Repository navigation
Comparing changes
Open a pull request
base repository: abetlen/llama-cpp-python
base: main
head repository: ByCzech/llama-cpp-python
compare: main
- 19 commits
- 17 files changed
- 2 contributors
Commits on Aug 27, 2026
-
feat: DEB packages - Add modular CPU, Vulkan, CUDA, ROCm and SYCL bac…
…kend packages * split llama-cpp-python into core and dynamically loadable backend packages * add CPU backend with multiple optimized CPU variants and OpenBLAS * add Vulkan backend * add co-installable CUDA 12.4, 12.9 and 13.3 backends using update-alternatives * add ROCm 7.14 HIP backend with support for current AMD GPU architectures * add Intel oneAPI SYCL backend with oneMKL, oneDNN and Level Zero support * add transitional packages for legacy package names * add backend alternatives handling and private shared-library dependency resolution * add CUDA 12.9 compatibility handling for Debian Trixie/glibc 2.41 * isolate external CUDA, ROCm and oneAPI toolchains from Debian system libraries * update package dependencies, shlibs handling, cleanup rules and runtime RPATHs
Jaroslav Mikulík committedAug 27, 2026 Configuration menu - View commit details
-
Copy full SHA for bb4a154 - Browse repository at this point
Copy the full SHA bb4a154View commit details -
fix: filled missing maintainer
Jaroslav Mikulík committedAug 27, 2026 Configuration menu - View commit details
-
Copy full SHA for e62d620 - Browse repository at this point
Copy the full SHA e62d620View commit details -
feat: DEB packages - Add modular CPU, Vulkan, CUDA, ROCm and SYCL bac…
…kend packages * split llama-cpp-python into core and dynamically loadable backend packages * add CPU backend with multiple optimized CPU variants and OpenBLAS * add Vulkan backend * add co-installable CUDA 12.4, 12.9 and 13.3 backends using update-alternatives * add ROCm 7.14 HIP backend with support for current AMD GPU architectures * add Intel oneAPI SYCL backend with oneMKL, oneDNN and Level Zero support * add transitional packages for legacy package names * add backend alternatives handling and private shared-library dependency resolution * add CUDA 12.9 compatibility handling for Debian Trixie/glibc 2.41 * isolate external CUDA, ROCm and oneAPI toolchains from Debian system libraries * update package dependencies, shlibs handling, cleanup rules and runtime RPATHs FIxed dependencies for CUDA 12.9
Jaroslav Mikulík committedAug 27, 2026 Configuration menu - View commit details
-
Copy full SHA for c6079e0 - Browse repository at this point
Copy the full SHA c6079e0View commit details -
Configuration menu - View commit details
-
Copy full SHA for 46244d0 - Browse repository at this point
Copy the full SHA 46244d0View commit details
Commits on Oct 10, 2026
-
Jaroslav Mikulík committedOct 10, 2026 Configuration menu - View commit details
-
Copy full SHA for c1bf8a4 - Browse repository at this point
Copy the full SHA c1bf8a4View commit details -
Jaroslav Mikulík committedOct 10, 2026 Configuration menu - View commit details
-
Copy full SHA for 9b63be4 - Browse repository at this point
Copy the full SHA 9b63be4View commit details -
Jaroslav Mikulík committedOct 10, 2026 Configuration menu - View commit details
-
Copy full SHA for 3e46029 - Browse repository at this point
Copy the full SHA 3e46029View commit details -
Jaroslav Mikulík committedOct 10, 2026 Configuration menu - View commit details
-
Copy full SHA for d8d1836 - Browse repository at this point
Copy the full SHA d8d1836View commit details -
Jaroslav Mikulík committedOct 10, 2026 Configuration menu - View commit details
-
Copy full SHA for 64ac075 - Browse repository at this point
Copy the full SHA 64ac075View commit details -
Configuration menu - View commit details
-
Copy full SHA for 4951955 - Browse repository at this point
Copy the full SHA 4951955View commit details -
Jaroslav Mikulík committedOct 10, 2026 Configuration menu - View commit details
-
Copy full SHA for cea8514 - Browse repository at this point
Copy the full SHA cea8514View commit details -
Merge remote-tracking branch 'refs/remotes/origin/main'
# Conflicts: # debian/changelog # debian/control # debian/patches/0001-debian-dynamic-backend-path.patch # debian/rules
Jaroslav Mikulík committedOct 10, 2026 Configuration menu - View commit details
-
Copy full SHA for acaa60a - Browse repository at this point
Copy the full SHA acaa60aView commit details -
fix: escape ROCm version checks in Debian rules
Make interprets an unescaped dollar followed by a closing parenthesis as a one-character variable reference, removing both characters before the recipe reaches the shell. This leaves an unterminated regex group and aborts the ROCm build before its compiler probe. Double both end-anchor dollars so grep receives the intended SDK 10.1 and HIP 7.16 expressions. Bump the Debian revision to 0:0.3.36-3. Validated actual make -n expansion and grep execution with matching versions and negative version-stream cases.
ByCzech committedOct 10, 2026 Configuration menu - View commit details
-
Copy full SHA for 46d384e - Browse repository at this point
Copy the full SHA 46d384eView commit details
Commits on Oct 11, 2026
-
fix: exclude unsupported gfx1250 kernels from the ROCm backend
ROCm Core SDK 10.1 ships gfx1250 device libraries, but the vendored GGML classifies all __GFX12__ targets as RDNA4. Its flash-attention kernels then use v_dot2_f32_f16 inline assembly and gfx12 WMMA builtins that the compiler rejects for gfx1250. Remove only gfx1250 from ROCM101_TARGETS, retaining the other 24 targets and gfx908/MI100 support. Keep flash attention enabled for the remaining targets and retain the pinned upstream sources. Bump the Debian revision to 0:0.3.36-4. Validated actual Make expansion, unchanged remaining target order, changelog parsing and mbox application. GPU compilation requires the external ROCm SDK and has not been run in this workspace.
ByCzech committedOct 11, 2026 Configuration menu - View commit details
-
Copy full SHA for aaf5ede - Browse repository at this point
Copy the full SHA aaf5edeView commit details -
fix: restore gfx1250 with compatible HIP kernel dispatch
Separate gfx1250 from RDNA4 in both HIP device macros and host feature selection. Keep native integer dot4 and wave32 tile kernels, use FP32 products for the FP16 dot2 fallback, and retain hipBLAS matrix hardware eligibility without selecting incompatible custom GGML WMMA kernels. Restore gfx1250 in the ROCm 10.1 target list. Carry the source changes as a Debian quilt patch so the upstream orig tarball stays unmodified. This follows the previous packaging patch that excluded gfx1250. Validation with the actual ROCm 10.1 clang compiler: - reproduce both original gfx1250 dot2 and WMMA compilation failures; - compile and link the complete ggml-hip backend for gfx1250; - compile the affected tile and MMA kernels for gfx1201 and gfx908; - compile a gfx1250 arithmetic/dispatch probe and run its host checks; - inspect generated gfx1250 ISA for native dot4 and FP32 products; - apply the Debian quilt series without fuzz and check the mbox with git am. No gfx1250 hardware inference or numerical validation was available. Native custom gfx1250 WMMA optimization remains separate work.
ByCzech committedOct 11, 2026 Configuration menu - View commit details
-
Copy full SHA for 4a22fe2 - Browse repository at this point
Copy the full SHA 4a22fe2View commit details -
fix: apply gfx1250 quilt patch across vendored revisions
The RDNA4 hunk included GCN APU predicates introduced after the user's vendored llama.cpp revision 4df29be. dpkg-source rejected that hunk even though the RDNA4 predicate being replaced is identical in both revisions. Match the unique RDNA4 predicate without unrelated GCN/CDNA context. Preserve the original compatibility implementation and the upstream GCN classification on each revision. Keep the original source tarball intact. Validation: - dpkg-source --before-build applies both quilt patches without fuzz on pristine headers from 4df29be and 0c1e570; - generated 0c1e570 headers are byte-identical to the previous tested fix; - ROCm 10.1 clang compiles the failing gfx1250 flash tile and MMA kernels on 4df29be, plus MMVQ and a Q4_0 MMQ kernel; - a host/device dispatch probe compiles and its host assertions pass; - the mbox applies with git am and reproduces the same tree. GPU execution and numerical validation remain untested.
ByCzech committedOct 11, 2026 Configuration menu - View commit details
-
Copy full SHA for 3e4b865 - Browse repository at this point
Copy the full SHA 3e4b865View commit details -
fix: synchronize vendored llama.cpp with the 0.3.36 bindings
The Python 0.3.36 module binds llama_adapter_lora_init_from_file_ptr at import time. The retained 4df29be submodule does not export that API; upstream 0.3.36 pins 0c1e570 instead. Update the gitlink to that exact revision so the Python binding, core runtime, and backend plugins can be rebuilt from one matching source tree. Reject source headers lacking the required LoRA FILE API before any compilation, including stale source tarballs. Import the staged package before the separate accelerator builds; leave the source directory and clear library overrides so the check uses the packaged runtime. Declare the Python runtime dependencies required by that import check. Validation: - verify the gitlink pinned by the upstream 0.3.36 tag; - confirm the missing C API is absent in 4df29be and present in 0c1e570; - run the actual expanded Make header check against both revisions; - run the actual expanded staged-import recipe against the previously built 0c1e570 core and the 0.3.36 Python modules; - check the mbox with git am, including the resulting gitlink and tree. Unapply quilt patches before updating the submodule working directory, then reapply them. Rebuild the core and all plugins together. An orig source archive must also contain the matching upstream submodule.
ByCzech committedOct 11, 2026 Configuration menu - View commit details
-
Copy full SHA for b9ce865 - Browse repository at this point
Copy the full SHA b9ce865View commit details -
fix: deleted obsolete file, update maintainer
Jaroslav Mikulík committedOct 11, 2026 Configuration menu - View commit details
-
Copy full SHA for 043f9d6 - Browse repository at this point
Copy the full SHA 043f9d6View commit details -
fix: keep upstream sources clean for quilt packaging
Jaroslav Mikulík committedOct 11, 2026 Configuration menu - View commit details
-
Copy full SHA for 4ca033b - Browse repository at this point
Copy the full SHA 4ca033bView commit details
This comparison is taking too long to generate.
Unfortunately it looks like we can’t render this comparison for you right now. It might be too big, or there might be something weird with your repository.
You can try running this command locally to see the comparison on your machine:
git diff main...main