summaryrefslogtreecommitdiffstats
Commit message (Expand)AuthorAgeFilesLines
* vc4: Replace the SSA-style SEL operators with conditional MOVs.Eric Anholt2016-01-066-201/+128
* vc4: Don't try the SF coalescing unless it's on a def.Eric Anholt2016-01-061-3/+3
* gallium/drivers/svga: Use unsigned for loop indexEdward O'Callaghan2016-01-062-7/+7
* gallium/drivers/r600: Use unsigned for loop indexEdward O'Callaghan2016-01-061-9/+9
* gallium/drivers/ilo: Use unsigned for loop indexEdward O'Callaghan2016-01-064-16/+16
* gallium: Use unsigned for loop indexEdward O'Callaghan2016-01-061-3/+3
* gallium/drivers: Remove unnecessary semicolonsEdward O'Callaghan2016-01-0610-11/+11
* gallium: Remove unnecessary semicolonsEdward O'Callaghan2016-01-068-8/+9
* llvmpipe: Optimize lp_rast_triangle_32_3_16 for POWER8Oded Gabbay2016-01-061-1/+141
* llvmpipe: Optimize BUILD_MASK(_LINEAR) for POWER8Oded Gabbay2016-01-061-40/+110
* llvmpipe: Optimize do_triangle_ccw for POWER8Oded Gabbay2016-01-061-0/+100
* llvmpipe: add POWER8 portability file - u_pwr8.hOded Gabbay2016-01-061-0/+310
* configure.ac: Detect if running on POWER8 archOded Gabbay2016-01-061-0/+55
* nir: Add a lower_fdiv option, turn fdiv into fmul/frcp.Kenneth Graunke2016-01-053-0/+3
* i965: Only turn on ARB_compute_shader if we can write registers.Kenneth Graunke2016-01-051-2/+3
* i965: Use rcp in brw_lower_texture_gradients rather than 1.0 / x.Kenneth Graunke2016-01-051-1/+1
* mesa: fix GL_MAX_NAME_LENGTH query for tessellation shadersTimothy Arceri2016-01-061-2/+6
* glsl: don't change the varying type in validation codeTimothy Arceri2016-01-061-5/+0
* glsl: move lowering after matching validationTimothy Arceri2016-01-061-11/+11
* glsl: only add outward facing varyings to resourse list for SSOTimothy Arceri2016-01-061-7/+10
* i965/gen9: Modify the conditions to use blitter on skl+Anuj Phogat2016-01-051-3/+9
* i965/gen9: Return false in place of assert in intelEmitCopyBlit()Anuj Phogat2016-01-051-3/+4
* i965/gen9: Remove regions overlap check in fast copy blitAnuj Phogat2016-01-051-5/+0
* i965/gen9: Don't use fast copy blit in case of non power of 2 cppAnuj Phogat2016-01-051-2/+4
* i915/i965: Fix typo in perf_debug messageIan Romanick2016-01-052-2/+2
* st/mesa: minor indentation fixesBrian Paul2016-01-051-4/+4
* draw: minor indentation fixBrian Paul2016-01-051-1/+1
* mesa: minor clean-up of some memcpy/sizeof() calls in m_matrix.cBrian Paul2016-01-051-7/+7
* util: add debug_dump_ubyte_rgba_bmp()Brian Paul2016-01-052-0/+63
* mesa: check for z=0 in _mesa_Vertex3dv()Brian Paul2016-01-051-1/+4
* svga: fix test for SVGA_NEW_STIPPLEBrian Paul2016-01-052-4/+8
* svga: add some comments in svga_state_vs.cBrian Paul2016-01-051-0/+3
* svga: change svga_hw_view_state::dirty to booleanBrian Paul2016-01-051-1/+1
* svga: avoid emitting redundant SetVertexBuffers() commandsBrian Paul2016-01-052-5/+26
* svga: check for no-ops in svga_bind_sampler_states()Brian Paul2016-01-051-1/+15
* i965: quieten compiler warning about out-of-bounds accessIlia Mirkin2016-01-051-0/+1
* build: enable st/va with nouveau driverJulien Isorce2016-01-051-0/+2
* nvc0: add support for st/vaJulien Isorce2016-01-053-50/+133
* nouveau: split nouveau_vp3_bsp in begin/next/endJulien Isorce2016-01-054-41/+77
* st/va: count number of slicesJulien Isorce2016-01-055-0/+25
* i965/wm: use binding size for ubo/ssbo when automatic size is unsetIlia Mirkin2016-01-051-4/+10
* Revert "i965/wm: use proper API buffer size for the surfaces."Ilia Mirkin2016-01-054-13/+5
* st/mesa: make KHR_debug output independent of context creation flags (v2)Nicolai Hähnle2016-01-044-57/+98
* nvc0: scale up inter_bo size so that it's 16M for a 4K videoIlia Mirkin2016-01-041-2/+5
* nv50,nvc0: fix crash when increasing bsp bo size for h264Ilia Mirkin2016-01-042-4/+4
* i965/wm: use proper API buffer size for the surfaces.Samuel Iglesias Gonsálvez2016-01-044-5/+13
* radeonsi: remove unused parameter from si_shader_binary_read_configMarek Olšák2016-01-033-10/+7
* radeonsi: move si_shader_binary_upload out of si_shader_binary_readMarek Olšák2016-01-033-11/+10
* gallium/radeon: dump LLVM module outside of radeon_llvm_compileMarek Olšák2016-01-034-9/+12
* gallium/radeon: always add +DumpCode to the LLVM target machine for LLVM <= 3.5Marek Olšák2016-01-034-6/+5