summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* scons: Add glsl_symbol_table.cppJosé Fonseca2010-08-261-0/+1
|
* i965: Add support for destination RelAddr writes in the VS.Eric Anholt2010-08-262-13/+48
| | | | Fixes: glsl-vs-varying-array
* i965: Fix the test for variable indexing of shader inputs.Eric Anholt2010-08-261-5/+11
| | | | | | Shader inputs appear in source registers, not dst registers. Catches unsupported shaders in glsl-fs-varying-array and Humus RaytracedShadows.
* glsl2: Remove a couple FINISHME comments that have already been resolvedIan Romanick2010-08-261-7/+1
|
* glsl: Move built-ins to live beyond the global scope.Kenneth Graunke2010-08-261-4/+20
| | | | | | | | | | Per the GLSL 1.20 specification (presumably a clarification of 1.10). Also, when creating user functions, make a new ir_function that shadows the built-in ir_function, rather than adding new signatures. User functions are supposed to hide built-ins, not overload them. Fixes piglit tests redeclaration-{04, 12, 14}.vert.
* glsl: Move is_built_in flag from ir_function_signature to ir_function.Kenneth Graunke2010-08-268-18/+13
| | | | | | Also rename it to "is_builtin" for consistency. Signed-off-by: Ian Romanick <[email protected]>
* glsl: Refactor variable declaration handling.Kenneth Graunke2010-08-261-36/+41
| | | | | | | | | | | | | Moving the check for an earlier variable declaration helps cleanly separate out the re-declaration vs. new declaration code a bit. With that in place, conflicts between variable names and structure types or function names aren't caught by the earlier "redeclaration" error message, so check the return type on glsl_symbol_table::add_variable and issue an error there. If one occurs, don't emit the initializer. Fixes redeclaration-01.vert and redeclaration-09.vert. Signed-off-by: Ian Romanick <[email protected]>
* glsl: Don't add overloads to existing structure constructors.Kenneth Graunke2010-08-263-4/+9
| | | | | | Instead, make a new ir_function and try to add it to the symbol table. Fixes piglit test redeclaration-08.vert.
* glsl: Remove name_declared_this_scope check when adding functions.Kenneth Graunke2010-08-261-9/+8
| | | | | | Instead, rely on the symbol table's rules. Fixes redeclaration-02.vert.
* glsl: Use a single shared namespace in the symbol table.Kenneth Graunke2010-08-2611-576/+686
| | | | | | | | | | | | As of 1.20, variable names, function names, and structure type names all share a single namespace, and should conflict with one another in the same scope, or hide each other in nested scopes. However, in 1.10, variables and functions can share the same name in the same scope. Structure types, however, conflict with/hide both. Fixes piglit tests redeclaration-06.vert, redeclaration-11.vert, redeclaration-19.vert, and struct-05.vert.
* st/mesa: add missing packed depth/stencil formats in st_format_datatype()Brian Paul2010-08-261-1/+3
| | | | Fixes llvmpipe regression from one of the prev commits.
* mesa: fix mixed-up function call nameBrian Paul2010-08-261-2/+2
|
* mesa: Identify packed depth/stencil buffers using the Format field.Nick Bowler2010-08-263-2/+21
| | | | | | | | | | | | | | | | | | | | Intel sometimes uses packed depth/stencil buffers even when only a depth buffer or only a stencil buffer was requested. Common code currently uses the _BaseFormat field to determine whether a depth/stencil wrapper is necessary. But unless the user explicitly requested a packed depth/stencil buffer, the _BaseFormat field does not encode this information, and the required wrappers are not created. The problem was introduced by commit 45e76d2665b38b ("mesa: remove a bunch of gl_renderbuffer fields"), which killed off the _ActualFormat field upon which the decision to create a wrapper used to be made. This patch changes the logic to use the Format field instead, which is more like the old code. Fixes fdo bug 27590. Signed-off-by: Nick Bowler <[email protected]> Signed-off-by: Brian Paul <[email protected]>
* intel: Merge identical cases in switch statement.Nick Bowler2010-08-261-11/+5
| | | | | Signed-off-by: Nick Bowler <[email protected]> Signed-off-by: Brian Paul <[email protected]>
* r600: fix vertex buffer size calculationAndre Maasikas2010-08-262-12/+2
| | | | | | | when we dont know max_index we cannot calculate vb size from count anymore - just use the bo size. Also added an assert to remind that we dont handle GL_INT GL_DOUBLE upload when we dont' know max_index - will fix later
* graw: Undo late loading of graw drivers.José Fonseca2010-08-2616-163/+227
| | | | | | | | | | Keith prefers a clean separation between graw applications and implementations, where apps do not link libgallium.a but instead get all functionality they need via graw interface. Although this is not incompatible with late loading of graw drivers, it it would make it very hard to maintain, as wrappers for every utility symbol exposed in graw would have to be written or generated somehow.
* st/mesa: Fix glEGLImageTargetTexture2DOES.Chia-I Wu2010-08-262-1/+10
| | | | | | | stObj->pt should be set in st_bind_surface, just as in st_TexImage. On the other hand, st_TexImage should unreference stObj->pt. It also needs to initialize the texture image again as _mesa_clear_texture_object clears the image.
* st/egl: Add support for EGL_MESA_image_drm.Chia-I Wu2010-08-264-0/+260
|
* st/egl: Add support for EGL_KHR_surfaceless_*.Chia-I Wu2010-08-262-9/+16
|
* st/mesa: Add support for surfaceless current contexts.Chia-I Wu2010-08-261-4/+27
| | | | | A surfaceless current context is a context that is made current without draw and read framebuffers. Such contexts can only render to FBOs.
* st/egl: Make KMS support optional in KMS backend.Chia-I Wu2010-08-261-22/+24
| | | | It should be called DRM backend now.
* r300g: Include missing header in r300_texture_desc.h.Vinson Lee2010-08-261-0/+1
| | | | | | Include p_format.h for enum pipe_format symbol. Fixes r300g build.
* gallium: Clean up header file inclusion in p_defines.h.Vinson Lee2010-08-261-1/+1
| | | | | Remove p_format.h. Include p_compiler.h for boolean and uint64_t symbols.
* gallium: Clean up header file inclusion in p_context.h.Vinson Lee2010-08-261-6/+24
| | | | | | Remove p_state.h. Include p_compiler.h for boolean symbol. Add needed forward declarations after removing p_state.h.
* graw: Include missing header in graw_dl.h.Vinson Lee2010-08-261-2/+3
| | | | Include p_state.h for pipe_shader_state symbol.
* util: Include missing header in u_simple_shaders.c.Vinson Lee2010-08-261-0/+1
| | | | Include p_state.h for PIPE_MAX_COLOR_BUFS symbol.
* tgsi: Include missing header in tgsi_sse2.h.Vinson Lee2010-08-261-1/+4
| | | | | Include p_compiler.h for boolean symbol. Clean up forward declarations.
* rtasm: Include missing header in rtasm_x86sse.h.Vinson Lee2010-08-261-0/+1
| | | | Include p_compiler.h for stdint.h uint*_t symbols.
* pipebuffer: Clean up header file inclusion in pb_bufmgr.h.Vinson Lee2010-08-261-2/+1
| | | | | Remove p_compiler.h and p_defines.h. Include pb_buffer.h for pb_size symbol.
* glsl2: Move ir_expression_flattening to using the rvalue visitor class.Eric Anholt2010-08-261-110/+14
| | | | | | The previous implementation was missing handling of some rvalues, such as "if" conditions, leading to glsl-mat-int-from-ctor-* not getting caught.
* glsl: Don't consider things with a type containing a sampler as an lvalue.Eric Anholt2010-08-251-0/+23
| | | | | | | | | | | | | | | We had ad-hoc handled some common cases by flagging sampler-typed variables as read_only, and rejected initializers of samplers. However, people could sneak them in in all sorts of surprising ways, like using whole-array or structure assignment. Fixes: glslparsertest/glsl2/sampler-01.frag glslparsertest/glsl2/sampler-03.frag glslparsertest/glsl2/sampler-04.frag glslparsertest/glsl2/sampler-06.frag Bug #27403.
* mesa: Disable the debug spam for x86 CPU detection with a compile-time flag.Eric Anholt2010-08-251-7/+13
| | | | | I don't know of any problems with CPU detection in years. Don't spam the user that was just looking to enable assertions with this stuff.
* mesa: Move the x86 detection debug messages next to the detection.Eric Anholt2010-08-252-33/+39
|
* mesa: Don't warn when the desired result of s3tc lib available occurs.Eric Anholt2010-08-251-1/+0
|
* util: Include missing header in u_blit.h.Vinson Lee2010-08-251-5/+7
| | | | | Include p_compiler.h for uint symbol. Clean up forward declarations.
* util: Include missing header in u_bitmask.h.Vinson Lee2010-08-251-0/+3
| | | | Include p_compiler.h for boolean symbol.
* util: Include missing header in u_dirty_surfaces.h.Vinson Lee2010-08-251-0/+4
| | | | Include p_state.h for pipe_surface symbol.
* util: Add include guard in u_split_prim.h.Vinson Lee2010-08-251-0/+6
|
* util: Include missing headers in u_split_prim.h.Vinson Lee2010-08-251-1/+4
| | | | | Include p_compiler.h for boolean symbol. Include u_debug.h for assert symbol.
* graw: Remove graw-null.José Fonseca2010-08-264-160/+1
| | | | | Pointless now that the graw tests can be built independently of any graw implementation.
* graw: Dynamically load graw libraries.José Fonseca2010-08-2612-66/+161
| | | | | | | | | This allows to build multiple graws libs simultaneously and avoid unnecessary rebuilds of the tests. Also remove graw_util.c from inside the graw implementation -- it was only being provided by one implementation, and graw tests were linking against gallium anyway.
* scons: Fix old script compatability logic.José Fonseca2010-08-261-10/+5
| | | | Sconscript could be invoked twice if specified in the command line.
* util: Include missing headers in u_tile.h.Vinson Lee2010-08-251-0/+3
| | | | | Include p_format.h for enum pipe_format symbol. Include p_state.h for pipe_box symbol.
* util: Clean up header file inclusion in u_upload_mgr.h.Vinson Lee2010-08-251-4/+1
| | | | | | Remove p_defines.h. Remove unnecessary forward declarations. Add forward declaration for pipe_context.
* r300g: fix constant buffer upload once again for r3xx->r4xxMarek Olšák2010-08-261-1/+1
|
* r600: Remove unnecessary headers.Vinson Lee2010-08-255-8/+0
|
* r300: Remove unnecessary header.Vinson Lee2010-08-251-1/+0
|
* Revert "mesa: Don't add 1 to GL_ACTIVE_UNIFORM_MAX_LENGTH."Eric Anholt2010-08-251-0/+2
| | | | | | | | | | This reverts commit 001a7bfdfc8b3c8930d5ced21982dbdfb8cd35b3. I hadn't found the section of the spec clarifying that the old behavior was right. Reverting fixes the new version of the testcase, and the Humus demos that could no longer find their uniforms. Bug #29782 Bug #29783
* glsl: fix crash with variable indexing into array in a structAras Pranckevicius2010-08-251-1/+7
| | | | Signed-off-by: Ian Romanick <[email protected]>
* i965: Fix detection of implicit MOVs to message regs in brw_optimize.c.Eric Anholt2010-08-251-24/+38
| | | | | | | Texcoords in AmbientApertureLighting were getting trashed since the move of math arguments to implied moves, due to the logic for detecting ALU message reg writes overriding the logic for SEND implicit message reg writes.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-21 08:17:15 +0000