summaryrefslogtreecommitdiffstats
path: root/src/glsl
Commit message (Collapse)AuthorAgeFilesLines
* glsl2: Port equal() and notEqual() to ir_unop_all_equal and ir_unop_any_nequalIan Romanick2010-09-133-240/+48
|
* glsl: introduce ir_binop_all_equal and ir_binop_any_equal, allow vector cmpsLuca Barbieri2010-09-137-22/+67
| | | | | | | | | | | | | | | | | Currently GLSL IR forbids any vector comparisons, and defines "ir_binop_equal" and "ir_binop_nequal" to compare all elements and give a single bool. This is highly unintuitive and prevents generation of optimal Mesa IR. Hence, first rename "ir_binop_equal" to "ir_binop_all_equal" and "ir_binop_nequal" to "ir_binop_any_nequal". Second, readd "ir_binop_equal" and "ir_binop_nequal" with the same semantics as less, lequal, etc. Third, allow all comparisons to acts on vectors. Signed-off-by: Ian Romanick <[email protected]>
* loop_unroll: unroll loops with (lowered) breaksLuca Barbieri2010-09-131-4/+89
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the loop ends with an if with one break or in a single break unroll it. Loops that end with a continue will have that continue removed by the redundant jump optimizer. Likewise loops that end with an if-statement with a break at the end of both branches will have the break pulled out after the if-statement. Loops of the form for (...) { do_something1(); if (cond) { do_something2(); break; } else { do_something3(); } } will be unrolled as do_something1(); if (cond) { do_something2(); } else { do_something3(); do_something1(); if (cond) { do_something2(); } else { do_something3(); /* Repeat inserting iterations here.*/ } } ir_lower_jumps can guarantee that all loops are put in this form and thus all loops are now potentially unrollable if an upper bound on the number of iterations can be found. Signed-off-by: Ian Romanick <[email protected]>
* glsl2: Add pass to remove redundant jumpsIan Romanick2010-09-136-1/+118
|
* glsl: Explain file naming conventionIan Romanick2010-09-131-0/+12
|
* loop_controls: fix analysis of already analyzed loopsLuca Barbieri2010-09-131-1/+8
| | | | | | The loop_controls pass didn't look at the counter values it put in ir_loop on previous iterations, so while the first iteration worked, subsequent ones couldn't determine max_iterations.
* glsl: add continue/break/return unification/elimination pass (v2)Luca Barbieri2010-09-136-250/+548
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Changes in v2: - Base class renamed to ir_control_flow_visitor - Tried to comply with coding style This is a new pass that supersedes ir_if_return and "lowers" jumps to if/else structures. Currently it causes no regressions on softpipe and nv40, but I'm not sure whether the piglit glsl tests are thorough enough, so consider this experimental. It can be asked to: 1. Pull jumps out of ifs where possible 2. Remove all "continue"s, replacing them with an "execute flag" 3. Replace all "break" with a single conditional one at the end of the loop 4. Replace all "return"s with a single return at the end of the function, for the main function and/or other functions This gives several great benefits: 1. All functions can be inlined after this pass 2. nv40 and other pre-DX10 chips without "continue" can be supported 3. nv30 and other pre-DX10 chips with no control flow at all are better supported Note that for full effect we should also teach the unroller to unroll loops with a fixed maximum number of iterations but with the canonical conditional "break" that this pass will insert if asked to. Continues are lowered by adding a per-loop "execute flag", initialized to TRUE, that when cleared inhibits all execution until the end of the loop. Breaks are lowered to continues, plus setting a "break flag" that is checked at the end of the loop, and trigger the unique "break". Returns are lowered to breaks/continues, plus adding a "return flag" that causes loops to break again out of their enclosing loops until all the loops are exited: then the "execute flag" logic will ignore everything until the end of the function. Note that "continue" and "return" can also be implemented by adding a dummy loop and using break. However, this is bad for hardware with limited nesting depth, and prevents further optimization, and thus is not currently performed.
* glsl: add ir_control_flow_visitorLuca Barbieri2010-09-131-0/+17
| | | | | | | | | | This is just a subclass of ir_visitor with empty implementations of all the visit methods for non-control flow nodes. Used to avoid duplicating that in ir_visitor subclasses. ir_hierarchical_visitor is another way to solve this, but is less natural for some applications.
* glsl2: Fix scons build for all platformsJakob Bornecrantz2010-09-101-0/+1
|
* glsl2: Implement noise[1234] built-in functions using ir_unop_noiseIan Romanick2010-09-095-104/+458
|
* glsl2: Add lowering pass to remove noise opcodesIan Romanick2010-09-094-1/+75
|
* glsl2: Add ir_unop_noiseIan Romanick2010-09-092-0/+5
|
* glsl/builtins: normalize of a negative scalar should be -1.0.Kenneth Graunke2010-09-092-2/+2
|
* glsl: add several EmitNo* options, and MaxUnrollIterationsLuca Barbieri2010-09-086-10/+12
| | | | | | | | | This increases the chance that GLSL programs will actually work. Note that continues and returns are not yet lowered, so linking will just fail if not supported. Signed-off-by: Ian Romanick <[email protected]>
* glsl: Add info about talloc and optimization passes to the README.Eric Anholt2010-09-081-0/+26
|
* glsl: Update README talking about multi-instruction operations.Eric Anholt2010-09-081-4/+4
| | | | | | The previous thing taking multiple instructions ended up being handled at the IR level, as we suggested would be the common result. Pick a new one.
* glsl/builtins: Set the API in the fake context.Kenneth Graunke2010-09-082-0/+2
| | | | Otherwise it gets used uninitialized.
* glsl2: Clear out profile pointers in _mesa_glsl_release_functionsIan Romanick2010-09-082-0/+2
| | | | | | Otherwise builtin_profiles contains dangling pointers the next time _mesa_read_profile is called. I suspect this may fix bugzilla #29847, but I was never able to reproduce it.
* glsl: Fix for scalar float built-in definitions.Kenneth Graunke2010-09-083-4/+4
| | | | These need abs, and we need more tests.
* glsl: regenerate builtinsEric Anholt2010-09-081-1/+1
|
* glsl: Fix typo in builtin step() using a wrong channel.Eric Anholt2010-09-081-1/+1
|
* ir_validate: Ensure ir_binop_dot is only used on vector types.Kenneth Graunke2010-09-081-0/+1
|
* glsl: Refresh automatically generated file builtin_function.cpp.Kenneth Graunke2010-09-081-11/+9
|
* glsl/builtins: Don't use ir_binop_dot on floating point values.Kenneth Graunke2010-09-084-6/+6
| | | | ir_binop_dot is only defined for vector types. Use ir_binop_mul.
* glsl/builtins: Simplify degenerate scalar float cases.Kenneth Graunke2010-09-083-5/+3
| | | | | | | The code being generated was just stupid, considering that: - normalize(x) = 1.0 - length(x) = x - distance(x, y) = x - y
* glsl2: Make sure _mesa_glsl_parse_state constructor gets a contextIan Romanick2010-09-082-2/+4
| | | | | Fix an major regression in dc754586. Too bad that change was obviously never tested.
* glsl: Support GLSL ES in the standalone compile.Chia-I Wu2010-09-081-1/+3
| | | | GLSL ES mode is enabled when --glsl-es is passed to glsl_compiler.
* glsl: Require a context in _mesa_glsl_parse_state.Chia-I Wu2010-09-082-62/+62
| | | | | Create a dummy context in the standalone compiler and pass it to _mesa_glsl_parse_state.
* glcpp: Fix build on non-GCC compilers.Kenneth Graunke2010-09-072-90/+92
|
* ast_to_hir: Mark arrays as lvalues in GLSL ES, but prohibit assignment.Kenneth Graunke2010-09-071-1/+7
| | | | | | This allows them to be passed as out/inout parameters, but still prevents them from being used as the target of an assignment. This is per section 5.8 of the GLSL ES 1.00 specification.
* glsl: Allow overloading of built-ins without hiding in GLSL ES.Kenneth Graunke2010-09-071-1/+1
| | | | The rules are explicitly different from desktop GLSL.
* glsl: Move is_builtin flag back to ir_function_signature.Kenneth Graunke2010-09-079-15/+28
| | | | | | | | | | | This effectively reverts b6f15869b324ae64a00d0fe46fa3c8c62c1edb6c. In desktop GLSL, defining a function with the same name as a built-in hides that built-in function completely, so there would never be built-in and user function signatures in the same ir_function. However, in GLSL ES, overloading built-ins is allowed, and does not hide the built-in signatures - so we're back to needing this.
* ast_to_hir: Reject embedded structure definitions in GLSL ES 1.00.Kenneth Graunke2010-09-071-0/+9
|
* ast_to_hir: Reject unsized array declarations in GLSL ES 1.00.Kenneth Graunke2010-09-071-9/+18
|
* ast_to_hir: Allow matrix-from-matrix constructors in GLSL ES.Kenneth Graunke2010-09-071-1/+1
| | | | Everything but 1.10 supports this, so just change the check to ==.
* linker: Fix assertion and cross-version checks for version 100.Kenneth Graunke2010-09-071-2/+3
| | | | | | Fixes an assert (min_version >= 110) which was no longer correct, and also prohibits linking ES2 shaders with non-ES2 shaders. I'm not positive this is correct, but the specification doesn't seem to say.
* glsl: Add built-in function profiles for GLSL ES 1.00.Kenneth Graunke2010-09-073-11/+2953
|
* glsl: Add built-in variables for GLSL ES 1.00.Kenneth Graunke2010-09-072-0/+82
|
* glsl: Split out types that are in 1.10 but not GLSL ES 1.00.Kenneth Graunke2010-09-073-15/+35
|
* glsl: Recognize GLSL ES 1.00 keywords.Kenneth Graunke2010-09-074-281/+287
|
* glsl: Define GL_ES preprocessor macro if API is OpenGL ES 2.0.Kenneth Graunke2010-09-078-227/+249
| | | | Also define it if #version 100 is encountered.
* glsl: Accept language version 100 and make it the default on ES2.Kenneth Graunke2010-09-074-242/+255
|
* glsl: Set default language version in mesa_glsl_parse_state constructor.Kenneth Graunke2010-09-074-260/+249
| | | | | | | | | | 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.
* glsl2: Forbid array-types in ?: operator in GLSL 1.10Ian Romanick2010-09-071-0/+11
| | | | Fixes bugzilla #30039.
* glsl2: Early return with visit_continue in ↵Ian Romanick2010-09-071-1/+1
| | | | | | | | | | | | loop_analysis::visit(ir_dereference_variable *) Returning early with visit_continue_with_parent prevented the then-statements and else-statements of if-statements such as the following from being processed: if (some_var) { ... } else { ... } Fixes piglit test case glsl-fs-loop-nested-if and bugzilla #30030.
* glsl: Change grammar rules for selection statements to match the spec.Kenneth Graunke2010-09-073-950/+913
| | | | | | Fixes piglit test case loop-06.vert. Unfortunately, causes 1 shift/reduce conflict.
* glsl2: check for _NumLinkedShaders being 0Török Edvin2010-09-071-1/+1
| | | | | | | Otherwise spring 0.82+.4.0 crashes when starting a game because prog->_LinkedShaders[0] is NULL. This also fixes piglit test cases glsl-link-empty-prog-0[12].
* glsl: Add new files to sconscript.José Fonseca2010-09-051-0/+3
|
* ir_reader: Only validate IR when a global 'debug' flag is set.Kenneth Graunke2010-09-051-1/+4
| | | | | | | | | This extra validation is very useful when working on the built-ins, but in general overkill - the results should stay the same unless the built-ins or ir_validate have changed. Also, validating all the built-in functions in every test case makes piglit run unacceptably slow.
* ir_reader: Run ir_validate on the generated IR.Kenneth Graunke2010-09-041-0/+2
| | | | It's just too easy to get something wrong in hand-written IR.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-04 17:24:46 +0000