summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* gallium: fix broken hashing for vertex translationBrian Paul2008-04-231-5/+13
| | | | | | | It seems we get hash collisions fairly easily and the code as it was didn't deal with that properly. I think we need a simpler hashing interface...
* draw: remove stupid debug (Keith)Keith Whitwell2008-04-231-1/+1
|
* draw: add missing break (Jakob)Keith Whitwell2008-04-231-1/+2
|
* fix non-i386 buildsBen Skeggs2008-04-231-0/+2
|
* gallium: added a flushing_vcache flag, test in draw_do_flush()Brian Paul2008-04-223-1/+4
| | | | Fixes broken polygon stipple, aaline, aapoint stages
* gallium: fix bad logic in bind_pstip_fragment_shader(): use &&, not ||Brian Paul2008-04-221-1/+1
|
* In case the 'func' is deleting the state move the iterator beforeZack Rusin2008-04-221-1/+1
| | | | calling it.
* Fix a crash. Rasterizer can be nullZack Rusin2008-04-221-2/+3
|
* Cache translate's structs for emits and fetches.Zack Rusin2008-04-222-14/+114
| | | | Results in a fair speed improvement.
* Fix reporting of clipped vertices.Zack Rusin2008-04-221-3/+2
|
* gallium: setup an identity viewportBrian Paul2008-04-221-4/+6
| | | | This fixes broken blits.
* gallium: update comment about bypass clipping/viewportBrian Paul2008-04-221-1/+1
|
* gallium: setup an identity viewportBrian Paul2008-04-221-30/+5
|
* gallium: move the vertex print/debug codeBrian Paul2008-04-221-8/+6
|
* gallium: remove assertionsBrian Paul2008-04-221-2/+0
| | | | | It's possible the current vs/fs is null when cso_save_vertex/fragment_shader() is called.
* Revert "softpipe: squash warning"Keith Whitwell2008-04-221-0/+1
| | | | | | This reverts commit 1f0f029ba6f22ef4ada01fcdc153da91571a7963. Which broke the debug build.
* draw: allow drivers to query pipeline state more easilyKeith Whitwell2008-04-227-23/+58
| | | | | | Also, provide a separate flag to say whether the driver can handle clipping/rhw tasks, in addition to the API flag which indicates they have already been done.
* softpipe: squash warningKeith Whitwell2008-04-221-1/+0
|
* cso: fix vs/fs confusionKeith Whitwell2008-04-221-1/+1
|
* softpipe: do something sensible on an error path, squash warningKeith Whitwell2008-04-221-0/+2
|
* pipebuffer: Temporarily allow simultaneous CPU writes.José Fonseca2008-04-221-1/+17
| | | | Also, fast path for re-fencing the same buffer multiple times with the same fence.
* draw: disable broken edgeflag code - didn't work & was killing performanceKeith Whitwell2008-04-221-6/+7
|
* xlib: shortcircuit no-op makecurrentKeith Whitwell2008-04-221-0/+7
|
* xlib: implement SP_NO_RAST env varKeith Whitwell2008-04-221-1/+14
|
* softpipe: implement SP_NO_RAST env varKeith Whitwell2008-04-224-0/+18
|
* softpipe: make NUM_ENTRIES 32 so that div/mods are easierKeith Whitwell2008-04-221-1/+1
|
* cso: Fix build on Win32.Michal Krol2008-04-221-2/+4
|
* make the api consistent (all set functions return pipe_error status)Zack Rusin2008-04-212-30/+35
|
* handle some of the possible allocation failures within the hash itselfZack Rusin2008-04-212-4/+49
|
* initial stab at error handling in csoZack Rusin2008-04-212-52/+80
|
* cso: disable not-really-working cso_set_*_shader() funcsKeith Whitwell2008-04-211-1/+8
|
* draw: squash a couple of memory leaksKeith Whitwell2008-04-212-0/+8
|
* use cso fs/vs handle functionsKeith Whitwell2008-04-212-4/+4
|
* cso: provide functions to bind fs/vs handles directlyKeith Whitwell2008-04-212-8/+28
|
* Use llvm 2.3 (2.2 won't work because of a lot of problems, e.g.Zack Rusin2008-04-218-78/+79
| | | | lack of constant vectors handling in execution engine)
* rtasm: debug some missing funcsKeith Whitwell2008-04-211-0/+3
|
* rtasm: quieten sse_enabled debugKeith Whitwell2008-04-211-1/+10
|
* rtasm: clean up debug dumping a littleKeith Whitwell2008-04-211-146/+140
|
* cso: propogate one easy error - many moreKeith Whitwell2008-04-211-0/+2
|
* translate: fail on x86 rtasm failKeith Whitwell2008-04-211-0/+5
|
* rtasm: propogate errors in x86 emitKeith Whitwell2008-04-212-5/+30
|
* actually write the resultsZack Rusin2008-04-211-8/+9
|
* draw: more propogation -- pstipple stage.Keith Whitwell2008-04-212-11/+33
|
* draw: propogate lots of errorsKeith Whitwell2008-04-2117-35/+120
|
* draw: propogate errors out of aapoint stageKeith Whitwell2008-04-213-12/+38
|
* draw: propogate errors out of aaline stageKeith Whitwell2008-04-212-12/+48
|
* draw: consolidate all the passthrough line/tri/point funcsKeith Whitwell2008-04-2116-241/+177
|
* gallium: Hash the fragment shader tokens, instead of pipe_shader_state.José Fonseca2008-04-221-6/+18
| | | | PS: pipe_shader_state should probably go away now that it is reduced to a single pointer.
* gallium: Include dependent header.José Fonseca2008-04-221-0/+2
|
* gallium: Set all state via cso_context in blit/gen_mipmap utils.José Fonseca2008-04-212-7/+15
| | | | | | | | cso_restore_* functions are implemented on top of cso_set_*, therefore they require full knowledge of the current pipe state to work correctly. Directly calling pipe's set_*_state functions will lead to undefined state. Also save and restore shaders.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-22 20:50:35 +0000