github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #9171] Crashes when running deepseek-r1:671b because K-shift #31731

New Issue
Closed
opened 2026-04-22 12:27:17 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-22 12:27:17 -05:00
Owner
Copy Link

Originally created by @myoldcat on GitHub (Feb 17, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9171

What is the issue?

When running the deepseek-r1:671b model, if the chain of thought is relatively long, the model crashes.
The log shows: The current context does not support K-shift

Relevant log output

Feb 17 11:29:30 ubuntu ollama[734313]: time=2025-02-17T11:29:30.763Z level=DEBUG source=cache.go:231 msg="context limit hit - shifting" id=0 limit=2048 input=2048 keep=5 discard=1021
Feb 17 11:29:30 ubuntu ollama[734313]: llama.cpp:11942: The current context does not support K-shift
Feb 17 11:29:30 ubuntu ollama[734313]: SIGSEGV: segmentation violation
Feb 17 11:29:30 ubuntu ollama[734313]: PC=0x7ff76a858cd7 m=14 sigcode=1 addr=0x22b403fbc
Feb 17 11:29:30 ubuntu ollama[734313]: signal arrived during cgo execution
Feb 17 11:29:30 ubuntu ollama[734313]: goroutine 214 gp=0xc000605340 m=14 mp=0xc000680708 [syscall]:
Feb 17 11:29:30 ubuntu ollama[734313]: runtime.cgocall(0x55f1fccbd3a0, 0xc00031dba0)
Feb 17 11:29:30 ubuntu ollama[734313]:         runtime/cgocall.go:167 +0x4b fp=0xc00031db78 sp=0xc00031db40 pc=0x55f1fc115c2b
Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama._Cfunc_llama_decode(0x7ff4058e12d0, {0x1, 0x7ff405d4a2d0, 0x0, 0x0, 0x7ff406400f20, 0x7ff406402f30, 0x7ff405ef58f0, 0x7ff38121c900})
Feb 17 11:29:30 ubuntu ollama[734313]:         _cgo_gotypes.go:545 +0x4f fp=0xc00031dba0 sp=0xc00031db78 pc=0x55f1fc4cabef
Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama.(*Context).Decode.func1(0x55f1fc4d9beb?, 0x7ff4058e12d0?)
Feb 17 11:29:30 ubuntu ollama[734313]:         github.com/ollama/ollama/llama/llama.go:163 +0xf5 fp=0xc00031dc90 sp=0xc00031dba0 pc=0x55f1fc4cd915
Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama.(*Context).Decode(0x55f1fdb43c20?, 0x0?)
Feb 17 11:29:30 ubuntu ollama[734313]:         github.com/ollama/ollama/llama/llama.go:163 +0x13 fp=0xc00031dcd8 sp=0xc00031dc90 pc=0x55f1fc4cd793
Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/runner.(*Server).processBatch(0xc0003557a0, 0xc0008ba060, 0xc00031df20)
Feb 17 11:29:30 ubuntu ollama[734313]:         github.com/ollama/ollama/llama/runner/runner.go:434 +0x23f fp=0xc00031dee0 sp=0xc00031dcd8 pc=0x55f1fc4d89df
Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/runner.(*Server).run(0xc0003557a0, {0x55f1fd2eaf40, 0xc0006930e0})
Feb 17 11:29:30 ubuntu ollama[734313]:         github.com/ollama/ollama/llama/runner/runner.go:342 +0x1d5 fp=0xc00031dfb8 sp=0xc00031dee0 pc=0x55f1fc4d8415
Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/runner.Execute.gowrap2()
Feb 17 11:29:30 ubuntu ollama[734313]:         github.com/ollama/ollama/llama/runner/runner.go:975 +0x28 fp=0xc00031dfe0 sp=0xc00031dfb8 pc=0x55f1fc4dd308

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

ollama 0.5.11

Originally created by @myoldcat on GitHub (Feb 17, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9171 ### What is the issue? When running the deepseek-r1:671b model, if the chain of thought is relatively long, the model crashes. The log shows: The current context does not support K-shift ### Relevant log output ```shell Feb 17 11:29:30 ubuntu ollama[734313]: time=2025-02-17T11:29:30.763Z level=DEBUG source=cache.go:231 msg="context limit hit - shifting" id=0 limit=2048 input=2048 keep=5 discard=1021 Feb 17 11:29:30 ubuntu ollama[734313]: llama.cpp:11942: The current context does not support K-shift Feb 17 11:29:30 ubuntu ollama[734313]: SIGSEGV: segmentation violation Feb 17 11:29:30 ubuntu ollama[734313]: PC=0x7ff76a858cd7 m=14 sigcode=1 addr=0x22b403fbc Feb 17 11:29:30 ubuntu ollama[734313]: signal arrived during cgo execution Feb 17 11:29:30 ubuntu ollama[734313]: goroutine 214 gp=0xc000605340 m=14 mp=0xc000680708 [syscall]: Feb 17 11:29:30 ubuntu ollama[734313]: runtime.cgocall(0x55f1fccbd3a0, 0xc00031dba0) Feb 17 11:29:30 ubuntu ollama[734313]: runtime/cgocall.go:167 +0x4b fp=0xc00031db78 sp=0xc00031db40 pc=0x55f1fc115c2b Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama._Cfunc_llama_decode(0x7ff4058e12d0, {0x1, 0x7ff405d4a2d0, 0x0, 0x0, 0x7ff406400f20, 0x7ff406402f30, 0x7ff405ef58f0, 0x7ff38121c900}) Feb 17 11:29:30 ubuntu ollama[734313]: _cgo_gotypes.go:545 +0x4f fp=0xc00031dba0 sp=0xc00031db78 pc=0x55f1fc4cabef Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama.(*Context).Decode.func1(0x55f1fc4d9beb?, 0x7ff4058e12d0?) Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/llama.go:163 +0xf5 fp=0xc00031dc90 sp=0xc00031dba0 pc=0x55f1fc4cd915 Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama.(*Context).Decode(0x55f1fdb43c20?, 0x0?) Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/llama.go:163 +0x13 fp=0xc00031dcd8 sp=0xc00031dc90 pc=0x55f1fc4cd793 Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/runner.(*Server).processBatch(0xc0003557a0, 0xc0008ba060, 0xc00031df20) Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/runner/runner.go:434 +0x23f fp=0xc00031dee0 sp=0xc00031dcd8 pc=0x55f1fc4d89df Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/runner.(*Server).run(0xc0003557a0, {0x55f1fd2eaf40, 0xc0006930e0}) Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/runner/runner.go:342 +0x1d5 fp=0xc00031dfb8 sp=0xc00031dee0 pc=0x55f1fc4d8415 Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/runner.Execute.gowrap2() Feb 17 11:29:30 ubuntu ollama[734313]: github.com/ollama/ollama/llama/runner/runner.go:975 +0x28 fp=0xc00031dfe0 sp=0xc00031dfb8 pc=0x55f1fc4dd308 ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version ollama 0.5.11
GiteaMirror added the bug label 2026-04-22 12:27:17 -05:00
GiteaMirror commented 2026-04-22 12:27:18 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 17, 2025):

#5975

<!-- gh-comment-id:2662890045 --> @rick-github commented on GitHub (Feb 17, 2025): #5975
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#31731
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?