Merge pull request #4513 from ammallya/main

Name change for hip
Change naming convention for hip
2026-01-10 15:18:11 -05:00 · 2025-03-18 12:49:47 -07:00 · 2025-03-18 12:49:09 -07:00 · 2025-03-18 12:48:31 -07:00 · 2025-03-18 11:27:49 -07:00 · 2025-03-18 11:26:53 -07:00
116 changed files with 4216 additions and 4827 deletions
--- a/.azuredevops/components/MIOpen.yml
+++ b/.azuredevops/components/MIOpen.yml
@@ -101,7 +101,7 @@ jobs:
        -DMIOPEN_BACKEND=HIP
        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/miopen-deps
-        -DGPU_TARGETS=$(JOB_GPU_TARGET)
+        -DAMDGPU_TARGETS=$(JOB_GPU_TARGET)
        -DMIOPEN_ENABLE_AI_KERNEL_TUNING=OFF
        -DMIOPEN_ENABLE_AI_IMMED_MODE_FALLBACK=OFF
        -DCMAKE_BUILD_TYPE=Release
@@ -129,8 +129,6 @@ jobs:
  variables:
  - group: common
  - template: /.azuredevops/variables-global.yml
-  - name: ROCM_PATH
-    value: $(Agent.BuildDirectory)/rocm
  pool: $(JOB_TEST_POOL)
  workspace:
    clean: all
--- a/.azuredevops/components/ROCR-Runtime.yml
+++ b/.azuredevops/components/ROCR-Runtime.yml
@@ -123,12 +123,16 @@ jobs:
      targetType: 'inline'
      workingDirectory: $(Build.SourcesDirectory)/rocrtst/suites/test_common
      script: |
+        sudo rm -rf $(Agent.BuildDirectory)/external/llvm-project
+        mkdir -p $(Agent.BuildDirectory)/external/llvm-project/clang/lib
+        sudo ln -s $(Agent.BuildDirectory)/rocm/llvm/lib/clang/20/include $(Agent.BuildDirectory)/external/llvm-project/clang/lib/Headers
        mkdir build && cd build
        cmake .. \
          -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm \
          -DTARGET_DEVICES=$(JOB_GPU_TARGET) \
          -DROCM_DIR=$(Agent.BuildDirectory)/rocm \
          -DLLVM_DIR=$(Agent.BuildDirectory)/rocm/llvm/bin \
+          -DOPENCL_DIR=$(Agent.BuildDirectory)/rocm/llvm/bin
          -DOPENCL_INC_DIR=$(Agent.BuildDirectory)/rocm/llvm/lib/clang/21/include
        make
        make rocrtst_kernels
--- a/.azuredevops/components/rccl.yml
+++ b/.azuredevops/components/rccl.yml
@@ -51,7 +51,6 @@ parameters:

 jobs:
 - job: rccl
-  timeoutInMinutes: 90
  variables:
  - group: common
  - template: /.azuredevops/variables-global.yml
@@ -79,6 +78,7 @@ jobs:
      checkoutRef: ${{ parameters.checkoutRef }}
      dependencyList: ${{ parameters.rocmDependencies }}
      gpuTarget: $(JOB_GPU_TARGET)
+  - script: chmod +x $(Agent.BuildDirectory)/rocm/bin/hipify-perl
  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
    parameters:
      extraBuildFlags: >-
@@ -88,7 +88,7 @@ jobs:
        -DCMAKE_BUILD_TYPE=Release
        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
        -DBUILD_TESTS=ON
-        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/rocm/share/rocm/cmake;$(Agent.BuildDirectory)/rocm/libexec/hipify
+        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(Agent.BuildDirectory)/rocm/share/rocm/cmake/
        -DAMDGPU_TARGETS=$(JOB_GPU_TARGET)
        -GNinja
  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
--- a/.azuredevops/components/rocPyDecode.yml
+++ b/.azuredevops/components/rocPyDecode.yml
@@ -84,7 +84,6 @@ jobs:
        echo "##vso[task.setvariable variable=PYBIND11_PATH;]$(python3 -c 'import pybind11; print(pybind11.get_cmake_dir())')"
  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
    parameters:
-      installEnabled: false
      extraBuildFlags: >-
        -DROCM_PATH=$(Agent.BuildDirectory)/rocm
        -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm;$(PYTHON_USER_SITE)/pybind11;$(PYTHON_DIST_PACKAGES)/pybind11;$(PYBIND11_PATH)
@@ -92,14 +91,6 @@ jobs:
        -DGPU_TARGETS=$(JOB_GPU_TARGET)
        -DCMAKE_INSTALL_PREFIX_PYTHON=$(Build.BinariesDirectory)
        -GNinja
-  - task: Bash@3
-    displayName: 'rocPyDecode install'
-    inputs:
-      targetType: inline
-      script: |
-        sudo cmake --build . --target install
-        sudo chown -R $(whoami):$(id -gn) $(Build.BinariesDirectory)
-      workingDirectory: $(Build.SourcesDirectory)/build
  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
    parameters:
      gpuTarget: $(JOB_GPU_TARGET)
@@ -114,8 +105,7 @@ jobs:
      script: |
        export ROCM_PATH=$(Agent.BuildDirectory)/rocm
        export HIP_INCLUDE_DIRS=$(Agent.BuildDirectory)/rocm/include/hip
-        sudo python3 setup.py bdist_wheel
-        sudo chown -R $(whoami):$(id -gn) $(find . -name "*.whl")
+        python3 setup.py bdist_wheel
      workingDirectory: $(Build.SourcesDirectory)
  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-prepare-package.yml
    parameters:
--- a/.azuredevops/components/rocminfo.yml
+++ b/.azuredevops/components/rocminfo.yml
@@ -56,7 +56,6 @@ jobs:
  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
    parameters:
      aptPackages: ${{ parameters.aptPackages }}
-      registerROCmPackages: true

 - job: rocminfo_testing
  dependsOn: rocminfo
@@ -103,6 +102,5 @@ jobs:
  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
    parameters:
      aptPackages: ${{ parameters.aptPackages }}
-      registerROCmPackages: true
      environment: test
      gpuTarget: $(JOB_GPU_TARGET)
--- a/.azuredevops/components/rpp.yml
+++ b/.azuredevops/components/rpp.yml
@@ -29,7 +29,6 @@ parameters:
    - clr
    - half
    - llvm-project
-    - rocm-cmake
    - rocminfo
    - ROCR-Runtime
 - name: rocmTestDependencies
@@ -80,7 +79,6 @@ jobs:
        -DHALF_INCLUDE_DIRS=$(Agent.BuildDirectory)/rocm/include
        -DCMAKE_BUILD_TYPE=Release
        -DGPU_TARGETS=$(JOB_GPU_TARGET)
-        -DROCM_PLATFORM_VERSION=$(NEXT_RELEASE_VERSION)
        -GNinja
  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
    parameters:
--- a/.azuredevops/tag-builds/TransferBench.yml
+++ b/.azuredevops/tag-builds/TransferBench.yml
@@ -0,0 +1,29 @@
+variables:
+- group: common
+- template: /.azuredevops/variables-global.yml
+
+parameters:
+- name: checkoutRef
+  type: string
+  default: refs/tags/$(LATEST_RELEASE_TAG)
+
+resources:
+  repositories:
+  - repository: pipelines_repo
+    type: github
+    endpoint: ROCm
+    name: ROCm/ROCm
+  - repository: release_repo
+    type: github
+    endpoint: ROCm
+    name: ROCm/TransferBench
+    ref: ${{ parameters.checkoutRef }}
+
+trigger: none
+pr: none
+
+jobs:
+  - template: ${{ variables.CI_COMPONENT_PATH }}/TransferBench.yml
+    parameters:
+      checkoutRepo: release_repo
+      checkoutRef: ${{ parameters.checkoutRef }}
--- a/.azuredevops/templates/steps/miopen-get-ck-build.yml
+++ b/.azuredevops/templates/steps/miopen-get-ck-build.yml
@@ -20,37 +20,41 @@ steps:
      ARTIFACT_NAME="composablekernel.${{ parameters.gpuTarget }}"
      EXIT_CODE=0

-      # Try to find an Azure build for the specific CK commit called out in MIOpen's requirements.txt
+      # The commits that MIOpen reference are all merge commits from CK/develop to CK/amd-develop
+      # These commits are present on CK/amd-develop but not on CK/develop
+      # Ex-CI only builds CK/develop, so we need to find a commit present on both CK/develop and CK/amd-develop
+
      CK_COMMIT=$(grep 'ROCm/composable_kernel' requirements.txt | sed -E 's/.*@([a-f0-9]{40}).*/\1/')
      echo "Fetching CK build ID for commit $CK_COMMIT"
-      CK_CHECKS_URL="$GH_API/composable_kernel/commits/${CK_COMMIT}/check-runs"
-      CK_BUILD_ID=$(curl -s $CK_CHECKS_URL | \
+      CK_COMMIT_URL="$GH_API/composable_kernel/commits/${CK_COMMIT}"
+      PARENT_COMMIT=$(curl -s $CK_COMMIT_URL | jq '.parents[1].sha' | tr -d '"')
+      echo "Found parent commit: $PARENT_COMMIT"
+      PARENT_CHECKS_URL="$GH_API/composable_kernel/commits/${PARENT_COMMIT}/check-runs"
+      CK_BUILD_ID=$(curl -s $PARENT_CHECKS_URL | \
        jq '.check_runs[] | select(.name == "composable_kernel" and .app.slug == "azure-pipelines") | .details_url' | \
        tr -d '"' | grep -oP 'buildId=\K\d+')

-      # If none found, use latest successful CK build instead
-      if [[ -z "$CK_BUILD_ID" ]]; then
+      if [ -z "$CK_BUILD_ID" ]; then
        echo "Did not find specific CK build ID"
        LATEST_BUILD_URL="$AZ_API/build/builds?definitions=$(COMPOSABLE_KERNEL_PIPELINE_ID)&statusFilter=completed&resultFilter=succeeded&\$top=1&api-version=7.1"
        CK_BUILD_ID=$(curl -s $LATEST_BUILD_URL | jq '.value[0].id')
        echo "Found latest CK build ID: $CK_BUILD_ID"
        EXIT_CODE=1
-      else
-        echo "Found specific CK build ID: $CK_BUILD_ID"
      fi

      AZURE_URL="$AZ_API/build/builds/$CK_BUILD_ID/artifacts?artifactName=$ARTIFACT_NAME&api-version=7.1"
      ARTIFACT_URL=$(curl -s $AZURE_URL | jq '.resource.downloadUrl' | tr -d '"')

-      # If using the specific CK commit and it doesn't have any valid artifacts, use latest successful CK build instead
-      if { [[ -z "$ARTIFACT_URL" ]] || [[ "$ARTIFACT_URL" == "null" ]]; } && [[ $EXIT_CODE -eq 0 ]]; then
-        echo "Did not find valid specific CK build artifact"
-        LATEST_BUILD_URL="$AZ_API/build/builds?definitions=$(COMPOSABLE_KERNEL_PIPELINE_ID)&statusFilter=completed&resultFilter=succeeded&\$top=1&api-version=7.1"
+      if [ -z "$ARTIFACT_URL" ]; then
+        echo "Did not find specific CK build artifact"
+        LATEST_BUILD_URL="$AZ_API/build/builds?definitions=$(COMPOSABLE_KERNEL_PIPELINE_ID)&status=completed&result=succeeded&\$top=1&api-version=7.1"
        CK_BUILD_ID=$(curl -s $LATEST_BUILD_URL | jq '.value[0].id')
        echo "Found latest CK build ID: $CK_BUILD_ID"
        AZURE_URL="$AZ_API/build/builds/$CK_BUILD_ID/artifacts?artifactName=$ARTIFACT_NAME&api-version=7.1"
        ARTIFACT_URL=$(curl -s $AZURE_URL | jq '.resource.downloadUrl' | tr -d '"')
        EXIT_CODE=2
+      elif [ $EXIT_CODE -eq 0 ]; then
+        echo "Found specific CK build ID: $CK_BUILD_ID"
      fi

      echo "Downloading CK artifact from $ARTIFACT_URL"
@@ -60,13 +64,9 @@ steps:
      tar -zxvf $(System.ArtifactsDirectory)/$ARTIFACT_NAME/*.tar.gz -C $(Agent.BuildDirectory)/rocm
      rm -r $(System.ArtifactsDirectory)/ck.zip $(System.ArtifactsDirectory)/$ARTIFACT_NAME

-      if [[ $EXIT_CODE -ne 0 ]]; then
+      if [ $EXIT_CODE -ne 0 ]; then
        BUILD_COMMIT=$(curl -s $AZ_API/build/builds/$CK_BUILD_ID | jq '.sourceVersion' | tr -d '"')
-        if [[ $EXIT_CODE -eq 1 ]]; then
-          echo "WARNING: couldn't find a CK build for commit $CK_COMMIT"
-        elif [[ $EXIT_CODE -eq 2 ]]; then
-          echo "WARNING: couldn't find a valid CK artifact for commit $CK_COMMIT"
-        fi
+        echo "WARNING: couldn't find a CK build for commit $CK_COMMIT"
        echo "Instead used latest CK build $CK_BUILD_ID for commit $BUILD_COMMIT"
      fi
      exit $EXIT_CODE
--- a/.gitignore
+++ b/.gitignore
@@ -11,7 +11,6 @@ _toc.yml
 docBin/
 _doxygen/
 _readthedocs/
-__pycache__/

 # avoid duplicating contributing.md due to conf.py
 docs/CHANGELOG.md
--- a/.wordlist.txt
+++ b/.wordlist.txt
@@ -117,7 +117,6 @@ FX
 Filesystem
 FindDb
 Flang
-FluxBenchmark
 Fortran
 Fuyu
 GALB
@@ -132,7 +131,6 @@ GDS
 GEMM
 GEMMs
 GFortran
-GFXIP
 Gemma
 GiB
 GIM
@@ -156,7 +154,6 @@ HCA
 HGX
 HIPCC
 HIPExtension
-HIPification
 HIPIFY
 HIPification
 HIPify
@@ -319,7 +316,6 @@ PipelineParallel
 PnP
 PowerEdge
 PowerShell
-Pretraining
 Profiler's
 PyPi
 Pytest
@@ -341,7 +337,6 @@ RNNs
 ROC
 ROCProfiler
 ROCT
-ROCTx
 ROCTracer
 ROCclr
 ROCdbgapi
@@ -482,7 +477,6 @@ ZenDNN
 accuracies
 activations
 addr
-ai
 alloc
 allocatable
 allocator
@@ -548,7 +542,6 @@ cTDP
 dataset
 datasets
 dataspace
-datatemplate
 datatype
 datatypes
 dbgapi
@@ -577,7 +570,6 @@ el
 embeddings
 enablement
 encodings
-endfor
 endpgm
 enqueue
 env
@@ -698,7 +690,6 @@ pageable
 pallas
 parallelization
 parallelizing
-param
 parameterization
 passthrough
 perfcounter
@@ -723,7 +714,6 @@ preprocessing
 preprocessor
 prequantized
 prerequisites
-pretraining
 profiler
 profilers
 protobuf
@@ -776,7 +766,6 @@ rocm
 rocminfo
 rocprim
 rocprof
-rocprofv
 rocprofiler
 rocr
 rocrand
@@ -816,7 +805,6 @@ supercomputing
 symlink
 symlinks
 sys
-tabindex
 td
 tensorfloat
 th
@@ -862,7 +850,6 @@ vectorizes
 virtualize
 virtualized
 vjxb
-vllm
 voxel
 walkthrough
 walkthroughs
--- a/README.md
+++ b/README.md
@@ -50,7 +50,7 @@ The following example shows how to use the repo tool to download the ROCm source
 ```bash
 mkdir -p ~/ROCm/
 cd ~/ROCm/
-export ROCM_VERSION=6.3.3
+export ROCM_VERSION=6.3.2
 ~/bin/repo init -u http://github.com/ROCm/ROCm.git -b roc-6.3.x -m tools/rocm-build/rocm-${ROCM_VERSION}.xml
 ~/bin/repo sync
 ```
@@ -77,8 +77,8 @@ The Build time will reduce significantly if we limit the GPU Architecture/s agai

 mkdir -p ~/WORKSPACE/      # Or any folder name other than WORKSPACE
 cd ~/WORKSPACE/
-export ROCM_VERSION=6.3.3
-~/bin/repo init -u http://github.com/ROCm/ROCm.git -b roc-6.3.x -m tools/rocm-build/rocm-${ROCM_VERSION}.xml
+export ROCM_VERSION=6.3.2
+~/bin/repo init -u http://github.com/ROCm/ROCm.git -b develop -m tools/rocm-build/rocm-${ROCM_VERSION}.xml
 ~/bin/repo sync

 # --------------------------------------
--- a/RELEASE.md
+++ b/RELEASE.md
@@ -10,7 +10,7 @@
 <!-- markdownlint-disable reference-links-images            -->
 <!-- markdownlint-disable no-missing-space-atx              -->
 <!-- spellcheck-disable                                     -->
-# ROCm 6.3.3 release notes
+# ROCm 6.3.2 release notes

 The release notes provide a summary of notable changes since the previous ROCm release.

@@ -24,6 +24,8 @@ The release notes provide a summary of notable changes since the previous ROCm r

 - [ROCm known issues](#rocm-known-issues)

+- [ROCm resolved issues](#rocm-resolved-issues)
+
 - [ROCm upcoming changes](#rocm-upcoming-changes)

 ```{note}
@@ -32,43 +34,35 @@ documentation to verify compatibility and system requirements.
 ```
 ## Release highlights

-The following are notable new features and improvements in ROCm 6.3.3. For changes to individual components, see
+The following are notable improvements in ROCm 6.3.2. For changes to individual components, see
 [Detailed component changes](#detailed-component-changes).

-### ROCm Offline Installer Creator updates
-
-The ROCm Offline Installer Creator 6.3.3 adds a new Post-Install Options menu, which includes a new ``udev`` option for adding GPU resources access for all users. It also moves the user-specific GPU access option (for the ``video,render`` group) from the Driver Options menu to the Post-Install Options menu. See the [ROCm Offline Installer Creator](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/rocm-offline-installer.html#post-install-options-menu) documentation for more information.
-
 ### ROCm documentation updates

 ROCm documentation continues to be updated to provide clearer and more comprehensive guidance for a wider variety of user needs and use cases.

-* [Tutorials for AI developers](https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/) have been added. These tutorials are Jupyter notebook-based, easy-to-follow documents. They are ideal for AI developers who want to learn about specific topics, including inference, fine-tuning, and training.
+* Documentation about ROCm compatibility with deep learning frameworks has been added. These topics outline ROCm-enabled features for each deep learning framework, key ROCm libraries that can influence the capabilities, validated Docker image tags, and features supported across the available ROCm and framework versions. For more information, see:

-* The [LLM inference performance validation guide for AMD Instinct MI300X](https://rocm.docs.amd.com/en/latest/how-to/rocm-for-ai/inference/vllm-benchmark.html)
-  now includes additional models for performance benchmarking. The accompanying ROCm vLLM Docker has been upgraded to ROCm 6.3.1.
- 
-* The HIP documentation has been updated with new resources for developers. To learn more about concurrency, parallelism, and stream management on devices and multiple GPUs, see [Asynchronous concurrent execution](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_runtime_api/asynchronous.html)
+    * [PyTorch compatibility](https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/pytorch-compatibility.html)

-* The following HIP documentation topics have been updated:
-    - [Virtual memory management](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_runtime_api/memory_management/virtual_memory.html)
-    - [Programming for HIP runtime compiler (RTC)](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_rtc.html)
-    - [HIP porting guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
-    - [Porting CUDA driver API](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_driver_api.html)
-    - [CUDA to HIP API function comparison](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/api_syntax.html)
+    * [TensorFlow compatibility](https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/tensorflow-compatibility.html)
+
+    * [JAX compatibility](https://rocm.docs.amd.com/en/latest/compatibility/ml-compatibility/jax-compatibility.html)
+
+* The [HIP C++ language extensions](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_cpp_language_extensions.html) and [Kernel language C++ support](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/kernel_language_cpp_support.html) topics have been reorganized to make them easier to find and review. The topics have also been enhanced with new content.

 ## Operating system and hardware support changes

-Operating system and hardware support remain unchanged in this release.
+ROCm 6.3.2 adds support for Azure Linux 3.0 (kernel: 6.6). Azure Linux is supported only on AMD Instinct accelerators. For more information, see [Azure Linux installation](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/quick-start.html).

 See the [Compatibility
-matrix](https://rocm.docs.amd.com/en/docs-6.3.3/compatibility/compatibility-matrix.html)
+matrix](https://rocm.docs.amd.com/en/latest/compatibility/compatibility-matrix.html)
 for more information about operating system and hardware compatibility.

 ## ROCm components

-The following table lists the versions of ROCm components for ROCm 6.3.3, including any version
-changes from 6.3.2 to 6.3.3. Click the component's updated version to go to a list of its changes.
+The following table lists the versions of ROCm components for ROCm 6.3.2, including any version
+changes from 6.3.1 to 6.3.2. Click the component's updated version to go to a list of its changes.
 Click {fab}`github` to go to the component's source code on GitHub.

 <div class="pst-scrollable-table-container">
@@ -90,47 +84,47 @@ Click {fab}`github` to go to the component's source code on GitHub.
            <tr>
                <th rowspan="9">Libraries</th>
                <th rowspan="9">Machine learning and computer vision</th>
-                <td><a href="https://rocm.docs.amd.com/projects/composable_kernel/en/docs-6.3.3/index.html">Composable Kernel</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/composable_kernel/en/docs-6.3.2/index.html">Composable Kernel</a></td>
                <td>1.1.0</td>
                <td><a href="https://github.com/ROCm/composable_kernel"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/AMDMIGraphX/en/docs-6.3.3/index.html">MIGraphX</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/AMDMIGraphX/en/docs-6.3.2/index.html">MIGraphX</a></td>
                <td>2.11.0</td>
                <td><a href="https://github.com/ROCm/AMDMIGraphX"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/MIOpen/en/docs-6.3.3/index.html">MIOpen</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/MIOpen/en/docs-6.3.2/index.html">MIOpen</a></td>
                <td>3.3.0</td>
                <td><a href="https://github.com/ROCm/MIOpen"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/MIVisionX/en/docs-6.3.3/index.html">MIVisionX</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/MIVisionX/en/docs-6.3.2/index.html">MIVisionX</a></td>
                <td>3.1.0</td>
                <td><a href="https://github.com/ROCm/MIVisionX"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocAL/en/docs-6.3.3/index.html">rocAL</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocAL/en/docs-6.3.2/index.html">rocAL</a></td>
                <td>2.1.0</td>
                <td><a href="https://github.com/ROCm/rocAL"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocDecode/en/docs-6.3.3/index.html">rocDecode</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocDecode/en/docs-6.3.2/index.html">rocDecode</a></td>
                <td>0.8.0</td>
                <td><a href="https://github.com/ROCm/rocDecode"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocJPEG/en/docs-6.3.3/index.html">rocJPEG</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocJPEG/en/docs-6.3.2/index.html">rocJPEG</a></td>
                <td>0.6.0</td>
                <td><a href="https://github.com/ROCm/rocJPEG"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocPyDecode/en/docs-6.3.3/index.html">rocPyDecode</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocPyDecode/en/docs-6.3.2/index.html">rocPyDecode</a></td>
                <td>0.2.0</td>
                <td><a href="https://github.com/ROCm/rocPyDecode"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rpp/en/docs-6.3.3/index.html">RPP</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rpp/en/docs-6.3.2/index.html">RPP</a></td>
                <td>1.9.1</td>
                <td><a href="https://github.com/ROCm/rpp"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
@@ -139,7 +133,7 @@ Click {fab}`github` to go to the component's source code on GitHub.
            <tr>
                <th rowspan="1"></th>
                <th rowspan="1">Communication</th>
-                <td><a href="https://rocm.docs.amd.com/projects/rccl/en/docs-6.3.3/index.html">RCCL</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rccl/en/docs-6.3.2/index.html">RCCL</a></td>
                <td>2.21.5</td>
                <td><a href="https://github.com/ROCm/rccl"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
@@ -148,82 +142,82 @@ Click {fab}`github` to go to the component's source code on GitHub.
            <tr>
                <th rowspan="16"></th>
                <th rowspan="16">Math</th>
-                <td><a href="https://rocm.docs.amd.com/projects/hipBLAS/en/docs-6.3.3/index.html">hipBLAS</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipBLAS/en/docs-6.3.2/index.html">hipBLAS</a></td>
                <td>2.3.0</td>
                <td><a href="https://github.com/ROCm/hipBLAS"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/hipBLASLt/en/docs-6.3.3/index.html">hipBLASLt</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipBLASLt/en/docs-6.3.2/index.html">hipBLASLt</a></td>
                <td>0.10.0</td>
                <td><a href="https://github.com/ROCm/hipBLASLt"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/hipFFT/en/docs-6.3.3/index.html">hipFFT</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipFFT/en/docs-6.3.2/index.html">hipFFT</a></td>
                <td>1.0.17</td>
                <td><a href="https://github.com/ROCm/hipFFT"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/hipfort/en/docs-6.3.3/index.html">hipfort</a></td>
-                <td>0.5.1</td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipfort/en/docs-6.3.2/index.html">hipfort</a></td>
+                <td>0.5.0&nbsp;&Rightarrow;&nbsp;<a href="#hipfort-0-5-1">0.5.1</a></td>
                <td><a href="https://github.com/ROCm/hipfort"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/hipRAND/en/docs-6.3.3/index.html">hipRAND</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipRAND/en/docs-6.3.2/index.html">hipRAND</a></td>
                <td>2.11.1</td>
                <td><a href="https://github.com/ROCm/hipRAND"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/hipSOLVER/en/docs-6.3.3/index.html">hipSOLVER</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipSOLVER/en/docs-6.3.2/index.html">hipSOLVER</a></td>
                <td>2.3.0</td>
                <td><a href="https://github.com/ROCm/hipSOLVER"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/hipSPARSE/en/docs-6.3.3/index.html">hipSPARSE</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipSPARSE/en/docs-6.3.2/index.html">hipSPARSE</a></td>
                <td>3.1.2</td>
                <td><a href="https://github.com/ROCm/hipSPARSE"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/hipSPARSELt/en/docs-6.3.3/index.html">hipSPARSELt</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipSPARSELt/en/docs-6.3.2/index.html">hipSPARSELt</a></td>
                <td>0.2.2</td>
                <td><a href="https://github.com/ROCm/hipSPARSELt"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocALUTION/en/docs-6.3.3/index.html">rocALUTION</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocALUTION/en/docs-6.3.2/index.html">rocALUTION</a></td>
                <td>3.2.1</td>
                <td><a href="https://github.com/ROCm/rocALUTION"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocBLAS/en/docs-6.3.3/index.html">rocBLAS</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocBLAS/en/docs-6.3.2/index.html">rocBLAS</a></td>
                <td>4.3.0</td>
                <td><a href="https://github.com/ROCm/rocBLAS"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocFFT/en/docs-6.3.3/index.html">rocFFT</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocFFT/en/docs-6.3.2/index.html">rocFFT</a></td>
                <td>1.0.31</td>
                <td><a href="https://github.com/ROCm/rocFFT"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocRAND/en/docs-6.3.3/index.html">rocRAND</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocRAND/en/docs-6.3.2/index.html">rocRAND</a></td>
                <td>3.2.0</td>
                <td><a href="https://github.com/ROCm/rocRAND"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocSOLVER/en/docs-6.3.3/index.html">rocSOLVER</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocSOLVER/en/docs-6.3.2/index.html">rocSOLVER</a></td>
                <td>3.27.0</td>
                <td><a href="https://github.com/ROCm/rocSOLVER"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocSPARSE/en/docs-6.3.3/index.html">rocSPARSE</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocSPARSE/en/docs-6.3.2/index.html">rocSPARSE</a></td>
                <td>3.3.0</td>
                <td><a href="https://github.com/ROCm/rocSPARSE"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocWMMA/en/docs-6.3.3/index.html">rocWMMA</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocWMMA/en/docs-6.3.2/index.html">rocWMMA</a></td>
                <td>1.6.0</td>
                <td><a href="https://github.com/ROCm/rocWMMA"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/Tensile/en/docs-6.3.3/src/index.html">Tensile</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/Tensile/en/docs-6.3.2/src/index.html">Tensile</a></td>
                <td>4.42.0</td>
                <td><a href="https://github.com/ROCm/Tensile"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
@@ -232,22 +226,22 @@ Click {fab}`github` to go to the component's source code on GitHub.
            <tr>
                <th rowspan="4"></th>
                <th rowspan="4">Primitives</th>
-                <td><a href="https://rocm.docs.amd.com/projects/hipCUB/en/docs-6.3.3/index.html">hipCUB</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipCUB/en/docs-6.3.2/index.html">hipCUB</a></td>
                <td>3.3.0</td>
                <td><a href="https://github.com/ROCm/hipCUB"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/hipTensor/en/docs-6.3.3/index.html">hipTensor</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/hipTensor/en/docs-6.3.2/index.html">hipTensor</a></td>
                <td>1.4.0</td>
                <td><a href="https://github.com/ROCm/hipTensor"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocPRIM/en/docs-6.3.3/index.html">rocPRIM</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocPRIM/en/docs-6.3.2/index.html">rocPRIM</a></td>
                <td>3.3.0</td>
                <td><a href="https://github.com/ROCm/rocPRIM"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocThrust/en/docs-6.3.3/index.html">rocThrust</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocThrust/en/docs-6.3.2/index.html">rocThrust</a></td>
                <td>3.3.0</td>
                <td><a href="https://github.com/ROCm/rocThrust"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
@@ -256,27 +250,27 @@ Click {fab}`github` to go to the component's source code on GitHub.
            <tr>
                <th rowspan="7">Tools</th>
                <th rowspan="7">System management</th>
-                <td><a href="https://rocm.docs.amd.com/projects/amdsmi/en/docs-6.3.3/index.html">AMD SMI</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/amdsmi/en/docs-6.3.2/index.html">AMD SMI</a></td>
                <td>24.7.1</td>
                <td><a href="https://github.com/ROCm/amdsmi"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rdc/en/docs-6.3.3/index.html">ROCm Data Center Tool</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rdc/en/docs-6.3.2/index.html">ROCm Data Center Tool</a></td>
                <td>0.3.0</td>
                <td><a href="https://github.com/ROCm/rdc"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocminfo/en/docs-6.3.3/index.html">rocminfo</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocminfo/en/docs-6.3.2/index.html">rocminfo</a></td>
                <td>1.0.0</td>
                <td><a href="https://github.com/ROCm/rocminfo"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocm_smi_lib/en/docs-6.3.3/index.html">ROCm SMI</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocm_smi_lib/en/docs-6.3.2/index.html">ROCm SMI</a></td>
                <td>7.4.0</td>
                <td><a href="https://github.com/ROCm/rocm_smi_lib"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/ROCmValidationSuite/en/docs-6.3.3/index.html">ROCmValidationSuite</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/ROCmValidationSuite/en/docs-6.3.2/index.html">ROCmValidationSuite</a></td>
                <td>1.1.0</td>
                <td><a href="https://github.com/ROCm/ROCmValidationSuite"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
@@ -285,38 +279,38 @@ Click {fab}`github` to go to the component's source code on GitHub.
            <tr>
                <th rowspan="6"></th>
                <th rowspan="6">Performance</th>
-                <td><a href="https://rocm.docs.amd.com/projects/rocm_bandwidth_test/en/docs-6.3.3/index.html">ROCm Bandwidth
+                <td><a href="https://rocm.docs.amd.com/projects/rocm_bandwidth_test/en/docs-6.3.2/index.html">ROCm Bandwidth
                        Test</a></td>
                <td>1.4.0</td>
                <td><a href="https://github.com/ROCm/rocm_bandwidth_test/"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocprofiler-compute/en/docs-6.3.3/index.html">ROCm Compute Profiler</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocprofiler-compute/en/docs-6.3.2/index.html">ROCm Compute Profiler</a></td>
                <td>3.0.0</td>
                <td><a href="https://github.com/ROCm/rocprofiler-compute"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocprofiler-systems/en/docs-6.3.3/index.html">ROCm Systems Profiler</a></td>
-                <td>0.1.1&nbsp;&Rightarrow;&nbsp;<a href="#rocm-systems-profiler-0-1-2">0.1.2</td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocprofiler-systems/en/docs-6.3.2/index.html">ROCm Systems Profiler</a></td>
+                <td>0.1.0&nbsp;&Rightarrow;&nbsp;<a href="#rocm-systems-profiler-0-1-1">0.1.1</td>
                <td><a href="https://github.com/ROCm/rocprofiler-systems"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocprofiler/en/docs-6.3.3/index.html">ROCProfiler</a></td>
-                <td>2.0.0</td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocprofiler/en/docs-6.3.2/index.html">ROCProfiler</a></td>
+                <td>2.0.0&nbsp;&Rightarrow;&nbsp;<a href="#rocprofiler-2-0-0">2.0.0</a></td>
                <td><a href="https://github.com/ROCm/ROCProfiler/"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocprofiler-sdk/en/docs-6.3.3/index.html">ROCprofiler-SDK</a></td>
-                <td>0.5.0</td>
+                <td><a href="https://rocm.docs.amd.com/projects/rocprofiler-sdk/en/docs-6.3.2/index.html">ROCprofiler-SDK</a></td>
+                <td>0.5.0&nbsp;&Rightarrow;&nbsp;<a href="#rocprofiler-sdk-0-5-0">0.5.0</a></td>
                <td><a href="https://github.com/ROCm/rocprofiler-sdk/"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr >
-                <td><a href="https://rocm.docs.amd.com/projects/roctracer/en/docs-6.3.3/index.html">ROCTracer</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/roctracer/en/docs-6.3.2/index.html">ROCTracer</a></td>
                <td>4.1.0</td>
                <td><a href="https://github.com/ROCm/ROCTracer/"><i
                            class="fab fa-github fa-lg"></i></a></td>
@@ -326,32 +320,32 @@ Click {fab}`github` to go to the component's source code on GitHub.
            <tr>
                <th rowspan="5"></th>
                <th rowspan="5">Development</th>
-                <td><a href="https://rocm.docs.amd.com/projects/HIPIFY/en/docs-6.3.3/index.html">HIPIFY</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/HIPIFY/en/docs-6.3.2/index.html">HIPIFY</a></td>
                <td>18.0.0</td>
                <td><a href="https://github.com/ROCm/HIPIFY/"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/ROCdbgapi/en/docs-6.3.3/index.html">ROCdbgapi</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/ROCdbgapi/en/docs-6.3.2/index.html">ROCdbgapi</a></td>
                <td>0.77.0</td>
                <td><a href="https://github.com/ROCm/ROCdbgapi/"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/ROCmCMakeBuildTools/en/docs-6.3.3/index.html">ROCm CMake</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/ROCmCMakeBuildTools/en/docs-6.3.2/index.html">ROCm CMake</a></td>
                <td>0.14.0</td>
                <td><a href="https://github.com/ROCm/rocm-cmake/"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/ROCgdb/en/docs-6.3.3/index.html">ROCm Debugger (ROCgdb)</a>
+                <td><a href="https://rocm.docs.amd.com/projects/ROCgdb/en/docs-6.3.2/index.html">ROCm Debugger (ROCgdb)</a>
                </td>
                <td>15.2</td>
                <td><a href="https://github.com/ROCm/ROCgdb/"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/rocr_debug_agent/en/docs-6.3.3/index.html">ROCr Debug Agent</a>
+                <td><a href="https://rocm.docs.amd.com/projects/rocr_debug_agent/en/docs-6.3.2/index.html">ROCr Debug Agent</a>
                </td>
                <td>2.0.3</td>
                <td><a href="https://github.com/ROCm/rocr_debug_agent/"><i
@@ -361,13 +355,13 @@ Click {fab}`github` to go to the component's source code on GitHub.
        <tbody class="rocm-components-compilers">
            <tr>
                <th rowspan="2" colspan="2">Compilers</th>
-                <td><a href="https://rocm.docs.amd.com/projects/HIPCC/en/docs-6.3.3/index.html">HIPCC</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/HIPCC/en/docs-6.3.2/index.html">HIPCC</a></td>
                <td>1.1.1</td>
                <td><a href="https://github.com/ROCm/llvm-project/"><i
                            class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/llvm-project/en/docs-6.3.3/index.html">llvm-project</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/llvm-project/en/docs-6.3.2/index.html">llvm-project</a></td>
                <td>18.0.0</td>
                <td><a href="https://github.com/ROCm/llvm-project/"><i
                            class="fab fa-github fa-lg"></i></a></td>
@@ -376,12 +370,12 @@ Click {fab}`github` to go to the component's source code on GitHub.
        <tbody class="rocm-components-runtimes">
            <tr>
                <th rowspan="2" colspan="2">Runtimes</th>
-                <td><a href="https://rocm.docs.amd.com/projects/HIP/en/docs-6.3.3/index.html">HIP</a></td>
-                <td>6.3.2</td>
+                <td><a href="https://rocm.docs.amd.com/projects/HIP/en/docs-6.3.2/index.html">HIP</a></td>
+                <td>6.3.1&nbsp;&Rightarrow;&nbsp;<a href="#hip-6-3-2">6.3.2</a></td>
                <td><a href="https://github.com/ROCm/HIP/"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
            <tr>
-                <td><a href="https://rocm.docs.amd.com/projects/ROCR-Runtime/en/docs-6.3.3/index.html">ROCr Runtime</a></td>
+                <td><a href="https://rocm.docs.amd.com/projects/ROCR-Runtime/en/docs-6.3.2/index.html">ROCr Runtime</a></td>
                <td>1.14.0</td>
                <td><a href="https://github.com/ROCm/ROCR-Runtime/"><i class="fab fa-github fa-lg"></i></a></td>
            </tr>
@@ -393,34 +387,112 @@ Click {fab}`github` to go to the component's source code on GitHub.

 The following sections describe key changes to ROCm components.

-### **ROCm Systems Profiler** (0.1.2)
+### **HIP** (6.3.2)
+
+#### Added
+
+* Tracking of Heterogeneous System Architecture (HSA) handlers:
+    - Adds an atomic counter to track the outstanding HSA handlers.
+    - Waits on CPU for the callbacks if the number exceeds the defined value.
+* Codes to capture Architected Queueing Language (AQL) packets for HIP graph memory copy node between host and device. HIP enqueues AQL packets during graph launch.
+* Control to use system pool implementation in runtime commands handling. By default, it is disabled.
+* A new path to avoid `WaitAny` calls in `AsyncEventsLoop`. The new path is selected by default.
+* Runtime control on decrement counter only if the event is popped. There is a new way to restore dead signals cleanup for the old path.
+* A new logic in runtime to track the age of events from the kernel mode driver.
+
+#### Optimized
+
+* HSA callback performance. The HIP runtime creates and submits commands in the queue and interacts with HSA through a callback function. HIP waits for the CPU status from HSA to optimize the handling of events, profiling, commands, and HSA signals for higher performance.
+* Runtime optimization which combines all logic of `WaitAny` in a single processing loop and avoids extra memory allocations or reference counting. The runtime won't spin on the CPU if all events are busy.
+* Multi-threaded dispatches for performance improvement.
+* Command submissions and processing between CPU and GPU by introducing a way to limit the software batch size.
+* Switch to `std::shared_mutex` in book/keep logic in streams from multiple threads simultaneously, for performance improvement in specific customer applications.
+* `std::shared_mutex` is used in memory object mapping, for performance improvement.

 #### Resolved issues

-* Fixed an error that prevented GPU hardware activity from being presented in certain workloads.  
+* Race condition in multi-threaded producer/consumer scenario with `hipMallocFromPoolAsync`.
+* Segmentation fault with `hipStreamLegacy` while using the API `hipStreamWaitEvent`.
+* Usage of `hipStreamLegacy` in HIP event record.
+* A soft hang in graph execution process from HIP user object. The fix handles the release of graph execution object properly considering synchronization on the device/stream. The user application now behaves the same with `hipUserObject` on both the AMD ROCm and NVIDIA CUDA platforms.
+
+### **hipfort** (0.5.1)
+
+#### Added
+
+* Support for building with LLVM Flang.
+
+#### Resolved issues
+
+* Fixed the exported `hipfort::hipsparse` CMake target.
+
+### **ROCm Systems Profiler** (0.1.1)
+
+#### Resolved issues
+
+* Fixed an error when building from source on some SUSE and RHEL systems when using the `ROCPROFSYS_BUILD_DYNINST` option.
+
+### **ROCProfiler** (2.0.0)
+
+#### Changed
+
+* Replaced `CU_UTILIZATION` metric with `SIMD_UTILIZATION` for better accuracy.
+
+#### Resolved issues
+
+* Fixed the `VALUBusy` and `SALUBusy` activity metrics for accuracy on MI300.
+
+### **ROCprofiler-SDK** (0.5.0)
+
+#### Added
+
+* Support for system-wide collection of SQ counters across all HSA processes.
+
+#### Changed
+
+* `rocprofiler_sample_device_counting_service` API updated to return counter output immediately, when called in synchronous mode.

 ## ROCm known issues

 ROCm known issues are noted on {fab}`github` [GitHub](https://github.com/ROCm/ROCm/labels/Verified%20Issue). For known
 issues related to individual components, review the [Detailed component changes](#detailed-component-changes).

-### Zero value is displayed in ROCTx aggregated statistics
+## ROCm resolved issues

-The ROCTx markers are standalone markers within the ROCProfiler-SDK library. Each marker reports only a single timestamp, which is recorded as the `start_timestamp` and `end_timestamp`. As a result, the value for aggregated statistics presented in `TotalDurationNs`, `maxNs`, and `minNs`, is zero. The zero value indicates that the actual execution time is not associated with the markers, which is an expected behavior. See [GitHub issue #4396](https://github.com/ROCm/ROCm/issues/4396).
+The following are previously known issues resolved in this release. For resolved issues related to
+individual components, review the [Detailed component changes](#detailed-component-changes).
+
+### TransferBench packages not functional
+
+Issue with TransferBench packages not being compiled properly has been fixed. For more information, See [GitHub issue #4081](https://github.com/ROCm/ROCm/issues/4081).
+
+### ROCm Compute Profiler CTest failure in CI
+
+When running the ROCm Compute Profiler (`rocprof-compute`) CTest in the Azure CI environment, the
+`rocprof-compute` execution test failed. This issue was due to an outdated test file that was not renamed
+(`omniperf` to `rocprof-compute`), and the `ROCM_PATH` environment variable not being set in
+the Azure CI environment, resulting in the tool being unable to extract chip information as expected.
+This issue has been fixed in the ROCm 6.3.2 release. See [GitHub issue #4085](https://github.com/ROCm/ROCm/issues/4085).
+
+### MIVisionX memory access fault in Canny edge detection
+
+An issue where Canny edge detection kernels accessed out-of-bounds memory locations while
+computing gradient intensities on edge pixels has been fixed. This issue was isolated to
+Canny-specific use cases on Instinct MI300 series accelerators. See [GitHub issue #4086](https://github.com/ROCm/ROCm/issues/4086).
+
+### AMD VCN instability with rocDecode
+
+A firmware crash on gfx942 devices when AMD Video Core Next (VCN) was used for rocDecode operations has been resolved.

 ## ROCm upcoming changes

 The following changes to the ROCm software stack are anticipated for future releases.

-### ROCTracer and ROCProfiler (rocprof and rocprofv2) deprecation
-
-Development and support for ROCTracer and ROCProfiler (`rocprof` and `rocprofv2`) will phase out in favor of ROCprofiler-SDK (`rocprofv3`) in upcoming ROCm releases. Going forward, only critical defect fixes will be addressed for older versions of profiling tools and libraries. Upgrade to the latest version of ROCprofiler-SDK (`rocprofv3`) library to ensure continued support and access to new features.
-
 ### AMDGPU wavefront size compiler macro deprecation

 The `__AMDGCN_WAVEFRONT_SIZE__` macro will be deprecated in an upcoming
 release. It is recommended to remove any use of this macro. For more information, see [AMDGPU
-support](https://rocm.docs.amd.com/projects/llvm-project/en/docs-6.3.3/LLVM/clang/html/AMDGPUSupport.html).
+support](https://rocm.docs.amd.com/projects/llvm-project/en/docs-6.3.2/LLVM/clang/html/AMDGPUSupport.html).

 ### HIPCC Perl scripts deprecation

--- a/default.xml
+++ b/default.xml
@@ -1,76 +1,78 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <manifest>
    <remote name="rocm-org" fetch="https://github.com/ROCm/" />
-    <default revision="refs/tags/rocm-6.3.3"
+    <default revision="refs/heads/amd-mainline"
     remote="rocm-org"
     sync-c="true"
     sync-j="4" />
 <!--list of projects for ROCm-->
-    <project name="ROCK-Kernel-Driver" />
-    <project name="ROCR-Runtime" />
-    <project name="amdsmi" />
-    <project name="rdc" />
-    <project name="rocm_bandwidth_test" />
-    <project name="rocm_smi_lib" />
-    <project name="rocm-core" />
-    <project name="rocm-examples" />
-    <project name="rocminfo" />
-    <project name="rocprofiler" />
-    <project name="rocprofiler-register" />
-    <project name="rocprofiler-sdk" />
-    <project name="rocprofiler-compute" />
-    <project name="rocprofiler-systems" />
-    <project name="roctracer" />
+    <project name="ROCm" revision="refs/heads/main" />
+    <project name="ROCK-Kernel-Driver" revision="refs/heads/master" /> 
+    <project name="ROCR-Runtime" revision="refs/heads/amd-master"/> 
+    <project name="amdsmi" /> 
+    <project name="rdc" /> 
+    <project name="rocm_bandwidth_test" revision="refs/heads/master" /> 
+    <project name="rocm_smi_lib" /> 
+    <project name="rocm-core" revision="refs/heads/amd-master" /> 
+    <project name="rocm-examples" revision="refs/heads/develop"/> 
+    <project name="rocminfo" revision="refs/heads/amd-master" /> 
+    <project name="rocprofiler" /> 
+    <project name="rocprofiler-register" /> 
+    <project name="rocprofiler-sdk" /> 
+    <project name="rocprofiler-compute" /> 
+    <project name="rocprofiler-systems" /> 
+    <project name="roctracer" /> 
 <!--HIP Projects-->
-    <project name="HIP" />
-    <project name="hip-tests" />
-    <project name="HIPIFY" />
-    <project name="clr" />
-    <project name="hipother" />
+    <project name="hip" />
+    <project name="hip-tests" /> 
+    <project name="HIPIFY" /> 
+    <project name="clr" /> 
+    <project name="hipother" /> 
 <!-- The following projects are all associated with the AMDGPU LLVM compiler -->
-    <project name="half" />
+    <project name="half" revision="refs/heads/rocm" />
    <project name="llvm-project" />
+    <project name="SPIRV-LLVM-Translator" path="llvm-project/llvm/projects/SPIRV-LLVM-Translator" />
 <!-- gdb projects -->
-    <project name="ROCdbgapi" />
-    <project name="ROCgdb" />
+    <project name="ROCdbgapi" /> 
+    <project name="ROCgdb" revision="refs/heads/amd-mainline-rocgdb-15"/> 
    <project name="rocr_debug_agent" />
 <!-- ROCm Libraries -->
-    <project groups="mathlibs" name="AMDMIGraphX" />
-    <project groups="mathlibs" name="MIOpen" />
-    <project groups="mathlibs" name="MIVisionX" />
-    <project groups="mathlibs" name="ROCmValidationSuite" />
-    <project groups="mathlibs" name="Tensile" />
-    <project groups="mathlibs" name="composable_kernel" />
-    <project groups="mathlibs" name="hipBLAS-common" />
-    <project groups="mathlibs" name="hipBLAS" />
-    <project groups="mathlibs" name="hipBLASLt" />
-    <project groups="mathlibs" name="hipCUB" />
-    <project groups="mathlibs" name="hipFFT" />
-    <project groups="mathlibs" name="hipRAND" />
-    <project groups="mathlibs" name="hipSOLVER" />
-    <project groups="mathlibs" name="hipSPARSE" />
-    <project groups="mathlibs" name="hipSPARSELt" />
-    <project groups="mathlibs" name="hipTensor" />
-    <project groups="mathlibs" name="hipfort" />
-    <project groups="mathlibs" name="rccl" />
-    <project groups="mathlibs" name="rocAL" />
-    <project groups="mathlibs" name="rocALUTION" />
-    <project groups="mathlibs" name="rocBLAS" />
-    <project groups="mathlibs" name="rocDecode" />
-    <project groups="mathlibs" name="rocJPEG" />
-    <project groups="mathlibs" name="rocPyDecode" />
-    <project groups="mathlibs" name="rocFFT" />
-    <project groups="mathlibs" name="rocPRIM" />
-    <project groups="mathlibs" name="rocRAND" />
-    <project groups="mathlibs" name="rocSOLVER" />
-    <project groups="mathlibs" name="rocSPARSE" />
-    <project groups="mathlibs" name="rocThrust" />
-    <project groups="mathlibs" name="rocWMMA" />
-    <project groups="mathlibs" name="rocm-cmake" />
-    <project groups="mathlibs" name="rpp" />
-    <project groups="mathlibs" name="TransferBench" />
+    <project groups="mathlibs" name="AMDMIGraphX" revision="refs/heads/master" /> 
+    <project groups="mathlibs" name="MIOpen" revision="refs/heads/amd-master" /> 
+    <project groups="mathlibs" name="MIVisionX" revision="refs/heads/master" /> 
+    <project groups="mathlibs" name="ROCmValidationSuite" revision="refs/heads/master" /> 
+    <project groups="mathlibs" name="Tensile" revision="refs/heads/master" /> 
+    <project groups="mathlibs" name="composable_kernel" revision="refs/heads/amd-master"/> 
+    <project groups="mathlibs" name="hipBLAS-common" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="hipBLAS" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="hipBLASLt" revision="refs/heads/mainline" />  
+    <project groups="mathlibs" name="hipCUB" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="hipFFT" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="hipRAND" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="hipSOLVER" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="hipSPARSE" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="hipSPARSELt" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="hipTensor" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="hipfort" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rccl" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocAL" revision="refs/heads/master" /> 
+    <project groups="mathlibs" name="rocALUTION" revision="refs/heads/mainline"/> 
+    <project groups="mathlibs" name="rocBLAS" revision="refs/heads/mainline"/> 
+    <project groups="mathlibs" name="rocDecode" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocJPEG" revision="refs/heads/mainline"/> 
+    <project groups="mathlibs" name="rocPyDecode" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocFFT" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocPRIM" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocRAND" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocSOLVER" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocSPARSE" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocThrust" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocWMMA" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rocm-cmake" revision="refs/heads/mainline" /> 
+    <project groups="mathlibs" name="rpp" revision="refs/heads/master" /> 
+    <project groups="mathlibs" name="TransferBench" /> 
 <!-- Projects for OpenMP-Extras -->
-    <project name="aomp" path="openmp-extras/aomp" />
-    <project name="aomp-extras" path="openmp-extras/aomp-extras" />
-    <project name="flang" path="openmp-extras/flang" />
+    <project name="aomp" path="openmp-extras/aomp" /> 
+    <project name="aomp-extras" path="openmp-extras/aomp-extras" /> 
+    <project name="flang" path="openmp-extras/flang" /> 
 </manifest>
--- a/docs/compatibility/compatibility-matrix-historical-6.0.csv
+++ b/docs/compatibility/compatibility-matrix-historical-6.0.csv
@@ -1,120 +1,120 @@
-ROCm Version,6.3.3,6.3.2,6.3.1,6.3.0,6.2.4,6.2.2,6.2.1,6.2.0, 6.1.2, 6.1.1, 6.1.0, 6.0.2, 6.0.0
-      :ref:`Operating systems & kernels <OS-kernel-versions>`,Ubuntu 24.04.2,Ubuntu 24.04.2,Ubuntu 24.04.2,Ubuntu 24.04.2,"Ubuntu 24.04.1, 24.04","Ubuntu 24.04.1, 24.04","Ubuntu 24.04.1, 24.04",Ubuntu 24.04,,,,,
-      ,Ubuntu 22.04.5,Ubuntu 22.04.5,Ubuntu 22.04.5,Ubuntu 22.04.5,"Ubuntu 22.04.5, 22.04.4","Ubuntu 22.04.5, 22.04.4","Ubuntu 22.04.5, 22.04.4","Ubuntu 22.04.5, 22.04.4","Ubuntu 22.04.4, 22.04.3","Ubuntu 22.04.4, 22.04.3","Ubuntu 22.04.4, 22.04.3","Ubuntu 22.04.4, 22.04.3, 22.04.2","Ubuntu 22.04.4, 22.04.3, 22.04.2"
-      ,,,,,,,,,"Ubuntu 20.04.6, 20.04.5","Ubuntu 20.04.6, 20.04.5","Ubuntu 20.04.6, 20.04.5","Ubuntu 20.04.6, 20.04.5","Ubuntu 20.04.6, 20.04.5"
-      ,"RHEL 9.5, 9.4","RHEL 9.5, 9.4","RHEL 9.5, 9.4","RHEL 9.5, 9.4","RHEL 9.4, 9.3","RHEL 9.4, 9.3","RHEL 9.4, 9.3","RHEL 9.4, 9.3","RHEL 9.4, 9.3, 9.2","RHEL 9.4, 9.3, 9.2","RHEL 9.4, 9.3, 9.2","RHEL 9.3, 9.2","RHEL 9.3, 9.2"
-      ,RHEL 8.10,RHEL 8.10,RHEL 8.10,RHEL 8.10,"RHEL 8.10, 8.9","RHEL 8.10, 8.9","RHEL 8.10, 8.9","RHEL 8.10, 8.9","RHEL 8.9, 8.8","RHEL 8.9, 8.8","RHEL 8.9, 8.8","RHEL 8.9, 8.8","RHEL 8.9, 8.8"
-      ,"SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP5, SP4","SLES 15 SP5, SP4","SLES 15 SP5, SP4","SLES 15 SP5, SP4","SLES 15 SP5, SP4"
-      ,,,,,,,,,CentOS 7.9,CentOS 7.9,CentOS 7.9,CentOS 7.9,CentOS 7.9
-      ,Oracle Linux 8.10 [#mi300x-past-60]_,Oracle Linux 8.10 [#mi300x-past-60]_,Oracle Linux 8.10 [#mi300x-past-60]_,Oracle Linux 8.10 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,,,
-,Debian 12 [#single-node-past-60]_,Debian 12 [#single-node-past-60]_,Debian 12 [#single-node-past-60]_,,,,,,,,,,
-,Azure Linux 3.0 [#mi300x-past-60]_,Azure Linux 3.0 [#mi300x-past-60]_,,,,,,,,,,,
-      ,.. _architecture-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      :doc:`Architecture <rocm-install-on-linux:reference/system-requirements>`,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3
-      ,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2
-      ,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA
-      ,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3
-      ,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2
-      ,.. _gpu-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      :doc:`GPU / LLVM target <rocm-install-on-linux:reference/system-requirements>`,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100
-      ,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030
-      ,gfx942,gfx942,gfx942,gfx942,gfx942 [#mi300_624-past-60]_,gfx942 [#mi300_622-past-60]_,gfx942 [#mi300_621-past-60]_,gfx942 [#mi300_620-past-60]_, gfx942 [#mi300_612-past-60]_, gfx942 [#mi300_611-past-60]_, gfx942 [#mi300_610-past-60]_, gfx942 [#mi300_602-past-60]_, gfx942 [#mi300_600-past-60]_
-      ,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a
-      ,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908
-,,,,,,,,,,,,,
-      FRAMEWORK SUPPORT,.. _framework-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      :doc:`PyTorch <../compatibility/ml-compatibility/pytorch-compatibility>`,"2.4, 2.3, 2.2, 1.13","2.4, 2.3, 2.2, 1.13","2.4, 2.3, 2.2, 1.13","2.4, 2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13"
-      :doc:`TensorFlow <../compatibility/ml-compatibility/tensorflow-compatibility>`,"2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.14.0, 2.13.1, 2.12.1","2.14.0, 2.13.1, 2.12.1"
-      :doc:`JAX <../compatibility/ml-compatibility/jax-compatibility>`,0.4.31,0.4.31,0.4.31,0.4.31,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26
-      `ONNX Runtime <https://onnxruntime.ai/docs/build/eps.html#amd-migraphx>`_,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.14.1,1.14.1
-      ,,,,,,,,,,,,,
-      THIRD PARTY COMMS,.. _thirdpartycomms-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      `UCC <https://github.com/ROCm/ucc>`_,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.2.0,>=1.2.0
-      `UCX <https://github.com/ROCm/ucx>`_,>=1.15.0,>=1.15.0,>=1.15.0,>=1.15.0,>=1.15.0,>=1.15.0,>=1.15.0,>=1.15.0,>=1.14.1,>=1.14.1,>=1.14.1,>=1.14.1,>=1.14.1
-      ,,,,,,,,,,,,,
-      THIRD PARTY ALGORITHM,.. _thirdpartyalgorithm-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      Thrust,2.3.2,2.3.2,2.3.2,2.3.2,2.2.0,2.2.0,2.2.0,2.2.0,2.1.0,2.1.0,2.1.0,2.0.1,2.0.1
-      CUB,2.3.2,2.3.2,2.3.2,2.3.2,2.2.0,2.2.0,2.2.0,2.2.0,2.1.0,2.1.0,2.1.0,2.0.1,2.0.1
-,,,,,,,,,,,,,
-      KMD & USER SPACE [#kfd_support-past-60]_,.. _kfd-userspace-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      Tested user space versions,"6.3.x, 6.2.x, 6.1.x","6.3.x, 6.2.x, 6.1.x","6.3.x, 6.2.x, 6.1.x","6.3.x, 6.2.x, 6.1.x","6.3.x, 6.2.x, 6.1.x, 6.0.x","6.3.x, 6.2.x, 6.1.x, 6.0.x","6.3.x, 6.2.x, 6.1.x, 6.0.x","6.3.x, 6.2.x, 6.1.x, 6.0.x","6.3.x, 6.2.x, 6.1.x, 6.0.x, 5.7.x","6.3.x, 6.2.x, 6.1.x, 6.0.x, 5.7.x","6.3.x, 6.2.x, 6.1.x, 6.0.x, 5.7.x","6.2.x, 6.1.x, 6.0.x, 5.7.x, 5.6.x","6.2.x, 6.1.x, 6.0.x, 5.7.x, 5.6.x"
-      ,,,,,,,,,,,,,
-      ML & COMPUTER VISION,.. _mllibs-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      :doc:`Composable Kernel <composable_kernel:index>`,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0
-      :doc:`MIGraphX <amdmigraphx:index>`,2.11.0,2.11.0,2.11.0,2.11.0,2.10.0,2.10.0,2.10.0,2.10.0,2.9.0,2.9.0,2.9.0,2.8.0,2.8.0
-      :doc:`MIOpen <miopen:index>`,3.3.0,3.3.0,3.3.0,3.3.0,3.2.0,3.2.0,3.2.0,3.2.0,3.1.0,3.1.0,3.1.0,3.0.0,3.0.0
-      :doc:`MIVisionX <mivisionx:index>`,3.1.0,3.1.0,3.1.0,3.1.0,3.0.0,3.0.0,3.0.0,3.0.0,2.5.0,2.5.0,2.5.0,2.5.0,2.5.0
-      :doc:`rocAL <rocal:index>`,2.1.0,2.1.0,2.1.0,2.1.0,2.0.0,2.0.0,2.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0
-      :doc:`rocDecode <rocdecode:index>`,0.8.0,0.8.0,0.8.0,0.8.0,0.6.0,0.6.0,0.6.0,0.6.0,0.6.0,0.5.0,0.5.0,N/A,N/A
-      :doc:`rocJPEG <rocjpeg:index>`,0.6.0,0.6.0,0.6.0,0.6.0,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A
-      :doc:`rocPyDecode <rocpydecode:index>`,0.2.0,0.2.0,0.2.0,0.2.0,0.1.0,0.1.0,0.1.0,0.1.0,N/A,N/A,N/A,N/A,N/A
-      :doc:`RPP <rpp:index>`,1.9.1,1.9.1,1.9.1,1.9.1,1.8.0,1.8.0,1.8.0,1.8.0,1.5.0,1.5.0,1.5.0,1.4.0,1.4.0
-      ,,,,,,,,,,,,,
-      COMMUNICATION,.. _commlibs-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      :doc:`RCCL <rccl:index>`,2.21.5,2.21.5,2.21.5,2.21.5,2.20.5,2.20.5,2.20.5,2.20.5,2.18.6,2.18.6,2.18.6,2.18.3,2.18.3
-      ,,,,,,,,,,,,,
-      MATH LIBS,.. _mathlibs-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      `half <https://github.com/ROCm/half>`_ ,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0
-      :doc:`hipBLAS <hipblas:index>`,2.3.0,2.3.0,2.3.0,2.3.0,2.2.0,2.2.0,2.2.0,2.2.0,2.1.0,2.1.0,2.1.0,2.0.0,2.0.0
-      :doc:`hipBLASLt <hipblaslt:index>`,0.10.0,0.10.0,0.10.0,0.10.0,0.8.0,0.8.0,0.8.0,0.8.0,0.7.0,0.7.0,0.7.0,0.6.0,0.6.0
-      :doc:`hipFFT <hipfft:index>`,1.0.17,1.0.17,1.0.17,1.0.17,1.0.16,1.0.15,1.0.15,1.0.14,1.0.14,1.0.14,1.0.14,1.0.13,1.0.13
-      :doc:`hipfort <hipfort:index>`,0.5.1,0.5.1,0.5.0,0.5.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0
-      :doc:`hipRAND <hiprand:index>`,2.11.1,2.11.1,2.11.1,2.11.0,2.11.1,2.11.0,2.11.0,2.11.0,2.10.16,2.10.16,2.10.16,2.10.16,2.10.16
-      :doc:`hipSOLVER <hipsolver:index>`,2.3.0,2.3.0,2.3.0,2.3.0,2.2.0,2.2.0,2.2.0,2.2.0,2.1.1,2.1.1,2.1.0,2.0.0,2.0.0
-      :doc:`hipSPARSE <hipsparse:index>`,3.1.2,3.1.2,3.1.2,3.1.2,3.1.1,3.1.1,3.1.1,3.1.1,3.0.1,3.0.1,3.0.1,3.0.0,3.0.0
-      :doc:`hipSPARSELt <hipsparselt:index>`,0.2.2,0.2.2,0.2.2,0.2.2,0.2.1,0.2.1,0.2.1,0.2.1,0.2.0,0.1.0,0.1.0,0.1.0,0.1.0
-      :doc:`rocALUTION <rocalution:index>`,3.2.1,3.2.1,3.2.1,3.2.1,3.2.1,3.2.0,3.2.0,3.2.0,3.1.1,3.1.1,3.1.1,3.0.3,3.0.3
-      :doc:`rocBLAS <rocblas:index>`,4.3.0,4.3.0,4.3.0,4.3.0,4.2.4,4.2.1,4.2.1,4.2.0,4.1.2,4.1.0,4.1.0,4.0.0,4.0.0
-      :doc:`rocFFT <rocfft:index>`,1.0.31,1.0.31,1.0.31,1.0.31,1.0.30,1.0.29,1.0.29,1.0.28,1.0.27,1.0.27,1.0.26,1.0.25,1.0.23
-      :doc:`rocRAND <rocrand:index>`,3.2.0,3.2.0,3.2.0,3.2.0,3.1.1,3.1.0,3.1.0,3.1.0,3.0.1,3.0.1,3.0.1,3.0.0,2.10.17
-      :doc:`rocSOLVER <rocsolver:index>`,3.27.0,3.27.0,3.27.0,3.27.0,3.26.2,3.26.0,3.26.0,3.26.0,3.25.0,3.25.0,3.25.0,3.24.0,3.24.0
-      :doc:`rocSPARSE <rocsparse:index>`,3.3.0,3.3.0,3.3.0,3.3.0,3.2.1,3.2.0,3.2.0,3.2.0,3.1.2,3.1.2,3.1.2,3.0.2,3.0.2
-      :doc:`rocWMMA <rocwmma:index>`,1.6.0,1.6.0,1.6.0,1.6.0,1.5.0,1.5.0,1.5.0,1.5.0,1.4.0,1.4.0,1.4.0,1.3.0,1.3.0
-      :doc:`Tensile <tensile:src/index>`,4.42.0,4.42.0,4.42.0,4.42.0,4.41.0,4.41.0,4.41.0,4.41.0,4.40.0,4.40.0,4.40.0,4.39.0,4.39.0
-      ,,,,,,,,,,,,,
-      PRIMITIVES,.. _primitivelibs-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      :doc:`hipCUB <hipcub:index>`,3.3.0,3.3.0,3.3.0,3.3.0,3.2.1,3.2.0,3.2.0,3.2.0,3.1.0,3.1.0,3.1.0,3.0.0,3.0.0
-      :doc:`hipTensor <hiptensor:index>`,1.4.0,1.4.0,1.4.0,1.4.0,1.3.0,1.3.0,1.3.0,1.3.0,1.2.0,1.2.0,1.2.0,1.1.0,1.1.0
-      :doc:`rocPRIM <rocprim:index>`,3.3.0,3.3.0,3.3.0,3.3.0,3.2.2,3.2.0,3.2.0,3.2.0,3.1.0,3.1.0,3.1.0,3.0.0,3.0.0
-      :doc:`rocThrust <rocthrust:index>`,3.3.0,3.3.0,3.3.0,3.3.0,3.1.1,3.1.0,3.1.0,3.0.1,3.0.1,3.0.1,3.0.1,3.0.0,3.0.0
-      ,,,,,,,,,,,,,
-      SUPPORT LIBS,,,,,,,,,,,,,
-      `hipother <https://github.com/ROCm/hipother>`_,6.3.42134,6.3.42134,6.3.42133,6.3.42131,6.2.41134,6.2.41134,6.2.41134,6.2.41133,6.1.40093,6.1.40092,6.1.40091,6.1.32831,6.1.32830
-      `rocm-core <https://github.com/ROCm/rocm-core>`_,6.3.3,6.3.2,6.3.1,6.3.0,6.2.4,6.2.2,6.2.1,6.2.0,6.1.2,6.1.1,6.1.0,6.0.2,6.0.0
-      `ROCT-Thunk-Interface <https://github.com/ROCm/ROCT-Thunk-Interface>`_,N/A [#ROCT-rocr-past-60]_,N/A [#ROCT-rocr-past-60]_,N/A [#ROCT-rocr-past-60]_,N/A [#ROCT-rocr-past-60]_,20240607.5.7,20240607.5.7,20240607.4.05,20240607.1.4246,20240125.5.08,20240125.5.08,20240125.3.30,20231016.2.245,20231016.2.245
-      ,,,,,,,,,,,,,
-      SYSTEM MGMT TOOLS,.. _tools-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      :doc:`AMD SMI <amdsmi:index>`,24.7.1,24.7.1,24.7.1,24.7.1,24.6.3,24.6.3,24.6.3,24.6.2,24.5.1,24.5.1,24.4.1,23.4.2,23.4.2
-      :doc:`ROCm Data Center Tool <rdc:index>`,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0
-      :doc:`rocminfo <rocminfo:index>`,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0
-      :doc:`ROCm SMI <rocm_smi_lib:index>`,7.4.0,7.4.0,7.4.0,7.4.0,7.3.0,7.3.0,7.3.0,7.3.0,7.2.0,7.0.0,7.0.0,6.0.2,6.0.0
-      :doc:`ROCm Validation Suite <rocmvalidationsuite:index>`,1.1.0,1.1.0,1.1.0,1.1.0,1.0.60204,1.0.60202,1.0.60201,1.0.60200,1.0.60102,1.0.60101,1.0.60100,1.0.60002,1.0.60000
-      ,,,,,,,,,,,,,
-      PERFORMANCE TOOLS,,,,,,,,,,,,,
-      :doc:`ROCm Bandwidth Test <rocm_bandwidth_test:index>`,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0
-      :doc:`ROCm Compute Profiler <rocprofiler-compute:index>`,3.0.0,3.0.0,3.0.0,3.0.0,2.0.1,2.0.1,2.0.1,2.0.1,N/A,N/A,N/A,N/A,N/A
-      :doc:`ROCm Systems Profiler <rocprofiler-systems:index>`,0.1.2,0.1.1,0.1.0,0.1.0,1.11.2,1.11.2,1.11.2,1.11.2,N/A,N/A,N/A,N/A,N/A
-      :doc:`ROCProfiler <rocprofiler:index>`,2.0.60303,2.0.60302,2.0.60301,2.0.60300,2.0.60204,2.0.60202,2.0.60201,2.0.60200,2.0.60102,2.0.60101,2.0.60100,2.0.60002,2.0.60000
-      :doc:`ROCprofiler-SDK <rocprofiler-sdk:index>`,0.5.0,0.5.0,0.5.0,0.5.0,0.4.0,0.4.0,0.4.0,0.4.0,N/A,N/A,N/A,N/A,N/A
-      :doc:`ROCTracer <roctracer:index>`,4.1.60303,4.1.60302,4.1.60301,4.1.60300,4.1.60204,4.1.60202,4.1.60201,4.1.60200,4.1.60102,4.1.60101,4.1.60100,4.1.60002,4.1.60000
-      ,,,,,,,,,,,,,
-      DEVELOPMENT TOOLS,,,,,,,,,,,,,
-      :doc:`HIPIFY <hipify:index>`,18.0.0.25012,18.0.0.25012,18.0.0.24491,18.0.0.24455,18.0.0.24392,18.0.0.24355,18.0.0.24355,18.0.0.24232,17.0.0.24193,17.0.0.24154,17.0.0.24103,17.0.0.24012,17.0.0.23483
-      :doc:`ROCm CMake <rocmcmakebuildtools:index>`,0.14.0,0.14.0,0.14.0,0.14.0,0.13.0,0.13.0,0.13.0,0.13.0,0.12.0,0.12.0,0.12.0,0.11.0,0.11.0
-      :doc:`ROCdbgapi <rocdbgapi:index>`,0.77.0,0.77.0,0.77.0,0.77.0,0.76.0,0.76.0,0.76.0,0.76.0,0.71.0,0.71.0,0.71.0,0.71.0,0.71.0
-      :doc:`ROCm Debugger (ROCgdb) <rocgdb:index>`,15.2.0,15.2.0,15.2.0,15.2.0,14.2.0,14.2.0,14.2.0,14.2.0,14.1.0,14.1.0,14.1.0,13.2.0,13.2.0
-      `rocprofiler-register <https://github.com/ROCm/rocprofiler-register>`_,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.3.0,0.3.0,0.3.0,N/A,N/A
-      :doc:`ROCr Debug Agent <rocr_debug_agent:index>`,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3
-      ,,,,,,,,,,,,,
-      COMPILERS,.. _compilers-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      `clang-ocl <https://github.com/ROCm/clang-ocl>`_,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,0.5.0,0.5.0,0.5.0,0.5.0,0.5.0
-      :doc:`hipCC <hipcc:index>`,1.1.1,1.1.1,1.1.1,1.1.1,1.1.1,1.1.1,1.1.1,1.1.1,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0
-      `Flang <https://github.com/ROCm/flang>`_,18.0.0.25012,18.0.0.25012,18.0.0.24491,18.0.0.24455,18.0.0.24392,18.0.0.24355,18.0.0.24355,18.0.0.24232,17.0.0.24193,17.0.0.24154,17.0.0.24103,17.0.0.24012,17.0.0.23483
-      :doc:`llvm-project <llvm-project:index>`,18.0.0.25012,18.0.0.25012,18.0.0.24491,18.0.0.24491,18.0.0.24392,18.0.0.24355,18.0.0.24355,18.0.0.24232,17.0.0.24193,17.0.0.24154,17.0.0.24103,17.0.0.24012,17.0.0.23483
-      `OpenMP <https://github.com/ROCm/llvm-project/tree/amd-staging/openmp>`_,18.0.0.25012,18.0.0.25012,18.0.0.24491,18.0.0.24491,18.0.0.24392,18.0.0.24355,18.0.0.24355,18.0.0.24232,17.0.0.24193,17.0.0.24154,17.0.0.24103,17.0.0.24012,17.0.0.23483
-,,,,,,,,,,,,,
-      RUNTIMES,.. _runtime-support-compatibility-matrix-past-60:,,,,,,,,,,,,
-      :doc:`AMD CLR <hip:understand/amd_clr>`,6.3.42134,6.3.42134,6.3.42133,6.3.42131,6.2.41134,6.2.41134,6.2.41134,6.2.41133,6.1.40093,6.1.40092,6.1.40091,6.1.32831,6.1.32830
-      :doc:`HIP <hip:index>`,6.3.42134,6.3.42134,6.3.42133,6.3.42131,6.2.41134,6.2.41134,6.2.41134,6.2.41133,6.1.40093,6.1.40092,6.1.40091,6.1.32831,6.1.32830
-      `OpenCL Runtime <https://github.com/ROCm/clr/tree/develop/opencl>`_,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0
-      :doc:`ROCr Runtime <rocr-runtime:index>`,1.14.0,1.14.0,1.14.0,1.14.0,1.14.0,1.14.0,1.14.0,1.13.0,1.13.0,1.13.0,1.13.0,1.12.0,1.12.0
+ROCm Version,6.3.2,6.3.1,6.3.0,6.2.4,6.2.2,6.2.1,6.2.0, 6.1.2, 6.1.1, 6.1.0, 6.0.2, 6.0.0
+      :ref:`Operating systems & kernels <OS-kernel-versions>`,Ubuntu 24.04.2,Ubuntu 24.04.2,Ubuntu 24.04.2,"Ubuntu 24.04.1, 24.04","Ubuntu 24.04.1, 24.04","Ubuntu 24.04.1, 24.04",Ubuntu 24.04,,,,,
+      ,Ubuntu 22.04.5,Ubuntu 22.04.5,Ubuntu 22.04.5,"Ubuntu 22.04.5, 22.04.4","Ubuntu 22.04.5, 22.04.4","Ubuntu 22.04.5, 22.04.4","Ubuntu 22.04.5, 22.04.4","Ubuntu 22.04.4, 22.04.3","Ubuntu 22.04.4, 22.04.3","Ubuntu 22.04.4, 22.04.3","Ubuntu 22.04.4, 22.04.3, 22.04.2","Ubuntu 22.04.4, 22.04.3, 22.04.2"
+      ,,,,,,,,"Ubuntu 20.04.6, 20.04.5","Ubuntu 20.04.6, 20.04.5","Ubuntu 20.04.6, 20.04.5","Ubuntu 20.04.6, 20.04.5","Ubuntu 20.04.6, 20.04.5"
+      ,"RHEL 9.5, 9.4","RHEL 9.5, 9.4","RHEL 9.5, 9.4","RHEL 9.4, 9.3","RHEL 9.4, 9.3","RHEL 9.4, 9.3","RHEL 9.4, 9.3","RHEL 9.4, 9.3, 9.2","RHEL 9.4, 9.3, 9.2","RHEL 9.4, 9.3, 9.2","RHEL 9.3, 9.2","RHEL 9.3, 9.2"
+      ,RHEL 8.10,RHEL 8.10,RHEL 8.10,"RHEL 8.10, 8.9","RHEL 8.10, 8.9","RHEL 8.10, 8.9","RHEL 8.10, 8.9","RHEL 8.9, 8.8","RHEL 8.9, 8.8","RHEL 8.9, 8.8","RHEL 8.9, 8.8","RHEL 8.9, 8.8"
+      ,"SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP5, SP4","SLES 15 SP5, SP4","SLES 15 SP5, SP4","SLES 15 SP5, SP4","SLES 15 SP5, SP4"
+      ,,,,,,,,CentOS 7.9,CentOS 7.9,CentOS 7.9,CentOS 7.9,CentOS 7.9
+      ,Oracle Linux 8.10 [#mi300x-past-60]_,Oracle Linux 8.10 [#mi300x-past-60]_,Oracle Linux 8.10 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,Oracle Linux 8.9 [#mi300x-past-60]_,,,
+,Debian 12 [#single-node-past-60]_,Debian 12 [#single-node-past-60]_,,,,,,,,,,
+,Azure Linux 3.0 [#mi300x-past-60]_,,,,,,,,,,,
+      ,.. _architecture-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      :doc:`Architecture <rocm-install-on-linux:reference/system-requirements>`,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3,CDNA3
+      ,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2,CDNA2
+      ,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA,CDNA
+      ,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3,RDNA3
+      ,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2,RDNA2
+      ,.. _gpu-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      :doc:`GPU / LLVM target <rocm-install-on-linux:reference/system-requirements>`,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100,gfx1100
+      ,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030,gfx1030
+      ,gfx942,gfx942,gfx942,gfx942 [#mi300_624-past-60]_,gfx942 [#mi300_622-past-60]_,gfx942 [#mi300_621-past-60]_,gfx942 [#mi300_620-past-60]_, gfx942 [#mi300_612-past-60]_, gfx942 [#mi300_611-past-60]_, gfx942 [#mi300_610-past-60]_, gfx942 [#mi300_602-past-60]_, gfx942 [#mi300_600-past-60]_
+      ,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a,gfx90a
+      ,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908,gfx908
+,,,,,,,,,,,,
+      FRAMEWORK SUPPORT,.. _framework-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      :doc:`PyTorch <../compatibility/ml-compatibility/pytorch-compatibility>`,"2.4, 2.3, 2.2, 1.13","2.4, 2.3, 2.2, 1.13","2.4, 2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.3, 2.2, 2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13","2.1, 2.0, 1.13"
+      :doc:`TensorFlow <../compatibility/ml-compatibility/tensorflow-compatibility>`,"2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.14.0, 2.13.1, 2.12.1","2.14.0, 2.13.1, 2.12.1"
+      :doc:`JAX <../compatibility/ml-compatibility/jax-compatibility>`,0.4.31,0.4.31,0.4.31,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26
+      `ONNX Runtime <https://onnxruntime.ai/docs/build/eps.html#amd-migraphx>`_,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.14.1,1.14.1
+      ,,,,,,,,,,,,
+      THIRD PARTY COMMS,.. _thirdpartycomms-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      `UCC <https://github.com/ROCm/ucc>`_,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.2.0,>=1.2.0
+      `UCX <https://github.com/ROCm/ucx>`_,>=1.15.0,>=1.15.0,>=1.15.0,>=1.15.0,>=1.15.0,>=1.15.0,>=1.15.0,>=1.14.1,>=1.14.1,>=1.14.1,>=1.14.1,>=1.14.1
+      ,,,,,,,,,,,,
+      THIRD PARTY ALGORITHM,.. _thirdpartyalgorithm-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      Thrust,2.3.2,2.3.2,2.3.2,2.2.0,2.2.0,2.2.0,2.2.0,2.1.0,2.1.0,2.1.0,2.0.1,2.0.1
+      CUB,2.3.2,2.3.2,2.3.2,2.2.0,2.2.0,2.2.0,2.2.0,2.1.0,2.1.0,2.1.0,2.0.1,2.0.1
+,,,,,,,,,,,,
+      KMD & USER SPACE [#kfd_support-past-60]_,.. _kfd-userspace-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      Tested user space versions,"6.3.x, 6.2.x, 6.1.x","6.3.x, 6.2.x, 6.1.x","6.3.x, 6.2.x, 6.1.x","6.3.x, 6.2.x, 6.1.x, 6.0.x","6.3.x, 6.2.x, 6.1.x, 6.0.x","6.3.x, 6.2.x, 6.1.x, 6.0.x","6.3.x, 6.2.x, 6.1.x, 6.0.x","6.3.x, 6.2.x, 6.1.x, 6.0.x, 5.7.x","6.3.x, 6.2.x, 6.1.x, 6.0.x, 5.7.x","6.3.x, 6.2.x, 6.1.x, 6.0.x, 5.7.x","6.2.x, 6.1.x, 6.0.x, 5.7.x, 5.6.x","6.2.x, 6.1.x, 6.0.x, 5.7.x, 5.6.x"
+      ,,,,,,,,,,,,
+      ML & COMPUTER VISION,.. _mllibs-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      :doc:`Composable Kernel <composable_kernel:index>`,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0,1.1.0
+      :doc:`MIGraphX <amdmigraphx:index>`,2.11.0,2.11.0,2.11.0,2.10.0,2.10.0,2.10.0,2.10.0,2.9.0,2.9.0,2.9.0,2.8.0,2.8.0
+      :doc:`MIOpen <miopen:index>`,3.3.0,3.3.0,3.3.0,3.2.0,3.2.0,3.2.0,3.2.0,3.1.0,3.1.0,3.1.0,3.0.0,3.0.0
+      :doc:`MIVisionX <mivisionx:index>`,3.1.0,3.1.0,3.1.0,3.0.0,3.0.0,3.0.0,3.0.0,2.5.0,2.5.0,2.5.0,2.5.0,2.5.0
+      :doc:`rocAL <rocal:index>`,2.1.0,2.1.0,2.1.0,2.0.0,2.0.0,2.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0
+      :doc:`rocDecode <rocdecode:index>`,0.8.0,0.8.0,0.8.0,0.6.0,0.6.0,0.6.0,0.6.0,0.6.0,0.5.0,0.5.0,N/A,N/A
+      :doc:`rocJPEG <rocjpeg:index>`,0.6.0,0.6.0,0.6.0,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A
+      :doc:`rocPyDecode <rocpydecode:index>`,0.2.0,0.2.0,0.2.0,0.1.0,0.1.0,0.1.0,0.1.0,N/A,N/A,N/A,N/A,N/A
+      :doc:`RPP <rpp:index>`,1.9.1,1.9.1,1.9.1,1.8.0,1.8.0,1.8.0,1.8.0,1.5.0,1.5.0,1.5.0,1.4.0,1.4.0
+      ,,,,,,,,,,,,
+      COMMUNICATION,.. _commlibs-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      :doc:`RCCL <rccl:index>`,2.21.5,2.21.5,2.21.5,2.20.5,2.20.5,2.20.5,2.20.5,2.18.6,2.18.6,2.18.6,2.18.3,2.18.3
+      ,,,,,,,,,,,,
+      MATH LIBS,.. _mathlibs-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      `half <https://github.com/ROCm/half>`_ ,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0
+      :doc:`hipBLAS <hipblas:index>`,2.3.0,2.3.0,2.3.0,2.2.0,2.2.0,2.2.0,2.2.0,2.1.0,2.1.0,2.1.0,2.0.0,2.0.0
+      :doc:`hipBLASLt <hipblaslt:index>`,0.10.0,0.10.0,0.10.0,0.8.0,0.8.0,0.8.0,0.8.0,0.7.0,0.7.0,0.7.0,0.6.0,0.6.0
+      :doc:`hipFFT <hipfft:index>`,1.0.17,1.0.17,1.0.17,1.0.16,1.0.15,1.0.15,1.0.14,1.0.14,1.0.14,1.0.14,1.0.13,1.0.13
+      :doc:`hipfort <hipfort:index>`,0.5.1,0.5.0,0.5.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0
+      :doc:`hipRAND <hiprand:index>`,2.11.1,2.11.1,2.11.0,2.11.1,2.11.0,2.11.0,2.11.0,2.10.16,2.10.16,2.10.16,2.10.16,2.10.16
+      :doc:`hipSOLVER <hipsolver:index>`,2.3.0,2.3.0,2.3.0,2.2.0,2.2.0,2.2.0,2.2.0,2.1.1,2.1.1,2.1.0,2.0.0,2.0.0
+      :doc:`hipSPARSE <hipsparse:index>`,3.1.2,3.1.2,3.1.2,3.1.1,3.1.1,3.1.1,3.1.1,3.0.1,3.0.1,3.0.1,3.0.0,3.0.0
+      :doc:`hipSPARSELt <hipsparselt:index>`,0.2.2,0.2.2,0.2.2,0.2.1,0.2.1,0.2.1,0.2.1,0.2.0,0.1.0,0.1.0,0.1.0,0.1.0
+      :doc:`rocALUTION <rocalution:index>`,3.2.1,3.2.1,3.2.1,3.2.1,3.2.0,3.2.0,3.2.0,3.1.1,3.1.1,3.1.1,3.0.3,3.0.3
+      :doc:`rocBLAS <rocblas:index>`,4.3.0,4.3.0,4.3.0,4.2.4,4.2.1,4.2.1,4.2.0,4.1.2,4.1.0,4.1.0,4.0.0,4.0.0
+      :doc:`rocFFT <rocfft:index>`,1.0.31,1.0.31,1.0.31,1.0.30,1.0.29,1.0.29,1.0.28,1.0.27,1.0.27,1.0.26,1.0.25,1.0.23
+      :doc:`rocRAND <rocrand:index>`,3.2.0,3.2.0,3.2.0,3.1.1,3.1.0,3.1.0,3.1.0,3.0.1,3.0.1,3.0.1,3.0.0,2.10.17
+      :doc:`rocSOLVER <rocsolver:index>`,3.27.0,3.27.0,3.27.0,3.26.2,3.26.0,3.26.0,3.26.0,3.25.0,3.25.0,3.25.0,3.24.0,3.24.0
+      :doc:`rocSPARSE <rocsparse:index>`,3.3.0,3.3.0,3.3.0,3.2.1,3.2.0,3.2.0,3.2.0,3.1.2,3.1.2,3.1.2,3.0.2,3.0.2
+      :doc:`rocWMMA <rocwmma:index>`,1.6.0,1.6.0,1.6.0,1.5.0,1.5.0,1.5.0,1.5.0,1.4.0,1.4.0,1.4.0,1.3.0,1.3.0
+      :doc:`Tensile <tensile:src/index>`,4.42.0,4.42.0,4.42.0,4.41.0,4.41.0,4.41.0,4.41.0,4.40.0,4.40.0,4.40.0,4.39.0,4.39.0
+      ,,,,,,,,,,,,
+      PRIMITIVES,.. _primitivelibs-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      :doc:`hipCUB <hipcub:index>`,3.3.0,3.3.0,3.3.0,3.2.1,3.2.0,3.2.0,3.2.0,3.1.0,3.1.0,3.1.0,3.0.0,3.0.0
+      :doc:`hipTensor <hiptensor:index>`,1.4.0,1.4.0,1.4.0,1.3.0,1.3.0,1.3.0,1.3.0,1.2.0,1.2.0,1.2.0,1.1.0,1.1.0
+      :doc:`rocPRIM <rocprim:index>`,3.3.0,3.3.0,3.3.0,3.2.2,3.2.0,3.2.0,3.2.0,3.1.0,3.1.0,3.1.0,3.0.0,3.0.0
+      :doc:`rocThrust <rocthrust:index>`,3.3.0,3.3.0,3.3.0,3.1.1,3.1.0,3.1.0,3.0.1,3.0.1,3.0.1,3.0.1,3.0.0,3.0.0
+      ,,,,,,,,,,,,
+      SUPPORT LIBS,,,,,,,,,,,,
+      `hipother <https://github.com/ROCm/hipother>`_,6.3.42134,6.3.42133,6.3.42131,6.2.41134,6.2.41134,6.2.41134,6.2.41133,6.1.40093,6.1.40092,6.1.40091,6.1.32831,6.1.32830
+      `rocm-core <https://github.com/ROCm/rocm-core>`_,6.3.2,6.3.1,6.3.0,6.2.4,6.2.2,6.2.1,6.2.0,6.1.2,6.1.1,6.1.0,6.0.2,6.0.0
+      `ROCT-Thunk-Interface <https://github.com/ROCm/ROCT-Thunk-Interface>`_,N/A [#ROCT-rocr-past-60]_,N/A [#ROCT-rocr-past-60]_,N/A [#ROCT-rocr-past-60]_,20240607.5.7,20240607.5.7,20240607.4.05,20240607.1.4246,20240125.5.08,20240125.5.08,20240125.3.30,20231016.2.245,20231016.2.245
+      ,,,,,,,,,,,,
+      SYSTEM MGMT TOOLS,.. _tools-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      :doc:`AMD SMI <amdsmi:index>`,24.7.1,24.7.1,24.7.1,24.6.3,24.6.3,24.6.3,24.6.2,24.5.1,24.5.1,24.4.1,23.4.2,23.4.2
+      :doc:`ROCm Data Center Tool <rdc:index>`,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0,0.3.0
+      :doc:`rocminfo <rocminfo:index>`,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0
+      :doc:`ROCm SMI <rocm_smi_lib:index>`,7.4.0,7.4.0,7.4.0,7.3.0,7.3.0,7.3.0,7.3.0,7.2.0,7.0.0,7.0.0,6.0.2,6.0.0
+      :doc:`ROCm Validation Suite <rocmvalidationsuite:index>`,1.1.0,1.1.0,1.1.0,1.0.60204,1.0.60202,1.0.60201,1.0.60200,1.0.60102,1.0.60101,1.0.60100,1.0.60002,1.0.60000
+      ,,,,,,,,,,,,
+      PERFORMANCE TOOLS,,,,,,,,,,,,
+      :doc:`ROCm Bandwidth Test <rocm_bandwidth_test:index>`,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0,1.4.0
+      :doc:`ROCm Compute Profiler <rocprofiler-compute:index>`,3.0.0,3.0.0,3.0.0,2.0.1,2.0.1,2.0.1,2.0.1,N/A,N/A,N/A,N/A,N/A
+      :doc:`ROCm Systems Profiler <rocprofiler-systems:index>`,0.1.1,0.1.0,0.1.0,1.11.2,1.11.2,1.11.2,1.11.2,N/A,N/A,N/A,N/A,N/A
+      :doc:`ROCProfiler <rocprofiler:index>`,2.0.60302,2.0.60301,2.0.60300,2.0.60204,2.0.60202,2.0.60201,2.0.60200,2.0.60102,2.0.60101,2.0.60100,2.0.60002,2.0.60000
+      :doc:`ROCprofiler-SDK <rocprofiler-sdk:index>`,0.5.0,0.5.0,0.5.0,0.4.0,0.4.0,0.4.0,0.4.0,N/A,N/A,N/A,N/A,N/A
+      :doc:`ROCTracer <roctracer:index>`,4.1.60302,4.1.60301,4.1.60300,4.1.60204,4.1.60202,4.1.60201,4.1.60200,4.1.60102,4.1.60101,4.1.60100,4.1.60002,4.1.60000
+      ,,,,,,,,,,,,
+      DEVELOPMENT TOOLS,,,,,,,,,,,,
+      :doc:`HIPIFY <hipify:index>`,18.0.0.25012,18.0.0.24491,18.0.0.24455,18.0.0.24392,18.0.0.24355,18.0.0.24355,18.0.0.24232,17.0.0.24193,17.0.0.24154,17.0.0.24103,17.0.0.24012,17.0.0.23483
+      :doc:`ROCm CMake <rocmcmakebuildtools:index>`,0.14.0,0.14.0,0.14.0,0.13.0,0.13.0,0.13.0,0.13.0,0.12.0,0.12.0,0.12.0,0.11.0,0.11.0
+      :doc:`ROCdbgapi <rocdbgapi:index>`,0.77.0,0.77.0,0.77.0,0.76.0,0.76.0,0.76.0,0.76.0,0.71.0,0.71.0,0.71.0,0.71.0,0.71.0
+      :doc:`ROCm Debugger (ROCgdb) <rocgdb:index>`,15.2.0,15.2.0,15.2.0,14.2.0,14.2.0,14.2.0,14.2.0,14.1.0,14.1.0,14.1.0,13.2.0,13.2.0
+      `rocprofiler-register <https://github.com/ROCm/rocprofiler-register>`_,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.4.0,0.3.0,0.3.0,0.3.0,N/A,N/A
+      :doc:`ROCr Debug Agent <rocr_debug_agent:index>`,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3,2.0.3
+      ,,,,,,,,,,,,
+      COMPILERS,.. _compilers-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      `clang-ocl <https://github.com/ROCm/clang-ocl>`_,N/A,N/A,N/A,N/A,N/A,N/A,N/A,0.5.0,0.5.0,0.5.0,0.5.0,0.5.0
+      :doc:`hipCC <hipcc:index>`,1.1.1,1.1.1,1.1.1,1.1.1,1.1.1,1.1.1,1.1.1,1.0.0,1.0.0,1.0.0,1.0.0,1.0.0
+      `Flang <https://github.com/ROCm/flang>`_,18.0.0.25012,18.0.0.24491,18.0.0.24455,18.0.0.24392,18.0.0.24355,18.0.0.24355,18.0.0.24232,17.0.0.24193,17.0.0.24154,17.0.0.24103,17.0.0.24012,17.0.0.23483
+      :doc:`llvm-project <llvm-project:index>`,18.0.0.25012,18.0.0.24491,18.0.0.24491,18.0.0.24392,18.0.0.24355,18.0.0.24355,18.0.0.24232,17.0.0.24193,17.0.0.24154,17.0.0.24103,17.0.0.24012,17.0.0.23483
+      `OpenMP <https://github.com/ROCm/llvm-project/tree/amd-staging/openmp>`_,18.0.0.25012,18.0.0.24491,18.0.0.24491,18.0.0.24392,18.0.0.24355,18.0.0.24355,18.0.0.24232,17.0.0.24193,17.0.0.24154,17.0.0.24103,17.0.0.24012,17.0.0.23483
+,,,,,,,,,,,,
+      RUNTIMES,.. _runtime-support-compatibility-matrix-past-60:,,,,,,,,,,,
+      :doc:`AMD CLR <hip:understand/amd_clr>`,6.3.42134,6.3.42133,6.3.42131,6.2.41134,6.2.41134,6.2.41134,6.2.41133,6.1.40093,6.1.40092,6.1.40091,6.1.32831,6.1.32830
+      :doc:`HIP <hip:index>`,6.3.42134,6.3.42133,6.3.42131,6.2.41134,6.2.41134,6.2.41134,6.2.41133,6.1.40093,6.1.40092,6.1.40091,6.1.32831,6.1.32830
+      `OpenCL Runtime <https://github.com/ROCm/clr/tree/develop/opencl>`_,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0,2.0.0
+      :doc:`ROCr Runtime <rocr-runtime:index>`,1.14.0,1.14.0,1.14.0,1.14.0,1.14.0,1.14.0,1.13.0,1.13.0,1.13.0,1.13.0,1.12.0,1.12.0
--- a/docs/compatibility/compatibility-matrix.rst
+++ b/docs/compatibility/compatibility-matrix.rst
@@ -23,7 +23,7 @@ compatibility and system requirements.
 .. container:: format-big-table

  .. csv-table::
-      :header: "ROCm Version", "6.3.3", "6.3.2", "6.2.0"
+      :header: "ROCm Version", "6.3.2", "6.3.1", "6.2.0"
      :stub-columns: 1

      :ref:`Operating systems & kernels <OS-kernel-versions>`,Ubuntu 24.04.2,Ubuntu 24.04.2,Ubuntu 24.04
@@ -33,7 +33,7 @@ compatibility and system requirements.
      ,"SLES 15 SP6, SP5","SLES 15 SP6, SP5","SLES 15 SP6, SP5"
      ,Oracle Linux 8.10 [#mi300x]_,Oracle Linux 8.10 [#mi300x]_,Oracle Linux 8.9 [#mi300x]_
      ,Debian 12 [#single-node]_,Debian 12 [#single-node]_,
-      ,Azure Linux 3.0 [#mi300x]_,Azure Linux 3.0 [#mi300x]_,
+      ,Azure Linux 3.0 [#mi300x]_,,
      ,.. _architecture-support-compatibility-matrix:,,
      :doc:`Architecture <rocm-install-on-linux:reference/system-requirements>`,CDNA3,CDNA3,CDNA3
      ,CDNA2,CDNA2,CDNA2
@@ -83,7 +83,7 @@ compatibility and system requirements.
      :doc:`hipBLAS <hipblas:index>`,2.3.0,2.3.0,2.2.0
      :doc:`hipBLASLt <hipblaslt:index>`,0.10.0,0.10.0,0.8.0
      :doc:`hipFFT <hipfft:index>`,1.0.17,1.0.17,1.0.14
-      :doc:`hipfort <hipfort:index>`,0.5.1,0.5.1,0.4.0
+      :doc:`hipfort <hipfort:index>`,0.5.1,0.5.0,0.4.0
      :doc:`hipRAND <hiprand:index>`,2.11.1,2.11.1,2.11.0
      :doc:`hipSOLVER <hipsolver:index>`,2.3.0,2.3.0,2.2.0
      :doc:`hipSPARSE <hipsparse:index>`,3.1.2,3.1.2,3.1.1
@@ -104,8 +104,8 @@ compatibility and system requirements.
      :doc:`rocThrust <rocthrust:index>`,3.3.0,3.3.0,3.0.1
      ,,,
      SUPPORT LIBS,,,
-      `hipother <https://github.com/ROCm/hipother>`_,6.3.42134,6.3.42134,6.2.41133
-      `rocm-core <https://github.com/ROCm/rocm-core>`_,6.3.3,6.3.2,6.2.0
+      `hipother <https://github.com/ROCm/hipother>`_,6.3.42134,6.3.42133,6.2.41133
+      `rocm-core <https://github.com/ROCm/rocm-core>`_,6.3.2,6.3.1,6.2.0
      `ROCT-Thunk-Interface <https://github.com/ROCm/ROCT-Thunk-Interface>`_,N/A [#ROCT-rocr]_,N/A [#ROCT-rocr]_,20240607.1.4246
      ,,,
      SYSTEM MGMT TOOLS,.. _tools-support-compatibility-matrix:,,
@@ -118,35 +118,34 @@ compatibility and system requirements.
      PERFORMANCE TOOLS,,,
      :doc:`ROCm Bandwidth Test <rocm_bandwidth_test:index>`,1.4.0,1.4.0,1.4.0
      :doc:`ROCm Compute Profiler <rocprofiler-compute:index>`,3.0.0,3.0.0,2.0.1
-      :doc:`ROCm Systems Profiler <rocprofiler-systems:index>`,0.1.2,0.1.1,1.11.2
-      :doc:`ROCProfiler <rocprofiler:index>`,2.0.60303,2.0.60302,2.0.60200
+      :doc:`ROCm Systems Profiler <rocprofiler-systems:index>`,0.1.1,0.1.0,1.11.2
+      :doc:`ROCProfiler <rocprofiler:index>`,2.0.60302,2.0.60301,2.0.60200
      :doc:`ROCprofiler-SDK <rocprofiler-sdk:index>`,0.5.0,0.5.0,0.4.0
-      :doc:`ROCTracer <roctracer:index>`,4.1.60303,4.1.60302,4.1.60200
+      :doc:`ROCTracer <roctracer:index>`,4.1.60302,4.1.60301,4.1.60200
      ,,,
      DEVELOPMENT TOOLS,,,
-      :doc:`HIPIFY <hipify:index>`,18.0.0.25012,18.0.0.25012,18.0.0.24232
+      :doc:`HIPIFY <hipify:index>`,18.0.0.25012,18.0.0.24491,18.0.0.24232
      :doc:`ROCm CMake <rocmcmakebuildtools:index>`,0.14.0,0.14.0,0.13.0
      :doc:`ROCdbgapi <rocdbgapi:index>`,0.77.0,0.77.0,0.76.0
      :doc:`ROCm Debugger (ROCgdb) <rocgdb:index>`,15.2.0,15.2.0,14.2.0
      `rocprofiler-register <https://github.com/ROCm/rocprofiler-register>`_,0.4.0,0.4.0,0.4.0
      :doc:`ROCr Debug Agent <rocr_debug_agent:index>`,2.0.3,2.0.3,2.0.3
      ,,,
-      COMPILERS,.. _compilers-support-compatibility-matrix:,,
+      COMPILERS,.. _compilers-support-compatibility-matrix:,..
      `clang-ocl <https://github.com/ROCm/clang-ocl>`_,N/A,N/A,N/A
      :doc:`hipCC <hipcc:index>`,1.1.1,1.1.1,1.1.1
-      `Flang <https://github.com/ROCm/flang>`_,18.0.0.25012,18.0.0.25012,18.0.0.24232
-      :doc:`llvm-project <llvm-project:index>`,18.0.0.25012,18.0.0.25012,18.0.0.24232
-      `OpenMP <https://github.com/ROCm/llvm-project/tree/amd-staging/openmp>`_,18.0.0.25012,18.0.0.25012,18.0.0.24232
+      `Flang <https://github.com/ROCm/flang>`_,18.0.0.25012,18.0.0.24491,18.0.0.24232
+      :doc:`llvm-project <llvm-project:index>`,18.0.0.25012,18.0.0.24491,18.0.0.24232
+      `OpenMP <https://github.com/ROCm/llvm-project/tree/amd-staging/openmp>`_,18.0.0.25012,18.0.0.24491,18.0.0.24232
      ,,,
-      RUNTIMES,.. _runtime-support-compatibility-matrix:,,
-      :doc:`AMD CLR <hip:understand/amd_clr>`,6.3.42134,6.3.42134,6.2.41133
-      :doc:`HIP <hip:index>`,6.3.42134,6.3.42134,6.2.41133
+      RUNTIMES,.. _runtime-support-compatibility-matrix:,..
+      :doc:`AMD CLR <hip:understand/amd_clr>`,6.3.42134,6.3.42133,6.2.41133
+      :doc:`HIP <hip:index>`,6.3.42134,6.3.42133,6.2.41133
      `OpenCL Runtime <https://github.com/ROCm/clr/tree/develop/opencl>`_,2.0.0,2.0.0,2.0.0
      :doc:`ROCr Runtime <rocr-runtime:index>`,1.14.0,1.14.0,1.13.0



-
 .. rubric:: Footnotes

 .. [#mi300x] Oracle Linux and Azure Linux are supported only on AMD Instinct MI300X.
--- a/docs/compatibility/ml-compatibility/jax-compatibility.rst
+++ b/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
--- a/docs/compatibility/ml-compatibility/pytorch-compatibility.rst
+++ b/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
 ********************************************************************************
@@ -58,7 +56,7 @@ Docker image compatibility

 AMD validates and publishes ready-made `PyTorch images <https://hub.docker.com/r/rocm/pytorch>`_
 with ROCm backends on Docker Hub. The following Docker image tags and
-associated inventories are validated for `ROCm 6.3.3 <https://repo.radeon.com/rocm/apt/6.3.3/>`_.
+associated inventories are validated for `ROCm 6.3.0 <https://repo.radeon.com/rocm/apt/6.3/>`_.
 Click the |docker-icon| icon to view the image on Docker Hub.

 .. list-table:: PyTorch Docker image components
@@ -79,26 +77,11 @@ Click the |docker-icon| icon to view the image on Docker Hub.

    * - .. raw:: html

-           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3.3_ubuntu24.04_py3.12_pytorch_release_2.4.0/images/sha256-6c798857b2c9526b44ba535710b93a1737546acea79b53a93c646195c272f1d5"><i class="fab fa-docker fa-lg"></i></a>
+           <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
-      - `3.12.9 <https://www.python.org/downloads/release/python-3129/>`_
-      - `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/2.13.0>`_
-      - `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/rocm6.3.3_ubuntu22.04_py3.10_pytorch_release_2.4.0/images/sha256-a09b21248133876fc8912a5ff4e6ee2c8d62b14120313e426b3dadda5702713d"><i class="fab fa-docker fa-lg"></i></a>
-
-      - `2.4.0 <https://github.com/ROCm/pytorch/tree/release/2.4>`_
-      - 22.04
-      - `3.10.16 <https://www.python.org/downloads/release/python-31016/>`_
+      - `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/2.13.0>`_
@@ -109,11 +92,11 @@ Click the |docker-icon| icon to view the image on Docker Hub.

    * - .. raw:: html

-           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3.3_ubuntu22.04_py3.9_pytorch_release_2.4.0/images/sha256-963187534467f0f9da77996762fc1d112a6faa5372277c348a505533e7876ec8"><i class="fab fa-docker fa-lg"></i></a>
+           <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
-      - `3.9.21 <https://www.python.org/downloads/release/python-3921/>`_
+      - `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/2.13.0>`_
@@ -124,11 +107,26 @@ Click the |docker-icon| icon to view the image on Docker Hub.

    * - .. raw:: html

-           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3.3_ubuntu22.04_py3.10_pytorch_release_2.3.0/images/sha256-952f2621bd2bf3078bef19061e05b209105a82a7908e7e6cdf85014938a4d93a"><i class="fab fa-docker fa-lg"></i></a>
+           <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
+      - `3.9.18 <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/2.13.0>`_
+      - `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
-      - `3.10.16 <https://www.python.org/downloads/release/python-31016/>`_
+      - `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/2.13.0>`_
@@ -139,7 +137,7 @@ Click the |docker-icon| icon to view the image on Docker Hub.

    * - .. raw:: html

-           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3.3_ubuntu22.04_py3.10_pytorch_release_2.2.1/images/sha256-a2fe20e170feb9e05da3e5728bb98e40d08567e137be8e6ba797962ed2852608"><i class="fab fa-docker fa-lg"></i></a>
+           <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
@@ -154,7 +152,7 @@ Click the |docker-icon| icon to view the image on Docker Hub.

    * - .. raw:: html

-           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3.3_ubuntu20.04_py3.9_pytorch_release_2.2.1/images/sha256-7f231937c897cca5f89e360be33c70a2017d60f62d1fbe81292be48c15fe345b"><i class="fab fa-docker fa-lg"></i></a>
+           <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
@@ -169,14 +167,14 @@ Click the |docker-icon| icon to view the image on Docker Hub.

    * - .. raw:: html

-           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3.3_ubuntu22.04_py3.9_pytorch_release_1.13.1/images/sha256-616a47758004f91951e2da6c1fe291f903de65a7b2318d4b18359b48fe3032f4"><i class="fab fa-docker fa-lg"></i></a>
+           <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
      - `3.9.21 <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>`_
-      - `2.19.0 <https://github.com/tensorflow/tensorboard/tree/2.19>`_
+      - `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>`_
@@ -184,7 +182,7 @@ Click the |docker-icon| icon to view the image on Docker Hub.

    * - .. raw:: html

-           <a href="https://hub.docker.com/layers/rocm/pytorch/rocm6.3.3_ubuntu20.04_py3.9_pytorch_release_1.13.1/images/sha256-a2cfb365aea58b84595e241ffdb0d5ef3e6566e98c10b5499f4aa29983a74ea2"><i class="fab fa-docker fa-lg"></i></a>
+           <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
@@ -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.
--- a/docs/compatibility/ml-compatibility/tensorflow-compatibility.rst
+++ b/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
 *******************************************************************************
@@ -56,7 +54,7 @@ Docker image compatibility
 AMD validates and publishes ready-made `TensorFlow images
 <https://hub.docker.com/r/rocm/tensorflow>`_ with ROCm backends on
 Docker Hub. The following Docker image tags and associated inventories are
-validated for `ROCm 6.3.3 <https://repo.radeon.com/rocm/apt/6.3.3/>`_. Click
+validated for `ROCm 6.3.1 <https://repo.radeon.com/rocm/apt/6.3.1/>`_. Click
 the |docker-icon| icon to view the image on Docker Hub.

 .. list-table:: TensorFlow Docker image components
@@ -70,47 +68,47 @@ the |docker-icon| icon to view the image on Docker Hub.

    * - .. raw:: html

-           <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>
+           <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.1-py3.12-tf2.17.0-dev/images/sha256-804121ee4985718277ba7dcec53c57bdade130a1ef42f544b6c48090ad379c17"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>

-      - `tensorflow-rocm 2.17.0 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/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/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/>`_
+      - `Python 3.12 <https://www.python.org/downloads/release/python-3124/>`_
      - `TensorBoard 2.17.1 <https://github.com/tensorflow/tensorboard/tree/2.17.1>`_

    * - .. raw:: html

-           <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>
+           <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.1-py3.10-tf2.17.0-dev/images/sha256-776837ffa945913f6c466bfe477810a11453d21d5b6afb200be1c36e48fbc08e"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>

-      - `tensorflow-rocm 2.17.0 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/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/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>`_
+      - `Python 3.10 <https://www.python.org/downloads/release/python-31012/>`_
+      - `TensorBoard 2.17.0 <https://github.com/tensorflow/tensorboard/tree/2.17.0>`_

    * - .. raw:: html

-           <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>
+           <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.1-py3.12-tf2.16.2-dev/images/sha256-c793e1483e30809c3c28fc5d7805bedc033c73da224f839fff370717cb100944"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>

-      - `tensorflow-rocm 2.16.2 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/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/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/>`_
+      - `Python 3.12 <https://www.python.org/downloads/release/python-3124/>`_
      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`_

    * - .. raw:: html

-           <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>
+           <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.1-py3.10-tf2.16.0-dev/images/sha256-263e78414ae85d7bcd52a025a94131d0a279872a45ed632b9165336dfdcd4443"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>

-      - `tensorflow-rocm 2.16.2 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/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/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/>`_
+      - `Python 3.10 <https://www.python.org/downloads/release/python-31012/>`_
      - `TensorBoard 2.16.2 <https://github.com/tensorflow/tensorboard/tree/2.16.2>`_

    * - .. raw:: html

-           <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>
+           <a href="https://hub.docker.com/layers/rocm/tensorflow/rocm6.3.1-py3.10-tf2.15.0-dev/images/sha256-479046a8477ca701a9494a813ab17e8ab4f6baa54641e65dc8d07629f1e6a880"><i class="fab fa-docker fa-lg"></i> rocm/tensorflow</a>

-      - `tensorflow-rocm 2.15.1 <https://repo.radeon.com/rocm/manylinux/rocm-rel-6.3.3/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/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/>`_
+      - `Python 3.10 <https://www.python.org/downloads/release/python-31012/>`_
      - `TensorBoard 2.15.2 <https://github.com/tensorflow/tensorboard/tree/2.15.2>`_

 Critical ROCm libraries for TensorFlow
@@ -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
--- a/docs/compatibility/pytorch-compatibility.rst
+++ b/docs/compatibility/pytorch-compatibility.rst
@@ -1,916 +0,0 @@
-.. 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>`_
--- a/docs/conceptual/gpu-arch.md
+++ b/docs/conceptual/gpu-arch.md
@@ -32,7 +32,7 @@ architecture.

 * [AMD Instinct™ MI250 microarchitecture](./gpu-arch/mi250.md)
 * [AMD Instinct MI200/CDNA2 ISA](https://www.amd.com/system/files/TechDocs/instinct-mi200-cdna2-instruction-set-architecture.pdf)
-* [White paper](https://www.amd.com/content/dam/amd/en/documents/instinct-business-docs/white-papers/amd-cdna2-white-paper.pdf)
+* [White paper](https://www.amd.com/system/files/documents/amd-cdna2-white-paper.pdf)
 * [Performance counters](./gpu-arch/mi300-mi200-performance-counters.rst)

 :::
@@ -45,7 +45,7 @@ architecture.

 * [AMD Instinct™ MI100 microarchitecture](./gpu-arch/mi100.md)
 * [AMD Instinct MI100/CDNA1 ISA](https://www.amd.com/system/files/TechDocs/instinct-mi100-cdna1-shader-instruction-set-architecture%C2%A0.pdf)
-* [White paper](https://www.amd.com/content/dam/amd/en/documents/instinct-business-docs/white-papers/amd-cdna-white-paper.pdf)
+* [White paper](https://www.amd.com/system/files/documents/amd-cdna-whitepaper.pdf)

 :::

@@ -55,6 +55,7 @@ architecture.
 * [AMD RDNA3 ISA](https://www.amd.com/system/files/TechDocs/rdna3-shader-instruction-set-architecture-feb-2023_0.pdf)
 * [AMD RDNA2 ISA](https://www.amd.com/system/files/TechDocs/rdna2-shader-instruction-set-architecture.pdf)
 * [AMD RDNA ISA](https://www.amd.com/system/files/TechDocs/rdna-shader-instruction-set-architecture.pdf)
+* [AMD RDNA Architecture White Paper](https://www.amd.com/system/files/documents/rdna-whitepaper.pdf)

 :::

--- a/docs/conf.py
+++ b/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")

@@ -32,15 +30,15 @@ if os.environ.get("READTHEDOCS", "") == "True":
 project = "ROCm Documentation"
 author = "Advanced Micro Devices, Inc."
 copyright = "Copyright (c) 2025 Advanced Micro Devices, Inc. All rights reserved."
-version = "6.3.3"
-release = "6.3.3"
+version = "6.3.2"
+release = "6.3.2"
 setting_all_article_info = True
 all_article_info_os = ["linux", "windows"]
 all_article_info_author = ""

 # pages with specific settings
 article_pages = [
-    {"file": "about/release-notes", "os": ["linux"], "date": "2025-02-19"},
+    {"file": "about/release-notes", "os": ["linux"], "date": "2025-01-28"},
    {"file": "compatibility/compatibility-matrix", "os": ["linux"]},
    {"file": "compatibility/ml-compatibility/pytorch-compatibility", "os": ["linux"]},
    {"file": "compatibility/ml-compatibility/tensorflow-compatibility", "os": ["linux"]},
@@ -51,9 +49,6 @@ 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/scale-model-training", "os": ["linux"]},

    {"file": "how-to/rocm-for-ai/fine-tuning/index", "os": ["linux"]},
@@ -68,7 +63,7 @@ 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 +86,11 @@ 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 +101,8 @@ 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"

--- a/docs/data/how-to/rocm-for-ai/inference/vllm-benchmark-models.yaml
+++ b/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
--- a/docs/extension/init.py
+++ b/docs/extension/init.py
--- a/docs/extension/how-to/rocm-for-ai/inference/vllm-benchmark.js
+++ b/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();
-});
--- a/docs/extension/version-ref.py
+++ b/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,
-    }
--- a/docs/how-to/rocm-for-ai/inference/vllm-benchmark.rst
+++ b/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
+:ref:`Docker container <vllm-benchmark-get-started>` 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>`_
--- a/docs/how-to/rocm-for-ai/training/benchmark-docker/megatron-lm.rst
+++ b/docs/how-to/rocm-for-ai/training/benchmark-docker/megatron-lm.rst
@@ -1,547 +0,0 @@
-:orphan:
-
-.. meta::
-   :description: How to train a model using Megatron-LM for ROCm.
-   :keywords: ROCm, AI, LLM, train, Megatron-LM, megatron, Llama, tutorial, docker, torch
-
-******************************************
-Training a model with Megatron-LM for ROCm
-******************************************
-
-The Megatron-LM framework for ROCm is a specialized fork of the robust Megatron-LM,
-designed to enable efficient training of large-scale language models on AMD
-GPUs. By leveraging AMD Instinct™ MI300X series accelerators, Megatron-LM delivers
-enhanced scalability, performance, and resource utilization for AI workloads.
-It is purpose-built to support models like Llama 2, Llama 3, Llama 3.1, and
-DeepSeek, enabling developers to train next-generation AI models more
-efficiently. See the GitHub repository at `<https://github.com/ROCm/Megatron-LM>`__.
-
-AMD provides a ready-to-use Docker image for MI300X accelerators containing
-essential components, including PyTorch, ROCm libraries, and Megatron-LM
-utilities. It contains the following software components to accelerate training
-workloads:
-
-+--------------------------+--------------------------------+
-| Software component       | Version                        |
-+==========================+================================+
-| ROCm                     | 6.3.0                          |
-+--------------------------+--------------------------------+
-| PyTorch                  | 2.7.0a0+git637433              |
-+--------------------------+--------------------------------+
-| Python                   | 3.10                           |
-+--------------------------+--------------------------------+
-| Transformer Engine       | 1.11                           |
-+--------------------------+--------------------------------+
-| Flash Attention          | 3.0.0                          |
-+--------------------------+--------------------------------+
-| hipBLASLt                | git258a2162                    |
-+--------------------------+--------------------------------+
-| Triton                   | 3.1                            |
-+--------------------------+--------------------------------+
-
-Supported features and models
-=============================
-
-Megatron-LM provides the following key features to train large language models efficiently:
-
- Transformer Engine (TE)
-
- APEX
-
- GEMM tuning
-
- Torch.compile
-
- 3D parallelism: TP + SP + CP
-
- Distributed optimizer
-
- Flash Attention (FA) 3
-
- Fused kernels
-
- Pre-training
-
-.. _amd-megatron-lm-model-support:
-
-The following models are pre-optimized for performance on the AMD Instinct MI300X accelerator.
-
-* Llama 2 7B
-
-* Llama 2 70B
-
-* Llama 3 8B
-
-* Llama 3 70B
-
-* Llama 3.1 8B
-
-* Llama 3.1 70B
-
-* DeepSeek-V2-Lite
-
-.. note::
-
-   Some models, such as Llama 3, require an external license agreement through
-   a third party (for example, Meta).
-
-System validation
-=================
-
-If you have already validated your system settings, skip this step. Otherwise,
-complete the :ref:`system validation and optimization steps <train-a-model-system-validation>`
-to set up your system before starting training.
-
-Disable NUMA auto-balancing
---------------------------
-
-Generally, application performance can benefit from disabling NUMA auto-balancing. However,
-it might be detrimental to performance with certain types of workloads.
-
-Run the command ``cat /proc/sys/kernel/numa_balancing`` to check your current NUMA (Non-Uniform
-Memory Access) settings. Output ``0`` indicates this setting is disabled. If there is no output or
-the output is ``1``, run the following command to disable NUMA auto-balancing.
-
-.. code-block:: shell
-
-   sudo sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
-
-See :ref:`mi300x-disable-numa` for more information.
-
-.. _mi300x-amd-megatron-lm-training:
-
-Environment setup
-=================
-
-The pre-built ROCm Megatron-LM environment allows users to quickly validate system performance, conduct
-training benchmarks, and achieve superior performance for models like Llama 3.1, Llama 2, and DeepSeek V2.
-
-Use the following instructions to set up the environment, configure the script to train models, and
-reproduce the benchmark results on the MI300X accelerators with the AMD Megatron-LM Docker
-image.
-
-.. _amd-megatron-lm-requirements:
- 
-Download the Docker image
-------------------------
-
-1. Use the following command to pull the Docker image from Docker Hub.
-
-   .. code-block:: shell
-
-      docker pull rocm/megatron-lm:v25.3
-
-2. Launch the Docker container.
-
-   .. code-block:: shell
-
-      docker run -it --device /dev/dri --device /dev/kfd --network host --ipc host --group-add video --cap-add SYS_PTRACE --security-opt seccomp=unconfined --privileged -v $HOME:$HOME -v  $HOME/.ssh:/root/.ssh --shm-size 64G --name megatron_training_env rocm/megatron-lm:v25.3
-
-3. Use these commands if you exit the ``megatron_training_env`` container and need to return to it.
-
-   .. code-block:: shell
-
-      docker start megatron_training_env
-      docker exec -it megatron_training_env bash
-
-The Docker container includes a pre-installed, verified version of Megatron-LM from the `release branch <https://github.com/ROCm/Megatron-LM/tree/megatron_release_v25.3>`_.
-
-.. _amd-megatron-lm-environment-setup:
-
-Configuration scripts
---------------------
-
-.. tab-set::
-
-   .. tab-item:: Llama
-      :sync: llama
-
-      If you're working with Llama 2 7B or Llama 2 70 B, use the ``train_llama2.sh`` configuration
-      script in the ``examples/llama`` directory of
-      `<https://github.com/ROCm/Megatron-LM/tree/megatron_release_v25.3/examples/llama>`__.
-      Likewise, if you're working with Llama 3 or Llama 3.1, then use ``train_llama3.sh`` and update
-      the configuration script accordingly.
-
-   .. tab-item:: DeepSeek V2
-      :sync: deepseek
-
-      Use the ``train_deepseek_v2.sh`` configuration script in the ``examples/deepseek_v2``
-      directory of
-      `<https://github.com/ROCm/Megatron-LM/tree/megatron_release_v25.3/examples/deepseek_v2>`__
-      and update the configuration script accordingly.
-
-Network interface
-^^^^^^^^^^^^^^^^^
-
-.. tab-set::
-
-   .. tab-item:: Llama
-      :sync: llama
-
-      To avoid connectivity issues in multi-node deployments, ensure the correct network interface
-      is set in your training scripts.
-
-      1. Run the following command (outside the container) to find the active network interface on your system.
-
-         .. code-block:: shell
-
-            ip a
-
-      2. Update the ``NCCL_SOCKET_IFNAME`` and ``GLOO_SOCKET_IFNAME`` variables with your system’s network interface. For
-         example:
-
-         .. code-block:: shell
-
-            export NCCL_SOCKET_IFNAME=ens50f0np0
-
-            export GLOO_SOCKET_IFNAME=ens50f0np0
-
-Dataset options
-^^^^^^^^^^^^^^^
-
-.. tab-set::
-
-   .. tab-item:: Llama
-      :sync: llama
-
-      You can use either mock data or real data for training.
-
-      * Mock data can be useful for testing and validation. Use the ``MOCK_DATA`` variable to toggle between mock and real data. The default
-        value is ``1`` for enabled.
-
-        .. code-block:: bash
-
-           MOCK_DATA=1
-
-      * If you're using a real dataset, update the ``DATA_PATH`` variable to point to the location of your dataset.
-
-        .. code-block:: bash
-
-           MOCK_DATA=0
-
-           DATA_PATH=${DATA_PATH:-"/data/bookcorpus_text_sentence"}  # Change to where your dataset is stored
-
-        Ensure that the files are accessible inside the Docker container.
-
-   .. tab-item:: DeepSeek V2
-      :sync: deepseek
-
-      If you don't already have the dataset, download the DeepSeek dataset using the following
-      commands:
-
-      .. code-block:: shell
-
-         mkdir deepseek-datasets
-         cd deepseek-datasets
-         wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/SlimPajama.json
-         wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-train.json
-         wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/alpaca_zh-valid.json
-         wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.bin
-         wget https://atp-modelzoo-wlcb-pai.oss-cn-wulanchabu.aliyuncs.com/release/models/pai-megatron-patch/deepseek-datasets/mmap_deepseekv2_datasets_text_document.idx
-
-      You can use either mock data or real data for training.
-
-      * Mock data can be useful for testing and validation. Use the ``MOCK_DATA`` variable to toggle between mock and real data. The default
-        value is ``1`` for enabled.
-
-        .. code-block:: bash
-
-           MOCK_DATA=1
-
-      * If you're using a real dataset, update the ``DATA_DIR`` variable to point to the location of your dataset.
-
-        .. code-block:: bash
-
-           MOCK_DATA=0
-
-           DATA_DIR="/root/data/deepseek-datasets"  # Change to where your dataset is stored
-
-        Ensure that the files are accessible inside the Docker container.
-
-Tokenizer
-^^^^^^^^^
-
-Tokenization is the process of converting raw text into tokens that can be processed by the model. For Llama
-models, this typically involves sub-word tokenization, where words are broken down into smaller units based on
-a fixed vocabulary. The tokenizer is trained along with the model on a large corpus of text, and it learns a
-fixed vocabulary that can represent a wide range of text from different domains. This allows Llama models to
-handle a variety of input sequences, including unseen words or domain-specific terms.
-
-.. tab-set::
-
-   .. tab-item:: Llama
-      :sync: llama
-
-      To train any of the Llama 2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support>`, use the ``Llama2Tokenizer``.
-
-      To train any of Llama 3 and Llama 3.1 models that this Docker image supports, use the ``HuggingFaceTokenizer``.
-      Set the Hugging Face model link in the ``TOKENIZER_MODEL`` variable.
-
-      For example, if you're using the Llama 3.1 8B model:
-
-      .. code-block:: shell
-
-         TOKENIZER_MODEL=meta-llama/Llama-3.1-8B
-
-   .. tab-item:: DeepSeek V2
-      :sync: deepseek
-
-      To train any of the DeepSeek V2 models that :ref:`this Docker image supports <amd-megatron-lm-model-support>`, use the ``DeepSeekV2Tokenizer``.
-
-Multi-node training
-^^^^^^^^^^^^^^^^^^^
-
-.. tab-set::
-
-   .. tab-item:: Llama
-      :sync: llama
-
-      If you're running multi-node training, update the following environment variables. They can
-      also be passed as command line arguments.
-
-      * Change ``localhost`` to the master node's hostname:
-
-        .. code-block:: shell
-
-           MASTER_ADDR="${MASTER_ADDR:-localhost}"
-
-      * Set the number of nodes you want to train on (for instance, ``2``, ``4``, ``8``):
-
-        .. code-block:: shell
-
-           NNODES="${NNODES:-1}"
-
-      * Set the rank of each node (0 for master, 1 for the first worker node, and so on):
-
-        .. code-block:: shell
-
-           NODE_RANK="${NODE_RANK:-0}"
-
-      * Set ``DATA_CACHE_PATH`` to a common directory accessible by all the nodes (for example, an
-        NFS directory) for multi-node runs:
-
-        .. code-block:: shell
-
-           DATA_CACHE_PATH=/root/cache # Set to a common directory for multi-node runs
-
-      * For multi-node runs, make sure the correct network drivers are installed on the nodes. If
-        inside a Docker, either install the drivers inside the Docker container or pass the network
-        drivers from the host while creating the Docker container.
-
-Start training on AMD Instinct accelerators
-===========================================
-
-The prebuilt Megatron-LM with ROCm training environment allows users to quickly validate
-system performance, conduct training benchmarks, and achieve superior
-performance for models like Llama 3.1 and Llama 2. This container should not be
-expected to provide generalized performance across all training workloads. You
-can expect the container to perform in the model configurations described in
-the following section, but other configurations are not validated by AMD.
-
-Use the following instructions to set up the environment, configure the script
-to train models, and reproduce the benchmark results on MI300X series
-accelerators with the AMD Megatron-LM Docker image.
-
-.. tab-set::
-
-   .. tab-item:: Llama
-      :sync: llama
-
-      .. tab-set::
-
-         .. tab-item:: Single node training
-            :sync: single-node
-
-            To run training on a single node, navigate to the Megatron-LM folder and use the
-            following command:
-
-            .. code-block:: shell
-
-               TEE_OUTPUT=1 MBS=2 BS=128 TP=1 TE_FP8=1 SEQ_LENGTH=8192 MODEL_SIZE=8 bash examples/llama/train_llama3.sh
-
-         .. tab-item:: Multi-node training
-            :sync: multi-node
-
-            To run training on multiple nodes, launch the Docker container on each node. For example, for a two node setup (``NODE0`` as the master node), use these commands.
-
-            * On the master node ``NODE0``:
-
-              .. code-block:: shell
-
-                 TEE_OUTPUT=1 MBS=2 BS=256 TP=1 TE_FP8=1 SEQ_LENGTH=8192 MODEL_SIZE=8 MASTER_ADDR=IP_NODE0 NNODES=2 NODE_RANK=0 bash examples/llama/train_llama3.sh
-
-            * On the worker node ``NODE1``:
-
-              .. code-block:: shell
-
-                 TEE_OUTPUT=1 MBS=2 BS=256 TP=1 TE_FP8=1 SEQ_LENGTH=8192 MODEL_SIZE=8 MASTER_ADDR=IP_NODE0 NNODES=2 NODE_RANK=1 bash examples/llama/train_llama3.sh
-
-
-   .. tab-item:: DeepSeek V2
-      :sync: deepseek
-
-      To run the training on a single node, go to ``/Megatron-LM`` folder and use the following command:
-
-      .. code-block:: shell
-
-         cd /workspace/Megatron-LM
-         GEMM_TUNING=1 PR=bf16 MBS=4 AC=none bash examples/deepseek_v2/train_deepseekv2.sh
-
-Key options
-----------
-
-.. _amd-megatron-lm-benchmark-test-vars:
-
-The benchmark tests support the following sets of variables:
-
-.. tab-set::
-
-   .. tab-item:: Llama
-      :sync: llama
-
-      ``TEE_OUTPUT``
-        ``1`` to enable training logs or ``0`` to disable.
-
-      ``TE_FP8``
-        ``0`` for BP16 (default) or ``1`` for FP8 GEMMs.
-
-      ``GEMM_TUNING``
-        ``1`` to enable GEMM tuning, which boosts performance by using the best GEMM kernels.
-
-      ``USE_FLASH_ATTN``
-        ``1`` to enable Flash Attention.
-
-      ``ENABLE_PROFILING``
-        ``1`` to enable PyTorch profiling for performance analysis.
-
-      ``transformer-impl``
-        ``transformer_engine`` to use the Transformer Engine (TE) or ``local`` to disable TE.
-
-      ``MODEL_SIZE``
-        ``8B`` or ``70B`` for Llama 3 and 3.1. ``7B`` or ``70B`` for Llama 2.
-
-      ``TOTAL_ITERS``
-        The total number of iterations -- ``10`` by default.
-
-      ``MOCK_DATA``
-        ``1`` to use mock data or ``0`` to use real data provided by you.
-
-      ``MBS``
-        Micro batch size.
-
-      ``BS``
-        Global batch size.
-
-      ``TP``
-        Tensor parallel (``1``, ``2``, ``4``, ``8``).
-
-      ``SEQ_LENGTH``
-        Input sequence length.
-
-   .. tab-item:: DeepSeek V2
-      :sync: deepseek
-
-      ``PR``
-        Precision for training. ``bf16`` for BF16 (default) or ``fp8`` for FP8 GEMMs.
-
-      ``GEMM_TUNING``
-        ``1`` to enable GEMM tuning, which boosts performance by using the best GEMM kernels.
-
-      ``TOTAL_ITERS``
-        The total number of iterations -- ``10`` by default.
-
-      ``MOCK_DATA``
-        ``1`` to use mock data or ``0`` to use real data provided by you.
-
-      ``MBS``
-        Micro batch size.
-
-      ``GBS``
-        Global batch size.
-
-Benchmarking examples
---------------------
-
-.. tab-set::
-
-   .. tab-item:: Llama
-      :sync: llama
-
-      .. tab-set::
-
-         .. tab-item:: Single node training
-            :sync: single-node
-
-            Use this command to run training with Llama 2 7B model on a single node. You can specify MBS, BS, FP,
-            datatype, and so on.
-
-            .. code-block:: bash
-
-               TEE_OUTPUT=1 MBS=5 BS=120 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
-               SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
-
-            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
-
-            See the sample output:
-
-            .. image:: ../../../../data/how-to/rocm-for-ai/llama2-7b-training-log-sample.png
-               :width: 800
-
-         .. tab-item:: Multi-node training
-            :sync: multi-node
-
-            Launch the Docker container on each node.
-
-            In this example, run training with Llama 2 7B model on 2 nodes with specific MBS, BS, FP, datatype, and
-            so on.
-
-            On the master node:
-
-            .. code-block:: bash
-
-               TEE_OUTPUT=1 MBS=4 BS=64 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
-               SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
-
-            On the worker node:
-
-            .. code-block:: bash
-
-               TEE_OUTPUT=1 MBS=4 BS=64 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
-               SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
-
-            You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
-
-            Sample output for 2-node training:
-
-            Master node:
-
-            .. image:: ../../../../data/how-to/rocm-for-ai/2-node-training-master.png
-               :width: 800
-
-            Worker node:
-
-            .. image:: ../../../../data/how-to/rocm-for-ai/2-node-training-worker.png
-               :width: 800
-
-Previous versions
-=================
-
-This table lists previous versions of the ROCm Megatron-LM Docker image for training
-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
-     - Megatron-LM version
-     - PyTorch version
-     - Resources
-
-   * - 6.1
-     - 24.12-dev
-     - 2.4.0
-     - 
-       * `Documentation <https://rocm.docs.amd.com/en/docs-6.3.0/how-to/rocm-for-ai/train-a-model.html>`_
-       * `Docker Hub <https://hub.docker.com/layers/rocm/megatron-lm/24.12-dev/images/sha256-5818c50334ce3d69deeeb8f589d83ec29003817da34158ebc9e2d112b929bf2e>`_
--- a/docs/how-to/rocm-for-ai/training/benchmark-docker/pytorch-training.rst
+++ b/docs/how-to/rocm-for-ai/training/benchmark-docker/pytorch-training.rst
@@ -1,341 +0,0 @@
-:orphan:
-
-.. meta::
-   :description: How to train a model using PyTorch for ROCm.
-   :keywords: ROCm, AI, LLM, train, PyTorch, torch, Llama, flux, tutorial, docker
-
-**************************************
-Training a model with PyTorch for ROCm
-**************************************
-
-PyTorch is an open-source machine learning framework that is widely used for
-model training with GPU-optimized components for transformer-based models.
-
-The PyTorch for ROCm training Docker (``rocm/pytorch-training:v25.3``) image
-provides a prebuilt optimized environment for fine-tuning and pretraining a
-model on AMD Instinct MI325X and MI300X accelerators. It includes the following
-software components to accelerate training workloads:
-
-+--------------------------+--------------------------------+
-| Software component       | Version                        |
-+==========================+================================+
-| ROCm                     | 6.3.0                          |
-+--------------------------+--------------------------------+
-| PyTorch                  | 2.7.0a0+git637433              |
-+--------------------------+--------------------------------+
-| Python                   | 3.10                           |
-+--------------------------+--------------------------------+
-| Transformer Engine       | 1.11                           |
-+--------------------------+--------------------------------+
-| Flash Attention          | 3.0.0                          |
-+--------------------------+--------------------------------+
-| hipBLASLt                | git258a2162                    |
-+--------------------------+--------------------------------+
-| Triton                   | 3.1                            |
-+--------------------------+--------------------------------+
-
-.. _amd-pytorch-training-model-support:
-
-Supported models
-================
-
-The following models are pre-optimized for performance on the AMD Instinct MI300X accelerator.
-
-* Llama 3.1 8B
-
-* Llama 3.1 70B
-
-* FLUX.1-dev
-
-.. note::
-
-   Only these models are supported in the following steps.
-
-   Some models, such as Llama 3, require an external license agreement through
-   a third party (for example, Meta).
-
-System validation
-=================
-
-If you have already validated your system settings, skip this step. Otherwise,
-complete the :ref:`system validation and optimization steps <train-a-model-system-validation>`
-to set up your system before starting training.
-
-Disable NUMA auto-balancing
---------------------------
-
-Generally, application performance can benefit from disabling NUMA auto-balancing. However,
-it might be detrimental to performance with certain types of workloads.
-
-Run the command ``cat /proc/sys/kernel/numa_balancing`` to check your current NUMA (Non-Uniform
-Memory Access) settings. Output ``0`` indicates this setting is disabled. If there is no output or
-the output is ``1``, run the following command to disable NUMA auto-balancing.
-
-.. code-block:: shell
-
-   sudo sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
-
-See :ref:`mi300x-disable-numa` for more information.
-
-Environment setup
-=================
-
-This Docker image is optimized for specific model configurations outlined
-below. Performance can vary for other training workloads, as AMD 
-doesn’t validate configurations and run conditions outside those described.
-
-Download the Docker image
-------------------------
-
-1. Use the following command to pull the Docker image from Docker Hub.
-
-   .. code-block:: shell
-
-      docker pull rocm/pytorch-training:v25.3
-
-2. Run the Docker container.
-
-   .. code-block:: shell
-
-      docker run -it --device /dev/dri --device /dev/kfd --network host --ipc host --group-add video --cap-add SYS_PTRACE --security-opt seccomp=unconfined --privileged -v $HOME:$HOME -v  $HOME/.ssh:/root/.ssh --shm-size 64G --name training_env rocm/pytorch-training:v25.3
-
-3. Use these commands if you exit the ``training_env`` container and need to return to it.
-
-   .. code-block:: shell
-
-      docker start training_env
-      docker exec -it training_env bash
-
-4. In the Docker container, clone the `<https://github.com/ROCm/MAD>`__ repository and navigate to the benchmark scripts directory.
-
-   .. code-block:: shell
-
-      git clone https://github.com/ROCm/MAD
-      cd MAD/scripts/pytorch-train
-
-Prepare training datasets and dependencies
------------------------------------------
-
-The following benchmarking examples may require downloading models and datasets
-from Hugging Face. To ensure successful access to gated repos, set your
-``HF_TOKEN``.
-
-Run the setup script to install libraries and datasets needed for benchmarking.
-
-.. code-block:: shell
-
-   ./pytorch_benchmark_setup.sh
-
-``pytorch_benchmark_setup.sh`` installs the following libraries:
-
-.. list-table::
-   :header-rows: 1
-
-   * - Library
-     - Benchmark model
-     - Reference
-
-   * - ``accelerate``
-     - Llama 3.1 8B, FLUX
-     - `Hugging Face Accelerate <https://huggingface.co/docs/accelerate/en/index>`_
-
-   * - ``datasets``
-     - Llama 3.1 8B, 70B, FLUX
-     - `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
-
-   * - ``torchdata``
-     - Llama 3.1 70B
-     - `TorchData <https://pytorch.org/data/beta/index.html>`_
-
-   * - ``tomli``
-     - Llama 3.1 70B
-     - `Tomli <https://pypi.org/project/tomli/>`_
-
-   * - ``tiktoken``
-     - Llama 3.1 70B
-     - `tiktoken <https://github.com/openai/tiktoken>`_
-
-   * - ``blobfile``
-     - Llama 3.1 70B
-     - `blobfile <https://pypi.org/project/blobfile/>`_
-
-   * - ``tabulate``
-     - Llama 3.1 70B
-     - `tabulate <https://pypi.org/project/tabulate/>`_
-
-   * - ``wandb``
-     - Llama 3.1 70B
-     - `Weights & Biases <https://github.com/wandb/wandb>`_
-
-   * - ``sentencepiece``
-     - Llama 3.1 70B, FLUX
-     - `SentencePiece <https://github.com/google/sentencepiece>`_ 0.2.0
-
-   * - ``tensorboard``
-     - Llama 3.1 70 B, FLUX
-     - `TensorBoard <https://www.tensorflow.org/tensorboard>`_ 2.18.0
-
-   * - ``csvkit``
-     - FLUX
-     - `csvkit <https://csvkit.readthedocs.io/en/latest/>`_ 2.0.1
-
-   * - ``deepspeed``
-     - FLUX
-     - `DeepSpeed <https://github.com/deepspeedai/DeepSpeed>`_ 0.16.2
-
-   * - ``diffusers``
-     - FLUX
-     - `Hugging Face Diffusers <https://huggingface.co/docs/diffusers/en/index>`_ 0.31.0
-
-   * - ``GitPython``
-     - FLUX
-     - `GitPython <https://github.com/gitpython-developers/GitPython>`_ 3.1.44
-
-   * - ``opencv-python-headless``
-     - FLUX
-     - `opencv-python-headless <https://pypi.org/project/opencv-python-headless/>`_ 4.10.0.84
-
-   * - ``peft``
-     - FLUX
-     - `PEFT <https://huggingface.co/docs/peft/en/index>`_ 0.14.0
-
-   * - ``protobuf``
-     - FLUX
-     - `Protocol Buffers <https://github.com/protocolbuffers/protobuf>`_ 5.29.2
-
-   * - ``pytest``
-     - FLUX
-     - `PyTest <https://docs.pytest.org/en/stable/>`_ 8.3.4
-
-   * - ``python-dotenv``
-     - FLUX
-     - `python-dotenv <https://pypi.org/project/python-dotenv/>`_ 1.0.1
-
-   * - ``seaborn``
-     - FLUX
-     - `Seaborn <https://seaborn.pydata.org/>`_ 0.13.2
-
-   * - ``transformers``
-     - FLUX
-     - `Transformers <https://huggingface.co/docs/transformers/en/index>`_ 4.47.0
-
-``pytorch_benchmark_setup.sh`` downloads the following models from Hugging Face:
-
-* `meta-llama/Llama-3.1-70B-Instruct <https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct>`_
-
-* `black-forest-labs/FLUX.1-dev <https://huggingface.co/black-forest-labs/FLUX.1-dev>`_
-
-Along with the following datasets:
-
-* `WikiText <https://huggingface.co/datasets/Salesforce/wikitext>`_
-
-* `bghira/pseudo-camera-10k <https://huggingface.co/datasets/bghira/pseudo-camera-10k>`_
-
-Start training on AMD Instinct accelerators
-===========================================
-
-The prebuilt PyTorch with ROCm training environment allows users to quickly validate
-system performance, conduct training benchmarks, and achieve superior
-performance for models like Llama 3.1 and Llama 2. This container should not be
-expected to provide generalized performance across all training workloads. You
-can expect the container to perform in the model configurations described in
-the following section, but other configurations are not validated by AMD.
-
-Use the following instructions to set up the environment, configure the script
-to train models, and reproduce the benchmark results on MI300X series
-accelerators with the AMD PyTorch training Docker image.
-
-Once your environment is set up, use the following commands and examples to start benchmarking.
-
-Pretraining
-----------
-
-To start the pretraining benchmark, use the following command with the
-appropriate options. See the following list of options and their descriptions.
-
-.. code-block:: shell
-
-   ./pytorch_benchmark_report.sh -t $training_mode -m $model_repo -p $datatype -s $sequence_length
-
-Options and available models
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-.. list-table::
-   :header-rows: 1
-
-   * - Name
-     - Options
-     - Description
-
-   * - ``$training_mode``
-     - ``pretrain``
-     - Benchmark pretraining
-
-   * -
-     - ``finetune_fw``
-     - Benchmark full weight fine-tuning (Llama 3.1 70B with BF16)
-
-   * -
-     - ``finetune_lora``
-     - Benchmark LoRA fine-tuning (Llama 3.1 70B with BF16)
-
-   * - ``$datatype``
-     - FP8 or BF16
-     - Only Llama 3.1 8B supports FP8 precision.
-
-   * - ``$model_repo``
-     - Llama-3.1-8B
-     - `Llama 3.1 8B <https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct>`_
-
-   * - 
-     - Llama-3.1-70B
-     - `Llama 3.1 70B <https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct>`_
-
-   * - 
-     - Flux
-     - `FLUX.1 [dev] <https://huggingface.co/black-forest-labs/FLUX.1-dev>`_
-
-Fine-tuning
-----------
-
-To start the fine-tuning benchmark, use the following command. It will run the benchmarking example of Llama 2 70B
-with the WikiText dataset using the AMD fork of `torchtune <https://github.com/AMD-AIG-AIMA/torchtune>`_.
-
-.. code-block:: shell
-
-   ./pytorch_benchmark_report.sh -t {finetune_fw, finetune_lora} -p BF16 -m Llama-3.1-70B
-
-Benchmarking examples
---------------------
-
-Here are some examples of how to use the command.
-
-* Example 1: Llama 3.1 70B with BF16 precision with `torchtitan <https://github.com/ROCm/torchtitan>`_.
-
-  .. code-block:: shell
-
-     ./pytorch_benchmark_report.sh -t pretrain -p BF16 -m Llama-3.1-70B -s 8192
-
-* Example 2: Llama 3.1 8B with FP8 precision using Transformer Engine (TE) and Hugging Face Accelerator.
-
-  .. code-block:: shell
-
-     ./pytorch_benchmark_report.sh -t pretrain -p FP8 -m Llama-3.1-70B -s 8192
-
-* Example 3: FLUX.1-dev with BF16 precision with FluxBenchmark.
-
-  .. code-block:: shell
-
-     ./pytorch_benchmark_report.sh -t pretrain -p BF16 -m Flux
-
-* Example 4: Torchtune full weight fine-tuning with Llama 3.1 70B
-
-  .. code-block:: shell
-
-     ./pytorch_benchmark_report.sh -t finetune_fw -p BF16 -m Llama-3.1-70B
-
-* Example 5: Torchtune LoRA fine-tuning with Llama 3.1 70B
-
-  .. code-block:: shell
-
-     ./pytorch_benchmark_report.sh -t finetune_lora -p BF16 -m Llama-3.1-70B
--- a/docs/how-to/rocm-for-ai/training/index.rst
+++ b/docs/how-to/rocm-for-ai/training/index.rst
@@ -19,10 +19,6 @@ training, fine-tuning, and inference. It leverages popular machine learning fram

 In this guide, you'll learn about:

- Training a model
+- :doc:`Training a model <train-a-model>`

-  - :doc:`Train a model with Megatron-LM <benchmark-docker/megatron-lm>`
-
-  - :doc:`Train a model with PyTorch <benchmark-docker/pytorch-training>`
-
- :doc:`Scaling model training <scale-model-training>`
+- :doc:`Scale model training <scale-model-training>`
--- a/docs/how-to/rocm-for-ai/training/prerequisite-system-validation.rst
+++ b/docs/how-to/rocm-for-ai/training/prerequisite-system-validation.rst
@@ -1,130 +0,0 @@
-:orphan:
-
-.. meta::
-   :description: Prerequisite system validation before using ROCm for AI.
-   :keywords: ROCm, AI, LLM, train, megatron, Llama, tutorial, docker, torch, pytorch, jax
-
-.. _train-a-model-system-validation:
-
-**********************************************
-Prerequisite system validation before training
-**********************************************
-
-Complete the following system validation and optimization steps to set up your system before starting training.
-
-Disable NUMA auto-balancing
---------------------------
-
-Generally, application performance can benefit from disabling NUMA auto-balancing. However,
-it might be detrimental to performance with certain types of workloads.
-
-Run the command ``cat /proc/sys/kernel/numa_balancing`` to check your current NUMA (Non-Uniform
-Memory Access) settings. Output ``0`` indicates this setting is disabled. If there is no output or
-the output is ``1``, run the following command to disable NUMA auto-balancing.
-
-.. code-block:: shell
-
-   sudo sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
-
-See :ref:`mi300x-disable-numa` for more information.
-
-Hardware verification with ROCm
-------------------------------
-
-Use the command ``rocm-smi --setperfdeterminism 1900`` to set the max clock speed up to 1900 MHz
-instead of the default 2100 MHz. This can reduce the chance of a PCC event lowering the attainable
-GPU clocks. This setting will not be required for new IFWI releases with the production PRC feature.
-You can restore this setting to its default value with the ``rocm-smi -r`` command.
-
-Run the command:
-
-.. code-block:: shell
-
-   rocm-smi --setperfdeterminism 1900
-
-See :ref:`mi300x-hardware-verification-with-rocm` for more information.
-
-RCCL Bandwidth Test for multi-node setups
-----------------------------------------
-
-ROCm Collective Communications Library (RCCL) is a standalone library of standard collective communication
-routines for GPUs. See the :doc:`RCCL documentation <rccl:index>` for more information. Before starting
-pretraining, running a RCCL bandwidth test helps ensure that the multi-GPU or multi-node setup is optimized
-for efficient distributed training.
-
-Running the RCCL bandwidth test helps verify that:
-
- The GPUs can communicate across nodes or within a single node.
-
- The interconnect (such as InfiniBand, Ethernet, or Infinite fabric) is functioning as expected and
-  provides adequate bandwidth for communication.
-
- No hardware setup or cabling issues could affect the communication between GPUs
-
-Tuning and optimizing hyperparameters
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-In distributed training, specific hyperparameters related to distributed communication can be tuned based on
-the results of the RCCL bandwidth test. These variables are already set in the Docker image:
-
-.. code-block:: shell
-
-   # force all RCCL streams to be high priority
-   export TORCH_NCCL_HIGH_PRIORITY=1
-
-   # specify which RDMA interfaces to use for communication
-   export NCCL_IB_HCA=rdma0,rdma1,rdma2,rdma3,rdma4,rdma5,rdma6,rdma7
-
-   # define the Global ID index used in RoCE mode
-   export NCCL_IB_GID_INDEX=3
-
-   # avoid data corruption/mismatch issue that existed in past releases
-   export RCCL_MSCCL_ENABLE=0
-
-Running the RCCL Bandwidth Test
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-It's recommended you run the RCCL bandwidth test before launching training. It ensures system
-performance is sufficient to launch training. RCCL is not included in the AMD Megatron-LM Docker
-image; follow the instructions in `<https://github.com/ROCm/rccl-tests>`__ to get started.
-See :ref:`mi300x-rccl` for more information.
-
-Run on 8 GPUs (``-g 8``), scanning from 8 bytes to 10 GB:
-
-.. code-block:: shell
-
-   ./build/all_reduce_perf -b 8 -e 10G -f 2 -g 8
-
-.. image:: ../../../data/how-to/rocm-for-ai/rccl-tests-8-gpu.png
-   :width: 800
-
-Using one MPI process per GPU and ``-g 1`` for performance-oriented runs on both single-node and multi-node is
-recommended. So, a run on 8 GPUs looks something like:
-
-.. code-block:: shell
-
-   mpirun -np 8 --bind-to numa ./build/all_reduce_perf -b 8 -e 10G -f 2 -g 1
-
-.. image:: ../../../data/how-to/rocm-for-ai/rccl-tests-1-mpi-process-per-gpu.png
-   :width: 800
-
-Running with one MPI process per GPU ensures a one-to-one mapping for CPUs and GPUs, which can be beneficial
-for smaller message sizes. This better represents the real-world use of RCCL in deep learning frameworks like
-PyTorch and TensorFlow.
-
-Use the following script to run the RCCL test for four MI300X GPU nodes. Modify paths and node addresses as needed.
-
-.. code-block::
-
-   /home/$USER/ompi_for_gpu/ompi/bin/mpirun -np 32 -H tw022:8,tw024:8,tw010:8, tw015:8 \
-   --mca pml ucx \
-   --mca btl ^openib \
-   -x NCCL_SOCKET_IFNAME=ens50f0np0 \
-   -x NCCL_IB_HCA=rdma0:1,rdma1:1,rdma2:1,rdma3:1,rdma4:1,rdma5:1,rdma6:1,rdma7:1 \
-   -x NCCL_IB_GID_INDEX=3 \
-   -x NCCL_MIN_NCHANNELS=40 \
-   -x NCCL_DEBUG=version \
-   $HOME/rccl-tests/build/all_reduce_perf -b 8 -e 8g -f 2 -g 1
-
-.. image:: ../../../data/how-to/rocm-for-ai/rccl-tests-4-mi300x-gpu-nodes.png
-   :width: 800
--- a/docs/how-to/rocm-for-ai/training/train-a-model.rst
+++ b/docs/how-to/rocm-for-ai/training/train-a-model.rst
@@ -0,0 +1,503 @@
+.. meta::
+   :description: How to train a model using ROCm Megatron-LM
+   :keywords: ROCm, AI, LLM, train, Megatron-LM, megatron, Llama, tutorial, docker, torch
+
+**************************************
+Training a model with ROCm Megatron-LM
+**************************************
+
+.. _amd-megatron-lm:
+
+The ROCm Megatron-LM framework is a specialized fork of the robust Megatron-LM, designed to
+enable efficient training of large-scale language models on AMD GPUs. By leveraging AMD Instinct™ MI300X
+accelerators, AMD Megatron-LM delivers enhanced scalability, performance, and resource utilization for AI
+workloads. It is purpose-built to :ref:`support models <amd-megatron-lm-model-support>`
+like Meta's Llama 2, Llama 3, and Llama 3.1, enabling developers to train next-generation AI models with greater
+efficiency. See the GitHub repository at `<https://github.com/ROCm/Megatron-LM>`__.
+
+For ease of use, AMD provides a ready-to-use Docker image for MI300X accelerators containing essential
+components, including PyTorch, PyTorch Lightning, ROCm libraries, and Megatron-LM utilities. It contains the
+following software to accelerate training workloads:
+
+--------------------------+--------------------------------+
+| Software component       | Version                        |
+==========================+================================+
+| ROCm                     | 6.1                            |
+--------------------------+--------------------------------+
+| PyTorch                  | 2.4.0                          |
+--------------------------+--------------------------------+
+| PyTorch Lightning        | 2.4.0                          |
+--------------------------+--------------------------------+
+| Megatron Core            | 0.9.0                          |
+--------------------------+--------------------------------+
+| Transformer Engine       | 1.5.0                          |
+--------------------------+--------------------------------+
+| Flash Attention          | v2.6                           |
+--------------------------+--------------------------------+
+| Transformers             | 4.44.0                         |
+--------------------------+--------------------------------+
+
+Supported features and models
+=============================
+
+Megatron-LM provides the following key features to train large language models efficiently:
+
+- Transformer Engine (TE)
+
+- APEX
+
+- GEMM tuning
+
+- Torch.compile
+
+- 3D parallelism: TP + SP + CP
+
+- Distributed optimizer
+
+- Flash Attention (FA) 2
+
+- Fused kernels
+
+- Pre-training
+
+.. _amd-megatron-lm-model-support:
+
+The following models are pre-optimized for performance on the AMD Instinct MI300X accelerator.
+
+* Llama 2 7B
+
+* Llama 2 70B
+
+* Llama 3 8B
+
+* Llama 3 70B
+
+* Llama 3.1 8B
+
+* Llama 3.1 70B
+
+Prerequisite system validation steps
+====================================
+
+Complete the following system validation and optimization steps to set up your system before starting training.
+
+Disable NUMA auto-balancing
+---------------------------
+
+Generally, application performance can benefit from disabling NUMA auto-balancing. However,
+it might be detrimental to performance with certain types of workloads.
+
+Run the command ``cat /proc/sys/kernel/numa_balancing`` to check your current NUMA (Non-Uniform
+Memory Access) settings. Output ``0`` indicates this setting is disabled. If there is no output or
+the output is ``1``, run the following command to disable NUMA auto-balancing.
+
+.. code-block:: shell
+
+   sudo sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
+
+See :ref:`mi300x-disable-numa` for more information.
+
+Hardware verification with ROCm
+-------------------------------
+
+Use the command ``rocm-smi --setperfdeterminism 1900`` to set the max clock speed up to 1900 MHz
+instead of the default 2100 MHz. This can reduce the chance of a PCC event lowering the attainable
+GPU clocks. This setting will not be required for new IFWI releases with the production PRC feature.
+You can restore this setting to its default value with the ``rocm-smi -r`` command.
+
+Run the command:
+
+.. code-block:: shell
+
+   rocm-smi --setperfdeterminism 1900
+
+See :ref:`mi300x-hardware-verification-with-rocm` for more information.
+
+RCCL Bandwidth Test
+-------------------
+
+ROCm Collective Communications Library (RCCL) is a standalone library of standard collective communication
+routines for GPUs. See the :doc:`RCCL documentation <rccl:index>` for more information. Before starting
+pre-training, running a RCCL bandwidth test helps ensure that the multi-GPU or multi-node setup is optimized
+for efficient distributed training.
+
+Running the RCCL bandwidth test helps verify that:
+
+- The GPUs can communicate across nodes or within a single node.
+
+- The interconnect (such as InfiniBand, Ethernet, or Infinite fabric) is functioning as expected and
+  provides adequate bandwidth for communication.
+
+- No hardware setup or cabling issues could affect the communication between GPUs
+
+Tuning and optimizing hyperparameters
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In distributed training, specific hyperparameters related to distributed communication can be tuned based on
+the results of the RCCL bandwidth test. These variables are already set in the Docker image:
+
+.. code-block:: shell
+
+   # force all RCCL streams to be high priority
+   export TORCH_NCCL_HIGH_PRIORITY=1
+
+   # specify which RDMA interfaces to use for communication
+   export NCCL_IB_HCA=rdma0,rdma1,rdma2,rdma3,rdma4,rdma5,rdma6,rdma7
+
+   # define the Global ID index used in RoCE mode
+   export NCCL_IB_GID_INDEX=3
+
+   # avoid data corruption/mismatch issue that existed in past releases
+   export RCCL_MSCCL_ENABLE=0
+
+Running the RCCL Bandwidth Test
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+It's recommended you run the RCCL bandwidth test before launching training. It ensures system
+performance is sufficient to launch training. RCCL is not included in the AMD Megatron-LM Docker
+image; follow the instructions in `<https://github.com/ROCm/rccl-tests>`__ to get started.
+See :ref:`mi300x-rccl` for more information.
+
+Run on 8 GPUs (``-g 8``), scanning from 8 bytes to 10 GB:
+
+.. code-block:: shell
+
+   ./build/all_reduce_perf -b 8 -e 10G -f 2 -g 8
+
+.. image:: ../../../data/how-to/rocm-for-ai/rccl-tests-8-gpu.png
+   :width: 800
+
+Using one MPI process per GPU and ``-g 1`` for performance-oriented runs on both single-node and multi-node is
+recommended. So, a run on 8 GPUs looks something like:
+
+.. code-block:: shell
+
+   mpirun -np 8 --bind-to numa ./build/all_reduce_perf -b 8 -e 10G -f 2 -g 1
+
+.. image:: ../../../data/how-to/rocm-for-ai/rccl-tests-1-mpi-process-per-gpu.png
+   :width: 800
+
+Running with one MPI process per GPU ensures a one-to-one mapping for CPUs and GPUs, which can be beneficial
+for smaller message sizes. This better represents the real-world use of RCCL in deep learning frameworks like
+PyTorch and TensorFlow.
+
+Use the following script to run the RCCL test for four MI300X GPU nodes. Modify paths and node addresses as needed.
+
+.. code-block::
+
+   /home/$USER/ompi_for_gpu/ompi/bin/mpirun -np 32 -H tw022:8,tw024:8,tw010:8, tw015:8 \
+   --mca pml ucx \
+   --mca btl ^openib \
+   -x NCCL_SOCKET_IFNAME=ens50f0np0 \
+   -x NCCL_IB_HCA=rdma0:1,rdma1:1,rdma2:1,rdma3:1,rdma4:1,rdma5:1,rdma6:1,rdma7:1 \
+   -x NCCL_IB_GID_INDEX=3 \
+   -x NCCL_MIN_NCHANNELS=40 \
+   -x NCCL_DEBUG=version \
+   $HOME/rccl-tests/build/all_reduce_perf -b 8 -e 8g -f 2 -g 1
+
+.. image:: ../../../data/how-to/rocm-for-ai/rccl-tests-4-mi300x-gpu-nodes.png
+   :width: 800
+
+.. _mi300x-amd-megatron-lm-training:
+
+Start training on MI300X accelerators
+=====================================
+
+The pre-built ROCm Megatron-LM environment allows users to quickly validate system performance, conduct
+training benchmarks, and achieve superior performance for models like Llama 2 and Llama 3.1.
+
+Use the following instructions to set up the environment, configure the script to train models, and
+reproduce the benchmark results on the MI300X accelerators with the AMD Megatron-LM Docker
+image.
+
+.. _amd-megatron-lm-requirements:
+
+Download the Docker image and required packages
+-----------------------------------------------
+
+1. Use the following command to pull the Docker image from Docker Hub.
+
+   .. code-block:: shell
+
+      docker pull rocm/megatron-lm:24.12-dev
+
+2. Launch the Docker container.
+
+   .. code-block:: shell
+
+      docker run -it --device /dev/dri --device /dev/kfd --network host --ipc host --group-add video --cap-add SYS_PTRACE --security-opt seccomp=unconfined --privileged -v $CACHE_DIR:/root/.cache --name megatron-dev-env rocm/megatron-lm:24.12-dev /bin/bash
+
+3. Clone the ROCm Megatron-LM repository to a local directory and install the required packages on the host machine.
+
+   .. code-block:: shell
+
+      git clone https://github.com/ROCm/Megatron-LM
+      cd Megatron-LM
+
+   .. note::
+
+      This release is validated with ``ROCm/Megatron-LM`` commit `bb93ccb <https://github.com/ROCm/Megatron-LM/tree/bb93ccbfeae6363c67b361a97a27c74ab86e7e92>`_.
+      Checking out this specific commit is recommended for a stable and reproducible environment.
+
+      .. code-block:: shell
+         
+         git checkout bb93ccbfeae6363c67b361a97a27c74ab86e7e92
+
+Prepare training datasets
+-------------------------
+
+If you already have the preprocessed data, you can skip this section.
+
+Use the following command to process datasets. We use GPT data as an example. You may change the merge table, use an
+end-of-document token, remove sentence splitting, and use the tokenizer type.
+
+.. code-block:: shell
+
+   python tools/preprocess_data.py \
+       --input my-corpus.json \
+       --output-prefix my-gpt2 \
+       --vocab-file gpt2-vocab.json \
+       --tokenizer-type GPT2BPETokenizer \
+       --merge-file gpt2-merges.txt \
+       --append-eod
+
+In this case, the automatically generated output files are named ``my-gpt2_text_document.bin`` and
+``my-gpt2_text_document.idx``.
+
+.. image:: ../../../data/how-to/rocm-for-ai/prep-training-datasets-my-gpt2-text-document.png
+   :width: 800
+
+.. _amd-megatron-lm-environment-setup:
+
+Environment setup
+-----------------
+
+In the ``examples/llama`` directory of Megatron-LM, if you're working with Llama 2 7B or Llama 2 70 B, use the
+``train_llama2.sh`` configuration script. Likewise, if you're working with Llama 3 or Llama 3.1, then use
+``train_llama3.sh`` and update the configuration script accordingly.
+
+Network interface
+^^^^^^^^^^^^^^^^^
+
+To avoid connectivity issues, ensure the correct network interface is set in your training scripts.
+
+1. Run the following command to find the active network interface on your system.
+
+   .. code-block:: shell
+
+      ip a
+
+2. Update the ``NCCL_SOCKET_IFNAME`` and ``GLOO_SOCKET_IFNAME`` variables with your system’s network interface. For
+   example:
+
+   .. code-block:: shell
+
+      export NCCL_SOCKET_IFNAME=ens50f0np0
+
+      export GLOO_SOCKET_IFNAME=ens50f0np0
+
+Dataset options
+^^^^^^^^^^^^^^^
+
+You can use either mock data or real data for training.
+
+* If you're using a real dataset, update the ``DATA_PATH`` variable to point to the location of your dataset.
+
+  .. code-block:: shell
+
+     DATA_DIR="/root/.cache/data" # Change to where your dataset is stored
+
+     DATA_PATH=${DATA_DIR}/bookcorpus_text_sentence
+
+  .. code-block:: shell
+
+     --data-path $DATA_PATH
+
+  Ensure that the files are accessible inside the Docker container.
+
+* Mock data can be useful for testing and validation. If you're using mock data, replace ``--data-path $DATA_PATH`` with the ``--mock-data`` option.
+
+  .. code-block:: shell
+
+     --mock-data
+
+Tokenizer
+^^^^^^^^^
+
+Tokenization is the process of converting raw text into tokens that can be processed by the model. For Llama
+models, this typically involves sub-word tokenization, where words are broken down into smaller units based on
+a fixed vocabulary. The tokenizer is trained along with the model on a large corpus of text, and it learns a
+fixed vocabulary that can represent a wide range of text from different domains. This allows Llama models to
+handle a variety of input sequences, including unseen words or domain-specific terms.
+
+To train any of the Llama 2 models that this Docker image supports, use the ``Llama2Tokenizer``.
+
+To train any of Llama 3 and Llama 3.1 models that this Docker image supports, use the ``HuggingFaceTokenizer``.
+Set the Hugging Face model link in the ``TOKENIZER_MODEL`` variable.
+
+For example, if you're using the Llama 3.1 8B model:
+
+.. code-block:: shell
+
+   TOKENIZER_MODEL=meta-llama/Llama-3.1-8B
+
+Run benchmark tests
+-------------------
+
+.. note::
+
+   If you're running **multi node training**, update the following environment variables. They can
+   also be passed as command line arguments.
+
+   * Change ``localhost`` to the master node's hostname:
+
+     .. code-block:: shell
+
+        MASTER_ADDR="${MASTER_ADDR:-localhost}"
+
+   * Set the number of nodes you want to train on (for instance, ``2``, ``4``, ``8``):
+
+     .. code-block:: shell
+
+        NNODES="${NNODES:-1}"
+
+   * Set the rank of each node (0 for master, 1 for the first worker node, and so on):
+
+     .. code-block:: shell
+
+        NODE_RANK="${NODE_RANK:-0}"
+
+* Use this command to run a performance benchmark test of any of the Llama 2 models that this Docker image supports (see :ref:`variables <amd-megatron-lm-benchmark-test-vars>`).
+
+  .. code-block:: shell
+
+     {variables} bash examples/llama/train_llama2.sh
+
+* Use this command to run a performance benchmark test of any of the Llama 3 and Llama 3.1 models that this Docker image supports (see :ref:`variables <amd-megatron-lm-benchmark-test-vars>`).
+
+  .. code-block:: shell
+
+     {variables} bash examples/llama/train_llama3.sh
+
+.. _amd-megatron-lm-benchmark-test-vars:
+
+The benchmark tests support the same set of variables:
+
+--------------------------+-----------------------+-----------------------+
+| Name                     | Options               | Description           |
+==========================+=======================+=======================+
+| ``TEE_OUTPUT``           | 0 or 1                | 0: disable training   |
+|                          |                       | log                   |
+|                          |                       |                       |
+|                          |                       | 1: enable training    |
+|                          |                       | log                   |
+--------------------------+-----------------------+-----------------------+
+| ``MBS``                  |                       | Micro batch size      |
+--------------------------+-----------------------+-----------------------+
+| ``BS``                   |                       | Batch size            |
+--------------------------+-----------------------+-----------------------+
+| ``TP``                   | 1, 2, 4, 8            | Tensor parallel       |
+--------------------------+-----------------------+-----------------------+
+| ``TE_FP8``               | 0 or 1                | Datatype.             |
+|                          |                       | If it is set to 1,    |
+|                          |                       | FP8.                  |
+|                          |                       |                       |
+|                          |                       | If it is set to 0.    |
+|                          |                       | BP16                  |
+--------------------------+-----------------------+-----------------------+
+| ``NO_TORCH_COMPILE``     | 0 or 1                | If it is set to 1,    |
+|                          |                       | enable torch.compile. |
+|                          |                       |                       |
+|                          |                       | If it is set to 0.    |
+|                          |                       | Disable torch.compile |
+|                          |                       | (default)             |
+--------------------------+-----------------------+-----------------------+
+| ``SEQ_LENGTH``           |                       | Input sequence length |
+--------------------------+-----------------------+-----------------------+
+| ``GEMM_TUNING``          | 0 or 1                | If it is set to 1,    |
+|                          |                       | enable gemm tuning.   |
+|                          |                       |                       |
+|                          |                       | If it is set to 0,    |
+|                          |                       | disable gemm tuning   |
+--------------------------+-----------------------+-----------------------+
+| ``USE_FLASH_ATTN``       | 0 or 1                | 0: disable flash      |
+|                          |                       | attention             |
+|                          |                       |                       |
+|                          |                       | 1: enable flash       |
+|                          |                       | attention             |
+--------------------------+-----------------------+-----------------------+
+| ``ENABLE_PROFILING``     | 0 or 1                | 0: disable torch      |
+|                          |                       | profiling             |
+|                          |                       |                       |
+|                          |                       | 1: enable torch       |
+|                          |                       | profiling             |
+--------------------------+-----------------------+-----------------------+
+| ``MODEL_SIZE``           |                       | The size of the mode: |
+|                          |                       | 7B/70B, etc.          |
+--------------------------+-----------------------+-----------------------+
+| ``TOTAL_ITERS``          |                       | Total number of       |
+|                          |                       | iterations            |
+--------------------------+-----------------------+-----------------------+
+| ``transformer-impl``     | transformer_engine or | Enable transformer    |
+|                          | local                 | engine by default     |
+--------------------------+-----------------------+-----------------------+
+
+Benchmarking examples
+^^^^^^^^^^^^^^^^^^^^^
+
+.. tab-set::
+
+   .. tab-item:: Single node training
+      :sync: single
+
+      Use this command to run training with Llama 2 7B model on a single node. You can specify MBS, BS, FP,
+      datatype, and so on.
+
+      .. code-block:: bash
+
+         TEE_OUTPUT=1 MBS=5 BS=120 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
+         SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
+
+      You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
+
+      See the sample output:
+
+      .. image:: ../../../data/how-to/rocm-for-ai/llama2-7b-training-log-sample.png
+         :width: 800
+
+   .. tab-item:: Multi node training
+      :sync: multi
+
+      Launch the Docker container on each node.
+
+      In this example, run training with Llama 2 7B model on 2 nodes with specific MBS, BS, FP, datatype, and
+      so on.
+
+      On the master node:
+
+      .. code-block:: bash
+
+         TEE_OUTPUT=1 MBS=4 BS=64 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
+         SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
+
+      On the worker node:
+
+      .. code-block:: bash
+
+         TEE_OUTPUT=1 MBS=4 BS=64 TP=8 TE_FP8=0 NO_TORCH_COMPILE=1
+         SEQ_LENGTH=4096 bash examples/llama/train_llama2.sh
+
+      You can find the training logs at the location defined in ``$TRAIN_LOG`` in the :ref:`configuration script <amd-megatron-lm-environment-setup>`.
+
+      Sample output for 2-node training:
+
+      Master node:
+
+      .. image:: ../../../data/how-to/rocm-for-ai/2-node-training-master.png
+         :width: 800
+
+      Worker node:
+
+      .. image:: ../../../data/how-to/rocm-for-ai/2-node-training-worker.png
+         :width: 800
+
--- a/docs/how-to/system-optimization/mi300x.rst
+++ b/docs/how-to/system-optimization/mi300x.rst
@@ -308,24 +308,6 @@ Otherwise, if the system has Intel host CPUs add this instead to

   intel_iommu=on iommu=pt

-``modprobe.blacklist=amdgpu``
-  For some system configurations, the ``amdgpu`` driver needs to be blocked during kernel initialization to avoid an issue where after boot, the GPUs are not listed when running the command ``rocm-smi`` or ``amd-smi``.
-
-  Alternatively, configuring the AMD recommended system optimized BIOS settings might remove the need for using this setting. Some manufacturers and users might not implement the recommended system optimized BIOS settings.
-
-If you experience the mentioned issue, then add this to ``GRUB_CMDLINE_LINUX``:
-
-.. code-block:: text
-
-  modprobe.blacklist=amdgpu
-
-After the change, the ``amdgpu`` module must be loaded to support the ROCm framework
-software tools and utilities. Run the following command to load the ``amdgpu`` module:
-
-.. code-block:: text
-
-  sudo modprobe amdgpu
-
 Update GRUB
 -----------

--- a/docs/reference/gpu-arch-specs.rst
+++ b/docs/reference/gpu-arch-specs.rst
@@ -32,8 +32,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - L1 Instruction Cache (KiB)
          - VGPR File (KiB)
          - SGPR File (KiB)
-          - GFXIP Major version
-          - GFXIP Minor version
        *
          - MI325X
          - CDNA3
@@ -49,8 +47,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 64 per 2 CUs
          - 512
          - 12.5
-          - 9
-          - 4
        *
          - MI300X
          - CDNA3
@@ -66,8 +62,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 64 per 2 CUs
          - 512
          - 12.5
-          - 9
-          - 4
        *
          - MI300A
          - CDNA3
@@ -83,8 +77,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 64 per 2 CUs
          - 512
          - 12.5
-          - 9
-          - 4
        *
          - MI250X
          - CDNA2
@@ -100,8 +92,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 2 CUs
          - 512
          - 12.5
-          - 9
-          - 0
        *
          - MI250
          - CDNA2
@@ -117,8 +107,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 2 CUs
          - 512
          - 12.5
-          - 9
-          - 0
        *
          - MI210
          - CDNA2
@@ -134,8 +122,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 2 CUs
          - 512
          - 12.5
-          - 9
-          - 0
        *
          - MI100
          - CDNA
@@ -151,8 +137,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 3 CUs
          - 256 VGPR and 256 AccVGPR
          - 12.5
-          - 9
-          - 0
        *
          - MI60
          - GCN5.1
@@ -168,8 +152,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 3 CUs
          - 256
          - 12.5
-          - 9
-          - 0
        *
          - MI50 (32GB)
          - GCN5.1
@@ -185,8 +167,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 3 CUs
          - 256
          - 12.5
-          - 9
-          - 0
        *
          - MI50 (16GB)
          - GCN5.1
@@ -202,8 +182,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 3 CUs
          - 256
          - 12.5
-          - 9
-          - 0
        *
          - MI25
          - GCN5.0
@@ -219,8 +197,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 3 CUs
          - 256
          - 12.5
-          - 9
-          - 0
        *
          - MI8
          - GCN3.0
@@ -236,8 +212,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 4 CUs
          - 256
          - 12.5
-          - 8
-          - 0
        *
          - MI6
          - GCN4.0
@@ -253,8 +227,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 4 CUs
          - 256
          - 12.5
-          - 8
-          - 0

  .. tab-item:: AMD Radeon PRO GPUs

@@ -266,7 +238,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - Model
          - Architecture
          - LLVM target name
-
          - VRAM (GiB)
          - Compute Units
          - Wavefront Size
@@ -279,8 +250,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - L0 Instruction Cache (KiB)
          - VGPR File (KiB)
          - SGPR File (KiB)
-          - GFXIP Major version
-          - GFXIP Minor version
        *
          - Radeon PRO V710
          - RDNA3
@@ -297,8 +266,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon PRO W7900 Dual Slot
          - RDNA3
@@ -315,8 +282,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon PRO W7900
          - RDNA3
@@ -333,8 +298,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon PRO W7800
          - RDNA3
@@ -351,8 +314,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon PRO W7700
          - RDNA3
@@ -369,8 +330,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon PRO W6800
          - RDNA2
@@ -387,8 +346,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon PRO W6600
          - RDNA2
@@ -405,8 +362,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon PRO V620
          - RDNA2
@@ -423,8 +378,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon Pro W5500
          - RDNA
@@ -441,8 +394,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 20
-          - 10
-          - 1
        *
          - Radeon Pro VII
          - GCN5.1
@@ -459,8 +410,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 3 CUs
          - 256
          - 12.5
-          - 9
-          - 0

  .. tab-item:: AMD Radeon GPUs

@@ -484,8 +433,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - L0 Instruction Cache (KiB)
          - VGPR File (KiB)
          - SGPR File (KiB)
-          - GFXIP Major version
-          - GFXIP Minor version
        *
          - Radeon RX 7900 XTX
          - RDNA3
@@ -502,8 +449,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon RX 7900 XT
          - RDNA3
@@ -520,8 +465,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon RX 7900 GRE
          - RDNA3
@@ -538,8 +481,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon RX 7800 XT
          - RDNA3
@@ -556,8 +497,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon RX 7700 XT
          - RDNA3
@@ -574,8 +513,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 768
          - 16
-          - 11
-          - 0
        *
          - Radeon RX 7600
          - RDNA3
@@ -592,8 +529,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 11
-          - 0
        *
          - Radeon RX 6950 XT
          - RDNA2
@@ -610,8 +545,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon RX 6900 XT
          - RDNA2
@@ -628,8 +561,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon RX 6800 XT
          - RDNA2
@@ -646,8 +577,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon RX 6800
          - RDNA2
@@ -664,8 +593,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon RX 6750 XT
          - RDNA2
@@ -682,8 +609,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon RX 6700 XT
          - RDNA2
@@ -700,8 +625,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon RX 6700
          - RDNA2
@@ -718,8 +641,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon RX 6650 XT
          - RDNA2
@@ -736,8 +657,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon RX 6600 XT
          - RDNA2
@@ -754,8 +673,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon RX 6600
          - RDNA2
@@ -772,8 +689,6 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32
          - 512
          - 16
-          - 10
-          - 3
        *
          - Radeon VII
          - GCN5.1
@@ -790,14 +705,12 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
          - 32 per 3 CUs
          - 256
          - 12.5
-          - 9
-          - 0

 Glossary
 ========

 For more information about the terms used, see the
-:ref:`specific documents and guides <gpu-arch-documentation>`, or
+:ref:`specific documents and guides <gpu-arch-documentation>`, or 
 :doc:`Understanding the HIP programming model<hip:understand/programming_model>`.

 **LLVM target name**
@@ -887,26 +800,6 @@ Purpose Vector Registers, used specifically in matrix instructions.
 Size of the Scalar General Purpose Register (SGPR) file. Holds data used in
 scalar instructions.

-**GFXIP**
-
-GFXIP (Graphics IP) is a versioning system used by AMD to identify the GPU
-architecture and its instruction set. It helps categorize different generations
-of GPUs and their feature sets.
-
-**GFXIP major version**
-
-Defines the GPU's core instruction set and architecture, which determines
-compatibility with software stacks such as HIP and OpenCL. For example, a GFXIP
-11 major version corresponds to the RDNA 3 (Navi 3x) architecture, influencing
-driver support and available compute features.
-
-**GFXIP minor version**
-
-Represents specific variations within a GFXIP major version and affects feature sets,
-optimizations, and driver behavior in software stacks such as HIP and OpenCL. Different
-GPU models within the same major version can have unique capabilities, impacting
-performance and supported instructions.
-
 **GCD**

 Graphics Compute Die.
--- a/docs/reference/precision-support.rst
+++ b/docs/reference/precision-support.rst
@@ -9,14 +9,16 @@
 Data types and precision support
 *************************************************************

-This topic lists the data types support on AMD GPUs, ROCm libraries along
-with corresponding :doc:`HIP <hip:index>` data types.
+This topic lists the supported data types of AMD GPUs and ROCm libraries.
+Corresponding :doc:`HIP <hip:index>` data types are also noted. 

 Integral types
-==============
+==========================================

 The signed and unsigned integral types supported by ROCm are listed in
-the following table.
+the following table, along with their corresponding HIP type and a short
+description.
+

 .. list-table::
    :header-rows: 1
@@ -46,9 +48,10 @@ the following table.
 .. _precision_support_floating_point_types:

 Floating-point types
-====================
+==========================================

-The floating-point types supported by ROCm are listed in the following table.
+The floating-point types supported by ROCm are listed in the following
+table, along with their corresponding HIP type and a short description.

 .. image:: ../data/about/compatibility/floating-point-data-types.png
    :alt: Supported floating-point types
@@ -63,18 +66,18 @@ The floating-point types supported by ROCm are listed in the following table.
      - Description
    *
      - float8 (E4M3)
-      - ``__hip_fp8_e4m3_fnuz``
+      - ``-``
      - An 8-bit floating-point number that mostly follows IEEE-754 conventions
-        and **S1E4M3** bit layout, as described in `8-bit Numerical Formats for Deep Neural Networks <https://arxiv.org/abs/2206.02915>`_,
-        with expanded range and no infinity or signed zero. NaN is represented
-        as negative zero.
+        and **S1E4M3** bit layout, as described in `8-bit Numerical Formats for Deep Neural Networks <https://arxiv.org/abs/2206.02915>`_ ,
+        with expanded range and no infinity or signed zero. NaN is
+        represented as negative zero.
    *
      - float8 (E5M2)
-      - ``__hip_fp8_e5m2_fnuz``
+      - ``-``
      - An 8-bit floating-point number mostly following IEEE-754 conventions and
-        **S1E5M2** bit layout, as described in `8-bit Numerical Formats for Deep Neural Networks <https://arxiv.org/abs/2206.02915>`_,
-        with expanded range and no infinity or signed zero. NaN is represented
-        as negative zero.
+        **S1E5M2** bit layout, as described in `8-bit Numerical Formats for Deep Neural Networks <https://arxiv.org/abs/2206.02915>`_ ,
+        with expanded range and no infinity or signed zero. NaN is
+        represented as negative zero.
    *
      - float16
      - ``half``
@@ -87,7 +90,7 @@ The floating-point types supported by ROCm are listed in the following table.
        format.
    *
      - tensorfloat32
-      - Not available
+      - ``-``
      - A floating-point number that occupies 32 bits or less of storage,
        providing improved range compared to half (16-bit) format, at
        (potentially) greater throughput than single-precision (32-bit) formats.
@@ -114,15 +117,12 @@ The floating-point types supported by ROCm are listed in the following table.

  * In some AMD documents and articles, float8 (E5M2) is referred to as bfloat8.

-  * The :doc:`low precision floating point types page <hip:reference/low_fp_types>`
-    describes how to use these types in HIP with examples.
+ROCm support icons
+==========================================

-Level of support definitions
-============================
-
-In the following sections, icons represent the level of support. These icons,
-described in the following table, are also used in the library data type support
-pages.
+In the following sections, icons represent the level of support. These
+icons, described in the following table, are also used in the library data type
+support pages.

 .. list-table::
    :header-rows: 1
@@ -130,11 +130,6 @@ pages.
    *
      -  Icon
      - Definition
-
-    *
-      - NA
-      - Not applicable
-
    *
      - ❌
      - Not supported
@@ -163,15 +158,16 @@ pages.
  * Any type can be emulated by software, but this page does not cover such
    cases.

-Data type support by Hardware Architecture
+Hardware data type support
 ==========================================

-The MI200 series GPUs, which include MI210, MI250, and MI250X, are based on the
-CDNA2 architecture. The MI300 series GPUs, consisting of MI300A, MI300X, and
-MI325X, are based on the CDNA3 architecture.
+The following tables provide information about AMD Instinct accelerators support
+for various data types. The MI200 series GPUs, which include MI210, MI250, and
+MI250X, are based on the CDNA2 architecture. The MI300 series GPUs, consisting
+of MI300A, MI300X, and MI325X, are built on the CDNA3 architecture.

 Compute units support
---------------------
+-------------------------------------------------------------------------------

 The following table lists data type support for compute units.

@@ -252,7 +248,7 @@ The following table lists data type support for compute units.
        - ✅

 Matrix core support
-------------------
+-------------------------------------------------------------------------------

 The following table lists data type support for AMD GPU matrix cores.

@@ -333,7 +329,7 @@ The following table lists data type support for AMD GPU matrix cores.
        - ✅

 Atomic operations support
-------------------------
+-------------------------------------------------------------------------------

 The following table lists data type support for atomic operations.

@@ -420,14 +416,14 @@ The following table lists data type support for atomic operations.
  performance impact when they frequently access the same memory address.

 Data type support in ROCm libraries
-===================================
+==========================================

 ROCm library support for int8, float8 (E4M3), float8 (E5M2), int16, float16,
 bfloat16, int32, tensorfloat32, float32, int64, and float64 is listed in the
 following tables.

 Libraries input/output type support
-----------------------------------
+-------------------------------------------------------------------------------

 The following tables list ROCm library support for specific input and output
 data types. Refer to the corresponding library data type support page for a
@@ -448,37 +444,37 @@ detailed description.
        - int32
        - int64
      *
-        - :doc:`hipSPARSELt <hipsparselt:reference/data-type-support>`
+        - hipSPARSELt (:doc:`details <hipsparselt:reference/data-type-support>`)
        - ✅/✅
        - ❌/❌
        - ❌/❌
        - ❌/❌
      *
-        - :doc:`rocRAND <rocrand:api-reference/data-type-support>`
-        - NA/✅
-        - NA/✅
-        - NA/✅
-        - NA/✅
+        - rocRAND (:doc:`details <rocrand:api-reference/data-type-support>`)
+        - -/✅
+        - -/✅
+        - -/✅
+        - -/✅
      *
-        - :doc:`hipRAND <hiprand:api-reference/data-type-support>`
-        - NA/✅
-        - NA/✅
-        - NA/✅
-        - NA/✅
+        - hipRAND (:doc:`details <hiprand:api-reference/data-type-support>`)
+        - -/✅
+        - -/✅
+        - -/✅
+        - -/✅
      *
-        -  :doc:`rocPRIM <rocprim:reference/data-type-support>`
+        - rocPRIM (:doc:`details <rocprim:reference/data-type-support>`)
        - ✅/✅
        - ✅/✅
        - ✅/✅
        - ✅/✅
      *
-        - :doc:`hipCUB <hipcub:api-reference/data-type-support>`
+        - hipCUB (:doc:`details <hipcub:api-reference/data-type-support>`)
        - ✅/✅
        - ✅/✅
        - ✅/✅
        - ✅/✅
      *
-        - :doc:`rocThrust <rocthrust:data-type-support>`
+        - rocThrust (:doc:`details <rocthrust:data-type-support>`)
        - ✅/✅
        - ✅/✅
        - ✅/✅
@@ -500,7 +496,7 @@ detailed description.
        - float32
        - float64
      *
-        - :doc:`hipSPARSELt <hipsparselt:reference/data-type-support>`
+        - hipSPARSELt (:doc:`details <hipsparselt:reference/data-type-support>`)
        - ❌/❌
        - ❌/❌
        - ✅/✅
@@ -509,25 +505,25 @@ detailed description.
        - ❌/❌
        - ❌/❌
      *
-        - :doc:`rocRAND <rocrand:api-reference/data-type-support>`
-        - NA/❌
-        - NA/❌
-        - NA/✅
-        - NA/❌
-        - NA/❌
-        - NA/✅
-        - NA/✅
+        - rocRAND (:doc:`details <rocrand:api-reference/data-type-support>`)
+        - -/❌
+        - -/❌
+        - -/✅
+        - -/❌
+        - -/❌
+        - -/✅
+        - -/✅
      *
-        - :doc:`hipRAND <hiprand:api-reference/data-type-support>`
-        - NA/❌
-        - NA/❌
-        - NA/✅
-        - NA/❌
-        - NA/❌
-        - NA/✅
-        - NA/✅
+        - hipRAND (:doc:`details <hiprand:api-reference/data-type-support>`)
+        - -/❌
+        - -/❌
+        - -/✅
+        - -/❌
+        - -/❌
+        - -/✅
+        - -/✅
      *
-        - :doc:`rocPRIM <rocprim:reference/data-type-support>`
+        - rocPRIM (:doc:`details <rocprim:reference/data-type-support>`)
        - ❌/❌
        - ❌/❌
        - ✅/✅
@@ -536,7 +532,7 @@ detailed description.
        - ✅/✅
        - ✅/✅
      *
-        - :doc:`hipCUB <hipcub:api-reference/data-type-support>`
+        - hipCUB (:doc:`details <hipcub:api-reference/data-type-support>`)
        - ❌/❌
        - ❌/❌
        - ✅/✅
@@ -545,7 +541,7 @@ detailed description.
        - ✅/✅
        - ✅/✅
      *
-        - :doc:`rocThrust <rocthrust:data-type-support>`
+        - rocThrust (:doc:`details <rocthrust:data-type-support>`)
        - ❌/❌
        - ❌/❌
        - ⚠️/⚠️
@@ -554,14 +550,9 @@ detailed description.
        - ✅/✅
        - ✅/✅

-.. note::
-   
-  As random number generation libraries, rocRAND and hipRAND only specify output
-  data types for the random values they generate, with no need for input data
-  types.

 Libraries internal calculations type support
--------------------------------------------
+-------------------------------------------------------------------------------

 The following tables list ROCm library support for specific internal data types.
 Refer to the corresponding library data type support page for a detailed
@@ -582,7 +573,7 @@ description.
        - int32
        - int64
      *
-        - :doc:`hipSPARSELt <hipsparselt:reference/data-type-support>`
+        - hipSPARSELt (:doc:`details <hipsparselt:reference/data-type-support>`)
        - ❌
        - ❌
        - ✅
@@ -605,7 +596,7 @@ description.
        - float32
        - float64
      *
-        - :doc:`hipSPARSELt <hipsparselt:reference/data-type-support>`
+        - hipSPARSELt (:doc:`details <hipsparselt:reference/data-type-support>`)
        - ❌
        - ❌
        - ❌
--- a/docs/release/versions.md
+++ b/docs/release/versions.md
@@ -10,7 +10,6 @@

 | Version | Release date |
 | ------- | ------------ |
-| [6.3.3](https://rocm.docs.amd.com/en/docs-6.3.3/) | February 19, 2025 |
 | [6.3.2](https://rocm.docs.amd.com/en/docs-6.3.2/) | January 28, 2025 |
 | [6.3.1](https://rocm.docs.amd.com/en/docs-6.3.1/) | December 20, 2024 |
 | [6.3.0](https://rocm.docs.amd.com/en/docs-6.3.0/) | December 3, 2024 |
--- a/docs/sphinx/_toc.yml.in
+++ b/docs/sphinx/_toc.yml.in
@@ -40,13 +40,11 @@ subtrees:
        title: Training
        subtrees:
        - entries:
-          - file: how-to/rocm-for-ai/training/benchmark-docker/megatron-lm
-            title: Train a model with Megatron-LM
-          - file: how-to/rocm-for-ai/training/benchmark-docker/pytorch-training
-            title: Train a model with PyTorch
+          - file: how-to/rocm-for-ai/training/train-a-model.rst
+            title: Train a model
          - file: how-to/rocm-for-ai/training/scale-model-training.rst
            title: Scale model training
-
+      
      - file: how-to/rocm-for-ai/fine-tuning/index.rst
        title: Fine-tuning LLMs
        subtrees:
@@ -154,7 +152,7 @@ subtrees:
        - entries:
          - url: https://www.amd.com/system/files/TechDocs/instinct-mi200-cdna2-instruction-set-architecture.pdf
            title: AMD Instinct MI200/CDNA2 ISA
-          - url: https://www.amd.com/content/dam/amd/en/documents/instinct-business-docs/white-papers/amd-cdna2-white-paper.pdf
+          - url: https://www.amd.com/system/files/documents/amd-cdna2-white-paper.pdf
            title: White paper
      - file: conceptual/gpu-arch/mi100.md
        title: MI100 microarchitecture
@@ -162,7 +160,7 @@ subtrees:
        - entries:
          - url: https://www.amd.com/system/files/TechDocs/instinct-mi100-cdna1-shader-instruction-set-architecture%C2%A0.pdf
            title: AMD Instinct MI100/CDNA1 ISA
-          - url: https://www.amd.com/content/dam/amd/en/documents/instinct-business-docs/white-papers/amd-cdna-white-paper.pdf
+          - url: https://www.amd.com/system/files/documents/amd-cdna-whitepaper.pdf
            title: White paper
  - file: conceptual/iommu.rst
    title: Input-Output Memory Management Unit (IOMMU)
--- a/docs/sphinx/requirements.in
+++ b/docs/sphinx/requirements.in
@@ -1,4 +1,3 @@
 rocm-docs-core==1.17.0
 sphinx-reredirects
 sphinx-sitemap
-sphinxcontrib.datatemplates==0.11.0
--- a/docs/sphinx/requirements.txt
+++ b/docs/sphinx/requirements.txt
@@ -1,15 +1,13 @@
 #
-# This file is autogenerated by pip-compile with Python 3.11
+# This file is autogenerated by pip-compile with Python 3.10
 # 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
@@ -25,7 +23,7 @@ beautifulsoup4==4.12.3
    # via pydata-sphinx-theme
 breathe==4.35.0
    # via rocm-docs-core
-certifi==2024.12.14
+certifi==2024.8.30
    # via requests
 cffi==1.17.1
    # via
@@ -39,14 +37,12 @@ click==8.1.7
    #   sphinx-external-toc
 comm==0.2.2
    # via ipykernel
-cryptography==44.0.0
+cryptography==43.0.3
    # 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
@@ -55,9 +51,11 @@ docutils==0.21.2
    #   myst-parser
    #   pydata-sphinx-theme
    #   sphinx
+exceptiongroup==1.2.2
+    # via ipython
 executing==2.2.0
    # via stack-data
-fastjsonschema==2.21.1
+fastjsonschema==2.20.0
    # via
    #   nbformat
    #   rocm-docs-core
@@ -65,6 +63,8 @@ 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
@@ -75,13 +75,13 @@ importlib-metadata==8.6.1
    #   myst-nb
 ipykernel==6.29.5
    # via myst-nb
-ipython==8.32.0
+ipython==8.31.0
    # via
    #   ipykernel
    #   myst-nb
 jedi==0.19.2
    # via ipython
-jinja2==3.1.4
+jinja2==3.1.5
    # via
    #   myst-parser
    #   sphinx
@@ -115,7 +115,7 @@ mdit-py-plugins==0.4.2
    # via myst-parser
 mdurl==0.1.2
    # via markdown-it-py
-myst-nb==1.2.0
+myst-nb==1.1.2
    # via rocm-docs-core
 myst-parser==4.0.0
    # via myst-nb
@@ -142,7 +142,7 @@ platformdirs==4.3.6
    # via jupyter-core
 prompt-toolkit==3.0.50
    # via ipython
-psutil==7.0.0
+psutil==6.1.1
    # via ipykernel
 ptyprocess==0.7.0
    # via pexpect
@@ -150,7 +150,7 @@ pure-eval==0.2.3
    # via stack-data
 pycparser==2.22
    # via cffi
-pydata-sphinx-theme==0.16.1
+pydata-sphinx-theme==0.16.0
    # via
    #   rocm-docs-core
    #   sphinx-book-theme
@@ -162,7 +162,7 @@ pygments==2.18.0
    #   ipython
    #   pydata-sphinx-theme
    #   sphinx
-pyjwt[crypto]==2.10.1
+pyjwt[crypto]==2.10.0
    # via pygithub
 pynacl==1.5.0
    # via pygithub
@@ -175,7 +175,7 @@ pyyaml==6.0.2
    #   myst-parser
    #   rocm-docs-core
    #   sphinx-external-toc
-pyzmq==26.2.1
+pyzmq==26.2.0
    # via
    #   ipykernel
    #   jupyter-client
@@ -215,8 +215,6 @@ 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
@@ -228,13 +226,11 @@ 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
@@ -243,16 +239,16 @@ 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==2.0.38
+sqlalchemy==2.0.37
    # 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
--- a/docs/sphinx/static/css/vllm-benchmark.css
+++ b/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;
-}
--- a/tools/autotag/components.xml
+++ b/tools/autotag/components.xml
@@ -1,7 +1,7 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <manifest>
    <remote name="rocm-org" fetch="https://github.com/ROCm/" />
-    <default revision="refs/tags/rocm-6.3.3"
+    <default revision="refs/tags/rocm-6.3.2"
     remote="rocm-org"
     sync-c="true"
     sync-j="4" />
--- a/tools/autotag/templates/highlights/6.3.3.md
+++ b/tools/autotag/templates/highlights/6.3.3.md
@@ -1,47 +0,0 @@
-# ROCm 6.3.3 release notes
-
-The release notes provide a summary of notable changes since the previous ROCm release.
-
- [Release highlights](#release-highlights)
-
- [Operating system and hardware support changes](#operating-system-and-hardware-support-changes)
-
- [ROCm components versioning](#rocm-components)
-
- [Detailed component changes](#detailed-component-changes)
-
- [ROCm known issues](#rocm-known-issues)
-
- [ROCm upcoming changes](#rocm-upcoming-changes)
-
-```{note}
-If you’re using Radeon™ PRO or Radeon GPUs in a workstation setting with a display connected, see the [Use ROCm on Radeon GPUs](https://rocm.docs.amd.com/projects/radeon/en/latest/docs/compatibility/native_linux/native_linux_compatibility.html)
-documentation to verify compatibility and system requirements.
-```
-## Release highlights
-
-The following are notable new features and improvements in ROCm 6.3.3. For changes to individual components, see
-[Detailed component changes](#detailed-component-changes).
-
-### ROCm Offline Installer Creator updates
-
-The ROCm Offline Installer Creator 6.3.3 adds a new Post-Install Options menu, which includes a new ``udev`` option for adding GPU resources access for all users. It also moves the user-specific GPU access option (for the ``video,render`` group) from the Driver Options menu to the Post-Install Options menu. See the [ROCm Offline Installer Creator](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/rocm-offline-installer.html#post-install-options-menu) documentation for more information.
-
-### ROCm documentation updates
-
-ROCm documentation continues to be updated to provide clearer and more comprehensive guidance for a wider variety of user needs and use cases.
-
-* [Tutorials for AI developers](https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/) have been added. These tutorials are Jupyter notebook-based, easy-to-follow documents. They are ideal for AI developers who want to learn about specific topics, including inference, fine-tuning, and training.
-
-* The [LLM inference performance validation guide for AMD Instinct MI300X](https://rocm.docs.amd.com/en/latest/how-to/rocm-for-ai/inference/vllm-benchmark.html)
-  now includes additional models for performance benchmarking. The accompanying ROCm vLLM Docker has been upgraded to ROCm 6.3.1.
- 
-* The HIP documentation has been updated with new resources for developers. To learn more about concurrency, parallelism, and stream management on devices and multiple GPUs, see [Asynchronous concurrent execution](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_runtime_api/asynchronous.html)
-
-* The following HIP documentation topics have been updated:
-    - [Virtual memory management](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_runtime_api/memory_management/virtual_memory.html)
-    - [Programming for HIP runtime compiler (RTC)](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_rtc.html)
-    - [HIP porting guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
-    - [Porting CUDA driver API](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_driver_api.html)
-    - [CUDA to HIP API function comparison](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/api_syntax.html)
-
--- a/tools/autotag/templates/known_issues/6.3.3.md
+++ b/tools/autotag/templates/known_issues/6.3.3.md
@@ -1,8 +0,0 @@
-## ROCm known issues
-
-ROCm known issues are noted on {fab}`github` [GitHub](https://github.com/ROCm/ROCm/labels/Verified%20Issue). For known
-issues related to individual components, review the [Detailed component changes](#detailed-component-changes).
-
-### Zero value is displayed in ROCTx aggregated statistics
-
-The ROCTx markers are standalone markers within the ROCProfiler-SDK library. Each marker reports only a single timestamp, which is recorded as the `start_timestamp` and `end_timestamp`. As a result, the value for aggregated statistics presented in `TotalDurationNs`, `maxNs`, and `minNs`, is zero. The zero value indicates that the actual execution time is not associated with the markers, which is an expected behavior.
--- a/tools/autotag/templates/resolved_issues/6.3.3.md
+++ b/tools/autotag/templates/resolved_issues/6.3.3.md
--- a/tools/autotag/templates/support/6.3.3.md
+++ b/tools/autotag/templates/support/6.3.3.md
@@ -1,7 +0,0 @@
-## Operating system and hardware support changes
-
-Operating system and hardware support remain unchanged in this release.
-
-See the [Compatibility
-matrix](https://rocm.docs.amd.com/en/docs-6.3.3/compatibility/compatibility-matrix.html)
-for more information about operating system and hardware compatibility.
--- a/tools/autotag/templates/upcoming_changes/6.3.3.md
+++ b/tools/autotag/templates/upcoming_changes/6.3.3.md
@@ -1,17 +0,0 @@
-## ROCm upcoming changes
-
-The following changes to the ROCm software stack are anticipated for future releases.
-
-### ROCTracer and ROCProfiler (rocprof and rocprofv2) deprecation
-
-Development and support for ROCTracer and ROCProfiler (`rocprof` and `rocprofv2`) will phase out in favor of ROCprofiler-SDK (`rocprofv3`) in upcoming ROCm releases. Going forward, only critical defect fixes will be addressed for older versions of profiling tools and libraries. Upgrade to the latest version of ROCprofiler-SDK (`rocprofv3`) library to ensure continued support and access to new features.
-
-### AMDGPU wavefront size compiler macro deprecation
-
-The `__AMDGCN_WAVEFRONT_SIZE__` macro will be deprecated in an upcoming
-release. It is recommended to remove any use of this macro. For more information, see [AMDGPU
-support](https://rocm.docs.amd.com/projects/llvm-project/en/docs-6.3.3/LLVM/clang/html/AMDGPUSupport.html).
-
-### HIPCC Perl scripts deprecation
-
-The HIPCC Perl scripts (`hipcc.pl` and `hipconfig.pl`) will be removed in an upcoming release.
--- a/tools/rocm-build/ROCm.mk
+++ b/tools/rocm-build/ROCm.mk
@@ -103,7 +103,7 @@ $(call adddep,composable_kernel,lightning hipcc hip_on_rocclr rocm-cmake)
 $(call adddep,half,rocm-cmake)
 $(call adddep,hipblas-common,lightning)
 $(call adddep,hipblas,hip_on_rocclr rocblas rocsolver lightning hipcc)
-$(call adddep,hipblaslt,hip_on_rocclr openmp_extras lightning hipcc hipblas-common rocm-dev)
+$(call adddep,hipblaslt,hip_on_rocclr openmp_extras lightning hipcc hipblas-common roctracer)
 $(call adddep,hipcub,hip_on_rocclr rocprim lightning hipcc)
 $(call adddep,hipfft,hip_on_rocclr openmp_extras rocfft rocrand hiprand lightning hipcc)
 $(call adddep,hipfort,rocblas hipblas rocsparse hipsparse rocfft hipfft rocrand hiprand rocsolver hipsolver lightning hipcc)
--- a/tools/rocm-build/build_amd_smi_lib.sh
+++ b/tools/rocm-build/build_amd_smi_lib.sh
@@ -10,7 +10,9 @@ printUsage() {
    echo "  -c,  --clean              Removes all amd_smi build artifacts"
    echo "  -r,  --release            Build non-debug version amd_smi (default is debug)"
    echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -s,  --static             Component/Build does not support static builds just accepting this param & ignore. No effect of the param on this build"
+    echo "  -w,  --wheel              Creates python wheel package of amd-smi. 
+                                      It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of type referred to by pkg_type"
    echo "  -p,  --package <type>     Specify packaging format"
    echo "  -h,  --help               Prints this help"
@@ -25,7 +27,6 @@ printUsage() {
 PROJ_NAME="amdsmi"
 PACKAGE_ROOT="$(getPackageRoot)"
 TARGET="build"
-
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE="$(getIncludePath)"
 AMDSMI_BUILD_DIR=$(getBuildPath $PROJ_NAME)
@@ -42,7 +43,7 @@ SHARED_LIBS="ON"
 CLEAN_OR_OUT=0;
 PKGTYPE="deb"

-VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
+VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
@@ -60,6 +61,8 @@ do
                ADDRESS_SANITIZER=true ; shift ;;
        (-s | --static)
                ack_and_skip_static ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; shift ;;
        (-o | --outdir)
                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
        (-p | --package)
@@ -99,6 +102,7 @@ build_amdsmi() {
            -DBUILD_SHARED_LIBS=$SHARED_LIBS \
            $(rocm_common_cmake_params) \
            $(rocm_cmake_params) \
+            ${GEN_NINJA} \
            -DENABLE_LDCONFIG=OFF \
            -DAMD_SMI_PACKAGE="${AMDSMI_PKG_NAME}" \
            -DCPACK_PACKAGE_VERSION_MAJOR="1" \
@@ -106,14 +110,14 @@ build_amdsmi() {
            -DCPACK_PACKAGE_VERSION_PATCH="0" \
            -DADDRESS_SANITIZER="$ADDRESS_SANITIZER" \
            -DBUILD_TESTS=ON \
-            "$AMD_SMI_LIB_ROOT"
+            -S "$AMD_SMI_LIB_ROOT"
        popd
    fi

    echo "Making amd_smi package:"
-    cmake --build "$AMDSMI_BUILD_DIR" -- $AMDSMI_MAKE_OPTS
-    cmake --build "$AMDSMI_BUILD_DIR" -- $AMDSMI_MAKE_OPTS install
-    cmake --build "$AMDSMI_BUILD_DIR" -- $AMDSMI_MAKE_OPTS package
+    cmake --build "$AMDSMI_BUILD_DIR" -- $DASH_JAY
+    cmake --build "$AMDSMI_BUILD_DIR" -- $DASH_JAY install
+    cmake --build "$AMDSMI_BUILD_DIR" -- $DASH_JAY package

    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$AMDSMI_PACKAGE_DEB_DIR" $AMDSMI_BUILD_DIR/*.deb
    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$AMDSMI_PACKAGE_RPM_DIR" $AMDSMI_BUILD_DIR/*.rpm
@@ -135,7 +139,7 @@ verifyEnvSetup

 case $TARGET in
    (clean) clean_amdsmi ;;
-    (build) build_amdsmi ;;
+    (build) build_amdsmi; build_wheel "$AMDSMI_BUILD_DIR" "$PROJ_NAME" ;;
    (outdir) print_output_directory ;;
    (*) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_amdmigraphx.sh
+++ b/tools/rocm-build/build_amdmigraphx.sh
@@ -1,41 +1,42 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src AMDMIGraphX

 build_amdmigraphx() {
    echo "Start build"

+    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
+        ack_and_skip_static
+    fi
+
    cd $COMPONENT_SRC

    if ! command -v rbuild &> /dev/null; then
        pip3 install https://github.com/RadeonOpenCompute/rbuild/archive/master.tar.gz
    fi

+    # Remove CK
+    xargs -d '\n' -a ${OUT_DIR}/ck.files rm -- || true
+
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
         set_asan_env_vars
         set_address_sanitizer_on
    fi

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx900;gfx906;gfx908;gfx90a;gfx1030;gfx1100;gfx1101;gfx1102;gfx942;gfx1200;gfx1201"
-    fi
    init_rocm_common_cmake_params

    mkdir -p ${BUILD_DIR} && rm -rf ${BUILD_DIR}/* && mkdir -p ${HOME}/amdmigraphx && rm -rf ${HOME}/amdmigraphx/*
    rbuild package -d "${HOME}/amdmigraphx" -B "${BUILD_DIR}" \
-        --cxx="${ROCM_PATH}/llvm/bin/clang++" \
-        --cc="${ROCM_PATH}/llvm/bin/clang" \
+        --cxx="$(set_build_variables __CLANG++__)" \
+        --cc="$(set_build_variables __CLANG__)" \
        "${rocm_math_common_cmake_params[@]}" \
        -DCMAKE_MODULE_LINKER_FLAGS="-Wl,--enable-new-dtags,--build-id=sha1,--rpath,$ROCM_LIB_RPATH" \
-	    -DGPU_TARGETS="${GPU_TARGETS}" \
        -DCMAKE_INSTALL_RPATH=""

-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
    cd $BUILD_DIR && cmake --build . -- install -j${PROC}

    show_build_cache_stats
@@ -50,7 +51,7 @@ clean_amdmigraphx() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_amdmigraphx ;;
+    build) build_amdmigraphx; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_amdmigraphx ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_comgr.sh
+++ b/tools/rocm-build/build_comgr.sh
@@ -14,6 +14,7 @@ printUsage() {
        type referred to by pkg_type"
    echo "  -h,  --help               Prints this help"
    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -w,  --wheel              Creates python wheel package of comgr. It needs to be used along with -r option"
    echo "  -l,  --link_llvm_static   Link to LLVM statically.  Default is to dynamically link to LLVM; requires that LLVM dylibs are created."
    echo
    echo "Possible values for <type>:"
@@ -24,6 +25,7 @@ printUsage() {
    return 0
 }

+## Build environment variables
 API_NAME=amd_comgr
 PROJ_NAME=$API_NAME
 LIB_NAME=lib${API_NAME}
@@ -33,8 +35,8 @@ PACKAGE_ROOT=$(getPackageRoot)
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE=$(getIncludePath)
 BUILD_DIR=$(getBuildPath $API_NAME)
-PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
-PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
+PACKAGE_DEB=$PACKAGE_ROOT/deb/$PROJ_NAME
+PACKAGE_RPM=$PACKAGE_ROOT/rpm/$PROJ_NAME
 PACKAGE_PREFIX=$ROCM_INSTALL_PATH
 BUILD_TYPE=Debug
 MAKE_OPTS="$DASH_JAY CTEST_OUTPUT_ON_FAILURE=1 -C $BUILD_DIR"
@@ -44,13 +46,17 @@ CLEAN_OR_OUT=0;
 PKGTYPE="deb"
 MAKETARGET="deb"

+# link to LLVM dynamicaly.  Default is to link to llvm dynamically.
+# temporarily set this default to OFF until we resolve Issues
 LINK_LLVM_DYLIB="OFF"

-VALID_STR=`getopt -o hcraslo:p: --long help,clean,release,address_sanitizer,static,link_llvm_static,outdir:,package: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswlo:p: --long help,clean,release,address_sanitizer,static,wheel,link_llvm_static,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
 do
+    #echo "parocessing $1"
    case "$1" in
        (-h | --help)
                printUsage ; exit 0;;
@@ -63,6 +69,8 @@ do
                set_address_sanitizer_on ; shift ;;
        (-s | --static)
                SHARED_LIBS="OFF" ; shift ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; shift ;;
        (-l | --link_llvm_static)
                LINK_LLVM_DYLIB="OFF"; shift ;;
        (-o | --outdir)
@@ -105,6 +113,7 @@ build() {
       echo " Building Shared Object "
    fi

+# Remove CTEST var once SWDEV-381396 is fixed
    cmake \
        $(rocm_cmake_params) \
        -DBUILD_SHARED_LIBS=$SHARED_LIBS \
@@ -143,7 +152,7 @@ verifyEnvSetup

 case $TARGET in
    (clean) clean ;;
-    (build) build ;;
+    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
    (outdir) print_output_directory ;;
    (*) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_composable_kernel.sh
+++ b/tools/rocm-build/build_composable_kernel.sh
@@ -2,10 +2,14 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+
+# Temporarily disable Address Sanitizer
+ENABLE_ADDRESS_SANITIZER=false

 set_component_src composable_kernel
-
+disable_debug_package_generation
+# Set the GPU_ARCH_LIST to the supported GPUs needed after https://github.com/ROCm/composable_kernel/pull/1536/
 GPU_ARCH_LIST="gfx908;gfx90a;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"

 build_miopen_ck() {
@@ -23,10 +27,17 @@ build_miopen_ck() {
        GPU_ARCH_LIST="gfx942"
        ack_and_skip_static
    fi
-
+    # if ENABLE_GPU_ARCH is set in env by Job parameter ENABLE_GPU_ARCH, then set GPU_ARCH_LIST to that value
+    # This needs to be removed when CK aligns with other component and uses -DAMDGPU_TARGET or _DGPO_TARGET
+    # then we can use set_gpu_arch from compure_helper.sh and get rid of all if clauses
+    if [ -n "$ENABLE_GPU_ARCH" ]; then
+        GPU_ARCH_LIST="$ENABLE_GPU_ARCH"
+    fi
+    # Latest CK requiring Python 3.8 as the minimum.
+    # Point CMake to that explicit location and adjust LD_LIBRARY_PATH.
    PYTHON_VERSION_WORKAROUND=''
    echo "DISTRO_ID: ${DISTRO_ID}"
-    if [ "$DISTRO_ID" = "rhel-8.8" ] || [ "$DISTRO_ID" = "sles-15.5" ] ; then
+    if [ "$DISTRO_ID" = "rhel-8.8" ] || [ "$DISTRO_NAME" == "sles" ] || [ "$DISTRO_ID" = "debian-10" ]; then
        EXTRA_PYTHON_PATH=/opt/Python-3.8.13
        PYTHON_VERSION_WORKAROUND="-DCK_USE_ALTERNATIVE_PYTHON=${EXTRA_PYTHON_PATH}/bin/python3.8"
        # For the python interpreter we need to export LD_LIBRARY_PATH.
@@ -38,23 +49,25 @@ build_miopen_ck() {
    init_rocm_common_cmake_params

    cmake \
+        ${GEN_NINJA} \
        -DBUILD_DEV=OFF \
        "${rocm_math_common_cmake_params[@]}" \
        ${PYTHON_VERSION_WORKAROUND} \
        -DCPACK_GENERATOR="${PKGTYPE^^}" \
-        -DCMAKE_CXX_COMPILER="${ROCM_PATH}/llvm/bin/clang++" \
-        -DCMAKE_C_COMPILER="${ROCM_PATH}/llvm/bin/clang" \
+        -DCMAKE_CXX_COMPILER=$(set_build_variables __CLANG++__) \
+        -DCMAKE_C_COMPILER=$(set_build_variables __CLANG__) \
        ${LAUNCHER_FLAGS} \
        -DGPU_ARCHS="${GPU_ARCH_LIST}" \
-        -DCMAKE_BUILD_TYPE=Release \
-        -DCMAKE_CXX_FLAGS=" -O3 " \
        "$COMPONENT_SRC"

    cmake --build . -- -j${PROC} package
    cmake --build "$BUILD_DIR" -- install
-    mkdir -p $PACKAGE_DIR && cp ./*.${PKGTYPE} $PACKAGE_DIR
+
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
 }

+# Use the function to unset the LDFLAGS and CXXFLAGS
+# specifically set for ASAN
 unset_asan_env_vars() {
    ASAN_CMAKE_PARAMS="false"
    export ADDRESS_SANITIZER="OFF"
@@ -77,7 +90,7 @@ clean_miopen_ck() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_miopen_ck ;;
+    build) build_miopen_ck; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_miopen_ck ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_dbgapi.sh
+++ b/tools/rocm-build/build_dbgapi.sh
@@ -11,6 +11,8 @@ printUsage() {
    echo "  -r,  --release            Make a release build instead of a debug build"
    echo "       --enable-assertions  Enable assertions"
    echo "  -a,  --address_sanitizer  Enable address sanitizer"
+    echo "  -w,  --wheel              Creates python wheel package of dbgapi. 
+                                      It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
            type referred to by pkg_type"
    echo "  -h,  --help               Prints this help"
@@ -25,6 +27,7 @@ printUsage() {
    return 0
 }

+## Build environment variables
 API_NAME=rocm-dbgapi
 AMD_DBGAPI_NAME=amd-dbgapi
 MAKEINSTALL_MANIFEST=makeinstall_manifest.txt
@@ -36,21 +39,24 @@ PACKAGE_ROOT=$(getPackageRoot)
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE=$(getIncludePath)
 BUILD_DIR=$(getBuildPath $API_NAME)
-PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
-PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
+PACKAGE_DEB=$(getPackageRoot)/deb/$PROJ_NAME
+PACKAGE_RPM=$(getPackageRoot)/rpm/$PROJ_NAME
+#PACKAGE_PREFIX=$ROCM_INSTALL_PATH
 BUILD_TYPE=Debug
-MAKE_OPTS=($DASH_JAY -C "$BUILD_DIR")
+MAKE_OPTS=($DASH_JAY -C "$BUILD_DIR") # Note that DASH_JAY might have a space after the -j
 SHARED_LIBS="ON"
 CLEAN_OR_OUT=0;
 MAKETARGET="deb"
 PKGTYPE="deb"
 DODOCSBUILD=true

-VALID_STR=$(getopt -o hcraso:p:M --long help,clean,release,enable-assertions,static,address_sanitizer,outdir:,package:skip_man_pages -- "$@")
+#parse the arguments
+VALID_STR=$(getopt -o hcraswo:p:M --long help,clean,release,enable-assertions,static,wheel,address_sanitizer,outdir:,package:skip_man_pages -- "$@")
 eval set -- "$VALID_STR"

 while true ;
 do
+    #echo "parocessing $1"
    case "$1" in
        (-h | --help)
                printUsage ; exit 0;;
@@ -66,12 +72,14 @@ do
                set_address_sanitizer_on ;;
        (-s | --static)
                ack_and_skip_static ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ;;
        (-o | --outdir)
                TARGET="outdir"; PKGTYPE=$2 ; ((CLEAN_OR_OUT|=2)) ; shift 1 ;;
        (-M | --skip_man_pages) DODOCSBUILD=false;;
        (-p | --package)
                MAKETARGET="$2" ; shift 1;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        (*)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
    esac
@@ -104,7 +112,7 @@ build() {
    then
           echo " No $ROCM_DBGAPI_ROOT/CMakeLists.txt file, skipping rocm-dbgapi" >&2
           echo " No $ROCM_DBGAPI_ROOT/CMakeLists.txt file, skipping rocm-dbgapi"
-       exit 0
+       exit 0 # THis is not an error
    fi
    echo "Building $PROJ_NAME"

@@ -121,11 +129,14 @@ build() {
    cmake --build "$BUILD_DIR" -- "${MAKE_OPTS[@]}"
    "$DODOCSBUILD" && cmake --build "$BUILD_DIR" -- "${MAKE_OPTS[@]}" doc
    cmake --build "$BUILD_DIR" -- "${MAKE_OPTS[@]}" install
-
+    #install_manifest.txt is created by make install and make package with same name unless
+    #component packaging is enabled. To avoid overwriting by make package,move the manifest file
+    #to a different name and can be used for build clean up
    mv "$BUILD_DIR/install_manifest.txt" "$BUILD_DIR/$MAKEINSTALL_MANIFEST"
    cmake --build "$BUILD_DIR" -- "${MAKE_OPTS[@]}" package

    mkdir -p "$PACKAGE_LIB"
+    # handle the library being in more than one place to avoid breaking the build.
    (
        shopt -s nullglob
        cp -R "$BUILD_DIR/lib/${LIB_NAME}"* "$BUILD_DIR/${LIB_NAME}"* "$PACKAGE_LIB"
@@ -151,7 +162,7 @@ verifyEnvSetup

 case $TARGET in
    (clean) clean ;;
-    (build) build ;;
+    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME";;
    (outdir) print_output_directory ;;
    (*) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_devicelibs.sh
+++ b/tools/rocm-build/build_devicelibs.sh
@@ -10,7 +10,9 @@ printUsage() {
    echo "  -c,  --clean              Clean output and delete all intermediate work"
    echo "  -r,  --release            Build a release version of the package"
    echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
+    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -w,  --wheel              Creates python wheel package of devicelibs.
+                                      It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
    type referred to by pkg_type"
    echo "  -h,  --help             Prints this help"
@@ -19,10 +21,11 @@ printUsage() {

    return 0
 }
+PROJ_NAME="devicelibs"
 PACKAGE_ROOT="$(getPackageRoot)"
 PACKAGE_BIN="$(getBinPath)"
 PACKAGE_LIB="$(getLibPath)"
-BUILD_PATH="$(getBuildPath devicelibs)"
+BUILD_PATH="$(getBuildPath $PROJ_NAME)"
 INSTALL_PATH="$(getPackageRoot)"
 LIGHTNING_BUILD_PATH="$(getBuildPath lightning)"
 DEB_PATH="$(getDebPath devicelibs)"
@@ -36,11 +39,13 @@ CLEAN_OR_OUT=0;
 PKGTYPE="deb"
 MAKETARGET="deb"

-VALID_STR=`getopt -o hcraso: --long help,clean,release,static,address_sanitizer,outdir: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswo: --long help,clean,release,static,wheel,address_sanitizer,outdir: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
 do
+    #echo "parocessing $1"
    case "$1" in
        (-h | --help)
                printUsage ; exit 0;;
@@ -53,6 +58,8 @@ 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 ;;
        --)     shift; break;; # end delimiter
@@ -71,7 +78,9 @@ fi


 clean_devicelibs() {
+    # Delete cmake output directory
    rm -rf "$BUILD_PATH"
+    # Delete the *.bc files from the PACKAGE_LIB directory
    rm -f $PACKAGE_LIB/hc*.bc
    rm -f $PACKAGE_LIB/irif*.bc
    rm -f $PACKAGE_LIB/ockl*.bc
@@ -80,6 +89,7 @@ clean_devicelibs() {
    rm -f $PACKAGE_LIB/opencl*.bc
    rm -f $PACKAGE_LIB/openmp*.bc
    rm -rf $PACKAGE_ROOT/amdgcn
+    # Delete any packages generated
    rm -rf "$DEB_PATH"
    rm -rf "$RPM_PATH"
 }
@@ -96,6 +106,7 @@ build_devicelibs() {
    if [ ! -e Makefile ]; then
        cmake $(rocm_cmake_params) \
              $(rocm_common_cmake_params) \
+              ${GEN_NINJA} \
              -DBUILD_SHARED_LIBS=$SHARED_LIBS \
              -DROCM_DEVICE_LIBS_BITCODE_INSTALL_LOC_NEW="$bitcodeInstallLoc/amdgcn" \
              -DROCM_DEVICE_LIBS_BITCODE_INSTALL_LOC_OLD="amdgcn" \
@@ -133,7 +144,7 @@ print_output_directory() {
 }
 case $TARGET in
    (clean) clean_devicelibs ;;
-    (build) build_devicelibs; package_devicelibs ;;
+    (build) build_devicelibs; package_devicelibs; build_wheel "$BUILD_PATH" "$PROJ_NAME" ;;
    (outdir) print_output_directory ;;
    (*) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_half.sh
+++ b/tools/rocm-build/build_half.sh
@@ -1,22 +1,31 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src half

 build_half() {
    echo "Start build"

+    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
+        ack_and_skip_static
+    fi
+
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
         set_asan_env_vars
         set_address_sanitizer_on
+         # ASAN packaging is not required for HALF, since its header only package
+         # Setting the asan_cmake_params to false will disable ASAN packaging
         ASAN_CMAKE_PARAMS="false"
    fi
    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"

    cmake \
        -DCMAKE_INSTALL_PREFIX="$ROCM_PATH" \
+        -DCPACK_PACKAGING_INSTALL_PREFIX=${ROCM_PATH}\
+        -DCPACK_SET_DESTDIR="OFF" \
+        -DCPACK_RPM_PACKAGE_RELOCATABLE="ON" \
        -DBUILD_FILE_REORG_BACKWARD_COMPATIBILITY=OFF \
        "$COMPONENT_SRC"

@@ -25,7 +34,7 @@ build_half() {
    cmake --build "$BUILD_DIR" -- install

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -39,7 +48,7 @@ clean_half() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_half ;;
+    build) build_half; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_half ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hip_on_rocclr.sh
+++ b/tools/rocm-build/build_hip_on_rocclr.sh
@@ -10,6 +10,8 @@ printUsage() {
    echo "  -r,  --release              Make a release build instead of a debug build"
    echo "  -a,  --address_sanitizer    Enable address sanitizer"
    echo "  -s,  --static               Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -w,  --wheel                Creates python wheel package of hip-on-rocclr.
+                                        It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>    Print path of output directory containing packages of type referred to by pkg_type"
    echo "  -t,  --offload-arch=<arch>  Specify arch for catch tests ex: --offload-arch=gfx1030 --offload-arch=gfx1100"
    echo "  -p,  --package <type>       Specify packaging format"
@@ -25,7 +27,6 @@ printUsage() {
 source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
 MAKEOPTS="$DASH_JAY"
 PROJ_NAME="hip-on-rocclr"
-
 BUILD_PATH="$(getBuildPath $PROJ_NAME)"

 TARGET="build"
@@ -36,7 +37,6 @@ PACKAGE_RPM="$PACKAGE_ROOT/rpm/$PROJ_NAME"
 PREFIX_PATH="$PACKAGE_ROOT"
 CORE_BUILD_DIR="$(getBuildPath hsa-core)"
 ROCclr_BUILD_DIR="$(getBuildPath rocclr)"
-HIPCC_BUILD_DIR="$(getBuildPath hipcc)"
 CATCH_BUILD_DIR="$(getBuildPath catch)"
 CATCH_SRC="$HIP_CATCH_TESTS_ROOT/catch"
 SAMPLES_SRC="$HIP_CATCH_TESTS_ROOT/samples"
@@ -53,9 +53,10 @@ MAKETARGET="deb"
 PKGTYPE="deb"
 OFFLOAD_ARCH=()

-DEFAULT_OFFLOAD_ARCH=(gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942 gfx1030 gfx1031 gfx1033 gfx1034 gfx1035 gfx1100 gfx1101 gfx1102 gfx1200 gfx1201)
+DEFAULT_OFFLOAD_ARCH=(gfx900 gfx906 gfx908 gfx90a gfx942 gfx1030 gfx1031 gfx1033 gfx1034 gfx1035 gfx1100 gfx1101 gfx1102 gfx1103 gfx1150 gfx1151 gfx1200 gfx1201)

-VALID_STR=`getopt -o hcrast:o: --long help,clean,release,address_sanitizer,static,offload-arch=:,outdir: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswt:o: --long help,clean,release,address_sanitizer,static,wheel,offload-arch=:,outdir: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
@@ -72,11 +73,13 @@ do
                set_address_sanitizer_on ; shift ;;
        (-s | --static)
                SHARED_LIBS="OFF" ; shift ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; shift ;;
        (-t | --offload-arch=)
                OFFLOAD_ARCH+=( "$2" ); ((CLEAN_OR_OUT|=2)); shift 2 ;;
        (-o | --outdir)
                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter

        (*)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
@@ -98,7 +101,10 @@ if [ $RET_CONFLICT -ge 30 ]; then
   exit $RET_CONFLICT
 fi

+
+
 clean_hip_on_rocclr() {
+    # Delete cmake output directory
    rm -rf "$BUILD_PATH"
    rm -rf "$PACKAGE_DEB"
    rm -rf "$PACKAGE_RPM"
@@ -106,17 +112,23 @@ clean_hip_on_rocclr() {
 }

 build_hip_on_rocclr() {
+    # TODO This if condition is a temporary workaround so that mainline builds dont error out
+    # until build migrated from hipamd to clr repo
    if [  -e "$CLR_ROOT/CMakeLists.txt" ]; then
+        # We are in a branch that has migrated to clr repo
        _HIP_CMAKELIST_DIR="$CLR_ROOT"
        _HIP_CMAKELIST_OPT="-DCLR_BUILD_HIP=ON -DCLR_BUILD_OCL=OFF"
        if [ -e "$HIPOTHER_ROOT/hipnv" ]; then
+            # We are in a branch that has hipnv headers migrated
            _HIP_CMAKELIST_OPT="$_HIP_CMAKELIST_OPT -DHIPNV_DIR=$HIPOTHER_ROOT/hipnv"
        fi
    elif [ ! -e "$HIPAMD_ROOT/CMakeLists.txt" ]; then
+        # We seem to have hit a branch in which both the old and the new repo don't exist
        echo "No $HIPAMD_ROOT/CMakeLists.txt file, skipping hip on rocclr" >&2
        echo "No $HIPAMD_ROOT/CMakeLists.txt file, skipping hip on rocclr"
-        exit 0
+        exit 0 # This is not an error
    else
+        # We are in a branch that has not yet migrated to clr repo yet
        _HIP_CMAKELIST_DIR="$HIPAMD_ROOT"
        _HIP_CMAKELIST_OPT=""
    fi
@@ -125,6 +137,7 @@ build_hip_on_rocclr() {
    mkdir -p "$BUILD_PATH"
    pushd "$BUILD_PATH"

+    # FIXME: Remove -DROCclr_DIR/LIBROCclr_STATIC_DIR
    if [ ! -e Makefile ]; then
        echo "Building HIP-On-ROCclr CMake environment"
        print_lib_type $SHARED_LIBS
@@ -140,7 +153,7 @@ build_hip_on_rocclr() {
            -DCMAKE_SKIP_BUILD_RPATH=TRUE \
            -DCPACK_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
            -DROCM_PATH="$ROCM_INSTALL_PATH" \
-            -DHIPCC_BIN_DIR="$HIPCC_BUILD_DIR" \
+            -DHIPCC_BIN_DIR="$ROCM_PATH/bin" \
            -DHIP_CATCH_TEST=1 \
            $_HIP_CMAKELIST_OPT \
            "$_HIP_CMAKELIST_DIR"
@@ -165,8 +178,13 @@ build_catch_tests() {
   rm -rf "$CATCH_BUILD_DIR"
   mkdir -p "$CATCH_BUILD_DIR"
   pushd "$CATCH_BUILD_DIR"
+   # use the newly built hip as HIP for catch
   export HIP_PATH="$ROCM_INSTALL_PATH"
   export ROCM_PATH="$ROCM_INSTALL_PATH"
+   # Note: The rocm_common_cmake_params will provide CMAKE_EXE_LINKER_FLAGS_INIT
+   # for binaries located in /opt/rocm/bin
+   # hip-catch binaries are located in each test folder under /opt/rocm/share/hip/catch_test/
+   # Append the EXE LINKER flags with ROCm library path
   cmake \
         -DCMAKE_BUILD_TYPE="${BUILD_TYPE}" \
         -DBUILD_SHARED_LIBS=$SHARED_LIBS \
@@ -177,6 +195,7 @@ build_catch_tests() {
         $(rocm_common_cmake_params) \
         -DCPACK_RPM_DEBUGINFO_PACKAGE=FALSE \
         -DCPACK_DEBIAN_DEBUGINFO_PACKAGE=FALSE \
+         -DCMAKE_EXE_LINKER_FLAGS_INIT=-Wl,--enable-new-dtags,--build-id=sha1,--rpath,$ROCM_EXE_RPATH:$ROCM_INSTALL_PATH/lib:/opt/rocm/lib \
         -DCPACK_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
         "$CATCH_SRC"

@@ -190,6 +209,7 @@ build_catch_tests() {
 }

 package_samples() {
+   # TODO: To be removed once the issue related to ASAN builds are fixed
   if [ "$ASAN_CMAKE_PARAMS" == "true" ] ; then
      echo "Disable the packaging of HIP samples" >&2
      return
@@ -198,13 +218,17 @@ package_samples() {
   if [ ! -e "$SAMPLES_SRC/CMakeLists.txt" ]; then
      echo "HIP samples source not found at: $SAMPLES_SRC" >&2
      echo "Using samples package from hip project: $BUILD_PATH" >&2
+      # TODO: change to return failure after hip-tests samples change is available in mainline
      return
   fi
-
+   # package samples
   rm -rf "$SAMPLES_BUILD_DIR"
   mkdir -p "$SAMPLES_BUILD_DIR"
   pushd "$SAMPLES_BUILD_DIR"
+   # The cmake path is different for asan and non-asan builds.
+   # Fetch after getting build type. Default will be non-asan build
   local CMAKE_PATH="$(getCmakePath)"
+   # use the newly built hip as HIP for samples
   export HIP_PATH="$ROCM_INSTALL_PATH"
   export ROCM_PATH="$ROCM_INSTALL_PATH"
   cmake \
@@ -291,10 +315,24 @@ print_output_directory() {
 }

 case $TARGET in
-    (clean) clean_hip_on_rocclr; clean_hip_tests ;;
-    (build) build_hip_on_rocclr; build_catch_tests; package_hip_on_rocclr; package_samples; copy_hip_tests;;
-   (outdir) print_output_directory ;;
-        (*) die "Invalid target $TARGET" ;;
+    (clean)
+        clean_hip_on_rocclr
+        clean_hip_tests
+        ;;
+    (build)
+        build_hip_on_rocclr
+        build_catch_tests
+        package_hip_on_rocclr
+        package_samples
+        build_wheel "$BUILD_PATH" "$PROJ_NAME"
+        copy_hip_tests
+        ;;
+    (outdir)
+        print_output_directory
+        ;;
+    (*)
+        die "Invalid target $TARGET"
+        ;;
 esac

 echo "Operation complete"
--- a/tools/rocm-build/build_hipblas-common.sh
+++ b/tools/rocm-build/build_hipblas-common.sh
@@ -2,26 +2,29 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipBLAS-common

 build_hipblas-common() {
    echo "Start build"

+    CXX=$(set_build_variables __C_++__)
    cd $COMPONENT_SRC
    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"

+    echo "C compiler: $CC"
+    echo "CXX compiler: $CXX"
    init_rocm_common_cmake_params
    cmake \
+        ${GEN_NINJA} \
        "${rocm_math_common_cmake_params[@]}" \
        "$COMPONENT_SRC"
    cmake --build "$BUILD_DIR" -- install
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
-
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
    show_build_cache_stats
 }

@@ -34,7 +37,7 @@ clean_hipblas-common() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_hipblas-common ;;
+    build) build_hipblas-common; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hipblas-common ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hipblas.sh
+++ b/tools/rocm-build/build_hipblas.sh
@@ -2,28 +2,32 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipBLAS

 build_hipblas() {
    echo "Start build"

-    CXX="g++"
+    CXX=$(set_build_variables __G_++__)
    CXX_FLAG=
    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
-        CXX="amdclang++"
-        CXX_FLAG="-DCMAKE_CXX_COMPILER=${ROCM_PATH}/llvm/bin/clang++"
+        CXX=$(set_build_variables __AMD_CLANG_++__)
+        CXX_FLAG=$(set_build_variables __CMAKE_CXX_PARAMS__)
    fi

    CLIENTS_SAMPLES="ON"
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
       set_asan_env_vars
       set_address_sanitizer_on
+       # fixme: remove CLIENTS_SAMPLES=OFF once SWDEV-417076 is fixed
       CLIENTS_SAMPLES="OFF"
    fi

    SHARED_LIBS="ON"
+    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
+        SHARED_LIBS="OFF"
+    fi

    echo "C compiler: $CC"
    echo "CXX compiler: $CXX"
@@ -39,10 +43,10 @@ build_hipblas() {
    init_rocm_common_cmake_params
    cmake \
        ${LAUNCHER_FLAGS} \
-        "${rocm_math_common_cmake_params[@]}" \
+	    "${rocm_math_common_cmake_params[@]}" \
        -DUSE_CUDA=OFF \
        -DBUILD_SHARED_LIBS=$SHARED_LIBS \
-        -DBUILD_CLIENTS_TESTS=ON \
+	    -DBUILD_CLIENTS_TESTS=ON \
        -DBUILD_CLIENTS_BENCHMARKS=ON \
        -DBUILD_CLIENTS_SAMPLES="${CLIENTS_SAMPLES}" \
        -DBUILD_ADDRESS_SANITIZER="${ADDRESS_SANITIZER}" \
@@ -54,7 +58,7 @@ build_hipblas() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -68,7 +72,7 @@ clean_hipblas() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_hipblas ;;
+    build) build_hipblas; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hipblas ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hipblaslt.sh
+++ b/tools/rocm-build/build_hipblaslt.sh
@@ -1,10 +1,20 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipBLASLt

+#not enabling the debug files by defalut
+disable_debug_package_generation
+# if ENABLE_GPU_ARCH is set in env by Job parameter ENABLE_GPU_ARCH, then set GFX_ARCH to that value
+if [ -n "$ENABLE_GPU_ARCH" ]; then
+    set_gpu_arch "$ENABLE_GPU_ARCH"
+else
+    # gfx90a:xnack+;gfx90a:xnack-;gfx942
+    set_gpu_arch "all"
+fi
+
 build_hipblaslt() {
    echo "Start build"

@@ -20,31 +30,31 @@ build_hipblaslt() {
    cd $COMPONENT_SRC
    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"

-    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
-       rebuild_lapack
-    fi
-
-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        # gfx90a:xnack+;gfx90a:xnack-;gfx940;gfx941;gfx942
-        GPU_TARGETS=all
+    # hipblaslt requiring Python 3.12.
+    # Point CMake to that explicit location and adjust LD_LIBRARY_PATH.
+    if [ "$DISTRO_ID" = "rhel-8.8" ] || [ "$DISTRO_NAME" == "sles" ] || \
+        [ "$DISTRO_ID" = "rhel-9.1" ] || [ "$DISTRO_ID" = "almalinux-8.10" ] || \
+        [ "$DISTRO_ID" = "debian-10" ]; then
+        EXTRA_PYTHON_PATH=/opt/Python-3.12.7
+        EXTRA_CMAKE_OPTIONS="-DPython_ROOT=/opt/Python-3.12.7"
+        export LD_LIBRARY_PATH=${EXTRA_PYTHON_PATH}/lib
    fi

    init_rocm_common_cmake_params
-    CXX=$(set_build_variables CXX)\
+    CXX=$(set_build_variables __CXX__)\
    cmake \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        ${LAUNCHER_FLAGS} \
-        "${rocm_math_common_cmake_params[@]}" \
+         "${rocm_math_common_cmake_params[@]}" \
        -DTensile_LOGIC= \
-        -DTensile_CODE_OBJECT_VERSION=default \
-        -DTensile_CPU_THREADS= \
+        -DTensile_CODE_OBJECT_VERSION=4 \
+        -DTensile_CPU_THREADS=$((PROC / 4)) \
        -DTensile_LIBRARY_FORMAT=msgpack \
        -DBUILD_CLIENTS_SAMPLES=ON \
        -DBUILD_CLIENTS_TESTS=ON \
+        -DLINK_BLIS=ON \
        -DBUILD_CLIENTS_BENCHMARKS=ON \
        -DBUILD_ADDRESS_SANITIZER="${ADDRESS_SANITIZER}" \
+        ${EXTRA_CMAKE_OPTIONS} \
        "$COMPONENT_SRC"

    cmake --build "$BUILD_DIR" -- -j${PROC}
@@ -52,8 +62,7 @@ build_hipblaslt() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
-
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
    show_build_cache_stats
 }

@@ -66,7 +75,7 @@ clean_hipblaslt() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_hipblaslt ;;
+    build) build_hipblaslt; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hipblaslt ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hipcc.sh
+++ b/tools/rocm-build/build_hipcc.sh
@@ -9,6 +9,8 @@ printUsage() {
    echo "Options:"
    echo "  -a,  --address_sanitizer  Enable address sanitizer"
    echo "  -c,  --clean              Clean output and delete all intermediate work"
+    echo "  -w,  --wheel              Creates python wheel package of hipcc. 
+                                      It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
                            type referred to by pkg_type"
    echo "  -r,  --release            Makes a release build"
@@ -21,6 +23,7 @@ printUsage() {
 }


+## Build environment variables
 API_NAME=hipcc
 PROJ_NAME=$API_NAME

@@ -29,10 +32,11 @@ MAKEOPTS="$DASH_JAY"
 BUILD_TYPE="Debug"
 SHARED_LIBS="ON"
 BUILD_DIR=$(getBuildPath $API_NAME)
-PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
-PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
+PACKAGE_DEB=$(getPackageRoot)/deb/$PROJ_NAME
+PACKAGE_RPM=$(getPackageRoot)/rpm/$PROJ_NAME
 PACKAGE_SRC="$(getSrcPath)"

+#parse the arguments
 VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,address_sanitizer,static,outdir,wheel:,package: -- "$@"`
 eval set -- "$VALID_STR"

@@ -45,13 +49,15 @@ do
            TARGET="clean" ;;
        (-o  | --outdir)
            TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 1 ;;
+        (-w  | --wheel)
+            WHEEL_PACKAGE=true ;;
        (-r  | --release)
            BUILD_TYPE="RelWithDebInfo" ;;
        (-s | --static)
            SHARED_LIBS="OFF" ;;
        (-h  | --help)
            printUsage ; exit 0 ;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        (*)
            echo "Invalid option [$1]" >&2; printUsage; exit 1 ;;
    esac
@@ -87,11 +93,12 @@ build() {
    fi

    cmake \
+        ${GEN_NINJA} \
        -DBUILD_SHARED_LIBS=$SHARED_LIBS \
        $(rocm_cmake_params) \
        $(rocm_common_cmake_params) \
        -DHIPCC_BACKWARD_COMPATIBILITY=OFF \
-        -DCMAKE_INSTALL_PREFIX="$OUT_DIR" \
+        -DCMAKE_INSTALL_PREFIX="$ROCM_PATH" \
        $HIPCC_ROOT
    popd

@@ -118,10 +125,20 @@ print_output_directory() {
 }

 case $TARGET in
-    (clean) clean ;;
-    (build) build ; copy_hipcc_sources ;;
-    (outdir) print_output_directory ;;
-    (*) die "Invalid target $TARGET" ;;
+    (clean)
+        clean
+        ;;
+    (build)
+        build
+        build_wheel "$BUILD_DIR" "$PROJ_NAME"
+        copy_hipcc_sources
+        ;;
+    (outdir)
+        print_output_directory
+        ;;
+    (*)
+        die "Invalid target $TARGET"
+        ;;
 esac

 echo "Operation complete"
--- a/tools/rocm-build/build_hipcub.sh
+++ b/tools/rocm-build/build_hipcub.sh
@@ -2,7 +2,7 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipCUB

@@ -17,26 +17,22 @@ build_hipcub() {
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
         set_asan_env_vars
         set_address_sanitizer_on
+         # ASAN packaging is not required for HIPCUB, since its header only package
+         # Setting the asan_cmake_params to false will disable ASAN packaging
         ASAN_CMAKE_PARAMS="false"
    fi

    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
    init_rocm_common_cmake_params

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908:xnack-;gfx90a:xnack-;gfx90a:xnack+;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
-    fi
-
-    CXX=$(set_build_variables CXX)\
+    CXX=$(set_build_variables __CXX__)\
    cmake \
+        ${GEN_NINJA} \
        ${LAUNCHER_FLAGS} \
-        "${rocm_math_common_cmake_params[@]}" \
+	    "${rocm_math_common_cmake_params[@]}" \
        -DCMAKE_MODULE_PATH="${ROCM_PATH}/lib/cmake/hip;${ROCM_PATH}/hip/cmake" \
        -Drocprim_DIR="${ROCM_PATH}/rocprim"  \
        -DBUILD_TEST=ON \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        "$COMPONENT_SRC"

    cmake --build "$BUILD_DIR" -- -j${PROC}
@@ -44,7 +40,7 @@ build_hipcub() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -58,7 +54,7 @@ clean_hipcub() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_hipcub ;;
+    build) build_hipcub; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hipcub ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hipfft.sh
+++ b/tools/rocm-build/build_hipfft.sh
@@ -2,13 +2,17 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipFFT

 build_hipfft() {
    echo "Start build"

+    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
+        ack_and_skip_static
+    fi
+
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
       set_asan_env_vars
       set_address_sanitizer_on
@@ -18,17 +22,10 @@ build_hipfft() {
    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
    init_rocm_common_cmake_params

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908;gfx90a;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
-    fi
-
    cmake \
-        -DCMAKE_CXX_COMPILER=$(set_build_variables CXX) \
+        "$(set_build_variables __CMAKE_CXX_PARAMS__)" \
        ${LAUNCHER_FLAGS} \
-        "${rocm_math_common_cmake_params[@]}" \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
+	    "${rocm_math_common_cmake_params[@]}" \
        -DCMAKE_MODULE_PATH="${ROCM_PATH}/lib/cmake/hip" \
        -DCMAKE_SKIP_BUILD_RPATH=TRUE \
        -DBUILD_ADDRESS_SANITIZER="${ADDRESS_SANITIZER}" \
@@ -41,7 +38,7 @@ build_hipfft() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR  && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -55,7 +52,7 @@ clean_hipfft() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_hipfft ;;
+    build) build_hipfft; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hipfft ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hipfort.sh
+++ b/tools/rocm-build/build_hipfort.sh
@@ -1,13 +1,18 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipfort

 build_hipfort() {
    echo "Start build"

+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+       set_asan_env_vars
+       set_address_sanitizer_on
+    fi
+
    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
        ack_and_skip_static
    fi
@@ -26,7 +31,7 @@ build_hipfort() {
        -DCMAKE_Fortran_FLAGS_DEBUG="" \
        ${LAUNCHER_FLAGS} \
        -DROCM_SYMLINK_LIBS=OFF \
-	-DCMAKE_INSTALL_PREFIX=${ROCM_PATH} \
+        -DCMAKE_INSTALL_PREFIX=${ROCM_PATH} \
        -DHIPFORT_AR="${ROCM_PATH}/${ROCM_LLVMDIR}/bin/llvm-ar" \
        -DHIPFORT_RANLIB="${ROCM_PATH}/${ROCM_LLVMDIR}/bin/llvm-ranlib" \
        "$COMPONENT_SRC"
@@ -36,7 +41,7 @@ build_hipfort() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -50,7 +55,7 @@ clean_hipfort() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_hipfort ;;
+    build) build_hipfort; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hipfort ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hipify_clang.sh
+++ b/tools/rocm-build/build_hipify_clang.sh
@@ -10,22 +10,33 @@ printUsage() {
    echo "  -c,  --clean              Clean output and delete all intermediate work"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
    type referred to by pkg_type"
-    echo "  -h,  --help             Prints this help"
-    echo "  -r,  --release          Make a release build"
+    echo "  -r,  --release            Make a release build"
+    echo "  -n,  --skip_hipify_tests  Skip hipify-clang testing"
    echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
+    echo "  -s,  --static             Component/Build does not support static builds just accepting this param & ignore. No effect of the param on this build"
+    echo "  -i,  --clang_headers      Install clang headers"
+    echo "  -w,  --wheel              Creates python wheel package of hipify-clang.
+                                      It needs to be used along with -r option"
+    echo "  -h,  --help               Prints this help"
    echo

    return 0
 }

+# Build Environmental Variables
+PROJ_NAME="hipify"
 TARGET="build"
-MAKEOPTS="$DASH_JAY"
+NINJAOPTS="$DASH_JAY"
 HIPIFY_CLANG_BUILD_DIR="$(getBuildPath $HIPIFY_ROOT)"
 BUILD_TYPE="Debug"
 PACKAGE_ROOT="$(getPackageRoot)"
 HIPIFY_CLANG_HASH=""
 LIGHTNING_PATH="$ROCM_INSTALL_PATH/llvm"
+LIGHTNING_BUILD_PATH="$PACKAGE_ROOT/build/lightning"
+RUN_HIPIFY_TESTS=true
+CUDA_DEFAULT_VERSION="12.3.2"
+CUDNN_DEFAULT_VERSION="9.2.0"
+GCC_MIN_VERSION="9.2"
 ADDRESS_SANITIZER=false
 INSTALL_CLANG_HEADERS="OFF"
 DEB_PATH="$(getDebPath hipify)"
@@ -36,7 +47,8 @@ MAKETARGET="deb"
 PKGTYPE="deb"


-VALID_STR=`getopt -o hcraso: --long help,clean,release,static,address_sanitizer,outdir: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcrnawsio: --long help,clean,release,skip_hipify_tests,wheel,static,address_sanitizer,clang_headers,outdir: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
@@ -48,15 +60,21 @@ do
                TARGET="clean" ; ((CLEAN_OR_OUT|=1)) ; shift ;;
        (-r | --release)
                BUILD_TYPE="RelWithDebInfo" ; shift ;;
+        (-n | --skip_hipify_tests)
+                RUN_HIPIFY_TESTS=false; shift ;;
        (-a | --address_sanitizer)
                set_asan_env_vars
                set_address_sanitizer_on
                ADDRESS_SANITIZER=true ; shift ;;
        (-s | --static)
                ack_and_skip_static ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; shift ;;
+        (-i | --clang_headers)
+                INSTALL_CLANG_HEADERS="ON" ; shift ;;
        (-o | --outdir)
                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        (*)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
    esac
@@ -79,6 +97,7 @@ clean_hipify() {
 }

 package_hipify() {
+    # set-up dirs
    if [ "$PACKAGEEXT" = "deb" ]; then
        rm -rf "$DEB_PATH"
        mkdir -p "$DEB_PATH"
@@ -89,8 +108,9 @@ package_hipify() {
        mkdir -p "$RPM_PATH"
    fi

+    # make the pkg
    pushd "$HIPIFY_CLANG_BUILD_DIR"
-    make $MAKEOPTS package_hipify-clang
+    ninja $NINJAOPTS package_hipify-clang
    popd

    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$DEB_PATH"  $HIPIFY_CLANG_BUILD_DIR/hipify*.deb
@@ -102,7 +122,38 @@ build_hipify() {
    mkdir -p "$HIPIFY_CLANG_BUILD_DIR"

    pushd "$HIPIFY_CLANG_BUILD_DIR"
+
+    # Check the installed GCC version
+    INSTALLED_GCC_VERSION=$(gcc --version | head -n 1 | sed -E 's/[^0-9]*([0-9]+\.[0-9]+).*/\1/')
+    if echo "$INSTALLED_GCC_VERSION $GCC_MIN_VERSION" | awk '{exit !($1 < $2)}'; then
+        RUN_HIPIFY_TESTS=false
+        echo "Minimum required GCC version: $GCC_MIN_VERSION"
+        echo "Installed GCC version $INSTALLED_GCC_VERSION does not meet the minimum GCC version requirement."
+        echo "Skipping hipify tests"
+    fi
+
+    if $ADDRESS_SANITIZER ; then
+        echo "Skipping hipify tests becasue of Address Sanitizer"
+        RUN_HIPIFY_TESTS=false
+    fi
+
+    if $RUN_HIPIFY_TESTS ; then
+        # TODO: Add option for user defined cuda version?
+        CUDA_VERSION="${CUDA_DEFAULT_VERSION}"
+        CUDNN_VERSION="${CUDNN_DEFAULT_VERSION}"
+
+        if [[ "$DISTRO_ID" == "rhel-8"* || "$DISTRO_NAME" == "sles" || "$DISTRO_ID" == "debian-10" ]]; then
+            EXTRA_PYTHON_PATH=/opt/Python-3.8.13
+            export LD_LIBRARY_PATH=${EXTRA_PYTHON_PATH}/lib:$LD_LIBRARY_PATH
+        fi
+
+        echo "Copy FileCheck into ROCM_INSTALL_PATH"
+        cp "$LIGHTNING_BUILD_PATH/bin/FileCheck" "$LIGHTNING_PATH/bin/FileCheck"
+    fi
+
    cmake \
+        ${GEN_NINJA} \
+        -DHIPIFY_CLANG_TESTS="$RUN_HIPIFY_TESTS" \
        -DCMAKE_BUILD_TYPE="$BUILD_TYPE" \
        $(rocm_common_cmake_params) \
        -DCMAKE_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
@@ -110,10 +161,20 @@ build_hipify() {
        -DCMAKE_PREFIX_PATH="$LIGHTNING_PATH" \
        -DADDRESS_SANITIZER="$ADDRESS_SANITIZER" \
        -DHIPIFY_INSTALL_CLANG_HEADERS="$INSTALL_CLANG_HEADERS" \
+        -DCUDA_TOOLKIT_ROOT_DIR="/usr/local/cuda-${CUDA_DEFAULT_VERSION}" \
+        -DCUDA_DNN_ROOT_DIR="/usr/local/cuDNN/${CUDNN_DEFAULT_VERSION}" \
+        -DCUDA_CUB_ROOT_DIR="/usr/local/cuda-${CUDA_DEFAULT_VERSION}" \
+        -DLLVM_EXTERNAL_LIT="${LIGHTNING_BUILD_PATH}/bin/llvm-lit" \
        $HIPIFY_ROOT

-    cmake --build . -- $MAKEOPTS install
+    if $RUN_HIPIFY_TESTS ; then
+        echo "Running hipify tests"
+        cmake --build . -- $NINJAOPTS test-hipify
+    fi
+
+    cmake --build . -- $NINJAOPTS install
    popd
+
    pushd "$HIPIFY_ROOT"
        HIPIFY_CLANG_HASH=`git describe --dirty --long --match [0-9]* --always`
    popd
@@ -132,10 +193,20 @@ print_output_directory() {
 }

 case $TARGET in
-    (clean) clean_hipify ;;
-    (build) build_hipify; package_hipify ;;
-    (outdir) print_output_directory ;;
-    (*) die "Invalid target $TARGET" ;;
+    (clean)
+        clean_hipify
+        ;;
+    (build)
+        build_hipify
+        package_hipify
+        build_wheel "$HIPIFY_CLANG_BUILD_DIR" "$PROJ_NAME"
+        ;;
+    (outdir)
+        print_output_directory
+        ;;
+    (*)
+        die "Invalid target $TARGET"
+        ;;
 esac

 echo "Operation complete"
--- a/tools/rocm-build/build_hiprand.sh
+++ b/tools/rocm-build/build_hiprand.sh
@@ -1,7 +1,7 @@
 #!/bin/bash
 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipRAND

@@ -41,17 +41,10 @@ build_hiprand() {

    init_rocm_common_cmake_params

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908:xnack-;gfx90a:xnack-;gfx90a:xnack+;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
-    fi
-
-    CXX=$(set_build_variables CXX)\
+    CXX=$(set_build_variables __CXX__)\
    cmake \
        ${LAUNCHER_FLAGS} \
 	    "${rocm_math_common_cmake_params[@]}" \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        -DBUILD_SHARED_LIBS=$SHARED_LIBS \
        -DBUILD_TEST=ON \
        -DBUILD_BENCHMARK=ON \
@@ -67,7 +60,7 @@ build_hiprand() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/  && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
 }

 clean_hiprand() {
@@ -89,7 +82,7 @@ print_output_directory() {
 }

 case $TARGET in
-    build) build_hiprand ;;
+    build) build_hiprand; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hiprand ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hipsolver.sh
+++ b/tools/rocm-build/build_hipsolver.sh
@@ -2,7 +2,7 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipSOLVER

@@ -10,12 +10,12 @@ build_hipsolver() {
    echo "Start build"

    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
-        CXX_FLAG="-DCMAKE_CXX_COMPILER=${ROCM_PATH}/llvm/bin/clang++"
+        CXX_FLAG=$(set_build_variables __CMAKE_CXX_PARAMS__)
    fi

    cd $COMPONENT_SRC

-    CXX="amdclang++"
+    CXX=$(set_build_variables __AMD_CLANG_++__)
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
       set_asan_env_vars
       set_address_sanitizer_on
@@ -39,7 +39,7 @@ build_hipsolver() {
    init_rocm_common_cmake_params
    cmake \
        -DUSE_CUDA=OFF \
-	-DCMAKE_CXX_COMPILER=${CXX} \
+	    -DCMAKE_CXX_COMPILER=${CXX} \
        ${LAUNCHER_FLAGS} \
        "${rocm_math_common_cmake_params[@]}" \
        -DBUILD_SHARED_LIBS=$SHARED_LIBS \
@@ -55,7 +55,7 @@ build_hipsolver() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -69,7 +69,7 @@ clean_hipsolver() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_hipsolver ;;
+    build) build_hipsolver; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hipsolver ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hipsparse.sh
+++ b/tools/rocm-build/build_hipsparse.sh
@@ -2,7 +2,7 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 PATH=${ROCM_PATH}/bin:$PATH
 set_component_src hipSPARSE
@@ -10,12 +10,12 @@ set_component_src hipSPARSE
 build_hipsparse() {
    echo "Start build"

-    CXX="g++"
+    CXX=$(set_build_variables __G_++__)
    CXX_FLAG=

    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
-        CXX="${ROCM_PATH}/llvm/bin/clang++"
-        CXX_FLAG="-DCMAKE_CXX_COMPILER=${ROCM_PATH}/llvm/bin/clang++"
+        CXX=$(set_build_variables __CXX__)
+        CXX_FLAG=$(set_build_variables __CMAKE_CXX_PARAMS__)
    fi

    cd $COMPONENT_SRC
@@ -54,7 +54,7 @@ build_hipsparse() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -68,7 +68,7 @@ clean_hipsparse() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_hipsparse ;;
+    build) build_hipsparse; build_wheel  ;;
    outdir) print_output_directory ;;
    clean) clean_hipsparse ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hipsparselt.sh
+++ b/tools/rocm-build/build_hipsparselt.sh
@@ -1,10 +1,18 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipSPARSELt

+disable_debug_package_generation
+# if ENABLE_GPU_ARCH is set in env by Job parameter ENABLE_GPU_ARCH, then set GFX_ARCH to that value
+if [ -n "$ENABLE_GPU_ARCH" ]; then
+    set_gpu_arch "$ENABLE_GPU_ARCH"
+else
+    # gfx90a:xnack+;gfx90a:xnack-;gfx942
+    set_gpu_arch "all"
+fi
 while [ "$1" != "" ];
 do
    case $1 in
@@ -18,6 +26,31 @@ do
    shift 1
 done

+create_blis_link()
+{
+    #find the pre-installed blis library and create the link under $BUILD_DIR/deps/blis
+    BLIS_REF_ROOT="$BUILD_DIR/deps/blis"
+    mkdir -p "$BLIS_REF_ROOT"/include
+    for blis_path in "/opt/AMD/aocl/aocl-linux-gcc-4.2.0/gcc" \
+                     "/opt/AMD/aocl/aocl-linux-gcc-4.1.0/gcc" \
+                     "/opt/AMD/aocl/aocl-linux-gcc-4.0.0/gcc";
+    do
+        if [ -e "${blis_path}/lib_ILP64/libblis-mt.a" ] ; then
+            ln -sf "${blis_path}/include_ILP64" "${BLIS_REF_ROOT}/include/blis"
+            ln -sf "${blis_path}/lib_ILP64" "${BLIS_REF_ROOT}/lib"
+            return
+        fi
+    done
+
+    if [[ -e "/usr/local/lib/libblis.a" ]]; then
+        ln -sf /usr/local/include/blis ${BLIS_REF_ROOT}/include/blis
+        ln -sf /usr/local/lib ${BLIS_REF_ROOT}/lib
+        return
+    fi
+    echo "error: BLIS lib not found" >&2
+    return 1
+}
+
 build_hipsparselt() {
    echo "Start build"

@@ -32,23 +65,19 @@ build_hipsparselt() {

    cd $COMPONENT_SRC
    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
-    init_rocm_common_cmake_params
-
-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        # gfx940;gfx941;gfx942
-        GPU_TARGETS=all
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+        create_blis_link
+        EXTRA_CMAKE_OPTIONS=("-DLINK_BLIS=ON" "-DBUILD_DIR=${BUILD_DIR}")
    fi

-    FC=gfortran \
-    CXX="${ROCM_PATH}/bin/hipcc" \
+    init_rocm_common_cmake_params
+
+    CXX=$(set_build_variables __CXX__) \
    cmake \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        ${LAUNCHER_FLAGS} \
-        "${rocm_math_common_cmake_params[@]}" \
+         "${rocm_math_common_cmake_params[@]}" \
        -DTensile_LOGIC= \
-        -DTensile_CODE_OBJECT_VERSION=default \
+        -DTensile_CODE_OBJECT_VERSION=4 \
        -DTensile_CPU_THREADS= \
        -DTensile_LIBRARY_FORMAT=msgpack \
        -DBUILD_CLIENTS_SAMPLES=ON \
@@ -56,13 +85,14 @@ build_hipsparselt() {
        -DBUILD_CLIENTS_BENCHMARKS=ON \
        -DCMAKE_INSTALL_PREFIX=${ROCM_PATH} \
        -DBUILD_ADDRESS_SANITIZER="${ADDRESS_SANITIZER}" \
+        ${EXTRA_CMAKE_OPTIONS[@]} \
        "$COMPONENT_SRC"

    cmake --build "$BUILD_DIR" -- -j${PROC}
    cmake --build "$BUILD_DIR" -- install
    cmake --build "$BUILD_DIR" -- package

-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    $SCCACHE_BIN -s || echo "Unable to display sccache stats"
 }
@@ -86,7 +116,7 @@ print_output_directory() {
 }

 case $TARGET in
-    build) build_hipsparselt ;;
+    build) build_hipsparselt; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hipsparselt ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_hiptensor.sh
+++ b/tools/rocm-build/build_hiptensor.sh
@@ -2,9 +2,10 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src hipTensor
+disable_debug_package_generation

 build_hiptensor() {
    echo "Start build hipTensor"
@@ -22,17 +23,11 @@ build_hiptensor() {
    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
    init_rocm_common_cmake_params

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908:xnack-;gfx90a:xnack-;gfx90a:xnack+;gfx940;gfx941;gfx942"
-    fi
-
    cmake \
        -B "${BUILD_DIR}" \
        "${rocm_math_common_cmake_params[@]}" \
-        $(set_build_variables CMAKE_C_CXX) \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
+        "$(set_build_variables __CMAKE_CC_PARAMS__)" \
+        "$(set_build_variables __CMAKE_CXX_PARAMS__)" \
        ${LAUNCHER_FLAGS} \
        "$COMPONENT_SRC"

@@ -41,7 +36,7 @@ build_hiptensor() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p "$PACKAGE_DIR" && cp ${BUILD_DIR}/*.${PKGTYPE} "$PACKAGE_DIR"
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -55,7 +50,7 @@ clean_hiptensor() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_hiptensor ;;
+    build) build_hiptensor; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_hiptensor ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_lightning.sh
+++ b/tools/rocm-build/build_lightning.sh
@@ -42,6 +42,7 @@ DEB_PATH="$(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"
@@ -59,7 +60,7 @@ MAKETARGET="deb"
 ASSERT_LLVM_VERSION_MAJOR=""
 ASSERT_LLVM_VERSION_MINOR=""

-SKIP_LIT_TESTS=0
+SKIP_LIT_TESTS=1
 BUILD_MANPAGES="ON"
 STATIC_FLAG=

@@ -149,6 +150,7 @@ 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"
@@ -330,6 +332,15 @@ 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
@@ -1147,4 +1158,9 @@ 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"
--- a/tools/rocm-build/build_miopen-deps.sh
+++ b/tools/rocm-build/build_miopen-deps.sh
@@ -1,7 +1,7 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src MIOpen

@@ -19,18 +19,18 @@ build_miopen_mlir() {
    mkdir build && cd build
    cmake \
        -G Ninja \
-        -DCMAKE_C_COMPILER="${ROCM_PATH}/llvm/bin/clang" \
-        -DCMAKE_CXX_COMPILER="${ROCM_PATH}/llvm/bin/clang++" \
+        -DCMAKE_C_COMPILER="$(set_build_variables __CLANG__)" \
+        -DCMAKE_CXX_COMPILER="$(set_build_variables __CLANG++__)" \
        -DCMAKE_BUILD_TYPE=Release \
-	-DCMAKE_PREFIX_PATH="${ROCM_PATH};${HOME}/miopen-deps" \
+        -DCMAKE_PREFIX_PATH="${ROCM_PATH};${HOME}/miopen-deps" \
        -DCMAKE_INSTALL_PREFIX="$ROCM_PATH" \
        -DBUILD_FAT_LIBROCKCOMPILER=1 \
-       ..
+        ..
    cmake --build . -- librockCompiler -j${PROC}
    cmake --build . -- install

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-
+    
    echo "Finished building rocMLIR"
 }

@@ -43,12 +43,6 @@ build_miopen_deps() {
    echo "Start build"
    cd "$COMPONENT_SRC"

-    # Update rocm-recipes for boost link
-    # https://github.com/ROCm/MIOpen/pull/3457/files
-    sed -i 's/329203d79f9fe77ae5d0d742af0966bc57f4dfc8/92c6695449c85887962f45509b376f2eb0d284f7/g' \
-        rbuild.ini \
-        install_deps.cmake \
-        dev-requirements.txt
    # Commenting the rocMLIR & composable_kernel from requirements.txt
    sed -i '/ROCm\/rocMLIR@\|ROCm\/composable_kernel@/s/^/#/' requirements.txt
    # Extract MLIR commit from requirements.txt
@@ -56,7 +50,7 @@ build_miopen_deps() {


    pip3 install https://github.com/RadeonOpenCompute/rbuild/archive/master.tar.gz
-    PATH="${PATH}:${ROCM_PATH}:${HOME}/.local/bin" rbuild prepare -d "$HOME/miopen-deps" --cxx=${ROCM_PATH}/llvm/bin/clang++ --cc ${ROCM_PATH}/llvm/bin/clang
+    PATH="${PATH}:${ROCM_PATH}:${HOME}/.local/bin" rbuild prepare -d "$HOME/miopen-deps" --cxx="$(set_build_variables __CLANG++__)" --cc "$(set_build_variables __CLANG__)"
    build_miopen_mlir "$MLIR_COMMIT"

    show_build_cache_stats
--- a/tools/rocm-build/build_miopen-hip.sh
+++ b/tools/rocm-build/build_miopen-hip.sh
@@ -1,7 +1,7 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src MIOpen

@@ -9,6 +9,8 @@ PACKAGE_DIR=${PACKAGE_DIR%\/*}/miopen-hip
 DEB_PATH=$PACKAGE_DIR
 RPM_PATH=$PACKAGE_DIR

+disable_debug_package_generation
+
 build_miopen_hip() {
    echo "Start build"

@@ -31,19 +33,19 @@ build_miopen_hip() {
        "${rocm_math_common_cmake_params[@]}" \
        -DMIOPEN_BACKEND=HIP \
        -DMIOPEN_OFFLINE_COMPILER_PATHS_V2=1 \
-        -DCMAKE_CXX_COMPILER="${ROCM_PATH}/llvm/bin/clang++" \
-        -DCMAKE_C_COMPILER="${ROCM_PATH}/llvm/bin/clang" \
-        -DCMAKE_PREFIX_PATH="${ROCM_PATH};${ROCM_PATH}/hip;${HOME}/miopen-deps" \
+        -DCMAKE_CXX_COMPILER=$(set_build_variables __CLANG++__) \
+        -DCMAKE_C_COMPILER=$(set_build_variables __CLANG__) \
+        -DCMAKE_PREFIX_PATH="${ROCM_PATH};${ROCM_PATH}/hip;/tmp/miopen-deps" \
        -DHIP_OC_COMPILER="${ROCM_PATH}/bin/clang-ocl" \
        -DMIOPEN_TEST_DISCRETE=OFF \
-        "$COMPONENT_SRC"
+	    "$COMPONENT_SRC"

    cmake --build "$BUILD_DIR" -- -j${PROC}
    cmake --build "$BUILD_DIR" -- install
    cmake --build "$BUILD_DIR" -- package

    rm -rf $BUILD_DIR/_CPack_Packages/ && find $BUILD_DIR -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -62,7 +64,7 @@ checkout_lfs() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_miopen_hip ;;
+    build) build_miopen_hip; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_miopen_hip ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_mivisionx.sh
+++ b/tools/rocm-build/build_mivisionx.sh
@@ -1,39 +1,44 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src MIVisionX
 BUILD_DEV=ON

 build_mivisionx() {
+
+    if [ "$DISTRO_ID" = "mariner-2.0" ] || [ "$DISTRO_ID" = "azurelinux-3.0" ] ; then
+        echo "Not building mivisionx for ${DISTRO_ID}. Exiting..."
+        return 0
+    fi
+
    echo "Start build"

    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
        ack_and_skip_static
    fi

+    CXX=$(set_build_variables __AMD_CLANG_++__)
    mkdir -p $BUILD_DIR && cd $BUILD_DIR
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
       set_asan_env_vars
       set_address_sanitizer_on
+       # Setting BUILD_DEV to OFF. This will prevent the installation of
+       # header files, other files in share,libexec folder. ASAN pkg doesn't need this
       BUILD_DEV=OFF
    fi

-    init_rocm_common_cmake_params
+    echo "C compiler: $CC"
+    echo "CXX compiler: $CXX"

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908;gfx90a;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
-    fi
+    init_rocm_common_cmake_params

    cmake \
        "${rocm_math_common_cmake_params[@]}" \
        -DROCM_PATH="$ROCM_PATH" \
        -DBUILD_DEV=$BUILD_DEV \
        -DCMAKE_INSTALL_LIBDIR=$(getInstallLibDir) \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        -DROCM_DEP_ROCMCORE=ON \
        -DROCAL_PYTHON=OFF \
        ${LAUNCHER_FLAGS} \
@@ -44,7 +49,7 @@ build_mivisionx() {
    cpack -G ${PKGTYPE^^}

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -58,7 +63,7 @@ clean_mivisionx() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_mivisionx ;;
+    build) build_mivisionx; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_mivisionx ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_omniperf.sh
+++ b/tools/rocm-build/build_omniperf.sh
@@ -0,0 +1,149 @@
+#!/bin/bash
+
+source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
+
+printUsage() {
+    echo
+    echo "Usage: ${BASH_SOURCE##*/} [options ...]"
+    echo
+    echo "Options:"
+    echo "  -c,  --clean              Clean output and delete all intermediate work"
+    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -p,  --package <type>     Specify packaging format"
+    echo "  -r,  --release            Make a release build instead of a debug build"
+    echo "  -a,  --address_sanitizer  Enable address sanitizer"
+    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
+                                      type referred to by pkg_type"
+    echo "  -w,  --wheel              Creates python wheel package of omniperf.
+                                      It needs to be used along with -r option"
+    echo "  -h,  --help               Prints this help"
+    echo
+    echo "Possible values for <type>:"
+    echo "  deb -> Debian format (default)"
+    echo "  rpm -> RPM format"
+    echo
+
+    return 0
+}
+
+## Build environment variables
+API_NAME="omniperf"
+PROJ_NAME="$API_NAME"
+LIB_NAME="lib${API_NAME}"
+TARGET="build"
+MAKETARGET="deb"
+PACKAGE_ROOT="$(getPackageRoot)"
+PACKAGE_LIB="$(getLibPath)"
+# PACKAGE_INCLUDE="$(getIncludePath)"
+BUILD_DIR="$(getBuildPath $API_NAME)"
+PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
+PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
+BUILD_TYPE="Debug"
+MAKE_OPTS="$DASH_JAY -C $BUILD_DIR"
+SHARED_LIBS="ON"
+CLEAN_OR_OUT=0;
+MAKETARGET="deb"
+PKGTYPE="deb"
+
+#parse the arguments
+VALID_STR=$(getopt -o hcraso:p:w --long help,clean,release,static,address_sanitizer,outdir:,package:,wheel -- "$@")
+eval set -- "$VALID_STR"
+
+while true ;
+do
+    #echo "parocessing $1"
+    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 $PROJ_NAME"
+    rm -rf "$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)"
+
+        #install python deps
+        #python3 -m pip install -t ${BUILD_DIR}/python-libs -r ${OMNIPERF_ROOT}/requirements.txt
+        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
+}
+
+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 "$TARGET" in
+    (clean) clean ;;
+    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
+esac
+
+echo "Operation complete"
--- a/tools/rocm-build/build_omnitrace.sh
+++ b/tools/rocm-build/build_omnitrace.sh
@@ -0,0 +1,243 @@
+#!/bin/bash
+
+source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
+
+printUsage() {
+    echo
+    echo "Usage: ${BASH_SOURCE##*/} [options ...]"
+    echo
+    echo "Options:"
+    echo "  -c,  --clean              Clean output and delete all intermediate work"
+    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -p,  --package <type>     Specify packaging format"
+    echo "  -r,  --release            Make a release build instead of a debug build"
+    echo "  -a,  --address_sanitizer  Enable address sanitizer"
+    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
+                                      type referred to by pkg_type"
+    echo "  -w,  --wheel              Creates python wheel package of omnitrace.
+                                      It needs to be used along with -r option"
+    echo "  -h,  --help               Prints this help"
+    echo
+    echo "Possible values for <type>:"
+    echo "  deb -> Debian format (default)"
+    echo "  rpm -> RPM format"
+    echo
+
+    return 0
+}
+
+## Build environment variables
+API_NAME="omnitrace"
+PROJ_NAME="$API_NAME"
+LIB_NAME="lib${API_NAME}"
+TARGET="build"
+MAKETARGET="deb"
+PACKAGE_ROOT="$(getPackageRoot)"
+PACKAGE_LIB="$(getLibPath)"
+# PACKAGE_INCLUDE="$(getIncludePath)"
+BUILD_DIR="$(getBuildPath $API_NAME)"
+PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
+PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
+# PACKAGE_PREFIX="$ROCM_INSTALL_PATH"
+BUILD_TYPE="Debug"
+MAKE_OPTS="-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
+    #echo "parocessing $1"
+    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
+        # set_asan_env_vars
+        # set_address_sanitizer_on
+
+        ASAN=1
+        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 $PROJ_NAME"
+    rm -rf "$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
+
+    cd $OMNITRACE_ROOT || exit
+
+    echo "Current submodule status"
+    git submodule status
+    echo "Cached (old) submodule status"
+    git submodule status --cached
+    cat .git/config
+
+    echo "Updating submodules"
+    git submodule init
+    # copy the new URL to your local config
+    git submodule sync --recursive
+    # force update the submodule from the new URL
+    git submodule update --init --recursive --force
+
+    echo "Updated submodule status"
+    git submodule status
+    cat .git/config
+
+    echo "Build directory: $BUILD_DIR"
+    pushd "$BUILD_DIR" || exit
+    print_lib_type $SHARED_LIBS
+
+    ELFUTIL_URL="https://compute-artifactory.amd.com/artifactory/rocm-generic-local/dev-tools/omnitrace/elfutils-0.188.tar.bz2"
+    BINUTIL_URL="https://compute-artifactory.amd.com/artifactory/rocm-generic-local/dev-tools/omnitrace/binutils-2.40.tar.gz"
+
+    echo "ROCm CMake Params: $(rocm_cmake_params)"
+    echo "ROCm Common CMake Params: $(rocm_common_cmake_params)"
+    echo "ELFUTIL_URL=$ELFUTIL_URL, BINUTIL_URL=$BINUTIL_URL"
+
+    if [ $ASAN == 1 ]; then
+        echo "Address Sanitizer path"
+
+        # Commenting out the below cmake command as it is not working as expected
+        # LD_LIBRARY_PATH=$ROCM_INSTALL_PATH/lib/asan:$LD_LIBRARY_PATH
+        # cmake \
+        #     $(rocm_cmake_params) \
+        #     $(rocm_common_cmake_params) \
+        #     -DOMNITRACE_BUILD_{LIBUNWIND,DYNINST}=ON \
+        #     -DDYNINST_BUILD_{TBB,BOOST,ELFUTILS,LIBIBERTY}=ON \
+        #     -DAMDDeviceLibs_DIR="${ROCM_INSTALL_PATH}/lib/asan/cmake/AMDDeviceLibs" \
+        #     -Dhip_DIR="${ROCM_INSTALL_PATH}/lib/asan/cmake/hip" \
+        #     -Dhip-lang_DIR="${ROCM_INSTALL_PATH}/lib/asan/cmake/hip-lang" \
+        #     -Damd_comgr_DIR="${ROCM_INSTALL_PATH}/lib/asan/cmake/amd_comgr" \
+        #     -Dhsa-runtime64_DIR="${ROCM_INSTALL_PATH}/lib/asan/cmake/hsa-runtime64" \
+        #     -Dhsakmt_DIR="${ROCM_INSTALL_PATH}/lib/asan/cmake/hsakmt" \
+        #     -DROCM_PATH="${ROCM_INSTALL_PATH}/lib/asan" \
+        #     -Drocprofiler_ROOT_DIR="${ROCM_INSTALL_PATH}/lib/asan" \
+        #     -DCMAKE_HIP_COMPILER_ROCM_ROOT="${ROCM_INSTALL_PATH}" \
+        #     -DCMAKE_PREFIX_PATH="${ROCM_INSTALL_PATH};${ROCM_INSTALL_PATH}/lib/asan" \
+        #     -DCMAKE_LIBRARY_PATH="${ROCM_INSTALL_PATH}/lib/asan" \
+        #     -DCPACK_DEBIAN_PACKAGE_SHLIBDEPS=OFF \
+        #     "$OMNITRACE_ROOT"
+
+    else
+        cmake \
+            $(rocm_cmake_params) \
+            $(rocm_common_cmake_params) \
+            -DOMNITRACE_BUILD_{LIBUNWIND,DYNINST}=ON \
+            -DDYNINST_BUILD_{TBB,BOOST,ELFUTILS,LIBIBERTY}=ON \
+            -DElfUtils_DOWNLOAD_URL="$ELFUTIL_URL" \
+            -D{DYNINST,TIMEMORY}_BINUTILS_DOWNLOAD_URL="$BINUTIL_URL" \
+            "$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 "$TARGET" in
+    clean) clean ;;
+    build) build_omnitrace; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+    outdir) print_output_directory ;;
+    *) die "Invalid target $TARGET" ;;
+esac
+
+echo "Operation complete"
--- a/tools/rocm-build/build_opencl_icd_loader.sh
+++ b/tools/rocm-build/build_opencl_icd_loader.sh
@@ -0,0 +1,144 @@
+#!/bin/bash
+source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
+PROJ_NAME="OpenCL-ICD-Loader"
+TARGET="build"
+MAKEOPTS="$DASH_JAY"
+BUILD_TYPE="Debug"
+OPENCL_ICD_LOADER_BUILD_DIR="$(getBuildPath ${PROJ_NAME})"
+PACKAGE_ROOT="$(getPackageRoot)"
+PACKAGE_DEB="$PACKAGE_ROOT/deb/${PROJ_NAME,,}"
+PACKAGE_RPM="$PACKAGE_ROOT/rpm/${PROJ_NAME,,}"
+CLEAN_OR_OUT=0;
+PKGTYPE="deb"
+MAKETARGET="deb"
+
+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 "  -w,  --wheel              Creates python wheel package of opencl-icd-loader. 
+                                      It needs to be used along with -r option"
+    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 $PROJ_NAME"
+    rm -rf "$OPENCL_ICD_LOADER_BUILD_DIR"
+    rm -rf "$PACKAGE_DEB"
+    rm -rf "$PACKAGE_RPM"
+    rm -rf "$PACKAGE_ROOT/opencl-icd-loader"
+}
+
+build_opencl_icd_loader() {
+    echo "Building $PROJ_NAME"
+    if [ ! -e "$OPENCL_ICD_LOADER_ROOT/CMakeLists.txt" ]
+       then
+       echo "No $OPENCL_ICD_LOADER_ROOT/CMakeLists.txt file, skipping opencl icd loader" >&2
+       echo "No $OPENCL_ICD_LOADER_ROOT/CMakeLists.txt file, skipping opencl icd loader"
+       exit 0 # This is not an error
+    fi
+
+    mkdir -p "$OPENCL_ICD_LOADER_BUILD_DIR"
+    pushd "$OPENCL_ICD_LOADER_BUILD_DIR"
+    if [ ! -e Makefile ]; then
+        cmake \
+            -S "$OPENCL_ICD_LOADER_ROOT" \
+            $(rocm_cmake_params) \
+            $(rocm_common_cmake_params) \
+            -DENABLE_OPENCL_LAYERS="OFF" \
+            -DOPENCL_ICD_LOADER_HEADERS_DIR="$OPENCL_HEADERS_ROOT"
+    fi
+    cmake --build . -- $MAKEOPTS
+    echo "Installing opencl-icd-loader"
+    cmake --build . -- $MAKEOPTS install
+    popd
+}
+
+package_opencl_icd_loader() {
+    echo "Packaging $PROJ_NAME"
+    pushd "$OPENCL_ICD_LOADER_BUILD_DIR"
+    cmake --build . -- package
+    mkdir -p $PACKAGE_DEB
+    mkdir -p $PACKAGE_RPM
+    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_DEB" *.deb
+    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$PACKAGE_RPM" *.rpm
+    popd
+}
+
+print_output_directory() {
+    case ${PKGTYPE} in
+        ("deb")
+            echo ${PACKAGE_DEB};;
+        ("rpm")
+            echo ${PACKAGE_RPM};;
+        (*)
+            echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
+    esac
+    exit
+}
+
+#parse the arguments
+VALID_STR=`getopt -o hcraswlo:p: --long help,clean,release,wheel,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 ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; 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;; # end delimiter
+        (*)
+            echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
+    esac
+done
+
+case $TARGET in
+    (clean)
+        clean_opencl_icd_loader
+        ;;
+    (build)
+        build_opencl_icd_loader
+        package_opencl_icd_loader
+        build_wheel "$OPENCL_ICD_LOADER_BUILD_DIR" "$PROJ_NAME"
+        ;;
+    (outdir)
+        print_output_directory
+        ;;
+    (*)
+        die "Invalid target $TARGET"
+        ;;
+esac
+
+echo "Operation complete"
--- a/tools/rocm-build/build_opencl_on_rocclr.sh
+++ b/tools/rocm-build/build_opencl_on_rocclr.sh
@@ -8,11 +8,13 @@ printUsage() {
    echo
    echo "Options:"
    echo "  -h,  --help               Prints this help"
-    echo "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
    echo "  -c,  --clean              Clean output and delete all intermediate work"
    echo "  -r,  --release            Make a release build instead of a debug build"
    echo "  -a,  --address_sanitizer  Enable address sanitizer"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of type referred to by pkg_type"
+    echo "  -s,  --static             Component/Build does not support static builds just accepting this param & ignore. No effect of the param on this build"
+    echo "  -w,  --wheel              Creates python wheel package of opencl. 
+                                      It needs to be used along with -r option"
    echo
    echo "Possible values for <type>:"
    echo "  deb -> Debian format (default)"
@@ -23,9 +25,7 @@ printUsage() {
 }
 PROJ_NAME="opencl-on-rocclr"
 MAKEOPTS="$DASH_JAY"
-
 BUILD_PATH="$(getBuildPath $PROJ_NAME)"
-
 TARGET="build"
 PACKAGE_ROOT="$(getPackageRoot)"
 PACKAGE_DEB="$PACKAGE_ROOT/deb/$PROJ_NAME"
@@ -39,7 +39,8 @@ MAKETARGET="deb"
 PKGTYPE="deb"


-VALID_STR=`getopt -o hcraso: --long help,clean,release,static,address_sanitizer,outdir: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcrawso: --long help,clean,release,static,wheel,address_sanitizer,outdir: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
@@ -56,9 +57,11 @@ do
                set_address_sanitizer_on ; shift ;;
        (-s | --static)
                ack_and_skip_static ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; shift ;;
        (-o | --outdir)
                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        (*)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
    esac
@@ -73,6 +76,7 @@ if [ $RET_CONFLICT -ge 30 ]; then
 fi

 clean_opencl_on_rocclr() {
+    # Delete cmake output directory
    rm -rf "$BUILD_PATH"
    rm -rf "$PACKAGE_DEB"
    rm -rf "$PACKAGE_RPM"
@@ -83,14 +87,19 @@ clean_opencl_on_rocclr() {
 }

 build_opencl_on_rocclr() {
+    # TODO This if condition is a temporary workaround so that mainline builds dont error out
+    # until build migrated from opencl to clr repo
    if [  -e "$CLR_ROOT/CMakeLists.txt" ]; then
+        # We are in a branch that has migrated to clr repo
        _OCL_CMAKELIST_DIR="$CLR_ROOT"
        _OCL_CMAKELIST_OPT="-DCLR_BUILD_HIP=OFF -DCLR_BUILD_OCL=ON"
    elif [ ! -e "$OPENCL_ON_ROCclr_ROOT/CMakeLists.txt" ]; then
+        # We seem to have hit a branch in which both the old and the new repo don't exist
        echo "No $OPENCL_ON_ROCclr_ROOT/CMakeLists.txt file, skipping opencl on rocclr" >&2
        echo "No $OPENCL_ON_ROCclr_ROOT/CMakeLists.txt file, skipping opencl on rocclr"
-        exit 0
+        exit 0 # This is not an error
    else
+         # We are in a branch that has not yet migrated to clr repo yet
         _OCL_CMAKELIST_DIR="$OPENCL_ON_ROCclr_ROOT"
         _OCL_CMAKELIST_OPT=""
    fi
@@ -99,10 +108,12 @@ build_opencl_on_rocclr() {
    mkdir -p "$BUILD_PATH"
    pushd "$BUILD_PATH"

+    # FIXME: Remove -DROCclr_DIR/LIBROCclr_STATIC_DIR
    if [ ! -e Makefile ]; then
        echo "Building OpenCL-On-ROCclr CMake environment"

        cmake \
+            ${GEN_NINJA} \
            $(rocm_cmake_params) \
            -DUSE_COMGR_LIBRARY=ON \
            $(rocm_common_cmake_params) \
@@ -147,10 +158,20 @@ print_output_directory() {
 }

 case $TARGET in
-    (clean) clean_opencl_on_rocclr ;;
-    (build) build_opencl_on_rocclr ; package_opencl_on_rocclr ;;
-    (outdir) print_output_directory ;;
-        (*) die "Invalid target $TARGET" ;;
+    (clean)
+        clean_opencl_on_rocclr
+        ;;
+    (build)
+        build_opencl_on_rocclr
+        package_opencl_on_rocclr
+        build_wheel "$BUILD_PATH" "$PROJ_NAME"
+        ;;
+    (outdir)
+        print_output_directory
+        ;;
+    (*)
+        die "Invalid target $TARGET"
+        ;;
 esac

 echo "Operation complete"
--- a/tools/rocm-build/build_openmp_extras.sh
+++ b/tools/rocm-build/build_openmp_extras.sh
@@ -10,6 +10,8 @@ printUsage() {
    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 "  -w,  --wheel              Creates python wheel package of openmp-extras. 
+                                      It needs to be used along with -r option"
    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"
@@ -23,9 +25,8 @@ printUsage() {

    return 0
 }
-
 PROJ_NAME="openmp-extras"
-packageMajorVersion="18.63"
+packageMajorVersion="19.65"
 packageMinorVersion="0"
 packageVersion="${packageMajorVersion}.${packageMinorVersion}.${ROCM_LIBPATCH_VERSION}"
 BUILD_PATH="$(getBuildPath $PROJ_NAME)"
@@ -35,9 +36,12 @@ TARGET="build"
 MAKEOPTS="$DASH_JAY"
 STATIC_PKG_DEPS="OFF"

+# Should only need to update this variable when moving
+# to the new ROCM_INSTALL_PATH.
 export INSTALL_PREFIX=${ROCM_INSTALL_PATH}

-VALID_STR=`getopt -o hcraso:p: --long help,clean,release,address_sanitizer,static,outdir,package: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,address_sanitizer,static,outdir,wheel:,package: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
@@ -52,17 +56,23 @@ do
        -a  | --address_sanitizer )
            set_asan_env_vars
            set_address_sanitizer_on
+            # The path will be appended to cmake prefix path for asan builds
+            # Required for finding cmake config files for asan builds
            export ROCM_CMAKECONFIG_PATH="$INSTALL_PREFIX/lib/asan/cmake"
            export VERBOSE=1
+            # openmp debug build of ompd uses python build, which defaults to gcc
            export LDSHARED="$INSTALL_PREFIX/lib/llvm/bin/clang -shared -Wl,-O1 -Wl,-Bsymbolic-functions -Wl,-z,relro -g -fwrapv -O2"
+            # SANITIZER is used in openmp-debug build scripts so that the asan C/CXX Flags are not overwritten
            export SANITIZER=1 ;;
        -o  | --outdir )
            shift 1; PKGTYPE=$1 ; TARGET="outdir" ;;
+        -w  | --wheel )
+                WHEEL_PACKAGE=true ;;
 	-s | --static )
            export STATIC_PKG_DEPS="ON" ;;
        -h  | --help )
            printUsage ; exit 0 ;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        *)
            MAKEARG=$@ ; break ;;
    esac
@@ -71,6 +81,7 @@ done


 clean_openmp_extras() {
+    # Delete cmake output and install directory
    rm -rf "$BUILD_PATH"
    rm -rf "$INSTALL_PREFIX/openmp-extras"
 }
@@ -93,6 +104,9 @@ build_openmp_extras() {
     echo BUILD_PATH: $BUILD_PATH
     echo "INSTALL_PREFIX:$INSTALL_PREFIX"
     export AOMP_STANDALONE_BUILD=0
+     # FIXME: Check the build scripts to see if support for DEVEL package
+     # is on. This can be removed once the old packaging logic is gone. This
+     # may return a non-zero, so for now do not error out the script.
     set +e
     checkDevel=$(grep "ENABLE_DEVEL_PACKAGE=ON" $AOMP_REPOS/aomp/bin/build_openmp.sh)
     set -e
@@ -104,6 +118,7 @@ build_openmp_extras() {
     fi
     export BUILD_AOMP=$BUILD_PATH

+     # EPSDB does not build hsa, we need to pick it up from ROCm install location.
     if [ "$EPSDB" == "1" ]; then
       export ROCM_DIR=$ROCM_INSTALL_PATH
     else
@@ -130,10 +145,17 @@ build_openmp_extras() {
       echo "--------------------------"
     fi

-     export AOMP_JENKINS_BUILD_LIST="extras openmp pgmath flang flang_runtime"
+     # can we stop building flang-classic ?
+     if [ -e $AOMP_REPOS/aomp/bin/disableClassic ]; then
+       export AOMP_JENKINS_BUILD_LIST="extras openmp"
+       export AOMP_SKIP_FLANG=1
+     else
+       export AOMP_JENKINS_BUILD_LIST="extras openmp pgmath flang flang_runtime"
+     fi
     echo "BEGIN Build of openmp-extras"
     "$AOMP_REPOS"/aomp/bin/build_aomp.sh $MAKEARG

+     # Create symlinks for omp headers. Stage 2 components need these.
     local llvm_ver=`$INSTALL_PREFIX/lib/llvm/bin/clang --print-resource-dir | sed 's^/llvm/lib/clang/^ ^' | awk '{print $2}'`
     if [ ! -e $ROCM_INSTALL_PATH/lib/llvm/lib/clang/$llvm_ver/include/omp.h ] ; then
       if [ ! -h $ROCM_INSTALL_PATH/lib/llvm/lib/clang/$llvm_ver/include/omp.h ] ; then
@@ -154,6 +176,7 @@ build_openmp_extras() {
 }

 package_openmp_extras_deb() {
+        # Debian packaging
        local packageName=$1
        local packageDeb="$packageDir/deb"
        local packageArch="amd64"
@@ -184,14 +207,20 @@ package_openmp_extras_deb() {
            debDependencies="$debDependencies, openmp-extras-runtime, hsa-rocr-static-dev"
          fi
        fi
-
+        # copyPath = /opt/rocm
+        # installPath = /opt/rocm/lib/llvm
+        # FIXME: openmp-extras/devel logic can be removed once the new packaging lands
+        # New packaging uses installed_files.txt the openmp-extras/devel
+        # directory will no longer exist.
        if [ -f "$BUILD_PATH"/build/installed_files.txt ] && [ ! -d "$INSTALL_PREFIX"/openmp-extras/devel ]; then
          if [ "$packageType" == "runtime" ]; then
+            # Cleanup previous packages
            rm -rf "$packageDir"
            rm -rf "$DEB_PATH"
            mkdir -p "$DEB_PATH"
            mkdir -p $packageDeb/openmp-extras
-
+            # Licensing
+            # Copy licenses into share/doc/openmp-extras
            mkdir -p $packageDeb/openmp-extras$copyPath/share/doc/openmp-extras
            cp -r $AOMP_REPOS/aomp/LICENSE $packageDeb/openmp-extras$copyPath/share/doc/openmp-extras/LICENSE.apache2
            cp -r $AOMP_REPOS/aomp-extras/LICENSE $packageDeb/openmp-extras$copyPath/share/doc/openmp-extras/LICENSE.mit
@@ -199,20 +228,29 @@ package_openmp_extras_deb() {
 	  else
            rm -rf $packageDeb/openmp-extras/*
            mkdir -p $packageDeb/openmp-extras$copyPath/bin
-            cp -r --parents "$installPath"/lib-debug/src $packageDeb/openmp-extras
+            if [ -d "$installPath"/lib-debug/src ]; then
+              cp -r --parents "$installPath"/lib-debug/src $packageDeb/openmp-extras
+            else
+              cp -r --parents "$ompdSrcDir" $packageDeb/openmp-extras
+            fi
 	  fi
        else
+	  # FIXME: Old packaging method, can delete once new packaging lands.
          if [ "$packageType" == "runtime" ]; then
+            # Cleanup previous packages
            rm -rf "$packageDir"
            rm -rf "$DEB_PATH"
            mkdir -p "$DEB_PATH"
            mkdir -p $packageDeb/openmp-extras$installPath
            mkdir -p $packageDeb/openmp-extras$installPath/lib/clang/$llvm_ver/include
+            # Licensing
+            # Copy licenses into share/doc/openmp-extras
            mkdir -p $packageDeb/openmp-extras$copyPath/share/doc/openmp-extras
            cp -r $AOMP_REPOS/aomp/LICENSE $packageDeb/openmp-extras$copyPath/share/doc/openmp-extras/LICENSE.apache2
            cp -r $AOMP_REPOS/aomp-extras/LICENSE $packageDeb/openmp-extras$copyPath/share/doc/openmp-extras/LICENSE.mit
            cp -r $AOMP_REPOS/flang/LICENSE.txt $packageDeb/openmp-extras$copyPath/share/doc/openmp-extras/LICENSE.flang
          else
+            # Clean packageDeb for devel build
            rm -rf $packageDeb/openmp-extras$installPath/*
            rm -rf $packageDeb/openmp-extras/bin
            rm -rf $packageDeb/openmp-extras$copyPath/share
@@ -224,10 +262,13 @@ package_openmp_extras_deb() {

        mkdir -p "$(dirname $controlFile)"

+        # Copy openmp-extras files, bin will turn into llvm/bin
 	if [ -f "$BUILD_PATH"/build/installed_files.txt ] && [ ! -d "$INSTALL_PREFIX"/openmp-extras/devel ]; then
 	  if [ "$packageType" == "runtime" ]; then
 	    cat "$BUILD_PATH"/build/installed_files.txt | grep -P '\.so|\.a' | cut -d":" -f2 | cut -d" " -f2 | xargs -I {} cp -d --parents {} "$packageDeb"/openmp-extras
-
+	    # libgomp and libiomp5 are not on the install_manifest.txt and need
+	    # to be manually copied. Waiting on trunk patch to flow into amd-staging
+	    # to ensure these symlinks are in the manifest.
 	    cp -d --parents "$installPath/lib/libgomp.so" "$packageDeb"/openmp-extras
 	    cp -d --parents "$installPath/lib/libiomp5.so" "$packageDeb"/openmp-extras
 	    cp -d --parents "$installPath/lib-debug/libgomp.so" "$packageDeb"/openmp-extras
@@ -236,9 +277,11 @@ package_openmp_extras_deb() {
 	    cat "$BUILD_PATH"/build/installed_files.txt | grep -Pv '\.so|\.a' | cut -d":" -f2 | cut -d" " -f2 | xargs -I {} cp -d --parents {} "$packageDeb"/openmp-extras
 	  fi
 	else
+	  # FIXME: Old packaging method, can delete once new packaging lands.
          cp -r "$AOMP"/"$packageType"/* "$packageDeb"/openmp-extras"$installPath"
 	fi

+       # Copy examples
        if [ "$packageType" == "devel" ]; then
          mkdir -p "$packageDeb"/openmp-extras"$copyPath"/share/openmp-extras/examples
          echo cp -r "$AOMP_REPOS"/aomp/examples/fortran "$packageDeb"/openmp-extras"$copyPath"/share/openmp-extras/examples
@@ -252,9 +295,11 @@ package_openmp_extras_deb() {
        fi

        if [ "$packageType" == "devel" ]; then
+          # Create symbolic links for openmp header files
          ln -s ../../../../include/omp.h $packageDeb/openmp-extras$installPath/lib/clang/$llvm_ver/include/omp.h
          ln -s ../../../../include/ompt.h $packageDeb/openmp-extras$installPath/lib/clang/$llvm_ver/include/ompt.h
          ln -s ../../../../include/omp-tools.h $packageDeb/openmp-extras$installPath/lib/clang/$llvm_ver/include/omp-tools.h
+         # Only create symlinks if file exists
          if [ ! -h "$packageDeb"/openmp-extras"$copyPath"/bin/aompcc ] && [ -e "$packageDeb"/openmp-extras"$installPath"/bin/aompcc ]; then
            ln -s ../lib/llvm/bin/aompcc "$packageDeb"/openmp-extras"$copyPath"/bin/aompcc
          fi
@@ -266,12 +311,14 @@ package_openmp_extras_deb() {
          fi
        fi

+        # Inspect
        ls -l "$packageDeb"/openmp-extras"$installPath"
 	if [ "$packageType" == "devel" ]; then
          ls -l "$packageDeb"/openmp-extras"$installPath"/bin
          ls -l "$packageDeb"/openmp-extras"$copyPath"/bin
 	fi

+        # Create control file
        {
          echo "Package: $packageName"
          echo "Architecture: $packageArch"
@@ -293,7 +340,9 @@ package_openmp_extras_deb() {
        "$DEB_PATH/${packageName}_${packageVersion}-${CPACK_DEBIAN_PACKAGE_RELEASE}_${packageArch}.deb"
 }

+# ASAN debian package
 package_openmp_extras_asan_deb() {
+        # Debian packaging
        local packageName=$1
        local packageDeb="$packageDir/deb"
        local packageArch="amd64"
@@ -308,6 +357,8 @@ package_openmp_extras_asan_deb() {
        rm -rf "$packageDir"
        rm -rf "$DEB_PATH"
        mkdir -p "$DEB_PATH"
+        # Licensing
+        # Copy licenses into share/doc/openmp-extras-asan
        local licenseDir="$packageDeb/openmp-extras$copyPath/share/doc/openmp-extras-asan"
        mkdir -p $licenseDir
        cp -r $AOMP_REPOS/aomp/LICENSE $licenseDir/LICENSE.apache2
@@ -317,15 +368,18 @@ package_openmp_extras_asan_deb() {
        mkdir -p "$(dirname $controlFile)"
        if [ -f "$BUILD_PATH"/build/installed_files.txt ] && [ ! -d "$INSTALL_PREFIX"/openmp-extras ]; then
 	  cat "$BUILD_PATH"/build/installed_files.txt | grep -P 'asan' | cut -d":" -f2 | cut -d" " -f2 | xargs -I {} cp -d --parents {} "$packageDeb"/openmp-extras
-
+	  # libgomp and libiomp5 are not on the install_manifest.txt and need
+	  # to be manually copied. Waiting on trunk patch to flow into amd-staging
+	  # to ensure these symlinks are in the manifest.
 	  cp -d --parents "$installPath/lib/asan/libgomp.so" "$packageDeb"/openmp-extras
 	  cp -d --parents "$installPath/lib/asan/libiomp5.so" "$packageDeb"/openmp-extras
 	  cp -d --parents "$installPath/lib-debug/asan/libgomp.so" "$packageDeb"/openmp-extras
 	  cp -d --parents "$installPath/lib-debug/asan/libiomp5.so" "$packageDeb"/openmp-extras
-
+        # FIXME: Old packaging method, can delete once new packaging lands.
        else
          mkdir -p $packageDeb/openmp-extras$installPath/lib/asan
          mkdir -p $packageDeb/openmp-extras$installPath/lib-debug/asan
+          # runtime folder have asan libraries. Copy to asan folder for packaging
          cp -r "$AOMP"/lib/asan/* "$packageDeb"/openmp-extras"$installPath"/lib/asan/
          cp -r "$AOMP"/lib-debug/asan/* "$packageDeb"/openmp-extras"$installPath"/lib-debug/asan/
          cp -r "$AOMP"/"$asanLibDir"/lib/asan/* "$packageDeb"/openmp-extras"$installPath"/lib/asan/
@@ -334,6 +388,7 @@ package_openmp_extras_asan_deb() {
          cp -r "$AOMP"/devel/lib-debug/asan/* "$packageDeb"/openmp-extras"$installPath"/lib-debug/asan/
        fi

+        # Create control file
        {
          echo "Package: $packageName"
          echo "Architecture: $packageArch"
@@ -352,6 +407,7 @@ package_openmp_extras_asan_deb() {


 package_openmp_extras_rpm() {
+        # RPM packaging
        local packageName=$1
        local packageRpm="$packageDir/rpm"
        local specFile="$packageDir/$packageName.spec"
@@ -376,6 +432,7 @@ package_openmp_extras_rpm() {
          fi
        fi

+        # Cleanup previous packages
        if [ "$packageType" == "runtime" ]; then
          rm -rf "$packageDir"
          rm -rf "$RPM_PATH"
@@ -386,6 +443,8 @@ package_openmp_extras_rpm() {
        mkdir -p "$(dirname $specFile)"

        {
+          # FIXME: Remove all conditions for empty packageType when
+          # devel and runtime packaging changes land in mainline.
          echo "%define is_runtime %( if [ $packageType == runtime ]; then echo "1" ; else echo "0"; fi )"
          echo "%define is_devel %( if [ $packageType == devel ]; then echo "1" ; else echo "0"; fi )"

@@ -403,6 +462,7 @@ package_openmp_extras_rpm() {
          echo "Obsoletes:  $rpmObsoletes"
          echo "%endif"
          echo "%define debug_package %{nil}"
+          # Redefining __os_install_post to remove stripping
          echo "%define __os_install_post %{nil}"
          echo "%description"
          echo "$packageSummaryLong"
@@ -419,6 +479,7 @@ package_openmp_extras_rpm() {
          echo "    mkdir -p \$RPM_BUILD_ROOT$copyPath/bin"
          echo "    mkdir -p \$RPM_BUILD_ROOT$installPath/lib/clang/$llvm_ver/include"
          echo "  %endif"
+          # FIXME: Old packaging method, can delete once new packaging lands.
          echo "else"
          echo "  %if %is_runtime"
          echo "    mkdir -p  \$RPM_BUILD_ROOT$installPath"
@@ -430,10 +491,13 @@ package_openmp_extras_rpm() {
          echo "  %endif"
          echo "fi"

+          # Copy openmp-extras files, bin will turn into llvm/bin
          echo "if [ -f $BUILD_PATH/build/installed_files.txt ] && [ ! -d $INSTALL_PREFIX/openmp-extras/devel ]; then"
          echo "  %if %is_runtime"
          echo "    cat $BUILD_PATH/build/installed_files.txt | grep -P '\.so|\.a' | cut -d':' -f2 | cut -d' ' -f2 | xargs -I {} cp -d --parents {} \$RPM_BUILD_ROOT"
-
+          # libgomp and libiomp5 are not on the install_manifest.txt and need
+          # to be manually copied. Waiting on trunk patch to flow into amd-staging
+          # to ensure these symlinks are in the manifest.
          echo "    cp -d --parents "$installPath/lib/libgomp.so" \$RPM_BUILD_ROOT"
          echo "    cp -d --parents "$installPath/lib/libiomp5.so" \$RPM_BUILD_ROOT"
          echo "    cp -d --parents "$installPath/lib-debug/libgomp.so" \$RPM_BUILD_ROOT"
@@ -442,14 +506,17 @@ package_openmp_extras_rpm() {
          echo "%if %is_devel"
          echo "  cat "$BUILD_PATH"/build/installed_files.txt | grep -Pv '\.so|\.a' | cut -d':' -f2 | cut -d' ' -f2 | xargs -I {} cp -d --parents {} \$RPM_BUILD_ROOT"
          echo "%endif"
-
+          # FIXME: Old packaging method, can delete once new packaging lands.
          echo "else"
          echo "  cp -r $AOMP/$packageType/* \$RPM_BUILD_ROOT$installPath"
+          # Devel does not have examples in the llvm dir
          echo "  %if %is_devel"
          echo "    rm -rf \$RPM_BUILD_ROOT$installPath/share"
          echo "  %endif"
          echo "fi"

+          # Create symbolic links from /opt/rocm/bin to /opt/rocm/lib/llvm/bin utilities
+          #echo "ls \$RPM_BUILD_ROOT$installPath"
          echo "%if %is_devel"
          echo "  if [ ! -h \$RPM_BUILD_ROOT$copyPath/bin/aompcc ] && [ -e \$RPM_BUILD_ROOT$installPath/bin/aompcc ]; then"
          echo "    ln -s ../lib/llvm/bin/aompcc \$RPM_BUILD_ROOT$copyPath/bin/aompcc"
@@ -462,18 +529,23 @@ package_openmp_extras_rpm() {
          echo "  fi"
          echo "  ls \$RPM_BUILD_ROOT$copyPath"

+          # Create symbolic links for openmp header files
          echo "  ln -s ../../../../include/omp.h  \$RPM_BUILD_ROOT/$installPath/lib/clang/$llvm_ver/include/omp.h"
          echo "  ln -s ../../../../include/ompt.h  \$RPM_BUILD_ROOT/$installPath/lib/clang/$llvm_ver/include/ompt.h"
          echo "  ln -s ../../../../include/omp-tools.h \$RPM_BUILD_ROOT/$installPath/lib/clang/$llvm_ver/include/omp-tools.h"
          echo "%endif"
          echo 'find $RPM_BUILD_ROOT \! -type d | sed "s|$RPM_BUILD_ROOT||"> files.list'
+          # Copy examples

          echo "%if %is_runtime"
+          # Licensing
+          # Copy licenses into share/doc/openmp-extras
          echo "  mkdir -p \$RPM_BUILD_ROOT$copyPath/share/doc/openmp-extras"
          echo "  cp -r $AOMP_REPOS/aomp/LICENSE \$RPM_BUILD_ROOT$copyPath/share/doc/openmp-extras/LICENSE.apache2"
          echo "  cp -r $AOMP_REPOS/aomp-extras/LICENSE \$RPM_BUILD_ROOT$copyPath/share/doc/openmp-extras/LICENSE.mit"
          echo "  cp -r $AOMP_REPOS/flang/LICENSE.txt \$RPM_BUILD_ROOT$copyPath/share/doc/openmp-extras/LICENSE.flang"
          echo "%else"
+          # Copy devel examples to share/openmp-extras/examples
          echo "  mkdir -p \$RPM_BUILD_ROOT$copyPath/share/openmp-extras/examples"
          echo "  cp -r $AOMP_REPOS/aomp/examples/fortran \$RPM_BUILD_ROOT$copyPath/share/openmp-extras/examples"
          echo "  cp -r $AOMP_REPOS/aomp/examples/openmp \$RPM_BUILD_ROOT$copyPath/share/openmp-extras/examples"
@@ -493,6 +565,12 @@ package_openmp_extras_rpm() {
          echo "  $copyPath/share/openmp-extras"
          echo "%endif"
          echo "%defattr(-,root,root,-)"
+          # Note: In some OS like SLES, during upgrade rocm-core is getting upgraded first and followed by other packages
+          # rocm-core cannot delete rocm-ver folder, since it contains files of other packages that are yet to be upgraded
+          # To remove rocm-ver folder after upgrade the spec file of other packages should contain the rocm-ver directory
+          # Otherwise after upgrade empty old rocm-ver folder will be left out.
+          # If empty remove /opt/rocm-ver folder and its subdirectories created by
+          # openmp-extras runtime and devel package
          echo "%if %is_runtime || %is_devel"
          echo "  $copyPath"
          echo "%endif"
@@ -502,15 +580,19 @@ package_openmp_extras_rpm() {
        mv $packageRpm/RPMS/x86_64/*.rpm $RPM_PATH
 }

+# ASAN RPM packaging
 package_openmp_extras_asan_rpm() {
+        # RPM packaging
        local packageName=$1
        local packageRpm="$packageDir/rpm"
        local specFile="$packageDir/$packageName.spec"
        local packageSummary="AddressSanitizer OpenMP Extras provides instrumented openmp and flang libraries."
        local packageSummaryLong="openmp-extras $packageVersion is based on LLVM 17 and is used for offloading to Radeon GPUs."
        local rpmRequires="hsa-rocr-asan, rocm-core-asan"
+        # After build,runtime folder will have ASAN libaries.
        local asanLibDir="runtime"

+        # Cleanup previous packages
        rm -rf "$packageDir"
        rm -rf "$RPM_PATH"
        mkdir -p "$RPM_PATH"
@@ -527,6 +609,7 @@ package_openmp_extras_asan_rpm() {
          echo "License:    MIT and ASL 2.0 and ASL 2.0 with exceptions"
          echo "Vendor:     Advanced Micro Devices, Inc."
          echo "Requires:   $rpmRequires"
+          # Redefining __os_install_post to prevent binary stripping
          echo "%define __os_install_post %{nil}"
          echo "%description"
          echo "%undefine _debugsource_packages"
@@ -539,11 +622,16 @@ package_openmp_extras_asan_rpm() {
          echo "%install"
          echo "if [ -f $BUILD_PATH/build/installed_files.txt ] && [ ! -d "$INSTALL_PREFIX"/openmp-extras ]; then"
          echo "  cat $BUILD_PATH/build/installed_files.txt | grep -P 'asan' | cut -d':' -f2 | cut -d' ' -f2 | xargs -I {} cp -d --parents {} \$RPM_BUILD_ROOT"
+          # libgomp and libiomp5 are not on the install_manifest.txt and need
+          # to be manually copied. Waiting on trunk patch to flow into amd-staging
+          # to ensure these symlinks are in the manifest.
          echo "  cp -d --parents "$installPath/lib/asan/libgomp.so" \$RPM_BUILD_ROOT"
          echo "  cp -d --parents "$installPath/lib/asan/libiomp5.so" \$RPM_BUILD_ROOT"
          echo "  cp -d --parents "$installPath/lib-debug/asan/libgomp.so" \$RPM_BUILD_ROOT"
          echo "  cp -d --parents "$installPath/lib-debug/asan/libiomp5.so" \$RPM_BUILD_ROOT"
+          # FIXME: Old packaging method, can delete once new packaging lands.
          echo "else"
+          # Copy openmp-extras ASAN libraries to ASAN folders
          echo "  mkdir -p  \$RPM_BUILD_ROOT$installPath/lib/asan"
          echo "  mkdir -p  \$RPM_BUILD_ROOT$installPath/lib-debug/asan"
          echo "  cp -r $AOMP/lib/asan/* \$RPM_BUILD_ROOT$installPath/lib/asan"
@@ -554,8 +642,11 @@ package_openmp_extras_asan_rpm() {
          echo "  cp -r $AOMP/devel/lib-debug/asan/* \$RPM_BUILD_ROOT$installPath/lib-debug/asan"
          echo "fi"

+          # Create symbolic links from /opt/rocm/bin to /opt/rocm/lib/llvm/bin utilities
          echo 'find $RPM_BUILD_ROOT \! -type d | sed "s|$RPM_BUILD_ROOT||"> files.list'

+          # Licensing
+          # Copy licenses into share/doc/openmp-extras
          echo "  mkdir -p \$RPM_BUILD_ROOT$copyPath/share/doc/openmp-extras-asan"
          echo "  cp -r $AOMP_REPOS/aomp/LICENSE \$RPM_BUILD_ROOT$copyPath/share/doc/openmp-extras-asan/LICENSE.apache2"
          echo "  cp -r $AOMP_REPOS/aomp-extras/LICENSE \$RPM_BUILD_ROOT$copyPath/share/doc/openmp-extras-asan/LICENSE.mit"
@@ -566,6 +657,12 @@ package_openmp_extras_asan_rpm() {
          echo "%files -f files.list"
          echo "  $copyPath/share/doc/openmp-extras-asan"
          echo "%defattr(-,root,root,-)"
+          # Note: In some OS like SLES, during upgrade rocm-core is getting upgraded first and followed by other packages
+          # rocm-core cannot delete rocm-ver folder, since it contains files of other packages that are yet to be upgraded
+          # To remove rocm-ver folder after upgrade the spec file of other packages should contain the rocm-ver directory
+          # Otherwise after upgrade empty old rocm-ver folder will be left out.
+          # If empty remove /opt/rocm-ver folder and its subdirectories created by
+          # openmp-extras runtime and devel package
          echo "  $copyPath"

        } > $specFile
@@ -581,21 +678,29 @@ package_openmp_extras() {
    local llvm_ver=`$INSTALL_PREFIX/lib/llvm/bin/clang --print-resource-dir | sed 's^/llvm/lib/clang/^ ^' | awk '{print $2}'`
    local debNames="openmp-extras-runtime openmp-extras-dev"
    local rpmNames="openmp-extras-runtime openmp-extras-devel"
+    ompdSrcDir="$INSTALL_PREFIX/lib/llvm/share/gdb/python/ompd/src"
    if [ "$SANITIZER" == "1" ]; then
      local asanPkgName="openmp-extras-asan"
-      if [[ $DISTRO_NAME =~ "Ubuntu" ]]; then
+      if [[ $DISTRO_NAME =~ "Ubuntu" ]] || [[ $DISTRO_NAME =~ "Debian" ]]; then
+        echo "Warning: Assuming DEBs"
        package_openmp_extras_asan_deb $asanPkgName
      else
+        echo "Warning: Assuming RPMs"
        package_openmp_extras_asan_rpm $asanPkgName
      fi
+      # For ASAN build, create only ASAN package
+      # Devel and runtime pkg should be created from non-asan build. So return the execution
      return 0
    fi
-
-    if [[ $DISTRO_NAME =~ "Ubuntu" ]]; then
+    # Only build deb in Ubuntu environment
+    if [[ $DISTRO_NAME =~ "Ubuntu" ]] || [[ $DISTRO_NAME =~ "Debian" ]]; then
+      echo "Warning: Assuming DEBs"
      for name in $debNames; do
        package_openmp_extras_deb $name
      done
+    # Only build RPM in CENTOS/SLES environment
    else
+      echo "Warning: Assuming RPMs"
      for name in $rpmNames; do
        package_openmp_extras_rpm $name
      done
@@ -603,6 +708,7 @@ package_openmp_extras() {
 }

 package_tests_deb(){
+    # Openmp-extras debian test packaging
    local packageDir="$BUILD_PATH/package"
    local packageDeb="$packageDir/deb"
    local packageArch="amd64"
@@ -615,6 +721,7 @@ package_tests_deb(){
    local installPath="$ROCM_INSTALL_PATH/share/openmp-extras/tests"
    local packageName="openmp-extras-tests"

+    # Cleanup previous packages
    rm -rf "$packageDir"

    mkdir -p $packageDeb/openmp-extras"$installPath"
@@ -622,9 +729,15 @@ package_tests_deb(){
        rm $(dirname $controlFile)
    fi
    mkdir -p "$(dirname $controlFile)"
+    # Copy openmp-extras files
    cp -r "$AOMP_REPOS/aomp/." "$packageDeb/openmp-extras/$installPath"
    rm -rf "$packageDeb"/openmp-extras"$installPath"/.git "$packageDeb"/openmp-extras"$installPath"/.github
-    cp "$OUT_DIR/build/lightning/bin/FileCheck" "$packageDeb/openmp-extras/$installPath/bin"
+    # Copy FileCheck
+    if [ -f "$OUT_DIR"/build/lightning/bin/FileCheck ]; then
+      cp "$OUT_DIR/build/lightning/bin/FileCheck" "$packageDeb/openmp-extras/$installPath/bin"
+    fi
+
+    # Create control file
    {
      echo "Package: $packageName"
      echo "Architecture: $packageArch"
@@ -642,16 +755,19 @@ package_tests_deb(){
 }

 package_tests_rpm(){
+    # RPM packaging
    AOMP_STANDALONE_BUILD=1 $AOMP_REPOS/aomp/bin/build_fixups.sh
    local copyPath="$ROCM_INSTALL_PATH"
    local packageDir="$BUILD_PATH/package"
    local packageRpm="$packageDir/rpm"
    local installPath="$ROCM_INSTALL_PATH/share/openmp-extras/tests"
    local packageName="openmp-extras-tests"
+    local rpmRequires="openmp-extras-devel, rocm-core"
    local specFile="$packageDir/$packageName.spec"
    local packageSummary="Tests for openmp-extras."
    local packageSummaryLong="Tests for openmp-extras $packageVersion is based on LLVM 18 and is used for offloading to Radeon GPUs."

+    # Cleanup previous packages
    rm -rf "$packageDir"
    mkdir -p "$packageRpm/openmp-extras/$installPath"
    {
@@ -679,7 +795,9 @@ package_tests_rpm(){
      echo "mkdir -p  \$RPM_BUILD_ROOT$installPath"
      echo "cp -R $AOMP_REPOS/aomp/. \$RPM_BUILD_ROOT$installPath"
      echo "rm -rf \$RPM_BUILD_ROOT$installPath/.git \$RPM_BUILD_ROOT$installPath/.github"
-      echo "cp $OUT_DIR/build/lightning/bin/FileCheck \$RPM_BUILD_ROOT$installPath/bin"
+      echo "if [ -f $OUT_DIR/build/lightning/bin/FileCheck ]; then"
+      echo "  cp $OUT_DIR/build/lightning/bin/FileCheck \$RPM_BUILD_ROOT$installPath/bin"
+      echo "fi"
      echo 'find $RPM_BUILD_ROOT \! -type d | sed "s|$RPM_BUILD_ROOT||"> files.list'

      echo "%clean"
@@ -698,8 +816,10 @@ package_tests_rpm(){

 package_tests() {
    local DISTRO_NAME=$(cat /etc/os-release | grep -e ^NAME=)
+    # Only build deb in Ubuntu environment
    if [[ $DISTRO_NAME =~ "Ubuntu" ]]; then
        package_tests_deb
+    # Only build RPM in CENTOS/SLES environment
    else
        package_tests_rpm
    fi
@@ -718,10 +838,21 @@ print_output_directory() {
 }

 case $TARGET in
-    (clean) clean_openmp_extras ;;
-    (build) build_openmp_extras; package_openmp_extras; package_tests ;;
-    (outdir) print_output_directory ;;
-    (*) die "Invalid target $TARGET" ;;
+    (clean)
+        clean_openmp_extras
+        ;;
+    (build)
+        build_openmp_extras
+        package_openmp_extras
+        package_tests
+        build_wheel "$BUILD_PATH" "$PROJ_NAME"
+        ;;
+    (outdir)
+        print_output_directory
+        ;;
+    (*)
+        die "Invalid target $TARGET"
+        ;;
 esac

 echo "Operation complete"
--- a/tools/rocm-build/build_rccl.sh
+++ b/tools/rocm-build/build_rccl.sh
@@ -1,12 +1,10 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src rccl

-ENABLE_ADDRESS_SANITIZER=false
-
 build_rccl() {
    echo "Start build"

@@ -24,19 +22,12 @@ build_rccl() {

    mkdir -p $BUILD_DIR && cd $BUILD_DIR

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908:xnack-;gfx90a:xnack-;gfx90a:xnack+;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
-    fi
-
    init_rocm_common_cmake_params

    CC=${ROCM_PATH}/bin/amdclang \
-    CXX=$(set_build_variables CXX) \
+    CXX=$(set_build_variables __CXX__) \
    cmake \
        "${rocm_math_common_cmake_params[@]}" \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        -DHIP_COMPILER=clang \
        -DCMAKE_PREFIX_PATH="${ROCM_PATH};${ROCM_PATH}/share/rocm/cmake/" \
        ${LAUNCHER_FLAGS} \
@@ -50,7 +41,7 @@ build_rccl() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -64,7 +55,7 @@ clean_rccl() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_rccl ;;
+    build) build_rccl; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_rccl ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_rdc.sh
+++ b/tools/rocm-build/build_rdc.sh
@@ -1,121 +1,45 @@
 #!/bin/bash

-source "$(dirname "$BASH_SOURCE")/compute_utils.sh"
+set -ex

-printUsage() {
-    echo
-    echo "Usage: $(basename $0) [-c|-r|-h] [makeopts]"
-    echo
-    echo "Options:"
-    echo "  -c,  --clean            Removes all rdc build artifacts, except grpc"
-    echo "  -g,  --clean_grpc       Removes the grpc files and artifacts"
-    echo "  -r,  --release          Build release version of RDC (default is debug)"
-    echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
-        type referred to by pkg_type"
-    echo "  -s,  --static           Component/Build does not support static builds just accepting this param & ignore. No effect of the param on this build"
-    echo "  -h,  --help             Prints this help"
-    echo
-    return 0
-}
+### Set up RPATH for RDC to simplify finding RDC libraries. Please note that
+### ROCm has a setup_env.sh script which already populates these variables.
+### These variables are then used inside compute_utils.sh.
+###
+### We need to append additional paths to these variables BEFORE sourcing
+### compute_utils.sh

-PACKAGE_ROOT="$(getPackageRoot)"
-RDC_BUILD_DIR=$(getBuildPath rdc)
-GRPC_BUILD_DIR=$(getBuildPath grpc)
-TARGET="build"
-PACKAGE_LIB=$(getLibPath)
-PACKAGE_INCLUDE="$(getIncludePath)"
-PACKAGE_BIN="$(getBinPath)"
-RDC_PACKAGE_DEB_DIR="$PACKAGE_ROOT/deb/rdc"
-RDC_PACKAGE_RPM_DIR="$PACKAGE_ROOT/rpm/rdc"
-BUILD_TYPE="Debug"
-MAKETARGET="deb"
-MAKEARG="$DASH_JAY O=$RDC_BUILD_DIR"
-SHARED_LIBS="ON"
-CLEAN_OR_OUT=0;
-CLEAN_GRPC="no"
-PKGTYPE="deb"
-RDC_MAKE_OPTS="$DASH_JAY O=$RDC_BUILD_DIR -C $RDC_BUILD_DIR"
-BUILD_DOCS="no"
-RDC_PKG_NAME_ROOT="rdc"
-RDC_PKG_NAME="${RDC_PKG_NAME_ROOT}"
-GRPC_PROTOC_ROOT="${RDC_BUILD_DIR}/grpc"
-GRPC_SEARCH_ROOT="/usr/grpc"
-GRPC_DESIRED_VERSION="1.61.0"
-
-RDC_LIB_RPATH='$ORIGIN'
-RDC_LIB_RPATH=$RDC_LIB_RPATH:'$ORIGIN/..'
-RDC_LIB_RPATH=$RDC_LIB_RPATH:'$ORIGIN/rdc/grpc/lib'
-RDC_LIB_RPATH=$RDC_LIB_RPATH:'$ORIGIN/grpc/lib'
-RDC_EXE_RPATH='$ORIGIN/../lib'
-RDC_EXE_RPATH=$RDC_EXE_RPATH:'$ORIGIN/../lib/rdc/grpc/lib'
-
-VALID_STR=`getopt -o hcgradso:p: --long help,clean,clean_grpc,release,documentation,static,address_sanitizer,outdir:,package: -- "$@"`
-eval set -- "$VALID_STR"
-
-while true ;
-do
-    case "$1" in
-        (-h | --help)
-                printUsage ; exit 0;;
-        (-c | --clean)
-                TARGET="clean" ; ((CLEAN_OR_OUT|=1)) ; shift ;;
-        (-g | --clean_grpc)
-                TARGET="clean_grpc" ; shift ;;
-        (-r | --release)
-                BUILD_TYPE="Release" ; shift ;;
-        (-a | --address_sanitizer)
-                set_asan_env_vars
-                set_address_sanitizer_on ; shift ;;
-        (-d | --documentation )
-                BUILD_DOCS="yes" ;;
-        (-s | --static)
-                ack_and_skip_static ;;
-        (-o | --outdir)
-                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
-        (-p | --package)
-                MAKETARGET="$2" ; shift 2;;
-        --)     shift; break;;
-        (*)
-                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
-    esac
-
-done
-
-RET_CONFLICT=1
-check_conflicting_options $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
-if [ $RET_CONFLICT -ge 30 ]; then
-   print_vars $API_NAME $TARGET $BUILD_TYPE $SHARED_LIBS $CLEAN_OR_OUT $PKGTYPE $MAKETARGET
-   exit $RET_CONFLICT
+# lib/rdc/librdc_rocp.so needs lib/librdc_bootstrap.so
+# this also covers the ASAN usecase
+ROCM_LIB_RPATH=$ROCM_LIB_RPATH:'$ORIGIN/..'
+# grpc
+ROCM_LIB_RPATH=$ROCM_LIB_RPATH:'$ORIGIN/rdc/grpc/lib'
+ROCM_LIB_RPATH=$ROCM_LIB_RPATH:'$ORIGIN/grpc/lib'
+# help RDC executables find RDC libraries
+# lib/librdc_bootstrap.so.0 and grpc
+ROCM_EXE_RPATH=$ROCM_EXE_RPATH:'$ORIGIN/../lib/rdc/grpc/lib'
+if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+    ROCM_EXE_RPATH="$ROCM_ASAN_EXE_RPATH:$ROCM_EXE_RPATH"
 fi

+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

-clean_rdc() {
-    rm -rf "$RDC_BUILD_DIR"
-    rm -rf "$RDC_PACKAGE_DEB_DIR"
-    rm -rf "$RDC_PACKAGE_RPM_DIR"
-    rm -rf "$RDC_BUILD_DIR/rdc"
+set_component_src rdc

-    rm -rf "$PACKAGE_INCLUDE/rdc"
-    rm -f $PACKAGE_LIB/librdc*
-    rm -f $PACKAGE_BIN/rdci
-    return 0
-}
-
-clean_grpc() {
-    rm -rf "$GRPC_BUILD_DIR"
-}
+# RDC
+# BUILD ARGUMENTS
+BUILD_DOCS="no"
+GRPC_PROTOC_ROOT="${BUILD_DIR}/grpc"
+GRPC_SEARCH_ROOT="/usr/grpc"
+GRPC_DESIRED_VERSION="1.67.1" # do not include 'v'
+# lib/librocm_smi64.so and lib/libamd_smi.so

+# check if exact version of gRPC is installed
 find_grpc() {
    grep -s -F "$GRPC_DESIRED_VERSION" ${GRPC_SEARCH_ROOT}/*/cmake/grpc/gRPCConfigVersion.cmake &&
        GRPC_PROTOC_ROOT=$GRPC_SEARCH_ROOT
 }

-rdc_backwards_compat_cmake_params() {
-    grep -q "RDC_CLIENT_INSTALL_PREFIX" "$RDC_ROOT/CMakeLists.txt" &&
-        echo "-DRDC_CLIENT_INSTALL_PREFIX=$PACKAGE_ROOT"
-}
-
 build_rdc() {
    if ! find_grpc; then
        echo "ERROR: GRPC SEARCH FAILED!"
@@ -125,81 +49,96 @@ build_rdc() {
    fi
    echo "gRPC [${GRPC_DESIRED_VERSION}] found!"

+    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
+        ack_and_skip_static
+    fi
+
+    CXX=$(set_build_variables __C_++__)
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+        set_asan_env_vars
+        set_address_sanitizer_on
+        # NOTE: Temp fix for ASAN failures SWDEV-515858
+        export ASAN_OPTIONS="detect_leaks=0:new_delete_type_mismatch=0"
+    fi
+
    echo "Building RDC"
    echo "RDC_BUILD_DIR: ${RDC_BUILD_DIR}"
    echo "GRPC_PROTOC_ROOT: ${GRPC_PROTOC_ROOT}"

-    export LD_PRELOAD="$ASAN_LIB_PATH"
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+        # NOTE: Temp workaround for libasan not being first in the library list.
+        # libasan not being first causes ADDRESS_SANITIZER builds to fail.
+        # This value is set by set_asan_env_vars. Which is only called when -a arg is passed.
+        export LD_PRELOAD="$ASAN_LIB_PATH"
+    fi

-    if [ ! -d "$RDC_BUILD_DIR/rdc_libs" ]; then
-        mkdir -p $RDC_BUILD_DIR
-        pushd $RDC_BUILD_DIR
+    echo "C compiler: $CC"
+    echo "CXX compiler: $CXX"
+    init_rocm_common_cmake_params
+    if [ ! -d "$BUILD_DIR/rdc_libs" ]; then
+        mkdir -p "$BUILD_DIR"
+        pushd "$BUILD_DIR"
        cmake \
            -DGRPC_ROOT="$GRPC_PROTOC_ROOT" \
            -DGRPC_DESIRED_VERSION="$GRPC_DESIRED_VERSION" \
-            -DCMAKE_MODULE_PATH="$RDC_ROOT/cmake_modules" \
-            $(rocm_cmake_params) \
-            $(rdc_backwards_compat_cmake_params) \
-            $(rocm_common_cmake_params) \
-            -DROCM_DIR=$ROCM_INSTALL_PATH \
-            -DRDC_PACKAGE="${RDC_PKG_NAME}" \
+            -DCMAKE_MODULE_PATH="$COMPONENT_SRC/cmake_modules" \
+            "${rocm_math_common_cmake_params[@]}" \
+            -DCPACK_GENERATOR="${PKGTYPE^^}" \
+            -DROCM_DIR=$ROCM_PATH \
            -DCPACK_PACKAGE_VERSION_MAJOR="1" \
            -DCPACK_PACKAGE_VERSION_MINOR="$ROCM_LIBPATCH_VERSION" \
            -DCPACK_PACKAGE_VERSION_PATCH="0" \
            -DADDRESS_SANITIZER="$ADDRESS_SANITIZER" \
            -DBUILD_TESTS=ON \
+            -DBUILD_PROFILER=ON \
+            -DBUILD_RVS=ON \
            -DCMAKE_SKIP_BUILD_RPATH=TRUE \
-            -DCMAKE_EXE_LINKER_FLAGS_INIT="-Wl,--no-as-needed,-z,origin,--enable-new-dtags,--build-id=sha1,--rpath,$RDC_EXE_RPATH" \
-            -DCMAKE_SHARED_LINKER_FLAGS_INIT="-Wl,--no-as-needed,-z,origin,--enable-new-dtags,--build-id=sha1,--rpath,$RDC_LIB_RPATH" \
-            "$RDC_ROOT"
+            "$COMPONENT_SRC"
        popd
    fi
    echo "Making rdc package:"
-    cmake --build "$RDC_BUILD_DIR" -- $RDC_MAKE_OPTS
-    cmake --build "$RDC_BUILD_DIR" -- $RDC_MAKE_OPTS install
+    cmake --build "$BUILD_DIR" -- -j${PROC}
+    cmake --build "$BUILD_DIR" -- install

-    unset LD_PRELOAD
-    cmake --build "$RDC_BUILD_DIR" -- $RDC_MAKE_OPTS package
-
-    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$RDC_PACKAGE_DEB_DIR" "$RDC_BUILD_DIR/$RDC_PKG_NAME"*.deb
-    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$RDC_PACKAGE_RPM_DIR" "$RDC_BUILD_DIR/$RDC_PKG_NAME"*.rpm
-
-    if [ ! -e $ROCM_INSTALL_PATH/include/rdc/rdc.h ]; then
-      cp -r "$ROCM_INSTALL_PATH/rdc/lib/." "$PACKAGE_LIB"
-      cp -r "$ROCM_INSTALL_PATH/rdc/bin/." "$PACKAGE_BIN"
-      cp -r "$ROCM_INSTALL_PATH/rdc/include/." "$PACKAGE_INCLUDE"
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+        # NOTE: Must disable LD_PRELOAD hack before packaging!
+        # cmake fails with cryptic error on RHEL:
+        #
+        #    AddressSanitizer:DEADLYSIGNAL
+        #    ==17083==ERROR: AddressSanitizer: stack-overflow on address ...
+        #
+        # The issue is likely in python3.6 cpack scripts
+        unset LD_PRELOAD
    fi

+    cmake --build "$BUILD_DIR" -- package
+
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+
    if [ "$BUILD_DOCS" = "yes" ]; then
      echo "Building Docs"
-      cmake --build "$RDC_BUILD_DIR" -- $RDC_MAKE_OPTS doc
-      pushd $RDC_BUILD_DIR/latex
+      cmake --build "$BUILD_DIR" -- doc
+      pushd "$BUILD_DIR"/latex
      cmake --build . --
-      mv refman.pdf "$ROCM_INSTALL_PATH/rdc/RDC_Manual.pdf"
+      mv refman.pdf "$ROCM_PATH/rdc/RDC_Manual.pdf"
      popd
    fi
 }

-print_output_directory() {
-    case ${PKGTYPE} in
-        ("deb")
-            echo ${RDC_PACKAGE_DEB_DIR};;
-        ("rpm")
-            echo ${RDC_PACKAGE_RPM_DIR};;
-        (*)
-            echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
-    esac
-    exit
+clean_rdc() {
+    echo "Cleaning RDC build directory: ${BUILD_DIR} ${PACKAGE_DIR}"
+    rm -rf "$BUILD_DIR" "$PACKAGE_DIR"
+    return 0
 }

-verifyEnvSetup
+stage2_command_args "$@"
+disable_debug_package_generation

 case $TARGET in
-    (clean) clean_rdc ;;
-    (clean_grpc) clean_grpc ;;
-    (build) build_rdc ;;
-    (outdir) print_output_directory ;;
-    (*) die "Invalid target $TARGET" ;;
+    clean) clean_rdc ;;
+    build) build_rdc ;;
+    outdir) print_output_directory ;;
+    *) die "Invalid target $TARGET" ;;
 esac

 echo "Operation complete"
--- a/tools/rocm-build/build_rocal.sh
+++ b/tools/rocm-build/build_rocal.sh
@@ -1,35 +1,49 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src rocAL

 build_rocal() {

-    echo "Start build"
-
-    # Enable ASAN
-    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
-        set_asan_env_vars
-        set_address_sanitizer_on
+    if [ "$DISTRO_ID" = "mariner-2.0" ] || [ "$DISTRO_ID" = "azurelinux-3.0" ] ; then
+        echo "Not building rocal for ${DISTRO_ID}. Exiting..."
+        return 0
    fi

-#    python3 ${COMPONENT_SRC}/rocAL-setup.py
+    echo "Start build"
+
+    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
+        ack_and_skip_static
+    fi
+
+    # Enable ASAN
+    # Temporarily disable ASAN for rocal - SWDEV-471302
+    #if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+    #    set_asan_env_vars
+    #    set_address_sanitizer_on
+    #fi
    pushd /tmp
+
    # PyBind11
+    rm -rf pybind11
    git clone -b v2.11.1  https://github.com/pybind/pybind11
    cd pybind11 && mkdir build && cd build
    cmake -DDOWNLOAD_CATCH=ON -DDOWNLOAD_EIGEN=ON ../
    make -j$(nproc) && sudo make install
    cd ../..
+
    # Turbo JPEG
+    rm -rf libjpeg-turbo
    git clone -b 3.0.2 https://github.com/libjpeg-turbo/libjpeg-turbo.git
    cd libjpeg-turbo && mkdir build && cd build
    cmake -DCMAKE_INSTALL_PREFIX=/usr -DCMAKE_BUILD_TYPE=RELEASE -DENABLE_STATIC=FALSE -DCMAKE_INSTALL_DEFAULT_LIBDIR=lib -DWITH_JPEG8=TRUE ..
    make -j$(nproc) && sudo make install
    cd ../..
+
    # RapidJSON
+    rm -rf rapidjson
    git clone https://github.com/Tencent/rapidjson.git
    cd rapidjson && mkdir build && cd build
    cmake .. && make -j$(nproc) && sudo make install
@@ -37,16 +51,19 @@ build_rocal() {

    mkdir -p $BUILD_DIR && cd $BUILD_DIR

-    cmake -DAMDRPP_PATH=$ROCM_PATH ${COMPONENT_SRC}
-    make -j${PROC}
+    # python3 ../rocAL-setup.py
+
+    if [[ "${DISTRO_ID}" == almalinux-8* ]]; then
+        cmake -DPYTHON_VERSION_SUGGESTED=3.8 -DAMDRPP_PATH=$ROCM_PATH ${COMPONENT_SRC}
+    else
+        cmake -DAMDRPP_PATH=$ROCM_PATH ${COMPONENT_SRC}
+    fi
+    make -j8
    cmake --build . --target PyPackageInstall
-    sudo make install
-    sudo make package
-    sudo chown -R $(id -u):$(id -g) ${BUILD_DIR}
+    make package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR
-    cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
    show_build_cache_stats
 }

@@ -59,7 +76,7 @@ clean_rocal() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_rocal ;;
+    build) build_rocal; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_rocal ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_rocalution.sh
+++ b/tools/rocm-build/build_rocalution.sh
@@ -2,7 +2,7 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 PATH=${ROCM_PATH}/bin:$PATH
 set_component_src rocALUTION
@@ -16,7 +16,7 @@ build_rocalution() {

    cd $COMPONENT_SRC

-    CXX="g++"
+    CXX=$(set_build_variables __G_++__)
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
        set_asan_env_vars
        set_address_sanitizer_on
@@ -28,17 +28,11 @@ build_rocalution() {
    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
    init_rocm_common_cmake_params

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908:xnack-;gfx90a:xnack-;gfx90a:xnack+;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
-    fi
-
    cmake \
+        ${GEN_NINJA} \
        -DSUPPORT_HIP=ON \
        ${LAUNCHER_FLAGS} \
        "${rocm_math_common_cmake_params[@]}" \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        -DBUILD_CLIENTS_SAMPLES=ON \
        -DBUILD_CLIENTS_TESTS=ON \
        -DBUILD_CLIENTS_BENCHMARKS=ON \
@@ -51,7 +45,7 @@ build_rocalution() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -65,7 +59,7 @@ clean_rocalution() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_rocalution ;;
+    build) build_rocalution; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_rocalution ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_rocblas.sh
+++ b/tools/rocm-build/build_rocblas.sh
@@ -2,12 +2,13 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src rocBLAS
 DEPS_DIR=${HOME}/rocblas

 stage2_command_args "$@"
+disable_debug_package_generation

 build_rocblas() {
    echo "Start build"
@@ -17,28 +18,37 @@ build_rocblas() {
        SHARED_LIBS="OFF"
    fi

+    #Removed GPU ARCHS from here as it will be part of compute_utils.sh ROCMOPS-7302 & ROCMOPS-8091
+    # Temporary workaround for rocBLAS to build with ASAN as suggested in #SWDEV-314505
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
-       set_asan_env_vars
-       set_address_sanitizer_on
-       export ASAN_OPTIONS="detect_leaks=0:verify_asan_link_order=0"
+        set_asan_env_vars
+        set_address_sanitizer_on
+        export ASAN_OPTIONS="detect_leaks=0:verify_asan_link_order=0"
+        # updating GPU_ARCHS for ASAN build to supported gpu arch only SWDEV-479178
+        #GPU_ARCHS="gfx90a:xnack+;gfx942:xnack+" #This will be part of compute_utils.sh ROCMOPS-7302 & ROCMOPS-8091
    fi
-
+    LAZY_LOADING=ON
+    SEPARATE_ARCHES=ON
    cd $COMPONENT_SRC

+    CXX=$(set_build_variables __AMD_CLANG_++__)
    mkdir -p $DEPS_DIR && cp -r /usr/blis $DEPS_DIR
    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx900;gfx906:xnack-;gfx908:xnack-;gfx90a:xnack+;gfx90a:xnack-;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
+    # if ENABLE_GPU_ARCH is set in env by Job parameter ENABLE_GPU_ARCH, then set GFX_ARCH to that value. This will override any of the case values above
+    if [ -n "$ENABLE_GPU_ARCH" ]; then
+        #setting gfx arch as part of rocm_common_cmake_params
+        set_gpu_arch "${ENABLE_GPU_ARCH}"
    fi
-    init_rocm_common_cmake_params

+    echo "C compiler: $CC"
+    echo "CXX compiler: $CXX"
+
+    init_rocm_common_cmake_params
    cmake \
        -DCMAKE_TOOLCHAIN_FILE=toolchain-linux.cmake \
        -DBUILD_DIR="${BUILD_DIR}" \
-	"${rocm_math_common_cmake_params[@]}" \
+        "${rocm_math_common_cmake_params[@]}" \
        -DROCM_DIR="${ROCM_PATH}" \
        ${LAUNCHER_FLAGS} \
        -DBUILD_SHARED_LIBS=$SHARED_LIBS \
@@ -47,11 +57,10 @@ build_rocblas() {
        -DBUILD_CLIENTS_BENCHMARKS=ON \
        -DBUILD_CLIENTS_SAMPLES=ON \
        -DLINK_BLIS=ON \
-        -DAMDGPU_TARGETS="${GPU_TARGETS}" \
        -DTensile_CODE_OBJECT_VERSION=default \
        -DTensile_LOGIC=asm_full \
-        -DTensile_SEPARATE_ARCHITECTURES=ON \
-        -DTensile_LAZY_LIBRARY_LOADING=ON \
+        -DTensile_SEPARATE_ARCHITECTURES="${SEPARATE_ARCHES}" \
+        -DTensile_LAZY_LIBRARY_LOADING="${LAZY_LOADING}" \
        -DTensile_LIBRARY_FORMAT=msgpack \
        -DBUILD_ADDRESS_SANITIZER="${ADDRESS_SANITIZER}" \
        -DTENSILE_VENV_UPGRADE_PIP=ON \
@@ -63,7 +72,7 @@ build_rocblas() {

    rm -rf _CPack_Packages/ && rm -rf ./library/src/build_tmp && find -name '*.o' -delete

-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -75,7 +84,7 @@ clean_rocblas() {
 }

 case $TARGET in
-    build) build_rocblas ;;
+    build) build_rocblas; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_rocblas ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_rocdecode.sh
+++ b/tools/rocm-build/build_rocdecode.sh
@@ -1,9 +1,17 @@
 #!/bin/bash
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
 set_component_src rocDecode
 BUILD_DEV=ON
 build_rocdecode() {
+    if [ "$DISTRO_ID" = "centos-7" ] || \
+       [ "$DISTRO_ID" = "mariner-2.0" ] || \
+       [ "$DISTRO_ID" = "azurelinux-3.0" ] || \
+       [ "$DISTRO_ID" = "debian-10" ]; then
+     echo "Not building rocDecode for ${DISTRO_ID}. Exiting..."
+     return 0
+    fi
+
    echo "Start build"

    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
@@ -11,18 +19,27 @@ build_rocdecode() {
    fi

    mkdir -p $BUILD_DIR && cd $BUILD_DIR
-#    python3 ${COMPONENT_SRC}/rocDecode-setup.py --developer OFF
-    
-    cmake -DROCM_DEP_ROCMCORE=ON ${COMPONENT_SRC}
-    make -j8
-    make install
-    make package
-    
+    # for i in {1..5}; do
+    #     python3 ../rocDecode-setup.py --developer OFF && break || {
+    #         echo "Attempt $i failed! Retrying in $((i * 30)) seconds..."
+    #         sleep $((i * 30))
+    #     }
+    # done
+    # python3 ${COMPONENT_SRC}/rocDecode-setup.py --developer OFF
+
+    init_rocm_common_cmake_params
+    cmake \
+        "${rocm_math_common_cmake_params[@]}" \
+        -DROCM_DEP_ROCMCORE=ON \
+        -DROCDECODE_ENABLE_ROCPROFILER_REGISTER=ON \
+        "${COMPONENT_SRC}"
+
    cmake --build "$BUILD_DIR" -- -j${PROC}
-    cpack -G ${PKGTYPE^^} -B ${BUILD_DIR}
+    cmake --build "$BUILD_DIR" -- install
+    cmake --build "$BUILD_DIR" -- package
+
    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR
-    cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
    show_build_cache_stats
 }
 clean_rocdecode() {
@@ -32,7 +49,7 @@ clean_rocdecode() {
 }
 stage2_command_args "$@"
 case $TARGET in
-    build) build_rocdecode ;;
+    build) build_rocdecode; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_rocdecode ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_rocfft.sh
+++ b/tools/rocm-build/build_rocfft.sh
@@ -1,7 +1,7 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 PATH=${ROCM_PATH}/bin:$PATH
 set_component_src rocFFT
@@ -9,6 +9,10 @@ set_component_src rocFFT
 build_rocfft() {
    echo "Start Build"

+    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
+        ack_and_skip_static
+    fi
+
    cd $COMPONENT_SRC

    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
@@ -18,17 +22,14 @@ build_rocfft() {
    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
    init_rocm_common_cmake_params

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908;gfx90a;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
-    fi
+    #Removed GPU ARCHS from here as it will be part of compute_utils.sh ROCMOPS-7302 & ROCMOPS-8091

-    CXX="${ROCM_PATH}/bin/hipcc" \
+    # Work around for HIP sources with C++ suffix, and force CXXFLAGS for both
+    # HIP and C++ compiles
+    CXX=$(set_build_variables __HIP_CC__) \
    cmake \
        ${LAUNCHER_FLAGS} \
        "${rocm_math_common_cmake_params[@]}" \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        -DUSE_HIP_CLANG=ON \
        -DHIP_COMPILER=clang  \
        -DBUILD_CLIENTS_SAMPLES=ON  \
@@ -41,7 +42,7 @@ build_rocfft() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -55,7 +56,7 @@ clean_rocfft() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_rocfft ;;
+    build) build_rocfft; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_rocfft ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_rocjpeg.sh
+++ b/tools/rocm-build/build_rocjpeg.sh
@@ -1,9 +1,13 @@
 #!/bin/bash
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
 set_component_src rocJPEG
 BUILD_DEV=ON
 build_rocjpeg() {
+    if [ "$DISTRO_ID" = "centos-7" ] || [ "$DISTRO_ID" = "sles-15.4" ] || [ "$DISTRO_ID" = "azurelinux-3.0" ]  || [ "$DISTRO_ID" = "debian-10" ]; then
+     echo "Not building rocJPEG for ${DISTRO_ID}. Exiting..." 
+     return 0
+    fi
    echo "Start build"

    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
@@ -13,16 +17,15 @@ build_rocjpeg() {
    mkdir -p $BUILD_DIR && cd $BUILD_DIR
    # python3 ../rocJPEG-setup.py

-    cmake -DROCM_DEP_ROCMCORE=ON "$COMPONENT_SRC"
-    make -j8
-    make install
-    make package
+    cmake ${GEN_NINJA} -DROCM_DEP_ROCMCORE=ON -DROCJPEG_ENABLE_ROCPROFILER_REGISTER=ON "$COMPONENT_SRC"
+    ninja -j8
+    ninja install
+    ninja package

    cmake --build "$BUILD_DIR" -- -j${PROC}
    cpack -G ${PKGTYPE^^}
    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR
-    cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
    show_build_cache_stats
 }
 clean_rocjpeg() {
--- a/tools/rocm-build/build_rocm-cmake.sh
+++ b/tools/rocm-build/build_rocm-cmake.sh
@@ -23,15 +23,15 @@ printUsage() {
    return 0
 }

+## ROCm cmake build (using Makefile) environment variables
+PROJ_NAME="rocm-cmake"
 TARGET="build"
 PACKAGE_ROOT="$(getPackageRoot)"
-PACKAGE_DEB="$(getPackageRoot)/deb/rocm-cmake"
-PACKAGE_RPM="$(getPackageRoot)/rpm/rocm-cmake"
-ROCM_CMAKE_BUILD_DIR="$(getBuildPath rocm-cmake)"

-ROCM_CMAKE_BUILD_DIR="$(getBuildPath rocm-cmake)"
-ROCM_CMAKE_PACKAGE_DEB="$(getPackageRoot)/deb/rocm-cmake"
-ROCM_CMAKE_PACKAGE_RPM="$(getPackageRoot)/rpm/rocm-cmake"
+## ROCm cmake build (using CMake) environment variables
+ROCM_CMAKE_BUILD_DIR="$(getBuildPath $PROJ_NAME)"
+ROCM_CMAKE_PACKAGE_DEB="$PACKAGE_ROOT/deb/$PROJ_NAME"
+ROCM_CMAKE_PACKAGE_RPM="$PACKAGE_ROOT/rpm/$PROJ_NAME"
 ROCM_CMAKE_BUILD_TYPE="debug"
 BUILD_TYPE="Debug"
 SHARED_LIBS="ON"
@@ -39,11 +39,13 @@ CLEAN_OR_OUT=0;
 PKGTYPE="deb"
 MAKETARGET="deb"

+#parse the arguments
 VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
 do
+    #echo "parocessing $1"
    case "$1" in
        (-h | --help)
                printUsage ; exit 0;;
@@ -55,11 +57,13 @@ 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)
                MAKETARGET="$2" ; shift 2;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        (*)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
    esac
@@ -115,10 +119,19 @@ print_output_directory() {
 }

 case $TARGET in
-    (clean) clean_rocm_cmake ;;
-    (build) build_rocm_cmake ;;
-    (outdir) print_output_directory ;;
-    (*) die "Invalid target $TARGET" ;;
+    (clean)
+        clean_rocm_cmake
+        ;;
+    (build)
+        build_rocm_cmake
+        build_wheel "$ROCM_CMAKE_BUILD_DIR" "$PROJ_NAME"
+        ;;
+    (outdir)
+        print_output_directory
+        ;;
+    (*)
+        die "Invalid target $TARGET"
+        ;;
 esac

 echo "Operation complete"
--- a/tools/rocm-build/build_rocm-core.sh
+++ b/tools/rocm-build/build_rocm-core.sh
@@ -13,8 +13,10 @@ printUsage() {
    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 "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -w,  --wheel              Creates python wheel package of rocm-core. 
+                                      It needs to be used along with -r option"
    echo "  -h,  --help               Prints this help"
-    echo "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
    echo
    echo "Possible values for <type>:"
    echo "  deb -> Debian format (default)"
@@ -24,6 +26,7 @@ printUsage() {
    return 0
 }

+## ROCm build (using CMake) environment variables
 PROJ_NAME="rocm-core"
 PACKAGE_ROOT="$(getPackageRoot)"
 ROCM_CORE_BUILD_DIR="$(getBuildPath rocm_core)"
@@ -38,7 +41,8 @@ MAKETARGET="deb"
 PKGTYPE="deb"
 ADDRESS_SANITIZER=false

-VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,address_sanitizer,outdir,wheel:,package: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
@@ -56,6 +60,8 @@ do
                ADDRESS_SANITIZER=true ; shift ;;
        (-s | --static)
                SHARED_LIBS="OFF" ; shift ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; shift ;;
        (-o | --outdir)
                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
        (-p | --package )
@@ -87,14 +93,11 @@ build_rocm_core() {
    if [ ! -d "$ROCM_CORE_BUILD_DIR" ]; then
       mkdir -p "$ROCM_CORE_BUILD_DIR"
    fi
-
    pushd "$ROCM_CORE_BUILD_DIR"
    cmake \
-            -DCMAKE_VERBOSE_MAKEFILE=1 \
+            $(rocm_cmake_params) \
            $(rocm_common_cmake_params) \
-            -DCMAKE_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
-            -DCPACK_PACKAGING_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
-            -DCPACK_GENERATOR="${CPACKGEN:-"DEB;RPM"}" \
+            -DBUILD_SHARED_LIBS=$SHARED_LIBS \
            -DCPACK_DEBIAN_PACKAGE_RELEASE=$CPACK_DEBIAN_PACKAGE_RELEASE \
            -DCPACK_RPM_PACKAGE_RELEASE=$CPACK_RPM_PACKAGE_RELEASE \
            -DROCM_VERSION="$ROCM_VERSION" \
@@ -109,24 +112,33 @@ build_rocm_core() {
 }

 print_output_directory() {
-     case ${PKGTYPE} in
+    case ${PKGTYPE} in
         ("deb")
             echo ${ROCM_CORE_PACKAGE_DEB};;
         ("rpm")
             echo ${ROCM_CORE_PACKAGE_RPM};;
         (*)
             echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
-     esac
-     exit
+    esac
+    exit
 }

 verifyEnvSetup

 case $TARGET in
-    (clean) clean_rocm_core ;;
-    (build) build_rocm_core ;;
-    (outdir) print_output_directory ;;
-        (*) die "Invalid target $TARGET" ;;
+    (clean)
+        clean_rocm_core
+        ;;
+    (build)
+        build_rocm_core
+        build_wheel "$ROCM_CORE_BUILD_DIR" "$PROJ_NAME"
+        ;;
+    (outdir)
+        print_output_directory
+        ;;
+    (*)
+        die "Invalid target $TARGET"
+        ;;
 esac

 echo "Operation complete"
--- a/tools/rocm-build/build_rocm-dev.sh
+++ b/tools/rocm-build/build_rocm-dev.sh
@@ -2,13 +2,18 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 stage2_command_args "$@"
+build_rocm-dev(){
+    $(dirname "${BASH_SOURCE[0]}")/build_rocm.sh -d
+    mv ${OUT_DIR}/${PKGTYPE}/meta ${OUT_DIR}/${PKGTYPE}/rocm-dev
+}
+

 case $TARGET in
-    build) echo "end of rocm-dev build..." ;;
-    outdir) ;;
-    clean) echo "Cleaning rocm-dev is not required..." ;;
+    build) build_rocm-dev ;;
+    outdir) echo "${OUT_DIR}/${PKGTYPE}/rocm-dev" ;;
+    clean) rm -rf ${OUT_DIR}/${PKGTYPE}/rocm-dev ;;
    *) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_rocm-gdb.sh
+++ b/tools/rocm-build/build_rocm-gdb.sh
@@ -1,7 +1,6 @@
 #!/bin/bash
-
 source "${BASH_SOURCE%/*}/compute_utils.sh" || return
-
+# Can't use -R or -r in here
 remove_make_r_flags

 printUsage() {
@@ -15,8 +14,10 @@ printUsage() {
    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 "  -s,  --static             Component/Build does not support static builds just accepting this param & ignore. No effect of the param on this build"
+    echo "  -w,  --wheel              Creates python wheel package of rocm-gdb.
+                                      It needs to be used along with -r option"
    echo "  -h,  --help               Prints this help"
-    echo "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
    echo
    echo "Possible values for <type>:"
    echo "  deb -> Debian format (default)"
@@ -32,23 +33,32 @@ toStdoutStderr(){
 }

 linkFiles(){
+    # Attempt to use hard links first for speed and save disk,
+    # if that fails do a copy
    cp -lfR "$1" "$2" || cp -fR "$1" "$2"
 }

+## Build environment variables
 PROJ_NAME=rocm-gdb
 TARGET=build
-MAKETARGET=deb
-BUILD_DIR=$(getBuildPath $PROJ_NAME)
-PACKAGE_DEB=$(getPackageRoot)/deb/$PROJ_NAME
-PACKAGE_RPM=$(getPackageRoot)/rpm/$PROJ_NAME
-MAKE_OPTS="$DASH_JAY"
+MAKETARGET=deb                                  # Not currently used
+BUILD_DIR=$(getBuildPath $PROJ_NAME)            # e.g. out/ubuntu.16.04/16.04/build/rocm-gdb
+PACKAGE_DEB=$(getPackageRoot)/deb/$PROJ_NAME    # e.g. out/ubuntu.16.04/16.04/deb
+PACKAGE_RPM=$(getPackageRoot)/rpm/$PROJ_NAME    # e.g. out/ubuntu.16.04/16.04/rpm
+MAKE_OPTS="$DASH_JAY"                           # e.g. -j 56
 BUG_URL="https://github.com/ROCm-Developer-Tools/ROCgdb/issues"
 SHARED_LIBS="ON"
 CLEAN_OR_OUT=0;
 MAKETARGET="deb"
 PKGTYPE="deb"
 LDFLAGS="$LDFLAGS -Wl,--enable-new-dtags"
-LIB_AMD_PYTHON="libamdpython.so"
+LIB_AMD_PYTHON="amdpythonlib.so"                #lib name which replaces required python lib
+LIB_AMD_PYTHON_DIR_PATH=${ROCM_INSTALL_PATH}/lib
+
+
+# A curated list of things to keep. It would be safer to have a list
+# of things to remove, as failing to remove something is usually much
+# less harmful than removeing too much.

 tokeep=(
    main${ROCM_INSTALL_PATH}/bin/rocgdb
@@ -77,13 +87,18 @@ tokeep=(
 keep_wanted_files(){
    (
        cd "$BUILD_DIR/package/"
+        # generate the keep pattern as one name per line
        printf -v keeppattern '%s\n' "${tokeep[@]}"
        find main/opt -type f | grep -xv "$keeppattern" | xargs -r rm
+        # prune empty directories
        find main/opt -type d -empty -delete
    )
    return 0
 }

+# Move to a function so that both package_deb and package_rpm can call
+# and remove the depenency of package_rpm_tests on package_deb.
+# Copy the required files to create a stand-alone testsuite.
 copy_testsuite_files() {
 (
 dest="$BUILD_DIR/package/tests${ROCM_INSTALL_PATH}/test/gdb/"
@@ -113,43 +128,57 @@ clean() {
    rm -rf $PACKAGE_RPM
 }

+# set the lexically bound variable VERSION to the current version
+# A passed parameter gives the default
+# Note that this might need to be stripped further as it might have
+# things like "-git" which is not acceptable as a version to rpm.
 get_version(){
    VERSION=$(sed -n 's/^.*char version[^"]*"\([^"]*\)".*;.*/\1/p' $BUILD_DIR/gdb/version.c || : )
    VERSION=${VERSION:-$1}
 }

 package_deb(){
+    # Package main binary.
+    # TODO package documentation when we build some.
    mkdir -p "$BUILD_DIR/package/main/DEBIAN"
+    # Extract version from build
    local VERSION
    get_version unknown
+    # add upgrade related version sub-field
    VERSION="${VERSION}.${ROCM_LIBPATCH_VERSION}"
+
    #create postinstall and prerm
-    grep -v '^# ' > "$BUILD_DIR/package/main/DEBIAN/preinst" <<EOF
+    grep -v '^# ' > "$BUILD_DIR/package/main/DEBIAN/postinst" <<EOF
 #!/bin/sh
-# Pre-installation script commands
-echo "Running pre-installation script..."
-mkdir -p ${ROCM_INSTALL_PATH}/lib
-PYTHON_LIB_INSTALLED=\$(ldconfig -p | awk '/libpython3/ { print \$NF; exit}')
-ln -s \$PYTHON_LIB_INSTALLED ${ROCM_INSTALL_PATH}/lib/$LIB_AMD_PYTHON
-echo "pre-installation done."
+
+# Post-installation script commands
+echo "Running post-installation script..."
+mkdir -p $LIB_AMD_PYTHON_DIR_PATH
+# Choosing the lowest version of the libpython3 installed.
+PYTHON_LIB_INSTALLED=\$(find /lib/ -name 'libpython3*.so' | head -n 1)
+echo "Installing rocm-gdb with [\$PYTHON_LIB_INSTALLED]."
+ln -s \$PYTHON_LIB_INSTALLED $LIB_AMD_PYTHON_DIR_PATH/$LIB_AMD_PYTHON
+echo "post-installation done."
 EOF
-    grep -v '^# ' > "$BUILD_DIR/package/main/DEBIAN/postrm" <<EOF
+
+
+    grep -v '^# ' > "$BUILD_DIR/package/main/DEBIAN/prerm" <<EOF
 #!/bin/sh
-# Post-uninstallation script commands
-echo "Running post-uninstallation script..."
-PYTHON_LINK_BY_OPENCL=\$(ldconfig -p | awk '/libpython3/ { print \$NF; exit}'  | awk -F'/' '{print \$NF}')
-rm -f ${ROCM_INSTALL_PATH}/lib/\$PYTHON_LINK_BY_OPENCL
-rm -f ${ROCM_INSTALL_PATH}/lib/$LIB_AMD_PYTHON
-if [ -L "${ROCM_INSTALL_PATH}/lib/$LIB_AMD_PYTHON" ]  || \
-   [ -L "${ROCM_INSTALL_PATH}/lib/\$PYTHON_LINK_BY_OPENCL" ] ; then
+
+# Pre-uninstallation script commands
+echo "Running pre-uninstallation script..."
+rm -f $LIB_AMD_PYTHON_DIR_PATH/$LIB_AMD_PYTHON
+if [ -L "$LIB_AMD_PYTHON_DIR_PATH/$LIB_AMD_PYTHON" ] ; then
        echo " some rocm-gdb requisite libs could not be removed"
 else
        echo " all requisite libs removed successfully "
 fi
-echo "post-uninstallation done."
+echo "pre-uninstallation done."
 EOF
-    chmod +x $BUILD_DIR/package/main/DEBIAN/postrm
-    chmod +x $BUILD_DIR/package/main/DEBIAN/preinst
+
+    chmod +x $BUILD_DIR/package/main/DEBIAN/prerm
+    chmod +x $BUILD_DIR/package/main/DEBIAN/postinst
+
    # Create control file, with variable substitution.
    # Lines with # at the start are removed, to allow for comments
    mkdir "$BUILD_DIR/debian"
@@ -167,8 +196,9 @@ Section: utils
 Architecture: amd64
 Essential: no
 Priority: optional
-Depends: \${shlibs:Depends}, rocm-dbgapi, rocm-core
+Depends: \${shlibs:Depends}, rocm-dbgapi, rocm-core, python3-dev
 EOF
+
    # Use dpkg-shlibdeps to list shlib dependencies, the result is placed
    # in $BUILD_DIR/debian/substvars.
    (
@@ -179,6 +209,7 @@ EOF
 	fi
 	dpkg-shlibdeps --ignore-missing-info  -e "$BUILD_DIR/package/main/${ROCM_INSTALL_PATH}/bin/rocgdb"
    )
+
    # Generate the final DEBIAN/control, and substitute the shlibs:Depends.
    # This is a bit unorthodox as we are only using bits and pieces of the
    # dpkg tools.
@@ -191,8 +222,10 @@ EOF
 	    -e "s/\\$\{shlibs:Depends\}/$SHLIB_DEPS/" \
 	    < debian/control > "$BUILD_DIR/package/main/DEBIAN/control"
    )
+
    mkdir -p "$OUT_DIR/deb/$PROJ_NAME"
    fakeroot dpkg-deb -Zgzip --build "$BUILD_DIR/package/main" "$OUT_DIR/deb/$PROJ_NAME"
+
    # Package the tests so they can be run on a test slave
    mkdir -p "$BUILD_DIR/package/tests/DEBIAN"
    mkdir -p "$BUILD_DIR/package/tests/${ROCM_INSTALL_PATH}/test/gdb"
@@ -213,21 +246,28 @@ Priority: optional
 # rocm-core as policy says everything to depend on rocm-core
 Depends: ${PROJ_NAME} (=${VERSION}-${CPACK_DEBIAN_PACKAGE_RELEASE}), dejagnu, rocm-core, make
 EOF
+
    copy_testsuite_files
    fakeroot dpkg-deb -Zgzip --build "$BUILD_DIR/package/tests" "$OUT_DIR/deb/$PROJ_NAME"
 }

 package_rpm(){
+    # TODO, use this to package the tests as well. In the mean time hard code the package
    set -- rocm-gdb
-    local packageDir="$BUILD_DIR/package_rpm/$1"
-    local specFile="$packageDir/$1.spec"
-    local packageRpm="$packageDir/rpm"
-
+    local packageDir="$BUILD_DIR/package_rpm/$1" # e.g. out/ubuntu-16.04/16.04/build/rocm-gdb/package_rpm/main
+    local specFile="$packageDir/$1.spec"         # The generated spec file
+    local packageRpm="$packageDir/rpm"           # The RPM infrastructure
+    # Extract version from build. If more than one line matches then
+    # expect failures. Solution is to find which version is the wanted one.
    local VERSION
    get_version 0.0.0
-
+    # add upgrade related version sub-field
    VERSION=${VERSION}.${ROCM_LIBPATCH_VERSION}

+    # get the __os_install_post macro, edit it to remove the python bytecode generation
+    # rpm --showrc shows the default macros with priority level (-14) in this case
+    # so remove everything before this macro, everything after it, remove the offending line,
+    # and make it available as the ospost variable to insert into spec file
    local ospost="$(echo '%define __os_install_post \'
                    rpm --showrc | sed '1,/^-14: __os_install_post/d;
                                       /^-14:/,$d;/^%{nil}/!s/$/ \\/;
@@ -238,6 +278,8 @@ package_rpm(){

    mkdir -p "$packageDir"

+    # Create the spec file.
+    # Allow comments in the generation of the specfile, may be overkill.
    grep -v '^## ' <<- EOF > $specFile
 ## Set up where this stuff goes
 %define _topdir $packageRpm
@@ -256,8 +298,7 @@ Version: ${VERSION//-/_}
 Release: ${CPACK_RPM_PACKAGE_RELEASE}%{?dist}
 License: GPL
 Prefix: ${ROCM_INSTALL_PATH}
-Requires: rocm-core
-Provides: $LIB_AMD_PYTHON()(64bit)
+Requires: rocm-core, rocm-dbgapi

 %description
 This is ROCgdb, the ROCm source-level debugger for Linux, based on
@@ -278,27 +319,6 @@ https://github.com/RadeonOpenCompute/ROCm
 ## into the local RPM_BUILD_ROOT and left the defaults take over. Need
 ## to quote the dollar signs as we want rpm to expand them when it is
 ## run, rather than the shell when we build the spec file.
-%pre
-# Post-install script commands
-echo "Running post-install script..."
-mkdir -p ${ROCM_INSTALL_PATH}/lib
-PYTHON_LIB_INSTALLED=\$(ldconfig -p | awk '/libpython3/ { print \$NF; exit}')
-ln -s \$PYTHON_LIB_INSTALLED ${ROCM_INSTALL_PATH}/lib/$LIB_AMD_PYTHON
-
-%postun
-# Post-uninstallation script commands
-echo "Running post-uninstallation script..."
-PYTHON_LINK_BY_OPENCL=\$(ldconfig -p | awk '/libpython3/ { print \$NF; exit}'  | awk -F'/' '{print \$NF}')
-rm -f ${ROCM_INSTALL_PATH}/lib/\$PYTHON_LINK_BY_OPENCL
-rm -f ${ROCM_INSTALL_PATH}/lib/$LIB_AMD_PYTHON
-if [ -L "${ROCM_INSTALL_PATH}/lib/$LIB_AMD_PYTHON" ]  || \
-   [ -L "${ROCM_INSTALL_PATH}/lib/\$PYTHON_LINK_BY_OPENCL" ] ; then
-        echo " some rocm-gdb requisite libs could not be removed"
-else
-        echo " all requisite libs removed successfully "
-fi
-echo "post-uninstallation done."
-
 %install
 rm -rf \$RPM_BUILD_ROOT
 mkdir -p \$RPM_BUILD_ROOT
@@ -307,24 +327,27 @@ cp -ar $BUILD_DIR/package/main/opt \$RPM_BUILD_ROOT/opt
 ## The file section is generated by walking the tree.
 %files
 EOF
-
+    # Now generate the files
    find $BUILD_DIR/package/main/opt -type d | sed "s:$BUILD_DIR/package/main:%dir :" >> $specFile
    find $BUILD_DIR/package/main/opt ! -type d | sed "s:$BUILD_DIR/package/main::" >> $specFile

    rpmbuild --define "_topdir $packageRpm" -ba $specFile
-
+    # Now copy it to final location
    mkdir -p "$PACKAGE_RPM"        # e.g. out/ubuntu-16.04/16.04/rpm/rocm-gdb
    mv $packageRpm/RPMS/x86_64/*.rpm "$PACKAGE_RPM"
 }

 package_rpm_tests(){
+    # TODO, use this to package the tests as well. In the mean time hard code the package
    set -- rocm-gdb-tests
-    local packageDir="$BUILD_DIR/package_rpm/$1"
-    local specFile="$packageDir/$1.spec"
-    local packageRpm="$packageDir/rpm"
-
+    local packageDir="$BUILD_DIR/package_rpm/$1" # e.g. out/ubuntu-16.04/16.04/build/rocm-gdb/package_rpm/main
+    local specFile="$packageDir/$1.spec"         # The generated spec file
+    local packageRpm="$packageDir/rpm"           # The RPM infrastructure
+    # Extract version from build. If more than one line matches then
+    # expect failures. Solution is to find which version is the wanted one.
    local VERSION
    get_version 0.0.0
+    # add upgrade related version sub-field
    VERSION=${VERSION}.${ROCM_LIBPATCH_VERSION}
    local RELEASE=${CPACK_RPM_PACKAGE_RELEASE}%{?dist}

@@ -333,6 +356,8 @@ package_rpm_tests(){

    mkdir -p "$packageRpm"

+    # Create the spec file.
+    # Allow comments in the generation of the specfile, may be overkill.
    local ospost="$(echo '%define __os_install_post \'
                    rpm --showrc | sed '1,/^-14: __os_install_post/d;
                                       /^-14:/,$d;/^%{nil}/!s/$/ \\/;
@@ -385,10 +410,15 @@ EOF

    copy_testsuite_files

+    # rpm wont resolve the dependency if the unversioned python (#!/usr/bin/python) is used.
+    # /usr/bin/python - this file is not owned by any package.
+    # So find and append the version to the /usr/bin/python as /usr/bin/python3
+    # By updating this, the python3 requirement will be provided by the python3 package.
    find $BUILD_DIR/package/tests/opt -type f -exec sed -i '1s:^#! */usr/bin/python\>:&3:' {} +

    rpmbuild --define "_topdir $packageRpm" -ba $specFile
-    mkdir -p "$PACKAGE_RPM"
+    # Now copy it to final location
+    mkdir -p "$PACKAGE_RPM"        # e.g. out/ubuntu-16.04/16.04/rpm/rocm-gdb
    mv $packageRpm/RPMS/x86_64/*.rpm "$PACKAGE_RPM"
 }

@@ -396,20 +426,29 @@ build() {
    if [ ! -e "$ROCM_GDB_ROOT/configure" ]
    then
        toStdoutStderr "No $ROCM_GDB_ROOT/configure file, skippping rocm-gdb"
-        exit 0
+        exit 0                        # This is not an error
    fi
    local pythonver=python3
    if [[ "$DISTRO_ID" == "ubuntu-18.04" ]]; then
        pythonver=python3.8
    fi
-
+    # Workaround for rocm-gdb failure due to texlive on RHEL-9 & CentOS-9 SWDEV-339596
    if [[ "$DISTRO_ID" == "centos-9" ]] || [[ "$DISTRO_ID" == "rhel-9.0" ]]; then
        fmtutil-user --missing
    fi
    echo "Building $PROJ_NAME"
    mkdir -p "$BUILD_DIR"
    cd "$BUILD_DIR" || die "Failed to cd to '$BUILD_DIR'"
-
+    # Build instructions taken from the README.ROCM with the addition
+    # of --with-pkgversion
+    #
+    # The "new" way of specifying the path to the amd-dbgapi library is by
+    # setting PKG_CONFIG_PATH.  The --with-amd-dbgapi flag is used to ensure
+    # that if amd-dbgapi is not found, configure fails.
+    #
+    # --with-rocm-dbgapi is kept for now, to ease the transition.  It can be
+    # removed once it is determined that we don't need to build source trees
+    # using the "old" way.
    $ROCM_GDB_ROOT/configure --program-prefix=roc --prefix="${ROCM_INSTALL_PATH}" \
 	--htmldir="\${prefix}/share/html" --pdfdir="\${prefix}/share/doc/rocgdb" \
 	--infodir="\${prefix}/share/info/rocgdb" \
@@ -441,24 +480,31 @@ build() {
 	LDFLAGS="$LDFLAGS"
    LD_RUN_PATH='${ORIGIN}/../lib' make $MAKE_OPTS

-    REPLACE_LIB_NAME=$(ldd -d $BUILD_DIR/gdb/gdb |awk '/libpython/{print $1}')
-    echo "Replacing $REPLACE_LIB_NAME with $LIB_AMD_PYTHON"
-    patchelf --replace-needed $REPLACE_LIB_NAME $LIB_AMD_PYTHON $BUILD_DIR/gdb/gdb
+    if [[ "$DISTRO_ID" == "ubuntu"* ]]; then
+        #changing the python lib requirement in the built gdb for ubuntu builds
+        REPLACE_LIB_NAME=$(ldd -d $BUILD_DIR/gdb/gdb |awk '/libpython/{print $1}')
+        echo "Replacing $REPLACE_LIB_NAME with $LIB_AMD_PYTHON"
+        patchelf --replace-needed $REPLACE_LIB_NAME $LIB_AMD_PYTHON $BUILD_DIR/gdb/gdb
+    fi

    mkdir -p $BUILD_DIR/package/main${ROCM_INSTALL_PATH}/{share/rocgdb,bin}
-
+    # Install gdb
    make $MAKE_OPTS -C gdb DESTDIR=$BUILD_DIR/package/main install install-pdf install-html
-
+    # Install binutils for coremerge and coremerge manpage.
    make $MAKE_OPTS -C binutils DESTDIR=$BUILD_DIR/package/main install
-
+    # Add in the AMD licences file
    linkFiles $ROCM_GDB_ROOT/gdb/NOTICES.txt $BUILD_DIR/package/main${ROCM_INSTALL_PATH}/share/doc/rocgdb
    keep_wanted_files
-
+    # If a variable CPACKGEN indicates only one type of packaging is required
+    # then don't bother making the other. This variable is set up in compute_utils.sh
+    # Default to building both package types if variable is not set
+    # Use "[ var = value ] || ..." rather than "[ var != value ] &&" so expression
+    # evaluates to true, and set -e does not cause script to exit.
    [ "${CPACKGEN}" = "DEB" ] || package_rpm && package_rpm_tests
    [ "${CPACKGEN}" = "RPM" ] || package_deb
 }

-
+# See if the code exists
 print_output_directory() {
    case ${PKGTYPE} in
        ("deb")
@@ -475,11 +521,14 @@ verifyEnvSetup

 main(){

-VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
+
 ASAN_BUILD="no"
 while true ;
 do
+    #echo "parocessing $1"
    case "$1" in
        (-h | --help)
                printUsage ; exit 0;;
@@ -490,9 +539,11 @@ do
        (-a | --address_sanitizer)
                set_asan_env_vars
                set_address_sanitizer_on
-                ASAN_BUILD="yes" ; shift ;;
+		ASAN_BUILD="yes" ; shift ;;
        (-s | --static)
                ack_and_skip_static ;;
+        (-w | --wheel)
+                echo " Wheel"; WHEEL_PACKAGE=true ; shift;;
        (-o | --outdir)
                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
        (-p | --package)                 #FIXME
@@ -505,6 +556,7 @@ do
    esac
 done

+# If building with Clang, we need to build in C++17 mode, to avoid some build problems
 if [[ $CXX == *"clang++" ]]
 then
    CXX="$CXX -std=gnu++17"
@@ -518,17 +570,28 @@ if [ $RET_CONFLICT -ge 30 ]; then
 fi

    case $TARGET in
-        ("clean") clean ;;
-        ("build") build ;;
-        ("outdir") print_output_directory ;;
-        (*) die "Invalid target $TARGET" ;;
+        ("clean")
+            clean
+            ;;
+        ("build")
+            build
+            build_wheel "$BUILD_DIR" "$PROJ_NAME"
+            ;;
+        ("outdir")
+            print_output_directory
+            ;;
+        (*)
+            die "Invalid target $TARGET"
+            ;;
    esac
+
    echo "Operation complete"
 }

+# If this script is not being sourced, then run it.
 if [ "$0" = "$BASH_SOURCE" ]
 then
    main "$@"
 else
-    set +e
+    set +e                       # Undo the damage from compute_utils.sh
 fi
--- a/tools/rocm-build/build_rocm_smi_lib.sh
+++ b/tools/rocm-build/build_rocm_smi_lib.sh
@@ -11,6 +11,8 @@ printUsage() {
    echo "  -r,  --release            Build non-debug version rocm_smi (default is debug)"
    echo "  -a,  --address_sanitizer  Enable address sanitizer"
    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -w,  --wheel              Creates python wheel package of rocm-smi. 
+                                      It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of type referred to by pkg_type"
    echo "  -p,  --package <type>     Specify packaging format"
    echo "  -32,                      Build 32b version (default is 64b)"
@@ -22,7 +24,10 @@ printUsage() {

    return 0
 }
+# RSMI ==> deb/rpm package target
+# ROCM_SMI ==> rocm_smi library target and header

+# ROCM_SMI
 PROJ_NAME="rsmi"
 PACKAGE_ROOT="$(getPackageRoot)"
 TARGET="build"
@@ -30,15 +35,18 @@ TARGET="build"
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE="$(getIncludePath)"

+# RSMI
 RSMI_BUILD_DIR=$(getBuildPath rsmi)
 RSMI_PACKAGE_DEB_DIR="$(getPackageRoot)/deb/$PROJ_NAME"
 RSMI_PACKAGE_RPM_DIR="$(getPackageRoot)/rpm/$PROJ_NAME"
 RSMI_BUILD_TYPE="debug"
 BUILD_TYPE="Debug"

+# BUILD ARGUMENTS
 MAKETARGET="deb"
 MAKEARG="$DASH_JAY O=$RSMI_BUILD_DIR"
 RSMI_MAKE_OPTS="$DASH_JAY O=$RSMI_BUILD_DIR -C $RSMI_BUILD_DIR"
+# The following should be 64 (default) 32 for 32b
 ROCM_SMI_BLD_BITS=64
 RSMI_PKG_NAME_ROOT="rocm-smi-lib"
 RSMI_PKG_NAME="${RSMI_PKG_NAME_ROOT}${ROCM_SMI_BLD_BITS}"
@@ -46,11 +54,13 @@ SHARED_LIBS="ON"
 CLEAN_OR_OUT=0;
 PKGTYPE="deb"

-VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
 do
+    #echo "parocessing $1"
    case "$1" in
        (-h | --help)
                printUsage ; exit 0;;
@@ -65,13 +75,15 @@ do
                ADDRESS_SANITIZER=true ; 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)
                MAKETARGET="$2" ; shift 2;;
        (-32)
                ROCM_SMI_BLD_BITS="32"; shift ;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        (*)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
    esac
@@ -109,6 +121,7 @@ build_rsmi() {
            -DBUILD_SHARED_LIBS=$SHARED_LIBS \
 	    $(rocm_common_cmake_params) \
            $(rocm_cmake_params) \
+            ${GEN_NINJA} \
            -DENABLE_LDCONFIG=OFF \
            -DROCM_SMI_PACKAGE="${RSMI_PKG_NAME}" \
            -DCPACK_PACKAGE_VERSION_MAJOR="1" \
@@ -116,14 +129,14 @@ build_rsmi() {
            -DCPACK_PACKAGE_VERSION_PATCH="0" \
            -DADDRESS_SANITIZER="$ADDRESS_SANITIZER" \
            -DBUILD_TESTS=ON \
-            "$ROCM_SMI_LIB_ROOT"
+            -S "$ROCM_SMI_LIB_ROOT"
        popd
    fi

    echo "Making rocm_smi package:"
-    cmake --build "$RSMI_BUILD_DIR" -- $RSMI_MAKE_OPTS
-    cmake --build "$RSMI_BUILD_DIR" -- $RSMI_MAKE_OPTS install
-    cmake --build "$RSMI_BUILD_DIR" -- $RSMI_MAKE_OPTS package
+    cmake --build "$RSMI_BUILD_DIR" -- $DASH_JAY
+    cmake --build "$RSMI_BUILD_DIR" -- $DASH_JAY install
+    cmake --build "$RSMI_BUILD_DIR" -- $DASH_JAY package

    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$RSMI_PACKAGE_DEB_DIR" $RSMI_BUILD_DIR/*.deb
    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$RSMI_PACKAGE_RPM_DIR" $RSMI_BUILD_DIR/*.rpm
@@ -145,10 +158,11 @@ verifyEnvSetup

 case $TARGET in
    (clean) clean_rsmi ;;
-    (build) build_rsmi ;;
+    (build) build_rsmi; build_wheel "$RSMI_BUILD_DIR" "$PROJ_NAME" ;;
    (outdir) print_output_directory ;;
    (*) die "Invalid target $TARGET" ;;
 esac

 echo "Operation complete"
 exit 0
+
--- a/tools/rocm-build/build_rocminfo.sh
+++ b/tools/rocm-build/build_rocminfo.sh
@@ -10,7 +10,9 @@ printUsage() {
    echo "  -c,  --clean              Removes all rocminfo build artifacts"
    echo "  -r,  --release            Build non-debug version rocminfo (default is debug)"
    echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
+    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -w,  --wheel              Creates python wheel package of rocminfo. 
+                                      It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
                                      type referred to by pkg_type"
    echo "  -h,  --help               Prints this help"
@@ -39,7 +41,8 @@ MAKETARGET="deb"
 PKGTYPE="deb"


-VALID_STR=`getopt -o hcraso:g: --long help,clean,release,static,address_sanitizer,outdir:,gpu_list: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswo:g: --long help,clean,release,static,wheel,address_sanitizer,outdir:,gpu_list: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
@@ -56,11 +59,13 @@ do
                set_address_sanitizer_on ; 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 ;;
        (-g | --gpu_list)
                GPU_LIST="$2" ; shift 2;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        (*)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
    esac
@@ -124,7 +129,7 @@ print_output_directory() {

 case $TARGET in
    (clean) clean_rocminfo ;;
-    (build) build_rocminfo ;;
+    (build) build_rocminfo; build_wheel "$ROCMINFO_BUILD_DIR" "$PROJ_NAME" ;;
   (outdir) print_output_directory ;;
        (*) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_rocmvalidationsuite.sh
+++ b/tools/rocm-build/build_rocmvalidationsuite.sh
@@ -1,10 +1,11 @@
 #!/bin/bash

 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src ROCmValidationSuite

+# Internal RVS libraries need the extra RPATH
 ROCM_RVS_LIB_RPATH="\$ORIGIN/.."

 build_rocmvalidationsuite() {
@@ -14,6 +15,7 @@ build_rocmvalidationsuite() {
        ack_and_skip_static
    fi

+    CXX=$(set_build_variables __CXX__)
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
       set_asan_env_vars
       set_address_sanitizer_on
@@ -22,12 +24,12 @@ build_rocmvalidationsuite() {
    cd "${COMPONENT_SRC}"
    mkdir -p "$BUILD_DIR"
    init_rocm_common_cmake_params
-
+    #  rocm_common_cmake_params provides the default rpath for rocm executables and libraries
+    # Override cmake shared linker flags, since rvs libraries requires RUNPATH - $ORIGIN:$ORIGIN/..
    cmake \
        "${rocm_math_common_cmake_params[@]}" \
        -DFETCH_ROCMPATH_FROM_ROCMCORE=ON \
        -DCMAKE_SHARED_LINKER_FLAGS_INIT="-Wl,--enable-new-dtags,--build-id=sha1,--rpath,$ROCM_LIB_RPATH:$ROCM_RVS_LIB_RPATH" \
-        -DRVS_BUILD_TESTS=FALSE \
        -B "$BUILD_DIR" \
        "$COMPONENT_SRC"

@@ -50,7 +52,7 @@ clean_rocmvalidationsuite() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_rocmvalidationsuite ;;
+    build) build_rocmvalidationsuite; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_rocmvalidationsuite ;;
    *) die "Invalid target ${TARGET}" ;;
--- a/tools/rocm-build/build_rocprim.sh
+++ b/tools/rocm-build/build_rocprim.sh
@@ -2,7 +2,7 @@

 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src rocPRIM

@@ -10,12 +10,18 @@ build_rocprim() {
    echo "Start build"

    cd $COMPONENT_SRC
+    #  Temporary Fix as suggested in #SWDEV-314510
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+       #Set ASAN flags
       set_asan_env_vars
       set_address_sanitizer_on
+       # ASAN packaging is not required for rocPRIM, since its header only package
+       # Setting the asan_cmake_params to false will disable ASAN packaging
       ASAN_CMAKE_PARAMS="false"
    fi

+    # Enable/Disable Static Flag to be used for
+    # Package Dependecies during static/non-static builds
    SHARED_LIBS="ON"
    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
        SHARED_LIBS="OFF"
@@ -23,18 +29,14 @@ build_rocprim() {

    mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908:xnack-;gfx90a:xnack-;gfx90a:xnack+;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
-    fi
+    #Removed GPU ARCHS from here as it will be part of compute_utils.sh ROCMOPS-7302 & ROCMOPS-8091

    init_rocm_common_cmake_params
-    CXX="${ROCM_PATH}/bin/hipcc" \
+    CXX=$(set_build_variables __HIP_CC__) \
    cmake \
+        ${GEN_NINJA} \
        ${LAUNCHER_FLAGS} \
        "${rocm_math_common_cmake_params[@]}" \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        -DBUILD_BENCHMARK=OFF \
        -DBUILD_TEST=ON \
        -DBUILD_SHARED_LIBS=$SHARED_LIBS \
@@ -46,7 +48,7 @@ build_rocprim() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -60,7 +62,7 @@ clean_rocprim() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_rocprim ;;
+    build) build_rocprim; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_rocprim ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_rocprofiler-compute.sh
+++ b/tools/rocm-build/build_rocprofiler-compute.sh
@@ -26,6 +26,7 @@ printUsage() {
    return 0
 }

+## Build environment variables
 API_NAME="rocprofiler-compute"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -43,11 +44,13 @@ CLEAN_OR_OUT=0;
 MAKETARGET="deb"
 PKGTYPE="deb"

+#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
+    #echo "parocessing $1"
    case "$1" in
        -h | --help)
                printUsage ; exit 0;;
@@ -66,7 +69,7 @@ do
                ack_and_skip_static ;;
        -w | --wheel)
                WHEEL_PACKAGE=true ; shift ;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        *)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
    esac
@@ -108,13 +111,14 @@ build() {
        cmake \
            $(rocm_cmake_params) \
            $(rocm_common_cmake_params) \
+            ${GEN_NINJA} \
            -DCHECK_PYTHON_DEPS=NO \
            -DPYTHON_DEPS=${BUILD_DIR}/python-libs \
-            "$ROCPROFILER_COMPUTE_ROOT"
+            -S "$ROCPROFILER_COMPUTE_ROOT"
    fi
-    make $MAKE_OPTS
-    make $MAKE_OPTS install
-    make $MAKE_OPTS package
+    ninja $MAKE_OPTS
+    ninja $MAKE_OPTS install
+    ninja $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
@@ -136,7 +140,7 @@ verifyEnvSetup

 case "$TARGET" in
    (clean) clean ;;
-    (build) build ;;
+    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
    (outdir) print_output_directory ;;
    (*) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_rocprofiler-register.sh
+++ b/tools/rocm-build/build_rocprofiler-register.sh
@@ -12,6 +12,7 @@ printUsage() {
    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 "  -w,  --wheel              Creates python wheel package of rocprofiler-register. It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
    type referred to by pkg_type"
    echo "  -h,  --help               Prints this help"
@@ -24,6 +25,7 @@ printUsage() {
    return 0
 }

+## Build environment variables
 API_NAME="rocprofiler-register"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -44,11 +46,13 @@ MAKETARGET="deb"
 PKGTYPE="deb"


-VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,address_sanitizer,wheeloutdir:,package: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
 do
+    #echo "parocessing $1"
    case "$1" in
        (-h | --help)
                printUsage ; exit 0;;
@@ -61,11 +65,13 @@ do
                set_address_sanitizer_on ; 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)
                MAKETARGET="$2" ; shift 2;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        (*)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
    esac
@@ -99,17 +105,18 @@ build() {
        cmake \
            $(rocm_cmake_params) \
            $(rocm_common_cmake_params) \
+            ${GEN_NINJA} \
            -DBUILD_SHARED_LIBS=$SHARED_LIBS \
            -DENABLE_LDCONFIG=OFF \
            -DROCPROFILER_REGISTER_BUILD_CI=1 \
            -DROCPROFILER_REGISTER_BUILD_TESTS=1 \
            -DROCPROFILER_REGISTER_BUILD_SAMPLES=1 \
-            "$ROCPROFILER_REGISTER_ROOT"
+            -S "$ROCPROFILER_REGISTER_ROOT"
        popd
    fi
-    make $MAKE_OPTS
-    make $MAKE_OPTS install
-    make $MAKE_OPTS package
+    ninja $MAKE_OPTS
+    ninja $MAKE_OPTS install
+    ninja $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
@@ -131,7 +138,7 @@ verifyEnvSetup

 case "$TARGET" in
    (clean) clean ;;
-    (build) build ;;
+    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
    (outdir) print_output_directory ;;
    (*) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_rocprofiler-sdk.sh
+++ b/tools/rocm-build/build_rocprofiler-sdk.sh
@@ -26,6 +26,7 @@ printUsage() {
    return 0
 }

+## Build environment variables
 API_NAME="rocprofiler-sdk"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -35,8 +36,8 @@ PACKAGE_ROOT="$(getPackageRoot)"
 PACKAGE_LIB="$(getLibPath)"
 PACKAGE_INCLUDE="$(getIncludePath)"
 BUILD_DIR="$(getBuildPath $API_NAME)"
-PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
-PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
+PACKAGE_DEB="$PACKAGE_ROOT/deb/$PROJ_NAME"
+PACKAGE_RPM="$PACKAGE_ROOT/rpm/$PROJ_NAME"
 PACKAGE_PREFIX="$ROCM_INSTALL_PATH"
 BUILD_TYPE="Debug"
 MAKE_OPTS="$DASH_JAY"
@@ -44,14 +45,16 @@ SHARED_LIBS="ON"
 CLEAN_OR_OUT=0
 MAKETARGET="deb"
 PKGTYPE="deb"
-
-GPU_LIST="gfx900;gfx906;gfx908;gfx90a;gfx940;gfx941;gfx942;gfx1030;gfx1031;gfx1100;gfx1101;gfx1102"
+# Handling GPU Targets for HSACO and HIP Executables
+GPU_LIST="gfx900;gfx906;gfx908;gfx90a;gfx942;gfx1030;gfx1031;gfx1100;gfx1101;gfx1102"
 ASAN=0

+#parse the arguments
 VALID_STR=$(getopt -o hcrawso:p: --long help,clean,release,static,address_sanitizer,wheel,outdir:,package: -- "$@")
 eval set -- "$VALID_STR"

 while true; do
+    #echo "parocessing $1"
    case "$1" in
        -h | --help)
            printUsage
@@ -194,7 +197,7 @@ verifyEnvSetup

 case "$TARGET" in
    clean) clean ;;
-    build) build_rocprofiler-sdk ;;
+    build) build_rocprofiler-sdk; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
    outdir) print_output_directory ;;
    *) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_rocprofiler-systems.sh
+++ b/tools/rocm-build/build_rocprofiler-systems.sh
@@ -26,6 +26,7 @@ printUsage() {
    return 0
 }

+## Build environment variables
 API_NAME="rocprofiler-systems"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -33,11 +34,11 @@ TARGET="build"
 MAKETARGET="deb"
 PACKAGE_ROOT="$(getPackageRoot)"
 PACKAGE_LIB="$(getLibPath)"
-
+# PACKAGE_INCLUDE="$(getIncludePath)"
 BUILD_DIR="$(getBuildPath $API_NAME)"
 PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
 PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
-
+# PACKAGE_PREFIX="$ROCM_INSTALL_PATH"
 BUILD_TYPE="Debug"
 MAKE_OPTS="-j 8"
 SHARED_LIBS="ON"
@@ -46,6 +47,7 @@ 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"

@@ -121,6 +123,10 @@ clean() {

 build_rocprofiler_systems() {
    echo "Building $PROJ_NAME"
+    if [ "$DISTRO_ID" = "mariner-2.0" ] || [ "$DISTRO_ID" = "azurelinux-3.0" ]; then
+        echo "Skip make and uploading packages for \"$PROJ_NAME\" on \"${DISTRO_ID}\" distro"
+        exit 0
+    fi

    if [ $ASAN == 1 ]; then
        echo "Skip make and uploading packages for rocprofiler-systems on ASAN build"
@@ -141,9 +147,9 @@ build_rocprofiler_systems() {

    echo "Updating submodules"
    git submodule init
-
+    # copy the new URL to your local config
    git submodule sync --recursive
-
+    # force update the submodule from the new URL
    git submodule update --init --recursive --force

    echo "Updated submodule status"
@@ -187,6 +193,7 @@ build_rocprofiler_systems() {

    else
        cmake \
+            ${GEN_NINJA} \
            $(rocm_cmake_params) \
            $(rocm_common_cmake_params) \
            -DROCPROFSYS_BUILD_{LIBUNWIND,DYNINST}=ON \
@@ -228,7 +235,7 @@ verifyEnvSetup

 case "$TARGET" in
    clean) clean ;;
-    build) build_rocprofiler_systems ;;
+    build) build_rocprofiler_systems; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
    outdir) print_output_directory ;;
    *) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_rocprofiler.sh
+++ b/tools/rocm-build/build_rocprofiler.sh
@@ -12,6 +12,8 @@ printUsage() {
    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 "  -w,  --wheel              Creates python wheel package of roc-profiler.
+                                      It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
    type referred to by pkg_type"
    echo "  -h,  --help               Prints this help"
@@ -24,6 +26,7 @@ printUsage() {
    return 0
 }

+## Build environment variables
 API_NAME="rocprofiler"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -33,8 +36,8 @@ PACKAGE_ROOT="$(getPackageRoot)"
 PACKAGE_LIB="$(getLibPath)"
 PACKAGE_INCLUDE="$(getIncludePath)"
 BUILD_DIR="$(getBuildPath $API_NAME)"
-PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
-PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
+PACKAGE_DEB="$PACKAGE_ROOT/deb/$PROJ_NAME"
+PACKAGE_RPM="$PACKAGE_ROOT/rpm/$PROJ_NAME"
 PACKAGE_PREFIX="$ROCM_INSTALL_PATH"
 BUILD_TYPE="Debug"
 MAKE_OPTS="$DASH_JAY -C $BUILD_DIR"
@@ -42,12 +45,15 @@ SHARED_LIBS="ON"
 CLEAN_OR_OUT=0
 MAKETARGET="deb"
 PKGTYPE="deb"
-GPU_LIST="gfx900,gfx906,gfx908,gfx90a,gfx940,gfx941,gfx942,gfx1030,gfx1031,gfx1100,gfx1101,gfx1102,gfx1200,gfx1201"
+# Handling GPU Targets for HSACO and HIP Executables
+GPU_LIST="gfx900,gfx906,gfx908,gfx90a,gfx942,gfx1030,gfx1031,gfx1100,gfx1101,gfx1102,gfx1200,gfx1201"

+#parse the arguments
 VALID_STR=$(getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@")
 eval set -- "$VALID_STR"

 while true; do
+    #echo "parocessing $1"
    case "$1" in
    -h | --help)
        printUsage
@@ -69,6 +75,9 @@ while true; do
        ;;
    -s | --static)
        ack_and_skip_static
+        ;;
+    -w | --wheel)
+        WHEEL_PACKAGE=true
        shift
        ;;
    -o | --outdir)
@@ -85,7 +94,7 @@ while true; do
    --)
        shift
        break
-        ;;
+        ;; # end delimiter
    *)
        echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] " >&2
        exit 20
@@ -115,9 +124,6 @@ clean() {

 build_rocprofiler() {
    echo "Building $PROJ_NAME"
-
-    sed -i 's/set(CPACK_GENERATOR "DEB" "RPM" "TGZ")/set(CPACK_GENERATOR "DEB" "TGZ")/' "${ROCPROFILER_ROOT}/CMakeLists.txt"
-
    PACKAGE_CMAKE="$(getCmakePath)"
    if [ ! -d "$BUILD_DIR" ]; then
        mkdir -p "$BUILD_DIR"
@@ -173,7 +179,7 @@ verifyEnvSetup

 case "$TARGET" in
 clean) clean ;;
-build) build_rocprofiler ;;
+build) build_rocprofiler; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
 outdir) print_output_directory ;;
 *) die "Invalid target $TARGET" ;;
 esac
--- a/tools/rocm-build/build_rocr_debug_agent.sh
+++ b/tools/rocm-build/build_rocr_debug_agent.sh
@@ -11,6 +11,8 @@ printUsage() {
    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 "  -w,  --wheel              Creates python wheel package of rocr-debug-agent. 
+                                      It needs to be used along with -r option"
    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
    type referred to by pkg_type"
    echo "  -h,  --help             Prints this help"
@@ -23,6 +25,7 @@ printUsage() {
    return 0
 }

+## Build environment variables
 API_NAME=rocm-debug-agent
 PROJ_NAME=$API_NAME
 LIB_NAME=lib${API_NAME}.so
@@ -33,12 +36,12 @@ PACKAGE_BIN="$(getBinPath)"
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE=$(getIncludePath)
 BUILD_DIR=$(getBuildPath $API_NAME)
-PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
-PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
+PACKAGE_DEB=$PACKAGE_ROOT/deb/$PROJ_NAME
+PACKAGE_RPM=$PACKAGE_ROOT/rpm/$PROJ_NAME
 PACKAGE_PREFIX=$ROCM_INSTALL_PATH
 BUILD_TYPE=Debug
-MAKE_OPTS="$DASH_JAY -C"

+## Test variables
 TEST_PACKAGE_DIR="$(getBinPath)/rocm-debug-agent-test"
 PACKAGE_UTILS=$(getUtilsPath)

@@ -54,11 +57,13 @@ MAKETARGET="deb"
 PKGTYPE="deb"


-VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
 do
+    #echo "parocessing $1"
    case "$1" in
        (-h | --help)
                printUsage ; exit 0;;
@@ -71,11 +76,13 @@ do
                set_address_sanitizer_on ; shift ;;
        (-s | --static)
                ack_and_skip_static ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; shift ;;
        (-o | --outdir)
                TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
        (-p | --package)
                MAKETARGET="$2" ; shift 2;;
-        --)     shift; break;;
+        --)     shift; break;; # end delimiter
        (*)
                echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
    esac
@@ -102,14 +109,17 @@ clean() {

 build() {
    echo "Building $PROJ_NAME"
-
+    # The cmake path is different for asan and non-asan builds.
+    # Fetch after getting build type. Default will be non-asan build
    PACKAGE_CMAKE="$(getCmakePath)"
+    # If rocm is installed to an unconventional location, --rocm-path needs to be set.
    export HIPCC_COMPILE_FLAGS_APPEND="--rocm-path=$ROCM_PATH"
    if [ ! -d "$BUILD_DIR" ]; then
        mkdir -p "$BUILD_DIR"
        pushd "$BUILD_DIR"

        cmake $(rocm_cmake_params) \
+            ${GEN_NINJA} \
            -DCMAKE_PREFIX_PATH="$PACKAGE_CMAKE/amd-dbgapi" \
            -DCMAKE_MODULE_PATH="$PACKAGE_CMAKE/hip" \
            $(rocm_common_cmake_params) \
@@ -118,15 +128,22 @@ build() {

        popd
    fi
-    cmake --build "$BUILD_DIR" -- $MAKE_OPTS $BUILD_DIR
-    cmake --build "$BUILD_DIR" -- $MAKE_OPTS $BUILD_DIR install
-    cmake --build "$BUILD_DIR" -- $MAKE_OPTS $BUILD_DIR package
+    cmake --build "$BUILD_DIR" -- $DASH_JAY
+    cmake --build "$BUILD_DIR" --target install -- $DASH_JAY
+    cmake --build "$BUILD_DIR" --target package -- $DASH_JAY

    mkdir -p $PACKAGE_LIB
    cp -R $BUILD_DIR/${LIB_NAME}* $PACKAGE_LIB

    copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DEB}" "$BUILD_DIR/${API_NAME}"*.deb
    copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_RPM}" "$BUILD_DIR/${API_NAME}"*.rpm
+
+    mkdir -p "$PACKAGE_UTILS"
+    progressCopy "$SCRIPT_ROOT/run_rocr_debug_agent_test.sh" "$PACKAGE_UTILS"
+
+    ## Copy run test py script
+    echo "copying run-test.py to $PACKAGE_BIN"
+    progressCopy "$ROCR_DEBUG_AGENT_ROOT/test/run-test.py" "$PACKAGE_BIN"
 }

 print_output_directory() {
@@ -145,7 +162,7 @@ verifyEnvSetup

 case $TARGET in
    (clean) clean ;;
-    (build) build ;;
+    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
    (outdir) print_output_directory ;;
    (*) die "Invalid target $target" ;;
 esac
--- a/tools/rocm-build/build_rocrand.sh
+++ b/tools/rocm-build/build_rocrand.sh
@@ -1,7 +1,7 @@
 #!/bin/bash
 set -ex

-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"

 set_component_src rocRAND

@@ -13,31 +13,37 @@ build_rocrand() {
        SHARED_LIBS="OFF"
    fi

+    #Removed GPU ARCHS from here as it will be part of compute_utils.sh ROCMOPS-7302 & ROCMOPS-8091
    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
-         set_asan_env_vars
-         set_address_sanitizer_on
+        set_asan_env_vars
+        set_address_sanitizer_on
+        # updating GPU_ARCHS for ASAN build to supported gpu arch only SWDEV-479178
+        #GPU_ARCHS="gfx90a:xnack+;gfx942:xnack+"#This will be part of compute_utils.sh ROCMOPS-7302 & ROCMOPS-8091
    fi
-
    cd $COMPONENT_SRC && mkdir "$BUILD_DIR"

+    git config --global --add safe.directory "$COMPONENT_SRC"
+    # Rename the remote set by the repo tool to origin as
+    # git submodule update looks for the remote origin.
    remote_name=$(git remote show | head -n 1)
+    echo "remote name: $remote_name"
    [ "$remote_name" == "origin" ] || git remote rename "$remote_name" origin
+    git remote -v
    git submodule update --init --force

-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx908:xnack-;gfx90a:xnack-;gfx90a:xnack+;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101"
+    # if GPU_ENABLE_GPU_ARCHARCH is set in env by Job parameter ENABLE_GPU_ARCH, then set GFX_ARCH to that value. This will override any of the case values above
+    if [ -n "$ENABLE_GPU_ARCH" ]; then
+        #setting gfx arch as part of rocm_common_cmake_params
+        set_gpu_arch "${ENABLE_GPU_ARCH}"
    fi

    init_rocm_common_cmake_params

-    CXX=$(set_build_variables CXX)\
+    CXX=$(set_build_variables __CXX__)\
    cmake \
        ${LAUNCHER_FLAGS} \
        "${rocm_math_common_cmake_params[@]}" \
        -DBUILD_SHARED_LIBS=$SHARED_LIBS \
-        -DAMDGPU_TARGETS=${GPU_TARGETS} \
        -DBUILD_TEST=ON \
        -DBUILD_BENCHMARK=ON \
        -DBUILD_CRUSH_TEST=ON \
@@ -53,7 +59,7 @@ build_rocrand() {
    cmake --build "$BUILD_DIR" -- package

    rm -rf _CPack_Packages/  && find -name '*.o' -delete
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"

    show_build_cache_stats
 }
@@ -67,7 +73,7 @@ clean_rocrand() {
 stage2_command_args "$@"

 case $TARGET in
-    build) build_rocrand ;;
+    build) build_rocrand; build_wheel ;;
    outdir) print_output_directory ;;
    clean) clean_rocrand ;;
    *) die "Invalid target $TARGET" ;;
--- a/tools/rocm-build/build_rocrsamples.sh
+++ b/tools/rocm-build/build_rocrsamples.sh
@@ -42,7 +42,7 @@ PKGTYPE="deb"


 #parse the arguments
-VALID_STR=`getopt -o hcrao:seg: --long help,clean,release,outdir:,static,address_sanitizer,emulator,gpu_list: -- "$@"`
+VALID_STR=`getopt -o hcrawo:seg: --long help,clean,release,outdir:,static,wheel,address_sanitizer,emulator,gpu_list: -- "$@"`
 eval set -- "$VALID_STR"

 while true ;
@@ -61,6 +61,8 @@ do
                exit ;;
        (-s | --static)
                ack_and_skip_static ;;
+        (-w | --wheel)
+                echo "wheel build option accepted and ignored" ; shift ;;
        (-e | --emulator )
                EMULATOR_BUILD=1 ; ((CLEAN_OR_OUT|=3)) ; shift ;;
        (-g | --gpu_list )
@@ -92,6 +94,7 @@ build_rocrsamples() {
    cmake -DTARGET_DEVICES=$GPU_LIST \
        $(rocm_cmake_params) \
        $(rocm_common_cmake_params) \
+        ${GEN_NINJA} \
        -DROCRTST_BLD_TYPE=$ROCRTST_SAMPLES_BUILD_TYPE \
        -DROCM_DIR=$PACKAGE_ROOT \
        -DLLVM_DIR="$ROCM_INSTALL_PATH/llvm/bin" \
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
ammallya	65b8498867	Merge pull request #4513 from ammallya/main Name change for hip	2025-03-18 12:49:47 -07:00
ammallya	937dcb376b	Change naming convention for hip	2025-03-18 12:49:09 -07:00
ammallya	ac944077ba	Promote new	2025-03-18 12:48:31 -07:00
ammallya	1688d56edc	Merge pull request #4512 from ammallya/main Reverting rbt and clr	2025-03-18 11:27:49 -07:00
ammallya	042b01a71b	Reverting hip_on_rocclr	2025-03-18 11:26:53 -07:00
ammallya	c26ff9528f	Revert RBT	2025-03-18 11:25:47 -07:00
ammallya	9baa927218	Merge pull request #4511 from ammallya/main Revert rocr.sh	2025-03-18 10:37:00 -07:00
ammallya	d7136806da	Revert rocr.sh	2025-03-18 10:36:33 -07:00
ammallya	61ada2eae9	Merge pull request #4510 from ammallya/main Reverting ROCm.mk	2025-03-18 09:51:02 -07:00
ammallya	9b34984b1b	Reverting ROCm.mk	2025-03-18 09:50:41 -07:00
ammallya	20a264a715	Merge pull request #4509 from ammallya/main Promoting Build Scripts	2025-03-18 09:37:21 -07:00
Ameya Keshava Mallya	3750e88066	Promoting Build Scripts	2025-03-18 16:36:14 +00:00
ammallya	beb5f5b48d	Merge pull request #4502 from ammallya/main Updating hipblaslt and hipsparselt build scripts	2025-03-14 14:21:06 -07:00
ammallya	65330eb26d	Fixed small typo	2025-03-14 14:20:19 -07:00
ammallya	23b90e4e6b	Update hipsparselt build script	2025-03-14 14:19:46 -07:00
ammallya	080c442789	Updating hipblaslt build script	2025-03-14 14:15:56 -07:00
ammallya	5d90797591	Merge pull request #4487 from ammallya/main Updated to reduce ram usage	2025-03-12 15:11:37 -07:00
ammallya	0417c2196d	Updated to reduce ram usage	2025-03-12 15:11:05 -07:00
dependabot[bot]	1f2a94bb3a	Build(deps): Bump rocm-docs-core from 1.15.0 to 1.17.0 in /docs/sphinx (#4413 ) Bumps [rocm-docs-core](https://github.com/ROCm/rocm-docs-core) from 1.15.0 to 1.17.0. - [Release notes](https://github.com/ROCm/rocm-docs-core/releases) - [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](https://github.com/ROCm/rocm-docs-core/compare/v1.15.0...v1.17.0) --- updated-dependencies: - dependency-name: rocm-docs-core dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2025-02-24 17:09:15 -07:00
ammallya	9c5c2c9ca1	Merge pull request #4363 from ammallya/main Added main branch with fixes to run mainline	2025-02-10 15:11:07 -08:00
Ameya Keshava Mallya	190f609ea2	Added main branch with fixes to run mainline	2025-02-10 22:40:33 +00:00