aboutsummaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/llvmpipe/lp_test_conv.c
Commit message (Collapse)AuthorAgeFilesLines
* llvmpipe: don't call LLVMCreateJITCompiler() twiceBrian Paul2010-07-291-12/+2
| | | | | | | | | Fixes a failed assertion with LLVM 2.6: <unnamed>::JITResolver::JITResolver(llvm::JIT&): Assertion `TheJITResolver == 0&& "Multiple JIT resolvers?"' failed. Though, not everyone seems to experience this problem.
* gallivm: Fix 4 x unorm8 -> 4 x float conversion.José Fonseca2010-07-021-1/+1
| | | | Also fix the test.
* gallivm: Allow to conversions to/from registers of different sizes.José Fonseca2010-07-011-4/+16
| | | | | | Allow for example to convert from 4 x float32 to 4 x unorm8 and vice versa. Uses code and ideas from Brian Paul.
* llvmpipe: new -s option to run single testBrian Paul2010-06-041-2/+24
| | | | | Put specific test code in the test_single() function and pass -s to execute that code.
* llvmpipe: Use pointer_to_func() instead of custom wrappers.José Fonseca2010-05-301-12/+2
|
* llvmpipe: silence cast warnings in test programsBrian Paul2010-05-131-2/+15
|
* llvmpipe: s/bool/boolean/ in test progsBrian Paul2010-05-121-2/+2
|
* llvmpipe: export the tgsi translation code to a common layerZack Rusin2010-02-081-4/+4
| | | | | | | the llvmpipe tgsi translation is a lot more complete than what was in gallivm so replacing the latter with the former. this is needed since the draw llvm paths will use the same code. effectively the proven llvmpipe code becomes gallivm.
* gallium: Rename ALIGN_STACK -> PIPE_ALIGN_STACK for consistency.José Fonseca2010-01-121-1/+1
|
* gallium: Simplify PIPE_ALIGN_VAR.José Fonseca2010-01-121-2/+2
| | | | | | gcc allows pre-fix variable attributes. Suggested by Ian Romanick.
* gallium: Generalize the alignment macros to other compilers and any alignment.José Fonseca2010-01-121-2/+2
|
* llvmpipe: Use C-style comment.Vinson Lee2010-01-011-1/+1
|
* llvmpipe: Ensure stack variables in unit tests are properly aligned.José Fonseca2009-11-091-2/+3
|
* llvmpipe: Avoid variable size arrays.José Fonseca2009-10-221-2/+2
| | | | Not really variable size, but MSVC still doesn't like them.
* llvmpipe: Make the code portable for MinGW.José Fonseca2009-09-161-2/+2
|
* llvmpipe: Make lp_type a regular union.José Fonseca2009-09-141-36/+36
| | | | | Union not worth the hassle of violating C99 or adding a name to the structure.
* llvmpipe: Get conversions working correctly for all integer types.José Fonseca2009-08-291-7/+7
|
* llvmpipe: Several fixes of the conversion test.José Fonseca2009-08-291-3/+7
|
* llvmpipe: Disassemble generated x86 code.José Fonseca2009-08-291-0/+4
|
* llvmpipe: Test only present functionality to allow regression testing.José Fonseca2009-08-291-0/+7
|
* llvmpipe: Handle more conversions.José Fonseca2009-08-291-20/+42
| | | | Not all successfully yet though..
* llvmpipe: Normalize the cycles with the number of channel.José Fonseca2009-08-291-5/+3
| | | | So that we have a comparable number for different formats.
* llvmpipe: Handle ubyte -> float conversion too.José Fonseca2009-08-291-2/+2
|
* llvmpipe: Bootstrap type conversions.José Fonseca2009-08-291-0/+392