ci : move more CPU jobs to self-hosted runners (#23715)

.github/workflows/build-self-hosted.yml

@@ -297,8 +297,8 @@ jobs:
           source ./openvino_toolkit/setupvars.sh
           GG_BUILD_OPENVINO=1 GGML_OPENVINO_DEVICE=GPU GG_BUILD_LOW_PERF=1 bash ./ci/run.sh ~/results/llama.cpp ~/mnt/llama.cpp
 
-  ggml-ci-arm64-cpu-low-perf:
+  ggml-ci-cpu-low-perf:
-    runs-on: [self-hosted, Linux, ARM64, CPU]
+    runs-on: [self-hosted, CPU]
 
     steps:
       - name: Clone
@@ -310,8 +310,8 @@ jobs:
         run: |
           LLAMA_ARG_THREADS=$(nproc) GG_BUILD_LOW_PERF=1 bash ./ci/run.sh ~/results/llama.cpp ~/mnt/llama.cpp
 
-  ggml-ci-arm64-cpu-high-perf:
+  ggml-ci-cpu-high-perf:
-    runs-on: [self-hosted, Linux, ARM64, CPU]
+    runs-on: [self-hosted, CPU]
 
     steps:
       - name: Clone

.github/workflows/build.yml

@@ -428,31 +428,33 @@ jobs:
 
 # TODO: simplify the following workflows using a matrix
 # TODO: run lighter CI on PRs and the full CI only on master (if needed)
-  ggml-ci-x64-cpu-low-perf:
-    runs-on: ubuntu-22.04
 
-    steps:
+# note: moved to build-self-hosted.yml - can remove from here when everything is stable
-      - name: Clone
+#  ggml-ci-x64-cpu-low-perf:
-        id: checkout
+#    runs-on: ubuntu-22.04
-        uses: actions/checkout@v6
+#
-
+#    steps:
-      - name: ccache
+#      - name: Clone
-        uses: ggml-org/ccache-action@v1.2.21
+#        id: checkout
-        with:
+#        uses: actions/checkout@v6
-          key: ggml-ci-x64-cpu-low-perf
+#
-          evict-old-files: 1d
+#      - name: ccache
-          save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
+#        uses: ggml-org/ccache-action@v1.2.21
-
+#        with:
-      - name: Dependencies
+#          key: ggml-ci-x64-cpu-low-perf
-        id: depends
+#          evict-old-files: 1d
-        run: |
+#          save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
-          sudo apt-get update
+#
-          sudo apt-get install build-essential
+#      - name: Dependencies
-
+#        id: depends
-      - name: Test
+#        run: |
-        id: ggml-ci
+#          sudo apt-get update
-        run: |
+#          sudo apt-get install build-essential
-          LLAMA_ARG_THREADS=$(nproc) GG_BUILD_LOW_PERF=1 bash ./ci/run.sh ./tmp/results ./tmp/mnt
+#
+#      - name: Test
+#        id: ggml-ci
+#        run: |
+#          LLAMA_ARG_THREADS=$(nproc) GG_BUILD_LOW_PERF=1 bash ./ci/run.sh ./tmp/results ./tmp/mnt
 
 # note: moved to build-self-hosted.yml - can remove from here when everything is stable
 #  ggml-ci-arm64-cpu-low-perf:
@@ -481,31 +483,32 @@ jobs:
 #        run: |
 #          LLAMA_ARG_THREADS=$(nproc) GG_BUILD_LOW_PERF=1 bash ./ci/run.sh ./tmp/results ./tmp/mnt
 
-  ggml-ci-x64-cpu-high-perf:
+# note: moved to build-self-hosted.yml - can remove from here when everything is stable
-    runs-on: ubuntu-22.04
+#  ggml-ci-x64-cpu-high-perf:
-
+#    runs-on: ubuntu-22.04
-    steps:
+#
-      - name: Clone
+#    steps:
-        id: checkout
+#      - name: Clone
-        uses: actions/checkout@v6
+#        id: checkout
-
+#        uses: actions/checkout@v6
-      - name: ccache
+#
-        uses: ggml-org/ccache-action@v1.2.21
+#      - name: ccache
-        with:
+#        uses: ggml-org/ccache-action@v1.2.21
-          key: ggml-ci-x64-cpu-high-perf
+#        with:
-          evict-old-files: 1d
+#          key: ggml-ci-x64-cpu-high-perf
-          save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
+#          evict-old-files: 1d
-
+#          save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
-      - name: Dependencies
+#
-        id: depends
+#      - name: Dependencies
-        run: |
+#        id: depends
-          sudo apt-get update
+#        run: |
-          sudo apt-get install build-essential
+#          sudo apt-get update
-
+#          sudo apt-get install build-essential
-      - name: Test
+#
-        id: ggml-ci
+#      - name: Test
-        run: |
+#        id: ggml-ci
-          LLAMA_ARG_THREADS=$(nproc) GG_BUILD_HIGH_PERF=1 bash ./ci/run.sh ./tmp/results ./tmp/mnt
+#        run: |
+#          LLAMA_ARG_THREADS=$(nproc) GG_BUILD_HIGH_PERF=1 bash ./ci/run.sh ./tmp/results ./tmp/mnt
 
 # note: moved to build-self-hosted.yml - can remove from here when everything is stable
 #  ggml-ci-arm64-cpu-high-perf:

ci/run.sh

@@ -66,6 +66,8 @@ fi
 
 if [ ! -z ${GG_BUILD_METAL} ]; then
     CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_METAL=ON"
+else
+    CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_METAL=OFF"
 fi
 
 if [ ! -z ${GG_BUILD_CUDA} ]; then
@@ -114,10 +116,7 @@ fi
 if [ ! -z ${GG_BUILD_VULKAN} ]; then
     CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_VULKAN=1"
 
-    # if on Mac, disable METAL
     if [[ "$OSTYPE" == "darwin"* ]]; then
-        CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_METAL=OFF -DGGML_BLAS=OFF"
-
         MACOS_RUNNER_CUSTOM_VULKAN_CMAKE_LOCATION="/usr/local/lib/cmake/vulkan"
         MACOS_RUNNER_CUSTOM_SPIRV_HEADERS_LOCATION="${MACOS_RUNNER_CUSTOM_VULKAN_CMAKE_LOCATION}/SPIRV-Headers/SPIRV-HeadersConfig.cmake"
         if [[ -f "${MACOS_RUNNER_CUSTOM_SPIRV_HEADERS_LOCATION}" || -h "${MACOS_RUNNER_CUSTOM_SPIRV_HEADERS_LOCATION}" ]]; then
@@ -133,7 +132,7 @@ if [ ! -z ${GG_BUILD_VULKAN} ]; then
 fi
 
 if [ ! -z ${GG_BUILD_WEBGPU} ]; then
-    CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_WEBGPU=1 -DGGML_METAL=OFF -DGGML_BLAS=OFF"
+    CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_WEBGPU=1"
 
     if [ ! -z "${GG_BUILD_WEBGPU_DAWN_PREFIX}" ]; then
         if [ -z "${CMAKE_PREFIX_PATH}" ]; then
@@ -167,6 +166,8 @@ fi
 
 if [ ! -z ${GG_BUILD_BLAS} ]; then
     CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_BLAS=ON -DGGML_BLAS_VENDOR=${GG_BUILD_BLAS_VENDOR:-OpenBLAS}"
+else
+    CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_BLAS=OFF"
 fi
 
 if [ ! -z ${GG_BUILD_OPENVINO} ]; then