NVIDIA / NVIDIA/TensorRT

CUDA error 700 launching __myl_MulMeanAddSqrtDivMulMul kernel.

Open
#4,690 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Module:Runtime
Dominant language
C++
Stars
13.4k
Forks
2.4k
Avg merge
5d 3h
Merged PRs (30d)
2

Description

Description

[01/24/2022-19:23:08] [I] Starting inference
[01/24/2022-19:23:08] [E] Error[1]: IExecutionContext::enqueueV3: Error Code 1: Myelin ([exec_instruction.cpp:exec:1087] CUDA error 700 launching __myl_MulMeanAddSqrtDivMulMul kernel.)
Cuda failure at ../common/sampleInference.cpp:660: an illegal memory access was encountered

Environment

TensorRT Version: 10.10.10

NVIDIA GPU: auto-thor-u

NVIDIA Driver Version:

CUDA Version:12.8

CUDNN Version:9.7.1

Operating System: ubuntu 2404 x86_64

Relevant Files

Model link:

Steps To Reproduce

plan is generate, but infer faliled.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start at ../common/sampleInference.cpp:660 and inspect the inference path around the reported CUDA failure. Reproduction depends on a model and setup details that are not provided; a useful result would reproduce or narrow the illegal memory access from the __myl_MulMeanAddSqrtDivMulMul kernel.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
machine-learning, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.