[GH-ISSUE #10991] ollama crash running Gemma3:27b #53759

Closed
opened 2026-04-29 04:42:14 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @mario-grgic on GitHub (Jun 6, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10991

What is the issue?

ollama crashes when running gemma3:27b. I did git bisect and it looks like the behaviour was introduced in the following commit:

aaa7818000c42a82fc030212c35ef83f9799efd7 is the first bad commit
commit aaa7818000c42a82fc030212c35ef83f9799efd7
Author: Jesse Gross <jesse@ollama.com>
Date:   Thu Apr 24 11:48:49 2025 -0700

    ggml: Export GPU UUIDs
    
    This enables matching up devices and information reported by the backend
    with system management libraries such as nvml to get accurate free
    memory reporting.

 llama/patches/0017-ggml-Export-GPU-UUIDs.patch   | 102 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 ml/backend.go                                    |   8 +++++
 ml/backend/ggml/ggml.go                          |   6 ++++
 ml/backend/ggml/ggml/include/ggml-backend.h      |   1 +
 ml/backend/ggml/ggml/src/ggml-cuda/ggml-cuda.cu  |  33 +++++++++++++++++++++
 ml/backend/ggml/ggml/src/ggml-metal/ggml-metal.m |   1 +
 6 files changed, 151 insertions(+)
 create mode 100644 llama/patches/0017-ggml-Export-GPU-UUIDs.patch

tag 0.9.0 works fine.

Here is the crash log:

$ ./ollama serve
time=2025-06-05T21:10:01.164-04:00 level=INFO source=routes.go:1242 msg="server config" env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Users/mariogrgic/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:true OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]"
time=2025-06-05T21:10:01.165-04:00 level=INFO source=images.go:479 msg="total blobs: 16"
time=2025-06-05T21:10:01.165-04:00 level=INFO source=images.go:486 msg="total unused blobs removed: 0"
[GIN-debug] [WARNING] Creating an Engine instance with the Logger and Recovery middleware already attached.

[GIN-debug] [WARNING] Running in "debug" mode. Switch to "release" mode in production.
 - using env:	export GIN_MODE=release
 - using code:	gin.SetMode(gin.ReleaseMode)

[GIN-debug] HEAD   /                         --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func1 (5 handlers)
[GIN-debug] GET    /                         --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func2 (5 handlers)
[GIN-debug] HEAD   /api/version              --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func3 (5 handlers)
[GIN-debug] GET    /api/version              --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func4 (5 handlers)
[GIN-debug] POST   /api/pull                 --> github.com/ollama/ollama/server.(*Server).PullHandler-fm (5 handlers)
[GIN-debug] POST   /api/push                 --> github.com/ollama/ollama/server.(*Server).PushHandler-fm (5 handlers)
[GIN-debug] HEAD   /api/tags                 --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (5 handlers)
[GIN-debug] GET    /api/tags                 --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (5 handlers)
[GIN-debug] POST   /api/show                 --> github.com/ollama/ollama/server.(*Server).ShowHandler-fm (5 handlers)
[GIN-debug] DELETE /api/delete               --> github.com/ollama/ollama/server.(*Server).DeleteHandler-fm (5 handlers)
[GIN-debug] POST   /api/create               --> github.com/ollama/ollama/server.(*Server).CreateHandler-fm (5 handlers)
[GIN-debug] POST   /api/blobs/:digest        --> github.com/ollama/ollama/server.(*Server).CreateBlobHandler-fm (5 handlers)
[GIN-debug] HEAD   /api/blobs/:digest        --> github.com/ollama/ollama/server.(*Server).HeadBlobHandler-fm (5 handlers)
[GIN-debug] POST   /api/copy                 --> github.com/ollama/ollama/server.(*Server).CopyHandler-fm (5 handlers)
[GIN-debug] GET    /api/ps                   --> github.com/ollama/ollama/server.(*Server).PsHandler-fm (5 handlers)
[GIN-debug] POST   /api/generate             --> github.com/ollama/ollama/server.(*Server).GenerateHandler-fm (5 handlers)
[GIN-debug] POST   /api/chat                 --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (5 handlers)
[GIN-debug] POST   /api/embed                --> github.com/ollama/ollama/server.(*Server).EmbedHandler-fm (5 handlers)
[GIN-debug] POST   /api/embeddings           --> github.com/ollama/ollama/server.(*Server).EmbeddingsHandler-fm (5 handlers)
[GIN-debug] POST   /v1/chat/completions      --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (6 handlers)
[GIN-debug] POST   /v1/completions           --> github.com/ollama/ollama/server.(*Server).GenerateHandler-fm (6 handlers)
[GIN-debug] POST   /v1/embeddings            --> github.com/ollama/ollama/server.(*Server).EmbedHandler-fm (6 handlers)
[GIN-debug] GET    /v1/models                --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (6 handlers)
[GIN-debug] GET    /v1/models/:model         --> github.com/ollama/ollama/server.(*Server).ShowHandler-fm (6 handlers)
time=2025-06-05T21:10:01.166-04:00 level=INFO source=routes.go:1295 msg="Listening on 127.0.0.1:11434 (version 0.0.0)"
time=2025-06-05T21:10:01.195-04:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=metal variant="" compute="" driver=0.0 name="" total="96.0 GiB" available="96.0 GiB"
[GIN] 2025/06/05 - 21:10:06 | 200 |     222.334µs |       127.0.0.1 | HEAD     "/"
[GIN] 2025/06/05 - 21:10:06 | 200 |   49.765208ms |       127.0.0.1 | POST     "/api/show"
time=2025-06-05T21:10:06.725-04:00 level=INFO source=sched.go:788 msg="new model will fit in available VRAM in single GPU, loading" model=/Users/mariogrgic/.ollama/models/blobs/sha256-e796792eba26c4d3b04b0ac5adb01a453dd9ec2dfd83b6c59cbf6fe5f30b0f68 gpu=0 parallel=2 available=103079215104 required="20.1 GiB"
time=2025-06-05T21:10:06.725-04:00 level=INFO source=server.go:135 msg="system memory" total="128.0 GiB" free="117.4 GiB" free_swap="0 B"
time=2025-06-05T21:10:06.726-04:00 level=INFO source=server.go:168 msg=offload library=metal layers.requested=-1 layers.model=63 layers.offload=63 layers.split="" memory.available="[96.0 GiB]" memory.gpu_overhead="0 B" memory.required.full="20.1 GiB" memory.required.partial="20.1 GiB" memory.required.kv="1.6 GiB" memory.required.allocations="[20.1 GiB]" memory.weights.total="15.4 GiB" memory.weights.repeating="14.3 GiB" memory.weights.nonrepeating="1.1 GiB" memory.graph.full="565.0 MiB" memory.graph.partial="565.0 MiB" projector.weights="795.9 MiB" projector.graph="1.0 GiB"
time=2025-06-05T21:10:06.756-04:00 level=INFO source=server.go:431 msg="starting llama server" cmd="/Volumes/DATA/dev/ollama/ollama.git/ollama runner --ollama-engine --model /Users/mariogrgic/.ollama/models/blobs/sha256-e796792eba26c4d3b04b0ac5adb01a453dd9ec2dfd83b6c59cbf6fe5f30b0f68 --ctx-size 8192 --batch-size 512 --n-gpu-layers 63 --threads 12 --parallel 2 --port 49300"
time=2025-06-05T21:10:06.757-04:00 level=INFO source=sched.go:483 msg="loaded runners" count=1
time=2025-06-05T21:10:06.757-04:00 level=INFO source=server.go:591 msg="waiting for llama runner to start responding"
time=2025-06-05T21:10:06.757-04:00 level=INFO source=server.go:625 msg="waiting for server to become available" status="llm server not responding"
time=2025-06-05T21:10:06.765-04:00 level=INFO source=runner.go:925 msg="starting ollama engine"
time=2025-06-05T21:10:06.766-04:00 level=INFO source=runner.go:983 msg="Server listening on 127.0.0.1:49300"
time=2025-06-05T21:10:06.793-04:00 level=INFO source=ggml.go:92 msg="" architecture=gemma3 file_type=Q4_K_M name="" description="" num_tensors=1247 num_key_values=37
time=2025-06-05T21:10:06.793-04:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang)
unexpected fault address 0x2000000000
fatal error: fault
[signal SIGBUS: bus error code=0x1 addr=0x2000000000 pc=0x100edea94]

goroutine 14 gp=0x14000103a40 m=8 mp=0x14000600008 [running]:
runtime.throw({0x101c05000?, 0x0?})
	/usr/local/go/src/runtime/panic.go:1101 +0x38 fp=0x1400004d100 sp=0x1400004d0d0 pc=0x100f4c788
runtime.sigpanic()
	/usr/local/go/src/runtime/signal_unix.go:922 +0x170 fp=0x1400004d160 sp=0x1400004d100 pc=0x100f4ebf0
indexbytebody()
	/usr/local/go/src/internal/bytealg/indexbyte_arm64.s:74 +0x64 fp=0x1400004d170 sp=0x1400004d170 pc=0x100edea94
runtime.findnull(0x1400004d208?)
	/usr/local/go/src/runtime/string.go:577 +0x78 fp=0x1400004d1d0 sp=0x1400004d170 pc=0x100f35ba8
runtime.gostring(0x2000000000)
	/usr/local/go/src/runtime/string.go:363 +0x20 fp=0x1400004d210 sp=0x1400004d1d0 pc=0x100f4fa60
github.com/ollama/ollama/ml/backend/ggml._Cfunc_GoString(...)
	_cgo_gotypes.go:300
github.com/ollama/ollama/ml/backend/ggml.New({0x16ef26ca2, 0x6e}, {0xc, 0x0, 0x3f, {0x0, 0x0, 0x0}, 0x0})
	/Volumes/DATA/dev/ollama/ollama.git/ml/backend/ggml/ggml.go:141 +0xaa4 fp=0x1400004dc00 sp=0x1400004d210 pc=0x1013180d4
github.com/ollama/ollama/ml.NewBackend({0x16ef26ca2, 0x6e}, {0xc, 0x0, 0x3f, {0x0, 0x0, 0x0}, 0x0})
	/Volumes/DATA/dev/ollama/ollama.git/ml/backend.go:209 +0x90 fp=0x1400004dc60 sp=0x1400004dc00 pc=0x10130dc90
github.com/ollama/ollama/model.New({0x16ef26ca2?, 0x0?}, {0xc, 0x0, 0x3f, {0x0, 0x0, 0x0}, 0x0})
	/Volumes/DATA/dev/ollama/ollama.git/model/model.go:102 +0x5c fp=0x1400004dd60 sp=0x1400004dc60 pc=0x10132464c
github.com/ollama/ollama/runner/ollamarunner.(*Server).initModel(0x140004b9b00, {0x16ef26ca2?, 0x0?}, {0xc, 0x0, 0x3f, {0x0, 0x0, 0x0}, 0x0}, ...)
	/Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:841 +0x80 fp=0x1400004ddc0 sp=0x1400004dd60 pc=0x10139d3a0
github.com/ollama/ollama/runner/ollamarunner.(*Server).load(0x140004b9b00, {0x102076dd0, 0x14000390410}, {0x16ef26ca2?, 0x0?}, {0xc, 0x0, 0x3f, {0x0, 0x0, ...}, ...}, ...)
	/Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:878 +0xa8 fp=0x1400004df20 sp=0x1400004ddc0 pc=0x10139d6b8
github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1()
	/Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:959 +0x88 fp=0x1400004dfd0 sp=0x1400004df20 pc=0x10139e738
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400004dfd0 sp=0x1400004dfd0 pc=0x100f54c64
created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1
	/Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:959 +0x840

goroutine 1 gp=0x140000021c0 m=nil [IO wait]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x100f625b0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x140001335e0 sp=0x140001335c0 pc=0x100f4c8a8
runtime.netpollblock(0x140004b5678?, 0xfd1170?, 0x1?)
	/usr/local/go/src/runtime/netpoll.go:575 +0x158 fp=0x14000133620 sp=0x140001335e0 pc=0x100f12728
internal/poll.runtime_pollWait(0x149b16f30, 0x72)
	/usr/local/go/src/runtime/netpoll.go:351 +0xa0 fp=0x14000133650 sp=0x14000133620 pc=0x100f4ba60
internal/poll.(*pollDesc).wait(0x14000055780?, 0x101f3fb40?, 0x0)
	/usr/local/go/src/internal/poll/fd_poll_runtime.go:84 +0x28 fp=0x14000133680 sp=0x14000133650 pc=0x100fcc988
internal/poll.(*pollDesc).waitRead(...)
	/usr/local/go/src/internal/poll/fd_poll_runtime.go:89
internal/poll.(*FD).Accept(0x14000055780)
	/usr/local/go/src/internal/poll/fd_unix.go:620 +0x24c fp=0x14000133730 sp=0x14000133680 pc=0x100fd125c
net.(*netFD).accept(0x14000055780)
	/usr/local/go/src/net/fd_unix.go:172 +0x28 fp=0x140001337f0 sp=0x14000133730 pc=0x1010403c8
net.(*TCPListener).accept(0x1400043b100)
	/usr/local/go/src/net/tcpsock_posix.go:159 +0x24 fp=0x14000133840 sp=0x140001337f0 pc=0x101054624
net.(*TCPListener).Accept(0x1400043b100)
	/usr/local/go/src/net/tcpsock.go:380 +0x2c fp=0x14000133880 sp=0x14000133840 pc=0x10105360c
net/http.(*onceCloseListener).Accept(0x102076d60?)
	<autogenerated>:1 +0x30 fp=0x140001338a0 sp=0x14000133880 pc=0x10122eb40
net/http.(*Server).Serve(0x1400011f100, {0x102074978, 0x1400043b100})
	/usr/local/go/src/net/http/server.go:3424 +0x290 fp=0x140001339d0 sp=0x140001338a0 pc=0x1012081e0
github.com/ollama/ollama/runner/ollamarunner.Execute({0x14000000270, 0xe, 0xf})
	/Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:984 +0xb78 fp=0x14000133ce0 sp=0x140001339d0 pc=0x10139e408
github.com/ollama/ollama/runner.Execute({0x14000000250?, 0x0?, 0x0?})
	/Volumes/DATA/dev/ollama/ollama.git/runner/runner.go:20 +0x120 fp=0x14000133d10 sp=0x14000133ce0 pc=0x10139ecb0
github.com/ollama/ollama/cmd.NewCLI.func2(0x1400011ef00?, {0x101c03c5e?, 0x4?, 0x101c03c62?})
	/Volumes/DATA/dev/ollama/ollama.git/cmd/cmd.go:1529 +0x54 fp=0x14000133d40 sp=0x14000133d10 pc=0x1019e9984
github.com/spf13/cobra.(*Command).execute(0x140004e6f08, {0x14000434f00, 0xf, 0xf})
	/Volumes/DATA/dev/ollama/ollama.git/gopath/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:940 +0x648 fp=0x14000133e60 sp=0x14000133d40 pc=0x1010ae968
github.com/spf13/cobra.(*Command).ExecuteC(0x140004cc908)
	/Volumes/DATA/dev/ollama/ollama.git/gopath/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:1068 +0x320 fp=0x14000133f20 sp=0x14000133e60 pc=0x1010af0b0
github.com/spf13/cobra.(*Command).Execute(...)
	/Volumes/DATA/dev/ollama/ollama.git/gopath/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:992
github.com/spf13/cobra.(*Command).ExecuteContext(...)
	/Volumes/DATA/dev/ollama/ollama.git/gopath/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:985
main.main()
	/Volumes/DATA/dev/ollama/ollama.git/main.go:12 +0x54 fp=0x14000133f40 sp=0x14000133f20 pc=0x1019ea464
runtime.main()
	/usr/local/go/src/runtime/proc.go:283 +0x284 fp=0x14000133fd0 sp=0x14000133f40 pc=0x100f19244
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000133fd0 sp=0x14000133fd0 pc=0x100f54c64

goroutine 2 gp=0x14000002c40 m=nil [force gc (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000084f90 sp=0x14000084f70 pc=0x100f4c8a8
runtime.goparkunlock(...)
	/usr/local/go/src/runtime/proc.go:441
runtime.forcegchelper()
	/usr/local/go/src/runtime/proc.go:348 +0xb8 fp=0x14000084fd0 sp=0x14000084f90 pc=0x100f19598
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000084fd0 sp=0x14000084fd0 pc=0x100f54c64
created by runtime.init.7 in goroutine 1
	/usr/local/go/src/runtime/proc.go:336 +0x24

goroutine 3 gp=0x14000003180 m=nil [GC sweep wait]:
runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000085760 sp=0x14000085740 pc=0x100f4c8a8
runtime.goparkunlock(...)
	/usr/local/go/src/runtime/proc.go:441
runtime.bgsweep(0x140000b0000)
	/usr/local/go/src/runtime/mgcsweep.go:316 +0x108 fp=0x140000857b0 sp=0x14000085760 pc=0x100f046c8
runtime.gcenable.gowrap1()
	/usr/local/go/src/runtime/mgc.go:204 +0x28 fp=0x140000857d0 sp=0x140000857b0 pc=0x100ef84c8
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000857d0 sp=0x140000857d0 pc=0x100f54c64
created by runtime.gcenable in goroutine 1
	/usr/local/go/src/runtime/mgc.go:204 +0x6c

goroutine 4 gp=0x14000003340 m=nil [GC scavenge wait]:
runtime.gopark(0x10000?, 0x101db4f78?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000085f60 sp=0x14000085f40 pc=0x100f4c8a8
runtime.goparkunlock(...)
	/usr/local/go/src/runtime/proc.go:441
runtime.(*scavengerState).park(0x102917f40)
	/usr/local/go/src/runtime/mgcscavenge.go:425 +0x5c fp=0x14000085f90 sp=0x14000085f60 pc=0x100f0215c
runtime.bgscavenge(0x140000b0000)
	/usr/local/go/src/runtime/mgcscavenge.go:658 +0xac fp=0x14000085fb0 sp=0x14000085f90 pc=0x100f026fc
runtime.gcenable.gowrap2()
	/usr/local/go/src/runtime/mgc.go:205 +0x28 fp=0x14000085fd0 sp=0x14000085fb0 pc=0x100ef8468
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000085fd0 sp=0x14000085fd0 pc=0x100f54c64
created by runtime.gcenable in goroutine 1
	/usr/local/go/src/runtime/mgc.go:205 +0xac

goroutine 5 gp=0x14000003c00 m=nil [finalizer wait]:
runtime.gopark(0x18000845c8?, 0x1000000000000?, 0xf8?, 0x45?, 0x1012314ac?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000084590 sp=0x14000084570 pc=0x100f4c8a8
runtime.runfinq()
	/usr/local/go/src/runtime/mfinal.go:196 +0x108 fp=0x140000847d0 sp=0x14000084590 pc=0x100ef74c8
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000847d0 sp=0x140000847d0 pc=0x100f54c64
created by runtime.createfing in goroutine 1
	/usr/local/go/src/runtime/mfinal.go:166 +0x80

goroutine 6 gp=0x140001ee700 m=nil [chan receive]:
runtime.gopark(0x14000233680?, 0x1400000e1b0?, 0x48?, 0x67?, 0x101014698?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x140000866f0 sp=0x140000866d0 pc=0x100f4c8a8
runtime.chanrecv(0x140000be310, 0x0, 0x1)
	/usr/local/go/src/runtime/chan.go:664 +0x42c fp=0x14000086770 sp=0x140000866f0 pc=0x100ee9f7c
runtime.chanrecv1(0x0?, 0x0?)
	/usr/local/go/src/runtime/chan.go:506 +0x14 fp=0x140000867a0 sp=0x14000086770 pc=0x100ee9b14
runtime.unique_runtime_registerUniqueMapCleanup.func2(...)
	/usr/local/go/src/runtime/mgc.go:1796
runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1799 +0x3c fp=0x140000867d0 sp=0x140000867a0 pc=0x100efb6ec
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000867d0 sp=0x140000867d0 pc=0x100f54c64
created by unique.runtime_registerUniqueMapCleanup in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1794 +0x78

goroutine 7 gp=0x140001eee00 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000086f10 sp=0x14000086ef0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000086fb0 sp=0x14000086f10 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000086fd0 sp=0x14000086fb0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000086fd0 sp=0x14000086fd0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 18 gp=0x14000504000 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000080710 sp=0x140000806f0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140000807b0 sp=0x14000080710 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140000807d0 sp=0x140000807b0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000807d0 sp=0x140000807d0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 34 gp=0x14000102380 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011a710 sp=0x1400011a6f0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011a7b0 sp=0x1400011a710 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011a7d0 sp=0x1400011a7b0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011a7d0 sp=0x1400011a7d0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 35 gp=0x14000102540 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011af10 sp=0x1400011aef0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011afb0 sp=0x1400011af10 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011afd0 sp=0x1400011afb0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011afd0 sp=0x1400011afd0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 8 gp=0x140001eefc0 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000087710 sp=0x140000876f0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140000877b0 sp=0x14000087710 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140000877d0 sp=0x140000877b0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000877d0 sp=0x140000877d0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 19 gp=0x140005041c0 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000080f10 sp=0x14000080ef0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000080fb0 sp=0x14000080f10 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000080fd0 sp=0x14000080fb0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000080fd0 sp=0x14000080fd0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 20 gp=0x14000504380 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000081710 sp=0x140000816f0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140000817b0 sp=0x14000081710 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140000817d0 sp=0x140000817b0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000817d0 sp=0x140000817d0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 36 gp=0x14000102700 m=nil [GC worker (idle)]:
runtime.gopark(0x26313f85800?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011b710 sp=0x1400011b6f0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011b7b0 sp=0x1400011b710 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011b7d0 sp=0x1400011b7b0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011b7d0 sp=0x1400011b7d0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 37 gp=0x140001028c0 m=nil [GC worker (idle)]:
runtime.gopark(0x102948780?, 0x1?, 0xc?, 0xa8?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011bf10 sp=0x1400011bef0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011bfb0 sp=0x1400011bf10 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011bfd0 sp=0x1400011bfb0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011bfd0 sp=0x1400011bfd0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 38 gp=0x14000102a80 m=nil [GC worker (idle)]:
runtime.gopark(0x26313daa359?, 0x0?, 0x0?, 0x0?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011c710 sp=0x1400011c6f0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011c7b0 sp=0x1400011c710 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011c7d0 sp=0x1400011c7b0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011c7d0 sp=0x1400011c7d0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 39 gp=0x14000102c40 m=nil [GC worker (idle)]:
runtime.gopark(0x26313da80ae?, 0x3?, 0x5e?, 0x1b?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011cf10 sp=0x1400011cef0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011cfb0 sp=0x1400011cf10 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011cfd0 sp=0x1400011cfb0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011cfd0 sp=0x1400011cfd0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 9 gp=0x140001ef180 m=nil [GC worker (idle)]:
runtime.gopark(0x26313f87e6a?, 0x1?, 0x18?, 0x5f?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000087f10 sp=0x14000087ef0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000087fb0 sp=0x14000087f10 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000087fd0 sp=0x14000087fb0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000087fd0 sp=0x14000087fd0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 10 gp=0x140001ef340 m=nil [GC worker (idle)]:
runtime.gopark(0x26313da7de9?, 0x3?, 0x53?, 0x8?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000116710 sp=0x140001166f0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140001167b0 sp=0x14000116710 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140001167d0 sp=0x140001167b0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140001167d0 sp=0x140001167d0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 11 gp=0x140001ef500 m=nil [GC worker (idle)]:
runtime.gopark(0x26313da7643?, 0x1?, 0x12?, 0xe3?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000116f10 sp=0x14000116ef0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000116fb0 sp=0x14000116f10 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000116fd0 sp=0x14000116fb0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000116fd0 sp=0x14000116fd0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 12 gp=0x140001ef6c0 m=nil [GC worker (idle)]:
runtime.gopark(0x26313da7a2b?, 0x1?, 0x6f?, 0xa9?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000117710 sp=0x140001176f0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140001177b0 sp=0x14000117710 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140001177d0 sp=0x140001177b0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140001177d0 sp=0x140001177d0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 13 gp=0x140001ef880 m=nil [GC worker (idle)]:
runtime.gopark(0x26313da7b4f?, 0x1?, 0xc3?, 0xb?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000117f10 sp=0x14000117ef0 pc=0x100f4c8a8
runtime.gcBgMarkWorker(0x140000bf730)
	/usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000117fb0 sp=0x14000117f10 pc=0x100efa95c
runtime.gcBgMarkStartWorkers.gowrap1()
	/usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000117fd0 sp=0x14000117fb0 pc=0x100efa848
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000117fd0 sp=0x14000117fd0 pc=0x100f54c64
created by runtime.gcBgMarkStartWorkers in goroutine 1
	/usr/local/go/src/runtime/mgc.go:1339 +0x140

goroutine 15 gp=0x14000103c00 m=nil [sync.WaitGroup.Wait]:
runtime.gopark(0x1029266a0?, 0x0?, 0x60?, 0x20?, 0x0?)
	/usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x140001196d0 sp=0x140001196b0 pc=0x100f4c8a8
runtime.goparkunlock(...)
	/usr/local/go/src/runtime/proc.go:441
runtime.semacquire1(0x140004b9b08, 0x0, 0x1, 0x0, 0x18)
	/usr/local/go/src/runtime/sema.go:188 +0x204 fp=0x14000119720 sp=0x140001196d0 pc=0x100f2d724
sync.runtime_SemacquireWaitGroup(0x0?)
	/usr/local/go/src/runtime/sema.go:110 +0x2c fp=0x14000119760 sp=0x14000119720 pc=0x100f4e31c
sync.(*WaitGroup).Wait(0x140004b9b00)
	/usr/local/go/src/sync/waitgroup.go:118 +0x70 fp=0x14000119780 sp=0x14000119760 pc=0x100f60540
github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0x140004b9b00, {0x102076dd0, 0x14000390410})
	/Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:355 +0x2c fp=0x140001197a0 sp=0x14000119780 pc=0x101399e9c
github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap2()
	/Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:960 +0x30 fp=0x140001197d0 sp=0x140001197a0 pc=0x10139e670
runtime.goexit({})
	/usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140001197d0 sp=0x140001197d0 pc=0x100f54c64
created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1
	/Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:960 +0x898
time=2025-06-05T21:10:06.799-04:00 level=ERROR source=server.go:457 msg="llama runner terminated" error="exit status 2"
time=2025-06-05T21:10:07.007-04:00 level=ERROR source=sched.go:489 msg="error loading llama server" error="llama runner process has terminated: error:fault"
[GIN] 2025/06/05 - 21:10:07 | 500 |   335.97975ms |       127.0.0.1 | POST     "/api/generate"

Relevant log output


OS

macOS

GPU

Apple

CPU

Apple

Ollama version

No response

Originally created by @mario-grgic on GitHub (Jun 6, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10991 ### What is the issue? ollama crashes when running gemma3:27b. I did git bisect and it looks like the behaviour was introduced in the following commit: ``` aaa7818000c42a82fc030212c35ef83f9799efd7 is the first bad commit commit aaa7818000c42a82fc030212c35ef83f9799efd7 Author: Jesse Gross <jesse@ollama.com> Date: Thu Apr 24 11:48:49 2025 -0700 ggml: Export GPU UUIDs This enables matching up devices and information reported by the backend with system management libraries such as nvml to get accurate free memory reporting. llama/patches/0017-ggml-Export-GPU-UUIDs.patch | 102 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ ml/backend.go | 8 +++++ ml/backend/ggml/ggml.go | 6 ++++ ml/backend/ggml/ggml/include/ggml-backend.h | 1 + ml/backend/ggml/ggml/src/ggml-cuda/ggml-cuda.cu | 33 +++++++++++++++++++++ ml/backend/ggml/ggml/src/ggml-metal/ggml-metal.m | 1 + 6 files changed, 151 insertions(+) create mode 100644 llama/patches/0017-ggml-Export-GPU-UUIDs.patch ``` tag 0.9.0 works fine. Here is the crash log: ``` $ ./ollama serve time=2025-06-05T21:10:01.164-04:00 level=INFO source=routes.go:1242 msg="server config" env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Users/mariogrgic/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:true OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]" time=2025-06-05T21:10:01.165-04:00 level=INFO source=images.go:479 msg="total blobs: 16" time=2025-06-05T21:10:01.165-04:00 level=INFO source=images.go:486 msg="total unused blobs removed: 0" [GIN-debug] [WARNING] Creating an Engine instance with the Logger and Recovery middleware already attached. [GIN-debug] [WARNING] Running in "debug" mode. Switch to "release" mode in production. - using env: export GIN_MODE=release - using code: gin.SetMode(gin.ReleaseMode) [GIN-debug] HEAD / --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func1 (5 handlers) [GIN-debug] GET / --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func2 (5 handlers) [GIN-debug] HEAD /api/version --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func3 (5 handlers) [GIN-debug] GET /api/version --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func4 (5 handlers) [GIN-debug] POST /api/pull --> github.com/ollama/ollama/server.(*Server).PullHandler-fm (5 handlers) [GIN-debug] POST /api/push --> github.com/ollama/ollama/server.(*Server).PushHandler-fm (5 handlers) [GIN-debug] HEAD /api/tags --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (5 handlers) [GIN-debug] GET /api/tags --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (5 handlers) [GIN-debug] POST /api/show --> github.com/ollama/ollama/server.(*Server).ShowHandler-fm (5 handlers) [GIN-debug] DELETE /api/delete --> github.com/ollama/ollama/server.(*Server).DeleteHandler-fm (5 handlers) [GIN-debug] POST /api/create --> github.com/ollama/ollama/server.(*Server).CreateHandler-fm (5 handlers) [GIN-debug] POST /api/blobs/:digest --> github.com/ollama/ollama/server.(*Server).CreateBlobHandler-fm (5 handlers) [GIN-debug] HEAD /api/blobs/:digest --> github.com/ollama/ollama/server.(*Server).HeadBlobHandler-fm (5 handlers) [GIN-debug] POST /api/copy --> github.com/ollama/ollama/server.(*Server).CopyHandler-fm (5 handlers) [GIN-debug] GET /api/ps --> github.com/ollama/ollama/server.(*Server).PsHandler-fm (5 handlers) [GIN-debug] POST /api/generate --> github.com/ollama/ollama/server.(*Server).GenerateHandler-fm (5 handlers) [GIN-debug] POST /api/chat --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (5 handlers) [GIN-debug] POST /api/embed --> github.com/ollama/ollama/server.(*Server).EmbedHandler-fm (5 handlers) [GIN-debug] POST /api/embeddings --> github.com/ollama/ollama/server.(*Server).EmbeddingsHandler-fm (5 handlers) [GIN-debug] POST /v1/chat/completions --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (6 handlers) [GIN-debug] POST /v1/completions --> github.com/ollama/ollama/server.(*Server).GenerateHandler-fm (6 handlers) [GIN-debug] POST /v1/embeddings --> github.com/ollama/ollama/server.(*Server).EmbedHandler-fm (6 handlers) [GIN-debug] GET /v1/models --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (6 handlers) [GIN-debug] GET /v1/models/:model --> github.com/ollama/ollama/server.(*Server).ShowHandler-fm (6 handlers) time=2025-06-05T21:10:01.166-04:00 level=INFO source=routes.go:1295 msg="Listening on 127.0.0.1:11434 (version 0.0.0)" time=2025-06-05T21:10:01.195-04:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=metal variant="" compute="" driver=0.0 name="" total="96.0 GiB" available="96.0 GiB" [GIN] 2025/06/05 - 21:10:06 | 200 | 222.334µs | 127.0.0.1 | HEAD "/" [GIN] 2025/06/05 - 21:10:06 | 200 | 49.765208ms | 127.0.0.1 | POST "/api/show" time=2025-06-05T21:10:06.725-04:00 level=INFO source=sched.go:788 msg="new model will fit in available VRAM in single GPU, loading" model=/Users/mariogrgic/.ollama/models/blobs/sha256-e796792eba26c4d3b04b0ac5adb01a453dd9ec2dfd83b6c59cbf6fe5f30b0f68 gpu=0 parallel=2 available=103079215104 required="20.1 GiB" time=2025-06-05T21:10:06.725-04:00 level=INFO source=server.go:135 msg="system memory" total="128.0 GiB" free="117.4 GiB" free_swap="0 B" time=2025-06-05T21:10:06.726-04:00 level=INFO source=server.go:168 msg=offload library=metal layers.requested=-1 layers.model=63 layers.offload=63 layers.split="" memory.available="[96.0 GiB]" memory.gpu_overhead="0 B" memory.required.full="20.1 GiB" memory.required.partial="20.1 GiB" memory.required.kv="1.6 GiB" memory.required.allocations="[20.1 GiB]" memory.weights.total="15.4 GiB" memory.weights.repeating="14.3 GiB" memory.weights.nonrepeating="1.1 GiB" memory.graph.full="565.0 MiB" memory.graph.partial="565.0 MiB" projector.weights="795.9 MiB" projector.graph="1.0 GiB" time=2025-06-05T21:10:06.756-04:00 level=INFO source=server.go:431 msg="starting llama server" cmd="/Volumes/DATA/dev/ollama/ollama.git/ollama runner --ollama-engine --model /Users/mariogrgic/.ollama/models/blobs/sha256-e796792eba26c4d3b04b0ac5adb01a453dd9ec2dfd83b6c59cbf6fe5f30b0f68 --ctx-size 8192 --batch-size 512 --n-gpu-layers 63 --threads 12 --parallel 2 --port 49300" time=2025-06-05T21:10:06.757-04:00 level=INFO source=sched.go:483 msg="loaded runners" count=1 time=2025-06-05T21:10:06.757-04:00 level=INFO source=server.go:591 msg="waiting for llama runner to start responding" time=2025-06-05T21:10:06.757-04:00 level=INFO source=server.go:625 msg="waiting for server to become available" status="llm server not responding" time=2025-06-05T21:10:06.765-04:00 level=INFO source=runner.go:925 msg="starting ollama engine" time=2025-06-05T21:10:06.766-04:00 level=INFO source=runner.go:983 msg="Server listening on 127.0.0.1:49300" time=2025-06-05T21:10:06.793-04:00 level=INFO source=ggml.go:92 msg="" architecture=gemma3 file_type=Q4_K_M name="" description="" num_tensors=1247 num_key_values=37 time=2025-06-05T21:10:06.793-04:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang) unexpected fault address 0x2000000000 fatal error: fault [signal SIGBUS: bus error code=0x1 addr=0x2000000000 pc=0x100edea94] goroutine 14 gp=0x14000103a40 m=8 mp=0x14000600008 [running]: runtime.throw({0x101c05000?, 0x0?}) /usr/local/go/src/runtime/panic.go:1101 +0x38 fp=0x1400004d100 sp=0x1400004d0d0 pc=0x100f4c788 runtime.sigpanic() /usr/local/go/src/runtime/signal_unix.go:922 +0x170 fp=0x1400004d160 sp=0x1400004d100 pc=0x100f4ebf0 indexbytebody() /usr/local/go/src/internal/bytealg/indexbyte_arm64.s:74 +0x64 fp=0x1400004d170 sp=0x1400004d170 pc=0x100edea94 runtime.findnull(0x1400004d208?) /usr/local/go/src/runtime/string.go:577 +0x78 fp=0x1400004d1d0 sp=0x1400004d170 pc=0x100f35ba8 runtime.gostring(0x2000000000) /usr/local/go/src/runtime/string.go:363 +0x20 fp=0x1400004d210 sp=0x1400004d1d0 pc=0x100f4fa60 github.com/ollama/ollama/ml/backend/ggml._Cfunc_GoString(...) _cgo_gotypes.go:300 github.com/ollama/ollama/ml/backend/ggml.New({0x16ef26ca2, 0x6e}, {0xc, 0x0, 0x3f, {0x0, 0x0, 0x0}, 0x0}) /Volumes/DATA/dev/ollama/ollama.git/ml/backend/ggml/ggml.go:141 +0xaa4 fp=0x1400004dc00 sp=0x1400004d210 pc=0x1013180d4 github.com/ollama/ollama/ml.NewBackend({0x16ef26ca2, 0x6e}, {0xc, 0x0, 0x3f, {0x0, 0x0, 0x0}, 0x0}) /Volumes/DATA/dev/ollama/ollama.git/ml/backend.go:209 +0x90 fp=0x1400004dc60 sp=0x1400004dc00 pc=0x10130dc90 github.com/ollama/ollama/model.New({0x16ef26ca2?, 0x0?}, {0xc, 0x0, 0x3f, {0x0, 0x0, 0x0}, 0x0}) /Volumes/DATA/dev/ollama/ollama.git/model/model.go:102 +0x5c fp=0x1400004dd60 sp=0x1400004dc60 pc=0x10132464c github.com/ollama/ollama/runner/ollamarunner.(*Server).initModel(0x140004b9b00, {0x16ef26ca2?, 0x0?}, {0xc, 0x0, 0x3f, {0x0, 0x0, 0x0}, 0x0}, ...) /Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:841 +0x80 fp=0x1400004ddc0 sp=0x1400004dd60 pc=0x10139d3a0 github.com/ollama/ollama/runner/ollamarunner.(*Server).load(0x140004b9b00, {0x102076dd0, 0x14000390410}, {0x16ef26ca2?, 0x0?}, {0xc, 0x0, 0x3f, {0x0, 0x0, ...}, ...}, ...) /Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:878 +0xa8 fp=0x1400004df20 sp=0x1400004ddc0 pc=0x10139d6b8 github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1() /Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:959 +0x88 fp=0x1400004dfd0 sp=0x1400004df20 pc=0x10139e738 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400004dfd0 sp=0x1400004dfd0 pc=0x100f54c64 created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1 /Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:959 +0x840 goroutine 1 gp=0x140000021c0 m=nil [IO wait]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x100f625b0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x140001335e0 sp=0x140001335c0 pc=0x100f4c8a8 runtime.netpollblock(0x140004b5678?, 0xfd1170?, 0x1?) /usr/local/go/src/runtime/netpoll.go:575 +0x158 fp=0x14000133620 sp=0x140001335e0 pc=0x100f12728 internal/poll.runtime_pollWait(0x149b16f30, 0x72) /usr/local/go/src/runtime/netpoll.go:351 +0xa0 fp=0x14000133650 sp=0x14000133620 pc=0x100f4ba60 internal/poll.(*pollDesc).wait(0x14000055780?, 0x101f3fb40?, 0x0) /usr/local/go/src/internal/poll/fd_poll_runtime.go:84 +0x28 fp=0x14000133680 sp=0x14000133650 pc=0x100fcc988 internal/poll.(*pollDesc).waitRead(...) /usr/local/go/src/internal/poll/fd_poll_runtime.go:89 internal/poll.(*FD).Accept(0x14000055780) /usr/local/go/src/internal/poll/fd_unix.go:620 +0x24c fp=0x14000133730 sp=0x14000133680 pc=0x100fd125c net.(*netFD).accept(0x14000055780) /usr/local/go/src/net/fd_unix.go:172 +0x28 fp=0x140001337f0 sp=0x14000133730 pc=0x1010403c8 net.(*TCPListener).accept(0x1400043b100) /usr/local/go/src/net/tcpsock_posix.go:159 +0x24 fp=0x14000133840 sp=0x140001337f0 pc=0x101054624 net.(*TCPListener).Accept(0x1400043b100) /usr/local/go/src/net/tcpsock.go:380 +0x2c fp=0x14000133880 sp=0x14000133840 pc=0x10105360c net/http.(*onceCloseListener).Accept(0x102076d60?) <autogenerated>:1 +0x30 fp=0x140001338a0 sp=0x14000133880 pc=0x10122eb40 net/http.(*Server).Serve(0x1400011f100, {0x102074978, 0x1400043b100}) /usr/local/go/src/net/http/server.go:3424 +0x290 fp=0x140001339d0 sp=0x140001338a0 pc=0x1012081e0 github.com/ollama/ollama/runner/ollamarunner.Execute({0x14000000270, 0xe, 0xf}) /Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:984 +0xb78 fp=0x14000133ce0 sp=0x140001339d0 pc=0x10139e408 github.com/ollama/ollama/runner.Execute({0x14000000250?, 0x0?, 0x0?}) /Volumes/DATA/dev/ollama/ollama.git/runner/runner.go:20 +0x120 fp=0x14000133d10 sp=0x14000133ce0 pc=0x10139ecb0 github.com/ollama/ollama/cmd.NewCLI.func2(0x1400011ef00?, {0x101c03c5e?, 0x4?, 0x101c03c62?}) /Volumes/DATA/dev/ollama/ollama.git/cmd/cmd.go:1529 +0x54 fp=0x14000133d40 sp=0x14000133d10 pc=0x1019e9984 github.com/spf13/cobra.(*Command).execute(0x140004e6f08, {0x14000434f00, 0xf, 0xf}) /Volumes/DATA/dev/ollama/ollama.git/gopath/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:940 +0x648 fp=0x14000133e60 sp=0x14000133d40 pc=0x1010ae968 github.com/spf13/cobra.(*Command).ExecuteC(0x140004cc908) /Volumes/DATA/dev/ollama/ollama.git/gopath/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:1068 +0x320 fp=0x14000133f20 sp=0x14000133e60 pc=0x1010af0b0 github.com/spf13/cobra.(*Command).Execute(...) /Volumes/DATA/dev/ollama/ollama.git/gopath/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:992 github.com/spf13/cobra.(*Command).ExecuteContext(...) /Volumes/DATA/dev/ollama/ollama.git/gopath/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:985 main.main() /Volumes/DATA/dev/ollama/ollama.git/main.go:12 +0x54 fp=0x14000133f40 sp=0x14000133f20 pc=0x1019ea464 runtime.main() /usr/local/go/src/runtime/proc.go:283 +0x284 fp=0x14000133fd0 sp=0x14000133f40 pc=0x100f19244 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000133fd0 sp=0x14000133fd0 pc=0x100f54c64 goroutine 2 gp=0x14000002c40 m=nil [force gc (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000084f90 sp=0x14000084f70 pc=0x100f4c8a8 runtime.goparkunlock(...) /usr/local/go/src/runtime/proc.go:441 runtime.forcegchelper() /usr/local/go/src/runtime/proc.go:348 +0xb8 fp=0x14000084fd0 sp=0x14000084f90 pc=0x100f19598 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000084fd0 sp=0x14000084fd0 pc=0x100f54c64 created by runtime.init.7 in goroutine 1 /usr/local/go/src/runtime/proc.go:336 +0x24 goroutine 3 gp=0x14000003180 m=nil [GC sweep wait]: runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000085760 sp=0x14000085740 pc=0x100f4c8a8 runtime.goparkunlock(...) /usr/local/go/src/runtime/proc.go:441 runtime.bgsweep(0x140000b0000) /usr/local/go/src/runtime/mgcsweep.go:316 +0x108 fp=0x140000857b0 sp=0x14000085760 pc=0x100f046c8 runtime.gcenable.gowrap1() /usr/local/go/src/runtime/mgc.go:204 +0x28 fp=0x140000857d0 sp=0x140000857b0 pc=0x100ef84c8 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000857d0 sp=0x140000857d0 pc=0x100f54c64 created by runtime.gcenable in goroutine 1 /usr/local/go/src/runtime/mgc.go:204 +0x6c goroutine 4 gp=0x14000003340 m=nil [GC scavenge wait]: runtime.gopark(0x10000?, 0x101db4f78?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000085f60 sp=0x14000085f40 pc=0x100f4c8a8 runtime.goparkunlock(...) /usr/local/go/src/runtime/proc.go:441 runtime.(*scavengerState).park(0x102917f40) /usr/local/go/src/runtime/mgcscavenge.go:425 +0x5c fp=0x14000085f90 sp=0x14000085f60 pc=0x100f0215c runtime.bgscavenge(0x140000b0000) /usr/local/go/src/runtime/mgcscavenge.go:658 +0xac fp=0x14000085fb0 sp=0x14000085f90 pc=0x100f026fc runtime.gcenable.gowrap2() /usr/local/go/src/runtime/mgc.go:205 +0x28 fp=0x14000085fd0 sp=0x14000085fb0 pc=0x100ef8468 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000085fd0 sp=0x14000085fd0 pc=0x100f54c64 created by runtime.gcenable in goroutine 1 /usr/local/go/src/runtime/mgc.go:205 +0xac goroutine 5 gp=0x14000003c00 m=nil [finalizer wait]: runtime.gopark(0x18000845c8?, 0x1000000000000?, 0xf8?, 0x45?, 0x1012314ac?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000084590 sp=0x14000084570 pc=0x100f4c8a8 runtime.runfinq() /usr/local/go/src/runtime/mfinal.go:196 +0x108 fp=0x140000847d0 sp=0x14000084590 pc=0x100ef74c8 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000847d0 sp=0x140000847d0 pc=0x100f54c64 created by runtime.createfing in goroutine 1 /usr/local/go/src/runtime/mfinal.go:166 +0x80 goroutine 6 gp=0x140001ee700 m=nil [chan receive]: runtime.gopark(0x14000233680?, 0x1400000e1b0?, 0x48?, 0x67?, 0x101014698?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x140000866f0 sp=0x140000866d0 pc=0x100f4c8a8 runtime.chanrecv(0x140000be310, 0x0, 0x1) /usr/local/go/src/runtime/chan.go:664 +0x42c fp=0x14000086770 sp=0x140000866f0 pc=0x100ee9f7c runtime.chanrecv1(0x0?, 0x0?) /usr/local/go/src/runtime/chan.go:506 +0x14 fp=0x140000867a0 sp=0x14000086770 pc=0x100ee9b14 runtime.unique_runtime_registerUniqueMapCleanup.func2(...) /usr/local/go/src/runtime/mgc.go:1796 runtime.unique_runtime_registerUniqueMapCleanup.gowrap1() /usr/local/go/src/runtime/mgc.go:1799 +0x3c fp=0x140000867d0 sp=0x140000867a0 pc=0x100efb6ec runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000867d0 sp=0x140000867d0 pc=0x100f54c64 created by unique.runtime_registerUniqueMapCleanup in goroutine 1 /usr/local/go/src/runtime/mgc.go:1794 +0x78 goroutine 7 gp=0x140001eee00 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000086f10 sp=0x14000086ef0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000086fb0 sp=0x14000086f10 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000086fd0 sp=0x14000086fb0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000086fd0 sp=0x14000086fd0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 18 gp=0x14000504000 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000080710 sp=0x140000806f0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140000807b0 sp=0x14000080710 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140000807d0 sp=0x140000807b0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000807d0 sp=0x140000807d0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 34 gp=0x14000102380 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011a710 sp=0x1400011a6f0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011a7b0 sp=0x1400011a710 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011a7d0 sp=0x1400011a7b0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011a7d0 sp=0x1400011a7d0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 35 gp=0x14000102540 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011af10 sp=0x1400011aef0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011afb0 sp=0x1400011af10 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011afd0 sp=0x1400011afb0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011afd0 sp=0x1400011afd0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 8 gp=0x140001eefc0 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000087710 sp=0x140000876f0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140000877b0 sp=0x14000087710 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140000877d0 sp=0x140000877b0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000877d0 sp=0x140000877d0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 19 gp=0x140005041c0 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000080f10 sp=0x14000080ef0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000080fb0 sp=0x14000080f10 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000080fd0 sp=0x14000080fb0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000080fd0 sp=0x14000080fd0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 20 gp=0x14000504380 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000081710 sp=0x140000816f0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140000817b0 sp=0x14000081710 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140000817d0 sp=0x140000817b0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000817d0 sp=0x140000817d0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 36 gp=0x14000102700 m=nil [GC worker (idle)]: runtime.gopark(0x26313f85800?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011b710 sp=0x1400011b6f0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011b7b0 sp=0x1400011b710 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011b7d0 sp=0x1400011b7b0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011b7d0 sp=0x1400011b7d0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 37 gp=0x140001028c0 m=nil [GC worker (idle)]: runtime.gopark(0x102948780?, 0x1?, 0xc?, 0xa8?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011bf10 sp=0x1400011bef0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011bfb0 sp=0x1400011bf10 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011bfd0 sp=0x1400011bfb0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011bfd0 sp=0x1400011bfd0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 38 gp=0x14000102a80 m=nil [GC worker (idle)]: runtime.gopark(0x26313daa359?, 0x0?, 0x0?, 0x0?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011c710 sp=0x1400011c6f0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011c7b0 sp=0x1400011c710 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011c7d0 sp=0x1400011c7b0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011c7d0 sp=0x1400011c7d0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 39 gp=0x14000102c40 m=nil [GC worker (idle)]: runtime.gopark(0x26313da80ae?, 0x3?, 0x5e?, 0x1b?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x1400011cf10 sp=0x1400011cef0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x1400011cfb0 sp=0x1400011cf10 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x1400011cfd0 sp=0x1400011cfb0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400011cfd0 sp=0x1400011cfd0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 9 gp=0x140001ef180 m=nil [GC worker (idle)]: runtime.gopark(0x26313f87e6a?, 0x1?, 0x18?, 0x5f?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000087f10 sp=0x14000087ef0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000087fb0 sp=0x14000087f10 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000087fd0 sp=0x14000087fb0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000087fd0 sp=0x14000087fd0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 10 gp=0x140001ef340 m=nil [GC worker (idle)]: runtime.gopark(0x26313da7de9?, 0x3?, 0x53?, 0x8?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000116710 sp=0x140001166f0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140001167b0 sp=0x14000116710 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140001167d0 sp=0x140001167b0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140001167d0 sp=0x140001167d0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 11 gp=0x140001ef500 m=nil [GC worker (idle)]: runtime.gopark(0x26313da7643?, 0x1?, 0x12?, 0xe3?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000116f10 sp=0x14000116ef0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000116fb0 sp=0x14000116f10 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000116fd0 sp=0x14000116fb0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000116fd0 sp=0x14000116fd0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 12 gp=0x140001ef6c0 m=nil [GC worker (idle)]: runtime.gopark(0x26313da7a2b?, 0x1?, 0x6f?, 0xa9?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000117710 sp=0x140001176f0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x140001177b0 sp=0x14000117710 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x140001177d0 sp=0x140001177b0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140001177d0 sp=0x140001177d0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 13 gp=0x140001ef880 m=nil [GC worker (idle)]: runtime.gopark(0x26313da7b4f?, 0x1?, 0xc3?, 0xb?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x14000117f10 sp=0x14000117ef0 pc=0x100f4c8a8 runtime.gcBgMarkWorker(0x140000bf730) /usr/local/go/src/runtime/mgc.go:1423 +0xdc fp=0x14000117fb0 sp=0x14000117f10 pc=0x100efa95c runtime.gcBgMarkStartWorkers.gowrap1() /usr/local/go/src/runtime/mgc.go:1339 +0x28 fp=0x14000117fd0 sp=0x14000117fb0 pc=0x100efa848 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000117fd0 sp=0x14000117fd0 pc=0x100f54c64 created by runtime.gcBgMarkStartWorkers in goroutine 1 /usr/local/go/src/runtime/mgc.go:1339 +0x140 goroutine 15 gp=0x14000103c00 m=nil [sync.WaitGroup.Wait]: runtime.gopark(0x1029266a0?, 0x0?, 0x60?, 0x20?, 0x0?) /usr/local/go/src/runtime/proc.go:435 +0xc8 fp=0x140001196d0 sp=0x140001196b0 pc=0x100f4c8a8 runtime.goparkunlock(...) /usr/local/go/src/runtime/proc.go:441 runtime.semacquire1(0x140004b9b08, 0x0, 0x1, 0x0, 0x18) /usr/local/go/src/runtime/sema.go:188 +0x204 fp=0x14000119720 sp=0x140001196d0 pc=0x100f2d724 sync.runtime_SemacquireWaitGroup(0x0?) /usr/local/go/src/runtime/sema.go:110 +0x2c fp=0x14000119760 sp=0x14000119720 pc=0x100f4e31c sync.(*WaitGroup).Wait(0x140004b9b00) /usr/local/go/src/sync/waitgroup.go:118 +0x70 fp=0x14000119780 sp=0x14000119760 pc=0x100f60540 github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0x140004b9b00, {0x102076dd0, 0x14000390410}) /Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:355 +0x2c fp=0x140001197a0 sp=0x14000119780 pc=0x101399e9c github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap2() /Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:960 +0x30 fp=0x140001197d0 sp=0x140001197a0 pc=0x10139e670 runtime.goexit({}) /usr/local/go/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140001197d0 sp=0x140001197d0 pc=0x100f54c64 created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1 /Volumes/DATA/dev/ollama/ollama.git/runner/ollamarunner/runner.go:960 +0x898 time=2025-06-05T21:10:06.799-04:00 level=ERROR source=server.go:457 msg="llama runner terminated" error="exit status 2" time=2025-06-05T21:10:07.007-04:00 level=ERROR source=sched.go:489 msg="error loading llama server" error="llama runner process has terminated: error:fault" [GIN] 2025/06/05 - 21:10:07 | 500 | 335.97975ms | 127.0.0.1 | POST "/api/generate" ``` ### Relevant log output ```shell ``` ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-29 04:42:14 -05:00
Author
Owner

@mario-grgic commented on GitHub (Jun 6, 2025):

Turns out I am building with go 1.24.2. Once I upgraded golang to 1.24.4, there is no crash.

<!-- gh-comment-id:2947657098 --> @mario-grgic commented on GitHub (Jun 6, 2025): Turns out I am building with go 1.24.2. Once I upgraded golang to 1.24.4, there is no crash.
Author
Owner

@jmorganca commented on GitHub (Jun 6, 2025):

@mario-grgic Thanks for the issue. Some changes don't break the go build cache properly and so a go clean -cache is required. Upgrading Go must have fixed this.

<!-- gh-comment-id:2947956058 --> @jmorganca commented on GitHub (Jun 6, 2025): @mario-grgic Thanks for the issue. Some changes don't break the go build cache properly and so a `go clean -cache` is required. Upgrading Go must have fixed this.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#53759