ci: target x86-64-v3 in benchmarks workflow#12003
ci: target x86-64-v3 in benchmarks workflow#12003xtqqczze wants to merge 1 commit intouutils:mainfrom
Conversation
|
GNU testsuite comparison: |
Merging this PR will improve performance by ×3.3
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ⚡ | Simulation | wc_lines_large_line_count[500000] |
2.6 ms | 2.5 ms | +6.25% |
| ⚡ | Simulation | wc_chars_large_line_count[100000] |
713.4 µs | 683 µs | +4.45% |
| ⚡ | Simulation | wc_lines_variable_length[(50, 500)] |
3.2 ms | 3 ms | +6.32% |
| ⚡ | Simulation | wc_bytes_lines_synthetic[2000] |
120.8 ms | 113.3 ms | +6.65% |
| ⚡ | Simulation | wc_lines_extreme_line_lengths[(100000, 200)] |
1.4 ms | 1.3 ms | +5.68% |
| ⚡ | Simulation | df_deep_directory |
368.3 µs | 344.9 µs | +6.79% |
| ⚡ | Simulation | true_consecutive_calls |
199.4 ns | 170.3 ns | +17.13% |
| ⚡ | Simulation | sort_ascii_c_locale |
16 ms | 15.1 ms | +5.98% |
| ⚡ | Simulation | merge_pre_sorted_files |
102.8 ms | 98.9 ms | +4.01% |
| ⚡ | Simulation | split_lines |
5.6 ms | 4.6 ms | +21.33% |
| ⚡ | Simulation | split_numeric_suffix |
5.8 ms | 4.8 ms | +20.46% |
| ⚡ | Simulation | false_consecutive_calls |
199.4 ns | 170.3 ns | +17.13% |
| ⚡ | Simulation | timeout_quick_exit |
198.1 µs | 191.9 µs | +3.22% |
| ⚡ | Simulation | cksum_sha3 |
1.8 s | 1.4 s | +30.39% |
| ⚡ | Simulation | cksum_sha384 |
1.1 s | 1 s | +7.26% |
| ⚡ | Simulation | cksum_shake256 |
1.9 s | 1.4 s | +29.41% |
| ⚡ | Simulation | cksum_sha224 |
1.8 s | 1.5 s | +19.36% |
| ⚡ | Simulation | cksum_sha256 |
1.8 s | 1.5 s | +19.36% |
| ⚡ | Simulation | cksum_sm3 |
1.6 s | 1.3 s | +19.3% |
| ⚡ | Simulation | cksum_crc |
16.8 ms | 14.7 ms | +14.87% |
| ... | ... | ... | ... | ... | ... |
ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.
Comparing xtqqczze:x86-64-v3 (380fcbc) with main (5aade31)2
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. ↩
-
No successful run was found on
main(dd72eca) during the generation of this report, so 5aade31 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩
|
This is misleading for people using binaries published at here and other distributions. |
|
I think we can replace individual binaries with v3. |
c30a1a1 to
768f310
Compare
fe31b12 to
d5ba7ee
Compare
Ubuntu 26.04 builds packages targeting x86-64-v3, so using this target for benchmarks better reflects real-world performance.