summaryrefslogtreecommitdiffstats
path: root/src/mesa/shader
Commit message (Collapse)AuthorAgeFilesLines
* mesa: be smarter about allocating vert/frag program constantsBrian Paul2009-05-061-25/+76
| | | | | | | Try to re-use constants/literals more often to make best use of the constant buffer space. See bug 21354.
* mesa: more complete fix for transform_invarient glitchesKeith Whitwell2009-05-051-2/+117
| | | | | | Add a new flag mvp_with_dp4 in the context, and use that to switch both ffvertex.c and programopt.c vertex transformation code to either DP4 or MUL/MAD implementations.
* mesa: also print program params/constants when dumping shaders to diskBrian Paul2009-05-041-0/+4
|
* mesa: remove some unfinished/devel codeBrian Paul2009-05-041-92/+0
|
* Merge branch 'const-buffer-changes'Brian Paul2009-05-015-32/+155
|\ | | | | | | | | | | | | | | Conflicts: src/mesa/drivers/dri/i965/brw_curbe.c src/mesa/drivers/dri/i965/brw_vs_emit.c src/mesa/drivers/dri/i965/brw_wm_glsl.c
| * mesa: added _mesa_check_soa_dependencies() functionBrian Paul2009-04-292-0/+53
| | | | | | | | | | This function will check an instruction to see if there's data dependencies between the dst and src registers if executed in an SOA manner.
| * mesa: refactor code and make _mesa_find_temp_intervals() publicBrian Paul2009-04-242-22/+144
| |
| * mesa: signal _NEW_PROGRAM_CONSTANTS instead of _NEW_PROGRAMBrian Paul2009-04-243-10/+11
| | | | | | | | | | Use _NEW_PROGRAM_CONSTANTS when changing constant/uniform buffer values. Binding a new program/shader sets both _NEW_PROGRAM and _NEW_PROGRAM_CONSTANTS.
* | mesa: added _mesa_check_soa_dependencies() functionBrian Paul2009-04-292-0/+53
| | | | | | | | | | This function will check an instruction to see if there's data dependencies between the dst and src registers if executed in an SOA manner.
* | Avoid a segfault in shader compilationRobert Ellison2009-04-271-0/+6
| | | | | | | | | | | | | | | | | | | | If a shader reaches an out-of-memory condition while adding a new function (reallocating the function list), a segfault will occur during cleanup (because the num_functions field is non-zero, but the functions pointer is NULL). This fixes that segfault by zeroing out the num_functions field if reallocation fails.
* | mesa: Fix buffer overflow when parsing generic vertex attributes.Brian Paul2009-04-221-3/+9
|/
* mesa: new _NEW_PROGRAM_CONSTANTS flagBrian Paul2009-04-213-9/+7
| | | | | | | | | | | | | | | This state flag will be used to indicate that vertex/fragment program constants have changed. _NEW_PROGRAM will be used to indicate changes to the vertex/fragment shader itself, or misc related state. _NEW_PROGRAM_CONSTANTS is also set whenever a program parameter that's tracking GL state has changed. For example, if the projection matrix is in the parameter list, calling glFrustum() will cause _NEW_PROGRAM_CONSTANTS to be set. This will let to remove the need for dynamic state atoms in some drivers. For now, we still set _NEW_PROGRAM in all the places we used to. We'll no longer set _NEW_PROGRAM in glUniform() after drivers/etc have been updated.
* mesa: print internal.current[i] attribBrian Paul2009-04-211-2/+7
|
* mesa: print parameter list dirty state flag maskBrian Paul2009-04-211-0/+1
|
* mesa: add switch case for GL_VERTEX_STATE_PROGRAM_NV in _mesa_new_program()Brian Paul2009-04-181-0/+1
| | | | Fixes bug seen in progs/tests/vptest1.c
* mesa: build a float[4] value in _mesa_add_sampler() to avoid random valuesBrian Paul2009-04-171-3/+4
|
* mesa: merge the prog_src_register::NegateBase and NegateAbs fieldsBrian Paul2009-04-148-82/+52
| | | | | | There's really no need for two negation fields. This came from the GL_NV_fragment_program extension. The new, unified Negate bitfield applies after the absolute value step.
* mesa: remove unused matrixType param from ctx->Driver.UniformMatrix() functionsBrian Paul2009-04-141-1/+1
|
* mesa: move #define for GL_PROGRAM_BINARY_LENGTH_OESBrian Paul2009-04-141-5/+0
|
* mesa: remove NV vertex/fragment program print/debug codeBrian Paul2009-04-144-407/+4
| | | | The code in prog_print.c can be used instead.
* mesa: minor datatype changes in optimization codeBrian Paul2009-04-081-6/+7
|
* glsl: enable the new linear scan register allocator codeBrian Paul2009-04-071-3/+3
| | | | | Seems to b working well enough to enable all the time. Optimizations can be disabled with "export MESA_GLSL=nopt" if needed.
* glsl: don't optimize program if MESA_GLSL=noptBrian Paul2009-04-071-1/+2
|
* mesa: in mesa_add_named_constant(), avoid adding duplicate constantsBrian Paul2009-04-031-7/+12
|
* mesa: fix parameter counting in ARB vertex/fragment program parsingBrian Paul2009-04-031-5/+2
| | | | Duplicated unnamed constants were getting counted more than once.
* mesa: replace >= with > when testing if we've exceeded max local paramsBrian Paul2009-04-031-2/+2
| | | | Now a program that uses 256 locals works as it should.
* mesa: for OPCODE_LIT, use _mesa_pow() instead of exp() and log()Brian Paul2009-04-031-2/+2
| | | | Also, s/pow/_mesa_pow/
* mesa: rename some gl_light fields to be clearerBrian Paul2009-04-031-2/+2
| | | | | EyeDirection -> SpotDirection _NormDirection -> _NormSpotDirection
* mesa: replace assertion with conditional in _mesa_opcode_string()Brian Paul2009-04-031-2/+4
|
* mesa: clean up formatting and use 'return' instead of 'break' consistantlyBrian Paul2009-04-031-51/+67
|
* glsl: fix segfault in linker when vertex or fragment shader was missingBrian Paul2009-04-021-9/+14
|
* glsl: implement compiling/linking of separate compilation unitsBrian Paul2009-04-015-23/+138
| | | | | | | | | A shader program may consist of multiple shaders (source code units). If we find there are unresolved functions after compiling the unit that defines main(), we'll concatenate all the respective vertex or fragment shaders then recompile. This isn't foolproof but should work in most cases.
* mesa: use correct tex unit lod bias for TXB instructionBrian Paul2009-04-011-2/+2
|
* glsl: fix texgen state variable tokens in emit_statevars()Brian Paul2009-04-011-8/+16
| | | | | This fixes broken variable indexing into the gl_Eye/ObjectPlaneS/T/R/Q arrays. See bug 20986.
* mesa: minor tweaks in append_token() for printing state var stringsBrian Paul2009-04-011-8/+8
|
* slang: ensure structure elements have their array length setAlan Hourihane2009-03-251-0/+1
|
* mesa: add new internal state var for window sizeBrian Paul2009-03-202-0/+14
| | | | Actually, window width - 1, height - 1
* mesa: linear scan register allocation for shader programsBrian Paul2009-03-201-21/+407
| | | | | | | | | | | | | | This is a check-point commit; not turned on yet. Use the linear scan register allocation algorithm to re-allocate temporary registers. This is done by computing the live intervals for registers and reallocating temps with that information. For some shaders this dramatically reduces the number of temp registers needed. For the time being we give up on a few cases such as relative-indexed temps and subroutine calls (but we inline most GLSL functions anyway).
* slang: Use _mesa_snprintf() wrapper.Michal Krol2009-03-201-1/+1
|
* slang: initialize the contextAlan Hourihane2009-03-191-2/+3
|
* slang: support uniform arraysAlan Hourihane2009-03-191-1/+13
|
* glsl: change GLSL #pragma initializationBrian Paul2009-03-193-14/+22
| | | | | | Initialize the shader's pragma settings before calling the compiler. Added pragma "Ignore" fields to allow overriding the #pragma directives found in shader source code.
* glsl: when debug pragma is on, emit comments about function calls/inlinesBrian Paul2009-03-193-0/+13
| | | | BTW, the debug pragma syntax is "#pragma debug(on)"
* slang: if/else/break & if/else/continue work for unrolled loopsAlan Hourihane2009-03-192-7/+38
|
* slang: redo the last commit for if/break & if/continue tests as itAlan Hourihane2009-03-181-1/+8
| | | | wasn't good enough for deeply nested if's.
* Revert "slang: if we detect an if/break or if/continue within a loop and we're"Alan Hourihane2009-03-181-28/+17
| | | | This reverts commit 752296b8f311c5e3844f3ce89d17ba57224ce5ba.
* slang: if we detect an if/break or if/continue within a loop and we'reAlan Hourihane2009-03-181-17/+28
| | | | trying to unroll, bail, and fallback to doing the real loop.
* mesa: update/fix doxygen commentsVinson Lee2009-03-172-2/+2
|
* mesa: Silence compiler warnings.Michal Krol2009-03-161-0/+1
|
* mesa: glUseProgram() debug code (disabled)Brian Paul2009-03-131-0/+11
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 21:42:48 +0000