This repository has been archived by the owner on Feb 1, 2023. It is now read-only.
ci(circleci): add benchmark comparisons #147
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
hannahhoward
force-pushed
the
feat/benchmark-ci
branch
22 times, most recently
from
4d3ce76
to
6dca0e9
Compare
hannahhoward
force-pushed
the
feat/benchmark-ci
branch
2 times, most recently
from
2a6239e
to
c0f6e39
Compare
Run benchmarks before and after, compare results
hannahhoward
force-pushed
the
feat/benchmark-ci
branch
2 times, most recently
from
b8cf00b
to
4238f53
Compare
hannahhoward
force-pushed
the
feat/benchmark-ci
branch
from
4238f53
to
31553b7
Compare
Stebalien
reviewed
Jul 3, 2019
bin/build.sh
Outdated
| done | ||
| } | ||
|
|
||
| build_all |
bin/run-benchmarks.sh
Outdated
| fi | ||
| ( cd "$dir"; go build -buildmode=$buildmode -o /dev/null "$pkg") | ||
| done | ||
| } |
There was a problem hiding this comment.
We can probably skip all of this logic and just rely on the standard test scripts.
Kubuxu
reviewed
Jul 3, 2019
bin/run-benchmarks.sh
Outdated
| display_and_run BENCHMARK_SEED="$$" ./bin/benchmark-to-file.sh benchmark-after.txt after | ||
| if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then | ||
| display_and_run ./bin/diff-benchmarks.sh benchmark-before.txt benchmark-after.txt | ||
| fi |
There was a problem hiding this comment.
Let's have \n at the ends of scripts. This is standard on unix systems.
|
shellcheck pointed these out: |
Remove unused code, follow conventions
|
@Stebalien + @Kubuxu incorporated suggestions though I wasn't sure what to do with "The benchmarks look good but I'd like to avoid duplicating the CI scripts." Anyway let me know if I can merge. |
Stebalien
approved these changes
Jul 3, 2019
There was a problem hiding this comment.
LGTM! ❤️
The "duplicate code" comment was referring to the code duplicated from https://raw.githubusercontent.com/ipfs/ci-helpers/master/travis-ci/run-standard-tests.sh
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Goals
Add Benchmarking to CI in Bitswap!
Implementation
This adds a new job to Bitswap's Travis Config that runs benchmarks across both master and the current branch based on ipfs/kubo#6446 but adopted for Travis instead of CircleCI (various script config changes, etc)
Also, required making a few patches to benchmarks so they aren't failing and also seeded on a similar random seed. These were merged in master so this PR could complete CI w/o failing. (#dontHateMe)
For Discussion
I can't get these to predictably and reliably pass without setting the threshold lower than <5% --- I've got it currently set at <25%, which is a bummer in that it's imprecise but still protects against order of magnitude regressions ala what happened with 0.4.21.
Also, this does mean CI takes somewhat longer, about 6 minutes.