summaryrefslogtreecommitdiffstats
path: root/src/mesa
Commit message (Collapse)AuthorAgeFilesLines
* mesa: code to unpack RGBA as uintsBrian Paul2010-11-032-0/+787
|
* intel: Remove leftover dri1 locking fields in the context.Eric Anholt2010-11-031-3/+0
|
* intel: Remove duplicated teximage miptree to object miptree promotion.Eric Anholt2010-11-031-15/+0
| | | | | intel_finalize_mipmap_tree() does this optimization too, just more aggressively.
* intel: Avoid taking logbase2 of several things that we max.Eric Anholt2010-11-031-5/+1
| | | | | | logbase2(max(width, height, depth)) == max(logbase2(width), logbase2(height), logbase2(depth)), but in 60 bytes less code.
* i965: Remove dead intel_structs.h file.Eric Anholt2010-11-032-264/+0
|
* intel: Remove the magic unaligned memcpy code.Eric Anholt2010-11-031-89/+0
| | | | | | | | In testing on Ironlake, the histogram of clocks/pixel results for the system memcpy and magic unaligned memcpy show no noticeable difference (and no statistically significant difference with the 5510 samples taken, though the stddev is large due to what looks like the cache effects from the different texture sizes used).
* intel: Annotate debug printout checks with unlikely().Eric Anholt2010-11-0334-129/+116
| | | | | | | This provides the optimizer with hints about code hotness, which we're quite certain about for debug printouts (or, rather, while we developers often hit the checks for debug printouts, we don't care about performance while doing so).
* mesa: silence new warnings in texobj.cBrian Paul2010-11-021-4/+2
| | | | | | Silences warning such as: main/texobj.c:442:40: warning: ISO C99 requires rest arguments to be used main/texobj.c:498:58: warning: ISO C99 requires rest arguments to be used
* savage: Remove unnecessary header.Vinson Lee2010-11-021-1/+0
|
* intel: For batch, use GTT mapping instead of writing to a malloc and copying.Eric Anholt2010-11-022-13/+9
| | | | | No measurable performance difference on cairo-perf-trace, but simplifies the code and should have cache benefit in general.
* mesa: Don't compute an unused texture completeness debug string.Eric Anholt2010-11-021-9/+12
| | | | | This showed up at about 1% on cairo-gl firefox-talos-gfx, where glClear() is called while a texture is incomplete.
* st/mesa: Reset the index buffer before destroying the pipe context.Tilman Sauerbeck2010-11-021-0/+2
| | | | Signed-off-by: Tilman Sauerbeck <[email protected]>
* mesa: Fix C++ includes in sampler.cppChad Versace2010-11-021-4/+4
| | | | | | | | | Some C++ header files were included in an extern "C" block. When building with Clang, this caused the build to fail due to namespace errors. (GCC did not report any errors.) Reviewed-by: Ian Romanick <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* st/mesa: unbind constant buffer when not in useKeith Whitwell2010-11-021-1/+4
| | | | | | | | | | | Important as more constant buffers per shader start to get used. Fix up r600 (tested) and nv50 (untested) to cope with this. Drivers previously didn't see unbinds of constant buffers often or ever, so this isn't always dealt with cleanly. For r600 just return and keep the reference. Will try to do better in a followup change.
* mesa: fix aux/accum comment and error message mixupsBrian Paul2010-11-021-2/+2
|
* mesa: remove always-false conditional in check_compatible()Brian Paul2010-11-021-3/+0
| | | | The two gl_config pointers can never be equal.
* dri/util: add a bunch of commentsBrian Paul2010-11-021-0/+24
|
* mesa: move the gl_config struct declarationBrian Paul2010-11-021-54/+62
| | | | | It was in the middle of the lighting-related structures before. Also add some info about field sizes in this structure.
* mesa: use GLubyte for edge flag arraysBrian Paul2010-11-021-2/+2
| | | | Fixes http://bugs.freedesktop.org/show_bug.cgi?id=31310
* i965: refresh wm push constant also for BRW_NEW_FRAMENT_PROGRAM on gen6Zhenyu Wang2010-11-021-1/+1
| | | | | | Fix compiz crash. https://bugs.freedesktop.org/show_bug.cgi?id=31124
* mesa: Allow contexts of different APIs to coexist.Chia-I Wu2010-11-025-159/+132
| | | | | | | | | | | | | | | | | | This effectively redoes 1741ddb747ca0be284315adb4b6fe67ddf292d03 in a way that allows contexts of different APIs to coexist. First, the changes to the remap table are reverted. The remap table (driDispatchRemapTable) is always initialized in the same way regardless of the context API. es_generator.py is updated to use a local remap table, whose sole purpose is to help initialize its dispatch table. The local remap table and the global one are always different, as they use different glapidispatch.h. But the dispatch tables initialized by both remap tables are always compatible with glapi (libGL.so). Finally, the semantics of one_time_init are changed to per-api one-time initialization.
* mesa: Select FEATURE_remap_table when multiple APIs are enabled.Chia-I Wu2010-11-023-3/+5
| | | | | | Core mesa should query glapi for the positions of the functions in _glapi_table when multiple APIs are supported. It does not know which glapitable.h glapi used.
* r300/compiler: Don't track readers into an IF block.Tom Stellard2010-11-011-0/+3
| | | | | This makes rc_get_readers_normal() more conservative than it needs to be, but it fixes some incorrect behavior in the optimization passes.
* scons: Fix MinGW cross-compilation.José Fonseca2010-11-011-70/+71
|
* scons: Revamp how to specify targets to build.José Fonseca2010-11-011-314/+315
| | | | | | | | | | | | | | | | | | | | | | Use scons target and dependency system instead of ad-hoc options. Now is simply a matter of naming what to build. For example: scons libgl-xlib scons libgl-gdi scons graw-progs scons llvmpipe and so on. And there is also the possibility of scepcified subdirs, e.g. scons src/gallium/drivers If nothing is specified then everything will be build. There might be some rough corners over the next days. Please bare with me.
* dri/nouveau: Re-emit the BO state when coming back from a software fallback.Francisco Jerez2010-10-311-2/+4
|
* dri/nouveau: Validate the framebuffer state on read buffer changes.Francisco Jerez2010-10-311-0/+7
|
* dri/nouveau: Fix type promotion issue on 32bit platforms.Francisco Jerez2010-10-311-1/+1
| | | | Fixes some VTX protection errors introduced by e89af209261.
* r300/compiler: Don't clobber presubtract sources during optimizationsTom Stellard2010-10-301-39/+37
| | | | https://bugs.freedesktop.org/show_bug.cgi?id=28294
* dri/nouveau: Pipeline glTexSubImage texture transfers.Francisco Jerez2010-10-312-19/+68
|
* dri/nouveau: Keep small DYNAMIC_DRAW vertex buffers in system ram.Francisco Jerez2010-10-314-13/+35
|
* dri/nouveau: Optimize VBO binding re-emission.Francisco Jerez2010-10-313-26/+84
|
* dri/nouveau: Split out array handling to its own file.Francisco Jerez2010-10-319-165/+244
|
* dri/nouveau: Use a macro to iterate over the bound vertex attributes.Francisco Jerez2010-10-315-112/+83
|
* dri/nouveau: Avoid recursion in nouveau_bo_context_reset().Francisco Jerez2010-10-311-5/+5
|
* dri/nouveau: Split out the scratch helpers to a separate file.Francisco Jerez2010-10-319-78/+165
|
* dri/nouveau: Tell the vbo module we want real hardware BOs.Francisco Jerez2010-10-311-0/+1
|
* dri/nouveau: Honor the access flags in nouveau_bufferobj_map_range.Francisco Jerez2010-10-311-32/+40
|
* dri/nouveau: Call _mesa_update_state() after framebuffer invalidation.Francisco Jerez2010-10-302-1/+6
| | | | | Previously nouveau_state_emit() was being called directly, sometimes that doesn't work because it doesn't update the derived GL context.
* dri/nv25: Bind a hierarchical depth buffer.Francisco Jerez2010-10-305-8/+34
|
* dri/nouveau: Don't assert(0) on compressed internal formats.Francisco Jerez2010-10-301-0/+5
|
* dri/nv20: Clear with the 3D engine.Francisco Jerez2010-10-301-0/+54
|
* st/mesa: Unreference the sampler view in st_bind_surface.Chia-I Wu2010-10-311-0/+1
| | | | | | | | Without this, update_textures may not pick up the new pipe_resource. It is actually update_textures that should check stObj->sampler_view->texture != stObj->pt, but let's follow st_TexImage and others for now.
* osmesa: fix renderbuffer memleak in OSMesaMakeCurrent()Brian Paul2010-10-301-4/+6
| | | | Fixes fd.o bug 31128.
* mesa: Make metaops use program refcounts instead of names.Eric Anholt2010-10-293-62/+80
| | | | | | | Fixes failure on restoring state when the program was active but deleted, and the name no longer exists. Bug #31194
* mesa: remove dead codeBrian Paul2010-10-291-41/+0
|
* mesa: Fix windows build (uint -> GLuint).José Fonseca2010-10-291-4/+4
|
* Merge branch 'glapi-reorg'Chia-I Wu2010-10-299-24/+4499
|\ | | | | | | | | | | | | | | | | Conflicts: src/mapi/glapi/glapi_sparc.S src/mapi/glapi/glapi_x86.S src/mapi/glapi/glapidispatch.h src/mapi/glapi/glapioffsets.h src/mapi/glapi/glprocs.h
| * mesa: Remove unnecessary glapitable.h includes.Chia-I Wu2010-10-276-9/+5
| | | | | | | | | | | | | | | | | | With 07b85457d95bcc70588584e9380c51cd63aa3a2b, glapitable.h is included by core mesa only to know the size of _glapi_table. It is not necessary as the same info is given by _gloffset_COUNT. This change makes _glapi_table opaque to core mesa. All operations on it are supposed to go through one of the SET/GET/CALL macros.
| * vbo: Use CALL_* macros.Chia-I Wu2010-10-271-8/+8
| | | | | | | | | | Use macros to access _glapi_table consistently. There is no functional change.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-29 11:39:02 +0000