Internal Error: MyelinCheckException: kernel_gen_ds.cpp:1147: CHECK(!redop->owner()->hardware_compute_capability_compatibility()) failed. In HCCC mode, we have disabled tiled reduce fusion under partition dag, so we should never enter here under HCCC mode.
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 13.4k
- Forks
- 2.4k
- Avg merge
- 5d 3h
- Merged PRs (30d)
- 2
Description
Description
When building a tensorRT engine with the flag trt.HardwareCompatibilityLevel.SAME_COMPUTE_CAPABILITY enabled, it fails with the error below:
Internal Error: MyelinCheckException: kernel_gen_ds.cpp:1147: CHECK(!redop->owner()->hardware_compute_capability_compatibility()) failed. In HCCC mode, we have disabled tiled reduce fusion under partition dag, so we should never enter here under HCCC mode.
[01/28/2026-08:55:25] [TRT] [E] Error Code: 9: Skipping tactic 0x0000000000000000 due to exception [myelin_graph.h:1182: attachExceptionMsgToGraph] MyelinCheckException: kernel_gen_ds.cpp:1147: CHECK(!redop->owner()->hardware_compute_capability_compatibility()) failed. In HCCC mode, we have disabled tiled reduce fusion under partition dag, so we should never enter here under HCCC mode. In compileGraph at /_src/optimizer/myelin/codeGenerator.cpp:1548
With the above flag disabled, everything works fine (but I wish this engine to be compatible on devices having the same compute compatibility).
Environment
TensorRT Version: 10.14.1
NVIDIA GPU: RTX4090
NVIDIA Driver Version: 535.230.02
CUDA Version: 12.2
CUDNN Version: 8.9
Operating System: Ubuntu 20.04 LTS
Python Version (if applicable): 3.10.12
Tensorflow Version (if applicable):
PyTorch Version (if applicable):
Baremetal or Container (if so, version):
Relevant Files
Model link:
Steps To Reproduce
Commands or scripts:
Have you tried the latest release?:
Attach the captured .json and .bin files from TensorRT's API Capture tool if you're on an x86_64 Unix system
Can this model run on other frameworks? For example run ONNX model with ONNXRuntime (polygraphy run <model.onnx> --onnxrt):
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the HCCC failure at kernel_gen_ds.cpp:1147 and the compileGraph call noted at myelin/codeGenerator.cpp:1548. Obtain a minimal reproduction or the missing API Capture .json and .bin files, then reproduce with TensorRT 10.14.1 and SAME_COMPUTE_CAPABILITY enabled. Done means the engine builds successfully in HCCC mode without this internal error.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100