github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #15151] compiling ollama in debian 12. seems to try to run yum from redhat using build_linux.sh #56213

New Issue
Closed
opened 2026-04-29 10:26:19 -05:00 by GiteaMirror · 7 comments
GiteaMirror commented 2026-04-29 10:26:19 -05:00
Owner
Copy Link

Originally created by @jcharth on GitHub (Mar 30, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15151

What is the issue?

i am trying to compile from source to use my m60 gpu. it is really hard? any sugestions

 => ERROR [linux/arm64 base-arm64 2/2] RUN yum install -y yum-utils epel-release     && dnf install -y clang ccache git     && yum-config-manager --add-repo https://develop  4.4s
 => CANCELED [linux/arm64 jetpack-6 2/5] RUN apt-get update && apt-get install -y curl ccache unzip     && curl -fsSL https://github.com/Kitware/CMake/releases/download/v3.  0.0s
 => CANCELED [linux/arm64 jetpack-5 2/5] RUN apt-get update && apt-get install -y curl ccache unzip     && curl -fsSL https://github.com/Kitware/CMake/releases/download/v3.  0.0s
 => CANCELED [linux/amd64 base-amd64 2/2] RUN dnf install -y yum-utils ccache gcc-toolset-11-gcc gcc-toolset-11-gcc-c++ gcc-toolset-11-binutils     && yum-config-manager --  0.0s
------
 > [linux/arm64 base-arm64 2/2] RUN yum install -y yum-utils epel-release     && dnf install -y clang ccache git     && yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/sbsa/cuda-rhel8.repo:
4.249 exec /bin/sh: exec format error
------

 2 warnings found (use docker --debug to expand):
 - FromPlatformFlagConstDisallowed: FROM --platform flag should not use constant value "linux/arm64" (line 95)
 - FromPlatformFlagConstDisallowed: FROM --platform flag should not use constant value "linux/arm64" (line 111)
Dockerfile:24
--------------------
  23 |     # install epel-release for ccache
  24 | >>> RUN yum install -y yum-utils epel-release \
  25 | >>>     && dnf install -y clang ccache git \
  26 | >>>     && yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/sbsa/cuda-rhel8.repo
  27 |     ENV CC=clang CXX=clang++
--------------------
ERROR: failed to build: failed to solve: process "/bin/sh -c yum install -y yum-utils epel-release     && dnf install -y clang ccache git     && yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/sbsa/cuda-rhel8.repo" did not complete successfully: exit code: 255

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @jcharth on GitHub (Mar 30, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15151 ### What is the issue? i am trying to compile from source to use my m60 gpu. it is really hard? any sugestions ``` => ERROR [linux/arm64 base-arm64 2/2] RUN yum install -y yum-utils epel-release && dnf install -y clang ccache git && yum-config-manager --add-repo https://develop 4.4s => CANCELED [linux/arm64 jetpack-6 2/5] RUN apt-get update && apt-get install -y curl ccache unzip && curl -fsSL https://github.com/Kitware/CMake/releases/download/v3. 0.0s => CANCELED [linux/arm64 jetpack-5 2/5] RUN apt-get update && apt-get install -y curl ccache unzip && curl -fsSL https://github.com/Kitware/CMake/releases/download/v3. 0.0s => CANCELED [linux/amd64 base-amd64 2/2] RUN dnf install -y yum-utils ccache gcc-toolset-11-gcc gcc-toolset-11-gcc-c++ gcc-toolset-11-binutils && yum-config-manager -- 0.0s ------ > [linux/arm64 base-arm64 2/2] RUN yum install -y yum-utils epel-release && dnf install -y clang ccache git && yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/sbsa/cuda-rhel8.repo: 4.249 exec /bin/sh: exec format error ------ 2 warnings found (use docker --debug to expand): - FromPlatformFlagConstDisallowed: FROM --platform flag should not use constant value "linux/arm64" (line 95) - FromPlatformFlagConstDisallowed: FROM --platform flag should not use constant value "linux/arm64" (line 111) Dockerfile:24 -------------------- 23 | # install epel-release for ccache 24 | >>> RUN yum install -y yum-utils epel-release \ 25 | >>> && dnf install -y clang ccache git \ 26 | >>> && yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/sbsa/cuda-rhel8.repo 27 | ENV CC=clang CXX=clang++ -------------------- ERROR: failed to build: failed to solve: process "/bin/sh -c yum install -y yum-utils epel-release && dnf install -y clang ccache git && yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/sbsa/cuda-rhel8.repo" did not complete successfully: exit code: 255 ``` ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-29 10:26:19 -05:00
GiteaMirror commented 2026-04-29 10:26:21 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 30, 2026):

What CPU achitecture?

<!-- gh-comment-id:4158297809 --> @rick-github commented on GitHub (Mar 30, 2026): What CPU achitecture?
GiteaMirror commented 2026-04-29 10:26:21 -05:00
Author
Owner
Copy Link

@jcharth commented on GitHub (Mar 30, 2026):

6.1.0-44-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.164-1 (2026-03-09) x86_64 GNU/Linux
it is running on a i7 3770 in an old dell 7010
it seems like it will only compile in redhat. may be i need to run ollama in fedora instead of debian.

<!-- gh-comment-id:4158775768 --> @jcharth commented on GitHub (Mar 30, 2026): 6.1.0-44-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.164-1 (2026-03-09) x86_64 GNU/Linux it is running on a i7 3770 in an old dell 7010 it seems like it will only compile in redhat. may be i need to run ollama in fedora instead of debian.
GiteaMirror commented 2026-04-29 10:26:22 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 30, 2026):

If you are compiling in docker, the host system shouldn't matter. Try:

docker build -t ollama/ollama:m60 .
<!-- gh-comment-id:4158810121 --> @rick-github commented on GitHub (Mar 30, 2026): If you are compiling in docker, the host system shouldn't matter. Try: ``` docker build -t ollama/ollama:m60 . ```
GiteaMirror commented 2026-04-29 10:26:22 -05:00
Author
Owner
Copy Link

@jcharth commented on GitHub (Mar 31, 2026):

i was able to use the snap go and it compiled without gpu support

ill give it a try i got an error almost at the end. it could be because of version cuda 12.6

cmake -B build
cmake --build build

/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp: In function ‘void ggml_vk_print_gpu_info(size_t)’:
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5053:5: error: ‘VkPhysicalDeviceCooperativeMatrixFeaturesKHR’ was not declared in this scope; did you mean ‘VkPhysicalDeviceCooperativeMatrixFeaturesNV’?
 5053 |     VkPhysicalDeviceCooperativeMatrixFeaturesKHR coopmat_features;
      |     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      |     VkPhysicalDeviceCooperativeMatrixFeaturesNV
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5054:5: error: ‘coopmat_features’ was not declared in this scope
 5054 |     coopmat_features.pNext = nullptr;
      |     ^~~~~~~~~~~~~~~~
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5055:30: error: ‘VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_COOPERATIVE_MATRIX_FEATURES_KHR’ was not declared in this scope; did you mean ‘VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_COOPERATIVE_MATRIX_FEATURES_NV’?
 5055 |     coopmat_features.sType = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_COOPERATIVE_MATRIX_FEATURES_KHR;
      |                              ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      |                              VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_COOPERATIVE_MATRIX_FEATURES_NV
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp: In function ‘void ggml_vk_instance_init()’:
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5168:21: error: ‘LayerSettingEXT’ is not a member of ‘vk’
 5168 |     std::vector<vk::LayerSettingEXT> settings = {
      |                     ^~~~~~~~~~~~~~~
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5168:36: error: template argument 1 is invalid
 5168 |     std::vector<vk::LayerSettingEXT> settings = {
      |                                    ^
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5168:36: error: template argument 2 is invalid
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5172:17: error: ‘vk::LayerSettingTypeEXT’ has not been declared
 5172 |             vk::LayerSettingTypeEXT::eBool32,
      |                 ^~~~~~~~~~~~~~~~~~~
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5176:5: error: too many braces around scalar initializer for type ‘int’
 5176 |     };
      |     ^
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5177:9: error: ‘LayerSettingsCreateInfoEXT’ is not a member of ‘vk’; did you mean ‘ImageViewMinLodCreateInfoEXT’?
 5177 |     vk::LayerSettingsCreateInfoEXT layer_setting_info(settings);
      |         ^~~~~~~~~~~~~~~~~~~~~~~~~~
      |         ImageViewMinLodCreateInfoEXT
/home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5178:108: error: ‘layer_setting_info’ was not declared in this scope; did you mean ‘layer_settings’?
 5178 |     vk::InstanceCreateInfo instance_create_info(vk::InstanceCreateFlags{}, &app_info, layers, extensions, &layer_setting_info);
      |                                                                                                            ^~~~~~~~~~~~~~~~~~
      |                                                                                                            layer_settings
gmake[2]: *** [ml/backend/ggml/ggml/src/ggml-vulkan/CMakeFiles/ggml-vulkan.dir/build.make:938: ml/backend/ggml/ggml/src/ggml-vulkan/CMakeFiles/ggml-vulkan.dir/ggml-vulkan.cpp.o] Error 1
gmake[1]: *** [CMakeFiles/Makefile2:629: ml/backend/ggml/ggml/src/ggml-vulkan/CMakeFiles/ggml-vulkan.dir/all] Error 2
gmake: *** [Makefile:136: all] Error 2
<!-- gh-comment-id:4158947692 --> @jcharth commented on GitHub (Mar 31, 2026): i was able to use the snap go and it compiled without gpu support ill give it a try i got an error almost at the end. it could be because of version cuda 12.6 cmake -B build cmake --build build ``` /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp: In function ‘void ggml_vk_print_gpu_info(size_t)’: /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5053:5: error: ‘VkPhysicalDeviceCooperativeMatrixFeaturesKHR’ was not declared in this scope; did you mean ‘VkPhysicalDeviceCooperativeMatrixFeaturesNV’? 5053 | VkPhysicalDeviceCooperativeMatrixFeaturesKHR coopmat_features; | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | VkPhysicalDeviceCooperativeMatrixFeaturesNV /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5054:5: error: ‘coopmat_features’ was not declared in this scope 5054 | coopmat_features.pNext = nullptr; | ^~~~~~~~~~~~~~~~ /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5055:30: error: ‘VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_COOPERATIVE_MATRIX_FEATURES_KHR’ was not declared in this scope; did you mean ‘VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_COOPERATIVE_MATRIX_FEATURES_NV’? 5055 | coopmat_features.sType = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_COOPERATIVE_MATRIX_FEATURES_KHR; | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_COOPERATIVE_MATRIX_FEATURES_NV /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp: In function ‘void ggml_vk_instance_init()’: /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5168:21: error: ‘LayerSettingEXT’ is not a member of ‘vk’ 5168 | std::vector<vk::LayerSettingEXT> settings = { | ^~~~~~~~~~~~~~~ /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5168:36: error: template argument 1 is invalid 5168 | std::vector<vk::LayerSettingEXT> settings = { | ^ /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5168:36: error: template argument 2 is invalid /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5172:17: error: ‘vk::LayerSettingTypeEXT’ has not been declared 5172 | vk::LayerSettingTypeEXT::eBool32, | ^~~~~~~~~~~~~~~~~~~ /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5176:5: error: too many braces around scalar initializer for type ‘int’ 5176 | }; | ^ /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5177:9: error: ‘LayerSettingsCreateInfoEXT’ is not a member of ‘vk’; did you mean ‘ImageViewMinLodCreateInfoEXT’? 5177 | vk::LayerSettingsCreateInfoEXT layer_setting_info(settings); | ^~~~~~~~~~~~~~~~~~~~~~~~~~ | ImageViewMinLodCreateInfoEXT /home/john/test/ollama/ml/backend/ggml/ggml/src/ggml-vulkan/ggml-vulkan.cpp:5178:108: error: ‘layer_setting_info’ was not declared in this scope; did you mean ‘layer_settings’? 5178 | vk::InstanceCreateInfo instance_create_info(vk::InstanceCreateFlags{}, &app_info, layers, extensions, &layer_setting_info); | ^~~~~~~~~~~~~~~~~~ | layer_settings gmake[2]: *** [ml/backend/ggml/ggml/src/ggml-vulkan/CMakeFiles/ggml-vulkan.dir/build.make:938: ml/backend/ggml/ggml/src/ggml-vulkan/CMakeFiles/ggml-vulkan.dir/ggml-vulkan.cpp.o] Error 1 gmake[1]: *** [CMakeFiles/Makefile2:629: ml/backend/ggml/ggml/src/ggml-vulkan/CMakeFiles/ggml-vulkan.dir/all] Error 2 gmake: *** [Makefile:136: all] Error 2 ```
GiteaMirror commented 2026-04-29 10:26:23 -05:00
Author
Owner
Copy Link

@jcharth commented on GitHub (Mar 31, 2026):

docker build -t ollama/ollama:m60 .
has frozen multiple times. i am trying again without Xorg running. i got 24gb of ram. i notice that the vulkan libraries are old in debian 12. normal cmake fails because of missing functions. i tried work arounds with no luck. I might need to use a different distribution to compile ollama or simply run ollama. i tried debian 13 and i did not like it much but i might try it again. what linux is best to compile ollama? any free ones?

<!-- gh-comment-id:4163724136 --> @jcharth commented on GitHub (Mar 31, 2026): docker build -t ollama/ollama:m60 . has frozen multiple times. i am trying again without Xorg running. i got 24gb of ram. i notice that the vulkan libraries are old in debian 12. normal cmake fails because of missing functions. i tried work arounds with no luck. I might need to use a different distribution to compile ollama or simply run ollama. i tried debian 13 and i did not like it much but i might try it again. what linux is best to compile ollama? any free ones?
GiteaMirror commented 2026-04-29 10:26:24 -05:00
Author
Owner
Copy Link

@jcharth commented on GitHub (Apr 1, 2026):

i was able to compile ollama in debian 13 with nvidia support. using the non-free repo and the 12.4 cuda toolkit. i did not fix the problem with my tesla m60. ill keep testing more things.

<!-- gh-comment-id:4166851584 --> @jcharth commented on GitHub (Apr 1, 2026): i was able to compile ollama in debian 13 with nvidia support. using the non-free repo and the 12.4 cuda toolkit. i did not fix the problem with my tesla m60. ill keep testing more things.
GiteaMirror commented 2026-04-29 10:26:24 -05:00
Author
Owner
Copy Link

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

got ollama running in both gpus. no need to compile. i had to switch bios to disable csm and enable 4g. i used gpumodeswitch to unable compute mode.

<!-- gh-comment-id:4223534926 --> @jcharth commented on GitHub (Apr 10, 2026): got ollama running in both gpus. no need to compile. i had to switch bios to disable csm and enable 4g. i used gpumodeswitch to unable compute mode.
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#56213
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?