Skip to content

HIP: bump requirement to rocm 6.1#15296

Merged
IMbackK merged 2 commits into
ggml-org:masterfrom
IMbackK:rocm61
Aug 13, 2025
Merged

HIP: bump requirement to rocm 6.1#15296
IMbackK merged 2 commits into
ggml-org:masterfrom
IMbackK:rocm61

Conversation

@IMbackK

@IMbackK IMbackK commented Aug 13, 2025

Copy link
Copy Markdown
Collaborator

Unfortunately due to issues in amd_hip_bf16.h, we can not longer support rocm <6.1

I am not happy about this, but at least rocm 6.1 is what is currently in debian stable, so i believe we still cover all reasonable installations of rocm. There where no changes to hardware support between rocm 5.5 and rocm 6.1 so no hardware is left behind either.

@github-actions github-actions Bot added Nvidia GPU Issues specific to Nvidia GPUs devops improvements to build systems and github actions ggml changes relating to the ggml tensor library for machine learning labels Aug 13, 2025
Comment thread ggml/src/ggml-cuda/common.cuh
@IMbackK IMbackK merged commit 29c8fbe into ggml-org:master Aug 13, 2025
47 checks passed
@CISC

CISC commented Aug 15, 2025

Copy link
Copy Markdown
Member

@IMbackK Not related to this, but I've seen a couple of times now that the windows-latest-cmake-hip build gets stuck on the Install step for hours, any idea?

https://github.com/ggml-org/llama.cpp/actions/runs/16989547440/job/48165519619

Nexesenex added a commit to Nexesenex/croco.cpp that referenced this pull request Oct 6, 2025
@ckastner

Copy link
Copy Markdown
Collaborator

Update: we shipped a fix for bf16 HIP bug in the last Debian stable update, 13.2 from mid-November. Specifically, with libamdhip64-dev version 5.7.1-6+deb13u1.

I undid the change of this PR in a b6441 release (had to use an older one for reasons), so back to 5.5, and it built and ran fine. Just in case this is useful to someone.

blime4 referenced this pull request in blime4/llama.cpp Feb 5, 2026
Seunghhon pushed a commit to Seunghhon/llama.cpp that referenced this pull request Apr 26, 2026
ljubomirj pushed a commit to ljubomirj/llama.cpp that referenced this pull request May 6, 2026
my-other-github-account pushed a commit to my-other-github-account/llama.cpp that referenced this pull request May 15, 2026
my-other-github-account pushed a commit to my-other-github-account/llama.cpp that referenced this pull request May 15, 2026
phibya pushed a commit to ziee-ai/llama.cpp that referenced this pull request May 29, 2026
fewtarius pushed a commit to fewtarius/llama.cpp that referenced this pull request May 30, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

devops improvements to build systems and github actions ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants