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 #12106] ArchLinux since 0.11.4 runner seg fault #8046

New Issue
Closed
opened 2026-04-12 20:17:39 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-12 20:17:39 -05:00
Owner
Copy Link

Originally created by @str0g on GitHub (Aug 28, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12106

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

Since 0.11.4 I am not able to run any model, runner is exiting with segmentation fault

Trace in attachment (version 0.17),
last.log

romc libs version 6.4.3
romc info
rocm_info.log

Relevant log output

OS

Linux

GPU

AMD

CPU

AMD

Ollama version

0.11.7

Originally created by @str0g on GitHub (Aug 28, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12106 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? Since 0.11.4 I am not able to run any model, runner is exiting with segmentation fault Trace in attachment (version 0.17), [last.log](https://github.com/user-attachments/files/22021327/last.log) romc libs version 6.4.3 romc info [rocm_info.log](https://github.com/user-attachments/files/22021386/rocm_info.log) ### Relevant log output ```shell ``` ### OS Linux ### GPU AMD ### CPU AMD ### Ollama version 0.11.7
GiteaMirror added the bug label 2026-04-12 20:17:39 -05:00
GiteaMirror commented 2026-04-12 20:17:41 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Aug 28, 2025):

sie 28 10:25:21 razer ollama[49992]: [GIN] 2025/08/28 - 10:25:21 | 200 | 17.696002893s |       127.0.0.1 | POST     "/api/generate"
sie 28 10:25:27 razer ollama[49992]: SIGSEGV: segmentation violation
sie 28 10:25:27 razer ollama[49992]: PC=0x7fb47d1c4640 m=0 sigcode=1 addr=0x28
sie 28 10:25:27 razer ollama[49992]: signal arrived during cgo execution
sie 28 10:25:27 razer ollama[49992]: goroutine 52 gp=0xc000582fc0 m=0 mp=0x559bf6aa1120 [syscall]:
sie 28 10:25:27 razer ollama[49992]: runtime.cgocall(0x559bf5b2da10, 0xc0004fdbd8)
sie 28 10:25:27 razer ollama[49992]:         /usr/lib/go/src/runtime/cgocall.go:167 +0x4b fp=0xc0004fdbb0 sp=0xc0004fdb78 pc=0x559bf4e5922b
sie 28 10:25:27 razer ollama[49992]: github.com/ollama/ollama/llama._Cfunc_llama_decode(0x7fb47b447a20, {0x4, 0x7fa9ac0c8250, 0x0, 0x7fb47bf1d080, 0x7fb47bf40530, 0x7fa9ac0b1610, 0x7fb47bf3ec80})
sie 28 10:25:27 razer ollama[49992]:         _cgo_gotypes.go:609 +0x4a fp=0xc0004fdbd8 sp=0xc0004fdbb0 pc=0x559bf520968a
sie 28 10:25:27 razer ollama[49992]: github.com/ollama/ollama/llama.(*Context).Decode.func1(...)
sie 28 10:25:27 razer ollama[49992]:         /build/ollama/src/ollama/llama/llama.go:133
sie 28 10:25:27 razer ollama[49992]: github.com/ollama/ollama/llama.(*Context).Decode(0xc000508d88?, 0x1?)
sie 28 10:25:27 razer ollama[49992]:         /build/ollama/src/ollama/llama/llama.go:133 +0xed fp=0xc0004fdcc0 sp=0xc0004fdbd8 pc=0x559bf520b8cd
sie 28 10:25:27 razer ollama[49992]: github.com/ollama/ollama/runner/llamarunner.(*Server).processBatch(0xc0004da360, 0xc0000ff900, 0xc000508f28)
sie 28 10:25:27 razer ollama[49992]:         /build/ollama/src/ollama/runner/llamarunner/runner.go:436 +0x209 fp=0xc0004fdee8 sp=0xc0004fdcc0 pc=0x559bf52c7709

Are you by any chance building from source? If so, can you go clean -cache and try again?

<!-- gh-comment-id:3234908755 --> @dhiltgen commented on GitHub (Aug 28, 2025): ``` sie 28 10:25:21 razer ollama[49992]: [GIN] 2025/08/28 - 10:25:21 | 200 | 17.696002893s | 127.0.0.1 | POST "/api/generate" sie 28 10:25:27 razer ollama[49992]: SIGSEGV: segmentation violation sie 28 10:25:27 razer ollama[49992]: PC=0x7fb47d1c4640 m=0 sigcode=1 addr=0x28 sie 28 10:25:27 razer ollama[49992]: signal arrived during cgo execution sie 28 10:25:27 razer ollama[49992]: goroutine 52 gp=0xc000582fc0 m=0 mp=0x559bf6aa1120 [syscall]: sie 28 10:25:27 razer ollama[49992]: runtime.cgocall(0x559bf5b2da10, 0xc0004fdbd8) sie 28 10:25:27 razer ollama[49992]: /usr/lib/go/src/runtime/cgocall.go:167 +0x4b fp=0xc0004fdbb0 sp=0xc0004fdb78 pc=0x559bf4e5922b sie 28 10:25:27 razer ollama[49992]: github.com/ollama/ollama/llama._Cfunc_llama_decode(0x7fb47b447a20, {0x4, 0x7fa9ac0c8250, 0x0, 0x7fb47bf1d080, 0x7fb47bf40530, 0x7fa9ac0b1610, 0x7fb47bf3ec80}) sie 28 10:25:27 razer ollama[49992]: _cgo_gotypes.go:609 +0x4a fp=0xc0004fdbd8 sp=0xc0004fdbb0 pc=0x559bf520968a sie 28 10:25:27 razer ollama[49992]: github.com/ollama/ollama/llama.(*Context).Decode.func1(...) sie 28 10:25:27 razer ollama[49992]: /build/ollama/src/ollama/llama/llama.go:133 sie 28 10:25:27 razer ollama[49992]: github.com/ollama/ollama/llama.(*Context).Decode(0xc000508d88?, 0x1?) sie 28 10:25:27 razer ollama[49992]: /build/ollama/src/ollama/llama/llama.go:133 +0xed fp=0xc0004fdcc0 sp=0xc0004fdbd8 pc=0x559bf520b8cd sie 28 10:25:27 razer ollama[49992]: github.com/ollama/ollama/runner/llamarunner.(*Server).processBatch(0xc0004da360, 0xc0000ff900, 0xc000508f28) sie 28 10:25:27 razer ollama[49992]: /build/ollama/src/ollama/runner/llamarunner/runner.go:436 +0x209 fp=0xc0004fdee8 sp=0xc0004fdcc0 pc=0x559bf52c7709 ``` Are you by any chance building from source? If so, can you `go clean -cache` and try again?
GiteaMirror commented 2026-04-12 20:17:42 -05:00
Author
Owner
Copy Link

@Kfftfuftur commented on GitHub (Aug 28, 2025):

Hello,

I have a similar problem, also on Arch.

According to pacman logs I upgraded to ollama and ollama-rocm 11.4 a few weeks ago, but ollama was running fine just yesterday.
For me it broke after upgrading other rocm related packages.

[2025-08-28T00:21:57+0200] [ALPM] upgraded rocm-core (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:09+0200] [ALPM] upgraded rocm-llvm (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:09+0200] [ALPM] upgraded rocm-device-libs (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:09+0200] [ALPM] upgraded comgr (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:13+0200] [ALPM] upgraded rocprofiler-register (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:13+0200] [ALPM] upgraded hsa-rocr (6.4.1-2 -> 6.4.3-1)
[2025-08-28T00:22:13+0200] [ALPM] upgraded rocminfo (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:13+0200] [ALPM] upgraded hip-runtime-amd (6.4.1-1 -> 6.4.3-2)
[2025-08-28T00:22:13+0200] [ALPM] upgraded hipblas-common (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:13+0200] [ALPM] upgraded roctracer (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:21+0200] [ALPM] upgraded rocblas (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:21+0200] [ALPM] upgraded rocprim (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:21+0200] [ALPM] upgraded rocsparse (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:22+0200] [ALPM] upgraded rocsolver (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:22+0200] [ALPM] upgraded hipblas (6.4.1-1 -> 6.4.3-1)
[2025-08-28T00:22:24+0200] [ALPM] upgraded rocm-smi-lib (6.4.1-1 -> 6.4.3-1)

ollama.log
rocminfo.log

<!-- gh-comment-id:3235119442 --> @Kfftfuftur commented on GitHub (Aug 28, 2025): Hello, I have a similar problem, also on Arch. According to pacman logs I upgraded to ollama and ollama-rocm 11.4 a few weeks ago, but ollama was running fine just yesterday. For me it broke after upgrading other rocm related packages. ``` [2025-08-28T00:21:57+0200] [ALPM] upgraded rocm-core (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:09+0200] [ALPM] upgraded rocm-llvm (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:09+0200] [ALPM] upgraded rocm-device-libs (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:09+0200] [ALPM] upgraded comgr (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:13+0200] [ALPM] upgraded rocprofiler-register (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:13+0200] [ALPM] upgraded hsa-rocr (6.4.1-2 -> 6.4.3-1) [2025-08-28T00:22:13+0200] [ALPM] upgraded rocminfo (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:13+0200] [ALPM] upgraded hip-runtime-amd (6.4.1-1 -> 6.4.3-2) [2025-08-28T00:22:13+0200] [ALPM] upgraded hipblas-common (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:13+0200] [ALPM] upgraded roctracer (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:21+0200] [ALPM] upgraded rocblas (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:21+0200] [ALPM] upgraded rocprim (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:21+0200] [ALPM] upgraded rocsparse (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:22+0200] [ALPM] upgraded rocsolver (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:22+0200] [ALPM] upgraded hipblas (6.4.1-1 -> 6.4.3-1) [2025-08-28T00:22:24+0200] [ALPM] upgraded rocm-smi-lib (6.4.1-1 -> 6.4.3-1) ``` [ollama.log](https://github.com/user-attachments/files/22034132/ollama.log) [rocminfo.log](https://github.com/user-attachments/files/22034131/rocminfo.log)
GiteaMirror commented 2026-04-12 20:17:43 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Aug 29, 2025):

@Kfftfuftur Our official binaries are compiled against ROCm 6.3.3 on Linux currently. I'm not sure how your install is set up, but mixing different versions of ROCm could lead to incompatibilities.

<!-- gh-comment-id:3235320454 --> @dhiltgen commented on GitHub (Aug 29, 2025): @Kfftfuftur Our official binaries are compiled against ROCm 6.3.3 on Linux currently. I'm not sure how your install is set up, but mixing different versions of ROCm could lead to incompatibilities.
GiteaMirror commented 2026-04-12 20:17:43 -05:00
Author
Owner
Copy Link

@Kfftfuftur commented on GitHub (Aug 29, 2025):

On Arch it is built using the PKGBUILD script:
https://gitlab.archlinux.org/archlinux/packaging/packages/ollama/-/blob/main/PKGBUILD?ref_type=heads

I've noticed there is already an open issue there, so I guess this issue can be closed.
https://gitlab.archlinux.org/archlinux/packaging/packages/ollama/-/issues/19

<!-- gh-comment-id:3235877779 --> @Kfftfuftur commented on GitHub (Aug 29, 2025): On Arch it is built using the PKGBUILD script: https://gitlab.archlinux.org/archlinux/packaging/packages/ollama/-/blob/main/PKGBUILD?ref_type=heads I've noticed there is already an open issue there, so I guess this issue can be closed. https://gitlab.archlinux.org/archlinux/packaging/packages/ollama/-/issues/19
GiteaMirror commented 2026-04-12 20:17:44 -05:00
Author
Owner
Copy Link

@str0g commented on GitHub (Aug 29, 2025):

Looks like on arch extra-testing its being fixed, so it was purely maintainers fault, thanks for your time guys!

<!-- gh-comment-id:3237282162 --> @str0g commented on GitHub (Aug 29, 2025): Looks like on arch extra-testing its being fixed, so it was purely maintainers fault, thanks for your time guys!
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#8046
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?