This is an automated email from the ASF dual-hosted git repository.
mshr pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git
The following commit(s) were added to refs/heads/main by this push:
new ac70260e5a [CI] Update system cuda version 12.4->12.8 (#18783)
ac70260e5a is described below
commit ac70260e5a148bf06238f2370ea7cd6e3c305a15
Author: Masahiro Hiramori <[email protected]>
AuthorDate: Mon Feb 16 01:31:25 2026 +0900
[CI] Update system cuda version 12.4->12.8 (#18783)
Fix cudnn errors due to version mismatch between system and torch
dependency
link to the jenkins log:
https://ci.tlcpack.ai/blue/organizations/jenkins/tvm-gpu/detail/PR-18775/11/pipeline
```
tests/python/relax/test_codegen_cudnn.py::test_conv2d_offload[data_shape1-weight_shape1-float32-True-none]
Could not load symbol cudnnGetLibConfig. Error:
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9: undefined symbol:
cudnnGetLibConfig
```
validated the fix with the following local ci script
```bash
python3 tests/scripts/ci.py gpu -d tvm.ci_gpu --tests
tests/python/relax/test_codegen_cudnn.py::test_conv2d_offload
```
---
docker/Dockerfile.ci_gpu | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/docker/Dockerfile.ci_gpu b/docker/Dockerfile.ci_gpu
index 380e5e4e7e..18b4d3828e 100644
--- a/docker/Dockerfile.ci_gpu
+++ b/docker/Dockerfile.ci_gpu
@@ -17,7 +17,7 @@
# CI docker GPU env
# tag: v0.60
-FROM nvidia/cuda:12.4.1-cudnn-devel-ubuntu22.04
+FROM nvidia/cuda:12.8.1-cudnn-devel-ubuntu22.04
COPY utils/apt-install-and-clear.sh /usr/local/bin/apt-install-and-clear
