Suppress TSAN false positives in crossbeam crate due to use of memory fences#8464
Conversation
… fences Signed-off-by: Adam Gutglick <[email protected]>
Merging this PR will not alter performance
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | Simulation | decompress_rd[f64, (100000, 0.0)] |
845.4 µs | 1,024 µs | -17.45% |
| ❌ | Simulation | decompress_rd[f32, (100000, 0.0)] |
499 µs | 586.5 µs | -14.91% |
| ❌ | Simulation | bitwise_not_vortex_buffer_mut[128] |
186.1 ns | 215.3 ns | -13.55% |
| ❌ | Simulation | bitwise_not_vortex_buffer_mut[1024] |
246.4 ns | 275.6 ns | -10.58% |
| ⚡ | Simulation | decompress_rd[f64, (100000, 0.01)] |
981.1 µs | 845.4 µs | +16.05% |
| ⚡ | Simulation | decompress_rd[f64, (100000, 0.1)] |
981.1 µs | 845.4 µs | +16.05% |
| ⚡ | Simulation | encode_varbin[(1000, 2)] |
176.1 µs | 157.7 µs | +11.66% |
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing adamg/tsan-suprressions (962576d) with develop (1726f5e)
Footnotes
-
10 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. ↩
Summary
This PR adds a few additional ThreadSanitizer suppressions to deal with false positives caused by
crossbeam. They have been surfaced by #8454 and in other cases in the past.