summaryrefslogtreecommitdiff
path: root/src/gallium/drivers
AgeCommit message (Collapse)Author
2010-08-15llvmpipe: eliminate last usage of step array in rast_tmp.hKeith Whitwell
For 16 and 64 pixel levels, calculate a mask which is linear in x and y (ie not in the swizzle layout). When iterating over full and partial masks, figure out position by manipulating the bit number set in the mask, rather than relying on postion arrays. Similarly, calculate the lower-level c values from dcdx, dcdy and the position rather than relying on the step array.
2010-08-15llvmpipe: don't refer to plane->step when dcdx or dcdy would doKeith Whitwell
2010-08-15llvmpipe: also use build_mask at 16, 64 pixel levelsKeith Whitwell
2010-08-15llvmpipe: version of block4 which doesn't need the full step arrayKeith Whitwell
No noticable slowdown with isosurf.
2010-08-15llvmpipe: reorganize block4 loop, nice speedupKeith Whitwell
isosurf 95->115 fps just by exchanging the two inner loops in this function...
2010-08-15r300g: mark HiZ/ZMask_clear atoms as non-dirty after emission in clearMarek Olšák
2010-08-15r300g: do not use HiZ if HiZ RAM is not properly initializedMarek Olšák
2010-08-15r300g: rename dirty_zmask -> zmask_in_useMarek Olšák
2010-08-15r300g: do not clear with blitter if we clear just the ZMask RAMMarek Olšák
This skips the blitter clear path entirely if the color is not cleared and the depth+stencil is cleared with the ZMask.
2010-08-15r300g: do not use fastfill if ZMask RAM is not properly initializedMarek Olšák
z_fastfill -> dirty_zmask[level].
2010-08-15r300g: separate num_cs_end_dwords out from prepare_for_renderingMarek Olšák
2010-08-13r300g: do not support separate depth/stencil clear in the driverMarek Olšák
It doesn't work well with Hyper-Z, so put the burden on the state tracker.
2010-08-13r300g: fix fastfill when color and Z clear are invoked separatelyMarek Olšák
This always restores the previous depth clear value after CBZB clear.
2010-08-13r600g: update shader capsMarek Olšák
Sent on ML by Владимир. These values are what fglrx returns.
2010-08-13r300g: disable depth clamp for nowMarek Olšák
It breaks Regnum Online in that it renders random triangles all over the screen. https://bugs.freedesktop.org/show_bug.cgi?id=29518
2010-08-13r600g: fix memory leaks running gears.Dave Airlie
I noticed gears memory usage was heading skywards, some r600 "states" aren't properly refcounted, and the ctx->state is never freed.
2010-08-12Revert "u_blitter: unify clear_depth_stencil and flush_depth_stencil"Marek Olšák
This reverts commit de4784e36505316c2a5ab34cc5b371d17f38d3c5.
2010-08-12r600g: fix typo in stencil translate.Dave Airlie
fixes piglit stencil-twoside and stencil-wrap
2010-08-12r600g: fix provoking-vertex piglit test.Dave Airlie
2010-08-12r600g: improve texture format checker.Dave Airlie
This takes the r300g texture format checker and fixes it up for r600g, it passes glean texSwizzle, pixelformats, and texture_srgb tests, however I think it L8S8_SRGB is broken as is L8_SRGB, need to investigate. Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-12u_blitter: unify clear_depth_stencil and flush_depth_stencilMarek Olšák
No need to enable depth test for clear.
2010-08-11r300/compiler: Implement the CONT opcode.Tom Stellard
2010-08-11r600g: accept empty frag prog shaderJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-08-11r600g: add src negation supportJerome Glisse
Should fix few glBitmap cases. Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-08-11r600g: add point/sprite rendering supportJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-08-11llvmpipe: Debug code to dump interpolation coefficients.José Fonseca
2010-08-11llvmpipe: Use single precision divide for one over area computation.José Fonseca
2010-08-11auxiliary: fix u_split_prim naming conventionLuca Barbieri
Current practice is to start identifiers with "util_" instead of "u_".
2010-08-11auxiliary: move Ben Skeggs' primitive splitter to common codeLuca Barbieri
This is a simple framework that handles splitting primitives in an abstract way. The user has to specify the primitive start, start index and count. Then, it can ask the primitive splitter to "draw" a chunk of the primitive, staying under a given vertex/index budget. The primitive splitter will then call user-supplied functions to emit a range of vertices/indices, as well as switch the edgeflag on or off. This is particularly useful for hardware that either has limits on the vertex count field, or where vertices are pushed on a FIFO or temporary buffer of limited size. Note that unlike other splitters, it does not manipulate data in any way, and merely asks a callback to do so, in vertex intervals.
2010-08-11r300g: initialize VAP_VTX_STATE_CNTLMarek Olšák
This got lost during the rasterizer rewrite.
2010-08-11r300g: implement gl_FrontFacingMarek Olšák
2010-08-10r300g: Remove unnecessary header.Vinson Lee
2010-08-10r300/compiler: Implement hardware assisted loops for vertex shaders.Tom Stellard
Single loops work, but nested loops do not.
2010-08-10r600g: avoid reemiting literal, avoid scheduling empty csJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-08-10llvmpipe: Always use floating-point operators for floating-point typesnobled
See: http://bugs.freedesktop.org/29404 http://bugs.freedesktop.org/29407 Signed-off-by: José Fonseca <jfonseca@vmware.com>
2010-08-09util: Move _mm_shuffle_epi8() to u_sse.h.José Fonseca
It's bound to be useful elsewhere.
2010-08-09r600g: fix r600 context structure, avoid segfault when no scissorJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-08-09r300g: do not print shader compiler errors by defaultMarek Olšák
2010-08-09r600g: fix some warningsMarek Olšák
2010-08-09r600g: fill out some missing caps and sort themMarek Olšák
The shader caps need additional corrections. (based on a patch from netkas at Phoronix)
2010-08-09r300g: fix hiz/zmask offset emissions.Dave Airlie
ofs is in dwords, so need to shift it for registers. Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-09nouveau: fix maps with PIPE_TRANSFER_UNSYNCHRONIZED | PIPE_TRANSFER_DONTBLOCKLuca Barbieri
In this case, we were incorrectly prioritizing PIPE_TRANSFER_DONTBLOCK over PIPE_TRANSFER_UNSYNCHRONIZED. This can lead to failure in the Mesa VBO draw paths that end up specifying both, but don't expect map to fail (in particular, the problem manifested as a leak of buffer objects in teapot with other changes).
2010-08-08r300g: remove a flushMarek Olšák
Ooops, it wasn't supposed to be there.
2010-08-08r300g: flush zmasks of zbuffers we are going to use as samplersMarek Olšák
It sometimes works, sometimes not. I guess we have the zmask offsets wrong.
2010-08-08r300g: do not allocate a zmask block for 3D textures and cubemapsMarek Olšák
2010-08-08r300g: take hiz/zmask offsets into a/c when clearing.Dave Airlie
Need to add a test for multi-hiz/zmask db in a single context. Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-07r300g: fix cbzb clears when hyperz is offMarek Olšák
2010-08-06nouveau: fix potential NULL-ptr dereference in nouveau_stateobj.hMaarten Maathuis
- This can only be triggered when DEBUG_NOUVEAU_STATEOBJ is active. - Also remove a redundant pointer assignment. Reported-by: Roy Spliet <r.spliet@student.tudelft.nl> Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
2010-08-06r600g: add PA_CL_CLIP_CNTL definitionJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-08-06r600g: fix rendering, only enable target we write tooJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>