summaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/llvmpipe
Commit message (Collapse)AuthorAgeFilesLines
* gallium: make all checks for PIPE_TEXTURE_2D check for PIPE_TEXTURE_RECT tooLuca Barbieri2010-08-202-0/+2
| | | | | | | Searched for them with: git grep -E '[!=]=.*PIPE_TEXTURE_2D|PIPE_TEXTURE_2D.*[!=]=|case.*PIPE_TEXTURE_2D' Behavior hasn't been changed.
* llvmpipe: special case triangles which fall in a single 16x16 blockKeith Whitwell2010-08-153-0/+106
| | | | | Check for these and route them to a dedicated handler with one fewer levels of recursive rasterization.
* llvmpipe: consolidate several loops in lp_rast_triangleKeith Whitwell2010-08-151-18/+15
|
* llvmpipe: remove all traces of step arrays, pos_tablesKeith Whitwell2010-08-153-114/+0
| | | | | No need to calculate these values any longer, nor to store them in the bin data. Improves isosurf a bit more, 115->123 fps.
* llvmpipe: eliminate last usage of step array in rast_tmp.hKeith Whitwell2010-08-152-16/+57
| | | | | | | | | | | | 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.
* llvmpipe: don't refer to plane->step when dcdx or dcdy would doKeith Whitwell2010-08-151-6/+6
|
* llvmpipe: also use build_mask at 16, 64 pixel levelsKeith Whitwell2010-08-151-28/+19
|
* llvmpipe: version of block4 which doesn't need the full step arrayKeith Whitwell2010-08-152-7/+27
| | | | No noticable slowdown with isosurf.
* llvmpipe: reorganize block4 loop, nice speedupKeith Whitwell2010-08-152-12/+19
| | | | | isosurf 95->115 fps just by exchanging the two inner loops in this function...
* llvmpipe: Debug code to dump interpolation coefficients.José Fonseca2010-08-111-1/+29
|
* llvmpipe: Use single precision divide for one over area computation.José Fonseca2010-08-111-1/+1
|
* llvmpipe: Always use floating-point operators for floating-point typesnobled2010-08-101-9/+8
| | | | | | | | See: http://bugs.freedesktop.org/29404 http://bugs.freedesktop.org/29407 Signed-off-by: José Fonseca <[email protected]>
* util: Move _mm_shuffle_epi8() to u_sse.h.José Fonseca2010-08-091-28/+1
| | | | It's bound to be useful elsewhere.
* gallium: remove stray semicolonsBrian Paul2010-08-061-1/+1
|
* llvmpipe: Only get no rast option onceJakob Bornecrantz2010-08-051-1/+5
|
* llvmpipe: Silence unused value warning.Vinson Lee2010-07-301-0/+1
|
* llvmpipe: Fix implicit declaration of lp_func_delete_body warnings.Vinson Lee2010-07-301-0/+1
|
* llvmpipe: delete function bodies after generating machine codeZack Rusin2010-07-301-0/+1
|
* llvmpipe: fix on-debug build breakageBrian Paul2010-07-291-1/+1
|
* llvmpipe: added some jit debug codeBrian Paul2010-07-292-0/+40
| | | | | If we crash in the jitted function we can examine jit_line and jit_state in gdb to learn more about the shader.
* llvmpipe: silence warnings in lp_test_sincos.cBrian Paul2010-07-291-2/+3
|
* llvmpipe: also test the new lp_build_assert() functionBrian Paul2010-07-291-0/+4
|
* llvmpipe: don't call LLVMCreateJITCompiler() twiceBrian Paul2010-07-295-45/+14
| | | | | | | | | Fixes a failed assertion with LLVM 2.6: <unnamed>::JITResolver::JITResolver(llvm::JIT&): Assertion `TheJITResolver == 0&& "Multiple JIT resolvers?"' failed. Though, not everyone seems to experience this problem.
* scons: Use the current python executable for code generation.José Fonseca2010-07-291-1/+2
| | | | Less susceptible to be broken.
* llvmpipe: Avoid corrupting the FPU stack with MMX instructions on 32bit OSes.José Fonseca2010-07-291-0/+5
| | | | | | | Unfortunately LLVM doesn't emit EMMS itself, and there is no easy/effective way to disable MMX. http://llvm.org/bugs/show_bug.cgi?id=3287
* gallium: Avoid void pointer arithmetic.Chia-I Wu2010-07-291-2/+2
| | | | This fixes fdo bug #29286.
* gallium: Keep only pipe_context::draw_vbo.Chia-I Wu2010-07-291-151/+0
| | | | | | | That is, remove pipe_context::draw_arrays, pipe_context::draw_elements, pipe_context::draw_arrays_instanced, pipe_context::draw_elements_instanced, pipe_context::draw_range_elements.
* gallium: Implement draw_vbo and set_index_buffer for all drivers.Chia-I Wu2010-07-293-27/+78
| | | | | | | | | | | | | | | Some drivers define a generic function that is called by all drawing functions. To implement draw_vbo for such drivers, either draw_vbo calls the generic function or the prototype of the generic function is changed to match draw_vbo. Other drivers have no such generic function. draw_vbo is implemented by calling either draw_arrays and draw_elements. For most drivers, set_index_buffer does not mark the state dirty for tracking. Instead, the index buffer state is emitted whenever draw_vbo is called, just like the case with draw_elements. It surely can be improved.
* llvmpipe: pass face+slice to llvmpipe_unswizzle_cbuf_tile()Brian Paul2010-07-271-2/+3
| | | | | | | | Cube map faces and 3D texture slices are treated the same in llvmpipe textures. Need to pass the sum of these fields to llvmpipe_unswizzle_cbuf_tile() as we do elsewhere. Fixes piglit fbo-3d test (fd.o bug 29135).
* llvmpipe: Use lp_build_select_bitwise() where appropriate.José Fonseca2010-07-271-11/+11
| | | | Fixes fdo 29269.
* llvmpipe: Partially fix resource texture from_handleJakob Bornecrantz2010-07-221-1/+35
|
* llvmpipe: Don't align values already alignedJakob Bornecrantz2010-07-221-2/+2
|
* llvmpipe: say no to depth clampMarek Olšák2010-07-211-0/+2
| | | | The other drivers just return 0 without the assert.
* llvmpipe: Remove dead initialization.Vinson Lee2010-07-181-1/+1
|
* llvmpipe: Remove unused variable in lp_test_sincos.Vinson Lee2010-07-171-2/+0
|
* llvmpipe: use single swizzled tileKeith Whitwell2010-07-167-100/+148
| | | | | | | | | | | | | Use a single swizzled tile per colorbuf (and per thread) to avoid accumulating large amounts of cached swizzled data. Now that the SSE3 code has been merged to master, the performance delta of this change is minimal, the main benefit is reduced memory usage due to no longer keeping swizzled copies of render targets. It's clear from the performance of the in-place version of this code that there is still quite a bit of time being spent swizzling & unswizzling, but it's not clear exactly how to reduce that.
* llvmpipe: Describe _mm_shuffle_epi8() with gcc extended inline assembly when ↵José Fonseca2010-07-161-3/+30
| | | | -mssse3 is not supported/enabled.
* llvmpipe: Only use -mssse3 on gcc 4.3+José Fonseca2010-07-161-1/+5
|
* llvmpipe: implement instanced drawing functionsBrian Paul2010-07-151-20/+108
| | | | | And express all the other drawing functions in terms of llvmpipe_draw_range_elements_instanced().
* llvmpipe: Remove redundant statement.José Fonseca2010-07-151-1/+0
| | | | Thanks to Vinson for spotting this.
* llvmpipe: delete lp_test_*.o files with make cleanBrian Paul2010-07-141-0/+3
|
* llvmpipe: Remove redundant alignments.José Fonseca2010-07-142-3/+1
| | | | | | | The lp_rast_shader_inputs' alignment is irrelevant now that it contains pointers instead of actual data. Likewise, lp_rast_triangle's size alignment is meaningless.
* llvmpipe: Addi ssse3 swizzling for B8G8R8A8_UNORM.Chris Li2010-07-143-3/+231
|
* llvmpipe: fix comment typoRoland Scheidegger2010-07-131-2/+2
|
* llvmpipe: move rasterizer to screen instead of setup contextRoland Scheidegger2010-07-134-13/+22
| | | | | there's no point of having this per context, so move to screen (and protect with a mutex).
* llvmpipe: Align texture data to the cache line.José Fonseca2010-07-131-2/+5
|
* llvmpipe: eliminate the set_state rasterizer commandKeith Whitwell2010-07-138-92/+22
| | | | | | Just put a pointer to the state in the tri->inputs struct. Remove some complex logic for eliminating unused statechanges in bins at the expense of a slightly larger triangle struct.
* llvmpipe: pass mask into fragment shaderKeith Whitwell2010-07-1314-790/+901
| | | | | | | | | Move this code back out to C for now, will generate separately. Shader now takes a mask parameter instead of C0/C1/C2/etc. Shader does not currently use that parameter and rasterizes whole pixel stamps always.
* llvmpipe: move fences from per-bin to per-threadKeith Whitwell2010-07-138-29/+58
| | | | | | | | | Rather than inserting an lp_rast_fence command at the end of each bin, have each rasterizer thread call this function directly once it has run out of work to do on a particular scene. This results in fewer calls to the mutex & related functions, but more importantly makes it easier to recognize empty bins.
* llvmpipe: Always swizzle/unswizzle whole tiles.José Fonseca2010-07-133-18/+18
| | | | | This was already the case, but the generated (un)swizzling code was not benefiting of that knowledge.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-20 18:36:23 +0000