From c47cf0f880bee2554fa955101c8a1016ee37798d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Daniel=20M=C3=BCller?= Date: Fri, 27 Dec 2024 13:34:18 -0800 Subject: [PATCH] Generate large benchmark files for nightly benchmarks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently we only generate the "large" benchmark files for the Criterion benchmarks in our 'bench' CI job, because that is when they are needed. However, with commit 9d1978e66160 ("Add benchmark for ELF "str2symtab" creation") it now one of the earlier nightly benchmarks that already requires one of those files. Thus, make sure to generate them just a bit earlier. Signed-off-by: Daniel Müller --- .github/workflows/test.yml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml index c917cea3..1636951c 100644 --- a/.github/workflows/test.yml +++ b/.github/workflows/test.yml @@ -390,14 +390,14 @@ jobs: shell: bash run: | echo '```' >> $GITHUB_STEP_SUMMARY - cargo bench --workspace --features=nightly -- bench_ | tee --append $GITHUB_STEP_SUMMARY + cargo bench --workspace --features=nightly,blazesym-dev/generate-large-test-files -- bench_ | tee --append $GITHUB_STEP_SUMMARY # We use bencher format here for better relation to the above # but also because it emits less other crap into our summary. # Note that because libtest does not understand the # `--output-format` option, we need to specify the benchmark # binary (`main`) here and have a different invocation for # libtest style benchmarks above. Sigh. - cargo bench --workspace --bench=main --bench=capi --features=blazesym-dev/generate-large-test-files,blazesym-dev/dont-generate-unit-test-files -- --output-format=bencher | tee --append $GITHUB_STEP_SUMMARY + cargo bench --workspace --bench=main --bench=capi --features=blazesym-dev/dont-generate-unit-test-files -- --output-format=bencher | tee --append $GITHUB_STEP_SUMMARY echo '```' >> $GITHUB_STEP_SUMMARY clippy: name: Lint with clippy