aboutsummaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers
Commit message (Collapse)AuthorAgeFilesLines
* r600g: make r[67]00 not bail out on PRED_SETNE_INT.Mathias Fröhlich2011-10-231-0/+1
|
* r600g: make if's use PRED_SETNE_INT no matter what.Dave Airlie2011-10-222-1/+2
| | | | | | | This is more correct for TGSI if, and with native ints enabled it fixes 25 piglit fails. Signed-off-by: Dave Airlie <[email protected]>
* r600g: print inst in hex in dumpsDave Airlie2011-10-221-6/+6
| | | | | | since we have them as hex in the headers. Signed-off-by: Dave Airlie <[email protected]>
* r600g: drop specific i2f it should be a trans only op2.Dave Airlie2011-10-221-27/+1
| | | | Signed-off-by: Dave Airlie <[email protected]>
* nv50/ir: import SM4 converterChristoph Bumiller2011-10-214-4/+2513
|
* nouveau,nvc0: fix/improve handling of multiple constant buffersChristoph Bumiller2011-10-218-35/+111
|
* nvc0: add support for linear and buffer textures and RTsChristoph Bumiller2011-10-218-82/+213
|
* nvc0: add support for clip distance shader outputsChristoph Bumiller2011-10-216-16/+88
|
* nvc0: handle more query typesChristoph Bumiller2011-10-213-49/+204
|
* nvc0: fix location of the PrimitiveID outputChristoph Bumiller2011-10-211-4/+6
|
* nvc0: prevent VERTEXID/INSTANCEID from consuming input slotsChristoph Bumiller2011-10-211-3/+15
|
* nvc0: fixes for program tessellation parametersChristoph Bumiller2011-10-212-17/+47
|
* nv50,nvc0: reset per-instance state for inactive vertex elementsChristoph Bumiller2011-10-212-0/+6
|
* nv50,nvc0: reset base element in draw_arraysChristoph Bumiller2011-10-212-0/+11
| | | | It affects VERTEX_BUFFER_FIRST,COUNT submission, too.
* nvc0: apply first_layer offset to all resources with array_sizeChristoph Bumiller2011-10-211-3/+1
| | | | Makes CUBE arrays work with d3d1x.
* nvc0: emit state to allow GP to select the RT layerChristoph Bumiller2011-10-211-1/+4
|
* nvc0: validate GP samplers, texturesChristoph Bumiller2011-10-211-0/+2
|
* nvc0: fix clear_render_target/depth_stencil regionChristoph Bumiller2011-10-211-19/+21
| | | | In all 3 dimensions (now clearing all layers too).
* nvc0: fix assertion that immediate array buffer fitsChristoph Bumiller2011-10-211-1/+1
|
* nv50/ir: use RDSV to fetch FrontFacing before loweringChristoph Bumiller2011-10-214-13/+33
|
* nv50/ir: fix textureGrad with offsets and in non-FPsChristoph Bumiller2011-10-214-3/+14
|
* nv50/ir: add wrap mode for shift operationsChristoph Bumiller2011-10-213-2/+5
| | | | D3D1x specifies that only the low 5 bit of the shift are used.
* nv50/ir: initialize RelocInfo to 0Christoph Bumiller2011-10-211-0/+2
|
* nvc0/ir: fix emission of cvt when register and type size differChristoph Bumiller2011-10-211-2/+3
|
* nv50/ir: fix argument count for CUBE_ARRAY texture targetChristoph Bumiller2011-10-212-4/+4
|
* nvc0/ir: GP emit address must end up in $r0Christoph Bumiller2011-10-211-0/+4
|
* nvc0/ir: TXQ requires different lowering from normal TEXChristoph Bumiller2011-10-211-1/+10
|
* nv50/ir: initialize default prog_info values for GP,TPChristoph Bumiller2011-10-211-0/+10
|
* nv50/ir: fix memory value equality checkChristoph Bumiller2011-10-211-1/+1
|
* nv50/ir: fix leak in removal of graph rootChristoph Bumiller2011-10-211-5/+5
|
* nvc0/ir: handle levelZero modifier in TEX emissionChristoph Bumiller2011-10-211-1/+9
|
* nvc0/ir: fix lowering of DIV F32Christoph Bumiller2011-10-211-0/+1
|
* nvc0/ir: fix xy confusion typo in readTessCoordChristoph Bumiller2011-10-211-1/+1
|
* r600g: cleanup r600_reset_blittable_to_compressedMarek Olšák2011-10-211-6/+3
|
* r300g: don't return NULL in resource_from_handle if the resource is too smallMarek Olšák2011-10-204-30/+22
| | | | | | | | | The DDX may allocate a buffer with a too small size. Instead of failing, let's pretend everything's alright. Such bugs should be fixed in the DDX, of course. NOTE: This is a candidate for the stable branches.
* softpipe: implement/fix TIMESTAMP queriesChristoph Bumiller2011-10-201-3/+8
|
* softpipe: remove unused vars in softpipe_clear()Brian Paul2011-10-181-3/+0
|
* svga: Plug a fence leakThomas Hellstrom2011-10-181-3/+3
| | | | | Signed-off-by: Thomas Hellstrom <[email protected]> Reviewed-by: Jakob Bornecrantz <[email protected]>
* i915g: Use the right shader limits.Stéphane Marchesin2011-10-173-5/+9
|
* i915g: Add TODO.Stéphane Marchesin2011-10-171-0/+2
|
* r600g: Use the bitfield define matching the register it is used for.Mathias Fröhlich2011-10-181-2/+2
| | | | Fix a typo that should result in the same code.
* r300/compiler: Try to eliminate REPL_ALPHA instructionsTom Stellard2011-10-161-33/+57
| | | | | | | Scalar instruction that need to write to the xyz components of a register must reserve the RGB instruction slot for a REPL_ALPHA instruction. With this commit, the scheduler will attempt to free the RGB slot by moving the write to the w component of a register.
* r300/compiler: More agressive RGB to Alpha conversionsTom Stellard2011-10-161-8/+10
|
* r300/compiler: Only convert RGB->Alpha when optimizations are enabledTom Stellard2011-10-162-1/+8
|
* llvmpipe: Use lp_build_ifloor_fract for exp2 calculation.José Fonseca2011-10-161-0/+10
| | | | | | | Instead of separate ifloor / fract calls. No change for SSE4.1 code, but less FP<->SI conversions on non SSE4.1 systems.
* Revert "r300g: fix rendering with a non-zero index bias in ↵Marek Olšák2011-10-161-3/+1
| | | | | | | | draw_elements_immediate" This reverts commit 5506f6ef966b8883e575a3f60ce96ad42ee6ffd2. It breaks more things than it fixes.
* nv50,nvc0: extend formats table for integer formatsChristoph Bumiller2011-10-154-1135/+731
|
* r300/compiler: Enable reg rename pass on r500 and run it before optimizationsTom Stellard2011-10-141-4/+1
| | | | | The scheduler and the register allocator are now smart enough to handle it.
* r300/compiler: Implement the texture semaphoreTom Stellard2011-10-146-49/+308
| | | | | | | | | | | The texture semaphore allows for prefetching of texture data. On my RV515, this increases the FPS of Lightsmark by 33% (This is with the reg_rename pass enabled, which is enabled in the next commit). There is a new env variable now called RADEON_TEX_GROUP, which allows you to specify the maximum number of texture lookups to do at once. The default is 8, but different values could produce better results for various application / card combinations.
* r300/compiler: Don't pair output writes with GPR writes in the schedulerTom Stellard2011-10-141-0/+7
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-14 00:30:07 +0000