github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #5908] GPU ID initialization incorrect - CPUs not always first in list #3687

New Issue
Closed
opened 2026-04-12 14:30:13 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 14:30:13 -05:00
Owner
Copy Link

Originally created by @7910f6ba7ee4 on GitHub (Jul 24, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/5908

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

In amd_linux.go, it is assumed that "CPUs are always first in the list" when calculating the gpu id:

// CPUs are always first in the list
gpuID := nodeID - cpuCount

however, this is not always the case when the number of topology nodes is greater than 10 (which I unfortunately have), as this results in, for example, the node ids (0, 1, 10, 11, 2, 3...), as the file glob returns things alphabetically. This is a problem when nodes 0-3 are CPUs and 4-11 are GPUs, as 10,11 get set to 8,9 with the current implementation, which results in a gap for gpu ids 6-7.

I am guessing this could be solved though a natural sorting of the list of filename strings before looping through them, or by first calculating all cpu ids before offsetting the gpu ids by the cpu count.

OS

Linux

GPU

AMD

CPU

No response

Ollama version

0.2.1

Originally created by @7910f6ba7ee4 on GitHub (Jul 24, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/5908 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? In `amd_linux.go`, it is assumed that "CPUs are always first in the list" when calculating the gpu id: ``` // CPUs are always first in the list gpuID := nodeID - cpuCount ``` however, this is not always the case when the number of topology nodes is greater than 10 (which I unfortunately have), as this results in, for example, the node ids (0, 1, 10, 11, 2, 3...), as the file glob returns things alphabetically. This is a problem when nodes 0-3 are CPUs and 4-11 are GPUs, as 10,11 get set to 8,9 with the current implementation, which results in a gap for gpu ids 6-7. I am guessing this could be solved though a natural sorting of the list of filename strings before looping through them, or by first calculating all cpu ids before offsetting the gpu ids by the cpu count. ### OS Linux ### GPU AMD ### CPU _No response_ ### Ollama version 0.2.1
GiteaMirror added the linuxamdbug labels 2026-04-12 14:30:13 -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 amd bug linux
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#3687
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?