Fix v0.4.0 mtmd helper compatibility - #72
Merged
Merged
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
The automatic llama.cpp v0.4.0 release failed compiling the TTS speaker-reference path:
mtmd_helper_bitmap_init_from_bufnow requires helper options. Detect that API at configure time and pass upstream defaults when available, retaining the four-argument path for the currently pinned v0.3.0.Add a model-free, real-image decode test and run the complete wrapper contract against both the pinned revision and immutable v0.4.0 commit
5266f24da75dc449bd56cbed7addb9c8e4a6a73ein PR CI. The originalwrapper-contractcheck name is retained.Closes #71.
Validation
tools/tts_smoke.cpp. Cancellation/reset, two sequence syntheses, finite/non-silent 24 kHz PCM, and WAV output passed. Output WAVs were byte-identical: SHA-25616dc619f392f905d32b9b03c588ba4d424218ff9830096207bb99673273c6cf9.8d18c94acb2addd042f97da63c98be144eafa76d0d9495177eab65130cf85129; projector6fd65188839bcd6ecc91b277ad471e22a0edfada4699a0fe82f1165c18cfcce2; speaker WAVde964f9d761766be44fdf648c961f9232cd53cb14373c20b6046106f06661203. TextHi, languageen, seed 1, Apple M4 Max CPU. This proves speaker-reference integration, not voice-fidelity or hardware-GPU qualification.git diff --check: PASS.80278f2c8cb96bb0080be16a95081323dee5c6b6; no remaining blocker. Reviewer independently reran the test executables and export checks for both revisions.The exported wrapper ABI, submodule pin, release policy, and downstream pins are unchanged. A complete successful release matrix is still required before any v0.4.0 artifact is published. The equivalent Web call-site prerequisite is tracked separately in leehack/llama-web-bridge#90.