github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-15 12:48:05 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #6336] AMD Discrete GPU Version info not found - Radeon RX Vega56 - gfx900 #3976

New Issue
Open
opened 2026-04-12 14:50:52 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-12 14:50:52 -05:00
Owner
Copy Link

Originally created by @safe049 on GitHub (Aug 13, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6336

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

What happened

Every time I start the ollama,it always shows these:

https://www.amd.com/en/support/linux-drivers" error="amdgpu version file missing: /sys/module/amdgpu/version stat /sys/module/amdgpu/version: no such file or directory"
time=2024-08-13T16:46:10.463+08:00 level=INFO source=amd_linux.go:348 msg="skipping rocm gfx compatibility check" HSA_OVERRIDE_GFX_VERSION=9.0.0
time=2024-08-13T16:46:10.463+08:00 level=INFO source=types.go:105 msg="inference compute" id=0 library=rocm compute=gfx900 driver=0.0 name=1002:687f total="8.0 GiB" available="7.1 GiB"

I am using Arch Linux system and i sure that i installed my drivers correctly
My GPU: Radeon RX Vega56 8GB

ROCm version:
rocm-core 6.0.2-2

What I expect

I can use GPU Support
Without this,i can only use my CPU.

OS

Linux

GPU

AMD

CPU

Intel

Ollama version

0.3.3

Originally created by @safe049 on GitHub (Aug 13, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6336 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? ## What happened Every time I start the ollama,it always shows these: ``` https://www.amd.com/en/support/linux-drivers" error="amdgpu version file missing: /sys/module/amdgpu/version stat /sys/module/amdgpu/version: no such file or directory" time=2024-08-13T16:46:10.463+08:00 level=INFO source=amd_linux.go:348 msg="skipping rocm gfx compatibility check" HSA_OVERRIDE_GFX_VERSION=9.0.0 time=2024-08-13T16:46:10.463+08:00 level=INFO source=types.go:105 msg="inference compute" id=0 library=rocm compute=gfx900 driver=0.0 name=1002:687f total="8.0 GiB" available="7.1 GiB" ``` I am using Arch Linux system and i sure that i installed my drivers correctly My GPU: _Radeon RX Vega56 8GB_ ROCm version: _rocm-core 6.0.2-2_ ## What I expect I can use GPU Support Without this,i can only use my CPU. ### OS Linux ### GPU AMD ### CPU Intel ### Ollama version 0.3.3
GiteaMirror added the feature requestamd labels 2026-04-12 14:50:52 -05:00
GiteaMirror commented 2026-04-12 14:50:53 -05:00
Author
Owner
Copy Link

@davide445 commented on GitHub (Aug 14, 2024):

Not sure if this is the problem but your GPU its not supported from ROCm 6.0.2 (or even the more recent releases)
https://rocm.docs.amd.com/projects/install-on-linux/en/docs-6.0.2/reference/system-requirements.html

<!-- gh-comment-id:2289329691 --> @davide445 commented on GitHub (Aug 14, 2024): Not sure if this is the problem but your GPU its not supported from ROCm 6.0.2 (or even the more recent releases) https://rocm.docs.amd.com/projects/install-on-linux/en/docs-6.0.2/reference/system-requirements.html
GiteaMirror commented 2026-04-12 14:50:53 -05:00
Author
Owner
Copy Link

@safe049 commented on GitHub (Aug 24, 2024):

Indeed it says not supported but it works pretty fine by setting the HSA override and other env settings:

# zshrc

export PATH="$PATH:/opt/rocm/bin"

export HSA_OVERRIDE_GFX_VERSION="9.0.0"
export ROC_ENABLE_PRE_VEGA=1

Currently there is a better solve which is run on the docker[rocm],it works pretty fine on docker.
BTW,if i try to use the rocm version ollama in arch linux repository,it crashes my system which kick me into the SDDM
and make my screen flashing around,after removing it everythings back to normal
but if use the docker version it just worked out of the box.

But I more prefer to use the non-docker one,so i expect it to be solved.

<!-- gh-comment-id:2307973392 --> @safe049 commented on GitHub (Aug 24, 2024): Indeed it says not supported but it works pretty fine by setting the HSA override and other env settings: ``` # zshrc export PATH="$PATH:/opt/rocm/bin" export HSA_OVERRIDE_GFX_VERSION="9.0.0" export ROC_ENABLE_PRE_VEGA=1 ``` Currently there is a better solve which is run on the docker[rocm],it works pretty fine on docker. BTW,if i try to use the rocm version ollama in arch linux repository,it crashes my system which kick me into the SDDM and make my screen flashing around,after removing it everythings back to normal but if use the docker version it just worked out of the box. But I more prefer to use the non-docker one,so i expect it to be solved.
GiteaMirror commented 2026-04-12 14:50:53 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Sep 5, 2024):

As pointed out above, the version of ROCm we compile against no longer supports this GPU. https://rocm.docs.amd.com/projects/install-on-linux/en/latest/reference/system-requirements.html

Our goal is to enable users to build from source to use older versions of ROCm, to a point, but supporting older versions may be tricky as dependencies creep in to new features and capabilities.

https://github.com/ollama/ollama/blob/main/docs/development.md#linux-rocm-amd

<!-- gh-comment-id:2332373852 --> @dhiltgen commented on GitHub (Sep 5, 2024): As pointed out above, the version of ROCm we compile against no longer supports this GPU. https://rocm.docs.amd.com/projects/install-on-linux/en/latest/reference/system-requirements.html Our goal is to enable users to build from source to use older versions of ROCm, to a point, but supporting older versions may be tricky as dependencies creep in to new features and capabilities. https://github.com/ollama/ollama/blob/main/docs/development.md#linux-rocm-amd
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 amd feature request
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#3976
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?