David Peter
f9aa299c0f
Switch to unit-safe quantities
2026-10-05 15:58:56 +02:00
David Peter
914d5cbc59
Units
2026-10-05 15:57:26 +02:00
David Peter
d4c6b53e40
Runs => Measurements
2026-10-05 15:57:26 +02:00
David Peter
8ba632c954
More simplifications
2026-10-05 15:57:12 +02:00
David Peter
f7e6d7286f
Simplify run creation
2026-10-05 15:56:53 +02:00
David Peter
c20fcc0bda
Real => wall clock
2026-10-05 15:56:53 +02:00
David Peter
af4422d212
Remove command_with_unused_parameters field
2026-10-05 15:56:52 +02:00
David Peter
ae0035ad56
Use wait4 instead of getrusage
2026-10-05 15:56:37 +02:00
David Peter
50232feb22
Move user and system time
2026-10-05 15:56:36 +02:00
David Peter
14d773998a
Introduce new BenchmarkRun struct
2026-10-05 15:56:36 +02:00
David Peter
463a889638
Simplify parameterized reference handling
2026-10-05 15:07:46 +02:00
David Peter
f7429e2fb2
Use parameterized benchmarks as references
2026-10-05 14:59:18 +02:00
David Peter
0f2871bd66
Address review feedback
2026-10-05 14:16:33 +02:00
Devon Hollowood
ce704016f9
Make time elapsed only show up initially
2026-10-05 14:14:15 +02:00
David Peter
8084a9eb7f
Adopt the remaining statistics changes from #790
2026-10-05 13:57:31 +02:00
David Peter
df49a04a66
Use typed quantities for relative-speed calculations
2026-10-05 13:41:20 +02:00
David Peter
e689520fb5
Merge pull request #969 from sharkdp/codex/result-measurements
...
Store typed Measurements in BenchmarkResult
2026-10-05 12:51:22 +02:00
David Peter
b23372c5e2
Store typed Measurements in BenchmarkResult
2026-10-05 12:46:10 +02:00
David Peter
30a2b61d4c
Merge pull request #807 from willcl-ark/expose-run-var
...
expose HYPERFINE_ITERATION to intermediate commands
2026-10-05 12:45:29 +02:00
David Peter
3558138cea
Further simplifications
2026-10-05 12:40:33 +02:00
David Peter
8c40de45f8
Store parameter usage metadata in benchmark results
2026-10-05 12:35:24 +02:00
willcl-ark
44c0cc231b
expose HYPERFINE_ITERATION to intermediate commands
...
Fixes : #781
Expose the `$HYPERFINE_ITERATION` variable to intermediate commands.
This will allow iteration-specific prepare/conclude, such as:
```bash
hyperfine --runs 5 --prepare 'echo touch file${HYPERFINE_ITERATION}.dat' \
'echo testing file${HYPERFINE_ITERATION}.dat' \
--conclude 'echo rm file${HYPERFINE_ITERATION}.dat' --show-output
```
2026-10-05 12:24:43 +02:00
David Peter
6f557e0494
Extract typed quantity formatting
2026-10-05 11:46:49 +02:00
David Peter
b5f58651a1
Extract quantity and command helpers
2026-10-05 11:14:09 +02:00
David Peter
aa14b8c0d9
Group units, formatting, and statistics under quantity
2026-10-05 10:46:51 +02:00
David Peter
bef66df7e4
Restore original quantity trait and comments
2026-10-05 10:43:13 +02:00
David Peter
a8e56efecc
Use uom quantities for benchmark measurements
2026-10-05 10:32:32 +02:00
David Peter
db3a3dedc7
Collect benchmark runs in Measurements
2026-10-05 10:10:09 +02:00
David Peter
35c94a0e36
Clarify internal measurement field names
2026-10-05 10:09:41 +02:00
David Peter
750fe11b85
Merge pull request #954 from sharkdp/codex/remove-statistical
...
Replace statistical with local floating-point helpers
2026-10-05 10:03:27 +02:00
David Peter
305e89d536
Consolidate min and max in the statistics module
2026-10-05 10:00:07 +02:00
David Peter
1fb03c3807
Replace statistical with local floating-point helpers
2026-10-05 09:47:33 +02:00
David Peter
6cb9566118
Unify timer and executor measurement results
2026-10-05 09:47:09 +02:00
David Peter
fe75d1cb1b
Merge pull request #853 from Rohan5commit/docs/fix-comment-typo
...
docs: fix typo in relative speed comment
2026-10-05 09:46:07 +02:00
c0b3834277
Handle broken pipe errors on stdout gracefully ( #932 )
...
Replace `println!` calls on the results paths with `writeln!` on
`io::stdout()` and propagate the `io::Error` to `main`, where
`ErrorKind::BrokenPipe` is turned into a quiet `exit(0)`. This matches
the conventional behavior of command-line tools when their output pipe
is closed early (e.g. `hyperfine ... | head -n 0`), and fixes panics in
all run-output paths: the per-benchmark header and results, the speed
comparison summary, and `-`-style exports to stdout. The warnings and
"Note" outputs to stderr are converted the same way.
---------
Co-authored-by: Mathjk <Mathjk@users.noreply.github.com >
Co-authored-by: David Peter <mail@david-peter.de >
2026-10-02 11:18:56 +02:00
David Peter
4d073e08ae
Dependency and MSRV update ( #934 )
2026-10-02 11:05:02 +02:00
rohan436
7720bfaba8
docs: fix typo in relative speed comment
2026-03-13 13:29:46 +08:00
David Peter
b5a6860d96
Support list of exit codes in --ignore-failure
2025-11-18 09:19:27 +01:00
David Peter
98d262c497
Fix warnings
2025-09-03 20:51:44 +02:00
Niklas Dewally
2166c9f02e
Add option to name reference command
...
Add --reference-name option to give a reference command a name.
This was discussed in the PR that added --reference, #744 , but
was unfortunately never implemented.
2025-09-03 20:45:39 +02:00
David Peter
ebcf794d97
Clippy
2024-12-29 14:07:21 +01:00
David Peter
038f369420
SortOrder field
2024-12-29 14:07:21 +01:00
David Peter
0705d6c7d0
Basic scheduler test
2024-12-29 14:07:21 +01:00
David Peter
6556b2bd46
Initial work on tracking memory usage
2024-12-29 12:18:55 +01:00
David Peter
99d24b7a2e
Multiple output policies
2024-11-11 22:24:25 +01:00
David Carlier
2899deb8fd
apply clippy advices on executor/commands lifetimes.
2024-11-11 08:47:33 +01:00
David Peter
83f04cdd31
Add iteration information to failure error message
2024-11-10 22:26:13 +01:00
Jan-Eric Nitschke
13547e973c
Change message for identical time and add test for it
2024-06-30 22:45:40 +02:00
Jan-Eric Nitschke
d4749b015f
Fix prepare and conclude behaviour and add tests.
2024-06-30 22:45:40 +02:00
Jan-Eric Nitschke
9806ed7694
Added option to manually specify a reference to compare the results to.
2024-06-30 22:45:40 +02:00
David Peter