Avatar for the vortex-data user
vortex-data
vortex
BlogDocsChangelog

fix(cuda): handle validity in GPU kernels

#7372Merged
Comparing
ad/cuda-validity
(
1aebccf
) with
develop
(
ae906c7
)
CodSpeed Performance Gauge
+30%
Improvement
2
Untouched
1085
Archived
1490

Benchmarks

2577 total
take_10k_random
encodings/fastlanes/benches/bitpacking_take.rs
CodSpeed Performance Gauge
+30%
251.2 µs193.3 µs
take_10k_contiguous
encodings/fastlanes/benches/bitpacking_take.rs
CodSpeed Performance Gauge
+23%
315 µs257 µs
bitwise_not_vortex_buffer_mut[128]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
+10%
333.6 ns304.4 ns
bitwise_not_vortex_buffer_mut[1024]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
+8%
395.3 ns366.1 ns
iter_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
+7%
890.3 ns831.9 ns
bitwise_not_vortex_buffer_mut[2048]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
+7%
458.9 ns429.7 ns
binary_search_std
vortex-array/benches/search_sorted.rs
CodSpeed Performance Gauge
+6%
515.3 ns486.1 ns
set_indices_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-5%
1.1 µs1.2 µs
encode_varbin[(1000, 4)]
vortex-array/benches/dict_compress.rs
CodSpeed Performance Gauge
+4%
177.2 µs169.7 µs
true_count_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-4%
648.9 ns678.1 ns
binary_search_vortex
vortex-array/benches/search_sorted.rs
CodSpeed Performance Gauge
+4%
727.2 ns698.1 ns
take_map[(0.1, 0.5)]
vortex-array/benches/take_patches.rs
CodSpeed Performance Gauge
-4%
1 ms1 ms
map_new_output[128]
vortex-buffer/benches/vortex_buffer.rs
CodSpeed Performance Gauge
+4%
851.9 ns822.8 ns
bitwise_or_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-4%
1.6 µs1.7 µs
bitwise_and_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-3%
1.7 µs1.7 µs
bitwise_or_arrow_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-3%
1.8 µs1.8 µs
bitwise_and_arrow_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-3%
1.8 µs1.8 µs
train_compressor[(10000, 16, 4)]
encodings/fsst/benches/fsst_compress.rs
CodSpeed Performance Gauge
-3%
3.5 ms3.7 ms
iter_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-3%
893.6 ns922.8 ns
true_count_arrow_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-3%
940.8 ns970 ns
value_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
+3%
1,011.1 ns981.9 ns
bitwise_or_arrow_buffer[2048]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-3%
1.9 µs2 µs
bitwise_and_arrow_buffer[2048]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-3%
2 µs2.1 µs
set_indices_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
+3%
1.1 µs1 µs
true_count_arrow_buffer[2048]
vortex-buffer/benches/vortex_bitbuffer.rs
CodSpeed Performance Gauge
-3%
1 µs1 µs

Commits

Click on a commit to change the comparison range
Base
develop
ae906c7
+29.94%
fix(cuda): propagate validity through dynamic dispatch and fix standalone kernel panics
1aebccf
3 months ago
by 0ax1
© 2026 CodSpeed Technology
Home Terms Privacy Docs