summaryrefslogtreecommitdiffstats
path: root/src/mesa/pipe/tgsi
Commit message (Collapse)AuthorAgeFilesLines
* Finish unifying the surface and texture tile caches.Brian2007-10-221-8/+2
|
* added case for TGSI_OPCODE_ENDBrian2007-10-201-0/+4
|
* added const qualifiersBrian2007-10-182-4/+4
|
* Corrected the file permissions in src/mesa/pipe.Oliver McFadden2007-10-187-0/+0
|
* fix logic for printing extended swizzlesBrian2007-10-151-13/+15
|
* added code to print extended swizzlesBrian2007-10-121-0/+20
|
* change exec_kilp() to only touch lowest four bits in kilmaskBrian2007-10-121-3/+3
|
* Merge tgsi_mesa_compile_fp_program() and tgsi_mesa_compile_vp_program() into ↵Brian2007-10-102-291/+16
| | | | tgsi_translate_mesa_program().
* start on a unified vertex/fragment program translation functionBrian2007-10-091-0/+194
|
* Generate DCLs for temporaries.Brian2007-10-091-1/+77
|
* Finish up decls for packed fragment program outputs.Brian2007-10-092-24/+31
|
* Pack fragment program outputs to be consistant with vertex programs.Brian2007-10-091-29/+2
| | | | | | | Previously, output[0] was always Z and output[1] was color. Now output[0] will be color if Z is not written. In shade_quad() use the semantic info to determine which quantity is in which output slot.
* update some commentsBrian2007-10-091-3/+3
|
* More work on flow control for CAL/RET.Brian2007-10-092-12/+31
|
* New ATTRIB interpolation type for vertex program inputs/declarations.Brian2007-10-094-4/+7
|
* Don't emit semantic info for vertex program inputs.Brian2007-10-091-7/+10
|
* use different temps in exec_tex() for biasBrian2007-10-031-5/+5
|
* updated commentsBrian2007-10-031-14/+4
|
* fix eq/neq bugsBrian2007-10-031-16/+4
|
* fix f/u mix-up in micro_trunc()Brian2007-10-031-4/+4
|
* move TEX code into exec_tex()Brian2007-10-031-89/+125
|
* clean-upsBrian2007-10-021-21/+12
|
* Implement/use fragment kill resultsBrian2007-10-022-8/+8
|
* Implement TGSI_OPCODE_SNE with micro_ne()Brian2007-10-021-1/+20
|
* Separate TGSI_OPCODE_KIL and TGSI_OPCODE_KILP (predicated).Brian2007-10-024-6/+31
| | | | | | These correspond to the NV and ARB-style fragment program KIL instructions. The former is really supposed to examine the NV condition codes but Mesa's GLSL compiler always emits unconditional KIL instructions.
* Push mask stacks upon CAL, pop upon RET.Brian2007-10-021-0/+19
| | | | Still need to handle conditional RET statements...
* added license and interpreter commentsBrian2007-10-021-0/+52
|
* Implement CONT statement.Brian2007-10-022-8/+26
|
* added micro_trunc(), re-order some codeBrian2007-10-021-32/+47
|
* Minor optimization: no-op CAL when ExecMask==0x0.Brian2007-10-021-3/+6
|
* Added TGSI_FILE_OUTPUT case in fetch_src_file_channel().Brian2007-10-021-0/+8
| | | | | GLSL output variables (such as vertex varyings) can be read (they're not write-only).
* Implement CAL/RET and a call stack for subroutines.Brian2007-10-023-22/+37
| | | | The glsl/brick.c shader demo runs now.
* Added TGSI_OPCODE_ENDBrian2007-10-024-28/+22
| | | | | Halt program execution when we get to END instruction. The GLSL compiler puts subroutines after the end instruction so we have to stop before then.
* Fold expand_program() into tgsi_exec_prepare()Brian2007-10-022-88/+48
|
* no more need for tgsi_exec_machine_run2(), remove dead codeBrian2007-10-022-59/+1
|
* Move tgsi machine state init/allocations so they're done less frequently.Brian2007-10-022-79/+105
| | | | | This, plus expanding all instructions ahead of time, seems to have improved the performance of program execution by 8x or so.
* checkpoint: unpack all instructions before interpretation. Actual looping ↵Brian2007-10-024-9/+130
| | | | works.
* Initial conditional execution support for loops and BRK instruction.Brian2007-09-292-94/+110
| | | | | Also, instead of passing cond mask to each micro op, just apply it in the store_dest() function.
* remove unused tgsi_exec_cond_* typesBrian2007-09-291-24/+0
|
* Initial implementation of IF/ELSE/ENDIF using conditional masking.Brian2007-09-282-191/+124
| | | | | | mach->CondMask controls writing to each of the 4 components in a quad. mach->CondMaskStack handles nested conditionals. Only a few of the micro ops (add/sub/mul/store) have been updated to obey CondMask at this time.
* Fix some compiler warnings with -pedanticKeith Whitwell2007-09-271-9/+9
|
* Enable codegen based whenever __i386__ is defined.Keith Whitwell2007-09-271-1/+1
|
* Translate mesa vertex/fragment programs to TGSI programs at same time to do ↵Brian2007-09-251-0/+4
| | | | | | | | | | | | | | | | | | proper linking. Previously, programs were translated independently during validation. The problem is the translation to TGSI format, which packs shader input/outputs into continuous slots, depends on which vertex program is being paired with which fragment shader. Now, we look at the outputs of the vertex program in conjunction with the inputs of the fragment shader to be sure the attributes match up correctly. The new 'linked_program_pair' class keeps track of the associations between vertex and fragment shaders. It's also the place where the TGSI tokens are kept since they're no longer per-program state but per-linkage. Still a few loose ends, like implementing some kind of hash/lookup table for linked_program_pairs.
* Enable SSE2 for FS.michal2007-09-242-3/+250
|
* added PSIZEBrian2007-09-211-0/+2
|
* Fix up some point size breakage. Start on fogcoord too.Brian2007-09-211-2/+3
|
* remove debug printfsBrian2007-09-211-2/+0
|
* remove SEMANTIC_TEXCOORDBrian2007-09-211-2/+0
|
* Get rid of TGSI_SEMANTIC_TEXCOORD and just use TGSI_SEMANTIC_GENERIC.Brian2007-09-211-3/+2
|
* minor renamingBrian2007-09-211-7/+7
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-21 09:56:45 +0000