Merge remote-tracking branch 'ggml-org/master' into allozaur/svelte-webui

Author: allozaur

.devops/cloud-v-pipeline

@@ -4,19 +4,15 @@ FROM ubuntu:$UBUNTU_VERSION AS build
 
 ARG TARGETARCH
 
-ARG GGML_CPU_ARM_ARCH=armv8-a
-
 RUN apt-get update && \
     apt-get install -y build-essential git cmake libcurl4-openssl-dev
 
 WORKDIR /app
 
 COPY . .
 
-RUN if [ "$TARGETARCH" = "amd64" ]; then \
+RUN if [ "$TARGETARCH" = "amd64" ] || [ "$TARGETARCH" = "arm64" ]; then \
         cmake -S . -B build -DCMAKE_BUILD_TYPE=Release -DGGML_NATIVE=OFF -DLLAMA_BUILD_TESTS=OFF -DGGML_BACKEND_DL=ON -DGGML_CPU_ALL_VARIANTS=ON; \
-    elif [ "$TARGETARCH" = "arm64" ]; then \
-        cmake -S . -B build -DCMAKE_BUILD_TYPE=Release -DGGML_NATIVE=OFF -DLLAMA_BUILD_TESTS=OFF -DGGML_CPU_ARM_ARCH=${GGML_CPU_ARM_ARCH}; \
     else \
         echo "Unsupported architecture"; \
         exit 1; \

.devops/cpu.Dockerfile

@@ -61,7 +61,7 @@ RUN apt-get update \
     python3 \
     python3-pip \
     && pip install --upgrade pip setuptools wheel \
-    && pip install -r requirements.txt \
+    && pip install --break-system-packages -r requirements.txt \
     && apt autoremove -y \
     && apt clean -y \
     && rm -rf /tmp/* /var/tmp/* \

.devops/cuda.Dockerfile

@@ -40,7 +40,7 @@ body:
     attributes:
         label: GGML backends
         description: Which GGML backends do you know to be affected?
-        options: [AMX, BLAS, CPU, CUDA, HIP, Metal, Musa, RPC, SYCL, Vulkan, OpenCL]
+        options: [AMX, BLAS, CPU, CUDA, HIP, Metal, Musa, RPC, SYCL, Vulkan, OpenCL, zDNN]
         multiple: true
     validations:
       required: true

.github/ISSUE_TEMPLATE/010-bug-compilation.yml

@@ -42,7 +42,7 @@ body:
     attributes:
         label: GGML backends
         description: Which GGML backends do you know to be affected?
-        options: [AMX, BLAS, CPU, CUDA, HIP, Metal, Musa, RPC, SYCL, Vulkan, OpenCL]
+        options: [AMX, BLAS, CPU, CUDA, HIP, Metal, Musa, RPC, SYCL, Vulkan, OpenCL, zDNN]
         multiple: true
     validations:
       required: true

.github/ISSUE_TEMPLATE/011-bug-results.yml

@@ -22,6 +22,11 @@ Vulkan:
         - any-glob-to-any-file:
             - ggml/include/ggml-vulkan.h
             - ggml/src/ggml-vulkan/**
+IBM zDNN:
+    - changed-files:
+        - any-glob-to-any-file:
+            - ggml/include/ggml-zdnn.h
+            - ggml/src/ggml-zdnn/**
 documentation:
     - changed-files:
         - any-glob-to-any-file:

.github/labeler.yml

@@ -0,0 +1,43 @@
+name: Build on RISCV Linux Machine by Cloud-V
+on:
+  workflow_dispatch:
+  workflow_call:
+
+jobs:
+  bianbu-riscv64-native: # Bianbu 2.2
+    runs-on: self-hosted
+
+    steps:
+      - name: Install prerequisites
+        run: |
+          sudo apt-get update || true
+          sudo apt-get install -y libatomic1
+      - uses: actions/checkout@v4
+      - name: Setup Riscv
+        run: |
+          sudo apt-get update || true
+          sudo apt-get install -y --no-install-recommends \
+                  build-essential \
+                  gcc-14-riscv64-linux-gnu \
+                  g++-14-riscv64-linux-gnu \
+                  cmake
+
+      - name: Build
+        run: |
+          cmake -B build -DLLAMA_CURL=OFF \
+                         -DCMAKE_BUILD_TYPE=Release \
+                         -DGGML_OPENMP=OFF \
+                         -DLLAMA_BUILD_EXAMPLES=ON \
+                         -DLLAMA_BUILD_TOOLS=ON \
+                         -DLLAMA_BUILD_TESTS=OFF \
+                         -DCMAKE_SYSTEM_NAME=Linux \
+                         -DCMAKE_SYSTEM_PROCESSOR=riscv64 \
+                         -DCMAKE_C_COMPILER=riscv64-linux-gnu-gcc-14 \
+                         -DCMAKE_CXX_COMPILER=riscv64-linux-gnu-g++-14 \
+                         -DCMAKE_POSITION_INDEPENDENT_CODE=ON \
+                         -DCMAKE_FIND_ROOT_PATH=/usr/lib/riscv64-linux-gnu \
+                         -DCMAKE_FIND_ROOT_PATH_MODE_PROGRAM=NEVER \
+                         -DCMAKE_FIND_ROOT_PATH_MODE_LIBRARY=ONLY \
+                         -DCMAKE_FIND_ROOT_PATH_MODE_INCLUDE=BOTH
+
+          cmake --build build --config Release -j $(nproc)

.github/workflows/build-riscv-native.yml

@@ -64,7 +64,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: macOS-latest-cmake-arm64
           evict-old-files: 1d
@@ -104,7 +104,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: macOS-latest-cmake-x64
           evict-old-files: 1d
@@ -144,7 +144,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: macOS-latest-cmake-arm64-webgpu
           evict-old-files: 1d
@@ -199,7 +199,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ubuntu-cpu-cmake
           evict-old-files: 1d
@@ -251,7 +251,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ubuntu-latest-cmake-sanitizer-${{ matrix.sanitizer }}
           evict-old-files: 1d
@@ -330,7 +330,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ubuntu-latest-cmake-rpc
           evict-old-files: 1d
@@ -363,7 +363,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ubuntu-22-cmake-vulkan
           evict-old-files: 1d
@@ -400,7 +400,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ubuntu-22-cmake-webgpu
           evict-old-files: 1d
@@ -443,7 +443,7 @@ jobs:
 
   ubuntu-22-cmake-hip:
     runs-on: ubuntu-22.04
-    container: rocm/dev-ubuntu-22.04:6.0.2
+    container: rocm/dev-ubuntu-22.04:6.1.2
 
     steps:
       - name: Clone
@@ -457,7 +457,7 @@ jobs:
           sudo apt-get install -y build-essential git cmake rocblas-dev hipblas-dev libcurl4-openssl-dev
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ubuntu-22-cmake-hip
           evict-old-files: 1d
@@ -471,16 +471,6 @@ jobs:
             -DGGML_HIP=ON
           cmake --build build --config Release -j $(nproc)
 
-      - name: Build with legacy HIP support
-        id: cmake_build_legacy_hip
-        run: |
-          cmake -B build2 -S . \
-            -DCMAKE_C_COMPILER=hipcc \
-            -DCMAKE_CXX_COMPILER=hipcc \
-            -DGGML_HIP_ROCWMMA_FATTN=ON \
-            -DGGML_HIP=ON
-          cmake --build build2 --config Release -j $(nproc)
-
   ubuntu-22-cmake-musa:
     runs-on: ubuntu-22.04
     container: mthreads/musa:rc4.2.0-devel-ubuntu22.04-amd64
@@ -497,7 +487,7 @@ jobs:
           apt-get install -y build-essential git cmake libcurl4-openssl-dev
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ubuntu-22-cmake-musa
           evict-old-files: 1d
@@ -542,7 +532,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ubuntu-22-cmake-sycl
           evict-old-files: 1d
@@ -590,7 +580,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ubuntu-22-cmake-sycl-fp16
           evict-old-files: 1d
@@ -621,7 +611,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: macOS-latest-cmake-ios
           evict-old-files: 1d
@@ -658,7 +648,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: macOS-latest-cmake-tvos
           evict-old-files: 1d
@@ -730,7 +720,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: macOS-latest-swift
           evict-old-files: 1d
@@ -776,7 +766,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: windows-msys2
           variant: ccache
@@ -844,7 +834,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: windows-latest-cmake-${{ matrix.build }}
           variant: ccache
@@ -958,7 +948,7 @@ jobs:
               apt install -y cmake build-essential ninja-build libgomp1 git libcurl4-openssl-dev
 
         - name: ccache
-          uses: hendrikmuhs/[email protected]
+          uses: ggml-org/[email protected]
           with:
             key: ubuntu-latest-cmake-cuda
             evict-old-files: 1d
@@ -987,7 +977,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: Install ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: windows-cuda-${{ matrix.cuda }}
           variant: ccache
@@ -1043,7 +1033,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: windows-latest-cmake-sycl
           variant: ccache
@@ -1080,7 +1070,8 @@ jobs:
           write-host "Downloading AMD HIP SDK Installer"
           Invoke-WebRequest -Uri "https://download.amd.com/developer/eula/rocm-hub/AMD-Software-PRO-Edition-24.Q3-WinSvr2022-For-HIP.exe" -OutFile "${env:RUNNER_TEMP}\rocm-install.exe"
           write-host "Installing AMD HIP SDK"
-          Start-Process "${env:RUNNER_TEMP}\rocm-install.exe" -ArgumentList '-install' -NoNewWindow -Wait
+          $proc = Start-Process "${env:RUNNER_TEMP}\rocm-install.exe" -ArgumentList '-install' -NoNewWindow -PassThru
+          $proc.WaitForExit(600000)
           write-host "Completed AMD HIP SDK installation"
 
       - name: Verify ROCm
@@ -1089,7 +1080,7 @@ jobs:
           & 'C:\Program Files\AMD\ROCm\*\bin\clang.exe' --version
 
       - name: Install ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: ${{ github.job }}
           evict-old-files: 1d
@@ -1123,6 +1114,11 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@v4
 
+      - name: Setup Xcode
+        uses: maxim-lobanov/setup-xcode@v1
+        with:
+          xcode-version: latest-stable
+
       - name: Build
         id: cmake_build
         run: |
@@ -1156,7 +1152,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: ccache
-        uses: hendrikmuhs/[email protected]
+        uses: ggml-org/[email protected]
         with:
           key: android-build
           evict-old-files: 1d