Repository navigation
Conversation
…0.79.0 The release counterparts of the -git pkgbases in #2: one split pkgbase (mesh-llm, skippy, CPU and Vulkan runtimes) and one per GPU runtime, built from the v0.79.0 tag tarball and the llama.cpp release it pins (b12199), with checksums. prepare() fails if the two pins drift apart. Each runtime package also installs llama.cpp's MIT license, and each directory carries the 0BSD LICENSE the AUR expects. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…y env Upstream keeps sccache and its linker drivers in .cargo/config.toml on purpose, so the packages override them with CARGO_BUILD_RUSTC_WRAPPER= and CARGO_TARGET_<TRIPLE>_LINKER=cc instead of editing the file with sed. One _cargo_env helper per PKGBUILD replaces the repeated exports. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… runtime version Upstream's runtime build fetches whisper.cpp from GitHub. The pinned commit is now a source, and the build points WHISPER_UPSTREAM_URL at the local clone; prepare() checks it against whisper-upstream.txt. MeshLLM only loads a runtime of its own release, so mesh-llm and skippy depend on skippy-native-runtime=$pkgver, and the Vulkan, CUDA and ROCm runtimes conflict with any other mesh-llm or skippy version. The runtimes need AVX2, FMA, F16C and BMI2 like upstream's (Mesh-LLM/mesh-llm#2691); an install script warns on CPUs without them. Each remaining workaround comment now names its upstream reference or says why there is none. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Important Review skippedAuto reviews are limited based on label configuration. 🏷️ Required labels (at least one): 1
Please check the settings in the CodeRabbit UI or the
|
llama.cpp b12199 -> b12340 (the commit 0.80.0 pins); whisper.cpp unchanged.
…and tool modes mesh-llm's check() asks the built mesh-llm to list every runtime of this pkgbase, as nixpkgs' runtime-discovery tests and upstream's package QA do. The CUDA and ROCm runtimes check that tools/mesh-llm-gpu-benchmark is executable: mesh-llm starts it itself and refuses it otherwise.
kleinbem
force-pushed
the
feat/runtime-discovery
branch
from
November 18, 2026 14:42
f126457 to
c55c0da
Compare
Owner
Author
|
Folded into #9, which now has this commit. |
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.
On top of #9. Mirrors the nixpkgs runtime-discovery tests:
check()in the mesh-llm pkgbase asks the builtmesh-llmto list the CPU and Vulkan runtimes it built, and the CUDA and ROCm pkgbases check thattools/mesh-llm-gpu-benchmarkis executable (mesh-llm starts it itself; Mesh-LLM/mesh-packaging#47 is the same bug in upstream's packages).The CUDA and ROCm pkgbases build no mesh-llm, so their discovery is not checked here.
CI builds all three pkgbases; that run is the test.
🤖 Generated with Claude Code