andrew

andrew pushed to main at weaselab/conflict-set

2024-07-01 04:21:07 +00:00

66bd799f05 Fix hasty change about end invariants

andrew closed issue weaselab/conflict-set#30

2024-07-01 04:17:49 +00:00

Get back to 100% test coverage

andrew pushed to main at weaselab/conflict-set

2024-07-01 04:17:49 +00:00

2646d5eaf1 Get back to 100% coverage

0367ba9856 Fast path for prefix reads

Compare 2 commits »

andrew pushed to main at weaselab/conflict-set

2024-06-30 22:38:27 +00:00

9dec45317e Try to fix code coverage in Jenkins

a68ad5dd17 Interface change! Return TooOld after 2e9 versions

Compare 2 commits »

andrew pushed to main at weaselab/conflict-set

2024-06-30 20:32:42 +00:00

8e3eacb54f Apply function multi versioning higher in call stack to save branches

andrew pushed to main at weaselab/conflict-set

2024-06-30 18:35:13 +00:00

0184e1d7f6 Remove incorrect comma in CMakeLists.txt

andrew pushed to main at weaselab/conflict-set

2024-06-30 05:54:14 +00:00

c52d50f4f9 Remove bounds from forEachInRange

andrew pushed to main at weaselab/conflict-set

2024-06-30 05:48:25 +00:00

447da11d59 Remove obsolete optimizations

andrew pushed to main at weaselab/conflict-set

2024-06-30 05:42:16 +00:00

daa8e02d4f Fixes from testing on an avx512f-capable machine

andrew pushed to main at weaselab/conflict-set

2024-06-30 05:22:00 +00:00

fd3ea2c2a8 clang-format fixes

andrew pushed to main at weaselab/conflict-set

2024-06-30 05:21:43 +00:00

0b839b9d7e Fixes for symbol multi-versioning with avx512f

andrew pushed to main at weaselab/conflict-set

2024-06-30 04:56:59 +00:00

11a022dcf7 Attempt at avx512f 32bit compare

andrew pushed to main at weaselab/conflict-set

2024-06-29 22:11:49 +00:00

94da4c72a5 Fix clang-format

andrew pushed to main at weaselab/conflict-set

2024-06-29 22:10:42 +00:00

461e07822a 32-bit x86 simd for the other scan16 too

andrew pushed to main at weaselab/conflict-set

2024-06-29 22:03:54 +00:00

75499543e7 Fix clang-format

andrew pushed to main at weaselab/conflict-set

2024-06-29 22:02:41 +00:00

81f44d352f SIMD scan16 for x86 + 32-bit versions

andrew pushed to main at weaselab/conflict-set

2024-06-29 06:07:17 +00:00

45da8fb996 Use the faster unvectorized implementation for Node3

4958a4cced Make always_inline function inline

Compare 2 commits »

andrew pushed to main at weaselab/conflict-set

2024-06-29 02:57:50 +00:00

587874841f Fix test that had a decreasing write version

648b0b9238 Add an always_inline, with explanatory comment

Compare 2 commits »

andrew pushed to main at weaselab/conflict-set

2024-06-29 02:48:28 +00:00

d3f4afa167 More SIMD for scanning Node256 with 32-bit versions

andrew pushed to main at weaselab/conflict-set

2024-06-29 00:29:24 +00:00

f762add4d6 Write vectorized 32-bit compare by hand for arm in scan16