github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-09 01:59:50 -05:00
Code Issues 14.6k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #14253] Cannot run any model - Ollama incompatible with kernel 6.19? #71339

New Issue
Closed
opened 2026-05-05 01:15:23 -05:00 by GiteaMirror · 24 comments
GiteaMirror commented 2026-05-05 01:15:23 -05:00
Owner
Copy Link

Originally created by @ndrewpj on GitHub (Feb 14, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/14253

What is the issue?

I cannot load and run models in ollama in docker. I have AMD Strix halo, according to the ollama log on loading the increased VRAM size is visible to ollama:

[GIN] 2026/02/14 - 12:51:23 | 500 | 816.630011ms | 127.0.0.1 | POST "/api/embed"

time=2026-02-14T12:53:02.765Z level=INFO source=routes.go:1636 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:8192 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE:q8_0 OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:2 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 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_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]"

time=2026-02-14T12:53:02.766Z level=INFO source=images.go:473 msg="total blobs: 30"

time=2026-02-14T12:53:02.766Z level=INFO source=images.go:480 msg="total unused blobs removed: 0"

time=2026-02-14T12:53:02.766Z level=INFO source=routes.go:1689 msg="Listening on [::]:11434 (version 0.16.1)"

time=2026-02-14T12:53:02.767Z level=INFO source=runner.go:67 msg="discovering available GPUs..."

time=2026-02-14T12:53:02.767Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 39845"

time=2026-02-14T12:53:03.120Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 38575"

time=2026-02-14T12:53:03.530Z level=INFO source=types.go:42 msg="inference compute" id=0 filter_id=0 library=ROCm compute=gfx1151 name=ROCm0 description="AMD Radeon Graphics" libdirs=ollama,rocm driver=60342.13 pci_id=0000:c5:00.0 type=iGPU total="120.4 GiB" available="118.3 GiB"

time=2026-02-14T12:53:03.530Z level=INFO source=routes.go:1739 msg="vram-based default context" total_vram="120.4 GiB" default_num_ctx=262144

Relevant log output

runtime/mgc.go:1339 +0x105

goroutine 53 gp=0xc000120540 m=nil [GC worker (idle)]:

runtime.gopark(0x55ef24d58560?, 0x1?, 0xc1?, 0x22?, 0x0?)

	runtime/proc.go:435 +0xce fp=0xc000119f38 sp=0xc000119f18 pc=0x55ef22a89f4e

runtime.gcBgMarkWorker(0xc00003f7a0)

	runtime/mgc.go:1423 +0xe9 fp=0xc000119fc8 sp=0xc000119f38 pc=0x55ef22a37b09

runtime.gcBgMarkStartWorkers.gowrap1()

	runtime/mgc.go:1339 +0x25 fp=0xc000119fe0 sp=0xc000119fc8 pc=0x55ef22a379e5

runtime.goexit({})

	runtime/asm_amd64.s:1700 +0x1 fp=0xc000119fe8 sp=0xc000119fe0 pc=0x55ef22a91ec1

created by runtime.gcBgMarkStartWorkers in goroutine 1

	runtime/mgc.go:1339 +0x105

goroutine 13 gp=0xc00021d6c0 m=nil [GC worker (idle)]:

runtime.gopark(0x55ef24d58560?, 0x1?, 0x78?, 0x2d?, 0x0?)

	runtime/proc.go:435 +0xce fp=0xc000507f38 sp=0xc000507f18 pc=0x55ef22a89f4e

runtime.gcBgMarkWorker(0xc00003f7a0)

	runtime/mgc.go:1423 +0xe9 fp=0xc000507fc8 sp=0xc000507f38 pc=0x55ef22a37b09

runtime.gcBgMarkStartWorkers.gowrap1()

	runtime/mgc.go:1339 +0x25 fp=0xc000507fe0 sp=0xc000507fc8 pc=0x55ef22a379e5

runtime.goexit({})

	runtime/asm_amd64.s:1700 +0x1 fp=0xc000507fe8 sp=0xc000507fe0 pc=0x55ef22a91ec1

created by runtime.gcBgMarkStartWorkers in goroutine 1

	runtime/mgc.go:1339 +0x105

goroutine 14 gp=0xc000582fc0 m=nil [sync.WaitGroup.Wait]:

runtime.gopark(0x0?, 0x0?, 0x20?, 0x61?, 0x0?)

	runtime/proc.go:435 +0xce fp=0xc0000bda90 sp=0xc0000bda70 pc=0x55ef22a89f4e

runtime.goparkunlock(...)

	runtime/proc.go:441

runtime.semacquire1(0xc000269198, 0x0, 0x1, 0x0, 0x18)

	runtime/sema.go:188 +0x229 fp=0xc0000bdaf8 sp=0xc0000bda90 pc=0x55ef22a69d49

sync.runtime_SemacquireWaitGroup(0x0?)

	runtime/sema.go:110 +0x25 fp=0xc0000bdb30 sp=0xc0000bdaf8 pc=0x55ef22a8b885

sync.(*WaitGroup).Wait(0xc000269190?)

	sync/waitgroup.go:118 +0x48 fp=0xc0000bdb58 sp=0xc0000bdb30 pc=0x55ef22a9d928

github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0xc0002690e0, {0x55ef24289ea0, 0xc0003b9810})

	github.com/ollama/ollama/runner/ollamarunner/runner.go:442 +0x45 fp=0xc0000bdfb8 sp=0xc0000bdb58 pc=0x55ef23088805

github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1()

	github.com/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x28 fp=0xc0000bdfe0 sp=0xc0000bdfb8 pc=0x55ef23092088

runtime.goexit({})

	runtime/asm_amd64.s:1700 +0x1 fp=0xc0000bdfe8 sp=0xc0000bdfe0 pc=0x55ef22a91ec1

created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1

	github.com/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x4c9

goroutine 66 gp=0xc000583500 m=nil [IO wait]:

runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0xb?)

	runtime/proc.go:435 +0xce fp=0xc00050d5d8 sp=0xc00050d5b8 pc=0x55ef22a89f4e

runtime.netpollblock(0x55ef22aad7f8?, 0x22a23506?, 0xef?)

	runtime/netpoll.go:575 +0xf7 fp=0xc00050d610 sp=0xc00050d5d8 pc=0x55ef22a4f0f7

internal/poll.runtime_pollWait(0x7fdf27fc6d98, 0x72)

	runtime/netpoll.go:351 +0x85 fp=0xc00050d630 sp=0xc00050d610 pc=0x55ef22a89165

internal/poll.(*pollDesc).wait(0xc0004f6a80?, 0xc00012f4b1?, 0x0)

	internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00050d658 sp=0xc00050d630 pc=0x55ef22b11487

internal/poll.(*pollDesc).waitRead(...)

	internal/poll/fd_poll_runtime.go:89

internal/poll.(*FD).Read(0xc0004f6a80, {0xc00012f4b1, 0x1, 0x1})

	internal/poll/fd_unix.go:165 +0x27a fp=0xc00050d6f0 sp=0xc00050d658 pc=0x55ef22b1277a

net.(*netFD).Read(0xc0004f6a80, {0xc00012f4b1?, 0x0?, 0x0?})

	net/fd_posix.go:55 +0x25 fp=0xc00050d738 sp=0xc00050d6f0 pc=0x55ef22b87cc5

net.(*conn).Read(0xc0000aea58, {0xc00012f4b1?, 0x0?, 0x0?})

	net/net.go:194 +0x45 fp=0xc00050d780 sp=0xc00050d738 pc=0x55ef22b96085

net/http.(*connReader).backgroundRead(0xc00012f4a0)

	net/http/server.go:690 +0x37 fp=0xc00050d7c8 sp=0xc00050d780 pc=0x55ef22d82bd7

net/http.(*connReader).startBackgroundRead.gowrap2()

	net/http/server.go:686 +0x25 fp=0xc00050d7e0 sp=0xc00050d7c8 pc=0x55ef22d82b05

runtime.goexit({})

	runtime/asm_amd64.s:1700 +0x1 fp=0xc00050d7e8 sp=0xc00050d7e0 pc=0x55ef22a91ec1

created by net/http.(*connReader).startBackgroundRead in goroutine 15

	net/http/server.go:686 +0xb6

rax    0x0

rbx    0xb5

rcx    0x7fdf6f373b2c

rdx    0x6

rdi    0xa9

rsi    0xb5

rbp    0x7fdf1e7fb340

rsp    0x7fdf1e7fb300

r8     0x0

r9     0x0

r10    0x8


r11    0x246

r12    0x6

r13    0x7fdec6e20890

r14    0x16

r15    0x7fdec6df6ba0

rip    0x7fdf6f373b2c


rflags 0x246

cs     0x33

fs     0x0

gs     0x0

time=2026-02-14T12:51:23.644Z level=ERROR source=server.go:304 msg="llama runner terminated" error="exit status 2"

time=2026-02-14T12:51:23.644Z level=ERROR source=server.go:1205 msg="do load request" error="Post \"http://127.0.0.1:45707/load\": EOF"

time=2026-02-14T12:51:23.644Z level=ERROR source=server.go:1205 msg="do load request" error="Post \"http://127.0.0.1:45707/load\": dial tcp 127.0.0.1:45707: connect: connection refused"

time=2026-02-14T12:51:23.644Z level=INFO source=sched.go:490 msg="Load failed" model=/root/.ollama/models/blobs/sha256-2b0cf8f17b4c723c27303015383c27ec4bf2d8314bb677d05e920dd70bb0f16b error="model failed to load, this may be due to resource limitations or an internal error, check ollama server logs for details"

[GIN] 2026/02/14 - 12:51:23 | 500 |  816.630011ms |       127.0.0.1 | POST     "/api/embed"

OS

Docker

GPU

AMD

CPU

AMD

Ollama version

0.16.2

Originally created by @ndrewpj on GitHub (Feb 14, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/14253 ### What is the issue? I cannot load and run models in ollama in docker. I have AMD Strix halo, according to the ollama log on loading the increased VRAM size is visible to ollama: [GIN] 2026/02/14 - 12:51:23 | 500 | 816.630011ms | 127.0.0.1 | POST "/api/embed" time=2026-02-14T12:53:02.765Z level=INFO source=routes.go:1636 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:8192 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE:q8_0 OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:2 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 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_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]" time=2026-02-14T12:53:02.766Z level=INFO source=images.go:473 msg="total blobs: 30" time=2026-02-14T12:53:02.766Z level=INFO source=images.go:480 msg="total unused blobs removed: 0" time=2026-02-14T12:53:02.766Z level=INFO source=routes.go:1689 msg="Listening on [::]:11434 (version 0.16.1)" time=2026-02-14T12:53:02.767Z level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2026-02-14T12:53:02.767Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 39845" time=2026-02-14T12:53:03.120Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 38575" time=2026-02-14T12:53:03.530Z level=INFO source=types.go:42 msg="inference compute" id=0 filter_id=0 library=ROCm compute=gfx1151 name=ROCm0 description="AMD Radeon Graphics" libdirs=ollama,rocm driver=60342.13 pci_id=0000:c5:00.0 type=iGPU total="120.4 GiB" available="118.3 GiB" time=2026-02-14T12:53:03.530Z level=INFO source=routes.go:1739 msg="vram-based default context" total_vram="120.4 GiB" default_num_ctx=262144 ### Relevant log output ```shell runtime/mgc.go:1339 +0x105 goroutine 53 gp=0xc000120540 m=nil [GC worker (idle)]: runtime.gopark(0x55ef24d58560?, 0x1?, 0xc1?, 0x22?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000119f38 sp=0xc000119f18 pc=0x55ef22a89f4e runtime.gcBgMarkWorker(0xc00003f7a0) runtime/mgc.go:1423 +0xe9 fp=0xc000119fc8 sp=0xc000119f38 pc=0x55ef22a37b09 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000119fe0 sp=0xc000119fc8 pc=0x55ef22a379e5 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000119fe8 sp=0xc000119fe0 pc=0x55ef22a91ec1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 13 gp=0xc00021d6c0 m=nil [GC worker (idle)]: runtime.gopark(0x55ef24d58560?, 0x1?, 0x78?, 0x2d?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000507f38 sp=0xc000507f18 pc=0x55ef22a89f4e runtime.gcBgMarkWorker(0xc00003f7a0) runtime/mgc.go:1423 +0xe9 fp=0xc000507fc8 sp=0xc000507f38 pc=0x55ef22a37b09 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000507fe0 sp=0xc000507fc8 pc=0x55ef22a379e5 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000507fe8 sp=0xc000507fe0 pc=0x55ef22a91ec1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 14 gp=0xc000582fc0 m=nil [sync.WaitGroup.Wait]: runtime.gopark(0x0?, 0x0?, 0x20?, 0x61?, 0x0?) runtime/proc.go:435 +0xce fp=0xc0000bda90 sp=0xc0000bda70 pc=0x55ef22a89f4e runtime.goparkunlock(...) runtime/proc.go:441 runtime.semacquire1(0xc000269198, 0x0, 0x1, 0x0, 0x18) runtime/sema.go:188 +0x229 fp=0xc0000bdaf8 sp=0xc0000bda90 pc=0x55ef22a69d49 sync.runtime_SemacquireWaitGroup(0x0?) runtime/sema.go:110 +0x25 fp=0xc0000bdb30 sp=0xc0000bdaf8 pc=0x55ef22a8b885 sync.(*WaitGroup).Wait(0xc000269190?) sync/waitgroup.go:118 +0x48 fp=0xc0000bdb58 sp=0xc0000bdb30 pc=0x55ef22a9d928 github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0xc0002690e0, {0x55ef24289ea0, 0xc0003b9810}) github.com/ollama/ollama/runner/ollamarunner/runner.go:442 +0x45 fp=0xc0000bdfb8 sp=0xc0000bdb58 pc=0x55ef23088805 github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1() github.com/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x28 fp=0xc0000bdfe0 sp=0xc0000bdfb8 pc=0x55ef23092088 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc0000bdfe8 sp=0xc0000bdfe0 pc=0x55ef22a91ec1 created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1 github.com/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x4c9 goroutine 66 gp=0xc000583500 m=nil [IO wait]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0xb?) runtime/proc.go:435 +0xce fp=0xc00050d5d8 sp=0xc00050d5b8 pc=0x55ef22a89f4e runtime.netpollblock(0x55ef22aad7f8?, 0x22a23506?, 0xef?) runtime/netpoll.go:575 +0xf7 fp=0xc00050d610 sp=0xc00050d5d8 pc=0x55ef22a4f0f7 internal/poll.runtime_pollWait(0x7fdf27fc6d98, 0x72) runtime/netpoll.go:351 +0x85 fp=0xc00050d630 sp=0xc00050d610 pc=0x55ef22a89165 internal/poll.(*pollDesc).wait(0xc0004f6a80?, 0xc00012f4b1?, 0x0) internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00050d658 sp=0xc00050d630 pc=0x55ef22b11487 internal/poll.(*pollDesc).waitRead(...) internal/poll/fd_poll_runtime.go:89 internal/poll.(*FD).Read(0xc0004f6a80, {0xc00012f4b1, 0x1, 0x1}) internal/poll/fd_unix.go:165 +0x27a fp=0xc00050d6f0 sp=0xc00050d658 pc=0x55ef22b1277a net.(*netFD).Read(0xc0004f6a80, {0xc00012f4b1?, 0x0?, 0x0?}) net/fd_posix.go:55 +0x25 fp=0xc00050d738 sp=0xc00050d6f0 pc=0x55ef22b87cc5 net.(*conn).Read(0xc0000aea58, {0xc00012f4b1?, 0x0?, 0x0?}) net/net.go:194 +0x45 fp=0xc00050d780 sp=0xc00050d738 pc=0x55ef22b96085 net/http.(*connReader).backgroundRead(0xc00012f4a0) net/http/server.go:690 +0x37 fp=0xc00050d7c8 sp=0xc00050d780 pc=0x55ef22d82bd7 net/http.(*connReader).startBackgroundRead.gowrap2() net/http/server.go:686 +0x25 fp=0xc00050d7e0 sp=0xc00050d7c8 pc=0x55ef22d82b05 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00050d7e8 sp=0xc00050d7e0 pc=0x55ef22a91ec1 created by net/http.(*connReader).startBackgroundRead in goroutine 15 net/http/server.go:686 +0xb6 rax 0x0 rbx 0xb5 rcx 0x7fdf6f373b2c rdx 0x6 rdi 0xa9 rsi 0xb5 rbp 0x7fdf1e7fb340 rsp 0x7fdf1e7fb300 r8 0x0 r9 0x0 r10 0x8  r11 0x246 r12 0x6 r13 0x7fdec6e20890 r14 0x16 r15 0x7fdec6df6ba0 rip 0x7fdf6f373b2c  rflags 0x246 cs 0x33 fs 0x0 gs 0x0 time=2026-02-14T12:51:23.644Z level=ERROR source=server.go:304 msg="llama runner terminated" error="exit status 2" time=2026-02-14T12:51:23.644Z level=ERROR source=server.go:1205 msg="do load request" error="Post \"http://127.0.0.1:45707/load\": EOF" time=2026-02-14T12:51:23.644Z level=ERROR source=server.go:1205 msg="do load request" error="Post \"http://127.0.0.1:45707/load\": dial tcp 127.0.0.1:45707: connect: connection refused" time=2026-02-14T12:51:23.644Z level=INFO source=sched.go:490 msg="Load failed" model=/root/.ollama/models/blobs/sha256-2b0cf8f17b4c723c27303015383c27ec4bf2d8314bb677d05e920dd70bb0f16b error="model failed to load, this may be due to resource limitations or an internal error, check ollama server logs for details" [GIN] 2026/02/14 - 12:51:23 | 500 | 816.630011ms | 127.0.0.1 | POST "/api/embed" ``` ### OS Docker ### GPU AMD ### CPU AMD ### Ollama version 0.16.2
GiteaMirror added the bug label 2026-05-05 01:15:23 -05:00
GiteaMirror commented 2026-05-05 01:15:26 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 14, 2026):

Include the part of the log leading up to the crash dump.

<!-- gh-comment-id:3901918518 --> @rick-github commented on GitHub (Feb 14, 2026): Include the part of the log leading up to the crash dump.
GiteaMirror commented 2026-05-05 01:15:27 -05:00
Author
Owner
Copy Link

@ndrewpj commented on GitHub (Feb 14, 2026):

Logs attached.

_ollama_logs.txt

<!-- gh-comment-id:3901975532 --> @ndrewpj commented on GitHub (Feb 14, 2026): Logs attached. [_ollama_logs.txt](https://github.com/user-attachments/files/25315476/_ollama_logs.txt)
GiteaMirror commented 2026-05-05 01:15:29 -05:00
Author
Owner
Copy Link

@pkoutsogilas commented on GitHub (Feb 15, 2026):

I have the same issue, I can't run even the smallest models.

OS:
Fedora 43 (6.18.9-200.fc43.x86_64)

GPU:
AMD Radeon 8060S Graphics (allocated fixed 96GB)

CPU:
AMD Ryzen AI Max 300 Series

Ollama version:
0.16.0

ollama-service.log

<!-- gh-comment-id:3904531397 --> @pkoutsogilas commented on GitHub (Feb 15, 2026): I have the same issue, I can't run even the smallest models. **OS:** Fedora 43 (6.18.9-200.fc43.x86_64) **GPU:** AMD Radeon 8060S Graphics (allocated fixed 96GB) **CPU:** AMD Ryzen AI Max 300 Series **Ollama version:** 0.16.0 [ollama-service.log](https://github.com/user-attachments/files/25325400/ollama-service.log)
GiteaMirror commented 2026-05-05 01:15:30 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 15, 2026):

I'm unable to duplicate on an AMD Ryzen Max+ 395. What's the output of rocm-smi -a and radeontop?

<!-- gh-comment-id:3904686779 --> @rick-github commented on GitHub (Feb 15, 2026): I'm unable to duplicate on an AMD Ryzen Max+ 395. What's the output of `rocm-smi -a` and `radeontop`?
GiteaMirror commented 2026-05-05 01:15:32 -05:00
Author
Owner
Copy Link

@pkoutsogilas commented on GitHub (Feb 15, 2026):

@rick-github:

I don't have rocm installed only ollama, this is the output, LM Studio works fine.

Output of rocm-smi -a:

============================ ROCm System Management Interface ============================
============================== Version of System Component ===============================
Driver version: 6.18.9-200.fc43.x86_64
==========================================================================================
=========================================== ID ===========================================
Fail to open libdrm_amdgpu.so: libdrm_amdgpu.so: cannot open shared object file: No such file or directory
GPU[0]          : get_name, Failed to load a library
GPU[0]          : Device Name:          N/A
GPU[0]          : Device ID:            0x1586
GPU[0]          : Device Rev:           0xc1
GPU[0]          : Subsystem ID:         0x000a
GPU[0]          : GUID:                 40551
==========================================================================================
======================================= Unique ID ========================================
GPU[0]          : Unique ID: 0x0
==========================================================================================
========================================= VBIOS ==========================================
GPU[0]          : VBIOS version: 113-STRXLGEN-001
==========================================================================================
====================================== Temperature =======================================
GPU[0]          : Temperature (Sensor edge) (C): 40.0
==========================================================================================
=============================== Current clock frequencies ================================
Exception caught: map::at
GPU[0]          : sclk clock level: 1: (881Mhz)
==========================================================================================
=================================== Current Fan Metric ===================================
GPU[0]          : Not supported
==========================================================================================
================================= Show Performance Level =================================
GPU[0]          : Performance Level: auto
==========================================================================================
==================================== OverDrive Level =====================================
GPU[0]          : get_overdrive_level_sclk, Not supported on the given system
==========================================================================================
==================================== OverDrive Level =====================================
GPU[0]          : get_mem_overdrive_level_mclk, Not supported on the given system
==========================================================================================
======================================= Power Cap ========================================
GPU[0]          : get_power_cap, Not supported on the given system
GPU[0]          : Max Graphics Package Power Unsupported
==========================================================================================
================================== Show Power Profiles ===================================
GPU[0]          : get_power_profiles, Not supported on the given system
==========================================================================================
=================================== Power Consumption ====================================
GPU[0]          : Current Socket Graphics Package Power (W): 13.041
==========================================================================================
============================== Supported clock frequencies ===============================
GPU[0]          : Clock [dcefclk] on device [0] exists but EMPTY! Likely driver error!
GPU[0]          : Clock [fclk] on device [0] exists but EMPTY! Likely driver error!
GPU[0]          : Clock [mclk] on device [0] exists but EMPTY! Likely driver error!
GPU[0]          : Supported sclk frequencies on GPU0
GPU[0]          : 0: 600Mhz
GPU[0]          : 1: 881Mhz *
GPU[0]          : 2: 2900Mhz
GPU[0]          : 
GPU[0]          : Clock [socclk] on device [0] exists but EMPTY! Likely driver error!
------------------------------------------------------------------------------------------
==========================================================================================
=================================== % time GPU is busy ===================================
GPU[0]          : GPU use (%): 2
==========================================================================================
=================================== Current Memory Use ===================================
GPU[0]          : GPU Memory Allocated (VRAM%): 1
GPU[0]          : Memory Activity: N/A
GPU[0]          : Not supported on the given system
==========================================================================================
===================================== Memory Vendor ======================================
GPU[0]          : get_vram_vendor, Not supported on the given system
==========================================================================================
================================== PCIe Replay Counter ===================================
GPU[0]          : PCIe Replay Count, Not supported on the given system
==========================================================================================
===================================== Serial Number ======================================
GPU[0]          : get_serial_number, Not supported on the given system
GPU[0]          : Serial Number: N/A
==========================================================================================
===================================== KFD Processes ======================================
No KFD PIDs currently running
==========================================================================================
================================== GPUs Indexed by PID ===================================
No KFD PIDs currently running
==========================================================================================
======================= GPU Memory clock frequencies and voltages ========================
GPU[0]          : OD_SCLK:
GPU[0]          : 0: 600Mhz
GPU[0]          : 1: 2900Mhz
GPU[0]          : OD_MCLK:
GPU[0]          : 0: 0Mhz
GPU[0]          : 1: 0Mhz
==========================================================================================
==================================== Current voltage =====================================
GPU[0]          : Voltage (mV): 0
==========================================================================================
======================================= PCI Bus ID =======================================
GPU[0]          : PCI Bus: 0000:C3:00.0
==========================================================================================
================================== Firmware Information ==================================
GPU[0]          : ASD firmware version:         0x21000104
GPU[0]          : ME firmware version:          32
GPU[0]          : MEC firmware version:         32
GPU[0]          : MES firmware version:         0x00000080
GPU[0]          : MES KIQ firmware version:     0x0000006f
GPU[0]          : PFP firmware version:         46
GPU[0]          : RLC firmware version:         290653446
GPU[0]          : SDMA firmware version:        17
GPU[0]          : SMC firmware version:         10.100.02.00
GPU[0]          : VCN firmware version:         0x0911801b
==========================================================================================
====================================== Product Info ======================================
Fail to open libdrm_amdgpu.so: libdrm_amdgpu.so: cannot open shared object file: No such file or directory
GPU[0]          : get_name, Failed to load a library
GPU[0]          : Card Series:          N/A
GPU[0]          : Card Model:           0x1586
GPU[0]          : Card Vendor:          Advanced Micro Devices, Inc. [AMD/ATI]
GPU[0]          : Card SKU:             STRXLGEN
GPU[0]          : Subsystem ID:         0x000a
GPU[0]          : Device Rev:           0xc1
GPU[0]          : Node ID:              1
GPU[0]          : GUID:                 40551
GPU[0]          : GFX Version:          gfx1151
==========================================================================================
======================================= Pages Info =======================================
GPU[0]          : ras, Not supported on the given system
================================= Show Valid sclk Range ==================================
GPU[0]          : Valid sclk range: 600Mhz - 2900Mhz
==========================================================================================
================================= Show Valid mclk Range ==================================
GPU[0]          : Valid mclk range: 0Mhz - 0Mhz
==========================================================================================
================================ Show Valid voltage Range ================================
WARNING: GPU[0] : Voltage curve regions unsupported.
==========================================================================================
================================== Voltage Curve Points ==================================
WARNING: GPU[0] : Voltage curve Points unsupported.
==========================================================================================
==================================== Consumed Energy =====================================
GPU[0]          : % Energy Counter, Unexpected data received
==========================================================================================
=============================== Current Compute Partition ================================
GPU[0]          : Not supported on the given system
==========================================================================================
================================ Current Memory Partition ================================
GPU[0]          : Not supported on the given system
==========================================================================================
====================================== GPU Metrics =======================================
GPU[0]          : get_gpu_metrics, Unexpected data received
ERROR: GPU[0]   : Failed to retrieve GPU metrics, metric version may not be supported for this device.
==========================================================================================
================================== End of ROCm SMI Log ===================================
<!-- gh-comment-id:3904736833 --> @pkoutsogilas commented on GitHub (Feb 15, 2026): @rick-github: I don't have rocm installed only ollama, this is the output, LM Studio works fine. Output of rocm-smi -a: ``` ============================ ROCm System Management Interface ============================ ============================== Version of System Component =============================== Driver version: 6.18.9-200.fc43.x86_64 ========================================================================================== =========================================== ID =========================================== Fail to open libdrm_amdgpu.so: libdrm_amdgpu.so: cannot open shared object file: No such file or directory GPU[0] : get_name, Failed to load a library GPU[0] : Device Name: N/A GPU[0] : Device ID: 0x1586 GPU[0] : Device Rev: 0xc1 GPU[0] : Subsystem ID: 0x000a GPU[0] : GUID: 40551 ========================================================================================== ======================================= Unique ID ======================================== GPU[0] : Unique ID: 0x0 ========================================================================================== ========================================= VBIOS ========================================== GPU[0] : VBIOS version: 113-STRXLGEN-001 ========================================================================================== ====================================== Temperature ======================================= GPU[0] : Temperature (Sensor edge) (C): 40.0 ========================================================================================== =============================== Current clock frequencies ================================ Exception caught: map::at GPU[0] : sclk clock level: 1: (881Mhz) ========================================================================================== =================================== Current Fan Metric =================================== GPU[0] : Not supported ========================================================================================== ================================= Show Performance Level ================================= GPU[0] : Performance Level: auto ========================================================================================== ==================================== OverDrive Level ===================================== GPU[0] : get_overdrive_level_sclk, Not supported on the given system ========================================================================================== ==================================== OverDrive Level ===================================== GPU[0] : get_mem_overdrive_level_mclk, Not supported on the given system ========================================================================================== ======================================= Power Cap ======================================== GPU[0] : get_power_cap, Not supported on the given system GPU[0] : Max Graphics Package Power Unsupported ========================================================================================== ================================== Show Power Profiles =================================== GPU[0] : get_power_profiles, Not supported on the given system ========================================================================================== =================================== Power Consumption ==================================== GPU[0] : Current Socket Graphics Package Power (W): 13.041 ========================================================================================== ============================== Supported clock frequencies =============================== GPU[0] : Clock [dcefclk] on device [0] exists but EMPTY! Likely driver error! GPU[0] : Clock [fclk] on device [0] exists but EMPTY! Likely driver error! GPU[0] : Clock [mclk] on device [0] exists but EMPTY! Likely driver error! GPU[0] : Supported sclk frequencies on GPU0 GPU[0] : 0: 600Mhz GPU[0] : 1: 881Mhz * GPU[0] : 2: 2900Mhz GPU[0] : GPU[0] : Clock [socclk] on device [0] exists but EMPTY! Likely driver error! ------------------------------------------------------------------------------------------ ========================================================================================== =================================== % time GPU is busy =================================== GPU[0] : GPU use (%): 2 ========================================================================================== =================================== Current Memory Use =================================== GPU[0] : GPU Memory Allocated (VRAM%): 1 GPU[0] : Memory Activity: N/A GPU[0] : Not supported on the given system ========================================================================================== ===================================== Memory Vendor ====================================== GPU[0] : get_vram_vendor, Not supported on the given system ========================================================================================== ================================== PCIe Replay Counter =================================== GPU[0] : PCIe Replay Count, Not supported on the given system ========================================================================================== ===================================== Serial Number ====================================== GPU[0] : get_serial_number, Not supported on the given system GPU[0] : Serial Number: N/A ========================================================================================== ===================================== KFD Processes ====================================== No KFD PIDs currently running ========================================================================================== ================================== GPUs Indexed by PID =================================== No KFD PIDs currently running ========================================================================================== ======================= GPU Memory clock frequencies and voltages ======================== GPU[0] : OD_SCLK: GPU[0] : 0: 600Mhz GPU[0] : 1: 2900Mhz GPU[0] : OD_MCLK: GPU[0] : 0: 0Mhz GPU[0] : 1: 0Mhz ========================================================================================== ==================================== Current voltage ===================================== GPU[0] : Voltage (mV): 0 ========================================================================================== ======================================= PCI Bus ID ======================================= GPU[0] : PCI Bus: 0000:C3:00.0 ========================================================================================== ================================== Firmware Information ================================== GPU[0] : ASD firmware version: 0x21000104 GPU[0] : ME firmware version: 32 GPU[0] : MEC firmware version: 32 GPU[0] : MES firmware version: 0x00000080 GPU[0] : MES KIQ firmware version: 0x0000006f GPU[0] : PFP firmware version: 46 GPU[0] : RLC firmware version: 290653446 GPU[0] : SDMA firmware version: 17 GPU[0] : SMC firmware version: 10.100.02.00 GPU[0] : VCN firmware version: 0x0911801b ========================================================================================== ====================================== Product Info ====================================== Fail to open libdrm_amdgpu.so: libdrm_amdgpu.so: cannot open shared object file: No such file or directory GPU[0] : get_name, Failed to load a library GPU[0] : Card Series: N/A GPU[0] : Card Model: 0x1586 GPU[0] : Card Vendor: Advanced Micro Devices, Inc. [AMD/ATI] GPU[0] : Card SKU: STRXLGEN GPU[0] : Subsystem ID: 0x000a GPU[0] : Device Rev: 0xc1 GPU[0] : Node ID: 1 GPU[0] : GUID: 40551 GPU[0] : GFX Version: gfx1151 ========================================================================================== ======================================= Pages Info ======================================= GPU[0] : ras, Not supported on the given system ================================= Show Valid sclk Range ================================== GPU[0] : Valid sclk range: 600Mhz - 2900Mhz ========================================================================================== ================================= Show Valid mclk Range ================================== GPU[0] : Valid mclk range: 0Mhz - 0Mhz ========================================================================================== ================================ Show Valid voltage Range ================================ WARNING: GPU[0] : Voltage curve regions unsupported. ========================================================================================== ================================== Voltage Curve Points ================================== WARNING: GPU[0] : Voltage curve Points unsupported. ========================================================================================== ==================================== Consumed Energy ===================================== GPU[0] : % Energy Counter, Unexpected data received ========================================================================================== =============================== Current Compute Partition ================================ GPU[0] : Not supported on the given system ========================================================================================== ================================ Current Memory Partition ================================ GPU[0] : Not supported on the given system ========================================================================================== ====================================== GPU Metrics ======================================= GPU[0] : get_gpu_metrics, Unexpected data received ERROR: GPU[0] : Failed to retrieve GPU metrics, metric version may not be supported for this device. ========================================================================================== ================================== End of ROCm SMI Log =================================== ```
GiteaMirror commented 2026-05-05 01:15:33 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 15, 2026):

GPU[0]          : Clock [dcefclk] on device [0] exists but EMPTY! Likely driver error!

Given the number of driver errors, missing library and unexpected data messages, I suspect a mismatch between driver and device.

<!-- gh-comment-id:3904747858 --> @rick-github commented on GitHub (Feb 15, 2026): ``` GPU[0] : Clock [dcefclk] on device [0] exists but EMPTY! Likely driver error! ``` Given the number of driver errors, missing library and unexpected data messages, I suspect a mismatch between driver and device.
GiteaMirror commented 2026-05-05 01:15:35 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 15, 2026):

Try sudo dnf install libdrm or sudo dnf install libdrm-amdgpu to install the missing library.

<!-- gh-comment-id:3904750404 --> @rick-github commented on GitHub (Feb 15, 2026): Try `sudo dnf install libdrm` or `sudo dnf install libdrm-amdgpu` to install the missing library.
GiteaMirror commented 2026-05-05 01:15:42 -05:00
Author
Owner
Copy Link

@pkoutsogilas commented on GitHub (Feb 15, 2026):

Here are the responses:

sudo dnf install libdrm
Updating and loading repositories:
Repositories loaded.
Package "libdrm-2.4.131-1.fc43.x86_64" is already installed.
Package "libdrm-2.4.131-1.fc43.i686" is already installed.

sudo dnf install libdrm-amdgpu
Updating and loading repositories:
Repositories loaded.
Failed to resolve the transaction:
No match for argument: libdrm-amdgpu
You can try to add to command line:
  --skip-unavailable to skip unavailable packages

After installing libdrm-devel package rocm-smi -a give a lot more info an no warning about missing library, but still cannot run olllama :(

Maybe time to try using a different distro to see if Fedora is problem.

<!-- gh-comment-id:3904763108 --> @pkoutsogilas commented on GitHub (Feb 15, 2026): Here are the responses: ``` sudo dnf install libdrm Updating and loading repositories: Repositories loaded. Package "libdrm-2.4.131-1.fc43.x86_64" is already installed. Package "libdrm-2.4.131-1.fc43.i686" is already installed. ``` ``` sudo dnf install libdrm-amdgpu Updating and loading repositories: Repositories loaded. Failed to resolve the transaction: No match for argument: libdrm-amdgpu You can try to add to command line: --skip-unavailable to skip unavailable packages ``` After installing libdrm-devel package rocm-smi -a give a lot more info an no warning about missing library, but still cannot run olllama :( Maybe time to try using a different distro to see if Fedora is problem.
GiteaMirror commented 2026-05-05 01:15:49 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 15, 2026):

My AMD is running Linux Mint, an Ubuntu variant. I've had very few issues with it, other than installing kernel 6.11.0 for an infrequent AMD GPU hang issue.

<!-- gh-comment-id:3904786765 --> @rick-github commented on GitHub (Feb 15, 2026): My AMD is running Linux Mint, an Ubuntu variant. I've had very few issues with it, other than installing kernel 6.11.0 for an infrequent AMD GPU hang issue.
GiteaMirror commented 2026-05-05 01:15:56 -05:00
Author
Owner
Copy Link

@pkoutsogilas commented on GitHub (Feb 15, 2026):

Thanks for the help, going to see if switching to another distro will solve my issue, I will keep you posted.

<!-- gh-comment-id:3904798738 --> @pkoutsogilas commented on GitHub (Feb 15, 2026): Thanks for the help, going to see if switching to another distro will solve my issue, I will keep you posted.
GiteaMirror commented 2026-05-05 01:15:58 -05:00
Author
Owner
Copy Link

@ndrewpj commented on GitHub (Feb 15, 2026):

I'm unable to duplicate on an AMD Ryzen Max+ 395. What's the output of rocm-smi -a and radeontop?

rocm-smi -a


============================ ROCm System Management Interface ============================
WARNING: AMD GPU device(s) is/are in a low-power state. Check power control/runtime_status

============================== Version of System Component ===============================
Driver version: 6.19.0-2-cachyos
==========================================================================================
=========================================== ID ===========================================
GPU[0]          : Device Name:          Radeon 8060S Graphics
GPU[0]          : Device ID:            0x1586
GPU[0]          : Device Rev:           0xc1
GPU[0]          : Subsystem ID:         -0x7fe3
GPU[0]          : GUID:                 4538
==========================================================================================
======================================= Unique ID ========================================
GPU[0]          : Unique ID: 0x0
==========================================================================================
========================================= VBIOS ==========================================
GPU[0]          : VBIOS version: 113-STRXLGEN-001
==========================================================================================
====================================== Temperature =======================================
GPU[0]          : Temperature (Sensor edge) (C): 38.0
==========================================================================================
=============================== Current clock frequencies ================================
Exception caught: map::at
GPU[0]          : mclk clock level: 2: (1000Mhz)
GPU[0]          : sclk clock level: 1: (698Mhz)
==========================================================================================
=================================== Current Fan Metric ===================================
GPU[0]          : Not supported
==========================================================================================
================================= Show Performance Level =================================
GPU[0]          : Performance Level: auto
==========================================================================================
==================================== OverDrive Level =====================================
GPU[0]          : get_overdrive_level_sclk, Not supported on the given system
==========================================================================================
==================================== OverDrive Level =====================================
GPU[0]          : get_mem_overdrive_level_mclk, Not supported on the given system
==========================================================================================
======================================= Power Cap ========================================
GPU[0]          : get_power_cap, Not supported on the given system
GPU[0]          : Max Graphics Package Power Unsupported
==========================================================================================
================================== Show Power Profiles ===================================
GPU[0]          : get_power_profiles, Not supported on the given system
==========================================================================================
=================================== Power Consumption ====================================
GPU[0]          : Current Socket Graphics Package Power (W): 22.085
==========================================================================================
============================== Supported clock frequencies ===============================
GPU[0]          : Clock [dcefclk] on device [0] exists but EMPTY! Likely driver error!
GPU[0]          : Clock [fclk] on device [0] exists but EMPTY! Likely driver error!
GPU[0]          : Supported mclk frequencies on GPU0
GPU[0]          : 0: 400Mhz
GPU[0]          : 1: 800Mhz
GPU[0]          : 2: 1000Mhz *
GPU[0]          : 
GPU[0]          : Supported sclk frequencies on GPU0
GPU[0]          : 0: 600Mhz
GPU[0]          : 1: 698Mhz *
GPU[0]          : 2: 2900Mhz
GPU[0]          : 
GPU[0]          : Clock [socclk] on device [0] exists but EMPTY! Likely driver error!
------------------------------------------------------------------------------------------
==========================================================================================
=================================== % time GPU is busy ===================================
GPU[0]          : GPU use (%): 1
==========================================================================================
=================================== Current Memory Use ===================================
GPU[0]          : GPU Memory Allocated (VRAM%): 94
GPU[0]          : Memory Activity: N/A
GPU[0]          : Not supported on the given system
==========================================================================================
===================================== Memory Vendor ======================================
GPU[0]          : get_vram_vendor, Not supported on the given system
==========================================================================================
================================== PCIe Replay Counter ===================================
GPU[0]          : PCIe Replay Count, Not supported on the given system
==========================================================================================
===================================== Serial Number ======================================
GPU[0]          : get_serial_number, Not supported on the given system
GPU[0]          : Serial Number: N/A
==========================================================================================
===================================== KFD Processes ======================================
No KFD PIDs currently running
==========================================================================================
================================== GPUs Indexed by PID ===================================
No KFD PIDs currently running
==========================================================================================
======================= GPU Memory clock frequencies and voltages ========================
GPU[0]          : OD_SCLK:
GPU[0]          : 0: 600Mhz
GPU[0]          : 1: 2900Mhz
GPU[0]          : OD_MCLK:
GPU[0]          : 0: 18446744073709Mhz
GPU[0]          : 1: 18446744073709Mhz
GPU[0]          : OD_RANGE:
GPU[0]          : SCLK:     600Mhz        2900Mhz
==========================================================================================
==================================== Current voltage =====================================
GPU[0]          : Voltage (mV): 0
==========================================================================================
======================================= PCI Bus ID =======================================
GPU[0]          : PCI Bus: 0000:C5:00.0
==========================================================================================
================================== Firmware Information ==================================
GPU[0]          : ASD firmware version:         0x21000104
GPU[0]          : ME firmware version:          32
GPU[0]          : MEC firmware version:         32
GPU[0]          : MES firmware version:         0x00000080
GPU[0]          : MES KIQ firmware version:     0x0000006f
GPU[0]          : PFP firmware version:         46
GPU[0]          : RLC firmware version:         290653446
GPU[0]          : SDMA firmware version:        17
GPU[0]          : SMC firmware version:         10.100.01.00
GPU[0]          : VCN firmware version:         0x0911801b
==========================================================================================
====================================== Product Info ======================================
GPU[0]          : Card Series:          Radeon 8060S Graphics
GPU[0]          : Card Model:           0x1586
GPU[0]          : Card Vendor:          Advanced Micro Devices, Inc. [AMD/ATI]
GPU[0]          : Card SKU:             STRXLGEN
GPU[0]          : Subsystem ID:         -0x7fe3
GPU[0]          : Device Rev:           0xc1
GPU[0]          : Node ID:              1
GPU[0]          : GUID:                 4538
GPU[0]          : GFX Version:          gfx1151
==========================================================================================
======================================= Pages Info =======================================
GPU[0]          : ras, Not supported on the given system
================================= Show Valid sclk Range ==================================
GPU[0]          : Valid sclk range: 600Mhz - 2900Mhz
==========================================================================================
================================= Show Valid mclk Range ==================================
GPU[0]          : Unable to display mclk range
==========================================================================================
================================ Show Valid voltage Range ================================
WARNING: GPU[0] : Voltage curve regions unsupported.
==========================================================================================
================================== Voltage Curve Points ==================================
==========================================================================================
==================================== Consumed Energy =====================================
GPU[0]          : % Energy Counter, Unexpected data received
==========================================================================================
=============================== Current Compute Partition ================================
GPU[0]          : Not supported on the given system
==========================================================================================
================================ Current Memory Partition ================================
GPU[0]          : Not supported on the given system
==========================================================================================
====================================== GPU Metrics =======================================
GPU[0]          : Failed to retrieve GPU metrics, metric version may not be supported for this device.
==========================================================================================
================================== End of ROCm SMI Log ===================================
<!-- gh-comment-id:3905033019 --> @ndrewpj commented on GitHub (Feb 15, 2026): > I'm unable to duplicate on an AMD Ryzen Max+ 395. What's the output of `rocm-smi -a` and `radeontop`? ``` rocm-smi -a ============================ ROCm System Management Interface ============================ WARNING: AMD GPU device(s) is/are in a low-power state. Check power control/runtime_status ============================== Version of System Component =============================== Driver version: 6.19.0-2-cachyos ========================================================================================== =========================================== ID =========================================== GPU[0] : Device Name: Radeon 8060S Graphics GPU[0] : Device ID: 0x1586 GPU[0] : Device Rev: 0xc1 GPU[0] : Subsystem ID: -0x7fe3 GPU[0] : GUID: 4538 ========================================================================================== ======================================= Unique ID ======================================== GPU[0] : Unique ID: 0x0 ========================================================================================== ========================================= VBIOS ========================================== GPU[0] : VBIOS version: 113-STRXLGEN-001 ========================================================================================== ====================================== Temperature ======================================= GPU[0] : Temperature (Sensor edge) (C): 38.0 ========================================================================================== =============================== Current clock frequencies ================================ Exception caught: map::at GPU[0] : mclk clock level: 2: (1000Mhz) GPU[0] : sclk clock level: 1: (698Mhz) ========================================================================================== =================================== Current Fan Metric =================================== GPU[0] : Not supported ========================================================================================== ================================= Show Performance Level ================================= GPU[0] : Performance Level: auto ========================================================================================== ==================================== OverDrive Level ===================================== GPU[0] : get_overdrive_level_sclk, Not supported on the given system ========================================================================================== ==================================== OverDrive Level ===================================== GPU[0] : get_mem_overdrive_level_mclk, Not supported on the given system ========================================================================================== ======================================= Power Cap ======================================== GPU[0] : get_power_cap, Not supported on the given system GPU[0] : Max Graphics Package Power Unsupported ========================================================================================== ================================== Show Power Profiles =================================== GPU[0] : get_power_profiles, Not supported on the given system ========================================================================================== =================================== Power Consumption ==================================== GPU[0] : Current Socket Graphics Package Power (W): 22.085 ========================================================================================== ============================== Supported clock frequencies =============================== GPU[0] : Clock [dcefclk] on device [0] exists but EMPTY! Likely driver error! GPU[0] : Clock [fclk] on device [0] exists but EMPTY! Likely driver error! GPU[0] : Supported mclk frequencies on GPU0 GPU[0] : 0: 400Mhz GPU[0] : 1: 800Mhz GPU[0] : 2: 1000Mhz * GPU[0] : GPU[0] : Supported sclk frequencies on GPU0 GPU[0] : 0: 600Mhz GPU[0] : 1: 698Mhz * GPU[0] : 2: 2900Mhz GPU[0] : GPU[0] : Clock [socclk] on device [0] exists but EMPTY! Likely driver error! ------------------------------------------------------------------------------------------ ========================================================================================== =================================== % time GPU is busy =================================== GPU[0] : GPU use (%): 1 ========================================================================================== =================================== Current Memory Use =================================== GPU[0] : GPU Memory Allocated (VRAM%): 94 GPU[0] : Memory Activity: N/A GPU[0] : Not supported on the given system ========================================================================================== ===================================== Memory Vendor ====================================== GPU[0] : get_vram_vendor, Not supported on the given system ========================================================================================== ================================== PCIe Replay Counter =================================== GPU[0] : PCIe Replay Count, Not supported on the given system ========================================================================================== ===================================== Serial Number ====================================== GPU[0] : get_serial_number, Not supported on the given system GPU[0] : Serial Number: N/A ========================================================================================== ===================================== KFD Processes ====================================== No KFD PIDs currently running ========================================================================================== ================================== GPUs Indexed by PID =================================== No KFD PIDs currently running ========================================================================================== ======================= GPU Memory clock frequencies and voltages ======================== GPU[0] : OD_SCLK: GPU[0] : 0: 600Mhz GPU[0] : 1: 2900Mhz GPU[0] : OD_MCLK: GPU[0] : 0: 18446744073709Mhz GPU[0] : 1: 18446744073709Mhz GPU[0] : OD_RANGE: GPU[0] : SCLK: 600Mhz 2900Mhz ========================================================================================== ==================================== Current voltage ===================================== GPU[0] : Voltage (mV): 0 ========================================================================================== ======================================= PCI Bus ID ======================================= GPU[0] : PCI Bus: 0000:C5:00.0 ========================================================================================== ================================== Firmware Information ================================== GPU[0] : ASD firmware version: 0x21000104 GPU[0] : ME firmware version: 32 GPU[0] : MEC firmware version: 32 GPU[0] : MES firmware version: 0x00000080 GPU[0] : MES KIQ firmware version: 0x0000006f GPU[0] : PFP firmware version: 46 GPU[0] : RLC firmware version: 290653446 GPU[0] : SDMA firmware version: 17 GPU[0] : SMC firmware version: 10.100.01.00 GPU[0] : VCN firmware version: 0x0911801b ========================================================================================== ====================================== Product Info ====================================== GPU[0] : Card Series: Radeon 8060S Graphics GPU[0] : Card Model: 0x1586 GPU[0] : Card Vendor: Advanced Micro Devices, Inc. [AMD/ATI] GPU[0] : Card SKU: STRXLGEN GPU[0] : Subsystem ID: -0x7fe3 GPU[0] : Device Rev: 0xc1 GPU[0] : Node ID: 1 GPU[0] : GUID: 4538 GPU[0] : GFX Version: gfx1151 ========================================================================================== ======================================= Pages Info ======================================= GPU[0] : ras, Not supported on the given system ================================= Show Valid sclk Range ================================== GPU[0] : Valid sclk range: 600Mhz - 2900Mhz ========================================================================================== ================================= Show Valid mclk Range ================================== GPU[0] : Unable to display mclk range ========================================================================================== ================================ Show Valid voltage Range ================================ WARNING: GPU[0] : Voltage curve regions unsupported. ========================================================================================== ================================== Voltage Curve Points ================================== ========================================================================================== ==================================== Consumed Energy ===================================== GPU[0] : % Energy Counter, Unexpected data received ========================================================================================== =============================== Current Compute Partition ================================ GPU[0] : Not supported on the given system ========================================================================================== ================================ Current Memory Partition ================================ GPU[0] : Not supported on the given system ========================================================================================== ====================================== GPU Metrics ======================================= GPU[0] : Failed to retrieve GPU metrics, metric version may not be supported for this device. ========================================================================================== ================================== End of ROCm SMI Log =================================== ```
GiteaMirror commented 2026-05-05 01:16:01 -05:00
Author
Owner
Copy Link

@ndrewpj commented on GitHub (Feb 15, 2026):

I also tried to download ollama to v.0.14.0 - same issue. So it looks like an incompatibility issue of some kind

<!-- gh-comment-id:3905075952 --> @ndrewpj commented on GitHub (Feb 15, 2026): I also tried to download ollama to v.0.14.0 - same issue. So it looks like an incompatibility issue of some kind
GiteaMirror commented 2026-05-05 01:16:06 -05:00
Author
Owner
Copy Link

@pkoutsogilas commented on GitHub (Feb 16, 2026):

So I tried to following:

  • CachyOS, with kernet 6.19, ollama failed to load a model.
  • Kubuntu 25.10, with kernetl 6,17, ollama sucessfully was able to load models.

It guess it's an issue with kernels > 6.17 and/or Linux firmware version.

<!-- gh-comment-id:3907125979 --> @pkoutsogilas commented on GitHub (Feb 16, 2026): So I tried to following: * CachyOS, with kernet 6.19, ollama failed to load a model. * Kubuntu 25.10, with kernetl 6,17, ollama sucessfully was able to load models. It guess it's an issue with kernels > 6.17 and/or Linux firmware version.
GiteaMirror commented 2026-05-05 01:16:07 -05:00
Author
Owner
Copy Link

@ndrewpj commented on GitHub (Feb 16, 2026):

So I tried to following:

  • CachyOS, with kernet 6.19, ollama failed to load a model.
  • Kubuntu 25.10, with kernetl 6,17, ollama sucessfully was able to load models.

It guess it's an issue with kernels > 6.17 and/or Linux firmware version.

Meanwhile llama.cpp is working fine with 6.19 kernel and ROCm 7.2

❯ ./llama-bench -m ~/models/Apriel-1.5-15b-Thinker-Q6_K.gguf
ggml_cuda_init: found 1 ROCm devices:
  Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model                          |       size |     params | backend    | ngl |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | --------------: | -------------------: |
| llama 34B Q6_K                 |  11.02 GiB |    14.43 B | ROCm       |  99 |           pp512 |        506.83 ± 1.50 |
| llama 34B Q6_K                 |  11.02 GiB |    14.43 B | ROCm       |  99 |           tg128 |         16.63 ± 0.03 |

build: d5dfc3302 (8069)
❯ uname -a
Linux halo 6.19.0-2-cachyos #1 SMP PREEMPT_DYNAMIC Fri, 13 Feb 2026 16:07:19 +0000 x86_64 GNU/Linux
<!-- gh-comment-id:3907411090 --> @ndrewpj commented on GitHub (Feb 16, 2026): > So I tried to following: > > * CachyOS, with kernet 6.19, ollama failed to load a model. > * Kubuntu 25.10, with kernetl 6,17, ollama sucessfully was able to load models. > > It guess it's an issue with kernels > 6.17 and/or Linux firmware version. Meanwhile llama.cpp is working fine with 6.19 kernel and ROCm 7.2 ``` ❯ ./llama-bench -m ~/models/Apriel-1.5-15b-Thinker-Q6_K.gguf ggml_cuda_init: found 1 ROCm devices: Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32 | model | size | params | backend | ngl | test | t/s | | ------------------------------ | ---------: | ---------: | ---------- | --: | --------------: | -------------------: | | llama 34B Q6_K | 11.02 GiB | 14.43 B | ROCm | 99 | pp512 | 506.83 ± 1.50 | | llama 34B Q6_K | 11.02 GiB | 14.43 B | ROCm | 99 | tg128 | 16.63 ± 0.03 | build: d5dfc3302 (8069) ❯ uname -a Linux halo 6.19.0-2-cachyos #1 SMP PREEMPT_DYNAMIC Fri, 13 Feb 2026 16:07:19 +0000 x86_64 GNU/Linux ```
GiteaMirror commented 2026-05-05 01:16:08 -05:00
Author
Owner
Copy Link

@pkoutsogilas commented on GitHub (Feb 16, 2026):

@rick-github : Is the kernel / linux-firmware something that could be further investigated? You could try to reproduce it with a newer kernel.

<!-- gh-comment-id:3907476465 --> @pkoutsogilas commented on GitHub (Feb 16, 2026): @rick-github : Is the kernel / linux-firmware something that could be further investigated? You could try to reproduce it with a newer kernel.
GiteaMirror commented 2026-05-05 01:16:09 -05:00
Author
Owner
Copy Link

@dougmaitelli commented on GitHub (Feb 18, 2026):

I have a similar problem, but not sure if it is related or not.
On Ubuntu 24.04, Ollama runs out-of-the-box (but on CPU only, not able to detect the Strix Halo GPU).

But as soon as I install linux-oem-24.04c and boot with kernel 6.17 Ollama breaks and fails to load any model. GPU is then detected but I can't run any models. Booting back to previous kernels works.

Important to note I have rocm installed (7.2) and I assume the problem is because Ollama is currently built with rocm-6

<!-- gh-comment-id:3923592378 --> @dougmaitelli commented on GitHub (Feb 18, 2026): I have a similar problem, but not sure if it is related or not. On Ubuntu 24.04, Ollama runs out-of-the-box (but on CPU only, not able to detect the Strix Halo GPU). But as soon as I install `linux-oem-24.04c` and boot with kernel `6.17` Ollama breaks and fails to load any model. GPU is then detected but I can't run any models. Booting back to previous kernels works. Important to note I have rocm installed (`7.2`) and I assume the problem is because Ollama is currently built with `rocm-6`
GiteaMirror commented 2026-05-05 01:16:10 -05:00
Author
Owner
Copy Link

@ndrewpj commented on GitHub (Feb 19, 2026):

I have a similar problem, but not sure if it is related or not. On Ubuntu 24.04, Ollama runs out-of-the-box (but on CPU only, not able to detect the Strix Halo GPU).

But as soon as I install linux-oem-24.04c and boot with kernel 6.17 Ollama breaks and fails to load any model. GPU is then detected but I can't run any models. Booting back to previous kernels works.

Important to note I have rocm installed (7.2) and I assume the problem is because Ollama is currently built with rocm-6

Your error is most probably caused by buggy linux-firmware package.

<!-- gh-comment-id:3925718147 --> @ndrewpj commented on GitHub (Feb 19, 2026): > I have a similar problem, but not sure if it is related or not. On Ubuntu 24.04, Ollama runs out-of-the-box (but on CPU only, not able to detect the Strix Halo GPU). > > But as soon as I install `linux-oem-24.04c` and boot with kernel `6.17` Ollama breaks and fails to load any model. GPU is then detected but I can't run any models. Booting back to previous kernels works. > > Important to note I have rocm installed (`7.2`) and I assume the problem is because Ollama is currently built with `rocm-6` Your error is most probably caused by buggy linux-firmware package.
GiteaMirror commented 2026-05-05 01:16:12 -05:00
Author
Owner
Copy Link

@pkoutsogilas commented on GitHub (Feb 19, 2026):

I currently just use llama using a toolbox from https://strix-halo-toolboxes.com/ and that works perfectly. Hopefully in the future ollama will work again.

<!-- gh-comment-id:3925898032 --> @pkoutsogilas commented on GitHub (Feb 19, 2026): I currently just use llama using a toolbox from https://strix-halo-toolboxes.com/ and that works perfectly. Hopefully in the future ollama will work again.
GiteaMirror commented 2026-05-05 01:16:14 -05:00
Author
Owner
Copy Link

@ndrewpj commented on GitHub (Feb 19, 2026):

I currently just use llama using a toolbox from https://strix-halo-toolboxes.com/ and that works perfectly. Hopefully in the future ollama will work again.

I often cannot get llama.cpp engaging LLM tools, Ollama usually works better.... although I like llama.cpp very much, llama-swap too

<!-- gh-comment-id:3929793930 --> @ndrewpj commented on GitHub (Feb 19, 2026): > I currently just use llama using a toolbox from https://strix-halo-toolboxes.com/ and that works perfectly. Hopefully in the future ollama will work again. I often cannot get llama.cpp engaging LLM tools, Ollama usually works better.... although I like llama.cpp very much, llama-swap too
GiteaMirror commented 2026-05-05 01:16:15 -05:00
Author
Owner
Copy Link

@fischerdr commented on GitHub (Feb 20, 2026):

I also current have this issue on fedora43 using the docker ollama-rocm
Linux flyyn 6.18.12-200.fc43.x86_64 #1 SMP PREEMPT_DYNAMIC Mon Feb 16 18:58:26 UTC 2026 x86_64 GNU/Linux
amd-gpu-firmware-20260110-1.fc43.noarch
amd-ucode-firmware-20260110-1.fc43.noarch
rocm-core-6.4.4-1.fc43.x86_64 <- in OS but I use dockers with 7.2
I use https://github.com/kyuz0/amd-strix-halo-toolboxes with 2 different X2-evo (1 64gig, 1 128gig)
I have no issues using llama-server/cli and get ~35-48tps
flyyn-rocminfo.txt
flyyn-rocm-smi.txt

edit:
added libdrm-devel.x86_64 and the library error went away but still no luck

edit 2: downgraded to kernel 6.17.1-300.fc43.x86_64
and able to run
podman run -d --replace --name ollama-rocm --device=/dev/kfd --device=/dev/dri --group-add video -e HSA_OVERRIDE_GFX_VERSION=11.0.0 -e HIP_VISIBLE_DEVICES=0 -e OLLAMA_FLASH_ATTENTION=true -v ollama:/root/.ollama -p 11434:11434 ollama/ollama:rocm

<!-- gh-comment-id:3936240774 --> @fischerdr commented on GitHub (Feb 20, 2026): I also current have this issue on fedora43 using the docker ollama-rocm Linux flyyn 6.18.12-200.fc43.x86_64 #1 SMP PREEMPT_DYNAMIC Mon Feb 16 18:58:26 UTC 2026 x86_64 GNU/Linux amd-gpu-firmware-20260110-1.fc43.noarch amd-ucode-firmware-20260110-1.fc43.noarch rocm-core-6.4.4-1.fc43.x86_64 <- in OS but I use dockers with 7.2 I use https://github.com/kyuz0/amd-strix-halo-toolboxes with 2 different X2-evo (1 64gig, 1 128gig) I have no issues using llama-server/cli and get ~35-48tps [flyyn-rocminfo.txt](https://github.com/user-attachments/files/25448680/flyyn-rocminfo.txt) [flyyn-rocm-smi.txt](https://github.com/user-attachments/files/25448738/flyyn-rocm-smi.txt) edit: added libdrm-devel.x86_64 and the library error went away but still no luck edit 2: downgraded to kernel 6.17.1-300.fc43.x86_64 and able to run podman run -d --replace --name ollama-rocm --device=/dev/kfd --device=/dev/dri --group-add video -e HSA_OVERRIDE_GFX_VERSION=11.0.0 -e HIP_VISIBLE_DEVICES=0 -e OLLAMA_FLASH_ATTENTION=true -v ollama:/root/.ollama -p 11434:11434 ollama/ollama:rocm
GiteaMirror commented 2026-05-05 01:16:16 -05:00
Author
Owner
Copy Link

@ndrewpj commented on GitHub (Feb 20, 2026):

yep, on CachyOS kernels 6.18.* and 6.19.* ollama gives the error, tried 6.17 - and it is working. So kernel incompatibility

<!-- gh-comment-id:3937107574 --> @ndrewpj commented on GitHub (Feb 20, 2026): yep, on CachyOS kernels 6.18.* and 6.19.* ollama gives the error, tried 6.17 - and it is working. So kernel incompatibility
GiteaMirror commented 2026-05-05 01:16:18 -05:00
Author
Owner
Copy Link

@ulope commented on GitHub (Apr 10, 2026):

We seem to have the same issue. It also behaves weirdly inconsistent (which suggests maybe user / config error as well).

On nearly identical hardware (Ryzen AI Max 395, but from different manufacturers) on one system only Kernel 6.14.0-1019-oem works while on another only 6.14.0-1013-oem does.

<!-- gh-comment-id:4224075843 --> @ulope commented on GitHub (Apr 10, 2026): We seem to have the same issue. It also behaves weirdly inconsistent (which suggests maybe user / config error as well). On nearly identical hardware (Ryzen AI Max 395, but from different manufacturers) on one system only Kernel `6.14.0-1019-oem` works while on another only `6.14.0-1013-oem` does.
GiteaMirror commented 2026-05-05 01:16:19 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 10, 2026):

AMD recommends linux kernel 6.18.4 or newer for AMD Ryzen AI Max series APUs.

<!-- gh-comment-id:4224111169 --> @rick-github commented on GitHub (Apr 10, 2026): AMD [recommends](https://rocm.docs.amd.com/en/latest/how-to/system-optimization/strixhalo.html#required-kernel-version) linux kernel 6.18.4 or newer for AMD Ryzen AI Max series APUs.
GiteaMirror commented 2026-05-05 01:16:20 -05:00
Author
Owner
Copy Link

@ndrewpj commented on GitHub (Apr 10, 2026):

Ollama runs on kernel 7.x so closing this one.

<!-- gh-comment-id:4224879373 --> @ndrewpj commented on GitHub (Apr 10, 2026): Ollama runs on kernel 7.x so closing this one.
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#71339
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?