WIP

Co-authored-by: srawat <120587655+SwRaw@users.noreply.github.com>

.azuredevops/ci-builds/aomp.yml

@@ -0,0 +1,29 @@
+variables:
+- group: common
+- template: /.azuredevops/variables-global.yml
+
+resources:
+  repositories:
+  - repository: release_repo
+    type: github
+    endpoint: ROCm
+    name: ROCm/aomp
+    ref: aomp-dev
+  - repository: llvm-project_repo
+    type: github
+    endpoint: ROCm
+    name: ROCm/llvm-project
+    ref: amd-staging
+  pipelines:
+  - pipeline: rocr-runtime_pipeline
+    source: \ROCR-Runtime
+    trigger: true
+# this job will only be triggered after successful build sequence of llvm-project and ROCR-Runtime
+
+trigger: none
+pr: none
+
+jobs:
+  - template: ${{ variables.CI_COMPONENT_PATH }}/aomp.yml
+    parameters:
+      checkoutRepo: release_repo

.azuredevops/components/AMDMIGraphX.yml

@@ -84,8 +84,8 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/clang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/clang
         -DCMAKE_BUILD_TYPE=Release
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm/llvm;$(Agent.BuildDirectory)/rocm

.azuredevops/components/MIOpen.yml

@@ -9,20 +9,29 @@ parameters:
   type: object
   default:
     - software-properties-common
+    - python3-pip
     - cmake
     - ninja-build
-    - composablekernel-dev
-    - half
-    - rocrand
-    - rocblas
     - libsqlite3-dev
     - libbz2-dev
     - nlohmann-json3-dev
     - libgtest-dev
+    - libdrm-dev
 - name: rocmDependencies
   type: object
   default:
     - rocMLIR
+    - rocRAND
+    - rocBLAS
+    - half
+    - composable_kernel
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - rocprofiler-register
+    - clr
+    - rocminfo
+    - roctracer
 
 jobs:
 - job: MIOpen
@@ -30,8 +39,6 @@ jobs:
   - group: common
   - template: /.azuredevops/variables-global.yml
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -70,9 +77,8 @@ jobs:
     parameters:
       extraBuildFlags: >-
         -DMIOPEN_BACKEND=HIP
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
-        -DCMAKE_PREFIX_PATH="$(Agent.BuildDirectory)/rocm"
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DMIOPEN_ENABLE_AI_KERNEL_TUNING=OFF
         -DMIOPEN_ENABLE_AI_IMMED_MODE_FALLBACK=OFF
         -DCMAKE_BUILD_TYPE=Release

.azuredevops/components/MIVisionX.yml

@@ -14,10 +14,6 @@ parameters:
     - wget
     - unzip
     - pkg-config
-    - half
-    - rocblas-dev
-    - miopen-hip-dev
-    - migraphx-dev
     - protobuf-compiler
     - libprotoc-dev
     - ffmpeg
@@ -25,10 +21,6 @@ parameters:
     - libavformat-dev
     - libavutil-dev
     - libswscale-dev
-    - rpp
-    - rpp-dev
-    - rocdecode
-    - rocdecode-dev
     - build-essential
     - libgtk2.0-dev
     - libavcodec-dev
@@ -41,6 +33,7 @@ parameters:
     - libtiff-dev
     - libdc1394-dev
     - libgmp-dev
+    - libopencv-dev
 - name: pipModules
   type: object
   default:
@@ -50,6 +43,21 @@ parameters:
     - google==3.0.0
     - protobuf==3.12.4
     - onnx==1.12.0
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocprofiler-register
+    - half
+    - rocBLAS
+    - MIOpen
+    - AMDMIGraphX
+    - rpp
+    - rocDecode
 
 jobs:
 - job: MIVisionX
@@ -58,8 +66,6 @@ jobs:
   - template: /.azuredevops/variables-global.yml
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -71,11 +77,23 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
         -DCMAKE_BUILD_TYPE=Release
-        -DROCM_PATH=/opt/rocm
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
         -DROCM_DEP_ROCMCORE=ON
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/ROCdbgapi.yml

@@ -10,6 +10,13 @@ parameters:
   default:
     - cmake
     - ninja-build
+- name: rocmDependencies
+  type: object
+  default:
+    - clr
+    - llvm-project
+    - rocminfo
+    - ROCR-Runtime
 
 jobs:
 - job: ROCdbgapi
@@ -18,8 +25,6 @@ jobs:
   - template: /.azuredevops/variables-global.yml
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -30,9 +35,22 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
         -DCMAKE_BUILD_TYPE=Release
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/ROCmValidationSuite.yml

@@ -10,22 +10,33 @@ parameters:
   default:
     - cmake
     - ninja-build
-    - rocblas
     - libyaml-cpp-dev
     - libpci-dev
     - libpci3
     - googletest
     - git
+- name: rocmDependencies
+  type: object
+  default:
+    - clr
+    - llvm-project
+    - rocBLAS
+    - rocm-cmake
+    - rocm_smi_lib
+    - rocminfo
+    - rocprofiler-register
+    - ROCR-Runtime
+    - ROCT-Thunk-Interface
 
 jobs:
 - job: ROCmValidationSuite
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Build.BinariesDirectory)/rocm
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -36,11 +47,23 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DROCM_PATH=/opt/rocm
-        -DCMAKE_PREFIX_PATH=/opt/rocm
-        -DCPACK_PACKAGING_INSTALL_PREFIX='$(Build.BinariesDirectory)'
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
+        -DCPACK_PACKAGING_INSTALL_PREFIX=$(Build.BinariesDirectory)
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/aomp.yml

@@ -5,6 +5,9 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+- name: offloadEnabled
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -94,18 +97,18 @@ jobs:
       cmakeBuildDir: $(Build.SourcesDirectory)/llvm-project/openmp/build
       installDir: $(Build.BinariesDirectory)/llvm
 # offload does not exist for recent releases, so use CI conditional
-  - ${{ if eq(parameters.checkoutRef, '') }}:
+  - ${{ if eq(parameters.offloadEnabled, true) }}:
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         componentName: llvm-offload
         extraBuildFlags: >-
           -DOPENMP_ENABLE_LIBOMPTARGET=1
-          -DOPENMP_TEST_C_COMPILER==$(Agent.BuildDirectory)/rocm/llvm/bin/clang
-          -DOPENMP_TEST_CXX_COMPILER==$(Agent.BuildDirectory)/rocm/llvm/bin/clang++
-          -DCMAKE_C_COMPILER==$(Agent.BuildDirectory)/rocm/llvm/bin/clang
-          -DCMAKE_CXX_COMPILER==$(Agent.BuildDirectory)/rocm/llvm/bin/clang++
+          -DOPENMP_TEST_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/clang
+          -DOPENMP_TEST_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/clang++
+          -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/clang
+          -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/clang++
           -DLIBOMPTARGET_AMDGCN_GFXLIST=gfx700;gfx701;gfx801;gfx803;gfx900;gfx902;gfx906;gfx908;gfx90a;gfx90c;gfx940;gfx941;gfx942;gfx1030;gfx1031;gfx1035;gfx1036;gfx1100;gfx1101;gfx1102;gfx1103
-          -DLLVM_DIR==$(Agent.BuildDirectory)/rocm/llvm
+          -DLLVM_DIR=$(Agent.BuildDirectory)/rocm/llvm
           -DLLVM_MAIN_INCLUDE_DIR=$(Build.SourcesDirectory)/llvm-project/llvm/include
           -DLIBOMPTARGET_LLVM_INCLUDE_DIRS=$(Build.SourcesDirectory)/llvm-project/llvm/include
           -DCUDA_TOOLKIT_ROOT_DIR=OFF

.azuredevops/components/composable_kernel.yml

@@ -11,15 +11,25 @@ parameters:
     - cmake
     - ninja-build
     - git
+    - python3-pip
+    - libdrm-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocprofiler-register
 
 jobs:
 - job: composable_kernel
+  timeoutInMinutes: 210
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
+  pool: ${{ variables.ULTRA_BUILD_POOL }}
   workspace:
     clean: all
   steps:
@@ -30,12 +40,25 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DCMAKE_BUILD_TYPE=Release
-        -DGPU_TARGETS=gfx1030;gfx1100
+        -DINSTANCES_ONLY=ON
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/half.yml

@@ -8,8 +8,17 @@ parameters:
 - name: aptPackages
   type: object
   default:
+    - python3-pip
     - cmake
     - ninja-build
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
 
 jobs:
 - job: half
@@ -18,8 +27,6 @@ jobs:
   - template: /.azuredevops/variables-global.yml
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -30,9 +37,22 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+  # manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DBUILD_FILE_REORG_BACKWARD_COMPATIBILITY=OFF
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/hipBLAS.yml

@@ -10,25 +10,33 @@ parameters:
   default:
     - cmake
     - ninja-build
-    - rocblas-dev
-    - rocsparse
-    - rocsolver-dev
     - gfortran
     - googletest
     - git
     - libgtest-dev
     - wget
+    - python3-pip
+    - libomp-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocprofiler-register
+    - rocBLAS
+    - rocSPARSE
+    - rocSOLVER
+    - aomp
 
 jobs:
 - job: hipBLAS
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  - name: LD_LIBRARY_PATH
-    value: '/lib:/usr/lib:/usr/local/lib'
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -51,20 +59,24 @@ jobs:
       targetType: inline
       script: sudo apt install --yes ./aocl-linux-aocc-4.1.0_1_amd64.deb
       workingDirectory: '$(Pipeline.Workspace)'
-  - task: Bash@3
-    displayName: 'ldconfig'
-    inputs:
-      targetType: inline
-      script: sudo ldconfig
-      workingDirectory: '/usr/local/lib'
-  - script: 'ls -1R /usr/local'
-    displayName: 'Artifact listing'
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_PREFIX_PATH=/opt/rocm
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DCMAKE_BUILD_TYPE=Release
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -DHIP_PLATFORM=amd
         -DBUILD_CLIENTS_TESTS=ON

.azuredevops/components/hipBLASLt.yml

@@ -8,25 +8,44 @@ parameters:
 - name: aptPackages
   type: object
   default:
-    - cmake
     - ninja-build
     - python3-venv
     - libmsgpack-dev
-    - hipblas-dev
     - git
+    - python3-pip
+    - libdrm-dev
 - name: pipModules
   type: object
   default:
     - joblib
+- name: rocmDependencies
+  type: object
+  default:
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocprofiler-register
+    - hipBLAS
 
 jobs:
 - job: hipBLASLt
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Build.BinariesDirectory)/rocm
+  - name: TENSILE_ROCM_ASSEMBLER_PATH
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
+  - name: CMAKE_CXX_COMPILER
+    value: $(Agent.BuildDirectory)/rocm/bin/hipcc
+  - name: TENSILE_ROCM_OFFLOAD_BUNDLER_PATH
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin/clang-offload-bundler
+  - name: TENSILE_ROCM_PATH
+    value: $(Agent.BuildDirectory)/rocm/bin/hipcc
+  - name: PATH
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin:$(Agent.BuildDirectory)/rocm/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -34,21 +53,36 @@ jobs:
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-cmake-latest.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
+  - script: sudo ln -s $(Agent.BuildDirectory)/rocm /opt/rocm
+    displayName: ROCm symbolic link
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
         -DCMAKE_BUILD_TYPE=Release
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
         -DAMDGPU_TARGETS=gfx90a
         -DTensile_LOGIC=
         -DTensile_CPU_THREADS=
         -DTensile_CODE_OBJECT_VERSION=default
         -DTensile_LIBRARY_FORMAT=msgpack
-        -DCMAKE_PREFIX_PATH="/opt/rocm"
+        -DCMAKE_PREFIX_PATH="$(Agent.BuildDirectory)/rocm"
         -GNinja
-  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/hipCUB.yml

@@ -10,9 +10,17 @@ parameters:
   default:
     - cmake
     - ninja-build
-    - rocprim
     - googletest
     - git
+    - python3-pip
+- name: rocmDependencies
+  type: object
+  default:
+    - clr
+    - llvm-project
+    - rocminfo
+    - rocPRIM
+    - ROCR-Runtime
 
 jobs:
 - job: hipCUB
@@ -20,8 +28,6 @@ jobs:
   - group: common
   - template: /.azuredevops/variables-global.yml
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -32,12 +38,24 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/amdclang
-        -DCMAKE_PREFIX_PATH="/opt/rocm"
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DBUILD_TEST=ON
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -GNinja

.azuredevops/components/hipFFT.yml

@@ -10,22 +10,32 @@ parameters:
   default:
     - cmake
     - ninja-build
-    - rocrand
-    - hiprand
-    - rocfft
     - libboost-program-options-dev
-    - googletest
+    - libgtest-dev
     - libfftw3-dev
-
+    - python3-pip
+    - libomp-14-dev
+# rocm dependencies should match dependencies-rocm.yml
+- name: rocmDependencies
+  type: object
+  default:
+    - rocRAND
+    - hipRAND
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocFFT
+    - aomp
 jobs:
 - job: hipFFT
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Build.BinariesDirectory)/rocm
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -36,16 +46,31 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
+        -DCMAKE_MODULE_PATH=$(Agent.BuildDirectory)/rocm/lib/cmake/hip
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DCMAKE_BUILD_TYPE=Release
         -DAMDGPU_TARGETS=gfx1030;gfx1100
+        -DUSE_HIP_CLANG=ON
+        -DHIP_COMPILER=clang
         -DBUILD_CLIENTS_TESTS=ON
-        -DBUILD_CLIENTS_BENCH=OFF
+        -DBUILD_CLIENTS_BENCHMARKS=OFF
         -DBUILD_CLIENTS_SAMPLES=OFF
-        -L
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/hipSOLVER.yml

@@ -10,15 +10,23 @@ parameters:
   default:
     - cmake
     - ninja-build
-    - rocblas
-    - rocsparse
-    - hipsparse
-    - rocsolver
     - libsuitesparse-dev
     - gfortran
     - git
     - googletest
     - libgtest-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - clr
+    - hipSPARSE
+    - llvm-project
+    - rocBLAS
+    - rocm-cmake
+    - rocminfo
+    - ROCR-Runtime
+    - rocSPARSE
+    - rocSOLVER
 
 jobs:
 - job: hipSOLVER
@@ -27,8 +35,6 @@ jobs:
   - template: /.azuredevops/variables-global.yml
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -39,6 +45,18 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
 # build external gtest and lapack
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
@@ -52,10 +70,10 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_PREFIX_PATH="/opt/rocm;$(Pipeline.Workspace)/deps-install"
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Pipeline.Workspace)/deps-install
         -DCMAKE_BUILD_TYPE=Release
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -DBUILD_CLIENTS_TESTS=ON
         -DUSE_CUDA=OFF

.azuredevops/components/hipSPARSE.yml

@@ -13,10 +13,18 @@ parameters:
     - libboost-program-options-dev
     - googletest
     - libfftw3-dev
-    - rocsparse
     - git
     - gfortran
     - libgtest-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - clr
+    - llvm-project
+    - rocminfo
+    - rocprofiler-register
+    - ROCR-Runtime
+    - rocSPARSE
 
 jobs:
 - job: hipSPARSE
@@ -25,8 +33,6 @@ jobs:
   - template: /.azuredevops/variables-global.yml
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -37,15 +43,35 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
         -DCMAKE_BUILD_TYPE=Release
-        -DCMAKE_PREFIX_PATH="/opt/rocm;/opt/rocm/share/rocm/cmake/"
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/rocm/share/rocm/cmake/
         -DBUILD_CLIENTS_TESTS=ON
         -DBUILD_CLIENTS_SAMPLES=OFF
-        -DBUILD_CLIENTS_BENCHMARKS=OFF
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
+    parameters:
+      artifactName: hipSPARSE
+      publish: false
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-prepare-package.yml
+    parameters:
+      sourceDir: $(Build.SourcesDirectory)/build/clients
+      contentsString: matrices/**
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
+    parameters:
+      artifactName: testMatrices

.azuredevops/components/hipSPARSELt.yml

@@ -8,25 +8,43 @@ parameters:
 - name: aptPackages
   type: object
   default:
-    - cmake
     - ninja-build
     - python3-venv
     - libmsgpack-dev
-    - hipsparse-dev
     - git
+    - python3-pip
 - name: pipModules
   type: object
   default:
     - joblib
+# rocm dependencies should match dependencies-rocm.yml
+- name: rocmDependencies
+  type: object
+  default:
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocprofiler-register
+    - hipSPARSE
+    - rocBLAS
 
 jobs:
 - job: hipSPARSELt
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Build.BinariesDirectory)/rocm
+  - name: TENSILE_ROCM_ASSEMBLER_PATH
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin/clang
+  - name: CMAKE_CXX_COMPILER
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin/hipcc
+  - name: TENSILE_ROCM_OFFLOAD_BUNDLER_PATH
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin/clang-offload-bundler
+  - name: PATH
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin:$(Agent.BuildDirectory)/rocm/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -34,21 +52,35 @@ jobs:
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-cmake-latest.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
         -DCMAKE_BUILD_TYPE=Release
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
         -DAMDGPU_TARGETS=all
         -DTensile_LOGIC=
         -DTensile_CPU_THREADS=
         -DTensile_CODE_OBJECT_VERSION=default
         -DTensile_LIBRARY_FORMAT=msgpack
-        -DCMAKE_PREFIX_PATH="/opt/rocm"
+        -DCMAKE_PREFIX_PATH="$(Agent.BuildDirectory)/rocm"
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/hipTensor.yml

@@ -10,12 +10,17 @@ parameters:
   default:
     - cmake
     - ninja-build
-    - composablekernel-dev
+    - python3-pip
     - git
 - name: rocmDependencies
   type: object
   default:
+    - rocm-cmake
     - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - composable_kernel
 
 jobs:
 - job: hipTensor
@@ -23,8 +28,6 @@ jobs:
   - group: common
   - template: /.azuredevops/variables-global.yml
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -50,12 +53,11 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/amdclang
-        -DCMAKE_PREFIX_PATH="$(Agent.BuildDirectory)/rocm/llvm"
-        -DROCM_PATH="$(Agent.BuildDirectory)/rocm"
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/rocm/llvm
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
         -DCMAKE_BUILD_TYPE=Release
         -DHIPTENSOR_BUILD_TESTS=ON
-        -DAMDGPU_TARGETS=gfx1030;gfx1100
-        -GNinja
+        -DAMDGPU_TARGETS=gfx90a
+      multithreadFlag: -- -j32
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/llvm-project.yml

@@ -14,6 +14,7 @@ parameters:
     - ninja-build
     - python-is-python3
     - zlib1g-dev
+    - pkg-config
 - name: rocmDependencies
   type: object
   default:
@@ -68,8 +69,6 @@ jobs:
         -DLIBCXXABI_INSTALL_STATIC_LIBRARY=OFF
         -DLLVM_BUILD_DOCS=OFF
         -DLLVM_ENABLE_SPHINX=OFF
-        -DSPHINX_WARNINGS_AS_ERRORS=OFF
-        -DSPHINX_OUTPUT_MAN=OFF
         -DLLVM_ENABLE_ASSERTIONS=OFF
         -DLLVM_ENABLE_Z3_SOLVER=OFF
         -DLLVM_ENABLE_ZLIB=ON
@@ -80,7 +79,6 @@ jobs:
         -DPACKAGE_VENDOR=AMD
         -DCLANG_LINK_FLANG_LEGACY=ON
         -DCMAKE_CXX_STANDARD=17
-        -DFLANG_INCLUDE_DOCS=OFF
         -DROCM_LLVM_BACKWARD_COMPAT_LINK=$(Build.BinariesDirectory)/llvm
         -DROCM_LLVM_BACKWARD_COMPAT_LINK_TARGET=./lib/llvm
         -GNinja

.azuredevops/components/rccl.yml

@@ -8,21 +8,37 @@ parameters:
 - name: aptPackages
   type: object
   default:
+    - python3-pip
     - cmake
     - libboost-program-options-dev
     - googletest
     - libfftw3-dev
     - git
     - ninja-build
+    - libstdc++-12-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocm_smi_lib
+    - rocprofiler-register
+    - rocm-core
+    - HIPIFY
+    - aomp
+    - aomp-extras
 
 jobs:
 - job: rccl
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Build.BinariesDirectory)/rocm
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -33,14 +49,29 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+  # manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
+  - script: chmod +x $(Agent.BuildDirectory)/rocm/bin/hipify-perl
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
+        -DHALF_INCLUDE_DIR=$(Agent.BuildDirectory)/rocm/include
         -DCMAKE_BUILD_TYPE=Release
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
         -DBUILD_TESTS=ON
-        -DCMAKE_PREFIX_PATH="/opt/rocm;/opt/rocm/share/rocm/cmake/"
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/rocm/share/rocm/cmake/
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rdc.yml

@@ -8,6 +8,7 @@ parameters:
 - name: aptPackages
   type: object
   default:
+    - python3-pip
     - cmake
     - ninja-build
     - git
@@ -17,6 +18,16 @@ parameters:
     - autoconf
     - libtool
     - pkg-config
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocm_smi_lib
+    - amdsmi
 
 jobs:
 - job: rdc
@@ -24,8 +35,6 @@ jobs:
   - group: common
   - template: /.azuredevops/variables-global.yml
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -36,6 +45,18 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+  # manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
 # Build grpc
   - task: Bash@3
     displayName: 'git clone grpc'
@@ -57,6 +78,7 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DGRPC_ROOT="$(Build.SourcesDirectory)/bin"
         -DBUILD_TESTS=ON
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocAL.yml

@@ -0,0 +1,138 @@
+parameters:
+- name: checkoutRepo
+  type: string
+  default: 'self'
+- name: checkoutRef
+  type: string
+  default: ''
+- name: aptPackages
+  type: object
+  default:
+    - python3-pip
+    - python3-protobuf
+    - cmake
+    - ninja-build
+    - libprotobuf-dev
+    - libprotoc-dev
+    - protobuf-compiler
+    - liblmdb-dev
+    - pkg-config
+    - ffmpeg
+    - libavcodec-dev
+    - libavformat-dev
+    - libavutil-dev
+    - libswscale-dev
+    - libturbojpeg-dev
+    - libjpeg-turbo-official=3.0.2-20240124
+    - libopencv-dev
+- name: pipModules
+  type: object
+  default:
+    - numpy
+    - opencv-python
+    - torch
+    - pillow
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocDecode
+    - half
+    - rpp
+    - MIVisionX
+    - aomp
+
+jobs:
+- job: rocAL
+  variables:
+  - group: common
+  - template: /.azuredevops/variables-global.yml
+  pool:
+    vmImage: ${{ variables.BASE_BUILD_POOL }}
+  workspace:
+    clean: all
+  steps:
+  - task: Bash@3
+    displayName: 'Register libjpeg-turbo packages'
+    inputs:
+      targetType: inline
+      script: |
+        sudo mkdir --parents --mode=0755 /etc/apt/keyrings
+        wget -q -O- https://packagecloud.io/dcommander/libjpeg-turbo/gpgkey | gpg --dearmor | sudo tee /etc/apt/trusted.gpg.d/libjpeg-turbo.gpg > /dev/null
+        echo "deb [signed-by=/etc/apt/trusted.gpg.d/libjpeg-turbo.gpg] https://packagecloud.io/dcommander/libjpeg-turbo/any/ any main" | sudo tee /etc/apt/sources.list.d/libjpeg-turbo.list
+        sudo apt update
+        apt-cache show libjpeg-turbo-official | grep Version
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
+    parameters:
+      aptPackages: ${{ parameters.aptPackages }}
+      pipModules: ${{ parameters.pipModules }}
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
+    parameters:
+      checkoutRepo: ${{ parameters.checkoutRepo }}
+  - task: Bash@3
+    displayName: 'Clone PyBind11'
+    inputs:
+      targetType: inline
+      script: git clone --depth 1 -b v2.11.1 https://github.com/pybind/pybind11
+      workingDirectory: '$(Build.SourcesDirectory)'
+  - task: Bash@3
+    displayName: 'Clone RapidJSON'
+    inputs:
+      targetType: inline
+      script: git clone --depth 1 https://github.com/Tencent/rapidjson.git
+      workingDirectory: '$(Build.SourcesDirectory)'
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
+    parameters:
+      componentName: PyBind11
+      cmakeBuildDir: '$(Build.SourcesDirectory)/pybind11/build'
+      customInstallPath: false
+      installEnabled: false
+      extraBuildFlags: >-
+        -DDOWNLOAD_CATCH=ON
+        -DDOWNLOAD_EIGEN=ON
+        -GNinja
+  - task: Bash@3
+    displayName: 'Install PyBind11'
+    inputs:
+      targetType: inline
+      script: sudo cmake --build . --target install
+      workingDirectory: '$(Build.SourcesDirectory)/pybind11/build'
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
+    parameters:
+      componentName: RapidJSON
+      cmakeBuildDir: '$(Build.SourcesDirectory)/rapidjson/build'
+      customInstallPath: false
+      installEnabled: false
+      extraBuildFlags: >-
+        -GNinja
+  - task: Bash@3
+    displayName: 'Install RapidJSON'
+    inputs:
+      targetType: inline
+      script: sudo cmake --build . --target install
+      workingDirectory: '$(Build.SourcesDirectory)/rapidjson/build'
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+  # manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
+    parameters:
+      extraBuildFlags: >-
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;/opt/libjpeg-turbo
+        -DCMAKE_INSTALL_PREFIX_PYTHON=$Python3_STDARCH
+        -DCMAKE_BUILD_TYPE=Release
+        -GNinja
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocALUTION.yml

@@ -15,16 +15,29 @@ parameters:
     - git
     - mpich
     - ninja-build
+- name: rocmDependencies
+  type: object
+  default:
+    - aomp
+    - clr
+    - llvm-project
+    - rocBLAS
+    - rocminfo
+    - rocPRIM
+    - rocprofiler-register
+    - ROCR-Runtime
+    - rocRAND
+    - rocSPARSE
 
 jobs:
 - job: rocALUTION
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Build.BinariesDirectory)/rocm
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -35,13 +48,25 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
         -DCMAKE_BUILD_TYPE=Release
-        -DCMAKE_PREFIX_PATH="/opt/rocm;/opt/rocm/share/rocm/cmake/"
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/rocm/share/rocm/cmake/
+        -DCMAKE_MODULE_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/rocm/lib/cmake/hip
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -DBUILD_CLIENTS_TESTS=ON
         -DBUILD_CLIENTS_BENCHMARKS=OFF

.azuredevops/components/rocBLAS.yml

@@ -18,19 +18,40 @@ parameters:
     - googletest
     - libgtest-dev
     - wget
+    - python3-pip
+    - libdrm-dev
 - name: pipModules
   type: object
   default:
     - joblib
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocprofiler-register
+    - rocm_smi_lib
+    - rocm-core
+    - aomp
+    - aomp-extras
 
 jobs:
 - job: rocBLAS
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Build.BinariesDirectory)/rocm
+  - name: TENSILE_ROCM_ASSEMBLER_PATH
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin/clang
+  - name: CMAKE_CXX_COMPILER
+    value: $(Agent.BuildDirectory)/rocm/bin/hipcc
+  - name: TENSILE_ROCM_OFFLOAD_BUNDLER_PATH
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin/clang-offload-bundler
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -42,23 +63,60 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - task: Bash@3
+      displayName: 'Download AOCL'
+      inputs:
+        targetType: inline
+        script: wget -nv https://download.amd.com/developer/eula/aocl/aocl-4-2/aocl-linux-gcc-4.2.0_1_amd64.deb
+        workingDirectory: '$(Pipeline.Workspace)'
+    - task: Bash@3
+      displayName: 'Install AOCL'
+      inputs:
+        targetType: inline
+        script: sudo apt install --yes ./aocl-linux-gcc-4.2.0_1_amd64.deb
+        workingDirectory: '$(Pipeline.Workspace)'
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - task: Bash@3
+      displayName: 'Download AOCL'
+      inputs:
+        targetType: inline
+        script: wget -nv https://download.amd.com/developer/eula/aocl/aocl-4-1/aocl-linux-aocc-4.1.0_1_amd64.deb
+        workingDirectory: '$(Pipeline.Workspace)'
+    - task: Bash@3
+      displayName: 'Install AOCL'
+      inputs:
+        targetType: inline
+        script: sudo apt install --yes ./aocl-linux-aocc-4.1.0_1_amd64.deb
+        workingDirectory: '$(Pipeline.Workspace)'
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
+  - script: echo $PATH
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
         -DCMAKE_TOOLCHAIN_FILE=toolchain-linux.cmake
-        -DCMAKE_PREFIX_PATH="/opt/rocm;$(Pipeline.Workspace)/deps-install"
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm/llvm;$(Agent.BuildDirectory)/rocm;$(Pipeline.Workspace)/deps-install
         -DCMAKE_BUILD_TYPE=Release
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -DTensile_CODE_OBJECT_VERSION=default
         -DTensile_LOGIC=asm_full
         -DTensile_SEPARATE_ARCHITECTURES=ON
         -DTensile_LAZY_LIBRARY_LOADING=ON
         -DTensile_LIBRARY_FORMAT=msgpack
-        -DTENSILE_VENV_UPGRADE_PIP=ON
         -DBUILD_CLIENTS_TESTS=ON
         -DBUILD_CLIENTS_BENCHMARKS=OFF
         -DBUILD_CLIENTS_SAMPLES=OFF
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocDecode.yml

@@ -8,6 +8,7 @@ parameters:
 - name: aptPackages
   type: object
   default:
+    - python3-pip
     - cmake
     - ninja-build
     - pkg-config
@@ -18,6 +19,16 @@ parameters:
     - libstdc++-12-dev
     - libva-dev
     - mesa-amdgpu-va-drivers
+    - libdrm-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocm-core
 
 jobs:
 - job: rocDecode
@@ -26,11 +37,21 @@ jobs:
   - template: /.azuredevops/variables-global.yml
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
+  # Since mesa-amdgpu-multimedia-devel is not directly available from apt, register it
+  - task: Bash@3
+    displayName: 'Register ROCm packages'
+    inputs:
+      targetType: inline
+      script: |
+        sudo mkdir --parents --mode=0755 /etc/apt/keyrings
+        wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/${{ variables.KEYRING_VERSION }}/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/${{ variables.KEYRING_VERSION }} jammy main" | sudo tee --append /etc/apt/sources.list.d/rocm.list
+        echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
+        sudo apt update
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
@@ -38,10 +59,24 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+  # manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DCMAKE_BUILD_TYPE=Release
-        -L
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocFFT.yml

@@ -10,20 +10,31 @@ parameters:
   default:
     - cmake
     - ninja-build
-    - rocrand
-    - hiprand
     - libboost-program-options-dev
     - libgtest-dev
     - libfftw3-dev
-
+    - python3-pip
+# rocm dependencies should match dependencies-rocm.yml
+- name: rocmDependencies
+  type: object
+  default:
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocprofiler-register
+    - hipRAND
+    - rocRAND
+    - rocm-cmake
+    - aomp
 jobs:
 - job: rocFFT
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Build.BinariesDirectory)/rocm
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -34,12 +45,24 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
-        -DCMAKE_PREFIX_PATH=/opt/rocm
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DCMAKE_BUILD_TYPE=Release
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -DUSE_HIP_CLANG=ON

.azuredevops/components/rocMLIR.yml

@@ -10,6 +10,13 @@ parameters:
   default:
     - cmake
     - ninja-build
+    - git
+    - python3-pip
+- name: rocmDependencies
+  type: object
+  default:
+    - llvm-project
+    - rocm-cmake
 
 jobs:
 - job: rocMLIR
@@ -17,8 +24,6 @@ jobs:
   - group: common
   - template: /.azuredevops/variables-global.yml
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -29,13 +34,25 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
         -DCMAKE_BUILD_TYPE=Release
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
-        -DCMAKE_PREFIX_PATH=/opt/rocm
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/clang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/clang
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DBUILD_FAT_LIBROCKCOMPILER=1
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocPRIM.yml

@@ -12,6 +12,15 @@ parameters:
     - ninja-build
     - libgtest-dev
     - git
+    - python3-pip
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
 
 jobs:
 - job: rocPRIM
@@ -19,8 +28,6 @@ jobs:
   - group: common
   - template: /.azuredevops/variables-global.yml
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -34,12 +41,24 @@ jobs:
 # ${{ }} are resolved during compile-time
 # so this next step is skipped completely until
 # we define explicit aptPackages needed to install
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+  # manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DBUILD_BENCHMARK=ON
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -DBUILD_TEST=ON
         -GNinja

.azuredevops/components/rocSOLVER.yml

@@ -10,15 +10,25 @@ parameters:
   default:
     - cmake
     - ninja-build
-    - rocblas
-    - rocsparse
-    - hipsparse
     - libsuitesparse-dev
     - gfortran
     - libfmt-dev
     - git
     - googletest
     - libgtest-dev
+    - python3-pip
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocBLAS
+    - rocPRIM
+    - rocSPARSE
+    - hipSPARSE
 
 jobs:
 - job: rocSOLVER
@@ -26,8 +36,6 @@ jobs:
   - group: common
   - template: /.azuredevops/variables-global.yml
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -44,6 +52,18 @@ jobs:
       targetType: inline
       script: git clone --depth 1 --branch v3.9.1 https://github.com/Reference-LAPACK/lapack
       workingDirectory: '$(Build.SourcesDirectory)'
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       componentName: lapack
@@ -59,11 +79,10 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_PREFIX_PATH="/opt/rocm;$(Pipeline.Workspace)/deps-install"
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Pipeline.Workspace)/deps-install
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
         -DAMDGPU_TARGETS=gfx1030;gfx1100
-        -DUSE_CUDA=OFF
         -DBUILD_CLIENTS_TESTS=ON
         -DBUILD_CLIENTS_BENCHMARKS=OFF
         -DBUILD_CLIENTS_SAMPLES=OFF

.azuredevops/components/rocSPARSE.yml

@@ -8,6 +8,7 @@ parameters:
 - name: aptPackages
   type: object
   default:
+    - python3-pip
     - cmake
     - ninja-build
     - libboost-program-options-dev
@@ -15,17 +16,28 @@ parameters:
     - libfftw3-dev
     - git
     - gfortran
-    - rocprim-dev
     - libgtest-dev
+    - libdrm-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocBLAS
+    - rocminfo
+    - rocPRIM
+    - rocprofiler-register
 
 jobs:
 - job: rocSPARSE
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Build.BinariesDirectory)/rocm
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -36,16 +48,40 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+  # manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/hipcc
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/hipcc
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
         -DCMAKE_BUILD_TYPE=Release
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -DBUILD_CLIENTS_SAMPLES=OFF
         -DBUILD_CLIENTS_TESTS=ON
         -DBUILD_CLIENTS_BENCHMARKS=OFF
-        -DCMAKE_MODULE_PATH="/opt/rocm/lib/cmake/hip;/opt/rocm/hip/cmake"
+        -DCMAKE_MODULE_PATH=$(Agent.BuildDirectory)/rocm/lib/cmake/hip;$(Agent.BuildDirectory)/rocm/hip/cmake
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
+    parameters:
+      artifactName: rocSPARSE
+      publish: false
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-prepare-package.yml
+    parameters:
+      sourceDir: $(Build.SourcesDirectory)/build/clients
+      contentsString: matrices/**
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
+    parameters:
+      artifactName: testMatrices

.azuredevops/components/rocThrust.yml

@@ -10,12 +10,20 @@ parameters:
   default:
     - cmake
     - ninja-build
-    - hiprand
-    - rocprim-dev
     - libboost-program-options-dev
     - googletest
     - libfftw3-dev
     - git
+    - python3-pip
+- name: rocmDependencies
+  type: object
+  default:
+    - clr
+    - hipRAND
+    - llvm-project
+    - rocminfo
+    - rocPRIM
+    - ROCR-Runtime
 
 jobs:
 - job: rocThrust
@@ -23,8 +31,6 @@ jobs:
   - group: common
   - template: /.azuredevops/variables-global.yml
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -35,14 +41,25 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
         -GNinja
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
-        -DROCM_PATH=/opt/rocm
-        -DCMAKE_PREFIX_PATH=/opt/rocm
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -DBUILD_TEST=ON
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocWMMA.yml

@@ -8,6 +8,7 @@ parameters:
 - name: aptPackages
   type: object
   default:
+    - python3-pip
     - cmake
     - ninja-build
     - libboost-program-options-dev
@@ -15,7 +16,18 @@ parameters:
     - googletest
     - libfftw3-dev
     - git
-    - rocblas
+    - libomp-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - llvm-project
+    - ROCR-Runtime
+    - clr
+    - rocminfo
+    - rocBLAS
+    - aomp
+    - rocm_smi_lib
 
 jobs:
 - job: rocWMMA
@@ -23,8 +35,6 @@ jobs:
   - group: common
   - template: /.azuredevops/variables-global.yml
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -35,11 +45,23 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+  # manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/bin/amdclang
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
         -DCMAKE_BUILD_TYPE=Release
         -DROCWMMA_BUILD_TESTS=ON
         -DROCWMMA_BUILD_SAMPLES=OFF

.azuredevops/components/rocm-examples.yml

@@ -5,6 +5,30 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+- name: aptPackages
+  type: object
+  default:
+    - libglfw3-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - AMDMIGraphX
+    - clr
+    - hipBLAS
+    - hipCUB
+    - HIPIFY
+    - hipRAND
+    - hipSOLVER
+    - hipSPARSE
+    - llvm-project
+    - rocBLAS
+    - rocPRIM
+    - rocprofiler-register
+    - ROCR-Runtime
+    - rocRAND
+    - rocSOLVER
+    - rocSPARSE
+    - rocThrust
 
 jobs:
 - job: rocm_examples
@@ -20,5 +44,28 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
+    parameters:
+      aptPackages: ${{ parameters.aptPackages }}
+  # CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+  # manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
+    parameters:
+      # https://github.com/ROCm/HIP/issues/2203
+      extraBuildFlags: >-
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
+        -DROCM_ROOT=$(Agent.BuildDirectory)/rocm
+        -DCMAKE_HIP_ARCHITECTURES=gfx1030;gfx1100
+        -DCMAKE_EXE_LINKER_FLAGS=-fgpu-rdc
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocm_bandwidth_test.yml

@@ -10,21 +10,33 @@ parameters:
   default:
     - cmake
     - ninja-build
+    - python3-pip
 - name: pipModules
   type: object
   default:
     - CppHeaderParser
     - argparse
+- name: rocmDependencies
+  type: object
+  default:
+    - clr
+    - llvm-project
+    - rocminfo
+    - rocprofiler-register
+    - ROCR-Runtime
+    - ROCT-Thunk-Interface
 
 jobs:
 - job: rocm_bandwidth_test
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: ROCR_INC_DIR
+    value: $(Agent.BuildDirectory)/rocm
+  - name: ROCR_LIB_DIR
+    value: $(Agent.BuildDirectory)/rocm
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -36,11 +48,23 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
         -DCMAKE_BUILD_TYPE=release
-        -DCMAKE_MODULE_PATH="$(Build.SourcesDirectory)/cmake_modules"
-        -DCMAKE_PREFIX_PATH=/opt/rocm
+        -DCMAKE_MODULE_PATH=$(Build.SourcesDirectory)/cmake_modules
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/rocm/include;$(Agent.BuildDirectory)/rocm/include/hsa
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocm_smi_lib.yml

@@ -24,4 +24,5 @@ jobs:
     parameters:
       extraBuildFlags: >-
         -DBUILD_TESTS=ON
+        -DROCM_DEP_ROCMCORE=ON
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocprofiler-register.yml

@@ -21,4 +21,17 @@ jobs:
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
+    parameters:
+      componentName: rocprofiler-register
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
+    parameters:
+      componentName: rocprofiler-register-tests
+      extraBuildFlags: >-
+        -DCMAKE_PREFIX_PATH=$(Build.BinariesDirectory)
+      cmakeBuildDir: 'tests/build'
+      installEnabled: false
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
+    parameters:
+      componentName: rocprofiler-register
+      testDir: 'tests/build'
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rocprofiler.yml

@@ -10,12 +10,13 @@ parameters:
   default:
     - cmake
     - libgtest-dev
+    - libdrm-dev
     - libdw-dev
     - libsystemd-dev
     - libelf-dev
     - libnuma-dev
     - libpciaccess-dev
-    - rocm-llvm-dev
+    - python3-pip
 - name: pipModules
   type: object
   default:
@@ -26,15 +27,31 @@ parameters:
     - lxml
     - barectf
     - pandas
+- name: rocmDependencies
+  type: object
+  default:
+    - clr
+    - llvm-project
+    - ROCdbgapi
+    - rocm-cmake
+    - rocm-core
+    - rocm_smi_lib
+    - rocminfo
+    - ROCR-Runtime
+    - rocprofiler-register
+    - ROCT-Thunk-Interface
+    - roctracer
 
 jobs:
 - job: rocprofiler
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
+  - name: HIP_ROCCLR_HOME
+    value: $(Agent.BuildDirectory)/rocm
+  - name: ROCM_PATH
+    value: $(Agent.BuildDirectory)/rocm
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -46,11 +63,46 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# Manually download aqlprofile, hard-coded 6.1.0 version
+  - task: Bash@3
+    displayName: 'Download aqlprofile'
+    inputs:
+      targetType: inline
+      script: wget -nv https://repo.radeon.com/rocm/misc/aqlprofile/ubuntu-22.04/hsa-amd-aqlprofile_1.0.0.60200.60200-crdnnh.14213~22.04_amd64.deb
+      workingDirectory: '$(Pipeline.Workspace)'
+  - task: Bash@3
+    displayName: 'Extract aqlprofile'
+    inputs:
+      targetType: inline
+      script: |
+        mkdir hsa-amd-aqlprofile
+        dpkg-deb -R hsa-amd-aqlprofile_1.0.0.60200.60200-crdnnh.14213~22.04_amd64.deb hsa-amd-aqlprofile
+      workingDirectory: '$(Pipeline.Workspace)'
+  - task: Bash@3
+    displayName: 'Move aqlprofile'
+    inputs:
+      targetType: inline
+      script: |
+        mkdir -p $(Agent.BuildDirectory)/rocm
+        cp -R hsa-amd-aqlprofile/opt/rocm-6.2.0-14213/* $(Agent.BuildDirectory)/rocm
+      workingDirectory: '$(Pipeline.Workspace)'
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_MODULE_PATH="$(Build.SourcesDirectory)/cmake_modules;/opt/rocm/lib/cmake"
-        -DCMAKE_PREFIX_PATH="/opt/rocm"
+        -DCMAKE_MODULE_PATH=$(Build.SourcesDirectory)/cmake_modules;$(Agent.BuildDirectory)/rocm/lib/cmake;$(Agent.BuildDirectory)/rocm/lib/cmake/hip
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DENABLE_LDCONFIG=OFF
         -DUSE_PROF_API=1
         -DGPU_TARGETS=gfx1030;gfx1100

.azuredevops/components/rocr_debug_agent.yml

@@ -12,6 +12,16 @@ parameters:
     - ninja-build
     - libelf-dev
     - libdw-dev
+- name: rocmDependencies
+  type: object
+  default:
+    - rocm-cmake
+    - clr
+    - llvm-project
+    - ROCdbgapi
+    - rocminfo
+    - ROCR-Runtime
+    - rocprofiler-register
 
 jobs:
 - job: rocr_debug_agent
@@ -20,8 +30,6 @@ jobs:
   - template: /.azuredevops/variables-global.yml
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -32,11 +40,24 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
         -DCMAKE_BUILD_TYPE=Release
-        -DROCM_PATH=/opt/rocm
-        -DCMAKE_MODULE_PATH=/opt/rocm/lib/cmake
+        -DCMAKE_MODULE_PATH=$(Agent.BuildDirectory)/rocm/lib/cmake;$(Agent.BuildDirectory)/rocm/lib/cmake/hip
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/roctracer.yml

@@ -67,10 +67,4 @@ jobs:
         -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
         -DGPU_TARGETS=gfx1030;gfx1100
         -GNinja
-#  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
-#  - task: Bash@3
-#    displayName: 'Tests'
-#    inputs:
-#      targetType: inline
-#      script: ./run.sh
-#      workingDirectory: build
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/components/rpp.yml

@@ -9,10 +9,18 @@ parameters:
   type: object
   default:
     - cmake
+    - libomp-dev # needed to pass flag step
     - ninja-build
     - clang
+- name: rocmDependencies
+  type: object
+  default:
+    - aomp # needed to pass build step
+    - clr
     - half
-    - libomp-dev
+    - llvm-project
+    - rocminfo
+    - ROCR-Runtime
 
 jobs:
 - job: rpp
@@ -21,8 +29,6 @@ jobs:
   - template: /.azuredevops/variables-global.yml
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
-  container:
-    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:
@@ -33,13 +39,27 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
+# CI case: download latest default branch build
+  - ${{ if eq(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: staging
+# manual build case: triggered by ROCm/ROCm repo
+  - ${{ if ne(parameters.checkoutRef, '') }}:
+    - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+      parameters:
+        dependencyList: ${{ parameters.rocmDependencies }}
+        dependencySource: tag-builds
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
-        -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/amdclang++
-        -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
+        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
+        -DHALF_INCLUDE_DIRS=$(Agent.BuildDirectory)/rocm/include
         -DCMAKE_BUILD_TYPE=Release
-        -DBUILD_CLIENTS=ON
         -DAMDGPU_TARGETS=gfx1030;gfx1100
         -GNinja
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml

.azuredevops/tag-builds/ROCgdb.yml

@@ -23,7 +23,7 @@ trigger: none
 pr: none
 
 jobs:
-  - template: ${{ variables.CI_COMPONENT_PATH }}/rocgdb.yml
+  - template: ${{ variables.CI_COMPONENT_PATH }}/ROCgdb.yml
     parameters:
       checkoutRepo: release_repo
       checkoutRef: ${{ parameters.checkoutRef }}

.azuredevops/tag-builds/rocAL.yml

@@ -0,0 +1,29 @@
+variables:
+- group: common
+- template: /.azuredevops/variables-global.yml
+
+parameters:
+- name: checkoutRef
+  type: string
+  default: refs/tags/$(LATEST_RELEASE_TAG)
+
+resources:
+  repositories:
+  - repository: pipelines_repo
+    type: github
+    endpoint: ROCm
+    name: ROCm/ROCm
+  - repository: release_repo
+    type: github
+    endpoint: ROCm
+    name: ROCm/rocAL
+    ref: ${{ parameters.checkoutRef }}
+
+trigger: none
+pr: none
+
+jobs:
+  - template: ${{ variables.CI_COMPONENT_PATH }}/rocAL.yml
+    parameters:
+      checkoutRepo: release_repo
+      checkoutRef: ${{ parameters.checkoutRef }}

.azuredevops/tag-builds/rocm-examples.yml

@@ -0,0 +1,29 @@
+variables:
+- group: common
+- template: /.azuredevops/variables-global.yml
+
+parameters:
+- name: checkoutRef
+  type: string
+  default: refs/tags/$(LATEST_RELEASE_TAG)
+
+resources:
+  repositories:
+  - repository: pipelines_repo
+    type: github
+    endpoint: ROCm
+    name: ROCm/ROCm
+  - repository: release_repo
+    type: github
+    endpoint: ROCm
+    name: ROCm/rocm-examples
+    ref: ${{ parameters.checkoutRef }}
+
+trigger: none
+pr: none
+
+jobs:
+  - template: ${{ variables.CI_COMPONENT_PATH }}/rocm-examples.yml
+    parameters:
+      checkoutRepo: release_repo
+      checkoutRef: ${{ parameters.checkoutRef }}

.azuredevops/templates/steps/artifact-download.yml

@@ -12,6 +12,7 @@ parameters:
 - name: defaultBranchList
   type: object
   default:
+    amdsmi: develop
     aomp: aomp-dev
     aomp-extras: aomp-dev
     AMDMIGraphX: develop
@@ -20,13 +21,20 @@ parameters:
     half: master
     HIP: develop
     hipBLAS: develop
+    hipCUB: develop
     hipRAND: develop
+    hipSOLVER: develop
     hipSPARSE: develop
     llvm-project: amd-staging
     MIOpen: develop
+    MIVisionX: develop
+    rdc: develop
     rocBLAS: develop
+    ROCdbgapi : amd-master
+    rocDecode: develop
     rocFFT: develop
     rocm-cmake: develop
+    rocm_smi_lib: develop
     rocminfo: master
     rocMLIR: develop
     rocPRIM: develop
@@ -36,6 +44,8 @@ parameters:
     rocSOLVER: develop
     rocSPARSE: develop
     ROCT-Thunk-Interface: master
+    rocThrust: develop
+    roctracer: amd-master
     rpp: master
 - name: componentsFailureOkay
   type: object

.azuredevops/templates/steps/build-cmake.yml

@@ -5,6 +5,9 @@ parameters:
 - name: extraBuildFlags
   type: string
   default: ''
+- name: multithreadFlag
+  type: string
+  default: ''
 - name: cmakeBuildDir
   type: string
   default: 'build'
@@ -17,6 +20,12 @@ parameters:
 - name: installDir
   type: string
   default: '$(Build.BinariesDirectory)'
+- name: customInstallPath
+  type: boolean
+  default: true
+- name: installEnabled
+  type: boolean
+  default: true
 
 steps:
 # create workingDirectory if it does not exist and change into it
@@ -25,19 +34,23 @@ steps:
   displayName: '${{parameters.componentName }} CMake Flags'
   inputs:
     workingDirectory: ${{ parameters.cmakeBuildDir }}
-    cmakeArgs: -DCMAKE_INSTALL_PREFIX=${{ parameters.installDir }} ${{ parameters.extraBuildFlags }} ..
+    ${{ if eq(parameters.customInstallPath, true) }}:
+      cmakeArgs: -DCMAKE_INSTALL_PREFIX=${{ parameters.installDir }} ${{ parameters.extraBuildFlags }} ..
+    ${{ else }}:
+      cmakeArgs: ${{ parameters.extraBuildFlags }} ..
 # equivalent to running make $cmakeTargetDir from $cmakeBuildDir
 # i.e., cd $cmakeBuildDir; make $cmakeTargetDir
 - task: CMake@1
   displayName: '${{parameters.componentName }} Build'
   inputs:
     workingDirectory: ${{ parameters.cmakeBuildDir }}
-    cmakeArgs: '--build ${{ parameters.cmakeTargetDir }}'
+    cmakeArgs: '--build ${{ parameters.cmakeTargetDir }} ${{ parameters.multithreadFlag }}'
     retryCountOnTaskFailure: 10
 # equivalent to running make $cmakeTarget from $cmakeBuildDir
 # e.g., make install
-- task: CMake@1
-  displayName: '${{parameters.componentName }} ${{ parameters.cmakeTarget }}'
-  inputs:
-    workingDirectory: ${{ parameters.cmakeBuildDir }}
-    cmakeArgs: '--build ${{ parameters.cmakeTargetDir }} --target ${{ parameters.cmakeTarget }}'
+- ${{ if eq(parameters.installEnabled, true) }}:
+  - task: CMake@1
+    displayName: '${{parameters.componentName }} ${{ parameters.cmakeTarget }}'
+    inputs:
+      workingDirectory: ${{ parameters.cmakeBuildDir }}
+      cmakeArgs: '--build ${{ parameters.cmakeTargetDir }} --target ${{ parameters.cmakeTarget }}'

.azuredevops/templates/steps/dependencies-cmake-latest.yml

@@ -0,0 +1,10 @@
+# replace cmake from apt install with newest version using snap install
+steps:
+- task: Bash@3
+  displayName: update cmake
+  inputs:
+    targetType: inline
+    script: |
+      sudo apt purge cmake
+      sudo snap install cmake --classic
+      hash -r

.azuredevops/templates/steps/dependencies-other.yml

@@ -12,23 +12,31 @@ steps:
   displayName: 'sudo apt-get update'
   inputs:
     targetType: inline
-    script: sudo apt-get update
+    script: sudo apt-get --yes update
+  env:
+    DEBIAN_FRONTEND: noninteractive
 - task: Bash@3
   displayName: 'sudo apt-get upgrade'
   inputs:
     targetType: inline
-    script: sudo apt-get update
+    script: sudo apt-get --yes upgrade
+  env:
+    DEBIAN_FRONTEND: noninteractive
 - task: Bash@3
   displayName: 'sudo apt-get fix'
   inputs:
     targetType: inline
     script: sudo apt --yes --fix-broken install
+  env:
+    DEBIAN_FRONTEND: noninteractive
 - ${{ if gt(length(parameters.aptPackages), 0) }}:
   - task: Bash@3
     displayName: 'sudo apt-get install ...'
     inputs:
       targetType: inline
       script: sudo apt-get --yes install ${{ join(' ', parameters.aptPackages) }}
+    env:
+      DEBIAN_FRONTEND: noninteractive
 - ${{ if gt(length(parameters.pipModules), 0) }}:
   - task: Bash@3
     displayName: 'pip install  ...'

.azuredevops/templates/steps/dependencies-rocm.yml

@@ -23,6 +23,7 @@ parameters:
 - name: stagingPipelineIdentifiers
   type: object
   default:
+    amdsmi: $(amdsmi-pipeline-id)
     aomp: $(aomp-pipeline-id)
     aomp-extras: $(aomp-extras-pipeline-id)
     AMDMIGraphX: $(amdmigraphx-pipeline-id)
@@ -30,13 +31,22 @@ parameters:
     composable_kernel: $(composable-kernel-pipeline-id)
     half: $(half-pipeline-id)
     hipBLAS: $(hipblas-pipeline-id)
+    hipCUB: $(hipcub-pipeline-id)
+    HIPIFY: $(hipify-pipeline-id)
     hipRAND: $(hiprand-pipeline-id)
+    hipSOLVER: $(hipsolver-pipeline-id)
     hipSPARSE: $(hipsparse-pipeline-id)
     llvm-project: $(llvm-project-pipeline-id)
     MIOpen: $(miopen-pipeline-id)
+    MIVisionX: $(mivisionx-pipeline-id)
+    rdc: $(rdc-pipeline-id)
     rocBLAS: $(rocblas-pipeline-id)
-    rocFFT: $(rotfft-pipeline-id)
+    ROCdbgapi : $(rocdbgapi-pipeline-id)
+    rocDecode: $(rocdecode-pipeline-id)
+    rocFFT: $(rocfft-pipeline-id)
     rocm-cmake: $(rocm-cmake-pipeline-id)
+    rocm-core: $(rocm-core-pipeline-id)
+    rocm_smi_lib: $(rocm-smi-lib-pipeline-id)
     rocminfo: $(rocminfo-pipeline-id)
     rocMLIR: $(rocmlir-pipeline-id)
     rocPRIM: $(rocprim-pipeline-id)
@@ -46,10 +56,13 @@ parameters:
     rocSOLVER: $(rocsolver-pipeline-id)
     rocSPARSE: $(rocsparse-pipeline-id)
     ROCT-Thunk-Interface: $(roct-thunk-interface-pipeline-id)
+    rocThrust: $(rocthrust-pipeline-id)
+    roctracer: $(roctracer-pipeline-id)
     rpp: $(rpp-pipeline-id)
 - name: taggedPipelineIdentifiers
   type: object
   default:
+    amdsmi: $(amdsmi-tagged-pipeline-id)
     aomp: $(aomp-tagged-pipeline-id)
     aomp-extras: $(aomp-extras-tagged-pipeline-id)
     AMDMIGraphX: $(amdmigraphx-tagged-pipeline-id)
@@ -57,13 +70,22 @@ parameters:
     composable_kernel: $(composable-kernel-tagged-pipeline-id)
     half: $(half-tagged-pipeline-id)
     hipBLAS: $(hipblas-tagged-pipeline-id)
+    hipCUB: $(hipcub-tagged-pipeline-id)
+    HIPIFY: $(hipify-tagged-pipeline-id)
     hipRAND: $(hiprand-tagged-pipeline-id)
+    hipSOLVER: $(hipsolver-tagged-pipeline-id)
     hipSPARSE: $(hipsparse-tagged-pipeline-id)
     llvm-project: $(llvm-project-tagged-pipeline-id)
     MIOpen: $(miopen-tagged-pipeline-id)
+    MIVisionX: $(mivisionx-tagged-pipeline-id)
+    rdc: $(rdc-tagged-pipeline-id)
     rocBLAS: $(rocblas-tagged-pipeline-id)
-    rocFFT: $(rotfft-tagged-pipeline-id)
+    ROCdbgapi : $(rocdbgapi-tagged-pipeline-id)
+    rocDecode: $(rocdecode-tagged-pipeline-id)
+    rocFFT: $(rocfft-tagged-pipeline-id)
     rocm-cmake: $(rocm-cmake-tagged-pipeline-id)
+    rocm-core: $(rocm-core-tagged-pipeline-id)
+    rocm_smi_lib: $(rocm-smi-lib-tagged-pipeline-id)
     rocminfo: $(rocminfo-tagged-pipeline-id)
     rocMLIR: $(rocmlir-tagged-pipeline-id)
     rocPRIM: $(rocprim-tagged-pipeline-id)
@@ -73,6 +95,8 @@ parameters:
     rocSOLVER: $(rocsolver-tagged-pipeline-id)
     rocSPARSE: $(rocsparse-tagged-pipeline-id)
     ROCT-Thunk-Interface: $(roct-thunk-interface-tagged-pipeline-id)
+    rocThrust: $(rocthrust-tagged-pipeline-id)
+    roctracer: $(roctracer-tagged-pipeline-id)
     rpp: $(rpp-tagged-pipeline-id)
 # set to true if you're calling this template file multiple files in same pipeline
 # only leave last call false to optimize sequence

.azuredevops/variables-global.yml

@@ -27,3 +27,5 @@ variables:
   value: rocm/dev-ubuntu-22.04
 - name: LATEST_DOCKER_VERSION
   value: 6.1
+- name: KEYRING_VERSION
+  value: 6.1

.gitignore

@@ -16,4 +16,4 @@ _readthedocs/
 docs/CHANGELOG.md
 docs/contribute/index.md
 docs/about/release-notes.md
-docs/about/CHANGELOG.md
+docs/about/changelog.md

.readthedocs.yaml

@@ -3,19 +3,20 @@
 
 version: 2
 
+sphinx:
+   configuration: docs/conf.py
+
+formats: [htmlzip]
+
+python:
+   install:
+   - requirements: docs/sphinx/requirements.txt
+
 build:
    os: ubuntu-22.04
    tools:
       python: "3.10"
    apt_packages:
      - "doxygen"
+     - "gfortran" # For pre-processing fortran sources
      - "graphviz" # For dot graphs in doxygen
-
-python:
-   install:
-   - requirements: docs/sphinx/requirements.txt
-
-sphinx:
-   configuration: docs/conf.py
-
-formats: []

.wordlist.txt

@@ -15,6 +15,7 @@ AOMP
 APIC
 APIs
 APU
+AQL
 ASIC
 ASICs
 ASan
@@ -62,6 +63,7 @@ CommonMark
 Concretized
 Conda
 ConnectX
+DENORM
 DGEMM
 DKMS
 DL
@@ -70,6 +72,7 @@ DNN
 DNNL
 DPM
 DRI
+DRM
 DW
 DWORD
 Dask
@@ -85,6 +88,7 @@ ELMo
 ENDPGM
 EPYC
 ESXi
+FFFFFFF
 FFT
 FFTs
 FFmpeg
@@ -122,6 +126,7 @@ GenAI
 GenZ
 GitHub
 Gitpod
+HBCC
 HBM
 HCA
 HIPCC
@@ -132,6 +137,7 @@ HPCG
 HPE
 HPL
 HSA
+HW
 HWE
 Haswell
 Higgs
@@ -157,11 +163,16 @@ Intra
 Ioffe
 JSON
 Jupyter
+KBytes
+KERNARG
 KFD
 KiB
+KMD
 KVM
 Keras
+Kernarg
 Khronos
+Ki
 LAPACK
 LCLK
 LDS
@@ -180,6 +191,7 @@ MiB
 MIGraphX
 MIOpen
 MIOpenGEMM
+MIPMAP
 MIVisionX
 MLM
 MMA
@@ -222,6 +234,7 @@ NousResearch's
 NumPy
 OAM
 OAMs
+OBJFILE
 OCP
 OEM
 OFED
@@ -237,6 +250,7 @@ OpenCV
 OpenFabrics
 OpenGL
 OpenMP
+OpenMPI
 OpenSSL
 OpenVX
 PCI
@@ -261,6 +275,7 @@ RCCL
 RDC
 RDMA
 RDNA
+RGP
 RHEL
 ROC
 ROCProfiler
@@ -274,6 +289,7 @@ ROCmCC
 ROCmSoftwarePlatform
 ROCmValidationSuite
 ROCr
+RPATH
 RST
 RW
 Radeon
@@ -302,10 +318,12 @@ SMEM
 SMI
 SMT
 SPI
+SQTT
 SQs
 SRAM
 SRAMECC
 SVD
+SVM
 SWE
 SerDes
 Shlens
@@ -343,6 +361,8 @@ UIF
 USM
 UTCL
 UTIL
+UNBUNDLER
+USWC
 Uncached
 Unhandled
 VALU
@@ -357,6 +377,8 @@ VSIX
 VSkipped
 Vanhoucke
 Vulkan
+WERROR
+WG
 WGP
 WGPs
 WX
@@ -388,6 +410,8 @@ allocator
 allocators
 amdgpu
 api
+arg
+args
 atmi
 atomics
 autogenerated
@@ -400,6 +424,7 @@ bfloat
 bilinear
 bitsandbytes
 blit
+bool
 boson
 bosons
 buildable
@@ -411,6 +436,10 @@ centos
 centric
 changelog
 chiplet
+clBuildProgram
+clCompileProgram
+clLinkProgram
+clr
 cmake
 cmd
 coalescable
@@ -426,6 +455,7 @@ convolutional
 convolves
 cpp
 csn
+cstring
 cuBLAS
 cuFFT
 cuLIB
@@ -443,6 +473,7 @@ deallocation
 denoise
 denoised
 denoises
+denorm
 denormalize
 deserializers
 detections
@@ -457,6 +488,7 @@ embeddings
 enablement
 endpgm
 encodings
+enqueue
 env
 epilog
 etcetera
@@ -480,7 +512,9 @@ heterogenous
 hipBLAS
 hipBLASLt
 hipCUB
+hipConfig
 hipFFT
+hipHostMalloc
 hipLIB
 hipRAND
 hipSOLVER
@@ -489,12 +523,15 @@ hipSPARSELt
 hipTensor
 hipamd
 hipblas
+hipcc
 hipcub
 hipfft
 hipfort
 hipify
+hiprtc
 hipsolver
 hipsparse
+hpc
 hpp
 hsa
 hsakmt
@@ -509,6 +546,7 @@ initializer
 inlining
 installable
 interprocedural
+interprocess
 intra
 invariants
 invocating
@@ -526,9 +564,12 @@ localscratch
 logits
 lossy
 macOS
+malloc
 matchers
+mem
 microarchitecture
 migraphx
+mipmap
 miopen
 miopengemm
 mivisionx
@@ -539,6 +580,8 @@ mvffr
 namespace
 namespaces
 numref
+nvcc
+nvidia
 ocl
 opencl
 opencv
@@ -559,6 +602,7 @@ prebuilt
 precompiled
 prefetch
 prefetchable
+prepinned
 preprocess
 preprocessed
 preprocessing
@@ -589,6 +633,7 @@ rocFFT
 rocLIB
 rocMLIR
 rocPRIM
+rocProfiler
 rocRAND
 rocSOLVER
 rocSPARSE
@@ -609,11 +654,13 @@ rocsolver
 rocsparse
 rocthrust
 roctracer
+rpath
 runtime
 runtimes
 sL
 scalability
 scalable
+sdma
 sendmsg
 serializers
 shader
@@ -624,13 +671,17 @@ smi
 softmax
 spack
 src
+stderr
 stochastically
 strided
+stubing
+suballocaitons
 subdirectory
 subexpression
 subfolder
 subfolders
 supercomputing
+td
 tensorfloat
 th
 tokenization
@@ -647,6 +698,8 @@ tqdm
 tracebacks
 txt
 uarch
+uint
+unbundler
 uncached
 uncorrectable
 uninstallation
@@ -677,12 +730,14 @@ wavefronts
 whitespaces
 workgroup
 workgroups
+workitems
 writeback
 writebacks
 wrreq
 wzo
 xargs
+xf
 xz
 yaml
 ysvmadyb
-zypper
+zypper

README.md

@@ -56,12 +56,103 @@ cd ~/ROCm/
 
 **Note:** Using this sample code will cause the repo tool to download the open source code associated with the specified ROCm release. Ensure that you have ssh-keys configured on your machine for your GitHub ID prior to the download as explained at [Connecting to GitHub with SSH](https://docs.github.com/en/authentication/connecting-to-github-with-ssh).
 
-### Building the ROCm source code
+## Building the ROCm source code
 
 Each ROCm component repository contains directions for building that component, such as the rocSPARSE documentation [Installation and Building for Linux](https://rocm.docs.amd.com/projects/rocSPARSE/en/latest/install/Linux_Install_Guide.html). Refer to the specific component documentation for instructions on building the repository.
 
 Each release of the ROCm software supports specific hardware and software configurations. Refer to [System requirements (Linux)](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/reference/system-requirements.html) for the current supported hardware and OS.
 
+## Build ROCm from source
+
+The Build will use as many processors as it can find to build in parallel. Some of the compiles can consume as much as 10GB of RAM, so make sure you have plenty of Swap Space !
+
+By default the ROCm build will compile for all supported GPU architectures and will take approximately 500 CPU hours.
+The Build time will reduce significantly if we limit the GPU Architecture/s against which we need to build by using the environment variable GPU_ARCHS as mentioned below.
+
+```bash
+# --------------------------------------
+# Step1: clone source code
+# --------------------------------------
+
+mkdir -p ~/WORKSPACE/      # Or any folder name other than WORKSPACE
+cd ~/WORKSPACE/
+export ROCM_VERSION=6.1.0   # or 6.1.1 6.1.2
+~/bin/repo init -u http://github.com/ROCm/ROCm.git -b roc-6.1.x -m tools/rocm-build/rocm-${ROCM_VERSION}.xml
+~/bin/repo sync
+
+# --------------------------------------
+# Step 2: Prepare build environment
+# --------------------------------------
+
+# Option 1: Start a docker container
+# Pulling required base docker images:
+# Ubuntu20.04 built from ROCm/tools/rocm-build/docker/ubuntu20/Dockerfile
+docker pull rocm/rocm-build-ubuntu-20.04:6.1
+# Ubuntu22.04 built from ROCm/tools/rocm-build/docker/ubuntu22/Dockerfile
+docker pull rocm/rocm-build-ubuntu-22.04:6.1
+
+# Start docker container and mount the source code folder:
+docker run -ti \
+    -e ROCM_VERSION=${ROCM_VERSION} \
+    -e CCACHE_DIR=$HOME/.ccache \
+    -e CCACHE_ENABLED=true \
+    -e DOCK_WORK_FOLD=/src \
+    -w /src \
+    -v $PWD:/src \
+    -v /etc/passwd:/etc/passwd \
+    -v /etc/shadow:/etc/shadow \
+    -v ${HOME}/.ccache:${HOME}/.ccache \
+    -u $(id -u):$(id -g) \
+    <replace_with_required_ubuntu_base_docker_image> bash
+
+# Option 2: Install required packages into the host machine
+# For ubuntu20.04 system
+cd ROCm/tools/rocm-build/docker/ubuntu20
+bash install-prerequisites.sh
+# For ubuntu22.04 system
+cd ROCm/tools/rocm-build/docker/ubuntu22
+bash install-prerequisities.sh
+
+# --------------------------------------
+# Step 3: Run build command line
+# --------------------------------------
+
+# Select GPU targets before building:
+# When GPU_ARCHS is not set, default GPU targets supported by ROCm6.1 will be used.
+# To build against a subset of GFX architectures you can use the below env variable.
+# Support MI300 (gfx940, gfx941, gfx942).
+export GPU_ARCHS="gfx942"               # Example
+export GPU_ARCHS="gfx940;gfx941;gfx942" # Example
+
+# Pick and run build commands in the docker container:
+# Build rocm-dev packages
+make -f ROCm/tools/rocm-build/ROCm.mk -j ${NPROC:-$(nproc)} rocm-dev
+# Build all ROCm packages
+make -f ROCm/tools/rocm-build/ROCm.mk -j ${NPROC:-$(nproc)} all
+# list all ROCm components to find required components
+make -f ROCm/tools/rocm-build/ROCm.mk list_components
+# Build a single ROCm packages
+make -f ROCm/tools/rocm-build/ROCm.mk T_rocblas
+
+# Find built packages in ubuntu20.04:
+out/ubuntu-20.04/20.04/deb/
+# Find built packages in ubuntu22.04:
+out/ubuntu-22.04/22.04/deb/
+
+# Find built logs in ubuntu20.04:
+out/ubuntu-20.04/20.04/logs/
+# Find built logs in ubuntu22.04:
+out/ubuntu-22.04/22.04/logs/
+# All logs pertaining to failed components, end with .errrors extension.
+out/ubuntu-22.04/22.04/logs/rocblas.errors          # Example
+# All logs pertaining to building components, end with .inprogress extension.
+out/ubuntu-22.04/22.04/logs/rocblas.inprogress  # Example
+# All logs pertaining to passed components, use the component names.
+out/ubuntu-22.04/22.04/logs/rocblas             # Example
+```
+
+Note: [Overview for ROCm.mk](tools/rocm-build/README.md)
+
 ## ROCm documentation
 
 This repository contains the [manifest file](https://gerrit.googlesource.com/git-repo/+/HEAD/docs/manifest-format.md)

RELEASE.md

@@ -1,4 +1,6 @@
-# ROCm 6.1.1 release notes
+# ROCm 6.1.2 release notes
+<!-- Do not edit this file! This file is autogenerated with -->
+<!--   tools/autotag/tag_script.py                          -->
 
 <!-- Disable lints since this is an auto-generated file.    -->
 <!-- markdownlint-disable blanks-around-headers             -->
@@ -9,155 +11,137 @@
 
 <!-- spellcheck-disable -->
 
-ROCm™ 6.1.1 introduces minor fixes and improvements to some tools and libraries.
+ROCm 6.1.2 includes enhancements to SMI tools and improvements to some libraries.
 
-## OS support
+### OS support
 
-* ROCm 6.1.1 now supports Oracle Linux. It has been tested against version 8.9 (kernel 5.15.0-205) with AMD Instinct MI300X accelerators.
+ROCm 6.1.2 has been tested against a pre-release version of Ubuntu 22.04.5 (kernel: 5.15 [GA], 6.8 [HWE]).
 
-* ROCm 6.1.1 has been tested against a pre-release version of Ubuntu 22.04.5 (kernel: 5.15 [GA], 6.8 [HWE]).
+### AMD SMI
 
-## AMD SMI
-
-AMD SMI for ROCm 6.1.1
-
-### Additions
-
-- Added deferred error correctable counts to `amd-smi metric -ecc -ecc-blocks`.
-
-### Changes
-
-- Updated the output of `amd-smi metric --ecc-blocks` to show counters available from blocks.
-- Updated the output of `amd-smi metric --clock` to reflect each engine.
-- Updated the output of `amd-smi topology --json` to align with output reported by host and guest systems.
-
-### Fixes
-
-- Fixed `amd-smi metric --clock`'s clock lock and deep sleep status.
-- Fixed an issue that would cause an error when resetting non-AMD GPUs.
-- Fixed `amd-smi metric --pcie` and `amdsmi_get_pcie_info()` when using RDNA3 (Navi 32 and Navi 31) hardware to prevent "UNKNOWN" reports.
-- Fixed the output results of `amd-smi process` when getting processes running on a device.
-
-### Removals
-
-- Removed the `amdsmi_get_gpu_process_info` API from the Python library. It was removed from the C library in an earlier release.
-
-### Known issues
-
-- `amd-smi bad-pages` can result in a `ValueError: Null pointer access` error when using some PMU firmware versions.
-
-```{note}
-See the [detailed changelog](https://github.com/ROCm/amdsmi/blob/docs/6.1.1/CHANGELOG.md) with code samples for more information.
-```
-
-## HIPCC
-
-HIPCC for ROCm 6.1.1
-
-### Changes
-
-- **Upcoming:** a future release will enable use of compiled binaries `hipcc.bin` and `hipconfig.bin` by default. No action is needed by users. You can continue calling high-level Perl scripts `hipcc` and `hipconfig`. `hipcc.bin` and `hipconfig.bin` will be invoked by the high-level Perl scripts. To revert to the previous behavior and invoke `hipcc.pl` and `hipconfig.pl`, set the `HIP_USE_PERL_SCRIPTS` environment variable to `1`.
-- **Upcoming:** a subsequent release will remove high-level Perl scripts `hipcc` and `hipconfig`. This release will remove the `HIP_USE_PERL_SCRIPTS` environment variable. It will rename `hipcc.bin` and `hipconfig.bin` to `hipcc` and `hipconfig` respectively. No action is needed by the users. To revert to the previous behavior, invoke `hipcc.pl` and `hipconfig.pl` explicitly.
-- **Upcoming:** a subsequent release will remove `hipcc.pl` and `hipconfig.pl`.
-
-## ROCm SMI
-
-ROCm SMI for ROCm 6.1.1
-
-### Additions
-
-* Added the capability to unlock mutex when a process is dead. Added related debug output.
-* Added the `Partition ID` field to the `rocm-smi` CLI.
-* Added `NODE`, `GUID`, and `GFX Version` fields to the CLI.
-* Documentation now includes C++ and Python tutorials, API guides, and reference material.
-
-### Changes
-
-* Some `rocm-smi` fields now display `N/A` instead of `unknown/unsupported` for consistency.
-* Changed stacked ID formatting in the `rocm-smi` CLI to make it easier to spot identifiers.
-
-### Fixes
-
-* Fixed HIP and ROCm SMI mismatch on GPU bus assignments.
-* Fixed memory leaks caused by not closing directories and creating maps nodes instead of using `.at()`.
-* Fixed initializing calls which reuse `rocmsmi.initializeRsmi()` bindings in the `rocmsmi` Python API.
-* Fixed an issue causing `rsmi_dev_activity_metric_get` gfx/memory to not update with GPU activity.
-
-### Known issues
-
-- ROCm SMI reports GPU utilization incorrectly for RDNA3 GPUs in some situations. See the issue on [GitHub](https://github.com/ROCm/ROCm/issues/3112).
-
-```{note}
-See the [detailed ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/docs/6.1.1/CHANGELOG.md) with code samples for more information.
-```
-
-## Library changes in ROCm 6.1.1
-
-| Library     | Version                                                                    |
-| ----------- | -------------------------------------------------------------------------- |
-| AMDMIGraphX | [2.9](https://github.com/ROCm/AMDMIGraphX/releases/tag/rocm-6.1.1)         |
-| hipBLAS     | [2.1.0](https://github.com/ROCm/hipBLAS/releases/tag/rocm-6.1.1)           |
-| hipBLASLt   | [0.7.0](https://github.com/ROCm/hipBLASLt/releases/tag/rocm-6.1.1)         |
-| hipCUB      | [3.1.0](https://github.com/ROCm/hipCUB/releases/tag/rocm-6.1.1)            |
-| hipFFT      | [1.0.14](https://github.com/ROCm/hipFFT/releases/tag/rocm-6.1.1)           |
-| hipRAND     | [2.10.17](https://github.com/ROCm/hipRAND/releases/tag/rocm-6.1.1)         |
-| hipSOLVER   | 2.1.0 ⇒ [2.1.1](https://github.com/ROCm/hipSOLVER/releases/tag/rocm-6.1.1) |
-| hipSPARSE   | [3.0.1](https://github.com/ROCm/hipSPARSE/releases/tag/rocm-6.1.1)         |
-| hipSPARSELt | [0.2.0](https://github.com/ROCm/hipSPARSELt/releases/tag/rocm-6.1.1)       |
-| hipTensor   | [1.2.0](https://github.com/ROCm/hipTensor/releases/tag/rocm-6.1.1)         |
-| MIOpen      | [3.1.0](https://github.com/ROCm/MIOpen/releases/tag/rocm-6.1.1)            |
-| MIVisionX   | [2.5.0](https://github.com/ROCm/MIVisionX/releases/tag/rocm-6.1.1)         |
-| rccl        | [2.18.6](https://github.com/ROCm/rccl/releases/tag/rocm-6.1.1)             |
-| rocALUTION  | [3.1.1](https://github.com/ROCm/rocALUTION/releases/tag/rocm-6.1.1)        |
-| rocBLAS     | [4.1.0](https://github.com/ROCm/rocBLAS/releases/tag/rocm-6.1.1)           |
-| rocDecode   | [0.5.0](https://github.com/ROCm/rocDecode/releases/tag/rocm-6.1.1)         |
-| rocFFT      | 1.0.26 ⇒ [1.0.27](https://github.com/ROCm/rocFFT/releases/tag/rocm-6.1.1)  |
-| rocm-cmake  | [0.12.0](https://github.com/ROCm/rocm-cmake/releases/tag/rocm-6.1.1)       |
-| rocPRIM     | [3.1.0](https://github.com/ROCm/rocPRIM/releases/tag/rocm-6.1.1)           |
-| rocRAND     | [3.0.1](https://github.com/ROCm/rocRAND/releases/tag/rocm-6.1.1)           |
-| rocSOLVER   | [3.25.0](https://github.com/ROCm/rocSOLVER/releases/tag/rocm-6.1.1)        |
-| rocSPARSE   | [3.1.2](https://github.com/ROCm/rocSPARSE/releases/tag/rocm-6.1.1)         |
-| rocThrust   | [3.0.1](https://github.com/ROCm/rocThrust/releases/tag/rocm-6.1.1)         |
-| rocWMMA     | [1.4.0](https://github.com/ROCm/rocWMMA/releases/tag/rocm-6.1.1)           |
-| rpp         | [1.5.0](https://github.com/ROCm/rpp/releases/tag/rocm-6.1.1)               |
-| Tensile     | [4.40.0](https://github.com/ROCm/Tensile/releases/tag/rocm-6.1.1)          |
-
-### hipBLASLt 0.7.0
-
-hipBLASLt 0.7.0 for ROCm 6.1.1
+AMD SMI for ROCm 6.1.2
 
 #### Additions
 
-- Added `hipblasltExtSoftmax` extension API.
-- Added `hipblasltExtLayerNorm` extension API.
-- Added `hipblasltExtAMax` extension API.
-- Added `GemmTuning` extension parameter to set split-k by user.
-- Added support for mixed precision datatype: fp16/fp8 in with fp16 outk.
+* Added process isolation and clean shader APIs and CLI commands.
+  * `amdsmi_get_gpu_process_isolation()`
+  * `amdsmi_set_gpu_process_isolation()`
+  * `amdsmi_set_gpu_clear_sram_data()`
+* Added the `MIN_POWER` metric to output provided by `amd-smi static --limit`.
 
-#### Deprecations
+#### Optimizations
 
-- **Upcoming**: `algoGetHeuristic()` ext API for GroupGemm will be deprecated in a future release of hipBLASLt.
-
-### hipSOLVER 2.1.1
-
-hipSOLVER 2.1.1 for ROCm 6.1.1
+* Updated the `amd-smi monitor --pcie` output to prevent delays with the `monitor` command.
 
 #### Changes
 
-- By default, `BUILD_WITH_SPARSE` is now set to OFF on Microsoft Windows.
+* Updated `amismi_get_power_cap_info` to return values in uW instead of W.
+* Updated Python library return types for `amdsmi_get_gpu_memory_reserved_pages` and `amdsmi_get_gpu_bad_page_info`.
+* Updated the output of `amd-smi metric --ecc-blocks` to show counters available from blocks.
 
 #### Fixes
 
-- Fixed benchmark client build when `BUILD_WITH_SPARSE` is OFF.
+* `amdsmi_get_gpu_board_info()` no longer returns junk character strings.
+* `amd-smi metric --power` now correctly details power output for RDNA3, RDNA2, and MI1x devices.
+* Fixed the `amdsmitstReadWrite.TestPowerCapReadWrite` test for RDNA3, RDNA2, and MI100 devices.
+* Fixed an issue with the `amdsmi_get_gpu_memory_reserved_pages` and `amdsmi_get_gpu_bad_page_info` Python interface calls.
 
-### rocFFT 1.0.27
+#### Removals
 
-rocFFT 1.0.27 for ROCm 6.1.1
+* Removed the `amdsmi_get_gpu_process_info` API from the Python library. It was removed from the C library in an earlier release.
+
+```{note}
+See the AMD SMI [detailed changelog](https://github.com/ROCm/amdsmi/blob/rocm-6.1.x/CHANGELOG.md) with code samples for more information.
+```
+
+### ROCm SMI
+
+ROCm SMI for ROCm 6.1.2
 
 #### Additions
 
-- Enable multi-GPU testing on systems without direct GPU-interconnects.
+* Added the ring hang event to the `amdsmi_evt_notification_type_t` enum.
 
 #### Fixes
 
-- Fixed kernel launch failure on execute of very large odd-length real-complex transforms.
+* Fixed an issue causing ROCm SMI to incorrectly report GPU utilization for RDNA3 GPUs. See the issue on [GitHub](https://github.com/ROCm/ROCm/issues/3112).
+* Fixed the parsing of `pp_od_clk_voltage` in `get_od_clk_volt_info` to work better with MI-series hardware.
+
+## Library changes in ROCm 6.1.2
+
+| Library | Version |
+|---------|---------|
+| AMDMIGraphX | [2.9](https://github.com/ROCm/AMDMIGraphX/releases/tag/rocm-6.1.2) |
+| composable_kernel | [0.2.0](https://github.com/ROCm/composable_kernel/releases/tag/rocm-6.1.2) |
+| hipBLAS | [2.1.0](https://github.com/ROCm/hipBLAS/releases/tag/rocm-6.1.2) |
+| hipBLASLt | [0.7.0](https://github.com/ROCm/hipBLASLt/releases/tag/rocm-6.1.2) |
+| hipCUB | [3.1.0](https://github.com/ROCm/hipCUB/releases/tag/rocm-6.1.2) |
+| hipFFT | [1.0.14](https://github.com/ROCm/hipFFT/releases/tag/rocm-6.1.2) |
+| hipRAND | [2.10.17](https://github.com/ROCm/hipRAND/releases/tag/rocm-6.1.2) |
+| hipSOLVER | [2.1.1](https://github.com/ROCm/hipSOLVER/releases/tag/rocm-6.1.2) |
+| hipSPARSE | [3.0.1](https://github.com/ROCm/hipSPARSE/releases/tag/rocm-6.1.2) |
+| hipSPARSELt | [0.2.0](https://github.com/ROCm/hipSPARSELt/releases/tag/rocm-6.1.2) |
+| hipTensor | [1.2.0](https://github.com/ROCm/hipTensor/releases/tag/rocm-6.1.2) |
+| MIOpen | [3.1.0](https://github.com/ROCm/MIOpen/releases/tag/rocm-6.1.2) |
+| MIVisionX | [2.5.0](https://github.com/ROCm/MIVisionX/releases/tag/rocm-6.1.2) |
+| rccl | [2.18.6](https://github.com/ROCm/rccl/releases/tag/rocm-6.1.2) |
+| rocALUTION | [3.1.1](https://github.com/ROCm/rocALUTION/releases/tag/rocm-6.1.2) |
+| rocBLAS | 4.1.0 ⇒ [4.1.2](https://github.com/ROCm/rocBLAS/releases/tag/rocm-6.1.2) |
+| rocDecode | 0.5.0 ⇒ [0.6.0](https://github.com/ROCm/rocDecode/releases/tag/rocm-6.1.2) |
+| rocFFT | [1.0.27](https://github.com/ROCm/rocFFT/releases/tag/rocm-6.1.2) |
+| rocm-cmake | [0.12.0](https://github.com/ROCm/rocm-cmake/releases/tag/rocm-6.1.2) |
+| rocPRIM | [3.1.0](https://github.com/ROCm/rocPRIM/releases/tag/rocm-6.1.2) |
+| rocRAND | [3.0.1](https://github.com/ROCm/rocRAND/releases/tag/rocm-6.1.2) |
+| rocSOLVER | [3.25.0](https://github.com/ROCm/rocSOLVER/releases/tag/rocm-6.1.2) |
+| rocSPARSE | [3.1.2](https://github.com/ROCm/rocSPARSE/releases/tag/rocm-6.1.2) |
+| rocThrust | [3.0.1](https://github.com/ROCm/rocThrust/releases/tag/rocm-6.1.2) |
+| rocWMMA | [1.4.0](https://github.com/ROCm/rocWMMA/releases/tag/rocm-6.1.2) |
+| rpp | [1.5.0](https://github.com/ROCm/rpp/releases/tag/rocm-6.1.2) |
+| Tensile | [4.40.0](https://github.com/ROCm/Tensile/releases/tag/rocm-6.1.2) |
+
+### RCCL
+
+RCCL 2.18.6 for ROCm 6.1.2
+
+#### Changes
+
+* Reduced `NCCL_TOPO_MAX_NODES` to limit stack usage and avoid stack overflow.
+
+### rocBLAS
+
+rocBLAS 4.1.2 for ROCm 6.1.2
+
+#### Optimizations
+
+* Tuned BBS TN and TT operations on the CDNA3 architecture.
+
+#### Fixes
+
+* Fixed an issue related to obtaining solutions for BF16 TT operations.
+
+### rocDecode
+
+rocDecode 0.6.0 for ROCm 6.1.2
+
+#### Additions
+
+* Added support for FFmpeg v5.x.
+
+#### Optimizations
+
+* Updated error checking in the `rocDecode-setup.py` script.
+
+#### Changes
+
+* Updated core dependencies.
+* Updated to support the use of public LibVA headers.
+
+#### Fixes
+
+* Fixed some package dependencies.
+
+## Upcoming changes
+
+* A future release will enable the use of HIPCC compiled binaries `hipcc.bin` and `hipconfig.bin` by default. No action is needed by users; you may continue calling high-level Perl scripts `hipcc` and `hipconfig`. `hipcc.bin` and `hipconfig.bin` will be invoked by the high-level Perl scripts. To revert to the previous behavior and invoke `hipcc.pl` and `hipconfig.pl`, set the `HIP_USE_PERL_SCRIPTS` environment variable to `1`.
+* A subsequent release will remove high-level HIPCC Perl scripts from `hipcc` and `hipconfig`. This release will remove the `HIP_USE_PERL_SCRIPTS` environment variable. It will rename `hipcc.bin` and `hipconfig.bin` to `hipcc` and `hipconfig` respectively. No action is needed by the users. To revert to the previous behavior, invoke `hipcc.pl` and `hipconfig.pl` explicitly.
+* A subsequent release will remove `hipcc.pl` and `hipconfig.pl` for HIPCC.

default.xml

@@ -1,12 +1,11 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <manifest>
     <remote name="rocm-org" fetch="https://github.com/ROCm/" />
-    <default revision="refs/tags/rocm-6.1.1"
+    <default revision="refs/tags/rocm-6.1.2"
      remote="rocm-org"
      sync-c="true"
      sync-j="4" />
 <!--list of projects for ROCm-->
-    <project path="ROCm-OpenCL-Runtime/api/opencl/khronos/icd" name="OpenCL-ICD-Loader" remote="KhronosGroup" />
     <project name="ROCK-Kernel-Driver" />
     <project name="ROCR-Runtime" />
     <project name="ROCT-Thunk-Interface" />

docs/about/compatibility/openmp.md

@@ -77,8 +77,7 @@ Obtain the value of `gpu-arch` by running the following command:
 
 [//]: # (dated link below, needs updating)
 
-See the complete list of compiler command-line references
-[here](https://github.com/ROCm/llvm-project/blob/amd-stg-open/clang/docs/CommandGuide/clang.rst).
+See the complete list of [compiler command-line references](https://github.com/ROCm/llvm-project/blob/amd-staging/openmp/docs/CommandLineArgumentReference.rst).
 
 ### Using `rocprof` with OpenMP
 

docs/conceptual/gpu-arch/mi250.md

@@ -33,8 +33,8 @@ Units (CU). The MI250 GCD has 104 active CUs. Each compute unit is further
 subdivided into four SIMD units that process SIMD instructions of 16 data
 elements per instruction (for the FP64 data type). This enables the CU to
 process 64 work items (a so-called “wavefront”) at a peak clock frequency of 1.7
-GHz. Therefore, the theoretical maximum FP64 peak performance per GCD is 22.6
-TFLOPS for vector instructions. This equates to 45.3 TFLOPS for vector instructions for both GCDs together. The MI250 compute units also provide specialized
+GHz. Therefore, the theoretical maximum FP64 peak performance per GCD is 45.3
+TFLOPS for vector instructions. The MI250 compute units also provide specialized
 execution units (also called matrix cores), which are geared toward executing
 matrix operations like matrix-matrix multiplications. For FP64, the peak
 performance of these units amounts to 90.5 TFLOPS.

docs/conceptual/using-gpu-sanitizer.md

@@ -424,8 +424,4 @@ Shadow byte legend (one shadow byte represents 8 application bytes):
 
 * Lack of detection on the GPU might also be due to the implementation not instrumenting accesses to all GPU specific address spaces. For example, in the current implementation accesses to "private" or "stack" variables on the GPU are not instrumented, and accesses to HIP shared variables (also known as "local data store" or "LDS") are also not instrumented.
 
-* It can also be the case that a memory fault is reported for an invalid address even with the instrumentation. This is usually caused by the invalid address being so wild that its shadow address is outside any memory region, and the fault actually occurs on the access to the shadow address. It is also possible to hit a memory fault for the `NULL` pointer. While address 0 does have a shadow location, it is not poisoned by the runtime.
-
-* There is currently a bug which can result in memory faults being reported when running instrumented device code which makes use of `malloc`, `free`, `new`, or `delete`.
-
-* There is currently a bug which can result in undefined symbols being reported at compile time when instrumented device code makes use of `new` and `delete`.
+* It can also be the case that a memory fault is hit for an invalid address even with the instrumentation. This is usually caused by the invalid address being so wild that its shadow address is outside any memory region, and the fault actually occurs on the access to the shadow address. It is also possible to hit a memory fault for the `NULL` pointer. While address 0 does have a shadow location, it is not poisoned by the runtime.

docs/conf.py

@@ -23,7 +23,7 @@ for template in templates:
 
 shutil.copy2('../RELEASE.md','./about/release-notes.md')
 # Keep capitalization due to similar linking on GitHub's markdown preview.
-shutil.copy2('../CHANGELOG.md','./about/CHANGELOG.md')
+shutil.copy2('../CHANGELOG.md','./about/changelog.md')
 
 latex_engine = "xelatex"
 latex_elements = {
@@ -38,8 +38,8 @@ latex_elements = {
 project = "ROCm Documentation"
 author = "Advanced Micro Devices, Inc."
 copyright = "Copyright (c) 2024 Advanced Micro Devices, Inc. All rights reserved."
-version = "6.1.1"
-release = "6.1.1"
+version = "6.1.2"
+release = "6.1.2"
 setting_all_article_info = True
 all_article_info_os = ["linux", "windows"]
 all_article_info_author = ""
@@ -49,60 +49,18 @@ article_pages = [
     {
         "file":"about/release-notes",
         "os":["linux", "windows"],
-        "date":"2024-05-08"
+        "date":"2024-06-04"
     },
     {
-        "file":"about/CHANGELOG",
+        "file":"about/changelog",
         "os":["linux", "windows"],
-        "date":"2024-05-08"
+        "date":"2024-06-04"
     },
 
-    {"file":"install/windows/install-quick", "os":["windows"]},
-    {"file":"install/linux/install-quick", "os":["linux"]},
-
-    {"file":"install/linux/install", "os":["linux"]},
-    {"file":"install/linux/install-options", "os":["linux"]},
-    {"file":"install/linux/prerequisites", "os":["linux"]},
-
-    {"file":"install/docker", "os":["linux"]},
-    {"file":"install/magma-install", "os":["linux"]},
-    {"file":"install/pytorch-install", "os":["linux"]},
-    {"file":"install/tensorflow-install", "os":["linux"]},
-
-    {"file":"install/windows/install", "os":["windows"]},
-    {"file":"install/windows/prerequisites", "os":["windows"]},
-    {"file":"install/windows/cli/index", "os":["windows"]},
-    {"file":"install/windows/gui/index", "os":["windows"]},
-
-    {"file":"about/compatibility/docker-image-support-matrix", "os":["linux"]},
-    {"file":"about/compatibility/user-kernel-space-compat-matrix", "os":["linux"]},
-
-    {"file":"reference/library-index", "os":["linux"]},
-
     {"file":"how-to/deep-learning-rocm", "os":["linux"]},
     {"file":"how-to/gpu-enabled-mpi", "os":["linux"]},
     {"file":"how-to/system-debugging", "os":["linux"]},
     {"file":"how-to/tuning-guides", "os":["linux", "windows"]},
-
-    {"file":"how-to/rocm-for-ai/index", "os":["linux"]},
-    {"file":"how-to/rocm-for-ai/install", "os":["linux"]},
-    {"file":"how-to/rocm-for-ai/train-a-model", "os":["linux"]},
-    {"file":"how-to/rocm-for-ai/deploy-your-model", "os":["linux"]},
-    {"file":"how-to/rocm-for-ai/hugging-face-models", "os":["linux"]},
-
-    {"file":"how-to/rocm-for-hpc/index", "os":["linux"]},
-
-    {"file":"how-to/llm-fine-tuning-optimization/index", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/overview", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/fine-tuning-and-inference", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/single-gpu-fine-tuning-and-inference", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/multi-gpu-fine-tuning-and-inference", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/llm-inference-frameworks", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/model-acceleration-libraries", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/model-quantization", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/optimizing-with-composable-kernel", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/optimizing-triton-kernel", "os":["linux"]},
-    {"file":"how-to/llm-fine-tuning-optimization/profiling-and-debugging", "os":["linux"]},
 ]
 
 exclude_patterns = ['temp']
@@ -126,5 +84,5 @@ html_theme_options = {
 }
 
 redirects = {
-     "reference/openmp/openmp": "../../about/compatibility/openmp.html"
+    "reference/openmp/openmp": "../../about/compatibility/openmp.html"
 }

docs/contribute/feedback.md

@@ -12,7 +12,8 @@ There are four standard ways to provide feedback on this repository.
 
 All contributions to ROCm documentation should arrive via the
 [GitHub Flow](https://docs.github.com/en/get-started/quickstart/github-flow)
-targeting the develop branch of the repository.
+targeting the develop branch of the repository. If you are unable to contribute
+via the GitHub Flow, feel free to email us at [rocm-feedback@amd.com](mailto:rocm-feedback@amd.com?subject=Documentation%20Feedback).
 
 For more in-depth information on creating a pull request (PR), see
 [Contributing](./contributing.md).
@@ -29,3 +30,7 @@ and follow along on via public announcements.
 
 Issues on existing or absent documentation can be filed in
 [GitHub Issues](https://github.com/ROCm/ROCm/issues).
+
+## Email
+
+Send other feedback or questions to [rocm-feedback@amd.com](mailto:rocm-feedback@amd.com?subject=Documentation%20Feedback).

docs/how-to/deep-learning-rocm.rst

@@ -8,44 +8,14 @@ Installing deep learning frameworks for ROCm
 
 ROCm provides a comprehensive ecosystem for deep learning development, including
 :ref:`libraries <artificial-intelligence-apis>` for optimized deep learning operations and ROCm-aware versions of popular
-deep learning frameworks and libraries such as PyTorch, TensorFlow, JAX, and MAGMA. ROCm works closely with these
+deep learning frameworks and libraries such as PyTorch, TensorFlow, and JAX. ROCm works closely with these
 frameworks to ensure that framework-specific optimizations take advantage of AMD accelerator and GPU architectures.
 
 The following guides cover installation processes for ROCm-aware deep learning frameworks.
 
-.. grid::
-
-   .. grid-item::
-      :columns: 3
-
-      :doc:`PyTorch for ROCm <rocm-install-on-linux:how-to/3rd-party/pytorch-install>`
-
-   .. grid-item::
-      :columns: 3
-
-      :doc:`TensorFlow for ROCm <rocm-install-on-linux:how-to/3rd-party/tensorflow-install>`
-
-   .. grid-item::
-      :columns: 3
-
-   .. grid-item::
-      :columns: 3
-
-   .. grid-item::
-      :columns: 3
-
-      :doc:`JAX for ROCm <rocm-install-on-linux:how-to/3rd-party/jax-install>`
-
-   .. grid-item::
-      :columns: 3
-
-      :doc:`MAGMA for ROCm <rocm-install-on-linux:how-to/3rd-party/magma-install>`
-
-   .. grid-item::
-      :columns: 3
-
-   .. grid-item::
-      :columns: 3
+* :doc:`PyTorch for ROCm <rocm-install-on-linux:how-to/3rd-party/pytorch-install>`
+* :doc:`TensorFlow for ROCm <rocm-install-on-linux:how-to/3rd-party/tensorflow-install>`
+* :doc:`JAX for ROCm <rocm-install-on-linux:how-to/3rd-party/jax-install>`
 
 The following chart steps through typical installation workflows for installing deep learning frameworks for ROCm.
 

docs/how-to/llm-fine-tuning-optimization/model-quantization.rst

@@ -154,13 +154,13 @@ kernels by configuring the ``exllama_config`` parameter as the following.
 .. code-block:: python
 
    from transformers import AutoModelForCausalLM, GPTQConfig
-   pretrained_model_dir = "meta-llama/Llama-2-7b"
-   gptq_config = GPTQConfig(bits=4, exllama_config={"version":2})
+   #pretrained_model_dir = "meta-llama/Llama-2-7b"
+   base_model_name = "NousResearch/Llama-2-7b-hf"
+   gptq_config = GPTQConfig(bits=4, dataset="c4", exllama_config={"version":2})
    quantized_model = AutoModelForCausalLM.from_pretrained(
-                           base_model_name, 
-                           device_map="auto", 
+                           base_model_name,
+                           device_map="auto",
                            quantization_config=gptq_config)
-
 bitsandbytes
 ============
 

docs/how-to/rocm-for-ai/index.rst

@@ -21,3 +21,6 @@ In this guide, you'll learn about:
 - :doc:`Running models from Hugging Face <hugging-face-models>`
 
 - :doc:`Deploying your model <deploy-your-model>`
+
+To learn about ROCm for HPC applications and scientific computing, see
+:doc:`../rocm-for-hpc/index`.

docs/how-to/rocm-for-ai/train-a-model.rst

@@ -137,4 +137,4 @@ The following developer blogs showcase examples of how to fine-tune a model on a
 * Recipes for fine-tuning Llama2 and 3 with ``llama-recipes``
 
   * `meta-llama/llama-recipes: Scripts for fine-tuning Meta Llama3 with composable FSDP & PEFT methods to cover
-    single/multi-node GPUs <https://github.com/meta-llama/llama-recipes/tree/main/recipes/quickstart/finetuning>`_
+    single/multi-node GPUs <https://github.com/meta-llama/llama-recipes/tree/main/recipes/finetuning>`_

docs/how-to/rocm-for-hpc/index.rst

@@ -0,0 +1,231 @@
+.. meta::
+   :description: How to use ROCm for HPC
+   :keywords: ROCm, AI, high performance computing, HPC
+
+******************
+Using ROCm for HPC
+******************
+
+The ROCm open-source software stack is optimized to extract high-performance
+computing (HPC) workload performance from AMD Instinct™ accelerators
+while maintaining compatibility with industry software frameworks.
+
+ROCm enhances support and access for developers by providing streamlined and
+improved tools that significantly increase productivity. Being open-source, ROCm
+fosters innovation, differentiation, and collaboration within the developer
+community, making it a powerful and accessible solution for leveraging the full
+potential of AMD accelerators' capabilities in diverse computational
+applications.
+
+* For more information, see :doc:`What is ROCm? <../../what-is-rocm>`.
+
+* For guidance on installing ROCm, see :doc:`rocm-install-on-linux:index`. See
+  the :doc:`../../compatibility/compatibility-matrix` for details on hardware
+  and operating system support.
+
+Some of the most popular HPC frameworks are part of the ROCm platform, including
+those to help parallelize operations across multiple accelerators and servers,
+handle memory hierarchies, and solve linear systems.
+
+.. image:: ../../data/how-to/rocm-for-hpc/hpc-stack-2024_6_20.png
+   :align: center
+   :alt: Software and hardware ecosystem surrounding ROCm and AMD Instinct for HPC
+
+The following catalog of GPU-accelerated solutions includes a vast set of
+platform-compatible HPC applications, including those for astrophysics, climate 
+and weather, computational chemistry, computational fluid dynamics, earth
+science, genomics, geophysics, molecular dynamics, and physics computing.
+
+Refer to the resources in the following table for instructions on building,
+running, and deploying these applications on ROCm-capable systems with AMD
+Instinct accelerators. Each build container provides parameters to specify
+different source code branches, release versions of ROCm, OpenMPI, UCX, and
+Ubuntu versions.
+
+.. _hpc-apps:
+
+..
+   Reduce font size of HPC app descriptions slightly.
+
+.. raw:: html
+
+   <style>
+     #hpc-apps-table tr td:last-child {
+       font-size: 0.9rem;
+     }
+   </style>
+
+.. container::
+   :name: hpc-apps-table
+
+   .. list-table::
+      :header-rows: 1
+      :stub-columns: 1
+      :widths: 2 2 5
+
+      * - Application domain
+        - HPC application
+        - Description
+
+      * - Physics
+        - `Chroma <https://github.com/amd/InfinityHub-CI/tree/main/chroma/>`_
+        - The Chroma package supports data-parallel programming constructs for lattice
+          field theory and in particular lattice QCD. It uses the SciDAC QDP++ data-parallel
+          programming (in C++) that presents a single high-level code image to the user,
+          but can generate highly optimized code for many architectural systems including
+          single node workstations, multi and many-core nodes, clusters of nodes via
+          QMP, and classic vector computers.
+
+      * -
+        - `Grid <https://github.com/amd/InfinityHub-CI/tree/main/grid/>`_
+        - Grid is a library for lattice QCD calculations that employs a high-level data parallel
+          approach while using a number of techniques to target multiple types of parallelism.
+          The library currently supports MPI, OpenMP and short vector parallelism. The SIMD
+          instructions sets covered include SSE, AVX, AVX2, FMA4, IMCI and AVX512. Recent
+          releases expanded this support to include GPU offloading.
+
+      * -
+        - `MILC <https://github.com/amd/InfinityHub-CI/tree/main/milc/>`_
+        - The MILC Code is a set of research codes developed by MIMD Lattice Computation
+          (MILC) collaboration for doing simulations of four dimensional SU(3) lattice gauge
+          theory on MIMD parallel machines scaling from single-processor workstations
+          to HPC systems. The MILC Code is publicly available for research purposes.
+          Publications of work done using this code or derivatives of this code should
+          acknowledge this use.
+
+      * -
+        - `PIConGPU <https://github.com/amd/InfinityHub-CI/tree/main/picongpu>`_
+        - PIConGPU (Particle-in-cell on Graphics Processing Units) is an Open Source
+          simulations framework for plasma and laser-plasma physics used to develop
+          advanced particle accelerators for radiation therapy of cancer, high energy
+          physics and photon science.
+
+      * - Astrophysics
+        - `Cholla <https://github.com/amd/InfinityHub-CI/tree/main/cholla/>`_
+        - An astrophysical simulation code developed for the extreme environments
+          encountered in astrophysical systems.
+
+      * - Geophysics
+        - `SPECFEM3D Cartesian <https://github.com/amd/InfinityHub-CI/tree/main/specfem3d>`_
+        - SPECFEM3D Cartesian simulates acoustic (fluid), elastic (solid), coupled
+          acoustic/elastic, poroelastic or seismic wave propagation in any type of
+          conforming mesh of hexahedra (structured or not.) It can, for instance,
+          model seismic waves propagating in sedimentary basins or any other
+          regional geological model following earthquakes. It can also be used
+          for non-destructive testing or for ocean acoustics.
+
+      * - Molecular dynamics
+        - `GROMACS with HIP (AMD implementation) <https://github.com/amd/InfinityHub-CI/tree/main/gromacs>`_
+        - GROMACS is a versatile package to perform molecular dynamics, i.e.
+          simulate the Newtonian equations of motion for systems with hundreds
+          to millions of particles. This AMD container is based on a released
+          version of GROMACS modified by AMD. This container only supports up
+          to a 8 GPU configuration
+
+      * -
+        - `LAMMPS <https://github.com/amd/InfinityHub-CI/tree/main/lammps>`_
+        - LAMMPS is a classical molecular dynamics code with a focus on materials
+          modeling. It's an acronym for Large-scale Atomic/Molecular Massively
+          Parallel Simulator.
+
+      * - Computational fluid dynamics
+        - `NEKO <https://github.com/amd/InfinityHub-CI/tree/main/neko>`_
+        - Neko is a portable framework for high-order spectral element flow simulations.
+          Written in modern Fortran, Neko adopts an object-oriented approach, allowing
+          multi-tier abstractions of the solver stack and facilitating various hardware
+          backends ranging from general-purpose processors, CUDA and HIP enabled
+          accelerators to SX-Aurora vector processors.
+
+      * -
+        - `nekRS <https://github.com/amd/InfinityHub-CI/tree/main/nekrs>`_
+        - nekRS is an open-source Navier Stokes solver based on the spectral element
+          method targeting classical processors and accelerators like GPUs. 
+
+      * - Computational chemistry
+        - `QUDA <https://github.com/amd/InfinityHub-CI/tree/main/quda>`_
+        - Library designed for efficient lattice QCD computations on
+          accelerators. It includes optimized Dirac operators and a variety of
+          fermion solvers and conjugate gradient (CG) implementations, enhancing
+          performance and accuracy in lattice QCD simulations.
+
+      * - Electronic structure
+        - `CP2K <https://github.com/amd/InfinityHub-CI/tree/main/cp2k>`_
+        - CP2K is a quantum chemistry and solid state physics software package that can
+          perform atomistic simulations of solid state, liquid, molecular, periodic, material,
+          crystal, and biological systems. This AMD container, based on a released version
+          of CP2K, is an AMD beta version with ongoing optimizations.
+
+      * - Quantum Monte Carlo Simulation
+        - `QMCPACK <https://github.com/amd/InfinityHub-CI/tree/main/qmcpack>`_
+        - QMCPACK is an open-source production-level many-body ab initio Quantum
+          Monte Carlo code for computing the electronic structure of atoms, molecules, 2D
+          nanomaterials and solids. The solid-state capabilities include metallic systems
+          as well as insulators. QMCPACK is expected to run well on workstations through
+          to the latest generation supercomputers. Besides high performance, particular
+          emphasis is placed on code quality and reproducibility.
+
+      * - Climate and weather
+        - `MPAS <https://github.com/amd/InfinityHub-CI/tree/main/mpas>`_
+        - The Model for Prediction Across Scales (MPAS) is a collaborative project for
+          developing atmosphere, ocean, and other earth-system simulation components
+          for use in climate, regional climate, and weather studies.
+
+      * - Benchmark
+        - `rocHPL <https://github.com/amd/InfinityHub-CI/tree/main/rochpl>`_
+        - HPL, or High-Performance Linpack, is a benchmark which solves a uniformly 
+          random system of linear equations and reports floating-point execution rate. 
+          This documentation supports the implementation of the HPL benchmark on 
+          top of AMD's ROCm platform.
+
+      * -
+        - `rocHPL-MxP <https://github.com/amd/InfinityHub-CI/tree/main/hpl-mxp>`_
+        - Benchmark that highlights the convergence of HPC and AI workloads by
+          solving a system of linear equations using novel, mixed-precision
+          algorithms.
+
+      * -
+        - `HPCG <https://github.com/amd/InfinityHub-CI/tree/main/hpcg>`_
+        - HPCG, or the High Performance Conjugate Gradient Benchmark complements
+          the High Performance LINPACK (HPL) benchmark. The computational and data
+          access patterns of HPCG are designed to closely match a broad set of important
+          applications not represented by HPL, and to incentivize computer system
+          designers to invest in capabilities that will benefit the collective performance
+          of these applications.
+
+      * - Tools and libraries
+        - `ROCm with GPU-aware MPI container <https://github.com/amd/InfinityHub-CI/tree/main/base-gpu-mpi-rocm-docker>`_
+        - Base container for GPU-aware MPI with ROCm for HPC applications. This
+          project provides a boilerplate for building and running a Docker
+          container with ROCm supporting GPU-aware MPI implementations using
+          OpenMPI or UCX.
+
+      * -
+        - `Kokkos <https://github.com/amd/InfinityHub-CI/tree/main/kokkos>`_
+        - Kokkos is a programming model in C++ for writing performance portable
+          applications for use across HPC platforms. It provides abstractions for both
+          parallel execution of code and data management. Kokkos is designed to target
+          complex node architectures with N-level memory hierarchies and multiple types
+          of execution resources.
+
+      * -
+        - `PyFR <https://github.com/amd/InfinityHub-CI/tree/main/pyfr>`_
+        - PyFR is an open-source Python based framework for solving advection-diffusion
+          type problems on streaming architectures using the Flux Reconstruction approach of
+          Huynh. The framework is designed to solve a range of governing systems on mixed
+          unstructured grids containing various element types. It is also designed to target a
+          range of hardware platforms via use of an in-built domain specific language derived
+          from the Mako templating engine.
+
+      * -
+        - `RAJA <https://github.com/amd/InfinityHub-CI/tree/main/raja>`_
+        - RAJA is a library of C++ software abstractions, primarily developed at Lawrence
+          Livermore National Laboratory (LLNL), that enables architecture and programming
+          model portability for HPC applications.
+
+      * -
+        - `Trilinos <https://github.com/amd/InfinityHub-CI/tree/main/trilinos>`_
+        - The Trilinos Project is an effort to develop algorithms and enabling technologies
+          within an object-oriented software framework for the solution of large-scale,
+          complex multi-physics engineering and scientific problems.
+
+To learn about ROCm for AI applications, see :doc:`../rocm-for-ai/index`.

docs/how-to/system-debugging.md

@@ -5,7 +5,7 @@
   ROCm">
 </head>
 
-# System debugging
+# System debugging guide
 
 ## ROCm language and system-level debug, flags, and environment variables
 

docs/how-to/tuning-guides.md

@@ -0,0 +1,108 @@
+<head>
+  <meta charset="UTF-8">
+  <meta name="description" content="AMD hardware optimization for specific workloads">
+  <meta name="keywords" content="high-performance computing, HPC, Instinct accelerators,
+  Radeon, tuning, tuning guide, AMD, ROCm">
+</head>
+
+# System optimization
+
+This guide outlines system setup and tuning suggestions for AMD hardware to optimize performance for specific types of
+workloads or use-cases.
+
+## High-performance computing
+
+High-performance computing (HPC) workloads have unique requirements. The default
+hardware and BIOS configurations for OEM platforms may not provide optimal
+performance for HPC workloads. To enable optimal HPC settings on a per-platform
+and per-workload level, this guide calls out:
+
+* BIOS settings that can impact performance
+* Hardware configuration best practices
+* Supported versions of operating systems
+* Workload-specific recommendations for optimal BIOS and operating system
+  settings
+
+There is also a discussion on the AMD Instinct™ software development
+environment, including information on how to install and run the DGEMM, STREAM,
+HPCG, and HPL benchmarks. This guidance provides a good starting point but is
+not exhaustively tested across all compilers.
+
+Prerequisites to understanding this document and to performing tuning of HPC
+applications include:
+
+* Experience in configuring servers
+* Administrative access to the server's Management Interface (BMC)
+* Administrative access to the operating system
+* Familiarity with the OEM server's BMC (strongly recommended)
+* Familiarity with the OS specific tools for configuration, monitoring, and
+  troubleshooting (strongly recommended)
+
+This document provides guidance on tuning systems with various AMD Instinct™
+accelerators for HPC workloads. This document is not an all-inclusive guide, and
+some items referred to may have similar, but different, names in various OEM
+systems (for example, OEM-specific BIOS settings). This document also provides
+suggestions on items that should be the initial focus of additional,
+application-specific tuning.
+
+This document is based on the AMD EPYC™ 7003-series processor family (former
+codename "Milan").
+
+While this guide is a good starting point, developers are encouraged to perform
+their own performance testing for additional tuning.
+
+:::::{grid} 1 1 2 2
+:gutter: 1
+
+:::{grid-item-card}
+**[AMD Instinct™ MI200](./tuning-guides/mi200)**
+
+This chapter goes through how to configure your AMD Instinct™ MI200 accelerated
+compute nodes to get the best performance out of them.
+
+* [Instruction Set Architecture (ISA)](https://www.amd.com/system/files/TechDocs/instinct-mi200-cdna2-instruction-set-architecture.pdf)
+* [White paper](https://www.amd.com/system/files/documents/amd-cdna2-white-paper.pdf)
+
+:::
+
+:::{grid-item-card}
+**[AMD Instinct™ MI100](./tuning-guides/mi100)**
+
+This chapter briefly reviews hardware aspects of the AMD Instinct™ MI100
+accelerators and the CDNA™ 1 architecture that is the foundation of these GPUs.
+
+* [ISA](https://www.amd.com/system/files/TechDocs/instinct-mi100-cdna1-shader-instruction-set-architecture%C2%A0.pdf)
+* [White paper](https://www.amd.com/system/files/documents/amd-cdna-whitepaper.pdf)
+
+:::
+
+:::::
+
+## Workstation
+
+Workstation workloads, much like high-performance computing, have a unique set of
+requirements, a blend of both graphics and compute, certification, stability and
+the list continues.
+
+The document covers specific software requirements and processes needed to use
+these GPUs for Single Root I/O Virtualization (SR-IOV) and machine learning
+(ML).
+
+The main purpose of this document is to help users utilize the RDNA 2 GPUs to
+their full potential.
+
+:::::{grid} 1 1 2 2
+:gutter: 1
+
+:::{grid-item-card}
+**[AMD Radeon™ PRO W6000 and V620](./tuning-guides/w6000-v620)**
+
+This chapter describes the AMD GPUs with RDNA™ 2 architecture, namely AMD Radeon
+PRO W6800 and AMD Radeon PRO V620
+
+* [AMD RDNA2 ISA](https://www.amd.com/system/files/TechDocs/rdna2-shader-instruction-set-architecture.pdf)
+* [White paper](https://www.amd.com/system/files/documents/rdna2-explained-radeon-pro-W6000.pdf)
+
+:::
+
+:::::

docs/how-to/tuning-guides.rst

@@ -1,107 +0,0 @@
-.. meta::
-   :description: AMD hardware optimization for specific workloads
-   :keywords: high-performance computing, HPC, Instinct accelerators, Radeon,
-              AMD, ROCm, system, EPYC, CPU, GPU, BIOS, OS
-
-*******************
-System optimization
-*******************
-
-System administrators can optimize the performance of their AMD hardware
-generally and based on specific workloads and use cases. This section outlines
-recommended system optimization options for AMD accelerators and GPUs, enabling
-administrators to maximize efficiency and performance.
-
-High-performance computing workloads
-====================================
-
-High-performance computing (HPC) workloads have unique requirements that may not
-be fully met by the default hardware and BIOS configurations of OEM platforms.
-To achieve optimal performance for HPC workloads, it is crucial to adjust
-settings at both the platform and workload levels. 
-
-The :ref:`AMD Instinct™ accelerator optimization guides <mi-optimization-guides>`
-in this section describe:
-
-* BIOS settings that can impact performance
-* Hardware configuration best practices
-* Supported versions of operating systems
-* Workload-specific recommendations for optimal BIOS and operating system
-  settings
-
-The guides might also discuss the AMD Instinct software development
-environment, including information on how to install and run the DGEMM, STREAM,
-HPCG, and HPL benchmarks. The guides provide a good starting point but is
-not tested exhaustively across all compilers.
-
-Knowledge prerequisites to better understand the following
-:ref:`Instinct system optimization guides <mi-optimization-guides>` and to
-perform tuning for HPC applications include:
-
-* Experience in configuring servers
-* Administrative access to the server's Management Interface (BMC)
-* Administrative access to the operating system
-* Familiarity with the OEM server's BMC (strongly recommended)
-* Familiarity with the OS specific tools for configuration, monitoring, and
-  troubleshooting (strongly recommended)
-
-While the following guides are a good starting point, developers are encouraged
-to perform their own performance testing for additional tuning per device and
-per workload.
-
-.. _mi-optimization-guides:
-
-.. list-table::
-   :header-rows: 1
-   :stub-columns: 1
-
-   * - Optimization guide
-
-     - Architecture reference
-
-     - White papers
-
-   * - :doc:`AMD Instinct MI200 <tuning-guides/mi200>`
-
-     - `AMD Instinct MI200 instruction set architecture <https://www.amd.com/system/files/TechDocs/instinct-mi200-cdna2-instruction-set-architecture.pdf>`_
-
-     - `CDNA 2 architecture <https://www.amd.com/system/files/documents/amd-cdna2-white-paper.pdf>`_
-
-   * - :doc:`AMD Instinct MI100 <tuning-guides/mi100>`
-
-     - `AMD Instinct MI100 instruction set architecture <https://www.amd.com/system/files/TechDocs/instinct-mi100-cdna1-shader-instruction-set-architecture%C2%A0.pdf>`_
-
-     - `CDNA architecture <https://www.amd.com/system/files/documents/amd-cdna-whitepaper.pdf>`_
-
-Workstation workloads
-=====================
-
-Workstation workloads, much like those for HPC, have a unique set of
-requirements: a blend of both graphics and compute, certification, stability and
-others.
-
-The document covers specific software requirements and processes needed to use
-these GPUs for Single Root I/O Virtualization (SR-IOV) and machine learning
-tasks.
-
-The main purpose of this document is to help users utilize the RDNA™ 2 GPUs to
-their full potential.
-
-.. _rdna-optimization-guides:
-
-.. list-table::
-   :header-rows: 1
-   :stub-columns: 1
-
-   * - Optimization guide
-
-     - Architecture reference
-
-     - White papers
-
-   * - :doc:`AMD Radeon PRO W6000 and V620 <tuning-guides/w6000-v620>`
-
-     - `AMD RDNA 2 instruction set architecture <https://www.amd.com/system/files/TechDocs/rdna2-shader-instruction-set-architecture.pdf>`_
-
-     - `RDNA 2 architecture <https://www.amd.com/system/files/documents/rdna2-explained-radeon-pro-W6000.pdf>`_
-

docs/how-to/tuning-guides/mi100.md

@@ -1,11 +1,11 @@
 <head>
   <meta charset="UTF-8">
   <meta name="description" content="MI100 high-performance computing and tuning guide">
-  <meta name="keywords" content="MI100, high-performance computing, HPC, BIOS
+  <meta name="keywords" content="MI100, high-performance computing, HPC, tuning, BIOS
   settings, NBIO, AMD, ROCm">
 </head>
 
-# AMD Instinct MI100 system optimization
+# MI100 high-performance computing and tuning guide
 
 ## System settings
 

docs/how-to/tuning-guides/mi200.md

@@ -1,11 +1,11 @@
 <head>
   <meta charset="UTF-8">
   <meta name="description" content="MI200 high-performance computing and tuning guide">
-  <meta name="keywords" content="MI200, high-performance computing, HPC, BIOS
+  <meta name="keywords" content="MI200, high-performance computing, HPC, tuning, BIOS
   settings, NBIO, AMD, ROCm">
 </head>
 
-# AMD Instinct MI200 system optimization
+# MI200 high-performance computing and tuning guide
 
 ## System settings
 

docs/how-to/tuning-guides/w6000-v620.md

@@ -1,11 +1,11 @@
 <head>
   <meta charset="UTF-8">
   <meta name="description" content="RDNA2 workstation tuning guide">
-  <meta name="keywords" content="RDNA2, workstation, BIOS settings, installation, AMD,
+  <meta name="keywords" content="RDNA2, workstation tuning, BIOS settings, installation, AMD,
   ROCm">
 </head>
 
-# AMD RDNA2 system optimization
+# RDNA2 workstation tuning guide
 
 ## System settings
 

docs/index.md

@@ -42,7 +42,6 @@ Our documentation is organized into the following categories:
   * {doc}`PyTorch for ROCm<rocm-install-on-linux:how-to/3rd-party/pytorch-install>`
   * {doc}`TensorFlow for ROCm<rocm-install-on-linux:how-to/3rd-party/tensorflow-install>`
   * {doc}`JAX for ROCm<rocm-install-on-linux:how-to/3rd-party/jax-install>`
-  * {doc}`MAGMA for ROCm<rocm-install-on-linux:how-to/3rd-party/magma-install>`
 :::
 
 :::{grid-item-card}
@@ -78,6 +77,7 @@ Our documentation is organized into the following categories:
   * [Development](#development-tools)
   * [Performance analysis](#performance-analysis)
   * [System](#system-tools)
+* [Environment Variables](./reference/env-variables.rst)
 * [Hardware specifications](./reference/gpu-arch-specs.rst)
 :::
 <!-- markdownlint-enable MD051 -->
@@ -88,18 +88,19 @@ Our documentation is organized into the following categories:
 :padding: 2
 
 * [Using ROCm for AI](./how-to/rocm-for-ai/index.rst)
+* [Using ROCm for HPC](./how-to/rocm-for-hpc/index.rst)
 * [Fine-tuning LLMs and inference optimization](./how-to/llm-fine-tuning-optimization/index.rst)
-* [System optimization](./how-to/tuning-guides.rst)
-  * [AMD Instinct MI200](./how-to/tuning-guides/mi200.md)
-  * [AMD Instinct MI100](./how-to/tuning-guides/mi100.md)
-  * [AMD Instinct RDNA2](./how-to/tuning-guides/w6000-v620.md)
-* [System level debugging](./how-to/system-debugging.md)
+* [System tuning for various architectures](./how-to/tuning-guides.md)
+  * [MI100](./how-to/tuning-guides/mi100.md)
+  * [MI200](./how-to/tuning-guides/mi200.md)
+  * [RDNA2](./how-to/tuning-guides/w6000-v620.md)
 * [GPU-enabled MPI](./how-to/gpu-enabled-mpi.rst)
 * [Using compiler features](./conceptual/compiler-topics.md)
   * [Using AddressSanitizer](./conceptual/using-gpu-sanitizer.md)
   * [Compiler disambiguation](./conceptual/compiler-disambiguation.md)
   * [OpenMP support in ROCm](./about/compatibility/openmp.md)
 * [Setting the number of CUs](./how-to/setting-cus)  
+* [System level debugging](./how-to/system-debugging.md)
 * [GitHub examples](https://github.com/amd/rocm-examples)
 :::
 

docs/reference/env-variables.rst

@@ -0,0 +1,920 @@
+.. meta::
+    :description: Environment variables reference
+    :keywords: AMD, ROCm, environment variables, environment, reference
+
+.. role:: cpp(code)
+   :language: cpp
+
+.. _env-variables-reference:
+
+*************************************************************
+ROCm environment variables
+*************************************************************
+
+The following table lists the most commonly used environment variables in the ROCm software stack. These variables help to perform simple tasks such as building a ROCm library or running applications on AMDGPUs.
+
+.. list-table::
+    :header-rows: 1
+    :widths: 70,30
+
+    * - **Environment variable**
+      - **Value**
+
+    * - | ``HIP_PATH``
+        | The path of the HIP SDK on Microsoft Windows.
+      - Default: ``C:/hip``
+
+    * - | ``HIP_DIR``
+        | The path of the HIP SDK on Microsoft Windows. This variable is ignored, if ``HIP_PATH`` is set.
+      - Default: ``C:/hip``
+
+    * - | ``ROCM_PATH``
+        | The path of the installed ROCm software stack on Linux.
+      - Default: ``/opt/rocm``
+
+    * - | ``HIP_PLATFORM``
+        | The platform targeted by HIP. If ``HIP_PLATFORM`` is not set, then HIPCC attempts to auto-detect the platform, if it can find NVCC.
+      - ``amd``, ``nvidia``
+
+CLR environment variables
+=========================
+
+AMD Common Language Runtime (:doc:`CLR <hip:understand/amd_clr>`) library contains source codes for AMD's compute languages runtimes: 
+
+*  ``hipamd``: Contains implementation of HIP language on the AMD platform.
+* ``opencl``: Contains implementation of `OpenCL™ <https://www.khronos.org/opencl/>`_ on AMD platform. It is hosted at `clr/opencl <https://github.com/ROCm/clr/tree/develop/opencl>`_.
+* ``rocclr``: Contains common runtime used in HIP and OpenCL. This is hosted at `clr/rocclr <https://github.com/ROCm/clr/tree/develop/rocclr>`_.
+
+The environment variables affecting the CLR library might affect HIP and OpenCL libraries or applications.
+
+The following table lists the environment variables that affect ``opencl`` and ``hipamd``:
+
+.. list-table::
+    :header-rows: 1
+    :widths: 70,30
+
+    * - **Environment variable**
+      - **Value**
+
+    * - | ``ROCM_LIBPATCH_VERSION``
+        | The ROCm version in the integer format. The format is
+        | :cpp:`MAJOR * 10000 + MINOR * 100 + PATCH`
+      - 50000, 60020...
+
+    * - | ``CPACK_DEBIAN_PACKAGE_RELEASE``
+        | This is the numbering of the Debian package itself, i.e., the version of the packaging and not the version of the content.
+      - 1, 2, 3...
+
+    * - | ``CPACK_RPM_PACKAGE_RELEASE``
+        | This is the numbering of the RPM package itself, i.e., the version of the packaging and not the version of the content.
+      - 1, 2, 3...
+
+The following table lists the environment variables that affect ``hipamd``:
+
+.. list-table::
+    :header-rows: 1
+    :widths: 70,30
+
+    * - **Environment variable**
+      - **Value**
+
+    * - | ``HIP_FORCE_QUEUE_PROFILING``
+        | Simulates the application to run in rocprof by forcing command queue profiling to ``on`` by default.
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``HSA_OVERRIDE_GFX_VERSION``
+        | Overrides the target version; used to enable HIP usage on unsupported hardware.
+      - 11.0.0, 10.3.0
+
+    * - | ``HSA_DISABLE_CACHE``
+        | Disables the L2 cache.
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``HSAKMT_DEBUG_LEVEL``
+        | When set to the highest level, the system prints memory allocation information.
+      - 1, 2, ... 7
+
+The following table lists the environment variables that affect ``rocclr``:
+
+.. https://github.com/ROCm/clr/blob/develop/rocclr/utils/flags.hpp
+
+.. list-table::
+    :header-rows: 1
+    :widths: 35,14,51
+
+    * - **Environment variable**
+      - **Default value**
+      - **Value**
+
+    * - | ``AMD_CPU_AFFINITY``
+        | Resets CPU affinity of any runtime threads
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``AMD_DIRECT_DISPATCH``
+        | Enables direct kernel dispatch. Currently available on Linux; under development for Windows.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``AMD_GPU_FORCE_SINGLE_FP_DENORM``
+        | Forces denormalization for single precision.
+      - ``-1``
+      - | -1: Don't force 
+        | 0: Disable
+        | 1: Enable
+
+    * - | ``AMD_LOG_LEVEL``
+        | Enables HIP log on various level.
+      - ``0``
+      - | 0: Disable log.
+        | 1: Enables log on error level.
+        | 2: Enables log on warning and lower levels.
+        | 3: Enables log on information and lower levels.
+        | 4: Enables log on debug and lower levels.
+
+    * - | ``AMD_LOG_LEVEL_FILE``
+        | Sets output file for ``AMD_LOG_LEVEL``.
+      - stderr output
+      - 
+
+    * - | ``AMD_LOG_MASK``
+        | Specifies HIP log filters. Here is the ` complete list of log masks <https://github.com/ROCm/clr/blob/develop/rocclr/utils/debug.hpp#L40>`_.
+      - ``0x7FFFFFFF``
+      - | 0x1: Log API calls.
+        | 0x2: Kernel and copy commands and barriers.
+        | 0x4: Synchronization and waiting for commands to finish.
+        | 0x8: Decode and display AQL packets.
+        | 0x10: Queue commands and queue contents.
+        | 0x20: Signal creation, allocation, pool.
+        | 0x40: Locks and thread-safety code.
+        | 0x80: Kernel creations and arguments, etc.
+        | 0x100: Copy debug.
+        | 0x200: Detailed copy debug.
+        | 0x400: Resource allocation, performance-impacting events.
+        | 0x800: Initialization and shutdown.
+        | 0x1000: Misc debug, not yet classified.
+        | 0x2000: Show raw bytes of AQL packet.
+        | 0x4000: Show code creation debug.
+        | 0x8000: More detailed command info, including barrier commands.
+        | 0x10000: Log message location.
+        | 0x20000: Memory allocation.
+        | 0x40000: Memory pool allocation, including memory in graphs.
+        | 0x80000: Timestamp details.
+        | 0xFFFFFFFF: Log always even mask flag is zero.
+
+    * - | ``AMD_OCL_BUILD_OPTIONS``
+        | Sets the options for ``clBuildProgram`` and ``clCompileProgram``. This variable overrides the previously set options.
+      - None
+      - 
+
+    * - | ``AMD_OCL_BUILD_OPTIONS_APPEND``
+        | Appends the options for ``clBuildProgram`` and ``clCompileProgram``.
+      - None
+      - 
+
+    * - | ``AMD_OCL_LINK_OPTIONS``
+        | Sets the options for ``clLinkProgram``.
+      - None
+      - 
+
+    * - | ``AMD_OCL_LINK_OPTIONS_APPEND``
+        | Appends the options for ``clLinkProgram``.
+      - None
+      - 
+
+    * - | ``AMD_OCL_WAIT_COMMAND``
+        | Enforces a wait for every submitted command.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``OCL_SET_SVM_SIZE``
+        | Sets shared virtual memory (SVM) space size in bytes for discrete GPUs.
+      - ``65536``
+      -
+
+    * - | ``OCL_STUB_PROGRAMS``
+        | Enables OCL programs stubing.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``OPENCL_VERSION``
+        | Force GPU OpenCL version.
+      - ``200``
+      - 
+
+    * - | ``AMD_OPT_FLUSH``
+        | Sets kernel flush option.
+      - ``0x1``
+      - | ``0x0`` = Uses system-scope fence operations.
+        | ``0x1`` = Uses device-scope fence operations when possible.
+
+    * - | ``AMD_SERIALIZE_COPY``
+        | Controls serialization of copies
+      - ``0``
+      - | 0: Disable
+        | 1: Waits for completion before enqueue.
+        | 2: Waits for completion after enqueue.
+        | 3: Both
+
+    * - | ``AMD_SERIALIZE_KERNEL``
+        | Serializes kernel enqueue.
+      - ``0``
+      - | 0: Disable
+        | 1: Waits for completion before enqueue.
+        | 2: Waits for completion after enqueue.
+        | 3: Both
+
+    * - | ``AMD_THREAD_TRACE_ENABLE``
+        | Enables thread trace extension.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``CL_KHR_FP64``
+        | Controls support for double precision.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``CQ_THREAD_STACK_SIZE``
+        | The default command queue thread stack size in Bytes.
+      - ``262144``: 256 KB
+      -
+
+    * - | ``CUDA_VISIBLE_DEVICES``
+        | The visible devices to HIP (whose indices are present in the sequence)
+      - None
+      - ``0,1,2``: List of the device indices. Depending on the number of devices in the system.
+
+    * - | ``DEBUG_CLR_GRAPH_PACKET_CAPTURE``
+        | Controls capturing of graph packets.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``DEBUG_CLR_LIMIT_BLIT_WG``
+        | Sets the limit for the number of workgroups in blit operations.
+      - ``16``
+      -
+
+    * - | ``DISABLE_DEFERRED_ALLOC``
+        | Controls deferred memory allocation on device.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_ADD_HBCC_SIZE``
+        | Adds HBCC size to the reported device memory.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_ANALYZE_HANG``
+        | Allows you to analyze GPU hang issue.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_BLIT_ENGINE_TYPE``
+        | Specifies blit engine type.
+      - ``0``
+      - | 0: Default
+        | 1: Host
+        | 2: CAL
+        | 3: Kernel
+
+    * - | ``GPU_CP_DMA_COPY_SIZE``
+        | Set maximum size of CP DMA copy in KB.
+      - ``1``
+      -
+
+    * - | ``GPU_DEBUG_ENABLE``
+        | Enables collection of extra information for debugger at the cost of performance.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_DEVICE_ORDINAL``
+        | Selects the device ordinal, which is a comma separated list of available devices.
+      - None
+      - A value of ``0,2`` exposes devices 1 and 3 in the system.
+
+    * - | ``GPU_DUMP_BLIT_KERNELS``
+        | Controls dumping of the kernels for blit manager.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_DUMP_CODE_OBJECT``
+        | Controls dumping of code object.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_ENABLE_COOP_GROUPS``
+        | Enables cooperative group launch.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_ENABLE_HW_P2P``
+        | Enables hardware peer to peer (P2P) path.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_ENABLE_LC``
+        | Enables LC path.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_ENABLE_PAL``
+        | Specifies platform abstraction library (PAL) backend.
+      - ``2``
+      - | 0: ROC
+        | 1: PAL
+        | 2: ROC or PAL
+
+    * - | ``GPU_ENABLE_WAVE32_MODE``
+        | Enables Wave32 compilation in hardware, if available.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_ENABLE_WGP_MODE``
+        | Enables WGP Mode in hardware, if available. Workgroups of waves are
+        | dispatched in one of the two modes: CU or WGP.
+      - ``1``
+      - | 0: CU mode. The waves of a workgroup are distributed across just two SIMD32’s.
+        | 1: WGP mode. The waves of a workgroup are distributed across all 4 SIMD32’s within a workgroup.
+
+    * - | ``GPU_FORCE_BLIT_COPY_SIZE``
+        | Specifies the threshold size in KB, under which blit is forced instead of system direct memory access (SDMA).
+      - ``0``
+      -
+
+    * - | ``GPU_FORCE_QUEUE_PROFILING``
+        | Forces command queue profiling.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_FLUSH_ON_EXECUTION``
+        | Submits commands to hardware on every operation.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_IMAGE_BUFFER_WAR``
+        | Enables image buffer workaround.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_IMAGE_DMA``
+        | Enables DRM DMA for image transfers.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_MAX_COMMAND_BUFFERS``
+        | Sets the maximum number of command buffers allocated per queue.
+      - ``8``
+      -
+
+    * - | ``GPU_MAX_HEAP_SIZE``
+        | Sets the maximum size of the GPU heap (in percentage) on the board memory.
+      - ``100``
+      -
+
+    * - | ``GPU_MAX_HW_QUEUES``
+        | Sets the maximum number of hardware queues to be allocated per device.
+      - ``4``
+      - This variable controls how many independent hardware queues HIP runtime can create per process, per device. If an application allocates more HIP streams than the specified value, then HIP runtime reuses the same hardware queues for the new streams in a round-robin manner. Note that this value doesn't apply to hardware queues that are created for CU-masked HIP streams or cooperative queues for HIP cooperative groups (single queue per device).
+
+    * - | ``GPU_MAX_REMOTE_MEM_SIZE``
+        | Sets the maximum size in KB for device memory substitution with the system.
+      - ``2``
+      -
+
+    * - | ``GPU_MAX_SUBALLOC_SIZE``
+        | Sets the maximum size for sub-allocations in KB.
+      - ``4096``
+      -
+
+    * - | ``GPU_MAX_USWC_ALLOC_SIZE``
+        | Sets the maximum uncacheable speculative write combining (USWC) allocation size in MB.
+      - ``2048``
+      - -1: No limit
+
+    * - | ``GPU_MAX_WORKGROUP_SIZE``
+        | Sets the maximum number of workitems in a workgroup for GPU.
+      - ``0``: Sets no limit on workitems.
+      -
+
+    * - | ``GPU_MIPMAP``
+        | Enables GPU mipmap extension.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_NUM_COMPUTE_RINGS``
+        | Sets the number of GPU compute rings.
+      - ``2``
+      - | 0: Disable
+        | Any other number corresponds to the number of compute rings.
+
+    * - | ``GPU_NUM_MEM_DEPENDENCY``
+        | Sets the number of memory objects for dependency tracking.
+      - ``256``
+      -
+
+    * - | ``GPU_PINNED_MIN_XFER_SIZE``
+        | Sets the minimum buffer size (in MB) for pinned read and write transfers.
+      - ``128``
+      -
+
+    * - | ``GPU_PINNED_XFER_SIZE``
+        | Sets the buffer size (in MB) for pinned read and write transfers.
+      - ``32``
+      -
+
+    * - | ``GPU_PRINT_CHILD_KERNEL``
+        | Specifies the number of child kernels to be printed.
+      - ``0``
+      -
+
+    * - | ``GPU_RESOURCE_CACHE_SIZE``
+        | Sets the resource cache size in MB.
+      - ``64``
+      -
+
+    * - | ``GPU_SINGLE_ALLOC_PERCENT``
+        | Sets the maximum size of a single allocation as a percentage of  the total.
+      - ``85``
+      - 
+
+    * - | ``GPU_STAGING_BUFFER_SIZE``
+        | Sets the GPU staging buffer size in MB.
+      - ``4``
+      -
+
+    * - | ``GPU_STREAMOPS_CP_WAIT``
+        | Forces the stream memory operation to wait on command processor (CP).
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_USE_DEVICE_QUEUE``
+        | Controls use of dedicated device queue for the actual submissions.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``GPU_WAVES_PER_SIMD``
+        | Forces the number of waves per SIMD.
+      - ``0``
+      - 1-10
+
+    * - | ``GPU_XFER_BUFFER_SIZE``
+        | Sets the transfer buffer size for image copy optimization in KB.
+      - ``0``
+      -
+        
+    * - | ``HIP_FORCE_DEV_KERNARG``
+        | Forces device memory for kernel arguments.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``HIP_HIDDEN_FREE_MEM``
+        | Specifies the amount of memory to hide from the free memory reported by ``hipMemGetInfo``.
+      - ``0``: Disable
+      -
+
+    * - | ``HIP_HOST_COHERENT``
+        | Specifies if the memory is coherent between the host and GPU in ``hipHostMalloc``.
+      - ``0``
+      - | 0: Memory is not coherent.
+        | 1: Memory is coherent.
+        | Environment variable has effect, if the following conditions are statisfied:
+        | - One of the ``hipHostMallocDefault``, ``hipHostMallocPortable``,  ``hipHostMallocWriteCombined`` or ``hipHostMallocNumaUser`` flag set to 1.
+        | - ``hipHostMallocCoherent``, ``hipHostMallocNonCoherent`` and ``hipHostMallocMapped`` flags set to 0.
+
+    * - | ``HIP_INITIAL_DM_SIZE``
+        | Sets the initial heap size for device malloc.
+      - ``8388608``: 8 MB
+      -
+
+    * - | ``HIP_LAUNCH_BLOCKING``
+        | Controls serialization of kernel execution.
+      - ``0``
+      - | 0: Disable. Kernel executes normally.
+        | 1: Enable. Serializes kernel execution; behaves similar to ``AMD_SERIALIZE_KERNEL``.
+
+    * - | ``HIP_MEM_POOL_SUPPORT``
+        | Enables memory pool support in HIP.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``HIP_MEM_POOL_USE_VM``
+        | Enables memory pool support in HIP.
+      - | ``0``: Default value on other OS.
+        | ``1``: Default value on Microsoft Windows.
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``HIP_USE_RUNTIME_UNBUNDLER``
+        | Controls use of runtime code object unbundler.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``HIP_VISIBLE_DEVICES``
+        | Specifies the indices of the devices allowed to be visible to HIP.
+      - None
+      - 0,1,2: Depending on the number of devices on the system.
+
+    * - | ``HIP_VMEM_MANAGE_SUPPORT``
+        | Enables virtual memory management support.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``HIPCC_VERBOSE``
+        | Controls the extra information to be displayed during the build such as compiler commands with flags, paths and arguments.
+      - ``0``
+      - | 0x1: Print detailed compiler commands.
+        | 0x2: Print HIP, ROCm and CUDA paths (``HIP_PATH``, ``ROCM_PATH``, ``HIP_CLANG_PATH``, ...). 
+        | 0x4: Print HIPCC arguments.
+
+    * - | ``HIPRTC_COMPILE_OPTIONS_APPEND``
+        | Sets compile options needed for ``hiprtc`` compilation.
+      - None
+      - ``--gpu-architecture=gfx906:sramecc+:xnack``, ``-fgpu-rdc``
+
+    * - | ``HIPRTC_LINK_OPTIONS_APPEND``
+        | Sets link options needed for ``hiprtc`` compilation.
+      - None
+      - 
+
+    * - | ``HIPRTC_USE_RUNTIME_UNBUNDLER``
+        | Forces runtime unbundler in hiprtc.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``HSA_KERNARG_POOL_SIZE``
+        | Sets the pool size for kernel arguments.
+      - ``1048576``: 1 MB
+      -
+
+    * - | ``HSA_LOCAL_MEMORY_ENABLE``
+        | Enables use of local memory on HSA device.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``PAL_DISABLE_SDMA``
+        | Disables SDMA for PAL.
+      - ``0``
+      - | 0: Enable SDMA for PAL.
+        | 1: Disable SDMA for PAL.
+
+    * - | ``PAL_MALL_POLICY``
+        | Controls the behaviour of allocations with respect to the MALL.
+      - ``0``
+      - | 0: MALL policy is decided by KMD.
+        | 1: Allocations are never put through the MALL.
+        | 2: Allocations will always be put through the MALL.
+
+    * - | ``PAL_ALWAYS_RESIDENT``
+        | Forces memory resources to become resident during allocation.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``PAL_EMBED_KERNEL_MD``
+        | Enables writing kernel metadata into command buffers.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``PAL_FORCE_ASIC_REVISION``
+        | Forces a specific ASIC revision on all devices.
+      - ``0``
+      -
+
+    * - | ``PAL_HIP_IPC_FLAG``
+        | Enables inter-process flag for device allocation in PAL HIP.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``PAL_PREPINNED_MEMORY_SIZE``
+        | Sets the size in KB of pre-pinned memory.
+      - ``64``
+      -
+
+    * - | ``PAL_RGP_DISP_COUNT``
+        | Sets the number of dispatches for RGP capture with SQTT.
+      - ``10000``
+      -
+
+    * - | ``REMOTE_ALLOC``
+        | Enables use of remote memory for the global heap allocation.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``ROC_ACTIVE_WAIT_TIMEOUT``
+        | Forces active wait of GPU interrupt for the timeout in us.
+      - ``0``
+      -
+
+    * - | ``ROC_AQL_QUEUE_SIZE``
+        | Sets the AQL queue size in bytes in the AQL packets.
+      - ``16384``: 16 KB
+      -
+
+    * - | ``ROC_CPU_WAIT_FOR_SIGNAL``
+        | Enable CPU wait for dependent HSA signals.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``ROC_ENABLE_LARGE_BAR``
+        | Enable large bar if supported by the device.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``ROC_GLOBAL_CU_MASK``
+        | Sets a global CU mask, entered as hex value for all queues. Each active bit represents one CU, e.g., ``0xf`` enables 4 CUs.
+      - None
+      - 
+
+    * - | ``ROC_HMM_FLAGS``
+        | Sets ROCm HMM configuration flags.
+      - ``0``: Disabled
+      - 
+
+    * - | ``ROC_P2P_SDMA_SIZE``
+        | Sets the minimum size in KB for peer to peer (P2P) transfer with SDMA.
+      - ``1024``: 1 MB
+      -
+
+    * - | ``ROC_SIGNAL_POOL_SIZE``
+        | Sets the initial size for HSA signal pool.
+      - ``32``
+      - 
+
+    * - | ``ROC_SKIP_KERNEL_ARG_COPY``
+        | Allows the runtime to skip kernel argument copy.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``ROC_SYSTEM_SCOPE_SIGNAL``
+        | Enable system scope for signals, uses interrupts.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``ROC_USE_FGS_KERNARG``
+        | Enables use of fine grain kernel arguments segment for supported ASICs.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``ROCPROFILER_REGISTER_ROOT``
+        | Sets the path to ``rocProfiler``.
+      - None
+      - 
+
+The following table lists the debug environment variables that affect ``rocclr`` of the CLR project. These environment variables can only be set during DEBUG build.
+
+.. list-table::
+    :header-rows: 1
+    :widths: 35,14,51
+
+    * - **Environment variable**
+      - **Default value**
+      - **Value**
+
+    * - | ``AMD_OCL_SUBST_OBJFILE``
+        | Specifies binary substitution config file for OpenCL.
+      - None
+      - 
+
+    * - | ``CPU_MEMORY_ALIGNMENT_SIZE``
+        | Sets the size in bytes for the default alignment of guarded memory on CPU.
+      - ``256``
+      -
+
+    * - | ``CPU_MEMORY_GUARD_PAGE_SIZE``
+        | Size of the CPU memory guard page in KB.
+      - ``64``: 64 KB
+      -
+
+    * - | ``CPU_MEMORY_GUARD_PAGES``
+        | Enables using guard pages for CPU memory.
+      - ``0``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``MEMOBJ_BASE_ADDR_ALIGN``
+        | Alignment of the base address of any allocate memory object.
+      - ``4096``: 4 KB
+      -
+
+    * - | ``PARAMETERS_MIN_ALIGNMENT``
+        | Specifies the minimum alignment required for the abstract parameters stack.
+      - 64 at ``__AVX512F__``, 32 at ``__AVX__`` and 16 in other cases
+      -
+
+ROCR-Runtime environment variables
+==================================
+
+.. https://github.com/ROCm/ROCR-Runtime/blob/master/src/core/util/flag.h
+.. We need to extend the following list.
+
+The following table lists the ROCR-Runtime environment variables:
+
+.. list-table::
+    :header-rows: 1
+    :widths: 35,14,51
+
+    * - **Environment variable**
+      - **Default value**
+      - **Value**
+
+    * - | ``ROCR_VISIBLE_DEVICES``
+        | Specifies a list of device indices or UUIDs to be exposed to the applications.
+      - None
+      - ``0,GPU-DEADBEEFDEADBEEF``
+
+    * - | ``HSA_SCRATCH_MEM``
+        | Specifies the maximum amount of scratch memory that can be used per process per GPU.
+      -
+      -
+
+    * - | ``HSA_XNACK``
+        | Enables XNACK.
+      - None
+      - 1: Enable
+
+    * - | ``HSA_CU_MASK``
+        | Sets the mask on a lower level of queue creation in the driver. 
+        | This mask is also applied to the queues being profiled.
+      - None
+      - ``1:0-8``
+
+    * - | ``HSA_ENABLE_SDMA``
+        | Enables the use of direct memory access (DMA) engines in all copy directions (Host-to-Device, Device-to-Host, Device-to-Device), when using any of the following APIs:
+        | ``hsa_memory_copy``, 
+        | ``hsa_amd_memory_fill``, 
+        | ``hsa_amd_memory_async_copy``, 
+        | ``hsa_amd_memory_async_copy_on_engine``.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+    * - | ``HSA_ENABLE_PEER_SDMA``
+        | Enables the use of DMA engines for Device-to-Device copies, when using any of the following APIs:
+        | ``hsa_memory_copy``,
+        | ``hsa_amd_memory_async_copy``,
+        | ``hsa_amd_memory_async_copy_on_engine``.
+      - ``1``
+      - | 0: Disable
+        | 1: Enable
+
+Note that this environment variable is ignored if ``HSA_ENABLE_SDMA`` is set to 0.
+
+rocPRIM environment variables
+=============================
+
+The following table lists the environment variables used in the rocPRIM library.
+
+.. list-table::
+    :header-rows: 1
+    :widths: 70,30
+
+    * - **Environment variable**
+      - **Default value**
+
+    * - | ``HIP_PATH``
+        | Specifies the path of the HIP SDK on Microsoft Windows.
+      - ``C:/hip``
+
+    * - | ``HIP_DIR``
+        | Specifies the path of the HIP SDK on Microsoft Windows. This variable is ignored, if ``HIP_PATH`` is set.
+      - ``C:/hip``
+
+    * - | ``VCPKG_PATH``
+        | Specifies the path of the ``vcpkg`` package manager on Microsoft Windows. This environment variable has no effect on Linux.
+      - ``C:/github/vcpkg``
+
+    * - | ``ROCM_PATH``
+        | Specifies the path of the installed ROCm software stack on Linux.
+      - ``/opt/rocm``
+
+    * - | ``ROCM_CMAKE_PATH``
+        | Specifies the path of the installed ROCm ``cmake`` file on Microsoft Windows.
+      - ``C:/hipSDK``
+
+    * - | ``HIPCC_COMPILE_FLAGS_APPEND``
+        | Enables extra ``amdclang++`` compiler flags on Linux. This environment variable is ignored if ``CXX`` environment variable is set.
+      - None
+
+    * - | ``ROCPRIM_USE_HMM``
+        | Enables the test suite to use unified memory, when set to 1 during the tests.
+      - None
+
+    * - | ``CTEST_RESOURCE_GROUP_0``
+        | Enables grouping of the tests for different CI steps. This environment variable is used by CI and is of little use to most users.
+      - None
+
+hipCUB environment variables
+============================
+
+The following table lists the environment variables used in the hipCUB library.
+
+.. list-table::
+    :header-rows: 1
+    :widths: 70,30
+
+    * - **Environment variable**
+      - **Default value**
+
+    * - | ``HIP_PATH``
+        | Specifies the path of the HIP SDK on Microsoft Windows.
+      - ``C:/hip``
+
+    * - | ``HIP_DIR``
+        | Specifies the path of the HIP SDK on Microsoft Windows. This variable is ignored, if ``HIP_PATH`` is set.
+      - ``C:/hip``
+
+    * - | ``VCPKG_PATH``
+        | Specifies the path of the ``vcpkg`` package manager on Microsoft Windows. This environment variable has no effect on Linux.
+      - ``C:/github/vcpkg``
+
+    * - | ``ROCM_PATH``
+        | Specifies the path of the installed ROCm software stack on Linux.
+      - ``/opt/rocm``
+
+    * - | ``HIPCC_COMPILE_FLAGS_APPEND``
+        | Enables extra ``amdclang`` or ``amdclang++`` compiler flags on Linux. This environment variable is ignored if ``CXX`` or ``CC`` environment variable is set.
+      - None
+
+    * - | ``HIPCUB_USE_HMM``
+        | Enables the test suite to use unified memory, when set to 1 during the tests.
+      - None
+
+    * - | ``CTEST_RESOURCE_GROUP_0``
+        | Enables grouping of the tests for different CI steps. This environment variable is used by CI and is of little use to most users.
+      - None
+
+rocThrust environment variables
+===============================
+
+The following table lists the environment variables used in the rocThrust library.
+
+.. list-table::
+    :header-rows: 1
+    :widths: 70,30
+
+    * - **Environment variable**
+      - **Default value**
+
+    * - | ``HIP_PATH``
+        | Specifies the path of the HIP SDK on Microsoft Windows.
+      - ``C:/hip``
+
+    * - | ``HIP_DIR``
+        | Specifies the path of the HIP SDK on Microsoft Windows. This variable is ignored, if ``HIP_PATH`` is set.
+      - ``C:/hip``
+
+    * - | ``VCPKG_PATH``
+        | Specifies the path of the ``vcpkg`` package manager on Microsoft Windows. This environment variable has no effect on Linux.
+      - ``C:/github/vcpkg``
+
+    * - | ``ROCM_PATH``
+        | Specifies the path of the installed ROCm software stack on Linux.
+      - ``/opt/rocm``
+
+    * - | ``ROCTHRUST_USE_HMM``
+        | Enables the test suite to use unified memory, when set to 1 during the tests.
+      - None
+
+    * - | ``CTEST_RESOURCE_GROUP_0``
+        | Enables grouping of the tests for different CI steps. This environment variable is used by CI and is of little use to most users.
+      - None

docs/reference/gpu-arch-specs.rst

@@ -233,6 +233,22 @@ The following tables provide an overview of the hardware specifications for AMD
           - L0 Instruction Cache (KiB)
           - VGPR File (KiB)
           - SGPR File (KiB)
+        *
+          - Radeon PRO W7900 Dual Slot
+          - RDNA3
+          - gfx1100
+          - 48
+          - 96
+          - 32
+          - 128
+          - 96
+          - 6
+          - 256
+          - 32
+          - 16
+          - 32
+          - 384
+          - 20
         *
           - Radeon PRO W7900
           - RDNA3

docs/reference/rocm-tools.md

@@ -53,6 +53,7 @@
 * {doc}`ROCm Data Center Tool <rdc:index>`
 * {doc}`ROCm SMI <rocm_smi_lib:index>`
 * {doc}`ROCm Validation Suite <rocmvalidationsuite:index>`
+* {doc}`TransferBench <transferbench:index>`
 :::
 
 ::::

docs/release/versions.md

@@ -8,6 +8,7 @@
 
 | Version | Release date |
 | ------- | ------------ |
+| [6.1.2](https://rocm.docs.amd.com/en/docs-6.1.2/) | June 4, 2024 |
 | [6.1.1](https://rocm.docs.amd.com/en/docs-6.1.1/) | May 8, 2024 |
 | [6.1.0](https://rocm.docs.amd.com/en/docs-6.1.0/) | Apr 16, 2024 |
 | [6.0.2](https://rocm.docs.amd.com/en/docs-6.0.2/) | Jan 31, 2024 |

docs/sphinx/_toc.yml.in

@@ -11,7 +11,7 @@ subtrees:
     title: Release notes
     subtrees:
     - entries:
-      - file: about/CHANGELOG.md
+      - file: about/changelog.md
         title: Changelog
   - url: https://github.com/ROCm/ROCm/labels/Verified%20Issue
     title: Known issues
@@ -44,6 +44,8 @@ subtrees:
       title: API libraries
     - file: reference/rocm-tools.md
       title: Tools
+    - file: reference/env-variables
+      title: Environment variables
     - file: reference/gpu-arch-specs.rst
       title: Hardware specifications
 
@@ -58,6 +60,8 @@ subtrees:
       - file: how-to/rocm-for-ai/train-a-model.rst
       - file: how-to/rocm-for-ai/hugging-face-models.rst
       - file: how-to/rocm-for-ai/deploy-your-model.rst
+  - file: how-to/rocm-for-hpc/index.rst
+    title: Using ROCm for HPC
   - file: how-to/llm-fine-tuning-optimization/index.rst
     title: Fine-tuning LLMs and inference optimization
     subtrees:
@@ -79,17 +83,16 @@ subtrees:
       - file: how-to/llm-fine-tuning-optimization/optimizing-triton-kernel.rst
         title: Optimizing Triton kernels
       - file: how-to/llm-fine-tuning-optimization/profiling-and-debugging.rst
-  - file: how-to/tuning-guides.rst
+  - file: how-to/tuning-guides.md
     title: System optimization
     subtrees:
     - entries:
-      - file: how-to/tuning-guides/mi200.md
-        title: AMD Instinct MI200
       - file: how-to/tuning-guides/mi100.md
-        title: AMD Instinct MI100
+        title: MI100
+      - file: how-to/tuning-guides/mi200.md
+        title: MI200
       - file: how-to/tuning-guides/w6000-v620.md
-        title: AMD RDNA 2
-  - file: how-to/system-debugging.md
+        title: RDNA2
   - file: how-to/gpu-enabled-mpi.rst
     title: Using MPI
   - file: conceptual/compiler-topics.md
@@ -104,6 +107,8 @@ subtrees:
         title: OpenMP support
   - file: how-to/setting-cus
     title: Setting the number of CUs  
+  - file: how-to/system-debugging.md
+    title: Debugging
   - url: https://github.com/amd/rocm-examples
     title: GitHub examples
 

docs/sphinx/requirements.in

@@ -1,2 +1,2 @@
-rocm-docs-core==1.8.0
-sphinx-reredirects
+rocm-docs-core==1.4.0
+sphinx-reredirects

docs/sphinx/requirements.txt

@@ -6,9 +6,9 @@
 #
 accessible-pygments==0.0.5
     # via pydata-sphinx-theme
-alabaster==1.0.0
+alabaster==0.7.16
     # via sphinx
-babel==2.16.0
+babel==2.15.0
     # via
     #   pydata-sphinx-theme
     #   sphinx
@@ -16,9 +16,9 @@ beautifulsoup4==4.12.3
     # via pydata-sphinx-theme
 breathe==4.35.0
     # via rocm-docs-core
-certifi==2024.8.30
+certifi==2024.2.2
     # via requests
-cffi==1.17.1
+cffi==1.16.0
     # via
     #   cryptography
     #   pynacl
@@ -26,7 +26,7 @@ charset-normalizer==3.3.2
     # via requests
 click==8.1.7
     # via sphinx-external-toc
-cryptography==43.0.1
+cryptography==42.0.7
     # via pyjwt
 deprecated==1.2.14
     # via pygithub
@@ -36,13 +36,13 @@ docutils==0.21.2
     #   myst-parser
     #   pydata-sphinx-theme
     #   sphinx
-fastjsonschema==2.20.0
+fastjsonschema==2.19.1
     # via rocm-docs-core
 gitdb==4.0.11
     # via gitpython
 gitpython==3.1.43
     # via rocm-docs-core
-idna==3.10
+idna==3.7
     # via requests
 imagesize==1.4.1
     # via sphinx
@@ -56,34 +56,34 @@ markdown-it-py==3.0.0
     #   myst-parser
 markupsafe==2.1.5
     # via jinja2
-mdit-py-plugins==0.4.2
+mdit-py-plugins==0.4.1
     # via myst-parser
 mdurl==0.1.2
     # via markdown-it-py
-myst-parser==4.0.0
+myst-parser==3.0.1
     # via rocm-docs-core
-packaging==24.1
+packaging==24.0
     # via
     #   pydata-sphinx-theme
     #   sphinx
 pycparser==2.22
     # via cffi
-pydata-sphinx-theme==0.15.4
+pydata-sphinx-theme==0.15.3
     # via
     #   rocm-docs-core
     #   sphinx-book-theme
-pygithub==2.4.0
+pygithub==2.3.0
     # via rocm-docs-core
 pygments==2.18.0
     # via
     #   accessible-pygments
     #   pydata-sphinx-theme
     #   sphinx
-pyjwt[crypto]==2.9.0
+pyjwt[crypto]==2.8.0
     # via pygithub
 pynacl==1.5.0
     # via pygithub
-pyyaml==6.0.2
+pyyaml==6.0.1
     # via
     #   myst-parser
     #   rocm-docs-core
@@ -92,15 +92,15 @@ requests==2.32.3
     # via
     #   pygithub
     #   sphinx
-rocm-docs-core==1.8.0
+rocm-docs-core==1.4.0
     # via -r requirements.in
 smmap==5.0.1
     # via gitdb
 snowballstemmer==2.2.0
     # via sphinx
-soupsieve==2.6
+soupsieve==2.5
     # via beautifulsoup4
-sphinx==8.0.2
+sphinx==7.3.7
     # via
     #   breathe
     #   myst-parser
@@ -112,37 +112,37 @@ sphinx==8.0.2
     #   sphinx-external-toc
     #   sphinx-notfound-page
     #   sphinx-reredirects
-sphinx-book-theme==1.1.3
+sphinx-book-theme==1.1.2
     # via rocm-docs-core
 sphinx-copybutton==0.5.2
     # via rocm-docs-core
-sphinx-design==0.6.1
+sphinx-design==0.6.0
     # via rocm-docs-core
 sphinx-external-toc==1.0.1
     # via rocm-docs-core
-sphinx-notfound-page==1.0.4
+sphinx-notfound-page==1.0.2
     # via rocm-docs-core
-sphinx-reredirects==0.1.5
+sphinx-reredirects==0.1.4
     # via -r requirements.in
-sphinxcontrib-applehelp==2.0.0
+sphinxcontrib-applehelp==1.0.8
     # via sphinx
-sphinxcontrib-devhelp==2.0.0
+sphinxcontrib-devhelp==1.0.6
     # via sphinx
-sphinxcontrib-htmlhelp==2.1.0
+sphinxcontrib-htmlhelp==2.0.5
     # via sphinx
 sphinxcontrib-jsmath==1.0.1
     # via sphinx
-sphinxcontrib-qthelp==2.0.0
+sphinxcontrib-qthelp==1.0.7
     # via sphinx
-sphinxcontrib-serializinghtml==2.0.0
+sphinxcontrib-serializinghtml==1.1.10
     # via sphinx
 tomli==2.0.1
     # via sphinx
-typing-extensions==4.12.2
+typing-extensions==4.12.0
     # via
     #   pydata-sphinx-theme
     #   pygithub
-urllib3==2.2.3
+urllib3==2.2.2
     # via
     #   pygithub
     #   requests

docs/what-is-rocm.rst

@@ -100,7 +100,7 @@ Tools
   ":doc:`AMD SMI <amdsmi:index>`", "C library for Linux that provides a user space interface for applications to monitor and control AMD devices"
   ":doc:`HIPIFY <hipify:index>`", "Translates CUDA source code into portable HIP C++"
   ":doc:`ROCdbgapi <rocdbgapi:index>`", "ROCm debugger API library"
-  ":doc:`ROCm compilers <./reference/rocmcc>`", "Clang/LLVM-based compiler"
+  ":doc:`ROCmCC <./reference/rocmcc>`", "Clang/LLVM-based compiler"
   ":doc:`rocminfo <rocminfo:index>`", "Reports system information"
   ":doc:`ROCProfiler <rocprofiler:index>`", "Profiling tool for HIP applications"
   ":doc:`ROCTracer <roctracer:index>`", "Intercepts runtime API calls and traces asynchronous activity"
@@ -111,6 +111,7 @@ Tools
   ":doc:`ROCm SMI <rocm_smi_lib:index>`", "C library for Linux that provides a user space interface for applications to monitor and control GPU applications"
   ":doc:`ROCm Validation Suite <rocmvalidationsuite:index>`", "Detects and troubleshoots common problems affecting AMD GPUs running in a high-performance computing environment"
   ":doc:`ROCr Debug Agent <rocr_debug_agent:index>`", "Prints the state of all AMD GPU wavefronts that caused a queue error by sending a SIGQUIT signal to the process while the program is running"
+  ":doc:`TransferBench <transferbench:index>`", "Utility to benchmark simultaneous transfers between user-specified devices (CPUs/GPUs)"
 
 Compilers
 -----------------------------------------------
@@ -119,7 +120,7 @@ Compilers
   :header: "Component", "Description"
 
   "`FLANG <https://github.com/ROCm/flang/>`_", "An out-of-tree Fortran compiler targeting LLVM"
-  "`hipCC <https://github.com/ROCm/HIPCC>`_ ", "Compiler driver utility that calls Clang or NVCC and passes the appropriate include and library options for the target compiler and HIP infrastructure"
+  ":doc:`hipCC <hipcc:index>`", "Compiler driver utility that calls Clang or NVCC and passes the appropriate include and library options for the target compiler and HIP infrastructure"
   "`LLVM (amdclang) <https://github.com/ROCm/llvm-project>`_ ", "Toolkit for the construction of highly optimized compilers, optimizers, and runtime environments"
 
 Runtimes
@@ -128,6 +129,6 @@ Runtimes
 .. csv-table::
   :header: "Component", "Description"
 
-  ":doc:`AMD Common Language Runtime (CLR) <hip:understand/amd_clr>`", "Contains source code for AMD's common language runtimes: HIP and OpenCL"
+  "`AMD Common Language Runtime (CLR) <https://github.com/ROCm/clr>`_", "Contains source code for AMD's common language runtimes: :doc:`HIP <hip:index>` and OpenCL"
   ":doc:`HIP <hip:index>`", "AMD's GPU programming language extension and the GPU runtime"
   ":doc:`ROCR-Runtime <rocr-runtime:index>`", "User-mode API interfaces and libraries necessary for host applications to launch compute kernels on available HSA ROCm kernel agents"

tools/autotag/templates/changelog.jinja

@@ -16,7 +16,7 @@
 
 <!-- spellcheck-disable -->
 
-This page contains the release notes for AMD ROCm Software.
+This page contains the release notes for AMD ROCm™ Software.
 
 {%- for version, release in releases %}
 

tools/autotag/templates/rocm_changes/6.1.1.md

@@ -17,7 +17,6 @@ AMD SMI for ROCm 6.1.1
 
 #### Changes
 
-* Updated the output of `amd-smi metric --ecc-blocks` to show counters available from blocks.
 * Updated the output of `amd-smi metric --clock` to reflect each engine.
 * Updated the output of `amd-smi topology --json` to align with output reported by host and guest systems.
 
@@ -28,10 +27,6 @@ AMD SMI for ROCm 6.1.1
 * Fixed `amd-smi metric --pcie` and `amdsmi_get_pcie_info()` when using RDNA3 (Navi 32 and Navi 31) hardware to prevent "UNKNOWN" reports.
 * Fixed the output results of `amd-smi process` when getting processes running on a device.
 
-#### Removals
-
-* Removed the `amdsmi_get_gpu_process_info` API from the Python library. It was removed from the C library in an earlier release.
-
 #### Known issues
 
 * `amd-smi bad-pages` can result in a `ValueError: Null pointer access` error when using certain PMU firmware versions.
@@ -48,7 +43,7 @@ HIPCC for ROCm 6.1.1
 
 * **Upcoming:** a future release will enable use of compiled binaries `hipcc.bin` and `hipconfig.bin` by default. No action is needed by users; you may continue calling high-level Perl scripts `hipcc` and `hipconfig`. `hipcc.bin` and `hipconfig.bin` will be invoked by the high-level Perl scripts. To revert to the previous behavior and invoke `hipcc.pl` and `hipconfig.pl`, set the `HIP_USE_PERL_SCRIPTS` environment variable to `1`.
 * **Upcoming:** a subsequent release will remove high-level Perl scripts `hipcc` and `hipconfig`. This release will remove the `HIP_USE_PERL_SCRIPTS` environment variable. It will rename `hipcc.bin` and `hipconfig.bin` to `hipcc` and `hipconfig` respectively. No action is needed by the users. To revert to the previous behavior, invoke `hipcc.pl` and `hipconfig.pl` explicitly.
-* **Upcoming:** a subsequent will remove `hipcc.pl` and `hipconfig.pl`.
+* **Upcoming:** a subsequent release will remove `hipcc.pl` and `hipconfig.pl`.
 
 ### ROCm SMI
 

tools/autotag/templates/rocm_changes/6.1.2.md

@@ -0,0 +1,66 @@
+
+ROCm 6.1.2 includes enhancements to SMI tools and improvements to some libraries.
+
+### OS support
+
+ROCm 6.1.2 has been tested against a pre-release version of Ubuntu 22.04.5 (kernel: 5.15 [GA], 6.8 [HWE]).
+
+### AMD SMI
+
+AMD SMI for ROCm 6.1.2
+
+#### Additions
+
+* Added process isolation and clean shader APIs and CLI commands.
+  * `amdsmi_get_gpu_process_isolation()`
+  * `amdsmi_set_gpu_process_isolation()`
+  * `amdsmi_set_gpu_clear_sram_data()`
+* Added the `MIN_POWER` metric to output provided by `amd-smi static --limit`.
+
+#### Optimizations
+
+* Updated the `amd-smi monitor --pcie` output to prevent delays with the `monitor` command.
+
+#### Changes
+
+* Updated `amismi_get_power_cap_info` to return values in uW instead of W.
+* Updated Python library return types for `amdsmi_get_gpu_memory_reserved_pages` and `amdsmi_get_gpu_bad_page_info`.
+* Updated the output of `amd-smi metric --ecc-blocks` to show counters available from blocks.
+
+#### Fixes
+
+* `amdsmi_get_gpu_board_info()` no longer returns junk character strings.
+* `amd-smi metric --power` now correctly details power output for RDNA3, RDNA2, and MI1x devices.
+* Fixed the `amdsmitstReadWrite.TestPowerCapReadWrite` test for RDNA3, RDNA2, and MI100 devices.
+* Fixed an issue with the `amdsmi_get_gpu_memory_reserved_pages` and `amdsmi_get_gpu_bad_page_info` Python interface calls.
+
+#### Removals
+
+* Removed the `amdsmi_get_gpu_process_info` API from the Python library. It was removed from the C library in an earlier release.
+
+```{note}
+See the AMD SMI [detailed changelog](https://github.com/ROCm/amdsmi/blob/rocm-6.1.x/CHANGELOG.md) with code samples for more information.
+```
+
+### HIPCC
+
+HIPCC for ROCm 6.1.2
+
+#### Changes
+
+* **Upcoming:** a future release will enable use of compiled binaries `hipcc.bin` and `hipconfig.bin` by default. No action is needed by users; you may continue calling high-level Perl scripts `hipcc` and `hipconfig`. `hipcc.bin` and `hipconfig.bin` will be invoked by the high-level Perl scripts. To revert to the previous behavior and invoke `hipcc.pl` and `hipconfig.pl`, set the `HIP_USE_PERL_SCRIPTS` environment variable to `1`.
+* **Upcoming:** a subsequent release will remove high-level Perl scripts `hipcc` and `hipconfig`. This release will remove the `HIP_USE_PERL_SCRIPTS` environment variable. It will rename `hipcc.bin` and `hipconfig.bin` to `hipcc` and `hipconfig` respectively. No action is needed by the users. To revert to the previous behavior, invoke `hipcc.pl` and `hipconfig.pl` explicitly.
+* **Upcoming:** a subsequent release will remove `hipcc.pl` and `hipconfig.pl`.
+
+### ROCm SMI
+
+ROCm SMI for ROCm 6.1.2
+
+#### Additions
+
+* Added the ring hang event to the `amdsmi_evt_notification_type_t` enum.
+
+#### Fixes
+
+* Fixed an issue causing ROCm SMI to incorrectly report GPU utilization for RDNA3 GPUs. See the issue on [GitHub](https://github.com/ROCm/ROCm/issues/3112).
+* Fixed the parsing of `pp_od_clk_voltage` in `get_od_clk_volt_info` to work better with MI-series hardware.

tools/rocm-build/README.md

@@ -0,0 +1,37 @@
+# Overview for ROCm.mk
+
+This Makefile builds the various projects that makes up ROCm in the correct order.
+It is expected to be run in an environment with the tooling set up. An easy way
+to do this is to use Docker.
+
+## Targets
+
+* all (default)
+* rocm-dev (a subset of all)
+* clean
+* list_components
+* help
+* T_foo
+* C_foo
+
+## Makefile Variables
+
+* PEVAL set to 1 to enable some Makefile debugging code.
+* RELEASE\_FLAG set to "" to avoid passing "-r" to builds, effect is package defined.
+* NOBUILD="foo bar" to avoid adding foo and bar into the dependencies of top level targets. They still may be
+  built if they are needed as dependencies of other top level targets.
+* toplevel
+
+## Makefile assumptions
+
+### Requirements for package "foo"
+
+#### program build\_foo.sh
+
+* Should take option "-c" to clean
+* Should take option "-r" to do a normal "RelWithDeb" build
+
+### For package "foo" we define some targets
+
+* T\_foo - The main build target, calls "build\_foo.sh -r"
+* C\_foo - Clean target, calls "build_foo.sh -c"

tools/rocm-build/ROCm.mk

@@ -0,0 +1,248 @@
+# Traditional first make target
+all:
+
+# Use bash as a shell
+# On Ubuntu sh is 'dash'
+SHELL:=bash
+
+# Allow RELEASE_FLAG to be overwritten
+RELEASE_FLAG?=-r
+
+# Set SANITIZER_FLAG for sanitizer
+ASAN_DEP:=
+ifeq (${ENABLE_ADDRESS_SANITIZER},true)
+	ASAN_DEP=lightning
+	SANITIZER_FLAG=-a
+endif
+
+export INFRA_REPO:=ROCm/tools/rocm-build
+
+OUT_DIR:=$(shell . ${INFRA_REPO}/envsetup.sh >/dev/null 2>&1 ; echo $${OUT_DIR})
+ROCM_INSTALL_PATH:=$(shell . ${INFRA_REPO}/envsetup.sh >/dev/null 2>&1 ; echo $${ROCM_INSTALL_PATH})
+
+$(info OUT_DIR=${OUT_DIR})
+$(info ROCM_INSTALL_PATH=${ROCM_INSTALL_PATH})
+
+# -------------------------------------------------------------------------
+# Internal stuff. Could be put in a different file to hide it.
+# Internal macros, they need to be defined before being used.
+
+# The internal "eval" allows parts of the Makefile to be generated.
+# Whilst it is possible to dump the effective Makefile, it can be
+# hard to see where parts come from. Set up the "peval" macro which
+# optionally prints out the generated makefile snippet and evaluate it.
+# Use "make PEVAL=1 all" to see the things being evaluated.
+ifeq (,${PEVAL})
+    define peval =
+    $(eval $1)
+    endef
+else
+    define peval =
+    $(eval $(info $1)$1)
+    endef
+endif
+
+# macro to add dependencies. Saves having to put all the OUT_DIR/logs in
+# The outer strip is to work around a gnu make 4.1 and earlier bug
+# It should not be needed.
+define adddep =
+$(strip $(call peval,components+= $(1) $(2))
+$(foreach comp,$(strip $2),$(call peval,${OUT_DIR}/logs/${1}: ${OUT_DIR}/logs/${comp}))
+)
+endef
+# End of internal stuff that is needed at the start of the file
+# -------------------------------------------------------------------------
+
+# Dependencies. These can be updated. Anything that is mentioned in
+# either the args to the adddep macro will be added to components.  as
+# an example there is no need for the adddep of lightning, as it
+# depends on nothing and at least one other component includes it.
+
+# Syntax. Up to the first comma everything is fixed. The "call" is a
+# keyword to gnu make. The "adddep" is the name of the variable containing
+# the macro.
+# The second comma delimited argument is the target.
+# The third comma delimited arg is the thing that the target depends on.
+# It is a space seperated list with zero or more elements.
+
+$(call adddep,amd_smi_lib,${ASAN_DEP})
+$(call adddep,aqlprofile,${ASAN_DEP} hsa)
+$(call adddep,clang-ocl,lightning rocm-cmake)
+$(call adddep,comgr,lightning devicelibs)
+$(call adddep,dbgapi,hsa comgr)
+$(call adddep,devicelibs,lightning)
+$(call adddep,hip_on_rocclr,${ASAN_DEP} rocclr rocprofiler-register)
+$(call adddep,hipcc,)
+$(call adddep,hipify_clang,hip_on_rocclr lightning)
+$(call adddep,hsa,${ASAN_DEP} thunk lightning devicelibs rocprofiler-register)
+$(call adddep,lightning,)
+$(call adddep,opencl_on_rocclr,${ASAN_DEP} rocclr)
+$(call adddep,openmp_extras,thunk lightning devicelibs hsa)
+$(call adddep,rdc,${ASAN_DEP} rocm_smi_lib hsa rocprofiler)
+$(call adddep,rocclr,${ASAN_DEP} hsa comgr hipcc rocprofiler-register)
+$(call adddep,rocm_bandwidth_test,${ASAN_DEP} hsa)
+$(call adddep,rocm_smi_lib,${ASAN_DEP})
+$(call adddep,rocm-cmake,${ASAN_DEP})
+$(call adddep,rocm-core,${ASAN_DEP})
+$(call adddep,rocm-gdb,dbgapi)
+$(call adddep,rocminfo,${ASAN_DEP} hsa)
+$(call adddep,rocprofiler-register,${ASAN_DEP})
+$(call adddep,rocprofiler,${ASAN_DEP} hsa roctracer aqlprofile opencl_on_rocclr hip_on_rocclr comgr dbgapi rocm_smi_lib)
+$(call adddep,rocr_debug_agent,${ASAN_DEP} hip_on_rocclr hsa dbgapi)
+$(call adddep,roctracer,${ASAN_DEP} hsa hip_on_rocclr)
+$(call adddep,thunk,${ASAN_DEP})
+
+# rocm-dev points to all possible last finish components of Stage1 build.
+rocm-dev-components :=rdc hipify_clang openmp_extras \
+	rocm-core amd_smi_lib hipcc clang-ocl \
+	rocm_bandwidth_test rocr_debug_agent rocm-gdb
+$(call adddep,rocm-dev,$(filter-out ${NOBUILD},${rocm-dev-components}))
+
+$(call adddep,amdmigraphx,hip_on_rocclr half rocblas miopen-hip lightning hipcc)
+$(call adddep,composable_kernel,lightning hipcc hip_on_rocclr rocm-cmake)
+$(call adddep,half,rocm-cmake)
+$(call adddep,hipblas,hip_on_rocclr rocblas rocsolver lightning hipcc)
+$(call adddep,hipblaslt,hip_on_rocclr openmp_extras hipblas lightning hipcc)
+$(call adddep,hipcub,hip_on_rocclr rocprim lightning hipcc)
+$(call adddep,hipfft,hip_on_rocclr openmp_extras rocfft rocrand hiprand lightning hipcc)
+$(call adddep,hipfort,rocblas hipblas rocsparse hipsparse rocfft hipfft rocrand hiprand rocsolver hipsolver lightning hipcc)
+$(call adddep,hiprand,hip_on_rocclr rocrand lightning hipcc)
+$(call adddep,hipsolver,hip_on_rocclr rocblas rocsolver rocsparse lightning hipcc hipsparse)
+$(call adddep,hipsparse,hip_on_rocclr rocsparse lightning hipcc)
+$(call adddep,hipsparselt,hip_on_rocclr hipsparse lightning hipcc openmp_extras)
+$(call adddep,hiptensor,hip_on_rocclr composable_kernel lightning hipcc)
+$(call adddep,miopen-deps,lightning hipcc)
+$(call adddep,miopen-hip,composable_kernel half hip_on_rocclr miopen-deps rocblas roctracer lightning hipcc)
+$(call adddep,mivisionx,amdmigraphx miopen-hip rpp lightning hipcc)
+$(call adddep,rccl,hip_on_rocclr hsa lightning hipcc rocm_smi_lib hipify_clang)
+$(call adddep,rocalution,rocblas rocsparse rocrand lightning hipcc)
+$(call adddep,rocblas,hip_on_rocclr openmp_extras lightning hipcc)
+$(call adddep,rocdecode,hip_on_rocclr lightning hipcc)
+$(call adddep,rocfft,hip_on_rocclr rocrand hiprand lightning hipcc openmp_extras)
+$(call adddep,rocmvalidationsuite,hip_on_rocclr hsa rocblas rocm-core lightning hipcc rocm_smi_lib)
+$(call adddep,rocprim,hip_on_rocclr lightning hipcc)
+$(call adddep,rocrand,hip_on_rocclr lightning hipcc)
+$(call adddep,rocsolver,hip_on_rocclr rocblas rocsparse lightning hipcc)
+$(call adddep,rocsparse,hip_on_rocclr rocprim lightning hipcc)
+$(call adddep,rocthrust,hip_on_rocclr rocprim lightning hipcc)
+$(call adddep,rocwmma,hip_on_rocclr rocblas lightning hipcc rocm-cmake rocm_smi_lib)
+$(call adddep,rpp,half lightning hipcc openmp_extras)
+
+
+# -------------------------------------------------------------------------
+# The rest of the file is internal
+# Do not pass jobserver params if -n build
+ifneq (,$(findstring n,${MAKEFLAGS}))
+RMAKE:=
+else
+RMAKE := +
+endif
+
+
+# disable the builtin rules
+.SUFFIXES:
+
+# Linear
+# include moredeps
+
+# A macro to define a toplevel target, add it to the 'all' target
+# Make it depend on the generated log. Generate the log of the build.
+
+# See if the macro is already defined, if so don't touch it.
+# As GNU make allows more than one makefile to be specified with "-f"
+# one could put an alternative definition of "toplevel" in a different
+# file or even the environment, and use the data in this file for other
+# purposes. Uses might include generating output in "dot" format for
+# showing the dependency graph, or having a wrapper script to run programs
+# to generate code quality tools.
+ifeq (${toplevel},)
+# { Start of test to see if toplevel is defined
+define toplevel =
+
+# The "target" make, this builds the package if it is out of date
+T_$1: ${OUT_DIR}/logs/$1 FRC
+	:              $1 built
+
+# The "upload" for $1, it uploads the packages for $1 to the central storage
+U_$1: T_$1 FRC
+	source $${INFRA_REPO}/envsetup.sh && $${INFRA_REPO}/upload_packages.sh "$1"
+	:		$1 uploaded
+
+# The "clean" for $1, it just marks the target as not existing so it will be built
+# in the future.
+C_$1: FRC
+	rm -f ${OUT_DIR}/logs/$1 ${OUT_DIR}/logs/$1.repackaged
+
+# parallel build {
+${OUT_DIR}/logs/$1: | ${OUT_DIR}/logs
+ifneq ($(wildcard ${OUT_DIR}/logs/$1.repackaged),)
+	@echo  Skipping build of $1 as it has already been repackaged
+	cat $$@.repackaged > $$@
+	rm -f $$@.repackaged
+else # } {
+	@echo  $1 started due to $$? | sed "s:${OUT_DIR}/logs/::g"
+# Build in a subshell so we get the time output
+# Pass in jobserver info using the RMAKE variable
+	${RMAKE}@( if set -x && source $${INFRA_REPO}/envsetup.sh && \
+	rm -f $$@.errors $$@ $$@.repackaged && \
+	$${INFRA_REPO}/build_$1.sh -c && source $${INFRA_REPO}/ccache-env-mathlib.sh && \
+	time bash -x $${INFRA_REPO}/build_$1.sh $${RELEASE_FLAG} $${SANITIZER_FLAG} && $${INFRA_REPO}/post_inst_pkg.sh "$1" ; \
+	then mv $$@.inprogress $$@ ; \
+	else mv $$@.inprogress $$@.errors ; echo Error in $1 >&2 ; exit 1 ;\
+	fi ) > $$@.inprogress 2>&1
+endif # }
+
+# end of toplevel macro
+endef
+# } End of test to see if toplevel is defined
+endif
+
+components:=$(sort $(components))
+
+# Create all the T_xxxx and C_xxxx targets
+$(call peval,$(foreach dep,$(strip ${components}),$(call toplevel,${dep})))
+
+# Add all the T_xxxx targets to "all"  except those listed in NOBUILD
+# Note this does not prohibit them from being built, it just means that
+# a build of "all" will not force them to be built directly
+# example command
+#  make -f jenkins-utils/scripts/Stage1.mk -j60
+
+##help all: Build everything
+all: $(addprefix T_,$(filter-out ${NOBUILD},${components}))
+	@echo All ROCm components built
+# Do not document this target
+upload: $(addprefix U_,${components})
+	@echo All ROCm components built and uploaded
+
+##help rocm-dev: Build a subset of ROCm
+rocm-dev: T_rocm-dev
+	@echo rocm-dev built
+
+${OUT_DIR}/logs:
+	sudo mkdir -p -m 775 "${ROCM_INSTALL_PATH}" && \
+	sudo chown -R "$(shell id -u):$(shell id -g)" "${ROCM_INSTALL_PATH}"
+	sudo chown -R "$(shell id -u):$(shell id -g)" "/home/$(shell id -un)"
+	mkdir -p "${@}"
+	mkdir -p ${HOME}/.ccache
+
+##help clean: remove the output directory and recreate it
+clean:
+	[ -n "${OUT_DIR}" ] && rm -rf "${OUT_DIR}"
+	mkdir -p ${OUT_DIR}/logs
+
+.SECONDARY: ${components:%=${OUT_DIR}/logs/%}
+
+.PHONY: all clean repack help list_components
+
+##help list_components: output the list of components
+##help : Hint make list_components | paste - - - | column -t
+list_components:
+	@echo "${components}" | sed 'y/ /\n/'
+
+##help help:	show this text
+help:
+	@sed -n 's/^##help //p' ${MAKEFILE_LIST} | \
+	    if type -t column > /dev/null ; then  column -s: -t ; else cat ; fi
+
+FRC:

tools/rocm-build/build_amd_smi_lib.sh

@@ -0,0 +1,145 @@
+#!/bin/bash
+
+source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
+
+printUsage() {
+    echo
+    echo "Usage: $(basename "${BASH_SOURCE}") [-c|-r|-h] [makeopts]"
+    echo
+    echo "Options:"
+    echo "  -c,  --clean              Removes all amd_smi build artifacts"
+    echo "  -r,  --release            Build non-debug version amd_smi (default is debug)"
+    echo "  -a,  --address_sanitizer  Enable address sanitizer"
+    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of type referred to by pkg_type"
+    echo "  -p,  --package <type>     Specify packaging format"
+    echo "  -h,  --help               Prints this help"
+    echo "Possible values for <type>:"
+    echo "  deb -> Debian format (default)"
+    echo "  rpm -> RPM format"
+    echo
+
+    return 0
+}
+
+PACKAGE_ROOT="$(getPackageRoot)"
+TARGET="build"
+
+PACKAGE_LIB=$(getLibPath)
+PACKAGE_INCLUDE="$(getIncludePath)"
+
+AMDSMI_BUILD_DIR=$(getBuildPath amdsmi)
+AMDSMI_PACKAGE_DEB_DIR="$(getPackageRoot)/deb/amdsmi"
+AMDSMI_PACKAGE_RPM_DIR="$(getPackageRoot)/rpm/amdsmi"
+AMDSMI_BUILD_TYPE="debug"
+BUILD_TYPE="Debug"
+
+MAKETARGET="deb"
+MAKEARG="$DASH_JAY O=$AMDSMI_BUILD_DIR"
+AMDSMI_MAKE_OPTS="$DASH_JAY O=$AMDSMI_BUILD_DIR -C $AMDSMI_BUILD_DIR"
+AMDSMI_PKG_NAME="amd-smi-lib"
+SHARED_LIBS="ON"
+CLEAN_OR_OUT=0;
+PKGTYPE="deb"
+
+VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
+eval set -- "$VALID_STR"
+
+while true ;
+do
+    case "$1" in
+        (-h | --help)
+                printUsage ; exit 0;;
+        (-c | --clean)
+                TARGET="clean" ; ((CLEAN_OR_OUT|=1)) ; shift ;;
+        (-r | --release)
+                BUILD_TYPE="RelWithDebInfo" ; shift ;;
+        (-a | --address_sanitizer)
+                set_asan_env_vars
+                set_address_sanitizer_on
+                # TODO - support standard option of passing cmake environment vars - CFLAGS,CXXFLAGS etc., to enable address sanitizer
+                ADDRESS_SANITIZER=true ; shift ;;
+        (-s | --static)
+                SHARED_LIBS="OFF" ; shift ;;
+        (-o | --outdir)
+                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
+        (-p | --package)
+                MAKETARGET="$2" ; shift 2;;
+        --)     shift; break;; # end delimiter
+        (*)
+                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
+    esac
+
+done
+
+RET_CONFLICT=1
+check_conflicting_options $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+if [ $RET_CONFLICT -ge 30 ]; then
+   print_vars $API_NAME $TARGET $BUILD_TYPE $SHARED_LIBS $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+   exit $RET_CONFLICT
+fi
+
+clean_amdsmi() {
+    rm -rf "$AMDSMI_BUILD_DIR"
+    rm -rf "$AMDSMI_PACKAGE_DEB_DIR"
+    rm -rf "$AMDSMI_PACKAGE_RPM_DIR"
+    rm -rf "$PACKAGE_ROOT/amd_smi"
+    rm -rf "$PACKAGE_INCLUDE/amd_smi"
+    rm -f $PACKAGE_LIB/libamd_smi.*
+    return 0
+}
+
+build_amdsmi() {
+    echo "Building AMDSMI"
+    echo "AMDSMI_BUILD_DIR: ${AMDSMI_BUILD_DIR}"
+    if [ ! -d "$AMDSMI_BUILD_DIR" ]; then
+        mkdir -p $AMDSMI_BUILD_DIR
+        pushd $AMDSMI_BUILD_DIR
+        print_lib_type $SHARED_LIBS
+        cmake \
+            -DBUILD_SHARED_LIBS=$SHARED_LIBS \
+            $(rocm_common_cmake_params) \
+            $(rocm_cmake_params) \
+            -DENABLE_LDCONFIG=OFF \
+            -DAMD_SMI_PACKAGE="${AMDSMI_PKG_NAME}" \
+            -DCPACK_PACKAGE_VERSION_MAJOR="1" \
+            -DCPACK_PACKAGE_VERSION_MINOR="$ROCM_LIBPATCH_VERSION" \
+            -DCPACK_PACKAGE_VERSION_PATCH="0" \
+            -DADDRESS_SANITIZER="$ADDRESS_SANITIZER" \
+            -DBUILD_TESTS=ON \
+            "$AMD_SMI_LIB_ROOT"
+        popd
+    fi
+
+    echo "Making amd_smi package:"
+    cmake --build "$AMDSMI_BUILD_DIR" -- $AMDSMI_MAKE_OPTS
+    cmake --build "$AMDSMI_BUILD_DIR" -- $AMDSMI_MAKE_OPTS install
+    cmake --build "$AMDSMI_BUILD_DIR" -- $AMDSMI_MAKE_OPTS package
+
+    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$AMDSMI_PACKAGE_DEB_DIR" $AMDSMI_BUILD_DIR/*.deb
+    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$AMDSMI_PACKAGE_RPM_DIR" $AMDSMI_BUILD_DIR/*.rpm
+}
+
+print_output_directory() {
+    case ${PKGTYPE} in
+        ("deb")
+            echo ${AMDSMI_PACKAGE_DEB_DIR};;
+        ("rpm")
+            echo ${AMDSMI_PACKAGE_RPM_DIR};;
+        (*)
+            echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
+    esac
+    exit
+}
+
+verifyEnvSetup
+
+case $TARGET in
+    (clean) clean_amdsmi ;;
+    (build) build_amdsmi ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
+esac
+
+echo "Operation complete"
+exit 0

tools/rocm-build/build_amdmigraphx.sh

@@ -0,0 +1,54 @@
+#!/bin/bash
+
+set -ex
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+
+set_component_src AMDMIGraphX
+
+build_amdmigraphx() {
+    echo "Start build"
+
+    cd $COMPONENT_SRC
+
+    pip3 install https://github.com/RadeonOpenCompute/rbuild/archive/master.tar.gz
+
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+         set_asan_env_vars
+         set_address_sanitizer_on
+    fi
+
+    if [ -n "$GPU_ARCHS" ]; then
+        GPU_TARGETS="$GPU_ARCHS"
+    else
+        GPU_TARGETS="gfx908;gfx90a;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101"
+    fi
+
+    mkdir -p ${BUILD_DIR} && rm -rf ${BUILD_DIR}/* && mkdir -p ${HOME}/amdmigraphx && rm -rf ${HOME}/amdmigraphx/*
+    rbuild package -d "${HOME}/amdmigraphx" -B "${BUILD_DIR}" \
+        --cxx="${ROCM_PATH}/llvm/bin/clang++" \
+        --cc="${ROCM_PATH}/llvm/bin/clang" \
+        $(rocm_common_cmake_params) \
+        -DCMAKE_MODULE_LINKER_FLAGS="-Wl,--enable-new-dtags -Wl,--rpath,$ROCM_LIB_RPATH" \
+	    -DGPU_TARGETS="${GPU_TARGETS}" \
+        -DCMAKE_INSTALL_RPATH=""
+
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    cd $BUILD_DIR && cmake --build . -- install -j${PROC}
+
+    show_build_cache_stats
+}
+
+clean_amdmigraphx() {
+    echo "Cleaning AMDMIGraphX build directory: ${BUILD_DIR} ${DEPS_DIR} ${PACKAGE_DIR}"
+    rm -rf "$BUILD_DIR" "$DEPS_DIR" "$PACKAGE_DIR"
+    echo "Done!"
+}
+
+stage2_command_args "$@"
+
+case $TARGET in
+    build) build_amdmigraphx ;;
+    outdir) print_output_directory ;;
+    clean) clean_amdmigraphx ;;
+    *) die "Invalid target $TARGET" ;;
+esac

tools/rocm-build/build_aqlprofile.sh

@@ -0,0 +1,150 @@
+#!/bin/bash
+
+source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
+
+printUsage() {
+    echo
+    echo "Usage: ${BASH_SOURCE##*/} [options ...]"
+    echo
+    echo "Options:"
+    echo "  -c,  --clean              Clean output and delete all intermediate work"
+    echo "  -p,  --package <type>     Specify packaging format"
+    echo "  -r,  --release            Make a release build instead of a debug build"
+    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
+            type referred to by pkg_type"
+    echo "  -h,  --help               Prints this help"
+    echo
+    echo "Possible values for <type>:"
+    echo "  deb -> Debian format (default)"
+    echo "  rpm -> RPM format"
+    echo
+
+    return 0
+}
+
+API_NAME="hsa-amd-aqlprofile"
+PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
+PROJ_NAME="$API_NAME"
+TARGET="build"
+BUILD_TYPE="Debug"
+MAKETARGET="deb"
+MAKE_OPTS="$DASH_JAY -C $BUILD_DIR"
+SHARED_LIBS="ON"
+CLEAN_OR_OUT=0
+MAKETARGET="deb"
+PKGTYPE="deb"
+
+VALID_STR=$(getopt -o hcro:p: --long help,clean,release,clean,outdir:,package: -- "$@")
+eval set -- "$VALID_STR"
+
+while true; do
+    case "$1" in
+    -h | --help)
+        printUsage
+        exit 0
+        ;;
+    -c | --clean)
+        TARGET="clean"
+        ((CLEAN_OR_OUT |= 1))
+        shift
+        ;;
+    -r | --release)
+        BUILD_TYPE="Release"
+        shift
+        ;;
+    -o | --outdir)
+        TARGET="outdir"
+        PKGTYPE=$2
+        OUT_DIR_SPECIFIED=1
+        ((CLEAN_OR_OUT |= 2))
+        shift 2
+        ;;
+    --)
+        shift
+        break
+        ;;
+    *)
+        echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] " >&2
+        exit 20
+        ;;
+    esac
+
+done
+
+copy_pkg_files_to_rocm() {
+    local comp_folder=$1
+    local comp_pkg_name=$2
+
+    cd "${OUT_DIR}/${PKGTYPE}/${comp_folder}"|| exit 2
+    if [ "${PKGTYPE}" = 'deb' ]; then
+        dpkg-deb -x ${comp_pkg_name}_*.deb pkg/
+    else
+        mkdir pkg && pushd pkg/ || exit 2
+        if [[ "${comp_pkg_name}" != *-dev* ]]; then
+            rpm2cpio ../${comp_pkg_name}-*.rpm | cpio -idmv
+        else
+            rpm2cpio ../${comp_pkg_name}el-*.rpm | cpio -idmv
+        fi
+        popd || exit 2
+    fi
+    ls ./pkg -alt
+    sudo cp -r ./pkg/*/rocm*/* "${ROCM_PATH}" || exit 2
+    rm -rf pkg/
+}
+
+clean() {
+    echo "Cleaning $PROJ_NAME package"
+    rm -rf "$PACKAGE_DEB"
+}
+
+build() {
+    echo "Downloading $PROJ_NAME" package
+    if [ "$DISTRO_NAME" = ubuntu ]; then
+        mkdir -p "$PACKAGE_DEB"
+        local rocm_ver=${ROCM_VERSION}
+        if [ ${ROCM_VERSION##*.} = 0 ]; then
+            rocm_ver=${ROCM_VERSION%.*}
+        fi
+        local url="https://repo.radeon.com/rocm/apt/${rocm_ver}/pool/main/h/${API_NAME}/"
+        local package
+        package=$(curl -s "$url" | grep -Po 'href="\K[^"]*' | grep "${DISTRO_RELEASE}" | head -n 1)
+
+        if [ -z "$package" ]; then
+            echo "No package found for Ubuntu version $DISTRO_RELEASE"
+            exit 1
+        fi
+
+        wget -t3 -P "$PACKAGE_DEB" "${url}${package}"
+        copy_pkg_files_to_rocm ${API_NAME} ${API_NAME}
+    else
+        echo "$DISTRO_ID is not supported..."
+        exit 2
+    fi
+
+    echo "Installing $PROJ_NAME" package
+}
+
+print_output_directory() {
+    case ${PKGTYPE} in
+    "deb")
+        echo ${PACKAGE_DEB}
+        ;;
+    "rpm")
+        echo ${PACKAGE_RPM}
+        ;;
+    *)
+        echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2
+        exit 1
+        ;;
+    esac
+    exit
+}
+
+case "$TARGET" in
+clean) clean ;;
+build) build ;;
+outdir) print_output_directory ;;
+*) die "Invalid target $TARGET" ;;
+esac
+
+echo "Operation complete"

tools/rocm-build/build_clang-ocl.sh

@@ -0,0 +1,136 @@
+#!/bin/bash
+
+source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
+
+printUsage() {
+    echo
+    echo "Usage: $(basename "${BASH_SOURCE}") [-c|-r|-h] [makeopts]"
+    echo
+    echo "Options:"
+    echo "  -c,  --clean              Removes all clang-ocl build artifacts"
+    echo "  -r,  --release            Build non-debug version clang-ocl (default is debug)"
+    echo "  -a,  --address_sanitizer  Enable address sanitizer"
+    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
+                                      type referred to by pkg_type"
+    echo "  -h,  --help               Prints this help"
+    echo "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
+    echo
+
+    return 0
+}
+
+
+TARGET="build"
+CLANG_OCL_DEST="$(getBinPath)"
+CLANG_OCL_SRC_ROOT="$CLANG_OCL_ROOT"
+CLANG_OCL_BUILD_DIR="$(getBuildPath clang-ocl)"
+
+MAKEARG="$DASH_JAY"
+PACKAGE_ROOT="$(getPackageRoot)"
+PACKAGE_UTILS="$(getUtilsPath)"
+CLANG_OCL_PACKAGE_DEB="$PACKAGE_ROOT/deb/clang-ocl"
+CLANG_OCL_PACKAGE_RPM="$PACKAGE_ROOT/rpm/clang-ocl"
+BUILD_TYPE="Debug"
+SHARED_LIBS="ON"
+CLEAN_OR_OUT=0;
+MAKETARGET="deb"
+PKGTYPE="deb"
+
+
+VALID_STR=`getopt -o hcraso:g: --long help,clean,release,clean,static,address_sanitizer,outdir:,gpu_list: -- "$@"`
+eval set -- "$VALID_STR"
+
+while true ;
+do
+    case "$1" in
+        (-h | --help)
+                printUsage ; exit 0;;
+        (-c | --clean)
+                TARGET="clean" ; ((CLEAN_OR_OUT|=1)) ; shift ;;
+        (-r | --release)
+                MAKEARG="$MAKEARG BUILD_TYPE=rel" ;  BUILD_TYPE="Release" ; shift ;;
+        (-a | --address_sanitizer)
+                set_asan_env_vars
+                set_address_sanitizer_on ; shift ;;
+        (-s | --static)
+                SHARED_LIBS="OFF" ; shift ;;
+        (-o | --outdir)
+                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
+        (-g | --gpu_list )
+                GPU_LIST=$2; shift 2 ;;
+        --)     shift; break;;
+        (*)
+                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
+    esac
+
+done
+
+RET_CONFLICT=1
+check_conflicting_options $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+if [ $RET_CONFLICT -ge 30 ]; then
+   print_vars $API_NAME $TARGET $BUILD_TYPE $SHARED_LIBS $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+   exit $RET_CONFLICT
+fi
+
+clean_clang-ocl() {
+    echo "Removing clang-ocl"
+    rm -rf $CLANG_OCL_DEST/clang-ocl
+    rm -rf $CLANG_OCL_BUILD_DIR
+    rm -rf $CLANG_OCL_PACKAGE_DEB
+    rm -rf $CLANG_OCL_PACKAGE_RPM
+}
+
+build_clang-ocl() {
+    if [ ! -d "$CLANG_OCL_BUILD_DIR" ]; then
+        mkdir -p $CLANG_OCL_BUILD_DIR
+        pushd $CLANG_OCL_BUILD_DIR
+
+        if [ -e $PACKAGE_ROOT/lib/bitcode/opencl.amdgcn.bc ]; then
+            BC_DIR="$ROCM_INSTALL_PATH/lib"
+        else
+            BC_DIR="$ROCM_INSTALL_PATH/amdgcn/bitcode"
+        fi
+
+        cmake \
+            $(rocm_cmake_params) \
+            -DDISABLE_CHECKS="ON" \
+            -DCLANG_BIN="$ROCM_INSTALL_PATH/llvm/bin" \
+            -DBITCODE_DIR="$BC_DIR" \
+	    $(rocm_common_cmake_params) \
+            -DCPACK_SET_DESTDIR="OFF" \
+            $CLANG_OCL_SRC_ROOT
+
+        echo "Making clang-ocl:"
+        cmake --build . -- $MAKEARG
+        cmake --build . -- $MAKEARG install
+        cmake --build . -- $MAKEARG package
+        popd
+    fi
+
+    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$CLANG_OCL_PACKAGE_DEB" $CLANG_OCL_BUILD_DIR/rocm-clang-ocl*.deb
+    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$CLANG_OCL_PACKAGE_RPM" $CLANG_OCL_BUILD_DIR/rocm-clang-ocl*.rpm
+}
+
+
+print_output_directory() {
+     case ${PKGTYPE} in
+         ("deb")
+             echo ${CLANG_OCL_PACKAGE_DEB};;
+         ("rpm")
+             echo ${CLANG_OCL_PACKAGE_RPM};;
+         (*)
+             echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
+     esac
+     exit
+}
+
+case $TARGET in
+    (clean) clean_clang-ocl ;;
+    (build) build_clang-ocl ;;
+   (outdir) print_output_directory ;;
+        (*) die "Invalid target $TARGET" ;;
+esac
+
+echo "Operation complete"
+exit 0
+

tools/rocm-build/build_comgr.sh

@@ -0,0 +1,151 @@
+#!/bin/bash
+source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
+
+printUsage() {
+    echo
+    echo "Usage: $(basename "${BASH_SOURCE}") [options ...]"
+    echo
+    echo "Options:"
+    echo "  -c,  --clean              Clean output and delete all intermediate work"
+    echo "  -p,  --package <type>     Specify packaging format"
+    echo "  -r,  --release            Make a release build instead of a debug build"
+    echo "  -a,  --address_sanitizer  Enable address sanitizer"
+    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
+        type referred to by pkg_type"
+    echo "  -h,  --help               Prints this help"
+    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -l,  --link_llvm_static   Link to LLVM statically.  Default is to dynamically link to LLVM; requires that LLVM dylibs are created."
+    echo
+    echo "Possible values for <type>:"
+    echo "  deb -> Debian format (default)"
+    echo "  rpm -> RPM format"
+    echo
+
+    return 0
+}
+
+API_NAME=amd_comgr
+PROJ_NAME=$API_NAME
+LIB_NAME=lib${API_NAME}
+TARGET=build
+MAKETARGET=deb
+PACKAGE_ROOT=$(getPackageRoot)
+PACKAGE_LIB=$(getLibPath)
+PACKAGE_INCLUDE=$(getIncludePath)
+BUILD_DIR=$(getBuildPath $API_NAME)
+PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
+PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
+PACKAGE_PREFIX=$ROCM_INSTALL_PATH
+BUILD_TYPE=Debug
+MAKE_OPTS="$DASH_JAY CTEST_OUTPUT_ON_FAILURE=1 -C $BUILD_DIR"
+VERBOSE_OPTS="AMD_COMGR_EMIT_VERBOSE_LOGS=1 AMD_COMGR_REDIRECT_LOGS=stdout"
+SHARED_LIBS="ON"
+CLEAN_OR_OUT=0;
+PKGTYPE="deb"
+MAKETARGET="deb"
+
+LINK_LLVM_DYLIB="OFF"
+
+VALID_STR=`getopt -o hcraslo:p: --long help,clean,release,address_sanitizer,static,link_llvm_static,outdir:,package: -- "$@"`
+eval set -- "$VALID_STR"
+
+while true ;
+do
+    case "$1" in
+        (-h | --help)
+                printUsage ; exit 0;;
+        (-c | --clean)
+                TARGET="clean" ; ((CLEAN_OR_OUT|=1)) ; shift ;;
+        (-r | --release)
+                BUILD_TYPE="RelWithDebInfo" ; shift ;;
+        (-a | --address_sanitizer)
+                set_asan_env_vars
+                set_address_sanitizer_on ; shift ;;
+        (-s | --static)
+                SHARED_LIBS="OFF" ; shift ;;
+        (-l | --link_llvm_static)
+                LINK_LLVM_DYLIB="OFF"; shift ;;
+        (-o | --outdir)
+                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
+        (-p | --package)
+                MAKETARGET="$2" ; shift 2;;
+        --)     shift; break;; # end delimiter
+        (*)
+                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
+    esac
+
+done
+
+RET_CONFLICT=1
+check_conflicting_options $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+if [ $RET_CONFLICT -ge 30 ]; then
+   print_vars $API_NAME $TARGET $BUILD_TYPE $SHARED_LIBS $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+   exit $RET_CONFLICT
+fi
+
+clean() {
+    echo "Cleaning $PROJ_NAME"
+
+    rm -rf $BUILD_DIR
+    rm -rf $PACKAGE_DEB
+    rm -rf $PACKAGE_RPM
+    rm -rf $PACKAGE_ROOT/${PROJ_NAME}
+    rm -rf $PACKAGE_LIB/${LIB_NAME}*
+}
+
+build() {
+    echo "Building $PROJ_NAME"
+
+    mkdir -p "$BUILD_DIR"
+    pushd "$BUILD_DIR"
+    if [ "$SHARED_LIBS" == "OFF" ]
+    then
+       echo " Building Archive "
+    else
+       echo " Building Shared Object "
+    fi
+
+    cmake \
+        $(rocm_cmake_params) \
+        -DBUILD_SHARED_LIBS=$SHARED_LIBS \
+        $(rocm_common_cmake_params) \
+        -DCMAKE_DISABLE_FIND_PACKAGE_hip=TRUE \
+        -DCMAKE_CTEST_ARGUMENTS="--exclude-regex;multithread_test" \
+        -DLLVM_LINK_LLVM_DYLIB=$LINK_LLVM_DYLIB \
+        -DLLVM_ENABLE_LIBCXX=$LINK_LLVM_DYLIB \
+        $COMGR_ROOT
+
+    popd
+    cmake --build "$BUILD_DIR" -- $MAKE_OPTS
+    cmake --build "$BUILD_DIR" -- $MAKE_OPTS $VERBOSE_OPTS test
+    cmake --build "$BUILD_DIR" -- $MAKE_OPTS install
+    cmake --build "$BUILD_DIR" -- $MAKE_OPTS package
+
+    mkdir -p $PACKAGE_LIB
+    cp -R $BUILD_DIR/${LIB_NAME}* $PACKAGE_LIB
+
+    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_DEB" $BUILD_DIR/comgr*.deb
+    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_RPM" $BUILD_DIR/comgr*.rpm
+}
+
+print_output_directory() {
+    case ${PKGTYPE} in
+        ("deb")
+            echo ${PACKAGE_DEB};;
+        ("rpm")
+            echo ${PACKAGE_RPM};;
+        (*)
+            echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
+    esac
+    exit
+}
+verifyEnvSetup
+
+case $TARGET in
+    (clean) clean ;;
+    (build) build ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
+esac
+
+echo "Operation complete"

tools/rocm-build/build_composable_kernel.sh

@@ -0,0 +1,185 @@
+#!/bin/bash
+
+set -ex
+
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+
+set_component_src composable_kernel
+
+build_miopen_ck() {
+    echo "Start Building Composable Kernel"
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+       set_asan_env_vars
+       set_address_sanitizer_on
+    fi
+
+    cd $COMPONENT_SRC
+    mkdir "$BUILD_DIR" && cd "$BUILD_DIR"
+
+    if [ -n "$GPU_ARCHS" ]; then
+        GPU_TARGETS="$GPU_ARCHS"
+    else
+        GPU_TARGETS="gfx908;gfx90a;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101"
+    fi
+
+    if [ "${ASAN_CMAKE_PARAMS}" == "true" ] ; then
+        cmake -DBUILD_DEV=OFF \
+            -DCMAKE_BUILD_TYPE=${BUILD_TYPE:-'RelWithDebInfo'} \
+            -DCMAKE_CXX_COMPILER=${ROCM_PATH}/llvm/bin/clang++ \
+            -DCMAKE_CXX_FLAGS=" -O3 " \
+            -DCMAKE_PREFIX_PATH="${ROCM_PATH%-*}/lib/cmake;${ROCM_PATH%-*}/$ASAN_LIBDIR;${ROCM_PATH%-*}/llvm;${ROCM_PATH%-*}" \
+            -DCMAKE_SHARED_LINKER_FLAGS_INIT="-Wl,--enable-new-dtags,--rpath,$ROCM_ASAN_LIB_RPATH" \
+            -DCMAKE_EXE_LINKER_FLAGS_INIT="-Wl,--enable-new-dtags,--rpath,$ROCM_ASAN_EXE_RPATH" \
+            -DCMAKE_VERBOSE_MAKEFILE=1 \
+            -DCMAKE_INSTALL_RPATH_USE_LINK_PATH=FALSE \
+            -DCMAKE_INSTALL_PREFIX=${ROCM_PATH} \
+            -DCMAKE_PACKAGING_INSTALL_PREFIX=${ROCM_PATH} \
+            -DBUILD_FILE_REORG_BACKWARD_COMPATIBILITY=OFF \
+            -DROCM_SYMLINK_LIBS=OFF \
+            -DCPACK_PACKAGING_INSTALL_PREFIX=${ROCM_PATH} \
+            -DROCM_DISABLE_LDCONFIG=ON \
+            -DROCM_PATH=${ROCM_PATH} \
+            -DCPACK_GENERATOR="${PKGTYPE^^}" \
+            ${LAUNCHER_FLAGS} \
+            -DINSTANCES_ONLY=ON \
+            -DENABLE_ASAN_PACKAGING=true \
+            -DAMDGPU_TARGETS=${GPU_TARGETS} \
+            "$COMPONENT_SRC"
+    else
+        cmake -DBUILD_DEV=OFF \
+            -DCMAKE_BUILD_TYPE=Release \
+            -DCMAKE_CXX_COMPILER=${ROCM_PATH}/llvm/bin/clang++ \
+            -DCMAKE_CXX_FLAGS=" -O3 " \
+            -DCMAKE_PREFIX_PATH=${ROCM_PATH%-*} \
+            -DCMAKE_SHARED_LINKER_FLAGS_INIT='-Wl,--enable-new-dtags,--rpath,$ORIGIN' \
+            -DCMAKE_EXE_LINKER_FLAGS_INIT='-Wl,--enable-new-dtags,--rpath,$ORIGIN/../lib' \
+            -DCMAKE_VERBOSE_MAKEFILE=1 \
+            -DCMAKE_INSTALL_RPATH_USE_LINK_PATH=FALSE \
+            -DCMAKE_INSTALL_PREFIX=${ROCM_PATH} \
+            -DCMAKE_PACKAGING_INSTALL_PREFIX=${ROCM_PATH} \
+            -DBUILD_FILE_REORG_BACKWARD_COMPATIBILITY=OFF \
+            -DROCM_SYMLINK_LIBS=OFF \
+            -DCPACK_PACKAGING_INSTALL_PREFIX=${ROCM_PATH} \
+            -DROCM_DISABLE_LDCONFIG=ON \
+            -DROCM_PATH=${ROCM_PATH} \
+            -DCPACK_GENERATOR="${PKGTYPE^^}" \
+            ${LAUNCHER_FLAGS} \
+            -DINSTANCES_ONLY=ON \
+            -DAMDGPU_TARGETS=${GPU_TARGETS} \
+            "$COMPONENT_SRC"
+    fi
+
+    cmake --build . -- -j${PROC} package
+    cmake --build "$BUILD_DIR" -- install
+    mkdir -p $PACKAGE_DIR && cp ./*.${PKGTYPE} $PACKAGE_DIR
+    rm -rf *
+}
+
+unset_asan_env_vars() {
+    ASAN_CMAKE_PARAMS="false"
+    export ADDRESS_SANITIZER="OFF"
+    export LD_LIBRARY_PATH=""
+    export ASAN_OPTIONS=""
+}
+
+set_address_sanitizer_off() {
+    export CFLAGS=""
+    export CXXFLAGS=""
+    export LDFLAGS=""
+}
+
+build_miopen_ckProf() {
+    ENABLE_ADDRESS_SANITIZER=false
+    echo "Start Building Composable Kernel Profiler"
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+       set_asan_env_vars
+       set_address_sanitizer_on
+    else
+       unset_asan_env_vars
+       set_address_sanitizer_off
+    fi
+
+    cd $COMPONENT_SRC
+    cd "$BUILD_DIR"
+    rm -rf *
+
+    architectures='gfx10 gfx11 gfx90 gfx94'
+    if [ -n "$GPU_ARCHS" ]; then
+        architectures=$(echo ${GPU_ARCHS} | awk -F';' '{for(i=1;i<=NF;i++) a[substr($i,1,5)]} END{for(i in a) printf i" "}')
+    else
+        architectures='gfx10 gfx11 gfx90 gfx94'
+    fi
+
+    for arch in ${architectures}
+        do
+            if [ "${ASAN_CMAKE_PARAMS}" == "true" ] ; then
+                cmake -DBUILD_DEV=OFF \
+                    -DCMAKE_PREFIX_PATH="${ROCM_PATH%-*}/lib/cmake;${ROCM_PATH%-*}/$ASAN_LIBDIR;${ROCM_PATH%-*}/llvm;${ROCM_PATH%-*}" \
+                    -DCMAKE_BUILD_TYPE=${BUILD_TYPE:-'RelWithDebInfo'} \
+                    -DCMAKE_SHARED_LINKER_FLAGS_INIT="-Wl,--enable-new-dtags,--rpath,$ROCM_ASAN_LIB_RPATH" \
+                    -DCMAKE_EXE_LINKER_FLAGS_INIT="-Wl,--enable-new-dtags,--rpath,$ROCM_ASAN_EXE_RPATH" \
+                    -DCMAKE_VERBOSE_MAKEFILE=1 \
+                    -DCMAKE_INSTALL_RPATH_USE_LINK_PATH=FALSE \
+                    -DCMAKE_INSTALL_PREFIX="${ROCM_PATH}" \
+                    -DCMAKE_PACKAGING_INSTALL_PREFIX="${ROCM_PATH}" \
+                    -DBUILD_FILE_REORG_BACKWARD_COMPATIBILITY=OFF \
+                    -DROCM_SYMLINK_LIBS=OFF \
+                    -DCPACK_PACKAGING_INSTALL_PREFIX="${ROCM_PATH}" \
+                    -DROCM_DISABLE_LDCONFIG=ON \
+                    -DROCM_PATH="${ROCM_PATH}" \
+                    -DCPACK_GENERATOR="${PKGTYPE^^}" \
+                    -DCMAKE_CXX_COMPILER="${ROCM_PATH}/llvm/bin/clang++" \
+                    -DCMAKE_C_COMPILER="${ROCM_PATH}/llvm/bin/clang" \
+                    ${LAUNCHER_FLAGS} \
+                    -DPROFILER_ONLY=ON \
+                    -DENABLE_ASAN_PACKAGING=true \
+                    -DGPU_ARCH="${arch}" \
+                    "$COMPONENT_SRC"
+            else
+                cmake -DBUILD_DEV=OFF \
+                    -DCMAKE_PREFIX_PATH="${ROCM_PATH%-*}" \
+                    -DCMAKE_BUILD_TYPE=Release \
+                    -DCMAKE_SHARED_LINKER_FLAGS_INIT='-Wl,--enable-new-dtags,--rpath,$ORIGIN' \
+                    -DCMAKE_EXE_LINKER_FLAGS_INIT='-Wl,--enable-new-dtags,--rpath,$ORIGIN/../lib' \
+                    -DCMAKE_VERBOSE_MAKEFILE=1 \
+                    -DCMAKE_INSTALL_RPATH_USE_LINK_PATH=FALSE \
+                    -DCMAKE_INSTALL_PREFIX="${ROCM_PATH}" \
+                    -DCMAKE_PACKAGING_INSTALL_PREFIX="${ROCM_PATH}" \
+                    -DBUILD_FILE_REORG_BACKWARD_COMPATIBILITY=OFF \
+                    -DROCM_SYMLINK_LIBS=OFF \
+                    -DCPACK_PACKAGING_INSTALL_PREFIX="${ROCM_PATH}" \
+                    -DROCM_DISABLE_LDCONFIG=ON \
+                    -DROCM_PATH="${ROCM_PATH}" \
+                    -DCPACK_GENERATOR="${PKGTYPE^^}" \
+                    -DCMAKE_CXX_COMPILER="${ROCM_PATH}/llvm/bin/clang++" \
+                    -DCMAKE_C_COMPILER="${ROCM_PATH}/llvm/bin/clang" \
+                    ${LAUNCHER_FLAGS} \
+                    -DPROFILER_ONLY=ON \
+                    -DGPU_ARCH="${arch}" \
+                    "$COMPONENT_SRC"
+            fi
+
+            cmake --build . -- -j${PROC} package
+            cp ./*ckprofiler*.${PKGTYPE} $PACKAGE_DIR
+            rm -rf *
+        done
+    rm -rf _CPack_Packages/ && find -name '*.o' -delete
+
+    echo "Finished building Composable Kernel"
+    show_build_cache_stats
+}
+
+clean_miopen_ck() {
+    echo "Cleaning MIOpen-CK build directory: ${BUILD_DIR} ${PACKAGE_DIR}"
+    rm -rf "$BUILD_DIR" "$PACKAGE_DIR"
+    echo "Done!"
+}
+
+stage2_command_args "$@"
+
+case $TARGET in
+    build) build_miopen_ck; build_miopen_ckProf;;
+    outdir) print_output_directory ;;
+    clean) clean_miopen_ck ;;
+    *) die "Invalid target $TARGET" ;;
+esac

tools/rocm-build/build_dbgapi.sh

@@ -0,0 +1,159 @@
+#!/bin/bash
+source "${BASH_SOURCE%/*}/compute_utils.sh"
+
+printUsage() {
+    echo
+    echo "Usage: $(basename "${BASH_SOURCE[0]}") [options ...]"
+    echo
+    echo "Options:"
+    echo "  -c,  --clean              Clean output and delete all intermediate work"
+    echo "  -p,  --package <type>     Specify packaging format"
+    echo "  -r,  --release            Make a release build instead of a debug build"
+    echo "       --enable-assertions  Enable assertions"
+    echo "  -a,  --address_sanitizer  Enable address sanitizer"
+    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
+            type referred to by pkg_type"
+    echo "  -h,  --help               Prints this help"
+    echo "  -M,  --skip_man_pages     Do not build the 'docs' target"
+    echo "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
+    echo
+    echo "Possible values for <type>:"
+    echo "  deb -> Debian format (default)"
+    echo "  rpm -> RPM format"
+    echo
+
+    return 0
+}
+
+API_NAME=rocm-dbgapi
+AMD_DBGAPI_NAME=amd-dbgapi
+MAKEINSTALL_MANIFEST=makeinstall_manifest.txt
+PROJ_NAME=$API_NAME
+LIB_NAME=lib${API_NAME}.so
+TARGET=build
+MAKETARGET=deb
+PACKAGE_ROOT=$(getPackageRoot)
+PACKAGE_LIB=$(getLibPath)
+PACKAGE_INCLUDE=$(getIncludePath)
+BUILD_DIR=$(getBuildPath $API_NAME)
+PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
+PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
+BUILD_TYPE=Debug
+MAKE_OPTS=($DASH_JAY -C "$BUILD_DIR")
+SHARED_LIBS="ON"
+CLEAN_OR_OUT=0;
+MAKETARGET="deb"
+PKGTYPE="deb"
+DODOCSBUILD=true
+
+VALID_STR=$(getopt -o hcraso:p:M --long help,clean,release,enable-assertions,static,address_sanitizer,outdir:,package:skip_man_pages -- "$@")
+eval set -- "$VALID_STR"
+
+while true ;
+do
+    case "$1" in
+        (-h | --help)
+                printUsage ; exit 0;;
+        (-c | --clean)
+                TARGET="clean" ; ((CLEAN_OR_OUT|=1)) ;;
+        (-r | --release)
+                ENABLE_ASSERTIONS=${ENABLE_ASSERTIONS:-"Off"} ;
+                BUILD_TYPE="RelWithDebInfo" ;;
+        ( --enable-assertions)
+                ENABLE_ASSERTIONS="On" ;;
+        (-a | --address_sanitizer)
+                set_asan_env_vars
+                set_address_sanitizer_on ;;
+        (-s | --static)
+                SHARED_LIBS="OFF" ;;
+        (-o | --outdir)
+                TARGET="outdir"; PKGTYPE=$2 ; ((CLEAN_OR_OUT|=2)) ; shift 1 ;;
+        (-M | --skip_man_pages) DODOCSBUILD=false;;
+        (-p | --package)
+                MAKETARGET="$2" ; shift 1;;
+        --)     shift; break;;
+        (*)
+                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
+    esac
+    shift
+
+done
+
+check_conflicting_options $CLEAN_OR_OUT "$PKGTYPE" "$MAKETARGET"
+if [ "$RET_CONFLICT" -ge 30 ]; then
+   print_vars "$API_NAME" "$TARGET" "$BUILD_TYPE" "$SHARED_LIBS" "$CLEAN_OR_OUT" "$PKGTYPE" "$MAKETARGET"
+   exit "$RET_CONFLICT"
+fi
+
+clean() {
+    echo "Cleaning $PROJ_NAME"
+    if [ -e "$BUILD_DIR/$MAKEINSTALL_MANIFEST" ] ; then
+        xargs rm -f < "$BUILD_DIR/$MAKEINSTALL_MANIFEST"
+    fi
+    rm -rf "$BUILD_DIR"
+    rm -rf "$PACKAGE_DEB"
+    rm -rf "$PACKAGE_RPM"
+    rm -rf "${PACKAGE_ROOT:?}/${PROJ_NAME}"
+    rm -rf "${PACKAGE_LIB:?}/${LIB_NAME}"*
+    rm -rf "${PACKAGE_LIB:?}/cmake/${AMD_DBGAPI_NAME}"
+    rm -rf "${PACKAGE_INCLUDE:?}/${AMD_DBGAPI_NAME}"
+}
+
+build() {
+    if [ ! -e "$ROCM_DBGAPI_ROOT/CMakeLists.txt" ]
+    then
+           echo " No $ROCM_DBGAPI_ROOT/CMakeLists.txt file, skipping rocm-dbgapi" >&2
+           echo " No $ROCM_DBGAPI_ROOT/CMakeLists.txt file, skipping rocm-dbgapi"
+       exit 0
+    fi
+    echo "Building $PROJ_NAME"
+
+    mkdir -p "$BUILD_DIR"
+    pushd "$BUILD_DIR" || exit 99
+
+    cmake \
+        $(rocm_cmake_params) \
+        $(rocm_common_cmake_params) \
+        -DENABLE_ASSERTIONS=${ENABLE_ASSERTIONS:-"On"} \
+        "$ROCM_DBGAPI_ROOT"
+
+    popd || exit 99
+    cmake --build "$BUILD_DIR" -- "${MAKE_OPTS[@]}"
+    "$DODOCSBUILD" && cmake --build "$BUILD_DIR" -- "${MAKE_OPTS[@]}" doc
+    cmake --build "$BUILD_DIR" -- "${MAKE_OPTS[@]}" install
+
+    mv "$BUILD_DIR/install_manifest.txt" "$BUILD_DIR/$MAKEINSTALL_MANIFEST"
+    cmake --build "$BUILD_DIR" -- "${MAKE_OPTS[@]}" package
+
+    mkdir -p "$PACKAGE_LIB"
+    (
+        shopt -s nullglob
+        cp -R "$BUILD_DIR/lib/${LIB_NAME}"* "$BUILD_DIR/${LIB_NAME}"* "$PACKAGE_LIB"
+    )
+    
+    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_DEB" "$BUILD_DIR/${API_NAME}"*.deb
+    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_RPM" "$BUILD_DIR/${API_NAME}"*.rpm
+}
+
+print_output_directory() {
+    case ${PKGTYPE} in
+        ("deb")
+            echo ${PACKAGE_DEB};;
+        ("rpm")
+            echo ${PACKAGE_RPM};;
+        (*)
+            echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
+    esac
+    exit
+}
+
+verifyEnvSetup
+
+case $TARGET in
+    (clean) clean ;;
+    (build) build ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
+esac
+
+echo "Operation complete"

tools/rocm-build/build_devicelibs.sh

@@ -0,0 +1,140 @@
+#!/bin/bash
+
+source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
+
+printUsage() {
+    echo
+    echo "Usage: $(basename "${BASH_SOURCE}") [options ...]"
+    echo
+    echo "Options:"
+    echo "  -c,  --clean              Clean output and delete all intermediate work"
+    echo "  -r,  --release            Build a release version of the package"
+    echo "  -a,  --address_sanitizer  Enable address sanitizer"
+    echo "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
+    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
+    type referred to by pkg_type"
+    echo "  -h,  --help             Prints this help"
+    echo
+    echo
+
+    return 0
+}
+PACKAGE_ROOT="$(getPackageRoot)"
+PACKAGE_BIN="$(getBinPath)"
+PACKAGE_LIB="$(getLibPath)"
+BUILD_PATH="$(getBuildPath devicelibs)"
+INSTALL_PATH="$(getPackageRoot)"
+LIGHTNING_BUILD_PATH="$(getBuildPath lightning)"
+DEB_PATH="$(getDebPath devicelibs)"
+RPM_PATH="$(getRpmPath devicelibs)"
+AMDGCN_LIB_PATH="$PACKAGE_ROOT/amdgcn/bitcode/"
+
+TARGET="build"
+MAKEOPTS="$DASH_JAY"
+SHARED_LIBS="ON"
+CLEAN_OR_OUT=0;
+PKGTYPE="deb"
+MAKETARGET="deb"
+
+VALID_STR=`getopt -o hcraso: --long help,clean,release,static,address_sanitizer,outdir: -- "$@"`
+eval set -- "$VALID_STR"
+
+while true ;
+do
+    case "$1" in
+        (-h | --help)
+                printUsage ; exit 0;;
+        (-c | --clean)
+                TARGET="clean" ; ((CLEAN_OR_OUT|=1)) ; shift ;;
+        (-r | --release)
+                BUILD_TYPE="Release" ; shift ;;
+        (-a | --address_sanitizer)
+                ASAN_CMAKE_PARAMS="true"
+                ack_and_ignore_asan ; shift ;;
+        (-s | --static)
+                SHARED_LIBS="OFF" ; shift ;;
+        (-o | --outdir)
+                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
+        --)     shift; break;; # end delimiter
+        (*)
+                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
+    esac
+
+done
+
+RET_CONFLICT=1
+check_conflicting_options $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+if [ $RET_CONFLICT -ge 30 ]; then
+   print_vars $API_NAME $TARGET $BUILD_TYPE $SHARED_LIBS $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+   exit $RET_CONFLICT
+fi
+
+
+clean_devicelibs() {
+    rm -rf "$BUILD_PATH"
+    rm -f $PACKAGE_LIB/hc*.bc
+    rm -f $PACKAGE_LIB/irif*.bc
+    rm -f $PACKAGE_LIB/ockl*.bc
+    rm -f $PACKAGE_LIB/oclc*.bc
+    rm -f $PACKAGE_LIB/ocml*.bc
+    rm -f $PACKAGE_LIB/opencl*.bc
+    rm -f $PACKAGE_LIB/openmp*.bc
+    rm -rf $PACKAGE_ROOT/amdgcn
+    rm -rf "$DEB_PATH"
+    rm -rf "$RPM_PATH"
+}
+
+build_devicelibs() {
+    mkdir -p "$BUILD_PATH"
+    pushd "$BUILD_PATH"
+
+    local clangResourceDir="$($LIGHTNING_BUILD_PATH/bin/clang -print-resource-dir)"
+    local clangResourceVer=${clangResourceDir#*lib/clang/}
+    local bitcodeInstallLoc="lib/llvm/lib/clang/${clangResourceVer}/lib"
+
+    export LLVM_BUILD="$LIGHTNING_BUILD_PATH"
+    if [ ! -e Makefile ]; then
+        cmake $(rocm_cmake_params) \
+              $(rocm_common_cmake_params) \
+              -DROCM_DEVICE_LIBS_BITCODE_INSTALL_LOC_NEW="$bitcodeInstallLoc/amdgcn" \
+              -DROCM_DEVICE_LIBS_BITCODE_INSTALL_LOC_OLD="amdgcn" \
+              "$DEVICELIBS_ROOT"
+
+        echo "CMake complete"
+    fi
+
+    echo "Building device-libs"
+    cmake --build . -- $MAKEOPTS
+    cmake --build . -- $MAKEOPTS install
+    popd
+}
+
+package_devicelibs() {
+    mkdir -p "$DEB_PATH"
+    mkdir -p "$RPM_PATH"
+    pushd "$BUILD_PATH"
+    cpack
+    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$DEB_PATH" *.deb
+    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$RPM_PATH" *.rpm 
+    popd
+}
+
+print_output_directory() {
+    case ${PKGTYPE} in
+        ("deb")
+            echo ${DEB_PATH};;
+        ("rpm")
+            echo ${RPM_PATH};;
+        (*)
+            echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
+    esac
+    exit
+}
+case $TARGET in
+    (clean) clean_devicelibs ;;
+    (build) build_devicelibs; package_devicelibs ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
+esac
+
+echo "Operation complete"

tools/rocm-build/build_half.sh

@@ -0,0 +1,46 @@
+#!/bin/bash
+
+set -ex
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+
+set_component_src half
+
+build_half() {
+    echo "Start build"
+
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+         set_asan_env_vars
+         set_address_sanitizer_on
+         ASAN_CMAKE_PARAMS="false"
+    fi
+    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
+
+    cmake \
+        -DCMAKE_INSTALL_PREFIX="$ROCM_PATH" \
+        -DBUILD_FILE_REORG_BACKWARD_COMPATIBILITY=OFF \
+        "$COMPONENT_SRC"
+
+    cmake --build "$BUILD_DIR" -- -j${PROC}
+    cmake --build "$BUILD_DIR" -- package
+    cmake --build "$BUILD_DIR" -- install
+
+    rm -rf _CPack_Packages/ && find -name '*.o' -delete
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+
+    show_build_cache_stats
+}
+
+clean_half() {
+    echo "Cleaning half build directory: ${BUILD_DIR} ${PACKAGE_DIR}"
+    rm -rf "$BUILD_DIR" "$PACKAGE_DIR"
+    echo "Done!"
+}
+
+stage2_command_args "$@"
+
+case $TARGET in
+    build) build_half ;;
+    outdir) print_output_directory ;;
+    clean) clean_half ;;
+    *) die "Invalid target $TARGET" ;;
+esac

tools/rocm-build/build_hip_on_rocclr.sh

@@ -0,0 +1,295 @@
+#!/bin/bash
+
+printUsage() {
+    echo
+    echo "Usage: $(basename "${BASH_SOURCE}") [options ...]"
+    echo
+    echo "Options:"
+    echo "  -h,  --help                 Prints this help"
+    echo "  -c,  --clean                Clean output and delete all intermediate work"
+    echo "  -r,  --release              Make a release build instead of a debug build"
+    echo "  -a,  --address_sanitizer    Enable address sanitizer"
+    echo "  -s,  --static               Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -o,  --outdir <pkg_type>    Print path of output directory containing packages of type referred to by pkg_type"
+    echo "  -t,  --offload-arch=<arch>  Specify arch for catch tests ex: --offload-arch=gfx1030 --offload-arch=gfx1100"
+    echo "  -p,  --package <type>       Specify packaging format"
+    echo
+    echo "Possible values for <type>:"
+    echo "  deb -> Debian format (default)"
+    echo "  rpm -> RPM format"
+    echo
+
+    return 0
+}
+
+source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
+MAKEOPTS="$DASH_JAY"
+
+BUILD_PATH="$(getBuildPath hip-on-rocclr)"
+
+TARGET="build"
+PACKAGE_ROOT="$(getPackageRoot)"
+PACKAGE_SRC="$(getSrcPath)"
+PACKAGE_DEB="$PACKAGE_ROOT/deb/hip-on-rocclr"
+PACKAGE_RPM="$PACKAGE_ROOT/rpm/hip-on-rocclr"
+PREFIX_PATH="$PACKAGE_ROOT"
+CORE_BUILD_DIR="$(getBuildPath hsa-core)"
+ROCclr_BUILD_DIR="$(getBuildPath rocclr)"
+HIPCC_BUILD_DIR="$(getBuildPath hipcc)"
+CATCH_BUILD_DIR="$(getBuildPath catch)"
+CATCH_SRC="$HIP_CATCH_TESTS_ROOT/catch"
+SAMPLES_SRC="$HIP_CATCH_TESTS_ROOT/samples"
+SAMPLES_BUILD_DIR="$(getBuildPath samples)"
+if [ ! -e "$CATCH_SRC/CMakeLists.txt" ]; then
+   echo "Using catch source from hip project" >&2
+   CATCH_SRC="$HIP_ON_ROCclr_ROOT/tests/catch"
+fi
+
+BUILD_TYPE="Debug"
+SHARED_LIBS="ON"
+CLEAN_OR_OUT=0;
+MAKETARGET="deb"
+PKGTYPE="deb"
+OFFLOAD_ARCH=()
+
+DEFAULT_OFFLOAD_ARCH=(gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942 gfx1030 gfx1031 gfx1033 gfx1034 gfx1035 gfx1100 gfx1101 gfx1102 gfx1103)
+
+VALID_STR=`getopt -o hcrast:o: --long help,clean,release,address_sanitizer,static,offload-arch=:,outdir: -- "$@"`
+eval set -- "$VALID_STR"
+
+while true ;
+do
+    case "$1" in
+        (-h | --help)
+                printUsage ; exit 0;;
+        (-c | --clean)
+                TARGET="clean" ; ((CLEAN_OR_OUT|=1)) ; shift ;;
+        (-r | --release)
+                BUILD_TYPE="RelWithDebInfo" ; shift ;;
+        (-a | --address_sanitizer)
+                set_asan_env_vars
+                set_address_sanitizer_on ; shift ;;
+        (-s | --static)
+                SHARED_LIBS="OFF" ; shift ;;
+        (-t | --offload-arch=)
+                OFFLOAD_ARCH+=( "$2" ); ((CLEAN_OR_OUT|=2)); shift 2 ;;
+        (-o | --outdir)
+                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
+        --)     shift; break;;
+
+        (*)
+                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
+    esac
+
+done
+
+if [ ${#OFFLOAD_ARCH[@]} = 0 ] ; then
+      OFFLOAD_ARCH=( "${DEFAULT_OFFLOAD_ARCH[@]}" )
+else
+    echo "Using user defined offload archs ${OFFLOAD_ARCH[@]} for catch tests";
+fi
+printf -v OFFLOAD_ARCH_STR -- '--offload-arch=%q ' "${OFFLOAD_ARCH[@]}"
+
+RET_CONFLICT=1
+check_conflicting_options $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+if [ $RET_CONFLICT -ge 30 ]; then
+   print_vars $API_NAME $TARGET $BUILD_TYPE $SHARED_LIBS $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
+   exit $RET_CONFLICT
+fi
+
+clean_hip_on_rocclr() {
+    rm -rf "$BUILD_PATH"
+    rm -rf "$PACKAGE_DEB"
+    rm -rf "$PACKAGE_RPM"
+    rm -rf "$OUT_DIR/hip"
+}
+
+build_hip_on_rocclr() {
+    if [  -e "$CLR_ROOT/CMakeLists.txt" ]; then
+        _HIP_CMAKELIST_DIR="$CLR_ROOT"
+        _HIP_CMAKELIST_OPT="-DCLR_BUILD_HIP=ON -DCLR_BUILD_OCL=OFF"
+        if [ -e "$HIPOTHER_ROOT/hipnv" ]; then
+            _HIP_CMAKELIST_OPT="$_HIP_CMAKELIST_OPT -DHIPNV_DIR=$HIPOTHER_ROOT/hipnv"
+        fi
+    elif [ ! -e "$HIPAMD_ROOT/CMakeLists.txt" ]; then
+        echo "No $HIPAMD_ROOT/CMakeLists.txt file, skipping hip on rocclr" >&2
+        echo "No $HIPAMD_ROOT/CMakeLists.txt file, skipping hip on rocclr"
+        exit 0
+    else
+        _HIP_CMAKELIST_DIR="$HIPAMD_ROOT"
+        _HIP_CMAKELIST_OPT=""
+    fi
+
+    echo "$_HIP_CMAKELIST_DIR"
+    mkdir -p "$BUILD_PATH"
+    pushd "$BUILD_PATH"
+
+    if [ ! -e Makefile ]; then
+        echo "Building HIP-On-ROCclr CMake environment"
+        print_lib_type $SHARED_LIBS
+
+        cmake $(rocm_cmake_params) \
+            -DBUILD_SHARED_LIBS=$SHARED_LIBS \
+            -DHIP_COMPILER=clang \
+            -DHIP_PLATFORM=amd \
+            -DHIP_COMMON_DIR="$HIP_ON_ROCclr_ROOT" \
+            $(rocm_common_cmake_params) \
+            -DCMAKE_HIP_ARCHITECTURES=OFF \
+            -DHSA_PATH="$ROCM_INSTALL_PATH" \
+            -DCMAKE_SKIP_BUILD_RPATH=TRUE \
+            -DCPACK_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
+            -DROCM_PATH="$ROCM_INSTALL_PATH" \
+            -DHIPCC_BIN_DIR="$HIPCC_BUILD_DIR" \
+            -DHIP_CATCH_TEST=1 \
+            $_HIP_CMAKELIST_OPT \
+            "$_HIP_CMAKELIST_DIR"
+
+        echo "CMake complete"
+    fi
+
+    echo "Build and Install HIP"
+    cmake --build . -- $MAKEOPTS install "VERBOSE=1"
+
+    popd
+}
+
+build_catch_tests() {
+   WORKSPACE=`pwd`
+   echo "Build catch2 tests independently"
+   if [ ! -e "$CATCH_SRC/CMakeLists.txt" ]; then
+      echo "catch source not found: $CATCH_SRC" >&2
+      exit
+   fi
+   # build catch
+   rm -rf "$CATCH_BUILD_DIR"
+   mkdir -p "$CATCH_BUILD_DIR"
+   pushd "$CATCH_BUILD_DIR"
+   export HIP_PATH="$ROCM_INSTALL_PATH"
+   export ROCM_PATH="$ROCM_INSTALL_PATH"
+   cmake \
+         -DCMAKE_BUILD_TYPE="${BUILD_TYPE}" \
+         -DHIP_PLATFORM=amd \
+         -DROCM_PATH="$ROCM_INSTALL_PATH" \
+         -DOFFLOAD_ARCH_STR="$OFFLOAD_ARCH_STR" \
+         $(rocm_common_cmake_params) \
+         -DCPACK_RPM_DEBUGINFO_PACKAGE=FALSE \
+         -DCPACK_DEBIAN_DEBUGINFO_PACKAGE=FALSE \
+         -DCPACK_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
+         "$CATCH_SRC"
+
+
+   make $MAKEOPTS build_tests
+   echo "Packaging catch tests"
+   make $MAKEOPTS package_test
+   copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_DEB" *.deb
+   copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_RPM" *.rpm
+   popd
+}
+
+package_samples() {
+   if [ "$ASAN_CMAKE_PARAMS" == "true" ] ; then
+      echo "Disable the packaging of HIP samples" >&2
+      return
+   fi
+   WORKSPACE=`pwd`
+   if [ ! -e "$SAMPLES_SRC/CMakeLists.txt" ]; then
+      echo "HIP samples source not found at: $SAMPLES_SRC" >&2
+      echo "Using samples package from hip project: $BUILD_PATH" >&2
+      return
+   fi
+
+   rm -rf "$SAMPLES_BUILD_DIR"
+   mkdir -p "$SAMPLES_BUILD_DIR"
+   pushd "$SAMPLES_BUILD_DIR"
+   local CMAKE_PATH="$(getCmakePath)"
+   export HIP_PATH="$ROCM_INSTALL_PATH"
+   export ROCM_PATH="$ROCM_INSTALL_PATH"
+   cmake \
+         -DROCM_PATH="$ROCM_INSTALL_PATH" \
+         $(rocm_common_cmake_params) \
+         -DCMAKE_MODULE_PATH="$CMAKE_PATH/hip" \
+         -DCPACK_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
+         "$SAMPLES_SRC"
+   echo "Packaging hip samples from hip-tests project"
+   make $MAKEOPTS package_samples
+   copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_DEB" *.deb
+   copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_RPM" *.rpm
+   popd
+}
+
+clean_hip_tests(){
+    rm -rf "$CATCH_BUILD_DIR"
+    rm -rf "$PACKAGE_SRC/hip-on-rocclr"
+    rm -rf "$PACKAGE_SRC/hipamd"
+    rm -rf "$PACKAGE_SRC/rocclr"
+    rm -rf "$PACKAGE_SRC/opencl-on-rocclr"
+    rm -rf "$PACKAGE_SRC/clr"
+    rm -rf "$PACKAGE_SRC/hip-tests"
+    rm -rf "$PACKAGE_SRC/hipother"
+}
+copy_hip_tests() {
+    clean_hip_tests
+
+    echo "Copy HIP & ROCclr Source and tests"
+
+    mkdir -p "$PACKAGE_SRC/hip-on-rocclr"
+    echo "Copying hip-on-rocclr"
+    progressCopy "$HIP_ON_ROCclr_ROOT" "$PACKAGE_SRC/hip-on-rocclr"
+
+    if [ -e "$CLR_ROOT/CMakeLists.txt" ]; then
+        mkdir -p "$PACKAGE_SRC/clr"
+        echo "Copying clr"
+        progressCopy "$CLR_ROOT" "$PACKAGE_SRC/clr"
+    else
+        mkdir -p "$PACKAGE_SRC/hipamd"
+        mkdir -p "$PACKAGE_SRC/rocclr"
+        mkdir -p "$PACKAGE_SRC/opencl-on-rocclr"
+        echo "Copying hipamd"
+        progressCopy "$HIPAMD_ROOT" "$PACKAGE_SRC/hipamd"
+        echo "Copying rocclr"
+        progressCopy "$ROCclr_ROOT" "$PACKAGE_SRC/rocclr"
+        echo "Copying opencl-on-rocclr"
+        progressCopy "$OPENCL_ON_ROCclr_ROOT" "$PACKAGE_SRC/opencl-on-rocclr"
+    fi
+
+    if [ -e "$HIPOTHER_ROOT/hipnv" ]; then
+        mkdir -p "$PACKAGE_SRC/hipother"
+        echo "Copying hipother"
+        progressCopy "$HIPOTHER_ROOT" "$PACKAGE_SRC/hipother"
+    fi
+
+    mkdir -p "$PACKAGE_SRC/hip-tests"
+    echo "Copying hip-tests"
+    progressCopy "$HIP_CATCH_TESTS_ROOT" "$PACKAGE_SRC/hip-tests"
+}
+
+package_hip_on_rocclr()
+{
+    echo "Packagin HIP-on-ROCclr"
+    pushd "$BUILD_PATH"
+    cmake --build . -- $MAKEOPTS package
+    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_DEB" *.deb
+    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_RPM" *.rpm
+    popd
+}
+
+print_output_directory() {
+     case ${PKGTYPE} in
+         ("deb")
+             echo ${PACKAGE_DEB};;
+         ("rpm")
+             echo ${PACKAGE_RPM};;
+         (*)
+             echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
+     esac
+     exit
+}
+
+case $TARGET in
+    (clean) clean_hip_on_rocclr; clean_hip_tests ;;
+    (build) build_hip_on_rocclr; build_catch_tests; package_hip_on_rocclr; package_samples; copy_hip_tests;;
+   (outdir) print_output_directory ;;
+        (*) die "Invalid target $TARGET" ;;
+esac
+
+echo "Operation complete"

tools/rocm-build/build_hipblas.sh

@@ -0,0 +1,65 @@
+#!/bin/bash
+
+set -ex
+
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+
+set_component_src hipBLAS
+
+build_hipblas() {
+    echo "Start build"
+
+    CXX="g++"
+    CLIENTS_SAMPLES="ON"
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+       set_asan_env_vars
+       set_address_sanitizer_on
+       CLIENTS_SAMPLES="OFF"
+    fi
+
+    echo "C compiler: $CC"
+    echo "CXX compiler: $CXX"
+    echo "FC compiler: $FC"
+
+    cd $COMPONENT_SRC
+    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
+
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+       rebuild_lapack
+    fi
+
+    cmake \
+        ${LAUNCHER_FLAGS} \
+	$(rocm_common_cmake_params) \
+        -DUSE_CUDA=OFF \
+        -DBUILD_CLIENTS_TESTS=ON \
+        -DBUILD_CLIENTS_BENCHMARKS=ON \
+        -DBUILD_CLIENTS_SAMPLES="${CLIENTS_SAMPLES}" \
+        -DCPACK_SET_DESTDIR=OFF \
+        -DBUILD_ADDRESS_SANITIZER="${ADDRESS_SANITIZER}" \
+        "$COMPONENT_SRC"
+
+    cmake --build "$BUILD_DIR" -- -j${PROC}
+    cmake --build "$BUILD_DIR" -- install
+    cmake --build "$BUILD_DIR" -- package
+
+    rm -rf _CPack_Packages/ && find -name '*.o' -delete
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+
+    show_build_cache_stats
+}
+
+clean_hipblas() {
+    echo "Cleaning hipBLAS build directory: ${BUILD_DIR} ${PACKAGE_DIR}"
+    rm -rf "$BUILD_DIR" "$PACKAGE_DIR"
+    echo "Done!"
+}
+
+stage2_command_args "$@"
+
+case $TARGET in
+    build) build_hipblas ;;
+    outdir) print_output_directory ;;
+    clean) clean_hipblas ;;
+    *) die "Invalid target $TARGET" ;;
+esac