-
Notifications
You must be signed in to change notification settings - Fork 2.2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
OP_REQUIRES failed at xla_compile_on_demand_op.cc:290 : UNIMPLEMENTED: Could not find compiler for platform CUDA: NOT_FOUND #2217
Comments
hi, the error suggests that there is an issue with some dynamics loops which can be implemented in the generate function ( |
@rb-23, Can you try passing |
This issue has been marked stale because it has no recent activity since 7 days. It will be closed if no further activity occurs. Thank you. |
Hi @singhniraj08 , i tried doing as you suggested. Unfortunately, the same error comes out: |
Tagging similar issue in past #2214 |
Bug Report
If this is a bug report, please fill out the following form in full:
System information
Describe the problem
Although CUDA and all other relevant libraries were linked in, when running inference on the model server, the CUDA compiler is not found. This does not happen if i try to run other models with the same containers.
Exact Steps to Reproduce
sudo singularity build.
The following are the container definition file for the tensorflow-serving container, as well as the base tensorflow container to run inference from.base_tensorflow_container.def:
tensorflow_container.def:
save_model.py:
singularity run --nv -B shaped_input_model:/models/model/1 -B /usr/local/cuda-11.8:/usr/local/cuda-11.8 tensorflow_container.sif --per_process_gpu_memory_fraction=0.5
singularity run --nv base_tensorflow_container.sif
and run inference using the python scriptinfer.py:
Source code / logs
output of infer.py
Summary: {'error': '2 root error(s) found.\n (0) UNIMPLEMENTED: Could not find compiler for platform CUDA: NOT_FOUND: could not find registered compiler for platform CUDA -- was support for that platform linked in?\n\t [[{{function_node while_body_26758}}{{node while/XlaDynamicUpdateSlice}}]]\n\t [[StatefulPartitionedCall/StatefulPartitionedCall/while/body/_1058/while/tf_bart_for_conditional_generation/model/decoder/assert_less/Assert/Const_1/_1674]]\n (1) UNIMPLEMENTED: Could not find compiler for platform CUDA: NOT_FOUND: could not find registered compiler for platform CUDA -- was support for that platform linked in?\n\t [[{{function_node while_body_26758}}{{node while/XlaDynamicUpdateSlice}}]]\n0 successful operations.\n0 derived errors ignored.'}
nvcc inside tensorflow_serving singularity container:
The text was updated successfully, but these errors were encountered: