90 Commits
Author SHA1 Message Date
Sreeram f660cb0ae8 Add ROCm 10.2.0 builds and stop one ROCm version from cancelling or withholding the others 2026-11-19 12:53:37 -08:00
Sreeram 0d18143f31 ci: pin Linux CPU/Vulkan builds to Ubuntu 24.04 and drop retired macos-14 leg 2026-11-18 16:53:09 -08:00
Sreeram 90e86d183d Add ROCm 10.1.0 builds (#19) 2026-10-09 13:13:33 -05:00
Harkirat Gill f5df645e22 ci: bundle HIP runtime libraries with Windows ROCm release (#18) 2026-10-02 14:13:27 -05:00
Mario Limonciello (AMD) 017cc8e5b5 Fix ROCm ccache build caching
Using mtime doesn't work with wheels, need to compare content.
2026-08-27 09:19:49 -05:00
Mario Limonciello 879c67a96f Update ROCm to 10.0 release 2026-08-27 09:18:42 -05:00
Ken VanDineandClaude Opus 5 aeaf146d5c ci: install only the CUDA components the Linux builds use
Both Linux CUDA jobs installed the cuda-toolkit-12-9 metapackage, which
apt reports as 3935 MB of archives per job. That is 16 jobs per nightly
run on the x64 matrix plus the arm64 leg, and it made "Install CUDA
Toolkit" the second-largest cost in those jobs at 36.5 min across the
x64 legs.

Most of it is never used. The metapackage pulls Nsight Systems and
Compute, cuFFT, cuSPARSE, cuSOLVER and NPP; ggml-cuda links only
CUDA::cudart, CUDA::cublas and CUDA::cuda_driver plus CCCL for
thrust/cub, and stable-diffusion.cpp links only ggml. Install that
subset instead, ~595 MB of direct archives, and add a guard that names
the missing file if NVIDIA ever repackages one of them rather than
failing partway through the compile.

Package names verified present in both developer.download.nvidia.com
repos the jobs use, ubuntu2204/x86_64 and ubuntu2404/sbsa.

The Windows CUDA installs already select sub-packages through
Jimver/cuda-toolkit and were left alone.

Co-Authored-By: Claude Opus 5 <[email protected]>
2026-07-28 22:07:42 -04:00
Ken VanDineandClaude Opus 5 382f9e0c9b ci: CUDA build improvements
Two changes, both measured against run 30332616539.

Cache the windows-latest-cmake build. Its cuda12 leg compiles nine CUDA
architectures in a single job and had no ccache step at all, making it
the second-longest job in the run at 90.6 min of Build. windows-latest-cuda
runs the same MSVC/Ninja/Release setup with ccache and averages 1.6 min
per leg at 97.96% hit rate. No cmake flags are needed: ggml's
src/CMakeLists.txt sets RULE_LAUNCH_COMPILE when it finds ccache on PATH
(GGML_CCACHE defaults to ON), which is how windows-latest-cuda already
gets its hits. The cuda12 leg gets max-size 2G because nine
architectures plus GGML_CPU_ALL_VARIANTS will not fit the 500M default
that suits the cpu and vulkan legs.

Parallelize xz when packing the Linux CUDA artifacts. tar -cJf drives a
single-threaded xz and was the largest cost in those jobs, 50.9 min
across the eight x64 legs plus 5.9 min on arm64, against only 12.8 min
of Build. Measured on a 2.9 GB binary payload: 12m41s single-threaded
versus 47.7s on 32 cores, with the archive 1.8% larger because threaded
mode compresses independent blocks. The runners are 4-core, so expect
roughly 3-4x there.

ccache itself was already healthy everywhere it was configured, at
98-100% hit rates, and the release job's artifact download is fine on a
full ubuntu-latest VM. Neither needed changing.

Co-Authored-By: Claude Opus 5 <[email protected]>
2026-07-28 22:01:27 -04:00
Mario Limonciello 65745b48ed Attempt to align to upstream jobs 2026-07-22 16:23:54 -05:00
Mario Limonciello d43a5df8d8 Align vulkan job with upstream 2026-07-22 11:11:41 -05:00
Mario Limonciello (AMD) bd6bbfa8f2 Add vulkan jobs 2026-07-21 23:29:05 -05:00
Mario Limonciello 506bb58f28 Update workflows for ROCm 7.14 2026-07-21 11:02:16 -05:00
Ken VanDineandClaude Sonnet 5 6296594e1b ci: prune stale release assets from families dropped in the current run
The release is reused across runs against the same tag (idempotent
create-or-get). If a family (e.g. rocm-therock) was complete and
published in an earlier run but the current run finds it incomplete and
drops it locally, its assets from that earlier run were never removed —
the release would keep looking complete for a family this run couldn't
actually reproduce.

The upload step now reconciles existing release assets against the
current ./artifact file set: anything not fully uploaded (a broken
partial upload, as before) or not present in this run's file set (stale
leftovers from a dropped family) gets deleted before uploading.

Addresses review feedback on lemonade-sdk/stable-diffusion.cpp#14. The
always() fix requested in the same review was already applied in
03e249d.

Co-Authored-By: Claude Sonnet 5 <[email protected]>
2026-07-02 10:31:42 -04:00
Ken VanDineandClaude Sonnet 5 40969fbb32 ci: run release job even when a build job fails
Without always(), GitHub Actions' default skip-if-a-needed-job-failed
behavior meant a single failed build job (e.g. one CUDA sm_* matrix leg,
several of which use fail-fast: false) could skip the release job
outright, before the per-family completeness check ever ran — silently
producing no release instead of a correct partial one.

The other issue found in the equivalent llama.cpp fix (a hard-fail step
that could block unrelated families) doesn't apply here: this workflow
has no cross-job merge step comparable to llama.cpp's Windows CPU-into-
other-zips step, so there's nothing else that needs the same treatment.

Follow-up to #14.

Co-Authored-By: Claude Sonnet 5 <[email protected]>
2026-07-02 10:31:42 -04:00
Ken VanDineandClaude Sonnet 5 f41bf4f420 ci: fail release if expected artifacts are missing, per backend family
Port the artifact-completeness gate from lemonade-sdk/llama.cpp#16 to
this repo's release job, adapted to its build matrix:

- cuda: 8 sm_* variants on ubuntu x64 + windows x64, plus sm_121-only
  on ubuntu arm64 (GB10) — 17 files total.
- rocm-therock: TheRock nightly ROCm 7.13.0 on windows + ubuntu.
- rocm-hipsdk: official/legacy HIP SDK ROCm on windows (7.1.1) +
  ubuntu (7.2.1) — kept separate from rocm-therock since they're
  different SDKs, not just different platforms of the same backend.
- cpu: ubuntu + windows avx2 builds.
- metal: macos-14 + macos-latest builds.

Each family is all-or-nothing: if any artifact in a family is
missing, every artifact for that family is dropped from the release
so a release never contains a partial/inconsistent set (e.g. CUDA
split across sm_* or platforms). Other families are unaffected by a
gap in one. A release with zero artifacts overall is never created.

Also brings the release/upload mechanism up to parity with
llama.cpp's: an idempotent create-or-get-release step (safe to
re-run after a partial failure) and a retrying, resumable asset
upload, replacing the old anzz1/action-create-release +
github-script@v3 flow.

Co-Authored-By: Claude Sonnet 5 <[email protected]>
2026-07-02 10:31:42 -04:00
Ken VanDine bd1c5f2d43 ci(windows-cuda): pass /bigobj to fix C1128 build failure
stable-diffusion.cpp's Windows CUDA build fails with:

  fatal error C1128: number of sections exceeded object file format
  limit: compile with /bigobj

src/stable-diffusion.cpp combined with the CUDA backend's generated
code exceeds MSVC's default COFF section limit per object file. The
windows-latest-cmake job already works around the same class of issue
with -DCMAKE_CXX_FLAGS='/bigobj'; apply the same flag to the CUDA
build.
2026-07-02 09:42:22 -04:00
Ken VanDineandCopilot Autofix powered by AI 8cd9abef0f Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <[email protected]>
2026-06-11 17:16:58 -04:00
Ken VanDineandCopilot 9e10190500 ci: bump Jimver/cuda-toolkit to v0.2.35 with CUDA 12.9.1
v0.2.22 does not include CUDA 12.9.x; v0.2.24 was the first release to
add it. Bump to latest v0.2.35 and use 12.9.1 (latest 12.9 patch).

Co-authored-by: Copilot <[email protected]>
2026-06-11 16:45:10 -04:00
Ken VanDineandCopilot f19563e9e3 ci: upgrade CUDA to 12.9 for sm_121 support
sm_121 (GB10/Blackwell) requires CUDA 12.9. Update ubuntu-latest-cuda,
windows-latest-cuda, and ubuntu-arm64-cuda jobs accordingly.

Co-authored-by: Copilot <[email protected]>
2026-06-11 16:15:03 -04:00
Ken VanDineandCopilot a8ec287fa5 ci: add sm_121 (GB10) CUDA arch for x86 and arm64
- Add sm_121 to ubuntu-latest-cuda and windows-latest-cuda matrices
- Add new ubuntu-arm64-cuda job targeting ubuntu-24.04-arm runner with sm_121
- Use CUDA sbsa (ARM64) package repo for arm64 CUDA Toolkit install
- Add ubuntu-arm64-cuda to release job needs

Co-authored-by: Copilot <[email protected]>
2026-06-11 15:29:38 -04:00
Ken VanDineandCopilot fb10de9a5c fix(cuda): add -DGGML_CUDA=ON to Windows CUDA build
The Windows CUDA cmake invocation was missing -DGGML_CUDA=ON, which
meant the ggml CUDA backend was never compiled in. Without it, the
build silently falls back to CPU even when the CUDA toolkit is
installed. This is the same fix already applied to the Linux CUDA
build in commit 6a68a98.

Fixes: lemonade-sdk/lemonade#2171 (sdcpp:cuda falls back to CPU on Windows + RTX 4090)

Co-authored-by: Copilot <[email protected]>
2026-06-09 21:42:34 -04:00
Ken VanDine b638fe9228 Merge pull request #11 from Phqen1x/lemonade
Standardize and fix CUDA artifact naming and packaging
2026-06-05 17:03:03 +02:00
Drew VanDineandClaude Sonnet 4.6 6a68a98599 ci: fix CUDA compilation and bundle all ggml shared libs
Add -DGGML_CUDA=ON and -DCMAKE_CUDA_COMPILER explicitly to the Linux
CUDA cmake invocation so CUDA is actually compiled instead of silently
falling back to CPU when auto-detection fails.

Widen the ggml artifact collection from libggml-cuda.so only to
libggml*.so* (Linux) and ggml*.dll (Windows) to capture all ggml shared
libraries (core, base backend, cuda backend) regardless of where cmake
places them in the build tree.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-06-05 09:08:23 -04:00
Ken VanDine c2cc188ff3 Merge pull request #10 from Phqen1x/bundle-ggml-cuda
ci: bundle libggml-cuda.so/ggml-cuda.dll in CUDA release artifacts
2026-06-04 22:35:49 +02:00
Drew VanDineandClaude Sonnet 4.6 df4efd59c6 ci: bundle libggml-cuda.so/ggml-cuda.dll in CUDA release artifacts
The ggml CUDA backend is built as a shared library plugin when
SD_BUILD_SHARED_LIBS=ON. Bundle it alongside the CUDA runtime libs so
that CUDA inference works out of the box without the plugin being
missing at runtime.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-06-04 16:22:51 -04:00
Ken VanDine e3937d9654 Merge pull request #8 from kenvandine/add_ci_check
ci: validate release build on pull requests
2026-06-01 14:43:50 +02:00
copilot-swe-agent[bot] c3af5c187b Fix: use least-privilege permissions in CI workflow 2026-06-01 12:41:37 +00:00
Ken VanDineandClaude Opus 4.8 429bf57c94 ci: validate release build on pull requests
Add a pull_request trigger to build.yml so the release build is
validated on PRs, mirroring lemonade-sdk/llama.cpp. The pack/upload
and release jobs remain gated on schedule/create_release, so nothing
is published on a PR run.

Scope the trigger to the workflow file itself: the build jobs clone
source from upstream (leejet/stable-diffusion.cpp), so only changes
to this workflow can affect a PR run's outcome.

On PRs, the CUDA matrices build a single representative arch (sm_89)
instead of all seven; packaging logic is identical across sm_*, and
the full matrix still runs on schedule/dispatch.

Co-Authored-By: Claude Opus 4.8 <[email protected]>
2026-06-01 07:41:33 -04:00
copilot-swe-agent[bot] 9cffd0da35 Merge origin/lemonade into fix-powershell-parser 2026-06-01 03:10:30 +00:00
Drew VanDineandClaude Sonnet 4.6 7437d050c0 ci: fix PowerShell parser error in Windows CUDA pack step
Use multi-line array literal format for $runtimeDllPatterns and remove
the trailing backslash from the Copy-Item destination path, which was
causing an 'Unexpected token }' parser error at the foreach closing brace.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 23:07:24 -04:00
Drew VanDine 3930b7bc5b Merge pull request #10 from Phqen1x/fix-cuda-clone-ref
ci: pin CUDA clone steps to ref: master
2026-05-31 22:55:33 -04:00
Drew VanDineandClaude Sonnet 4.6 fa19b4115b ci: pin CUDA clone steps to ref: master
Without an explicit ref, actions/checkout makes a GitHub API call to
determine the default branch of leejet/stable-diffusion.cpp. With 14
parallel CUDA jobs all doing this simultaneously this call can fail,
presenting as a checkout error. Pinning to master skips the API call.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 22:53:11 -04:00
Ken VanDineandCopilot Autofix powered by AI 73bb113fc7 Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <[email protected]>
2026-05-31 22:38:08 -04:00
Drew VanDineandClaude Sonnet 4.6 8018705329 ci: match lemonade-sdk/llama.cpp CUDA library bundling
Linux:
- Add libcudart.so*, libnvJitLink.so* to bundled runtime libs
- Use cp -av (preserves symlinks) instead of cp -v
- RPATH step now patches all ELF files via file(1) instead of glob

Windows:
- Add curand and nvjitlink to CUDA toolkit sub-packages
- Copy cudart64, cublas64, cublasLt64, curand64, nvJitLink DLLs
  from CUDA_PATH\bin before zipping

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 22:09:36 -04:00
Drew VanDineandClaude Sonnet 4.6 cb80321973 ci: bundle CUDA runtime libs for portability on both platforms
Linux: add libcudart.so* to the bundled runtime libraries so binaries
can run on systems without a matching CUDA toolkit installed.

Windows: add id to cuda-toolkit step and robocopy cudart64/cublas64/
cublasLt64 DLLs into the zip so the build runs on machines without a
pre-installed CUDA runtime.

Addresses Copilot review comments on lemonade-sdk/stable-diffusion.cpp#7.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 21:55:41 -04:00
Drew VanDineandClaude Sonnet 4.6 c69f4790fa ci: revert Windows CUDA artifacts back to .zip
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 20:03:15 -04:00
Drew VanDineandClaude Sonnet 4.6 b55463d29b ci: package Windows CUDA artifacts as .7z (matches Lemonade expectation)
Lemonade's sd_server.cpp builds the Windows CUDA download URL with a
.7z extension. The CI was producing .zip, causing a 404 on download
which presented as an extraction failure.

Also adds .7z to the release upload script allowlist.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 19:58:49 -04:00
Drew VanDineandClaude Sonnet 4.6 36778b2765 ci: standardize CUDA artifact names and compression formats
Linux:   sd-master-{hash}-ubuntu-cuda-sm_XX-x64.tar.xz
Windows: sd-master-{hash}-windows-cuda-sm_XX-x64.zip

Also adds .tar.xz support to the release upload script.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 16:10:52 -04:00
Drew VanDineandClaude Sonnet 4.6 cb32062435 ci: revert CUDA artifact names to original sd-*-bin-* format
Restores the filenames introduced in the initial CUDA PR:
  sd-master-{hash}-bin-Linux-Ubuntu-22.04-x86_64-cuda-{sm}.zip
  sd-master-{hash}-bin-win-cuda-{sm}-x64.zip

Reverts the intermediate ubuntu-cuda-* and windows-cuda-* name
experiments, and drops the .tar.xz branch from the release upload script.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 16:04:17 -04:00
Drew VanDineandClaude Sonnet 4.6 adb7fd07ec ci: use sd-{branch}-{hash}-ubuntu/windows-cuda-{sm}-x64 artifact names
Renames CUDA artifacts to match the format Lemonade SDK expects:
  sd-master-148f69f-ubuntu-cuda-sm_89-x64.tar.xz
  sd-master-148f69f-windows-cuda-sm_89-x64.zip

Also restores the sd-* download pattern in the release job and keeps
.tar.xz support in the upload script.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 15:54:35 -04:00
Drew VanDineandClaude Sonnet 4.6 a0d011290f ci: fix CUDA artifact names to match Lemonade SDK expectations
Linux CUDA artifacts are now named ubuntu-cuda-sm_XX-x64.tar.xz and
Windows CUDA artifacts are named windows-cuda-sm_XX-x64.zip, matching
the filenames Lemonade SDK constructs when downloading from releases.

Also removes the sd-* pattern filter from the release download step so
CUDA artifacts (which don't carry the sd- prefix) are included, and
adds .tar.xz support to the release upload script.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 15:47:35 -04:00
Drew VanDineandClaude Sonnet 4.6 3a5581f80d ci: add Linux and Windows CUDA builds per compute capability
Adds ubuntu-latest-cuda and windows-latest-cuda jobs covering sm_75
through sm_120, modeled after the llama.cpp CUDA CI. Linux installs
CUDA 12.8 via apt, bundles cublas runtime libs, and patches RPATHs.
Windows uses Jimver/[email protected] with MSVC + Ninja Multi-Config.
Both jobs are wired into the release job.

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
2026-05-31 11:55:05 -04:00
George Sofianos 148f69fe31 fix: add RDNA1 + RDNA2 targets for ROCm 7.13 (#6) 2026-05-17 19:33:06 +03:00
Mario Limonciello (AMD) 8acff7156f Update to 7.13 2026-05-14 23:45:31 -05:00
geramyloveless a466f7e46d ci: pin Linux x86_64 cpu+rocm builds to AVX2+FMA+F16C (drop -march=native)
ggml's CMake defaults GGML_NATIVE=ON, which adds -march=native and silently
enables every extension the build runner's CPU advertises. Recent Azure
ubuntu-latest hosts are Xeon Platinum 8370C (AVX-512 capable), so the
master-593 release of libstable-diffusion.so was built with 5,396 zmm /
1,582 k-mask AVX-512 instructions and SIGILLs immediately on any consumer
or CI host without AVX-512.

Pin the Linux x86_64 cpu and ROCm builds to GGML_NATIVE=OFF +
GGML_AVX2=ON + GGML_FMA=ON + GGML_F16C=ON — a portable Haswell-era
baseline that downstream consumers (Lemonade, end users) can rely on.

The Windows AVX2 cpu build and Windows ROCm builds already use
GGML_NATIVE=OFF; this brings Linux into line.

Verified by inspecting libstable-diffusion.so:
  master-569 (no AVX-512): 0 zmm refs, 0 k-mask refs
  master-593 (AVX-512):    5,396 zmm refs, 1,582 k-mask refs

The crash this fixes manifests in lemonade-sdk/lemonade CI as
sd-server "exit code: -1" within 200ms on the GitHub-hosted ollama
test job, while the self-hosted ROCm runner (full AVX-512) succeeds.
2026-05-06 12:40:46 -07:00
Geramy Loveless 7f65f2acce ci: add macOS arm64 (Metal) build (#2)
* ci: add macOS arm64 (Metal) build for stable-diffusion.cpp

Adds a macos-arm64-cmake job covering [macos-14, macos-latest] that builds
leejet/stable-diffusion.cpp with Metal enabled and produces a release-ready
sd-*-bin-Darwin-macOS-<ver>-arm64.zip artifact, mirroring the existing Linux
job's naming convention. Wired into the release job so macOS arm64 binaries
are attached alongside Linux/Windows on scheduled and manual releases.

* ci(macos): bake @loader_path rpath so dylib loads next to binary

Without this, sd-cli ships with the GitHub runner's absolute build path
hardcoded in its rpath and fails at runtime with 'Library not loaded:
@rpath/libstable-diffusion.dylib'. Mirrors llama.cpp's release.yml flags.

* ci(macos): use stable artifact filename without OS version

Drop the dynamic Darwin-macOS-{sw_vers}-arm64 suffix in favor of a stable
sd-{branch}-{hash}-bin-Darwin-arm64.zip name (with -latest variant for
macos-latest). The previous naming embedded the runner image's macOS
point release (e.g. 14.8.5, 15.7.4), which churns silently as GitHub
patches the image and would break downstream consumers (Lemonade C++)
that download by exact filename. Matches the fixed-name pattern already
used by the Windows builds.

* ci: grant contents:write so the release job can publish artifacts

Without this, the workflow's GITHUB_TOKEN defaults to read-only on forks
and the 'Create release' / 'Upload release' steps fail with HTTP 403
'Resource not accessible by integration'. Same fix the Linux/Windows
release flow needs to function on Geramy's fork; safe on upstream where
the org-default token already has write access (this just makes it
explicit and forks-portable).

* ci(macos): name artifact …Darwin-arm64-metal.zip to align with backend ID

Lemonade's recipe matrix is adopting a 'metal' backend identifier for
macOS arm64 (mirroring the existing llamacpp/metal pattern). Renaming
the artifact's suffix to match keeps Lemonade C++'s download URL
construction trivial (filename = sd-{ver}-bin-Darwin-arm64-metal.zip)
and signals to consumers that this build has Metal acceleration enabled
(GGML_METAL=ON GGML_METAL_USE_BF16=ON GGML_METAL_EMBED_LIBRARY=ON).
2026-05-01 01:00:13 -05:00
Mario Limonciello (AMD) 8d8f55a3f1 Revert "Pack amdhip64_7.dll"
This reverts commit 06dd7b4041.
2026-04-16 22:07:43 -05:00
Mario Limonciello (AMD) b977b00edc Revert "Fix path"
This reverts commit 0134eac6d2.
2026-04-16 22:07:42 -05:00
Mario Limonciello 0134eac6d2 Fix path 2026-04-16 19:06:02 -05:00
Mario Limonciello 06dd7b4041 Pack amdhip64_7.dll 2026-04-16 18:56:42 -05:00