Skip to content

Commit 1808006

Browse files
leofangatmnp
andauthored
ci: disable sccache preprocessor cache mode in wheel build (#141)
The preprocessor cache mode produces minimal/zero hits across CI runs because pip's per-run build-isolation overlay path leaks into compiler flags. See NVIDIA/cuda-python#2220 for the full investigation. LLVM 7 / ToT builds are unaffected — they use the S3 backend configured via ci/setup-sccache.sh and don't go through cibuildwheel. Co-authored-by: Atmn Patel <[email protected]>
1 parent 4976ff7 commit 1808006

1 file changed

Lines changed: 0 additions & 2 deletions

File tree

‎.github/workflows/build-wheel.yml‎

Lines changed: 0 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -56,7 +56,6 @@ jobs:
5656
| sudo tar -C /usr/local/bin -xz --strip-components=1 --wildcards '*/sccache'
5757
sccache --version
5858
echo "SCCACHE_PATH=/usr/local/bin/sccache" >> "$GITHUB_ENV"
59-
echo "SCCACHE_GHA_USE_PREPROCESSOR_CACHE_MODE=true" >> "$GITHUB_ENV"
6059
6160
# xref: https://github.com/orgs/community/discussions/42856#discussioncomment-7678867
6261
- name: Adding addtional GHA cache-related env vars
@@ -134,7 +133,6 @@ jobs:
134133
ACTIONS_RESULTS_URL=${{ env.ACTIONS_RESULTS_URL }}
135134
ACTIONS_CACHE_URL=${{ env.ACTIONS_CACHE_URL }}
136135
ACTIONS_CACHE_SERVICE_V2=${{ env.ACTIONS_CACHE_SERVICE_V2 }}
137-
SCCACHE_GHA_USE_PREPROCESSOR_CACHE_MODE=${{ env.SCCACHE_GHA_USE_PREPROCESSOR_CACHE_MODE }}
138136
CIBW_ENVIRONMENT_WINDOWS: >
139137
MLIR_DIR="$(cygpath -w ${{ env.LLVM_MODERN_INSTALL_DIR }}/lib/cmake/mlir)"
140138
LIBLLVM7="$(cygpath -w ${{ env.LLVM7_INSTALL_DIR }}/bin/LLVM-C.dll)"

0 commit comments

Comments
 (0)