summaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers
Commit message (Collapse)AuthorAgeFilesLines
* llvmpipe: Don't mention deprecated drivers=trace option.José Fonseca2010-03-011-2/+2
| | | | Now trace always built.
* r300g: atomize texture and sampler statesMarek Olšák2010-02-289-135/+153
|
* r300g: decouple vertex stream setup (PSC) and VS output mapping (VAP_OUT)Marek Olšák2010-02-287-66/+101
| | | | | | | | | | | Formerly known as vertex_format_state. These two are completely unrelated when using HWTCL and decoupling them makes the design less SWTCL-centric. When bypass_vs_clip_and_viewport gets removed, the PSC setup will no longer be a derived state. This change shouldn't make unbreaking SWTCL harder.
* i965g: Fix texture creationJakob Bornecrantz2010-02-281-1/+1
|
* r300g: put validating buffers after flushingMarek Olšák2010-02-283-53/+37
| | | | Also cleaning up the nasty validation process.
* r300g: use the atom size directly during emissionMarek Olšák2010-02-284-30/+31
|
* r300g: atomize invalidation of texture cachesMarek Olšák2010-02-285-8/+11
|
* r300g: atomize PVS flushMarek Olšák2010-02-285-27/+29
| | | | The first non-state atom. It's better and cleaner to have it.
* r300g: add size parameter to the atom emit functionsMarek Olšák2010-02-285-28/+47
| | | | | Maintaining a closer relationship between the atom size and what's passed in BEGIN_CS.
* r300g: move the emission of GA_POINT_MINMAX into emit_fb_stateMarek Olšák2010-02-273-23/+7
| | | | | The only practical limits are the ones derived from the currently-set framebuffer state.
* r300g: put the emission of R300_US_OUT_FMT_UNUSED backMarek Olšák2010-02-272-2/+7
| | | | It wasn't such a good idea to remove it. :/
* r300g: mark rasterizer_state as dirty only when it's not NULLMarek Olšák2010-02-271-1/+1
|
* r300g: always emit the correct max vertex index to avoid DRM errorsMarek Olšák2010-02-273-2/+13
| | | | Fixing bizarre reports that a vertex buffer is not large enough.
* r300g: remove pointless "while"Marek Olšák2010-02-271-1/+1
|
* nv40: fix include of nv04_surface_2d.hChristoph Bumiller2010-02-261-1/+1
|
* gallium/nouveau: clean-up #includesBrian Paul2010-02-265-22/+22
|
* softpipe: fix up #includesBrian Paul2010-02-262-6/+8
|
* gallium/trace: remove unneeded path prefixes on #includesBrian Paul2010-02-261-5/+5
|
* gallium/identity: remove unneeded path prefixes on #includesBrian Paul2010-02-261-5/+5
|
* util: Code generate functions to pack and unpack a single pixel.José Fonseca2010-02-262-5/+6
| | | | | | | | | | | | | | | | | | | | | Should work correctly for all pixel formats except SRGB formats. Generated code made much simpler by defining the pixel format as a C structure. For example this is the generated structure for PIPE_FORMAT_B6UG5SR5S_NORM: union util_format_b6ug5sr5s_norm { uint16_t value; struct { int r:5; int g:5; unsigned b:6; } chan; }; Not used everywhere yet because it seems compiled code is slower than bitshift arithmetic by some misterious reason. So we should generate bitshift arithmetic at least for the simple UNORM pixel formats.
* util: Factor some code into u_format_parse.pyJosé Fonseca2010-02-261-2/+2
|
* util: Use python names consistent with u_format.hJosé Fonseca2010-02-261-35/+35
|
* llvmpipe: Fix lp_tile_r5g6b5_unorm_write_4ub.José Fonseca2010-02-261-1/+2
| | | | Was using 32bit integer pointers instead of 16bits.
* llvmpipe: Fix scons dependencies.José Fonseca2010-02-261-0/+6
|
* r300g: Atomize vertex shader.Corbin Simpson2010-02-248-60/+58
|
* r300g: Move CALLOCs to correct place.Corbin Simpson2010-02-241-8/+7
|
* r300g: prevent CS overflow when emitting the draw packetsMarek Olšák2010-02-243-18/+55
| | | | Signed-off-by: Corbin Simpson <[email protected]>
* llvmpipe: clean up rasterization threads upon context destructionBrian Paul2010-02-242-0/+18
| | | | Fixes glean hang, bug 26536.
* r300g: remove invalid .PHONY lineBrian Paul2010-02-241-2/+0
| | | | | The $(COMPILER_ARCHIVE) target is not a phony target. This solves the unconditional re-making of libr300.a
* llvmpipe: more lp_rasterizer_task parameter passingBrian Paul2010-02-244-114/+89
|
* llvmpipe: pass fewer parameters to rasterization functionsBrian Paul2010-02-243-15/+9
|
* llvmpipe: added some assertionsBrian Paul2010-02-241-0/+5
|
* llvmpipe: SRGB textures not supported yetBrian Paul2010-02-241-1/+0
|
* llvmpipe: vertex shader textures not supported yetBrian Paul2010-02-241-1/+1
|
* r300: Update for UTIL_FORMAT_LAYOUT_xxx changes.José Fonseca2010-02-242-16/+4
|
* llvmpipe: Update for UTIL_FORMAT_LAYOUT_xxx changes.José Fonseca2010-02-242-7/+5
|
* i915g: fix a commentBrian Paul2010-02-241-1/+1
|
* llvmpipe: Fix breakeage.José Fonseca2010-02-241-24/+13
| | | | | | | Due to the u_format.csv changes the unrolled code generator was being used for formats it didn't support. Bring some bugfixes from u_format_access.py too.
* llvmpipe: Handle correctly 24bit formats.José Fonseca2010-02-241-44/+46
|
* llvmpipe: Kill unused UTIL_FORMAT_LAYOUT_SCALAR.José Fonseca2010-02-241-4/+2
|
* llvmpipe: add lp_tile_image.c to sourcesBrian Paul2010-02-232-0/+2
|
* llvmpipe: whole image tile/untile functionsBrian Paul2010-02-232-0/+183
|
* llvmpipe: remove Z16 format override hackBrian Paul2010-02-231-6/+0
|
* llvmpipe: Z16 format is not supportedBrian Paul2010-02-231-0/+4
|
* llvmpipe: added assertions to verify that scene is emptyBrian Paul2010-02-232-0/+5
|
* llvmpipe: added some assertionsBrian Paul2010-02-231-0/+3
|
* r300g: fix texture swizzling for the SRGB formatsMarek Olšák2010-02-241-4/+20
|
* i965g: Add fallthrough comments in switch statement.Vinson Lee2010-02-221-3/+3
| | | | Silences warnings about missing break statements in static analysis.
* r300g: fix draw_elements for "start" != 0Marek Olšák2010-02-231-7/+2
| | | | Reported-by: Andre Maasikas <[email protected]>
* svga: Upload user buffers only once.José Fonseca2010-02-224-29/+46
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 18:36:48 +0000