github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-17 13:47:08 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #4256] Ollama 0.1.34 not working on AGX Orin 64gb while no issue on 0.1.30 #2657

New Issue
Closed
opened 2026-04-12 12:59:53 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 12:59:53 -05:00
Owner
Copy Link

Originally created by @casimirextreme on GitHub (May 8, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4256

What is the issue?

It was perfectly working on 0.1.30 that I've compiled. Now either the official release nor my compiled version is working

Ollama "serve" hangs with:
CUDA error: the resource allocation failed current device: 0, in function cublas_handle at /data/llm/llama.cpp/ggml-cuda/common.cuh:526 cublasCreate_v2(&cublas_handles[device]) GGML_ASSERT: /data/llm/llama.cpp/ggml-cuda.cu:60: !"CUDA error" Could not attach to process. If your uid matches the uid of the target process, check the setting of /proc/sys/kernel/yama/ptrace_scope, or try again as the root user. For more details, see /etc/sysctl.d/10-ptrace.conf ptrace: Inappropriate ioctl for device. No stack. The program is not being run.

I've added the logs of both versions:
agx-orin-0.1.34.txt
agx-orin-0.1.30.txt

It was compiled with:
OLLAMA_SKIP_CPU_GENERATE="1" CMAKE_CUDA_ARCHITECTURES="72;87"
Is there any known issue with Jetson AGX orin?

OS

Linux

GPU

Nvidia

CPU

Other

Ollama version

0.1.34

Originally created by @casimirextreme on GitHub (May 8, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4256 ### What is the issue? It was perfectly working on 0.1.30 that I've compiled. Now either the official release nor my compiled version is working Ollama "serve" hangs with: `CUDA error: the resource allocation failed current device: 0, in function cublas_handle at /data/llm/llama.cpp/ggml-cuda/common.cuh:526 cublasCreate_v2(&cublas_handles[device]) GGML_ASSERT: /data/llm/llama.cpp/ggml-cuda.cu:60: !"CUDA error" Could not attach to process. If your uid matches the uid of the target process, check the setting of /proc/sys/kernel/yama/ptrace_scope, or try again as the root user. For more details, see /etc/sysctl.d/10-ptrace.conf ptrace: Inappropriate ioctl for device. No stack. The program is not being run.` I've added the logs of both versions: [agx-orin-0.1.34.txt](https://github.com/ollama/ollama/files/15247743/agx-orin-0.1.34.txt) [agx-orin-0.1.30.txt](https://github.com/ollama/ollama/files/15247744/agx-orin-0.1.30.txt) It was compiled with: `OLLAMA_SKIP_CPU_GENERATE="1" CMAKE_CUDA_ARCHITECTURES="72;87" ` Is there any known issue with Jetson AGX orin? ### OS Linux ### GPU Nvidia ### CPU Other ### Ollama version 0.1.34
GiteaMirror added the bug label 2026-04-12 12:59:53 -05:00
GiteaMirror commented 2026-04-12 12:59:55 -05:00
Author
Owner
Copy Link

@casimirextreme commented on GitHub (May 8, 2024):

Found the issue. The docker image used to compile was not the correct one.
One should use nvcr.io/nvidia/l4t-cuda:12.2.12-devel
for the AGX Orin 64gb

<!-- gh-comment-id:2100877897 --> @casimirextreme commented on GitHub (May 8, 2024): Found the issue. The docker image used to compile was not the correct one. One should use nvcr.io/nvidia/l4t-cuda:12.2.12-devel for the AGX Orin 64gb
GiteaMirror commented 2026-04-12 12:59:55 -05:00
Author
Owner
Copy Link

@qubeck commented on GitHub (May 16, 2024):

Hello,
I ran in to the same problem with 0.1.38 on Jetson Xavier NX.
Is the official build for arm64 going to be fixed again or is that particular CUDA version not a priority for some reason?

<!-- gh-comment-id:2115554678 --> @qubeck commented on GitHub (May 16, 2024): Hello, I ran in to the same problem with 0.1.38 on Jetson Xavier NX. Is the official build for arm64 going to be fixed again or is that particular CUDA version not a priority for some reason?
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#2657
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?