Merge branch 'develop' into roc-6.3.x

Fix miopen-deps build issue by updating rocm-recipes for boost link

.azuredevops/components/ROCR-Runtime.yml

@@ -133,7 +133,6 @@ jobs:
           -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
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml

.azuredevops/components/hipBLASLt.yml

@@ -59,10 +59,14 @@ jobs:
     value: $(Build.BinariesDirectory)/rocm
   - name: TENSILE_ROCM_ASSEMBLER_PATH
     value: $(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
+  - name: CMAKE_CXX_COMPILER
+    value: $(Agent.BuildDirectory)/rocm/bin/hipcc
   - name: TENSILE_ROCM_OFFLOAD_BUNDLER_PATH
     value: $(Agent.BuildDirectory)/rocm/llvm/bin/clang-offload-bundler
-  - name: ROCM_PATH
-    value: $(Agent.BuildDirectory)/rocm
+  - name: TENSILE_ROCM_PATH
+    value: $(Agent.BuildDirectory)/rocm/bin/hipcc
+  - name: PATH
+    value: $(Agent.BuildDirectory)/rocm/llvm/bin:$(Agent.BuildDirectory)/rocm/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin
   - name: DAY_STRING
     value: $[format('{0:ddMMyyyy}', pipeline.startTime)]
   pool: ${{ variables.ULTRA_BUILD_POOL }}
@@ -150,8 +154,9 @@ jobs:
       extraEnvVars:
         - HIP_ROCCLR_HOME:::/home/user/workspace/rocm
         - TENSILE_ROCM_ASSEMBLER_PATH:::/home/user/workspace/rocm/llvm/bin/amdclang
+        - CMAKE_CXX_COMPILER:::/home/user/workspace/rocm/bin/hipcc
         - TENSILE_ROCM_OFFLOAD_BUNDLER_PATH:::/home/user/workspace/rocm/llvm/bin/clang-offload-bundler
-        - ROCM_PATH:::/home/user/workspace/rocm
+        - TENSILE_ROCM_PATH:::/home/user/workspace/rocm/bin/hipcc
       extraCopyDirectories:
         - deps
 

.azuredevops/components/rocBLAS.yml

@@ -64,10 +64,10 @@ jobs:
     value: $(Build.BinariesDirectory)/rocm
   - name: TENSILE_ROCM_ASSEMBLER_PATH
     value: $(Agent.BuildDirectory)/rocm/llvm/bin/clang
+  - name: CMAKE_CXX_COMPILER
+    value: $(Agent.BuildDirectory)/rocm/bin/hipcc
   - name: TENSILE_ROCM_OFFLOAD_BUNDLER_PATH
     value: $(Agent.BuildDirectory)/rocm/llvm/bin/clang-offload-bundler
-  - name: ROCM_PATH
-    value: $(Agent.BuildDirectory)/rocm
   pool: ${{ variables.MEDIUM_BUILD_POOL }}
   workspace:
     clean: all
@@ -96,8 +96,8 @@ jobs:
         -DCMAKE_TOOLCHAIN_FILE=toolchain-linux.cmake
         -DCMAKE_PREFIX_PATH=$(Agent.BuildDirectory)/rocm/llvm;$(Agent.BuildDirectory)/rocm
         -DCMAKE_BUILD_TYPE=Release
-        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/bin/amdclang++
-        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/bin/amdclang
+        -DCMAKE_CXX_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
+        -DCMAKE_C_COMPILER=$(Agent.BuildDirectory)/rocm/bin/hipcc
         -DGPU_TARGETS=$(JOB_GPU_TARGET)
         -DTensile_CODE_OBJECT_VERSION=default
         -DTensile_LOGIC=asm_full
@@ -125,8 +125,8 @@ jobs:
       extraEnvVars:
         - HIP_ROCCLR_HOME:::/home/user/workspace/rocm
         - TENSILE_ROCM_ASSEMBLER_PATH:::/home/user/workspace/rocm/llvm/bin/clang
+        - CMAKE_CXX_COMPILER:::/home/user/workspace/rocm/bin/hipcc
         - TENSILE_ROCM_OFFLOAD_BUNDLER_PATH:::/home/user/workspace/rocm/llvm/bin/clang-offload-bundler
-        - ROCM_PATH:::/home/user/workspace/rocm
 
 - job: rocBLAS_testing
   dependsOn: rocBLAS

.azuredevops/components/rocDecode.yml

@@ -49,10 +49,21 @@ jobs:
   workspace:
     clean: all
   steps:
+  # Since mesa-amdgpu-multimedia-devel is not directly available from apt, register it
+  - task: Bash@3
+    displayName: 'Register ROCm packages'
+    inputs:
+      targetType: inline
+      script: |
+        sudo mkdir --parents --mode=0755 /etc/apt/keyrings
+        wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/${{ variables.KEYRING_VERSION }}/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/${{ variables.KEYRING_VERSION }} jammy main" | sudo tee --append /etc/apt/sources.list.d/rocm.list
+        echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
+        sudo apt update
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
-      registerROCmPackages: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
@@ -93,10 +104,21 @@ jobs:
         JOB_GPU_TARGET: gfx942
         JOB_TEST_POOL: ${{ variables.GFX942_TEST_POOL }}
   steps:
+  # Since mesa-amdgpu-multimedia-devel is not directly available from apt, register it
+  - task: Bash@3
+    displayName: 'Register ROCm packages'
+    inputs:
+      targetType: inline
+      script: |
+        sudo mkdir --parents --mode=0755 /etc/apt/keyrings
+        wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/${{ variables.KEYRING_VERSION }}/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/${{ variables.KEYRING_VERSION }} jammy main" | sudo tee --append /etc/apt/sources.list.d/rocm.list
+        echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
+        sudo apt update
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
-      registerROCmPackages: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-aqlprofile.yml

.azuredevops/components/rocJPEG.yml

@@ -48,10 +48,21 @@ jobs:
       gfx942:
         JOB_GPU_TARGET: gfx942
   steps:
+  # Since mesa-amdgpu-multimedia-devel is not directly available from apt, register it
+  - task: Bash@3
+    displayName: 'Register ROCm packages'
+    inputs:
+      targetType: inline
+      script: |
+        sudo mkdir --parents --mode=0755 /etc/apt/keyrings
+        wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/${{ variables.KEYRING_VERSION }}/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/${{ variables.KEYRING_VERSION }} jammy main" | sudo tee --append /etc/apt/sources.list.d/rocm.list
+        echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
+        sudo apt update
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
-      registerROCmPackages: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
@@ -95,10 +106,21 @@ jobs:
         JOB_GPU_TARGET: gfx942
         JOB_TEST_POOL: ${{ variables.GFX942_TEST_POOL }}
   steps:
+  # Since mesa-amdgpu-multimedia-devel is not directly available from apt, register it
+  - task: Bash@3
+    displayName: 'Register ROCm packages'
+    inputs:
+      targetType: inline
+      script: |
+        sudo mkdir --parents --mode=0755 /etc/apt/keyrings
+        wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/${{ variables.KEYRING_VERSION }}/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/${{ variables.KEYRING_VERSION }} jammy main" | sudo tee --append /etc/apt/sources.list.d/rocm.list
+        echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
+        sudo apt update
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
-      registerROCmPackages: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-aqlprofile.yml

.azuredevops/components/rocWMMA.yml

@@ -80,7 +80,6 @@ jobs:
         -DROCWMMA_BUILD_SAMPLES=OFF
         -DGPU_TARGETS=$(JOB_GPU_TARGET)
         -DCMAKE_BUILD_WITH_INSTALL_RPATH=ON
-        -DROCM_PLATFORM_VERSION=$(NEXT_RELEASE_VERSION)
         -GNinja
 # gfx1030 not supported in documentation
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml

.azuredevops/components/rocm-core.yml

@@ -36,7 +36,7 @@ jobs:
         -DCPACK_GENERATOR=DEB
         -DCPACK_DEBIAN_PACKAGE_RELEASE="local.9999~99.99"
         -DCPACK_RPM_PACKAGE_RELEASE="local.9999"
-        -DROCM_VERSION="$(NEXT_RELEASE_VERSION)"
+        -DROCM_VERSION="$(next-release)"
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-links.yml

.azuredevops/components/rocminfo.yml

@@ -35,7 +35,7 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-other.yml
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
-      registerROCmPackages: true
+      registerRadeon: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:

.azuredevops/components/rocprofiler-sdk.yml

@@ -37,8 +37,6 @@ parameters:
     - clr
     - llvm-project
     - rccl
-    - rocDecode
-    - rocJPEG
     - rocm-cmake
     - rocm-core
     - rocminfo
@@ -62,7 +60,7 @@ jobs:
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
-      registerROCmPackages: true
+      registerRadeon: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
@@ -101,7 +99,6 @@ jobs:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
       gpuTarget: $(JOB_GPU_TARGET)
-      registerROCmPackages: true
 
 - job: rocprofiler_sdk_testing
   dependsOn: rocprofiler_sdk
@@ -122,7 +119,7 @@ jobs:
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
-      registerROCmPackages: true
+      registerRadeon: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
@@ -158,4 +155,3 @@ jobs:
       pipModules: ${{ parameters.pipModules }}
       environment: test
       gpuTarget: $(JOB_GPU_TARGET)
-      registerROCmPackages: true

.azuredevops/components/rocprofiler-systems.yml

@@ -46,19 +46,17 @@ parameters:
 - name: rocmDependencies
   type: object
   default:
-    - amdsmi
     - aomp
     - clr
     - llvm-project
     - rccl
     - rocDecode
-    - rocJPEG
     - rocm-core
-    - rocminfo
     - rocm_smi_lib
+    - rocminfo
+    - ROCR-Runtime
     - rocprofiler-register
     - rocprofiler-sdk
-    - ROCR-Runtime
 
 jobs:
 - job: rocprofiler_systems
@@ -77,7 +75,7 @@ jobs:
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
-      registerROCmPackages: true
+      registerRadeon: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:
@@ -161,7 +159,7 @@ jobs:
     parameters:
       aptPackages: ${{ parameters.aptPackages }}
       pipModules: ${{ parameters.pipModules }}
-      registerROCmPackages: true
+      registerRadeon: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/checkout.yml
     parameters:

.azuredevops/dependencies/half560.yml

@@ -19,7 +19,7 @@ jobs:
   pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   container:
-    image: rocm/dev-ubuntu-22.04:${{ variables.LATEST_RELEASE_VERSION }}
+    image: ${{ variables.DOCKER_IMAGE_NAME }}:${{ variables.LATEST_DOCKER_VERSION }}
   workspace:
     clean: all
   steps:

.azuredevops/tag-builds/TransferBench.yml

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

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

@@ -4,13 +4,13 @@ steps:
   inputs:
     targetType: inline
     script: |
-      export packageName=$(curl -s https://repo.radeon.com/rocm/apt/$(REPO_RADEON_VERSION)/pool/main/h/hsa-amd-aqlprofile/ | grep -oP "href=\"\K[^\"]*$(lsb_release -rs)[^\"]*\.deb")
+      export packageName=$(curl -s https://repo.radeon.com/rocm/apt/latest/pool/main/h/hsa-amd-aqlprofile/ | grep -oP "href=\"\K[^\"]*$(lsb_release -rs)[^\"]*\.deb")
       echo "##vso[task.setvariable variable=packageName;isreadonly=true]$packageName"
 - task: Bash@3
   displayName: 'Download aqlprofile'
   inputs:
     targetType: inline
-    script: wget -nv https://repo.radeon.com/rocm/apt/$(REPO_RADEON_VERSION)/pool/main/h/hsa-amd-aqlprofile/$(packageName)
+    script: wget -nv https://repo.radeon.com/rocm/apt/latest/pool/main/h/hsa-amd-aqlprofile/$(packageName)
     workingDirectory: '$(Pipeline.Workspace)'
 - task: Bash@3
   displayName: 'Extract aqlprofile'

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

@@ -6,21 +6,21 @@ parameters:
 - name: pipModules
   type: object
   default: []
-- name: registerROCmPackages
+- name: registerRadeon
   type: boolean
   default: false
 
 steps:
-- ${{ if eq(parameters.registerROCmPackages, true) }}:
+- ${{ if eq(parameters.registerRadeon, true) }}:
   - task: Bash@3
-    displayName: 'Register AMDGPU & ROCm repos'
+    displayName: 'Register repo.radeon packages'
     inputs:
       targetType: inline
       script: |
         sudo mkdir --parents --mode=0755 /etc/apt/keyrings
         wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null
-        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/$(REPO_RADEON_VERSION)/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
-        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/$(REPO_RADEON_VERSION) jammy main" | sudo tee --append /etc/apt/sources.list.d/rocm.list
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/latest/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
+        echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/latest jammy main" | sudo tee --append /etc/apt/sources.list.d/rocm.list
         echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
         sudo apt update
 # firefox takes time to upgrade and is not needed for CI workloads, hold version

.azuredevops/templates/steps/docker-container.yml

@@ -154,8 +154,8 @@ steps:
         script: |
           echo "RUN mkdir --parents --mode=0755 /etc/apt/keyrings" >> Dockerfile
           echo "RUN wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | gpg --dearmor | tee /etc/apt/keyrings/rocm.gpg > /dev/null" >> Dockerfile
-          echo "RUN echo \"deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/$(REPO_RADEON_VERSION)/ubuntu jammy main\" | tee /etc/apt/sources.list.d/amdgpu.list" >> Dockerfile
-          echo "RUN echo \"deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/$(REPO_RADEON_VERSION) jammy main\" | tee --append /etc/apt/sources.list.d/rocm.list" >> Dockerfile
+          echo "RUN echo \"deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/latest/ubuntu jammy main\" | tee /etc/apt/sources.list.d/amdgpu.list" >> Dockerfile
+          echo "RUN echo \"deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/latest jammy main\" | tee --append /etc/apt/sources.list.d/rocm.list" >> Dockerfile
           echo "RUN printf 'Package: *\\nPin: release o=repo.radeon.com\\nPin-Priority: 600' > /etc/apt/preferences.d/rocm-pin-600" >> Dockerfile
           echo "RUN DEBIAN_FRONTEND=noninteractive apt-get --yes update" >> Dockerfile
   - ${{ if eq(parameters.registerCUDAPackages, true) }}:

.azuredevops/variables-global.yml

@@ -27,14 +27,14 @@ variables:
   value: rocm-ci_larger_base_disk_pool
 - name: GFX942_TEST_POOL
   value: gfx942_test_pool
-- name: LATEST_RELEASE_VERSION
-  value: 6.3.2
-- name: REPO_RADEON_VERSION
-  value: 6.3.2
-- name: NEXT_RELEASE_VERSION
-  value: 6.4.0
 - name: LATEST_RELEASE_TAG
-  value: rocm-6.3.2
+  value: rocm-6.1.0
+- name: DOCKER_IMAGE_NAME
+  value: rocm/dev-ubuntu-22.04
+- name: LATEST_DOCKER_VERSION
+  value: 6.1
+- name: KEYRING_VERSION
+  value: 6.3
 - name: AMDMIGRAPHX_GFX942_TEST_PIPELINE_ID
   value: 197
 - name: AMDMIGRAPHX_PIPELINE_ID
@@ -151,6 +151,8 @@ variables:
   value: 105
 - name: HIPTENSOR_TAGGED_PIPELINE_ID
   value: 56
+- name: LAST_RELEASE
+  value: 6.1.0
 - name: LLVM_PROJECT_PIPELINE_ID
   value: 2
 - name: LLVM_PROJECT_TAGGED_PIPELINE_ID
@@ -181,6 +183,10 @@ variables:
   value: 100
 - name: RDC_TAGGED_PIPELINE_ID
   value: 59
+- name: REIMAGE_ORG
+  value: AGS-ROCm-CI
+- name: REIMAGE_REPO
+  value: cirrascale-reimage-automation
 - name: ROCAL_PIPELINE_ID
   value: 151
 - name: ROCALUTION_GFX942_TEST_PIPELINE_ID

.github/workflows/issue_retrieval.yml

@@ -2,7 +2,7 @@ name: Issue retrieval
 
 on:
   issues:
-    types: [opened, edited]
+    types: [opened]
 
 jobs:
   auto-retrieve:
@@ -15,7 +15,7 @@ jobs:
           app_id: ${{ secrets.ACTION_APP_ID }}
           private_key: ${{ secrets.ACTION_PEM }}
       - name: 'Retrieve Issue'
-        uses: harkgill-amd/rocm_issue_management@main
+        uses: abhimeda/rocm_issue_management@main
         with:
           authentication-token: ${{ steps.generate_token.outputs.token }}
           github-organization: 'ROCm'

.wordlist.txt

@@ -74,7 +74,6 @@ Conda
 ConnectX
 CuPy
 Dashboarding
-DBRX
 DDR
 DF
 DGEMM
@@ -93,7 +92,6 @@ DataFrame
 DataLoader
 DataParallel
 Debian
-DeepSeek
 DeepSpeed
 Dependabot
 Deprecations
@@ -131,7 +129,6 @@ GDS
 GEMM
 GEMMs
 GFortran
-Gemma
 GiB
 GIM
 GL
@@ -154,6 +151,7 @@ HCA
 HGX
 HIPCC
 HIPExtension
+HIPification
 HIPIFY
 HIPification
 HIPify

RELEASE.md

@@ -29,7 +29,8 @@ The release notes provide a summary of notable changes since the previous ROCm r
 - [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)
+If you’re using Radeon™ PRO or Radeon GPUs in a workstation setting with a
+display connected, continue to use ROCm 6.2.3. See the [Use ROCm on Radeon GPUs](https://rocm.docs.amd.com/projects/radeon/en/latest/index.html)
 documentation to verify compatibility and system requirements.
 ```
 ## Release highlights

default.xml

@@ -1,78 +1,76 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <manifest>
     <remote name="rocm-org" fetch="https://github.com/ROCm/" />
-    <default revision="refs/heads/amd-mainline"
+    <default revision="refs/tags/rocm-6.3.2"
      remote="rocm-org"
      sync-c="true"
      sync-j="4" />
 <!--list of projects for ROCm-->
-    <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" /> 
+    <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" />
 <!--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" revision="refs/heads/rocm" />
+    <project name="half" />
     <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" revision="refs/heads/amd-mainline-rocgdb-15"/> 
+    <project name="ROCdbgapi" />
+    <project name="ROCgdb" />
     <project name="rocr_debug_agent" />
 <!-- ROCm Libraries -->
-    <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" /> 
+    <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" />
 <!-- 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>

docs/about/license.md

@@ -62,7 +62,7 @@ additional licenses. Please review individual repositories for more information.
 | [rocJPEG](https://github.com/ROCm/rocJPEG/) | [MIT](https://github.com/ROCm/rocJPEG/blob/develop/LICENSE) |
 | [ROCK-Kernel-Driver](https://github.com/ROCm/ROCK-Kernel-Driver/) | [GPL 2.0 WITH Linux-syscall-note](https://github.com/ROCm/ROCK-Kernel-Driver/blob/master/COPYING) |
 | [rocminfo](https://github.com/ROCm/rocminfo/) | [The University of Illinois/NCSA](https://github.com/ROCm/rocminfo/blob/amd-staging/License.txt) |
-| [ROCm Bandwidth Test](https://github.com/ROCm/rocm_bandwidth_test/) | [MIT](https://github.com/ROCm/rocm_bandwidth_test/blob/master/LICENSE.txt) |
+| [ROCm Bandwidth Test](https://github.com/ROCm/rocm_bandwidth_test/) | [The University of Illinois/NCSA](https://github.com/ROCm/rocm_bandwidth_test/blob/master/LICENSE.txt) |
 | [ROCm CMake](https://github.com/ROCm/rocm-cmake/) | [MIT](https://github.com/ROCm/rocm-cmake/blob/develop/LICENSE) |
 | [ROCm Communication Collectives Library (RCCL)](https://github.com/ROCm/rccl/) | [Custom](https://github.com/ROCm/rccl/blob/develop/LICENSE.txt) |
 | [ROCm-Core](https://github.com/ROCm/rocm-core) | [MIT](https://github.com/ROCm/rocm-core/blob/master/copyright) |

docs/compatibility/compatibility-matrix-historical-6.0.csv

@@ -7,7 +7,7 @@ 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
       ,"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]_,,,,,,,,,,
+,Debian 12 [#mi300x-past-60]_,Debian 12 [#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

docs/compatibility/compatibility-matrix.rst

@@ -32,7 +32,7 @@ compatibility and system requirements.
       ,RHEL 8.10,RHEL 8.10,"RHEL 8.10, 8.9"
       ,"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]_,
+      ,Debian 12 [#mi300x]_,Debian 12 [#mi300x]_,
       ,Azure Linux 3.0 [#mi300x]_,,
       ,.. _architecture-support-compatibility-matrix:,,
       :doc:`Architecture <rocm-install-on-linux:reference/system-requirements>`,CDNA3,CDNA3,CDNA3
@@ -148,8 +148,7 @@ compatibility and system requirements.
 
 .. rubric:: Footnotes
 
-.. [#mi300x] Oracle Linux and Azure Linux are supported only on AMD Instinct MI300X.
-.. [#single-node] Debian 12 is supported only on AMD Instinct MI300X for single-node functionality. 
+.. [#mi300x] Oracle Linux, Debian, and Azure Linux are supported only on AMD Instinct MI300X.
 .. [#mi300_620] **For ROCm 6.2.0** - MI300X (gfx942) is supported on listed operating systems *except* Ubuntu 22.04.5 [6.8 HWE] and Ubuntu 22.04.4 [6.5 HWE].
 .. [#kfd_support] ROCm provides forward and backward compatibility between the AMD Kernel-mode GPU Driver (KMD) and its user space software for +/- 2 releases. These are the compatibility combinations that are currently supported.
 .. [#ROCT-rocr] Starting from ROCm 6.3.0, the ROCT Thunk Interface is included as part of the ROCr runtime package.
@@ -216,8 +215,7 @@ Expand for full historical view of:
 
    .. rubric:: Footnotes
 
-   .. [#mi300x-past-60] Oracle Linux and Azure Linux are supported only on AMD Instinct MI300X.
-   .. [#single-node-past-60] Debian 12 is supported only on AMD Instinct MI300X for single-node functionality. 
+   .. [#mi300x-past-60] Oracle Linux, Debian, and Azure Linux are supported only on AMD Instinct MI300X.
    .. [#mi300_624-past-60] **For ROCm 6.2.4** - MI300X (gfx942) is supported on listed operating systems *except* Ubuntu 22.04.5 [6.8 HWE] and Ubuntu 22.04.4 [6.5 HWE].
    .. [#mi300_622-past-60] **For ROCm 6.2.2** - MI300X (gfx942) is supported on listed operating systems *except* Ubuntu 22.04.5 [6.8 HWE] and Ubuntu 22.04.4 [6.5 HWE].
    .. [#mi300_621-past-60] **For ROCm 6.2.1** - MI300X (gfx942) is supported on listed operating systems *except* Ubuntu 22.04.5 [6.8 HWE] and Ubuntu 22.04.4 [6.5 HWE].

docs/compatibility/pytorch-compatibility.rst

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

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

@@ -16,9 +16,6 @@ Throughout the following topics, this guide discusses the goals and :ref:`challe
 model <fine-tuning-llms-concept-challenge>` like Llama 2. In the
 sections that follow, you'll find practical guides on libraries and tools to accelerate your fine-tuning.
 
-The AI Developer Hub contains `AMD ROCm tutorials <https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/>`_ for
-training, fine-tuning, and inference. It leverages popular machine learning frameworks on AMD GPUs.
-
 - :doc:`Conceptual overview of fine-tuning LLMs <overview>`
 
 - :doc:`Fine-tuning and inference <fine-tuning-and-inference>` using a

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

@@ -12,9 +12,6 @@ You can use ROCm to perform distributed training, which enables you to train mod
  
 Overall, ROCm can be used to improve the performance and efficiency of your AI applications. With its training, fine-tuning, and inference support, ROCm provides a complete solution for optimizing AI workflows and achieving the optimum results possible on AMD GPUs. 
 
-The AI Developer Hub contains `AMD ROCm tutorials <https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/>`_ for
-training, fine-tuning, and inference. It leverages popular machine learning frameworks on AMD GPUs.
-
 In this guide, you'll learn how to use ROCm for AI:
 
 - :doc:`Training <training/index>`

docs/how-to/rocm-for-ai/inference-optimization/model-acceleration-libraries.rst

@@ -277,7 +277,7 @@ Installing FBGEMM_GPU
 Installing FBGEMM_GPU consists of the following steps:
 
 *  Set up an isolated Miniconda environment
-*  Install ROCm using Docker or the :doc:`package manager <rocm-install-on-linux:install/install-methods/package-manager-index>`
+*  Install ROCm using Docker or the :doc:`package manager <rocm-install-on-linux:install/native-install/index>`
 *  Install the nightly `PyTorch <https://pytorch.org/>`_ build
 *  Complete the pre-build and build tasks
   

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

@@ -11,9 +11,6 @@ Understanding the ROCm™ software platform’s architecture and capabilities is
 
 Throughout the following topics, this section provides a comprehensive guide to setting up and deploying AI inference on AMD GPUs. This includes instructions on how to install ROCm, how to use Hugging Face Transformers to manage pre-trained models for natural language processing (NLP) tasks, how to validate vLLM on AMD Instinct™ MI300X accelerators and illustrate how to deploy trained models in production environments. 
 
-The AI Developer Hub contains `AMD ROCm tutorials <https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/>`_ for
-training, fine-tuning, and inference. It leverages popular machine learning frameworks on AMD GPUs.
-
 - :doc:`Installing ROCm and machine learning frameworks <install>`
 
 - :doc:`Running models from Hugging Face <hugging-face-models>`

docs/how-to/rocm-for-ai/inference/install.rst

@@ -26,7 +26,7 @@ If you’re using a Radeon GPU for graphics-accelerated applications, refer to t
 
 ROCm supports multiple :doc:`installation methods <rocm-install-on-linux:install/install-overview>`:
 
-* :doc:`Using your Linux distribution's package manager <rocm-install-on-linux:install/install-methods/package-manager-index>`
+* :doc:`Using your Linux distribution's package manager <rocm-install-on-linux:install/native-install/index>`
 
 * :doc:`Using the AMDGPU installer <rocm-install-on-linux:install/amdgpu-install>`
 

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

@@ -10,22 +10,49 @@ LLM inference performance validation on AMD Instinct MI300X
 .. _vllm-benchmark-unified-docker:
 
 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:
+a prebuilt, optimized environment designed 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:
 
-* `ROCm 6.3.1 <https://github.com/ROCm/ROCm>`_
+* `ROCm 6.2.1 <https://github.com/ROCm/ROCm>`_
 
-* `vLLM 0.6.6 <https://docs.vllm.ai/en/latest>`_
+* `vLLM 0.6.4 <https://docs.vllm.ai/en/latest>`_
 
-* `PyTorch 2.7.0 (2.7.0a0+git3a58512) <https://github.com/pytorch/pytorch>`_
+* `PyTorch 2.5.0 <https://github.com/pytorch/pytorch>`_
+
+* Tuning files (in CSV format)
 
 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>`.
+performance numbers on the MI300X accelerator. This topic also provides tips on
+optimizing performance with popular AI models.
+
+.. hlist::
+   :columns: 6
+
+   * Llama 3.1 8B
+
+   * Llama 3.1 70B
+
+   * Llama 3.1 405B
+
+   * Llama 2 7B
+
+   * Llama 2 70B
+
+   * Mixtral 8x7B
+
+   * Mixtral 8x22B
+
+   * Mixtral 7B
+
+   * Qwen2 7B
+
+   * Qwen2 72B
+
+   * JAIS 13B
+
+   * JAIS 30B
 
 .. _vllm-benchmark-vllm:
 
@@ -64,9 +91,9 @@ MI300X accelerator with the prebuilt vLLM Docker image.
 
    .. code-block:: shell
 
-      docker pull rocm/vllm:rocm6.3.1_mi300_ubuntu22.04_py3.12_vllm_0.6.6
+      docker pull rocm/vllm:rocm6.2_mi300_ubuntu20.04_py3.9_vllm_0.6.4
 
-Once the setup is complete, choose between two options to reproduce the
+Once setup is complete, you can choose between two options to reproduce the
 benchmark results:
 
 -  :ref:`MAD-integrated benchmarking <vllm-benchmark-mad>`
@@ -103,89 +130,45 @@ 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.
 
-.. _vllm-benchmark-mad-models:
-
 Available models
 ----------------
 
-.. list-table::
-   :header-rows: 1
-   :widths: 2, 3
+.. hlist::
+   :columns: 3
 
-   * - Model name
-     - Tag
+   * ``pyt_vllm_llama-3.1-8b``
 
-   * - `Llama 3.1 8B <https://huggingface.co/meta-llama/Llama-3.1-8B>`_
-     - ``pyt_vllm_llama-3.1-8b``
+   * ``pyt_vllm_llama-3.1-70b``
 
-   * - `Llama 3.1 70B <https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct>`_
-     - ``pyt_vllm_llama-3.1-70b``
+   * ``pyt_vllm_llama-3.1-405b``
 
-   * - `Llama 3.1 405B <https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct>`_
-     - ``pyt_vllm_llama-3.1-405b``
+   * ``pyt_vllm_llama-2-7b``
 
-   * - `Llama 3.2 11B Vision <https://huggingface.co/meta-llama/Llama-3.2-11B-Vision-Instruct>`_
-     - ``pyt_vllm_llama-3.2-11b-vision-instruct``
+   * ``pyt_vllm_llama-2-70b``
 
-   * - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-7b-chat-hf>`_
-     - ``pyt_vllm_llama-2-7b``
+   * ``pyt_vllm_mixtral-8x7b``
 
-   * - `Llama 2 70B <https://huggingface.co/meta-llama/Llama-2-70b-chat-hf>`_
-     - ``pyt_vllm_llama-2-70b``
+   * ``pyt_vllm_mixtral-8x22b``
 
-   * - `Mixtral MoE 8x7B <https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1>`_
-     - ``pyt_vllm_mixtral-8x7b``
+   * ``pyt_vllm_mistral-7b``
 
-   * - `Mixtral MoE 8x22B <https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1>`_
-     - ``pyt_vllm_mixtral-8x22b``
+   * ``pyt_vllm_qwen2-7b``
 
-   * - `Mistral 7B <https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3>`_
-     - ``pyt_vllm_mistral-7b``
+   * ``pyt_vllm_qwen2-72b``
 
-   * - `Qwen2 7B <https://huggingface.co/Qwen/Qwen2-7B-Instruct>`_
-     - ``pyt_vllm_qwen2-7b``
+   * ``pyt_vllm_jais-13b``
 
-   * - `Qwen2 72B <https://huggingface.co/Qwen/Qwen2-72B-Instruct>`_
-     - ``pyt_vllm_qwen2-72b``
+   * ``pyt_vllm_jais-30b``
 
-   * - `JAIS 13B <https://huggingface.co/core42/jais-13b-chat>`_
-     - ``pyt_vllm_jais-13b``
+   * ``pyt_vllm_llama-3.1-8b_fp8``
 
-   * - `JAIS 30B <https://huggingface.co/core42/jais-30b-chat-v3>`_
-     - ``pyt_vllm_jais-30b``
+   * ``pyt_vllm_llama-3.1-70b_fp8``
 
-   * - `DBRX Instruct <https://huggingface.co/databricks/dbrx-instruct>`_
-     - ``pyt_vllm_dbrx-instruct``
+   * ``pyt_vllm_llama-3.1-405b_fp8``
 
-   * - `Gemma 2 27B <https://huggingface.co/google/gemma-2-27b>`_
-     - ``pyt_vllm_gemma-2-27b``
+   * ``pyt_vllm_mixtral-8x7b_fp8``
 
-   * - `C4AI Command R+ 08-2024 <https://huggingface.co/CohereForAI/c4ai-command-r-plus-08-2024>`_
-     - ``pyt_vllm_c4ai-command-r-plus-08-2024``
-
-   * - `DeepSeek MoE 16B <https://huggingface.co/deepseek-ai/deepseek-moe-16b-chat>`_
-     - ``pyt_vllm_deepseek-moe-16b-chat``
-
-   * - `Llama 3.1 70B FP8 <https://huggingface.co/amd/Llama-3.1-70B-Instruct-FP8-KV>`_
-     - ``pyt_vllm_llama-3.1-70b_fp8``
-
-   * - `Llama 3.1 405B FP8 <https://huggingface.co/amd/Llama-3.1-405B-Instruct-FP8-KV>`_
-     - ``pyt_vllm_llama-3.1-405b_fp8``
-
-   * - `Mixtral MoE 8x7B FP8 <https://huggingface.co/amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV>`_
-     - ``pyt_vllm_mixtral-8x7b_fp8``
-
-   * - `Mixtral MoE 8x22B FP8 <https://huggingface.co/amd/Mixtral-8x22B-Instruct-v0.1-FP8-KV>`_
-     - ``pyt_vllm_mixtral-8x22b_fp8``
-
-   * - `Mistral 7B FP8 <https://huggingface.co/amd/Mistral-7B-v0.1-FP8-KV>`_
-     - ``pyt_vllm_mistral-7b_fp8``
-
-   * - `DBRX Instruct FP8 <https://huggingface.co/amd/dbrx-instruct-FP8-KV>`_
-     - ``pyt_vllm_dbrx_fp8``
-
-   * - `C4AI Command R+ 08-2024 FP8 <https://huggingface.co/amd/c4ai-command-r-plus-FP8-KV>`_
-     - ``pyt_vllm_command-r-plus_fp8``
+   * ``pyt_vllm_mixtral-8x22b_fp8``
 
 .. _vllm-benchmark-standalone:
 
@@ -198,8 +181,8 @@ snippet.
 
 .. code-block::
 
-   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
+   docker pull rocm/vllm:rocm6.2_mi300_ubuntu20.04_py3.9_vllm_0.6.4
+   docker run -it --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 128G --security-opt seccomp=unconfined --security-opt apparmor=unconfined --cap-add=SYS_PTRACE -v $(pwd):/workspace --env HUGGINGFACE_HUB_CACHE=/workspace --name vllm_v0.6.4 rocm/vllm:rocm6.2_mi300_ubuntu20.04_py3.9_vllm_0.6.4
 
 In the Docker container, clone the ROCm MAD repository and navigate to the
 benchmark scripts directory at ``~/MAD/scripts/vllm``.
@@ -241,8 +224,8 @@ See the :ref:`examples <vllm-benchmark-run-benchmark>` for more information.
 
 .. _vllm-benchmark-standalone-options:
 
-Options and available models
-----------------------------
+Options
+-------
 
 .. list-table::
    :header-rows: 1
@@ -265,100 +248,72 @@ Options and available models
      - Measure both throughput and latency
 
    * - ``$model_repo``
-     - ``meta-llama/Llama-3.1-8B-Instruct``
-     - `Llama 3.1 8B <https://huggingface.co/meta-llama/Llama-3.1-8B>`_
+     - ``meta-llama/Meta-Llama-3.1-8B-Instruct``
+     - 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/Meta-Llama-3.1-70B-Instruct``
+     - Llama 3.1 70B
 
    * -
-     - ``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/Meta-Llama-3.1-405B-Instruct``
+     - Llama 3.1 405B
 
    * -
      - ``meta-llama/Llama-2-7b-chat-hf``
-     - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-7b-chat-hf>`_
+     - Llama 2 7B
 
    * -
      - ``meta-llama/Llama-2-70b-chat-hf``
-     - `Llama 2 7B <https://huggingface.co/meta-llama/Llama-2-70b-chat-hf>`_
+     - Llama 2 70B
 
    * -
      - ``mistralai/Mixtral-8x7B-Instruct-v0.1``
-     - `Mixtral MoE 8x7B <https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1>`_
+     - Mixtral 8x7B
 
    * -
      - ``mistralai/Mixtral-8x22B-Instruct-v0.1``
-     - `Mixtral MoE 8x22B <https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1>`_
+     - Mixtral 8x22B
 
    * -
      - ``mistralai/Mistral-7B-Instruct-v0.3``
-     - `Mistral 7B <https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3>`_
+     - Mixtral 7B
 
    * -
      - ``Qwen/Qwen2-7B-Instruct``
-     - `Qwen2 7B <https://huggingface.co/Qwen/Qwen2-7B-Instruct>`_
+     - Qwen2 7B
 
    * -
      - ``Qwen/Qwen2-72B-Instruct``
-     - `Qwen2 72B <https://huggingface.co/Qwen/Qwen2-72B-Instruct>`_
+     - Qwen2 72B
 
    * -
      - ``core42/jais-13b-chat``
-     - `JAIS 13B <https://huggingface.co/core42/jais-13b-chat>`_
+     - JAIS 13B
 
    * -
      - ``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>`_
+     - JAIS 30B
 
    * - ``$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>`_
+     - ``amd/Meta-Llama-3.1-8B-Instruct-FP8-KV``
+     - Llama 3.1 8B
 
    * - (``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/Meta-Llama-3.1-70B-Instruct-FP8-KV``
+     - Llama 3.1 70B
+
+   * -
+     - ``amd/Meta-Llama-3.1-405B-Instruct-FP8-KV``
+     - Llama 3.1 405B
 
    * -
      - ``amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV``
-     - `Mixtral MoE 8x7B FP8 <https://huggingface.co/amd/Mixtral-8x7B-Instruct-v0.1-FP8-KV>`_
+     - Mixtral 8x7B
 
    * -
      - ``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>`_
+     - Mixtral 8x22B
 
    * - ``$num_gpu``
      - 1 or 8
@@ -380,34 +335,34 @@ 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.
+Use this command to benchmark the latency of the Llama 3.1 8B model on one GPU 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
+   ./vllm_benchmark_report.sh -s latency -m meta-llama/Meta-Llama-3.1-8B-Instruct -g 1 -d float16
+   ./vllm_benchmark_report.sh -s latency -m amd/Meta-Llama-3.1-8B-Instruct-FP8-KV -g 1 -d float8
 
 Find the latency reports at:
 
-- ``./reports_float16/summary/Llama-3.1-70B-Instruct_latency_report.csv``
+- ``./reports_float16/summary/Meta-Llama-3.1-8B-Instruct_latency_report.csv``
 
-- ``./reports_float8/summary/Llama-3.1-70B-Instruct-FP8-KV_latency_report.csv``
+- ``./reports_float8/summary/Meta-Llama-3.1-8B-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.
+Use this command to benchmark the throughput of the Llama 3.1 8B model on one GPU 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
+   ./vllm_benchmark_report.sh -s throughput -m meta-llama/Meta-Llama-3.1-8B-Instruct -g 1 -d float16
+   ./vllm_benchmark_report.sh -s throughput -m amd/Meta-Llama-3.1-8B-Instruct-FP8-KV -g 1 -d float8
 
 Find the throughput reports at:
 
-- ``./reports_float16/summary/Llama-3.1-70B-Instruct_throughput_report.csv``
+- ``./reports_float16/summary/Meta-Llama-3.1-8B-Instruct_throughput_report.csv``
 
-- ``./reports_float8/summary/Llama-3.1-70B-Instruct-FP8-KV_throughput_report.csv``
+- ``./reports_float8/summary/Meta-Llama-3.1-8B-Instruct-FP8-KV_throughput_report.csv``
 
 .. raw:: html
 
@@ -439,40 +394,17 @@ Further reading
   MI300X accelerators, see :doc:`../../system-optimization/mi300x`.
 
 - To learn how to run LLM models from Hugging Face or your own model, see
-  :doc:`Running models from Hugging Face <hugging-face-models>`.
+  :doc:`Using ROCm for AI <../index>`.
 
 - To learn how to optimize inference on LLMs, see
   :doc:`Inference optimization <../inference-optimization/index>`.
 
+<<<<<<< HEAD:docs/how-to/performance-validation/mi300x/vllm-benchmark.rst
+=======
 - To learn how to fine-tune LLMs, see
   :doc:`Fine-tuning LLMs <../fine-tuning/index>`.
 
-Previous versions
-=================
+>>>>>>> develop:docs/how-to/rocm-for-ai/inference/vllm-benchmark.rst
+- To compare with the previous version of the ROCm vLLM Docker image for performance validation, refer to
+  `LLM inference performance validation on AMD Instinct MI300X (ROCm 6.2.0) <https://rocm.docs.amd.com/en/docs-6.2.0/how-to/performance-validation/mi300x/vllm-benchmark.html>`_.
 
-This table lists previous versions of the ROCm vLLM Docker image for inference
-performance validation. For detailed information about available models for
-benchmarking, see the version-specific documentation.
-
-.. list-table::
-   :header-rows: 1
-   :stub-columns: 1
-
-   * - ROCm version
-     - vLLM version
-     - PyTorch version
-     - Resources
-
-   * - 6.2.1
-     - 0.6.4
-     - 2.5.0
-     - 
-       * `Documentation <https://rocm.docs.amd.com/en/docs-6.3.0/how-to/performance-validation/mi300x/vllm-benchmark.html>`_
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.2_mi300_ubuntu20.04_py3.9_vllm_0.6.4/images/sha256-ccbb74cc9e7adecb8f7bdab9555f7ac6fc73adb580836c2a35ca96ff471890d8>`_
-
-   * - 6.2.0
-     - 0.4.3
-     - 2.4.0
-     -
-       * `Documentation <https://rocm.docs.amd.com/en/docs-6.2.0/how-to/performance-validation/mi300x/vllm-benchmark.html>`_
-       * `Docker Hub <https://hub.docker.com/layers/rocm/vllm/rocm6.2_mi300_ubuntu22.04_py3.9_vllm_7c5fd50/images/sha256-9e4dd4788a794c3d346d7d0ba452ae5e92d39b8dfac438b2af8efdc7f15d22c0>`_

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

@@ -14,9 +14,6 @@ Training models on AMD GPUs with the ROCm™ software platform allows you to use
  
 The ROCm software platform makes it easier to train models on AMD GPUs while maintaining compatibility with existing code and tools. The platform also provides features like multi-GPU support, allowing for scaling and parallelization of model training across multiple GPUs to enhance performance. 
 
-The AI Developer Hub contains `AMD ROCm tutorials <https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/>`_ for
-training, fine-tuning, and inference. It leverages popular machine learning frameworks on AMD GPUs.
-
 In this guide, you'll learn about:
 
 - :doc:`Training a model <train-a-model>`

docs/how-to/system-optimization/w6000-v620.md

@@ -12,7 +12,7 @@ myst:
 This chapter reviews system settings that are required to configure the system
 for ROCm virtualization on RDNA2-based AMD Radeon™ PRO GPUs. Installing ROCm on
 Bare Metal follows the routine ROCm
-{doc}`installation procedure<rocm-install-on-linux:install/install-methods/package-manager-index>`.
+{doc}`installation procedure<rocm-install-on-linux:install/native-install/index>`.
 
 To enable ROCm virtualization on V620, one has to setup Single Root I/O
 Virtualization (SR-IOV) in the BIOS via setting found in the following
@@ -166,4 +166,4 @@ First, assign GPU virtual function (VF) to VM using the following steps.
 Then start the VM.
 
 Finally install ROCm on the virtual machine (VM). For detailed instructions,
-refer to the {doc}`Linux install guide<rocm-install-on-linux:install/install-methods/package-manager-index>`.
+refer to the {doc}`Linux install guide<rocm-install-on-linux:install/native-install/index>`.

docs/index.md

@@ -38,7 +38,6 @@ ROCm documentation is organized into the following categories:
 :class-body: rocm-card-banner rocm-hue-12
 
 * [Use ROCm for AI](./how-to/rocm-for-ai/index.rst)
-* [AI tutorials](https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/)
 * [Use ROCm for HPC](./how-to/rocm-for-hpc/index.rst)
 * [System optimization](./how-to/system-optimization/index.rst)
 * [AMD Instinct MI300X performance validation and tuning](./how-to/tuning-guides/mi300x/index.rst)

docs/reference/precision-support.rst

@@ -1,23 +1,19 @@
 .. meta::
-  :description: Supported data types of AMD GPUs and libraries in ROCm.
-  :keywords: precision, data types, HIP types, int8, float8, float8 (E4M3),
-             float8 (E5M2), bfloat8, float16, half, bfloat16, tensorfloat32,
-             float, float32, float64, double, AMD data types, HIP data types,
-             ROCm precision, ROCm data types
+  :description: Supported data types in ROCm
+  :keywords: int8, float8, float8 (E4M3), float8 (E5M2), bfloat8, float16, half, bfloat16, tensorfloat32, float,
+   float32, float64, double, AMD, ROCm, AMDGPU
 
 *************************************************************
-Data types and precision support
+Precision support
 *************************************************************
 
-This topic lists the supported data types of AMD GPUs and ROCm libraries.
-Corresponding :doc:`HIP <hip:index>` data types are also noted. 
+Use the following sections to identify data types and HIP types ROCm™ supports.
 
 Integral types
 ==========================================
 
-The signed and unsigned integral types supported by ROCm are listed in
-the following table, along with their corresponding HIP type and a short
-description.
+The signed and unsigned integral types that are supported by ROCm are listed in the following table,
+together with their corresponding HIP type and a short description.
 
 
 .. list-table::
@@ -50,8 +46,8 @@ description.
 Floating-point types
 ==========================================
 
-The floating-point types supported by ROCm are listed in the following
-table, along with their corresponding HIP type and a short description.
+The floating-point types that are supported by ROCm are listed in the following table, together with
+their corresponding HIP type and a short description.
 
 .. image:: ../data/about/compatibility/floating-point-data-types.png
     :alt: Supported floating-point types
@@ -67,62 +63,43 @@ table, along with their corresponding HIP type and a short description.
     *
       - float8 (E4M3)
       - ``-``
-      - 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.
+      - 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 with no infinity or signed zero. NaN is represented as negative zero.
     *
       - float8 (E5M2)
       - ``-``
-      - 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.
+      - 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 with no infinity or signed zero. NaN is represented as negative zero.
     *
       - float16
       - ``half``
-      - A 16-bit floating-point number that conforms to the IEEE 754-2008
-        half-precision storage format.
+      - A 16-bit floating-point number that conforms to the IEEE 754-2008 half-precision storage format.
     *
       - bfloat16
       - ``bfloat16``
-      - A shortened 16-bit version of the IEEE 754 single-precision storage
-        format.
+      - A shortened 16-bit version of the IEEE 754 single-precision storage format.
     *
       - tensorfloat32
       - ``-``
-      - 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.
+      - 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.
     *
       - float32
       - ``float``
-      - A 32-bit floating-point number that conforms to the IEEE 754
-        single-precision storage format.
+      - A 32-bit floating-point number that conforms to the IEEE 754 single-precision storage format.
     *
       - float64
       - ``double``
-      - A 64-bit floating-point number that conforms to the IEEE 754
-        double-precision storage format.
+      - A 64-bit floating-point number that conforms to the IEEE 754 double-precision storage format.
 
 .. note::
 
-  * The float8 and tensorfloat32 types are internal types used in calculations
-    in Matrix Cores and can be stored in any type of the same size.
-
-  * The encodings for FP8 (E5M2) and FP8 (E4M3) that the
-    MI300 series natively supports differ from the FP8 (E5M2) and FP8 (E4M3)
-    encodings used in NVIDIA H100
-    (`FP8 Formats for Deep Learning <https://arxiv.org/abs/2209.05433>`_).
-
+  * The float8 and tensorfloat32 types are internal types used in calculations in Matrix Cores and can be stored in any type of the same size.
+  * The encodings for FP8 (E5M2) and FP8 (E4M3) that are natively supported by MI300 differ from the FP8 (E5M2) and FP8 (E4M3) encodings used in H100 (`FP8 Formats for Deep Learning <https://arxiv.org/abs/2209.05433>`_).
   * In some AMD documents and articles, float8 (E5M2) is referred to as bfloat8.
 
 ROCm support icons
 ==========================================
 
-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, we use icons to represent the level of support. These icons, described in the
+following table, are also used on the library data type support pages.
 
 .. list-table::
     :header-rows: 1
@@ -144,27 +121,14 @@ support pages.
 
 .. note::
 
-  * Full support means that the type is supported natively or with hardware
-    emulation.
+  * Full support means that the type is supported natively or with hardware emulation.
+  * Native support means that the operations for that type are implemented in hardware. Types that are not natively supported are emulated with the available hardware. The performance of non-natively supported types can differ from the full instruction throughput rate. For example, 16-bit integer operations can be performed on the 32-bit integer ALUs at full rate; however, 64-bit integer operations might need several instructions on the 32-bit integer ALUs.
+  * Any type can be emulated by software, but this page does not cover such cases.
 
-  * Native support means that the operations for that type are implemented in
-    hardware. Types that are not natively supported are emulated with the
-    available hardware. The performance of non-natively supported types can
-    differ from the full instruction throughput rate. For example, 16-bit
-    integer operations can be performed on the 32-bit integer ALUs at full rate;
-    however, 64-bit integer operations might need several instructions on the
-    32-bit integer ALUs.
-
-  * Any type can be emulated by software, but this page does not cover such
-    cases.
-
-Hardware data type support
+Hardware type support
 ==========================================
 
-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.
+AMD GPU hardware support for data types is listed in the following tables.
 
 Compute units support
 -------------------------------------------------------------------------------
@@ -411,23 +375,21 @@ The following table lists data type support for atomic operations.
 
 .. note::
 
-  You can emulate atomic operations using software for cases that are not
-  natively supported. Software-emulated atomic operations have a high negative
-  performance impact when they frequently access the same memory address.
+  For cases that are not natively supported, you can emulate atomic operations using software.
+  Software-emulated atomic operations have high negative performance impact when they frequently
+  access the same memory address.
 
-Data type support in ROCm libraries
+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.
+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
-detailed description.
+The following tables list ROCm library support for specific input and output data types. For a detailed
+description, refer to the corresponding library data type support page.
 
 .. tab-set::
 
@@ -554,9 +516,8 @@ detailed description.
 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
-description.
+The following tables list ROCm library support for specific internal data types. For a detailed
+description, refer to the corresponding library data type support page.
 
 .. tab-set::
 

docs/sphinx/_toc.yml.in

@@ -89,10 +89,7 @@ subtrees:
             title: Profile and debug
           - file: how-to/rocm-for-ai/inference-optimization/workload.rst
             title: Workload tuning
-
-      - url: https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/
-        title: AI tutorials
-
+  
   - file: how-to/rocm-for-hpc/index.rst
     title: Use ROCm for HPC
   - file: how-to/system-optimization/index.rst
@@ -129,7 +126,6 @@ subtrees:
   - url: https://github.com/amd/rocm-examples
     title: ROCm examples
 
-
 - caption: Conceptual
   entries:
   - file: conceptual/gpu-arch.md

docs/sphinx/requirements.in

@@ -1,3 +1,3 @@
-rocm-docs-core==1.17.0
+rocm-docs-core==1.13.0
 sphinx-reredirects
 sphinx-sitemap

docs/sphinx/requirements.txt

@@ -8,13 +8,6 @@ accessible-pygments==0.0.5
     # via pydata-sphinx-theme
 alabaster==1.0.0
     # via sphinx
-asttokens==3.0.0
-    # via stack-data
-attrs==25.1.0
-    # via
-    #   jsonschema
-    #   jupyter-cache
-    #   referencing
 babel==2.16.0
     # via
     #   pydata-sphinx-theme
@@ -32,17 +25,9 @@ cffi==1.17.1
 charset-normalizer==3.4.0
     # via requests
 click==8.1.7
-    # via
-    #   jupyter-cache
-    #   sphinx-external-toc
-comm==0.2.2
-    # via ipykernel
+    # via sphinx-external-toc
 cryptography==43.0.3
     # via pyjwt
-debugpy==1.8.12
-    # via ipykernel
-decorator==5.1.1
-    # via ipython
 deprecated==1.2.15
     # via pygithub
 docutils==0.21.2
@@ -51,103 +36,34 @@ docutils==0.21.2
     #   myst-parser
     #   pydata-sphinx-theme
     #   sphinx
-exceptiongroup==1.2.2
-    # via ipython
-executing==2.2.0
-    # via stack-data
 fastjsonschema==2.20.0
-    # via
-    #   nbformat
-    #   rocm-docs-core
+    # via rocm-docs-core
 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
     # via sphinx
-importlib-metadata==8.6.1
-    # via
-    #   jupyter-cache
-    #   myst-nb
-ipykernel==6.29.5
-    # via myst-nb
-ipython==8.31.0
-    # via
-    #   ipykernel
-    #   myst-nb
-jedi==0.19.2
-    # via ipython
 jinja2==3.1.5
     # via
     #   myst-parser
     #   sphinx
-jsonschema==4.23.0
-    # via nbformat
-jsonschema-specifications==2024.10.1
-    # via jsonschema
-jupyter-cache==1.0.1
-    # via myst-nb
-jupyter-client==8.6.3
-    # via
-    #   ipykernel
-    #   nbclient
-jupyter-core==5.7.2
-    # via
-    #   ipykernel
-    #   jupyter-client
-    #   nbclient
-    #   nbformat
 markdown-it-py==3.0.0
     # via
     #   mdit-py-plugins
     #   myst-parser
 markupsafe==3.0.2
     # via jinja2
-matplotlib-inline==0.1.7
-    # via
-    #   ipykernel
-    #   ipython
 mdit-py-plugins==0.4.2
     # via myst-parser
 mdurl==0.1.2
     # via markdown-it-py
-myst-nb==1.1.2
-    # via rocm-docs-core
 myst-parser==4.0.0
-    # via myst-nb
-nbclient==0.10.2
-    # via
-    #   jupyter-cache
-    #   myst-nb
-nbformat==5.10.4
-    # via
-    #   jupyter-cache
-    #   myst-nb
-    #   nbclient
-nest-asyncio==1.6.0
-    # via ipykernel
+    # via rocm-docs-core
 packaging==24.2
-    # via
-    #   ipykernel
-    #   sphinx
-parso==0.8.4
-    # via jedi
-pexpect==4.9.0
-    # via ipython
-platformdirs==4.3.6
-    # via jupyter-core
-prompt-toolkit==3.0.50
-    # via ipython
-psutil==6.1.1
-    # via ipykernel
-ptyprocess==0.7.0
-    # via pexpect
-pure-eval==0.2.3
-    # via stack-data
+    # via sphinx
 pycparser==2.22
     # via cffi
 pydata-sphinx-theme==0.16.0
@@ -159,42 +75,23 @@ pygithub==2.5.0
 pygments==2.18.0
     # via
     #   accessible-pygments
-    #   ipython
     #   pydata-sphinx-theme
     #   sphinx
 pyjwt[crypto]==2.10.0
     # via pygithub
 pynacl==1.5.0
     # via pygithub
-python-dateutil==2.9.0.post0
-    # via jupyter-client
 pyyaml==6.0.2
     # via
-    #   jupyter-cache
-    #   myst-nb
     #   myst-parser
     #   rocm-docs-core
     #   sphinx-external-toc
-pyzmq==26.2.0
-    # via
-    #   ipykernel
-    #   jupyter-client
-referencing==0.36.2
-    # via
-    #   jsonschema
-    #   jsonschema-specifications
 requests==2.32.3
     # via
     #   pygithub
     #   sphinx
-rocm-docs-core==1.17.0
+rocm-docs-core==1.13.0
     # via -r requirements.in
-rpds-py==0.22.3
-    # via
-    #   jsonschema
-    #   referencing
-six==1.17.0
-    # via python-dateutil
 smmap==5.0.1
     # via gitdb
 snowballstemmer==2.2.0
@@ -204,7 +101,6 @@ soupsieve==2.6
 sphinx==8.1.3
     # via
     #   breathe
-    #   myst-nb
     #   myst-parser
     #   pydata-sphinx-theme
     #   rocm-docs-core
@@ -241,43 +137,15 @@ sphinxcontrib-qthelp==2.0.0
     # via sphinx
 sphinxcontrib-serializinghtml==2.0.0
     # via sphinx
-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
-    #   jupyter-client
-traitlets==5.14.3
-    # via
-    #   comm
-    #   ipykernel
-    #   ipython
-    #   jupyter-client
-    #   jupyter-core
-    #   matplotlib-inline
-    #   nbclient
-    #   nbformat
 typing-extensions==4.12.2
     # via
-    #   ipython
-    #   myst-nb
     #   pydata-sphinx-theme
     #   pygithub
-    #   referencing
-    #   sqlalchemy
 urllib3==2.2.3
     # via
     #   pygithub
     #   requests
-wcwidth==0.2.13
-    # via prompt-toolkit
 wrapt==1.17.0
     # via deprecated
-zipp==3.21.0
-    # via importlib-metadata

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 roctracer)
+$(call adddep,hipblaslt,hip_on_rocclr openmp_extras lightning hipcc hipblas-common rocm-dev)
 $(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)

tools/rocm-build/build_amd_smi_lib.sh

@@ -10,9 +10,7 @@ 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             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 "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
     echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of type referred to by pkg_type"
     echo "  -p,  --package <type>     Specify packaging format"
     echo "  -h,  --help               Prints this help"
@@ -27,6 +25,7 @@ printUsage() {
 PROJ_NAME="amdsmi"
 PACKAGE_ROOT="$(getPackageRoot)"
 TARGET="build"
+
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE="$(getIncludePath)"
 AMDSMI_BUILD_DIR=$(getBuildPath $PROJ_NAME)
@@ -43,7 +42,7 @@ SHARED_LIBS="ON"
 CLEAN_OR_OUT=0;
 PKGTYPE="deb"
 
-VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
+VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
@@ -61,8 +60,6 @@ do
                 ADDRESS_SANITIZER=true ; shift ;;
         (-s | --static)
                 ack_and_skip_static ;;
-        (-w | --wheel)
-                WHEEL_PACKAGE=true ; shift ;;
         (-o | --outdir)
                 TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
         (-p | --package)
@@ -102,7 +99,6 @@ build_amdsmi() {
             -DBUILD_SHARED_LIBS=$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" \
@@ -110,14 +106,14 @@ build_amdsmi() {
             -DCPACK_PACKAGE_VERSION_PATCH="0" \
             -DADDRESS_SANITIZER="$ADDRESS_SANITIZER" \
             -DBUILD_TESTS=ON \
-            -S "$AMD_SMI_LIB_ROOT"
+            "$AMD_SMI_LIB_ROOT"
         popd
     fi
 
     echo "Making amd_smi package:"
-    cmake --build "$AMDSMI_BUILD_DIR" -- $DASH_JAY
-    cmake --build "$AMDSMI_BUILD_DIR" -- $DASH_JAY install
-    cmake --build "$AMDSMI_BUILD_DIR" -- $DASH_JAY package
+    cmake --build "$AMDSMI_BUILD_DIR" -- $AMDSMI_MAKE_OPTS
+    cmake --build "$AMDSMI_BUILD_DIR" -- $AMDSMI_MAKE_OPTS install
+    cmake --build "$AMDSMI_BUILD_DIR" -- $AMDSMI_MAKE_OPTS package
 
     copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$AMDSMI_PACKAGE_DEB_DIR" $AMDSMI_BUILD_DIR/*.deb
     copy_if RPM "${CPACKGEN:-"DEB;RPM"}" "$AMDSMI_PACKAGE_RPM_DIR" $AMDSMI_BUILD_DIR/*.rpm
@@ -139,7 +135,7 @@ verifyEnvSetup
 
 case $TARGET in
     (clean) clean_amdsmi ;;
-    (build) build_amdsmi; build_wheel "$AMDSMI_BUILD_DIR" "$PROJ_NAME" ;;
+    (build) build_amdsmi ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_amdmigraphx.sh

@@ -1,42 +1,41 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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 "${BUILD_DIR}" \
-        --cxx="$(set_build_variables __CLANG++__)" \
-        --cc="$(set_build_variables __CLANG__)" \
+        --cxx="${ROCM_PATH}/llvm/bin/clang++" \
+        --cc="${ROCM_PATH}/llvm/bin/clang" \
         "${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=""
 
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
     cd $BUILD_DIR && cmake --build . -- install -j${PROC}
 
     show_build_cache_stats
@@ -51,7 +50,7 @@ clean_amdmigraphx() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_amdmigraphx; build_wheel ;;
+    build) build_amdmigraphx ;;
     outdir) print_output_directory ;;
     clean) clean_amdmigraphx ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_comgr.sh

@@ -14,7 +14,6 @@ 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>:"
@@ -25,7 +24,6 @@ printUsage() {
     return 0
 }
 
-## Build environment variables
 API_NAME=amd_comgr
 PROJ_NAME=$API_NAME
 LIB_NAME=lib${API_NAME}
@@ -35,8 +33,8 @@ PACKAGE_ROOT=$(getPackageRoot)
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE=$(getIncludePath)
 BUILD_DIR=$(getBuildPath $API_NAME)
-PACKAGE_DEB=$PACKAGE_ROOT/deb/$PROJ_NAME
-PACKAGE_RPM=$PACKAGE_ROOT/rpm/$PROJ_NAME
+PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
+PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
 PACKAGE_PREFIX=$ROCM_INSTALL_PATH
 BUILD_TYPE=Debug
 MAKE_OPTS="$DASH_JAY CTEST_OUTPUT_ON_FAILURE=1 -C $BUILD_DIR"
@@ -46,17 +44,13 @@ 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"
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswlo:p: --long help,clean,release,address_sanitizer,static,wheel,link_llvm_static,outdir:,package: -- "$@"`
+VALID_STR=`getopt -o hcraslo:p: --long help,clean,release,address_sanitizer,static,link_llvm_static,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
 do
-    #echo "parocessing $1"
     case "$1" in
         (-h | --help)
                 printUsage ; exit 0;;
@@ -69,8 +63,6 @@ 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)
@@ -113,7 +105,6 @@ build() {
        echo " Building Shared Object "
     fi
 
-# Remove CTEST var once SWDEV-381396 is fixed
     cmake \
         $(rocm_cmake_params) \
         -DBUILD_SHARED_LIBS=$SHARED_LIBS \
@@ -152,7 +143,7 @@ verifyEnvSetup
 
 case $TARGET in
     (clean) clean ;;
-    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+    (build) build ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_composable_kernel.sh

@@ -2,14 +2,10 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
-
-# Temporarily disable Address Sanitizer
-ENABLE_ADDRESS_SANITIZER=false
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 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() {
@@ -27,17 +23,10 @@ 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" ] || [ "$DISTRO_NAME" == "sles" ] || [ "$DISTRO_ID" = "debian-10" ]; then
+    if [ "$DISTRO_ID" = "rhel-8.8" ] || [ "$DISTRO_ID" = "sles-15.5" ] ; 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.
@@ -49,25 +38,23 @@ 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=$(set_build_variables __CLANG++__) \
-        -DCMAKE_C_COMPILER=$(set_build_variables __CLANG__) \
+        -DCMAKE_CXX_COMPILER="${ROCM_PATH}/llvm/bin/clang++" \
+        -DCMAKE_C_COMPILER="${ROCM_PATH}/llvm/bin/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
-
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ./*.${PKGTYPE} $PACKAGE_DIR
 }
 
-# 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"
@@ -90,7 +77,7 @@ clean_miopen_ck() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_miopen_ck; build_wheel ;;
+    build) build_miopen_ck ;;
     outdir) print_output_directory ;;
     clean) clean_miopen_ck ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_dbgapi.sh

@@ -11,8 +11,6 @@ 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"
@@ -27,7 +25,6 @@ printUsage() {
     return 0
 }
 
-## Build environment variables
 API_NAME=rocm-dbgapi
 AMD_DBGAPI_NAME=amd-dbgapi
 MAKEINSTALL_MANIFEST=makeinstall_manifest.txt
@@ -39,24 +36,21 @@ PACKAGE_ROOT=$(getPackageRoot)
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE=$(getIncludePath)
 BUILD_DIR=$(getBuildPath $API_NAME)
-PACKAGE_DEB=$(getPackageRoot)/deb/$PROJ_NAME
-PACKAGE_RPM=$(getPackageRoot)/rpm/$PROJ_NAME
-#PACKAGE_PREFIX=$ROCM_INSTALL_PATH
+PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
+PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
 BUILD_TYPE=Debug
-MAKE_OPTS=($DASH_JAY -C "$BUILD_DIR") # Note that DASH_JAY might have a space after the -j
+MAKE_OPTS=($DASH_JAY -C "$BUILD_DIR")
 SHARED_LIBS="ON"
 CLEAN_OR_OUT=0;
 MAKETARGET="deb"
 PKGTYPE="deb"
 DODOCSBUILD=true
 
-#parse the arguments
-VALID_STR=$(getopt -o hcraswo:p:M --long help,clean,release,enable-assertions,static,wheel,address_sanitizer,outdir:,package:skip_man_pages -- "$@")
+VALID_STR=$(getopt -o hcraso:p:M --long help,clean,release,enable-assertions,static,address_sanitizer,outdir:,package:skip_man_pages -- "$@")
 eval set -- "$VALID_STR"
 
 while true ;
 do
-    #echo "parocessing $1"
     case "$1" in
         (-h | --help)
                 printUsage ; exit 0;;
@@ -72,14 +66,12 @@ 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;; # end delimiter
+        --)     shift; break;;
         (*)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
     esac
@@ -112,7 +104,7 @@ 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 # THis is not an error
+       exit 0
     fi
     echo "Building $PROJ_NAME"
 
@@ -129,14 +121,11 @@ 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"
@@ -162,7 +151,7 @@ verifyEnvSetup
 
 case $TARGET in
     (clean) clean ;;
-    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME";;
+    (build) build ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_devicelibs.sh

@@ -10,9 +10,7 @@ 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             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 "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
     echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
     type referred to by pkg_type"
     echo "  -h,  --help             Prints this help"
@@ -21,11 +19,10 @@ printUsage() {
 
     return 0
 }
-PROJ_NAME="devicelibs"
 PACKAGE_ROOT="$(getPackageRoot)"
 PACKAGE_BIN="$(getBinPath)"
 PACKAGE_LIB="$(getLibPath)"
-BUILD_PATH="$(getBuildPath $PROJ_NAME)"
+BUILD_PATH="$(getBuildPath devicelibs)"
 INSTALL_PATH="$(getPackageRoot)"
 LIGHTNING_BUILD_PATH="$(getBuildPath lightning)"
 DEB_PATH="$(getDebPath devicelibs)"
@@ -39,13 +36,11 @@ CLEAN_OR_OUT=0;
 PKGTYPE="deb"
 MAKETARGET="deb"
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswo: --long help,clean,release,static,wheel,address_sanitizer,outdir: -- "$@"`
+VALID_STR=`getopt -o hcraso: --long help,clean,release,static,address_sanitizer,outdir: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
 do
-    #echo "parocessing $1"
     case "$1" in
         (-h | --help)
                 printUsage ; exit 0;;
@@ -58,8 +53,6 @@ do
                 ack_and_ignore_asan ; shift ;;
         (-s | --static)
                 SHARED_LIBS="OFF" ; shift ;;
-        (-w | --wheel)
-                WHEEL_PACKAGE=true ; shift ;;
         (-o | --outdir)
                 TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
         --)     shift; break;; # end delimiter
@@ -78,9 +71,7 @@ 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
@@ -89,7 +80,6 @@ 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"
 }
@@ -106,7 +96,6 @@ 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" \
@@ -144,7 +133,7 @@ print_output_directory() {
 }
 case $TARGET in
     (clean) clean_devicelibs ;;
-    (build) build_devicelibs; package_devicelibs; build_wheel "$BUILD_PATH" "$PROJ_NAME" ;;
+    (build) build_devicelibs; package_devicelibs ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_half.sh

@@ -1,31 +1,22 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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"
 
@@ -34,7 +25,7 @@ build_half() {
     cmake --build "$BUILD_DIR" -- install
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -48,7 +39,7 @@ clean_half() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_half; build_wheel ;;
+    build) build_half ;;
     outdir) print_output_directory ;;
     clean) clean_half ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hip_on_rocclr.sh

@@ -10,8 +10,6 @@ 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"
@@ -27,6 +25,7 @@ printUsage() {
 source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
 MAKEOPTS="$DASH_JAY"
 PROJ_NAME="hip-on-rocclr"
+
 BUILD_PATH="$(getBuildPath $PROJ_NAME)"
 
 TARGET="build"
@@ -37,6 +36,7 @@ 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"
@@ -53,10 +53,9 @@ MAKETARGET="deb"
 PKGTYPE="deb"
 OFFLOAD_ARCH=()
 
-DEFAULT_OFFLOAD_ARCH=(gfx900 gfx906 gfx908 gfx90a gfx942 gfx1030 gfx1031 gfx1033 gfx1034 gfx1035 gfx1100 gfx1101 gfx1102 gfx1103 gfx1150 gfx1151 gfx1200 gfx1201)
+DEFAULT_OFFLOAD_ARCH=(gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942 gfx1030 gfx1031 gfx1033 gfx1034 gfx1035 gfx1100 gfx1101 gfx1102 gfx1200 gfx1201)
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswt:o: --long help,clean,release,address_sanitizer,static,wheel,offload-arch=:,outdir: -- "$@"`
+VALID_STR=`getopt -o hcrast:o: --long help,clean,release,address_sanitizer,static,offload-arch=:,outdir: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
@@ -73,13 +72,11 @@ 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;; # end delimiter
+        --)     shift; break;;
 
         (*)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
@@ -101,10 +98,7 @@ 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"
@@ -112,23 +106,17 @@ 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 $HIPAMD_ROOT/CMakeLists.txt file, skipping hip on rocclr"
-        exit 0 # This is not an error
+        exit 0
     else
-        # We are in a branch that has not yet migrated to clr repo yet
         _HIP_CMAKELIST_DIR="$HIPAMD_ROOT"
         _HIP_CMAKELIST_OPT=""
     fi
@@ -137,7 +125,6 @@ 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 $SHARED_LIBS
@@ -153,7 +140,7 @@ build_hip_on_rocclr() {
             -DCMAKE_SKIP_BUILD_RPATH=TRUE \
             -DCPACK_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
             -DROCM_PATH="$ROCM_INSTALL_PATH" \
-            -DHIPCC_BIN_DIR="$ROCM_PATH/bin" \
+            -DHIPCC_BIN_DIR="$HIPCC_BUILD_DIR" \
             -DHIP_CATCH_TEST=1 \
             $_HIP_CMAKELIST_OPT \
             "$_HIP_CMAKELIST_DIR"
@@ -178,13 +165,8 @@ 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 \
@@ -195,7 +177,6 @@ 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"
 
@@ -209,7 +190,6 @@ build_catch_tests() {
 }
 
 package_samples() {
-   # TODO: To be removed once the issue related to ASAN builds are fixed
    if [ "$ASAN_CMAKE_PARAMS" == "true" ] ; then
       echo "Disable the packaging of HIP samples" >&2
       return
@@ -218,17 +198,13 @@ package_samples() {
    if [ ! -e "$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 \
@@ -315,24 +291,10 @@ print_output_directory() {
 }
 
 case $TARGET in
-    (clean)
-        clean_hip_on_rocclr
-        clean_hip_tests
-        ;;
-    (build)
-        build_hip_on_rocclr
-        build_catch_tests
-        package_hip_on_rocclr
-        package_samples
-        build_wheel "$BUILD_PATH" "$PROJ_NAME"
-        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; copy_hip_tests;;
+   (outdir) print_output_directory ;;
+        (*) die "Invalid target $TARGET" ;;
 esac
 
 echo "Operation complete"

tools/rocm-build/build_hipblas-common.sh

@@ -2,29 +2,26 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+
     show_build_cache_stats
 }
 
@@ -37,7 +34,7 @@ clean_hipblas-common() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_hipblas-common; build_wheel ;;
+    build) build_hipblas-common ;;
     outdir) print_output_directory ;;
     clean) clean_hipblas-common ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hipblas.sh

@@ -2,32 +2,28 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src hipBLAS
 
 build_hipblas() {
     echo "Start build"
 
-    CXX=$(set_build_variables __G_++__)
+    CXX="g++"
     CXX_FLAG=
     if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
-        CXX=$(set_build_variables __AMD_CLANG_++__)
-        CXX_FLAG=$(set_build_variables __CMAKE_CXX_PARAMS__)
+        CXX="amdclang++"
+        CXX_FLAG="-DCMAKE_CXX_COMPILER=${ROCM_PATH}/llvm/bin/clang++"
     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"
@@ -43,10 +39,10 @@ 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}" \
@@ -58,7 +54,7 @@ build_hipblas() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -72,7 +68,7 @@ clean_hipblas() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_hipblas; build_wheel ;;
+    build) build_hipblas ;;
     outdir) print_output_directory ;;
     clean) clean_hipblas ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hipblaslt.sh

@@ -1,20 +1,10 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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"
 
@@ -30,31 +20,31 @@ build_hipblaslt() {
     cd $COMPONENT_SRC
     mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
 
-    # 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
+    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
     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[@]}" \
+        "${rocm_math_common_cmake_params[@]}" \
         -DTensile_LOGIC= \
-        -DTensile_CODE_OBJECT_VERSION=4 \
-        -DTensile_CPU_THREADS=$((PROC / 4)) \
+        -DTensile_CODE_OBJECT_VERSION=default \
+        -DTensile_CPU_THREADS= \
         -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}
@@ -62,7 +52,8 @@ build_hipblaslt() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+
     show_build_cache_stats
 }
 
@@ -75,7 +66,7 @@ clean_hipblaslt() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_hipblaslt; build_wheel ;;
+    build) build_hipblaslt ;;
     outdir) print_output_directory ;;
     clean) clean_hipblaslt ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hipcc.sh

@@ -9,8 +9,6 @@ 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"
@@ -23,7 +21,6 @@ printUsage() {
 }
 
 
-## Build environment variables
 API_NAME=hipcc
 PROJ_NAME=$API_NAME
 
@@ -32,11 +29,10 @@ MAKEOPTS="$DASH_JAY"
 BUILD_TYPE="Debug"
 SHARED_LIBS="ON"
 BUILD_DIR=$(getBuildPath $API_NAME)
-PACKAGE_DEB=$(getPackageRoot)/deb/$PROJ_NAME
-PACKAGE_RPM=$(getPackageRoot)/rpm/$PROJ_NAME
+PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
+PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
 PACKAGE_SRC="$(getSrcPath)"
 
-#parse the arguments
 VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,address_sanitizer,static,outdir,wheel:,package: -- "$@"`
 eval set -- "$VALID_STR"
 
@@ -49,15 +45,13 @@ 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;; # end delimiter
+        --)     shift; break;;
         (*)
             echo "Invalid option [$1]" >&2; printUsage; exit 1 ;;
     esac
@@ -93,12 +87,11 @@ build() {
     fi
 
     cmake \
-        ${GEN_NINJA} \
         -DBUILD_SHARED_LIBS=$SHARED_LIBS \
         $(rocm_cmake_params) \
         $(rocm_common_cmake_params) \
         -DHIPCC_BACKWARD_COMPATIBILITY=OFF \
-        -DCMAKE_INSTALL_PREFIX="$ROCM_PATH" \
+        -DCMAKE_INSTALL_PREFIX="$OUT_DIR" \
         $HIPCC_ROOT
     popd
 
@@ -125,20 +118,10 @@ print_output_directory() {
 }
 
 case $TARGET in
-    (clean)
-        clean
-        ;;
-    (build)
-        build
-        build_wheel "$BUILD_DIR" "$PROJ_NAME"
-        copy_hipcc_sources
-        ;;
-    (outdir)
-        print_output_directory
-        ;;
-    (*)
-        die "Invalid target $TARGET"
-        ;;
+    (clean) clean ;;
+    (build) build ; copy_hipcc_sources ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
 esac
 
 echo "Operation complete"

tools/rocm-build/build_hipcub.sh

@@ -2,7 +2,7 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src hipCUB
 
@@ -17,22 +17,26 @@ 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
 
-    CXX=$(set_build_variables __CXX__)\
+    if [ -n "$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)\
     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}
@@ -40,7 +44,7 @@ build_hipcub() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -54,7 +58,7 @@ clean_hipcub() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_hipcub; build_wheel ;;
+    build) build_hipcub ;;
     outdir) print_output_directory ;;
     clean) clean_hipcub ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hipfft.sh

@@ -2,17 +2,13 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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
@@ -22,10 +18,17 @@ 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 \
-        "$(set_build_variables __CMAKE_CXX_PARAMS__)" \
+        -DCMAKE_CXX_COMPILER=$(set_build_variables CXX) \
         ${LAUNCHER_FLAGS} \
-	    "${rocm_math_common_cmake_params[@]}" \
+        "${rocm_math_common_cmake_params[@]}" \
+        -DAMDGPU_TARGETS=${GPU_TARGETS} \
         -DCMAKE_MODULE_PATH="${ROCM_PATH}/lib/cmake/hip" \
         -DCMAKE_SKIP_BUILD_RPATH=TRUE \
         -DBUILD_ADDRESS_SANITIZER="${ADDRESS_SANITIZER}" \
@@ -38,7 +41,7 @@ build_hipfft() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR  && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -52,7 +55,7 @@ clean_hipfft() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_hipfft; build_wheel ;;
+    build) build_hipfft ;;
     outdir) print_output_directory ;;
     clean) clean_hipfft ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hipfort.sh

@@ -1,18 +1,13 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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
@@ -31,7 +26,7 @@ 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"
@@ -41,7 +36,7 @@ build_hipfort() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -55,7 +50,7 @@ clean_hipfort() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_hipfort; build_wheel ;;
+    build) build_hipfort ;;
     outdir) print_output_directory ;;
     clean) clean_hipfort ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hipify_clang.sh

@@ -10,33 +10,22 @@ 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 "  -r,  --release            Make a release build"
-    echo "  -n,  --skip_hipify_tests  Skip hipify-clang testing"
+    echo "  -h,  --help             Prints this help"
+    echo "  -r,  --release          Make a release build"
     echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    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 "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
     echo
 
     return 0
 }
 
-# Build Environmental Variables
-PROJ_NAME="hipify"
 TARGET="build"
-NINJAOPTS="$DASH_JAY"
+MAKEOPTS="$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)"
@@ -47,8 +36,7 @@ MAKETARGET="deb"
 PKGTYPE="deb"
 
 
-#parse the arguments
-VALID_STR=`getopt -o hcrnawsio: --long help,clean,release,skip_hipify_tests,wheel,static,address_sanitizer,clang_headers,outdir: -- "$@"`
+VALID_STR=`getopt -o hcraso: --long help,clean,release,static,address_sanitizer,outdir: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
@@ -60,21 +48,15 @@ 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;; # end delimiter
+        --)     shift; break;;
         (*)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
     esac
@@ -97,7 +79,6 @@ clean_hipify() {
 }
 
 package_hipify() {
-    # set-up dirs
     if [ "$PACKAGEEXT" = "deb" ]; then
         rm -rf "$DEB_PATH"
         mkdir -p "$DEB_PATH"
@@ -108,9 +89,8 @@ package_hipify() {
         mkdir -p "$RPM_PATH"
     fi
 
-    # make the pkg
     pushd "$HIPIFY_CLANG_BUILD_DIR"
-    ninja $NINJAOPTS package_hipify-clang
+    make $MAKEOPTS package_hipify-clang
     popd
 
     copy_if DEB "${CPACKGEN:-"DEB;RPM"}" "$DEB_PATH"  $HIPIFY_CLANG_BUILD_DIR/hipify*.deb
@@ -122,38 +102,7 @@ build_hipify() {
     mkdir -p "$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" \
@@ -161,20 +110,10 @@ 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
 
-    if $RUN_HIPIFY_TESTS ; then
-        echo "Running hipify tests"
-        cmake --build . -- $NINJAOPTS test-hipify
-    fi
-
-    cmake --build . -- $NINJAOPTS install
+    cmake --build . -- $MAKEOPTS install
     popd
-
     pushd "$HIPIFY_ROOT"
         HIPIFY_CLANG_HASH=`git describe --dirty --long --match [0-9]* --always`
     popd
@@ -193,20 +132,10 @@ print_output_directory() {
 }
 
 case $TARGET in
-    (clean)
-        clean_hipify
-        ;;
-    (build)
-        build_hipify
-        package_hipify
-        build_wheel "$HIPIFY_CLANG_BUILD_DIR" "$PROJ_NAME"
-        ;;
-    (outdir)
-        print_output_directory
-        ;;
-    (*)
-        die "Invalid target $TARGET"
-        ;;
+    (clean) clean_hipify ;;
+    (build) build_hipify; package_hipify ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
 esac
 
 echo "Operation complete"

tools/rocm-build/build_hiprand.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src hipRAND
 
@@ -41,10 +41,17 @@ build_hiprand() {
 
     init_rocm_common_cmake_params
 
-    CXX=$(set_build_variables __CXX__)\
+    if [ -n "$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)\
     cmake \
         ${LAUNCHER_FLAGS} \
 	    "${rocm_math_common_cmake_params[@]}" \
+        -DAMDGPU_TARGETS=${GPU_TARGETS} \
         -DBUILD_SHARED_LIBS=$SHARED_LIBS \
         -DBUILD_TEST=ON \
         -DBUILD_BENCHMARK=ON \
@@ -60,7 +67,7 @@ build_hiprand() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/  && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 }
 
 clean_hiprand() {
@@ -82,7 +89,7 @@ print_output_directory() {
 }
 
 case $TARGET in
-    build) build_hiprand; build_wheel ;;
+    build) build_hiprand ;;
     outdir) print_output_directory ;;
     clean) clean_hiprand ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hipsolver.sh

@@ -2,7 +2,7 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src hipSOLVER
 
@@ -10,12 +10,12 @@ build_hipsolver() {
     echo "Start build"
 
     if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
-        CXX_FLAG=$(set_build_variables __CMAKE_CXX_PARAMS__)
+        CXX_FLAG="-DCMAKE_CXX_COMPILER=${ROCM_PATH}/llvm/bin/clang++"
     fi
 
     cd $COMPONENT_SRC
 
-    CXX=$(set_build_variables __AMD_CLANG_++__)
+    CXX="amdclang++"
     if [ "${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
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -69,7 +69,7 @@ clean_hipsolver() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_hipsolver; build_wheel ;;
+    build) build_hipsolver ;;
     outdir) print_output_directory ;;
     clean) clean_hipsolver ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hipsparse.sh

@@ -2,7 +2,7 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 PATH=${ROCM_PATH}/bin:$PATH
 set_component_src hipSPARSE
@@ -10,12 +10,12 @@ set_component_src hipSPARSE
 build_hipsparse() {
     echo "Start build"
 
-    CXX=$(set_build_variables __G_++__)
+    CXX="g++"
     CXX_FLAG=
 
     if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
-        CXX=$(set_build_variables __CXX__)
-        CXX_FLAG=$(set_build_variables __CMAKE_CXX_PARAMS__)
+        CXX="${ROCM_PATH}/llvm/bin/clang++"
+        CXX_FLAG="-DCMAKE_CXX_COMPILER=${ROCM_PATH}/llvm/bin/clang++"
     fi
 
     cd $COMPONENT_SRC
@@ -54,7 +54,7 @@ build_hipsparse() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -68,7 +68,7 @@ clean_hipsparse() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_hipsparse; build_wheel  ;;
+    build) build_hipsparse ;;
     outdir) print_output_directory ;;
     clean) clean_hipsparse ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hipsparselt.sh

@@ -1,18 +1,10 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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
@@ -26,31 +18,6 @@ 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"
 
@@ -65,19 +32,23 @@ build_hipsparselt() {
 
     cd $COMPONENT_SRC
     mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
-    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
-        create_blis_link
-        EXTRA_CMAKE_OPTIONS=("-DLINK_BLIS=ON" "-DBUILD_DIR=${BUILD_DIR}")
-    fi
-
     init_rocm_common_cmake_params
 
-    CXX=$(set_build_variables __CXX__) \
+    if [ -n "$GPU_ARCHS" ]; then
+        GPU_TARGETS="$GPU_ARCHS"
+    else
+        # gfx940;gfx941;gfx942
+        GPU_TARGETS=all
+    fi
+
+    FC=gfortran \
+    CXX="${ROCM_PATH}/bin/hipcc" \
     cmake \
+        -DAMDGPU_TARGETS=${GPU_TARGETS} \
         ${LAUNCHER_FLAGS} \
-         "${rocm_math_common_cmake_params[@]}" \
+        "${rocm_math_common_cmake_params[@]}" \
         -DTensile_LOGIC= \
-        -DTensile_CODE_OBJECT_VERSION=4 \
+        -DTensile_CODE_OBJECT_VERSION=default \
         -DTensile_CPU_THREADS= \
         -DTensile_LIBRARY_FORMAT=msgpack \
         -DBUILD_CLIENTS_SAMPLES=ON \
@@ -85,14 +56,13 @@ build_hipsparselt() {
         -DBUILD_CLIENTS_BENCHMARKS=ON \
         -DCMAKE_INSTALL_PREFIX=${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
 
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     $SCCACHE_BIN -s || echo "Unable to display sccache stats"
 }
@@ -116,7 +86,7 @@ print_output_directory() {
 }
 
 case $TARGET in
-    build) build_hipsparselt; build_wheel ;;
+    build) build_hipsparselt ;;
     outdir) print_output_directory ;;
     clean) clean_hipsparselt ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_hiptensor.sh

@@ -2,10 +2,9 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src hipTensor
-disable_debug_package_generation
 
 build_hiptensor() {
     echo "Start build hipTensor"
@@ -23,11 +22,17 @@ 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[@]}" \
-        "$(set_build_variables __CMAKE_CC_PARAMS__)" \
-        "$(set_build_variables __CMAKE_CXX_PARAMS__)" \
+        $(set_build_variables CMAKE_C_CXX) \
+        -DAMDGPU_TARGETS=${GPU_TARGETS} \
         ${LAUNCHER_FLAGS} \
         "$COMPONENT_SRC"
 
@@ -36,7 +41,7 @@ build_hiptensor() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p "$PACKAGE_DIR" && cp ${BUILD_DIR}/*.${PKGTYPE} "$PACKAGE_DIR"
 
     show_build_cache_stats
 }
@@ -50,7 +55,7 @@ clean_hiptensor() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_hiptensor; build_wheel ;;
+    build) build_hiptensor ;;
     outdir) print_output_directory ;;
     clean) clean_hiptensor ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_lightning.sh

@@ -42,7 +42,6 @@ DEB_PATH="$(getDebPath $PROJ_NAME)"
 RPM_PATH="$(getRpmPath $PROJ_NAME)"
 INSTALL_PATH="${ROCM_INSTALL_PATH}/lib/llvm"
 LLVM_ROOT_LCL="${LLVM_ROOT}"
-ROCM_WHEEL_DIR="${BUILD_PATH}/_wheel"
 
 TARGET="all"
 MAKEOPTS="$DASH_JAY"
@@ -60,7 +59,7 @@ MAKETARGET="deb"
 ASSERT_LLVM_VERSION_MAJOR=""
 ASSERT_LLVM_VERSION_MINOR=""
 
-SKIP_LIT_TESTS=1
+SKIP_LIT_TESTS=0
 BUILD_MANPAGES="ON"
 STATIC_FLAG=
 
@@ -150,7 +149,6 @@ ENABLE_RUNTIMES="$ENABLE_RUNTIMES;libcxx;libcxxabi"
 BOOTSTRAPPING_BUILD_LIBCXX=1
 
 clean_lightning() {
-    rm -rf "$ROCM_WHEEL_DIR"
     rm -rf "$BUILD_PATH"
     rm -rf "$DEB_PATH"
     rm -rf "$RPM_PATH"
@@ -332,15 +330,6 @@ build_lightning() {
     echo "End Workaround for race condition"
     cmake --build . -- $MAKEOPTS
 
-    case "$DISTRO_ID" in
-    (rhel*|centos*)
-       RHEL_BUILD=1
-       ;;
-    (*)
-       RHEL_BUILD=0
-       ;;
-     esac
-
     if [ $SKIP_LIT_TESTS -eq 0 ]; then
         if [ $RHEL_BUILD -eq 1 ]; then
             cmake --build . -- $MAKEOPTS check-lld check-mlir
@@ -1158,9 +1147,4 @@ case $TARGET in
     (*) die "Invalid target $TARGET" ;;
 esac
 
-if [[ $WHEEL_PACKAGE == true ]]; then
-    echo "Wheel Package build started !!!!"
-    create_wheel_package
-fi
-
 echo "Operation complete"

tools/rocm-build/build_miopen-deps.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src MIOpen
 
@@ -19,18 +19,18 @@ build_miopen_mlir() {
     mkdir build && cd build
     cmake \
         -G Ninja \
-        -DCMAKE_C_COMPILER="$(set_build_variables __CLANG__)" \
-        -DCMAKE_CXX_COMPILER="$(set_build_variables __CLANG++__)" \
+        -DCMAKE_C_COMPILER="${ROCM_PATH}/llvm/bin/clang" \
+        -DCMAKE_CXX_COMPILER="${ROCM_PATH}/llvm/bin/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"
 }
 
@@ -43,6 +43,12 @@ 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
@@ -50,7 +56,7 @@ 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="$(set_build_variables __CLANG++__)" --cc "$(set_build_variables __CLANG__)"
+    PATH="${PATH}:${ROCM_PATH}:${HOME}/.local/bin" rbuild prepare -d "$HOME/miopen-deps" --cxx=${ROCM_PATH}/llvm/bin/clang++ --cc ${ROCM_PATH}/llvm/bin/clang
     build_miopen_mlir "$MLIR_COMMIT"
 
     show_build_cache_stats

tools/rocm-build/build_miopen-hip.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src MIOpen
 
@@ -9,8 +9,6 @@ PACKAGE_DIR=${PACKAGE_DIR%\/*}/miopen-hip
 DEB_PATH=$PACKAGE_DIR
 RPM_PATH=$PACKAGE_DIR
 
-disable_debug_package_generation
-
 build_miopen_hip() {
     echo "Start build"
 
@@ -33,19 +31,19 @@ build_miopen_hip() {
         "${rocm_math_common_cmake_params[@]}" \
         -DMIOPEN_BACKEND=HIP \
         -DMIOPEN_OFFLINE_COMPILER_PATHS_V2=1 \
-        -DCMAKE_CXX_COMPILER=$(set_build_variables __CLANG++__) \
-        -DCMAKE_C_COMPILER=$(set_build_variables __CLANG__) \
-        -DCMAKE_PREFIX_PATH="${ROCM_PATH};${ROCM_PATH}/hip;/tmp/miopen-deps" \
+        -DCMAKE_CXX_COMPILER="${ROCM_PATH}/llvm/bin/clang++" \
+        -DCMAKE_C_COMPILER="${ROCM_PATH}/llvm/bin/clang" \
+        -DCMAKE_PREFIX_PATH="${ROCM_PATH};${ROCM_PATH}/hip;${HOME}/miopen-deps" \
         -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 $BUILD_DIR -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -64,7 +62,7 @@ checkout_lfs() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_miopen_hip; build_wheel ;;
+    build) build_miopen_hip ;;
     outdir) print_output_directory ;;
     clean) clean_miopen_hip ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_mivisionx.sh

@@ -1,44 +1,39 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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
 
-    echo "C compiler: $CC"
-    echo "CXX compiler: $CXX"
-
     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 \
         "${rocm_math_common_cmake_params[@]}" \
         -DROCM_PATH="$ROCM_PATH" \
         -DBUILD_DEV=$BUILD_DEV \
         -DCMAKE_INSTALL_LIBDIR=$(getInstallLibDir) \
+        -DAMDGPU_TARGETS=${GPU_TARGETS} \
         -DROCM_DEP_ROCMCORE=ON \
         -DROCAL_PYTHON=OFF \
         ${LAUNCHER_FLAGS} \
@@ -49,7 +44,7 @@ build_mivisionx() {
     cpack -G ${PKGTYPE^^}
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -63,7 +58,7 @@ clean_mivisionx() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_mivisionx; build_wheel ;;
+    build) build_mivisionx ;;
     outdir) print_output_directory ;;
     clean) clean_mivisionx ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_omniperf.sh

@@ -1,149 +0,0 @@
-#!/bin/bash
-
-source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
-
-printUsage() {
-    echo
-    echo "Usage: ${BASH_SOURCE##*/} [options ...]"
-    echo
-    echo "Options:"
-    echo "  -c,  --clean              Clean output and delete all intermediate work"
-    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
-    echo "  -p,  --package <type>     Specify packaging format"
-    echo "  -r,  --release            Make a release build instead of a debug build"
-    echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
-                                      type referred to by pkg_type"
-    echo "  -w,  --wheel              Creates python wheel package of omniperf.
-                                      It needs to be used along with -r option"
-    echo "  -h,  --help               Prints this help"
-    echo
-    echo "Possible values for <type>:"
-    echo "  deb -> Debian format (default)"
-    echo "  rpm -> RPM format"
-    echo
-
-    return 0
-}
-
-## Build environment variables
-API_NAME="omniperf"
-PROJ_NAME="$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"

tools/rocm-build/build_omnitrace.sh

@@ -1,243 +0,0 @@
-#!/bin/bash
-
-source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
-
-printUsage() {
-    echo
-    echo "Usage: ${BASH_SOURCE##*/} [options ...]"
-    echo
-    echo "Options:"
-    echo "  -c,  --clean              Clean output and delete all intermediate work"
-    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
-    echo "  -p,  --package <type>     Specify packaging format"
-    echo "  -r,  --release            Make a release build instead of a debug build"
-    echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
-                                      type referred to by pkg_type"
-    echo "  -w,  --wheel              Creates python wheel package of omnitrace.
-                                      It needs to be used along with -r option"
-    echo "  -h,  --help               Prints this help"
-    echo
-    echo "Possible values for <type>:"
-    echo "  deb -> Debian format (default)"
-    echo "  rpm -> RPM format"
-    echo
-
-    return 0
-}
-
-## Build environment variables
-API_NAME="omnitrace"
-PROJ_NAME="$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"

tools/rocm-build/build_opencl_icd_loader.sh

@@ -1,144 +0,0 @@
-#!/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"

tools/rocm-build/build_opencl_on_rocclr.sh

@@ -8,13 +8,11 @@ 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)"
@@ -25,7 +23,9 @@ 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"
@@ -39,8 +39,7 @@ MAKETARGET="deb"
 PKGTYPE="deb"
 
 
-#parse the arguments
-VALID_STR=`getopt -o hcrawso: --long help,clean,release,static,wheel,address_sanitizer,outdir: -- "$@"`
+VALID_STR=`getopt -o hcraso: --long help,clean,release,static,address_sanitizer,outdir: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
@@ -57,11 +56,9 @@ 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;; # end delimiter
+        --)     shift; break;;
         (*)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
     esac
@@ -76,7 +73,6 @@ 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"
@@ -87,19 +83,14 @@ 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 $OPENCL_ON_ROCclr_ROOT/CMakeLists.txt file, skipping opencl on rocclr"
-        exit 0 # This is not an error
+        exit 0
     else
-         # We are in a branch that has not yet migrated to clr repo yet
          _OCL_CMAKELIST_DIR="$OPENCL_ON_ROCclr_ROOT"
          _OCL_CMAKELIST_OPT=""
     fi
@@ -108,12 +99,10 @@ 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) \
@@ -158,20 +147,10 @@ print_output_directory() {
 }
 
 case $TARGET in
-    (clean)
-        clean_opencl_on_rocclr
-        ;;
-    (build)
-        build_opencl_on_rocclr
-        package_opencl_on_rocclr
-        build_wheel "$BUILD_PATH" "$PROJ_NAME"
-        ;;
-    (outdir)
-        print_output_directory
-        ;;
-    (*)
-        die "Invalid target $TARGET"
-        ;;
+    (clean) clean_opencl_on_rocclr ;;
+    (build) build_opencl_on_rocclr ; package_opencl_on_rocclr ;;
+    (outdir) print_output_directory ;;
+        (*) die "Invalid target $TARGET" ;;
 esac
 
 echo "Operation complete"

tools/rocm-build/build_openmp_extras.sh

@@ -10,8 +10,6 @@ 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"
@@ -25,8 +23,9 @@ printUsage() {
 
     return 0
 }
+
 PROJ_NAME="openmp-extras"
-packageMajorVersion="19.65"
+packageMajorVersion="18.63"
 packageMinorVersion="0"
 packageVersion="${packageMajorVersion}.${packageMinorVersion}.${ROCM_LIBPATCH_VERSION}"
 BUILD_PATH="$(getBuildPath $PROJ_NAME)"
@@ -36,12 +35,9 @@ TARGET="build"
 MAKEOPTS="$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}
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,address_sanitizer,static,outdir,wheel:,package: -- "$@"`
+VALID_STR=`getopt -o hcraso:p: --long help,clean,release,address_sanitizer,static,outdir,package: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
@@ -56,23 +52,17 @@ 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=$1 ; TARGET="outdir" ;;
-        -w  | --wheel )
-                WHEEL_PACKAGE=true ;;
 	-s | --static )
             export STATIC_PKG_DEPS="ON" ;;
         -h  | --help )
             printUsage ; exit 0 ;;
-        --)     shift; break;; # end delimiter
+        --)     shift; break;;
         *)
             MAKEARG=$@ ; break ;;
     esac
@@ -81,7 +71,6 @@ done
 
 
 clean_openmp_extras() {
-    # Delete cmake output and install directory
     rm -rf "$BUILD_PATH"
     rm -rf "$INSTALL_PREFIX/openmp-extras"
 }
@@ -104,9 +93,6 @@ 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" $AOMP_REPOS/aomp/bin/build_openmp.sh)
      set -e
@@ -118,7 +104,6 @@ build_openmp_extras() {
      fi
      export BUILD_AOMP=$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
@@ -145,17 +130,10 @@ build_openmp_extras() {
        echo "--------------------------"
      fi
 
-     # can we stop building flang-classic ?
-     if [ -e $AOMP_REPOS/aomp/bin/disableClassic ]; then
-       export AOMP_JENKINS_BUILD_LIST="extras openmp"
-       export AOMP_SKIP_FLANG=1
-     else
-       export AOMP_JENKINS_BUILD_LIST="extras openmp pgmath flang flang_runtime"
-     fi
+     export AOMP_JENKINS_BUILD_LIST="extras openmp pgmath flang flang_runtime"
      echo "BEGIN Build of openmp-extras"
      "$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
@@ -176,7 +154,6 @@ build_openmp_extras() {
 }
 
 package_openmp_extras_deb() {
-        # Debian packaging
         local packageName=$1
         local packageDeb="$packageDir/deb"
         local packageArch="amd64"
@@ -207,20 +184,14 @@ 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
@@ -228,29 +199,20 @@ package_openmp_extras_deb() {
 	  else
             rm -rf $packageDeb/openmp-extras/*
             mkdir -p $packageDeb/openmp-extras$copyPath/bin
-            if [ -d "$installPath"/lib-debug/src ]; then
-              cp -r --parents "$installPath"/lib-debug/src $packageDeb/openmp-extras
-            else
-              cp -r --parents "$ompdSrcDir" $packageDeb/openmp-extras
-            fi
+            cp -r --parents "$installPath"/lib-debug/src $packageDeb/openmp-extras
 	  fi
         else
-	  # FIXME: Old packaging method, can delete once new packaging lands.
           if [ "$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
@@ -262,13 +224,10 @@ 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" "$packageDeb"/openmp-extras
@@ -277,11 +236,9 @@ package_openmp_extras_deb() {
 	    cat "$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
@@ -295,11 +252,9 @@ 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
@@ -311,14 +266,12 @@ 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: $packageArch"
@@ -340,9 +293,7 @@ package_openmp_extras_deb() {
         "$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"
@@ -357,8 +308,6 @@ 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 $licenseDir/LICENSE.apache2
@@ -368,18 +317,15 @@ package_openmp_extras_asan_deb() {
         mkdir -p "$(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/
@@ -388,7 +334,6 @@ 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"
@@ -407,7 +352,6 @@ package_openmp_extras_asan_deb() {
 
 
 package_openmp_extras_rpm() {
-        # RPM packaging
         local packageName=$1
         local packageRpm="$packageDir/rpm"
         local specFile="$packageDir/$packageName.spec"
@@ -432,7 +376,6 @@ package_openmp_extras_rpm() {
           fi
         fi
 
-        # Cleanup previous packages
         if [ "$packageType" == "runtime" ]; then
           rm -rf "$packageDir"
           rm -rf "$RPM_PATH"
@@ -443,8 +386,6 @@ package_openmp_extras_rpm() {
         mkdir -p "$(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 )"
 
@@ -462,7 +403,6 @@ 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"
@@ -479,7 +419,6 @@ 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"
@@ -491,13 +430,10 @@ 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"
@@ -506,17 +442,14 @@ 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"
@@ -529,23 +462,18 @@ 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"
@@ -565,12 +493,6 @@ 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"
@@ -580,19 +502,15 @@ 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"
@@ -609,7 +527,6 @@ 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"
@@ -622,16 +539,11 @@ 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"
@@ -642,11 +554,8 @@ 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"
@@ -657,12 +566,6 @@ 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"
 
         } > $specFile
@@ -678,29 +581,21 @@ package_openmp_extras() {
     local llvm_ver=`$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" ]] || [[ $DISTRO_NAME =~ "Debian" ]]; then
-        echo "Warning: Assuming DEBs"
+      if [[ $DISTRO_NAME =~ "Ubuntu" ]]; then
         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
-    # Only build deb in Ubuntu environment
-    if [[ $DISTRO_NAME =~ "Ubuntu" ]] || [[ $DISTRO_NAME =~ "Debian" ]]; then
-      echo "Warning: Assuming DEBs"
+
+    if [[ $DISTRO_NAME =~ "Ubuntu" ]]; then
       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
@@ -708,7 +603,6 @@ package_openmp_extras() {
 }
 
 package_tests_deb(){
-    # Openmp-extras debian test packaging
     local packageDir="$BUILD_PATH/package"
     local packageDeb="$packageDir/deb"
     local packageArch="amd64"
@@ -721,7 +615,6 @@ 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"
@@ -729,15 +622,9 @@ 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
-    # 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
+    cp "$OUT_DIR/build/lightning/bin/FileCheck" "$packageDeb/openmp-extras/$installPath/bin"
     {
       echo "Package: $packageName"
       echo "Architecture: $packageArch"
@@ -755,19 +642,16 @@ 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/$installPath"
     {
@@ -795,9 +679,7 @@ package_tests_rpm(){
       echo "mkdir -p  \$RPM_BUILD_ROOT$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 "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 "cp $OUT_DIR/build/lightning/bin/FileCheck \$RPM_BUILD_ROOT$installPath/bin"
       echo 'find $RPM_BUILD_ROOT \! -type d | sed "s|$RPM_BUILD_ROOT||"> files.list'
 
       echo "%clean"
@@ -816,10 +698,8 @@ 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
@@ -838,21 +718,10 @@ print_output_directory() {
 }
 
 case $TARGET in
-    (clean)
-        clean_openmp_extras
-        ;;
-    (build)
-        build_openmp_extras
-        package_openmp_extras
-        package_tests
-        build_wheel "$BUILD_PATH" "$PROJ_NAME"
-        ;;
-    (outdir)
-        print_output_directory
-        ;;
-    (*)
-        die "Invalid target $TARGET"
-        ;;
+    (clean) clean_openmp_extras ;;
+    (build) build_openmp_extras; package_openmp_extras; package_tests ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
 esac
 
 echo "Operation complete"

tools/rocm-build/build_rccl.sh

@@ -1,10 +1,12 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src rccl
 
+ENABLE_ADDRESS_SANITIZER=false
+
 build_rccl() {
     echo "Start build"
 
@@ -22,12 +24,19 @@ 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/" \
         ${LAUNCHER_FLAGS} \
@@ -41,7 +50,7 @@ build_rccl() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -55,7 +64,7 @@ clean_rccl() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rccl; build_wheel ;;
+    build) build_rccl ;;
     outdir) print_output_directory ;;
     clean) clean_rccl ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rdc.sh

@@ -1,45 +1,121 @@
 #!/bin/bash
 
-set -ex
+source "$(dirname "$BASH_SOURCE")/compute_utils.sh"
 
-### 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
+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
+}
 
-# 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:$ROCM_EXE_RPATH"
+PACKAGE_ROOT="$(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
 fi
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
 
-set_component_src rdc
+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"
 
-# RDC
-# BUILD ARGUMENTS
-BUILD_DOCS="no"
-GRPC_PROTOC_ROOT="${BUILD_DIR}/grpc"
-GRPC_SEARCH_ROOT="/usr/grpc"
-GRPC_DESIRED_VERSION="1.67.1" # do not include 'v'
-# lib/librocm_smi64.so and lib/libamd_smi.so
+    rm -rf "$PACKAGE_INCLUDE/rdc"
+    rm -f $PACKAGE_LIB/librdc*
+    rm -f $PACKAGE_BIN/rdci
+    return 0
+}
+
+clean_grpc() {
+    rm -rf "$GRPC_BUILD_DIR"
+}
 
-# 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!"
@@ -49,96 +125,81 @@ 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}"
 
-    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
+    export LD_PRELOAD="$ASAN_LIB_PATH"
 
-    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"
+    if [ ! -d "$RDC_BUILD_DIR/rdc_libs" ]; then
+        mkdir -p $RDC_BUILD_DIR
+        pushd $RDC_BUILD_DIR
         cmake \
             -DGRPC_ROOT="$GRPC_PROTOC_ROOT" \
             -DGRPC_DESIRED_VERSION="$GRPC_DESIRED_VERSION" \
-            -DCMAKE_MODULE_PATH="$COMPONENT_SRC/cmake_modules" \
-            "${rocm_math_common_cmake_params[@]}" \
-            -DCPACK_GENERATOR="${PKGTYPE^^}" \
-            -DROCM_DIR=$ROCM_PATH \
+            -DCMAKE_MODULE_PATH="$RDC_ROOT/cmake_modules" \
+            $(rocm_cmake_params) \
+            $(rdc_backwards_compat_cmake_params) \
+            $(rocm_common_cmake_params) \
+            -DROCM_DIR=$ROCM_INSTALL_PATH \
+            -DRDC_PACKAGE="${RDC_PKG_NAME}" \
             -DCPACK_PACKAGE_VERSION_MAJOR="1" \
             -DCPACK_PACKAGE_VERSION_MINOR="$ROCM_LIBPATCH_VERSION" \
             -DCPACK_PACKAGE_VERSION_PATCH="0" \
             -DADDRESS_SANITIZER="$ADDRESS_SANITIZER" \
             -DBUILD_TESTS=ON \
-            -DBUILD_PROFILER=ON \
-            -DBUILD_RVS=ON \
             -DCMAKE_SKIP_BUILD_RPATH=TRUE \
-            "$COMPONENT_SRC"
+            -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" \
+            "$RDC_ROOT"
         popd
     fi
     echo "Making rdc package:"
-    cmake --build "$BUILD_DIR" -- -j${PROC}
-    cmake --build "$BUILD_DIR" -- install
+    cmake --build "$RDC_BUILD_DIR" -- $RDC_MAKE_OPTS
+    cmake --build "$RDC_BUILD_DIR" -- $RDC_MAKE_OPTS install
 
-    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
-        # NOTE: Must disable LD_PRELOAD hack before packaging!
-        # cmake fails with cryptic error on RHEL:
-        #
-        #    AddressSanitizer:DEADLYSIGNAL
-        #    ==17083==ERROR: AddressSanitizer: stack-overflow on address ...
-        #
-        # The issue is likely in python3.6 cpack scripts
-        unset LD_PRELOAD
+    unset LD_PRELOAD
+    cmake --build "$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"
     fi
 
-    cmake --build "$BUILD_DIR" -- package
-
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
-
     if [ "$BUILD_DOCS" = "yes" ]; then
       echo "Building Docs"
-      cmake --build "$BUILD_DIR" -- doc
-      pushd "$BUILD_DIR"/latex
+      cmake --build "$RDC_BUILD_DIR" -- $RDC_MAKE_OPTS doc
+      pushd $RDC_BUILD_DIR/latex
       cmake --build . --
-      mv refman.pdf "$ROCM_PATH/rdc/RDC_Manual.pdf"
+      mv refman.pdf "$ROCM_INSTALL_PATH/rdc/RDC_Manual.pdf"
       popd
     fi
 }
 
-clean_rdc() {
-    echo "Cleaning RDC build directory: ${BUILD_DIR} ${PACKAGE_DIR}"
-    rm -rf "$BUILD_DIR" "$PACKAGE_DIR"
-    return 0
+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
 }
 
-stage2_command_args "$@"
-disable_debug_package_generation
+verifyEnvSetup
 
 case $TARGET in
-    clean) clean_rdc ;;
-    build) build_rdc ;;
-    outdir) print_output_directory ;;
-    *) die "Invalid target $TARGET" ;;
+    (clean) clean_rdc ;;
+    (clean_grpc) clean_grpc ;;
+    (build) build_rdc ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
 esac
 
 echo "Operation complete"

tools/rocm-build/build_rocal.sh

@@ -1,49 +1,35 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src rocAL
 
 build_rocal() {
 
-    if [ "$DISTRO_ID" = "mariner-2.0" ] || [ "$DISTRO_ID" = "azurelinux-3.0" ] ; then
-        echo "Not building rocal for ${DISTRO_ID}. Exiting..."
-        return 0
-    fi
-
     echo "Start build"
 
-    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
-        ack_and_skip_static
+    # Enable ASAN
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+        set_asan_env_vars
+        set_address_sanitizer_on
     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
+#    python3 ${COMPONENT_SRC}/rocAL-setup.py
     pushd /tmp
-
     # PyBind11
-    rm -rf pybind11
     git clone -b v2.11.1  https://github.com/pybind/pybind11
     cd pybind11 && mkdir build && cd build
     cmake -DDOWNLOAD_CATCH=ON -DDOWNLOAD_EIGEN=ON ../
     make -j$(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
@@ -51,19 +37,16 @@ build_rocal() {
 
     mkdir -p $BUILD_DIR && cd $BUILD_DIR
 
-    # 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 ${COMPONENT_SRC}
-    fi
-    make -j8
+    cmake -DAMDRPP_PATH=$ROCM_PATH ${COMPONENT_SRC}
+    make -j${PROC}
     cmake --build . --target PyPackageInstall
-    make package
+    sudo make install
+    sudo make package
+    sudo chown -R $(id -u):$(id -g) ${BUILD_DIR}
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR
+    cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
     show_build_cache_stats
 }
 
@@ -76,7 +59,7 @@ clean_rocal() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rocal; build_wheel ;;
+    build) build_rocal ;;
     outdir) print_output_directory ;;
     clean) clean_rocal ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rocalution.sh

@@ -2,7 +2,7 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 PATH=${ROCM_PATH}/bin:$PATH
 set_component_src rocALUTION
@@ -16,7 +16,7 @@ build_rocalution() {
 
     cd $COMPONENT_SRC
 
-    CXX=$(set_build_variables __G_++__)
+    CXX="g++"
     if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
         set_asan_env_vars
         set_address_sanitizer_on
@@ -28,11 +28,17 @@ 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=${GPU_TARGETS} \
         -DBUILD_CLIENTS_SAMPLES=ON \
         -DBUILD_CLIENTS_TESTS=ON \
         -DBUILD_CLIENTS_BENCHMARKS=ON \
@@ -45,7 +51,7 @@ build_rocalution() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -59,7 +65,7 @@ clean_rocalution() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rocalution; build_wheel ;;
+    build) build_rocalution ;;
     outdir) print_output_directory ;;
     clean) clean_rocalution ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rocblas.sh

@@ -2,13 +2,12 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src rocBLAS
 DEPS_DIR=${HOME}/rocblas
 
 stage2_command_args "$@"
-disable_debug_package_generation
 
 build_rocblas() {
     echo "Start build"
@@ -18,37 +17,28 @@ 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"
-        # 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
+       set_asan_env_vars
+       set_address_sanitizer_on
+       export ASAN_OPTIONS="detect_leaks=0:verify_asan_link_order=0"
     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 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}"
+    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"
     fi
-
-    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 \
@@ -57,10 +47,11 @@ build_rocblas() {
         -DBUILD_CLIENTS_BENCHMARKS=ON \
         -DBUILD_CLIENTS_SAMPLES=ON \
         -DLINK_BLIS=ON \
+        -DAMDGPU_TARGETS="${GPU_TARGETS}" \
         -DTensile_CODE_OBJECT_VERSION=default \
         -DTensile_LOGIC=asm_full \
-        -DTensile_SEPARATE_ARCHITECTURES="${SEPARATE_ARCHES}" \
-        -DTensile_LAZY_LIBRARY_LOADING="${LAZY_LOADING}" \
+        -DTensile_SEPARATE_ARCHITECTURES=ON \
+        -DTensile_LAZY_LIBRARY_LOADING=ON \
         -DTensile_LIBRARY_FORMAT=msgpack \
         -DBUILD_ADDRESS_SANITIZER="${ADDRESS_SANITIZER}" \
         -DTENSILE_VENV_UPGRADE_PIP=ON \
@@ -72,7 +63,7 @@ build_rocblas() {
 
     rm -rf _CPack_Packages/ && rm -rf ./library/src/build_tmp && find -name '*.o' -delete
 
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -84,7 +75,7 @@ clean_rocblas() {
 }
 
 case $TARGET in
-    build) build_rocblas; build_wheel ;;
+    build) build_rocblas ;;
     outdir) print_output_directory ;;
     clean) clean_rocblas ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rocdecode.sh

@@ -1,17 +1,9 @@
 #!/bin/bash
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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
@@ -19,27 +11,18 @@ build_rocdecode() {
     fi
 
     mkdir -p $BUILD_DIR && cd $BUILD_DIR
-    # 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}"
-
+#    python3 ${COMPONENT_SRC}/rocDecode-setup.py --developer OFF
+    
+    cmake -DROCM_DEP_ROCMCORE=ON ${COMPONENT_SRC}
+    make -j8
+    make install
+    make package
+    
     cmake --build "$BUILD_DIR" -- -j${PROC}
-    cmake --build "$BUILD_DIR" -- install
-    cmake --build "$BUILD_DIR" -- package
-
+    cpack -G ${PKGTYPE^^} -B ${BUILD_DIR}
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR
+    cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
     show_build_cache_stats
 }
 clean_rocdecode() {
@@ -49,7 +32,7 @@ clean_rocdecode() {
 }
 stage2_command_args "$@"
 case $TARGET in
-    build) build_rocdecode; build_wheel ;;
+    build) build_rocdecode ;;
     outdir) print_output_directory ;;
     clean) clean_rocdecode ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rocfft.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 PATH=${ROCM_PATH}/bin:$PATH
 set_component_src rocFFT
@@ -9,10 +9,6 @@ set_component_src rocFFT
 build_rocfft() {
     echo "Start Build"
 
-    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
-        ack_and_skip_static
-    fi
-
     cd $COMPONENT_SRC
 
     if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
@@ -22,14 +18,17 @@ build_rocfft() {
     mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
     init_rocm_common_cmake_params
 
-    #Removed GPU ARCHS from here as it will be part of compute_utils.sh ROCMOPS-7302 & ROCMOPS-8091
+    if [ -n "$GPU_ARCHS" ]; then
+        GPU_TARGETS="$GPU_ARCHS"
+    else
+        GPU_TARGETS="gfx908;gfx90a;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
+    fi
 
-    # Work around for HIP sources with C++ suffix, and force CXXFLAGS for both
-    # HIP and C++ compiles
-    CXX=$(set_build_variables __HIP_CC__) \
+    CXX="${ROCM_PATH}/bin/hipcc" \
     cmake \
         ${LAUNCHER_FLAGS} \
         "${rocm_math_common_cmake_params[@]}" \
+        -DAMDGPU_TARGETS=${GPU_TARGETS} \
         -DUSE_HIP_CLANG=ON \
         -DHIP_COMPILER=clang  \
         -DBUILD_CLIENTS_SAMPLES=ON  \
@@ -42,7 +41,7 @@ build_rocfft() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -56,7 +55,7 @@ clean_rocfft() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rocfft; build_wheel ;;
+    build) build_rocfft ;;
     outdir) print_output_directory ;;
     clean) clean_rocfft ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rocjpeg.sh

@@ -1,13 +1,9 @@
 #!/bin/bash
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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
@@ -17,15 +13,16 @@ build_rocjpeg() {
     mkdir -p $BUILD_DIR && cd $BUILD_DIR
     # python3 ../rocJPEG-setup.py
 
-    cmake ${GEN_NINJA} -DROCM_DEP_ROCMCORE=ON -DROCJPEG_ENABLE_ROCPROFILER_REGISTER=ON "$COMPONENT_SRC"
-    ninja -j8
-    ninja install
-    ninja package
+    cmake -DROCM_DEP_ROCMCORE=ON "$COMPONENT_SRC"
+    make -j8
+    make install
+    make package
 
     cmake --build "$BUILD_DIR" -- -j${PROC}
     cpack -G ${PKGTYPE^^}
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR
+    cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
     show_build_cache_stats
 }
 clean_rocjpeg() {

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 (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/$PROJ_NAME"
+ROCM_CMAKE_BUILD_DIR="$(getBuildPath rocm-cmake)"
+ROCM_CMAKE_PACKAGE_DEB="$(getPackageRoot)/deb/rocm-cmake"
+ROCM_CMAKE_PACKAGE_RPM="$(getPackageRoot)/rpm/rocm-cmake"
 ROCM_CMAKE_BUILD_TYPE="debug"
 BUILD_TYPE="Debug"
 SHARED_LIBS="ON"
@@ -39,13 +39,11 @@ 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;;
@@ -57,13 +55,11 @@ 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;; # end delimiter
+        --)     shift; break;;
         (*)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
     esac
@@ -119,19 +115,10 @@ print_output_directory() {
 }
 
 case $TARGET in
-    (clean)
-        clean_rocm_cmake
-        ;;
-    (build)
-        build_rocm_cmake
-        build_wheel "$ROCM_CMAKE_BUILD_DIR" "$PROJ_NAME"
-        ;;
-    (outdir)
-        print_output_directory
-        ;;
-    (*)
-        die "Invalid target $TARGET"
-        ;;
+    (clean) clean_rocm_cmake ;;
+    (build) build_rocm_cmake ;;
+    (outdir) print_output_directory ;;
+    (*) die "Invalid target $TARGET" ;;
 esac
 
 echo "Operation complete"

tools/rocm-build/build_rocm-core.sh

@@ -13,10 +13,8 @@ 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)"
@@ -26,7 +24,6 @@ printUsage() {
     return 0
 }
 
-## ROCm build (using CMake) environment variables
 PROJ_NAME="rocm-core"
 PACKAGE_ROOT="$(getPackageRoot)"
 ROCM_CORE_BUILD_DIR="$(getBuildPath rocm_core)"
@@ -41,8 +38,7 @@ MAKETARGET="deb"
 PKGTYPE="deb"
 ADDRESS_SANITIZER=false
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,address_sanitizer,outdir,wheel:,package: -- "$@"`
+VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
@@ -60,8 +56,6 @@ do
                 ADDRESS_SANITIZER=true ; shift ;;
         (-s | --static)
                 SHARED_LIBS="OFF" ; shift ;;
-        (-w | --wheel)
-                WHEEL_PACKAGE=true ; shift ;;
         (-o | --outdir)
                 TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
         (-p | --package )
@@ -93,11 +87,14 @@ build_rocm_core() {
     if [ ! -d "$ROCM_CORE_BUILD_DIR" ]; then
        mkdir -p "$ROCM_CORE_BUILD_DIR"
     fi
+
     pushd "$ROCM_CORE_BUILD_DIR"
     cmake \
-            $(rocm_cmake_params) \
+            -DCMAKE_VERBOSE_MAKEFILE=1 \
             $(rocm_common_cmake_params) \
-            -DBUILD_SHARED_LIBS=$SHARED_LIBS \
+            -DCMAKE_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
+            -DCPACK_PACKAGING_INSTALL_PREFIX="$ROCM_INSTALL_PATH" \
+            -DCPACK_GENERATOR="${CPACKGEN:-"DEB;RPM"}" \
             -DCPACK_DEBIAN_PACKAGE_RELEASE=$CPACK_DEBIAN_PACKAGE_RELEASE \
             -DCPACK_RPM_PACKAGE_RELEASE=$CPACK_RPM_PACKAGE_RELEASE \
             -DROCM_VERSION="$ROCM_VERSION" \
@@ -112,33 +109,24 @@ 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 $TARGET in
-    (clean)
-        clean_rocm_core
-        ;;
-    (build)
-        build_rocm_core
-        build_wheel "$ROCM_CORE_BUILD_DIR" "$PROJ_NAME"
-        ;;
-    (outdir)
-        print_output_directory
-        ;;
-    (*)
-        die "Invalid target $TARGET"
-        ;;
+    (clean) clean_rocm_core ;;
+    (build) build_rocm_core ;;
+    (outdir) print_output_directory ;;
+        (*) die "Invalid target $TARGET" ;;
 esac
 
 echo "Operation complete"

tools/rocm-build/build_rocm-dev.sh

@@ -2,18 +2,13 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.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 $TARGET in
-    build) build_rocm-dev ;;
-    outdir) echo "${OUT_DIR}/${PKGTYPE}/rocm-dev" ;;
-    clean) rm -rf ${OUT_DIR}/${PKGTYPE}/rocm-dev ;;
+    build) echo "end of rocm-dev build..." ;;
+    outdir) ;;
+    clean) echo "Cleaning rocm-dev is not required..." ;;
     *) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_rocm-gdb.sh

@@ -1,6 +1,7 @@
 #!/bin/bash
+
 source "${BASH_SOURCE%/*}/compute_utils.sh" || return
-# Can't use -R or -r in here
+
 remove_make_r_flags
 
 printUsage() {
@@ -14,10 +15,8 @@ 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)"
@@ -33,32 +32,23 @@ 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                                  # 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
+MAKETARGET=deb
+BUILD_DIR=$(getBuildPath $PROJ_NAME)
+PACKAGE_DEB=$(getPackageRoot)/deb/$PROJ_NAME
+PACKAGE_RPM=$(getPackageRoot)/rpm/$PROJ_NAME
+MAKE_OPTS="$DASH_JAY"
 BUG_URL="https://github.com/ROCm-Developer-Tools/ROCgdb/issues"
 SHARED_LIBS="ON"
 CLEAN_OR_OUT=0;
 MAKETARGET="deb"
 PKGTYPE="deb"
 LDFLAGS="$LDFLAGS -Wl,--enable-new-dtags"
-LIB_AMD_PYTHON="amdpythonlib.so"                #lib name which replaces required python lib
-LIB_AMD_PYTHON_DIR_PATH=${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.
+LIB_AMD_PYTHON="libamdpython.so"
 
 tokeep=(
     main${ROCM_INSTALL_PATH}/bin/rocgdb
@@ -87,18 +77,13 @@ tokeep=(
 keep_wanted_files(){
     (
         cd "$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/"
@@ -128,57 +113,43 @@ 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/postinst" <<EOF
+    grep -v '^# ' > "$BUILD_DIR/package/main/DEBIAN/preinst" <<EOF
 #!/bin/sh
-
-# 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."
+# 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."
 EOF
-
-
-    grep -v '^# ' > "$BUILD_DIR/package/main/DEBIAN/prerm" <<EOF
+    grep -v '^# ' > "$BUILD_DIR/package/main/DEBIAN/postrm" <<EOF
 #!/bin/sh
-
-# 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
+# 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 "pre-uninstallation done."
+echo "post-uninstallation done."
 EOF
-
-    chmod +x $BUILD_DIR/package/main/DEBIAN/prerm
-    chmod +x $BUILD_DIR/package/main/DEBIAN/postinst
-
+    chmod +x $BUILD_DIR/package/main/DEBIAN/postrm
+    chmod +x $BUILD_DIR/package/main/DEBIAN/preinst
     # Create control file, with variable substitution.
     # Lines with # at the start are removed, to allow for comments
     mkdir "$BUILD_DIR/debian"
@@ -196,9 +167,8 @@ Section: utils
 Architecture: amd64
 Essential: no
 Priority: optional
-Depends: \${shlibs:Depends}, rocm-dbgapi, rocm-core, python3-dev
+Depends: \${shlibs:Depends}, rocm-dbgapi, rocm-core
 EOF
-
     # Use dpkg-shlibdeps to list shlib dependencies, the result is placed
     # in $BUILD_DIR/debian/substvars.
     (
@@ -209,7 +179,6 @@ EOF
 	fi
 	dpkg-shlibdeps --ignore-missing-info  -e "$BUILD_DIR/package/main/${ROCM_INSTALL_PATH}/bin/rocgdb"
     )
-
     # Generate the final DEBIAN/control, and substitute the shlibs:Depends.
     # This is a bit unorthodox as we are only using bits and pieces of the
     # dpkg tools.
@@ -222,10 +191,8 @@ EOF
 	    -e "s/\\$\{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"
@@ -246,28 +213,21 @@ 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/$1" # e.g. out/ubuntu-16.04/16.04/build/rocm-gdb/package_rpm/main
-    local specFile="$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 packageDir="$BUILD_DIR/package_rpm/$1"
+    local specFile="$packageDir/$1.spec"
+    local packageRpm="$packageDir/rpm"
+
     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/$/ \\/;
@@ -278,8 +238,6 @@ 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
@@ -298,7 +256,8 @@ Version: ${VERSION//-/_}
 Release: ${CPACK_RPM_PACKAGE_RELEASE}%{?dist}
 License: GPL
 Prefix: ${ROCM_INSTALL_PATH}
-Requires: rocm-core, rocm-dbgapi
+Requires: rocm-core
+Provides: $LIB_AMD_PYTHON()(64bit)
 
 %description
 This is ROCgdb, the ROCm source-level debugger for Linux, based on
@@ -319,6 +278,27 @@ 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
@@ -327,27 +307,24 @@ 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/$1" # e.g. out/ubuntu-16.04/16.04/build/rocm-gdb/package_rpm/main
-    local specFile="$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 packageDir="$BUILD_DIR/package_rpm/$1"
+    local specFile="$packageDir/$1.spec"
+    local packageRpm="$packageDir/rpm"
+
     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}
 
@@ -356,8 +333,6 @@ 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/$/ \\/;
@@ -410,15 +385,10 @@ 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 $specFile
-    # Now copy it to final location
-    mkdir -p "$PACKAGE_RPM"        # e.g. out/ubuntu-16.04/16.04/rpm/rocm-gdb
+    mkdir -p "$PACKAGE_RPM"
     mv $packageRpm/RPMS/x86_64/*.rpm "$PACKAGE_RPM"
 }
 
@@ -426,29 +396,20 @@ build() {
     if [ ! -e "$ROCM_GDB_ROOT/configure" ]
     then
         toStdoutStderr "No $ROCM_GDB_ROOT/configure file, skippping rocm-gdb"
-        exit 0                        # This is not an error
+        exit 0
     fi
     local pythonver=python3
     if [[ "$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" \
@@ -480,31 +441,24 @@ build() {
 	LDFLAGS="$LDFLAGS"
     LD_RUN_PATH='${ORIGIN}/../lib' make $MAKE_OPTS
 
-    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 $BUILD_DIR/gdb/gdb
-    fi
+    REPLACE_LIB_NAME=$(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
 
     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")
@@ -521,14 +475,11 @@ verifyEnvSetup
 
 main(){
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
+VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
-
 ASAN_BUILD="no"
 while true ;
 do
-    #echo "parocessing $1"
     case "$1" in
         (-h | --help)
                 printUsage ; exit 0;;
@@ -539,11 +490,9 @@ 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
@@ -556,7 +505,6 @@ 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"
@@ -570,28 +518,17 @@ if [ $RET_CONFLICT -ge 30 ]; then
 fi
 
     case $TARGET in
-        ("clean")
-            clean
-            ;;
-        ("build")
-            build
-            build_wheel "$BUILD_DIR" "$PROJ_NAME"
-            ;;
-        ("outdir")
-            print_output_directory
-            ;;
-        (*)
-            die "Invalid target $TARGET"
-            ;;
+        ("clean") clean ;;
+        ("build") build ;;
+        ("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                       # Undo the damage from compute_utils.sh
+    set +e
 fi

tools/rocm-build/build_rocm_smi_lib.sh

@@ -11,8 +11,6 @@ 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)"
@@ -24,10 +22,7 @@ printUsage() {
 
     return 0
 }
-# RSMI ==> deb/rpm package target
-# ROCM_SMI ==> rocm_smi library target and header
 
-# ROCM_SMI
 PROJ_NAME="rsmi"
 PACKAGE_ROOT="$(getPackageRoot)"
 TARGET="build"
@@ -35,18 +30,15 @@ TARGET="build"
 PACKAGE_LIB=$(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}"
@@ -54,13 +46,11 @@ SHARED_LIBS="ON"
 CLEAN_OR_OUT=0;
 PKGTYPE="deb"
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
+VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
 do
-    #echo "parocessing $1"
     case "$1" in
         (-h | --help)
                 printUsage ; exit 0;;
@@ -75,15 +65,13 @@ 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;; # end delimiter
+        --)     shift; break;;
         (*)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
     esac
@@ -121,7 +109,6 @@ build_rsmi() {
             -DBUILD_SHARED_LIBS=$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" \
@@ -129,14 +116,14 @@ build_rsmi() {
             -DCPACK_PACKAGE_VERSION_PATCH="0" \
             -DADDRESS_SANITIZER="$ADDRESS_SANITIZER" \
             -DBUILD_TESTS=ON \
-            -S "$ROCM_SMI_LIB_ROOT"
+            "$ROCM_SMI_LIB_ROOT"
         popd
     fi
 
     echo "Making rocm_smi package:"
-    cmake --build "$RSMI_BUILD_DIR" -- $DASH_JAY
-    cmake --build "$RSMI_BUILD_DIR" -- $DASH_JAY install
-    cmake --build "$RSMI_BUILD_DIR" -- $DASH_JAY package
+    cmake --build "$RSMI_BUILD_DIR" -- $RSMI_MAKE_OPTS
+    cmake --build "$RSMI_BUILD_DIR" -- $RSMI_MAKE_OPTS install
+    cmake --build "$RSMI_BUILD_DIR" -- $RSMI_MAKE_OPTS package
 
     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
@@ -158,11 +145,10 @@ verifyEnvSetup
 
 case $TARGET in
     (clean) clean_rsmi ;;
-    (build) build_rsmi; build_wheel "$RSMI_BUILD_DIR" "$PROJ_NAME" ;;
+    (build) build_rsmi ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $TARGET" ;;
 esac
 
 echo "Operation complete"
 exit 0
-

tools/rocm-build/build_rocminfo.sh

@@ -10,9 +10,7 @@ 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             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 "  -s,  --static             Supports static CI by accepting this param & not bailing out. No effect of the param though"
     echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of
                                       type referred to by pkg_type"
     echo "  -h,  --help               Prints this help"
@@ -41,8 +39,7 @@ MAKETARGET="deb"
 PKGTYPE="deb"
 
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswo:g: --long help,clean,release,static,wheel,address_sanitizer,outdir:,gpu_list: -- "$@"`
+VALID_STR=`getopt -o hcraso:g: --long help,clean,release,static,address_sanitizer,outdir:,gpu_list: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
@@ -59,13 +56,11 @@ 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;; # end delimiter
+        --)     shift; break;;
         (*)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
     esac
@@ -129,7 +124,7 @@ print_output_directory() {
 
 case $TARGET in
     (clean) clean_rocminfo ;;
-    (build) build_rocminfo; build_wheel "$ROCMINFO_BUILD_DIR" "$PROJ_NAME" ;;
+    (build) build_rocminfo ;;
    (outdir) print_output_directory ;;
         (*) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_rocmvalidationsuite.sh

@@ -1,11 +1,10 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src ROCmValidationSuite
 
-# Internal RVS libraries need the extra RPATH
 ROCM_RVS_LIB_RPATH="\$ORIGIN/.."
 
 build_rocmvalidationsuite() {
@@ -15,7 +14,6 @@ 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
@@ -24,12 +22,12 @@ 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"
 
@@ -52,7 +50,7 @@ clean_rocmvalidationsuite() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rocmvalidationsuite; build_wheel ;;
+    build) build_rocmvalidationsuite ;;
     outdir) print_output_directory ;;
     clean) clean_rocmvalidationsuite ;;
     *) die "Invalid target ${TARGET}" ;;

tools/rocm-build/build_rocprim.sh

@@ -2,7 +2,7 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src rocPRIM
 
@@ -10,18 +10,12 @@ 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"
@@ -29,14 +23,18 @@ build_rocprim() {
 
     mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
 
-    #Removed GPU ARCHS from here as it will be part of compute_utils.sh ROCMOPS-7302 & ROCMOPS-8091
+    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
-    CXX=$(set_build_variables __HIP_CC__) \
+    CXX="${ROCM_PATH}/bin/hipcc" \
     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 \
@@ -48,7 +46,7 @@ build_rocprim() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -62,7 +60,7 @@ clean_rocprim() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rocprim; build_wheel ;;
+    build) build_rocprim ;;
     outdir) print_output_directory ;;
     clean) clean_rocprim ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rocprofiler-compute.sh

@@ -26,7 +26,6 @@ printUsage() {
     return 0
 }
 
-## Build environment variables
 API_NAME="rocprofiler-compute"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -44,13 +43,11 @@ 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;;
@@ -69,7 +66,7 @@ do
                 ack_and_skip_static ;;
         -w | --wheel)
                 WHEEL_PACKAGE=true ; shift ;;
-        --)     shift; break;; # end delimiter
+        --)     shift; break;;
         *)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
     esac
@@ -111,14 +108,13 @@ build() {
         cmake \
             $(rocm_cmake_params) \
             $(rocm_common_cmake_params) \
-            ${GEN_NINJA} \
             -DCHECK_PYTHON_DEPS=NO \
             -DPYTHON_DEPS=${BUILD_DIR}/python-libs \
-            -S "$ROCPROFILER_COMPUTE_ROOT"
+            "$ROCPROFILER_COMPUTE_ROOT"
     fi
-    ninja $MAKE_OPTS
-    ninja $MAKE_OPTS install
-    ninja $MAKE_OPTS package
+    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
@@ -140,7 +136,7 @@ verifyEnvSetup
 
 case "$TARGET" in
     (clean) clean ;;
-    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+    (build) build ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_rocprofiler-register.sh

@@ -12,7 +12,6 @@ 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"
@@ -25,7 +24,6 @@ printUsage() {
     return 0
 }
 
-## Build environment variables
 API_NAME="rocprofiler-register"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -46,13 +44,11 @@ MAKETARGET="deb"
 PKGTYPE="deb"
 
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,address_sanitizer,wheeloutdir:,package: -- "$@"`
+VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
 do
-    #echo "parocessing $1"
     case "$1" in
         (-h | --help)
                 printUsage ; exit 0;;
@@ -65,13 +61,11 @@ 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;; # end delimiter
+        --)     shift; break;;
         (*)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
     esac
@@ -105,18 +99,17 @@ 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 \
-            -S "$ROCPROFILER_REGISTER_ROOT"
+            "$ROCPROFILER_REGISTER_ROOT"
         popd
     fi
-    ninja $MAKE_OPTS
-    ninja $MAKE_OPTS install
-    ninja $MAKE_OPTS package
+    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
@@ -138,7 +131,7 @@ verifyEnvSetup
 
 case "$TARGET" in
     (clean) clean ;;
-    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+    (build) build ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_rocprofiler-sdk.sh

@@ -26,7 +26,6 @@ printUsage() {
     return 0
 }
 
-## Build environment variables
 API_NAME="rocprofiler-sdk"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -36,8 +35,8 @@ PACKAGE_ROOT="$(getPackageRoot)"
 PACKAGE_LIB="$(getLibPath)"
 PACKAGE_INCLUDE="$(getIncludePath)"
 BUILD_DIR="$(getBuildPath $API_NAME)"
-PACKAGE_DEB="$PACKAGE_ROOT/deb/$PROJ_NAME"
-PACKAGE_RPM="$PACKAGE_ROOT/rpm/$PROJ_NAME"
+PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
+PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
 PACKAGE_PREFIX="$ROCM_INSTALL_PATH"
 BUILD_TYPE="Debug"
 MAKE_OPTS="$DASH_JAY"
@@ -45,16 +44,14 @@ SHARED_LIBS="ON"
 CLEAN_OR_OUT=0
 MAKETARGET="deb"
 PKGTYPE="deb"
-# Handling GPU Targets for HSACO and HIP Executables
-GPU_LIST="gfx900;gfx906;gfx908;gfx90a;gfx942;gfx1030;gfx1031;gfx1100;gfx1101;gfx1102"
+
+GPU_LIST="gfx900;gfx906;gfx908;gfx90a;gfx940;gfx941;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
@@ -197,7 +194,7 @@ verifyEnvSetup
 
 case "$TARGET" in
     clean) clean ;;
-    build) build_rocprofiler-sdk; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+    build) build_rocprofiler-sdk ;;
     outdir) print_output_directory ;;
     *) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_rocprofiler-systems.sh

@@ -26,7 +26,6 @@ printUsage() {
     return 0
 }
 
-## Build environment variables
 API_NAME="rocprofiler-systems"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -34,11 +33,11 @@ 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"
@@ -47,7 +46,6 @@ 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"
 
@@ -123,10 +121,6 @@ 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"
@@ -147,9 +141,9 @@ 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"
@@ -193,7 +187,6 @@ build_rocprofiler_systems() {
 
     else
         cmake \
-            ${GEN_NINJA} \
             $(rocm_cmake_params) \
             $(rocm_common_cmake_params) \
             -DROCPROFSYS_BUILD_{LIBUNWIND,DYNINST}=ON \
@@ -235,7 +228,7 @@ verifyEnvSetup
 
 case "$TARGET" in
     clean) clean ;;
-    build) build_rocprofiler_systems; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+    build) build_rocprofiler_systems ;;
     outdir) print_output_directory ;;
     *) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_rocprofiler.sh

@@ -12,8 +12,6 @@ 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"
@@ -26,7 +24,6 @@ printUsage() {
     return 0
 }
 
-## Build environment variables
 API_NAME="rocprofiler"
 PROJ_NAME="$API_NAME"
 LIB_NAME="lib${API_NAME}"
@@ -36,8 +33,8 @@ PACKAGE_ROOT="$(getPackageRoot)"
 PACKAGE_LIB="$(getLibPath)"
 PACKAGE_INCLUDE="$(getIncludePath)"
 BUILD_DIR="$(getBuildPath $API_NAME)"
-PACKAGE_DEB="$PACKAGE_ROOT/deb/$PROJ_NAME"
-PACKAGE_RPM="$PACKAGE_ROOT/rpm/$PROJ_NAME"
+PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
+PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
 PACKAGE_PREFIX="$ROCM_INSTALL_PATH"
 BUILD_TYPE="Debug"
 MAKE_OPTS="$DASH_JAY -C $BUILD_DIR"
@@ -45,15 +42,12 @@ SHARED_LIBS="ON"
 CLEAN_OR_OUT=0
 MAKETARGET="deb"
 PKGTYPE="deb"
-# Handling GPU Targets for HSACO and HIP Executables
-GPU_LIST="gfx900,gfx906,gfx908,gfx90a,gfx942,gfx1030,gfx1031,gfx1100,gfx1101,gfx1102,gfx1200,gfx1201"
+GPU_LIST="gfx900,gfx906,gfx908,gfx90a,gfx940,gfx941,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
@@ -75,9 +69,6 @@ while true; do
         ;;
     -s | --static)
         ack_and_skip_static
-        ;;
-    -w | --wheel)
-        WHEEL_PACKAGE=true
         shift
         ;;
     -o | --outdir)
@@ -94,7 +85,7 @@ while true; do
     --)
         shift
         break
-        ;; # end delimiter
+        ;;
     *)
         echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] " >&2
         exit 20
@@ -124,6 +115,9 @@ 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"
@@ -179,7 +173,7 @@ verifyEnvSetup
 
 case "$TARGET" in
 clean) clean ;;
-build) build_rocprofiler; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+build) build_rocprofiler ;;
 outdir) print_output_directory ;;
 *) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_rocr_debug_agent.sh

@@ -11,8 +11,6 @@ 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"
@@ -25,7 +23,6 @@ printUsage() {
     return 0
 }
 
-## Build environment variables
 API_NAME=rocm-debug-agent
 PROJ_NAME=$API_NAME
 LIB_NAME=lib${API_NAME}.so
@@ -36,12 +33,12 @@ PACKAGE_BIN="$(getBinPath)"
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE=$(getIncludePath)
 BUILD_DIR=$(getBuildPath $API_NAME)
-PACKAGE_DEB=$PACKAGE_ROOT/deb/$PROJ_NAME
-PACKAGE_RPM=$PACKAGE_ROOT/rpm/$PROJ_NAME
+PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
+PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
 PACKAGE_PREFIX=$ROCM_INSTALL_PATH
 BUILD_TYPE=Debug
+MAKE_OPTS="$DASH_JAY -C"
 
-## Test variables
 TEST_PACKAGE_DIR="$(getBinPath)/rocm-debug-agent-test"
 PACKAGE_UTILS=$(getUtilsPath)
 
@@ -57,13 +54,11 @@ MAKETARGET="deb"
 PKGTYPE="deb"
 
 
-#parse the arguments
-VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
+VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
 do
-    #echo "parocessing $1"
     case "$1" in
         (-h | --help)
                 printUsage ; exit 0;;
@@ -76,13 +71,11 @@ 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 ;;
         (-p | --package)
                 MAKETARGET="$2" ; shift 2;;
-        --)     shift; break;; # end delimiter
+        --)     shift; break;;
         (*)
                 echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] ">&2 ; exit 20;;
     esac
@@ -109,17 +102,14 @@ 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) \
@@ -128,22 +118,15 @@ build() {
 
         popd
     fi
-    cmake --build "$BUILD_DIR" -- $DASH_JAY
-    cmake --build "$BUILD_DIR" --target install -- $DASH_JAY
-    cmake --build "$BUILD_DIR" --target package -- $DASH_JAY
+    cmake --build "$BUILD_DIR" -- $MAKE_OPTS $BUILD_DIR
+    cmake --build "$BUILD_DIR" -- $MAKE_OPTS $BUILD_DIR install
+    cmake --build "$BUILD_DIR" -- $MAKE_OPTS $BUILD_DIR package
 
     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() {
@@ -162,7 +145,7 @@ verifyEnvSetup
 
 case $TARGET in
     (clean) clean ;;
-    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+    (build) build ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $target" ;;
 esac

tools/rocm-build/build_rocrand.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src rocRAND
 
@@ -13,37 +13,31 @@ 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
-        # 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
+         set_asan_env_vars
+         set_address_sanitizer_on
     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 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}"
+    if [ -n "$GPU_ARCHS" ]; then
+        GPU_TARGETS="$GPU_ARCHS"
+    else
+        GPU_TARGETS="gfx908:xnack-;gfx90a:xnack-;gfx90a:xnack+;gfx940;gfx941;gfx942;gfx1030;gfx1100;gfx1101"
     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 \
@@ -59,7 +53,7 @@ build_rocrand() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/  && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -73,7 +67,7 @@ clean_rocrand() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rocrand; build_wheel ;;
+    build) build_rocrand ;;
     outdir) print_output_directory ;;
     clean) clean_rocrand ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rocrsamples.sh

@@ -42,7 +42,7 @@ PKGTYPE="deb"
 
 
 #parse the arguments
-VALID_STR=`getopt -o hcrawo:seg: --long help,clean,release,outdir:,static,wheel,address_sanitizer,emulator,gpu_list: -- "$@"`
+VALID_STR=`getopt -o hcrao:seg: --long help,clean,release,outdir:,static,address_sanitizer,emulator,gpu_list: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
@@ -61,8 +61,6 @@ 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 )
@@ -94,7 +92,6 @@ 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" \

tools/rocm-build/build_rocsolver.sh

@@ -2,7 +2,7 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src rocSOLVER
 
@@ -15,38 +15,39 @@ build_rocsolver() {
         SHARED_LIBS="OFF"
     fi
 
-    EXTRA_TESTS="ON"
     if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
-        set_asan_env_vars
-        set_address_sanitizer_on
-        rebuild_lapack
-        EXTRA_TESTS="OFF"
-        # 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
+       set_asan_env_vars
+       set_address_sanitizer_on
     fi
+
     cd $COMPONENT_SRC
 
     mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
 
-    # 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}"
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
+       rebuild_lapack
+    fi
+
+    if [ -n "$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"
     fi
 
     init_rocm_common_cmake_params
-    CXX=$(set_build_variables __HIP_CC__) \
+    CXX="${ROCM_PATH}/bin/hipcc" \
     cmake \
         ${LAUNCHER_FLAGS} \
         "${rocm_math_common_cmake_params[@]}" \
         -DBUILD_SHARED_LIBS=$SHARED_LIBS \
         -Drocblas_DIR="${ROCM_PATH}/rocblas/lib/cmake/rocblas" \
+        -DAMDGPU_TARGETS="${GPU_TARGETS}" \
         -DBUILD_CLIENTS_TESTS=ON \
         -DBUILD_ADDRESS_SANITIZER="${ADDRESS_SANITIZER}" \
         -DBUILD_CLIENTS_BENCHMARKS=ON \
         -DBUILD_CLIENTS_SAMPLES=ON \
         -DBUILD_TESTING=ON \
-        -DBUILD_CLIENTS_EXTRA_TESTS="${EXTRA_TESTS}" \
+        -DBUILD_CLIENTS_EXTRA_TESTS=ON \
         "$COMPONENT_SRC"
 
     cmake --build "$BUILD_DIR" -- -j${PROC}
@@ -68,7 +69,7 @@ clean_rocsolver() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rocsolver; build_wheel ;;
+    build) build_rocsolver ;;
     outdir) print_output_directory ;;
     clean) clean_rocsolver ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rocsparse.sh

@@ -1,20 +1,21 @@
-#!/bin/bash
+
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 PATH=${ROCM_PATH}/bin:$PATH
 set_component_src rocSPARSE
 
 build_rocsparse() {
     echo "Start build"
+
     cd $COMPONENT_SRC
+
     if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
-        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
+       set_asan_env_vars
+       set_address_sanitizer_on
     fi
+
     SHARED_LIBS="ON"
     if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
         SHARED_LIBS="OFF"
@@ -24,17 +25,19 @@ build_rocsparse() {
 
     mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
 
-    # 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}"
+    if [ -n "$GPU_ARCHS" ]; then
+        GPU_TARGETS="$GPU_ARCHS"
+    else
+        GPU_TARGETS="gfx900;gfx906:xnack-;gfx908:xnack-;gfx90a:xnack+;gfx90a:xnack-;gfx942;gfx1030;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
     fi
 
-    init_rocm_common_cmake_params
     ROCSPARSE_TEST_MIRROR=$MIRROR \
-    CXX=$(set_build_variables __CXX__)\
-    CC=$(set_build_variables __CC__)\
+    export CXX=$(set_build_variables CXX)\
+    export CC=$(set_build_variables CC)\
+
+    init_rocm_common_cmake_params
     cmake \
+        -DAMDGPU_TARGETS=${GPU_TARGETS} \
         ${LAUNCHER_FLAGS} \
         "${rocm_math_common_cmake_params[@]}" \
         -DBUILD_SHARED_LIBS=$SHARED_LIBS \
@@ -51,7 +54,7 @@ build_rocsparse() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
 
     show_build_cache_stats
 }
@@ -65,7 +68,7 @@ clean_rocsparse() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rocsparse; build_wheel ;;
+    build) build_rocsparse ;;
     outdir) print_output_directory ;;
     clean) clean_rocsparse ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rocthrust.sh

@@ -2,7 +2,7 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 build_release=true
 set_component_src rocThrust
@@ -19,17 +19,15 @@ build_rocthrust() {
     if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
          set_asan_env_vars
          set_address_sanitizer_on
-         # ASAN packaging is not required for rocThrust, 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
-    CXX=$(set_build_variables __CXX__)\
+
+    CXX=$(set_build_variables CXX)\
     cmake \
-        ${GEN_NINJA} \
         ${LAUNCHER_FLAGS} \
         "${rocm_math_common_cmake_params[@]}" \
         -DCMAKE_MODULE_PATH="${ROCM_PATH}/lib/cmake/hip;${ROCM_PATH}/hip/cmake" \
@@ -42,7 +40,8 @@ build_rocthrust() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+
 
     show_build_cache_stats
 }
@@ -57,7 +56,7 @@ stage2_command_args "$@"
 
 case $TARGET in
     (clean) clean_rocthrust ;;
-    (build) build_rocthrust; build_wheel ;;
+    (build) build_rocthrust ;;
    (outdir) print_output_directory ;;
         (*) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_roctracer.sh

@@ -12,22 +12,19 @@ printUsage() {
     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 roc-tracer.
-                                      It needs to be used along with -r option"
     echo "  -h,  --help               Prints this help"
     echo
 
     return 0
 }
 
-## Build environment variables
 API_NAME="roctracer"
 PROJ_NAME="$API_NAME"
 PACKAGE_ROOT="$(getPackageRoot)"
 PACKAGE_INCLUDE="$(getIncludePath)"
 BUILD_DIR="$(getBuildPath $API_NAME)"
-PACKAGE_DEB="$PACKAGE_ROOT/deb/$PROJ_NAME"
-PACKAGE_RPM="$PACKAGE_ROOT/rpm/$PROJ_NAME"
+PACKAGE_DEB="$(getPackageRoot)/deb/$API_NAME"
+PACKAGE_RPM="$(getPackageRoot)/rpm/$API_NAME"
 PACKAGE_PREFIX="$ROCM_INSTALL_PATH"
 
 export HIP_PATH="${ROCM_INSTALL_PATH}"
@@ -42,15 +39,12 @@ CLEAN_OR_OUT=0
 MAKETARGET="deb"
 PKGTYPE="deb"
 
-# Handling GPU Targets for HSACO and HIP Executables
-GPU_LIST="gfx900;gfx906;gfx908;gfx90a;gfx942;gfx1030;gfx1031;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
+GPU_LIST="gfx900;gfx906;gfx908;gfx90a;gfx940;gfx941;gfx942;gfx1030;gfx1031;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
 
-#parse the arguments
-VALID_STR=$(getopt -o hcraswo: --long help,clean,release,static,wheel,address_sanitizer,outdir: -- "$@")
+VALID_STR=$(getopt -o hcraso: --long help,clean,release,static,address_sanitizer,outdir: -- "$@")
 eval set -- "$VALID_STR"
 
 while true; do
-    #echo "parocessing $1"
     case "$1" in
     -h | --help)
         printUsage
@@ -72,9 +66,6 @@ while true; do
         ;;
     -s | --static)
         ack_and_skip_static
-        ;;
-    -w | --wheel)
-        WHEEL_PACKAGE=true
         shift
         ;;
     -o | --outdir)
@@ -87,7 +78,7 @@ while true; do
     --)
         shift
         break
-        ;; # end delimiter
+        ;;
     *)
         echo " This should never come but just incase : UNEXPECTED ERROR Parm : [$1] " >&2
         exit 20
@@ -112,12 +103,15 @@ clean() {
     rm -rf "$PACKAGE_ROOT/${PROJ_NAME}"
 }
 
+copy_libs_to_workspace() {
+    if [ "$ASAN_CMAKE_PARAMS" != "true" ]; then
+        cp -R "${ROCM_INSTALL_PATH}/lib/roctracer/" "${OUT_DIR}/lib/roctracer/"
+    fi
+}
+
 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)"
-    # Find HIP in the build tree
     export ROCM_PATH="${ROCM_INSTALL_PATH}"
     export HIPCC_COMPILE_FLAGS_APPEND="--rocm-path=$ROCM_PATH"
 
@@ -127,11 +121,11 @@ build() {
         print_lib_type $SHARED_LIBS
 
         export HIPCC_COMPILE_FLAGS_APPEND="--rocm-path=$ROCM_PATH --offload-arch=gfx900 --offload-arch=gfx906  --offload-arch=gfx908 \
-                                                            --offload-arch=gfx90a --offload-arch=gfx942 --offload-arch=gfx1030 --offload-arch=gfx1031 \
+                                                            --offload-arch=gfx90a --offload-arch=gfx940 --offload-arch=gfx941 \
+                                                            --offload-arch=gfx942 --offload-arch=gfx1030 --offload-arch=gfx1031 \
                                                             --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 \
                                                             --offload-arch=gfx1200 --offload-arch=gfx1201"
         cmake $(rocm_cmake_params) \
-            ${GEN_NINJA} \
             -DCPACK_PACKAGING_INSTALL_PREFIX="$PACKAGE_PREFIX/$PROJ_NAME" \
             -DCMAKE_MODULE_PATH="$PACKAGE_CMAKE/hip" \
             -DCMAKE_HIP_ARCHITECTURES=OFF \
@@ -180,7 +174,7 @@ verifyEnvSetup
 
 case $TARGET in
 clean) clean ;;
-build) build ; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
+build) build && copy_libs_to_workspace ;;
 outdir) print_output_directory ;;
 *) die "$BASH_SOURCE Invalid target $TARGET - exiting" ;;
 esac

tools/rocm-build/build_rocwmma.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src rocWMMA
 
@@ -12,7 +12,6 @@ build_rocwmma() {
         ack_and_skip_static
     fi
 
-    # To check if rocwmma source is present
     if [ ! -e $COMPONENT_SRC/CMakeLists.txt ]; then
         echo "Skipping rocWMMA as source is not available"
         mkdir -p $COMPONENT_SRC
@@ -22,14 +21,12 @@ build_rocwmma() {
     if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
          set_asan_env_vars
          set_address_sanitizer_on
-         # ASAN packaging is not required for rocWMMA, 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
-    CXX=$(set_build_variables __CXX__)\
+    CXX=$(set_build_variables CXX)\
     cmake \
         "${rocm_math_common_cmake_params[@]}" \
         ${LAUNCHER_FLAGS} \
@@ -44,7 +41,7 @@ build_rocwmma() {
     cmake --build "$BUILD_DIR" -- package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
     show_build_cache_stats
 }
 
@@ -57,7 +54,7 @@ clean_rocwmma() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rocwmma; build_wheel ;;
+    build) build_rocwmma ;;
     outdir) print_output_directory ;;
     clean) clean_rocwmma ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_rpp.sh

@@ -2,14 +2,14 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src rpp
 DEPS_DIR="$RPP_DEPS_LOCATION"
 
 LLVM_LIBDIR="${ROCM_PATH}/llvm/lib"
 ROCM_LLVM_LIB_RPATH="\$ORIGIN/llvm/lib"
-# RPP specific exe linker parameters
+
 rpp_specific_cmake_params() {
     local rpp_cmake_params
     if [ "${ASAN_CMAKE_PARAMS}" == "true" ] ; then
@@ -27,26 +27,21 @@ build_rpp() {
         ack_and_skip_static
     fi
 
-    # To check if RPP source is present
     if [ ! -e $COMPONENT_SRC/CMakeLists.txt ]; then
         echo "Skipping RPP build as source is not available"
         mkdir -p $COMPONENT_SRC
         exit 0
     fi
 
-    CXX=$(set_build_variables __AMD_CLANG_++__)
-    # Enable ASAN
     if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
         set_asan_env_vars
         set_address_sanitizer_on
     fi
 
-    echo "C compiler: $CC"
-    echo "CXX compiler: $CXX"
     mkdir -p $BUILD_DIR && cd $BUILD_DIR
+
     init_rocm_common_cmake_params
-    # rocm_common_cmake_params provides the default rpath for rocm executables and libraries
-    # Override cmake shared linker flags to add RPATH for boost libraries
+
     cmake \
         "${rocm_math_common_cmake_params[@]}" \
         ${LAUNCHER_FLAGS} \
@@ -62,7 +57,8 @@ build_rpp() {
     cpack -G ${PKGTYPE^^}
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR
+    cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
     show_build_cache_stats
 }
 
@@ -75,7 +71,7 @@ clean_rpp() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_rpp; build_wheel ;;
+    build) build_rpp ;;
     outdir) print_output_directory ;;
     clean) clean_rpp ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_transferbench.sh

@@ -2,12 +2,9 @@
 
 set -ex
 
-source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
 
 set_component_src TransferBench
-# ROCMOPS-8587 : this variable will be removing nic enablement from the compilation of 
-# the tool that will create complications for the general rocm release 
-export DISABLE_NIC_EXEC=1
 
 build_transferbench() {
     echo "Start build"
@@ -16,15 +13,18 @@ build_transferbench() {
         ack_and_skip_static
     fi
 
+    sed -i 's/^\(\s*set\s*(CMAKE_RUNTIME_OUTPUT_DIRECTORY.*\)$/#\1/'  "${COMPONENT_SRC}/CMakeLists.txt"
+
     mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR"
     init_rocm_common_cmake_params
 
-    CXX=$(set_build_variables __HIP_CC__) \
+    CXX="$ROCM_PATH"/bin/hipcc \
     cmake "${rocm_math_common_cmake_params[@]}" "$COMPONENT_SRC"
     make package
 
     rm -rf _CPack_Packages/ && find -name '*.o' -delete
-    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
+    mkdir -p $PACKAGE_DIR
+    cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
     show_build_cache_stats
 }
 
@@ -37,7 +37,7 @@ clean_transferbench() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_transferbench; build_wheel ;;
+    build) build_transferbench ;;
     outdir) print_output_directory ;;
     clean) clean_transferbench ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/compute_helper.sh

@@ -0,0 +1,323 @@
+#!/bin/bash
+
+# compute_helper.sh is created to add common functions similar to compute_utils.sh in stg1
+# stg2 build scripts can make use of common function
+# TODO:  Component build scripts can be optimized by adding common function in this file(ex for ASAN, Debugging  etc)
+#        All build scripts should use the common function.
+
+set -e
+set -o pipefail
+# Set the LLVM directory path with respect to ROCM_PATH
+# LLVM is installed in $ROCM_PATH/lib/llvm
+ROCM_LLVMDIR="lib/llvm"
+# Set ADDRESS_SANITIZER to OFF by default
+export ADDRESS_SANITIZER="OFF"
+
+DISABLE_DEBUG_PACKAGE="false"
+export CFLAGS="$CFLAGS $SET_DWARF_VERSION_4  "
+export CXXFLAGS="$CXXFLAGS $SET_DWARF_VERSION_4 "
+
+# Print message to stderr
+#   param message string to print on exit
+# Example: printErr "file not found"
+printErr() {
+    echo "$@" 1>&2
+}
+
+# Print message to stderr and terminate current program
+#   param message string to print on exit
+# Example: die "Your program" has terminated
+die() {
+    printErr "FATAL: $@"
+    exit 1
+}
+
+# Die if first argument is empty
+#   param string to validate
+#   param error message
+# Example: die "$VARIABLE" "Your program" has terminated
+dieIfEmpty() {
+    if [ "$1" == "" ]; then
+        shift
+        die "$@"
+    fi
+}
+
+ack_and_skip_static() {
+    echo "static build is not enabled for this component..skipping"
+    exit 0
+}
+
+# Copy a file or directory to target location and show single line progress
+progressCopy() {
+    if [ -d "$1" ]; then
+        rsync -a "$1"/* "$2"
+    else
+        rsync -a "$1" "$2"
+    fi
+}
+
+# Get OS identification string
+# xargs will remove the trailing whitespaces
+getOsVersion() {
+    lsb_release -d | cut -d: -f2 | xargs
+}
+
+# Get kernel identification string
+getKernelVersion() {
+    uname -r
+}
+
+# Trim excessive whitespace from a string
+strTrim() {
+    echo "$@" | xargs
+}
+
+# Return whether the booted OS is Fedora or not
+isFedora() {
+    grep -iq fedora /etc/os-release
+}
+
+# Return whether the booted system is EFI or not
+isEFI() {
+    [ -d "/sys/firmware/efi" ]
+}
+
+print_lib_type() {
+   if [ "$1" == "OFF" ];
+   then
+       echo " Building Archive "
+   else
+       echo " Building Shared Object "
+   fi
+}
+
+# Get CMAKE build flags for CMAKE build trigger
+set_build_variables() {
+    local cmake_cxx_flag_params
+    local cmake_cc_flag_params
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ] ; then
+        cmake_cxx_flag_params="$ROCM_PATH/llvm/bin/clang++"
+        cmake_cc_flag_params="$ROCM_PATH/llvm/bin/clang"
+        cmake_comp_params=(
+            "-DCMAKE_C_COMPILER=$ROCM_PATH/llvm/bin/clang"
+            "-DCMAKE_CXX_COMPILER=$ROCM_PATH/llvm/bin/clang++"
+        )
+    else
+        cmake_cxx_flag_params="$ROCM_PATH/bin/hipcc"
+        cmake_cc_flag_params="$ROCM_PATH/bin/hipcc"
+        cmake_comp_params=(
+            "-DCMAKE_C_COMPILER=$ROCM_PATH/bin/hipcc"
+            "-DCMAKE_CXX_COMPILER=$ROCM_PATH/bin/hipcc"
+        )
+    fi
+
+    case "$1" in
+    ("CXX")
+      printf "%s" "${cmake_cxx_flag_params}"
+      ;;
+     ("CC")
+      printf "%s" "${cmake_cc_flag_params}"
+      ;;
+     ("CMAKE_C_CXX")
+      printf '%s ' "${cmake_comp_params[@]}"
+      ;;
+     (*)
+      exit 1
+      ;;
+    esac
+    exit
+}
+
+# Get the directory containing the cmake config files
+getCmakePath() {
+    local rocmInstallPath=${ROCM_PATH}
+    local cmakePath="lib/cmake"
+    dieIfEmpty "$rocmInstallPath"
+    echo "$rocmInstallPath/$cmakePath"
+}
+
+getInstallLibDir() {
+    local libDir="lib"
+    if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ] ; then
+        libDir="lib/asan"
+    fi
+    echo "$libDir"
+}
+
+set_asan_env_vars() {
+    # Flag to set cmake build params for ASAN builds
+    ASAN_CMAKE_PARAMS="true"
+    export ADDRESS_SANITIZER="ON"
+    LLVM_BIN_DIR="${ROCM_PATH}/llvm/bin"
+    export CC="$LLVM_BIN_DIR/clang"
+    export CXX="$LLVM_BIN_DIR/clang++"
+    export FC="$LLVM_BIN_DIR/flang"
+    export PATH="$LLVM_BIN_DIR/:$PATH"
+    # get exact path to ASAN lib containing clang version
+    ASAN_LIB_PATH=$(clang --print-file-name=libclang_rt.asan-x86_64.so)
+    export LD_LIBRARY_PATH="${ASAN_LIB_PATH%/*}:${ROCM_PATH}/llvm/lib:${LD_LIBRARY_PATH:+:$LD_LIBRARY_PATH}"
+    export ASAN_OPTIONS="detect_leaks=0"
+}
+
+set_address_sanitizer_on() {
+    export CFLAGS="-fsanitize=address -shared-libasan $SET_DWARF_VERSION_4 -g -gz"
+    export CXXFLAGS="-fsanitize=address -shared-libasan $SET_DWARF_VERSION_4 -g -gz"
+    export LDFLAGS="-Wl,--enable-new-dtags -fuse-ld=lld  -fsanitize=address -shared-libasan -g -gz -Wl,--build-id=sha1 -L${ROCM_PATH}/llvm/lib -L${ROCM_PATH}/lib/asan"
+}
+
+rebuild_lapack() {
+    wget -nv -O lapack-3.9.1.tar.gz \
+        http://compute-artifactory.amd.com/artifactory/rocm-generic-thirdparty-deps/ubuntu/lapack-v3.9.1.tar.gz
+    sh -c "echo 'd0085d2caf997ff39299c05d4bacb6f3d27001d25a4cc613d48c1f352b73e7e0 *lapack-3.9.1.tar.gz' | sha256sum -c"
+    tar xzf lapack-3.9.1.tar.gz --one-top-level=lapack-src --strip-components 1
+    rm lapack-3.9.1.tar.gz
+
+    cmake -Slapack-src -Blapack-bld \
+        ${LAUNCHER_FLAGS} \
+        -DBUILD_TESTING=OFF \
+        -DCBLAS=ON \
+        -DLAPACKE=OFF
+    cmake --build lapack-bld -- -j${PROC}
+    cmake --build lapack-bld -- install
+    rm -r lapack-src lapack-bld
+}
+
+# debug function #dumping values in case of error to solve the same
+print_vars() {
+echo " Status of Vars in $1 build "
+echo " TARGET= $2 "
+echo " BUILD_TYPE = $3 "
+echo " SHARED_LIBS = $4 "
+echo " CLEAN_OR_OUT = $5 "
+echo " PKGTYPE= $6 "
+echo " MAKETARGET = $7 "
+}
+
+rocm_math_common_cmake_params=()
+init_rocm_common_cmake_params(){
+  local retCmakeParams=${1:-rocm_math_common_cmake_params}
+  local SET_BUILD_TYPE=${BUILD_TYPE:-'RelWithDebInfo'}
+  local ASAN_LIBDIR="lib/asan"
+  local CMAKE_PATH=$(getCmakePath)
+# Common cmake parameters can be set
+# component build scripts can use this function
+  local cmake_params
+  if [ "${ASAN_CMAKE_PARAMS}" == "true" ] ; then
+    cmake_params=(
+        "-DCMAKE_PREFIX_PATH=$CMAKE_PATH;${ROCM_PATH}/$ASAN_LIBDIR;$ROCM_PATH/llvm;$ROCM_PATH"
+        "-DCMAKE_SHARED_LINKER_FLAGS_INIT=-Wl,--enable-new-dtags,--build-id=sha1,--rpath,$ROCM_ASAN_LIB_RPATH"
+        "-DCMAKE_EXE_LINKER_FLAGS_INIT=-Wl,--enable-new-dtags,--build-id=sha1,--rpath,$ROCM_ASAN_EXE_RPATH"
+        "-DENABLE_ASAN_PACKAGING=true"
+    )
+  else
+    cmake_params=(
+        "-DCMAKE_PREFIX_PATH=${ROCM_PATH}/llvm;${ROCM_PATH}"
+        "-DCMAKE_SHARED_LINKER_FLAGS_INIT=-Wl,--enable-new-dtags,--build-id=sha1,--rpath,$ROCM_LIB_RPATH"
+        "-DCMAKE_EXE_LINKER_FLAGS_INIT=-Wl,--enable-new-dtags,--build-id=sha1,--rpath,$ROCM_EXE_RPATH"
+    )
+  fi
+
+  cmake_params+=(
+      "-DCMAKE_VERBOSE_MAKEFILE=1"
+      "-DCMAKE_BUILD_TYPE=${SET_BUILD_TYPE}"
+      "-DCMAKE_INSTALL_RPATH_USE_LINK_PATH=FALSE"
+      "-DCMAKE_INSTALL_PREFIX=${ROCM_PATH}"
+      "-DCMAKE_PACKAGING_INSTALL_PREFIX=${ROCM_PATH}"
+      "-DBUILD_FILE_REORG_BACKWARD_COMPATIBILITY=OFF"
+      "-DROCM_SYMLINK_LIBS=OFF"
+      "-DCPACK_PACKAGING_INSTALL_PREFIX=${ROCM_PATH}"
+      "-DROCM_DISABLE_LDCONFIG=ON"
+      "-DCPACK_SET_DESTDIR=OFF"
+      "-DCPACK_RPM_PACKAGE_RELOCATABLE=ON"
+      "-DROCM_PATH=${ROCM_PATH}"
+  )
+
+  #TODO :remove if clause once debug related issues are fixed
+  if [ "${DISABLE_DEBUG_PACKAGE}" == "true" ] ; then
+    SET_BUILD_TYPE=${BUILD_TYPE:-'Release'}
+    cmake_params+=(
+        "-DCPACK_DEBIAN_DEBUGINFO_PACKAGE=FALSE"
+        "-DCPACK_RPM_DEBUGINFO_PACKAGE=FALSE"
+        "-DCPACK_RPM_INSTALL_WITH_EXEC=FALSE"
+        "-DCMAKE_BUILD_TYPE=${SET_BUILD_TYPE}"
+    )
+  elif [ "$SET_BUILD_TYPE" == "RelWithDebInfo" ] || [ "$SET_BUILD_TYPE" == "Debug" ]; then
+    # RelWithDebinfo optimization level -O2 is having performance impact
+    # So overriding the same to -O3
+    cmake_params+=(
+        "-DCPACK_DEBIAN_DEBUGINFO_PACKAGE=TRUE"
+        "-DCPACK_RPM_DEBUGINFO_PACKAGE=TRUE"
+        "-DCPACK_RPM_INSTALL_WITH_EXEC=TRUE"
+        "-DCMAKE_CXX_FLAGS_RELWITHDEBINFO=-O3 -g -DNDEBUG"
+    )
+  fi
+    eval "${retCmakeParams}=( \"\${cmake_params[@]}\" ) "
+}
+
+disable_debug_package_generation(){
+    case "$DISTRO_ID" in
+       (sles*|rhel*)
+           export CFLAGS=${CFLAGS//"$SET_DWARF_VERSION_4"}
+           export CXXFLAGS=${CXXFLAGS//"$SET_DWARF_VERSION_4"}
+           ;;
+      (*)
+           ;;
+    esac
+    DISABLE_DEBUG_PACKAGE="true"
+}
+
+# Setup a number of variables to specify where to find the source
+# where to do the build and where to put the packages
+# Note the PACKAGE_DIR downcases the package name
+# This could be extended to do different things based on $1
+set_component_src(){
+    COMPONENT_SRC="$LIBS_WORK_DIR/$1"
+
+    BUILD_DIR="$OUT_DIR/build/$1"
+    DEB_PATH="$OUT_DIR/${PKGTYPE}/${1,,}"
+    RPM_PATH="$OUT_DIR/${PKGTYPE}/${1,,}"
+    PACKAGE_DIR="$OUT_DIR/${PKGTYPE}/${1,,}"
+}
+
+# Standard definition of function to print the package location.  If
+# for some reason a custom version is needed then it can overwrite
+# this definition
+# TODO: Don't use a global PKGTYPE, pass the value in as a parameter
+print_output_directory() {
+    case ${1:-$PKGTYPE} in
+        ("deb")
+            echo ${DEB_PATH};;
+        ("rpm")
+            echo ${RPM_PATH};;
+        (*)
+            echo "Invalid package type \"${PKGTYPE}\" provided for -o" >&2; exit 1;;
+    esac
+    exit
+}
+
+# Standard argument processing
+# Here to avoid repetition
+TARGET="build"			# default target
+stage2_command_args(){
+    while [ "$1" != "" ];
+    do
+	case $1 in
+            -o  | --outdir )
+		shift 1; PKGTYPE=$1 ; TARGET="outdir" ;;
+            -c  | --clean )
+		TARGET="clean" ;;
+            *)
+		break ;;
+	esac
+	shift 1
+    done
+}
+
+show_build_cache_stats(){
+    if [ "$CCACHE_ENABLED" = "true" ] ; then
+	if ! ccache -s ; then
+	    echo "Unable to display ccache stats"
+	fi
+    fi
+}