summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2007-10-24Add the "add" opcode and assert one vp if an opcode isn't supportedZack Rusin
2007-10-24Refactor the tgsi->llvm storage translatorZack Rusin
2007-10-24Implement basic opcode translation and storage translation.Zack Rusin
2007-10-24Draw first triangle. Start on the llvm builder.Zack Rusin
2007-10-24Cleanup some of the testing code. Implement first pass at actuallyZack Rusin
running shaders in llvm.
2007-10-24Execution engine is a singleton, for now keep it in the pipe.Zack Rusin
2007-10-24Generate the base shader.Zack Rusin
2007-10-24Implement the conversion and do the initial execution pass.Zack Rusin
2007-10-24Stub out some conversion.Zack Rusin
2007-10-24Initial stab at LLVM integration.Zack Rusin
2007-10-23Fix vertex cache bug that allows multiple vertices to fall into the same slot.Brian
Need to set the slot's bit in draw->vcache.referenced even when there was a cache hit since flushing the primitive buffer will have cleared the bitfield but not the cache's vertex indexes. Fixes a bug found when drawing long triangle fans but could be hit by other prim types as well. An alternate fix would be to call draw_vertex_cache_invalidate() from draw_vertex_cache_unreference().
2007-10-23added commentBrian
2007-10-23In get_vertex(), slot was computed using & 31. Replace with % VCACHE_SIZE.Brian
Also, assert that index is not too large before indexing array.
2007-10-23adjust coords in wide_line() to be conformantBrian
2007-10-23properly init dst reg's CondMask/Swizzle fieldsBrian
2007-10-23bump up MAX_INSTRUCTIONS and add an assertion to catch emitting too many ↵Brian
instructions
2007-10-23Don't pad renderbuffers to multiple of two pixels anymore.Brian
This was only needed to avoid out-of-bounds memory accesses with the 2x2 quad_read/write() functions which no longer exist.
2007-10-22don't apply ColorMask to main gcBrian
2007-10-22fix masking bug, memory leakBrian
2007-10-22don't use GL types for get/put_tile() paramsBrian
2007-10-22new flag to control psize (from vertex shader or fixed size)Brian
2007-10-22add support for sprite texcoord modesBrian
2007-10-22implement point sprite modeBrian
2007-10-22add point_sprite flag to rasterizer stateBrian
2007-10-22remove unused varBrian
2007-10-22tweak point corners to pass conform testBrian
2007-10-22plug the wide prims code into the pipelineBrian
2007-10-22update to working conditionBrian
2007-10-22Finish unifying the surface and texture tile caches.Brian
2007-10-22pull clip/ module wide and stippled lines/points codekeithw
2007-10-21Start implementing cache routines for textures.Brian
First step to consolidating surface/texture caching...
2007-10-21rename some varsBrian
2007-10-21silence warningBrian
2007-10-20Remove obsolete read/write_quad() functionsBrian
2007-10-20init sp->sbuf_cache to avoid possible segfaultBrian
2007-10-20Call softpipe_unmap_surfaces() in softpipe_flush().Brian
This fixes a DRM BO failure upon swapbuffers caused by the color buffer still being mapped. This is a bit heavy handed since we don't always need to unmap buffers when flushing. Need to pass a flag to flush() or design a new function.
2007-10-20In region_unmap(), check if region is mapped before decrementing refcount.Brian
2007-10-20flush the pipe before accum opsBrian
2007-10-20unmap regions when reallocating renderbuffer storageBrian
2007-10-20renderbuffer tweaks in update_framebuffer_state()Brian
2007-10-20use combined depth/stencil buffer when possibleBrian
2007-10-20Convert Z/stencil ops to use cached tiles like colors.Brian
Also, quite a bit of re-org of the tile caches and surface mapping/unmapping. Leave surfaces mapped between primitives now.
2007-10-20added case for TGSI_OPCODE_ENDBrian
2007-10-19get/put_tile_raw() funcs for 16/32bpp surfacesBrian
2007-10-19added get/put_tile_raw() methodsBrian
2007-10-19don't alloc region in xmesa_surface_alloc(), fixes a mem leakBrian
2007-10-19disable debug printfBrian
2007-10-19call Driver.Flush() in _mesa_notifySwapBuffers()Brian
2007-10-19Initial implementation of surface tile caching.Brian
Instead of using read/write_quad() functions, do framebuffer accesses via get/put_tile(). A cache of tiles is used to avoid frequent get/put() calls. Only implemented for color buffers right now.
2007-10-19Tile cache functions.Brian