Age | Commit message (Collapse) | Author | |
---|---|---|---|
2007-10-26 | convert OPCODE_END -> TGSI_OPCODE_RET | Brian | |
2007-10-26 | if we hit RET w/ empty call stack, halt | Brian | |
2007-10-26 | Initial support for immediate values in TGSI programs. | Brian | |
These can be evaluated at compile time. Code disabled pending clarifications of TGSI immediate data structures. | |||
2007-10-26 | Fix swizzle fetching in tgsi_util_get_full_src_register_extswizzle. Shorten | Zack Rusin | |
the swizzle coding in LLVM compilation using it and fix some warnings. | |||
2007-10-25 | tex comments | Brian | |
2007-10-25 | added some comments | Brian | |
2007-10-25 | update TILE_TOP/BOTTOM_* values to match softpipe | Brian | |
2007-10-25 | s/GLfloat/float/, s/GLint/int/, etc | Brian | |
2007-10-24 | print CAL label | Brian | |
2007-10-24 | fix bug in RET code | Brian | |
2007-10-22 | Finish unifying the surface and texture tile caches. | Brian | |
2007-10-20 | added case for TGSI_OPCODE_END | Brian | |
2007-10-18 | added const qualifiers | Brian | |
2007-10-18 | Corrected the file permissions in src/mesa/pipe. | Oliver McFadden | |
2007-10-15 | fix logic for printing extended swizzles | Brian | |
2007-10-12 | added code to print extended swizzles | Brian | |
2007-10-12 | change exec_kilp() to only touch lowest four bits in kilmask | Brian | |
2007-10-10 | Merge tgsi_mesa_compile_fp_program() and tgsi_mesa_compile_vp_program() into ↵ | Brian | |
tgsi_translate_mesa_program(). | |||
2007-10-09 | start on a unified vertex/fragment program translation function | Brian | |
2007-10-09 | Generate DCLs for temporaries. | Brian | |
2007-10-09 | Finish up decls for packed fragment program outputs. | Brian | |
2007-10-09 | Pack fragment program outputs to be consistant with vertex programs. | Brian | |
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. | |||
2007-10-09 | update some comments | Brian | |
2007-10-09 | More work on flow control for CAL/RET. | Brian | |
2007-10-09 | New ATTRIB interpolation type for vertex program inputs/declarations. | Brian | |
2007-10-09 | Don't emit semantic info for vertex program inputs. | Brian | |
2007-10-03 | use different temps in exec_tex() for bias | Brian | |
2007-10-03 | updated comments | Brian | |
2007-10-03 | fix eq/neq bugs | Brian | |
2007-10-03 | fix f/u mix-up in micro_trunc() | Brian | |
2007-10-03 | move TEX code into exec_tex() | Brian | |
2007-10-02 | clean-ups | Brian | |
2007-10-02 | Implement/use fragment kill results | Brian | |
2007-10-02 | Implement TGSI_OPCODE_SNE with micro_ne() | Brian | |
2007-10-02 | Separate TGSI_OPCODE_KIL and TGSI_OPCODE_KILP (predicated). | Brian | |
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. | |||
2007-10-02 | Push mask stacks upon CAL, pop upon RET. | Brian | |
Still need to handle conditional RET statements... | |||
2007-10-02 | added license and interpreter comments | Brian | |
2007-10-02 | Implement CONT statement. | Brian | |
2007-10-02 | added micro_trunc(), re-order some code | Brian | |
2007-10-02 | Minor optimization: no-op CAL when ExecMask==0x0. | Brian | |
2007-10-02 | Added TGSI_FILE_OUTPUT case in fetch_src_file_channel(). | Brian | |
GLSL output variables (such as vertex varyings) can be read (they're not write-only). | |||
2007-10-02 | Implement CAL/RET and a call stack for subroutines. | Brian | |
The glsl/brick.c shader demo runs now. | |||
2007-10-02 | Added TGSI_OPCODE_END | Brian | |
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. | |||
2007-10-02 | Fold expand_program() into tgsi_exec_prepare() | Brian | |
2007-10-02 | no more need for tgsi_exec_machine_run2(), remove dead code | Brian | |
2007-10-02 | Move tgsi machine state init/allocations so they're done less frequently. | Brian | |
This, plus expanding all instructions ahead of time, seems to have improved the performance of program execution by 8x or so. | |||
2007-10-02 | checkpoint: unpack all instructions before interpretation. Actual looping ↵ | Brian | |
works. | |||
2007-09-29 | Initial conditional execution support for loops and BRK instruction. | Brian | |
Also, instead of passing cond mask to each micro op, just apply it in the store_dest() function. | |||
2007-09-29 | remove unused tgsi_exec_cond_* types | Brian | |
2007-09-28 | Initial implementation of IF/ELSE/ENDIF using conditional masking. | Brian | |
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. |