summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Merge texmem-0-3-branch.Keith Whitwell2006-11-0120-518/+1735
|
* Add LIT instruction to fragment program.Jerome Glisse2006-11-012-17/+88
|
* Import texmem i915 driver to its new location as i915tex.Keith Whitwell2006-11-0173-0/+25933
|
* Add CMPH instruction to fragprogJerome Glisse2006-11-011-0/+2
| | | | | CMPH a0, a1, a2 -> if a2 > 0.5 return a1 else return a0 Guessed by examinating LIT instruction handling of FGLRX.
* Patch from <[email protected]> -- pass program stringKeith Whitwell2006-11-011-0/+2
| | | | notification callback through to tnl/ module, fixes glean crash.
* change DEBUG_FRAG conditionsBrian Paul2006-11-011-79/+82
|
* fix some bugs in interpolate_specular()Brian Paul2006-11-011-23/+29
|
* Added a few new ctx->Const. fields for shader-related limits.Brian Paul2006-11-014-16/+30
|
* make use of Parameters->StateFlags in _swrast_update_fragment_program()Brian Paul2006-10-311-7/+4
|
* minor clean-ups, simplifications and new commentsBrian Paul2006-10-311-26/+21
|
* fix MINGW32 problemsBrian Paul2006-10-311-2/+4
|
* We need to call _mesa_load_state_parameters() whenever a fragment programBrian Paul2006-10-312-6/+15
| | | | | | is enabled and we've changed any GL state that might be referenced by a fragment program (such as state.fog.params). Do this during swrast validation, not in _swrast_exec_fragment_program().
* signal _NEW_LIGHT in _tnl_copy_to_current()Brian Paul2006-10-311-0/+2
|
* In TEX instructions, use precomputed lambda only if using fragment.texcoord[k]Brian Paul2006-10-311-34/+33
| | | | | to sample texture[k]. Use zero otherwise. Not foolproof, but a good compromise.
* For TEX instructions use lambda=0. When sampling from texture unit K weBrian Paul2006-10-311-44/+71
| | | | | | | | | | were using the partial derivatives of texcoord[K] but the coordinate used for texture sampling may be something totally different (and texcoord[K] might not be a real texture coord at all). Net result was a bogus LOD is sometimes used, often resulting in using the smallest mipmap level (a constant color). Just use zero for now (undef LAMBDA_ZERO to override). Plus, some additional debug code.
* Fix bug #8799.Ian Romanick2006-10-302-22/+8
| | | | | | Properly resize bith the drawable and the readable in mgaUpdateRects. Eliminate the use of the deprecated GetBufferSize interface. Bump driver date.
* Refactor mgaXMesaSetFrontClipRects and mgaXMesaSetBackClipRects.Ian Romanick2006-10-301-33/+8
| | | | | | Combine mgaXMesaSetFrontClipRects and mgaXMesaSetBackClipRects into a single new function called mga_set_cliprects. This enables a small refactor in mgaDDDrawBuffer.
* Get DRI drawable from framebuffer object.Ian Romanick2006-10-303-10/+2
| | | | | | | | | | In the software rasterization fallbacks, get the __DRIdrawablePrivate pointer from the driRenderbuffer that is passed in. This eliminates the need for the mesa_drawable pointer in MGA context structure. This partially fixes bug #8799. wincopy no longer produces a black window for the destination window, but it does produce incorrect results if the destination window is resized.
* code tweaks, remove old commentsBrian Paul2006-10-301-65/+16
|
* fix _mesa_problem() textBrian Paul2006-10-301-2/+2
|
* Move gl_vertex_program_machine struct out of mtypes.h and put into nvvertexec.h.Brian Paul2006-10-305-154/+177
| | | | | Massage nvvertexec.c code to work more like s_nvfragprog.c - another step toward unifying vertex/fragment program execution.
* remove debug codeBrian Paul2006-10-301-7/+0
|
* Rather than always allocating 100 instructions, put instructions into aBrian Paul2006-10-291-18/+42
| | | | | temporary buffer, then copy into a malloc'd array of the correct size. According to Glean texCombine, we never need more than 21 instructions.
* New _mesa_exec_vertex_state_program() function. Start of some re-org.Brian Paul2006-10-293-9/+28
|
* fix wrong inequalityBrian Paul2006-10-291-1/+1
|
* Change _mesa_init_instruction() to initialize an array of instructions.Brian Paul2006-10-297-27/+32
|
* Program instructions are initialized before parsing so no need to setBrian Paul2006-10-291-7/+1
| | | | | | Abs, NegateAbs, CondMask and CondSwizzle fields to defaults here. CondMask was getting mis-set. Removing the incorrect assignment improves fragment program speed by 15%.
* Added PROGRAM_CONSTANT to get_register_pointer().Brian Paul2006-10-281-0/+15
| | | | More DEBUG_FRAG code.
* special case END in _mesa_print_instruction()Brian Paul2006-10-281-0/+3
|
* Implement _mesa_append_fog_code() for fragment program fog options.Brian Paul2006-10-282-19/+178
|
* If DEBUG, check that all array indices really do fall in [start,end] inBrian Paul2006-10-261-1/+19
| | | | the DrawRangeElements() call. Warn the user if that's not true.
* remove code for inserting mvp transform into position-invariant vertex progs ↵Roland Scheidegger2006-10-261-96/+7
| | | | and instead use _mesa_insert_mvp_code().
* Previously, if a fragment program referenced fragment.fogcoord but theBrian Paul2006-10-251-2/+3
| | | | | program didn't use the ARB_fog_linear option, the fragment program's fragment.fogcoord register wasn't loaded properly.
* change some bogus comments regarding the base e exponential function used ↵Roland Scheidegger2006-10-252-7/+6
| | | | for fog in vertex programs.
* fix (per-vertex) fog when using ARB_vp by incorporating fog factor ↵Roland Scheidegger2006-10-246-67/+153
| | | | computation into the vertex program (not yet fixed for swtnl). Simplify (and correct) the VTX_TCL_OUTPUT_VTXFMT handling when using vertex programs, turns out it's solely driven by the needs of the past-vertex stage of the pipeline, this should fix lockups with ill-specified applications using vertex programs (for instance applications enabling fog but not writing to fog coord output will now get (conformant) undefined results instead of lockups).
* Add missing call to glutInit.Ian Romanick2006-10-241-0/+1
|
* s/GLchan/GLubyte/Brian Paul2006-10-241-4/+4
|
* whitespace clean-upBrian Paul2006-10-242-4/+4
|
* Emit cliprects in the userspace driver as required, rather thanKeith Whitwell2006-10-2310-66/+81
| | | | | | | | passing them to the kernel. This works because all drawing commands in the 965 driver are emitted with the lock held and the batchbuffer is always flushed prior to releasing the lock. This allows multiple cliprects to be dealt with, without replaying entire batchbuffers and redundantly re-emitting state.
* assorted clean-upsBrian Paul2006-10-221-30/+24
|
* don't save color/z if fragment is killedBrian Paul2006-10-221-18/+20
|
* add minstall script to MAIN_FILES (bug 8720)Brian Paul2006-10-221-0/+1
|
* minor tweaks good for about 12% faster executionBrian Paul2006-10-221-66/+71
|
* silence used var warningsBrian Paul2006-10-221-1/+2
|
* Color clamping fixes.Brian Paul2006-10-226-8/+52
|
* move/improve comments, clean up code in a few placesBrian Paul2006-10-211-21/+16
|
* fix another logic error in computing deferredTextureBrian Paul2006-10-211-5/+9
|
* For deferredTexture, need to check if the fragment program/shader writesBrian Paul2006-10-211-6/+31
| | | | the fragment Z.
* move interpolate_varying() callBrian Paul2006-10-211-4/+2
|
* don't use params passed to Clear()Brian Paul2006-10-211-2/+6
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-30 07:25:25 +0000