feat[gpu]: slice support for CUDA dyn dispatch
#6705Comparing
ad/dyn-dispatch-slice
(a6411c1
) withdevelop
(8e29827
)0%
- Untouched
- 954
- Skipped
- 1466
Benchmarks
2420 total
bitwise_not_vortex_buffer_mut[128]
vortex-buffer/benches/vortex_bitbuffer.rs+7%
471.9 ns442.8 ns
bitwise_not_vortex_buffer_mut[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+5%
623.6 ns594.4 ns
bitwise_not_vortex_buffer_mut[2048]
vortex-buffer/benches/vortex_bitbuffer.rs+4%
788.6 ns759.4 ns
map_each[BufferMut<i32>, 128]
vortex-buffer/benches/vortex_buffer.rs+4%
799.7 ns770.6 ns
true_count_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
1.1 µs1.1 µs
push_n_vortex_buffer[u8, 128]
vortex-buffer/benches/vortex_buffer.rs+3%
2.4 µs2.3 µs
true_count_vortex_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.2 µs1.2 µs
push_n_vortex_buffer[u8, 1024]
vortex-buffer/benches/vortex_buffer.rs+2%
2.7 µs2.7 µs
true_count_vortex_buffer[2048]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.4 µs1.4 µs
push_n_vortex_buffer[u8, 2048]
vortex-buffer/benches/vortex_buffer.rs+2%
3.1 µs3.1 µs
chunked_bool_canonical_into[(1000, 10)]
vortex-array/benches/chunk_array_builder.rs+2%
68.2 µs67 µs
push_arrow_buffer[128]
vortex-buffer/benches/vortex_buffer.rs+2%
1.7 µs1.7 µs
decode_varbin[(1000, 4)]
vortex-array/benches/dict_compress.rs+2%
34.4 µs33.8 µs
decompress[u16, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+2%
18.5 µs18.2 µs
from_iter[Buffer<i32>, 128]
vortex-buffer/benches/vortex_buffer.rs+1%
4.1 µs4.1 µs
take_map[(0.005, 0.1)]
vortex-array/benches/take_patches.rs+1%
286.5 µs282.6 µs
bitwise_or_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
5 µs5 µs
decompress_alp[f64, (1000, 0.01, 0.95)]
encodings/alp/benches/alp_compress.rs+1%
17.5 µs17.3 µs
decompress_alp[f64, (1000, 0.1, 0.25)]
encodings/alp/benches/alp_compress.rs+1%
17.8 µs17.6 µs
encode_primitives[i64, (1000, 512)]
vortex-array/benches/dict_compress.rs+1%
227.7 µs225.3 µs
push_n_vortex_buffer[u16, 128]
vortex-buffer/benches/vortex_buffer.rs+1%
2.7 µs2.7 µs
encode_primitives[u8, (1000, 2)]
vortex-array/benches/dict_compress.rs+1%
92.1 µs91.1 µs
bitwise_or_arrow_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
5.5 µs5.5 µs
compress[(10000, 16)]
encodings/runend/benches/run_end_compress.rs+1%
176.1 µs174.2 µs
decompress_alp[f64, (1000, 0.1, 0.95)]
encodings/alp/benches/alp_compress.rs+1%
19.7 µs19.5 µs
Commits
Click on a commit to change the comparison rangeBase
develop
8e29827
+0.03%
feat[gpu]: slice support for CUDA dyn dispatch
a6411c1
6 days ago
by 0ax1