summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* glsl2: Implement ir_function::clone and ir_function_signature::cloneIan Romanick2010-07-121-6/+43
|
* glsl2: Move temp declaration to correct side of if-statement in IRIan Romanick2010-07-121-7/+7
|
* glsl2: Add missing fields in ir_variable::cloneIan Romanick2010-07-121-0/+8
|
* glsl2: Add declarations for temporaries to instruction streamIan Romanick2010-07-121-0/+3
| | | | | | | | Temporary variables added for &&, ||, and ?: were not being added to the instruction stream. This resulted in either test failures or Valgrind being angry after the original IR tree was destroyed by talloc_free. The talloc_free caused the ir_variables to be destroyed even though they were still referenced.
* glsl2: Store the gl_type of the array's element type in the array.Eric Anholt2010-07-121-0/+5
| | | | | Fixes glsl-fs-uniform-array-1, glsl-vs-uniform-array-1, and the -2 tests on software.
* Revert "ir_to_mesa: Add support for adding/subtracting matrices."Eric Anholt2010-07-121-39/+3
| | | | | This reverts commit b4d0c0e0ee983ee614b047799c3e01221a353c98. Now that ir_mat_op_to_vec is landed, this change is no longer needed.
* glsl2: Add a new pass at the IR level to break down matrix ops to vector ops.Eric Anholt2010-07-125-0/+197
| | | | | | | This will be used by the Mesa IR and likely most HW backends, as it allows other optimizations to occur that might not otherwise. Fixes glsl-vs-mat-sub-1, glsl-vs-mat-div-1.
* glsl2: Flatten out expressions that are the child of an assignment rhs.Eric Anholt2010-07-121-7/+19
| | | | | | | This feels a little odd, but it will be useful for ir_mat_to_vec, where I want to see a plain assignment of the expression to a variable, not to a writemasked array dereference with a call as the array index.
* glsl2: Check when inlining a bare function call that it actually is.Eric Anholt2010-07-121-0/+6
| | | | | | | It would be easy to miss an entry either of the two visitors involved that would result in trying to ir->remove() the call to remove it from the instruction stream when really it's part of an expression tree that wasn't flattened.
* ast_function: Fix non-float constructors with matrix arguments.Kenneth Graunke2010-07-091-22/+55
| | | | | | | | | | Previously, code like ivec4(mat2(...)) would fail because the compiler would naively try to convert a mat2 to an imat2...which doesn't exist. Now, a separate pass breaks such matrices down to their columns, which can be converted from vec2 to ivec2. Fixes piglit tests constructor-11.vert, constructor-14.vert, constructor-15.vert, and CorrectConstFolding2.frag.
* ast_function: Move error return earlier and don't indent the world.Kenneth Graunke2010-07-091-142/+139
| | | | This has no functional changes.
* ast_function: Remove unnecessary check for empty constructors.Kenneth Graunke2010-07-091-9/+0
| | | | | This case is already caught by a later check that ensures sufficient components were provided, based on the type.
* glsl2: Use new foreach_list_safe abstraction.Kenneth Graunke2010-07-091-7/+3
|
* glsl2: Add foreach_list_safe which works even when mutating the list.Kenneth Graunke2010-07-091-0/+14
| | | | | | | | In particular, with foreach_list_safe, one can remove and free the current node without crashes; if new nodes are added after the current node, they will be properly visited as well. Signed-off-by: Ian Romanick <[email protected]>
* glsl2: Remove generate_temporary and global temporary counter.Kenneth Graunke2010-07-084-32/+5
| | | | | | | Most places in the code simply use a static name, which works because names are never used to look up an ir_variable. generate_temporary is simply unnecessary (and looks like it would leak memory, and isn't thread safe...)
* mesa: Extend register lifetimes to the end of the largest loop required.Eric Anholt2010-07-071-17/+23
| | | | | | | | | | | | | Previously, a register defined at main scope and used in a loop in a loop could end up getting marked as needed only from the definition outside of the loops to the end of the inner loop, and we would cleverly slot in something else in its register in the end of the outer loop. Fixes glsl-vs-loop-nested and glsl-fs-loop-nested on glsl2. This doesn't happen much on master because the original compiler does its own register allocation, so we find little we can do with linear scan register (re)allocation.
* mesa: Fix documentation of BranchTarget for BRK.Eric Anholt2010-07-071-1/+1
| | | | It was changed in 2009 and the comment wasn't updated.
* glsl2: Add support for gl_PointCoord in 1.20.Eric Anholt2010-07-072-0/+11
| | | | Fixes glsl-fs-pointcoord on swrast (remains broken on 965, like master)
* ir_to_mesa: Fix the assertion on LHS array derefs to DWIM.Eric Anholt2010-07-071-7/+10
| | | | | This allows array derefs of matrices now, which makes idr's GLSL demo happy.
* ir_to_mesa: Add support for constant matrices (untested).Eric Anholt2010-07-071-2/+33
|
* ir_to_mesa: Add support for adding/subtracting matrices.Eric Anholt2010-07-071-3/+39
| | | | This isn't really tested, but didn't break normal vector add/sub.
* ir_to_mesa: Don't assert that we can't assign matrices. It should work now.Eric Anholt2010-07-071-1/+0
|
* ir_to_mesa: Only allocate a vector per column of a matrix.Eric Anholt2010-07-071-1/+1
|
* ir_to_mesa: Add support for assignment of aggregates.Eric Anholt2010-07-071-3/+12
|
* ir_to_mesa: Add support for matrix * matrix.Eric Anholt2010-07-071-16/+27
|
* ir_reader: Don't emit ir_function multiple times.Kenneth Graunke2010-07-071-8/+8
|
* linker: Use bit-0 instead of VERT_BIT_GENERIC0Ian Romanick2010-07-071-1/+1
| | | | | | Uses of the bits for allocation are offset by 16, and VERT_BIT_GENERIC0 already has the 16 offset. As a result, it was preventing the wrong thing from being allocated.
* glsl: Fix the setup of refract()'s output for vec3/vec4 and k < 0.0.Eric Anholt2010-07-072-4/+4
| | | | caught by valgrind.
* glsl2: Fix ir_div_to_mul_rcp for integer division.Eric Anholt2010-07-071-9/+47
| | | | | | rcp of an integer value did not produce the result you're looking for. Instead, do the a * rcp(b) as float and truncate after. This mostly fixes glsl-fs-loop-nested.
* glsl2: Actually add the declaration of _post_incdec_temp.Eric Anholt2010-07-071-0/+1
|
* ir_constant_expression: Fix loop increments.Kenneth Graunke2010-07-071-2/+2
|
* glsl2: Initialize yylineno and yycolumn so line numbers are sane.Kenneth Graunke2010-07-072-0/+3
|
* glsl2: Put the initializer in the instruction stream after the declarationIan Romanick2010-07-071-2/+11
|
* exec_list: Add method to append one complete list to anotherIan Romanick2010-07-071-0/+24
|
* Revert "glsl2: Put the declaration in the instruction stream before its ↵Ian Romanick2010-07-071-2/+2
| | | | | | | | | initializer." This change causes segfaults in other tests. A fix for both sets of segfaults is coming. This reverts commit d4d630b72c7b7f38074addda0f1b819608247d93.
* ir_to_mesa: Add support for gl_TextureMatrix access.Eric Anholt2010-07-071-19/+84
| | | | Fixes glsl-vs-texturematrix-1, and glsl-vs-texturematrix-2 on swrast.
* ir_to_mesa: Fill in some uninitialized fields that sometimes contained junk.Eric Anholt2010-07-071-2/+4
|
* glsl2: Add a pass to simplify if statements returning from both sides.Eric Anholt2010-07-074-0/+126
| | | | | | | | | | | | | | | This allows function inlining making the following tests work even without function calls implemented: glsl-fs-functions-2 glsl-fs-functions-3 glsl-vs-functions glsl-vs-functions-2 glsl-vs-functions-3 glsl-vs-vec4-indexing-5 (Note that those tests were designed to trigger actual function calls, and this defeats them. However, those testcases ended up catching the bug in the previous commit.)
* glsl2: Clean up vec_index_to_cond_assign after the clone return type change.Eric Anholt2010-07-071-2/+2
|
* glsl2: Don't forget to walk the parameters to a function in the hv.Eric Anholt2010-07-071-0/+4
| | | | | Fixes segfaults from use after free after the steal of ir nodes and free of the compile context.
* glsl2: Avoid null deref in scalar constant unop expressions.Eric Anholt2010-07-071-1/+6
|
* glsl2: Put the declaration in the instruction stream before its initializer.Eric Anholt2010-07-061-2/+2
| | | | | This fixes a regression in the generated code from when I did the ir_validate.cpp-driven rework of assignments.
* glsl2: Add pass for supporting variable vector indexing in rvalues.Eric Anholt2010-07-065-0/+206
| | | | | | | The Mesa IR needs this to support vector indexing correctly, and hardware backends such as 915 would want this behavior as well. Fixes glsl-vs-vec4-indexing-2.
* ir_to_mesa: Add support for conditional moves.Eric Anholt2010-07-061-5/+27
| | | | Nothing generates conditional moves yet.
* glsl2: Clone methods return the type of the thing being clonedIan Romanick2010-07-065-41/+45
| | | | This is as opposed to returning the type of the base class of the hierarchy.
* ir_constant_expression: Declare loop counting variables in the loops.Kenneth Graunke2010-07-061-24/+23
| | | | Fixes "name lookup of 'c' changed" warning.
* glsl2: Update TODO.Kenneth Graunke2010-07-061-11/+3
|
* ir_constant_expression: Add support for dot products.Kenneth Graunke2010-07-061-0/+20
|
* ir_constant_expression: Add support for matrix multiplication.Kenneth Graunke2010-07-061-3/+25
| | | | | | | Also handles matrix/vector and vector/matrix multiplication. Fixes piglit tests const-matrix-multiply-01.frag, const-matrix-multiply-02.frag, and const-vec-mat.frag.
* ir_constant_expression: Support scalar * vector and scalar * matrix.Kenneth Graunke2010-07-061-7/+10
| | | | | | | The test here is slightly different since we need to keep matrix multiplication separate. Fixes piglit tests const-vec-scalar-03.frag and const-mat-scalar-03.frag.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-15 07:58:47 +0000