returning a null wasn't adding any value. v2: use without_array() in _mesa_ast_array_index_to_hir() and don't use without_array() in lower_clip_distance_visitor() as we want to make sure the array is 2D. Reviewed-by: Matt Turner <mattst88@gmail.com>
2015-04-30 20:45:54 +10:00
_mesa_ast_set_aggregate_type(type->fields.array, expr);
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
}
/* If the aggregate is a struct, recursively set its fields' types. */
glsl: Simplify aggregate type inference to prepare for ARB_arrays_of_arrays. Most of the time it is not necessary to perform type inference to compile GLSL; the type of every expression can be inferred from the contents of the expression itself (and previous type declarations). The exception is aggregate initializers: their type is determined by the LHS of the variable being assigned to. For example, in the statement: mat2 foo = { { 1, 2 }, { 3, 4 } }; the type of { 1, 2 } is only known to be vec2 (as opposed to, say, ivec2, uvec2, int[2], or a struct) because of the fact that the result is being assigned to a mat2. Previous to this patch, we handled this situation by doing some type inference during parsing: when parsing a declaration like the one above, we would call _mesa_set_aggregate_type(), which would infer the type of each aggregate initializer and store it in the corresponding ast_aggregate_initializer::constructor_type field. Since this happened at parse time, we couldn't do the type inference using glsl_type objects; we had to use ast_type_specifiers, which are much more awkward to work with. Things are about to get more complicated when we add support for ARB_arrays_of_arrays. This patch simplifies things by postponing the call to _mesa_set_aggregate_type() until ast-to-hir time, when we have access to glsl_type objects. As a side benefit, we only need to have one call to _mesa_set_aggregate_type() now, instead of six. Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-01-21 15:41:26 -08:00
} else if (type->is_record()) {
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
exec_node *expr_node = ai->expressions.head;
glsl: Simplify aggregate type inference to prepare for ARB_arrays_of_arrays. Most of the time it is not necessary to perform type inference to compile GLSL; the type of every expression can be inferred from the contents of the expression itself (and previous type declarations). The exception is aggregate initializers: their type is determined by the LHS of the variable being assigned to. For example, in the statement: mat2 foo = { { 1, 2 }, { 3, 4 } }; the type of { 1, 2 } is only known to be vec2 (as opposed to, say, ivec2, uvec2, int[2], or a struct) because of the fact that the result is being assigned to a mat2. Previous to this patch, we handled this situation by doing some type inference during parsing: when parsing a declaration like the one above, we would call _mesa_set_aggregate_type(), which would infer the type of each aggregate initializer and store it in the corresponding ast_aggregate_initializer::constructor_type field. Since this happened at parse time, we couldn't do the type inference using glsl_type objects; we had to use ast_type_specifiers, which are much more awkward to work with. Things are about to get more complicated when we add support for ARB_arrays_of_arrays. This patch simplifies things by postponing the call to _mesa_set_aggregate_type() until ast-to-hir time, when we have access to glsl_type objects. As a side benefit, we only need to have one call to _mesa_set_aggregate_type() now, instead of six. Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-01-21 15:41:26 -08:00
/* Iterate through the struct's fields. */
for (unsigned i = 0; !expr_node->is_tail_sentinel() && i < type->length;
i++, expr_node = expr_node->next) {
ast_expression *expr = exec_node_data(ast_expression, expr_node,
link);
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
glsl: Simplify aggregate type inference to prepare for ARB_arrays_of_arrays. Most of the time it is not necessary to perform type inference to compile GLSL; the type of every expression can be inferred from the contents of the expression itself (and previous type declarations). The exception is aggregate initializers: their type is determined by the LHS of the variable being assigned to. For example, in the statement: mat2 foo = { { 1, 2 }, { 3, 4 } }; the type of { 1, 2 } is only known to be vec2 (as opposed to, say, ivec2, uvec2, int[2], or a struct) because of the fact that the result is being assigned to a mat2. Previous to this patch, we handled this situation by doing some type inference during parsing: when parsing a declaration like the one above, we would call _mesa_set_aggregate_type(), which would infer the type of each aggregate initializer and store it in the corresponding ast_aggregate_initializer::constructor_type field. Since this happened at parse time, we couldn't do the type inference using glsl_type objects; we had to use ast_type_specifiers, which are much more awkward to work with. Things are about to get more complicated when we add support for ARB_arrays_of_arrays. This patch simplifies things by postponing the call to _mesa_set_aggregate_type() until ast-to-hir time, when we have access to glsl_type objects. As a side benefit, we only need to have one call to _mesa_set_aggregate_type() now, instead of six. Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-01-21 15:41:26 -08:00
if (expr->oper == ast_aggregate) {
_mesa_ast_set_aggregate_type(type->fields.structure[i].type, expr);
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
}
}
glsl: Simplify aggregate type inference to prepare for ARB_arrays_of_arrays. Most of the time it is not necessary to perform type inference to compile GLSL; the type of every expression can be inferred from the contents of the expression itself (and previous type declarations). The exception is aggregate initializers: their type is determined by the LHS of the variable being assigned to. For example, in the statement: mat2 foo = { { 1, 2 }, { 3, 4 } }; the type of { 1, 2 } is only known to be vec2 (as opposed to, say, ivec2, uvec2, int[2], or a struct) because of the fact that the result is being assigned to a mat2. Previous to this patch, we handled this situation by doing some type inference during parsing: when parsing a declaration like the one above, we would call _mesa_set_aggregate_type(), which would infer the type of each aggregate initializer and store it in the corresponding ast_aggregate_initializer::constructor_type field. Since this happened at parse time, we couldn't do the type inference using glsl_type objects; we had to use ast_type_specifiers, which are much more awkward to work with. Things are about to get more complicated when we add support for ARB_arrays_of_arrays. This patch simplifies things by postponing the call to _mesa_set_aggregate_type() until ast-to-hir time, when we have access to glsl_type objects. As a side benefit, we only need to have one call to _mesa_set_aggregate_type() now, instead of six. Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-01-21 15:41:26 -08:00
/* If the aggregate is a matrix, set its columns' types. */
} else if (type->is_matrix()) {
for (exec_node *expr_node = ai->expressions.head;
!expr_node->is_tail_sentinel();
expr_node = expr_node->next) {
ast_expression *expr = exec_node_data(ast_expression, expr_node,
link);
if (expr->oper == ast_aggregate)
_mesa_ast_set_aggregate_type(type->column_type(), expr);
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
}
}
}
glsl: Move interface block processing to glsl_parser_extras.cpp No functional changes. Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-10 10:00:12 +02:00
void
_mesa_ast_process_interface_block(YYLTYPE *locp,
_mesa_glsl_parse_state *state,
ast_interface_block *const block,
glsl: Pass ast_type_qualifier by const reference. Coverity noticed that we were passing this by value, and it's 152 bytes. Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
2015-11-23 14:22:48 -08:00
const struct ast_type_qualifier &q)
glsl: Move interface block processing to glsl_parser_extras.cpp No functional changes. Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-10 10:00:12 +02:00
{
glsl: allow default qualifiers for shader storage block definitions This kind of definitions: layout(xxx) buffer; was not supported by commit 84fc5fece006. Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-03 12:00:16 +02:00
if (q.flags.q.buffer) {
if (!state->has_shader_storage_buffer_objects()) {
_mesa_glsl_error(locp, state,
"#version 430 / GL_ARB_shader_storage_buffer_object "
"required for defining shader storage blocks");
} else if (state->ARB_shader_storage_buffer_object_warn) {
_mesa_glsl_warning(locp, state,
"#version 430 / GL_ARB_shader_storage_buffer_object "
"required for defining shader storage blocks");
}
} else if (q.flags.q.uniform) {
glsl: Move interface block processing to glsl_parser_extras.cpp No functional changes. Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-10 10:00:12 +02:00
if (!state->has_uniform_buffer_objects()) {
_mesa_glsl_error(locp, state,
"#version 140 / GL_ARB_uniform_buffer_object "
"required for defining uniform blocks");
} else if (state->ARB_uniform_buffer_object_warn) {
_mesa_glsl_warning(locp, state,
"#version 140 / GL_ARB_uniform_buffer_object "
"required for defining uniform blocks");
}
} else {
if (state->es_shader || state->language_version < 150) {
_mesa_glsl_error(locp, state,
"#version 150 required for using "
"interface blocks");
}
}
/* From the GLSL 1.50.11 spec, section 4.3.7 ("Interface Blocks"):
* "It is illegal to have an input block in a vertex shader
* or an output block in a fragment shader"
*/
if ((state->stage == MESA_SHADER_VERTEX) && q.flags.q.in) {
_mesa_glsl_error(locp, state,
"`in' interface block is not allowed for "
"a vertex shader");
} else if ((state->stage == MESA_SHADER_FRAGMENT) && q.flags.q.out) {
_mesa_glsl_error(locp, state,
"`out' interface block is not allowed for "
"a fragment shader");
}
/* Since block arrays require names, and both features are added in
* the same language versions, we don't have to explicitly
* version-check both things.
*/
if (block->instance_name != NULL) {
state->check_version(150, 300, locp, "interface blocks with "
"an instance name are not allowed");
}
uint64_t interface_type_mask;
struct ast_type_qualifier temp_type_qualifier;
glsl: allow default qualifiers for shader storage block definitions This kind of definitions: layout(xxx) buffer; was not supported by commit 84fc5fece006. Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-03 12:00:16 +02:00
/* Get a bitmask containing only the in/out/uniform/buffer
glsl: Move interface block processing to glsl_parser_extras.cpp No functional changes. Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-10 10:00:12 +02:00
* flags, allowing us to ignore other irrelevant flags like
* interpolation qualifiers.
*/
temp_type_qualifier.flags.i = 0;
temp_type_qualifier.flags.q.uniform = true;
temp_type_qualifier.flags.q.in = true;
temp_type_qualifier.flags.q.out = true;
glsl: allow default qualifiers for shader storage block definitions This kind of definitions: layout(xxx) buffer; was not supported by commit 84fc5fece006. Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com> Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-03 12:00:16 +02:00
temp_type_qualifier.flags.q.buffer = true;
glsl: Move interface block processing to glsl_parser_extras.cpp No functional changes. Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-10 10:00:12 +02:00
interface_type_mask = temp_type_qualifier.flags.i;
/* Get the block's interface qualifier. The interface_qualifier
* production rule guarantees that only one bit will be set (and
* it will be in/out/uniform).
*/
uint64_t block_interface_qualifier = q.flags.i;
block->layout.flags.i |= block_interface_qualifier;
if (state->stage == MESA_SHADER_GEOMETRY &&
glsl: only apply default stream to output blocks This is needed to allow invalid qualifier checks on inputs. Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
2016-02-26 10:59:14 +11:00
state->has_explicit_attrib_stream() &&
block->layout.flags.q.out) {
glsl: Move interface block processing to glsl_parser_extras.cpp No functional changes. Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-10 10:00:12 +02:00
/* Assign global layout's stream value. */
block->layout.flags.q.stream = 1;
block->layout.flags.q.explicit_stream = 0;
block->layout.stream = state->out_qualifier->stream;
}
foreach_list_typed (ast_declarator_list, member, link, &block->declarations) {
ast_type_qualifier& qualifier = member->type->qualifier;
if ((qualifier.flags.i & interface_type_mask) == 0) {
/* GLSLangSpec.1.50.11, 4.3.7 (Interface Blocks):
* "If no optional qualifier is used in a member declaration, the
* qualifier of the variable is just in, out, or uniform as declared
* by interface-qualifier."
*/
qualifier.flags.i |= block_interface_qualifier;
} else if ((qualifier.flags.i & interface_type_mask) !=
block_interface_qualifier) {
/* GLSLangSpec.1.50.11, 4.3.7 (Interface Blocks):
* "If optional qualifiers are used, they can include interpolation
* and storage qualifiers and they must declare an input, output,
* or uniform variable consistent with the interface qualifier of
* the block."
*/
_mesa_glsl_error(locp, state,
"uniform/in/out qualifier on "
"interface block member does not match "
"the interface block");
}
glsl: Allow invariant qualifer in block members in desktop OpenGL. Feedback from Khronos is that 'invariant' should be allowed on block members for desktop OpenGL. Fix piglit regression added by fe1e89a0: invariant-qualifier-in-out-block-01.vert v2: - Allow it for in/out blocks in OpenGL ES too, so when OES_shader_io_blocks is supported we don't need to do any change (Timothy) Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89330 Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
2016-02-11 07:15:50 +01:00
if (!(q.flags.q.in || q.flags.q.out) && qualifier.flags.q.invariant)
glsl: Move interface block processing to glsl_parser_extras.cpp No functional changes. Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-10 10:00:12 +02:00
_mesa_glsl_error(locp, state,
glsl: Allow invariant qualifer in block members in desktop OpenGL. Feedback from Khronos is that 'invariant' should be allowed on block members for desktop OpenGL. Fix piglit regression added by fe1e89a0: invariant-qualifier-in-out-block-01.vert v2: - Allow it for in/out blocks in OpenGL ES too, so when OES_shader_io_blocks is supported we don't need to do any change (Timothy) Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89330 Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
2016-02-11 07:15:50 +01:00
"invariant qualifiers can be used only "
"in interface block members for shader "
"inputs or outputs");
glsl: Move interface block processing to glsl_parser_extras.cpp No functional changes. Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
2015-09-10 10:00:12 +02:00
}
}
glsl: Add support for C-style initializers. Required by GL_ARB_shading_language_420pack. Parts based on work done by Todd Previte and Ken Graunke, implementing basic support for C-style initializers of arrays. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:29:16 -07:00
Initial commit. lol
2010-02-22 13:19:34 -08:00
void
_mesa_ast_type_qualifier_print(const struct ast_type_qualifier *q)
{
glsl: add ast/parser support for subroutine parsing storage (v3.2) This is the guts of the GLSL parser and AST support for shader subroutines. The code creates a subroutine type in the parser, and uses that there to validate the identifiers. The parser also distinguishes between subroutine types/function prototypes /uniforms and subroutine defintions for functions. Then in the AST conversion it recreates the types, and stores the subroutine definition info or subroutine info into the ir_function along with a side lookup table in the parser state. It also converts subroutine calls into the enhanced ir_call. v2: move to handling method calls in function handling not in field selection. v3: merge Chris's previous parser patches in here, to make it clearer what's changed in one place. v3.1: add more documentation, drop unused include v3.2: drop is_subroutine_def Reviewed-by: Chris Forbes <chrisf@ijw.co.nz> Signed-off-by: Dave Airlie <airlied@redhat.com>
2015-06-01 10:55:47 +10:00
if (q->flags.q.subroutine)
printf("subroutine ");
if (q->flags.q.subroutine_def) {
printf("subroutine (");
q->subroutine_list->print();
printf(")");
}
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.constant)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("const ");
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.invariant)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("invariant ");
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.attribute)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("attribute ");
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.varying)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("varying ");
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.in && q->flags.q.out)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("inout ");
else {
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.in)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("in ");
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.out)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("out ");
}
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.centroid)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("centroid ");
glsl: Add frontend support for `sample` auxiliary storage qualifier Signed-off-by: Chris Forbes <chrisf@ijw.co.nz> Reviewed-by: Francisco Jerez <currojerez@riseup.net>
2013-11-29 21:21:56 +13:00
if (q->flags.q.sample)
printf("sample ");
glsl: add the patch in/out qualifier (v2) v2: Dropped some unrelated reordering in glsl_parser.yy as Ken suggested. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-03-05 13:43:17 +01:00
if (q->flags.q.patch)
printf("patch ");
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.uniform)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("uniform ");
glsl: Implement parser support for 'buffer' qualifier This is used to identify shader storage buffer interface blocks where buffer variables are declared. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2015-05-13 10:53:46 +02:00
if (q->flags.q.buffer)
printf("buffer ");
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.smooth)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("smooth ");
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.flat)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("flat ");
glsl: Wrap ast_type_qualifier contents in a struct in a union This will ease adding non-bit fields in the near future.
2010-10-05 16:38:47 -07:00
if (q->flags.q.noperspective)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("noperspective ");
}
void
ast_node::print(void) const
{
Remove ast_node::type It isn't a type (is was enum specifying the kind of node), it was unused, and it was easily confused with actual type fields. Kill with fire.
2010-04-02 11:03:47 -07:00
printf("unhandled node ");
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
ast_node::ast_node(void)
{
ast: Initialize location data in constructor of all ast_node objects. This prevents using uninitialized data in _msea_glsl_error in some cases, (including at least 6 piglit tests). Thanks to valgrind for pointing out the problem!
2010-07-29 16:39:36 -07:00
this->location.source = 0;
glsl: Extend ast location structure to hande end token position. Reviewed-by: Carl Worth <cworth@cworth.org>
2014-02-05 21:18:09 +06:00
this->location.first_line = 0;
this->location.first_column = 0;
this->location.last_line = 0;
this->location.last_column = 0;
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
static void
glsl: remove remaining is_array variables Previously the reason we needed is_array was because we used array_size == NULL to represent both non-arrays and unsized arrays. Now that we use a non-NULL array_specifier to represent an unsized array, is_array is redundant. Signed-off-by: Timothy Arceri <t_arceri@yahoo.com.au> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-01-23 23:22:01 +11:00
ast_opt_array_dimensions_print(const ast_array_specifier *array_specifier)
Initial commit. lol
2010-02-22 13:19:34 -08:00
{
glsl: remove remaining is_array variables Previously the reason we needed is_array was because we used array_size == NULL to represent both non-arrays and unsized arrays. Now that we use a non-NULL array_specifier to represent an unsized array, is_array is redundant. Signed-off-by: Timothy Arceri <t_arceri@yahoo.com.au> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-01-23 23:22:01 +11:00
if (array_specifier)
array_specifier->print();
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
void
ast_compound_statement::print(void) const
{
printf("{\n");
glsl: remove trailing spaces in various files Acked-by: Kenneth Graunke <kenneth@whitecape.org>
2015-11-24 12:40:53 +01:00
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed(ast_node, ast, link, &this->statements) {
Convert ast_node use of simple_node to exec_list and exec_node
2010-05-10 11:17:53 -07:00
ast->print();
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
printf("}\n");
}
ast_compound_statement::ast_compound_statement(int new_scope,
ast_node *statements)
{
this->new_scope = new_scope;
if (statements != NULL) {
Convert ast_node use of simple_node to exec_list and exec_node
2010-05-10 11:17:53 -07:00
this->statements.push_degenerate_list_at_head(&statements->link);
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
}
void
ast_expression::print(void) const
{
switch (oper) {
case ast_assign:
case ast_mul_assign:
case ast_div_assign:
case ast_mod_assign:
case ast_add_assign:
case ast_sub_assign:
case ast_ls_assign:
case ast_rs_assign:
case ast_and_assign:
case ast_xor_assign:
case ast_or_assign:
subexpressions[0]->print();
Add ast_expression_bin subclass of ast_expression The ast_expression_bin subclass is used for all binary expressions such as addition, subtraction, and comparisons. Several other subclasses are soon to follow.
2010-02-22 19:10:25 -08:00
printf("%s ", operator_string(oper));
Initial commit. lol
2010-02-22 13:19:34 -08:00
subexpressions[1]->print();
break;
case ast_field_selection:
subexpressions[0]->print();
printf(". %s ", primary_expression.identifier);
break;
case ast_plus:
case ast_neg:
case ast_bit_not:
case ast_logic_not:
case ast_pre_inc:
case ast_pre_dec:
Add ast_expression_bin subclass of ast_expression The ast_expression_bin subclass is used for all binary expressions such as addition, subtraction, and comparisons. Several other subclasses are soon to follow.
2010-02-22 19:10:25 -08:00
printf("%s ", operator_string(oper));
Initial commit. lol
2010-02-22 13:19:34 -08:00
subexpressions[0]->print();
break;
case ast_post_inc:
case ast_post_dec:
subexpressions[0]->print();
Add ast_expression_bin subclass of ast_expression The ast_expression_bin subclass is used for all binary expressions such as addition, subtraction, and comparisons. Several other subclasses are soon to follow.
2010-02-22 19:10:25 -08:00
printf("%s ", operator_string(oper));
Initial commit. lol
2010-02-22 13:19:34 -08:00
break;
case ast_conditional:
subexpressions[0]->print();
printf("? ");
subexpressions[1]->print();
printf(": ");
AST dump: fixed printing of conditionals. ast_expression::print() had an incorrect index into the subexpressions array, so (a ? b : c) was being incorrectly rendered as (a ? b : b). Signed-off-by: Brian Paul <brianp@vmware.com>
2011-06-03 10:02:32 -07:00
subexpressions[2]->print();
Initial commit. lol
2010-02-22 13:19:34 -08:00
break;
case ast_array_index:
subexpressions[0]->print();
printf("[ ");
subexpressions[1]->print();
printf("] ");
break;
case ast_function_call: {
subexpressions[0]->print();
printf("( ");
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed (ast_node, ast, link, &this->expressions) {
if (&ast->link != this->expressions.get_head())
Fix function call parameter printer to omit extraneous leading comma The output of all test cases was verified to be the same using diff.
2010-05-14 16:06:41 -07:00
printf(", ");
Convert ast_node use of simple_node to exec_list and exec_node
2010-05-10 11:17:53 -07:00
ast->print();
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
printf(") ");
break;
}
case ast_identifier:
printf("%s ", primary_expression.identifier);
break;
case ast_int_constant:
printf("%d ", primary_expression.int_constant);
break;
case ast_uint_constant:
printf("%u ", primary_expression.uint_constant);
break;
case ast_float_constant:
printf("%f ", primary_expression.float_constant);
break;
glsl/ast: Support double floats Signed-off-by: Dave Airlie <airlied@redhat.com> Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2015-02-05 12:04:58 +02:00
case ast_double_constant:
printf("%f ", primary_expression.double_constant);
break;
Initial commit. lol
2010-02-22 13:19:34 -08:00
case ast_bool_constant:
printf("%s ",
primary_expression.bool_constant
? "true" : "false");
break;
case ast_sequence: {
printf("( ");
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed (ast_node, ast, link, & this->expressions) {
if (&ast->link != this->expressions.get_head())
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf(", ");
Convert ast_node use of simple_node to exec_list and exec_node
2010-05-10 11:17:53 -07:00
ast->print();
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
printf(") ");
break;
}
Add ast_expression_bin subclass of ast_expression The ast_expression_bin subclass is used for all binary expressions such as addition, subtraction, and comparisons. Several other subclasses are soon to follow.
2010-02-22 19:10:25 -08:00
glsl: Add infrastructure for aggregate initializers. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:27:50 -07:00
case ast_aggregate: {
printf("{ ");
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed (ast_node, ast, link, & this->expressions) {
if (&ast->link != this->expressions.get_head())
glsl: Add infrastructure for aggregate initializers. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-29 19:27:50 -07:00
printf(", ");
ast->print();
}
printf("} ");
break;
}
Add ast_expression_bin subclass of ast_expression The ast_expression_bin subclass is used for all binary expressions such as addition, subtraction, and comparisons. Several other subclasses are soon to follow.
2010-02-22 19:10:25 -08:00
default:
assert(0);
break;
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
}
ast_expression::ast_expression(int oper,
ast_expression *ex0,
ast_expression *ex1,
glsl: Switch ast_node to the non-zeroing allocator. All member variables of ast_node are already being initialized from its constructor, but some of its derived classes were leaving members uninitialized -- Fix them. Using rzalloc makes it more likely that we will start relying on the allocator to zero out all memory if the class is ever extended with new member variables. That's bad because it ties objects to some specific allocation scheme, and gives unpredictable results when an object is created with a different allocator -- Stack allocation, array allocation, or aggregation inside a different object are some of the useful possibilities that come to my mind. v2: Use NULL initialization instead of default construction for pointers. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-09-20 15:36:38 -07:00
ast_expression *ex2) :
primary_expression()
Initial commit. lol
2010-02-22 13:19:34 -08:00
{
this->oper = ast_operators(oper);
this->subexpressions[0] = ex0;
this->subexpressions[1] = ex1;
this->subexpressions[2] = ex2;
glsl: Track descriptions of some expressions that can't be l-values Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2011-12-23 09:56:03 -08:00
this->non_lvalue_description = NULL;
glsl: add is_lhs bool on ast_expression Useful to know if a expression is the recipient of an assignment or not, that would be used to (for example) raise warnings of "use of uninitialized variable" without getting a false positive when assigning first a variable. By default the value is false, and it is assigned to true on the following cases: * The lhs assignments subexpression * At ast_array_index, on the array itself. * While handling the method on an array, to avoid the warning calling array.length * When computed the cached test expression at test_to_hir, to avoid a duplicate warning on the test expression of a switch. set_is_lhs setter is added, because in some cases (like ast_field_selection) the value need to be propagated on the expression tree. To avoid doing the propatagion if not needed, it skips if no primary_expression.identifier is available. v2: use a new bool on ast_expression, instead of a new parameter on ast_expression::hir (Timothy Arceri) v3: fix style and some typos on comments, initialize is_lhs default value on constructor, to avoid a c++11 feature (Ian Romanick) v4: some tweaks on comments (Timothy Arceri) Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94129 Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
2016-02-25 11:11:54 +01:00
this->is_lhs = false;
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
void
ast_expression_statement::print(void) const
{
if (expression)
expression->print();
printf("; ");
}
ast_expression_statement::ast_expression_statement(ast_expression *ex) :
expression(ex)
{
/* empty */
}
void
ast_function::print(void) const
{
return_type->print();
printf(" %s (", identifier);
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed(ast_node, ast, link, & this->parameters) {
Convert ast_node use of simple_node to exec_list and exec_node
2010-05-10 11:17:53 -07:00
ast->print();
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
printf(")");
}
ast_function::ast_function(void)
glsl: Initialize ast_function member variables. Fixes "Uninitialized pointer field" defect reported by Coverity. Signed-off-by: Vinson Lee <vlee@freedesktop.org>
2013-02-10 01:38:53 +01:00
: return_type(NULL), identifier(NULL), is_definition(false),
signature(NULL)
Initial commit. lol
2010-02-22 13:19:34 -08:00
{
Convert ast_node use of simple_node to exec_list and exec_node
2010-05-10 11:17:53 -07:00
/* empty */
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
void
ast_fully_specified_type::print(void) const
{
_mesa_ast_type_qualifier_print(& qualifier);
specifier->print();
}
void
ast_parameter_declarator::print(void) const
{
type->print();
if (identifier)
printf("%s ", identifier);
glsl: remove remaining is_array variables Previously the reason we needed is_array was because we used array_size == NULL to represent both non-arrays and unsized arrays. Now that we use a non-NULL array_specifier to represent an unsized array, is_array is redundant. Signed-off-by: Timothy Arceri <t_arceri@yahoo.com.au> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-01-23 23:22:01 +11:00
ast_opt_array_dimensions_print(array_specifier);
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
void
ast_function_definition::print(void) const
{
prototype->print();
body->print();
}
void
ast_declaration::print(void) const
{
printf("%s ", identifier);
glsl: remove remaining is_array variables Previously the reason we needed is_array was because we used array_size == NULL to represent both non-arrays and unsized arrays. Now that we use a non-NULL array_specifier to represent an unsized array, is_array is redundant. Signed-off-by: Timothy Arceri <t_arceri@yahoo.com.au> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-01-23 23:22:01 +11:00
ast_opt_array_dimensions_print(array_specifier);
Initial commit. lol
2010-02-22 13:19:34 -08:00
if (initializer) {
printf("= ");
initializer->print();
}
}
glsl: remove remaining is_array variables Previously the reason we needed is_array was because we used array_size == NULL to represent both non-arrays and unsized arrays. Now that we use a non-NULL array_specifier to represent an unsized array, is_array is redundant. Signed-off-by: Timothy Arceri <t_arceri@yahoo.com.au> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-01-23 23:22:01 +11:00
ast_declaration::ast_declaration(const char *identifier,
glsl: Add ARB_arrays_of_arrays support to yacc definition and ast Adds array specifier object to hold array information Signed-off-by: Timothy Arceri <t_arceri@yahoo.com.au> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-01-23 23:16:41 +11:00
ast_array_specifier *array_specifier,
Initial commit. lol
2010-02-22 13:19:34 -08:00
ast_expression *initializer)
{
this->identifier = identifier;
glsl: Add ARB_arrays_of_arrays support to yacc definition and ast Adds array specifier object to hold array information Signed-off-by: Timothy Arceri <t_arceri@yahoo.com.au> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2014-01-23 23:16:41 +11:00
this->array_specifier = array_specifier;
Initial commit. lol
2010-02-22 13:19:34 -08:00
this->initializer = initializer;
}
void
ast_declarator_list::print(void) const
{
assert(type || invariant);
if (type)
type->print();
glsl: Add support for `precise` redeclarations This works like glsl-1.20+'s invariant redeclarations, but with fewer restrictions, since `precise` is allowed on pretty much anything. Signed-off-by: Chris Forbes <chrisf@ijw.co.nz> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-04-27 16:03:54 +12:00
else if (invariant)
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf("invariant ");
glsl: Add support for `precise` redeclarations This works like glsl-1.20+'s invariant redeclarations, but with fewer restrictions, since `precise` is allowed on pretty much anything. Signed-off-by: Chris Forbes <chrisf@ijw.co.nz> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-04-27 16:03:54 +12:00
else
printf("precise ");
Initial commit. lol
2010-02-22 13:19:34 -08:00
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed (ast_node, ast, link, & this->declarations) {
if (&ast->link != this->declarations.get_head())
Initial commit. lol
2010-02-22 13:19:34 -08:00
printf(", ");
Convert ast_node use of simple_node to exec_list and exec_node
2010-05-10 11:17:53 -07:00
ast->print();
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
printf("; ");
}
ast_declarator_list::ast_declarator_list(ast_fully_specified_type *type)
{
this->type = type;
glsl2: Initialize ast_declarator_list::invariant in constructor
2010-07-01 17:10:11 -07:00
this->invariant = false;
glsl: Add support for `precise` redeclarations This works like glsl-1.20+'s invariant redeclarations, but with fewer restrictions, since `precise` is allowed on pretty much anything. Signed-off-by: Chris Forbes <chrisf@ijw.co.nz> Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-04-27 16:03:54 +12:00
this->precise = false;
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
void
ast_jump_statement::print(void) const
{
switch (mode) {
case ast_continue:
printf("continue; ");
break;
case ast_break:
printf("break; ");
break;
case ast_return:
printf("return ");
if (opt_return_value)
opt_return_value->print();
printf("; ");
break;
case ast_discard:
printf("discard; ");
break;
}
}
ast_jump_statement::ast_jump_statement(int mode, ast_expression *return_value)
glsl: Initialize ast_jump_statement::opt_return_value. opt_return_value was not initialized if mode != ast_return. Fixes "Uninitialized pointer field" defect reported by Coverity. Signed-off-by: Vinson Lee <vlee@freedesktop.org> Reviewed-by: Chad Versace <chad.versace@linux.intel.com>
2013-07-14 00:57:22 -07:00
: opt_return_value(NULL)
Initial commit. lol
2010-02-22 13:19:34 -08:00
{
this->mode = ast_jump_modes(mode);
if (mode == ast_return)
opt_return_value = return_value;
}
void
ast_selection_statement::print(void) const
{
printf("if ( ");
condition->print();
printf(") ");
then_statement->print();
if (else_statement) {
printf("else ");
else_statement->print();
}
}
ast_selection_statement::ast_selection_statement(ast_expression *condition,
ast_node *then_statement,
ast_node *else_statement)
{
this->condition = condition;
this->then_statement = then_statement;
this->else_statement = else_statement;
}
glsl: Reference data structure ctors in grammar We now tie the grammar to the ctors of the ASTs they reference. This requires that we actually have definitions of the ctors. In addition, we also need to define "print" and "hir" methods for the AST classes. The Print methods are pretty simple to flesh out. However, at this stage of the development, we simply stub out the "hir" methods and flesh them out later. Also, since actual class instances get returned by the productions in the grammar, we also need to designate the type of the productions that reference those instances. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2011-11-07 15:11:04 -08:00
void
ast_switch_statement::print(void) const
{
printf("switch ( ");
test_expression->print();
printf(") ");
body->print();
}
ast_switch_statement::ast_switch_statement(ast_expression *test_expression,
ast_node *body)
{
this->test_expression = test_expression;
this->body = body;
}
void
ast_switch_body::print(void) const
{
printf("{\n");
if (stmts != NULL) {
stmts->print();
}
printf("}\n");
}
ast_switch_body::ast_switch_body(ast_case_statement_list *stmts)
{
this->stmts = stmts;
}
void ast_case_label::print(void) const
{
if (test_value != NULL) {
printf("case ");
test_value->print();
printf(": ");
} else {
printf("default: ");
}
}
ast_case_label::ast_case_label(ast_expression *test_value)
{
this->test_value = test_value;
}
void ast_case_label_list::print(void) const
{
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed(ast_node, ast, link, & this->labels) {
glsl: Reference data structure ctors in grammar We now tie the grammar to the ctors of the ASTs they reference. This requires that we actually have definitions of the ctors. In addition, we also need to define "print" and "hir" methods for the AST classes. The Print methods are pretty simple to flesh out. However, at this stage of the development, we simply stub out the "hir" methods and flesh them out later. Also, since actual class instances get returned by the productions in the grammar, we also need to designate the type of the productions that reference those instances. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2011-11-07 15:11:04 -08:00
ast->print();
}
printf("\n");
}
ast_case_label_list::ast_case_label_list(void)
{
}
void ast_case_statement::print(void) const
{
labels->print();
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed(ast_node, ast, link, & this->stmts) {
glsl: Reference data structure ctors in grammar We now tie the grammar to the ctors of the ASTs they reference. This requires that we actually have definitions of the ctors. In addition, we also need to define "print" and "hir" methods for the AST classes. The Print methods are pretty simple to flesh out. However, at this stage of the development, we simply stub out the "hir" methods and flesh them out later. Also, since actual class instances get returned by the productions in the grammar, we also need to designate the type of the productions that reference those instances. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2011-11-07 15:11:04 -08:00
ast->print();
printf("\n");
}
}
ast_case_statement::ast_case_statement(ast_case_label_list *labels)
{
this->labels = labels;
}
void ast_case_statement_list::print(void) const
{
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed(ast_node, ast, link, & this->cases) {
glsl: Reference data structure ctors in grammar We now tie the grammar to the ctors of the ASTs they reference. This requires that we actually have definitions of the ctors. In addition, we also need to define "print" and "hir" methods for the AST classes. The Print methods are pretty simple to flesh out. However, at this stage of the development, we simply stub out the "hir" methods and flesh them out later. Also, since actual class instances get returned by the productions in the grammar, we also need to designate the type of the productions that reference those instances. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2011-11-07 15:11:04 -08:00
ast->print();
}
}
ast_case_statement_list::ast_case_statement_list(void)
{
}
Initial commit. lol
2010-02-22 13:19:34 -08:00
void
ast_iteration_statement::print(void) const
{
switch (mode) {
case ast_for:
printf("for( ");
if (init_statement)
init_statement->print();
printf("; ");
if (condition)
condition->print();
printf("; ");
if (rest_expression)
rest_expression->print();
printf(") ");
body->print();
break;
case ast_while:
printf("while ( ");
if (condition)
condition->print();
printf(") ");
body->print();
break;
case ast_do_while:
printf("do ");
body->print();
printf("while ( ");
if (condition)
condition->print();
printf("); ");
break;
}
}
ast_iteration_statement::ast_iteration_statement(int mode,
ast_node *init,
ast_node *condition,
ast_expression *rest_expression,
ast_node *body)
{
this->mode = ast_iteration_modes(mode);
this->init_statement = init;
this->condition = condition;
this->rest_expression = rest_expression;
this->body = body;
}
void
ast_struct_specifier::print(void) const
{
printf("struct %s { ", name);
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed(ast_node, ast, link, &this->declarations) {
Convert ast_node use of simple_node to exec_list and exec_node
2010-05-10 11:17:53 -07:00
ast->print();
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
printf("} ");
}
glsl: Use (const char *) in AST nodes rather than plain (char *). Nothing actually relied on them being mutable, and there was at least one cast which discarded const qualifiers. The next patch would have introduced many more. Casting away const qualifiers should be avoided if at all possible. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2012-03-29 23:17:31 -07:00
ast_struct_specifier::ast_struct_specifier(const char *identifier,
glsl: Don't hide the type of struct_declaration_list. I've been trying to derive from this for UBO support, and the slightly obfuscated types were putting me over the edge. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2012-04-26 10:16:52 -07:00
ast_declarator_list *declarator_list)
Initial commit. lol
2010-02-22 13:19:34 -08:00
{
glsl: Properly handle nested structure types. Fixes piglit test CorrectFull.frag.
2010-09-18 11:11:09 +02:00
if (identifier == NULL) {
glsl: protect anonymous struct id with a mutex There may be two contexts compiling shaders at the same time, and we want the anonymous struct id to be globally unique. Signed-off-by: Chia-I Wu <olv@lunarg.com> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-08-20 14:40:28 +08:00
static mtx_t mutex = _MTX_INITIALIZER_NP;
glsl: Properly handle nested structure types. Fixes piglit test CorrectFull.frag.
2010-09-18 11:11:09 +02:00
static unsigned anon_count = 1;
glsl: protect anonymous struct id with a mutex There may be two contexts compiling shaders at the same time, and we want the anonymous struct id to be globally unique. Signed-off-by: Chia-I Wu <olv@lunarg.com> Reviewed-by: Brian Paul <brianp@vmware.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-08-20 14:40:28 +08:00
unsigned count;
mtx_lock(&mutex);
count = anon_count++;
mtx_unlock(&mutex);
identifier = ralloc_asprintf(this, "#anon_struct_%04x", count);
glsl: Properly handle nested structure types. Fixes piglit test CorrectFull.frag.
2010-09-18 11:11:09 +02:00
}
Initial commit. lol
2010-02-22 13:19:34 -08:00
name = identifier;
Convert ast_node use of simple_node to exec_list and exec_node
2010-05-10 11:17:53 -07:00
this->declarations.push_degenerate_list_at_head(&declarator_list->link);
glsl: Add an is_declaration field to ast_struct_specifier. Will be used in a later commit to differentiate between a structure type declaration and a variable declaration of a struct type. I.e., the difference between struct S { float x; }; (is_declaration = true) and S s; (is_declaration = false) Also note that is_declaration = true for struct S { float x; } s; Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-26 15:53:12 -07:00
is_declaration = true;
Initial commit. lol
2010-02-22 13:19:34 -08:00
}
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
glsl: add ast/parser support for subroutine parsing storage (v3.2) This is the guts of the GLSL parser and AST support for shader subroutines. The code creates a subroutine type in the parser, and uses that there to validate the identifiers. The parser also distinguishes between subroutine types/function prototypes /uniforms and subroutine defintions for functions. Then in the AST conversion it recreates the types, and stores the subroutine definition info or subroutine info into the ir_function along with a side lookup table in the parser state. It also converts subroutine calls into the enhanced ir_call. v2: move to handling method calls in function handling not in field selection. v3: merge Chris's previous parser patches in here, to make it clearer what's changed in one place. v3.1: add more documentation, drop unused include v3.2: drop is_subroutine_def Reviewed-by: Chris Forbes <chrisf@ijw.co.nz> Signed-off-by: Dave Airlie <airlied@redhat.com>
2015-06-01 10:55:47 +10:00
void ast_subroutine_list::print(void) const
{
foreach_list_typed (ast_node, ast, link, & this->declarations) {
if (&ast->link != this->declarations.get_head())
printf(", ");
ast->print();
}
}
glsl: Export the compiler's GS layout qualifiers to the gl_shader. Next step is to validate them at link time. v2 (Paul Berry <stereotype441@gmail.com>): Don't attempt to export the layout qualifiers in the event of a compile error, since some of them are set up by ast_to_hir(), and ast_to_hir() isn't guaranteed to have run in the event of a compile error. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> v3 (Paul Berry <stereotype441@gmail.com>): Use PRIM_UNKNOWN to represent "not set in this shader". Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-12 17:21:44 -07:00
static void
set_shader_inout_layout(struct gl_shader *shader,
struct _mesa_glsl_parse_state *state)
{
glsl: add tessellation shader parsing support (v2) v2: Fixed things that Ken suggested. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-03-20 22:44:43 +01:00
/* Should have been prevented by the parser. */
if (shader->Stage == MESA_SHADER_TESS_CTRL) {
assert(!state->in_qualifier->flags.i);
} else if (shader->Stage == MESA_SHADER_TESS_EVAL) {
assert(!state->out_qualifier->flags.i);
} else if (shader->Stage != MESA_SHADER_GEOMETRY) {
glsl: convert GS input primitive to use ast_type_qualifier We introduce a new merge_in_qualifier ast_type_qualifier which allows specialized handling of merging input layout qualifiers. By merging layout qualifiers into state->in_qualifier, we allow multiple input qualifiers. For example, the primitive type can be specified specified separately from the invocations count (ARB_gpu_shader5). state->gs_input_prim_type is moved into state->in_qualifier->prim_type state->gs_input_prim_type_specified is still processed separately so we can determine when the input primitive is specified. This is important since certain scenerios are not supported until after the primitive type has been specified in the shader code. v4: * Merge with compute shader input layout qualifiers Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
2014-02-02 17:49:15 -08:00
assert(!state->in_qualifier->flags.i);
glsl: Export the compiler's GS layout qualifiers to the gl_shader. Next step is to validate them at link time. v2 (Paul Berry <stereotype441@gmail.com>): Don't attempt to export the layout qualifiers in the event of a compile error, since some of them are set up by ast_to_hir(), and ast_to_hir() isn't guaranteed to have run in the event of a compile error. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> v3 (Paul Berry <stereotype441@gmail.com>): Use PRIM_UNKNOWN to represent "not set in this shader". Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-12 17:21:44 -07:00
assert(!state->out_qualifier->flags.i);
}
glsl/cs: Handle compute shader local_size_{x,y,z} declaration. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 09:09:31 -08:00
if (shader->Stage != MESA_SHADER_COMPUTE) {
/* Should have been prevented by the parser. */
assert(!state->cs_input_local_size_specified);
glsl: Export the compiler's GS layout qualifiers to the gl_shader. Next step is to validate them at link time. v2 (Paul Berry <stereotype441@gmail.com>): Don't attempt to export the layout qualifiers in the event of a compile error, since some of them are set up by ast_to_hir(), and ast_to_hir() isn't guaranteed to have run in the event of a compile error. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> v3 (Paul Berry <stereotype441@gmail.com>): Use PRIM_UNKNOWN to represent "not set in this shader". Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-12 17:21:44 -07:00
}
glsl: Link error if fs defines conflicting qualifiers for gl_FragCoord GLSL 1.50 spec says: "If gl_FragCoord is redeclared in any fragment shader in a program, it must be redeclared in all the fragment shaders in that program that have a static use gl_FragCoord. All redeclarations of gl_FragCoord in all fragment shaders in a single program must have the same set of qualifiers." This patch causes the shader link to fail if we have multiple fragment shaders with conflicting layout qualifiers for gl_FragCoord. V2: Restructure the code and add conditions to correctly handle the following case: fragment shader 1: layout(origin_upper_left) in vec4 gl_FragCoord; void main() { foo(); gl_FragColor = gl_FragData; } fragment shader 2: layout(pixel_center_integer) in vec4 gl_FragCoord; void foo() { } V3: Allow linking in the following case: fragment shader 1: void main() { foo(); gl_FragColor = gl_FragCoord; } fragment shader 2: in vec4 gl_FragCoord; void foo() { ... } Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com> Cc: <mesa-stable@lists.freedesktop.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-02-05 15:01:58 -08:00
if (shader->Stage != MESA_SHADER_FRAGMENT) {
/* Should have been prevented by the parser. */
assert(!state->fs_uses_gl_fragcoord);
assert(!state->fs_redeclares_gl_fragcoord);
assert(!state->fs_pixel_center_integer);
assert(!state->fs_origin_upper_left);
glsl: Keep track of the early_fragment_tests flag in gl_shader. And rename _mesa_glsl_parse_state::early_fragment_tests to fs_early_fragment_tests for consistency with other FS-specific flags in the same struct. Reviewed-by: Matt Turner <mattst88@gmail.com>
2015-01-28 17:42:37 +02:00
assert(!state->fs_early_fragment_tests);
glsl: Link error if fs defines conflicting qualifiers for gl_FragCoord GLSL 1.50 spec says: "If gl_FragCoord is redeclared in any fragment shader in a program, it must be redeclared in all the fragment shaders in that program that have a static use gl_FragCoord. All redeclarations of gl_FragCoord in all fragment shaders in a single program must have the same set of qualifiers." This patch causes the shader link to fail if we have multiple fragment shaders with conflicting layout qualifiers for gl_FragCoord. V2: Restructure the code and add conditions to correctly handle the following case: fragment shader 1: layout(origin_upper_left) in vec4 gl_FragCoord; void main() { foo(); gl_FragColor = gl_FragData; } fragment shader 2: layout(pixel_center_integer) in vec4 gl_FragCoord; void foo() { } V3: Allow linking in the following case: fragment shader 1: void main() { foo(); gl_FragColor = gl_FragCoord; } fragment shader 2: in vec4 gl_FragCoord; void foo() { ... } Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com> Cc: <mesa-stable@lists.freedesktop.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-02-05 15:01:58 -08:00
}
glsl/cs: Handle compute shader local_size_{x,y,z} declaration. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 09:09:31 -08:00
switch (shader->Stage) {
glsl: add tessellation shader parsing support (v2) v2: Fixed things that Ken suggested. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-03-20 22:44:43 +01:00
case MESA_SHADER_TESS_CTRL:
shader->TessCtrl.VerticesOut = 0;
glsl: add support for complie-time constant expressions This patch replaces the old interger constant qualifiers with either the new ast_layout_expression type if the qualifier requires merging or ast_expression if the qualifier can't have mulitple declarations or if all but the newest qualifier is simply ignored. We also update the process_qualifier_constant() helper to be similar to the one in the ast_layout_expression class, but in this case it will be used to process the ast_expression qualifiers. Global shader layout qualifier validation is moved out of the parser in this change as we now need to evaluate any constant expression before doing the validation. V2: Fix minimum value check for vertices (Emil) Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
2015-11-14 15:13:28 +11:00
if (state->tcs_output_vertices_specified) {
unsigned vertices;
if (state->out_qualifier->vertices->
process_qualifier_constant(state, "vertices", &vertices,
false)) {
YYLTYPE loc = state->out_qualifier->vertices->get_location();
if (vertices > state->Const.MaxPatchVertices) {
_mesa_glsl_error(&loc, state, "vertices (%d) exceeds "
"GL_MAX_PATCH_VERTICES", vertices);
}
shader->TessCtrl.VerticesOut = vertices;
}
}
glsl: add tessellation shader parsing support (v2) v2: Fixed things that Ken suggested. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-03-20 22:44:43 +01:00
break;
case MESA_SHADER_TESS_EVAL:
shader->TessEval.PrimitiveMode = PRIM_UNKNOWN;
if (state->in_qualifier->flags.q.prim_type)
shader->TessEval.PrimitiveMode = state->in_qualifier->prim_type;
shader->TessEval.Spacing = 0;
if (state->in_qualifier->flags.q.vertex_spacing)
shader->TessEval.Spacing = state->in_qualifier->vertex_spacing;
shader->TessEval.VertexOrder = 0;
if (state->in_qualifier->flags.q.ordering)
shader->TessEval.VertexOrder = state->in_qualifier->ordering;
shader->TessEval.PointMode = -1;
if (state->in_qualifier->flags.q.point_mode)
shader->TessEval.PointMode = state->in_qualifier->point_mode;
break;
glsl/cs: Handle compute shader local_size_{x,y,z} declaration. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 09:09:31 -08:00
case MESA_SHADER_GEOMETRY:
shader->Geom.VerticesOut = 0;
glsl: add support for complie-time constant expressions This patch replaces the old interger constant qualifiers with either the new ast_layout_expression type if the qualifier requires merging or ast_expression if the qualifier can't have mulitple declarations or if all but the newest qualifier is simply ignored. We also update the process_qualifier_constant() helper to be similar to the one in the ast_layout_expression class, but in this case it will be used to process the ast_expression qualifiers. Global shader layout qualifier validation is moved out of the parser in this change as we now need to evaluate any constant expression before doing the validation. V2: Fix minimum value check for vertices (Emil) Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
2015-11-14 15:13:28 +11:00
if (state->out_qualifier->flags.q.max_vertices) {
unsigned qual_max_vertices;
if (state->out_qualifier->max_vertices->
process_qualifier_constant(state, "max_vertices",
&qual_max_vertices, true)) {
shader->Geom.VerticesOut = qual_max_vertices;
}
}
glsl/cs: Handle compute shader local_size_{x,y,z} declaration. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 09:09:31 -08:00
if (state->gs_input_prim_type_specified) {
glsl: convert GS input primitive to use ast_type_qualifier We introduce a new merge_in_qualifier ast_type_qualifier which allows specialized handling of merging input layout qualifiers. By merging layout qualifiers into state->in_qualifier, we allow multiple input qualifiers. For example, the primitive type can be specified specified separately from the invocations count (ARB_gpu_shader5). state->gs_input_prim_type is moved into state->in_qualifier->prim_type state->gs_input_prim_type_specified is still processed separately so we can determine when the input primitive is specified. This is important since certain scenerios are not supported until after the primitive type has been specified in the shader code. v4: * Merge with compute shader input layout qualifiers Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
2014-02-02 17:49:15 -08:00
shader->Geom.InputType = state->in_qualifier->prim_type;
glsl/cs: Handle compute shader local_size_{x,y,z} declaration. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 09:09:31 -08:00
} else {
shader->Geom.InputType = PRIM_UNKNOWN;
}
if (state->out_qualifier->flags.q.prim_type) {
shader->Geom.OutputType = state->out_qualifier->prim_type;
} else {
shader->Geom.OutputType = PRIM_UNKNOWN;
}
glsl/linker: produce gl_shader_program Geom.Invocations Grab the parsed invocation count, check for consistency during linking, and finally save the result in gl_shader_program Geom.Invocations. Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Paul Berry <stereotype441@gmail.com> Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
2014-01-25 02:17:21 -08:00
shader->Geom.Invocations = 0;
glsl: add support for complie-time constant expressions This patch replaces the old interger constant qualifiers with either the new ast_layout_expression type if the qualifier requires merging or ast_expression if the qualifier can't have mulitple declarations or if all but the newest qualifier is simply ignored. We also update the process_qualifier_constant() helper to be similar to the one in the ast_layout_expression class, but in this case it will be used to process the ast_expression qualifiers. Global shader layout qualifier validation is moved out of the parser in this change as we now need to evaluate any constant expression before doing the validation. V2: Fix minimum value check for vertices (Emil) Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
2015-11-14 15:13:28 +11:00
if (state->in_qualifier->flags.q.invocations) {
unsigned invocations;
if (state->in_qualifier->invocations->
process_qualifier_constant(state, "invocations",
&invocations, false)) {
YYLTYPE loc = state->in_qualifier->invocations->get_location();
if (invocations > MAX_GEOMETRY_SHADER_INVOCATIONS) {
_mesa_glsl_error(&loc, state,
"invocations (%d) exceeds "
"GL_MAX_GEOMETRY_SHADER_INVOCATIONS",
invocations);
}
shader->Geom.Invocations = invocations;
}
}
glsl/cs: Handle compute shader local_size_{x,y,z} declaration. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 09:09:31 -08:00
break;
case MESA_SHADER_COMPUTE:
if (state->cs_input_local_size_specified) {
for (int i = 0; i < 3; i++)
shader->Comp.LocalSize[i] = state->cs_input_local_size[i];
} else {
for (int i = 0; i < 3; i++)
shader->Comp.LocalSize[i] = 0;
}
break;
glsl: Link error if fs defines conflicting qualifiers for gl_FragCoord GLSL 1.50 spec says: "If gl_FragCoord is redeclared in any fragment shader in a program, it must be redeclared in all the fragment shaders in that program that have a static use gl_FragCoord. All redeclarations of gl_FragCoord in all fragment shaders in a single program must have the same set of qualifiers." This patch causes the shader link to fail if we have multiple fragment shaders with conflicting layout qualifiers for gl_FragCoord. V2: Restructure the code and add conditions to correctly handle the following case: fragment shader 1: layout(origin_upper_left) in vec4 gl_FragCoord; void main() { foo(); gl_FragColor = gl_FragData; } fragment shader 2: layout(pixel_center_integer) in vec4 gl_FragCoord; void foo() { } V3: Allow linking in the following case: fragment shader 1: void main() { foo(); gl_FragColor = gl_FragCoord; } fragment shader 2: in vec4 gl_FragCoord; void foo() { ... } Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com> Cc: <mesa-stable@lists.freedesktop.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-02-05 15:01:58 -08:00
case MESA_SHADER_FRAGMENT:
shader->redeclares_gl_fragcoord = state->fs_redeclares_gl_fragcoord;
shader->uses_gl_fragcoord = state->fs_uses_gl_fragcoord;
shader->pixel_center_integer = state->fs_pixel_center_integer;
shader->origin_upper_left = state->fs_origin_upper_left;
glsl: Apply the link error conditions to GL_ARB_fragment_coord_conventions Link error conditions added in previous patch are equally applicable to GL_ARB_fragment_coord_conventions implementation. Extension's spec says: "If gl_FragCoord is redeclared in any fragment shader in a program, it must be redeclared in all the fragment shaders in that program that have a static use of gl_FragCoord. All redeclarations of gl_FragCoord in all fragment shaders in a single program must have the same set of qualifiers." Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com> Cc: <mesa-stable@lists.freedesktop.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-02-10 14:12:40 -08:00
shader->ARB_fragment_coord_conventions_enable =
state->ARB_fragment_coord_conventions_enable;
glsl: Keep track of the early_fragment_tests flag in gl_shader. And rename _mesa_glsl_parse_state::early_fragment_tests to fs_early_fragment_tests for consistency with other FS-specific flags in the same struct. Reviewed-by: Matt Turner <mattst88@gmail.com>
2015-01-28 17:42:37 +02:00
shader->EarlyFragmentTests = state->fs_early_fragment_tests;
glsl: Link error if fs defines conflicting qualifiers for gl_FragCoord GLSL 1.50 spec says: "If gl_FragCoord is redeclared in any fragment shader in a program, it must be redeclared in all the fragment shaders in that program that have a static use gl_FragCoord. All redeclarations of gl_FragCoord in all fragment shaders in a single program must have the same set of qualifiers." This patch causes the shader link to fail if we have multiple fragment shaders with conflicting layout qualifiers for gl_FragCoord. V2: Restructure the code and add conditions to correctly handle the following case: fragment shader 1: layout(origin_upper_left) in vec4 gl_FragCoord; void main() { foo(); gl_FragColor = gl_FragData; } fragment shader 2: layout(pixel_center_integer) in vec4 gl_FragCoord; void foo() { } V3: Allow linking in the following case: fragment shader 1: void main() { foo(); gl_FragColor = gl_FragCoord; } fragment shader 2: in vec4 gl_FragCoord; void foo() { ... } Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com> Cc: <mesa-stable@lists.freedesktop.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-02-05 15:01:58 -08:00
break;
glsl/cs: Handle compute shader local_size_{x,y,z} declaration. Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
2014-01-06 09:09:31 -08:00
default:
/* Nothing to do. */
break;
glsl: Export the compiler's GS layout qualifiers to the gl_shader. Next step is to validate them at link time. v2 (Paul Berry <stereotype441@gmail.com>): Don't attempt to export the layout qualifiers in the event of a compile error, since some of them are set up by ast_to_hir(), and ast_to_hir() isn't guaranteed to have run in the event of a compile error. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> v3 (Paul Berry <stereotype441@gmail.com>): Use PRIM_UNKNOWN to represent "not set in this shader". Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-06-12 17:21:44 -07:00
}
}
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
extern "C" {
void
_mesa_glsl_compile_shader(struct gl_context *ctx, struct gl_shader *shader,
bool dump_ast, bool dump_hir)
{
struct _mesa_glsl_parse_state *state =
glsl: Change _mesa_glsl_parse_state ctor to use gl_shader_stage enum. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> v2: Also rename "target" param to "stage". Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-07 09:46:10 -08:00
new(shader) _mesa_glsl_parse_state(ctx, shader->Stage, shader);
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
const char *source = shader->Source;
glsl: Don't allocate a name for ir_var_temporary variables Valgrind massif results for a trimmed apitrace of dota2: n time(i) total(B) useful-heap(B) extra-heap(B) stacks(B) Before (32-bit): 74 40,578,719,715 67,762,208 62,263,404 5,498,804 0 After (32-bit): 52 40,565,579,466 66,359,800 61,187,818 5,171,982 0 Before (64-bit): 74 37,129,541,061 95,195,160 87,369,671 7,825,489 0 After (64-bit): 76 37,134,691,404 93,271,352 85,900,223 7,371,129 0 A real savings of 1.0MiB on 32-bit and 1.4MiB on 64-bit. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-07-08 19:03:52 -07:00
if (ctx->Const.GenerateTemporaryNames)
glsl: Initialize static temporaries_allocate_names once per process. Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
2014-11-21 18:04:21 -08:00
(void) p_atomic_cmpxchg(&ir_variable::temporaries_allocate_names,
false, true);
glsl: Don't allocate a name for ir_var_temporary variables Valgrind massif results for a trimmed apitrace of dota2: n time(i) total(B) useful-heap(B) extra-heap(B) stacks(B) Before (32-bit): 74 40,578,719,715 67,762,208 62,263,404 5,498,804 0 After (32-bit): 52 40,565,579,466 66,359,800 61,187,818 5,171,982 0 Before (64-bit): 74 37,129,541,061 95,195,160 87,369,671 7,825,489 0 After (64-bit): 76 37,134,691,404 93,271,352 85,900,223 7,371,129 0 A real savings of 1.0MiB on 32-bit and 1.4MiB on 64-bit. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-07-08 19:03:52 -07:00
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
state->error = glcpp_preprocess(state, &source, &state->info_log,
&ctx->Extensions, ctx);
if (!state->error) {
_mesa_glsl_lexer_ctor(state, source);
_mesa_glsl_parse(state);
_mesa_glsl_lexer_dtor(state);
}
if (dump_ast) {
glsl: Use foreach_list_typed when possible. Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-06-24 22:02:24 -07:00
foreach_list_typed(ast_node, ast, link, &state->translation_unit) {
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
ast->print();
}
printf("\n\n");
}
ralloc_free(shader->ir);
shader->ir = new(shader) exec_list;
if (!state->error && !state->translation_unit.is_empty())
_mesa_ast_to_hir(shader->ir, state);
if (!state->error) {
validate_ir_tree(shader->ir);
/* Print out the unoptimized IR. */
if (dump_hir) {
glsl: Add a file argument to the IR printer. While we want to be able to print to stdout for glsl_compiler, for debugging drivers we want to be able to dump to stderr because that's where other driver debug (like LIBGL_DEBUG) tends to go, and because some apps actually close stdout to shut up their own messages (such as the X Server, or NWN). Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-02-20 18:00:23 -08:00
_mesa_print_ir(stdout, shader->ir, state);
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
}
}
if (!state->error && !shader->ir->is_empty()) {
struct gl_shader_compiler_options *options =
mesa: move ShaderCompilerOptions into gl_constants Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
2014-08-03 04:31:56 +02:00
&ctx->Const.ShaderCompilerOptions[shader->Stage];
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
glsl/ir: add subroutine lowering pass (v2.3) This lowers the enhanced ir_call using the lookaside table of subroutines into an if ladder. This initially was done at the AST level but it caused some ordering issues so a separate pass was required. v2: clone return value derefs. v2.1: update for subroutine->int convert. v2.2: add a clone for the array index Reviewed-by: Chris Forbes <chrisf@ijw.co.nz> Signed-off-by: Dave Airlie <airlied@redhat.com>
2015-04-23 13:34:14 +10:00
lower_subroutine(shader->ir, state);
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
/* Do some optimization at compile time to reduce shader IR size
* and reduce later work if the same shader is linked multiple times
*/
glsl: Drop do_common_optimization's max_unroll_iterations parameter. Now that we pass in gl_shader_compiler_options, it makes sense to just use options->MaxUnrollIterations, rather than passing a separate parameter. Half of the invocations already passed options->MaxUnrollIterations, while the other half passed in a hardcoded value of 32. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Eric Anholt <eric@anholt.net>
2014-04-08 15:43:46 -07:00
while (do_common_optimization(shader->ir, false, false, options,
glsl: Pass ctx->Const.NativeIntegers to do_common_optimization(). The next few patches will introduce an optimization that only works when integers are not represented as floating point values. v2: Re-word-wrap a line, as requested by Ian Romanick. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-04-06 23:25:00 -07:00
ctx->Const.NativeIntegers))
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
;
validate_ir_tree(shader->ir);
glsl: Eliminate unused built-in variables after compilation After compilation (and before linking) we can eliminate quite a few built-in variables. Basically, any uniform or constant (e.g., gl_MaxVertexTextureImageUnits) that isn't used (with one exception) can be eliminated. System values, vertex shader inputs (with one exception), and fragment shader outputs that are not used and not re-declared in the shader text can also be removed. gl_ModelViewProjectMatrix and gl_Vertex are used by the built-in function ftransform. There are some complications with eliminating these variables (see the comment in the patch), so they are not eliminated. Valgrind massif results for a trimmed apitrace of dota2: n time(i) total(B) useful-heap(B) extra-heap(B) stacks(B) Before (32-bit): 46 40,661,487,174 75,116,800 68,854,065 6,262,735 0 After (32-bit): 50 40,564,927,443 69,185,408 63,683,871 5,501,537 0 Before (64-bit): 64 37,200,329,700 104,872,672 96,514,546 8,358,126 0 After (64-bit): 59 36,822,048,449 96,526,888 89,113,000 7,413,888 0 A real savings of 4.9MiB on 32-bit and 7.0MiB on 64-bit. v2: Don't remove any built-in with Transpose in the name. v3: Fix comment typo noticed by Anuj. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Suggested-by: Eric Anholt <eric@anholt.net> Reviewed-by: Matt Turner <mattst88@gmail.com> Acked-by: Anuj Phogat <anuj.phogat@gmail.com> Cc: Eric Anholt <eric@anholt.net>
2014-05-28 17:09:45 -07:00
enum ir_variable_mode other;
switch (shader->Stage) {
case MESA_SHADER_VERTEX:
other = ir_var_shader_in;
break;
case MESA_SHADER_FRAGMENT:
other = ir_var_shader_out;
break;
default:
/* Something invalid to ensure optimize_dead_builtin_uniforms
* doesn't remove anything other than uniforms or constants.
*/
other = ir_var_mode_count;
break;
}
optimize_dead_builtin_variables(shader->ir, other);
validate_ir_tree(shader->ir);
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
}
if (shader->InfoLog)
ralloc_free(shader->InfoLog);
glsl: call set_shader_inout_layout() earlier This will allow us to add error checking to this function in a later patch, if we don't move it the error messages will go missing. Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
2015-11-14 14:05:30 +11:00
if (!state->error)
set_shader_inout_layout(shader, state);
glsl: Rebuild the symbol table without unreachable symbols Previously we had to keep unreachable global symbols in the symbol table because the symbol table is used during linking. Having the symbol table retain pointers to freed memory... what could possibly go wrong? At the same time, this meant that we kept live references to tons of memory that was no longer needed. New strategy: destroy the old symbol table, and make a new one from the reachable symbols. Valgrind massif results for a trimmed apitrace of dota2: n time(i) total(B) useful-heap(B) extra-heap(B) stacks(B) Before (32-bit): 59 40,642,425,451 76,337,968 69,720,886 6,617,082 0 After (32-bit): 46 40,661,487,174 75,116,800 68,854,065 6,262,735 0 Before (64-bit): 79 37,179,441,771 106,986,512 98,112,095 8,874,417 0 After (64-bit): 64 37,200,329,700 104,872,672 96,514,546 8,358,126 0 A real savings of 846KiB on 32-bit and 1.5MiB on 64-bit. v2: (by Kenneth Graunke) Just add the ir_function from the IR stream, rather than looking it up in the symbol table; they're now identical. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-07-14 15:48:34 -07:00
shader->symbols = new(shader->ir) glsl_symbol_table;
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
shader->CompileStatus = !state->error;
shader->InfoLog = state->info_log;
shader->Version = state->language_version;
shader->IsES = state->es_shader;
glsl: Simplify the built-in function linking code. Previously, we stored an array of up to 16 additional shaders to link, as well as a count of how many each shader actually needed. Since the built-in functions rewrite, all the built-ins are stored in a single shader. So all we need is a boolean indicating whether a shader needs to link against built-ins or not. During linking, we can avoid creating the temporary array if none of the shaders being linked need built-ins. Otherwise, it's simply a copy of the array that has one additional element. This is much simpler. This patch saves approximately 128 bytes of memory per gl_shader object. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-11-23 12:11:34 -08:00
shader->uses_builtin_functions = state->uses_builtin_functions;
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
/* Retain any live IR, but trash the rest. */
reparent_ir(shader->ir, shader->ir);
glsl: Rebuild the symbol table without unreachable symbols Previously we had to keep unreachable global symbols in the symbol table because the symbol table is used during linking. Having the symbol table retain pointers to freed memory... what could possibly go wrong? At the same time, this meant that we kept live references to tons of memory that was no longer needed. New strategy: destroy the old symbol table, and make a new one from the reachable symbols. Valgrind massif results for a trimmed apitrace of dota2: n time(i) total(B) useful-heap(B) extra-heap(B) stacks(B) Before (32-bit): 59 40,642,425,451 76,337,968 69,720,886 6,617,082 0 After (32-bit): 46 40,661,487,174 75,116,800 68,854,065 6,262,735 0 Before (64-bit): 79 37,179,441,771 106,986,512 98,112,095 8,874,417 0 After (64-bit): 64 37,200,329,700 104,872,672 96,514,546 8,358,126 0 A real savings of 846KiB on 32-bit and 1.5MiB on 64-bit. v2: (by Kenneth Graunke) Just add the ir_function from the IR stream, rather than looking it up in the symbol table; they're now identical. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-07-14 15:48:34 -07:00
/* Destroy the symbol table. Create a new symbol table that contains only
* the variables and functions that still exist in the IR. The symbol
* table will be used later during linking.
*
* There must NOT be any freed objects still referenced by the symbol
* table. That could cause the linker to dereference freed memory.
*
* We don't have to worry about types or interface-types here because those
* are fly-weights that are looked up by glsl_type.
*/
foreach_in_list (ir_instruction, ir, shader->ir) {
switch (ir->ir_type) {
case ir_type_function:
shader->symbols->add_function((ir_function *) ir);
break;
glsl: Never put ir_var_temporary variables in the symbol table Later patches will give every ir_var_temporary the same name in release builds. Adding a bunch of variables named "compiler_temp" to the symbol table can only cause problems. No change Valgrind massif results for a trimmed apitrace of dota2. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-07-08 18:53:09 -07:00
case ir_type_variable: {
ir_variable *const var = (ir_variable *) ir;
if (var->data.mode != ir_var_temporary)
shader->symbols->add_variable(var);
glsl: Rebuild the symbol table without unreachable symbols Previously we had to keep unreachable global symbols in the symbol table because the symbol table is used during linking. Having the symbol table retain pointers to freed memory... what could possibly go wrong? At the same time, this meant that we kept live references to tons of memory that was no longer needed. New strategy: destroy the old symbol table, and make a new one from the reachable symbols. Valgrind massif results for a trimmed apitrace of dota2: n time(i) total(B) useful-heap(B) extra-heap(B) stacks(B) Before (32-bit): 59 40,642,425,451 76,337,968 69,720,886 6,617,082 0 After (32-bit): 46 40,661,487,174 75,116,800 68,854,065 6,262,735 0 Before (64-bit): 79 37,179,441,771 106,986,512 98,112,095 8,874,417 0 After (64-bit): 64 37,200,329,700 104,872,672 96,514,546 8,358,126 0 A real savings of 846KiB on 32-bit and 1.5MiB on 64-bit. v2: (by Kenneth Graunke) Just add the ir_function from the IR stream, rather than looking it up in the symbol table; they're now identical. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-07-14 15:48:34 -07:00
break;
glsl: Never put ir_var_temporary variables in the symbol table Later patches will give every ir_var_temporary the same name in release builds. Adding a bunch of variables named "compiler_temp" to the symbol table can only cause problems. No change Valgrind massif results for a trimmed apitrace of dota2. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-07-08 18:53:09 -07:00
}
glsl: Rebuild the symbol table without unreachable symbols Previously we had to keep unreachable global symbols in the symbol table because the symbol table is used during linking. Having the symbol table retain pointers to freed memory... what could possibly go wrong? At the same time, this meant that we kept live references to tons of memory that was no longer needed. New strategy: destroy the old symbol table, and make a new one from the reachable symbols. Valgrind massif results for a trimmed apitrace of dota2: n time(i) total(B) useful-heap(B) extra-heap(B) stacks(B) Before (32-bit): 59 40,642,425,451 76,337,968 69,720,886 6,617,082 0 After (32-bit): 46 40,661,487,174 75,116,800 68,854,065 6,262,735 0 Before (64-bit): 79 37,179,441,771 106,986,512 98,112,095 8,874,417 0 After (64-bit): 64 37,200,329,700 104,872,672 96,514,546 8,358,126 0 A real savings of 846KiB on 32-bit and 1.5MiB on 64-bit. v2: (by Kenneth Graunke) Just add the ir_function from the IR stream, rather than looking it up in the symbol table; they're now identical. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-07-14 15:48:34 -07:00
default:
break;
}
}
glsl/cs: Initialize gl_GlobalInvocationID in main() We initialize gl_GlobalInvocationID based on the extension spec formula: gl_GlobalInvocationID = gl_WorkGroupID * gl_WorkGroupSize + gl_LocalInvocationID https://www.opengl.org/registry/specs/ARB/compute_shader.txt Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Cc: Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
2015-08-17 14:35:44 -07:00
_mesa_glsl_initialize_derived_variables(shader);
glsl: Rebuild the symbol table without unreachable symbols Previously we had to keep unreachable global symbols in the symbol table because the symbol table is used during linking. Having the symbol table retain pointers to freed memory... what could possibly go wrong? At the same time, this meant that we kept live references to tons of memory that was no longer needed. New strategy: destroy the old symbol table, and make a new one from the reachable symbols. Valgrind massif results for a trimmed apitrace of dota2: n time(i) total(B) useful-heap(B) extra-heap(B) stacks(B) Before (32-bit): 59 40,642,425,451 76,337,968 69,720,886 6,617,082 0 After (32-bit): 46 40,661,487,174 75,116,800 68,854,065 6,262,735 0 Before (64-bit): 79 37,179,441,771 106,986,512 98,112,095 8,874,417 0 After (64-bit): 64 37,200,329,700 104,872,672 96,514,546 8,358,126 0 A real savings of 846KiB on 32-bit and 1.5MiB on 64-bit. v2: (by Kenneth Graunke) Just add the ir_function from the IR stream, rather than looking it up in the symbol table; they're now identical. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2014-07-14 15:48:34 -07:00
delete state->symbols;
mesa: Move the common _mesa_glsl_compile_shader() code to glsl/. This code had no relation to ir_to_mesa.cpp, since it was also used by intel and state_tracker, and most of it was duplicated with the standalone compiler (which has periodically drifted from the Mesa copy). v2: Split from the ir_to_mesa to shaderapi.c changes. Acked-by: Paul Berry <stereotype441@gmail.com> (v1) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2013-06-12 15:49:43 -07:00
ralloc_free(state);
}
} /* extern "C" */
glsl: Add uniform_locations_assigned parameter to do_dead_code opt pass Setting this flag prevents declarations of uniforms from being removed from the IR. Since the IR is directly used by several API functions that query uniforms in shaders, uniform declarations cannot be removed after the locations have been set. However, it should still be safe to reorder the declarations (this is not tested). Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=41980 Tested-by: Brian Paul <brianp@vmware.com> Reviewed-by: Bryan Cain <bryancain3@gmail.com> Cc: Vinson Lee <vlee@vmware.com> Cc: José Fonseca <jfonseca@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
2011-10-21 11:17:39 -07:00
/**
* Do the set of common optimizations passes
*
* \param ir List of instructions to be optimized
* \param linked Is the shader linked? This enables
* optimizations passes that remove code at
* global scope and could cause linking to
* fail.
* \param uniform_locations_assigned Have locations already been assigned for
* uniforms? This prevents the declarations
* of unused uniforms from being removed.
* The setting of this flag only matters if
* \c linked is \c true.
* \param max_unroll_iterations Maximum number of loop iterations to be
glsl: fix incorrect comment about do_common_optimization
2013-01-31 23:29:53 +01:00
* unrolled. Setting to 0 disables loop
* unrolling.
glsl: Pass struct shader_compiler_options into do_common_optimization. do_common_optimization may need to make choices about whether to emit certain kinds of instructions. gl_context::ShaderCompilerOptions contains exactly that information, so it makes sense to pass it in. Rather than passing the whole array, pass the structure for the stage that's currently being worked on. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Eric Anholt <eric@anholt.net>
2013-04-17 17:30:22 -07:00
* \param options The driver's preferred shader options.
glsl: Add uniform_locations_assigned parameter to do_dead_code opt pass Setting this flag prevents declarations of uniforms from being removed from the IR. Since the IR is directly used by several API functions that query uniforms in shaders, uniform declarations cannot be removed after the locations have been set. However, it should still be safe to reorder the declarations (this is not tested). Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=41980 Tested-by: Brian Paul <brianp@vmware.com> Reviewed-by: Bryan Cain <bryancain3@gmail.com> Cc: Vinson Lee <vlee@vmware.com> Cc: José Fonseca <jfonseca@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
2011-10-21 11:17:39 -07:00
*/
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
bool
glsl: Add uniform_locations_assigned parameter to do_dead_code opt pass Setting this flag prevents declarations of uniforms from being removed from the IR. Since the IR is directly used by several API functions that query uniforms in shaders, uniform declarations cannot be removed after the locations have been set. However, it should still be safe to reorder the declarations (this is not tested). Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=41980 Tested-by: Brian Paul <brianp@vmware.com> Reviewed-by: Bryan Cain <bryancain3@gmail.com> Cc: Vinson Lee <vlee@vmware.com> Cc: José Fonseca <jfonseca@vmware.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
2011-10-21 11:17:39 -07:00
do_common_optimization(exec_list *ir, bool linked,
bool uniform_locations_assigned,
glsl: Pass ctx->Const.NativeIntegers to do_common_optimization(). The next few patches will introduce an optimization that only works when integers are not represented as floating point values. v2: Re-word-wrap a line, as requested by Ian Romanick. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
2014-04-06 23:25:00 -07:00
const struct gl_shader_compiler_options *options,
bool native_integers)
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
{
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
const bool debug = false;
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
GLboolean progress = GL_FALSE;
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
#define OPT(PASS, ...) do { \
if (debug) { \
fprintf(stderr, "START GLSL optimization %s\n", #PASS); \
const bool opt_progress = PASS(__VA_ARGS__); \
progress = opt_progress || progress; \
if (opt_progress) \
_mesa_print_ir(stderr, ir, NULL); \
fprintf(stderr, "GLSL optimization %s: %s progress\n", \
#PASS, opt_progress ? "made" : "no"); \
} else { \
progress = PASS(__VA_ARGS__) || progress; \
} \
} while (false)
OPT(lower_instructions, ir, SUB_TO_ADD_NEG);
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
if (linked) {
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
OPT(do_function_inlining, ir);
OPT(do_dead_functions, ir);
OPT(do_structure_splitting, ir);
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
}
glsl: Add a pass to propagate the "invariant" and "precise" qualifiers Reviewed-by: Francisco Jerez <currojerez@riseup.net>
2016-03-17 13:58:40 -07:00
propagate_invariance(ir);
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
OPT(do_if_simplification, ir);
OPT(opt_flatten_nested_if_blocks, ir);
OPT(opt_conditional_discard, ir);
OPT(do_copy_propagation, ir);
OPT(do_copy_propagation_elements, ir);
glsl: Add a pass to flip matrix/vector multiplies to use dot products. This pass flips (matrix * vector) operations to (vector * matrixTranspose) for certain built-in matrices (currently gl_ModelViewProjectionMatrix and gl_TextureMatrix). This is equivalent, but results in dot products rather than multiplies and adds. On some hardware, this is more efficient. This pass is conditionalized on ctx->mvp_with_dp4, the flag drivers set to indicate they prefer dot products. Improves performance in Lightsmark by 1.01131% +/- 0.162069% (n = 10) on a Haswell GT2 system. Passes Piglit on Ivybridge. v2: Use struct gl_shader_compiler_options instead of plumbing through another boolean flag for this purpose. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Eric Anholt <eric@anholt.net>
2013-04-17 17:30:25 -07:00
mesa: rename PreferDP4 to OptimizeForAOS. This flag was really just a proxy for determining whether the backend was vector (AOS) or scalar (SOA). It will be used to apply a future optimization only for vector backends. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2014-01-03 14:48:53 -08:00
if (options->OptimizeForAOS && !linked)
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
OPT(opt_flip_matrices, ir);
glsl: Add a pass to flip matrix/vector multiplies to use dot products. This pass flips (matrix * vector) operations to (vector * matrixTranspose) for certain built-in matrices (currently gl_ModelViewProjectionMatrix and gl_TextureMatrix). This is equivalent, but results in dot products rather than multiplies and adds. On some hardware, this is more efficient. This pass is conditionalized on ctx->mvp_with_dp4, the flag drivers set to indicate they prefer dot products. Improves performance in Lightsmark by 1.01131% +/- 0.162069% (n = 10) on a Haswell GT2 system. Passes Piglit on Ivybridge. v2: Use struct gl_shader_compiler_options instead of plumbing through another boolean flag for this purpose. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Eric Anholt <eric@anholt.net>
2013-04-17 17:30:25 -07:00
glsl: Vectorize multiple scalar assignments Reduces vertex shader instruction counts in DOTA2 by 6.42%, L4D2 by 4.61%, and CS:GO by 5.71%. total instructions in shared programs: 1500153 -> 1498191 (-0.13%) instructions in affected programs: 59919 -> 57957 (-3.27%) Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-12-21 11:28:05 -08:00
if (linked && options->OptimizeForAOS) {
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
OPT(do_vectorize, ir);
glsl: Vectorize multiple scalar assignments Reduces vertex shader instruction counts in DOTA2 by 6.42%, L4D2 by 4.61%, and CS:GO by 5.71%. total instructions in shared programs: 1500153 -> 1498191 (-0.13%) instructions in affected programs: 59919 -> 57957 (-3.27%) Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
2013-12-21 11:28:05 -08:00
}
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
if (linked)
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
OPT(do_dead_code, ir, uniform_locations_assigned);
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
else
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
OPT(do_dead_code_unlinked, ir);
OPT(do_dead_code_local, ir);
OPT(do_tree_grafting, ir);
OPT(do_constant_propagation, ir);
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
if (linked)
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
OPT(do_constant_variable, ir);
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
else
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
OPT(do_constant_variable_unlinked, ir);
OPT(do_constant_folding, ir);
OPT(do_minmax_prune, ir);
OPT(do_rebalance_tree, ir);
OPT(do_algebraic, ir, native_integers, options);
OPT(do_lower_jumps, ir);
OPT(do_vec_index_to_swizzle, ir);
OPT(lower_vector_insert, ir, false);
OPT(do_swizzle_swizzle, ir);
OPT(do_noop_swizzle, ir);
OPT(optimize_split_arrays, ir, linked);
OPT(optimize_redundant_jumps, ir);
glsl2: Add pass to remove redundant jumps
2010-09-13 14:25:26 -07:00
glsl2: Perform initial bits of loop analysis during compilation
2010-08-26 16:45:22 -07:00
loop_state *ls = analyze_loop_variables(ir);
glsl: Skip the rest of loop unrolling if no loops were found. Shaves 1.6% (+/- 1.0%) off of ff_fragment_shader glean texCombine time (n=5).
2011-01-17 22:07:55 -08:00
if (ls->loop_found) {
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
OPT(set_loop_controls, ir, ls);
OPT(unroll_loops, ir, ls, options);
glsl: Skip the rest of loop unrolling if no loops were found. Shaves 1.6% (+/- 1.0%) off of ff_fragment_shader glean texCombine time (n=5).
2011-01-17 22:07:55 -08:00
}
glsl2: Perform initial bits of loop analysis during compilation
2010-08-26 16:45:22 -07:00
delete ls;
glsl: Enable debug prints for do_common_optimization Signed-off-by: Jordan Justen <jordan.l.justen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2016-01-19 23:23:23 -08:00
#undef OPT
glsl2: Move the common optimization passes to a helper function. These are passes that we expect all codegen to be happy with. The other lowering passes for Mesa IR are moved to the Mesa IR generator.
2010-08-10 13:06:49 -07:00
return progress;
}
glsl2: Free the shader compiler at dri screen destruction. Hooray, we can valgrind again without adding suppressions. This also adds an interface for use by an implementation of glReleaseShaderCompiler().
2010-08-18 16:56:39 -07:00
extern "C" {
/**
* To be called at GL teardown time, this frees compiler datastructures.
*
* After calling this, any previously compiled shaders and shader
* programs would be invalid. So this should happen at approximately
* program exit.
*/
void
_mesa_destroy_shader_compiler(void)
{
_mesa_destroy_shader_compiler_caches();
_mesa_glsl_release_types();
}
/**
* Releases compiler caches to trade off performance for memory.
*
* Intended to be used with glReleaseShaderCompiler().
*/
void
_mesa_destroy_shader_compiler_caches(void)
{
glsl: Switch to the new built-in function module. All built-ins are now handled by the new code; the old system is dead. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Matt Turner <mattst88@gmail.com> Reviewed-by: Paul Berry <stereotype441@gmail.com>
2013-09-01 20:48:45 -07:00
_mesa_glsl_release_builtin_functions();
glsl2: Free the shader compiler at dri screen destruction. Hooray, we can valgrind again without adding suppressions. This also adds an interface for use by an implementation of glReleaseShaderCompiler().
2010-08-18 16:56:39 -07:00
}
}
Reference in a new issue Copy permalink