summaryrefslogtreecommitdiffstats
path: root/src/gallium/auxiliary/gallivm
Commit message (Expand)AuthorAgeFilesLines
...
* gallivm: honor d3d10 floating point rules for shadow comparisonsRoland Scheidegger2013-08-081-3/+17
* gallivm: don't clamp reference value for shadow comparison for float formatsRoland Scheidegger2013-08-081-4/+17
* gallivm: propagate scalar_lod to emit_size_query tooRoland Scheidegger2013-08-084-0/+8
* gallivm: fix out-of-bounds behavior for fetch/ldRoland Scheidegger2013-08-083-30/+88
* gallivm: fix comment wrt srgb accuracy.Roland Scheidegger2013-08-081-2/+4
* gallivm: Fix build - Remove TargetOptions.RealignStack for llvm>=3.4Laurent Carlier2013-08-061-0/+2
* gallivm: obey clarified shift behaviorRoland Scheidegger2013-08-022-8/+24
* gallivm: use nearest rounding for float->unorm24 conversionRoland Scheidegger2013-07-311-4/+6
* gallivm: (trivial) git rid of assertion in float->uint conversion codeRoland Scheidegger2013-07-291-2/+3
* gallivm: fix float->SNORM conversionRoland Scheidegger2013-07-272-11/+47
* gallivm: handle texel swizzles correctly for d3d10-style sample opcodesRoland Scheidegger2013-07-271-0/+27
* gallivm: Remove NoFramePointerElimNonLeaf for LLVM >= 3.4.Vinson Lee2013-07-251-0/+2
* gallivm: Remove dead code in lp_build_compare_ext.Vinson Lee2013-07-241-6/+0
* gallivm: add a log function that handles edge casesZack Rusin2013-07-192-0/+21
* gallivm: export unordered/ordered cmp to a common functionZack Rusin2013-07-191-283/+158
* gallivm: handle -inf, inf and nan's in sin/cos instructionsZack Rusin2013-07-192-0/+49
* gallivm: add a version of log2 which handles edge casesZack Rusin2013-07-193-6/+65
* gallivm: fix edge cases in exp2Zack Rusin2013-07-191-3/+7
* gallivm: handle nan's in min/maxZack Rusin2013-07-196-52/+482
* gallivm: (trivial) simplify lp_build_cos/lp_build_sin a tiny bitRoland Scheidegger2013-07-171-7/+6
* llvmpipe: use MCJIT on ARM and AArch64Kyle McMartin2013-07-171-1/+1
* llvmpipe: support sRGB framebuffersRoland Scheidegger2013-07-162-4/+54
* gallivm: (trivial) use constant instead of exp2f() functionRoland Scheidegger2013-07-141-2/+3
* gallivm: handle srgb-to-linear and linear-to-srgb conversionsRoland Scheidegger2013-07-135-7/+331
* gallivm: better support for fast rsqrtRoland Scheidegger2013-07-132-16/+63
* gallium: fixup definitions of the rsq and sqrtZack Rusin2013-07-111-5/+0
* gallivm: Eliminate redundant lp_build_select calls.José Fonseca2013-07-121-12/+2
* tgsi: rename the TGSI fragment kill opcodesBrian Paul2013-07-124-17/+15
* tgsi: fix-up KILP commentsBrian Paul2013-07-121-2/+1
* util: treat denorm'ed floats like zeroZack Rusin2013-07-091-0/+1
* gallivm: (trivial) fix using one lod instead of per-quad lod for texel fetchRoland Scheidegger2013-07-051-1/+2
* gallivm: Remove bogus assert.José Fonseca2013-07-051-4/+1
* gallivm: (trivial) fix bogus assertion for per-element lod with 1d resourcesRoland Scheidegger2013-07-052-2/+1
* gallivm: do per-pixel lod calculations for explicit lodRoland Scheidegger2013-07-046-124/+188
* gallivm: Simplify intrinsic name construction.José Fonseca2013-07-021-23/+10
* gallivm: Fix MSVC build.José Fonseca2013-07-021-8/+7
* gallivm: Fix indirect immediate registers.José Fonseca2013-07-021-2/+2
* draw/gallivm: export overflow arithmetic to its own fileZack Rusin2013-06-282-0/+222
* gallium: Fix llvmpipe on big-endian machinesAdam Jackson2013-06-2410-208/+104
* gallivm: (trivial) remove duplicated code block (including comment)Roland Scheidegger2013-06-131-7/+0
* llvmpipe: Use saturating add/sub for UNORM formatsRichard Sandiford2013-06-101-0/+8
* gallivm: work around slow code generated for interleaving 128bit vectorsRoland Scheidegger2013-06-081-0/+22
* tgsi: replace tgsi_file_names tgsi_file_names[] with tgsi_file_name() functionBrian Paul2013-06-071-1/+1
* gallium: System z supportRichard Sandiford2013-06-051-1/+1
* gallivm: enhance special sse2 4x4f and 2x8f -> 1x16ub conversionRoland Scheidegger2013-06-051-32/+58
* gallivm: (trivial) fix lp_build_concat_nRoland Scheidegger2013-06-051-1/+5
* gallivm: fix out-of-bounds access with mirror_clamp_to_edge address modeRoland Scheidegger2013-06-011-6/+7
* llvmpipe: handle z32s8x24 depth/stencil formatRoland Scheidegger2013-05-182-1/+7
* gallivm: handle z32s8x24 format for samplingRoland Scheidegger2013-05-181-8/+51
* gallivm: Eliminate 8.8 fixed point intermediates from AoS sampling path.José Fonseca2013-05-174-240/+184
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-25 05:09:33 +0000