The uncompressed file size of a CUDA build is now 199MB (ggml.dll on Windows, on build b3240), whereas the build was just 80MB on a release from a few days ago (like b3166 for example). I think there might be another issue other than duplicate linking that might cause this (given...
cmake .. -DGGML_CUDA=ON \ > -DBUILD_SHARED_LIBS=OFF \ > -DLLAMA_CURL=ON \ > -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-12.2 -- The C compiler identification is GNU 10.5.0 -- The CXX compiler identification is GNU 9.4.0 -- Detecting C compiler ABI info -- Detecting C compiler...
windows-msys2 (CLANG64, clang-x86_64, Release) windows-latest-cmake (noavx-x64, -DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON -DGGML_A... windows-latest-cmake (avx2-x64, -DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON) windows-latest-cmake (avx-x64, ...
opened #12098 JohannesGaessler:cuda-fix-v100-force-mmq Status Queued Total duration – Artifacts – build.yml on: pull_request Matrix: ubuntu-cpu-cmake 2 jobs completed Show all jobs Matrix: windows-2019-cmake-cuda 0/2 jobs completed Show all jobs Matrix: windows-latest-cmake-hip-re...
ubuntu-latest-cmake-cuda windows-latest-cmake-hip ios-xcode-build android-build Matrix: macOS-latest-swift 1/3 job completed Show all jobs Matrix: openEuler-latest-cmake-cann 0/1 jobs completed Show all jobs Matrix: ubuntu-latest-cmake-sanitizer 0/3 jobs completed Show all jobs ...
Git commit git rev-parse HEAD d2fe216 Operating systems Linux GGML backends CUDA Problem description & steps to reproduce device: A800 cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON -DLLAMA_CURL=ON -- ccache fo...
time=2025-02-21T23:08:26.976+08:00 level=DEBUG source=amd_windows.go:34 msg="unable to load amdhip64_6.dll, please make sure to upgrade to the latest amd driver: The file cannot be accessed by the system." releasing cuda driver library releasing nvml library time=2025-02-21T23:08:...
Name and Version $ .\llama-cli.exe --version ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 ROCm devices: Device 0: AMD Radeon RX 7900 XTX, compute capability 11.0, VMM: no vers...
@@ -23,15 +23,16 @@ RUN if [ "${CUDA_DOCKER_ARCH}" != "default" ]; then \ export CMAKE_ARGS="-DCMAKE_CUDA_ARCHITECTURES=${CUDA_DOCKER_ARCH}"; \ fi && \ cmake -B build -DGGML_CUDA=ON -DLLAMA_CURL=ON ${CMAKE_ARGS} -DCMAKE_EXE_LINKER_FLAGS=-Wl,--allow-shlib-undef...
cmake -B build \ -DGGML_CUDA=ON \ -DGGML_VULKAN=1 \ -DCMAKE_INSTALL_PREFIX='/usr/local' \ -DGGML_ALL_WARNINGS=OFF \ -DGGML_ALL_WARNINGS_3RD_PARTY=OFF \ -DBUILD_SHARED_LIBS=ON \ -DGGML_STATIC=OFF \ -DGGML_LTO=ON \ -DGGML_RPC=ON \ -DLLAMA_CURL=ON \ -DGGML_CUDA=ON...