mirror of
https://github.com/mostlygeek/llama-swap.git
synced 2026-06-09 14:56:34 +02:00
Benson Wong
ccfba0df28
Build Containers / build-and-push (intel) (push) Failing after 47s
Build Containers / build-and-push (rocm) (push) Failing after 11m36s
Build Containers / build-and-push (musa) (push) Failing after 11m38s
Build Containers / build-and-push (cuda13) (push) Failing after 13m19s
Build Containers / build-and-push (cuda) (push) Failing after 14m56s
Build Containers / build-and-push (cpu) (push) Failing after 14m58s
Build Containers / build-and-push (vulkan) (push) Failing after 14m49s
Build Containers / delete-untagged-containers (push) Failing after 14m59s
Replace TARGETARCH build-arg with runtime arch detection via uname -m. BuildKit's TARGETARCH injection was unreliable for the multi-arch cpu build, causing the arm64 image variant to download and embed the x86_64 llama-swap binary — resulting in "exec format error" on arm64 hosts. With QEMU user-space emulation, uname -m correctly returns aarch64 inside an arm64 container build, so the download always fetches the right binary for the actual target architecture. Also adds --fail to curl so HTTP 404s produce a build error instead of silently embedding an HTML error page. fixes #818 Co-authored-by: Claude <noreply@anthropic.com>