Fix white paper links

Merge pull request #4393 from ROCm/docs_fix_arch

.azuredevops/components/AMDMIGraphX.yml

@@ -84,8 +84,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/MIOpen.yml

@@ -67,8 +67,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/MIVisionX.yml

@@ -77,8 +77,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/ROCmValidationSuite.yml

@@ -67,8 +67,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/Tensile.yml

@@ -87,6 +87,7 @@ jobs:
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
+      gpuTarget: $(JOB_GPU_TARGET)
 
 - job: Tensile_testing
   timeoutInMinutes: 90

.azuredevops/components/TransferBench.yml

@@ -42,8 +42,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/composable_kernel.yml

@@ -48,8 +48,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/hip-tests.yml

@@ -52,8 +52,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/hipBLAS.yml

@@ -63,8 +63,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/hipBLASLt.yml

@@ -72,8 +72,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:
@@ -158,7 +156,6 @@ jobs:
         - deps
 
 - job: hipBLASLt_testing
-  timeoutInMinutes: 120
   dependsOn: hipBLASLt
   condition: and(succeeded(), eq(variables.ENABLE_GFX942_TESTS, 'true'), not(containsValue(split(variables.DISABLED_GFX942_TESTS, ','), variables['Build.DefinitionName'])))
   variables:

.azuredevops/components/hipCUB.yml

@@ -43,8 +43,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/hipFFT.yml

@@ -54,8 +54,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/hipRAND.yml

@@ -45,8 +45,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/hipSOLVER.yml

@@ -57,8 +57,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/hipSPARSE.yml

@@ -52,8 +52,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/hipSPARSELt.yml

@@ -105,7 +105,6 @@ jobs:
         -DCMAKE_BUILD_TYPE=Release
         -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
         -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
-        -DCMAKE_Fortran_COMPILER=f95
         -DAMDGPU_TARGETS=$(JOB_GPU_TARGET)
         -DTensile_LOGIC=
         -DTensile_CPU_THREADS=

.azuredevops/components/hipTensor.yml

@@ -42,8 +42,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/hipfort.yml

@@ -51,8 +51,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/llvm-project.yml

@@ -29,7 +29,7 @@ jobs:
     value: '$(Build.BinariesDirectory)/amdgcn/bitcode'
   - name: HIP_PATH
     value: '$(Agent.BuildDirectory)/rocm'
-  pool: ${{ variables.ULTRA_BUILD_POOL }}
+  pool: ${{ variables.MEDIUM_BUILD_POOL }}
   workspace:
     clean: all
   steps:
@@ -51,7 +51,7 @@ jobs:
       extraBuildFlags: >-
         -DCMAKE_PREFIX_PATH="$(Build.BinariesDirectory)/llvm;$(Build.BinariesDirectory)"
         -DCMAKE_BUILD_TYPE=Release
-        -DLLVM_ENABLE_PROJECTS=clang;lld;clang-tools-extra;mlir;flang
+        -DLLVM_ENABLE_PROJECTS=clang;lld;clang-tools-extra;mlir
         -DLLVM_ENABLE_RUNTIMES=compiler-rt;libunwind;libcxx;libcxxabi
         -DCLANG_ENABLE_AMDCLANG=ON
         -DLLVM_TARGETS_TO_BUILD=AMDGPU;X86
@@ -85,7 +85,7 @@ jobs:
       componentName: check-llvm
       testDir: 'llvm/build'
       testExecutable: './bin/llvm-lit'
-      testParameters: '-q --xunit-xml-output=llvm_test_output.xml --filter-out="live-debug-values-spill-tracking" ./test'
+      testParameters: '-q --xunit-xml-output=llvm_test_output.xml ./test'
       testOutputFile: llvm_test_output.xml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
     parameters:

.azuredevops/components/omnitrace.yml

@@ -0,0 +1,140 @@
+# largely referenced from: https://github.com/ROCm/omnitrace/blob/main/.github/workflows/ubuntu-jammy.yml
+parameters:
+- name: checkoutRepo
+  type: string
+  default: 'self'
+- name: checkoutRef
+  type: string
+  default: ''
+- name: aptPackages
+  type: object
+  default:
+    - autoconf
+    - autotools-dev
+    - bison
+    - build-essential
+    - bzip2
+    - clang
+    - cmake
+    - environment-modules
+    - g++-12
+    - libdrm-dev
+    - libfabric-dev
+    - libiberty-dev
+    - libpapi-dev
+    - libpfm4-dev
+    - libtool
+    - libopenmpi-dev
+    - m4
+    - openmpi-bin
+    - software-properties-common
+    - python3-pip
+    - texinfo
+    - zlib1g-dev
+- name: pipModules
+  type: object
+  default:
+    - numpy
+    - perfetto
+    - dataclasses
+- name: rocmDependencies
+  type: object
+  default:
+    - aomp
+    - clr
+    - llvm-project
+    - rccl
+    - rocm-core
+    - rocm_smi_lib
+    - rocminfo
+    - ROCR-Runtime
+    - rocprofiler
+    - rocprofiler-register
+    - roctracer
+
+jobs:
+- job: omnitrace
+  variables:
+  - group: common
+  - template: /.azuredevops/variables-global.yml
+  pool: ${{ variables.MEDIUM_BUILD_POOL }}
+  workspace:
+    clean: all
+  strategy:
+    matrix:
+      gfx942:
+        JOB_GPU_TARGET: gfx942
+  steps:
+  - 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 }}
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
+    parameters:
+      checkoutRef: ${{ parameters.checkoutRef }}
+      dependencyList: ${{ parameters.rocmDependencies }}
+      gpuTarget: $(JOB_GPU_TARGET)
+  - task: Bash@3
+    displayName: ROCm symbolic link
+    inputs:
+      targetType: inline
+      script: |
+        sudo rm -rf /opt/rocm
+        sudo ln -s $(Agent.BuildDirectory)/rocm /opt/rocm
+  - task: Bash@3
+    displayName: Add ROCm binaries to PATH
+    inputs:
+      targetType: inline
+      script: echo "##vso[task.prependpath]$(Agent.BuildDirectory)/rocm/bin"
+  - task: Bash@3
+    displayName: Add ROCm compilers to PATH
+    inputs:
+      targetType: inline
+      script: echo "##vso[task.prependpath]$(Agent.BuildDirectory)/rocm/llvm/bin"
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
+    parameters:
+# build flags reference: https://rocm.docs.amd.com/projects/omnitrace/en/latest/install/install.html
+      extraBuildFlags: >-
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm
+        -DOMNITRACE_BUILD_TESTING=ON
+        -DOMNITRACE_BUILD_DYNINST=ON
+        -DOMNITRACE_BUILD_LIBUNWIND=ON
+        -DDYNINST_BUILD_TBB=ON
+        -DDYNINST_BUILD_ELFUTILS=ON
+        -DDYNINST_BUILD_LIBIBERTY=ON
+        -DDYNINST_BUILD_BOOST=ON
+        -DOMNITRACE_USE_PAPI=ON
+        -DOMNITRACE_USE_MPI=ON
+        -DAMDGPU_TARGETS=$(JOB_GPU_TARGET)
+      multithreadFlag: -- -j32
+  - task: Bash@3
+    displayName: Set up omnitrace env
+    inputs:
+      targetType: inline
+      script: source share/omnitrace/setup-env.sh
+      workingDirectory: build
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
+    parameters:
+      componentName: omnitrace
+  - task: Bash@3
+    displayName: Remove ROCm binaries from PATH
+    condition: always()
+    inputs:
+      targetType: inline
+      script: echo "##vso[task.setvariable variable=PATH]$(echo $PATH | sed -e 's;:$(Agent.BuildDirectory)/rocm/bin;;' -e 's;^/;;' -e 's;/$;;')"
+  - task: Bash@3
+    displayName: Remove ROCm compilers from PATH
+    condition: always()
+    inputs:
+      targetType: inline
+      script: echo "##vso[task.setvariable variable=PATH]$(echo $PATH | sed -e 's;:$(Agent.BuildDirectory)/rocm/llvm/bin;;' -e 's;^/;;' -e 's;/$;;')"
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
+    parameters:
+      gpuTarget: $(JOB_GPU_TARGET)
+  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
+    parameters:
+      gpuTarget: $(JOB_GPU_TARGET)

.azuredevops/components/rccl.yml

@@ -64,8 +64,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rdc.yml

@@ -65,8 +65,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocAL.yml

@@ -73,8 +73,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - task: Bash@3
     displayName: 'Register libjpeg-turbo packages'

.azuredevops/components/rocALUTION.yml

@@ -60,8 +60,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocBLAS.yml

@@ -75,8 +75,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocFFT.yml

@@ -55,8 +55,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocJPEG.yml

@@ -47,8 +47,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocPRIM.yml

@@ -42,8 +42,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocPyDecode.yml

@@ -48,8 +48,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocRAND.yml

@@ -45,8 +45,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocSOLVER.yml

@@ -58,8 +58,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocSPARSE.yml

@@ -56,8 +56,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocThrust.yml

@@ -47,8 +47,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocWMMA.yml

@@ -57,8 +57,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocm-cmake.yml

@@ -8,6 +8,7 @@ parameters:
 - name: aptPackages
   type: object
   default:
+    - cmake
     - doxygen
     - doxygen-doc
     - ninja-build
@@ -17,9 +18,7 @@ parameters:
   type: object
   default:
     - cget
-    - cmake==3.20.5
     - ninja
-    - rocm-docs-core
 
 jobs:
 - job: rocm_cmake
@@ -34,29 +33,21 @@ jobs:
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
-  - task: Bash@3
-    displayName: Add CMake to PATH
-    inputs:
-      targetType: inline
-      script: echo "##vso[task.prependpath]$(python3 -m site --user-base)/bin"
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
       checkoutRepo: ${{ parameters.checkoutRepo }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
-  - task: Bash@3
-    displayName: CTest setup
-    inputs:
-      targetType: inline
-      script: |
-        python -m pip install -r $(Build.SourcesDirectory)/docs/requirements.txt
-        python -m pip install -r $(Build.SourcesDirectory)/test/docsphinx/docs/.sphinx/requirements.txt
-        git config --global user.email "you@example.com"
-        git config --global user.name "Your Name"
+# extra steps for ctest suite
+  - script: |
+      python -m pip install -r $(Build.SourcesDirectory)/docs/requirements.txt
+      python -m pip install -r $(Build.SourcesDirectory)/test/docsphinx/docs/.sphinx/requirements.txt
+      git config --global user.email "you@example.com"
+      git config --global user.name "Your Name"
+    displayName: "ctest setup"
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
     parameters:
       componentName: rocm-cmake
-      testParameters: '-E "pass-version-parent" -VV --output-on-failure --force-new-ctest-process --output-junit test_output.xml'
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-links.yml
@@ -65,3 +56,4 @@ jobs:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
       environment: combined
+      gpuTarget: $(JOB_GPU_TARGET)

.azuredevops/components/rocm-examples.yml

@@ -75,8 +75,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocm_smi_lib.yml

@@ -10,7 +10,6 @@ parameters:
   default:
     - cmake
     - libdrm-dev
-    - pkg-config
     - python3-pip
 
 jobs:
@@ -40,6 +39,7 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
+      gpuTarget: $(JOB_GPU_TARGET)
 
 - job: rocm_smi_lib_testing
   dependsOn: rocm_smi_lib

.azuredevops/components/rocprofiler-compute.yml

@@ -59,8 +59,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocprofiler-sdk.yml

@@ -57,8 +57,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocprofiler-systems.yml

@@ -72,8 +72,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocprofiler.yml

@@ -57,8 +57,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/components/rocr_debug_agent.yml

@@ -69,6 +69,7 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
+      gpuTarget: $(JOB_GPU_TARGET)
 
 - job: rocr_debug_agent_testing
   dependsOn: rocr_debug_agent

.azuredevops/components/roctracer.yml

@@ -11,7 +11,6 @@ parameters:
     - cmake
     - doxygen
     - graphviz
-    - libdrm-amdgpu-dev
     - ninja-build
     - python3-pip
 - name: pipModules
@@ -50,14 +49,11 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
-      registerROCmPackages: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
@@ -89,7 +85,6 @@ jobs:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
       gpuTarget: $(JOB_GPU_TARGET)
-      registerROCmPackages: true
 
 - job: roctracer_testing
   dependsOn: roctracer
@@ -109,8 +104,6 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
-      pipModules: ${{ parameters.pipModules }}
-      registerROCmPackages: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
     parameters:
@@ -135,4 +128,3 @@ jobs:
       pipModules: ${{ parameters.pipModules }}
       environment: test
       gpuTarget: $(JOB_GPU_TARGET)
-      registerROCmPackages: true

.azuredevops/components/rpp.yml

@@ -57,8 +57,6 @@ jobs:
     matrix:
       gfx942:
         JOB_GPU_TARGET: gfx942
-      gfx90a:
-        JOB_GPU_TARGET: gfx90a
   steps:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:

.azuredevops/tag-builds/TransferBench.yml

@@ -1,29 +0,0 @@
-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/TransferBench
-    ref: ${{ parameters.checkoutRef }}
-
-trigger: none
-pr: none
-
-jobs:
-  - template: ${{ variables.CI_COMPONENT_PATH }}/TransferBench.yml
-    parameters:
-      checkoutRepo: release_repo
-      checkoutRef: ${{ parameters.checkoutRef }}

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

@@ -222,13 +222,13 @@ parameters:
       hasGpuTarget: false
     rocm-examples:
       pipelineId: $(ROCM_EXAMPLES_PIPELINE_ID)
-      stagingBranch: amd-staging
-      mainlineBranch: amd-mainline
+      stagingBranch: develop
+      mainlineBranch: develop
       hasGpuTarget: true
     rocminfo:
       pipelineId: $(ROCMINFO_PIPELINE_ID)
       stagingBranch: amd-staging
-      mainlineBranch: amd-mainline
+      mainlineBranch: amd-master
       hasGpuTarget: false
     rocMLIR:
       pipelineId: $(ROCMLIR_PIPELINE_ID)
@@ -262,7 +262,7 @@ parameters:
       hasGpuTarget: true
     rocprofiler-compute:
       pipelineId: $(ROCPROFILER_COMPUTE_PIPELINE_ID)
-      stagingBranch: develop
+      stagingBranch: amd-staging
       mainlineBranch: amd-mainline
       hasGpuTarget: true
     rocprofiler-register:

.azuredevops/templates/steps/test.yml

@@ -33,6 +33,7 @@ parameters:
     - aomp
     - HIPIFY
     - MIVisionX
+    - rocm-cmake
     - rocm_smi_lib
     - rocprofiler-sdk
     - roctracer

.azuredevops/variables-global.yml

@@ -28,13 +28,13 @@ variables:
 - name: GFX942_TEST_POOL
   value: gfx942_test_pool
 - name: LATEST_RELEASE_VERSION
-  value: 6.3.3
+  value: 6.3.2
 - name: REPO_RADEON_VERSION
-  value: 6.3.3
+  value: 6.3.2
 - name: NEXT_RELEASE_VERSION
   value: 6.4.0
 - name: LATEST_RELEASE_TAG
-  value: rocm-6.3.3
+  value: rocm-6.3.2
 - name: AMDMIGRAPHX_GFX942_TEST_PIPELINE_ID
   value: 197
 - name: AMDMIGRAPHX_PIPELINE_ID

.gitignore

@@ -11,7 +11,6 @@ _toc.yml
 docBin/
 _doxygen/
 _readthedocs/
-__pycache__/
 
 # avoid duplicating contributing.md due to conf.py
 docs/CHANGELOG.md

.wordlist.txt

@@ -156,6 +156,7 @@ HCA
 HGX
 HIPCC
 HIPExtension
+HIPification
 HIPIFY
 HIPification
 HIPify
@@ -481,7 +482,6 @@ ZenDNN
 accuracies
 activations
 addr
-ai
 alloc
 allocatable
 allocator
@@ -547,7 +547,6 @@ cTDP
 dataset
 datasets
 dataspace
-datatemplate
 datatype
 datatypes
 dbgapi
@@ -576,7 +575,6 @@ el
 embeddings
 enablement
 encodings
-endfor
 endpgm
 enqueue
 env
@@ -697,7 +695,6 @@ pageable
 pallas
 parallelization
 parallelizing
-param
 parameterization
 passthrough
 perfcounter
@@ -815,7 +812,6 @@ supercomputing
 symlink
 symlinks
 sys
-tabindex
 td
 tensorfloat
 th
@@ -861,7 +857,6 @@ vectorizes
 virtualize
 virtualized
 vjxb
-vllm
 voxel
 walkthrough
 walkthroughs

CONTRIBUTING.md

@@ -66,10 +66,11 @@ project-specific steps. Refer to each repository's PR process for any additional
   during our release cycle, as coordinated by the maintainer
 * We'll inform you once your change is committed
 
-> [!IMPORTANT]
-> By creating a PR, you agree to allow your contribution to be licensed under the
-> terms of the LICENSE.txt file in the corresponding repository. Different repositories may use different
-> licenses.
+:::{important}
+By creating a PR, you agree to allow your contribution to be licensed under the
+terms of the LICENSE.txt file in the corresponding repository. Different repositories may use different
+licenses.
+:::
 
 You can look up each license on the [ROCm licensing](https://rocm.docs.amd.com/en/latest/about/license.html) page.
 

README.md

@@ -116,7 +116,7 @@ bash install-prerequisites.sh
 # For ubuntu22.04 system
 cd ROCm/tools/rocm-build/docker/ubuntu22
 cp * /tmp && cd /tmp
-bash install-prerequisites.sh
+bash install-prerequisities.sh
 # For ubuntu24.04 system
 cd ROCm/tools/rocm-build/docker/ubuntu24
 cp * /tmp && cd /tmp

docs/compatibility/ml-compatibility/jax-compatibility.rst

@@ -4,8 +4,6 @@
    :description: JAX compatibility
    :keywords: GPU, JAX compatibility
 
-.. version-set:: rocm_version latest
-
 *******************************************************************************
 JAX compatibility
 *******************************************************************************
@@ -121,8 +119,7 @@ Critical ROCm libraries for JAX
 
 The functionality of JAX with ROCm is determined by its underlying library
 dependencies. These critical ROCm components affect the capabilities,
-performance, and feature set available to developers. The versions described
-are available in ROCm :version:`rocm_version`.
+performance, and feature set available to developers.
 
 .. list-table::
     :header-rows: 1
@@ -132,7 +129,7 @@ are available in ROCm :version:`rocm_version`.
       - Purpose
       - Used in
     * - `hipBLAS <https://github.com/ROCm/hipBLAS>`_
-      - :version-ref:`hipBLAS rocm_version`
+      - 2.3.0
       - Provides GPU-accelerated Basic Linear Algebra Subprograms (BLAS) for
         matrix and vector operations.
       - Matrix multiplication in ``jax.numpy.matmul``, ``jax.lax.dot`` and
@@ -141,7 +138,7 @@ are available in ROCm :version:`rocm_version`.
         ``jax.numpy.einsum`` with matrix-multiplication patterns algebra
         operations.
     * - `hipBLASLt <https://github.com/ROCm/hipBLASLt>`_
-      - :version-ref:`hipBLASLt rocm_version`
+      - 0.10.0
       - hipBLASLt is an extension of hipBLAS, providing additional
         features like epilogues fused into the matrix multiplication kernel or
         use of integer tensor cores.
@@ -150,7 +147,7 @@ are available in ROCm :version:`rocm_version`.
         operations, mixed-precision support, and hardware-specific
         optimizations.
     * - `hipCUB <https://github.com/ROCm/hipCUB>`_
-      - :version-ref:`hipCUB rocm_version`
+      - 3.3.0
       - Provides a C++ template library for parallel algorithms for reduction,
         scan, sort and select.
       - Reduction functions (``jax.numpy.sum``, ``jax.numpy.mean``,
@@ -158,23 +155,23 @@ are available in ROCm :version:`rocm_version`.
         (``jax.numpy.cumsum``, ``jax.numpy.cumprod``) and sorting
         (``jax.numpy.sort``, ``jax.numpy.argsort``).
     * - `hipFFT <https://github.com/ROCm/hipFFT>`_
-      - :version-ref:`hipFFT rocm_version`
+      - 1.0.17
       - Provides GPU-accelerated Fast Fourier Transform (FFT) operations.
       - Used in functions like ``jax.numpy.fft``.
     * - `hipRAND <https://github.com/ROCm/hipRAND>`_
-      - :version-ref:`hipRAND rocm_version`
+      - 2.11.0
       - Provides fast random number generation for GPUs.
       - The ``jax.random.uniform``, ``jax.random.normal``,
         ``jax.random.randint`` and ``jax.random.split``.
     * - `hipSOLVER <https://github.com/ROCm/hipSOLVER>`_
-      - :version-ref:`hipSOLVER rocm_version`
+      - 2.3.0
       - Provides GPU-accelerated solvers for linear systems, eigenvalues, and
         singular value decompositions (SVD).
       - Solving linear systems (``jax.numpy.linalg.solve``), matrix
         factorizations, SVD (``jax.numpy.linalg.svd``) and eigenvalue problems
         (``jax.numpy.linalg.eig``).
     * - `hipSPARSE <https://github.com/ROCm/hipSPARSE>`_
-      - :version-ref:`hipSPARSE rocm_version`
+      - 3.1.2
       - Accelerates operations on sparse matrices, such as sparse matrix-vector
         or matrix-matrix products.
       - Sparse matrix multiplication (``jax.numpy.matmul``), sparse
@@ -182,28 +179,28 @@ are available in ROCm :version:`rocm_version`.
         (``jax.experimental.sparse.dot``), sparse linear system solvers and
         sparse data handling.
     * - `hipSPARSELt <https://github.com/ROCm/hipSPARSELt>`_
-      - :version-ref:`hipSPARSELt rocm_version`
+      - 0.2.2
       - Accelerates operations on sparse matrices, such as sparse matrix-vector
         or matrix-matrix products.
       - Sparse matrix multiplication (``jax.numpy.matmul``), sparse
         matrix-vector and matrix-matrix products
         (``jax.experimental.sparse.dot``) and sparse linear system solvers.
     * - `MIOpen <https://github.com/ROCm/MIOpen>`_
-      - :version-ref:`MIOpen rocm_version`
+      - 3.3.0
       - Optimized for deep learning primitives such as convolutions, pooling,
         normalization, and activation functions.
       - Speeds up convolutional neural networks (CNNs), recurrent neural
         networks (RNNs), and other layers. Used in operations like
         ``jax.nn.conv``, ``jax.nn.relu``, and ``jax.nn.batch_norm``.
     * - `RCCL <https://github.com/ROCm/rccl>`_
-      - :version-ref:`RCCL rocm_version`
+      - 2.21.5
       - Optimized for multi-GPU communication for operations like  all-reduce,
         broadcast, and scatter.
       - Distribute computations across multiple GPU with ``pmap`` and
         ``jax.distributed``. XLA automatically uses rccl when executing
         operations across multiple GPUs on AMD hardware.
     * - `rocThrust <https://github.com/ROCm/rocThrust>`_
-      - :version-ref:`rocThrust rocm_version`
+      - 3.3.0
       - Provides a C++ template library for parallel algorithms like sorting,
         reduction, and scanning.
       - Reduction operations like ``jax.numpy.sum``, ``jax.pmap`` for

docs/compatibility/ml-compatibility/pytorch-compatibility.rst

@@ -4,8 +4,6 @@
     :description: PyTorch compatibility
     :keywords: GPU, PyTorch compatibility
 
-.. version-set:: rocm_version latest
-
 ********************************************************************************
 PyTorch compatibility
 ********************************************************************************
@@ -202,8 +200,7 @@ Critical ROCm libraries for PyTorch
 
 The functionality of PyTorch with ROCm is determined by its underlying library
 dependencies. These critical ROCm components affect the capabilities,
-performance, and feature set available to developers. The versions described
-are available in ROCm :version:`rocm_version`.
+performance, and feature set available to developers.
 
 .. list-table::
     :header-rows: 1
@@ -213,28 +210,28 @@ are available in ROCm :version:`rocm_version`.
       - Purpose
       - Used in
     * - `Composable Kernel <https://github.com/ROCm/composable_kernel>`_
-      - :version-ref:`"Composable Kernel" rocm_version`
+      - 1.1.0
       - Enables faster execution of core operations like matrix multiplication
         (GEMM), convolutions and transformations.
       - Speeds up ``torch.permute``, ``torch.view``, ``torch.matmul``,
         ``torch.mm``, ``torch.bmm``, ``torch.nn.Conv2d``, ``torch.nn.Conv3d``
         and ``torch.nn.MultiheadAttention``.
     * - `hipBLAS <https://github.com/ROCm/hipBLAS>`_
-      - :version-ref:`hipBLAS rocm_version`
+      - 2.3.0
       - Provides GPU-accelerated Basic Linear Algebra Subprograms (BLAS) for
         matrix and vector operations.
       - Supports operations like matrix multiplication, matrix-vector products,
         and tensor contractions. Utilized in both dense and batched linear
         algebra operations.
     * - `hipBLASLt <https://github.com/ROCm/hipBLASLt>`_
-      - :version-ref:`hipBLASLt rocm_version`
+      - 0.10.0
       - hipBLASLt is an extension of the hipBLAS library, providing additional
         features like epilogues fused into the matrix multiplication kernel or
         use of integer tensor cores.
       - It accelerates operations like ``torch.matmul``, ``torch.mm``, and the
         matrix multiplications used in convolutional and linear layers.
     * - `hipCUB <https://github.com/ROCm/hipCUB>`_
-      - :version-ref:`hipCUB rocm_version`
+      - 3.3.0
       - Provides a C++ template library for parallel algorithms for reduction,
         scan, sort and select.
       - Supports operations like ``torch.sum``, ``torch.cumsum``, ``torch.sort``
@@ -242,93 +239,93 @@ are available in ROCm :version:`rocm_version`.
         irregular shapes often involve scanning, sorting, and filtering, which
         hipCUB handles efficiently.
     * - `hipFFT <https://github.com/ROCm/hipFFT>`_
-      - :version-ref:`hipFFT rocm_version`
+      - 1.0.17
       - Provides GPU-accelerated Fast Fourier Transform (FFT) operations.
       - Used in functions like the ``torch.fft`` module.
     * - `hipRAND <https://github.com/ROCm/hipRAND>`_
-      - :version-ref:`hipRAND rocm_version`
+      - 2.11.0
       - Provides fast random number generation for GPUs.
       - The ``torch.rand``, ``torch.randn`` and stochastic layers like
         ``torch.nn.Dropout``.
     * - `hipSOLVER <https://github.com/ROCm/hipSOLVER>`_
-      - :version-ref:`hipSOLVER rocm_version`
+      - 2.3.0
       - Provides GPU-accelerated solvers for linear systems, eigenvalues, and
         singular value decompositions (SVD).
       - Supports functions like ``torch.linalg.solve``,
         ``torch.linalg.eig``, and ``torch.linalg.svd``.
     * - `hipSPARSE <https://github.com/ROCm/hipSPARSE>`_
-      - :version-ref:`hipSPARSE rocm_version`
+      - 3.1.2
       - Accelerates operations on sparse matrices, such as sparse matrix-vector
         or matrix-matrix products.
       - Sparse tensor operations ``torch.sparse``.
     * - `hipSPARSELt <https://github.com/ROCm/hipSPARSELt>`_
-      - :version-ref:`hipSPARSELt rocm_version`
+      - 0.2.2
       - Accelerates operations on sparse matrices, such as sparse matrix-vector
         or matrix-matrix products.
       - Sparse tensor operations ``torch.sparse``.
     * - `hipTensor <https://github.com/ROCm/hipTensor>`_
-      - :version-ref:`hipTensor rocm_version`
+      - 1.4.0
       - Optimizes for high-performance tensor operations, such as contractions.
       - Accelerates tensor algebra, especially in deep learning and scientific
         computing.
     * - `MIOpen <https://github.com/ROCm/MIOpen>`_
-      - :version-ref:`MIOpen rocm_version`
+      - 3.3.0
       - Optimizes deep learning primitives such as convolutions, pooling,
         normalization, and activation functions.
       - Speeds up convolutional neural networks (CNNs), recurrent neural
         networks (RNNs), and other layers. Used in operations like
         ``torch.nn.Conv2d``, ``torch.nn.ReLU``, and ``torch.nn.LSTM``.
     * - `MIGraphX <https://github.com/ROCm/AMDMIGraphX>`_
-      - :version-ref:`MIGraphX rocm_version`
+      - 2.11.0
       - Adds graph-level optimizations, ONNX models and mixed precision support
         and enable Ahead-of-Time (AOT) Compilation.
       - Speeds up inference models and executes ONNX models for
         compatibility with other frameworks.
         ``torch.nn.Conv2d``, ``torch.nn.ReLU``, and ``torch.nn.LSTM``.
     * - `MIVisionX <https://github.com/ROCm/MIVisionX>`_
-      - :version-ref:`MIVisionX rocm_version`
+      - 3.1.0
       - Optimizes acceleration for computer vision and AI workloads like
         preprocessing, augmentation, and inferencing.
       - Faster data preprocessing and augmentation pipelines for datasets like
         ImageNet or COCO and easy to integrate into PyTorch's ``torch.utils.data``
         and ``torchvision`` workflows.
     * - `rocAL <https://github.com/ROCm/rocAL>`_
-      - :version-ref:`rocAL rocm_version`
+      - 2.1.0
       - Accelerates the data pipeline by offloading intensive preprocessing and
         augmentation tasks. rocAL is part of MIVisionX.
       - Easy to integrate into PyTorch's ``torch.utils.data`` and
         ``torchvision`` data load workloads.
     * - `RCCL <https://github.com/ROCm/rccl>`_
-      - :version-ref:`RCCL rocm_version`
+      - 2.21.5
       - Optimizes for multi-GPU communication for operations like AllReduce and
         Broadcast.
       - Distributed data parallel training (``torch.nn.parallel.DistributedDataParallel``).
         Handles communication in multi-GPU setups.
     * - `rocDecode <https://github.com/ROCm/rocDecode>`_
-      - :version-ref:`rocDecode rocm_version`
+      - 0.8.0
       - Provides hardware-accelerated data decoding capabilities, particularly
         for image, video, and other dataset formats.
       - Can be integrated in ``torch.utils.data``, ``torchvision.transforms``
         and ``torch.distributed``.
     * - `rocJPEG <https://github.com/ROCm/rocJPEG>`_
-      - :version-ref:`rocJPEG rocm_version`
+      - 0.6.0
       - Provides hardware-accelerated JPEG image decoding and encoding.
       - GPU accelerated ``torchvision.io.decode_jpeg`` and
         ``torchvision.io.encode_jpeg`` and can be integrated in
         ``torch.utils.data`` and ``torchvision``.
     * - `RPP <https://github.com/ROCm/RPP>`_
-      - :version-ref:`RPP rocm_version`
+      - 1.9.1
       - Speeds up data augmentation, transformation, and other preprocessing steps.
       - Easy to integrate into PyTorch's ``torch.utils.data`` and
         ``torchvision`` data load workloads.
     * - `rocThrust <https://github.com/ROCm/rocThrust>`_
-      - :version-ref:`rocThrust rocm_version`
+      - 3.3.0
       - Provides a C++ template library for parallel algorithms like sorting,
         reduction, and scanning.
       - Utilized in backend operations for tensor computations requiring
         parallel processing.
     * - `rocWMMA <https://github.com/ROCm/rocWMMA>`_
-      - :version-ref:`rocWMMA rocm_version`
+      - 1.6.0
       - Accelerates warp-level matrix-multiply and matrix-accumulate to speed up matrix
         multiplication (GEMM) and accumulation operations with mixed precision
         support.

docs/compatibility/ml-compatibility/tensorflow-compatibility.rst

@@ -4,8 +4,6 @@
     :description: TensorFlow compatibility
     :keywords: GPU, TensorFlow compatibility
 
-.. version-set:: rocm_version latest
-
 *******************************************************************************
 TensorFlow compatibility
 *******************************************************************************
@@ -72,7 +70,7 @@ the |docker-icon| icon to view the image on Docker Hub.
 
            <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.3-py3.12-tf2.17-dev/images/sha256-fd2653f436880366cc874aa24264ca9dabd892d76ccb63fb807debba459bcaaf"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>
 
-      - `tensorflow-rocm 2.17.0 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3/tensorflow_rocm-2.17.0-cp312-cp312-manylinux_2_28_x86_64.whl>`__
+      - `tensorflow-rocm 2.17.0 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/tensorflow_rocm-2.17.0-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - dev
       - `Python 3.12.4 <https://www.python.org/downloads/release/python-3124/>`_
       - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`_
@@ -81,7 +79,7 @@ the |docker-icon| icon to view the image on Docker Hub.
 
            <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.3-py3.10-tf2.17-dev/images/sha256-8a5eb7443798935dd269575e2abae847b702e1dfb06766ab84f081a6314d8b95"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>
 
-      - `tensorflow-rocm 2.17.0 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3/tensorflow_rocm-2.17.0-cp310-cp310-manylinux_2_28_x86_64.whl>`__
+      - `tensorflow-rocm 2.17.0 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/tensorflow_rocm-2.17.0-cp310-cp310-manylinux_2_28_x86_64.whl>`__
       - dev
       - `Python 3.10.16 <https://www.python.org/downloads/release/python-31016/>`_
       - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`_
@@ -90,7 +88,7 @@ the |docker-icon| icon to view the image on Docker Hub.
 
            <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.3-py3.12-tf2.16-dev/images/sha256-8fc939b10cdd6d2b11407474880d4c8ab2b52ab6e2d1743c921fc2adbfd0422f"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>
 
-      - `tensorflow-rocm 2.16.2 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3/tensorflow_rocm-2.16.2-cp312-cp312-manylinux_2_28_x86_64.whl>`__
+      - `tensorflow-rocm 2.16.2 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/tensorflow_rocm-2.16.2-cp312-cp312-manylinux_2_28_x86_64.whl>`__
       - dev
       - `Python 3.12.4 <https://www.python.org/downloads/release/python-3124/>`_
       - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`_
@@ -99,7 +97,7 @@ the |docker-icon| icon to view the image on Docker Hub.
 
            <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.3-py3.10-tf2.16-dev/images/sha256-a4cc6ab23d59fdf5459ceac1f0a603e6c16ae7f885d30e42c0c2b3ac60c2ad10"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>
 
-      - `tensorflow-rocm 2.16.2 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3/tensorflow_rocm-2.16.2-cp310-cp310-manylinux_2_28_x86_64.whl>`__
+      - `tensorflow-rocm 2.16.2 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/tensorflow_rocm-2.16.2-cp310-cp310-manylinux_2_28_x86_64.whl>`__
       - dev
       - `Python 3.10.16 <https://www.python.org/downloads/release/python-31016/>`_
       - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`_
@@ -108,7 +106,7 @@ the |docker-icon| icon to view the image on Docker Hub.
 
            <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.3-py3.10-tf2.15-dev/images/sha256-60887c488421184adcb60b9ed4f72a8bd7bdb64d238e50943ca7cbde38e4aa48"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>
 
-      - `tensorflow-rocm 2.15.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3/tensorflow_rocm-2.15.1-cp310-cp310-manylinux_2_28_x86_64.whl>`_
+      - `tensorflow-rocm 2.15.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/tensorflow_rocm-2.15.1-cp310-cp310-manylinux_2_28_x86_64.whl>`_
       - dev
       - `Python 3.10.16 <https://www.python.org/downloads/release/python-31016/>`_
       - `TensorBoard 2.15.2 <https://github.com/tensorflow/tensorboard/tree/2.15.2>`_
@@ -119,8 +117,7 @@ Critical ROCm libraries for TensorFlow
 TensorFlow depends on multiple components and the supported features of those
 components can affect the TensorFlow ROCm supported feature set. The versions
 in the following table refer to the first TensorFlow version where the ROCm
-library was introduced as a dependency. The versions described
-are available in ROCm :version:`rocm_version`.
+library was introduced as a dependency.
 
 .. list-table::
     :widths: 25, 10, 35, 30
@@ -131,43 +128,43 @@ are available in ROCm :version:`rocm_version`.
       - Purpose
       - Used in
     * - `hipBLAS <https://github.com/ROCm/hipBLAS>`_
-      - :version-ref:`hipBLAS rocm_version`
+      - 2.3.0
       - Provides GPU-accelerated Basic Linear Algebra Subprograms (BLAS) for
         matrix and vector operations.
       - Accelerates operations like ``tf.matmul``, ``tf.linalg.matmul``, and
         other matrix multiplications commonly used in neural network layers.
     * - `hipBLASLt <https://github.com/ROCm/hipBLASLt>`_
-      - :version-ref:`hipBLASLt rocm_version`
+      - 0.10.0
       - Extends hipBLAS with additional optimizations like fused kernels and
         integer tensor cores.
       - Optimizes matrix multiplications and linear algebra operations used in
         layers like dense, convolutional, and RNNs in TensorFlow.
     * - `hipCUB <https://github.com/ROCm/hipCUB>`_
-      - :version-ref:`hipCUB rocm_version`
+      - 3.3.0
       - Provides a C++ template library for parallel algorithms for reduction,
         scan, sort and select.
       - Supports operations like ``tf.reduce_sum``, ``tf.cumsum``, ``tf.sort``
         and other tensor operations in TensorFlow, especially those involving
         scanning, sorting, and filtering.
     * - `hipFFT <https://github.com/ROCm/hipFFT>`_
-      - :version-ref:`hipFFT rocm_version`
+      - 1.0.17
       - Accelerates Fast Fourier Transforms (FFT) for signal processing tasks.
       - Used for operations like signal processing, image filtering, and
         certain types of neural networks requiring FFT-based transformations.
     * - `hipSOLVER <https://github.com/ROCm/hipSOLVER>`_
-      - :version-ref:`hipSOLVER rocm_version`
+      - 2.3.0
       - Provides GPU-accelerated direct linear solvers for dense and sparse
         systems.
       - Optimizes linear algebra functions such as solving systems of linear
         equations, often used in optimization and training tasks.
     * - `hipSPARSE <https://github.com/ROCm/hipSPARSE>`_
-      - :version-ref:`hipSPARSE rocm_version`
+      - 3.1.2
       - Optimizes sparse matrix operations for efficient computations on sparse
         data.
       - Accelerates sparse matrix operations in models with sparse weight
         matrices or activations, commonly used in neural networks.
     * - `MIOpen <https://github.com/ROCm/MIOpen>`_
-      - :version-ref:`MIOpen rocm_version`
+      - 3.3.0
       - Provides optimized deep learning primitives such as convolutions,
         pooling,
         normalization, and activation functions.
@@ -175,13 +172,13 @@ are available in ROCm :version:`rocm_version`.
         in TensorFlow for layers like ``tf.nn.conv2d``, ``tf.nn.relu``, and
         ``tf.nn.lstm_cell``.
     * - `RCCL <https://github.com/ROCm/rccl>`_
-      - :version-ref:`RCCL rocm_version`
+      - 2.21.5
       - Optimizes for multi-GPU communication for operations like AllReduce and
         Broadcast.
       - Distributed data parallel training (``tf.distribute.MirroredStrategy``).
         Handles communication in multi-GPU setups.
     * - `rocThrust <https://github.com/ROCm/rocThrust>`_
-      - :version-ref:`rocThrust rocm_version`
+      - 3.3.0
       - Provides a C++ template library for parallel algorithms like sorting,
         reduction, and scanning.
       - Reduction operations like ``tf.reduce_sum``, ``tf.cumsum`` for computing

docs/compatibility/pytorch-compatibility.rst

@@ -0,0 +1,916 @@
+.. meta::
+    :description: PyTorch compatibility
+    :keywords: GPU, PyTorch compatibility
+
+********************************************************************************
+PyTorch compatibility
+********************************************************************************
+
+`PyTorch <https://pytorch.org/>`_ is an open-source tensor library designed for
+deep learning. PyTorch on ROCm provides mixed-precision and large-scale training
+using `MIOpen <https://github.com/ROCm/MIOpen>`_ and
+`RCCL <https://github.com/ROCm/rccl>`_ libraries.
+
+ROCm support for PyTorch is upstreamed into the official PyTorch repository. Due to independent
+compatibility considerations, this results in two distinct release cycles for PyTorch on ROCm:
+
+- ROCm PyTorch release:
+
+  - Provides the latest version of ROCm but doesn't immediately support the latest stable PyTorch
+    version.
+
+  - Offers :ref:`Docker images <pytorch-docker-compat>` with ROCm and PyTorch
+    pre-installed.
+
+  - ROCm PyTorch repository: `<https://github.com/rocm/pytorch>`__
+
+  - See the :doc:`ROCm PyTorch installation guide <rocm-install-on-linux:install/3rd-party/pytorch-install>` to get started.
+
+- Official PyTorch release:
+
+  - Provides the latest stable version of PyTorch but doesn't immediately support the latest ROCm version.
+
+  - Official PyTorch repository: `<https://github.com/pytorch/pytorch>`__
+
+  - See the `Nightly and latest stable version installation guide <https://pytorch.org/get-started/locally/>`_
+    or `Previous versions <https://pytorch.org/get-started/previous-versions/>`_ to get started.
+
+The upstream PyTorch includes an automatic HIPification solution that automatically generates HIP
+source code from the CUDA backend. This approach allows PyTorch to support ROCm without requiring
+manual code modifications.
+
+ROCm's development is aligned with the stable release of PyTorch while upstream PyTorch testing uses
+the stable release of ROCm to maintain consistency.
+
+.. _pytorch-docker-compat:
+
+Docker image compatibility
+================================================================================
+
+AMD validates and publishes ready-made `PyTorch <https://hub.docker.com/r/rocm/pytorch>`_
+images with ROCm backends on Docker Hub. The following Docker image tags and
+associated inventories are validated for `ROCm 6.3.0 <https://repo.radeon.com/rocm/apt/6.3/>`_.
+
+.. list-table:: PyTorch Docker image components
+    :header-rows: 1
+    :class: docker-image-compatibility
+
+    * - Docker
+      - PyTorch
+      - Ubuntu
+      - Python
+      - Apex
+      - torchvision
+      - TensorBoard
+      - MAGMA
+      - UCX
+      - OMPI
+      - OFED
+
+    * - .. raw:: html
+
+           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3_ubuntu24.04_py3.12_pytorch_release_2.4.0/images/sha256-98ddf20333bd01ff749b8092b1190ee369a75d3b8c71c2fac80ffdcb1a98d529?context=explore"><i class="fab fa-docker fa-lg"></i></a>
+
+      - `2.4.0 <https://github.com/ROCm/pytorch/tree/release/2.4>`_
+      - 24.04
+      - `3.12 <https://www.python.org/downloads/release/python-3128/>`_
+      - `1.4.0 <https://github.com/ROCm/apex/tree/release/1.4.0>`_
+      - `0.19.0 <https://github.com/pytorch/vision/tree/v0.19.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `1.10.0 <https://github.com/openucx/ucx/tree/v1.10.0>`_
+      - `4.0.7 <https://github.com/open-mpi/ompi/tree/v4.0.7>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+
+    * - .. raw:: html
+
+           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3_ubuntu22.04_py3.10_pytorch_release_2.4.0/images/sha256-402c9b4f1a6b5a81c634a1932b56cbe01abb699cfcc7463d226276997c6cf8ea?context=explore"><i class="fab fa-docker fa-lg"></i></a>
+
+      - `2.4.0 <https://github.com/ROCm/pytorch/tree/release/2.4>`_
+      - 22.04
+      - `3.10 <https://www.python.org/downloads/release/python-31016/>`_
+      - `1.4.0 <https://github.com/ROCm/apex/tree/release/1.4.0>`_
+      - `0.19.0 <https://github.com/pytorch/vision/tree/v0.19.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `1.10.0 <https://github.com/openucx/ucx/tree/v1.10.0>`_
+      - `4.0.7 <https://github.com/open-mpi/ompi/tree/v4.0.7>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+
+    * - .. raw:: html
+
+           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3_ubuntu22.04_py3.9_pytorch_release_2.4.0/images/sha256-e0608b55d408c3bfe5c19fdd57a4ced3e0eb3a495b74c309980b60b156c526dd?context=explore"><i class="fab fa-docker fa-lg"></i></a>
+
+      - `2.4.0 <https://github.com/ROCm/pytorch/tree/release/2.4>`_
+      - 22.04
+      - `3.9 <https://www.python.org/downloads/release/python-3918/>`_
+      - `1.4.0 <https://github.com/ROCm/apex/tree/release/1.4.0>`_
+      - `0.19.0 <https://github.com/pytorch/vision/tree/v0.19.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `1.10.0 <https://github.com/openucx/ucx/tree/v1.10.0>`_
+      - `4.0.7 <https://github.com/open-mpi/ompi/tree/v4.0.7>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+
+    * - .. raw:: html
+
+           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3_ubuntu22.04_py3.10_pytorch_release_2.3.0/images/sha256-652cf25263d05b1de548222970aeb76e60b12de101de66751264709c0d0ff9d8?context=explore"><i class="fab fa-docker fa-lg"></i></a>
+
+      - `2.3.0 <https://github.com/ROCm/pytorch/tree/release/2.3>`_
+      - 22.04
+      - `3.10 <https://www.python.org/downloads/release/python-31016/>`_
+      - `1.3.0 <https://github.com/ROCm/apex/tree/release/1.3.0>`_
+      - `0.18.0 <https://github.com/pytorch/vision/tree/v0.18.0>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `1.14.1 <https://github.com/openucx/ucx/tree/v1.14.1>`_
+      - `4.1.5 <https://github.com/open-mpi/ompi/tree/v4.1.5>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+
+    * - .. raw:: html
+
+           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3_ubuntu22.04_py3.10_pytorch_release_2.2.1/images/sha256-051976f26beab8f9aa65d999e3ad546c027b39240a0cc3ee81b114a9024f2912?context=explore"><i class="fab fa-docker fa-lg"></i></a>
+
+      - `2.2.1 <https://github.com/ROCm/pytorch/tree/release/2.2>`_
+      - 22.04
+      - `3.10 <https://www.python.org/downloads/release/python-31016/>`_
+      - `1.2.0 <https://github.com/ROCm/apex/tree/release/1.2.0>`_
+      - `0.17.1 <https://github.com/pytorch/vision/tree/v0.17.1>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `1.14.1 <https://github.com/openucx/ucx/tree/v1.14.1>`_
+      - `4.1.5 <https://github.com/open-mpi/ompi/tree/v4.1.5>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+
+    * - .. raw:: html
+
+           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3_ubuntu20.04_py3.9_pytorch_release_2.2.1/images/sha256-88c839a364d109d3748c100385bfa100d28090d25118cc723fd0406390ab2f7e?context=explore"><i class="fab fa-docker fa-lg"></i></a>
+
+      - `2.2.1 <https://github.com/ROCm/pytorch/tree/release/2.2>`_
+      - 20.04
+      - `3.9 <https://www.python.org/downloads/release/python-3921/>`_
+      - `1.2.0 <https://github.com/ROCm/apex/tree/release/1.2.0>`_
+      - `0.17.1 <https://github.com/pytorch/vision/tree/v0.17.1>`_
+      - `2.13.0 <https://github.com/tensorflow/tensorboard/tree/2.13.0>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `1.10.0 <https://github.com/openucx/ucx/tree/v1.10.0>`_
+      - `4.0.3 <https://github.com/open-mpi/ompi/tree/v4.0.3>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+
+    * - .. raw:: html
+
+           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3_ubuntu22.04_py3.9_pytorch_release_1.13.1/images/sha256-994424ed07a63113f79dd9aa72159124c00f5fbfe18127151e6658f7d0b6f821?context=explore"><i class="fab fa-docker fa-lg"></i></a>
+
+      - `1.13.1 <https://github.com/ROCm/pytorch/tree/release/1.13>`_
+      - 22.04
+      - `3.9 <https://www.python.org/downloads/release/python-3921/>`_
+      - `1.0.0 <https://github.com/ROCm/apex/tree/release/1.0.0>`_
+      - `0.14.0 <https://github.com/pytorch/vision/tree/v0.14.0>`_
+      - `2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `1.14.1 <https://github.com/openucx/ucx/tree/v1.14.1>`_
+      - `4.1.5 <https://github.com/open-mpi/ompi/tree/v4.1.5>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+
+    * - .. raw:: html
+
+           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3_ubuntu20.04_py3.9_pytorch_release_1.13.1/images/sha256-7b8139fe40a9aeb4bca3aecd15c22c1fa96e867d93479fa3a24fdeeeeafa1219?context=explore"><i class="fab fa-docker fa-lg"></i></a>
+
+      - `1.13.1 <https://github.com/ROCm/pytorch/tree/release/1.13>`_
+      - 20.04
+      - `3.9 <https://www.python.org/downloads/release/python-3921/>`_
+      - `1.0.0 <https://github.com/ROCm/apex/tree/release/1.0.0>`_
+      - `0.14.0 <https://github.com/pytorch/vision/tree/v0.14.0>`_
+      - `2.18.0 <https://github.com/tensorflow/tensorboard/tree/2.18>`_
+      - `master <https://bitbucket.org/icl/magma/src/master/>`_
+      - `1.10.0 <https://github.com/openucx/ucx/tree/v1.10.0>`_
+      - `4.0.3 <https://github.com/open-mpi/ompi/tree/v4.0.3>`_
+      - `5.3-1.0.5.0 <https://content.mellanox.com/ofed/MLNX_OFED-5.3-1.0.5.0/MLNX_OFED_LINUX-5.3-1.0.5.0-ubuntu20.04-x86_64.tgz>`_
+
+Critical ROCm libraries for PyTorch
+================================================================================
+
+The functionality of PyTorch with ROCm is shaped by its underlying library
+dependencies. These critical ROCm components affect the capabilities,
+performance, and feature set available to developers.
+
+.. list-table::
+    :header-rows: 1
+
+    * - ROCm library
+      - Version
+      - Purpose
+      - Used in
+    * - `Composable Kernel <https://github.com/ROCm/composable_kernel>`_
+      - 1.1.0
+      - Enables faster execution of core operations like matrix multiplication
+        (GEMM), convolutions and transformations.
+      - Speeds up ``torch.permute``, ``torch.view``, ``torch.matmul``,
+        ``torch.mm``, ``torch.bmm``, ``torch.nn.Conv2d``, ``torch.nn.Conv3d``
+        and ``torch.nn.MultiheadAttention``. 
+    * - `hipBLAS <https://github.com/ROCm/hipBLAS>`_
+      - 2.3.0
+      - Provides GPU-accelerated Basic Linear Algebra Subprograms (BLAS) for
+        matrix and vector operations.
+      - Supports operations like matrix multiplication, matrix-vector products,
+        and tensor contractions. Utilized in both dense and batched linear
+        algebra operations.
+    * - `hipBLASLt <https://github.com/ROCm/hipBLASLt>`_
+      - 0.10.0
+      - hipBLASLt is an extension of the hipBLAS library, providing additional
+        features like epilogues fused into the matrix multiplication kernel or
+        use of integer tensor cores.
+      - It accelerates operations like ``torch.matmul``, ``torch.mm``, and the
+        matrix multiplications used in convolutional and linear layers.
+    * - `hipCUB <https://github.com/ROCm/hipCUB>`_
+      - 3.3.0
+      - Provides a C++ template library for parallel algorithms for reduction,
+        scan, sort and select.
+      - Supports operations like ``torch.sum``, ``torch.cumsum``, ``torch.sort``
+        and ``torch.topk``. Operations on sparse tensors or tensors with
+        irregular shapes often involve scanning, sorting, and filtering, which
+        hipCUB handles efficiently.
+    * - `hipFFT <https://github.com/ROCm/hipFFT>`_
+      - 1.0.17
+      - Provides GPU-accelerated Fast Fourier Transform (FFT) operations.
+      - Used in functions like the ``torch.fft`` module.
+    * - `hipRAND <https://github.com/ROCm/hipRAND>`_
+      - 2.11.0
+      - Provides fast random number generation for GPUs.
+      - The ``torch.rand``, ``torch.randn`` and stochastic layers like 
+        ``torch.nn.Dropout``.
+    * - `hipSOLVER <https://github.com/ROCm/hipSOLVER>`_
+      - 2.3.0
+      - Provides GPU-accelerated solvers for linear systems, eigenvalues, and
+        singular value decompositions (SVD).
+      - Supports functions like ``torch.linalg.solve``,
+        ``torch.linalg.eig``, and ``torch.linalg.svd``.
+    * - `hipSPARSE <https://github.com/ROCm/hipSPARSE>`_
+      - 3.1.2
+      - Accelerates operations on sparse matrices, such as sparse matrix-vector
+        or matrix-matrix products.
+      - Sparse tensor operations ``torch.sparse``.
+    * - `hipSPARSELt <https://github.com/ROCm/hipSPARSELt>`_
+      - 0.2.2
+      - Accelerates operations on sparse matrices, such as sparse matrix-vector
+        or matrix-matrix products.
+      - Sparse tensor operations ``torch.sparse``.
+    * - `hipTensor <https://github.com/ROCm/hipTensor>`_
+      - 1.4.0
+      - Optimizes for high-performance tensor operations, such as contractions.
+      - Accelerates tensor algebra, especially in deep learning and scientific
+        computing.
+    * - `MIOpen <https://github.com/ROCm/MIOpen>`_
+      - 3.3.0
+      - Optimizes deep learning primitives such as convolutions, pooling,
+        normalization, and activation functions.
+      - Speeds up convolutional neural networks (CNNs), recurrent neural
+        networks (RNNs), and other layers. Used in operations like
+        ``torch.nn.Conv2d``, ``torch.nn.ReLU``, and ``torch.nn.LSTM``.
+    * - `MIGraphX <https://github.com/ROCm/AMDMIGraphX>`_
+      - 2.11.0
+      - Add graph-level optimizations, ONNX models and mixed precision support
+        and enable Ahead-of-Time (AOT) Compilation.
+      - Speeds up inference models and executes ONNX models for
+        compatibility with other frameworks.
+        ``torch.nn.Conv2d``, ``torch.nn.ReLU``, and ``torch.nn.LSTM``.
+    * - `MIVisionX <https://github.com/ROCm/MIVisionX>`_
+      - 3.1.0
+      - Optimizes acceleration for computer vision and AI workloads like
+        preprocessing, augmentation, and inferencing.
+      - Faster data preprocessing and augmentation pipelines for datasets like
+        ImageNet or COCO and easy to integrate into PyTorch's ``torch.utils.data``
+        and ``torchvision`` workflows.
+    * - `rocAL <https://github.com/ROCm/rocAL>`_
+      - 2.1.0
+      - Accelerates the data pipeline by offloading intensive preprocessing and
+        augmentation tasks. rocAL is part of MIVisionX.
+      - Easy to integrate into PyTorch's ``torch.utils.data`` and
+        ``torchvision`` data load workloads.
+    * - `RCCL <https://github.com/ROCm/rccl>`_
+      - 2.21.5
+      - Optimizes for multi-GPU communication for operations like AllReduce and
+        Broadcast.
+      - Distributed data parallel training (``torch.nn.parallel.DistributedDataParallel``).
+        Handles communication in multi-GPU setups.
+    * - `rocDecode <https://github.com/ROCm/rocDecode>`_
+      - 0.8.0
+      - Provide hardware-accelerated data decoding capabilities, particularly
+        for image, video, and other dataset formats.
+      - Can be integrated in ``torch.utils.data``, ``torchvision.transforms``
+        and ``torch.distributed``.
+    * - `rocJPEG <https://github.com/ROCm/rocJPEG>`_
+      - 0.6.0
+      - Provide hardware-accelerated JPEG image decoding and encoding.
+      - GPU accelerated ``torchvision.io.decode_jpeg`` and
+        ``torchvision.io.encode_jpeg`` and can be integrated in
+        ``torch.utils.data`` and ``torchvision``.
+    * - `RPP <https://github.com/ROCm/RPP>`_
+      - 1.9.1
+      - Speed up data augmentation, transformation, and other preprocessing step.
+      - Easy to integrate into PyTorch's ``torch.utils.data`` and
+        ``torchvision`` data load workloads.
+    * - `rocThrust <https://github.com/ROCm/rocThrust>`_
+      - 3.3.0
+      - Provides a C++ template library for parallel algorithms like sorting,
+        reduction, and scanning.
+      - Utilized in backend operations for tensor computations requiring
+        parallel processing.
+    * - `rocWMMA <https://github.com/ROCm/rocWMMA>`_
+      - 1.6.0
+      - Accelerates warp-level matrix-multiply and matrix-accumulate to speed up matrix
+        multiplication (GEMM) and accumulation operations with mixed precision
+        support.
+      - Linear layers (``torch.nn.Linear``), convolutional layers
+        (``torch.nn.Conv2d``), attention layers, general tensor operations that
+        involve matrix products, such as ``torch.matmul``, ``torch.bmm``, and
+        more.
+
+Supported and unsupported features
+================================================================================
+
+The following section maps GPU-accelerated PyTorch features to their supported
+ROCm and PyTorch versions.
+
+torch
+--------------------------------------------------------------------------------
+
+`torch <https://pytorch.org/docs/stable/index.html>`_ is the central module of
+PyTorch, providing data structures for multi-dimensional tensors and
+implementing mathematical operations on them. It also includes utilities for
+efficient serialization of tensors and arbitrary data types, along with various
+other tools.
+
+Tensor data types
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The data type of a tensor is specified using the ``dtype`` attribute or argument, and PyTorch supports a wide range of data types for different use cases.
+
+The following table lists `torch.Tensor <https://pytorch.org/docs/stable/tensors.html>`_'s single data types:
+
+.. list-table::
+    :header-rows: 1
+
+    * - Data type
+      - Description
+      - Since PyTorch
+      - Since ROCm
+    * - ``torch.float8_e4m3fn``
+      - 8-bit floating point, e4m3
+      - 2.3
+      - 5.5
+    * - ``torch.float8_e5m2``
+      - 8-bit floating point, e5m2
+      - 2.3
+      - 5.5
+    * - ``torch.float16`` or ``torch.half``
+      - 16-bit floating point
+      - 0.1.6
+      - 2.0
+    * - ``torch.bfloat16``
+      - 16-bit floating point
+      - 1.6
+      - 2.6
+    * - ``torch.float32`` or ``torch.float``
+      - 32-bit floating point
+      - 0.1.12_2
+      - 2.0
+    * - ``torch.float64`` or ``torch.double``
+      - 64-bit floating point
+      - 0.1.12_2
+      - 2.0
+    * - ``torch.complex32`` or ``torch.chalf``
+      - PyTorch provides native support for 32-bit complex numbers
+      - 1.6
+      - 2.0
+    * - ``torch.complex64`` or ``torch.cfloat``
+      - PyTorch provides native support for 64-bit complex numbers
+      - 1.6
+      - 2.0
+    * - ``torch.complex128`` or ``torch.cdouble``
+      - PyTorch provides native support for 128-bit complex numbers
+      - 1.6
+      - 2.0
+    * - ``torch.uint8``
+      - 8-bit integer (unsigned)
+      - 0.1.12_2
+      - 2.0
+    * - ``torch.uint16``
+      - 16-bit integer (unsigned)
+      - 2.3
+      - Not natively supported
+    * - ``torch.uint32``
+      - 32-bit integer (unsigned)
+      - 2.3
+      - Not natively supported
+    * - ``torch.uint64``
+      - 32-bit integer (unsigned)
+      - 2.3
+      - Not natively supported
+    * - ``torch.int8``
+      - 8-bit integer (signed)
+      - 1.12
+      - 5.0
+    * - ``torch.int16`` or ``torch.short``
+      - 16-bit integer (signed)
+      - 0.1.12_2
+      - 2.0
+    * - ``torch.int32`` or ``torch.int``
+      - 32-bit integer (signed)
+      - 0.1.12_2
+      - 2.0
+    * - ``torch.int64`` or ``torch.long``
+      - 64-bit integer (signed)
+      - 0.1.12_2
+      - 2.0
+    * - ``torch.bool``
+      - Boolean
+      - 1.2
+      - 2.0
+    * - ``torch.quint8``
+      - Quantized 8-bit integer (unsigned)
+      - 1.8
+      - 5.0
+    * - ``torch.qint8``
+      - Quantized 8-bit integer (signed)
+      - 1.8
+      - 5.0
+    * - ``torch.qint32``
+      - Quantized 32-bit integer (signed)
+      - 1.8
+      - 5.0
+    * - ``torch.quint4x2``
+      - Quantized 4-bit integer (unsigned)
+      - 1.8
+      - 5.0
+
+.. note::
+
+  Unsigned types aside from ``uint8`` are currently only have limited support in
+  eager mode (they primarily exist to assist usage with ``torch.compile``).
+
+  The :doc:`ROCm precision support page <rocm:reference/precision-support>`
+  collected the native HW support of different data types.
+
+torch.cuda
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``torch.cuda`` in PyTorch is a module that provides utilities and functions for
+managing and utilizing AMD and NVIDIA GPUs. It enables GPU-accelerated
+computations, memory management, and efficient execution of tensor operations,
+leveraging ROCm and CUDA as the underlying frameworks.
+
+.. list-table::
+    :header-rows: 1
+
+    * - Data type
+      - Description
+      - Since PyTorch
+      - Since ROCm
+    * - Device management
+      - Utilities for managing and interacting with GPUs.
+      - 0.4.0
+      - 3.8
+    * - Tensor operations on GPU
+      - Perform tensor operations such as addition and matrix multiplications on
+        the GPU.
+      - 0.4.0
+      - 3.8
+    * - Streams and events
+      - Streams allow overlapping computation and communication for optimized
+        performance, events enable synchronization.
+      - 1.6.0
+      - 3.8
+    * - Memory management
+      - Functions to manage and inspect memory usage like
+        ``torch.cuda.memory_allocated()``, ``torch.cuda.max_memory_allocated()``,
+        ``torch.cuda.memory_reserved()`` and ``torch.cuda.empty_cache()``.
+      - 0.3.0
+      - 1.9.2
+    * - Running process lists of memory management
+      - Return a human-readable printout of the running processes and their GPU
+        memory use for a given device with functions like 
+        ``torch.cuda.memory_stats()`` and ``torch.cuda.memory_summary()``.
+      - 1.8.0
+      - 4.0
+    * - Communication collectives
+      - A set of APIs that enable efficient communication between multiple GPUs,
+        allowing for distributed computing and data parallelism.
+      - 1.9.0
+      - 5.0
+    * - ``torch.cuda.CUDAGraph``
+      - Graphs capture sequences of GPU operations to minimize kernel launch
+        overhead and improve performance.
+      - 1.10.0
+      - 5.3
+    * - TunableOp
+      - A mechanism that allows certain operations to be more flexible and
+        optimized for performance. It enables automatic tuning of kernel
+        configurations and other settings to achieve the best possible
+        performance based on the specific hardware (GPU) and workload.
+      - 2.0
+      - 5.4
+    * - NVIDIA Tools Extension (NVTX)
+      - Integration with NVTX for profiling and debugging GPU performance using
+        NVIDIA's Nsight tools.
+      - 1.8.0
+      - ❌
+    * - Lazy loading NVRTC
+      - Delays JIT compilation with NVRTC until the code is explicitly needed.
+      - 1.13.0
+      - ❌
+    * - Jiterator (beta)
+      - Jiterator allows asynchronous data streaming into computation streams
+        during training loops.
+      - 1.13.0
+      - 5.2
+
+.. Need to validate and extend.
+
+torch.backends.cuda
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``torch.backends.cuda`` is a PyTorch module that provides configuration options
+and flags to control the behavior of CUDA or ROCm operations. It is part of the
+PyTorch backend configuration system, which allows users to fine-tune how
+PyTorch interacts with the CUDA or ROCm environment.
+
+.. list-table::
+    :header-rows: 1
+
+    * - Data type
+      - Description
+      - Since PyTorch
+      - Since ROCm
+    * - ``cufft_plan_cache``
+      - Manages caching of GPU FFT plans to optimize repeated FFT computations.
+      - 1.7.0
+      - 5.0
+    * - ``matmul.allow_tf32``
+      - Enables or disables the use of TensorFloat-32 (TF32) precision for
+        faster matrix multiplications on GPUs with Tensor Cores.
+      - 1.10.0
+      - ❌
+    * - ``matmul.allow_fp16_reduced_precision_reduction``
+      - Reduced precision reductions (e.g., with fp16 accumulation type) are
+        allowed with fp16 GEMMs.
+      - 2.0
+      - ❌
+    * - ``matmul.allow_bf16_reduced_precision_reduction``
+      - Reduced precision reductions are allowed with bf16 GEMMs.
+      - 2.0
+      - ❌
+    * - ``enable_cudnn_sdp``
+      - Globally enables cuDNN SDPA's kernels within SDPA.
+      - 2.0
+      - ❌
+    * - ``enable_flash_sdp``
+      - Globally enables or disables FlashAttention for SDPA.
+      - 2.1
+      - ❌
+    * - ``enable_mem_efficient_sdp``
+      - Globally enables or disables Memory-Efficient Attention for SDPA.
+      - 2.1
+      - ❌
+    * - ``enable_math_sdp``
+      - Globally enables or disables the PyTorch C++ implementation within SDPA.
+      - 2.1
+      - ❌
+
+.. Need to validate and extend.
+
+torch.backends.cudnn
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Supported ``torch`` options:
+
+.. list-table::
+    :header-rows: 1
+
+    * - Data type
+      - Description
+      - Since PyTorch
+      - Since ROCm
+    * - ``allow_tf32``
+      - TensorFloat-32 tensor cores may be used in cuDNN convolutions on NVIDIA
+        Ampere or newer GPUs.
+      - 1.12.0
+      - ❌
+    * - ``deterministic``
+      - A bool that, if True, causes cuDNN to only use deterministic
+        convolution algorithms.
+      - 1.12.0
+      - 6.0
+
+Automatic mixed precision: torch.amp
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+PyTorch that automates the process of using both 16-bit (half-precision,
+float16) and 32-bit (single-precision, float32) floating-point types in model
+training and inference.
+
+.. list-table::
+    :header-rows: 1
+
+    * - Data type
+      - Description
+      - Since PyTorch
+      - Since ROCm
+    * - Autocasting
+      - Instances of autocast serve as context managers or decorators that allow
+        regions of your script to run in mixed precision.
+      - 1.9
+      - 2.5
+    * - Gradient scaling
+      - To prevent underflow, “gradient scaling” multiplies the network’s
+        loss(es) by a scale factor and invokes a backward pass on the scaled
+        loss(es). Gradients flowing backward through the network are then
+        scaled by the same factor. In other words, gradient values have a
+        larger magnitude, so they don’t flush to zero.
+      - 1.9
+      - 2.5
+    * - CUDA op-specific behavior
+      - These ops always go through autocasting whether they are invoked as part
+        of a ``torch.nn.Module``, as a function, or as a ``torch.Tensor`` method. If
+        functions are exposed in multiple namespaces, they go through
+        autocasting regardless of the namespace.
+      - 1.9
+      - 2.5
+
+Distributed library features
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The PyTorch distributed library includes a collective of parallelism modules, a
+communications layer, and infrastructure for launching and debugging large
+training jobs. See :ref:`rocm-for-ai-pytorch-distributed` for more information.
+
+The Distributed Library feature in PyTorch provides tools and APIs for building
+and running distributed machine learning workflows. It allows training models
+across multiple processes, GPUs, or nodes in a cluster, enabling efficient use
+of computational resources and scalability for large-scale tasks.
+
+.. list-table::
+    :header-rows: 1
+
+    * - Features
+      - Description
+      - Since PyTorch
+      - Since ROCm
+    * - TensorPipe
+      - TensorPipe is a point-to-point communication library integrated into
+        PyTorch for distributed training. It is designed to handle tensor data
+        transfers efficiently between different processes or devices, including
+        those on separate machines.
+      - 1.8
+      - 5.4
+    * - Gloo
+      - Gloo is designed for multi-machine and multi-GPU setups, enabling
+        efficient communication and synchronization between processes. Gloo is
+        one of the default backends for PyTorch's Distributed Data Parallel
+        (DDP) and RPC frameworks, alongside other backends like NCCL and MPI.
+      - 1.0
+      - 2.0
+
+torch.compiler
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. list-table::
+    :header-rows: 1
+
+    * - Features
+      - Description
+      - Since PyTorch
+      - Since ROCm
+    * - ``torch.compiler`` (AOT Autograd)
+      - Autograd captures not only the user-level code, but also backpropagation,
+        which results in capturing the backwards pass “ahead-of-time”. This
+        enables acceleration of both forwards and backwards pass using
+        ``TorchInductor``.
+      - 2.0
+      - 5.3
+    * - ``torch.compiler`` (TorchInductor)
+      - The default ``torch.compile`` deep learning compiler that generates fast
+        code for multiple accelerators and backends. You need to use a backend
+        compiler to make speedups through ``torch.compile`` possible. For AMD,
+        NVIDIA, and Intel GPUs, it leverages OpenAI Triton as the key building block.
+      - 2.0
+      - 5.3
+
+torchaudio
+--------------------------------------------------------------------------------
+
+The `torchaudio <https://pytorch.org/audio/stable/index.html>`_ library provides
+utilities for processing audio data in PyTorch, such as audio loading,
+transformations, and feature extraction.
+
+To ensure GPU-acceleration with ``torchaudio.transforms``, you need to move audio
+data (waveform tensor) explicitly to GPU using ``.to('cuda')``.
+
+The following ``torchaudio`` features are GPU-accelerated.
+
+.. list-table::
+    :header-rows: 1
+
+    * - Features
+      - Description
+      - Since torchaudio version
+      - Since ROCm
+    * - ``torchaudio.transforms.Spectrogram``
+      - Generate spectrogram of an input waveform using STFT.
+      - 0.6.0
+      - 4.5
+    * - ``torchaudio.transforms.MelSpectrogram``
+      - Generate the mel-scale spectrogram of raw audio signals.
+      - 0.9.0
+      - 4.5
+    * - ``torchaudio.transforms.MFCC``
+      - Extract of MFCC features.
+      - 0.9.0
+      - 4.5
+    * - ``torchaudio.transforms.Resample``
+      - Resample a signal from one frequency to another
+      - 0.9.0
+      - 4.5
+
+torchvision
+--------------------------------------------------------------------------------
+
+The `torchvision <https://pytorch.org/vision/stable/index.html>`_ library
+provide datasets, model architectures, and common image transformations for
+computer vision.
+
+The following ``torchvision`` features are GPU-accelerated.
+
+.. list-table::
+    :header-rows: 1
+
+    * - Features
+      - Description
+      - Since torchvision version
+      - Since ROCm
+    * - ``torchvision.transforms.functional``
+      - Provides GPU-compatible transformations for image preprocessing like
+        resize, normalize, rotate and crop.
+      - 0.2.0
+      - 4.0
+    * - ``torchvision.ops``
+      - GPU-accelerated operations for object detection and segmentation tasks.
+        ``torchvision.ops.roi_align``, ``torchvision.ops.nms`` and
+        ``box_convert``.
+      - 0.6.0
+      - 3.3
+    * - ``torchvision.models`` with ``.to('cuda')``
+      - ``torchvision`` provides several pre-trained models (ResNet, Faster
+        R-CNN, Mask R-CNN, ...) that can run on CUDA for faster inference and
+        training.
+      - 0.1.6
+      - 2.x
+    * - ``torchvision.io``
+      - Video decoding and frame extraction using GPU acceleration with NVIDIA’s
+        NVDEC and nvJPEG (rocJPEG) on CUDA-enabled GPUs.
+      - 0.4.0
+      - 6.3
+
+torchtext
+--------------------------------------------------------------------------------
+
+The `torchtext <https://pytorch.org/text/stable/index.html>`_ library provides
+utilities for processing and working with text data in PyTorch, including
+tokenization, vocabulary management, and text embeddings. torchtext supports
+preprocessing pipelines and integration with PyTorch models, simplifying the
+implementation of natural language processing (NLP) tasks.
+
+To leverage GPU acceleration in torchtext, you need to move tensors
+explicitly to the GPU using ``.to('cuda')``.
+
+* torchtext does not implement its own kernels. ROCm support is enabled by linking against ROCm libraries.
+
+* Only official release exists.
+
+torchtune
+--------------------------------------------------------------------------------
+
+The `torchtune <https://pytorch.org/torchtune/stable/index.html>`_ library for
+authoring, fine-tuning and experimenting with LLMs.
+
+* Usage: It works out-of-the-box, enabling developers to fine-tune ROCm PyTorch solutions.
+
+* Only official release exists.
+
+torchserve
+--------------------------------------------------------------------------------
+
+The `torchserve <https://pytorch.org/torchserve/>`_ is a PyTorch domain library
+for common sparsity and parallelism primitives needed for large-scale recommender
+systems.
+
+* torchtext does not implement its own kernels. ROCm support is enabled by linking against ROCm libraries.
+
+* Only official release exists.
+
+torchrec
+--------------------------------------------------------------------------------
+
+The `torchrec <https://pytorch.org/torchrec/>`_ is a PyTorch domain library for
+common sparsity and parallelism primitives needed for large-scale recommender
+systems.
+
+* torchrec does not implement its own kernels. ROCm support is enabled by linking against ROCm libraries.
+
+* Only official release exists.
+
+Unsupported PyTorch features
+----------------------------
+
+The following are GPU-accelerated PyTorch features not currently supported by ROCm.
+
+.. list-table::
+    :widths: 30, 60, 10
+    :header-rows: 1
+
+    * - Data type
+      - Description
+      - Since PyTorch
+    * - APEX batch norm
+      - Use APEX batch norm instead of PyTorch batch norm.
+      - 1.6.0
+    * - ``torch.backends.cuda`` / ``matmul.allow_tf32``
+      - A bool that controls whether TensorFloat-32 tensor cores may be used in
+        matrix multiplications.
+      - 1.7
+    * - ``torch.cuda`` / NVIDIA Tools Extension (NVTX)
+      - Integration with NVTX for profiling and debugging GPU performance using
+        NVIDIA's Nsight tools.
+      - 1.7.0
+    * - ``torch.cuda`` / Lazy loading NVRTC
+      - Delays JIT compilation with NVRTC until the code is explicitly needed.
+      - 1.8.0
+    * - ``torch-tensorrt``
+      - Integrate TensorRT library for optimizing and deploying PyTorch models.
+        ROCm does not have equialent library for TensorRT.
+      - 1.9.0
+    * - ``torch.backends`` / ``cudnn.allow_tf32``
+      - TensorFloat-32 tensor cores may be used in cuDNN convolutions.
+      - 1.10.0
+    * - ``torch.backends.cuda`` / ``matmul.allow_fp16_reduced_precision_reduction``
+      - Reduced precision reductions with fp16 accumulation type are
+        allowed with fp16 GEMMs.
+      - 2.0
+    * - ``torch.backends.cuda`` / ``matmul.allow_bf16_reduced_precision_reduction``
+      - Reduced precision reductions are allowed with bf16 GEMMs.
+      - 2.0
+    * - ``torch.nn.functional`` / ``scaled_dot_product_attention`` 
+      - Flash attention backend for SDPA to accelerate attention computation in
+        transformer-based models.
+      - 2.0
+    * - ``torch.backends.cuda`` / ``enable_cudnn_sdp``
+      - Globally enables cuDNN SDPA's kernels within SDPA.
+      - 2.0
+    * - ``torch.backends.cuda`` / ``enable_flash_sdp``
+      - Globally enables or disables FlashAttention for SDPA.
+      - 2.1
+    * - ``torch.backends.cuda`` / ``enable_mem_efficient_sdp``
+      - Globally enables or disables Memory-Efficient Attention for SDPA.
+      - 2.1
+    * - ``torch.backends.cuda`` / ``enable_math_sdp``
+      - Globally enables or disables the PyTorch C++ implementation within SDPA.
+      - 2.1
+    * - Dynamic parallelism
+      - PyTorch itself does not directly expose dynamic parallelism as a core
+        feature. Dynamic parallelism allow GPU threads to launch additional
+        threads which can be reached using custom operations via the
+        ``torch.utils.cpp_extension`` module.
+      - Not a core feature
+    * - Unified memory support in PyTorch
+      - Unified Memory is not directly exposed in PyTorch's core API, it can be
+        utilized effectively through custom CUDA extensions or advanced
+        workflows.
+      - Not a core feature
+
+Use cases and recommendations
+================================================================================
+
+* :doc:`Using ROCm for AI: training a model </how-to/rocm-for-ai/train-a-model>` provides
+  guidance on how to leverage the ROCm platform for training AI models. It covers the steps, tools, and best practices
+  for optimizing training workflows on AMD GPUs using PyTorch features.
+
+* :doc:`Single-GPU fine-tuning and inference </how-to/llm-fine-tuning-optimization/single-gpu-fine-tuning-and-inference>`
+  describes and demonstrates how to use the ROCm platform for the fine-tuning and inference of
+  machine learning models, particularly large language models (LLMs), on systems with a single AMD
+  Instinct MI300X accelerator. This page provides a detailed guide for setting up, optimizing, and
+  executing fine-tuning and inference workflows in such environments.
+
+* :doc:`Multi-GPU fine-tuning and inference optimization </how-to/llm-fine-tuning-optimization/multi-gpu-fine-tuning-and-inference>`
+  describes and demonstrates the fine-tuning and inference of machine learning models on systems
+  with multi MI300X accelerators.
+
+* The :doc:`Instinct MI300X workload optimization guide </how-to/tuning-guides/mi300x/workload>` provides detailed
+  guidance on optimizing workloads for the AMD Instinct MI300X accelerator using ROCm. This guide is aimed at helping
+  users achieve optimal performance for deep learning and other high-performance computing tasks on the MI300X
+  accelerator.
+
+* The :doc:`Inception with PyTorch documentation </conceptual/ai-pytorch-inception>`
+  describes how PyTorch integrates with ROCm for AI workloads It outlines the use of PyTorch on the ROCm platform and
+  focuses on how to efficiently leverage AMD GPU hardware for training and inference tasks in AI applications.
+
+For more use cases and recommendations, see `ROCm PyTorch blog posts <https://rocm.blogs.amd.com/blog/tag/pytorch.html>`_

docs/conf.py

@@ -6,8 +6,6 @@
 
 import os
 import shutil
-import sys
-from pathlib import Path
 
 shutil.copy2("../RELEASE.md", "./about/release-notes.md")
 
@@ -52,8 +50,8 @@ article_pages = [
     {"file": "how-to/rocm-for-ai/training/index", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/train-a-model", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/prerequisite-system-validation", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/benchmark-docker/megatron-lm", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/benchmark-docker/pytorch-training", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/train-a-model/benchmark-docker/megatron-lm", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/training/train-a-model/benchmark-docker/pytorch-training", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/scale-model-training", "os": ["linux"]},
 
     {"file": "how-to/rocm-for-ai/fine-tuning/index", "os": ["linux"]},
@@ -68,7 +66,7 @@ article_pages = [
     {"file": "how-to/rocm-for-ai/inference/llm-inference-frameworks", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/vllm-benchmark", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/deploy-your-model", "os": ["linux"]},
-
+    
     {"file": "how-to/rocm-for-ai/inference-optimization/index", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference-optimization/model-quantization", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference-optimization/model-acceleration-libraries", "os": ["linux"]},
@@ -91,16 +89,11 @@ article_pages = [
 
 external_toc_path = "./sphinx/_toc.yml"
 
-# Add the _extensions directory to Python's search path
-sys.path.append(str(Path(__file__).parent / 'extension'))
-
-extensions = ["rocm_docs", "sphinx_reredirects", "sphinx_sitemap", "sphinxcontrib.datatemplates", "version-ref"]
-
-compatibility_matrix_file = str(Path(__file__).parent / 'compatibility/compatibility-matrix-historical-6.0.csv')
+extensions = ["rocm_docs", "sphinx_reredirects", "sphinx_sitemap"]
 
 external_projects_current_project = "rocm"
 
-# Uncomment if facing rate limit exceed issue with local build
+# Uncomment if facing rate limit exceed issue with local build 
 # external_projects_remote_repository = ""
 
 html_baseurl = os.environ.get("READTHEDOCS_CANONICAL_URL", "https://rocm-stg.amd.com/")
@@ -111,9 +104,8 @@ if os.environ.get("READTHEDOCS", "") == "True":
 html_theme = "rocm_docs_theme"
 html_theme_options = {"flavor": "rocm-docs-home"}
 
-html_static_path = ["sphinx/static/css", "extension/how-to/rocm-for-ai/inference"]
-html_css_files = ["rocm_custom.css", "rocm_rn.css", "vllm-benchmark.css"]
-html_js_files = ["vllm-benchmark.js"]
+html_static_path = ["sphinx/static/css"]
+html_css_files = ["rocm_custom.css", "rocm_rn.css"]
 
 html_title = "ROCm Documentation"
 

docs/data/how-to/rocm-for-ai/inference/vllm-benchmark-models.yaml

@@ -1,153 +0,0 @@
-vllm_benchmark:
-  unified_docker:
-    latest:
-      pull_tag: rocm/vllm:rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6
-      docker_hub_url: https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6/images/sha256-9a12ef62bbbeb5a4c30a01f702c8e025061f575aa129f291a49fbd02d6b4d6c9
-      rocm_version: 6.3.1
-      vllm_version: 0.6.6
-      pytorch_version: 2.7.0 (2.7.0a0+git3a58512)
-  model_groups:
-    - group: Llama
-      tag: llama
-      models:
-      - model: Llama 3.1 8B
-        mad_tag: pyt_vllm_llama-3.1-8b
-        model_repo: meta-llama/Llama-3.1-8B-Instruct
-        url: https://huggingface.co/meta-llama/Llama-3.1-8B
-        precision: float16
-      - model: Llama 3.1 70B
-        mad_tag: pyt_vllm_llama-3.1-70b
-        model_repo: meta-llama/Llama-3.1-70B-Instruct
-        url: https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct
-        precision: float16
-      - model: Llama 3.1 405B
-        mad_tag: pyt_vllm_llama-3.1-405b
-        model_repo: meta-llama/Llama-3.1-405B-Instruct
-        url: https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct
-        precision: float16
-      - model: Llama 3.2 11B Vision
-        mad_tag: pyt_vllm_llama-3.2-11b-vision-instruct
-        model_repo: meta-llama/Llama-3.2-11B-Vision-Instruct
-        url: https://huggingface.co/meta-llama/Llama-3.2-11B-Vision-Instruct
-        precision: float16
-      - model: Llama 2 7B
-        mad_tag: pyt_vllm_llama-2-7b
-        model_repo: meta-llama/Llama-2-7b-chat-hf
-        url: https://huggingface.co/meta-llama/Llama-2-7b-chat-hf
-        precision: float16
-      - model: Llama 2 70B
-        mad_tag: pyt_vllm_llama-2-70b
-        model_repo: meta-llama/Llama-2-70b-chat-hf
-        url: https://huggingface.co/meta-llama/Llama-2-70b-chat-hf
-        precision: float16
-      - model: Llama 3.1 70B FP8
-        mad_tag: pyt_vllm_llama-3.1-70b_fp8
-        model_repo: amd/Llama-3.1-70B-Instruct-FP8-KV
-        url: https://huggingface.co/amd/Llama-3.1-70B-Instruct-FP8-KV
-        precision: float8
-      - model: Llama 3.1 405B FP8
-        mad_tag: pyt_vllm_llama-3.1-405b_fp8
-        model_repo: amd/Llama-3.1-405B-Instruct-FP8-KV
-        url: https://huggingface.co/amd/Llama-3.1-405B-Instruct-FP8-KV
-        precision: float8
-    - group: Mistral
-      tag: mistral
-      models:
-      - model: Mixtral MoE 8x7B
-        mad_tag: pyt_vllm_mixtral-8x7b
-        model_repo: mistralai/Mixtral-8x7B-Instruct-v0.1
-        url: https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1
-        precision: float16
-      - model: Mixtral MoE 8x22B
-        mad_tag: pyt_vllm_mixtral-8x22b
-        model_repo: mistralai/Mixtral-8x22B-Instruct-v0.1
-        url: https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1
-        precision: float16
-      - model: Mistral 7B
-        mad_tag: pyt_vllm_mistral-7b
-        model_repo: mistralai/Mistral-7B-Instruct-v0.3
-        url: https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3
-        precision: float16
-      - model: Mixtral MoE 8x7B FP8
-        mad_tag: pyt_vllm_mixtral-8x7b_fp8
-        model_repo: amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV
-        url: https://huggingface.co/amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV
-        precision: float8
-      - model: Mixtral MoE 8x22B FP8
-        mad_tag: pyt_vllm_mixtral-8x22b_fp8
-        model_repo: amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV
-        url: https://huggingface.co/amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV
-        precision: float8
-      - model: Mistral 7B FP8
-        mad_tag: pyt_vllm_mistral-7b_fp8
-        model_repo: amd/Mistral-7B-v0.1-FP8-KV
-        url: https://huggingface.co/amd/Mistral-7B-v0.1-FP8-KV
-        precision: float8
-    - group: Qwen
-      tag: qwen
-      models:
-      - model: Qwen2 7B
-        mad_tag: pyt_vllm_qwen2-7b
-        model_repo: Qwen/Qwen2-7B-Instruct
-        url: https://huggingface.co/Qwen/Qwen2-7B-Instruct
-        precision: float16
-      - model: Qwen2 72B
-        mad_tag: pyt_vllm_qwen2-72b
-        model_repo: Qwen/Qwen2-72B-Instruct
-        url: https://huggingface.co/Qwen/Qwen2-72B-Instruct
-        precision: float16
-    - group: JAIS
-      tag: jais
-      models:
-      - model: JAIS 13B
-        mad_tag: pyt_vllm_jais-13b
-        model_repo: core42/jais-13b-chat
-        url: https://huggingface.co/core42/jais-13b-chat
-        precision: float16
-      - model: JAIS 30B
-        mad_tag: pyt_vllm_jais-30b
-        model_repo: core42/jais-30b-chat-v3
-        url: https://huggingface.co/core42/jais-30b-chat-v3
-        precision: float16
-    - group: DBRX
-      tag: dbrx
-      models:
-      - model: DBRX Instruct
-        mad_tag: pyt_vllm_dbrx-instruct
-        model_repo: databricks/dbrx-instruct
-        url: https://huggingface.co/databricks/dbrx-instruct
-        precision: float16
-      - model: DBRX Instruct FP8
-        mad_tag: pyt_vllm_dbrx_fp8
-        model_repo: amd/dbrx-instruct-FP8-KV
-        url: https://huggingface.co/amd/dbrx-instruct-FP8-KV
-        precision: float8
-    - group: Gemma
-      tag: gemma
-      models:
-      - model: Gemma 2 27B
-        mad_tag: pyt_vllm_gemma-2-27b
-        model_repo: google/gemma-2-27b
-        url: https://huggingface.co/google/gemma-2-27b
-        precision: float16
-    - group: Cohere
-      tag: cohere
-      models:
-      - model: C4AI Command R+ 08-2024
-        mad_tag: pyt_vllm_c4ai-command-r-plus-08-2024
-        model_repo: CohereForAI/c4ai-command-r-plus-08-2024
-        url: https://huggingface.co/CohereForAI/c4ai-command-r-plus-08-2024
-        precision: float16
-      - model: C4AI Command R+ 08-2024 FP8
-        mad_tag: pyt_vllm_command-r-plus_fp8
-        model_repo: amd/c4ai-command-r-plus-FP8-KV
-        url: https://huggingface.co/amd/c4ai-command-r-plus-FP8-KV
-        precision: float8
-    - group: DeepSeek
-      tag: deepseek
-      models:
-      - model: DeepSeek MoE 16B
-        mad_tag: pyt_vllm_deepseek-moe-16b-chat
-        model_repo: deepseek-ai/deepseek-moe-16b-chat
-        url: https://huggingface.co/deepseek-ai/deepseek-moe-16b-chat
-        precision: float16

docs/extension/how-to/rocm-for-ai/inference/vllm-benchmark.js

@@ -1,212 +0,0 @@
-function ready(proc) {
-  // Check if page is loaded. If so, init.
-  if (document.readyState !== "loading") {
-    proc();
-  } else {
-    // Otherwise, wait for DOMContentLoaded event.
-    document.addEventListener("DOMContentLoaded", proc);
-  }
-}
-
-ready(() => {
-  const ModelPicker = {
-    // Selector strings for DOM elements
-    SELECTORS: {
-      CONTAINER: "#vllm-benchmark-ud-params-picker",
-      MODEL_GROUP_BTN: 'div[data-param-k="model-group"][data-param-v]',
-      MODEL_PARAM_BTN: 'div[data-param-k="model"][data-param-v]',
-      MODEL_DOC: "div.model-doc",
-    },
-    CSS_CLASSES: {
-      HIDDEN: "hidden",
-    },
-    ATTRIBUTES: {
-      PARAM_KEY: "data-param-k", // URL search parameter key (i.e., "model")
-      PARAM_VALUE: "data-param-v", // URL search param value (e.g., "pyt_vllm_llama-3.1-8b", "pyt_vllm_llama-3.1-70b") -- these are MAD model tags
-      PARAM_GROUP: "data-param-group", // Model group (e.g., "llama", "mistral")
-      PARAM_STATE: "data-param-state", // Selection state
-    },
-
-    // Cache DOM elements
-    elements: {
-      container: null,
-      modelGroups: null,
-      modelParams: null,
-      modelDocs: null,
-    },
-
-    data: {
-      availableModels: new Set(),
-      modelsByGroup: new Map(),
-      modelToGroupMap: new Map(),
-      formattedModelClassMap: new Map(), //TODO
-    },
-
-    init() {
-      this.elements.container = document.querySelector(
-        this.SELECTORS.CONTAINER,
-      );
-      if (!this.elements.container) return;
-
-      this.cacheDOMElements();
-      if (!this.validateElements()) return;
-
-      this.buildModelData();
-      this.bindEvents();
-      this.initializeState();
-    },
-
-    cacheDOMElements() {
-      const { CONTAINER, MODEL_GROUP_BTN, MODEL_PARAM_BTN, MODEL_DOC } =
-        this.SELECTORS;
-      this.elements = {
-        container: document.querySelector(CONTAINER),
-        modelGroups: document.querySelectorAll(MODEL_GROUP_BTN),
-        modelParams: document.querySelectorAll(MODEL_PARAM_BTN),
-        modelDocs: document.querySelectorAll(MODEL_DOC),
-      };
-    },
-
-    validateElements() {
-      const { modelGroups, modelParams } = this.elements;
-      if (!modelGroups.length || !modelParams.length) {
-        console.warn("Model picker is missing required elements");
-        return false;
-      }
-      return true;
-    },
-
-    buildModelData() {
-      const { PARAM_VALUE, PARAM_GROUP } = this.ATTRIBUTES;
-
-      this.elements.modelParams.forEach((model) => {
-        const modelTag = model.getAttribute(PARAM_VALUE);
-        const groupTag = model.getAttribute(PARAM_GROUP);
-
-        if (!modelTag || !groupTag) return;
-
-        this.data.availableModels.add(modelTag);
-        this.data.modelToGroupMap.set(modelTag, groupTag);
-
-        // FIXME: this is because Sphinx auto-formats class names to use dashes
-        this.data.formattedModelClassMap.set(
-          modelTag,
-          modelTag.replace(/[^a-zA-Z0-9]/g, "-"),
-        );
-
-        if (!this.data.modelsByGroup.has(groupTag)) {
-          this.data.modelsByGroup.set(groupTag, []);
-        }
-        this.data.modelsByGroup.get(groupTag).push(modelTag);
-      });
-    },
-
-    // Event listeners for user interactions
-    bindEvents() {
-      const handleInteraction = (event) => {
-        const target = event.target.closest(`[${this.ATTRIBUTES.PARAM_KEY}]`);
-        if (!target) return;
-
-        const paramType = target.getAttribute(this.ATTRIBUTES.PARAM_KEY);
-        const paramValue = target.getAttribute(this.ATTRIBUTES.PARAM_VALUE);
-
-        if (paramType === "model") {
-          const groupTag = target.getAttribute(this.ATTRIBUTES.PARAM_GROUP);
-          if (groupTag) this.updateUI(paramValue, groupTag);
-        } else if (paramType === "model-group") {
-          const firstModelInGroup = this.data.modelsByGroup.get(paramValue)
-            ?.[0];
-          if (firstModelInGroup) this.updateUI(firstModelInGroup, paramValue);
-        }
-      };
-
-      this.elements.container.addEventListener("click", handleInteraction);
-      this.elements.container.addEventListener("keydown", (event) => {
-        if (event.key === "Enter" || event.key === " ") {
-          event.preventDefault();
-          handleInteraction(event);
-        }
-      });
-    },
-
-    // Update the page based on the selected model
-    updateUI(modelTag, groupTag) {
-      const validModel = this.setModelSearchParam(modelTag);
-
-      // Update model group buttons
-      this.elements.modelGroups.forEach((group) => {
-        const isSelected =
-          group.getAttribute(this.ATTRIBUTES.PARAM_VALUE) === groupTag;
-        group.setAttribute(
-          this.ATTRIBUTES.PARAM_STATE,
-          isSelected ? "selected" : "",
-        );
-        group.setAttribute("aria-selected", isSelected.toString());
-      });
-
-      // Update model buttons
-      this.elements.modelParams.forEach((model) => {
-        const isInSelectedGroup =
-          model.getAttribute(this.ATTRIBUTES.PARAM_GROUP) === groupTag;
-        const isSelectedModel =
-          model.getAttribute(this.ATTRIBUTES.PARAM_VALUE) === validModel;
-
-        model.classList.toggle(this.CSS_CLASSES.HIDDEN, !isInSelectedGroup);
-        model.setAttribute(
-          this.ATTRIBUTES.PARAM_STATE,
-          isSelectedModel ? "selected" : "",
-        );
-        model.setAttribute("aria-selected", isSelectedModel.toString());
-      });
-
-      // Update visibility of doc sections
-      const formattedClass = this.data.formattedModelClassMap.get(validModel);
-      if (formattedClass) {
-        this.elements.modelDocs.forEach((doc) => {
-          doc.classList.toggle(
-            this.CSS_CLASSES.HIDDEN,
-            !doc.classList.contains(formattedClass),
-          );
-        });
-      }
-    },
-
-    // Get the current model from the URL search parameters.
-    getModelSearchParam() {
-      return new URLSearchParams(location.search).get("model");
-    },
-
-    // Set the model in the URL search parameters, or fallback to the first available one.
-    setModelSearchParam(modelTag) {
-      const defaultModel = [...this.data.availableModels][0];
-      const model = this.data.availableModels.has(modelTag)
-        ? modelTag
-        : defaultModel;
-
-      const searchParams = new URLSearchParams(location.search);
-      searchParams.set("model", model);
-
-      history.replaceState(
-        {},
-        "",
-        `${location.pathname}?${searchParams.toString()}`,
-      );
-      return model;
-    },
-
-    // Initialize the UI state based on the current URL search parameter or default values.
-    initializeState() {
-      const currentModel = this.getModelSearchParam();
-      const validModel = this.setModelSearchParam(currentModel);
-
-      const initialGroup = this.data.modelToGroupMap.get(validModel) ??
-        [...this.data.modelsByGroup.keys()][0];
-
-      if (initialGroup) {
-        this.updateUI(validModel, initialGroup);
-      }
-    },
-  };
-
-  ModelPicker.init();
-});

docs/extension/version-ref.py

@@ -1,266 +0,0 @@
-from docutils import nodes
-from docutils.parsers.rst import Directive
-from sphinx.util import logging
-import csv
-from io import StringIO
-import re
-import shlex
-
-logger = logging.getLogger(__name__)
-
-class VersionReference(nodes.Inline, nodes.TextElement):
-    """Represents an inline version reference."""
-    pass
-
-class VersionSetDirective(Directive):
-    """Directive for setting version references within a page scope."""
-    required_arguments = 2  # name and value
-    optional_arguments = 0
-
-    def run(self):
-        env = self.state.document.settings.env
-        if not hasattr(env, 'doc_version_refs'):
-            env.doc_version_refs = {}
-        current_doc = env.docname
-        if current_doc not in env.doc_version_refs:
-            env.doc_version_refs[current_doc] = {}
-
-        name, value = self.arguments
-        if name.lower() == 'latest':
-            logger.warning('Cannot override the "latest" keyword with version-set')
-            return []
-
-        # Handle 'latest' value by getting the actual version
-        if value.lower() == 'latest':
-            data = getattr(env, 'compatibility_matrix', None)
-            if data:
-                latest_version = get_latest_rocm_version(data)
-                if latest_version:
-                    value = latest_version
-
-        env.doc_version_refs[current_doc][name] = value
-        return []
-
-def clean_library_name(name):
-    """Extract library name from RST formatting."""
-    # Handle :doc: format
-    doc_match = re.search(r':doc:`([^<]+)(?:\s+<[^>]+>)?`', name)
-    if doc_match:
-        return doc_match.group(1).strip()
-
-    # Handle other link formats
-    link_match = re.search(r'`([^<]+)(?:\s+<[^>]+>)?`_?', name)
-    if link_match:
-        return link_match.group(1).strip()
-
-    return name.strip()
-
-def get_latest_rocm_version(data):
-    """Get the latest ROCm version from the matrix headers."""
-    if not data or len(data) == 0:
-        return None
-
-    # Get all column names except 'ROCm Version'
-    columns = [col for col in data[0].keys() if col != 'ROCm Version']
-    # Return the first column name (assumed to be the latest version)
-    return columns[0] if columns else None
-
-def version_role(name, rawtext, text, lineno, inliner, options={}, content=[]):
-    """
-    Role function to print version value.
-    Usage: :version:`version_name`
-    """
-    try:
-        version_name = text.strip()
-        env = inliner.document.settings.env
-
-        if hasattr(env, 'doc_version_refs'):
-            current_doc = env.docname
-            if current_doc in env.doc_version_refs:
-                doc_refs = env.doc_version_refs[current_doc]
-                if version_name in doc_refs:
-                    version = doc_refs[version_name]
-                    node = nodes.Text(version)
-                    return [node], []
-
-        msg = inliner.reporter.warning(
-            f'No version defined for name {version_name}',
-            line=lineno
-        )
-        return [], [msg]
-
-    except Exception as e:
-        msg = inliner.reporter.error(
-            f'Error looking up version: {str(e)}',
-            line=lineno
-        )
-        prb = inliner.problematic(rawtext, rawtext, msg)
-        return [prb], [msg]
-
-def version_ref_role(name, rawtext, text, lineno, inliner, options={}, content=[]):
-    """
-    Role function for version references.
-    Usage: :version-ref:`library_name release`
-           :version-ref:`"library name" release`
-           :version-ref:`library_name latest`
-           :version-ref:`rocm latest`
-    """
-    try:
-        # Parse the text - handle both quoted and unquoted formats
-        if '"' in text:
-            parts = shlex.split(text)
-        else:
-            parts = text.split()
-
-        if len(parts) != 2:
-            msg = inliner.reporter.error(
-                'Version reference must be in format "library_name release" or "\\"library name\\" release"',
-                line=lineno
-            )
-            prb = inliner.problematic(rawtext, rawtext, msg)
-            return [prb], [msg]
-
-        library_name, release = parts
-        env = inliner.document.settings.env
-
-        # Check if release is a version reference in current document
-        if hasattr(env, 'doc_version_refs'):
-            current_doc = env.docname
-            if current_doc in env.doc_version_refs:
-                doc_refs = env.doc_version_refs[current_doc]
-                if release in doc_refs:
-                    release = doc_refs[release]
-
-        # Handle special case for "rocm latest"
-        if library_name.lower() == 'rocm' and release.lower() == 'latest':
-            data = getattr(env, 'compatibility_matrix', None)
-            if not data:
-                raise ValueError("Compatibility matrix not found in environment")
-
-            latest_version = get_latest_rocm_version(data)
-            if latest_version:
-                node = VersionReference()
-                node += nodes.Text(latest_version)
-                return [node], []
-            else:
-                msg = inliner.reporter.warning(
-                    'No ROCm versions found in compatibility matrix',
-                    line=lineno
-                )
-                return [], [msg]
-
-        version = lookup_version(inliner, library_name, release)
-
-        if version:
-            node = VersionReference()
-            node += nodes.Text(version)
-            return [node], []
-        else:
-            msg = inliner.reporter.warning(
-                f'No version found for library {library_name} in release {release}',
-                line=lineno
-            )
-            return [], [msg]
-
-    except Exception as e:
-        msg = inliner.reporter.error(
-            f'Error looking up version: {str(e)}',
-            line=lineno
-        )
-        prb = inliner.problematic(rawtext, rawtext, msg)
-        return [prb], [msg]
-
-def lookup_version(inliner, library_name, release):
-    """Look up the version in the compatibility matrix."""
-    env = inliner.document.settings.env
-    data = getattr(env, 'compatibility_matrix', None)
-
-    if not data:
-        raise ValueError("Compatibility matrix not found in environment")
-
-    # Handle the 'latest' keyword
-    if release.lower() == 'latest':
-        latest_version = get_latest_rocm_version(data)
-        if not latest_version:
-            return None
-        release = latest_version
-
-    # For ROCm, check if the version exists in column headers
-    if library_name.lower() == 'rocm':
-        columns = [col for col in data[0].keys() if col != 'ROCm Version']
-        if release in columns:
-            return release
-        return None
-
-    # Find the library version
-    for row in data:
-        row_lib_name = clean_library_name(row['ROCm Version'])
-        if row_lib_name == library_name:
-            # Get the version, removing any whitespace
-            version = row.get(release, '').strip()
-            if version:
-                return version
-
-    # If not found, try a case-insensitive search
-    for row in data:
-        row_lib_name = clean_library_name(row['ROCm Version'])
-        if row_lib_name.lower() == library_name.lower():
-            version = row.get(release, '').strip()
-            if version:
-                return version
-
-    return None
-
-def visit_version_reference(self, node):
-    self.body.append(f'<span class="version-reference">')
-
-def depart_version_reference(self, node):
-    self.body.append('</span>')
-
-def load_compatibility_matrix(app):
-    """Load the compatibility matrix content from CSV."""
-    if not app.config.compatibility_matrix_file:
-        logger.warning('No compatibility matrix file configured')
-        return
-
-    try:
-        with open(app.config.compatibility_matrix_file, 'r', encoding='utf-8') as f:
-            reader = csv.DictReader(f)
-            app.env.compatibility_matrix = list(reader)
-            logger.info('Successfully loaded compatibility matrix')
-
-            # Debug: print first few rows with their library names
-            for row in list(app.env.compatibility_matrix)[:5]:
-                if 'ROCm Version' in row:
-                    lib_name = clean_library_name(row['ROCm Version'])
-                    logger.debug(f"Loaded library: {lib_name}")
-
-    except Exception as e:
-        logger.error(f'Error loading compatibility matrix: {str(e)}')
-
-def purge_version_refs(app, env, docname):
-    """Remove version references for a document when it is purged"""
-    if hasattr(env, 'doc_version_refs'):
-        if docname in env.doc_version_refs:
-            del env.doc_version_refs[docname]
-
-def setup(app):
-    app.add_node(VersionReference,
-                 html=(visit_version_reference, depart_version_reference))
-    app.add_role('version-ref', version_ref_role)
-    app.add_role('version', version_role)
-    app.add_directive('version-set', VersionSetDirective)
-
-    # Add a config value for the compatibility matrix file path
-    app.add_config_value('compatibility_matrix_file', None, 'env')
-
-    # Connect to the builder-inited event to load the matrix
-    app.connect('builder-inited', load_compatibility_matrix)
-
-    # Connect to env-purge-doc event to clean up document-specific version refs
-    app.connect('env-purge-doc', purge_version_refs)
-
-    return {
-        'parallel_read_safe': True,
-        'parallel_write_safe': True,
-    }

docs/how-to/rocm-for-ai/inference/vllm-benchmark.rst

@@ -9,266 +9,422 @@ LLM inference performance validation on AMD Instinct MI300X
 
 .. _vllm-benchmark-unified-docker:
 
-.. datatemplate:yaml:: /data/how-to/rocm-for-ai/inference/vllm-benchmark-models.yaml
+The `ROCm vLLM Docker <https://hub.docker.com/r/rocm/vllm/tags>`_ image offers
+a prebuilt, optimized environment for validating large language model (LLM)
+inference performance on the AMD Instinct™ MI300X accelerator. This ROCm vLLM
+Docker image integrates vLLM and PyTorch tailored specifically for the MI300X
+accelerator and includes the following components:
 
-   {% set unified_docker = data.vllm_benchmark.unified_docker.latest %}
-   {% set model_groups = data.vllm_benchmark.model_groups %}
+* `ROCm 6.3.1 <https://github.com/ROCm/ROCm>`_
 
-   The `ROCm vLLM Docker <{{ unified_docker.docker_hub_url }}>`_ image offers
-   a prebuilt, optimized environment for validating large language model (LLM)
-   inference performance on the AMD Instinct™ MI300X accelerator. This ROCm vLLM
-   Docker image integrates vLLM and PyTorch tailored specifically for the MI300X
-   accelerator and includes the following components:
+* `vLLM 0.6.6 <https://docs.vllm.ai/en/latest>`_
 
-   * `ROCm {{ unified_docker.rocm_version }} <https://github.com/ROCm/ROCm>`_
+* `PyTorch 2.7.0 (2.7.0a0+git3a58512) <https://github.com/pytorch/pytorch>`_
 
-   * `vLLM {{ unified_docker.vllm_version }} <https://docs.vllm.ai/en/latest>`_
+With this Docker image, you can quickly validate the expected inference
+performance numbers for the MI300X accelerator. This topic also provides tips on
+optimizing performance with popular AI models. For more information, see the lists of
+:ref:`available models for MAD-integrated benchmarking <vllm-benchmark-mad-models>`
+and :ref:`standalone benchmarking <vllm-benchmark-standalone-options>`.
 
-   * `PyTorch {{ unified_docker.pytorch_version }} <https://github.com/pytorch/pytorch>`_
+.. _vllm-benchmark-vllm:
 
-   With this Docker image, you can quickly validate the expected inference
-   performance numbers for the MI300X accelerator. This topic also provides tips on
-   optimizing performance with popular AI models.
+.. note::
 
-   .. _vllm-benchmark-available-models:
+   vLLM is a toolkit and library for LLM inference and serving. AMD implements
+   high-performance custom kernels and modules in vLLM to enhance performance.
+   See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
+   more information.
 
-   Available models
-   ================
+Getting started
+===============
 
-   .. raw:: html
+Use the following procedures to reproduce the benchmark results on an
+MI300X accelerator with the prebuilt vLLM Docker image.
 
-      <div id="vllm-benchmark-ud-params-picker" class="container-fluid">
-        <div class="row">
-          <div class="col-2 me-2 model-param-head">Model</div>
-          <div class="row col-10">
-   {% for model_group in model_groups %}
-            <div class="col-3 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
-   {% endfor %}
-          </div>
-        </div>
+.. _vllm-benchmark-get-started:
 
-        <div class="row mt-1">
-          <div class="col-2 me-2 model-param-head">Model variant</div>
-          <div class="row col-10">
-   {% for model_group in model_groups %}
-      {% set models = model_group.models %}
-      {% for model in models %}
-         {% if models|length % 3 == 0 %}
-            <div class="col-4 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
-         {% else %}
-            <div class="col-6 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
-         {% endif %}
-      {% endfor %}
-   {% endfor %}
-          </div>
-        </div>
-      </div>
+1. Disable NUMA auto-balancing.
 
-   .. _vllm-benchmark-vllm:
+   To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
+   might hang until the periodic balancing is finalized. For more information,
+   see :ref:`AMD Instinct MI300X system optimization <mi300x-disable-numa>`.
 
-   {% for model_group in model_groups %}
-      {% for model in model_group.models %}
+   .. code-block:: shell
 
-   .. container:: model-doc {{model.mad_tag}}
+      # disable automatic NUMA balancing
+      sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
+      # check if NUMA balancing is disabled (returns 0 if disabled)
+      cat /proc/sys/kernel/numa_balancing
+      0
 
-      .. note::
+2. Download the :ref:`ROCm vLLM Docker image <vllm-benchmark-unified-docker>`.
 
-         See the `{{ model.model }} model card on Hugging Face <{{ model.url }}>`_ to learn more about your selected model.
-         Some models require access authorization prior to use via an external license agreement through a third party.
+   Use the following command to pull the Docker image from Docker Hub.
 
-      {% endfor %}
-   {% endfor %}
+   .. code-block:: shell
 
+      docker pull rocm/vllm:rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6
 
-   .. note::
+Once the setup is complete, choose between two options to reproduce the
+benchmark results:
 
-      vLLM is a toolkit and library for LLM inference and serving. AMD implements
-      high-performance custom kernels and modules in vLLM to enhance performance.
-      See :ref:`fine-tuning-llms-vllm` and :ref:`mi300x-vllm-optimization` for
-      more information.
+-  :ref:`MAD-integrated benchmarking <vllm-benchmark-mad>`
 
-   Getting started
-   ===============
+-  :ref:`Standalone benchmarking <vllm-benchmark-standalone>`
 
-   Use the following procedures to reproduce the benchmark results on an
-   MI300X accelerator with the prebuilt vLLM Docker image.
+.. _vllm-benchmark-mad:
 
-   .. _vllm-benchmark-get-started:
+MAD-integrated benchmarking
+===========================
 
-   1. Disable NUMA auto-balancing.
+Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
+directory and install the required packages on the host machine.
 
-      To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
-      might hang until the periodic balancing is finalized. For more information,
-      see :ref:`AMD Instinct MI300X system optimization <mi300x-disable-numa>`.
+.. code-block:: shell
 
-      .. code-block:: shell
+   git clone https://github.com/ROCm/MAD
+   cd MAD
+   pip install -r requirements.txt
 
-         # disable automatic NUMA balancing
-         sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
-         # check if NUMA balancing is disabled (returns 0 if disabled)
-         cat /proc/sys/kernel/numa_balancing
-         0
+Use this command to run a performance benchmark test of the Llama 3.1 8B model
+on one GPU with ``float16`` data type in the host machine.
 
-   2. Download the `ROCm vLLM Docker image <{{ unified_docker.docker_hub_url }}>`_.
+.. code-block:: shell
 
-      Use the following command to pull the Docker image from Docker Hub.
+   export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
+   python3 tools/run_models.py --tags pyt_vllm_llama-3.1-8b --keep-model-dir --live-output --timeout 28800
 
-      .. code-block:: shell
+ROCm MAD launches a Docker container with the name
+``container_ci-pyt_vllm_llama-3.1-8b``. The latency and throughput reports of the
+model are collected in the following path: ``~/MAD/reports_float16/``.
 
-         docker pull {{ unified_docker.pull_tag }}
+Although the following models are preconfigured to collect latency and
+throughput performance data, you can also change the benchmarking parameters.
+Refer to the :ref:`Standalone benchmarking <vllm-benchmark-standalone>` section.
 
-   Benchmarking
-   ============
+.. _vllm-benchmark-mad-models:
 
-   Once the setup is complete, choose between two options to reproduce the
-   benchmark results:
+Available models
+----------------
 
-   .. _vllm-benchmark-mad:
+.. list-table::
+   :header-rows: 1
+   :widths: 2, 3
 
-   {% for model_group in model_groups %}
-      {% for model in model_group.models %}
+   * - Model name
+     - Tag
 
-   .. container:: model-doc {{model.mad_tag}}
+   * - `Llama 3.1 8B <https://huggingface.co/meta-llama/Llama-3.1-8B>`_
+     - ``pyt_vllm_llama-3.1-8b``
 
-      .. tab-set::
+   * - `Llama 3.1 70B <https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct>`_
+     - ``pyt_vllm_llama-3.1-70b``
 
-         .. tab-item:: MAD-integrated benchmarking
+   * - `Llama 3.1 405B <https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct>`_
+     - ``pyt_vllm_llama-3.1-405b``
 
-            Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
-            directory and install the required packages on the host machine.
+   * - `Llama 3.2 11B Vision <https://huggingface.co/meta-llama/Llama-3.2-11B-Vision-Instruct>`_
+     - ``pyt_vllm_llama-3.2-11b-vision-instruct``
 
-            .. code-block:: shell
+   * - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-7b-chat-hf>`_
+     - ``pyt_vllm_llama-2-7b``
 
-               git clone https://github.com/ROCm/MAD
-               cd MAD
-               pip install -r requirements.txt
+   * - `Llama 2 70B <https://huggingface.co/meta-llama/Llama-2-70b-chat-hf>`_
+     - ``pyt_vllm_llama-2-70b``
 
-            Use this command to run the performance benchmark test on the `{{model.model}} <{{ model.url }}>`_ model
-            using one GPU with the ``{{model.precision}}`` data type on the host machine.
+   * - `Mixtral MoE 8x7B <https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1>`_
+     - ``pyt_vllm_mixtral-8x7b``
 
-            .. code-block:: shell
+   * - `Mixtral MoE 8x22B <https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1>`_
+     - ``pyt_vllm_mixtral-8x22b``
 
-               export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
-               python3 tools/run_models.py --tags {{model.mad_tag}} --keep-model-dir --live-output --timeout 28800
+   * - `Mistral 7B <https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3>`_
+     - ``pyt_vllm_mistral-7b``
 
-            MAD launches a Docker container with the name
-            ``container_ci-{{model.mad_tag}}``. The latency and throughput reports of the
-            model are collected in the following path: ``~/MAD/reports_{{model.precision}}/``.
+   * - `Qwen2 7B <https://huggingface.co/Qwen/Qwen2-7B-Instruct>`_
+     - ``pyt_vllm_qwen2-7b``
 
-            Although the :ref:`available models <vllm-benchmark-available-models>` are preconfigured
-            to collect latency and throughput performance data, you can also change the benchmarking
-            parameters. See the standalone benchmarking tab for more information.
+   * - `Qwen2 72B <https://huggingface.co/Qwen/Qwen2-72B-Instruct>`_
+     - ``pyt_vllm_qwen2-72b``
 
-         .. tab-item:: Standalone benchmarking
+   * - `JAIS 13B <https://huggingface.co/core42/jais-13b-chat>`_
+     - ``pyt_vllm_jais-13b``
 
-            Run the vLLM benchmark tool independently by starting the
-            `Docker container <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6/images/sha256-9a12ef62bbbeb5a4c30a01f702c8e025061f575aa129f291a49fbd02d6b4d6c9>`_
-            as shown in the following snippet.
+   * - `JAIS 30B <https://huggingface.co/core42/jais-30b-chat-v3>`_
+     - ``pyt_vllm_jais-30b``
 
-            .. code-block::
+   * - `DBRX Instruct <https://huggingface.co/databricks/dbrx-instruct>`_
+     - ``pyt_vllm_dbrx-instruct``
 
-               docker pull rocm/vllm:rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6
-               docker run -it --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 16G --security-opt seccomp=unconfined --security-opt apparmor=unconfined --cap-add=SYS_PTRACE -v $(pwd):/workspace --env HUGGINGFACE_HUB_CACHE=/workspace --name vllm_v0.6.6 rocm/vllm:rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6
+   * - `Gemma 2 27B <https://huggingface.co/google/gemma-2-27b>`_
+     - ``pyt_vllm_gemma-2-27b``
 
-            In the Docker container, clone the ROCm MAD repository and navigate to the
-            benchmark scripts directory at ``~/MAD/scripts/vllm``.
+   * - `C4AI Command R+ 08-2024 <https://huggingface.co/CohereForAI/c4ai-command-r-plus-08-2024>`_
+     - ``pyt_vllm_c4ai-command-r-plus-08-2024``
 
-            .. code-block::
+   * - `DeepSeek MoE 16B <https://huggingface.co/deepseek-ai/deepseek-moe-16b-chat>`_
+     - ``pyt_vllm_deepseek-moe-16b-chat``
 
-               git clone https://github.com/ROCm/MAD
-               cd MAD/scripts/vllm
+   * - `Llama 3.1 70B FP8 <https://huggingface.co/amd/Llama-3.1-70B-Instruct-FP8-KV>`_
+     - ``pyt_vllm_llama-3.1-70b_fp8``
 
-            To start the benchmark, use the following command with the appropriate options.
+   * - `Llama 3.1 405B FP8 <https://huggingface.co/amd/Llama-3.1-405B-Instruct-FP8-KV>`_
+     - ``pyt_vllm_llama-3.1-405b_fp8``
 
-            .. code-block::
+   * - `Mixtral MoE 8x7B FP8 <https://huggingface.co/amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV>`_
+     - ``pyt_vllm_mixtral-8x7b_fp8``
 
-               ./vllm_benchmark_report.sh -s $test_option -m {{model.model_repo}} -g $num_gpu -d {{model.precision}}
+   * - `Mixtral MoE 8x22B FP8 <https://huggingface.co/amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV>`_
+     - ``pyt_vllm_mixtral-8x22b_fp8``
 
-            .. list-table::
-               :header-rows: 1
-               :align: center
+   * - `Mistral 7B FP8 <https://huggingface.co/amd/Mistral-7B-v0.1-FP8-KV>`_
+     - ``pyt_vllm_mistral-7b_fp8``
 
-               * - Name
-                 - Options
-                 - Description
+   * - `DBRX Instruct FP8 <https://huggingface.co/amd/dbrx-instruct-FP8-KV>`_
+     - ``pyt_vllm_dbrx_fp8``
 
-               * - ``$test_option``
-                 - latency
-                 - Measure decoding token latency
+   * - `C4AI Command R+ 08-2024 FP8 <https://huggingface.co/amd/c4ai-command-r-plus-FP8-KV>`_
+     - ``pyt_vllm_command-r-plus_fp8``
 
-               * -
-                 - throughput
-                 - Measure token generation throughput
+.. _vllm-benchmark-standalone:
 
-               * -
-                 - all
-                 - Measure both throughput and latency
+Standalone benchmarking
+=======================
 
-               * - ``$num_gpu``
-                 - 1 or 8
-                 - Number of GPUs
+You can run the vLLM benchmark tool independently by starting the
+`Docker container <https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6/images/sha256-9a12ef62bbbeb5a4c30a01f702c8e025061f575aa129f291a49fbd02d6b4d6c9>`_
+as shown in the following snippet.
 
-               * - ``$datatype``
-                 - ``float16`` or ``float8``
-                 - Data type
+.. code-block::
 
-            .. note::
+   docker pull rocm/vllm:rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6
+   docker run -it --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 16G --security-opt seccomp=unconfined --security-opt apparmor=unconfined --cap-add=SYS_PTRACE -v $(pwd):/workspace --env HUGGINGFACE_HUB_CACHE=/workspace --name vllm_v0.6.6 rocm/vllm:rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6
 
-               The input sequence length, output sequence length, and tensor parallel (TP) are
-               already configured. You don't need to specify them with this script.
+In the Docker container, clone the ROCm MAD repository and navigate to the
+benchmark scripts directory at ``~/MAD/scripts/vllm``.
 
-            .. note::
+.. code-block::
 
-               If you encounter the following error, pass your access-authorized Hugging
-               Face token to the gated models.
+   git clone https://github.com/ROCm/MAD
+   cd MAD/scripts/vllm
 
-               .. code-block::
+Command
+-------
 
-                  OSError: You are trying to access a gated repo.
+To start the benchmark, use the following command with the appropriate options.
+See :ref:`Options <vllm-benchmark-standalone-options>` for the list of
+options and their descriptions.
 
-                  # pass your HF_TOKEN
-                  export HF_TOKEN=$your_personal_hf_token
+.. code-block:: shell
 
-            Here are some examples of running the benchmark with various options.
+   ./vllm_benchmark_report.sh -s $test_option -m $model_repo -g $num_gpu -d $datatype
 
-            * Latency benchmark
+See the :ref:`examples <vllm-benchmark-run-benchmark>` for more information.
 
-              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with the ``{{model.precision}}`` data type.
+.. note::
 
-              .. code-block::
+   The input sequence length, output sequence length, and tensor parallel (TP) are
+   already configured. You don't need to specify them with this script.
 
-                 ./vllm_benchmark_report.sh -s latency -m {{model.model_repo}} -g 8 -d {{model.precision}}
+.. note::
 
-              Find the latency report at ``./reports_{{model.precision}}_vllm_rocm{{unified_docker.rocm_version}}/summary/{{model.model_repo.split('/', 1)[1] if '/' in model.model_repo else model.model_repo}}_latency_report.csv``.
+   If you encounter the following error, pass your access-authorized Hugging
+   Face token to the gated models.
 
-            * Throughput benchmark
+   .. code-block:: shell
 
-              Use this command to throughput the latency of the {{model.model}} model on eight GPUs with the ``{{model.precision}}`` data type.
+      OSError: You are trying to access a gated repo.
 
-              .. code-block:: shell
+      # pass your HF_TOKEN
+      export HF_TOKEN=$your_personal_hf_token
 
-                 ./vllm_benchmark_report.sh -s latency -m {{model.model_repo}} -g 8 -d {{model.precision}}
+.. _vllm-benchmark-standalone-options:
 
-              Find the throughput report at ``./reports_{{model.precision}}_vllm_rocm{{unified_docker.rocm_version}}/summary/{{model.model_repo.split('/', 1)[1] if '/' in model.model_repo else model.model_repo}}_throughput_report.csv``.
+Options and available models
+----------------------------
 
-            .. raw:: html
+.. list-table::
+   :header-rows: 1
+   :align: center
 
-               <style>
-               mjx-container[jax="CHTML"][display="true"] {
-                  text-align: left;
-                  margin: 0;
-               }
-               </style>
+   * - Name
+     - Options
+     - Description
 
-            .. note::
+   * - ``$test_option``
+     - latency
+     - Measure decoding token latency
 
-               Throughput is calculated as:
+   * -
+     - throughput
+     - Measure token generation throughput
 
-               - .. math:: throughput\_tot = requests \times (\mathsf{\text{input lengths}} + \mathsf{\text{output lengths}}) / elapsed\_time
+   * -
+     - all
+     - Measure both throughput and latency
 
-               - .. math:: throughput\_gen = requests \times \mathsf{\text{output lengths}} / elapsed\_time
-      {% endfor %}
-   {% endfor %}
+   * - ``$model_repo``
+     - ``meta-llama/Llama-3.1-8B-Instruct``
+     - `Llama 3.1 8B <https://huggingface.co/meta-llama/Llama-3.1-8B>`_
+
+   * - (``float16``)
+     - ``meta-llama/Llama-3.1-70B-Instruct``
+     - `Llama 3.1 70B <https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct>`_
+
+   * -
+     - ``meta-llama/Llama-3.1-405B-Instruct``
+     - `Llama 3.1 405B <https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct>`_
+
+   * -
+     - ``meta-llama/Llama-3.2-11B-Vision-Instruct``
+     - `Llama 3.2 11B Vision <https://huggingface.co/meta-llama/Llama-3.2-11B-Vision-Instruct>`_
+
+   * -
+     - ``meta-llama/Llama-2-7b-chat-hf``
+     - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-7b-chat-hf>`_
+
+   * -
+     - ``meta-llama/Llama-2-70b-chat-hf``
+     - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-70b-chat-hf>`_
+
+   * -
+     - ``mistralai/Mixtral-8x7B-Instruct-v0.1``
+     - `Mixtral MoE 8x7B <https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1>`_
+
+   * -
+     - ``mistralai/Mixtral-8x22B-Instruct-v0.1``
+     - `Mixtral MoE 8x22B <https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1>`_
+
+   * -
+     - ``mistralai/Mistral-7B-Instruct-v0.3``
+     - `Mistral 7B <https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3>`_
+
+   * -
+     - ``Qwen/Qwen2-7B-Instruct``
+     - `Qwen2 7B <https://huggingface.co/Qwen/Qwen2-7B-Instruct>`_
+
+   * -
+     - ``Qwen/Qwen2-72B-Instruct``
+     - `Qwen2 72B <https://huggingface.co/Qwen/Qwen2-72B-Instruct>`_
+
+   * -
+     - ``core42/jais-13b-chat``
+     - `JAIS 13B <https://huggingface.co/core42/jais-13b-chat>`_
+
+   * -
+     - ``core42/jais-30b-chat-v3``
+     - `JAIS 30B <https://huggingface.co/core42/jais-30b-chat-v3>`_
+
+   * -
+     - ``databricks/dbrx-instruct``
+     - `DBRX Instruct <https://huggingface.co/databricks/dbrx-instruct>`_
+
+   * -
+     - ``google/gemma-2-27b``
+     - `Gemma 2 27B <https://huggingface.co/google/gemma-2-27b>`_
+
+   * -
+     - ``CohereForAI/c4ai-command-r-plus-08-2024``
+     - `C4AI Command R+ 08-2024 <https://huggingface.co/CohereForAI/c4ai-command-r-plus-08-2024>`_
+
+   * -
+     - ``deepseek-ai/deepseek-moe-16b-chat``
+     - `DeepSeek MoE 16B <https://huggingface.co/deepseek-ai/deepseek-moe-16b-chat>`_
+
+   * - ``$model_repo``
+     - ``amd/Llama-3.1-70B-Instruct-FP8-KV``
+     - `Llama 3.1 70B FP8 <https://huggingface.co/amd/Llama-3.1-70B-Instruct-FP8-KV>`_
+
+   * - (``float8``)
+     - ``amd/Llama-3.1-405B-Instruct-FP8-KV``
+     - `Llama 3.1 405B FP8 <https://huggingface.co/amd/Llama-3.1-405B-Instruct-FP8-KV>`_
+
+   * -
+     - ``amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV``
+     - `Mixtral MoE 8x7B FP8 <https://huggingface.co/amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV>`_
+
+   * -
+     - ``amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV``
+     - `Mixtral MoE 8x22B FP8 <https://huggingface.co/amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV>`_
+
+   * -
+     - ``amd/Mistral-7B-v0.1-FP8-KV``
+     - `Mistral 7B FP8 <https://huggingface.co/amd/Mistral-7B-v0.1-FP8-KV>`_
+
+   * -
+     - ``amd/dbrx-instruct-FP8-KV``
+     - `DBRX Instruct FP8 <https://huggingface.co/amd/dbrx-instruct-FP8-KV>`_
+
+   * -
+     - ``amd/c4ai-command-r-plus-FP8-KV``
+     - `C4AI Command R+ 08-2024 FP8 <https://huggingface.co/amd/c4ai-command-r-plus-FP8-KV>`_
+
+   * - ``$num_gpu``
+     - 1 or 8
+     - Number of GPUs
+
+   * - ``$datatype``
+     - ``float16`` or ``float8``
+     - Data type
+
+.. _vllm-benchmark-run-benchmark:
+
+Running the benchmark on the MI300X accelerator
+-----------------------------------------------
+
+Here are some examples of running the benchmark with various options.
+See :ref:`Options <vllm-benchmark-standalone-options>` for the list of
+options and their descriptions.
+
+Example 1: latency benchmark
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+ 
+Use this command to benchmark the latency of the Llama 3.1 70B model on eight GPUs with the ``float16`` and ``float8`` data types.
+
+.. code-block::
+
+   ./vllm_benchmark_report.sh -s latency -m meta-llama/Llama-3.1-70B-Instruct -g 8 -d float16
+   ./vllm_benchmark_report.sh -s latency -m amd/Llama-3.1-70B-Instruct-FP8-KV -g 8 -d float8
+
+Find the latency reports at:
+
+- ``./reports_float16/summary/Llama-3.1-70B-Instruct_latency_report.csv``
+
+- ``./reports_float8/summary/Llama-3.1-70B-Instruct-FP8-KV_latency_report.csv``
+
+Example 2: throughput benchmark
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Use this command to benchmark the throughput of the Llama 3.1 70B model on eight GPUs with the ``float16`` and ``float8`` data types.
+
+.. code-block:: shell
+
+   ./vllm_benchmark_report.sh -s throughput -m meta-llama/Llama-3.1-70B-Instruct -g 8 -d float16
+   ./vllm_benchmark_report.sh -s throughput -m amd/Llama-3.1-70B-Instruct-FP8-KV -g 8 -d float8
+
+Find the throughput reports at:
+
+- ``./reports_float16/summary/Llama-3.1-70B-Instruct_throughput_report.csv``
+
+- ``./reports_float8/summary/Llama-3.1-70B-Instruct-FP8-KV_throughput_report.csv``
+
+.. raw:: html
+
+   <style>
+   mjx-container[jax="CHTML"][display="true"] {
+       text-align: left;
+       margin: 0;
+   }
+   </style>
+
+.. note::
+
+   Throughput is calculated as:
+
+   - .. math:: throughput\_tot = requests \times (\mathsf{\text{input lengths}} + \mathsf{\text{output lengths}}) / elapsed\_time
+
+   - .. math:: throughput\_gen = requests \times \mathsf{\text{output lengths}} / elapsed\_time
 
 Further reading
 ===============
@@ -290,3 +446,33 @@ Further reading
 
 - To learn how to fine-tune LLMs, see
   :doc:`Fine-tuning LLMs <../fine-tuning/index>`.
+
+Previous versions
+=================
+
+This table lists previous versions of the ROCm vLLM Docker image for inference
+performance validation. For detailed information about available models for
+benchmarking, see the version-specific documentation.
+
+.. list-table::
+   :header-rows: 1
+   :stub-columns: 1
+
+   * - ROCm version
+     - vLLM version
+     - PyTorch version
+     - Resources
+
+   * - 6.2.1
+     - 0.6.4
+     - 2.5.0
+     - 
+       * `Documentation <https://rocm.docs.amd.com/en/docs-6.3.0/how-to/performance-validation/mi300x/vllm-benchmark.html>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.2_mi300_ubuntu20.04_py3.9_vllm_0.6.4/images/sha256-ccbb74cc9e7adecb8f7bdab9555f7ac6fc73adb580836c2a35ca96ff471890d8>`_
+
+   * - 6.2.0
+     - 0.4.3
+     - 2.4.0
+     -
+       * `Documentation <https://rocm.docs.amd.com/en/docs-6.2.0/how-to/performance-validation/mi300x/vllm-benchmark.html>`_
+       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.2_mi300_ubuntu22.04_py3.9_vllm_7c5fd50/images/sha256-9e4dd4788a794c3d346d7d0ba452ae5e92d39b8dfac438b2af8efdc7f15d22c0>`_

docs/sphinx/requirements.in

@@ -1,4 +1,3 @@
 rocm-docs-core==1.17.0
 sphinx-reredirects
 sphinx-sitemap
-sphinxcontrib.datatemplates==0.11.0

docs/sphinx/requirements.txt

@@ -1,13 +1,15 @@
 #
-# This file is autogenerated by pip-compile with Python 3.10
+# This file is autogenerated by pip-compile with Python 3.11
 # by the following command:
 #
-#    pip-compile docs/sphinx/requirements.in
+#    pip-compile requirements.in
 #
 accessible-pygments==0.0.5
     # via pydata-sphinx-theme
 alabaster==1.0.0
     # via sphinx
+appnope==0.1.4
+    # via ipykernel
 asttokens==3.0.0
     # via stack-data
 attrs==25.1.0
@@ -23,7 +25,7 @@ beautifulsoup4==4.12.3
     # via pydata-sphinx-theme
 breathe==4.35.0
     # via rocm-docs-core
-certifi==2024.8.30
+certifi==2024.12.14
     # via requests
 cffi==1.17.1
     # via
@@ -37,14 +39,12 @@ click==8.1.7
     #   sphinx-external-toc
 comm==0.2.2
     # via ipykernel
-cryptography==44.0.1
+cryptography==44.0.0
     # via pyjwt
 debugpy==1.8.12
     # via ipykernel
 decorator==5.1.1
     # via ipython
-defusedxml==0.7.1
-    # via sphinxcontrib-datatemplates
 deprecated==1.2.15
     # via pygithub
 docutils==0.21.2
@@ -53,11 +53,9 @@ docutils==0.21.2
     #   myst-parser
     #   pydata-sphinx-theme
     #   sphinx
-exceptiongroup==1.2.2
-    # via ipython
 executing==2.2.0
     # via stack-data
-fastjsonschema==2.20.0
+fastjsonschema==2.21.1
     # via
     #   nbformat
     #   rocm-docs-core
@@ -65,8 +63,6 @@ gitdb==4.0.11
     # via gitpython
 gitpython==3.1.43
     # via rocm-docs-core
-greenlet==3.1.1
-    # via sqlalchemy
 idna==3.10
     # via requests
 imagesize==1.4.1
@@ -77,13 +73,13 @@ importlib-metadata==8.6.1
     #   myst-nb
 ipykernel==6.29.5
     # via myst-nb
-ipython==8.31.0
+ipython==8.32.0
     # via
     #   ipykernel
     #   myst-nb
 jedi==0.19.2
     # via ipython
-jinja2==3.1.5
+jinja2==3.1.4
     # via
     #   myst-parser
     #   sphinx
@@ -117,7 +113,7 @@ mdit-py-plugins==0.4.2
     # via myst-parser
 mdurl==0.1.2
     # via markdown-it-py
-myst-nb==1.1.2
+myst-nb==1.2.0
     # via rocm-docs-core
 myst-parser==4.0.0
     # via myst-nb
@@ -144,7 +140,7 @@ platformdirs==4.3.6
     # via jupyter-core
 prompt-toolkit==3.0.50
     # via ipython
-psutil==6.1.1
+psutil==7.0.0
     # via ipykernel
 ptyprocess==0.7.0
     # via pexpect
@@ -152,7 +148,7 @@ pure-eval==0.2.3
     # via stack-data
 pycparser==2.22
     # via cffi
-pydata-sphinx-theme==0.16.0
+pydata-sphinx-theme==0.16.1
     # via
     #   rocm-docs-core
     #   sphinx-book-theme
@@ -164,7 +160,7 @@ pygments==2.18.0
     #   ipython
     #   pydata-sphinx-theme
     #   sphinx
-pyjwt[crypto]==2.10.0
+pyjwt[crypto]==2.10.1
     # via pygithub
 pynacl==1.5.0
     # via pygithub
@@ -177,8 +173,7 @@ pyyaml==6.0.2
     #   myst-parser
     #   rocm-docs-core
     #   sphinx-external-toc
-    #   sphinxcontrib-datatemplates
-pyzmq==26.2.0
+pyzmq==26.2.1
     # via
     #   ipykernel
     #   jupyter-client
@@ -218,8 +213,6 @@ sphinx==8.1.3
     #   sphinx-notfound-page
     #   sphinx-reredirects
     #   sphinx-sitemap
-    #   sphinxcontrib-datatemplates
-    #   sphinxcontrib-runcmd
 sphinx-book-theme==1.1.3
     # via rocm-docs-core
 sphinx-copybutton==0.5.2
@@ -231,13 +224,11 @@ sphinx-external-toc==1.0.1
 sphinx-notfound-page==1.0.4
     # via rocm-docs-core
 sphinx-reredirects==0.1.5
-    # via -r docs/sphinx/requirements.in
+    # via -r requirements.in
 sphinx-sitemap==2.6.0
-    # via -r docs/sphinx/requirements.in
+    # via -r requirements.in
 sphinxcontrib-applehelp==2.0.0
     # via sphinx
-sphinxcontrib-datatemplates==0.11.0
-    # via -r docs/sphinx/requirements.in
 sphinxcontrib-devhelp==2.0.0
     # via sphinx
 sphinxcontrib-htmlhelp==2.1.0
@@ -246,18 +237,14 @@ sphinxcontrib-jsmath==1.0.1
     # via sphinx
 sphinxcontrib-qthelp==2.0.0
     # via sphinx
-sphinxcontrib-runcmd==0.2.0
-    # via sphinxcontrib-datatemplates
 sphinxcontrib-serializinghtml==2.0.0
     # via sphinx
-sqlalchemy==2.0.37
+sqlalchemy==2.0.38
     # via jupyter-cache
 stack-data==0.6.3
     # via ipython
 tabulate==0.9.0
     # via jupyter-cache
-tomli==2.1.0
-    # via sphinx
 tornado==6.4.2
     # via
     #   ipykernel

docs/sphinx/static/css/vllm-benchmark.css

@@ -1,102 +0,0 @@
-/* ------------------ Compatibility options grid ------------------ */
-html {
-  --compat-border-radius: 2px;
-  --compat-accent-color: var(--pst-color-primary);
-  --compat-bg-color: var(--pst-color-on-background);
-  --compat-fg-color: var(--pst-color-primary-text);
-  --compat-head-color: var(--pst-color-surface);
-  --compat-param-hover-color: var(--pst-color-link-hover);
-  --compat-param-selected-color: var(--pst-color-primary);
-}
-
-html[data-theme="light"] {
-  --compat-border-color: var(--pst-gray-500);
-  --compat-param-disabled-color: var(--pst-gray-300);
-}
-
-html[data-theme="dark"] {
-  --compat-border-color: var(--pst-gray-600);
-  --compat-param-disabled-color: var(--pst-gray-600);
-}
-
-div#vllm-benchmark-ud-params-picker.container-fluid {
-  padding: 0 0 1rem 0;
-}
-
-div[data-param-k="model"] {
-  background-color: var(--compat-bg-color);
-  padding: 2px;
-  border: solid 1px var(--compat-border-color);
-  font-weight: 500;
-  cursor: pointer;
-}
-
-div[data-param-k="model"][data-param-state="selected"] {
-  background-color: var(--compat-param-selected-color);
-  color: var(--compat-fg-color);
-}
-
-div[data-param-k="model"][data-param-state="latest-version"] {
-  background-color: var(--compat-param-selected-color);
-  color: var(--compat-fg-color);
-}
-
-div[data-param-k="model"][data-param-state="disabled"] {
-  background-color: var(--compat-param-disabled-color);
-  text-decoration: line-through;
-  /* text-decoration-color: var(--pst-color-danger); */
-  cursor: auto;
-}
-
-div[data-param-k="model"]:not([data-param-state]):hover {
-  background-color: var(--compat-param-hover-color);
-}
-
-div[data-param-k="model-group"] {
-  background-color: var(--compat-bg-color);
-  padding: 2px;
-  border: solid 1px var(--compat-border-color);
-  font-weight: 500;
-  cursor: pointer;
-}
-
-div[data-param-k="model-group"][data-param-state="selected"] {
-  background-color: var(--compat-param-selected-color);
-  color: var(--compat-fg-color);
-}
-
-div[data-param-k="model-group"][data-param-state="latest-version"] {
-  background-color: var(--compat-param-selected-color);
-  color: var(--compat-fg-color);
-}
-
-div[data-param-k="model-group"][data-param-state="disabled"] {
-  background-color: var(--compat-param-disabled-color);
-  text-decoration: line-through;
-  /* text-decoration-color: var(--pst-color-danger); */
-  cursor: auto;
-}
-
-div[data-param-k="model-group"]:not([data-param-state]):hover {
-  background-color: var(--compat-param-hover-color);
-}
-
-.model-param-head {
-  background-color: var(--compat-head-color);
-  padding: 0.15rem 0.15rem 0.15rem 0.67rem;
-  /* margin: 2px; */
-  border-right: solid 2px var(--compat-accent-color);
-  font-weight: 600;
-}
-
-.model-param {
-  /* padding: 2px; */
-  /* margin: 0 2px 0 2px; */
-  /* margin: 2px; */
-  border: solid 1px var(--compat-border-color);
-  font-weight: 500;
-}
-
-.hidden {
-  display: none !important;
-}

tools/rocm-build/build_composable_kernel.sh

@@ -68,85 +68,6 @@ set_address_sanitizer_off() {
     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" "}')
-    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"

tools/rocm-build/build_lightning.sh

@@ -42,7 +42,6 @@ DEB_PATH="$(getDebPath $PROJ_NAME)"
 RPM_PATH="$(getRpmPath $PROJ_NAME)"
 INSTALL_PATH="${ROCM_INSTALL_PATH}/lib/llvm"
 LLVM_ROOT_LCL="${LLVM_ROOT}"
-ROCM_WHEEL_DIR="${BUILD_PATH}/_wheel"
 
 TARGET="all"
 MAKEOPTS="$DASH_JAY"
@@ -150,7 +149,6 @@ ENABLE_RUNTIMES="$ENABLE_RUNTIMES;libcxx;libcxxabi"
 BOOTSTRAPPING_BUILD_LIBCXX=1
 
 clean_lightning() {
-    rm -rf "$ROCM_WHEEL_DIR"
     rm -rf "$BUILD_PATH"
     rm -rf "$DEB_PATH"
     rm -rf "$RPM_PATH"
@@ -332,15 +330,6 @@ build_lightning() {
     echo "End Workaround for race condition"
     cmake --build . -- $MAKEOPTS
 
-    case "$DISTRO_ID" in
-    (rhel*|centos*)
-       RHEL_BUILD=1
-       ;;
-    (*)
-       RHEL_BUILD=0
-       ;;
-     esac
-
     if [ $SKIP_LIT_TESTS -eq 0 ]; then
         if [ $RHEL_BUILD -eq 1 ]; then
             cmake --build . -- $MAKEOPTS check-lld check-mlir
@@ -1158,9 +1147,4 @@ case $TARGET in
     (*) die "Invalid target $TARGET" ;;
 esac
 
-if [[ $WHEEL_PACKAGE == true ]]; then
-    echo "Wheel Package build started !!!!"
-    create_wheel_package
-fi
-
 echo "Operation complete"

tools/rocm-build/build_omniperf.sh

@@ -1,171 +0,0 @@
-#!/bin/bash
-
-source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
-
-printUsage() {
-    echo
-    echo "Usage: ${BASH_SOURCE##*/} [options ...]"
-    echo
-    echo "Options:"
-    echo "  -c,  --clean              Clean output and delete all intermediate work"
-    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
-    echo "  -p,  --package <type>     Specify packaging format"
-    echo "  -r,  --release            Make a release build instead of a debug build"
-    echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
-                                      type referred to by pkg_type"
-    echo "  -w,  --wheel              Creates python wheel package of omniperf.
-                                      It needs to be used along with -r option"
-    echo "  -h,  --help               Prints this help"
-    echo
-    echo "Possible values for <type>:"
-    echo "  deb -> Debian format (default)"
-    echo "  rpm -> RPM format"
-    echo
-
-    return 0
-}
-
-API_NAME="omniperf"
-PROJ_NAME="$API_NAME"
-LIB_NAME="lib${API_NAME}"
-TARGET="build"
-MAKETARGET="deb"
-PACKAGE_ROOT="$(getPackageRoot)"
-PACKAGE_LIB="$(getLibPath)"
-BUILD_DIR="$(getBuildPath $API_NAME)"
-PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
-PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
-ROCM_WHEEL_DIR="${BUILD_DIR}/_wheel"
-BUILD_TYPE="Debug"
-MAKE_OPTS="$DASH_JAY -C $BUILD_DIR"
-SHARED_LIBS="ON"
-CLEAN_OR_OUT=0;
-MAKETARGET="deb"
-PKGTYPE="deb"
-WHEEL_PACKAGE=false
-
-
-#parse the arguments
-VALID_STR=$(getopt -o hcraso:p:w --long help,clean,release,static,address_sanitizer,outdir:,package:,wheel -- "$@")
-eval set -- "$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)
-                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 ;;
-        -p | --package)
-                MAKETARGET="$2" ; shift 2 ;;
-        -w | --wheel)
-                WHEEL_PACKAGE=true ; shift ;;
-        --)     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 $PROJ_NAME"
-    rm -rf "$ROCM_WHEEL_DIR"
-    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 $PROJ_NAME"
-    if [ "$DISTRO_ID" = centos-7 ]; then
-        echo "Skip make and uploading packages for Omniperf on Centos7 distro, due to python dependency"
-        exit 0
-    fi
-
-    if [ ! -d "$BUILD_DIR" ]; then
-        mkdir -p "$BUILD_DIR"
-        pushd "$BUILD_DIR" || exit
-
-        echo "ROCm CMake Params: $(rocm_cmake_params)"
-        echo "ROCm Common CMake Params: $(rocm_common_cmake_params)"
-
-        print_lib_type $SHARED_LIBS
-        cmake \
-            $(rocm_cmake_params) \
-            $(rocm_common_cmake_params) \
-            -DCHECK_PYTHON_DEPS=NO \
-            -DPYTHON_DEPS=${BUILD_DIR}/python-libs \
-            -DMOD_INSTALL_PATH=${BUILD_DIR}/modulefiles \
-            "$OMNIPERF_ROOT"
-    fi
-
-    make $MAKE_OPTS
-    make $MAKE_OPTS install
-    make $MAKE_OPTS package
-
-    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_DEB" "$BUILD_DIR/${API_NAME}"*.deb
-    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_RPM" "$BUILD_DIR/${API_NAME}"*.rpm
-}
-
-create_wheel_package() {
-    echo "Creating Omniperf wheel package"
-
-    # Copy the setup.py generator to build folder
-    mkdir -p "$ROCM_WHEEL_DIR"
-    cp -f "$SCRIPT_ROOT"/generate_setup_py.py "$ROCM_WHEEL_DIR"
-    cp -f "$SCRIPT_ROOT"/repackage_wheel.sh "$ROCM_WHEEL_DIR"
-    cd "$ROCM_WHEEL_DIR" || exit
-
-    # Currently only supports python3.6
-    ./repackage_wheel.sh "$BUILD_DIR"/*.rpm python3.6
-
-    # Copy the wheel created to RPM folder which will be uploaded to artifactory
-    copy_if WHL "WHL" "$PACKAGE_RPM" "$ROCM_WHEEL_DIR"/dist/*.whl
-}
-
-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
-
-if [[ $WHEEL_PACKAGE == true ]]; then
-    echo "Wheel Package build started !!!!"
-    create_wheel_package
-fi
-
-echo "Operation complete"

tools/rocm-build/build_omnitrace.sh

@@ -1,191 +0,0 @@
-#!/bin/bash
-
-source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
-
-printUsage() {
-    echo
-    echo "Usage: ${BASH_SOURCE##*/} [options ...]"
-    echo
-    echo "Options:"
-    echo "  -c,  --clean              Clean output and delete all intermediate work"
-    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
-    echo "  -p,  --package <type>     Specify packaging format"
-    echo "  -r,  --release            Make a release build instead of a debug build"
-    echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
-                                      type referred to by pkg_type"
-    echo "  -w,  --wheel              Creates python wheel package of omnitrace.
-                                      It needs to be used along with -r option"
-    echo "  -h,  --help               Prints this help"
-    echo
-    echo "Possible values for <type>:"
-    echo "  deb -> Debian format (default)"
-    echo "  rpm -> RPM format"
-    echo
-
-    return 0
-}
-
-API_NAME="omnitrace"
-PROJ_NAME="$API_NAME"
-LIB_NAME="lib${API_NAME}"
-TARGET="build"
-MAKETARGET="deb"
-PACKAGE_ROOT="$(getPackageRoot)"
-PACKAGE_LIB="$(getLibPath)"
-BUILD_DIR="$(getBuildPath $API_NAME)"
-PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
-PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
-BUILD_TYPE="Debug"
-MAKE_OPTS="-j 8"
-SHARED_LIBS="ON"
-CLEAN_OR_OUT=0
-MAKETARGET="deb"
-PKGTYPE="deb"
-ASAN=0
-
-#parse the arguments
-VALID_STR=$(getopt -o hcraso:p:w --long help,clean,release,address_sanitizer,static,outdir:,package:,wheel -- "$@")
-eval set -- "$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)
-        ack_and_ignore_asan
-
-        ASAN=1
-        shift
-        ;;
-    -s | --static)
-        SHARED_LIBS="OFF"
-        shift
-        ;;
-    -o | --outdir)
-        TARGET="outdir"
-        PKGTYPE=$2
-        ((CLEAN_OR_OUT |= 2))
-        shift 2
-        ;;
-    -p | --package)
-        MAKETARGET="$2"
-        shift 2
-        ;;
-    -w | --wheel)
-	   echo "omnitrace: wheel build option accepted and ignored"
-       shift
-       ;;
-    --)
-        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() {
-    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_omnitrace() {
-    echo "Building $PROJ_NAME"
-    if [ "$DISTRO_ID" = "mariner-2.0" ] || [ "$DISTRO_ID" = "ubuntu-24.04" ] || [ "$DISTRO_ID" = "azurelinux-3.0" ]; then
-        echo "Skip make and uploading packages for Omnitrace on \"${DISTRO_ID}\" distro"
-        exit 0
-    fi
-
-    if [ $ASAN == 1 ]; then
-        echo "Skip make and uploading packages for Omnitrace on ASAN build"
-        exit 0
-    fi
-    if [ ! -d "$BUILD_DIR" ]; then
-        mkdir -p "$BUILD_DIR"
-        echo "Created build directory: $BUILD_DIR"
-    fi
-
-    echo "Build directory: $BUILD_DIR"
-    pushd "$BUILD_DIR" || exit
-    print_lib_type $SHARED_LIBS
-
-    echo "ROCm CMake Params: $(rocm_cmake_params)"
-    echo "ROCm Common CMake Params: $(rocm_common_cmake_params)"
-
-
-    if [ $ASAN == 1 ]; then
-        echo "Address Sanitizer path"
-
-    else
-        cmake \
-            $(rocm_cmake_params) \
-            $(rocm_common_cmake_params) \
-            -DOMNITRACE_BUILD_{LIBUNWIND,DYNINST}=ON \
-            -DDYNINST_BUILD_{TBB,BOOST,ELFUTILS,LIBIBERTY}=ON \
-            "$OMNITRACE_ROOT"
-    fi
-
-
-    popd || exit
-
-    echo "Make Options: $MAKE_OPTS"
-    cmake --build "$BUILD_DIR" --target all -- $MAKE_OPTS
-    cmake --build "$BUILD_DIR" --target install -- $MAKE_OPTS
-    cmake --build "$BUILD_DIR" --target package -- $MAKE_OPTS
-
-    copy_if DEB "${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_omnitrace ;;
-outdir) print_output_directory ;;
-*) die "Invalid target $TARGET" ;;
-esac
-
-echo "Operation complete"

tools/rocm-build/build_opencl_icd_loader.sh

@@ -1,141 +0,0 @@
-#!/bin/bash
-
-source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
-PROJ_NAME=OpenCL-ICD-Loader
-TARGET="build"
-MAKEOPTS="$DASH_JAY"
-BUILD_TYPE="Debug"
-PACKAGE_ROOT="$(getPackageRoot)"
-PACKAGE_DEB="$PACKAGE_ROOT/deb/${PROJ_NAME,,}"
-PACKAGE_RPM="$PACKAGE_ROOT/rpm/${PROJ_NAME,,}"
-CLEAN_OR_OUT=0;
-PKGTYPE="deb"
-MAKETARGET="deb"
-API_NAME="rocm-opencl-icd-loader"
-
-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 "  -h,  --help               Prints this help"
-    echo "  -o,  --outdir             Print path of output directory containing packages"
-    echo "  -s,  --static             Component/Build does not support static builds just accepting this param & ignore. No effect of the param on this build"
-    echo
-    echo "Possible values for <type>:"
-    echo "  deb -> Debian format (default)"
-    echo "  rpm -> RPM format"
-    echo
-    return 0
-}
-
-RET_CONFLICT=1
-check_conflicting_options $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
-if [ $RET_CONFLICT -ge 30 ]; then
-   print_vars $TARGET $BUILD_TYPE $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
-   exit $RET_CONFLICT
-fi
-
-clean_opencl_icd_loader() {
-    echo "Cleaning $PROJ_NAME"
-    rm -rf "$PACKAGE_DEB"
-    rm -rf "$PACKAGE_RPM"
-    rm -rf "$PACKAGE_ROOT/${PROJ_NAME,,}"
-}
-
-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
-    cp -r ./pkg/*/rocm*/* "${ROCM_PATH}" || exit 2
-    rm -rf pkg/
-}
-
-build_opencl_icd_loader() {
-    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/${rocm_ver}/pool/main/r/${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 ${PROJ_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
-}
-
-VALID_STR=`getopt -o hcraswlo:p: --long help,clean,release,outdir:,package: -- "$@"`
-eval set -- "$VALID_STR"
-while true ;
-do
-    case "$1" in
-        (-c  | --clean )
-            TARGET="clean" ; ((CLEAN_OR_OUT|=1)) ; shift ;;
-        (-r  | --release )
-            BUILD_TYPE="RelWithDebInfo" ; shift ;;
-        (-h  | --help )
-            printUsage ; exit 0 ;;
-        (-a  | --address_sanitizer)
-            ack_and_ignore_asan ; shift ;;
-        (-o  | --outdir)
-            TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
-        (-p | --package)
-            MAKETARGET="$2" ; shift 2;;
-	(-s | --static)
-            echo "-s parameter accepted but ignored" ; shift ;;
-        --)     shift; break;;
-        (*)
-            echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
-    esac
-done
-
-case $TARGET in
-    (clean) clean_opencl_icd_loader ;;
-    (build) build_opencl_icd_loader ;;
-    (outdir) print_output_directory ;;
-    (*) die "Invalid target $TARGET" ;;
-esac
-
-echo "Operation complete"

tools/rocm-build/build_rocm-cmake.sh

@@ -32,7 +32,6 @@ ROCM_CMAKE_BUILD_DIR="$(getBuildPath rocm-cmake)"
 ROCM_CMAKE_BUILD_DIR="$(getBuildPath rocm-cmake)"
 ROCM_CMAKE_PACKAGE_DEB="$(getPackageRoot)/deb/rocm-cmake"
 ROCM_CMAKE_PACKAGE_RPM="$(getPackageRoot)/rpm/rocm-cmake"
-ROCM_WHEEL_DIR="${ROCM_CMAKE_BUILD_DIR}/_wheel"
 ROCM_CMAKE_BUILD_TYPE="debug"
 BUILD_TYPE="Debug"
 SHARED_LIBS="ON"
@@ -56,8 +55,6 @@ do
                 ack_and_ignore_asan ; shift ;;
         (-s | --static)
                 SHARED_LIBS="OFF" ; shift ;;
-        (-w | --wheel)
-            WHEEL_PACKAGE=true ; shift ;;
         (-o | --outdir)
                 TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
         (-p | --package)
@@ -78,7 +75,6 @@ fi
 
 
 clean_rocm_cmake() {
-    rm -rf "$ROCM_WHEEL_DIR"
     rm -rf $ROCM_CMAKE_BUILD_DIR
     rm -rf $ROCM_CMAKE_PACKAGE_DEB
     rm -rf $ROCM_CMAKE_PACKAGE_RPM
@@ -106,19 +102,6 @@ build_rocm_cmake() {
     copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$ROCM_CMAKE_PACKAGE_RPM" $ROCM_CMAKE_BUILD_DIR/rocm-cmake*.rpm
 }
 
-create_wheel_package() {
-    echo "Creating rocm-cmake wheel package"
-    # Copy the setup.py generator to build folder
-    mkdir -p $ROCM_WHEEL_DIR
-    cp -f $SCRIPT_ROOT/generate_setup_py.py $ROCM_WHEEL_DIR
-    cp -f $SCRIPT_ROOT/repackage_wheel.sh $ROCM_WHEEL_DIR
-    cd $ROCM_WHEEL_DIR
-    # Currently only supports python3.6
-    ./repackage_wheel.sh $ROCM_CMAKE_BUILD_DIR/rocm-cmake*.rpm python3.6
-    # Copy the wheel created to RPM folder which will be uploaded to artifactory
-    copy_if WHL "WHL" "$ROCM_CMAKE_PACKAGE_RPM" "$ROCM_WHEEL_DIR"/dist/*.whl
-}
-
 print_output_directory() {
     case ${PKGTYPE} in
         ("deb")
@@ -138,9 +121,4 @@ case $TARGET in
     (*) die "Invalid target $TARGET" ;;
 esac
 
-if [[ $WHEEL_PACKAGE == true ]]; then
-    echo "Wheel Package build started !!!!"
-    create_wheel_package
-fi
-
 echo "Operation complete"

tools/rocm-build/docker/ubuntu20/packages

@@ -7,7 +7,6 @@ bison
 bridge-utils
 build-essential
 bzip2
-ccache
 check
 chrpath
 cifs-utils
@@ -121,11 +120,9 @@ python3-yaml
 python3.8-dev
 re2c
 redis-tools
-# Eventually we should be able to remove rpm for debian builds.
 rpm
 rsync
 ssh
-# This makes life more pleasent inside the container
 strace
 sudo
 systemtap-sdt-dev

tools/rocm-build/docker/ubuntu22/install-prerequisities.sh

@@ -1,285 +0,0 @@
-#! /usr/bin/bash
-
-set -x
-
-apt-get -y update 
-DEBIAN_FRONTEND=noninteractive DEBCONF_NONINTERACTIVE_SEEN=true apt-get install --no-install-recommends -y $(sed 's/#.*//' /tmp/packages)
-apt-get clean 
-rm -rf /var/cache/apt/ /var/lib/apt/lists/* /etc/apt/apt.conf.d/01proxy
-
-#Install  2.17.1 version of git as we are seeing issues with 2.25 , where it was not allowing to add git submodules if the user is different for parent git directory
-curl -o git.tar.gz https://cdn.kernel.org/pub/software/scm/git/git-2.17.1.tar.gz 
-tar -zxf git.tar.gz 
-cd git-* 
-make prefix=/usr/local all    
-make prefix=/usr/local install
-git --version
-
-#install argparse and CppHeaderParser python modules for roctracer and rocprofiler
-#install rocm-docs-core for the docs-as-code project. Only needed on one OS
-# CppHeader needs setuptools. setuptools needs wheel.
-# Looks like I need them as seperate commands
-# Sigh, install both python2 and python 3 version
-pip3 install --no-cache-dir setuptools wheel tox
-pip3 install --no-cache-dir CppHeaderParser argparse requests lxml barectf recommonmark jinja2==3.0.0 websockets matplotlib numpy scipy minimal msgpack pytest sphinx joblib PyYAML rocm-docs-core cmake==3.25.2 pandas myst-parser
-
-# Allow sudo for everyone user
-echo 'ALL ALL=(ALL) NOPASSWD:ALL' > /etc/sudoers.d/everyone
-
-# Install OCaml packages to build LLVM's OCaml bindings to be used in lightning compiler test pipeline
-wget -nv https://sourceforge.net/projects/opam.mirror/files/2.1.4/opam-2.1.4-x86_64-linux -O /usr/local/bin/opam 
-chmod +x /usr/local/bin/opam
-opam init --yes --disable-sandboxing
-opam install ctypes --yes
-
-# Install and modify git-repo (#!/usr/bin/env python -> #!/usr/bin/env python3)
-curl https://storage.googleapis.com/git-repo-downloads/repo > /usr/bin/repo
-chmod a+x /usr/bin/repo
-
-# Build ccache from the source
-cd /tmp 
-git clone https://github.com/ccache/ccache -b v4.7.5 
-cd ccache 
-mkdir build 
-cd build 
-cmake -DCMAKE_BUILD_TYPE=Release .. 
-make 
-make install 
-cd /tmp 
-rm -rf ccache
-
-# Install sharp from MLNX_OFED_LINUX as dependency for rccl-rdma-sharp-plugins
-cd /var/tmp
-mkdir mlnx 
-wget -O mlnx/tar.tgz https://content.mellanox.com/ofed/MLNX_OFED-24.01-0.3.3.1/MLNX_OFED_LINUX-24.01-0.3.3.1-ubuntu22.04-x86_64.tgz 
-tar -xz -C mlnx -f mlnx/tar.tgz 
-apt-key add mlnx/*/RPM-GPG-KEY-Mellanox
-echo "deb [arch=amd64] file:$(echo $PWD/mlnx/*/DEBS) ./" > /etc/apt/sources.list.d/sharp.list
-apt update
-apt install -y sharp 
-apt clean
-rm -rf /var/cache/apt/ /var/lib/apt/lists/* mlnx /etc/apt/sources.list.d/sharp.list
-
-apt update
-apt -y install libunwind-dev
-apt -y install libgoogle-glog-dev
-
-# Install python3.8 from source
-curl -LO https://www.python.org/ftp/python/3.8.13/Python-3.8.13.tar.xz
-tar -xvf Python-3.8.13.tar.xz
-pwd
-ls /var/tmp/
-ls Python-3.8.13
-mv Python-3.8.13 /opt/
-apt install build-essential zlib1g-dev libncurses5-dev libgdbm-dev libnss3-dev libssl-dev libsqlite3-dev libreadline-dev libffi-dev curl libbz2-dev pkg-config make -y 
-cd /opt/Python-3.8.13/ 
-./configure --enable-optimizations --enable-shared
-make 
-make -j 6 
-make altinstall 
-ldconfig /opt/Python3.8.13
-python3.8 --version
-
-# roctracer and rocprofiler needs this python3.8
-python3.8 -m pip install setuptools wheel
-python3.8 -m pip install CppHeaderParser argparse requests lxml PyYAML joblib
-
-#Install older version of hwloc-devel package for rocrtst
-curl -lO https://download.open-mpi.org/release/hwloc/v1.11/hwloc-1.11.13.tar.bz2
-tar -xvf hwloc-1.11.13.tar.bz2
-cd hwloc-1.11.13
-./configure
-make
-make install
-cp /usr/local/lib/libhwloc.so.5 /usr/lib
-hwloc-info --version
-
-# Install gtest
-mkdir -p /tmp/gtest
-cd /tmp/gtest
-wget https://github.com/google/googletest/archive/refs/tags/v1.14.0.zip -O googletest.zip
-unzip googletest.zip
-cd googletest-1.14.0/ 
-mkdir build
-cd build
-cmake .. 
-make -j$(nproc) 
-make install
-rm -rf /tmp/gtest
-
-## Install gRPC from source
-## RDC Pre-requisites
-GRPC_ARCHIVE=grpc-1.61.0.tar.gz
-mkdir /tmp/grpc
-mkdir /usr/grpc 
-cd /tmp 
-git clone --recurse-submodules -b v1.61.0 https://github.com/grpc/grpc
-cd grpc
-mkdir -p build 
-cd build
-cmake  -DgRPC_INSTALL=ON -DBUILD_SHARED_LIBS=ON -DgRPC_BUILD_TESTS=OFF -DCMAKE_INSTALL_PREFIX=/usr/grpc -DCMAKE_BUILD_TYPE=Release -DCMAKE_CXX_STANDARD=14 -DCMAKE_SHARED_LINKER_FLAGS_INIT=-Wl,--enable-new-dtags,--build-id=sha1,--rpath,'$ORIGIN' .. 
-make -j $(nproc) install 
-rm -rf /tmp/grpc
-
-## rocBLAS Pre-requisites
-## Download prebuilt AMD multithreaded blis (2.0)
-## Reference : https://github.com/ROCmSoftwarePlatform/rocBLAS/blob/develop/install.sh#L403
-mkdir -p /tmp/blis 
-cd /tmp/blis
-wget -O - https://github.com/amd/blis/releases/download/2.0/aocl-blis-mt-ubuntu-2.0.tar.gz | tar xfz - 
-mv amd-blis-mt /usr/blis 
-cd / 
-rm -rf /tmp/blis
-
-## rocBLAS Pre-requisites(SWDEV-404612)
-## Download aocl-linux-gcc-4.2.0_1_amd64.deb
-mkdir -p /tmp/aocl 
-cd /tmp/aocl 
-wget -nv https://download.amd.com/developer/eula/aocl/aocl-4-2/aocl-linux-gcc-4.2.0_1_amd64.deb 
-apt install ./aocl-linux-gcc-4.2.0_1_amd64.deb 
-rm -rf /tmp/aocl
-
-## hipBLAS Pre-requisites
-## lapack(3.9.1v)
-## Reference https://github.com/ROCmSoftwarePlatform/rocSOLVER/blob/develop/install.sh#L174
-lapack_version=3.9.1
-lapack_srcdir=lapack-$lapack_version
-lapack_blddir=lapack-$lapack_version-bld
-mkdir -p /tmp/lapack
-cd /tmp/lapack
-rm -rf "$lapack_srcdir" "$lapack_blddir" 
-wget -O - https://github.com/Reference-LAPACK/lapack/archive/refs/tags/v3.9.1.tar.gz | tar xzf - 
-cmake -H$lapack_srcdir -B$lapack_blddir -DCMAKE_BUILD_TYPE=Release -DCMAKE_Fortran_FLAGS=-fno-optimize-sibling-calls -DBUILD_TESTING=OFF -DCBLAS=ON -DLAPACKE=OFF
-make -j$(nproc) -C "$lapack_blddir"
-make -C "$lapack_blddir" install
-cd $lapack_blddir
-cp -r ./include/* /usr/local/include/ 
-cp -r ./lib/* /usr/local/lib 
-cd / 
-rm -rf /tmp/lapack
-
-## rocSOLVER Pre-requisites
-## FMT(7.1.3v)
-## Reference https://github.com/ROCmSoftwarePlatform/rocSOLVER/blob/develop/install.sh#L152
-fmt_version=7.1.3
-fmt_srcdir=fmt-$fmt_version
-fmt_blddir=fmt-$fmt_version-bld
-mkdir -p /tmp/fmt
-cd /tmp/fmt
-rm -rf "$fmt_srcdir" "$fmt_blddir"
-wget -O - https://github.com/fmtlib/fmt/archive/refs/tags/7.1.3.tar.gz | tar xzf -
-cmake -H$fmt_srcdir -B$fmt_blddir -DCMAKE_BUILD_TYPE=Release -DCMAKE_POSITION_INDEPENDENT_CODE=ON -DCMAKE_CXX_STANDARD=17 -DCMAKE_CXX_EXTENSIONS=OFF -DCMAKE_CXX_STANDARD_REQUIRED=ON -DFMT_DOC=OFF -DFMT_TEST=OFF
-make -j$(nproc) -C "$fmt_blddir"
-make -C "$fmt_blddir" install
-
-# Build and install libjpeg-turbo
-mkdir -p /tmp/libjpeg-turbo 
-cd /tmp/libjpeg-turbo 
-wget -nv https://github.com/rrawther/libjpeg-turbo/archive/refs/heads/2.0.6.2.zip -O libjpeg-turbo-2.0.6.2.zip
-unzip libjpeg-turbo-2.0.6.2.zip 
-cd libjpeg-turbo-2.0.6.2
-mkdir build
-cd build
-cmake -DCMAKE_INSTALL_PREFIX=/usr -DCMAKE_BUILD_TYPE=RELEASE -DENABLE_STATIC=FALSE -DCMAKE_INSTALL_DEFAULT_LIBDIR=lib .. 
-make -j$(nproc) install 
-rm -rf /tmp/libjpeg-turbo
-
-# Get released ninja from source
-mkdir -p /tmp/ninja 
-cd /tmp/ninja 
-wget -nv https://codeload.github.com/Kitware/ninja/zip/refs/tags/v1.11.1.g95dee.kitware.jobserver-1 -O ninja.zip
-unzip ninja.zip
-cd ninja-1.11.1.g95dee.kitware.jobserver-1
-./configure.py --bootstrap 
-cp ninja /usr/local/bin/ 
-rm -rf /tmp/ninja
-
-# Install  FFmpeg and dependencies
-# Build NASM
-mkdir -p /tmp/nasm-2.15.05 
-cd /tmp
-wget -qO- "https://distfiles.macports.org/nasm/nasm-2.15.05.tar.bz2" | tar -xvj 
-cd nasm-2.15.05
-./autogen.sh
-./configure --prefix="/usr/local"
-make -j$(nproc) install
-rm -rf /tmp/nasm-2.15.05
-
-# Build YASM
-mkdir -p /tmp/yasm-1.3.0 
-cd /tmp
-wget -qO- "http://www.tortall.net/projects/yasm/releases/yasm-1.3.0.tar.gz" | tar -xvz 
-cd yasm-1.3.0
-./configure --prefix="/usr/local"
-make -j$(nproc) install
-rm -rf /tmp/yasm-1.3.0
-
-# Build x264
-mkdir -p /tmp/x264-snapshot-20191217-2245-stable
-cd /tmp
-wget -qO-  "https://download.videolan.org/pub/videolan/x264/snapshots/x264-snapshot-20191217-2245-stable.tar.bz2" | tar -xvj
-cd /tmp/x264-snapshot-20191217-2245-stable
-PKG_CONFIG_PATH="/usr/local/lib/pkgconfig" ./configure --prefix="/usr/local" --enable-shared
-make -j$(nproc) install
-rm -rf /tmp/x264-snapshot-20191217-2245-stable
-
-# Build x265
-mkdir -p /tmp/x265_2.7
-cd /tmp
-wget -qO- "https://get.videolan.org/x265/x265_2.7.tar.gz" | tar -xvz
-cd  /tmp/x265_2.7/build/linux
-cmake -G "Unix Makefiles" -DCMAKE_INSTALL_PREFIX="/usr/local" -DENABLE_SHARED:bool=on ../../source
-make -j$(nproc) install
-rm -rf /tmp/x265_2.7
-
-# Build fdk-aac
-mkdir -p /tmp/fdk-aac-2.0.2
-cd /tmp
-wget -qO- "https://sourceforge.net/projects/opencore-amr/files/fdk-aac/fdk-aac-2.0.2.tar.gz" | tar -xvz
-cd /tmp/fdk-aac-2.0.2
-autoreconf -fiv
-./configure --prefix="/usr/local" --enable-shared --disable-static
-make -j$(nproc) install
-rm -rf /tmp/fdk-aac-2.0.2
-
-# Build FFmpeg
-cd /tmp
-git clone -b release/4.4 https://git.ffmpeg.org/ffmpeg.git ffmpeg
-cd ffmpeg
-PKG_CONFIG_PATH="/usr/local/lib/pkgconfig"
-./configure  --prefix="/usr/local" --extra-cflags="-I/usr/local/include"   --extra-ldflags="-L/usr/local/lib"  --extra-libs=-lpthread  --extra-libs=-lm  --enable-shared   --disable-static   --enable-libx264  --enable-libx265  --enable-libfdk-aac  --enable-gpl --enable-nonfree
-make -j$(nproc) install
-rm -rf /tmp/ffmpeg
-
-cp /tmp/local-pin-600 /etc/apt/preferences.d
-
-command -v lbzip2 
-ln -sf $(command -v lbzip2) /usr/local/bin/compressor || ln -sf $(command -v bzip2) /usr/local/bin/compressor
-
-# Install Google Benchmark
-mkdir -p /tmp/Gbenchmark 
-cd /tmp/Gbenchmark 
-wget -qO- https://github.com/google/benchmark/archive/refs/tags/v1.6.1.tar.gz | tar xz 
-cmake -Sbenchmark-1.6.1 -Bbuild -DCMAKE_BUILD_TYPE=Release -DBUILD_SHARED_LIBS=OFF -DBENCHMARK_ENABLE_TESTING=OFF -DCMAKE_CXX_STANDARD=14 
-make -j -C build 
-cd /tmp/Gbenchmark/build
-make install
-
-# Build boost-1.85.0 from source for RPP
-# Installing in a non-standard location since the test packages of hipFFT and rocFFT pick up the version of
-# the installed Boost library and declare a package dependency on that specific version of Boost.
-# For example, if this was installed in the standard location it would declare a dependency on libboost-dev(el)1.85.0
-# which is not available as a package in any distro.
-# Once this is fixed, we can remove the Boost package from the requirements list and install this
-# in the standard location
-mkdir -p /tmp/boost-1.85.0 
-cd /tmp/boost-1.85.0 
-wget -nv https://sourceforge.net/projects/boost/files/boost/1.85.0/boost_1_85_0.tar.bz2 -O ./boost_1_85_0.tar.bz2 
-tar -xf boost_1_85_0.tar.bz2 --use-compress-program="/usr/local/bin/compressor" 
-cd boost_1_85_0 
-./bootstrap.sh --prefix=${RPP_DEPS_LOCATION} --with-python=python3 
-./b2 stage -j$(nproc) threading=multi link=shared cxxflags="-std=c++11" 
-./b2 install threading=multi link=shared --with-system --with-filesystem 
-./b2 stage -j$(nproc) threading=multi link=static cxxflags="-std=c++11 -fpic" cflags="-fpic"
-./b2 install threading=multi link=static --with-system --with-filesystem 
-rm -rf /tmp/boost-1.85.0

tools/rocm-build/docker/ubuntu22/packages

@@ -7,7 +7,6 @@ bison
 bridge-utils
 build-essential
 bzip2
-ccache
 check
 chrpath
 cifs-utils