perf(element): SIMD-accelerated slice scans (AVX2 + NEON) wired into join and lattice hotpaths
#3MergedComparing
perf-element
(10136c3
) withmain
(7095220
)-55%
- Archived
- 475
Benchmarks
475 total
The benchmarks below were run, but are now archived.
If they were deleted in another branch, consider rebasing to remove them from the report.
If they were added back in this branch, you can restore them to include them in future performance reports.
Learn more about archiving benchmarks
If they were deleted in another branch, consider rebasing to remove them from the report.
If they were added back in this branch, you can restore them to include them in future performance reports.
Learn more about archiving benchmarks
mixed_vs_singleton
benches/lattice.rs::benches::bench_refines::lattice::refinesArchived
+9%
2.1 µs2 µs
meta_singleton
benches/ty.rs::benches::bench_accessors::TypeId::accessorArchived
N/A
< 1 ns< 1 ns
singleton_vs_small_union
benches/lattice.rs::benches::bench_refines::lattice::refinesArchived
+2%
1.3 µs1.2 µs
deep_nested_to_bool
benches/cast.rs::benches::bench_cast::cast::castArchived
+2%
2.7 µs2.7 µs
miss_keyed_array_payload
benches/inspect.rs::benches::bench_any::inspect::anyArchived
0%
1.3 µs1.3 µs
miss_small_union
benches/inspect.rs::benches::bench_any::inspect::anyArchived
0%
468.1 ns468.1 ns
miss_named_object
benches/inspect.rs::benches::bench_any::inspect::anyArchived
0%
684.2 ns684.2 ns
miss_wide_union
benches/inspect.rs::benches::bench_any::inspect::anyArchived
+3%
958.3 ns929.2 ns
wide_union
benches/interner.rs::benches::bench_get_type::Interner::get_typeArchived
0%
153.3 ns153.3 ns
empty
benches/join.rs::benches::bench_compute_with_structural::join::compute_with(structural)Archived
0%
1.2 µs1.2 µs
identity
benches/meet.rs::benches::bench_compute::meet::computeArchived
+19%
370 ns311.7 ns
meta_small
benches/ty.rs::benches::bench_accessors::TypeId::accessorArchived
N/A
< 1 ns< 1 ns
small_union
benches/interner.rs::benches::bench_get_type::Interner::get_typeArchived
0%
153.3 ns153.3 ns
3_distinct_ints
benches/join.rs::benches::bench_compute_with_structural::join::compute_with(structural)Archived
-9%
2.6 µs2.8 µs
hit_int_tiny
benches/inspect.rs::benches::bench_any::inspect::anyArchived
0%
340.6 ns340.6 ns
miss_deep_nested
benches/inspect.rs::benches::bench_any::inspect::anyArchived
0%
1.1 µs1.1 µs
wide
benches/interner.rs::benches::bench_intern_element_list::Interner::intern_element_listArchived
+1%
2.5 µs2.5 µs
singleton_vs_singleton
benches/lattice.rs::benches::bench_overlaps::lattice::overlapsArchived
+7%
898.6 ns840.3 ns
meta_wide
benches/ty.rs::benches::bench_accessors::TypeId::accessorArchived
N/A
< 1 ns< 1 ns
singleton_vs_mixed
benches/meet.rs::benches::bench_narrow::meet::narrowArchived
+1%
2.8 µs2.8 µs
wide_vs_singleton
benches/meet.rs::benches::bench_narrow::meet::narrowArchived
0%
33.4 µs33.5 µs
could_be_falsy_tiny
benches/predicates.rs::benches::bench_predicates::predicatesArchived
+5%
644.4 ns615.3 ns
is_constant_foldable_small_union
benches/predicates.rs::benches::bench_predicates::predicatesArchived
-9%
309.7 ns338.9 ns
is_float_wide_union
benches/predicates.rs::benches::bench_predicates::predicatesArchived
0%
247.8 ns247.8 ns
is_never_deep_nested
benches/predicates.rs::benches::bench_predicates::predicatesArchived
N/A
< 1 ns< 1 ns
Commits
Click on a commit to change the comparison rangeBase
main
7095220
-55.34%
perf(element): SIMD-accelerated slice scans (AVX2 + NEON) wired into join and lattice hotpaths
10136c3
14 days ago
by azjezz