matt/ollama37
1
0
Fork 0
mirror of https://github.com/dogkeeper886/ollama37.git synced 2025-12-15 02:07:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

Optimize container images for startup (#6547)

Browse Source 
* Optimize container images for startup

This change adjusts how to handle runner payloads to support
container builds where we keep them extracted in the filesystem.
This makes it easier to optimize the cpu/cuda vs cpu/rocm images for
size, and should result in faster startup times for container images.

* Refactor payload logic and add buildx support for faster builds

* Move payloads around

* Review comments

* Converge to buildx based helper scripts

* Use docker buildx action for release
This commit is contained in:
Daniel Hiltgen
2024-09-12 12:10:30 -07:00
committed by GitHub
parent fef257c5c5
commit cd5c8f6471
32 changed files with 861 additions and 689 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
llm/generate/gen_darwin.sh
View File

@@ -39,7 +39,8 @@ case "${GOARCH}" in
#
init_vars
CMAKE_DEFS="${COMMON_CPU_DEFS} -DGGML_ACCELERATE=off -DGGML_BLAS=off -DGGML_AVX=off -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off ${CMAKE_DEFS}"
BUILD_DIR="../build/darwin/${ARCH}/cpu"
RUNNER=cpu
BUILD_DIR="../build/darwin/${ARCH}/${RUNNER}"
echo "Building LCD CPU"
build
sign ${BUILD_DIR}/bin/ollama_llama_server
@@ -51,7 +52,8 @@ case "${GOARCH}" in
#
init_vars
CMAKE_DEFS="${COMMON_CPU_DEFS} -DGGML_ACCELERATE=off -DGGML_BLAS=off -DGGML_AVX=on -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off ${CMAKE_DEFS}"
BUILD_DIR="../build/darwin/${ARCH}/cpu_avx"
RUNNER=cpu_avx
BUILD_DIR="../build/darwin/${ARCH}/${RUNNER}"
echo "Building AVX CPU"
build
sign ${BUILD_DIR}/bin/ollama_llama_server
@@ -63,7 +65,8 @@ case "${GOARCH}" in
#
init_vars
CMAKE_DEFS="${COMMON_CPU_DEFS} -DGGML_ACCELERATE=on -DGGML_BLAS=off -DGGML_AVX=on -DGGML_AVX2=on -DGGML_AVX512=off -DGGML_FMA=on -DGGML_F16C=on ${CMAKE_DEFS}"
BUILD_DIR="../build/darwin/${ARCH}/cpu_avx2"
RUNNER=cpu_avx2
BUILD_DIR="../build/darwin/${ARCH}/${RUNNER}"
echo "Building AVX2 CPU"
EXTRA_LIBS="${EXTRA_LIBS} -framework Accelerate -framework Foundation"
build
@@ -84,7 +87,8 @@ case "${GOARCH}" in
if [ -z "$OLLAMA_SKIP_METAL_GENERATE" ]; then
init_vars
CMAKE_DEFS="${COMMON_DARWIN_DEFS} -DCMAKE_SYSTEM_PROCESSOR=${ARCH} -DCMAKE_OSX_ARCHITECTURES=${ARCH} ${CMAKE_DEFS}"
BUILD_DIR="../build/darwin/${ARCH}/metal"
RUNNER="metal"
BUILD_DIR="../build/darwin/${ARCH}/${RUNNER}"
EXTRA_LIBS="${EXTRA_LIBS} -framework Accelerate -framework Foundation -framework Metal -framework MetalKit -framework MetalPerformanceShaders"
build
sign ${BUILD_DIR}/bin/ollama_llama_server