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

[PR #5582] [CLOSED] Remove nested runner payloads from linux #11836

New Issue
Closed
opened 2026-04-12 23:40:18 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 23:40:18 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/5582
Author: @dhiltgen
Created: 7/9/2024
Status: ❌ Closed

Base: main ← Head: no_linux_payloads

📝 Commits (1)

  • a043647 Remove nested runner payloads from linux

📊 Changes

11 files changed (+114 additions, -106 deletions)

View changed files

📝 Dockerfile (+15 -19)
📝 envconfig/config.go (+4 -4)
📝 gpu/assets.go (+26 -24)
📝 gpu/gpu.go (+4 -4)
📝 llm/generate/gen_common.sh (+14 -1)
📝 llm/generate/gen_linux.sh (+34 -42)
📝 llm/llm_linux.go (+1 -1)
📝 llm/payload.go (+1 -1)
📝 llm/server.go (+1 -1)
📝 scripts/build_linux.sh (+4 -6)
📝 scripts/install.sh (+10 -3)

📄 Description

This adjusts linux to follow the same model we use for windows with a discrete archive (zip/tgz) to cary the primary executable, subprocess runners, and dependent libraries.

Darwin retain the payload model where the go binary is fully self contained.

Marking draft as it still needs some more testing and CI will need adjusting, but initial happy path looks good.

For comparison, the current (v0.2.1) artifacts are:

  • ollama-linux-amd64 - 467 MB
  • ollama-linux-amd64-rocm.tgz - 1.14 GB

With this PR:

% ls -lh ./dist/ollama-linux-amd64.tgz
-rw-r--r--  1 daniel  staff   1.1G Jul  9 13:19 ./dist/ollama-linux-amd64.tgz

After extracting on a system:

% ls -F
cuda/  ollama*  ollama_runners/  rocm/
% du -sh .
4.6G	.

Note: I opted to include rocm into the single artifact, since we do the same on Windows, and that simplifies the overall logic. That said, it's the brunt of the extracted size so this may be an area worth optimizing. One option might be to simply exclude the rocm directory during the extract in the install script if no AMD GPUs are detected.

% du -sh rocm
3.6G	rocm

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/ollama/ollama/pull/5582 **Author:** [@dhiltgen](https://github.com/dhiltgen) **Created:** 7/9/2024 **Status:** ❌ Closed **Base:** `main` ← **Head:** `no_linux_payloads` --- ### 📝 Commits (1) - [`a043647`](https://github.com/ollama/ollama/commit/a043647bb7cb8a7e08c6965b433eef09d5f7d47d) Remove nested runner payloads from linux ### 📊 Changes **11 files changed** (+114 additions, -106 deletions) <details> <summary>View changed files</summary> 📝 `Dockerfile` (+15 -19) 📝 `envconfig/config.go` (+4 -4) 📝 `gpu/assets.go` (+26 -24) 📝 `gpu/gpu.go` (+4 -4) 📝 `llm/generate/gen_common.sh` (+14 -1) 📝 `llm/generate/gen_linux.sh` (+34 -42) 📝 `llm/llm_linux.go` (+1 -1) 📝 `llm/payload.go` (+1 -1) 📝 `llm/server.go` (+1 -1) 📝 `scripts/build_linux.sh` (+4 -6) 📝 `scripts/install.sh` (+10 -3) </details> ### 📄 Description This adjusts linux to follow the same model we use for windows with a discrete archive (zip/tgz) to cary the primary executable, subprocess runners, and dependent libraries. Darwin retain the payload model where the go binary is fully self contained. Marking draft as it still needs some more testing and CI will need adjusting, but initial happy path looks good. For comparison, the current (v0.2.1) artifacts are: - `ollama-linux-amd64` - 467 MB - `ollama-linux-amd64-rocm.tgz` - 1.14 GB With this PR: ``` % ls -lh ./dist/ollama-linux-amd64.tgz -rw-r--r-- 1 daniel staff 1.1G Jul 9 13:19 ./dist/ollama-linux-amd64.tgz ``` After extracting on a system: ``` % ls -F cuda/ ollama* ollama_runners/ rocm/ % du -sh . 4.6G . ``` Note: I opted to include rocm into the single artifact, since we do the same on Windows, and that simplifies the overall logic. That said, it's the brunt of the extracted size so this may be an area worth optimizing. One option might be to simply exclude the rocm directory during the extract in the install script if no AMD GPUs are detected. ``` % du -sh rocm 3.6G rocm ``` --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
GiteaMirror added the pull-request label 2026-04-12 23:40:18 -05:00
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 pull-request
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#11836
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?