github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7491] cuda runner fails to build correctly without CUDA_PATH set #51276

New Issue
Closed
opened 2026-04-28 19:11:37 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-28 19:11:37 -05:00
Owner
Copy Link

Originally created by @auiphc on GitHub (Nov 4, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7491

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

I'm following this https://github.com/ollama/ollama/blob/main/docs/development.md#windows to build ollama on windows but I got a segmentation fault:

(base) PS C:\Users\50728\Documents\code\localcai-server\ollama> make -j 5
/usr/bin/make -C llama all
make[1]: Entering directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama'
rm -rf C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/default C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/cuda_v12 C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/metal C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/cpu C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/cpu_avx C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/cpu_avx2
/usr/bin/make -f make/Makefile.default
/usr/bin/make -f make/Makefile.cuda_v12
make[2]: Entering directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama'
make[2]: Entering directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama'
make[2]: Nothing to be done for 'all'.
make[2]: Leaving directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama'
/bin/nvcc --shared -lcuda -LC:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/dist/windows-amd64/lib/ollama  -lcublas  -lcudart  -lcublasLt C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/acc.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/arange.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/argsort.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/binbcast.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/clamp.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/concat.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/conv-transpose-1d.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/convert.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/cpy.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/cross-entropy-loss.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/diagmask.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/dmmv.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/getrows.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/im2col.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/mmq.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/mmvq.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/norm.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/opt-step-adamw.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/out-prod.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/pad.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/pool2d.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/quantize.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/rope.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/rwkv-wkv.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/scale.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/softmax.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/sum.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/sumrows.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/tsembd.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/unary.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/upscale.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq1_s.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq2_s.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq2_xs.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq3_s.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq4_nl.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq4_xs.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q2_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q3_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q4_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q4_1.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q4_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q5_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q5_1.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q5_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q6_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q8_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-backend.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-alloc.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-quants.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/sgemm.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-aarch64.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/fattn-tile-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/fattn-tile-f32.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/fattn.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cuda_v12.obj -o C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/runners/cuda_v12/ggml_cuda_v12.dll
Internal error
make[2]: *** [make/gpu.make:88: C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/runners/cuda_v12/ggml_cuda_v12.dll] Segmentation fault
make[2]: Leaving directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama'
make[1]: *** [Makefile:41: cuda_v12] Error 2
make[1]: Leaving directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama'
make: *** [Makefile:4: all] Error 2

My environment is:

(base) PS C:\Users\50728\Documents\code\localcai-server\ollama> make --version
GNU Make 4.4.1
Built for x86_64-pc-msys
Copyright (C) 1988-2023 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <https://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
(base) PS C:\Users\50728\Documents\code\localcai-server\ollama> nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Thu_Mar_28_02:30:10_Pacific_Daylight_Time_2024
Cuda compilation tools, release 12.4, V12.4.131
Build cuda_12.4.r12.4/compiler.34097967_0
(base) PS C:\Users\50728\Documents\code\localcai-server\ollama> cl
用于 x64 的 Microsoft (R) C/C++ 优化编译器 19.41.34123 版
版权所有(C) Microsoft Corporation。保留所有权利。

What I have done to slove previous issue is:

  1. link nvcc to msys2 ln -s "/c/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v12.4/bin/nvcc" /usr/bin/nvcc since make fails to find nvcc /bin/sh: line 1: /bin/nvcc: No such file or directory
  2. copy all files from C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\include to ./llama since its fails to find cuda_runtime.h

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

No response

Originally created by @auiphc on GitHub (Nov 4, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7491 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? I'm following this https://github.com/ollama/ollama/blob/main/docs/development.md#windows to build ollama on windows but I got a segmentation fault: ``` (base) PS C:\Users\50728\Documents\code\localcai-server\ollama> make -j 5 /usr/bin/make -C llama all make[1]: Entering directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama' rm -rf C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/default C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/cuda_v12 C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/metal C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/cpu C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/cpu_avx C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/build/windows/amd64/cpu_avx2 /usr/bin/make -f make/Makefile.default /usr/bin/make -f make/Makefile.cuda_v12 make[2]: Entering directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama' make[2]: Entering directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama' make[2]: Nothing to be done for 'all'. make[2]: Leaving directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama' /bin/nvcc --shared -lcuda -LC:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/dist/windows-amd64/lib/ollama -lcublas -lcudart -lcublasLt C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/acc.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/arange.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/argsort.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/binbcast.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/clamp.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/concat.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/conv-transpose-1d.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/convert.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/cpy.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/cross-entropy-loss.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/diagmask.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/dmmv.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/getrows.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/im2col.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/mmq.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/mmvq.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/norm.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/opt-step-adamw.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/out-prod.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/pad.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/pool2d.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/quantize.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/rope.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/rwkv-wkv.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/scale.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/softmax.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/sum.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/sumrows.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/tsembd.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/unary.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/upscale.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq1_s.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq2_s.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq2_xs.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq3_s.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq4_nl.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-iq4_xs.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q2_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q3_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q4_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q4_1.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q4_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q5_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q5_1.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q5_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q6_k.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/mmq-instance-q8_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-backend.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-alloc.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-quants.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/sgemm.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-aarch64.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/fattn-tile-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/fattn-tile-f32.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/fattn.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cuda_v12.obj C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cuda_v12.obj -o C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/runners/cuda_v12/ggml_cuda_v12.dll Internal error make[2]: *** [make/gpu.make:88: C:/Users/50728/DOCUME~1/code/LOCALC~1/ollama/llama/build/windows-amd64/runners/cuda_v12/ggml_cuda_v12.dll] Segmentation fault make[2]: Leaving directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama' make[1]: *** [Makefile:41: cuda_v12] Error 2 make[1]: Leaving directory '/c/Users/50728/Documents/code/localcai-server/ollama/llama' make: *** [Makefile:4: all] Error 2 ``` My environment is: ``` (base) PS C:\Users\50728\Documents\code\localcai-server\ollama> make --version GNU Make 4.4.1 Built for x86_64-pc-msys Copyright (C) 1988-2023 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <https://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. (base) PS C:\Users\50728\Documents\code\localcai-server\ollama> nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2024 NVIDIA Corporation Built on Thu_Mar_28_02:30:10_Pacific_Daylight_Time_2024 Cuda compilation tools, release 12.4, V12.4.131 Build cuda_12.4.r12.4/compiler.34097967_0 (base) PS C:\Users\50728\Documents\code\localcai-server\ollama> cl 用于 x64 的 Microsoft (R) C/C++ 优化编译器 19.41.34123 版 版权所有(C) Microsoft Corporation。保留所有权利。 ``` What I have done to slove previous issue is: 1. link nvcc to msys2 `ln -s "/c/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v12.4/bin/nvcc" /usr/bin/nvcc` since make fails to find nvcc `/bin/sh: line 1: /bin/nvcc: No such file or directory` 2. copy all files from `C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\include` to `./llama` since its fails to find `cuda_runtime.h` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version _No response_
GiteaMirror added the buildbugwindows labels 2026-04-28 19:11:41 -05:00
GiteaMirror commented 2026-04-28 19:11:52 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Nov 4, 2024):

Typically when the NVIDIA CUDA installer runs, it sets up CUDA_PATH as a system-wide environment variable, which we're leveraging in the current makefiles. It looks like we need to try a bit harder to find the correct NVCC if that variable isn't set.

As a workaround, if you set CUDA_PATH in your environment it should build correctly.

$env:CUDA_PATH="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4"
make -j 5
<!-- gh-comment-id:2455186244 --> @dhiltgen commented on GitHub (Nov 4, 2024): Typically when the NVIDIA CUDA installer runs, it sets up CUDA_PATH as a system-wide environment variable, which we're leveraging in the current makefiles. It looks like we need to try a bit harder to find the correct NVCC if that variable isn't set. As a workaround, if you set `CUDA_PATH` in your environment it should build correctly. ```powershell $env:CUDA_PATH="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4" make -j 5 ```
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 build windows
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#51276
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?