Repository navigation
Bump llama.cpp to b11612 and prepare for patch 0012 drop - #536
Merged
Merged
Conversation
One step over 38 upstream commits; the bulk of the 6.4 MB diff is data (docs/ops/zDNN.csv, the NFC tables of src/unicode-data.cpp, the consolidated context-state tests). The priority list has one API change, common_chat_params::prompt as common_chat_prompt parts (#30389), which nothing here reads. No project source change. Patches: 0001 loses its tests/test-state-restore-fragmented.cpp hunk (the file went with #30023), 0014 is regenerated around upstream's new common_log_set_stdo() with the sink keeping precedence over the console switch (pinned by a new test case), the other eleven apply unchanged. check-patches.py 13/101/0, ctest 608/608, the model-free native Java tests 27/27 over the b11612 library. Patch 0012 stays: upstream merged the same even-split fallback as ggml-org/llama.cpp#29677 (df18bee11) on 2026-10-14, but 15 commits past b11612, so the pin still carries the bare division. The patch header, its CLAUDE.md row, test_model_split.cpp and a new TODO.md section record the fix with its link and the rule for the bump that crosses it: drop 0012 instead of refreshing it, and replace the guard, which links the two functions only the patch provides, by a test over upstream's own load path rather than keeping a residue of the patch for it. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017kVXQfPe1ACLAuk29Y4d2H
bernardladenthin
had a problem deploying
to
startgate
October 14, 2026 10:31 — with 
GitHub Actions
Error
GitHub Actions
Error
bernardladenthin
had a problem deploying
to
maven-central
October 14, 2026 10:31 — with
GitHub Actions
Failure
GitHub Actions
Failure
bernardladenthin
had a problem deploying
to
maven-central
October 14, 2026 10:31 — with
GitHub Actions
Failure
GitHub Actions
Failure
|
5 of 6 tasks
This branch had an error being deployed
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
2 participants
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.
Summary
0012at the next bump past b11612, where upstream merged the fix independently (model: fix "invalid vector subscript" when loading on a full GPU ggml-org/llama.cpp#29677)0012header with status and drop instructionscommon_log_set_stdo()(b11612 #30023)0014hunks to apply cleanly against b11612Test plan
test_common_log_callback.cppnew test added)Related issues / PRs
Upstream: ggml-org/llama.cpp#29677 (merged 2026-10-14, commit df18bee11)
Checklist
CONTRIBUTING.mdandCODE_OF_CONDUCT.mdhttps://claude.ai/code/session_017kVXQfPe1ACLAuk29Y4d2H