projects
/
mesa.git
/ search
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
|
commitdiff
|
tree
first ⋅ prev ⋅
next
i965/fs: legalize [u]int64 to 32-bit data conversions in lower_d2x
2017-02-08
Roland Scheidegger
mesa: (trivial) include <inttypes.h> for PRIx64 macros
commit
|
commitdiff
|
tree
2017-01-24
Roland Scheidegger
glsl: fix compile errors with mingw due to missing...
commit
|
commitdiff
|
tree
2017-01-24
Roland Scheidegger
gallivm: don't try to use fast rcp for fdiv
commit
|
commitdiff
|
tree
2017-01-24
Roland Scheidegger
gallivm: (trivial) fix ddiv cpu implementation
commit
|
commitdiff
|
tree
2017-01-24
Roland Scheidegger
tgsi: implement ddiv opcode
commit
|
commitdiff
|
tree
2017-01-18
Roland Scheidegger
gallivm: (trivial) fix copy/paste bug with big endian...
commit
|
commitdiff
|
tree
2017-01-06
Roland Scheidegger
llvmpipe: do transpose/untwiddle after conversion for...
commit
|
commitdiff
|
tree
2017-01-06
Roland Scheidegger
gallivm: generalize 4x4f->1x16ub special case conversion
commit
|
commitdiff
|
tree
2017-01-06
Roland Scheidegger
llvmpipe: use alpha from already converted color if...
commit
|
commitdiff
|
tree
2017-01-06
Roland Scheidegger
llvmpipe: use scalar load instead of vectors for small...
commit
|
commitdiff
|
tree
2017-01-05
Roland Scheidegger
gallivm: (trivial) fix typo bug with small AoS format...
commit
|
commitdiff
|
tree
2017-01-05
Roland Scheidegger
gallivm: implement aos unpack (to unorm8) for small...
commit
|
commitdiff
|
tree
2017-01-05
Roland Scheidegger
gallivm: optimize lp_build_unpack_arith_rgba_aos slightly
commit
|
commitdiff
|
tree
2017-01-05
Roland Scheidegger
gallivm: use 2 srcs for 32->16bit conversions in lp_bld_conv...
commit
|
commitdiff
|
tree
2017-01-05
Roland Scheidegger
llvmpipe: (trivial) minimally simplify mask construction
commit
|
commitdiff
|
tree
2016-12-21
Roland Scheidegger
draw: use SoA fetch, not AoS one
commit
|
commitdiff
|
tree
2016-12-21
Roland Scheidegger
gallivm: generalize the compressed format soa fetch...
commit
|
commitdiff
|
tree
2016-12-21
Roland Scheidegger
gallivm: provide soa fetch path handling formats with...
commit
|
commitdiff
|
tree
2016-12-21
Roland Scheidegger
gallivm: optimize gather a bit, by using supplied destinatio...
commit
|
commitdiff
|
tree
2016-12-21
Roland Scheidegger
gallivm: optimize SoA AoS fallback fetch path a little
commit
|
commitdiff
|
tree
2016-12-21
Roland Scheidegger
gallivm: (trivial) handle non-aligned fetch for lp_build_fet...
commit
|
commitdiff
|
tree
2016-12-06
Roland Scheidegger
gallivm: optimize 16bit->32bit gather path a bit
commit
|
commitdiff
|
tree
2016-12-06
Roland Scheidegger
gallivm: handle 16bit float fetches in lp_build_fetch_rgba_soa
commit
|
commitdiff
|
tree
2016-12-06
Roland Scheidegger
util: (trivial) ETC1 meets the criteria for fitting...
commit
|
commitdiff
|
tree
2016-12-06
Roland Scheidegger
glsl: (trivial) fix type typo
commit
|
commitdiff
|
tree
2016-12-06
Roland Scheidegger
main: allow NEAREST_MIPMAP_NEAREST for stencil texturing
commit
|
commitdiff
|
tree
2016-12-06
Roland Scheidegger
glsl: fix ldexp lowering if bitfield insert lowering...
commit
|
commitdiff
|
tree
2016-11-21
Roland Scheidegger
draw: drop some overflow computations
commit
|
commitdiff
|
tree
2016-11-21
Roland Scheidegger
draw: simplify fetch some more
commit
|
commitdiff
|
tree
2016-11-21
Roland Scheidegger
draw: unify linear and elts draw jit functions
commit
|
commitdiff
|
tree
2016-11-21
Roland Scheidegger
draw: use same argument order for jit draw linear ...
commit
|
commitdiff
|
tree
2016-11-21
Roland Scheidegger
draw: drop unnecessary index overflow handling from...
commit
|
commitdiff
|
tree
2016-11-21
Roland Scheidegger
draw: simplify vsplit elts code a bit
commit
|
commitdiff
|
tree
2016-11-18
Roland Scheidegger
draw: finally optimize bool clip mask generation
commit
|
commitdiff
|
tree
2016-11-18
Roland Scheidegger
draw: use vectorized calculations for fetch (v2)
commit
|
commitdiff
|
tree
2016-11-09
Roland Scheidegger
Revert "draw: use vectorized calculations for fetch"
commit
|
commitdiff
|
tree
2016-11-08
Roland Scheidegger
draw: use vectorized calculations for fetch
commit
|
commitdiff
|
tree
2016-11-08
Roland Scheidegger
gallivm: introduce 32x32->64bit lp_build_mul_32_lohi...
commit
|
commitdiff
|
tree
2016-11-04
Roland Scheidegger
draw: fix undefined input handling some more...
commit
|
commitdiff
|
tree
2016-10-18
Roland Scheidegger
draw: improve vertex fetch (v2)
commit
|
commitdiff
|
tree
2016-10-18
Roland Scheidegger
draw: improved handling of undefined inputs
commit
|
commitdiff
|
tree
2016-10-18
Roland Scheidegger
gallivm: print out time for jitting functions with...
commit
|
commitdiff
|
tree
2016-10-18
Roland Scheidegger
gallivm: Use native packs and unpacks for the lerps
commit
|
commitdiff
|
tree
2016-10-12
Roland Scheidegger
draw: initialize shader inputs
commit
|
commitdiff
|
tree
2016-10-04
Roland Scheidegger
gallivm: Use 8 wide AoS sampling on AVX2.
commit
|
commitdiff
|
tree
2016-09-06
Roland Scheidegger
util: (trivial) add <stdint.h> include to slab.c
commit
|
commitdiff
|
tree
2016-08-29
Roland Scheidegger
softpipe: (trivial) honor render_condition_enabled...
commit
|
commitdiff
|
tree
2016-08-29
Roland Scheidegger
llvmpipe: (trivial) honor render_condition_enabled...
commit
|
commitdiff
|
tree
2016-08-20
Roland Scheidegger
llvmpipe: fix issues with depth clamp
commit
|
commitdiff
|
tree
2016-08-20
Roland Scheidegger
llvmpipe: fix depth clamping wrt reversed near/far...
commit
|
commitdiff
|
tree
2016-07-30
Roland Scheidegger
Revert "gallium/util: fix resource leak"
commit
|
commitdiff
|
tree
2016-06-20
Roland Scheidegger
gallivm: don't use integer min/max sse intrinsics with...
commit
|
commitdiff
|
tree
2016-06-14
Roland Scheidegger
gallium/util: don't use blocksize for minify for assertions
commit
|
commitdiff
|
tree
2016-06-14
Roland Scheidegger
llvmpipe: hack-fix bugs due to bogus bind flags
commit
|
commitdiff
|
tree
2016-05-27
Roland Scheidegger
gallivm: eliminate a unnecessary AND with unorm lerps
commit
|
commitdiff
|
tree
2016-05-27
Roland Scheidegger
gallium/util: use enum pipe_prim_type instead of unsigned...
commit
|
commitdiff
|
tree
2016-05-11
Roland Scheidegger
gallivm: improve dumping of bitcode
commit
|
commitdiff
|
tree
2016-05-10
Roland Scheidegger
gallivm: print declarations of intrinsics with GALLIVM_DEBUG=ir
commit
|
commitdiff
|
tree
2016-05-10
Roland Scheidegger
gallivm: use InternalLinkage instead of PrivateLinkage...
commit
|
commitdiff
|
tree
2016-05-10
Roland Scheidegger
gallivm: disable avx512 features
commit
|
commitdiff
|
tree
2016-04-26
Roland Scheidegger
gallivm: make sampling more robust against bogus coordinates
commit
|
commitdiff
|
tree
2016-04-21
Roland Scheidegger
gallivm: fix bogus argument order to lp_build_sample_mipmap...
commit
|
commitdiff
|
tree
2016-04-21
Roland Scheidegger
glsl: add forgotten textureOffset function for sampler2DArra...
commit
|
commitdiff
|
tree
2016-04-17
Roland Scheidegger
gallivm: don't use vector selects with llvm 3.7
commit
|
commitdiff
|
tree
2016-04-15
Roland Scheidegger
llvmpipe: (trivial) initialize src1_alpha var to NULL
commit
|
commitdiff
|
tree
2016-04-13
Roland Scheidegger
gallivm: use llvm.nearbyint instead of llvm.round.
commit
|
commitdiff
|
tree
2016-04-12
Roland Scheidegger
i965/tiled_memcpy: Provide SSE2 for RGBA8 <-> BGRA8...
commit
|
commitdiff
|
tree
2016-03-30
Roland Scheidegger
tgsi: (trivial) only verify target for is_tex instructions
commit
|
commitdiff
|
tree
2016-03-15
Roland Scheidegger
llvmpipe: fix lp_rast_plane alignment on 32bit
commit
|
commitdiff
|
tree
2016-03-15
Roland Scheidegger
draw: fix line stippling
commit
|
commitdiff
|
tree
2016-03-15
Roland Scheidegger
softpipe: fix misleading TGSI_QUAD_SIZE usage
commit
|
commitdiff
|
tree
2016-03-15
Roland Scheidegger
softpipe: fix anisotropic filtering crash
commit
|
commitdiff
|
tree
2016-02-18
Roland Scheidegger
gallivm, tgsi: provide fake sample_i_ms implementations
commit
|
commitdiff
|
tree
2016-02-03
Roland Scheidegger
llvmpipe: use scissor_planes_needed helper function
commit
|
commitdiff
|
tree
2016-02-02
Roland Scheidegger
Revert "i965: Provide sse2 version for rgba8 <-> bgra8...
commit
|
commitdiff
|
tree
2016-02-02
Roland Scheidegger
gallivm: add PK2H/UP2H support
commit
|
commitdiff
|
tree
2016-02-02
Roland Scheidegger
gallivm: add PK2H/UP2H support
commit
|
commitdiff
|
tree
2016-02-02
Roland Scheidegger
tgsi: add PK2H/UP2H support
Reviewed-by:
Roland Scheidegger
<sroland@vmvware.com>
commit
|
commitdiff
|
tree
2016-02-02
Roland Scheidegger
llvmpipe: drop scissor planes early if the tri is fully...
commit
|
commitdiff
|
tree
2016-02-02
Roland Scheidegger
llvmpipe: minor cleanup of sse2 for calc_fixed_position
commit
|
commitdiff
|
tree
2016-02-02
Roland Scheidegger
llvmpipe: use vector loads for (optimized) tri raster...
commit
|
commitdiff
|
tree
2016-02-02
Roland Scheidegger
i965: Provide sse2 version for rgba8 <-> bgra8 swizzle
commit
|
commitdiff
|
tree
2016-02-02
Roland Scheidegger
mesa: fix typo in python scripts
commit
|
commitdiff
|
tree
2016-01-20
Roland Scheidegger
llvmpipe: warn about illegal use of objects in different...
commit
|
commitdiff
|
tree
2016-01-20
Roland Scheidegger
llvmpipe,i915: add back NEW_RASTERIZER dependency when...
commit
|
commitdiff
|
tree
2016-01-20
Roland Scheidegger
llvmpipe: turn depth clears into full depth/stencil...
commit
|
commitdiff
|
tree
2016-01-15
Roland Scheidegger
llvmpipe: ditch additional ref counting for vertex...
commit
|
commitdiff
|
tree
2016-01-15
Roland Scheidegger
llvmpipe: fix "leaking" textures
commit
|
commitdiff
|
tree
2016-01-13
Roland Scheidegger
llvmpipe: (trivial) use cast wrapper for __m128d to...
commit
|
commitdiff
|
tree
2016-01-13
Roland Scheidegger
llvmpipe: avoid most 64 bit math in rasterization
commit
|
commitdiff
|
tree
2016-01-13
Roland Scheidegger
llvmpipe: scale up bounding box planes to subpixel...
commit
|
commitdiff
|
tree
2016-01-13
Roland Scheidegger
llvmpipe: add sse code for fixed position calculation
commit
|
commitdiff
|
tree
2016-01-13
Roland Scheidegger
draw: fix key comparison with uninitialized value
commit
|
commitdiff
|
tree
2016-01-07
Roland Scheidegger
llvmpipe: do 64bit plane calculations in the sse path
commit
|
commitdiff
|
tree
2016-01-07
Roland Scheidegger
llvmpipe: don't store eo as 64bit int
commit
|
commitdiff
|
tree
2016-01-07
Roland Scheidegger
llvmpipe: use aligned data for the assembly program...
commit
|
commitdiff
|
tree
2016-01-07
Roland Scheidegger
draw: initialize prim header flags when clipping lines
commit
|
commitdiff
|
tree
2016-01-07
Roland Scheidegger
draw: fix line stippling with unfilled prims
commit
|
commitdiff
|
tree
2016-01-07
Roland Scheidegger
softpipe: tell draw about the vertex layout we want
commit
|
commitdiff
|
tree
2016-01-07
Roland Scheidegger
llvmpipe: use ints not unsigned for slots
commit
|
commitdiff
|
tree
next