mirror of
https://github.com/dogkeeper886/ollama37.git
synced 2025-12-09 23:37:06 +00:00
ef14fb5b26
This commit represents a complete rework after pulling the latest changes from official ollama/ollama repository and re-applying Tesla K80 compatibility patches. ## Key Changes ### CUDA Compute Capability 3.7 Support (Tesla K80) - Added sm_37 (compute 3.7) to CMAKE_CUDA_ARCHITECTURES in CMakeLists.txt - Updated CMakePresets.json to include compute 3.7 in "CUDA 11" preset - Using 37-virtual (PTX with JIT compilation) for maximum compatibility ### Legacy Toolchain Compatibility - **NVIDIA Driver**: 470.256.02 (last version supporting Kepler/K80) - **CUDA Version**: 11.4.4 (last CUDA 11.x supporting compute 3.7) - **GCC Version**: 10.5.0 (required by CUDA 11.4 host_config.h) ### CPU Architecture Trade-offs Due to GCC 10.5 limitation, sacrificed newer CPU optimizations: - Alderlake CPU variant enabled WITHOUT AVX_VNNI (requires GCC 11+) - Still supports: SSE4.2, AVX, F16C, AVX2, BMI2, FMA - Performance impact: ~3-7% on newer CPUs (acceptable for K80 compatibility) ### Build System Updates - Modified ml/backend/ggml/ggml/src/ggml-cuda/CMakeLists.txt for compute 3.7 - Added -Wno-deprecated-gpu-targets flag to suppress warnings - Updated ml/backend/ggml/ggml/src/CMakeLists.txt for Alderlake without AVX_VNNI ### Upstream Sync Merged latest llama.cpp changes including: - Enhanced KV cache management with ISWA and hybrid memory support - Improved multi-modal support (mtmd framework) - New model architectures (Gemma3, Llama4, Qwen3, etc.) - GPU backend improvements for CUDA, Metal, and ROCm - Updated quantization support and GGUF format handling ### Documentation - Updated CLAUDE.md with comprehensive build instructions - Documented toolchain constraints and CPU architecture trade-offs - Removed outdated CI/CD workflows (tesla-k80-*.yml) - Cleaned up temporary development artifacts ## Rationale This fork maintains Tesla K80 GPU support (compute 3.7) which was dropped in official Ollama due to legacy driver/CUDA requirements. The toolchain constraint creates a deadlock: - K80 → Driver 470 → CUDA 11.4 → GCC 10 → No AVX_VNNI We accept the loss of cutting-edge CPU optimizations to enable running modern LLMs on legacy but still capable Tesla K80 hardware (12GB VRAM per GPU). 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
24 lines
1.1 KiB
Bash
24 lines
1.1 KiB
Bash
# Common environment setup across build*.sh scripts
|
|
|
|
export VERSION=${VERSION:-$(git describe --tags --first-parent --abbrev=7 --long --dirty --always | sed -e "s/^v//g")}
|
|
export GOFLAGS="'-ldflags=-w -s \"-X=github.com/ollama/ollama/version.Version=$VERSION\" \"-X=github.com/ollama/ollama/server.mode=release\"'"
|
|
# TODO - consider `docker buildx ls --format=json` to autodiscover platform capability
|
|
PLATFORM=${PLATFORM:-"linux/arm64,linux/amd64"}
|
|
DOCKER_ORG=${DOCKER_ORG:-"ollama"}
|
|
FINAL_IMAGE_REPO=${FINAL_IMAGE_REPO:-"${DOCKER_ORG}/ollama"}
|
|
OLLAMA_COMMON_BUILD_ARGS="--build-arg=VERSION \
|
|
--build-arg=GOFLAGS \
|
|
--build-arg=OLLAMA_CUSTOM_CPU_DEFS \
|
|
--build-arg=OLLAMA_SKIP_CUDA_GENERATE \
|
|
--build-arg=OLLAMA_SKIP_CUDA_12_GENERATE \
|
|
--build-arg=CUDA_V12_ARCHITECTURES \
|
|
--build-arg=OLLAMA_SKIP_ROCM_GENERATE \
|
|
--build-arg=OLLAMA_FAST_BUILD \
|
|
--build-arg=CUSTOM_CPU_FLAGS \
|
|
--build-arg=GPU_RUNNER_CPU_FLAGS \
|
|
--build-arg=PARALLEL \
|
|
--build-arg=AMDGPU_TARGETS"
|
|
|
|
echo "Building Ollama"
|
|
echo "VERSION=$VERSION"
|
|
echo "PLATFORM=$PLATFORM" |