Age | Commit message (Collapse) | Author | |
---|---|---|---|
2008-10-09 | cell: massage the emit functions to get better instruction scheduling | Brian Paul | |
2008-10-09 | cell: more accurate comments | Brian Paul | |
2008-10-10 | gallium: silence warning | Alan Hourihane | |
2008-10-09 | Gallivm: cleanup soa storage. | Stephane Marchesin | |
2008-10-08 | cell: implement function calls from shader code. fslight demo runs now. | Brian Paul | |
Used for SIN, COS, EXP2, LOG2, POW instructions. TEX next. Fixed some bugs in MIN, MAX, DP3, DP4, DPH instructions. In rtasm code: Special-case spe_lqd(), spe_stqd() functions so they take byte offsets but low-order 4 bits are shifted out. This makes things consistant with SPU assembly language conventions. Added spe_get_registers_used() function. | |||
2008-10-08 | cell: implement more built-in shader functions, link spu code with -lm | Brian Paul | |
2008-10-08 | cell: increase SPU_MAX_FRAGMENT_PROGRAM_INSTS | Brian Paul | |
2008-10-08 | gallium: asst. clean-ups | Brian Paul | |
Don't use register qualifier. Doxygen-ize comments. Remove 'extern'. | |||
2008-10-08 | gallium: better instruction printing for SPE code | Brian Paul | |
2008-10-07 | mesa: use the shaderutil.c helper functions | Brian Paul | |
2008-10-07 | cell: add support for fragment shader constant buffers | Brian Paul | |
2008-10-07 | cell: fix incorrect extended swizzle term code in get_src_reg() | Brian Paul | |
2008-10-07 | cell: fix formatting | Brian Paul | |
2008-10-07 | cell: remove old code | Brian Paul | |
2008-10-07 | Gallivm: reorder the functions alphabetically so I can work on it. | Stephane Marchesin | |
2008-10-07 | Merge branch 'gallium-0.2' of ↵ | Stephane Marchesin | |
git+ssh://marcheu@git.freedesktop.org/git/mesa/mesa into gallium-0.2 | |||
2008-10-07 | Progs: hook the glsl identity example into the makefile. | Stephane Marchesin | |
2008-10-07 | cell: memset() key to zero | Brian Paul | |
2008-10-07 | cell: use new keymap to save/re-use fragment ops code | Brian Paul | |
2008-10-07 | gallium: added general-purpose key->data map/lookup container | Brian Paul | |
2008-10-07 | Progs: add a trivial glsl test, useful for gallium driver bringup/debug. | Stephane Marchesin | |
2008-10-07 | Gallivm: don't say hello, it's rude. | Stephane Marchesin | |
2008-10-07 | Merge branch 'gallium-0.2' of ↵ | Stephane Marchesin | |
git+ssh://marcheu@git.freedesktop.org/git/mesa/mesa into gallium-0.2 | |||
2008-10-07 | Gallivm: fix the constant layout, this gets a bunch of progs/ working. ↵ | Stephane Marchesin | |
Notably, gears doesn't. | |||
2008-10-07 | gallium: Introduce PIPE_ARCH_SSE define for SSE support. | José Fonseca | |
Besides meaning x86 and x86-64 architecture, it also depends on SSE2 support enabled on gcc. This fixes the linux-debug build. | |||
2008-10-06 | Mesa: fix the case where there are no vertex attributes. | Stephane Marchesin | |
2008-10-03 | CELL: changes to generate SPU code for stenciling | Robert Ellison | |
This set of code changes are for stencil code generation support. Both one-sided and two-sided stenciling are supported. In addition to the raw code generation changes, these changes had to be made elsewhere in the system: - Added new "register set" feature to the SPE assembly generation. A "register set" is a way to allocate multiple registers and free them all at the same time, delegating register allocation management to the spe_function unit. It's quite useful in complex register allocation schemes (like stenciling). - Added and improved SPE macro calculations. These are operations between registers and unsigned integer immediates. In many cases, the calculation can be performed with a single instruction; the macros will generate the single instruction if possible, or generate a register load and register-to-register operation if not. These macro functions are: spe_load_uint() (which has new ways to load a value in a single instruction), spe_and_uint(), spe_xor_uint(), spe_compare_equal_uint(), and spe_compare_greater_uint(). - Added facing to fragment generation. While rendering, the rasterizer needs to be able to determine front- and back-facing fragments, in order to correctly apply two-sided stencil. That requires these changes: - Added front_winding field to the cell_command_render block, so that the state tracker could communicate to the rasterizer what it considered to be the front-facing direction. - Added fragment facing as an input to the fragment function. - Calculated facing is passed during emit_quad(). | |||
2008-10-02 | draw: modify prefetching slightly | Keith Whitwell | |
2008-10-02 | draw: don't keep refetching constant inputs | Keith Whitwell | |
2008-10-02 | rtasm: add prefetch instructions | Keith Whitwell | |
2008-10-02 | draw: add streamlined paths for fetching linear verts | Keith Whitwell | |
2008-10-02 | mesa: Fix compiler warnings on Windows. | Michal Krol | |
2008-10-02 | Gallivm: add slt. glxgears should be running, except it isn't. | Stephane Marchesin | |
2008-10-02 | Gallivm: port to llvm 2.4. | Stephane Marchesin | |
2008-10-02 | Gallivm: fix off-by-one. | Stephane Marchesin | |
2008-10-02 | Gallivm: need to link with libstdc++ for llvm. | Stephane Marchesin | |
2008-10-02 | Gallivm: more instructions. | Stephane Marchesin | |
2008-10-02 | Gallivm: make it compile again, add some opcodes. | Stephane Marchesin | |
2008-10-02 | mesa: fix temp register allocation problems. | Brian Paul | |
Complex texcombine modes were running out of registers (>32 registers for 8 tex units). | |||
2008-10-02 | egl: check for null ptr/name | Brian Paul | |
2008-10-02 | egl: remove space after -L flag | Brian Paul | |
2008-10-01 | Add -msse and -msse2 to the *-x86 configs. | José Fonseca | |
2008-10-01 | util: No-op u_sse.h outside PIPE_ARCH_X86/X86_64. | José Fonseca | |
2008-10-01 | tgsi: Include p_config.h. | José Fonseca | |
2008-09-30 | cell: Fixed usage of MAX_INSTRUCTIONS to use new MAX_PROGRAM_INSTRUCTIONS ↵ | Jonathan White | |
instead of old MAX_NV_XXX definitions in order to allow Cell TGSI fragment program generator to work again. | |||
2008-09-30 | cell: Moved X86 checks to wrap #include section so that Cell targets will ↵ | Jonathan White | |
compile again. | |||
2008-10-01 | util: Fix util_fast_pow/exp2/log2. | José Fonseca | |
- Use a lookup table for log2. - Compute (float) (1 << ipart) by tweaking with the exponent directly to avoid integer overflow and float conversion. - Also table negative exponents to avoid float division and branching. - Implement util_fast_exp as function of util_fast_exp2. | |||
2008-09-30 | tgsi: SSE2 optimized exp2, log2 and pow implementations. | José Fonseca | |
Special care must be taken when calling compiler generated SSE2 functions from the runtime generated SSE2: saving the xmm registers, and notify gcc the stack is not 16byte aligned. It would be more efficient to keep the stack pointer 16byte aligned, but too hairy, and not consistent in all x86 architectures. This has been tested in linux x86 and windows x86 userspace. Not tested on x86-64 because it is broken for other reasons (even without this change). | |||
2008-09-30 | util: Header for SSE2 intrinsics portability. | José Fonseca | |
2008-09-29 | rtasm: Implement immediate group 1 instructions. Fix SIB emition. | José Fonseca | |