Repository navigation
Build option for private HIP kernels on HRX (-DONEBIT_GPU_PRIVATE) - #307
Merged
Merged
Conversation
|
Docs7 for 1bit-monster/engine
Commit |
bong-water-water-bong
marked this pull request as ready for review
October 5, 2026 11:21
bong-water-water-bong
enabled auto-merge (squash)
October 5, 2026 11:21
ONEBIT_GPU_PRIVATE=<gpu-kernels checkout> (needs ONEBIT_HRX) forwards GGML_HRX_HIP_ADDON_DIR=<checkout>/addons/hip-hrx to the HRX llama.cpp external project, which compiles that add-on's kernels and matchers into ggml-hrx. Configure fails if the checkout has no addons/hip-hrx/kernels or the llama.cpp pin has no GGML_HRX_HIP_ADDON_DIR hook. Without the option nothing changes. docs/hrx.md: "Private HIP kernels". Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
bong-water-water-bong
force-pushed
the
1bit/gpu-private-hook
branch
from
October 5, 2026 11:23
da5a6ea to
ce7620f
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stacked on #306. Draft: do not merge yet. The release is on hold.
-DONEBIT_GPU_PRIVATE=<checkout>builds a private HIP kernel add-on into the HRX llama.cpp build. It passesGGML_HRX_HIP_ADDON_DIR=<checkout>/addons/hip-hrxto thellama_hrxExternalProject. This mirrors-DONEBIT_NPU_PRIVATE.ONEBIT_HRX=ON, and configure fails clearly without it.🤖 Generated with Claude Code