summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* tgsi: Cleanup dot-product opcodes in interpreter.Michal Krol2010-01-081-97/+130
|
* nv20: Fix build for the latest nouveau_class.h changes.Francisco Jerez2010-01-081-2/+2
|
* scons: Don't build xlib when dri is enabled.José Fonseca2010-01-081-0/+4
| | | | Hopefully adddresses fdo 25828.
* lvmpipe: Initialize all coordinates.José Fonseca2010-01-081-0/+3
| | | | Fixes assertion failure with fp-incomplete-tex (fdo 24298).
* scons: Output a meaningful message when xlib libGL.so can't be built.José Fonseca2010-01-081-45/+57
|
* st/xorg: remove unused xorg_exa_get_pixmap_handle functionKeith Whitwell2010-01-072-30/+0
|
* nv50: preallocate TEMPs written first time in a subroutineChristoph Bumiller2010-01-071-2/+10
| | | | | Otherwise we risk overwriting them with temporary GPRs if they're not used immediately after the CALL.
* nv50: handle TGSI_OPCODE_SAD,UADDChristoph Bumiller2010-01-071-0/+96
|
* nv50: handle TGSI_OPCODE_IMAX,IMIN,UMAX,UMINChristoph Bumiller2010-01-071-10/+49
|
* nv50: handle integer SET operationsChristoph Bumiller2010-01-071-21/+34
|
* nv50: handle TGSI_OPCODE_SHL,ISHR,USHRChristoph Bumiller2010-01-071-0/+42
|
* nv50: handle TGSI_OPCODE_F2I,F2U,I2F,U2F plus src modsChristoph Bumiller2010-01-071-87/+131
|
* tgsi: add assertion to verify legal register fileBrian Paul2010-01-071-0/+1
| | | | | | This assertion fails with piglit glsl-vs-mov-after-deref test because we're double freeing the memory. It seems there's some confusion between what's placed in the hash table and what isn't.
* tgsi: fix incorrectly placed braces, add more braces to be clearBrian Paul2010-01-071-2/+4
|
* tgsi: move register checking code before check_register_usage()Brian Paul2010-01-071-4/+4
| | | | | | check_register_usage() frees the scan_register *reg data so we were reading from freed memory. This fixes a valgrind error found with piglit's glsl-vs-mov-after-deref test.
* gallium/util: remove useless assertionsBrian Paul2010-01-071-6/+0
|
* gallium/util: update comment: s/src_pitch/src_strideBrian Paul2010-01-071-1/+1
|
* gallium/util: make src_y unsigned like the other x/y paramsBrian Paul2010-01-072-2/+2
|
* llvmpipe: Remove TGSI sampler pointers from JIT context too.José Fonseca2010-01-073-19/+9
|
* llvmpipe: Remove more loose ends of TGSI exec sampling.José Fonseca2010-01-076-127/+4
|
* i965g: updates for draw retvalKeith Whitwell2010-01-071-11/+9
|
* llvmpipe: Axe texture sampling code inherited from softpipe.José Fonseca2010-01-075-1727/+0
| | | | | | | Was used only as a reference, since texture sampling is now code generated. Already axed in the lp-binning branch too. This fixes the llvmpipe build after recent sampling changes.
* gallium: Fix texture sampling with explicit LOD in softpipe.Michal Krol2010-01-075-53/+113
|
* gallium: Pass per-element (not per-quad) LOD bias values down to texture ↵Michal Krol2010-01-075-98/+123
| | | | sampler.
* tgsi: Cleanup exec_tex().Michal Krol2010-01-071-49/+40
|
* svga: Fix fs key size computation and key comparison.Michal Krol2010-01-072-5/+9
| | | | | This also allows us to have texture_target back as a bitfield and save us a few bytes.
* r300g: add back-face color VS outputsMarek Olšák2010-01-062-7/+37
| | | | | Required for two-sided lighting. BTW we are approaching softpipe-level quality. ;)
* r300g: fix SWTCL stream locations of texture coordinatesMarek Olšák2010-01-061-1/+1
| | | | It might have caused hardlocks when TCL was bypassed, not sure.
* r300g: if no DS buffer is set, disable reading from and writing to itMarek Olšák2010-01-062-2/+10
| | | | It fixes another "Bad CS" issue.
* r300g: disable the rasterization of WPOS if it's unused by the FSMarek Olšák2010-01-064-34/+58
|
* r300g: add WPOSMarek Olšák2010-01-066-3/+124
|
* r300g: consolidate paramsMarek Olšák2010-01-061-3/+6
|
* r300g: fix outputting depth in the FSMarek Olšák2010-01-062-3/+18
|
* r300g: optimize blending by conditionally disabling reads from the colorbufferMarek Olšák2010-01-062-3/+30
|
* r300g: optimize blending by discarding pixels that don't change the colorbufferMarek Olšák2010-01-062-3/+162
| | | | Let's get some performance out of the blender.
* r300g: fix blending when SRC_ALPHA_SATURATE is usedMarek Olšák2010-01-061-2/+7
|
* util/blitter: allow NULL CSOs to be savedMarek Olšák2010-01-061-10/+17
|
* softpipe: fix draw return valueKeith Whitwell2010-01-061-1/+1
|
* r300g: add missing is_anistropic paramBrian Paul2010-01-061-1/+2
|
* svga: make texture_target a ubyte, not a bitfieldBrian Paul2010-01-061-1/+1
| | | | | gcc won't let us take the address of a bitfield in the svga_fs_key_size() function.
* gallium/docs: finish up description of pipe_rasterizer_state fieldsBrian Paul2010-01-061-29/+94
|
* util: Fix format descriptor table entries.Michal Krol2010-01-061-11/+11
|
* pipe_sampler_state::compare_mode is not a boolean enable flag.Michal Krol2010-01-066-9/+9
| | | | It's a 1-bit enum.
* util: Handle the remaining format cases in pipe_tile_raw_to_rgba().Michal Krol2010-01-061-23/+4
|
* Make sure we use only signed/unsigned ints with bitfields.Michal Krol2010-01-061-14/+12
| | | | Seems to be the only way to stay fully portable.
* i965g: fix invalid assertion in emit_xpd(), againBrian Paul2010-01-061-1/+1
|
* Merge remote branch 'origin/mesa_7_7_branch'José Fonseca2010-01-066-34/+42
|\ | | | | | | | | | | | | Conflicts: configs/default src/gallium/auxiliary/pipebuffer/pb_buffer_fenced.c src/mesa/main/version.h
| * pipebuffer: Multi-threading fixes for fencing.José Fonseca2010-01-061-111/+170
| | | | | | | | | | | | | | | | | | I had this patch on my hard drive for long time. It doesn't fully address SVGA multi-threading issues, but causes no regressions, so decided to commit while it still applies cleanly. Attention: merging this into master will cause issues due to recent changes in reference counting to fix strict aliasing rules violation.
| * svga: Remove stale references to delete shader results.José Fonseca2010-01-052-0/+14
| | | | | | | | | | To ensure that a new result that happens to have the same address of the old one will be detected as a change.
| * svga: Rebind shaders when tokens change.José Fonseca2010-01-053-18/+10
| | | | | | | | | | | | | | | | | | | | Even shader ID doesn't change. Otherwise the token change is not effective. This reverts regression from commit cdb445f3a9285e2d8f042a07021ade78b94e0156. Regression could be seen in Quake3's loading screen -- the upper right corner of the screen would be black.