Repository navigation
Bump Rust toolchain to 1.95 - #9333
Conversation
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
Merging this PR will degrade performance by 3.37%
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | Simulation | set_indices_vortex_buffer[65536] |
121.2 µs | 182.6 µs | -33.64% |
| ❌ | Simulation | set_indices_vortex_buffer[16384] |
31.6 µs | 47.1 µs | -32.93% |
| ❌ | Simulation | compact_sliced[(16384, 10)] |
1.4 ms | 2 ms | -28.6% |
| ❌ | Simulation | set_indices_vortex_buffer[2048] |
5.4 µs | 7.5 µs | -27.89% |
| ❌ | Simulation | compact_sliced[(4096, 10)] |
384.9 µs | 528.4 µs | -27.16% |
| ❌ | Simulation | copy_nullable[65536] |
1.3 ms | 1.7 ms | -26.06% |
| ❌ | Simulation | in_place_non_nullable[65536] |
49.4 µs | 65.6 µs | -24.68% |
| ❌ | Simulation | set_indices_vortex_buffer[1024] |
3.6 µs | 4.7 µs | -24.29% |
| ❌ | Simulation | decompress[u64, (10000, 4)] |
311.9 µs | 405 µs | -22.98% |
| ❌ | Simulation | primitive_i64_vortex |
3.3 ms | 4 ms | -18.25% |
| ❌ | Simulation | div_decimal_i64_nonnull |
943 µs | 1,115.9 µs | -15.5% |
| ❌ | Simulation | set_indices[(16384, 0.01)] |
13.5 µs | 16 µs | -15.31% |
| ❌ | Simulation | new_raw_prim_test_between[i32, 2048] |
77 µs | 88.7 µs | -13.18% |
| ❌ | Simulation | set_indices[(16384, 0.1)] |
36.1 µs | 41.5 µs | -12.96% |
| ❌ | Simulation | set_indices[(16384, 0.5)] |
137.3 µs | 156.1 µs | -12.05% |
| ❌ | Simulation | set_indices[(16384, 0.9)] |
237.9 µs | 269.4 µs | -11.69% |
| ❌ | Simulation | set_indices[(16384, 0.99)] |
256 µs | 289.2 µs | -11.49% |
| ❌ | Simulation | compress_alp[f32, (10000, 0.1, 1.0)] |
422.5 µs | 475.8 µs | -11.19% |
| ❌ | Simulation | decompress[u64, (1000, 16)] |
65 µs | 73.2 µs | -11.14% |
| ❌ | Simulation | compress_alp[f32, (10000, 0.01, 1.0)] |
395.7 µs | 443.3 µs | -10.74% |
| ... | ... | ... | ... | ... | ... |
ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing adamg/rust-1-94 (296ff1c) with develop (c2f0adb)
Footnotes
-
46 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
|
@connortsui20 I was told you would like 1.95 |
|
|
||
| match path.extension().and_then(|e| e.to_str()) { | ||
| Some("cuh") | Some("h") => { | ||
| Some("cuh") | Some("h") if !is_generated => { |
The branch was 20 commits behind, including the Rust toolchain bump from 1.91.0 to 1.97.1 (#9333). CodSpeed builds the base from develop and the head from this branch, so the two were compiled by toolchains six releases apart, which is what the `cold_misaligned` instruction-count difference measured. Signed-off-by: "Joe Isaacs" <joe.isaacs@live.co.uk>
The branch was 20 commits behind, including the Rust toolchain bump from 1.91.0 to 1.97.1 (#9333). CodSpeed builds the base from develop and the head from this branch, so the two were compiled by toolchains six releases apart, which is what the `cold_misaligned` instruction-count difference measured. Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Rationale for this change
The upcoming Apache DataFusion release includes this requirement. This version is 3 releases behind the latest (as of this writing 1.97.1, with 1.98 being released in about 10 days).According to libs.rs, 67% of crates.io requests for the top 4000 crates use this version or newer.~
Started as an upgrade to 1.94, but realized that we'll need 1.95 for the full DF 55 upgrade because of geoarrow.