github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #11916] Race condition with GPU driver using AMD GPU on Linux #69968

New Issue
Open
opened 2026-05-04 19:57:17 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-05-04 19:57:17 -05:00
Owner
Copy Link

Originally created by @importantblimp on GitHub (Aug 15, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11916

What is the issue?

When Ollama is started before the GPU driver has created /sys/class/kfd/kfd/topology/nodes/*/properties Ollama falls back to using the CPU, and doesn't re-attempt to use the GPU.

This leads to inconsistent performance when (e.g.) running ollama serve on startup.
Sometimes Ollama uses the GPU and sometimes it uses the CPU, depending on whether ollama serve runs first or the driver creates the paths first.

This is an issue on master and is an issue because the GPU information is only read once.

Relevant log output

Line 3 is caused by /sys/class/kfd/kfd/topology/nodes/*/properties not existing yet:

source=gpu.go:217 msg="looking for compatible GPUs"
source=amd_linux.go:61 msg="ollama recommends running the 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"
source=amd_linux.go:402 msg="no compatible amdgpu devices detected"
source=gpu.go:377 msg="no compatible GPUs were discovered"

Re-running ollama serve after the amdgpu kernelModule has loaded:

source=gpu.go:217 msg="looking for compatible GPUs"
source=amd_linux.go:61 msg="ollama recommends running the 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"
source=amd_linux.go:389 msg="skipping rocm gfx compatibility check" HSA_OVERRIDE_GFX_VERSION=10.3.0
source=types.go:130 msg="inference compute" id=0 library=rocm variant="" compute=gfx1031 driver=0.0 name=1002:73df total="10.0 GiB" available="8.6 GiB"

OS

Linux

GPU

AMD

CPU

AMD

Ollama version

0.7.0

Originally created by @importantblimp on GitHub (Aug 15, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11916 ### What is the issue? When Ollama is started before the GPU driver has created `/sys/class/kfd/kfd/topology/nodes/*/properties` Ollama falls back to using the CPU, and doesn't re-attempt to use the GPU. This leads to inconsistent performance when (e.g.) running `ollama serve` on startup. Sometimes Ollama uses the GPU and sometimes it uses the CPU, depending on whether `ollama serve` runs first or the driver creates the paths first. This is an issue on [master](https://github.com/ollama/ollama/blob/main/discover/amd_linux.go#L27) and is an issue because the GPU information is only read once. ### Relevant log output Line 3 is caused by `/sys/class/kfd/kfd/topology/nodes/*/properties` not existing yet: ``` source=gpu.go:217 msg="looking for compatible GPUs" source=amd_linux.go:61 msg="ollama recommends running the 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" source=amd_linux.go:402 msg="no compatible amdgpu devices detected" source=gpu.go:377 msg="no compatible GPUs were discovered" ``` Re-running `ollama serve` after the `amdgpu` kernelModule has loaded: ``` source=gpu.go:217 msg="looking for compatible GPUs" source=amd_linux.go:61 msg="ollama recommends running the 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" source=amd_linux.go:389 msg="skipping rocm gfx compatibility check" HSA_OVERRIDE_GFX_VERSION=10.3.0 source=types.go:130 msg="inference compute" id=0 library=rocm variant="" compute=gfx1031 driver=0.0 name=1002:73df total="10.0 GiB" available="8.6 GiB" ``` ### OS Linux ### GPU AMD ### CPU AMD ### Ollama version 0.7.0
GiteaMirror added the bug label 2026-05-04 19:57:17 -05:00
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#69968
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?