summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2010-11-21r300/compiler: Add a more efficient version of rc_find_free_temporary()Tom Stellard
2010-11-21r300/compiler: Enable rename_reg pass for r500 cardsTom Stellard
In addition, the rename_reg pass has been rewritten to use rc_get_readers().
2010-11-21r300/compiler: Use presubtract operations as much as possibleTom Stellard
Previously, presubtract operations where only being used by instructions with less than three source source registers.
2010-11-21r300/compiler: Convert RGB to alpha in the schedulerTom Stellard
2010-11-21r300/compiler: Track readers through branches in rc_get_readers()Tom Stellard
2010-11-21r300/compiler: Handle BREAK and CONTINUE in rc_get_readers()Tom Stellard
2010-11-21r300/compiler: Add rc_get_readers()Tom Stellard
2010-11-21r300/compiler: Ignore alpha dest register when replicating the resultTom Stellard
When the result of the alpha instruction is being replicated to the RGB destination register, we do not need to use alpha's destination register. This fixes an invalid "Too many hardware temporaries used" error in the case where a transcendent operation writes to a temporary register greater than max_temp_regs. NOTE: This is a candidate for the 7.9 branch.
2010-11-21r300/compiler: Use zero as the register index for unused sourcesTom Stellard
This fixes an invalid "Too many hardware temporaries used" error in the case where a source reads from a temporary register with an index greater than max_temp_regs and then the source is marked as unused before the register allocation pass. NOTE: This is a candidate for the 7.9 branch.
2010-11-21r300/compiler: Fix instruction scheduling within IF blocksTom Stellard
Reads of registers that where not written to within the same block were not being tracked. So in a situations like this: 0: IF 1: ADD t0, t1, t2 2: MOV t2, t1 Instruction 2 didn't know that instruction 1 read from t2, so in some cases instruction 2 was being scheduled before instruction 1. NOTE: This is a candidate for the 7.9 branch.
2010-11-21r300/compiler: Fix register allocator's handling of loopsTom Stellard
NOTE: This is a candidate for the 7.9 branch.
2010-11-21r300/compiler: Make sure presubtract sources use supported swizzlesTom Stellard
NOTE: This is a candidate for the 7.9 branch.
2010-11-21r600: Remove unnecessary header.Vinson Lee
2010-11-21docs: add GL 4.1 statusMarek Olšák
2010-11-21st/mesa: enable ARB_explicit_attrib_location and EXT_separate_shader_objectsMarek Olšák
Gallium drivers pass all piglit tests for the two (there are 12 tests for separate_shader_objects and 5 tests for explicit_attrib_location), and I was told the extensions don't need any driver-specific code. I made them dependent on PIPE_CAP_GLSL. Signed-off-by: Brian Paul <brianp@vmware.com>
2010-11-21mesa: fix get_texture_dimensions() for texture array targetsBrian Paul
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=31779
2010-11-21docs: update some GL 3.0 statusBrian Paul
2010-11-21mesa: hook up GL 3.x entrypointsBrian Paul
Fix up some details in the xml files and regenerate dispatch files.
2010-11-21glapi: rename GL3.xml to GL3x.xml as it covers all GL 3.x versionsBrian Paul
2010-11-21mesa: fix error msg typoBrian Paul
2010-11-21i915g: kill idws->poolDaniel Vetter
The drm winsys only ever handles one gem memory manager. Rip out the unnecessary complication. Reviewed-by: Jakob Bornecrantz <wallbraker@gmail.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jakob Bornecrantz <wallbraker@gmail.com>
2010-11-21i915g: kill buf->map_gttDaniel Vetter
Not using the gtt is considered harmful for performance. And for partial uploads there's always drm_intel_bo_subdata. Reviewed-by: Jakob Bornecrantz <wallbraker@gmail.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jakob Bornecrantz <wallbraker@gmail.com>
2010-11-21i915g: kill RGBA/X formatsDaniel Vetter
It's intel, so always little endian! Reviewed-by: Jakob Bornecrantz <wallbraker@gmail.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jakob Bornecrantz <wallbraker@gmail.com>
2010-11-21i915g: add pineview pci idsDaniel Vetter
Reviewed-by: Jakob Bornecrantz <wallbraker@gmail.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jakob Bornecrantz <wallbraker@gmail.com>
2010-11-21i915g: s/hw_tiled/tilingDaniel Vetter
More in line with other intel drivers. Change to use enum by Jakob Bornecrantz. Reviewed-by: Jakob Bornecrantz <wallbraker@gmail.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jakob Bornecrantz <wallbraker@gmail.com>
2010-11-21i915g: rip out ->sw_tiledDaniel Vetter
It looks like this was meant to facilitate unfenced access to textures/ color/renderbuffers. It's totally incomplete and fundamentally broken on a few levels: - broken: The kernel needs to about every tiled bo to fix up bit17 swizzling on swap-in. - unflexible: fenced/unfenced relocs from execbuffer2 do the same, much simpler. - unneeded: with relaxed fencing tiled gem bos are as memory-efficient as this trick. Hence kill it. Reviewed-by: Jakob Bornecrantz <wallbraker@gmail.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jakob Bornecrantz <wallbraker@gmail.com>
2010-11-21r300g: silence guard band cap errorsJoakim Sindholt
Somebody should find out what these are. It can be found on Windows getting a D3DCAPS9 from IDirect3D9::GetCaps() and reading the GuardBand* values.
2010-11-21st/vega: Fix vgReadPixels with a subrectangle.Chia-I Wu
Fix a crash when the subrectangle is not inside the fb. Fix wrong pipe transfer when sx > 0 or sy + height != fb->height. This fixes "readpixels" demo.
2010-11-21st/vega: Set wrap_r for mask and blend samplers.Chia-I Wu
These two samplers use non-normalized texture coordinates. wrap_r cannot be PIPE_TEX_WRAP_REPEAT (the default). This fixes sp_tex_sample.c:1790:get_linear_unorm_wrap: Assertion `0' failed assertion failure.
2010-11-21st/vega: vegaLookupSingle should validate the state.Chia-I Wu
Fix "lookup" demo crash.
2010-11-21tgsi: Add STENCIL to text parser.Chia-I Wu
Fix OpenVG "filter" demo Program received signal SIGSEGV, Segmentation fault. 0xb7153dc9 in str_match_no_case (pcur=0xbfffe564, str=0x0) at tgsi/tgsi_text.c:86 86 while (*str != '\0' && *str == uprcase( *cur )) {
2010-11-20mesa: Clean up header file inclusion in stencil.h.Vinson Lee
2010-11-20mesa: Clean up header file inclusion in shared.h.Vinson Lee
2010-11-20mesa: Clean up header file inclusion in shaderapi.h.Vinson Lee
2010-11-20mesa: Clean up header file inclusion in scissor.h.Vinson Lee
2010-11-20mesa: Clean up header file inclusion in renderbuffer.h.Vinson Lee
2010-11-20mesa: Clean up header file inclusion in readpix.h.Vinson Lee
2010-11-20mesa: Clean up header file inclusion in rastpos.h.Vinson Lee
2010-11-20mesa: Clean up header file inclusion in polygon.h.Vinson Lee
2010-11-20intel: Remove unnecessary header.Vinson Lee
2010-11-20r600: Remove unnecesary header.Vinson Lee
2010-11-20swrast: Remove unnecessary header.Vinson Lee
2010-11-20st/mesa: Remove unnecessary headers.Vinson Lee
2010-11-20scons: Define IN_DRI_DRIVER.Chia-I Wu
The define is required for DRI drivers. It is not needed for libgl-xlib, but the overhead it introduces should be minor.
2010-11-20nvfx: only expose one rt on nv30Xavier Chantry
We do not know how to use more, GL_ARB_draw_buffers is not exposed on blob.
2010-11-20r600g: Fix location for clip plane registersOwen W. Taylor
The stride between the different clip plane registers was incorrect. https://bugs.freedesktop.org/show_bug.cgi?id=31788 agd5f: fix evergreen as well.
2010-11-20r300g: fix rendering with no vertex elementsMarek Olšák
Fixes glsl-vs-point-size, although I meant to fix glsl-novertexdata. Since swrast fails glsl-novertexdata too, I guess it's a core issue.
2010-11-19i965: Remove duplicate MRF writes in the FS backend.Eric Anholt
This is quite common for multitexture sampling, and not only cuts down on the second and later set of MOVs, but typically also allows compute-to-MRF on the first set. No statistically siginficant performance difference in nexuiz (n=3), but it reduces instruction count in one of its shaders and seems like a good idea.
2010-11-19i965: Improve compute-to-mrf.Eric Anholt
We were skipping it if the instruction producing the value we were going to compute-to-mrf used its result reg as a source reg. This meant that the typical "write interpolated color to fragment color" or "texture from interpolated texcoord" shader didn't compute-to-MRF. Just don't check for the interference cases until after we've checked if this is the instruction we wanted to compute-to-MRF. Improves nexuiz high-settings performance on my laptop 0.48% +- 0.08% (n=3).
2010-11-19ir_to_mesa: Detect and emit MOV_SATs for saturate constructs.Eric Anholt
The goal here is to avoid regressing performance on ir_to_mesa drivers for fixed function fragment shaders requiring saturates.