gemini-code-assist[bot] commented on code in PR #19886:
URL: https://github.com/apache/tvm/pull/19886#discussion_r3471860220
##########
ci/scripts/package/manylinux_build_libtvm_runtime_cuda.sh:
##########
@@ -29,6 +29,10 @@ repo_root="$(cd "$(dirname "${BASH_SOURCE[0]}")/../../.." &&
pwd)"
build_dir="${repo_root}/build-wheel-cuda"
parallel="$(getconf _NPROCESSORS_ONLN 2>/dev/null || echo 4)"
+# The image ships the toolkit but not the libcuda driver stub; install it from
+# the image's CUDA repo so the sidecar links libcuda.so.1 (sync version with
tag).
+dnf -y install cuda-driver-devel-13-1
Review Comment:
Hardcoding the CUDA version as `13-1` is fragile and will break if the base
image's CUDA version changes (or if `13-1` is a typo for `12-1` or another
version). Instead of hardcoding, we can dynamically detect the preinstalled
CUDA version from `nvcc` and install the matching `cuda-driver-devel` package.
```suggestion
cuda_version=$(/usr/local/cuda/bin/nvcc --version | sed -n 's/.*release
\([0-9]\+\.[0-9]\+\).*/\1/p' | tr '.' '-')
dnf -y install "cuda-driver-devel-${cuda_version}"
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
