Merge branch 'release/0.10' into yutji/rocm-6.0-dockerfile

.github/workflows/build-image.yml

@@ -29,18 +29,22 @@ jobs:
           dockerfile: cuda12.2
           tags: superbench/main:cuda12.2
           runner: [self-hosted, rocm-build]
+          build_args: "NUM_MAKE_JOBS=64"
         - name: cuda11.1.1
           dockerfile: cuda11.1.1
           tags: superbench/main:cuda11.1.1,superbench/superbench:latest
           runner: ubuntu-latest
+          build_args: "NUM_MAKE_JOBS=8"
         - name: rocm5.7
           dockerfile: rocm5.7.x
           tags: superbench/main:rocm5.7
           runner: [self-hosted, rocm-build]
+          build_args: "NUM_MAKE_JOBS=64"
         - name: rocm6.0
           dockerfile: rocm6.0.x
           tags: superbench/main:rocm6.0
           runner: [self-hosted, rocm-build]
+          build_args: "NUM_MAKE_JOBS=64"
     steps:
       - name: Checkout
         uses: actions/checkout@v2
@@ -80,7 +84,7 @@ jobs:
           fi
           DOCKERFILE=dockerfile/${{ matrix.dockerfile }}.dockerfile
 
-          BUILD_ARGS="NUM_MAKE_JOBS=8"
+          BUILD_ARGS=${{ matrix.build_args }}
           if [[ "${{ matrix.extra_args }}" ]]; then
             BUILD_ARGS="${BUILD_ARGS} ${{ matrix.extra_args }}"
           fi

dockerfile/rocm5.7.x.dockerfile

@@ -17,6 +17,7 @@ RUN apt-get update && \
     apt-get -q install -y --no-install-recommends  \
     autoconf \
     automake \
+    bc \
     build-essential \
     curl \
     dmidecode \
@@ -27,6 +28,7 @@ RUN apt-get update && \
     libaio-dev \
     libboost-program-options-dev \
     libcap2 \
+    libcurl4-openssl-dev \
     libnuma-dev \
     libpci-dev \
     libssl-dev \
@@ -38,6 +40,7 @@ RUN apt-get update && \
     openssh-client \
     openssh-server \
     pciutils \
+    python3-mpi4py \
     rsync \
     sudo \
     util-linux \
@@ -46,11 +49,11 @@ RUN apt-get update && \
     && \
     rm -rf /tmp/*
 
-ARG NUM_MAKE_JOBS=16
+ARG NUM_MAKE_JOBS=
 
 # Check if CMake is installed and its version
 RUN cmake_version=$(cmake --version 2>/dev/null | grep -oP "(?<=cmake version )(\d+\.\d+)" || echo "0.0") && \
-    required_version="3.26.4" && \
+    required_version="3.24.1" && \
     if [ "$(printf "%s\n" "$required_version" "$cmake_version" | sort -V | head -n 1)" != "$required_version" ]; then \
     echo "existing cmake version is ${cmake_version}" && \
     cd /tmp && \
@@ -100,21 +103,9 @@ RUN if ! command -v ofed_info >/dev/null 2>&1; then \
     rm -rf MLNX_OFED_LINUX-${OFED_VERSION}* ; \
     fi
 
-# Install UCX
-ENV UCX_VERSION=1.14.1
-RUN if [ -z "$(ls -A /opt/ucx)" ]; then \
-    echo "/opt/ucx is empty. Installing UCX..."; \
-    cd /tmp && \
-    git clone https://github.com/openucx/ucx.git -b v${UCX_VERSION} && \
-    cd ucx && \
-    ./autogen.sh && \
-    mkdir build && \
-    cd build && \
-    ../configure -prefix=$UCX_DIR --with-rocm=/opt/rocm --without-knem && \
-    make -j $(nproc) && make -j $(nproc) install && rm -rf /tmp/ucx-${UCX_VERSION} ; \
-    else \
-    echo "/opt/ucx is not empty. Skipping UCX installation."; \
-    fi
+# Add target file to help determine which device(s) to build for
+ENV ROCM_PATH=/opt/rocm
+RUN bash -c 'echo -e "gfx90a:xnack-\ngfx90a:xnac+\ngfx940\ngfx941\ngfx942\ngfx1030\ngfx1100\ngfx1101\ngfx1102\n" >> ${ROCM_PATH}/bin/target.lst'
 
 # Install OpenMPI
 ENV OPENMPI_VERSION=4.1.x
@@ -127,7 +118,7 @@ RUN [ -d /usr/local/bin/mpirun ] || { \
     ./autogen.pl && \
     mkdir build && \
     cd build && \
-    ../configure --prefix=/usr/local  --enable-orterun-prefix-by-default  --enable-mpirun-prefix-by-default  --enable-prte-prefix-by-default --enable-mca-no-build=btl-uct --with-ucx=/opt/ucx --with-rocm=/opt/rocm && \
+    ../configure --prefix=/usr/local  --enable-orterun-prefix-by-default  --enable-mpirun-prefix-by-default  --enable-prte-prefix-by-default --with-rocm=/opt/rocm && \
     make -j $(nproc) && \
     make -j $(nproc) install && \
     ldconfig && \
@@ -148,12 +139,18 @@ RUN cd /opt/ &&  \
     cd rccl && \
     mkdir build && \
     cd build && \
-    CXX=/opt/rocm/bin/hipcc cmake -DCMAKE_PREFIX_PATH=/opt/rocm/ .. && \
+    CXX=/opt/rocm/bin/hipcc cmake -DHIP_COMPILER=clang -DCMAKE_BUILD_TYPE=Release -DCMAKE_VERBOSE_MAKEFILE=1 \
+        -DCMAKE_PREFIX_PATH="${ROCM_PATH}/hsa;${ROCM_PATH}/hip;${ROCM_PATH}/share/rocm/cmake/;${ROCM_PATH}" \
+        .. && \
     make -j${NUM_MAKE_JOBS}
 
+# Install AMD SMI Python Library
+RUN cd /opt/rocm/share/amd_smi && \
+    python3 -m pip install --user .
+
 ENV PATH="/opt/superbench/bin:/usr/local/bin/:/opt/rocm/hip/bin/:/opt/rocm/bin/:${PATH}" \
     LD_PRELOAD="/opt/rccl/build/librccl.so:$LD_PRELOAD" \
-    LD_LIBRARY_PATH="/opt/ucx/lib:/usr/local/lib/:/opt/rocm/lib:${LD_LIBRARY_PATH}" \
+    LD_LIBRARY_PATH="/usr/local/lib/:/opt/rocm/lib:${LD_LIBRARY_PATH}" \
     SB_HOME=/opt/superbench \
     SB_MICRO_PATH=/opt/superbench \
     ANSIBLE_DEPRECATION_WARNINGS=FALSE \
@@ -163,13 +160,17 @@ RUN echo PATH="$PATH" > /etc/environment && \
     echo LD_LIBRARY_PATH="$LD_LIBRARY_PATH" >> /etc/environment && \
     echo SB_MICRO_PATH="$SB_MICRO_PATH" >> /etc/environment
 
+RUN apt install rocm-cmake -y && \
+    python3 -m pip install --upgrade pip wheel setuptools==65.7
+
 WORKDIR ${SB_HOME}
 
+ADD third_party third_party
+RUN make RCCL_HOME=/opt/rccl/build/ MPI_HOME=/usr/local ROCBLAS_BRANCH=release/rocm-rel-5.7.1.1 HIPBLASLT_BRANCH=release-staging/rocm-rel-5.7 ROCM_VER=rocm-5.5.0 -C third_party rocm -o cpu_hpl -o cpu_stream -o megatron_lm
+
 ADD . .
-RUN apt install rocm-cmake -y && \
-    python3 -m pip install --upgrade pip wheel setuptools==65.7 && \
-    python3 -m pip install .[amdworker]  && \
+#ENV USE_HIPBLASLT_DATATYPE=1
+ENV CXX=/opt/rocm/bin/hipcc
+RUN python3 -m pip install .[amdworker]  && \
+    make cppbuild  && \
     make postinstall
-RUN make cppbuild
-ADD third_party third_party
-RUN make RCCL_HOME=/opt/rccl/build/ ROCBLAS_BRANCH=release/rocm-rel-5.7.1.1 HIPBLASLT_BRANCH=release-staging/rocm-rel-5.7 ROCM_VER=rocm-5.5.0 -C third_party rocm -o cpu_hpl -o cpu_stream -o megatron_lm

docs/user-tutorial/benchmarks/micro-benchmarks.md

@@ -58,17 +58,18 @@ Large scale matmul operation using `torch.matmul` with one GPU.
 |--------------------------------|-----------|--------------------------------|
 | pytorch-matmul/nosharding_time | time (ms) | Time of pure matmul operation. |
 
-### `cublaslt-gemm`
+### `cublaslt-gemm` / `hipblaslt-gemm`
 
 #### Introduction
 
-Measure the GEMM performance of [`cublasLtMatmul`](https://docs.nvidia.com/cuda/cublas/#cublasltmatmul).
+Measure the GEMM performance of [`cublasLtMatmul`](https://docs.nvidia.com/cuda/cublas/#cublasltmatmul) or [`hipblasLt-bench`](https://github.com/ROCm/hipBLASLt/blob/develop/clients/benchmarks/README.md).
 
 #### Metrics
 
-| Name                                                     | Unit           | Description                     |
-|----------------------------------------------------------|----------------|---------------------------------|
-| cublaslt-gemm/${dtype}\_${batch}\_${m}\_${n}\_${k}_flops | FLOPS (TFLOPS) | TFLOPS of measured GEMM kernel. |
+| Name                                                      | Unit           | Description                     |
+|-----------------------------------------------------------|----------------|---------------------------------|
+| cublaslt-gemm/${dtype}\_${batch}\_${m}\_${n}\_${k}_flops  | FLOPS (TFLOPS) | TFLOPS of measured GEMM kernel. |
+| hipblaslt-gemm/${dtype}\_${batch}\_${m}\_${n}\_${k}_flops | FLOPS (TFLOPS) | TFLOPS of measured GEMM kernel. |
 
 ### `cublas-function`
 

superbench/benchmarks/micro_benchmarks/dist_inference_cpp/CMakeLists.txt

@@ -33,6 +33,11 @@ else()
         set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -O2 -DROCM_USE_FLOAT16=1")
         if(DEFINED ENV{USE_HIPBLASLT_DATATYPE})
             set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -DUSE_HIPBLASLT_DATATYPE=1")
+        elseif(DEFINED ENV{USE_HIP_DATATYPE})
+            set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -DUSE_HIP_DATATYPE=1")
+        endif()
+        if(DEFINED ENV{USE_HIPBLAS_COMPUTETYPE})
+            set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -DUSE_HIPBLAS_COMPUTETYPE=1")
         endif()
         target_link_libraries(dist_inference MPI::MPI_CXX rccl hipblaslt hip::device)
     else()

superbench/benchmarks/micro_benchmarks/dist_inference_cpp/dist_inference.cu

@@ -48,10 +48,18 @@ using cublasLtHalf = hipblasLtHalf;
 #if defined(USE_HIPBLASLT_DATATYPE)
 #define DIST_INF_HIP_DATATYPE_R_16F HIPBLASLT_R_16F
 #define DIST_INF_HIP_DATATYPE_R_32F HIPBLASLT_R_32F
+#elif defined(USE_HIP_DATATYPE)
+#define DIST_INF_HIP_DATATYPE_R_16F HIP_R_16F
+#define DIST_INF_HIP_DATATYPE_R_32F HIP_R_32F
 #else
 #define DIST_INF_HIP_DATATYPE_R_16F HIPBLAS_R_16F
 #define DIST_INF_HIP_DATATYPE_R_32F HIPBLAS_R_32F
 #endif
+#if defined(USE_HIPBLAS_COMPUTETYPE)
+#define DIST_INF_HIP_COMPUTETYPE_F32 HIPBLAS_COMPUTE_32F
+#else
+#define DIST_INF_HIP_COMPUTETYPE_F32 HIPBLASLT_COMPUTE_F32
+#endif
 #else
 #include <cublasLt.h>
 #include <cuda_fp16.h>
@@ -244,8 +252,10 @@ void TestModel(int64_t m, int64_t n, int64_t k, float alpha, float beta, int32_t
     CHECK_CUBLASLT_ERROR(hipblasLtMatrixLayoutCreate(&matF, DIST_INF_HIP_DATATYPE_R_16F, k, n, k));
     CHECK_CUBLASLT_ERROR(hipblasLtMatrixLayoutCreate(&matG, DIST_INF_HIP_DATATYPE_R_16F, k, n, k));
 
-    CHECK_CUBLASLT_ERROR(hipblasLtMatmulDescCreate(&matmul1, HIPBLASLT_COMPUTE_F32, DIST_INF_HIP_DATATYPE_R_32F));
-    CHECK_CUBLASLT_ERROR(hipblasLtMatmulDescCreate(&matmul2, HIPBLASLT_COMPUTE_F32, DIST_INF_HIP_DATATYPE_R_32F));
+    CHECK_CUBLASLT_ERROR(
+        hipblasLtMatmulDescCreate(&matmul1, DIST_INF_HIP_COMPUTETYPE_F32, DIST_INF_HIP_DATATYPE_R_32F));
+    CHECK_CUBLASLT_ERROR(
+        hipblasLtMatmulDescCreate(&matmul2, DIST_INF_HIP_COMPUTETYPE_F32, DIST_INF_HIP_DATATYPE_R_32F));
 
     hipblasOperation_t trans = HIPBLAS_OP_N;
     CHECK_CUBLASLT_ERROR(

superbench/benchmarks/micro_benchmarks/hipblaslt_function.py

@@ -103,7 +103,8 @@ def _process_raw_result(self, cmd_idx, raw_output):
                 raise ValueError('Invalid result')
 
             self._result.add_result(
-                f'{self._precision_in_commands[cmd_idx]}_{fields[3]}_{"_".join(fields[4:7])}_flops', float(fields[-2])
+                f'{self._precision_in_commands[cmd_idx]}_{fields[3]}_{"_".join(fields[4:7])}_flops',
+                float(fields[-2]) / 1000
             )
         except BaseException as e:
             self._result.set_return_code(ReturnCode.MICROBENCHMARK_RESULT_PARSING_FAILURE)

superbench/benchmarks/micro_benchmarks/rocm_common.cmake

@@ -45,8 +45,7 @@ message(STATUS "CMAKE HIP ARCHITECTURES: ${CMAKE_HIP_ARCHITECTURES}")
 
 if(EXISTS ${HIP_PATH})
     # Search for hip in common locations
-    list(APPEND CMAKE_PREFIX_PATH ${HIP_PATH} ${ROCM_PATH})
-    set(CMAKE_PREFIX_PATH /opt/rocm ROCM_PATH)
+    list(APPEND CMAKE_PREFIX_PATH ${HIP_PATH} ${ROCM_PATH} ${ROCM_PATH}/hsa ${ROCM_PATH}/hip ${ROCM_PATH}/share/rocm/cmake/)
     set(CMAKE_CXX_COMPILER "${HIP_PATH}/bin/hipcc")
     set(CMAKE_MODULE_PATH "${HIP_PATH}/cmake" ${CMAKE_MODULE_PATH})
     set(CMAKE_MODULE_PATH "${HIP_PATH}/lib/cmake/hip" ${CMAKE_MODULE_PATH})