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

Co-authored-by: Jeff Daily <jeff.daily@amd.com>

.azuredevops/components/AMDMIGraphX.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -27,12 +32,12 @@ parameters:
   type: object
   default:
     - https://github.com/RadeonOpenCompute/rbuild/archive/master.tar.gz
-    - onnx>=1.14.1
-    - numpy>=1.21.6
-    - typing>=3.7.4
-    - pytest>=6.0.1
-    - packaging>=23.0
-    - protobuf>=3.20.2
+    - "onnx>=1.14.1"
+    - "numpy>=1.21.6"
+    - "typing>=3.7.4"
+    - "pytest>=6.0.1"
+    - "packaging>=23.0"
+    - "protobuf>=3.20.2"
 - name: rocmDependencies
   type: object
   default:
@@ -115,6 +120,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -145,7 +151,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/HIP.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -42,7 +47,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -67,6 +72,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependenciesAMD }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
 # compile clr
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
@@ -99,7 +105,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -125,6 +131,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependenciesNvidia }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - script: 'ls -1R $(Agent.BuildDirectory)/rocm'
     displayName: 'Artifact listing'
 # compile clr

.azuredevops/components/HIPIFY.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:

.azuredevops/components/MIOpen.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -77,7 +82,7 @@ jobs:
     - template: /.azuredevops/variables-global.yml
     - name: ROCM_PATH
       value: $(Agent.BuildDirectory)/rocm
-    pool: ${{ variables.MEDIUM_BUILD_POOL }}
+    pool: ${{ variables.HIGH_BUILD_POOL }}
     workspace:
       clean: all
     steps:
@@ -97,6 +102,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - task: Bash@3
       displayName: Build and install other dependencies
       inputs:
@@ -107,7 +113,7 @@ jobs:
           mkdir -p $(Agent.BuildDirectory)/miopen-deps
           export CXX=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang++
           export CC=$(Agent.BuildDirectory)/rocm/llvm/bin/amdclang
-          cmake -P install_deps.cmake --prefix $(Agent.BuildDirectory)/miopen-deps
+          cmake -P install_deps.cmake --prefix $(Agent.BuildDirectory)/miopen-deps --generator Ninja
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -142,7 +148,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/MIVisionX.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -103,6 +108,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -129,7 +135,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/ROCR-Runtime.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -42,7 +47,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -58,6 +63,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
@@ -78,7 +84,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/ROCdbgapi.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -24,7 +29,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -40,6 +45,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-

.azuredevops/components/ROCgdb.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 # reference: https://github.com/ROCm/ROCgdb/blob/amd-staging/README-ROCM.md
 - name: aptPackages
   type: object
@@ -46,7 +51,8 @@ jobs:
     condition:
       and(
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
@@ -69,6 +75,7 @@ jobs:
       parameters:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-autotools.yml
       parameters:
         configureFlags: >-

.azuredevops/components/ROCmValidationSuite.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -96,6 +101,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -126,7 +132,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/Tensile.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -46,7 +51,7 @@ jobs:
   - template: /.azuredevops/variables-global.yml
   - name: ROCM_PATH
     value: $(Agent.BuildDirectory)/rocm
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -63,6 +68,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
   - task: Bash@3
     displayName: Create wheel file
@@ -104,7 +110,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/TransferBench.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -67,6 +72,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -92,7 +98,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/amdsmi.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -58,7 +63,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/aomp-extras.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -23,7 +28,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -39,6 +44,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       componentName: aomp-extras

.azuredevops/components/aomp.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 # reference:
 # https://github.com/ROCm/aomp/blob/aomp-dev/docs/SOURCEINSTALL_PREREQUISITE.md
 - name: aptPackages
@@ -108,6 +113,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       componentName: extras
@@ -176,7 +182,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/composable_kernel.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -71,6 +76,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - script: |
         mkdir -p $(CCACHE_DIR)
         echo "##vso[task.prependpath]/usr/lib/ccache"
@@ -117,7 +123,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/half.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -25,7 +30,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -41,6 +46,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-

.azuredevops/components/hip-tests.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -75,6 +80,7 @@ jobs:
       parameters:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     # compile hip-tests
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
@@ -109,7 +115,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/hipBLAS-common.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -29,7 +34,7 @@ jobs:
   - name: ROCM_PATH
     value: $(Agent.BuildDirectory)/rocm
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -45,6 +50,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-

.azuredevops/components/hipBLAS.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -89,6 +94,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -121,13 +127,12 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    - name: AMD_COMGR_CACHE
-      value: 0
     pool: ${{ job.target }}_test_pool
     workspace:
       clean: all

.azuredevops/components/hipBLASLt.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -22,7 +27,7 @@ parameters:
   type: object
   default:
     - joblib
-    - packaging>=22.0
+    - "packaging>=22.0"
     - --upgrade
 - name: rocmDependencies
   type: object
@@ -98,6 +103,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - task: Bash@3
       displayName: Add ROCm binaries to PATH
       inputs:
@@ -179,15 +185,14 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
     - name: ROCM_PATH
       value: $(Agent.BuildDirectory)/rocm
-    - name: AMD_COMGR_CACHE
-      value: 0
     pool: ${{ job.target }}_test_pool
     workspace:
       clean: all

.azuredevops/components/hipCUB.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -67,6 +72,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -94,7 +100,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/hipFFT.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -79,6 +84,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -112,7 +118,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/hipRAND.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -70,6 +75,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -101,7 +107,8 @@ jobs:
     condition:
         and(succeeded(),
           eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-          not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+          not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+          eq(${{ parameters.aggregatePipeline }}, False)
         )
     variables:
     - group: common

.azuredevops/components/hipSOLVER.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -82,6 +87,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
   # build external gtest and lapack
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
@@ -122,13 +128,12 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    - name: AMD_COMGR_CACHE
-      value: 0
     pool: ${{ job.target }}_test_pool
     workspace:
       clean: all

.azuredevops/components/hipSPARSE.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -77,6 +82,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -116,13 +122,12 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    - name: AMD_COMGR_CACHE
-      value: 0
     pool: ${{ job.target }}_test_pool
     workspace:
       clean: all

.azuredevops/components/hipSPARSELt.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -91,6 +96,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
   # Build and install gtest and lapack
   # $(Pipeline.Workspace)/deps is a temporary folder for the build process
   # $(Pipeline.Workspace)/s/deps is part of the hipSPARSELt repo
@@ -150,7 +156,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/hipTensor.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -66,6 +71,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -95,13 +101,12 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    - name: AMD_COMGR_CACHE
-      value: 0
     pool: ${{ job.target }}_test_pool
     workspace:
       clean: all

.azuredevops/components/hipfort.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -76,6 +81,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -111,7 +117,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/llvm-project.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -45,6 +50,7 @@ jobs:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
       skipLlvmSymlink: true
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       componentName: rocm-llvm

.azuredevops/components/rccl.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -72,7 +77,7 @@ jobs:
     - template: /.azuredevops/variables-global.yml
     - name: HIP_ROCCLR_HOME
       value: $(Build.BinariesDirectory)/rocm
-    pool: ${{ variables.MEDIUM_BUILD_POOL }}
+    pool: ${{ variables.HIGH_BUILD_POOL }}
     workspace:
       clean: all
     steps:
@@ -90,6 +95,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -124,7 +130,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rdc.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -89,6 +94,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
   # Build grpc
     - task: Bash@3
       displayName: 'git clone grpc'
@@ -135,7 +141,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocAL.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -151,6 +156,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -182,7 +188,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocALUTION.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -85,6 +90,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -117,7 +123,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocBLAS.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -101,6 +106,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -145,13 +151,12 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    - name: AMD_COMGR_CACHE
-      value: 0
     pool: ${{ job.target }}_test_pool
     workspace:
       clean: all

.azuredevops/components/rocDecode.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -73,6 +78,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
@@ -92,7 +98,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocFFT.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -79,6 +84,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -113,7 +119,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocJPEG.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -75,6 +80,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -83,6 +89,8 @@ jobs:
           -GNinja
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
+      parameters:
+        gpuTarget: ${{ job.target }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-links.yml
     # - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
     #   parameters:
@@ -96,7 +104,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
@@ -115,6 +124,8 @@ jobs:
         registerROCmPackages: true
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/preamble.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/local-artifact-download.yml
+      parameters:
+        gpuTarget: ${{ job.target }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-aqlprofile.yml
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/dependencies-rocm.yml
       parameters:
@@ -140,4 +151,3 @@ jobs:
         environment: test
         gpuTarget: ${{ job.target }}
         registerROCmPackages: true
-        optSymLink: true

.azuredevops/components/rocMLIR.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -62,6 +67,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
@@ -87,7 +93,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocPRIM.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -66,6 +71,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -93,7 +99,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocPyDecode.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -77,6 +82,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - task: Bash@3
       displayName: 'Save Python Package Paths'
       inputs:
@@ -152,7 +158,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocRAND.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -70,6 +75,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -98,7 +104,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocSOLVER.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -88,6 +93,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         componentName: lapack
@@ -131,13 +137,12 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    - name: AMD_COMGR_CACHE
-      value: 0
     pool: ${{ job.target }}_test_pool
     workspace:
       clean: all

.azuredevops/components/rocSPARSE.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -29,6 +34,7 @@ parameters:
     - rocminfo
     - rocPRIM
     - rocprofiler-register
+    - roctracer
 - name: rocmTestDependencies
   type: object
   default:
@@ -80,6 +86,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -126,13 +133,12 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    - name: AMD_COMGR_CACHE
-      value: 0
     pool: ${{ job.target }}_test_pool
     workspace:
       clean: all

.azuredevops/components/rocThrust.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -71,6 +76,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -98,7 +104,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocWMMA.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -65,7 +70,7 @@ jobs:
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    pool: ${{ variables.MEDIUM_BUILD_POOL }}
+    pool: ${{ variables.HIGH_BUILD_POOL }}
     workspace:
       clean: all
     steps:
@@ -81,6 +86,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -113,13 +119,12 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
     - template: /.azuredevops/variables-global.yml
-    - name: AMD_COMGR_CACHE
-      value: 0
     pool: ${{ job.target }}_test_pool
     workspace:
       clean: all

.azuredevops/components/rocm-cmake.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -26,7 +31,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all

.azuredevops/components/rocm-core.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -17,7 +22,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all

.azuredevops/components/rocm-examples.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -100,6 +105,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         # https://github.com/ROCm/HIP/issues/2203
@@ -136,7 +142,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocm_bandwidth_test.yml

@@ -5,6 +5,12 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
+
 - name: aptPackages
   type: object
   default:
@@ -49,7 +55,7 @@ jobs:
     value: $(Agent.BuildDirectory)/rocm
   - name: ROCR_LIB_DIR
     value: $(Agent.BuildDirectory)/rocm
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -66,6 +72,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
@@ -90,7 +97,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocm_smi_lib.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -28,7 +33,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -59,7 +64,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocminfo.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -38,7 +43,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -55,6 +60,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
       skipLlvmSymlink: true
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
@@ -72,7 +78,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocprofiler-compute.yml

@@ -5,10 +5,17 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
     - cmake
+    - libdw-dev
+    - libtbb-dev
     - locales
     - ninja-build
     - python3-pip
@@ -17,10 +24,10 @@ parameters:
   default:
     - astunparse==1.6.2
     - colorlover
-    - dash>=1.12.0
+    - "dash>=1.12.0"
     - matplotlib
-    - numpy>=1.17.5
-    - pandas>=1.4.3
+    - "numpy>=1.17.5"
+    - "pandas>=1.4.3"
     - pymongo
     - pyyaml
     - tabulate
@@ -114,6 +121,7 @@ jobs:
         dependencyList: ${{ parameters.rocmDependencies }}
         dependencySource: ${{ job.dependencySource }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -140,7 +148,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common
@@ -182,12 +191,9 @@ jobs:
       displayName: Add ROCm binaries to PATH
       inputs:
         targetType: inline
-        script: echo "##vso[task.prependpath]$(Agent.BuildDirectory)/rocm/bin"
-    - task: Bash@3
-      displayName: Add ROCm compilers to PATH
-      inputs:
-        targetType: inline
-        script: echo "##vso[task.prependpath]$(Agent.BuildDirectory)/rocm/llvm/bin"
+        script: |
+          echo "##vso[task.prependpath]$(Agent.BuildDirectory)/rocm/bin"
+          echo "##vso[task.prependpath]$(Agent.BuildDirectory)/rocm/llvm/bin"
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -206,18 +212,6 @@ jobs:
         componentName: rocprofiler-compute
         testDir: $(Build.BinariesDirectory)/libexec/rocprofiler-compute
         testExecutable: ROCM_PATH=$(Agent.BuildDirectory)/rocm ctest
-    - task: Bash@3
-      displayName: Remove ROCm binaries from PATH
-      condition: always()
-      inputs:
-        targetType: inline
-        script: echo "##vso[task.setvariable variable=PATH]$(echo $PATH | sed -e 's;:$(Agent.BuildDirectory)/rocm/bin;;' -e 's;^/;;' -e 's;/$;;')"
-    - task: Bash@3
-      displayName: Remove ROCm compilers from PATH
-      condition: always()
-      inputs:
-        targetType: inline
-        script: echo "##vso[task.setvariable variable=PATH]$(echo $PATH | sed -e 's;:$(Agent.BuildDirectory)/rocm/llvm/bin;;' -e 's;^/;;' -e 's;/$;;')"
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/docker-container.yml
       parameters:
         aptPackages: ${{ parameters.aptPackages }}

.azuredevops/components/rocprofiler-register.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -17,7 +22,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -32,20 +37,14 @@ jobs:
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       componentName: rocprofiler-register
-      extraBuildFlags: >-
-        -GNinja
-  - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
-    parameters:
-      componentName: rocprofiler-register-tests
       extraBuildFlags: >-
         -DCMAKE_PREFIX_PATH=$(Build.BinariesDirectory)
+        -DROCPROFILER_REGISTER_BUILD_TESTS=ON
+        -DROCPROFILER_REGISTER_BUILD_SAMPLES=ON
         -GNinja
-      cmakeBuildDir: 'tests/build'
-      installEnabled: false
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/test.yml
     parameters:
       componentName: rocprofiler-register
-      testDir: 'tests/build'
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-upload.yml
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/artifact-links.yml

.azuredevops/components/rocprofiler-sdk.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -31,7 +36,7 @@ parameters:
     - pandas
     - perfetto
     - pycobertura
-    - pytest>=6.2.5
+    - "pytest>=6.2.5"
     - pyyaml
 - name: rocmDependencies
   type: object
@@ -89,6 +94,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - task: Bash@3
       displayName: Add Python site-packages binaries to path
       inputs:
@@ -125,7 +131,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocprofiler-systems.yml

@@ -6,6 +6,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -16,6 +21,7 @@ parameters:
     - bzip2
     - clang
     - cmake
+    - chrpath
     - environment-modules
     - ffmpeg
     - g++-12
@@ -103,6 +109,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - task: Bash@3
       displayName: Add ROCm binaries to PATH
       inputs:
@@ -124,6 +131,7 @@ jobs:
           -DDYNINST_BUILD_BOOST=ON
           -DROCPROFSYS_USE_PAPI=ON
           -DROCPROFSYS_USE_MPI=ON
+          -DCMAKE_CXX_FLAGS=-I$(Agent.BuildDirectory)/rocm/include/rocjpeg
           -DGPU_TARGETS=${{ job.target }}
           -GNinja
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/manifest.yml
@@ -147,7 +155,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     timeoutInMinutes: 180
     variables:
@@ -197,6 +206,7 @@ jobs:
           -DDYNINST_BUILD_BOOST=ON
           -DROCPROFSYS_USE_PAPI=ON
           -DROCPROFSYS_USE_MPI=ON
+          -DCMAKE_CXX_FLAGS=-I$(Agent.BuildDirectory)/rocm/include/rocjpeg
           -DGPU_TARGETS=${{ job.target }}
           -GNinja
     - task: Bash@3

.azuredevops/components/rocprofiler.yml

@@ -5,6 +5,12 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
+
 - name: aptPackages
   type: object
   default:
@@ -83,6 +89,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -115,7 +122,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rocr_debug_agent.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -49,7 +54,7 @@ jobs:
   variables:
   - group: common
   - template: /.azuredevops/variables-global.yml
-  pool: 
+  pool:
     vmImage: ${{ variables.BASE_BUILD_POOL }}
   workspace:
     clean: all
@@ -65,6 +70,7 @@ jobs:
     parameters:
       checkoutRef: ${{ parameters.checkoutRef }}
       dependencyList: ${{ parameters.rocmDependencies }}
+      aggregatePipeline: ${{ parameters.aggregatePipeline }}
   - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
     parameters:
       extraBuildFlags: >-
@@ -86,7 +92,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/roctracer.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -77,6 +82,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -107,7 +113,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/components/rpp.yml

@@ -5,6 +5,11 @@ parameters:
 - name: checkoutRef
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 - name: aptPackages
   type: object
   default:
@@ -84,6 +89,7 @@ jobs:
         checkoutRef: ${{ parameters.checkoutRef }}
         dependencyList: ${{ parameters.rocmDependencies }}
         gpuTarget: ${{ job.target }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
     - template: ${{ variables.CI_TEMPLATE_PATH }}/steps/build-cmake.yml
       parameters:
         extraBuildFlags: >-
@@ -113,7 +119,8 @@ jobs:
     condition:
       and(succeeded(),
         eq(variables['ENABLE_${{ upper(job.target) }}_TESTS'], 'true'),
-        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName']))
+        not(containsValue(split(variables['DISABLED_${{ upper(job.target) }}_TESTS'], ','), variables['Build.DefinitionName'])),
+        eq(${{ parameters.aggregatePipeline }}, False)
       )
     variables:
     - group: common

.azuredevops/nightly/pytorch.yml

@@ -68,7 +68,7 @@ parameters:
   default:
     - cmake
     - astunparse
-    - expecttest>=0.2.1
+    - "expecttest>=0.2.1"
     - hypothesis
     - numpy
     - psutil
@@ -76,8 +76,8 @@ parameters:
     - requests
     - setuptools==75.8.0
     - types-dataclasses
-    - typing-extensions>=4.8.0
-    - sympy>=1.13.0
+    - "typing-extensions>=4.8.0"
+    - "sympy>=1.13.0"
     - filelock
     - networkx
     - jinja2
@@ -85,8 +85,8 @@ parameters:
     - lintrunner
     - ninja
     - packaging
-    - optree>=0.13.0
-    - click>=8.0.3
+    - "optree>=0.13.0"
+    - "click>=8.0.3"
   # list for vision
     - auditwheel
     - future

.azuredevops/scripts/download-manifest-artifacts/download-manifest-artifacts.py

@@ -0,0 +1,68 @@
+#!/usr/bin/env python3
+
+import json
+import requests
+import argparse
+from pathlib import Path
+
+def get_builds(entries, gpu_target, output):
+    already_downloaded = {}
+    for entry in entries:
+        already_downloaded = _get_builds(entry, gpu_target, already_downloaded, output)
+
+def _get_builds(entry, gpu_target, already_downloaded, output):
+    print()
+    print(f"{entry['buildNumber']} - {entry['buildId']} - {entry['repoName']}")
+    if already_downloaded.get(entry['buildId']):
+        print('Skipping, already downloaded from build ' + entry['buildId'])
+        return already_downloaded
+
+    artifacts_url = f"https://dev.azure.com/ROCm-CI/ROCm-CI/_apis/build/builds/{entry['buildId']}/artifacts?api-version=7.1"
+    artifacts = requests.get(artifacts_url).json()
+    for artifact in artifacts['value']:
+        if 'gfx' in artifact['name'] and gpu_target not in artifact['name']:
+            continue
+
+        print('Artifact name: ' + artifact['name'])
+        print('File size: ~' +
+              str(round(int(artifact['resource']['properties']['artifactsize'])/1000000, 2)) + ' MB')
+        download_url = f"{artifact['resource']['downloadUrl']}"
+        download = requests.get(download_url)
+
+        zip_file = Path(output) / f"{artifact['name']}.zip"
+        with open(zip_file, 'wb') as f:
+            f.write(download.content)
+        already_downloaded[entry['buildId']] = True
+
+    return already_downloaded
+
+def main():
+    parser = argparse.ArgumentParser(description="Command line tool for downloading external ci artifacts")
+    parser.add_argument('--target', type=str, dest="target", choices=["gfx90a", "gfx942"], help="Target gfx")
+    parser.add_argument('--manifest', type=str, dest="manifest", help='JSON manifest url or path to local manifest')
+    parser.add_argument('--output_dir', type=str, dest="output", help='Path to download directory')
+    args = parser.parse_args()
+
+    manifest = args.manifest
+    gpu_target = args.target
+
+    if not gpu_target:
+        print("Enter the GPU target (gfx942, gfx90a)")
+        gpu_target = input()
+
+    if not manifest:
+        print("Enter the manifest file (URL or local path)")
+        manifest = input()
+
+    if 'http' in manifest:
+        data = requests.get(manifest).json()
+    else:
+        with open(manifest, 'r') as f:
+            data = json.load(f)
+
+    entries = [e for e in data['current']]
+    entries.extend([e for e in data['dependencies']])
+    get_builds(entries, gpu_target, args.output)
+
+if __name__ == "__main__":
+    main()

.azuredevops/scripts/download-manifest-artifacts/requirements.txt

@@ -0,0 +1 @@
+requests

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

@@ -12,6 +12,11 @@ parameters:
 - name: fileFilter
   type: string
   default: ''
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 
 steps:
 - task: Bash@3
@@ -27,17 +32,23 @@ steps:
 - task: DownloadPipelineArtifact@2
   displayName: Download ${{ parameters.componentName }}
   inputs:
-    buildType: 'specific'
-    project: ROCm-CI
-    definition: ${{ parameters.pipelineId }}
-    specificBuildWithTriggering: true
-    itemPattern: '**/*${{ parameters.fileFilter }}*'
-    # aomp is a special case, since the trigger file is under ROCm/ROCm instead of the component repo
-    ${{ if notIn(parameters.componentName, 'aomp') }}:
-      buildVersionToDownload: latestFromBranch # default is 'latest'
-    branchName: refs/heads/${{ parameters.branchName }}
-    allowPartiallySucceededBuilds: $(allowPartiallySucceededBuilds)
-    targetPath: '$(Pipeline.Workspace)/d'
+    ${{ if eq(parameters.aggregatePipeline, false) }}:
+      buildType: 'specific'
+      project: ROCm-CI
+      definition: ${{ parameters.pipelineId }}
+      specificBuildWithTriggering: true
+      itemPattern: '**/*${{ parameters.fileFilter }}*'
+      # aomp is a special case, since the trigger file is under ROCm/ROCm instead of the component repo
+      ${{ if notIn(parameters.componentName, 'aomp') }}:
+        buildVersionToDownload: latestFromBranch # default is 'latest'
+      branchName: refs/heads/${{ parameters.branchName }}
+      allowPartiallySucceededBuilds: $(allowPartiallySucceededBuilds)
+      targetPath: '$(Pipeline.Workspace)/d'
+    ${{ else }}:
+      buildType: 'current'
+      itemPattern: '**/${{ parameters.componentName }}*${{ parameters.fileFilter }}*'
+      allowPartiallySucceededBuilds: $(allowPartiallySucceededBuilds)
+      targetPath: '$(Pipeline.Workspace)/d'
 - task: ExtractFiles@1
   displayName: Extract ${{ parameters.componentName }}
   inputs:

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

@@ -31,6 +31,11 @@ parameters:
 - name: setupHIPLibrarySymlinks
   type: boolean
   default: false
+# set to true if doing full build of ROCm stack
+# and dependencies are pulled from same pipeline
+- name: aggregatePipeline
+  type: boolean
+  default: false
 
 - name: componentVarList
   type: object
@@ -354,6 +359,7 @@ steps:
       parameters:
         componentName: ${{ split(dependency, ':')[0] }}
         pipelineId: ${{ parameters.componentVarList[split(dependency, ':')[0]].pipelineId }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
         ${{ if parameters.componentVarList[split(dependency, ':')[0]].hasGpuTarget }}:
           fileFilter: "${{ split(dependency, ':')[1] }}*${{ parameters.gpuTarget }}"
         # dependencySource = staging
@@ -383,6 +389,7 @@ steps:
       parameters:
         componentName: ${{ dependency }}
         pipelineId: ${{ parameters.componentVarList[dependency].pipelineId }}
+        aggregatePipeline: ${{ parameters.aggregatePipeline }}
         ${{ if parameters.componentVarList[dependency].hasGpuTarget }}:
           fileFilter: ${{ parameters.gpuTarget }}
         # dependencySource = staging

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

@@ -0,0 +1,82 @@
+# This template creates and uploads a Docker image from the current environment
+# It uses `docker commit` to do so, which can provide more accurate images than the standard template
+# It requires the following conditions:
+# - Job must be run inside a Docker container
+# - The container's external name must be the same as the container's internal hostname
+# - Docker must be installed inside said container and given sufficient permissions
+# Currently, it is only usable for test jobs run on our self-managed systems
+# Jobs run on Azure VMs will not be able to use this template (most if not all build jobs)
+
+parameters:
+- name: gpuTarget
+  type: string
+  default: ''
+- name: environment
+  type: string
+  default: build
+  values:
+    - build
+    - test
+    - combined
+    - amd
+    - nvidia
+- name: extraPaths
+  type: string
+  default: ''
+- name: extraEnvVars
+  type: object
+  default: []
+- name: forceDockerCreation
+  type: boolean
+  default: false
+
+steps:
+  - task: Bash@3
+    displayName: Commit container and initialize Dockerfile
+    condition: or(and(failed(), not(contains(variables['DOCKER_SKIP_GFX'], variables['JOB_GPU_TARGET']))), ${{ eq(parameters.forceDockerCreation, true) }})
+    inputs:
+      workingDirectory: $(Pipeline.Workspace)
+      targetType: inline
+      script: |
+        docker commit $(hostname) az-ci-temp-image:latest
+        echo "FROM az-ci-temp-image:latest" > Dockerfile
+        echo "RUN sudo groupmod -g $(getent group render | awk -F: '{print $3}') render" >> Dockerfile
+        echo "RUN sudo groupmod -g $(getent group docker | awk -F: '{print $3}') docker" >> Dockerfile
+        echo "ENV PATH='$PATH:${{ parameters.extraPaths }}'" >> Dockerfile
+        echo "ENTRYPOINT [\"/bin/bash\"]" >> Dockerfile
+  - ${{ each extraEnvVar in parameters.extraEnvVars }}:
+    - task: Bash@3
+      displayName: Add extra environment variables
+      condition: or(and(failed(), not(contains(variables['DOCKER_SKIP_GFX'], variables['JOB_GPU_TARGET']))), ${{ eq(parameters.forceDockerCreation, true) }})
+      inputs:
+        workingDirectory: $(Pipeline.Workspace)
+        targetType: inline
+        script: echo "ENV ${{ split(extraEnvVar, ':::')[0] }}='${{ split(extraEnvVar, ':::')[1] }}'" >> Dockerfile
+  - task: Bash@3
+    displayName: Print Dockerfile
+    condition: or(and(failed(), not(contains(variables['DOCKER_SKIP_GFX'], variables['JOB_GPU_TARGET']))), ${{ eq(parameters.forceDockerCreation, true) }})
+    inputs:
+      workingDirectory: $(Pipeline.Workspace)
+      targetType: inline
+      script: cat Dockerfile
+  - task: Docker@2
+    displayName: Build and upload Docker image
+    condition: or(and(failed(), not(contains(variables['DOCKER_SKIP_GFX'], variables['JOB_GPU_TARGET']))), ${{ eq(parameters.forceDockerCreation, true) }})
+    inputs:
+      containerRegistry: 'ContainerService3'
+      ${{ if ne(parameters.gpuTarget, '') }}:
+        repository: '$(Build.DefinitionName)-${{ parameters.environment }}-${{ parameters.gpuTarget }}'
+      ${{ else }}:
+        repository: '$(Build.DefinitionName)-${{ parameters.environment }}'
+      Dockerfile: '$(Pipeline.Workspace)/Dockerfile'
+      buildContext: '$(Pipeline.Workspace)'
+  - task: Bash@3
+    condition: or(and(failed(), not(contains(variables['DOCKER_SKIP_GFX'], variables['JOB_GPU_TARGET']))), ${{ eq(parameters.forceDockerCreation, true) }})
+    displayName: "!! Docker Image URL !!"
+    inputs:
+      workingDirectory: $(Pipeline.Workspace)
+      targetType: inline
+      ${{ if ne(parameters.gpuTarget, '') }}:
+        script: echo "rocmexternalcicd.azurecr.io/$(Build.DefinitionName)-${{ parameters.environment }}-${{ parameters.gpuTarget }}:$(Build.BuildId)" | tr '[:upper:]' '[:lower:]'
+      ${{ else }}:
+        script: echo "rocmexternalcicd.azurecr.io/$(Build.DefinitionName)-${{ parameters.environment }}:$(Build.BuildId)" | tr '[:upper:]' '[:lower:]'

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

@@ -334,7 +334,7 @@ steps:
   - task: Docker@2
     condition: or(and(failed(), ${{ not(containsValue(parameters.dockerSkipGfx, parameters.gpuTarget)) }}), ${{ eq(parameters.forceDockerCreation, true) }})
     inputs:
-      containerRegistry: 'ContainerService'
+      containerRegistry: 'ContainerService3'
       ${{ if ne(parameters.gpuTarget, '') }}:
         repository: '$(Build.DefinitionName)-${{ parameters.environment }}-${{ parameters.gpuTarget }}'
       ${{ else }}:
@@ -348,6 +348,6 @@ steps:
       workingDirectory: $(Pipeline.Workspace)
       targetType: inline
       ${{ if ne(parameters.gpuTarget, '') }}:
-        script: echo "rocmexternalcicd.azurecr.io/$(Build.DefinitionName)-${{ parameters.environment }}-${{ parameters.gpuTarget }}:$(Build.BuildId)"
+        script: echo "rocmexternalcicd.azurecr.io/$(Build.DefinitionName)-${{ parameters.environment }}-${{ parameters.gpuTarget }}:$(Build.BuildId)" | tr '[:upper:]' '[:lower:]'
       ${{ else }}:
-        script: echo "rocmexternalcicd.azurecr.io/$(Build.DefinitionName)-${{ parameters.environment }}:$(Build.BuildId)"
+        script: echo "rocmexternalcicd.azurecr.io/$(Build.DefinitionName)-${{ parameters.environment }}:$(Build.BuildId)" | tr '[:upper:]' '[:lower:]'

.azuredevops/templates/steps/miopen-get-ck-build.yml

@@ -54,7 +54,7 @@ steps:
       fi
 
       echo "Downloading CK artifact from $ARTIFACT_URL"
-      wget -nv $ARTIFACT_URL -O $(System.ArtifactsDirectory)/ck.zip
+      wget --tries=5 --waitretry=10 --retry-connrefused -nv $ARTIFACT_URL -O $(System.ArtifactsDirectory)/ck.zip
       unzip $(System.ArtifactsDirectory)/ck.zip -d $(System.ArtifactsDirectory)
       mkdir -p $(Agent.BuildDirectory)/rocm
       tar -zxvf $(System.ArtifactsDirectory)/$ARTIFACT_NAME/*.tar.gz -C $(Agent.BuildDirectory)/rocm

.azuredevops/variables-global.yml

@@ -30,13 +30,13 @@ variables:
 - name: GFX90A_TEST_POOL
   value: gfx90a_test_pool
 - name: LATEST_RELEASE_VERSION
-  value: 6.3.4
-- name: REPO_RADEON_VERSION
-  value: 6.3.4
-- name: NEXT_RELEASE_VERSION
   value: 6.4.0
+- name: REPO_RADEON_VERSION
+  value: 6.4
+- name: NEXT_RELEASE_VERSION
+  value: 6.5.0
 - name: LATEST_RELEASE_TAG
-  value: rocm-6.3.4
+  value: rocm-6.4.0
 - name: DOCKER_SKIP_GFX
   value: gfx90a
 - name: AMDMIGRAPHX_GFX942_TEST_PIPELINE_ID

.wordlist.txt

@@ -34,7 +34,6 @@ Autocast
 BARs
 BLAS
 BMC
-BabelStream
 Blit
 Blockwise
 Bluefield
@@ -139,7 +138,6 @@ GDR
 GDS
 GEMM
 GEMMs
-GFLOPS
 GFortran
 GFXIP
 Gemma
@@ -228,7 +226,6 @@ LM
 LSAN
 LSan
 LTS
-LanguageCrossEntropy
 LoRA
 MEM
 MERCHANTABILITY
@@ -246,7 +243,6 @@ MMIOH
 MMU
 MNIST
 MPI
-MPT
 MSVC
 MVAPICH
 MVFFR
@@ -263,7 +259,6 @@ Meta's
 Miniconda
 MirroredStrategy
 Mixtral
-MosaicML
 Multicore
 Multithreaded
 MyEnvironment
@@ -334,7 +329,6 @@ PipelineParallel
 PnP
 PowerEdge
 PowerShell
-Pretrained
 Pretraining
 Profiler's
 PyPi
@@ -643,7 +637,6 @@ hipSPARSELt
 hipTensor
 hipamd
 hipblas
-hipcc
 hipcub
 hipfft
 hipfort

CHANGELOG.md

@@ -6,7 +6,7 @@ different versions of the ROCm software stack and its components.
 
 ## ROCm 6.4.0
 
-See the [ROCm 6.4.0 release notes](https://rocm.docs.amd.com/en/docs-6.4.0/about/release-notes.html)
+See the [ROCm 6.4.0 release notes](https://rocm-stg.amd.com/en/latest/about/release-notes.html)
 for a complete overview of this release.
 
 ### **AMD SMI** (25.3.0)
@@ -125,8 +125,7 @@ Some workaround options are as follows:
 - The `pasid` field in struct `amdsmi_process_info_t` will be deprecated in a future ROCm release.
 
 ```{note}
-See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples,
-and in-depth descriptions.
+See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
 ```
 
 ### **AMDMIGraphX** (2.12.0)
@@ -736,8 +735,7 @@ The following lists the backward incompatible changes planned for upcoming major
 - Fixed `rsmi_dev_target_graphics_version_get`, `rocm-smi --showhw`, and `rocm-smi --showprod` not displaying graphics version correctly for Instinct MI200 series, MI100 series, and RDNA3-based GPUs. 
 
 ```{note}
-See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples,
-and in-depth descriptions.
+See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
 ```
 
 ### **ROCm Systems Profiler** (1.0.0)
@@ -745,10 +743,6 @@ and in-depth descriptions.
 #### Added 
 
 - Support for VA-API and rocDecode tracing.
-- Aggregation of MPI data collected across distributed nodes and ranks. The data is concatenated into a single proto file.
-
-#### Changed
-- Backend refactored to use [ROCprofiler-SDK](https://github.com/ROCm/rocprofiler-sdk) rather than [ROCProfiler](https://github.com/ROCm/rocprofiler) and [ROCTracer](https://github.com/ROCm/ROCTracer).
 
 #### Resolved issues
 
@@ -759,9 +753,9 @@ and in-depth descriptions.
 - Fixed interruption in config file generation.
 
 - Fixed segmentation fault while running rocprof-sys-instrument.
-- Fixed an issue where running `rocprof-sys-causal` or using the `-I all` option with `rocprof-sys-sample` caused the system to become non-responsive.
 
-- Fixed an issue where sampling multi-GPU Python workloads caused the system to stop responding.
+#### Changed
+- Backend refactored to use [ROCprofiler-SDK](https://github.com/ROCm/rocprofiler-sdk) rather than [ROCProfiler](https://github.com/ROCm/rocprofiler) and [ROCTracer](https://github.com/ROCm/ROCTracer).
 
 ### **rocPRIM** (3.4.0)
 

RELEASE.md

@@ -253,19 +253,14 @@ Click {fab}`github` to go to the component's source code on GitHub.
         </tbody>
         <tbody class="rocm-components-libs rocm-components-communication tbody-reverse-zebra">
             <tr>
-                <th rowspan="2"></th>
-                <th rowspan="2">Communication</th>
+                <th rowspan="1"></th>
+                <th rowspan="1">Communication</th>
                 <td><a href="https://rocm.docs.amd.com/projects/rccl/en/docs-6.4.0/index.html">RCCL</a></td>
                 <td>2.21.5&nbsp;&Rightarrow;&nbsp;<a href="#rccl-2-22-3">2.22.3</a></td>
                 <td><a href="https://github.com/ROCm/rccl"><i class="fab fa-github fa-lg"></i></a></td>
             </tr>
-            <tr>
-            <td><a href="https://rocm.docs.amd.com/projects/rocSHMEM/en/docs-6.4.0/index.html">rocSHMEM</a></td>
-                <td>2.0.0</td>
-                <td><a href="https://github.com/ROCm/rocSHMEM"><i class="fab fa-github fa-lg"></i></a></td>
-            </tr>
         </tbody>
-        <tbody class="rocm-components-libs rocm-components-math tbody-reverse-zebra">
+        <tbody class="rocm-components-libs rocm-components-math">
             <tr>
                 <th rowspan="16"></th>
                 <th rowspan="16">Math</th>
@@ -349,7 +344,7 @@ Click {fab}`github` to go to the component's source code on GitHub.
                 <td><a href="https://github.com/ROCm/Tensile"><i class="fab fa-github fa-lg"></i></a></td>
             </tr>
         </tbody>
-        <tbody class="rocm-components-libs rocm-components-primitives tbody-reverse-zebra">
+        <tbody class="rocm-components-libs rocm-components-primitives">
             <tr>
                 <th rowspan="4"></th>
                 <th rowspan="4">Primitives</th>
@@ -373,7 +368,7 @@ Click {fab}`github` to go to the component's source code on GitHub.
                 <td><a href="https://github.com/ROCm/rocThrust"><i class="fab fa-github fa-lg"></i></a></td>
             </tr>
         </tbody>
-        <tbody class="rocm-components-tools rocm-components-system tbody-reverse-zebra">
+        <tbody class="rocm-components-tools rocm-components-system">
             <tr>
                 <th rowspan="7">Tools</th>
                 <th rowspan="7">System management</th>
@@ -402,7 +397,7 @@ Click {fab}`github` to go to the component's source code on GitHub.
                 <td><a href="https://github.com/ROCm/ROCmValidationSuite"><i class="fab fa-github fa-lg"></i></a></td>
             </tr>
         </tbody>
-        <tbody class="rocm-components-tools rocm-components-perf">
+        <tbody class="rocm-components-tools rocm-components-perf tbody-reverse-zebra">
             <tr>
                 <th rowspan="6"></th>
                 <th rowspan="6">Performance</th>
@@ -443,7 +438,7 @@ Click {fab}`github` to go to the component's source code on GitHub.
                             class="fab fa-github fa-lg"></i></a></td>
             </tr>
         </tbody>
-        <tbody class="rocm-components-tools rocm-components-dev">
+        <tbody class="rocm-components-tools rocm-components-dev tbody-reverse-zebra">
             <tr>
                 <th rowspan="5"></th>
                 <th rowspan="5">Development</th>
@@ -479,7 +474,7 @@ Click {fab}`github` to go to the component's source code on GitHub.
                             class="fab fa-github fa-lg"></i></a></td>
             </tr>
         </tbody>
-        <tbody class="rocm-components-compilers tbody-reverse-zebra">
+        <tbody class="rocm-components-compilers">
             <tr>
                 <th rowspan="2" colspan="2">Compilers</th>
                 <td><a href="https://rocm.docs.amd.com/projects/HIPCC/en/docs-6.4.0/index.html">HIPCC</a></td>
@@ -494,7 +489,7 @@ Click {fab}`github` to go to the component's source code on GitHub.
                             class="fab fa-github fa-lg"></i></a></td>
             </tr>
         </tbody>
-        <tbody class="rocm-components-runtimes tbody-reverse-zebra">
+        <tbody class="rocm-components-runtimes">
             <tr>
                 <th rowspan="2" colspan="2">Runtimes</th>
                 <td><a href="https://rocm.docs.amd.com/projects/HIP/en/docs-6.4.0/index.html">HIP</a></td>
@@ -634,8 +629,7 @@ Some workaround options are as follows:
 - The `pasid` field in struct `amdsmi_process_info_t` will be deprecated in a future ROCm release.
 
 ```{note}
-See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples,
-and in-depth descriptions.
+See the full [AMD SMI changelog](https://github.com/ROCm/amdsmi/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
 ```
 
 ### **AMDMIGraphX** (2.12.0)
@@ -767,17 +761,17 @@ and in-depth descriptions.
 
 #### Changed
 
-* The `roc-obj` tools have been deprecated and will be removed in a future release.
+* `roc-obj` tools is deprecated and will be removed in an upcoming release.
 
-   -	`llvm-objdump`, `llvm-objcopy`, and `llvm-readobj` will be enhanced to provide similar functionality as that provided by the `roc-obj` tools . The LLVM tools are available in the `rocm-llvm` pkg.
-   -	While not related to the deprecation, also note that the `roc-obj` tools’ package dependency on Perl has been changed to recommended. It is the user’s responsibility to install Perl to use these tools.
+    - Perl package installation is not required, and users will need to install this themselves if they want to.
+    - Support for ROCm Object tooling has moved into `llvm-objdump` provided by package `rocm-llvm`.
 
 * SDMA retainer logic is removed for engine selection in operation of runtime buffer copy.
 
 #### Optimized
 
 * `hipGraphLaunch` parallelism is improved for complex data-parallel graphs.
-* Round-robin queue mechanism is updated for command scheduling. For multi-streams execution, HSA queue from null stream lock is freed and won't occupy the queue ID after the kernel in the stream is finished.
+* Make the round-robin queue selection in command scheduling. For multi-streams execution, HSA queue from null stream lock is freed and won't occupy the queue ID after the kernel in the stream is finished.
 * The HIP runtime doesn't free bitcode object before code generation. It adds a cache, which allows compiled code objects to be reused instead of recompiling. This improves performance on multi-GPU systems.
 * Runtime now uses unified copy approach:
 
@@ -786,11 +780,6 @@ and in-depth descriptions.
     - The default environment variable `GPU_FORCE_BLIT_COPY_SIZE` is set to `16`, which limits the kernel copy to sizes less than 16 KB, while copies larger than that would be handled by `SDMA` engine.
     - Blit code is refactored, and ASAN instrumentation is cleaned up.
 
-* HIP runtime uses signals without interrupts:
-
-    - In active wait mode, uses signals without interrupts by default.
-    - Only when a callback is required, switches to the interrupts.
-
 #### Resolved issues
 
 * Out-of-memory error on Microsoft Windows. When the user calls `hipMalloc` for device memory allocation while specifying a size larger than the available device memory, the HIP runtime fixes the error in the API implementation, allocating the available device memory plus system memory (shared virtual memory).
@@ -801,15 +790,13 @@ and in-depth descriptions.
 
 The following lists the backward incompatible changes planned for upcoming major ROCm releases.
 
-* Signature changes in APIs to match corresponding CUDA APIs,
+* Signature changes in APIs to correspond with NVIDIA CUDA APIs,
 
     - `hiprtcCreateProgram`
     - `hiprtcCompileProgram`
     - `hipCtxGetApiVersion`
 
 * Behavior of `hipPointerGetAttributes` is changed to match corresponding CUDA API in version 11 and later releases.
-* Behavior of `hipFree` is changed to match corresponding CUDA API `cudaFree`.
-* HIP vector constructor changes for `hipComplex`.
 * Return error/value code updates in the following hip APIs to match the corresponding CUDA APIs,
 
     - `hipModuleLaunchKernel`
@@ -1252,8 +1239,7 @@ The following lists the backward incompatible changes planned for upcoming major
 - Fixed `rsmi_dev_target_graphics_version_get`, `rocm-smi --showhw`, and `rocm-smi --showprod` not displaying graphics version correctly for Instinct MI200 series, MI100 series, and RDNA3-based GPUs. 
 
 ```{note}
-See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples,
-and in-depth descriptions.
+See the full [ROCm SMI changelog](https://github.com/ROCm/rocm_smi_lib/blob/release/rocm-rel-6.4/CHANGELOG.md) for details, examples, and in-depth descriptions.
 ```
 
 ### **ROCm Systems Profiler** (1.0.0)
@@ -1261,11 +1247,6 @@ and in-depth descriptions.
 #### Added 
 
 - Support for VA-API and rocDecode tracing.
-- Aggregation of MPI data collected across distributed nodes and ranks. The data is concatenated into a single proto file.
-
-
-#### Changed
-- Backend refactored to use [ROCprofiler-SDK](https://github.com/ROCm/rocprofiler-sdk) rather than [ROCProfiler](https://github.com/ROCm/rocprofiler) and [ROCTracer](https://github.com/ROCm/ROCTracer).
 
 #### Resolved issues
 
@@ -1276,9 +1257,9 @@ and in-depth descriptions.
 - Fixed interruption in config file generation.
 
 - Fixed segmentation fault while running rocprof-sys-instrument.
-- Fixed an issue where running `rocprof-sys-causal` or using the `-I all` option with `rocprof-sys-sample` caused the system to become non-responsive.
 
-- Fixed an issue where sampling multi-GPU Python workloads caused the system to stop responding.
+#### Changed
+- Backend refactored to use [ROCprofiler-SDK](https://github.com/ROCm/rocprofiler-sdk) rather than [ROCProfiler](https://github.com/ROCm/rocprofiler) and [ROCTracer](https://github.com/ROCm/ROCTracer).
 
 ### **rocPRIM** (3.4.0)
 
@@ -1732,7 +1713,7 @@ and will be disabled in a future release.
 
 * The `__AMDGCN_WAVEFRONT_SIZE__` macro and `__AMDGCN_WAVEFRONT_SIZE` alias will be removed in an upcoming release.
   It is recommended to remove any use of this macro. For more information, see
-  [AMDGPU support](https://rocm.docs.amd.com/projects/llvm-project/en/docs-6.4.0/LLVM/clang/html/AMDGPUSupport.html).
+  [AMDGPU support](https://rocm.docs.amd.com/projects/llvm-project/en/docs-6.3.2/LLVM/clang/html/AMDGPUSupport.html).
 * `warpSize` will only be available as a non-`constexpr` variable. Where required,
   the wavefront size should be queried via the `warpSize` variable in device code,
   or via `hipGetDeviceProperties` in host code. Neither of these will result in a compile-time constant. 
@@ -1770,5 +1751,4 @@ There are a number of upcoming changes planned for HIP runtime API in an upcomin
 that are not backward compatible with prior releases. Most of these changes increase 
 alignment between HIP and CUDA APIs or behavior. Some of the upcoming changes are to 
 clean up header files, remove namespace collision, and have a clear separation between 
-`hipRTC` and HIP runtime. For more information, see [HIP Upcoming changes](#hip-6-4-0)
-or [HIP 7.0 Is Coming: What You Need to Know to Stay Ahead](https://rocm.blogs.amd.com/ecosystems-and-partners/transition-to-hip-7.0:-guidance-on-upcoming-compatibility-changes/README.html).
+`hipRTC` and HIP runtime. For more information refer to [HIP Upcoming changes](#hip-6-4-0).

docs/about/license.md

@@ -81,7 +81,6 @@ additional licenses. Please review individual repositories for more information.
 | [rocRAND](https://github.com/ROCm/rocRAND/) | [MIT](https://github.com/ROCm/rocRAND/blob/develop/LICENSE.txt) |
 | [ROCr Debug Agent](https://github.com/ROCm/rocr_debug_agent/) | [The University of Illinois/NCSA](https://github.com/ROCm/rocr_debug_agent/blob/amd-staging/LICENSE.txt) |
 | [ROCR-Runtime](https://github.com/ROCm/ROCR-Runtime/) | [The University of Illinois/NCSA](https://github.com/ROCm/ROCR-Runtime/blob/amd-staging/LICENSE.txt) |
-| [rocSHMEM](https://github.com/ROCm/rocSHMEM/) | [MIT](https://github.com/ROCm/rocSHMEM/blob/develop/LICENSE.md) |
 | [rocSOLVER](https://github.com/ROCm/rocSOLVER/) | [BSD-2-Clause](https://github.com/ROCm/rocSOLVER/blob/develop/LICENSE.md) |
 | [rocSPARSE](https://github.com/ROCm/rocSPARSE/) | [MIT](https://github.com/ROCm/rocSPARSE/blob/develop/LICENSE.md) |
 | [rocThrust](https://github.com/ROCm/rocThrust/) | [Apache 2.0](https://github.com/ROCm/rocThrust/blob/develop/LICENSE) |

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

@@ -27,6 +27,7 @@ ROCm Version,6.4.0,6.3.3,6.3.2,6.3.1,6.3.0,6.2.4,6.2.2,6.2.1,6.2.0, 6.1.5, 6.1.2
       :doc:`TensorFlow <../compatibility/ml-compatibility/tensorflow-compatibility>`,"2.18.1, 2.17.1, 2.16.2","2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.17.0, 2.16.2, 2.15.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.16.1, 2.15.1, 2.14.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.15.0, 2.14.0, 2.13.1","2.14.0, 2.13.1, 2.12.1","2.14.0, 2.13.1, 2.12.1"
       :doc:`JAX <../compatibility/ml-compatibility/jax-compatibility>`,0.4.35,0.4.31,0.4.31,0.4.31,0.4.31,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26,0.4.26
       `ONNX Runtime <https://onnxruntime.ai/docs/build/eps.html#amd-migraphx>`_,1.2,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.17.3,1.14.1,1.14.1
+,,,,,,,,,,,,,,,
       ,,,,,,,,,,,,,,,
       THIRD PARTY COMMS,.. _thirdpartycomms-support-compatibility-matrix-past-60:,,,,,,,,,,,,,,
       `UCC <https://github.com/ROCm/ucc>`_,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.3.0,>=1.2.0,>=1.2.0
@@ -52,7 +53,6 @@ ROCm Version,6.4.0,6.3.3,6.3.2,6.3.1,6.3.0,6.2.4,6.2.2,6.2.1,6.2.0, 6.1.5, 6.1.2
       ,,,,,,,,,,,,,,,
       COMMUNICATION,.. _commlibs-support-compatibility-matrix-past-60:,,,,,,,,,,,,,,
       :doc:`RCCL <rccl:index>`,2.22.3,2.21.5,2.21.5,2.21.5,2.21.5,2.20.5,2.20.5,2.20.5,2.20.5,2.18.6,2.18.6,2.18.6,2.18.6,2.18.3,2.18.3
-      :doc:`rocSHMEM <rocSHMEM:index>`,2.0.0,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A,N/A
       ,,,,,,,,,,,,,,,
       MATH LIBS,.. _mathlibs-support-compatibility-matrix-past-60:,,,,,,,,,,,,,,
       `half <https://github.com/ROCm/half>`_ ,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0,1.12.0

docs/compatibility/compatibility-matrix.rst

@@ -77,7 +77,6 @@ compatibility and system requirements.
       ,,,
       COMMUNICATION,.. _commlibs-support-compatibility-matrix:,,
       :doc:`RCCL <rccl:index>`,2.22.3,2.21.5,2.20.5
-      :doc:`rocSHMEM <rocSHMEM:index>`,2.0.0,N/A,N/A
       ,,,
       MATH LIBS,.. _mathlibs-support-compatibility-matrix:,,
       `half <https://github.com/ROCm/half>`_ ,1.12.0,1.12.0,1.12.0

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

@@ -14,18 +14,17 @@ JAX provides a NumPy-like API, which combines automatic differentiation and the
 Accelerated Linear Algebra (XLA) compiler to achieve high-performance machine
 learning at scale.
 
-JAX uses composable transformations of Python and NumPy through just-in-time
-(JIT) compilation, automatic vectorization, and parallelization. To learn about
-JAX, including profiling and optimizations, see the official `JAX documentation
+JAX uses composable transformations of Python and NumPy through just-in-time (JIT) compilation,
+automatic vectorization, and parallelization. To learn about JAX, including profiling and
+optimizations, see the official `JAX documentation
 <https://jax.readthedocs.io/en/latest/notebooks/quickstart.html>`_.
 
-ROCm support for JAX is upstreamed, and users can build the official source code
-with ROCm support:
+ROCm support for JAX is upstreamed and users can build the official source code with ROCm
+support:
 
 - ROCm JAX release:
 
-  - Offers AMD-validated and community :ref:`Docker images <jax-docker-compat>`
-    with ROCm and JAX preinstalled.
+  - Offers AMD-validated and community :ref:`Docker images <jax-docker-compat>` with ROCm and JAX pre-installed.
 
   - ROCm JAX repository: `ROCm/jax <https://github.com/ROCm/jax>`_
 
@@ -37,8 +36,8 @@ with ROCm support:
   - Official JAX repository: `jax-ml/jax <https://github.com/jax-ml/jax>`_
 
   - See the `AMD GPU (Linux) installation section
-    <https://jax.readthedocs.io/en/latest/installation.html#amd-gpu-linux>`_ in
-    the JAX documentation.
+    <https://jax.readthedocs.io/en/latest/installation.html#amd-gpu-linux>`_ in the JAX
+    documentation.
 
 .. note::
 
@@ -47,44 +46,6 @@ with ROCm support:
    `Community ROCm JAX Docker images <https://hub.docker.com/r/rocm/jax-community>`_
    follow upstream JAX releases and use the latest available ROCm version.
 
-Use cases and recommendations
-================================================================================
-
-* The `nanoGPT in JAX <https://rocm.blogs.amd.com/artificial-intelligence/nanoGPT-JAX/README.html>`_
-  blog explores the implementation and training of a Generative Pre-trained
-  Transformer (GPT) model in JAX, inspired by Andrej Karpathy’s JAX-based
-  nanoGPT. Comparing how essential GPT components—such as self-attention 
-  mechanisms and optimizers—are realized in JAX and JAX, also highlights
-  JAX’s unique features.
-
-* The `Optimize GPT Training: Enabling Mixed Precision Training in JAX using
-  ROCm on AMD GPUs <https://rocm.blogs.amd.com/artificial-intelligence/jax-mixed-precision/README.html>`_
-  blog post provides a comprehensive guide on enhancing the training efficiency
-  of GPT models by implementing mixed precision techniques in JAX, specifically
-  tailored for AMD GPUs utilizing the ROCm platform.
-
-* The `Supercharging JAX with Triton Kernels on AMD GPUs <https://rocm.blogs.amd.com/artificial-intelligence/jax-triton/README.html>`_
-  blog demonstrates how to develop a custom fused dropout-activation kernel for
-  matrices using Triton, integrate it with JAX, and benchmark its performance
-  using ROCm.
-
-* The `Distributed fine-tuning with JAX on AMD GPUs <https://rocm.blogs.amd.com/artificial-intelligence/distributed-sft-jax/README.html>`_
-  outlines the process of fine-tuning a Bidirectional Encoder Representations
-  from Transformers (BERT)-based large language model (LLM) using JAX for a text
-  classification task. The blog post discuss techniques for parallelizing the
-  fine-tuning across multiple AMD GPUs and assess the model's performance on a
-  holdout dataset. During the fine-tuning, a BERT-base-cased transformer model
-  and the General Language Understanding Evaluation (GLUE) benchmark dataset was
-  used on a multi-GPU setup.
-
-* The `MI300X workload optimization guide <https://rocm.docs.amd.com/en/latest/how-to/tuning-guides/mi300x/workload.html>`_
-  provides detailed guidance on optimizing workloads for the AMD Instinct MI300X
-  accelerator using ROCm. The page is aimed at helping users achieve optimal
-  performance for deep learning and other high-performance computing tasks on
-  the MI300X GPU.
-
-For more use cases and recommendations, see `ROCm JAX blog posts <https://rocm.blogs.amd.com/blog/tag/jax.html>`_.
-
 .. _jax-docker-compat:
 
 Docker image compatibility
@@ -96,7 +57,7 @@ Docker image compatibility
 
 AMD validates and publishes ready-made `ROCm JAX Docker images <https://hub.docker.com/r/rocm/jax>`_
 with ROCm backends on Docker Hub. The following Docker image tags and
-associated inventories represent the latest JAX version from the official Docker Hub and are validated for
+associated inventories are validated for
 `ROCm 6.4.0 <https://repo.radeon.com/rocm/apt/6.4/>`_. Click the |docker-icon|
 icon to view the image on Docker Hub.
 
@@ -160,12 +121,13 @@ associated inventories are tested for `ROCm 6.3.2 <https://repo.radeon.com/rocm/
       - Ubuntu 22.04
       - `3.10.16 <https://www.python.org/downloads/release/python-31016/>`_
 
-Key ROCm libraries for JAX
+Critical ROCm libraries for JAX
 ================================================================================
 
-JAX functionality on ROCm is determined by its underlying library
-dependencies. These ROCm components affect the capabilities, performance, and
-feature set available to developers.
+The functionality of JAX with ROCm is determined by its underlying library
+dependencies. These critical ROCm components affect the capabilities,
+performance, and feature set available to developers. The versions described
+are available in ROCm :version:`rocm_version`.
 
 .. list-table::
     :header-rows: 1
@@ -253,10 +215,10 @@ feature set available to developers.
         distributed training, which involves parallel reductions or
         operations like ``jax.numpy.cumsum`` can use rocThrust.
 
-Supported features
+Supported and unsupported features
 ===============================================================================
 
-The following table maps the public JAX API modules to their supported
+The following table maps GPU-accelerated JAX modules to their supported
 ROCm and JAX versions.
 
 .. list-table::
@@ -264,8 +226,8 @@ ROCm and JAX versions.
 
     * - Module
       - Description
-      - As of JAX
-      - As of ROCm
+      - Since JAX
+      - Since ROCm
     * - ``jax.numpy``
       - Implements the NumPy API, using the primitives in ``jax.lax``.
       - 0.1.56
@@ -293,11 +255,21 @@ ROCm and JAX versions.
         devices.
       - 0.3.20
       - 5.1.0
+    * - ``jax.dlpack``
+      - For exchanging tensor data between JAX and other libraries that support the
+        DLPack standard.
+      - 0.1.57
+      - 5.0.0
     * - ``jax.distributed``
       - Enables the scaling of computations across multiple devices on a single
         machine or across multiple machines.
       - 0.1.74
       - 5.0.0
+    * - ``jax.dtypes``
+      - Provides utilities for working with and managing data types in JAX
+        arrays and computations.
+      - 0.1.66
+      - 5.0.0
     * - ``jax.image``
       - Contains image manipulation functions like resize, scale and translation.
       - 0.1.57
@@ -311,10 +283,27 @@ ROCm and JAX versions.
         array.
       - 0.1.57
       - 5.0.0
+    * - ``jax.profiler``
+      - Contains JAX’s tracing and time profiling features.
+      - 0.1.57
+      - 5.0.0
     * - ``jax.stages``
       - Contains interfaces to stages of the compiled execution process.
       - 0.3.4
       - 5.0.0
+    * - ``jax.tree``
+      - Provides utilities for working with tree-like container data structures.
+      - 0.4.26
+      - 5.6.0
+    * - ``jax.tree_util``
+      - Provides utilities for working with nested data structures, or
+        ``pytrees``.
+      - 0.1.65
+      - 5.0.0
+    * - ``jax.typing``
+      - Provides JAX-specific static type annotations.
+      - 0.3.18
+      - 5.1.0
     * - ``jax.extend``
       - Provides modules for access to JAX internal machinery module. The
         ``jax.extend`` module defines a library view of some of JAX’s internal
@@ -350,8 +339,8 @@ A SciPy-like API for scientific computing.
     :header-rows: 1
 
     * - Module
-      - As of JAX
-      - As of ROCm
+      - Since JAX
+      - Since ROCm
     * - ``jax.scipy.cluster``
       - 0.3.11
       - 5.1.0
@@ -396,8 +385,8 @@ jax.scipy.stats module
    :header-rows: 1
 
    * - Module
-     - As of JAX
-     - As of ROCm
+     - Since JAX
+     - Since ROCm
    * - ``jax.scipy.stats.bernouli``
      - 0.1.56
      - 5.0.0
@@ -480,8 +469,8 @@ Modules for JAX extensions.
     :header-rows: 1
 
     * - Module
-      - As of JAX
-      - As of ROCm
+      - Since JAX
+      - Since ROCm
     * - ``jax.extend.ffi``
       - 0.4.30
       - 6.0.0
@@ -495,25 +484,190 @@ Modules for JAX extensions.
       - 0.4.15
       - 5.5.0
 
-Unsupported JAX features
-===============================================================================
+jax.experimental module
+-------------------------------------------------------------------------------
 
-The following GPU-accelerated JAX features are not supported by ROCm for
-the listed supported JAX versions.
+Experimental modules and APIs.
+
+.. list-table::
+    :header-rows: 1
+
+    * - Module
+      - Since JAX
+      - Since ROCm
+    * - ``jax.experimental.checkify``
+      - 0.1.75
+      - 5.0.0
+    * - ``jax.experimental.compilation_cache.compilation_cache``
+      - 0.1.68
+      - 5.0.0
+    * - ``jax.experimental.custom_partitioning``
+      - 0.4.0
+      - 5.3.0
+    * - ``jax.experimental.jet``
+      - 0.1.56
+      - 5.0.0
+    * - ``jax.experimental.key_reuse``
+      - 0.4.26
+      - 5.6.0
+    * - ``jax.experimental.mesh_utils``
+      - 0.1.76
+      - 5.0.0
+    * - ``jax.experimental.multihost_utils``
+      - 0.3.2
+      - 5.0.0
+    * - ``jax.experimental.pallas``
+      - 0.4.15
+      - 5.5.0
+    * - ``jax.experimental.pjit``
+      - 0.1.61
+      - 5.0.0
+    * - ``jax.experimental.serialize_executable``
+      - 0.4.0
+      - 5.3.0
+    * - ``jax.experimental.shard_map``
+      - 0.4.3
+      - 5.3.0
+    * - ``jax.experimental.sparse``
+      - 0.1.75
+      - 5.0.0
+
+.. list-table::
+    :header-rows: 1
+
+    * - API
+      - Since JAX
+      - Since ROCm
+    * - ``jax.experimental.enable_x64``
+      - 0.1.60
+      - 5.0.0
+    * - ``jax.experimental.disable_x64``
+      - 0.1.60
+      - 5.0.0
+
+jax.experimental.pallas module
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Module for Pallas, a JAX extension for custom kernels.
+
+.. list-table::
+    :header-rows: 1
+
+    * - Module
+      - Since JAX
+      - Since ROCm
+    * - ``jax.experimental.pallas.mosaic_gpu``
+      - 0.4.31
+      - 6.1.3
+    * - ``jax.experimental.pallas.tpu``
+      - 0.4.15
+      - 5.5.0
+    * - ``jax.experimental.pallas.triton``
+      - 0.4.32
+      - 6.1.3
+
+jax.experimental.sparse module
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Experimental support for sparse matrix operations.
+
+.. list-table::
+    :header-rows: 1
+
+    * - Module
+      - Since JAX
+      - Since ROCm
+    * - ``jax.experimental.sparse.linalg``
+      - 0.3.15
+      - 5.2.0
+    * - ``jax.experimental.sparse.sparsify``
+      - 0.3.25
+      - ❌
+
+.. list-table::
+    :header-rows: 1
+
+    * - ``sparse`` data structure API
+      - Since JAX
+      - Since ROCm
+    * - ``jax.experimental.sparse.BCOO``
+      - 0.1.72
+      - 5.0.0
+    * - ``jax.experimental.sparse.BCSR``
+      - 0.3.20
+      - 5.1.0
+    * - ``jax.experimental.sparse.CSR``
+      - 0.1.75
+      - 5.0.0
+    * - ``jax.experimental.sparse.NM``
+      - 0.4.27
+      - 5.6.0
+    * - ``jax.experimental.sparse.COO``
+      - 0.1.75
+      - 5.0.0
+
+Unsupported JAX features
+------------------------
+
+The following are GPU-accelerated JAX features not currently supported by
+ROCm.
 
 .. list-table::
     :header-rows: 1
 
     * - Feature
       - Description
-
+      - Since JAX
     * - Mixed Precision with TF32
       - Mixed precision with TF32 is used for matrix multiplications,
         convolutions, and other linear algebra operations, particularly in
         deep learning workloads like CNNs and transformers.
-
+      - 0.2.25
+    * - RNN support
+      - Currently only LSTM with double bias is supported with float32 input
+        and weight.
+      - 0.3.25
     * - XLA int4 support
       - 4-bit integer (int4) precision in the XLA compiler.
+      - 0.4.0
+    * - ``jax.experimental.sparsify``
+      - Converts a dense matrix to a sparse matrix representation.
+      - Experimental
 
-    * - MOSAIC (GPU)
-      - Mosaic is a library of kernel-building abstractions for JAX's Pallas system
+Use cases and recommendations
+================================================================================
+
+* The `nanoGPT in JAX <https://rocm.blogs.amd.com/artificial-intelligence/nanoGPT-JAX/README.html>`_
+  blog explores the implementation and training of a Generative Pre-trained
+  Transformer (GPT) model in JAX, inspired by Andrej Karpathy’s PyTorch-based
+  nanoGPT. By comparing how essential GPT components—such as self-attention
+  mechanisms and optimizers—are realized in PyTorch and JAX, also highlight
+  JAX’s unique features.
+
+* The `Optimize GPT Training: Enabling Mixed Precision Training in JAX using
+  ROCm on AMD GPUs <https://rocm.blogs.amd.com/artificial-intelligence/jax-mixed-precision/README.html>`_
+  blog post provides a comprehensive guide on enhancing the training efficiency
+  of GPT models by implementing mixed precision techniques in JAX, specifically
+  tailored for AMD GPUs utilizing the ROCm platform.
+
+* The `Supercharging JAX with Triton Kernels on AMD GPUs <https://rocm.blogs.amd.com/artificial-intelligence/jax-triton/README.html>`_
+  blog demonstrates how to develop a custom fused dropout-activation kernel for
+  matrices using Triton, integrate it with JAX, and benchmark its performance
+  using ROCm.
+
+* The `Distributed fine-tuning with JAX on AMD GPUs <https://rocm.blogs.amd.com/artificial-intelligence/distributed-sft-jax/README.html>`_
+  outlines the process of fine-tuning a Bidirectional Encoder Representations
+  from Transformers (BERT)-based large language model (LLM) using JAX for a text
+  classification task. The blog post discuss techniques for parallelizing the
+  fine-tuning across multiple AMD GPUs and assess the model's performance on a
+  holdout dataset. During the fine-tuning, a BERT-base-cased transformer model
+  and the General Language Understanding Evaluation (GLUE) benchmark dataset was
+  used on a multi-GPU setup.
+
+* The `MI300X workload optimization guide <https://rocm.docs.amd.com/en/latest/how-to/tuning-guides/mi300x/workload.html>`_
+  provides detailed guidance on optimizing workloads for the AMD Instinct MI300X
+  accelerator using ROCm. The page is aimed at helping users achieve optimal
+  performance for deep learning and other high-performance computing tasks on
+  the MI300X GPU.
+
+For more use cases and recommendations, see `ROCm JAX blog posts <https://rocm.blogs.amd.com/blog/tag/jax.html>`_.

docs/conf.py

@@ -51,15 +51,12 @@ article_pages = [
     {"file": "how-to/deep-learning-rocm", "os": ["linux"]},
 
     {"file": "how-to/rocm-for-ai/index", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/install", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/system-health-check", "os": ["linux"]},
 
     {"file": "how-to/rocm-for-ai/training/index", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/train-a-model", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/prerequisite-system-validation", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/benchmark-docker/megatron-lm", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/benchmark-docker/pytorch-training", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/training/scale-model-training", "os": ["linux"]},
 
     {"file": "how-to/rocm-for-ai/fine-tuning/index", "os": ["linux"]},
@@ -69,6 +66,7 @@ article_pages = [
     {"file": "how-to/rocm-for-ai/fine-tuning/multi-gpu-fine-tuning-and-inference", "os": ["linux"]},
 
     {"file": "how-to/rocm-for-ai/inference/index", "os": ["linux"]},
+    {"file": "how-to/rocm-for-ai/inference/install", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/hugging-face-models", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/llm-inference-frameworks", "os": ["linux"]},
     {"file": "how-to/rocm-for-ai/inference/vllm-benchmark", "os": ["linux"]},

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

@@ -1,8 +1,8 @@
 vllm_benchmark:
   unified_docker:
     latest:
-      pull_tag: rocm/vllm:rocm6.3.1_instinct_vllm0.8.3_20250415
-      docker_hub_url: https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_instinct_vllm0.8.3_20250415/images/sha256-ad9062dea3483d59dedb17c67f7c49f30eebd6eb37c3fac0a171fb19696cc845
+      pull_tag: rocm/vllm:rocm6.3.1_instinct_vllm0.8.3_20250410
+      docker_hub_url: https://hub.docker.com/layers/rocm/vllm/rocm6.3.1_instinct_vllm0.8.3_20250410/images/sha256-a0b55c6c0f3fa5d437fb54a66e32a108306c36d4776e570dfd0ae902719bd190
       rocm_version: 6.3.1
       vllm_version: 0.8.3
       pytorch_version: 2.7.0 (dev nightly)

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

@@ -1,178 +1,15 @@
 .. meta::
    :description: How to use model quantization techniques to speed up inference.
-   :keywords: ROCm, LLM, fine-tuning, usage, tutorial, quantization, Quark, GPTQ, transformers, bitsandbytes
+   :keywords: ROCm, LLM, fine-tuning, usage, tutorial, quantization, GPTQ, transformers, bitsandbytes
 
 *****************************
 Model quantization techniques
 *****************************
 
 Quantization reduces the model size compared to its native full-precision version, making it easier to fit large models
-onto accelerators or GPUs with limited memory usage. This section explains how to perform LLM quantization using AMD Quark, GPTQ
+onto accelerators or GPUs with limited memory usage. This section explains how to perform LLM quantization using GPTQ
 and bitsandbytes on AMD Instinct hardware.
 
-.. _quantize-llms-quark:
-
-AMD Quark
-=========
-
-`AMD Quark <https://quark.docs.amd.com/latest/>`_ offers the leading efficient and scalable quantization solution tailored to AMD Instinct GPUs. It supports ``FP8`` and ``INT8`` quantization for activations, weights, and KV cache, 
-including ``FP8`` attention. For very large models, it employs a two-level ``INT4-FP8`` scheme—storing weights in ``INT4`` while computing with ``FP8``—for nearly 4× compression without sacrificing accuracy. 
-Quark scales efficiently across multiple GPUs, efficiently handling ultra-large models like Llama-3.1-405B. Quantized ``FP8`` models like Llama, Mixtral, and Grok-1 are available under the `AMD organization on Hugging Face <https://huggingface.co/collections/amd/quark-quantized-ocp-fp8-models-66db7936d18fcbaf95d4405c>`_, and can be deployed directly via `vLLM <https://github.com/vllm-project/vllm/tree/main/vllm>`_.
-
-Installing Quark
--------------------
-
-The latest release of Quark can be installed with pip
-
-.. code-block:: shell
-
-    pip install amd-quark
-
-For detailed installation instructions, refer to the `Quark documentation <https://quark.docs.amd.com/latest/install.html>`_.
-
-
-Using Quark for quantization
------------------------------
-
-#. First, load the pre-trained model and its corresponding tokenizer using the Hugging Face ``transformers`` library.
-
-   .. code-block:: python
-
-      from transformers import AutoTokenizer, AutoModelForCausalLM
-
-      MODEL_ID = "meta-llama/Llama-2-70b-chat-hf"
-      MAX_SEQ_LEN = 512
-
-      model = AutoModelForCausalLM.from_pretrained(
-          MODEL_ID, device_map="auto", torch_dtype="auto",
-      )
-      model.eval()
-
-      tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, model_max_length=MAX_SEQ_LEN)
-      tokenizer.pad_token = tokenizer.eos_token
-
-#. Prepare the calibration DataLoader (static quantization requires calibration data).
-
-   .. code-block:: python
-
-      from datasets import load_dataset
-      from torch.utils.data import DataLoader
-
-      BATCH_SIZE = 1
-      NUM_CALIBRATION_DATA = 512
-
-      dataset = load_dataset("mit-han-lab/pile-val-backup", split="validation")
-      text_data = dataset["text"][:NUM_CALIBRATION_DATA]
-
-      tokenized_outputs = tokenizer(
-      text_data, return_tensors="pt", padding=True, truncation=True, max_length=MAX_SEQ_LEN
-      )
-      calib_dataloader = DataLoader(
-      tokenized_outputs['input_ids'], batch_size=BATCH_SIZE, drop_last=True
-      )
-
-#. Define the quantization configuration. See the comments in the following code snippet for descriptions of each configuration option.
-
-   .. code-block:: python
-
-      from quark.torch.quantization import (Config, QuantizationConfig,
-                                           FP8E4M3PerTensorSpec)
-
-      # Define fp8/per-tensor/static spec.
-      FP8_PER_TENSOR_SPEC = FP8E4M3PerTensorSpec(observer_method="min_max",
-          is_dynamic=False).to_quantization_spec()
-
-      # Define global quantization config, input tensors and weight apply FP8_PER_TENSOR_SPEC.
-      global_quant_config = QuantizationConfig(input_tensors=FP8_PER_TENSOR_SPEC,
-          weight=FP8_PER_TENSOR_SPEC)
-
-      # Define quantization config for kv-cache layers, output tensors apply FP8_PER_TENSOR_SPEC.
-      KV_CACHE_SPEC = FP8_PER_TENSOR_SPEC
-      kv_cache_layer_names_for_llama = ["*k_proj", "*v_proj"]
-      kv_cache_quant_config = {name :
-          QuantizationConfig(input_tensors=global_quant_config.input_tensors,
-                             weight=global_quant_config.weight,
-                             output_tensors=KV_CACHE_SPEC)
-          for name in kv_cache_layer_names_for_llama}
-      layer_quant_config = kv_cache_quant_config.copy()
-
-      EXCLUDE_LAYERS = ["lm_head"]
-      quant_config = Config(
-          global_quant_config=global_quant_config,
-          layer_quant_config=layer_quant_config,
-          kv_cache_quant_config=kv_cache_quant_config,
-          exclude=EXCLUDE_LAYERS)
-
-#. Quantize the model and export
-
-   .. code-block:: python
-
-      import torch
-      from quark.torch import ModelQuantizer, ModelExporter
-      from quark.torch.export import ExporterConfig, JsonExporterConfig
-
-      # Apply quantization.
-      quantizer = ModelQuantizer(quant_config)
-      quant_model = quantizer.quantize_model(model, calib_dataloader)
-
-      # Freeze quantized model to export.
-      freezed_model = quantizer.freeze(model)
-
-      # Define export config.
-      LLAMA_KV_CACHE_GROUP = ["*k_proj", "*v_proj"]
-      export_config = ExporterConfig(json_export_config=JsonExporterConfig())
-      export_config.json_export_config.kv_cache_group = LLAMA_KV_CACHE_GROUP
-
-      EXPORT_DIR = MODEL_ID.split("/")[1] + "-w-fp8-a-fp8-kvcache-fp8-pertensor"
-      exporter = ModelExporter(config=export_config, export_dir=EXPORT_DIR)
-      with torch.no_grad():
-          exporter.export_safetensors_model(freezed_model,
-              quant_config=quant_config, tokenizer=tokenizer)
-
-Evaluating the quantized model with vLLM
-----------------------------------------
-
-The exported Quark-quantized model can be loaded directly by vLLM for inference. You need to specify the model path and inform vLLM about the quantization method (``quantization='quark'``) and the KV cache data type (``kv_cache_dtype='fp8'``).
-Use the ``LLM`` interface to load the model:
-
-.. code-block:: python
-
-   from vllm import LLM, SamplingParamsinterface
-
-   # Sample prompts.
-   prompts = [
-       "Hello, my name is",
-       "The president of the United States is",
-       "The capital of France is",
-       "The future of AI is",
-   ]
-   # Create a sampling params object.
-   sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
-
-   # Create an LLM.
-   llm = LLM(model="Llama-2-70b-chat-hf-w-fp8-a-fp8-kvcache-fp8-pertensor",
-             kv_cache_dtype='fp8',quantization='quark')
-   # Generate texts from the prompts. The output is a list of RequestOutput objects
-   # that contain the prompt, generated text, and other information.
-   outputs = llm.generate(prompts, sampling_params)
-   # Print the outputs.
-   print("\nGenerated Outputs:\n" + "-" * 60)
-   for output in outputs:
-       prompt = output.prompt
-       generated_text = output.outputs[0].text
-       print(f"Prompt:    {prompt!r}")
-       print(f"Output:    {generated_text!r}")
-       print("-" * 60)
-
-You can also evaluate the quantized model's accuracy on standard benchmarks using the `lm-evaluation-harness <https://github.com/EleutherAI/lm-evaluation-harness>`_. Pass the necessary vLLM arguments to ``lm_eval`` via ``--model_args``.
-
-.. code-block:: shell
-
-   lm_eval --model vllm \
-     --model_args pretrained=Llama-2-70b-chat-hf-w-fp8-a-fp8-kvcache-fp8-pertensor,kv_cache_dtype='fp8',quantization='quark' \
-     --tasks gsm8k
-
-This provides a standardized way to measure the performance impact of quantization.
 .. _fine-tune-llms-gptq:
 
 GPTQ
@@ -196,7 +33,7 @@ The AutoGPTQ library implements the GPTQ algorithm.
    .. code-block:: shell
 
       # This will install pre-built wheel for a specific ROCm version.
-
+      
       pip install auto-gptq --no-build-isolation --extra-index-url https://huggingface.github.io/autogptq-index/whl/rocm573/
 
    Or, install AutoGPTQ from source for the appropriate ROCm version (for example, ROCm 6.1).
@@ -206,10 +43,10 @@ The AutoGPTQ library implements the GPTQ algorithm.
       # Clone the source code.
       git clone https://github.com/AutoGPTQ/AutoGPTQ.git
       cd AutoGPTQ
-
+      
       # Speed up the compilation by specifying PYTORCH_ROCM_ARCH to target device.
       PYTORCH_ROCM_ARCH=gfx942 ROCM_VERSION=6.1 pip install .
-
+      
       # Show the package after the installation 
 
 #. Run ``pip show auto-gptq`` to print information for the installed ``auto-gptq`` package. Its output should look like
@@ -275,7 +112,7 @@ Using GPTQ with Hugging Face Transformers
    .. code-block:: python
 
       from transformers import AutoModelForCausalLM, AutoTokenizer, GPTQConfig
-
+      
       base_model_name = " NousResearch/Llama-2-7b-hf"
       tokenizer = AutoTokenizer.from_pretrained(base_model_name)
       gptq_config = GPTQConfig(bits=4, dataset="c4", tokenizer=tokenizer)
@@ -375,10 +212,10 @@ To get started with bitsandbytes primitives, use the following code as reference
 .. code-block:: python
 
    import bitsandbytes as bnb
-
+   
    # Use Int8 Matrix Multiplication
    bnb.matmul(..., threshold=6.0)
-
+   
    # Use bitsandbytes 8-bit Optimizers
    adam = bnb.optim.Adam8bit(model.parameters(), lr=0.001, betas=(0.9, 0.995))
 
@@ -390,14 +227,14 @@ To load a Transformers model in 4-bit, set ``load_in_4bit=true`` in ``BitsAndByt
 .. code-block:: python
 
    from transformers import AutoModelForCausalLM, BitsAndBytesConfig
-
+   
    base_model_name = "NousResearch/Llama-2-7b-hf"
    quantization_config = BitsAndBytesConfig(load_in_4bit=True)
    bnb_model_4bit = AutoModelForCausalLM.from_pretrained(
            base_model_name, 
            device_map="auto", 
            quantization_config=quantization_config)
-
+   
    # Check the memory footprint with get_memory_footprint method
    print(bnb_model_4bit.get_memory_footprint())
 
@@ -406,9 +243,9 @@ To load a model in 8-bit for inference, use the ``load_in_8bit`` option.
 .. code-block:: python
 
    from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
-
+   
    base_model_name = "NousResearch/Llama-2-7b-hf"
-
+   
    tokenizer = AutoTokenizer.from_pretrained(base_model_name)
    quantization_config = BitsAndBytesConfig(load_in_8bit=True)
    tokenizer = AutoTokenizer.from_pretrained(base_model_name)
@@ -416,7 +253,7 @@ To load a model in 8-bit for inference, use the ``load_in_8bit`` option.
            base_model_name, 
            device_map="auto", 
            quantization_config=quantization_config)
-
+   
    prompt = "What is a large language model?"
    inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
    generated_ids = model.generate(**inputs)

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

@@ -20,8 +20,6 @@ training, fine-tuning, and inference. It leverages popular machine learning fram
 
 - :doc:`LLM inference frameworks <llm-inference-frameworks>`
 
-- :doc:`vLLM inference performance testing <vllm-benchmark>`
-
-- :doc:`PyTorch inference performance testing <pytorch-inference-benchmark>`
+- :doc:`Performance testing <vllm-benchmark>`
 
 - :doc:`Deploying your model <deploy-your-model>`

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

@@ -30,7 +30,7 @@ ROCm supports multiple :doc:`installation methods <rocm-install-on-linux:install
 
 * :doc:`Using the AMDGPU installer <rocm-install-on-linux:install/amdgpu-install>`
 
-* :ref:`Multi-version installation <rocm-install-on-linux:installation-types>`
+* :ref:`Multi-version installation <rocm-install-on-linux:installation-types>`.
 
 .. grid:: 1
 
@@ -59,8 +59,4 @@ images with the framework pre-installed.
 
 * :doc:`JAX for ROCm <rocm-install-on-linux:install/3rd-party/jax-install>`
 
-Next steps
-==========
-
-After installing ROCm and your desired ML libraries -- and before running AI workloads -- conduct system health benchmarks
-to test the optimal performance of your AMD hardware. See :doc:`system-health-check` to get started.
+The sections that follow in :doc:`Training a model <../training/train-a-model>` are geared for a ROCm with PyTorch installation.

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

@@ -62,52 +62,47 @@ PyTorch inference performance testing
       {% endfor %}
    {% endfor %}
 
-   System validation
-   =================
+   Getting started
+   ===============
 
-   Before running AI workloads, it's important to validate that your AMD hardware is configured
-   correctly and performing optimally.
+   Use the following procedures to reproduce the benchmark results on an
+   MI300X series accelerator with the prebuilt PyTorch Docker image.
 
-   To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
-   might hang until the periodic balancing is finalized. For more information,
-   see the :ref:`system validation steps <rocm-for-ai-system-optimization>`.
+   .. _pytorch-benchmark-get-started:
 
-   .. code-block:: shell
+   1. Disable NUMA auto-balancing.
 
-      # disable automatic NUMA balancing
-      sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
-      # check if NUMA balancing is disabled (returns 0 if disabled)
-      cat /proc/sys/kernel/numa_balancing
-      0
+      To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
+      might hang until the periodic balancing is finalized. For more information,
+      see :ref:`AMD Instinct MI300X system optimization <mi300x-disable-numa>`.
 
-   To test for optimal performance, consult the recommended :ref:`System health benchmarks
-   <rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
-   system's configuration.
+      .. code-block:: shell
 
-   Pull the Docker image
-   =====================
+         # disable automatic NUMA balancing
+         sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
+         # check if NUMA balancing is disabled (returns 0 if disabled)
+         cat /proc/sys/kernel/numa_balancing
+         0
 
    .. container:: model-doc pyt_chai1_inference
 
-      Use the following command to pull the `ROCm PyTorch Docker image <https://hub.docker.com/layers/rocm/pytorch/rocm6.2.3_ubuntu22.04_py3.10_pytorch_release_2.3.0_triton_llvm_reg_issue/images/sha256-b736a4239ab38a9d0e448af6d4adca83b117debed00bfbe33846f99c4540f79b>`_ from Docker Hub.
+      2. Use the following command to pull the `ROCm PyTorch Docker image <https://hub.docker.com/layers/rocm/pytorch/latest/images/sha256-05b55983e5154f46e7441897d0908d79877370adca4d1fff4899d9539d6c4969>`_ from Docker Hub.
 
-      .. code-block:: shell
+         .. code-block:: shell
 
-         docker pull rocm/pytorch:rocm6.2.3_ubuntu22.04_py3.10_pytorch_release_2.3.0_triton_llvm_reg_issue
+            docker pull rocm/pytorch:rocm6.2.3_ubuntu22.04_py3.10_pytorch_release_2.3.0_triton_llvm_reg_issue
 
-      .. note::
+         .. note::
 
-         The Chai-1 benchmark uses a specifically selected Docker image using ROCm 6.2.3 and PyTorch 2.3.0 to address an accuracy issue.
+            The Chai-1 benchmark uses a specifically selected Docker image using ROCm 6.2.3 and PyTorch 2.3.0 to address an accuracy issue.
 
    .. container:: model-doc pyt_clip_inference
 
-      Use the following command to pull the `ROCm PyTorch Docker image <https://hub.docker.com/layers/rocm/pytorch/latest/images/sha256-05b55983e5154f46e7441897d0908d79877370adca4d1fff4899d9539d6c4969>`_ from Docker Hub.
+      2. Use the following command to pull the `ROCm PyTorch Docker image <https://hub.docker.com/layers/rocm/pytorch/rocm6.2.3_ubuntu22.04_py3.10_pytorch_release_2.3.0_triton_llvm_reg_issue/images/sha256-b736a4239ab38a9d0e448af6d4adca83b117debed00bfbe33846f99c4540f79b>`_ from Docker Hub.
 
-      .. code-block:: shell
+         .. code-block:: shell
 
-         docker pull rocm/pytorch:latest
-
-   .. _pytorch-benchmark-get-started:
+            docker pull rocm/pytorch:latest
 
    Benchmarking
    ============

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

@@ -111,37 +111,35 @@ vLLM inference performance testing
    For information on experimental features and known issues related to ROCm optimization efforts on vLLM,
    see the developer's guide at `<https://github.com/ROCm/vllm/blob/main/docs/dev-docker/README.md>`__.
 
-   System validation
-   =================
+   Getting started
+   ===============
 
-   Before running AI workloads, it's important to validate that your AMD hardware is configured
-   correctly and performing optimally.
+   Use the following procedures to reproduce the benchmark results on an
+   MI300X accelerator with the prebuilt vLLM Docker image.
 
-   To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
-   might hang until the periodic balancing is finalized. For more information,
-   see the :ref:`system validation steps <rocm-for-ai-system-optimization>`.
+   .. _vllm-benchmark-get-started:
 
-   .. code-block:: shell
+   1. Disable NUMA auto-balancing.
 
-      # disable automatic NUMA balancing
-      sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
-      # check if NUMA balancing is disabled (returns 0 if disabled)
-      cat /proc/sys/kernel/numa_balancing
-      0
+      To optimize performance, disable automatic NUMA balancing. Otherwise, the GPU
+      might hang until the periodic balancing is finalized. For more information,
+      see :ref:`AMD Instinct MI300X system optimization <mi300x-disable-numa>`.
 
-   To test for optimal performance, consult the recommended :ref:`System health benchmarks
-   <rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
-   system's configuration.
+      .. code-block:: shell
 
-   Pull the Docker image
-   =====================
+         # disable automatic NUMA balancing
+         sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
+         # check if NUMA balancing is disabled (returns 0 if disabled)
+         cat /proc/sys/kernel/numa_balancing
+         0
 
-   Download the `ROCm vLLM Docker image <{{ unified_docker.docker_hub_url }}>`_.
-   Use the following command to pull the Docker image from Docker Hub.
+   2. Download the `ROCm vLLM Docker image <{{ unified_docker.docker_hub_url }}>`_.
 
-   .. code-block:: shell
+      Use the following command to pull the Docker image from Docker Hub.
 
-      docker pull {{ unified_docker.pull_tag }}
+      .. code-block:: shell
+
+         docker pull {{ unified_docker.pull_tag }}
 
    Benchmarking
    ============
@@ -278,7 +276,7 @@ vLLM inference performance testing
 
             * Latency benchmark
 
-              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to benchmark the latency of the {{model.model}} model on eight GPUs with the ``{{model.precision}}`` data type.
 
               .. code-block::
 
@@ -288,11 +286,11 @@ vLLM inference performance testing
 
             * Throughput benchmark
 
-              Use this command to benchmark the throughput of the {{model.model}} model on eight GPUs with ``{{model.precision}}`` precision.
+              Use this command to throughput the latency of the {{model.model}} model on eight GPUs with the ``{{model.precision}}`` data type.
 
               .. code-block:: shell
 
-                 ./vllm_benchmark_report.sh -s throughput -m {{model.model_repo}} -g 8 -d {{model.precision}}
+                 ./vllm_benchmark_report.sh -s latency -m {{model.model_repo}} -g 8 -d {{model.precision}}
 
               Find the throughput report at ``./reports_{{model.precision}}_vllm_rocm{{unified_docker.rocm_version}}/summary/{{model.model_repo.split('/', 1)[1] if '/' in model.model_repo else model.model_repo}}_throughput_report.csv``.
 

docs/how-to/rocm-for-ai/system-health-check.rst

@@ -1,104 +0,0 @@
-.. meta::
-   :description: System health checks with RVS, RCCL tests, BabelStream, and TransferBench to validate AMD hardware performance running AI workloads.
-   :keywords: gpu, accelerator, system, health, validation, bench, perf, performance, rvs, rccl, babel, mi300x, mi325x, flops, bandwidth, rbt, training, inference
-
-.. _rocm-for-ai-system-health-bench:
-
-************************
-System health benchmarks
-************************
-
-Before running AI workloads, it is important to validate that your AMD hardware is configured correctly and is performing optimally. This topic outlines several system health benchmarks you can use to test key aspects like GPU compute capabilities (FLOPS), memory bandwidth, and interconnect performance. Many of these tests are part of the ROCm Validation Suite (RVS).
-
-ROCm Validation Suite (RVS) tests
-=================================
-
-RVS provides a collection of tests, benchmarks, and qualification tools, each
-targeting a specific subsystem of the system under test. It includes tests for
-GPU stress and memory bandwidth.
-
-.. _healthcheck-install-rvs:
-
-Install ROCm Validation Suite
------------------------------
-
-To get started, install RVS. For example, on an Ubuntu system with ROCm already
-installed, run the following command:
-
-.. code-block:: shell
-
-   sudo apt update
-   sudo apt install rocm-validation-suite
-
-See the `ROCm Validation Suite installation instructions <https://rocm.docs.amd.com/projects/ROCmValidationSuite/en/latest/install/installation.html>`_,
-and `System validation tests <https://instinct.docs.amd.com/projects/system-acceptance/en/latest/mi300x/system-validation.html#system-validation-tests>`_
-in the Instinct documentation for more detailed instructions.
-
-Benchmark, stress, and qualification tests
-------------------------------------------
-
-The GPU stress test runs various GEMM computations as workloads to stress the GPU FLOPS performance and check whether it
-meets the configured target GFLOPS.
-
-Run the benchmark, stress, and qualification tests included with RVS. See the `Benchmark, stress, qualification
-<https://instinct.docs.amd.com/projects/system-acceptance/en/latest/mi300x/system-validation.html#benchmark-stress-qualification>`_
-section of the Instinct documentation for usage instructions.
-
-BabelStream test
-----------------
-
-BabelStream is a synthetic GPU benchmark based on the STREAM benchmark for
-CPUs, measuring memory transfer rates to and from global device memory.
-BabelStream tests are included with the RVS package as part of the `BABEL module
-<https://rocm.docs.amd.com/projects/ROCmValidationSuite/en/latest/conceptual/rvs-modules.html#babel-benchmark-test-babel-module>`_.
-
-For more information, see `Performance benchmarking
-<https://instinct.docs.amd.com/projects/system-acceptance/en/latest/mi300x/performance-bench.html#babelstream-benchmarking-results>`_
-in the Instinct documentation.
-
-RCCL tests
-==========
-
-The ROCm Communication Collectives Library (RCCL) enables efficient multi-GPU
-communication. The `<https://github.com/ROCm/rccl-tests>`__ suite benchmarks
-the performance and verifies the correctness of these collective operations.
-This helps ensure optimal scaling for multi-accelerator tasks.
-
-1. To get started, build RCCL-tests using the official instructions in the README at
-   `<https://github.com/ROCm/rccl-tests?tab=readme-ov-file#build>`__ or use the
-   following commands:
-
-   .. code-block:: shell
-
-      git clone https://github.com/ROCm/rccl-tests.git
-      cd rccl-tests
-      make
-
-2. Run the suggested RCCL tests -- see `RCCL benchmarking
-   <https://instinct.docs.amd.com/projects/system-acceptance/en/latest/mi300x/performance-bench.html#rccl-benchmarking-results>`_
-   in the Instinct performance benchmarking documentation for instructions.
-
-TransferBench test
-==================
-
-TransferBench is a standalone utility for benchmarking simultaneous data
-transfer performance between various devices in the system, including
-CPU-to-GPU and GPU-to-GPU (peer-to-peer). This helps identify potential
-bottlenecks in data movement between the host system and the GPUs, or between
-GPUs, which can impact end-to-end latency.
-
-.. _healthcheck-install-transferbench:
-
-1. To get started, use the instructions in the `TransferBench documentation
-   <https://rocm.docs.amd.com/projects/TransferBench/en/latest/install/install.html#install-transferbench>`_
-   or use the following commands:
-
-   .. code:: shell
-
-      git clone https://github.com/ROCm/TransferBench.git
-      cd TransferBench
-      CC=hipcc make
-
-2. Run the suggested TransferBench tests -- see `TransferBench benchmarking
-   <https://instinct.docs.amd.com/projects/system-acceptance/en/latest/mi300x/performance-bench.html#transferbench-benchmarking-results>`_
-   in the Instinct performance benchmarking documentation for instructions.

docs/how-to/rocm-for-ai/training/benchmark-docker/jax-maxtext.rst

@@ -12,7 +12,7 @@ ROCm is an optimized fork of the upstream
 `<https://github.com/AI-Hypercomputer/maxtext>`__ enabling efficient AI workloads
 on AMD MI300X series accelerators.
 
-The MaxText for ROCm training Docker (``rocm/jax-training:maxtext-v25.5``) image
+The MaxText for ROCm training Docker (``rocm/jax-training:maxtext-v25.4``) image
 provides a prebuilt environment for training on AMD Instinct MI300X and MI325X accelerators,
 including essential components like JAX, XLA, ROCm libraries, and MaxText utilities.
 It includes the following software components:
@@ -20,15 +20,15 @@ It includes the following software components:
 +--------------------------+--------------------------------+
 | Software component       | Version                        |
 +==========================+================================+
-| ROCm                     | 6.3.4                          |
+| ROCm                     | 6.3.0                          |
 +--------------------------+--------------------------------+
-| JAX                      | 0.4.35                         |
+| JAX                      | 0.4.31                         |
 +--------------------------+--------------------------------+
-| Python                   | 3.10.12                        |
+| Python                   | 3.10                           |
 +--------------------------+--------------------------------+
-| Transformer Engine       | 1.12.0.dev0+b8b92dc            |
+| Transformer Engine       | 1.12.0.dev0+f81a3eb            |
 +--------------------------+--------------------------------+
-| hipBLASLt                | 0.13.0-ae9c477a                |
+| hipBLASLt                | git78ec8622                    |
 +--------------------------+--------------------------------+
 
 Supported features and models
@@ -48,8 +48,6 @@ MaxText provides the following key features to train large language models effic
 
 The following models are pre-optimized for performance on AMD Instinct MI300X series accelerators.
 
-* Llama 3.3 70B
-
 * Llama 3.1 8B
 
 * Llama 3.1 70B
@@ -79,18 +77,11 @@ across different input sequences. Support for packed input format is planned for
 System validation
 =================
 
-Before running AI workloads, it's important to validate that your AMD hardware is configured
-correctly and performing optimally.
-
-If you have already validated your system settings, including aspects like NUMA auto-balancing, you
-can skip this step. Otherwise, complete the procedures in the :ref:`System validation and
-optimization <rocm-for-ai-system-optimization>` guide to properly configure your system settings
+If you have already validated your system settings, including NUMA
+auto-balancing, skip this step. Otherwise, complete the :ref:`system validation
+and optimization steps <train-a-model-system-validation>` to set up your system
 before starting training.
 
-To test for optimal performance, consult the recommended :ref:`System health benchmarks
-<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
-system's configuration.
-
 Environment setup
 =================
 
@@ -124,7 +115,7 @@ with RDMA, skip ahead to :ref:`amd-maxtext-download-docker`.
 
    a. Master address
 
-      Change ``localhost`` to the master node's resolvable hostname or IP address:
+      Change `localhost` to the master node's resolvable hostname or IP address:
 
       .. code-block:: bash
 
@@ -182,22 +173,20 @@ with RDMA, skip ahead to :ref:`amd-maxtext-download-docker`.
 
 .. _amd-maxtext-download-docker:
 
-Pull the Docker image
----------------------
+Download the Docker image
+-------------------------
 
 1. Use the following command to pull the Docker image from Docker Hub.
 
    .. code-block:: shell
 
-      docker pull rocm/jax-training:maxtext-v25.5
+      docker pull rocm/jax-training:maxtext-v25.4
 
-2. Use the following command to launch the Docker container. Note that the benchmarking scripts
-   used in the :ref:`following section <amd-maxtext-get-started>` automatically launch the Docker container
-   and execute the benchmark.
+2. Run the Docker container.
 
    .. code-block:: shell
 
-      docker run -it --device /dev/dri --device /dev/kfd --network host --ipc host --group-add video --cap-add SYS_PTRACE --security-opt seccomp=unconfined --privileged -v $HOME/.ssh:/root/.ssh --shm-size 128G --name maxtext_training rocm/jax-training:maxtext-v25.5
+      docker run -it --device /dev/dri --device /dev/kfd --network host --ipc host --group-add video --cap-add SYS_PTRACE --security-opt seccomp=unconfined --privileged -v $HOME/.ssh:/root/.ssh --shm-size 128G --name maxtext_training rocm/jax-training:maxtext-v25.4
 
 .. _amd-maxtext-get-started:
 
@@ -230,9 +219,7 @@ Single node training benchmarking examples
 
   Run the single node training benchmark:
 
-  .. code-block:: shell
-
-     IMAGE="rocm/jax-training:maxtext-v25.5" bash ./llama2_7b.sh
+  IMAGE="rocm/jax-training:maxtext-v25.4" bash ./llama2_7b.sh
 
 * Example 2: Single node training with Llama 2 70B
 
@@ -246,7 +233,7 @@ Single node training benchmarking examples
 
   .. code-block:: shell
 
-     IMAGE="rocm/jax-training:maxtext-v25.5" bash ./llama2_70b.sh
+     IMAGE="rocm/jax-training:maxtext-v25.4" bash ./llama2_70b.sh
 
 * Example 3: Single node training with Llama 3 8B
 
@@ -260,7 +247,7 @@ Single node training benchmarking examples
 
   .. code-block:: shell
 
-     IMAGE="rocm/jax-training:maxtext-v25.5" bash ./llama3_8b.sh
+     IMAGE="rocm/jax-training:maxtext-v25.4" bash ./llama3_8b.sh
 
 * Example 4: Single node training with Llama 3 70B
 
@@ -274,23 +261,9 @@ Single node training benchmarking examples
 
   .. code-block:: shell
 
-     IMAGE="rocm/jax-training:maxtext-v25.5" bash ./llama3_70b.sh
+     IMAGE="rocm/jax-training:maxtext-v25.4" bash ./llama3_70b.sh
 
-* Example 5: Single node training with Llama 3.3 70B
-
-  Download the benchmarking script:
-
-  .. code-block:: shell
-
-     wget https://raw.githubusercontent.com/ROCm/maxtext/refs/heads/main/benchmarks/gpu-rocm/llama3.3_70b.sh
-
-  Run the single node training benchmark:
-
-  .. code-block:: shell
-
-     IMAGE="rocm/jax-training:maxtext-v25.5" bash ./llama3.3_70b.sh
-
-* Example 6: Single node training with DeepSeek V2 16B
+* Example 5: Single node training with DeepSeek V2 16B
 
   Download the benchmarking script:
 
@@ -302,7 +275,7 @@ Single node training benchmarking examples
 
   .. code-block:: shell
 
-     IMAGE="rocm/jax-training:maxtext-v25.5" bash ./deepseek_v2_16b.sh
+     IMAGE="rocm/jax-training:maxtext-v25.4" bash ./deepseek_v2_16b.sh
 
   .. note::
 
@@ -370,26 +343,3 @@ own cluster setup.
   .. code-block:: shell
 
      sbatch -N <num_nodes> llama3_70b_multinode.sh
-
-Previous versions
-=================
-
-This table lists previous versions of the ROCm JAX MaxText Docker image for training
-performance testing. For detailed information about available models for
-benchmarking, see the version-specific documentation.
-
-.. list-table::
-   :header-rows: 1
-   :stub-columns: 1
-
-   * - Image version
-     - ROCm version
-     - JAX version
-     - Resources
-
-   * - 25.4
-     - 6.3.0
-     - 0.4.31
-     - 
-       * `Documentation <https://rocm.docs.amd.com/en/docs-6.3.3/how-to/rocm-for-ai/training/benchmark-docker/jax-maxtext.html>`_
-       * `Docker Hub <https://hub.docker.com/layers/rocm/jax-training/maxtext-v25.4/images/sha256-fb3eb71cd74298a7b3044b7130cf84113f14d518ff05a2cd625c11ea5f6a7b01>`_

docs/how-to/rocm-for-ai/training/benchmark-docker/megatron-lm.rst

@@ -103,18 +103,11 @@ popular AI models.
 System validation
 =================
 
-Before running AI workloads, it's important to validate that your AMD hardware is configured
-correctly and performing optimally.
-
-If you have already validated your system settings, including aspects like NUMA auto-balancing, you
-can skip this step. Otherwise, complete the procedures in the :ref:`System validation and
-optimization <rocm-for-ai-system-optimization>` guide to properly configure your system settings
+If you have already validated your system settings, including NUMA
+auto-balancing, skip this step. Otherwise, complete the :ref:`system validation
+and optimization steps <train-a-model-system-validation>` to set up your system
 before starting training.
 
-To test for optimal performance, consult the recommended :ref:`System health benchmarks
-<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
-system's configuration.
-
 .. _mi300x-amd-megatron-lm-training:
 
 Environment setup

docs/how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry.rst

@@ -1,175 +0,0 @@
-.. meta::
-   :description: How to train a model using LLM Foundry for ROCm.
-   :keywords: ROCm, AI, LLM, train, PyTorch, torch, Llama, flux, tutorial, docker
-
-******************************************
-Training MPT-30B with LLM Foundry and ROCm
-******************************************
-
-MPT-30B is a 30-billion parameter decoder-style transformer-based model from
-the Mosaic Pretrained Transformer (MPT) family -- learn more about it in
-MosaicML's research blog `MPT-30B: Raising the bar for open-source foundation
-models <https://www.databricks.com/blog/mpt-30b>`_.
-
-ROCm and `<https://github.com/ROCm/MAD>`__ provide a pre-configured training
-environment for the MPT-30B model using the ``rocm/pytorch-training:v25.5``
-base `Docker image <https://hub.docker.com/layers/rocm/pytorch-training/v25.5/images/sha256-d47850a9b25b4a7151f796a8d24d55ea17bba545573f0d50d54d3852f96ecde5>`_
-and the `LLM Foundry <https://github.com/mosaicml/llm-foundry>`_ framework.
-This environment packages the following software components to train
-on AMD Instinct MI300X series accelerators:
-
-+--------------------------+--------------------------------+
-| Software component       | Version                        |
-+==========================+================================+
-| ROCm                     | 6.3.4                          |
-+--------------------------+--------------------------------+
-| PyTorch                  | 2.7.0a0+git6374332             |
-+--------------------------+--------------------------------+
-| Flash Attention          | 3.0.0.post1                    |
-+--------------------------+--------------------------------+
-
-Using this image, you can build, run, and test the training process
-for MPT-30B with access to detailed logs and performance metrics.
-
-System validation
-=================
-
-Before running AI workloads, it's important to validate that your AMD hardware is configured
-correctly and performing optimally.
-
-If you have already validated your system settings, including aspects like NUMA auto-balancing, you
-can skip this step. Otherwise, complete the procedures in the :ref:`System validation and
-optimization <rocm-for-ai-system-optimization>` guide to properly configure your system settings
-before starting training.
-
-To test for optimal performance, consult the recommended :ref:`System health benchmarks
-<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
-system's configuration.
-
-Getting started
-===============
-
-The following procedures help you set up the training environment in a
-reproducible Docker container. This training environment is tailored for
-training MPT-30B using LLM Foundry and the specific model configurations outlined.
-Other configurations and run conditions outside those described in this
-document are not validated.
-
-.. tab-set::
-
-   .. tab-item:: MAD-integrated benchmarking
-
-      On your host machine, clone the ROCm Model Automation and Dashboarding
-      (`<https://github.com/ROCm/MAD>`__) repository to a local directory and
-      install the required packages.
-
-      .. code-block:: shell
-
-         git clone https://github.com/ROCm/MAD
-         cd MAD
-         pip install -r requirements.txt
-
-      Use this command to initiate the MPT-30B training benchmark.
-
-      .. code-block:: shell
-
-         python3 tools/run_models.py --tags pyt_mpt30b_training --keep-model-dir --live-output --clean-docker-cache
-
-      .. tip::
-
-         If you experience data download failures, set the
-         ``MAD_SECRETS_HFTOKEN`` variable to your Hugging Face access token. See
-         `User access tokens <https://huggingface.co/docs/hub/security-tokens>`_
-         for details.
-
-         .. code-block:: shell
-
-            export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
-
-      .. note::
-
-         For improved performance (training throughput), consider enabling TunableOp.
-         By default, ``pyt_mpt30b_training`` runs with TunableOp disabled. To enable it,
-         run ``tools/run_models.py`` with the ``--tunableop on`` argument or edit the
-         ``models.json`` configuration before running training.
-
-         Although this might increase the initial training time, it can result in a performance gain.
-
-   .. tab-item:: Standalone benchmarking
-
-      To set up the training environment, clone the
-      `<https://github.com/ROCm/MAD>`__ repo and build the Docker image. In
-      this snippet, the image is named ``mosaic_mpt30_image``.
-
-      .. code-block:: shell
-
-         git clone https://github.com/ROCm/MAD
-         cd MAD
-
-         docker build --build-arg MAD_SYSTEM_GPU_ARCHITECTURE=gfx942 -f docker/pyt_mpt30b_training.ubuntu.amd.Dockerfile -t mosaic_mpt30_image .
-
-      Start a ``mosaic_mpt30_image`` container using the following command.
-
-      .. code-block:: shell
-
-         docker run -it --device=/dev/kfd --device=/dev/dri --group-add=video --ipc=host --shm-size=8G mosaic_mpt30_image
-
-      In the Docker container, clone the `<https://github.com/ROCm/MAD>`__
-      repository and navigate to the benchmark scripts directory at
-      ``/workspace/MAD/scripts/pyt_mpt30b_training``.
-
-      .. code-block:: shell
-
-         git clone https://github.com/ROCm/MAD
-         cd MAD/scripts/pyt_mpt30b_training
-
-      To initiate the training process, use the following command. This script uses the hyperparameters defined in
-      ``mpt-30b-instruct.yaml``.
-
-      .. code-block:: shell
-
-         source run.sh
-
-      .. note::
-
-         For improved performance (training throughput), consider enabling TunableOp.
-         To enable it, add the ``--tunableop on`` flag.
-
-         .. code-block:: shell
-
-            source run.sh --tunableop on
-
-         Although this might increase the initial training time, it can result in a performance gain.
-
-Interpreting the output
-=======================
-
-The training output will be displayed in the terminal and simultaneously saved
-to the ``output.txt`` file in the current directory. Key performance metrics will
-also be extracted and appended to the ``perf_pyt_mpt30b_training.csv`` file.
-
-Key performance metrics include:
-
-- Training logs: Real-time display of loss metrics, accuracy, and training progress.
-
-- Model checkpoints: Periodically saved model snapshots for potential resume or evaluation.
-
-- Performance metrics: Detailed summaries of training speed and training loss metrics.
-
-  - Performance (throughput/samples_per_sec)
-
-    Overall throughput, measuring the total samples processed per second. Higher values indicate better hardware utilization.
-
-  - Performance per device (throughput/samples_per_sec)
-
-    Throughput on a per-device basis, showing how each GPU or CPU is performing.
-
-  - Language Cross Entropy (metrics/train/LanguageCrossEntropy)
-
-    Measures prediction accuracy. Lower cross entropy suggests the model’s output is closer to the expected distribution.
-
-  - Training loss (loss/train/total)
-
-    Overall training loss. A decreasing trend indicates the model is learning effectively.
-
-

docs/how-to/rocm-for-ai/training/benchmark-docker/pytorch-training.rst

@@ -77,18 +77,11 @@ popular AI models.
 System validation
 =================
 
-Before running AI workloads, it's important to validate that your AMD hardware is configured
-correctly and performing optimally.
-
-If you have already validated your system settings, including aspects like NUMA auto-balancing, you
-can skip this step. Otherwise, complete the procedures in the :ref:`System validation and
-optimization <rocm-for-ai-system-optimization>` guide to properly configure your system settings
+If you have already validated your system settings, including NUMA
+auto-balancing, skip this step. Otherwise, complete the :ref:`system validation
+and optimization steps <train-a-model-system-validation>` to set up your system
 before starting training.
 
-To test for optimal performance, consult the recommended :ref:`System health benchmarks
-<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
-system's configuration.
-
 This Docker image is optimized for specific model configurations outlined
 below. Performance can vary for other training workloads, as AMD 
 doesn’t validate configurations and run conditions outside those described.
@@ -450,7 +443,7 @@ benchmarking, see the version-specific documentation.
      - 6.3.0
      - 2.7.0a0+git637433
      - 
-       * `Documentation <https://rocm.docs.amd.com/en/docs-6.3.3/how-to/rocm-for-ai/training/benchmark-docker/pytorch-training.html>`_
+       * `Documentation <https://rocm.docs.amd.com/en/docs-6.3.4/how-to/rocm-for-ai/training/benchmark-docker/pytorch-training.html>`_
        * `Docker Hub <https://hub.docker.com/layers/rocm/pytorch-training/v25.4/images/sha256-fa98a9aa69968e654466c06f05aaa12730db79b48b113c1ab4f7a5fe6920a20b>`_
 
    * - v25.3

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

@@ -21,12 +21,8 @@ In this guide, you'll learn about:
 
 - Training a model
 
-  - :doc:`With Megatron-LM <benchmark-docker/megatron-lm>`
+  - :doc:`Train a model with Megatron-LM <benchmark-docker/megatron-lm>`
 
-  - :doc:`With PyTorch <benchmark-docker/pytorch-training>`
-
-  - :doc:`With JAX MaxText <benchmark-docker/jax-maxtext>`
-
-  - :doc:`With LLM Foundry <benchmark-docker/mpt-llm-foundry>`
+  - :doc:`Train a model with PyTorch <benchmark-docker/pytorch-training>`
 
 - :doc:`Scaling model training <scale-model-training>`

docs/how-to/rocm-for-ai/training/prerequisite-system-validation.rst

@@ -5,13 +5,12 @@
    :keywords: ROCm, AI, LLM, train, megatron, Llama, tutorial, docker, torch, pytorch, jax
 
 .. _train-a-model-system-validation:
-.. _rocm-for-ai-system-optimization:
 
-**********************************************************
-Prerequisite system validation before running AI workloads
-**********************************************************
+**********************************************
+Prerequisite system validation before training
+**********************************************
 
-Complete the following system validation and optimization steps to set up your system before starting training and inference.
+Complete the following system validation and optimization steps to set up your system before starting training.
 
 Disable NUMA auto-balancing
 ---------------------------
@@ -27,8 +26,7 @@ the output is ``1``, run the following command to disable NUMA auto-balancing.
 
    sudo sh -c 'echo 0 > /proc/sys/kernel/numa_balancing'
 
-See `Disable NUMA auto-balancing <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html#disable-numa-auto-balancing>`_
-in the Instinct documentation for more information.
+See :ref:`mi300x-disable-numa` for more information.
 
 Hardware verification with ROCm
 -------------------------------
@@ -44,8 +42,7 @@ Run the command:
 
    rocm-smi --setperfdeterminism 1900
 
-See `Hardware verfication for ROCm <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html#hardware-verification-with-rocm>`_
-in the Instinct documentation for more information.
+See :ref:`mi300x-hardware-verification-with-rocm` for more information.
 
 RCCL Bandwidth Test for multi-node setups
 -----------------------------------------

docs/reference/api-libraries.md

@@ -45,7 +45,6 @@
 (communication-libraries)=
 
 * {doc}`RCCL <rccl:index>`
-* {doc}`rocSHMEM <rocSHMEM:index>`
 :::
 
 :::{grid-item-card} Math

docs/reference/gpu-arch-specs.rst

@@ -296,7 +296,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -314,7 +314,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -332,7 +332,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -350,7 +350,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -368,7 +368,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -386,7 +386,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -404,7 +404,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -422,7 +422,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -440,7 +440,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -519,7 +519,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -537,7 +537,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -555,7 +555,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -573,7 +573,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -591,7 +591,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 768
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -609,7 +609,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 11
           - 0
         *
@@ -627,7 +627,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -645,7 +645,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -663,7 +663,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -681,7 +681,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -699,7 +699,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -717,7 +717,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -735,7 +735,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -753,7 +753,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -771,7 +771,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *
@@ -789,7 +789,7 @@ For more information about ROCm hardware compatibility, see the ROCm `Compatibil
           - 16
           - 32
           - 512
-          - 32
+          - 16
           - 10
           - 3
         *

docs/sphinx/_toc.yml.in

@@ -12,14 +12,14 @@ subtrees:
   - file: compatibility/compatibility-matrix.rst
     title: Compatibility matrix
     entries:
-    - url: https://rocm.docs.amd.com/projects/install-on-linux/en/${branch}/reference/system-requirements.html
+    - url: https://rocm.docs.amd.com/projects/install-on-linux/en/latest/reference/system-requirements.html
       title: Linux system requirements
     - url: https://rocm.docs.amd.com/projects/install-on-windows/en/${branch}/reference/system-requirements.html
       title: Windows system requirements
 
 - caption: Install
   entries:
-  - url: https://rocm.docs.amd.com/projects/install-on-linux/en/${branch}/
+  - url: https://rocm.docs.amd.com/projects/install-on-linux/en/latest/
     title: ROCm on Linux
   - url: https://rocm.docs.amd.com/projects/install-on-windows/en/${branch}/
     title: HIP SDK on Windows
@@ -36,10 +36,6 @@ subtrees:
     title: Use ROCm for AI
     subtrees:
     - entries:
-      - file: how-to/rocm-for-ai/install.rst
-        title: Installation
-      - file: how-to/rocm-for-ai/system-health-check.rst
-        title: System health benchmarks
       - file: how-to/rocm-for-ai/training/index.rst
         title: Training
         subtrees:
@@ -50,8 +46,6 @@ subtrees:
             title: Train a model with PyTorch
           - file: how-to/rocm-for-ai/training/benchmark-docker/jax-maxtext
             title: Train a model with JAX MaxText
-          - file: how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry
-            title: Train a model with LLM Foundry
           - file: how-to/rocm-for-ai/training/scale-model-training.rst
             title: Scale model training
 
@@ -74,6 +68,8 @@ subtrees:
         title: Inference
         subtrees:
         - entries:
+          - file: how-to/rocm-for-ai/inference/install.rst
+            title: Installation
           - file: how-to/rocm-for-ai/inference/hugging-face-models.rst
             title: Run models from Hugging Face
           - file: how-to/rocm-for-ai/inference/llm-inference-frameworks.rst

docs/sphinx/requirements.txt

@@ -2,55 +2,54 @@
 # This file is autogenerated by pip-compile with Python 3.10
 # by the following command:
 #
-#    pip-compile requirements.in
+#    pip-compile docs/sphinx/requirements.in
 #
 accessible-pygments==0.0.5
     # via pydata-sphinx-theme
 alabaster==1.0.0
     # via sphinx
-appnope==0.1.4
-    # via ipykernel
 asttokens==3.0.0
     # via stack-data
-attrs==25.3.0
+attrs==25.1.0
     # via
     #   jsonschema
     #   jupyter-cache
     #   referencing
-babel==2.17.0
+babel==2.16.0
     # via
     #   pydata-sphinx-theme
     #   sphinx
-beautifulsoup4==4.13.3
+beautifulsoup4==4.12.3
     # via pydata-sphinx-theme
-breathe==4.36.0
+breathe==4.35.0
     # via rocm-docs-core
-certifi==2025.1.31
+certifi==2024.8.30
     # via requests
 cffi==1.17.1
     # via
     #   cryptography
     #   pynacl
-charset-normalizer==3.4.1
+charset-normalizer==3.4.0
     # via requests
-click==8.1.8
+click==8.1.7
     # via
     #   jupyter-cache
     #   sphinx-external-toc
 comm==0.2.2
     # via ipykernel
-cryptography==44.0.2
+cryptography==44.0.1
     # via pyjwt
-debugpy==1.8.14
+debugpy==1.8.12
     # via ipykernel
-decorator==5.2.1
+decorator==5.1.1
     # via ipython
 defusedxml==0.7.1
     # via sphinxcontrib-datatemplates
-deprecated==1.2.18
+deprecated==1.2.15
     # via pygithub
 docutils==0.21.2
     # via
+    #   breathe
     #   myst-parser
     #   pydata-sphinx-theme
     #   sphinx
@@ -58,14 +57,16 @@ exceptiongroup==1.2.2
     # via ipython
 executing==2.2.0
     # via stack-data
-fastjsonschema==2.21.1
+fastjsonschema==2.20.0
     # via
     #   nbformat
     #   rocm-docs-core
-gitdb==4.0.12
+gitdb==4.0.11
     # via gitpython
-gitpython==3.1.44
+gitpython==3.1.43
     # via rocm-docs-core
+greenlet==3.1.1
+    # via sqlalchemy
 idna==3.10
     # via requests
 imagesize==1.4.1
@@ -76,7 +77,7 @@ importlib-metadata==8.6.1
     #   myst-nb
 ipykernel==6.29.5
     # via myst-nb
-ipython==8.35.0
+ipython==8.31.0
     # via
     #   ipykernel
     #   myst-nb
@@ -116,9 +117,9 @@ mdit-py-plugins==0.4.2
     # via myst-parser
 mdurl==0.1.2
     # via markdown-it-py
-myst-nb==1.2.0
+myst-nb==1.1.2
     # via rocm-docs-core
-myst-parser==4.0.1
+myst-parser==4.0.0
     # via myst-nb
 nbclient==0.10.2
     # via
@@ -134,17 +135,16 @@ nest-asyncio==1.6.0
 packaging==24.2
     # via
     #   ipykernel
-    #   pydata-sphinx-theme
     #   sphinx
 parso==0.8.4
     # via jedi
 pexpect==4.9.0
     # via ipython
-platformdirs==4.3.7
+platformdirs==4.3.6
     # via jupyter-core
 prompt-toolkit==3.0.50
     # via ipython
-psutil==7.0.0
+psutil==6.1.1
     # via ipykernel
 ptyprocess==0.7.0
     # via pexpect
@@ -152,19 +152,19 @@ pure-eval==0.2.3
     # via stack-data
 pycparser==2.22
     # via cffi
-pydata-sphinx-theme==0.15.4
+pydata-sphinx-theme==0.16.0
     # via
     #   rocm-docs-core
     #   sphinx-book-theme
-pygithub==2.6.1
+pygithub==2.5.0
     # via rocm-docs-core
-pygments==2.19.1
+pygments==2.18.0
     # via
     #   accessible-pygments
     #   ipython
     #   pydata-sphinx-theme
     #   sphinx
-pyjwt[crypto]==2.10.1
+pyjwt[crypto]==2.10.0
     # via pygithub
 pynacl==1.5.0
     # via pygithub
@@ -178,7 +178,7 @@ pyyaml==6.0.2
     #   rocm-docs-core
     #   sphinx-external-toc
     #   sphinxcontrib-datatemplates
-pyzmq==26.4.0
+pyzmq==26.2.0
     # via
     #   ipykernel
     #   jupyter-client
@@ -192,13 +192,13 @@ requests==2.32.3
     #   sphinx
 rocm-docs-core==1.18.2
     # via -r requirements.in
-rpds-py==0.24.0
+rpds-py==0.22.3
     # via
     #   jsonschema
     #   referencing
 six==1.17.0
     # via python-dateutil
-smmap==5.0.2
+smmap==5.0.1
     # via gitdb
 snowballstemmer==2.2.0
     # via sphinx
@@ -220,7 +220,7 @@ sphinx==8.1.3
     #   sphinx-sitemap
     #   sphinxcontrib-datatemplates
     #   sphinxcontrib-runcmd
-sphinx-book-theme==1.1.4
+sphinx-book-theme==1.1.3
     # via rocm-docs-core
 sphinx-copybutton==0.5.2
     # via rocm-docs-core
@@ -228,7 +228,7 @@ sphinx-design==0.6.1
     # via rocm-docs-core
 sphinx-external-toc==1.0.1
     # via rocm-docs-core
-sphinx-notfound-page==1.1.0
+sphinx-notfound-page==1.0.4
     # via rocm-docs-core
 sphinx-reredirects==0.1.6
     # via -r requirements.in
@@ -250,13 +250,13 @@ sphinxcontrib-runcmd==0.2.0
     # via sphinxcontrib-datatemplates
 sphinxcontrib-serializinghtml==2.0.0
     # via sphinx
-sqlalchemy==2.0.40
+sqlalchemy==2.0.37
     # via jupyter-cache
 stack-data==0.6.3
     # via ipython
 tabulate==0.9.0
     # via jupyter-cache
-tomli==2.2.1
+tomli==2.1.0
     # via sphinx
 tornado==6.4.2
     # via
@@ -272,22 +272,21 @@ traitlets==5.14.3
     #   matplotlib-inline
     #   nbclient
     #   nbformat
-typing-extensions==4.13.2
+typing-extensions==4.12.2
     # via
-    #   beautifulsoup4
     #   ipython
     #   myst-nb
     #   pydata-sphinx-theme
     #   pygithub
     #   referencing
     #   sqlalchemy
-urllib3==2.4.0
+urllib3==2.2.3
     # via
     #   pygithub
     #   requests
 wcwidth==0.2.13
     # via prompt-toolkit
-wrapt==1.17.2
+wrapt==1.17.0
     # via deprecated
 zipp==3.21.0
     # via importlib-metadata

docs/what-is-rocm.rst

@@ -10,7 +10,7 @@ ROCm is a software stack, composed primarily of open-source software, that
 provides the tools for programming AMD Graphics Processing Units (GPUs), from
 low-level kernels to high-level end-user applications.
 
-.. image:: data/rocm-software-stack-6_4_0.jpg
+.. image:: data/rocm-software-stack-6_3_2.jpg
   :width: 800
   :alt: AMD's ROCm software stack and enabling technologies.
   :align: center
@@ -52,7 +52,6 @@ Communication
   :header: "Component", "Description"
 
   ":doc:`RCCL <rccl:index>`", "Standalone library that provides multi-GPU and multi-node collective communication primitives"
-  ":doc:`rocSHMEM <rocSHMEM:index>`", "An intra-kernel networking library that provides GPU-centric networking through an OpenSHMEM-like interface"
 
 Math
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

tools/rocm-build/ROCm.mk

@@ -12,17 +12,36 @@ RELEASE_FLAG?=-r
 ASAN_DEP:=
 ifeq (${ENABLE_ADDRESS_SANITIZER},true)
 	ASAN_DEP=lightning
-	SANITIZER_FLAG=-a
+	SANITIZER_FLAG=--address_sanitizer
+endif
+
+# Set STATIC_FLAG for static builds
+ifeq (${ENABLE_STATIC_BUILDS},true)
+	STATIC_FLAG=-s
 endif
 
 export INFRA_REPO:=ROCm/tools/rocm-build
 
+# # commannds to be run at makefile read time
+# # should only output $$OUT_DIR to stdout
+# # In an ideal world this would be another target
+# define INITBUILD
+# source ${INFRA_REPO}/envsetup.sh >/dev/null 2>&1;
+# [ -w "$${ROCM_INSTALL_PATH}" ] || sudo mkdir -p -m 775 "$${ROCM_INSTALL_PATH}" ;
+# sudo chown "$$(id -u):$$(id -g)" "$${ROCM_INSTALL_PATH}" "/home/$$(id -un)" ;
+# mkdir -p ${HOME}/.ccache ;
+# echo $${OUT_DIR} ;
+# endef
+
 OUT_DIR:=$(shell . ${INFRA_REPO}/envsetup.sh >/dev/null 2>&1 ; echo $${OUT_DIR})
 ROCM_INSTALL_PATH:=$(shell . ${INFRA_REPO}/envsetup.sh >/dev/null 2>&1 ; echo $${ROCM_INSTALL_PATH})
 
 $(info OUT_DIR=${OUT_DIR})
 $(info ROCM_INSTALL_PATH=${ROCM_INSTALL_PATH})
 
+# define SILENT to be empty to see the runner invocation
+SILENT?= @
+
 # -------------------------------------------------------------------------
 # Internal stuff. Could be put in a different file to hide it.
 # Internal macros, they need to be defined before being used.
@@ -47,7 +66,8 @@ endif
 # It should not be needed.
 define adddep =
 $(strip $(call peval,components+= $(1) $(2))
-$(foreach comp,$(strip $2),$(call peval,${OUT_DIR}/logs/${1}: ${OUT_DIR}/logs/${comp}))
+$(call peval,$(1)_DEPS += $(2))
+$(foreach comp,$(strip $2),$(call peval,${OUT_DIR}/logs/${1}.txt: ${OUT_DIR}/logs/${comp}.txt))
 )
 endef
 # End of internal stuff that is needed at the start of the file
@@ -67,6 +87,7 @@ endef
 
 $(call adddep,amd_smi_lib,${ASAN_DEP})
 $(call adddep,aqlprofile,${ASAN_DEP} rocr)
+$(call adddep,aqlprofiletest,lightning rocminfo aqlprofile opencl_on_rocclr hip_on_rocclr)
 $(call adddep,comgr,lightning devicelibs)
 $(call adddep,dbgapi,rocr comgr)
 $(call adddep,devicelibs,lightning)
@@ -83,27 +104,29 @@ $(call adddep,rocm-core,${ASAN_DEP})
 $(call adddep,rocm-gdb,dbgapi)
 $(call adddep,rocminfo,${ASAN_DEP} rocr)
 $(call adddep,rocprofiler-register,${ASAN_DEP})
-$(call adddep,rocprofiler-sdk,${ASAN_DEP} rocr aqlprofile opencl_on_rocclr hip_on_rocclr comgr)
+$(call adddep,rocprofiler-sdk,${ASAN_DEP} rocr aqlprofile opencl_on_rocclr hip_on_rocclr comgr rccl rocdecode)
 $(call adddep,rocprofiler-systems,${ASAN_DEP} hipcc rocr hip_on_rocclr rocm_smi_lib rocprofiler roctracer rocprofiler-sdk)
 $(call adddep,rocprofiler,${ASAN_DEP} rocr roctracer aqlprofile opencl_on_rocclr hip_on_rocclr comgr)
 $(call adddep,rocprofiler-compute,${ASAN_DEP})
 $(call adddep,rocr,${ASAN_DEP} lightning rocm_smi_lib devicelibs rocprofiler-register)
 $(call adddep,rocr_debug_agent,${ASAN_DEP} hip_on_rocclr rocr dbgapi)
+$(call adddep,rocrsamples,lightning devicelibs rocr )
 $(call adddep,roctracer,${ASAN_DEP} rocr hip_on_rocclr)
 
+
 # rocm-dev points to all possible last finish components of Stage1 build.
-rocm-dev-components :=amd_smi_lib aqlprofile comgr dbgapi devicelibs hip_on_rocclr hipcc hipify_clang \
+rocm-dev-components :=amd_smi_lib aqlprofile aqlprofiletest comgr dbgapi devicelibs hip_on_rocclr hipcc hipify_clang \
 	lightning rocprofiler-compute opencl_on_rocclr openmp_extras rocm_bandwidth_test rocm_smi_lib \
 	rocm-cmake rocm-core rocm-gdb rocminfo rocprofiler-register rocprofiler-sdk rocprofiler-systems \
-	rocprofiler rocr rocr_debug_agent roctracer
-$(call adddep,rocm-dev,$(filter-out ${NOBUILD} kernel_ubuntu,${rocm-dev-components}))
+	rocprofiler rocr rocr_debug_agent rocrsamples roctracer
+$(call adddep,rocm-dev,$(filter-out ${NOBUILD},${rocm-dev-components}))
 
 $(call adddep,amdmigraphx,hip_on_rocclr half rocblas miopen-hip lightning hipcc hiptensor)
 $(call adddep,composable_kernel,lightning hipcc hip_on_rocclr rocm-cmake)
 $(call adddep,half,rocm-cmake)
 $(call adddep,hipblas-common,lightning)
 $(call adddep,hipblas,hip_on_rocclr rocblas rocsolver lightning hipcc)
-$(call adddep,hipblaslt,hip_on_rocclr openmp_extras lightning hipcc hipblas-common rocm-dev)
+$(call adddep,hipblaslt,hip_on_rocclr openmp_extras lightning hipcc hipblas-common roctracer)
 $(call adddep,hipcub,hip_on_rocclr rocprim lightning hipcc)
 $(call adddep,hipfft,hip_on_rocclr openmp_extras rocfft rocrand hiprand lightning hipcc)
 $(call adddep,hipfort,rocblas hipblas rocsparse hipsparse rocfft hipfft rocrand hiprand rocsolver hipsolver lightning hipcc)
@@ -113,26 +136,34 @@ $(call adddep,hipsparse,hip_on_rocclr rocsparse lightning hipcc)
 $(call adddep,hipsparselt,hip_on_rocclr hipsparse lightning hipcc openmp_extras)
 $(call adddep,hiptensor,hip_on_rocclr composable_kernel lightning hipcc)
 $(call adddep,miopen-deps,lightning hipcc)
-$(call adddep,miopen-hip,composable_kernel half hip_on_rocclr miopen-deps hipblas hipblaslt rocrand roctracer lightning hipcc)
+$(call adddep,miopen-hip,rocm-core composable_kernel half hip_on_rocclr miopen-deps hipblas hipblaslt rocrand roctracer lightning hipcc)
 $(call adddep,mivisionx,amdmigraphx miopen-hip rpp lightning hipcc)
 $(call adddep,rccl,rocm-core hip_on_rocclr rocr lightning hipcc rocm_smi_lib hipify_clang)
-$(call adddep,rdc,rocm_smi_lib rocprofiler rocmvalidationsuite)
+$(call adddep,rdc,amd_smi_lib rocprofiler-sdk rocm_smi_lib rocprofiler  rocmvalidationsuite)
 $(call adddep,rocalution,rocblas rocsparse rocrand lightning hipcc)
-$(call adddep,rocblas,hip_on_rocclr openmp_extras lightning hipcc hipblaslt)
+$(call adddep,rocblas,rocminfo hip_on_rocclr openmp_extras lightning hipcc hipblaslt)
 $(call adddep,rocal,mivisionx)
-$(call adddep,rocdecode,hip_on_rocclr lightning hipcc amdmigraphx)
+$(call adddep,rocdecode,hip_on_rocclr lightning hipcc)
 $(call adddep,rocfft,hip_on_rocclr rocrand hiprand lightning hipcc openmp_extras)
-$(call adddep,rocjpeg,hip_on_rocclr lightning hipcc rocm-dev)
+$(call adddep,rocjpeg,hip_on_rocclr lightning hipcc)
 $(call adddep,rocmvalidationsuite,hip_on_rocclr rocr hipblas hiprand hipblaslt rocm-core lightning hipcc rocm_smi_lib)
 $(call adddep,rocprim,hip_on_rocclr lightning hipcc)
 $(call adddep,rocrand,hip_on_rocclr lightning hipcc)
+$(call adddep,rocshmem,rccl )
 $(call adddep,rocsolver,hip_on_rocclr rocblas rocsparse rocprim lightning hipcc)
 $(call adddep,rocsparse,hip_on_rocclr rocprim lightning hipcc)
 $(call adddep,rocthrust,hip_on_rocclr rocprim lightning hipcc)
 $(call adddep,rocwmma,hip_on_rocclr rocblas lightning hipcc rocm-cmake rocm_smi_lib)
 $(call adddep,rpp,half lightning hipcc openmp_extras)
 $(call adddep,transferbench,hip_on_rocclr lightning hipcc)
+ifneq ($(filter rocm-dev upload-rocm-dev, ${MAKECMDGOALS}),)
+	components = $(rocm-dev-components)
+endif
+$(call adddep,rocm,$(filter-out ${NOBUILD} rocm,${components}))
 
+ifeq ($(DISTRO_NAME),rhel)
+    WHL_GEN :=
+endif
 
 # -------------------------------------------------------------------------
 # The rest of the file is internal
@@ -165,7 +196,7 @@ ifeq (${toplevel},)
 define toplevel =
 
 # The "target" make, this builds the package if it is out of date
-T_$1: ${OUT_DIR}/logs/$1 FRC
+T_$1: ${OUT_DIR}/logs/$1.txt FRC
 	:              $1 built
 
 # The "upload" for $1, it uploads the packages for $1 to the central storage
@@ -176,25 +207,26 @@ U_$1: T_$1 FRC
 # The "clean" for $1, it just marks the target as not existing so it will be built
 # in the future.
 C_$1: FRC
-	rm -f ${OUT_DIR}/logs/$1 ${OUT_DIR}/logs/$1.repackaged
+	rm -f ${OUT_DIR}/logs/$1.txt ${OUT_DIR}/logs/$1.repackaged
 
-# parallel build {
-${OUT_DIR}/logs/$1: | ${OUT_DIR}/logs
-ifneq ($(wildcard ${OUT_DIR}/logs/$1.repackaged),)
+# parallel build
+${OUT_DIR}/logs/$1.txt: | ${OUT_DIR}/logs
+ifneq ($(wildcard ${OUT_DIR}/logs/$1.repackaged),) # {
 	@echo  Skipping build of $1 as it has already been repackaged
-	cat $$@.repackaged > $$@
-	rm -f $$@.repackaged
+	cat $${@:.txt=.repackaged} > $$@
+	rm -f $${@:.txt=.repackaged}
 else # } {
 	@echo  $1 started due to $$? | sed "s:${OUT_DIR}/logs/::g"
 # Build in a subshell so we get the time output
 # Pass in jobserver info using the RMAKE variable
-	${RMAKE}@( if set -x && source $${INFRA_REPO}/envsetup.sh && \
-	rm -f $$@.errors $$@ $$@.repackaged && \
-	$${INFRA_REPO}/build_$1.sh -c && \
-	time bash -x $${INFRA_REPO}/build_$1.sh $${RELEASE_FLAG} $${SANITIZER_FLAG} && $${INFRA_REPO}/post_inst_pkg.sh "$1" ; \
-	then mv $$@.inprogress $$@ ; \
-	else mv $$@.inprogress $$@.errors ; echo Error in $1 >&2 ; exit 1 ;\
-	fi ) > $$@.inprogress 2>&1
+# Allow project specific flags e..g. ROCMBUILD_lightning.
+	${RMAKE}${SILENT}( if set -x && source $${INFRA_REPO}/envsetup.sh && \
+	rm -f $${@:$1.txt=1.Errors.$1.txt} $$@ $${@:.txt=.repackaged} && \
+	$${INFRA_REPO}/runner $1 $${RELEASE_FLAG} $${SANITIZER_FLAG} $${STATIC_FLAG} ${ROCMBUILD_$1}; \
+	then mv $${@:$1.txt=2.Inprogress.$1.txt} $$@ ; \
+	else mv $${@:$1.txt=2.Inprogress.$1.txt} $${@:$1.txt=1.Errors.$1.txt} ;\
+		echo Error in $1 >&2 ; exit 1 ;\
+	fi ) > $${@:$1.txt=2.Inprogress.$1.txt} 2>&1
 endif # }
 
 # end of toplevel macro
@@ -227,22 +259,45 @@ upload-rocm-dev: $(addprefix U_,$(filter-out ${NOBUILD},${components}))
 rocm-dev: $(addprefix T_,$(filter-out ${NOBUILD},${components}))
 	@echo rocm-dev built
 
+ifeq ($(DISTRO_NAME),almalinux)
+	@sudo chmod -R 777 "/home/builder"
+endif
+
+# This code is broken. It stops us exiting a container and
+# starting a new one and continueing the build. The attempt
+# is to have run-once code.
 ${OUT_DIR}/logs:
 	sudo mkdir -p -m 775 "${ROCM_INSTALL_PATH}" && \
 	sudo chown -R "$(shell id -u):$(shell id -g)" "/opt"
-	sudo chown -R "$(shell id -u):$(shell id -g)" "/home/$(shell id -un)"
+	sudo chown "$(shell id -u):$(shell id -g)" "/home/$(shell id -un)"
 	mkdir -p "${@}"
 	mkdir -p ${HOME}/.ccache
 
 ##help clean: remove the output directory and recreate it
 clean:
 	[ -n "${OUT_DIR}" ] && rm -rf "${OUT_DIR}"
-	mkdir -p ${OUT_DIR}/logs
+#	mkdir -p ${OUT_DIR}/logs
 
 .SECONDARY: ${components:%=${OUT_DIR}/logs/%}
 
 .PHONY: all clean repack help list_components
 
+# get_all_deps: Recursively get all dependencies for a given component.
+# Usage: $(call get_all_deps,component_name,)
+# - component_name: The name of the component to get dependencies for.
+# - The second parameter is an internal parameter used to track already
+#   processed components to avoid circular dependencies.
+define get_all_deps
+$(if $(filter $(1),$(2)),,\
+	$(sort $(1) $(foreach d,$($(1)_DEPS),$(call get_all_deps,$d,$(1) $(2))))
+)
+endef
+
+##help deps_<component>: output the dependencies for <component>
+deps_%:
+	@echo "=== Dependencies for [$*] ==="
+	@echo "$(filter-out $*,$(call get_all_deps,$*,))"
+
 ##help list_components: output the list of components
 ##help : Hint make list_components | paste - - - | column -t
 list_components:

tools/rocm-build/build_amd_smi_lib.sh

@@ -1,4 +1,3 @@
-#!/bin/bash
 
 source "$(dirname "${BASH_SOURCE}")/compute_utils.sh"
 
@@ -10,7 +9,9 @@ printUsage() {
     echo "  -c,  --clean              Removes all amd_smi build artifacts"
     echo "  -r,  --release            Build non-debug version amd_smi (default is debug)"
     echo "  -a,  --address_sanitizer  Enable address sanitizer"
-    echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -s,  --static             Component/Build does not support static builds just accepting this param & ignore. No effect of the param on this build"
+    echo "  -w,  --wheel              Creates python wheel package of amd-smi. 
+                                      It needs to be used along with -r option"
     echo "  -o,  --outdir <pkg_type>  Print path of output directory containing packages of type referred to by pkg_type"
     echo "  -p,  --package <type>     Specify packaging format"
     echo "  -h,  --help               Prints this help"
@@ -25,7 +26,6 @@ printUsage() {
 PROJ_NAME="amdsmi"
 PACKAGE_ROOT="$(getPackageRoot)"
 TARGET="build"
-
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE="$(getIncludePath)"
 AMDSMI_BUILD_DIR=$(getBuildPath $PROJ_NAME)
@@ -42,7 +42,7 @@ SHARED_LIBS="ON"
 CLEAN_OR_OUT=0;
 PKGTYPE="deb"
 
-VALID_STR=`getopt -o hcraso:p: --long help,clean,release,static,address_sanitizer,outdir:,package: -- "$@"`
+VALID_STR=`getopt -o hcraswo:p: --long help,clean,release,static,wheel,address_sanitizer,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
@@ -60,6 +60,8 @@ do
                 ADDRESS_SANITIZER=true ; shift ;;
         (-s | --static)
                 ack_and_skip_static ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; shift ;;
         (-o | --outdir)
                 TARGET="outdir"; PKGTYPE=$2 ; OUT_DIR_SPECIFIED=1 ; ((CLEAN_OR_OUT|=2)) ; shift 2 ;;
         (-p | --package)
@@ -135,7 +137,7 @@ verifyEnvSetup
 
 case $TARGET in
     (clean) clean_amdsmi ;;
-    (build) build_amdsmi ;;
+    (build) build_amdsmi; build_wheel "$AMDSMI_BUILD_DIR" "$PROJ_NAME" ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $TARGET" ;;
 esac

tools/rocm-build/build_amdmigraphx.sh

@@ -1,41 +1,42 @@
 #!/bin/bash
 
 set -ex
-source "$(dirname "${BASH_SOURCE[0]}")/compute_helper.sh"
+source "$(dirname "${BASH_SOURCE[0]}")/compute_utils.sh"
 
 set_component_src AMDMIGraphX
 
 build_amdmigraphx() {
     echo "Start build"
 
+    if [ "${ENABLE_STATIC_BUILDS}" == "true" ]; then
+        ack_and_skip_static
+    fi
+
     cd $COMPONENT_SRC
 
     if ! command -v rbuild &> /dev/null; then
         pip3 install https://github.com/RadeonOpenCompute/rbuild/archive/master.tar.gz
     fi
 
+    # Remove CK
+    xargs -d '\n' -a ${OUT_DIR}/ck.files rm -- || true
+
     if [ "${ENABLE_ADDRESS_SANITIZER}" == "true" ]; then
          set_asan_env_vars
          set_address_sanitizer_on
     fi
 
-    if [ -n "$GPU_ARCHS" ]; then
-        GPU_TARGETS="$GPU_ARCHS"
-    else
-        GPU_TARGETS="gfx900;gfx906;gfx908;gfx90a;gfx1030;gfx1100;gfx1101;gfx1102;gfx942;gfx1200;gfx1201"
-    fi
     init_rocm_common_cmake_params
 
     mkdir -p ${BUILD_DIR} && rm -rf ${BUILD_DIR}/* && mkdir -p ${HOME}/amdmigraphx && rm -rf ${HOME}/amdmigraphx/*
     rbuild package -d "${HOME}/amdmigraphx" -B "${BUILD_DIR}" \
-        --cxx="${ROCM_PATH}/llvm/bin/clang++" \
-        --cc="${ROCM_PATH}/llvm/bin/clang" \
+        --cxx="$(set_build_variables __CLANG++__)" \
+        --cc="$(set_build_variables __CLANG__)" \
         "${rocm_math_common_cmake_params[@]}" \
         -DCMAKE_MODULE_LINKER_FLAGS="-Wl,--enable-new-dtags,--build-id=sha1,--rpath,$ROCM_LIB_RPATH" \
-	    -DGPU_TARGETS="${GPU_TARGETS}" \
         -DCMAKE_INSTALL_RPATH=""
 
-    mkdir -p $PACKAGE_DIR && cp ${BUILD_DIR}/*.${PKGTYPE} $PACKAGE_DIR
+    copy_if "${PKGTYPE}" "${CPACKGEN:-"DEB;RPM"}" "${PACKAGE_DIR}" "${BUILD_DIR}"/*."${PKGTYPE}"
     cd $BUILD_DIR && cmake --build . -- install -j${PROC}
 
     show_build_cache_stats
@@ -50,7 +51,7 @@ clean_amdmigraphx() {
 stage2_command_args "$@"
 
 case $TARGET in
-    build) build_amdmigraphx ;;
+    build) build_amdmigraphx; build_wheel ;;
     outdir) print_output_directory ;;
     clean) clean_amdmigraphx ;;
     *) die "Invalid target $TARGET" ;;

tools/rocm-build/build_clang-ocl.sh

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

tools/rocm-build/build_comgr.sh

@@ -14,6 +14,7 @@ printUsage() {
         type referred to by pkg_type"
     echo "  -h,  --help               Prints this help"
     echo "  -s,  --static             Build static lib (.a).  build instead of dynamic/shared(.so) "
+    echo "  -w,  --wheel              Creates python wheel package of comgr. It needs to be used along with -r option"
     echo "  -l,  --link_llvm_static   Link to LLVM statically.  Default is to dynamically link to LLVM; requires that LLVM dylibs are created."
     echo
     echo "Possible values for <type>:"
@@ -24,6 +25,7 @@ printUsage() {
     return 0
 }
 
+## Build environment variables
 API_NAME=amd_comgr
 PROJ_NAME=$API_NAME
 LIB_NAME=lib${API_NAME}
@@ -33,8 +35,8 @@ PACKAGE_ROOT=$(getPackageRoot)
 PACKAGE_LIB=$(getLibPath)
 PACKAGE_INCLUDE=$(getIncludePath)
 BUILD_DIR=$(getBuildPath $API_NAME)
-PACKAGE_DEB=$(getPackageRoot)/deb/$API_NAME
-PACKAGE_RPM=$(getPackageRoot)/rpm/$API_NAME
+PACKAGE_DEB=$PACKAGE_ROOT/deb/$PROJ_NAME
+PACKAGE_RPM=$PACKAGE_ROOT/rpm/$PROJ_NAME
 PACKAGE_PREFIX=$ROCM_INSTALL_PATH
 BUILD_TYPE=Debug
 MAKE_OPTS="$DASH_JAY CTEST_OUTPUT_ON_FAILURE=1 -C $BUILD_DIR"
@@ -44,13 +46,17 @@ CLEAN_OR_OUT=0;
 PKGTYPE="deb"
 MAKETARGET="deb"
 
+# link to LLVM dynamicaly.  Default is to link to llvm dynamically.
+# temporarily set this default to OFF until we resolve Issues
 LINK_LLVM_DYLIB="OFF"
 
-VALID_STR=`getopt -o hcraslo:p: --long help,clean,release,address_sanitizer,static,link_llvm_static,outdir:,package: -- "$@"`
+#parse the arguments
+VALID_STR=`getopt -o hcraswlo:p: --long help,clean,release,address_sanitizer,static,wheel,link_llvm_static,outdir:,package: -- "$@"`
 eval set -- "$VALID_STR"
 
 while true ;
 do
+    #echo "parocessing $1"
     case "$1" in
         (-h | --help)
                 printUsage ; exit 0;;
@@ -63,6 +69,8 @@ do
                 set_address_sanitizer_on ; shift ;;
         (-s | --static)
                 SHARED_LIBS="OFF" ; shift ;;
+        (-w | --wheel)
+                WHEEL_PACKAGE=true ; shift ;;
         (-l | --link_llvm_static)
                 LINK_LLVM_DYLIB="OFF"; shift ;;
         (-o | --outdir)
@@ -105,6 +113,7 @@ build() {
        echo " Building Shared Object "
     fi
 
+# Remove CTEST var once SWDEV-381396 is fixed
     cmake \
         $(rocm_cmake_params) \
         -DBUILD_SHARED_LIBS=$SHARED_LIBS \
@@ -143,7 +152,7 @@ verifyEnvSetup
 
 case $TARGET in
     (clean) clean ;;
-    (build) build ;;
+    (build) build; build_wheel "$BUILD_DIR" "$PROJ_NAME" ;;
     (outdir) print_output_directory ;;
     (*) die "Invalid target $TARGET" ;;
 esac