So there's at least one CI sanity checking all benchmarks.
Related to #32277.
<!--e57a25ab6845829454e8d69fc972939a-->
The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.
<!--006a51241073e994b41acfe9ec718e94-->
For details see: https://corecheck.dev/bitcoin/bitcoin/pulls/32288.
<!--021abf342d371248e50ceaed478a90ca-->
See the guideline for information on the review process.
If your review is incorrectly listed, please react with 👎 to this comment and the bot will ignore it on the next update.
<!--174a7506f384e20aa4161008e828411d-->
Reviewers, this pull request conflicts with the following ones:
If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first.
utACK 3f602f056001dc36657f01a938f62ea722f78854 - was thinking of pushing something similar
So there's at least one CI sanity checking all benchmarks.
Related to #32277.
357 | @@ -358,7 +358,7 @@ jobs: 358 | ./src/univalue/unitester.exe 359 | 360 | - name: Run benchmarks 361 | - run: ./bin/bench_bitcoin.exe -sanity-check -priority-level=high 362 | + run: ./bin/bench_bitcoin.exe -sanity-check
Why not drop -priority-level=high here as well:https://github.com/bitcoin/bitcoin/blob/7a3afe6787bc36fd98684eac020f9abdbfae6f0a/src/bench/CMakeLists.txt#L82-L84 ?
Will it have a significant impact on execution time?
I'm just changing CI, so that we catch regressions. Someone could follow up with changing this for all developers & users.
But this introduces another inconsistency between Windows and other platforms. Initially, run: ./bin/bench_bitcoin.exe -sanity-check -priority-level=high was written to mimic the bench_sanity_check_high_priority test task behaviour.
Can you explain your issue with running more tests in this CI?
If it only takes a couple of extra seconds, why run a different set of tests for each platform?
Windows should, to the best of our ability, be treated the same as any other platform.
If it only takes a couple of extra seconds,
I haven't measured, for all users, developers and CIs. The point of this change is to immediately start catching regressions in at least one CI.
The point of this change is to immediately start catching regressions in at least one CI.
Ok. Resolved.
The suggestion sounds reasonable. I'd be happy to review a pull doing it.
utACK 27f11217ca63e0f8f78f14db139150052dcd9962
ACK 27f11217ca63e0f8f78f14db139150052dcd9962.
utACK 27f11217ca63e0f8f78f14db139150052dcd9962
Seems related: https://github.com/l0rinc/bitcoin/actions/runs/14519683955/job/40737543737?pr=8
Running with -sanity-check option, output is being suppressed as benchmark results will be useless. terminate called after throwing an instance of 'std::filesystem::__cxx11::filesystem_error' what(): filesystem error: cannot remove all: The process cannot access the file because it is being used by another process [C:\Users\RUNNER~1\AppData\Local\Temp\test_common bitcoin\WalletMigration\af1d7c4664d64d17a328] [C:\Users\RUNNER~1\AppData\Local\Temp\test_common bitcoin\WalletMigration\af1d7c4664d64d17a328\regtest\wallet.dat]