Update default.xml for ROCm 4.5.2 release

Update manifest for ROCm 4.5 release.

README.md

@@ -1,24 +1,133 @@
 
-# AMD ROCm Release Notes v3.7.0
 
-This page describes the features, fixed issues, and information about downloading and installing the ROCm software.
-It also covers known issues and deprecated features in this release.
+# AMD ROCm™ v4.5.2 Release Notes 
 
-- [Supported Operating Systems and Documentation Updates](#Supported-Operating-Systems-and-Documentation-Updates)
-  * [Supported Operating Systems](#Supported-Operating-Systems)
+This document describes the features, fixed issues, and information about downloading and installing the AMD ROCm™ software. It also covers known issues and deprecations in this release.
+
+## INSTALLATION GUIDE UPDATES FOR ROCM V4.5.2 
+
+In this release, users have the option to install the kernel mode driver using the Installer method. Some of the ROCm-specific use cases that the installer currently supports are:    
+
+- OpenCL (ROCr/KFD based) runtime  
+
+- HIP runtimes  
+
+- ROCm libraries and applications  
+
+- ROCm Compiler and device libraries  
+
+- ROCr runtime and thunk  
+
+- Kernel mode driver  
+
+For more details, refer to the AMD ROCm Installation Guide v4.5.2 at, 
+
+https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation_new.html
+
+
+## HIP ENHANCEMENTS
+
+
+The ROCm v4.5.2 release consists of the following HIP enhancement. 
+
+### Changes to the roc-obj-ls Tool 
+
+The roc-obj-ls tool is corrected in ROCm v4.5.2, and the command roc-obj-ls <exe> | roc-obj-extract is no longer the preferred syntax.   
+
+Use the roc-obj tool with the following correct command:  
+
+```   
+
+        roc-obj <exe> 
+ 
+ ```   
+
+For example, 
+
+Extract all ROCm code objects from a list of executables 
+ 
+ ```   
+
+       roc-obj <executable>... 
+ 
+ ```   
+
+Extract all ROCm code objects from a list of executables, and disassemble them 
+ 
+ ```   
+
+       roc-obj --disassemble <executable>... 
+ 
+       # or 
+
+       roc-obj -d <executable>... 
+
+ ```   
+
+Extract all ROCm code objects from a list of executables into dir/ 
+ 
+ ```   
+
+       roc-obj --outdir dir/ <executable>... 
+
+       # or 
+
+       roc-obj -o dir/ <executable>... 
+
+ ```   
+
+Extract only ROCm code objects matching regex over Target ID 
+ 
+ ```   
+
+       roc-obj --target-id gfx9 <executable>... 
+
+        # or 
+
+       roc-obj -t gfx9 <executable>... 
+
+```    
+
+For more information, refer to the HIP Programming Guide at:  
+
+https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD_HIP_Programming_Guide.pdf
+
+
+ ## OPENMP DEFECT FIX
+
+Previously, ROCProfiler crashed when the following ROCProfiler options were used in OpenMP programs: 
+
+* --stats  
+
+* --hsa-trace 
+
+This issue is fixed in the OpenMP plugin by ensuring that the contents of a kernel dispatch packet are not accessed after publishing it. The issue is also fixed in ROCTracer by ensuring that the registered exit function is called before the runtime library is closed. 
+ 
+ 
+
+# AMD ROCm™ v4.5 Release Notes 
+
+This document describes the features, fixed issues, and information about downloading and installing the AMD ROCm™ software. It also covers known issues and deprecations in this release.
+
+- Supported Operating Environments and Documentation Updates
+  * [Supported Operating Environments](#Supported-Operating-Environments)
+  * [ROCm Installation Updates](#ROCm-Installation-Updates)
   * [AMD ROCm Documentation Updates](#AMD-ROCm-Documentation-Updates)
-   
-- [What\'s New in This Release](#Whats-New-in-This-Release)
-  * [AOMP Enhancements](#AOMP-Enhancements)
-  * [Compatibility with NVIDIA Communications Collective Library v2\.7 API](#Compatibility-with-NVIDIA-Communications-Collective-Library-v27-API)
-  * [Singular Value Decomposition of Bi\-diagonal Matrices](#Singular-Value-Decomposition-of-Bi-diagonal-Matrices)
-  * [rocSPARSE_gemmi\() Operations for Sparse Matrices](#rocSPARSE_gemmi-Operations-for-Sparse-Matrices)
-    
- 
-- [Known Issues](#Known-Issues)
 
-- [Deploying ROCm](#Deploying-ROCm)
- 
+   
+- What\'s New in This Release
+  * [HIP Enhancements](#HIP-Enhancements)
+  * [Unified Memory Support in ROCm](#Unified-Memory-Support-in-ROCm)
+  * [System Management Interface](#System-Management-Interface) 
+  * [ROCm Math and Communication Libraries](#ROCm-Math-and-Communication-Libraries)
+  * [OpenMP Enhancements](#OpenMP-Enhancements)   
+
+- Known Issues in This Release
+  * [Known Issues in This Release](#Known-Issues-in-This-Release)
+
+- Deprecations in This Release
+  * [Deprecations](#Deprecations)
+
 - [Hardware and Software Support](#Hardware-and-Software-Support)
 
 - [Machine Learning and High Performance Computing Software Stack for AMD GPU](#Machine-Learning-and-High-Performance-Computing-Software-Stack-for-AMD-GPU)
@@ -27,318 +136,555 @@ It also covers known issues and deprecated features in this release.
   
 
 
-# Supported Operating Systems 
 
-## Support for Ubuntu 20.04 
+# ROCm Installation Updates
+
+This document describes the features, fixed issues, and information
+about downloading and installing the AMD ROCm™ software.
+
+It also covers known issues and deprecations in this release.
+
+## Supported Operating Environments
+
+The AMD ROCm platform supports the following operating systems:
+
+| OS                | Kernel              |
+| ---               | ---                 |
+| SLES15 SP3        | 5.3.18-24.49        |
+| RHEL 7.9          | 3.10.0-1160.6.1.el7 |
+| CentOS 7.9        | 3.10.0-1127         |
+| RHEL 8.4          | 4.18.0-193.1.1.el8  |
+| CentOS 8.3        | 4.18.0-193.el8      |
+| Ubuntu 18.04.5    | 5.4.0-71-generic    |
+| Ubuntu 20.04.3HWE | 5.8.0-48-generic    |
 
 
-In this release, AMD ROCm extends support to Ubuntu 20.04, including dual-kernel.
 
-## List of Supported Operating Systems
+## Enhanced Installation Process for ROCm v4.5
 
-The AMD ROCm v3.7.x platform is designed to support the following operating systems:
+In addition to the installation method using the native Package Manager, AMD ROCm v4.5 introduces added methods to install ROCm. With this
+release, the ROCm installation uses the *amdgpu-install* and *amdgpu-uninstall* scripts. 
 
-* Ubuntu 20.04 and 18.04.4 (Kernel 5.3)	
-* CentOS 7.8 & RHEL 7.8 (Kernel 3.10.0-1127) (Using devtoolset-7 runtime support)
-* CentOS 8.2 & RHEL 8.2 (Kernel 4.18.0 ) (devtoolset is not required)
-* SLES 15 SP1
+The *amdgpu-install* script streamlines the installation process by:
 
-## Fresh Installation of AMD ROCm v3.7 Recommended
-A fresh and clean installation of AMD ROCm v3.7 is recommended. An upgrade from previous releases to AMD ROCm v3.7 is not supported.
+-   Abstracting the distribution-specific package installation logic
 
-For more information, refer to the AMD ROCm Installation Guide at:
-https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html
+-   Performing the repository set-up
 
-**Note**: AMD ROCm release v3.3 or prior releases are not fully compatible with AMD ROCm v3.5 and higher versions. You must perform a fresh ROCm installation if you want to upgrade from AMD ROCm v3.3 or older to 3.5 or higher versions and vice-versa.
+-   Allowing user to specify the use case and automating the installation of all the required packages,
+
+-   Performing post-install checks to verify whether the installation was performed successfully
+
+-   Installing the uninstallation script
+
+The *amdgpu-uninstall* script allows the removal of the entire ROCm stack by using a single command.
+
+Some of the ROCm-specific use cases that the installer currently supports are:
+
+-   OpenCL (ROCr/KFD based) runtime
+
+-   HIP runtimes
+
+-   ROCm libraries and applications
+
+-   ROCm Compiler and device libraries
+
+-   ROCr runtime and thunk
+
+**Note:** Graphics use cases are not supported in this release.
+
+For more details, refer to the AMD ROCm Installation Guide v4.5 at,
+
+https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation_new.html
 
 
-# AMD ROCm Documentation Updates
+# AMD ROCm Documentation Updates 
 
 ## AMD ROCm Installation Guide 
 
-The AMD ROCm Installation Guide in this release includes:
+The AMD ROCm Installation Guide in this release includes the following
+updates:
 
-* Updated Supported Environments
-* HIP Installation Instructions
+-   New - Installation Guide for ROCm v4.5
 
-https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html
+    https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation_new.html
 
 
-## AMD ROCm - HIP Documentation Updates
+## AMD Instinct™ High Performance Computing and Tuning
 
-### Texture and Surface Functions
-The documentation for Texture and Surface functions is updated and available at:
+- New - AMD Instinct™ High Performance Computing and Tuning Guide 
 
-https://rocmdocs.amd.com/en/latest/Programming_Guides/Kernel_language.html
+   https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD%20Instinct%E2%84%A2High%20Performance%20Computing%20and%20Tuning%20Guide.pdf
+   
+   
+## HIP Documentation
 
-### Warp Shuffle Functions
-The documentation for Warp Shuffle functions is updated and available at:
+-  HIP installation instructions
 
-https://rocmdocs.amd.com/en/latest/Programming_Guides/Kernel_language.html
+   https://rocmdocs.amd.com/en/latest/Installation_Guide/HIP-Installation.html
 
-### Compiler Defines and Environment Variables
-The documentation for the updated HIP Porting Guide is available at:
+-   HIP Programming Guide
 
-https://rocmdocs.amd.com/en/latest/Programming_Guides/HIP-porting-guide.html#hip-porting-guide
+    https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD_HIP_Programming_Guide.pdf
 
+-   HIP API Guide
 
-## AMD ROCm Debug Agent
+    https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD-HIP-API-4.5.pdf
 
-ROCm Debug Agent Library 
+-   HIP-Supported CUDA API Reference Guide
 
-https://rocmdocs.amd.com/en/latest/ROCm_Tools/rocm-debug-agent.html
+    https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD_HIP_Supported_CUDA_API_Reference_Guide.pdf
 
+-   AMD ROCm Compiler Reference Guide
 
-## General AMD ROCm Documentatin Links
+    https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD_Compiler_Reference_Guide_v4.5.pdf
 
-Access the following links for more information:
+-   HIP FAQ
 
-* For AMD ROCm documentation, see 
+    <https://rocmdocs.amd.com/en/latest/Programming_Guides/HIP-FAQ.html#hip-faq>
+   
 
-  https://rocmdocs.amd.com/en/latest/
+## System Interface Management
 
-* For installation instructions on supped platforms, see
+-   System Interface Management (SMI)
 
-  https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html
+    https://rocmdocs.amd.com/en/latest/ROCm_System_Managment/ROCm-System-Managment.html
+    
 
-* For AMD ROCm binary structure, see
+## AMD ROCm Data Center Tool
 
-  https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html#build-amd-rocm
+- AMD ROCm Data Center Tool API Guide
 
-* For AMD ROCm Release History, see
+  https://github.com/RadeonOpenCompute/ROCm/blob/master/RDC_API_Manual_4.5.pdf
+  
+- AMD ROCm Data Center Tool User Guide
 
-  https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html#amd-rocm-version-history
+  https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD_ROCm_DataCenter_Tool_User_Guide_v4.5.pdf
 
+## ROCm SMI API Guide
 
+-   ROCm SMI API Guide
 
-# What\'s New in This Release
+    https://github.com/RadeonOpenCompute/ROCm/blob/master/ROCm_SMI_Manual_4.5.pdf
 
-## AOMP ENHANCEMENTS
+## ROC Debugger User and API Guide 
 
-AOMP is a scripted build of LLVM. It supports OpenMP target offload on AMD GPUs. Since AOMP is a Clang/LLVM compiler, it also supports GPU offloading with HIP, CUDA, and OpenCL.
+-   ROCDebugger User Guide
 
-The following enhancements are made for AOMP in this release: 
-* OpenMP 5.0 is enabled by default. You can use -fopenmp-version=45 for OpenMP 4.5 compliance
-* Restructured to include the ROCm compiler
-* B=Bitcode search path using hip policy HIP_DEVICE_LIB_PATH and hip-devic-lib command line option to enable global_free for kmpc_impl_free
+    https://github.com/RadeonOpenCompute/ROCm/blob/master/ROCDebugger_User_Guide.pdf
 
-Restructured hostrpc, including:
-* Replaced hostcall register functions with handlePayload(service, payload). Note, handlPayload has a simple switch to call the correct service handler function.
-* Removed the WITH_HSA macro
-* Moved the hostrpc stubs and host fallback functions into a single library and the include file. This enables the stubs openmp cpp source instead of hip and reorganizes the directory openmp/libomptarget/hostrpc.
-* Moved hostrpc_invoke.cl to DeviceRTLs/amdgcn.
-* Generalized the vargs processing in printf to work for any vargs function to execute on the host, including a vargs function that uses a function pointer.
-* Reorganized files, added global_allocate and global_free.
-* Fixed llvm TypeID enum to match the current upstream llvm TypeID.
-* Moved strlen_max function inside the declare target #ifdef _DEVICE_GPU in hostrpc.cpp to resolve linker failure seen in pfspecifier_str smoke test.
-* Fixed AOMP_GIT_CHECK_BRANCH in aomp_common_vars to not block builds in Red Hat if the repository is on a specific commit hash.
-* Simplified and reduced the size of openmp host runtime
-* Switched to default OpenMP 5.0
+-   Debugger API Guide
 
-For more information, see https://github.com/ROCm-Developer-Tools/aomp
-     
+    https://github.com/RadeonOpenCompute/ROCm/blob/master/ROCDebugger_API_Guide.pdf
+    
+## OpenMP Documentation
 
-## ROCm COMMUNICATIONS COLLECTIVE LIBRARY
+- Updated OpenMP documentation 
 
-### Compatibility with NVIDIA Communications Collective Library v2\.7 API
+  https://rocmdocs.amd.com/en/latest/Programming_Guides/openmp_support.html
+  
 
-ROCm Communications Collective Library (RCCL) is now compatible with the NVIDIA Communications Collective Library (NCCL) v2.7 API.
+## AMD ROCm General Documentation Links
 
-RCCL (pronounced "Rickle") is a stand-alone library of standard collective communication routines for GPUs, implementing all-reduce, all-gather, reduce, broadcast, reduce-scatter, gather, scatter, and all-to-all. There is also initial support for direct GPU-to-GPU send and receive operations. It has been optimized to achieve high bandwidth on platforms using PCIe, xGMI as well as networking using InfiniBand Verbs or TCP/IP sockets. RCCL supports an arbitrary number of GPUs installed in a single node or multiple nodes, and can be used in either single- or multi-process (e.g., MPI) applications.
+-   For AMD ROCm documentation, see
 
-The collective operations are implemented using ring and tree algorithms and have been optimized for throughput and latency. For best performance, small operations can be either batched into larger operations or aggregated through the API.
+    https://rocmdocs.amd.com/en/latest/
 
-For more information about RCCL APIs and compatibility with NCCL v2.7, see
-https://rccl.readthedocs.io/en/develop/index.html
+-   For installation instructions on supported platforms, see
 
+    https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html
 
-## Singular Value Decomposition of Bi\-diagonal Matrices
+-   For AMD ROCm binary structure, see
 
-Rocsolver_bdsqr now computes the Singular Value Decomposition (SVD) of bi-diagonal matrices. It is an auxiliary function for the SVD of general matrices (function rocsolver_gesvd). 
+    https://rocmdocs.amd.com/en/latest/Installation_Guide/Software-Stack-for-AMD-GPU.html
 
-BDSQR computes the singular value decomposition (SVD) of a n-by-n bidiagonal matrix B.
+-   For AMD ROCm release history, see
 
-The SVD of B has the following form:
+    https://rocmdocs.amd.com/en/latest/Current_Release_Notes/ROCm-Version-History.html
 
-B = Ub * S * Vb'
-where 
-•	S is the n-by-n diagonal matrix of singular values of B
-•	the columns of Ub are the left singular vectors of B
-•	the columns of Vb are its right singular vectors
 
-The computation of the singular vectors is optional; this function accepts input matrices U (of size nu-by-n) and V (of size n-by-nv) that are overwritten with U*Ub and Vb’*V. If nu = 0 no left vectors are computed; if nv = 0 no right vectors are computed.
+# What's New in This Release 
 
-Optionally, this function can also compute Ub’*C for a given n-by-nc input matrix C.
+## HIP Enhancements
 
-PARAMETERS
+The ROCm v4.5 release consists of the following HIP enhancements:
 
-•	[in] handle: rocblas_handle.
+### HIP Direct Dispatch
 
-•	[in] uplo: rocblas_fill.
+The conventional producer-consumer model where the host thread(producer) enqueues commands to a command queue (per stream), which is then
+processed by a separate, per-stream worker thread (consumer) created by the runtime, is no longer applicable.
 
-Specifies whether B is upper or lower bidiagonal.
+In this release, for Direct Dispatch, the runtime directly queues a packet to the AQL queue (user mode queue to GPU) in Dispatch and some of
+the synchronization. This new functionality indicates the total latency of the HIP Dispatch API and the latency to launch the first wave on the
+GPU.
 
-•	[in] n: rocblas_int. n >= 0.
+In addition, eliminating the threads in runtime has reduced the variance in the dispatch numbers as the thread scheduling delays and
+atomics/locks synchronization latencies are reduced.
 
-The number of rows and columns of matrix B.
+This feature can be disabled by setting the following environment variable,
 
-•	[in] nv: rocblas_int. nv >= 0. 
+```     
+     AMD_DIRECT_DISPATCH=0
 
-The number of columns of matrix V.
+```
 
-•	[in] nu: rocblas_int. nu >= 0. 
+### Support for HIP Graph 
 
-The number of rows of matrix U.
+ROCm v4.5 extends support for HIP Graph. For details, refer to the HIP API Guide at
 
-•	[in] nc: rocblas_int. nu >= 0. 
+https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD-HIP-API-4.5.pdf
 
-The number of columns of matrix C.
+### Enhanced *launch_bounds* Check Error Log Message 
 
-•	[inout] D: pointer to real type. Array on the GPU of dimension n.
+When a kernel is launched with HIP APIs, for example, hipModuleLaunchKernel(), HIP validates to check that input kernel
+dimension size is not larger than specified launch_bounds.
 
-On entry, the diagonal elements of B. On exit, if info = 0, the singular values of B in decreasing order; if info > 0, the diagonal elements of a bidiagonal matrix orthogonally equivalent to B.
+If exceeded, HIP returns launch failure if AMD_LOG_LEVEL is set with the proper value. Users can find more information in the error log message,
+including launch parameters of kernel dim size, launch bounds, and the name of the faulting kernel. It is helpful to figure out the faulting
+kernel. Besides, the kernel dim size and launch bounds values will also assist in debugging such failures.
 
-•	[inout] E: pointer to real type. Array on the GPU of dimension n-1.
+For more details, refer to the HIP Programming Guide at
 
-On entry, the off-diagonal elements of B. On exit, if info > 0, the off-diagonal elements of a bidiagonal matrix orthogonally equivalent to B (if info = 0 this matrix converges to zero).
+https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD_HIP_Programming_Guide.pdf
 
-•	[inout] V: pointer to type. Array on the GPU of dimension ldv*nv.
 
-On entry, the matrix V. On exit, it is overwritten with Vb’*V. (Not referenced if nv = 0).
+### HIP Runtime Compilation
 
-•	[in] ldv: rocblas_int. ldv >= n if nv > 0, or ldv >=1 if nv = 0.
+HIP now supports runtime compilation (hipRTC), the usage of which will provide the possibility of optimizations and performance improvement
+compared with other APIs via regular offline static compilation. 
 
-Specifies the leading dimension of V.
+hipRTC APIs accept HIP source files in character string format as input parameters and create handles of programs by compiling the HIP source
+files without spawning separate processes.
 
-•	[inout] U: pointer to type. Array on the GPU of dimension ldu*n.
+For more details on hipRTC APIs, refer to the HIP API Guide at
 
-On entry, the matrix U. On exit, it is overwritten with U*Ub. (Not referenced if nu = 0).
+https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD-HIP-API-4.5.pdf
 
-•	[in] ldu: rocblas_int. ldu >= nu.
+### New Flag for Backwards Compatibility on float/double atomicAdd Function
 
-Specifies the leading dimension of U.
+In the ROCm4.5 release, a new compilation flag is introduced as an option in the CMAKE file. This flag ensures backwards compatibility in
+float/double atomicAdd functions.
 
-•	[inout] C: pointer to type. Array on the GPU of dimension ldc*nc.
+```   
+     \_\_HIP_USE_CMPXCHG_FOR_FP_ATOMICS
+```   
 
-On entry, the matrix C. On exit, it is overwritten with Ub’*C. (Not referenced if nc = 0).
+This compilation flag is not set("0") by default, so the HIP runtime uses the current float/double atomicAdd functions.
 
-•	[in] ldc: rocblas_int. ldc >= n if nc > 0, or ldc >=1 if nc = 0.
+If this compilation flag is set to "1" with the CMAKE option, the existing float/double atomicAdd functions is used for compatibility with
+compilers that do not support floating point atomics.
 
-Specifies the leading dimension of C.
+```  
+     D\_\_HIP_USE_CMPXCHG_FOR_FP_ATOMICS=1
+```  
+For details on how to build the HIP runtime, refer to the HIP Programming Guide at
 
-•	[out] info: pointer to a rocblas_int on the GPU.
+https://github.com/RadeonOpenCompute/ROCm/blob/master/AMD_HIP_Programming_Guide.pdf
 
-If info = 0, successful exit. If info = i > 0, i elements of E have not converged to zero.
 
-For more information, see
-https://rocsolver.readthedocs.io/en/latest/userguide_api.html#rocsolver-type-bdsqr
 
+### Updated HIP Version Definition
 
-### rocSPARSE_gemmi\() Operations for Sparse Matrices
+The HIP version definition is updated as follows:
 
-This enhancement provides a dense matrix sparse matrix multiplication using the CSR storage format.
-rocsparse_gemmi multiplies the scalar αα with a dense m×km×k matrix AA and the sparse k×nk×n matrix BB defined in the CSR storage format, and adds the result to the dense m×nm×n matrix CC that is multiplied by the scalar ββ, such that
-C:=α⋅op(A)⋅op(B)+β⋅CC:=α⋅op(A)⋅op(B)+β⋅C
-with
+```  
+     HIP_VERSION=HIP_VERSION_MAJOR * 10000000 + HIP_VERSION_MINOR *
+     100000 + HIP_VERSION_PATCH)
+```  
 
-op(A)=⎧⎩⎨⎪⎪A,AT,AH,if trans_A == rocsparse_operation_noneif trans_A == rocsparse_operation_transposeif trans_A == rocsparse_operation_conjugate_transposeop(A)={A,if trans_A == rocsparse_operation_noneAT,if trans_A == rocsparse_operation_transposeAH,if trans_A == rocsparse_operation_conjugate_transpose
+The HIP version can be queried from the following HIP API call,
 
-and
+```  
+     hipRuntimeGetVersion(&runtimeVersion);
+```  
 
-op(B)=⎧⎩⎨⎪⎪B,BT,BH,if trans_B == rocsparse_operation_noneif trans_B == rocsparse_operation_transposeif trans_B == rocsparse_operation_conjugate_transposeop(B)={B,if trans_B == rocsparse_operation_noneBT,if trans_B == rocsparse_operation_transposeBH,if trans_B == rocsparse_operation_conjugate_transpose
-Note: This function is non-blocking and executed asynchronously with the host. It may return before the actual computation has finished.
+The version returned is always greater than the versions in the previous ROCm releases.
 
-For more information and examples, see
-https://rocsparse.readthedocs.io/en/master/usermanual.html#rocsparse-gemmi
- 
+**Note:** The version definition of the HIP runtime is different from that of CUDA. The function returns the HIP runtime version on the AMD
+platform, while on the NVIDIA platform, it returns the CUDA runtime version. There is no mapping or a correlation between the HIP and CUDA
+versions.
+
+
+
+### Planned HIP Enhancements and Fixes
+
+#### Changes to hiprtc implementation to match nvrtc behavior
+
+In this release, there are changes to the *hiprtc* implementation to
+match the *nvrtc* behavior.
+
+**Impact:** Applications can no longer explicitly include HIP runtime header files. Minor code changes are required to remove the HIP runtime
+header files.
+
+#### HIP device attribute enumeration
+
+In a future release, there will be a breaking change in the HIP device attribute enumeration. Enum values are being rearranged to accommodate
+future enhancements and additions.
+
+**Impact:** This will require users to rebuild their applications. No code changes are required.
+
+#### Changes to behavior of hipGetLastError() and hipPeekAtLastError() to match CUDA behavior available
+
+In a later release, changes to behavior of hipGetLastError() and hipPeekAtLastError() to match CUDA behavior will be available.
+
+**Impact:** Applications relying on the previous behavior will be impacted and may require some code changes.
+
+## Unified Memory Support in ROCm
+
+Unified memory allows applications to map and migrate data between CPU and GPU seamlessly without explicitly copying it between different
+allocations. This enables a more complete implementation of *hipMallocManaged*, *hipMemAdvise*, *hipMemPrefetchAsync* and related
+APIs. Without unified memory, these APIs only support system memory. With unified memory, the driver can automatically migrate such memory to
+GPU memory for faster access.
+
+### Supported Operating Systems and Versions
+
+This feature is only supported on recent Linux kernels. Currently, it works on Ubuntu versions with 5.6 or newer kernels and the DKMS driver
+from ROCm. Current releases of RHEL and SLES do not support this feature yet. Future releases of those distributions will add support for this.
+The unified memory feature is also supported in the KFD driver included with upstream kernels starting from Linux 5.14.
+
+Unified memory only works on GFXv9 and later GPUs, including Vega10 and MI100. Fiji, Polaris and older GPUs are not supported. To check whether
+unified memory is enabled, look in the kernel log for this message:
+
+```  
+     \$ dmesg \| grep \"HMM registered"
+```  
+
+If unified memory is enabled, there should be a message like "HMM registered xyzMB device memory". If unified memory is not supported on
+your GPU or kernel version, this message is missing.
+
+### Unified Memory Support and XNACK
+
+Unified memory support comes in two flavours, XNACK-enabled and XNACK-disabled. XNACK refers to the ability of the GPU to handle page
+faults gracefully and retry a memory access. In XNACK-enabled mode, the GPU can handle retry after page-faults, which enables mapping and
+migrating data on demand, as well as memory overcommitment. In XNACK-disabled mode, all memory must be resident and mapped in the GPU
+page tables when the GPU is executing application code. Any migrations involve temporary preemption of the GPU queues by the driver. Both page
+fault handling and preemptions, happen automatically and are transparent to the applications.
+
+XNACK-enabled mode only has experimental support. XNACK-enabled mode requires compiling shader code differently. By default, the ROCm
+compiler builds code that works in both modes. Code can be optimized for one specific mode with compiler options:
+
+OpenCL:
+
+```  
+     clang \... -mcpu=gfx908:xnack+:sramecc- \... // xnack on, sramecc off\
+     clang \... -mcpu=gfx908:xnack-:sramecc+ \... // xnack off, sramecc on
+```  
+
+HIP:
+```  
+     clang \... \--cuda-gpu-arch=gfx906:xnack+ \... // xnack on\
+     clang \... \--cuda-gpu-arch=gfx906:xnack- \... // xnack off
+```  
+Not all the math libraries included in ROCm support XNACK-enabled mode on current hardware. Applications will fail to run if their shaders are
+compiled in the incorrect mode.
+
+On current hardware, the XNACK mode can be chosen at boot-time by a module parameter amdgpu.noretry. The default is XNACK-disabled
+(amdgpu.noretry=1).
+
+## System Management Interface 
+
+### Enhanced ROCm SMI --setpoweroverdrive Functionality
+
+The ROCm System Management Interface (SMI) *--setpoweroverdrive* functionality is used to lower the power cap on a device without needing
+to enable the OverDrive functionality in the driver. Similarly, even with the OverDrive driver functionality enabled, it is possible to
+request a lower power cap than the card's default.
+
+Currently, any use of the *--setpoweroverdrive* functionality in rocm-smi prints an out-of-spec warning to the screen and requires the
+user to agree that using this functionality potentially voids their warranty. However, this warning should only be printed when users are
+trying to set the power cap to higher than the card's default, which requires the OverDrive driver functionality to be enabled.
+
+For example:
+
+The default power cap is 225.0W before any changes.
+
+```
+
+     \[atitest\@rhel85 smi\]\$ ./rocm_smi.py --resetpoweroverdrive
+     ======================= ROCm System Management Interface =======================
+
+     ========================== Reset GPU Power OverDrive ===========================
+
+     GPU\[0\] : Successfully reset Power OverDrive to: 225W
+     ============================ End of ROCm SMI Log ============================
+
+     Now, after using --setpoweroverdrive to lower the power cap to 123
+     watts:
+
+     \[atitest\@rhel85 smi\]\$ ./rocm_smi.py --setpoweroverdrive 123
+     ======================= ROCm System Management Interface =======================
+
+     =========================== Set GPU Power OverDrive ============================
+
+     GPU\[0\] : Successfully set power to: 123W
+     ======================= End of ROCm SMI Log ==============================
+
+     Setting a power cap lower than the default of 225.0W (in this case,
+     123W) does not give a warning.
+
+     To verify that the power is set to the correct value:
+
+     \[atitest\@rhel85 smi\]\$ ./rocm_smi.py --showmaxpower
+     ======================= ROCm System Management Interface =======================
+
+     ======================== Power Cap ===================================
+
+     GPU\[0\] : Max Graphics Package Power (W): 123.0
+     ========================End of ROCm SMI Log ==============================
+```
+
+
+## OpenMP Enhancements
+
+The ROCm installation includes an LLVM-based implementation, which fully supports OpenMP 4.5 standard and a subset of the OpenMP 5.0 standard.
+Fortran and C/C++ compilers and corresponding runtime libraries are included. Along with host APIs, the OpenMP compilers support offloading
+code and data onto GPU devices.
+
+For more information, refer to
+
+<https://rocmdocs.amd.com/en/latest/Programming_Guides/openmp_support.html>
+
+## ROCm Math and Communication Libraries
+
+In this release, ROCm Math and Communication Libraries consists of the
+following enhancements and fixes:
+
+| Library   | Changes                                                  |
+| ---       | ---                                                      |
+| rocBLAS | **Optimizations** <ul><li>Improved performance of non-batched and batched syr for all sizes and data types</li><li>Improved performance of non-batched and batched hemv for all sizes and data types</li><li>Improved performance of non-batched and batched symv for all sizes and data types</li><li>Improved memory utilization in rocblas-bench, rocblas-test gemm functions, increasing possible runtime sizes.</li></ul>**Changes** <ul><li>Update from C++14 to C++17.</li>  <li>Packaging split into a runtime package (called rocblas) and a development package (called rocblas-dev for .deb packages, and rocblas-devel for .rpm packages). The development package depends on runtime. The runtime package suggests the development package for all supported OSes except CentOS 7 to aid in the transition. The 'suggest' feature in packaging is introduced as a deprecated feature and will be removed in a future ROCm release.</li></ul> **Fixed**<ul><li>For function geam avoid overflow in offset calculation.</li>  <li> For function syr avoid overflow in offset calculation.</li> <li>For function gemv (Transpose-case) avoid overflow in offset calculation.</li> <li>For functions ssyrk and dsyrk, allow conjugate-transpose case to match legacy BLAS. Behavior is the same as the transpose case.</li></ul> |
+| hipBLAS| **Added**<ul><li>More support for hipblas-bench</li></ul>**Fixed**<ul><li>Avoid large offset overflow for gemv and hemv in hipblas-test</li></ul>**Changed**<ul><li>Packaging split into a runtime package called hipblas and a development package called hipblas-devel. The development package depends on runtime. The runtime package suggests the development package for all supported OSes except CentOS 7 to aid in the transition. The 'suggests' feature in packaging is a transitional feature and will be removed in a future rocm release.</li></ul> |
+| rocFFT | **Optimizations**<ul><li>Optimized SBCC kernels of length 52, 60, 72, 80, 84, 96, 104, 108, 112, 160, 168, 208, 216, 224, 240 with new kernel generator.</li></ul>**Added**<ul><li>Split 2D device code into separate libraries.</li> </ul>**Changed**<ul><li>Packaging split into a runtime package called rocfft and a development package called rocfft-devel. The development package depends on runtime. The runtime package suggests the development package for all supported OSes except CentOS 7 to aid in the transition. The suggests feature in packaging is introduced as a deprecated feature and will be removed in a future rocm release.</li></ul>**Fixed**<ul><li>Fixed a few validation failures of even-length R2C inplace. 2D, 3D cubics sizes such as 100^2 (or ^3), 200^2 (or ^3), 256^2 (or ^3)...etc. We don't combine the three kernels (stockham-r2c-transpose). We only combine two kernels (r2c-transpose) instead.</li></ul> |
+| hipFFT | **Changed**  <ul><li>Packaging split into a runtime package called hipfft and a development package called hipfft-devel. The development package depends on runtime. The runtime package suggests the development package for all supported OSes except CentOS 7 to aid in the transition. The 'suggests' feature in packaging is a transitional feature and will be removed in a future rocm release.</li></ul> |
+| rocSPARSE | **Added** <ul><li>Triangular solve for multiple right-hand sides using BSR format</li> <li>SpMV for BSRX format</li> <li>SpMM in CSR format enhanced to work with transposed A</li> <li>Matrix coloring for CSR matrices </li><li>Added batched tridiagonal solve (gtsv_strided_batch)</li></ul> **Improved** <ul><li>Fixed a bug with gemvi on Navi21 </li><li>Optimization for pivot based gtsv</li></ul> |
+| hipSPARSE | **Added** <ul><li>Triangular solve for multiple right-hand sides using BSR format</li> <li>SpMV for BSRX format</li> <li>SpMM in CSR format enhanced to work with transposed A</li> <li>Matrix coloring for CSR matrices </li>  <li>Added batched tridiagonal solve (gtsv_strided_batch)</li></ul> **Improved** <ul><li>Fixed a bug with gemvi on Navi21</li> <li>Optimization for pivot based gtsv</li></ul> |
+| rocALUTION | **Changed** <ul><li>Packaging split into a runtime package called rocalution and a development package called rocalution-devel. The development package depends on runtime. The runtime package suggests the development package for all supported OSes except CentOS 7 to aid in the transition. The 'suggests' feature in packaging is a transitional feature and will be removed in a future rocm release.</li></ul> **Improved** <ul><li>(A)MG solving phase optimization</li></ul> |
+| rocTHRUST | **Changed**  <ul><li>Packaging changed to a development package (called rocthrust-dev for .deb packages, and rocthrust-devel for .rpm packages). As rocThrust is a header-only library, there is no runtime package. To aid in the transition, the development package sets the "provides" field to provide the package rocthrust, so that existing packages depending on rocthrust can continue to work. The 'provides' feature is a transitional feature and will be removed in a future ROCm release.</li></ul> |
+| rocSOLVER | **Added** <ul><li>RQ factorization routines:</li><li>GERQ2, GERQF (with batched and strided_batchedversions)</li>  <li>Linear solvers for general square systems:</li> <li>GESV (with batched and strided_batched versions)</li><li>Linear solvers for symmetric/hermitian positive definite systems:</li> <li>POTRS (with batched and strided_batched versions)</li> <li>POSV (with batched and strided_batched versions) </li> <li>Inverse of symmetric/hermitian positive definite matrices:</li><li>POTRI (with batched and strided_batched versions)</li> <li>General matrix inversion without pivoting:  </li>  <li>GETRI_NPVT (with batched and strided_batched versions)</li> <li>GETRI_NPVT_OUTOFPLACE (with batched and  strided_batched versions)</li></ul>**Optimized**<ul><li>Improved performance of LU factorization (especially for large matrix sizes)</li> <li>Changed</li>  <li>Raised reference LAPACK version used for rocSOLVER test and benchmark clients to v3.9.1</li>  <li>Minor CMake improvements for users building from source without install.sh:</li> <li>Removed fmt::fmt from rocsolver\'s public usage requirements</li> <li>Enabled small-size optimizations by default </li>  <li>Split packaging into a runtime package ('rocsolver') and a development package ('rocsolver-devel'). The development package depends on the runtime package. To aid in the transition, the runtime package suggests the development package (except on CentOS 7). This use of the 'suggests feature' is transitional and will be removed in a future ROCm release.</li></ul> **Fixed** <ul><li>Use of the GCC / Clang __attribute__((deprecated(...))) extension is now guarded by compiler detection macros. |
+| hipSOLVER | The following functions were added in this release:<ul><li>gesv</li><ul><li>hipsolverSSgesv_bufferSize, hipsolverDDgesv_bufferSize, hipsolverCCgesv_bufferSize, hipsolverZZgesv_bufferSize</li><li>hipsolverSSgesv, hipsolverDDgesv, hipsolverCCgesv, hipsolverZZgesv</li></ul><li>potrs</li><ul><li>hipsolverSpotrs_bufferSize, hipsolverDpotrs_bufferSize, hipsolverCpotrs_bufferSize, hipsolverZpotrs_bufferSize</li><li>hipsolverSpotrs, hipsolverDpotrs, hipsolverCpotrs, hipsolverZpotrs</li></ul><li>potrsBatched</li><ul><li>hipsolverSpotrsBatched_bufferSize, hipsolverDpotrsBatched_bufferSize, hipsolverCpotrsBatched_bufferSize, hipsolverZpotrsBatched_bufferSize</li><li>hipsolverSpotrsBatched, hipsolverDpotrsBatched, hipsolverCpotrsBatched, hipsolverZpotrsBatched</li></ul><li>potri</li><ul><li>hipsolverSpotri_bufferSize, hipsolverDpotri_bufferSize, hipsolverCpotri_bufferSize, hipsolverZpotri_bufferSize</li><li>hipsolverSpotri, hipsolverDpotri, hipsolverCpotri, hipsolverZpotri</li></ul></ul></li></ul> |
+| RCCL | **Added** <ul><li>Compatibility with NCCL 2.9.9 </li></ul>**Changed**  <ul><li>Packaging split into a runtime package called rccl and a development package called rccl-devel. The development package depends on runtime. The runtime package suggests the development package for all supported OSes except CentOS 7 to aid in the transition. The 'suggests' feature in packaging is a tranistional feature and will be removed in a future rocm release.</li></ul> |
+| hipCUB | **Changed**  <ul><li>Packaging changed to a development package (called hipcub-dev for .deb packages, and hipcub-devel for .rpm packages). As hipCUB is a header-only library, there is no runtime package. To aid in the transition, the development package sets the "provides" field to provide the package hipcub, so that existing packages depending on hipcub can continue to work. This provides feature is introduced as a deprecated feature and will be removed in a future ROCm release.</li></ul> |
+| rocPRIM| **Added** <ul><li>bfloat16 support added.</li></ul> **Changed**  <ul><li>Packaging split into a runtime package called rocprim and a development package called rocprim-devel. The development package depends on runtime. The runtime package suggests the development package for all supported OSes except CentOS 7 to aid in the transition. The suggests feature in packaging is introduced as a deprecated feature and will be removed in a future rocm release.</li> <li>As rocPRIM is a header-only library, the runtime package is an empty placeholder used to aid in the transition. This package is also a deprecated feature and will be removed in a future rocm release.</li></ul> **Deprecated** <ul><li>The warp_size() function is now deprecated; please switch to host_warp_size() and device_warp_size() for host and device references respectively.</li></ul> |
+| rocRAND| **Changed**  <ul><li>Packaging split into a runtime package called rocrand and a development package called rocrand-devel. The development package depends on runtime. The runtime package suggests the development package for all supported OSes except CentOS 7 to aid in the transition. The suggests feature in packaging is introduced as a deprecated feature and will be removed in a future rocm release.</li></ul> **Fixed** <ul><li>Fix for mrg_uniform_distribution_double generating incorrect range of values</li> <li>Fix for order of state calls for log_normal, normal, and uniform</li></ul> **Known issues**  <ul><li>kernel_xorwow test is failing for certain GPU architectures.</li></ul> |
+
+For more information about ROCm Libraries, refer to the documentation at
+
+<https://rocmdocs.amd.com/en/latest/ROCm_Libraries/ROCm_Libraries.html>
+
+# Known Issues in This Release 
 
-# Known Issues 
 The following are the known issues in this release.
 
-## (AOMP) ‘Undefined Hidden Symbol’ Linker Error Causes Compilation Failure in HIP
+## clinfo and rocminfo Do Not Display Marketing Name
 
-The HIP example device_lib fails to compile due to unreferenced symbols with Link Time Optimization resulting in ‘undefined hidden symbol’ errors. 
+clinfo and rocminfo display a blank field for Marketing Name. 
 
-This issue is under investigation and there is no known workaround at this time.
+This is due to a missing package that is not yet available from ROCm. This package will be distributed in future ROCm releases.
+
+## Compiler Support for Function Pointers and Virtual Functions
+
+A known issue in the compiler support for function pointers and virtual functions on the GPU may cause undefined behavior due to register
+corruption. 
+
+A temporary workaround is to compile the affected application with this option:
+
+```
+     -mllvm -amdgpu-fixed-function-abi=1
+     
+ ```
+
+**Note:** This is an internal compiler flag and may be removed without notice once the issue is addressed in a future release.
 
 
-## MIGraphX Fails for fp16 Datatype
-The MIGraphX functionality does not work for the fp16 datatype.  
+## Debugger Process Exit May Cause ROCgdb Internal Error
 
-The following workaround is recommended: 
+If the debugger process exits during debugging, ROCgdb may report internal errors. This issue occurs as it attempts to access the AMD GPU
+state for the exited process. To recover, users must restart ROCgdb.
+ 
+As a workaround, users can set breakpoints to prevent the debugged process from exiting. For example, users can set breakpoints at the last
+statement of the main function and in the abort() and exit() functions. This temporary solution allows the application to be re-run without
+restarting ROCgdb.
 
-Use the AMD ROCm v3.3 of MIGraphX 
+This issue is currently under investigation and will be fixed in a future release.
 
-Or
+For more information, refer to the ROCgdb User Guide at,
 
-Build MIGraphX v3.7 from the source using AMD ROCm v3.3
+https://github.com/RadeonOpenCompute/ROCm/blob/master/ROCDebugger_User_Guide.pdf
 
-## Missing Google Test Installation May Cause RCCL Unit Test Compilation Failure
-Users of the RCCL install.sh script may encounter an RCCL unit test compilation error. It is recommended to use CMAKE directly instead of install.sh to compile RCCL. Ensure Google Test 1.10+ is available in the CMAKE search path.
+## Cache Issues with ROCProfiler
+
+When the same kernel is launched back-to-back multiple times on a GPU, a cache flush is executed each time the kernel finishes when profiler data is collected. The cache flush is inserted by ROCprofiler for each kernel. This prevents kernel from being cached, instead it is being read each time it is launched. As a result the cache hit rate from rocprofiler is reported as 0% or very low.
+
+This issue is under investigation and will be fixed in a future release. 
+
+## Stability Issue on LAMMPS-KOKKOS Applications
+
+On mGPU machines, lammps-kokkos applications experience a stability issue (AMD Instinct MI100™). 
+
+As a workaround, perform a Translation LookAside Buffer (TLB) flush. 
+
+The issue is under active investigation and will be resolved in a future release.
 
 
-As a workaround, use the latest RCCL from the GitHub development branch at:
-https://github.com/ROCmSoftwarePlatform/rccl/pull/237
 
+# Deprecations
 
-## Issue with Peer-to-Peer Transfers 
-Using peer-to-peer (P2P) transfers on systems without the hardware P2P assistance may produce incorrect results.
+## AMD Instinct MI25 End of Life
 
-Ensure the hardware supports peer-to-peer transfers and enable the peer-to-peer setting in the hardware to resolve this issue. 
+ROCm release v4.5 is the final release to support AMD Instinct MI25. AMD Instinct MI25 has reached End of Life (EOL). ROCm 4.5 represents the
+last certified release for software and driver support. AMD will continue to provide technical support and issue resolution for AMD
+Instinct MI25 on ROCm v4.5 for a period of 12 months from the software GA date. 
 
+## Planned Deprecation for Code Object Versions 2 and 3 
 
-## Partial Loss of Tracing Events for Large Applications
-An internal tracing buffer allocation issue can cause a partial loss of some tracing events for large applications.
+With the ROCm v4.5 release, the generation of code object versions 2 and 3 is being deprecated and may be removed in a future release. This deprecation notice does not impact support for the execution of AMD GPU code object versions.
 
-As a workaround, rebuild the roctracer/rocprofiler libraries from the GitHub ‘roc-3.7’ branch at:
-•	https://github.com/ROCm-Developer-Tools/rocprofiler
-•	https://github.com/ROCm-Developer-Tools/roctracer
-
-
-## GPU Kernel C++ Names Not Demangled
-GPU kernel C++ names in the profiling traces and stats produced by ‘—hsa-trace’ option are not demangled.
-
-As a workaround, users may choose to demangle the GPU kernel C++ names as required.
-
-
-## ‘rocprof’ option ‘--parallel-kernels’ Not Supported in This Release
-
-‘rocprof’ option ‘--parallel-kernels’ is available in the options list, however,  it is not fully validated and supported in this release.
-
-## Random Soft Hang Observed When Running ResNet-Based Models
-
-A random soft hang is observed when running ResNet-based models for a loop run of more than 25 to 30 hours.  The issue is observed on both PyTorch and TensorFlow frameworks.
-You can terminate the unresponsive process to temporarily resolve the issue.
-
-There is no known workaround at this time.
+The -mcode-object-version Clang option can be used to instruct the compiler to generate a specific AMD GPU code object version. In ROCm v4.5, the compiler can generate AMD GPU code object version 2, 3, and 4, with version 4 being the default if not specified. 
 
 
 
 # Deploying ROCm
-AMD hosts both Debian and RPM repositories for the ROCm v3.7.x packages. 
+
+AMD hosts both Debian and RPM repositories for the ROCm packages. 
 
 For more information on ROCM installation on all platforms, see
 
 https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html
 
+
+# Machine Learning and High Performance Computing Software Stack for AMD GPU
+
+For an updated version of the software stack for AMD GPU, see
+
+https://rocmdocs.amd.com/en/latest/Installation_Guide/Software-Stack-for-AMD-GPU.html
+
+
+
 # Hardware and Software Support
 ROCm is focused on using AMD GPUs to accelerate computational tasks such as machine learning, engineering workloads, and scientific computing.
 In order to focus our development efforts on these domains of interest, ROCm supports a targeted set of hardware configurations which are detailed further in this section.
 
+**Note:** The AMD ROCm™ open software platform is a compute stack for headless system deployments. GUI-based software applications are currently not supported.
+
 #### Supported GPUs
 Because the ROCm Platform has a focus on particular computational domains, we offer official support for a selection of AMD GPUs that are designed to offer good performance and price in these domains.
 
+**Note:** The integrated GPUs of Ryzen are not officially supported targets for ROCm.
+
 ROCm officially supports AMD GPUs that use following chips:
 
-  * GFX8 GPUs
-    * "Fiji" chips, such as on the AMD Radeon R9 Fury X and Radeon Instinct MI8
-    * "Polaris 10" chips, such as on the AMD Radeon RX 580 and Radeon Instinct MI6
-  * GFX9 GPUs
-    * "Vega 10" chips, such as on the AMD Radeon RX Vega 64 and Radeon Instinct MI25
-    * "Vega 7nm" chips, such as on the Radeon Instinct MI50, Radeon Instinct MI60 or AMD Radeon VII
+* GFX9 GPUs
+
+  - "Vega 10" chips, such as on the AMD Radeon RX Vega 64 and Radeon Instinct MI25
+  
+  - "Vega 7nm" chips, such as on the Radeon Instinct MI50, Radeon Instinct MI60 or AMD Radeon VII, Radeon Pro VII
+
+* CDNA GPUs
+
+  - MI100 chips such as on the AMD Instinct™ MI100
+
 
 ROCm is a collection of software ranging from drivers and runtimes to libraries and developer tools.
 Some of this software may work with more GPUs than the "officially supported" list above, though AMD does not make any official claims of support for these devices on the ROCm software platform.
+
 The following list of GPUs are enabled in the ROCm software, though full support is not guaranteed:
 
   * GFX8 GPUs
@@ -354,7 +700,7 @@ As described [below](#limited-support), "Carrizo", "Bristol Ridge", and "Raven R
 However, they are not enabled in the HIP runtime, and may not work due to motherboard or OEM hardware limitations.
 As such, they are not yet officially supported targets for ROCm.
 
-For a more detailed list of hardware support, please see [the following documentation](https://rocm.github.io/hardware.html).
+For a more detailed list of hardware support, please see [the following documentation](https://en.wikipedia.org/wiki/List_of_AMD_graphics_processing_units).
 
 #### Supported CPUs
 As described above, GFX8 GPUs require PCIe 3.0 with PCIe atomics in order to run ROCm.
@@ -394,9 +740,10 @@ does not require or take advantage of PCIe Atomics. However, we still recommend
 from the list provided above for compatibility purposes.
 
 #### Not supported or limited support under ROCm
+
 ##### Limited support
 
-* ROCm 2.9.x should support PCIe 2.0 enabled CPUs such as the AMD Opteron, Phenom, Phenom II, Athlon, Athlon X2, Athlon II and older Intel Xeon and Intel Core Architecture and Pentium CPUs. However, we have done very limited testing on these configurations, since our test farm has been catering to CPUs listed above. This is where we need community support. _If you find problems on such setups, please report these issues_.
+* ROCm 4.x should support PCIe 2.0 enabled CPUs such as the AMD Opteron, Phenom, Phenom II, Athlon, Athlon X2, Athlon II and older Intel Xeon and Intel Core Architecture and Pentium CPUs. However, we have done very limited testing on these configurations, since our test farm has been catering to CPUs listed above. This is where we need community support. _If you find problems on such setups, please report these issues_.
 * Thunderbolt 1, 2, and 3 enabled breakout boxes should now be able to work with ROCm. Thunderbolt 1 and 2 are PCIe 2.0 based, and thus are only supported with GPUs that do not require PCIe 3.1.0 atomics (e.g. Vega 10). However, we have done no testing on this configuration and would need community support due to limited access to this type of equipment.
 * AMD "Carrizo" and "Bristol Ridge" APUs are enabled to run OpenCL, but do not yet support HIP or our libraries built on top of these compilers and runtimes.
   * As of ROCm 2.1, "Carrizo" and "Bristol Ridge" require the use of upstream kernel drivers.
@@ -409,11 +756,13 @@ from the list provided above for compatibility purposes.
 
 ##### Not supported
 
-* "Tonga", "Iceland", "Vega M", and "Vega 12" GPUs are not supported in ROCm 2.9.x
+* "Tonga", "Iceland", "Vega M", and "Vega 12" GPUs are not supported.
 * We do not support GFX8-class GPUs (Fiji, Polaris, etc.) on CPUs that do not have PCIe 3.0 with PCIe atomics.
   * As such, we do not support AMD Carrizo and Kaveri APUs as hosts for such GPUs.
   * Thunderbolt 1 and 2 enabled GPUs are not supported by GFX8 GPUs on ROCm. Thunderbolt 1 & 2 are based on PCIe 2.0.
 
+In the default ROCm configuration, GFX8 and GFX9 GPUs require PCI Express 3.0 with PCIe atomics. The ROCm platform leverages these advanced capabilities to allow features such as user-level submission of work from the host to the GPU. This includes PCIe atomic Fetch and Add, Compare and Swap, Unconditional Swap, and AtomicOp Completion.
+
 #### ROCm support in upstream Linux kernels
 
 As of ROCm 1.9.0, the ROCm user-level software is compatible with the AMD drivers in certain upstream Linux kernels.
@@ -440,9 +789,16 @@ For users that have the option of using either AMD's or the upstreamed driver, t
 |      |   | Does not include most up-to-date firmware |
 
 
+# Disclaimer
+
+The information presented in this document is for informational purposes only and may contain technical inaccuracies, omissions, and typographical errors.  The information contained herein is subject to change and may be rendered inaccurate for many reasons, including but not limited to product and roadmap changes, component and motherboard versionchanges, new  model and/or product releases, product differences between differing manufacturers, software changes, BIOS flashes, firmware upgrades, or the like. Any computer system has risks of security vulnerabilities that cannot be completely prevented or mitigated.AMD  assumes  no  obligation  to  update  or  otherwise  correct  or  revise  this  information.  However, AMD reserves the right to revise this information and to make changes from time to time to the content hereof without obligation of AMD to notify any person of such revisions or changes.THIS INFORMATION IS PROVIDED ‘AS IS.” AMD MAKES NO REPRESENTATIONS OR WARRANTIES WITH RESPECT  TO  THE  CONTENTS  HEREOF  AND  ASSUMES  NO  RESPONSIBILITY  FOR  ANY  INACCURACIES, ERRORS, OR OMISSIONS THAT MAY APPEAR IN THIS INFORMATION. AMD SPECIFICALLY DISCLAIMS ANY IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR ANY PARTICULAR PURPOSE.  IN NO EVENT WILL AMD BE LIABLE TO ANY PERSON FOR ANY RELIANCE, DIRECT, INDIRECT, SPECIAL,  OR  OTHER  CONSEQUENTIAL  DAMAGES  ARISING  FROM  THE  USE  OF  ANY  INFORMATION CONTAINED HEREIN, EVEN IF AMD IS EXPRESSLY ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.AMD,  the  AMD  Arrow  logo, and combinations thereof are trademarks of Advanced Micro Devices, Inc.Other product names  used  in  this  publication  are  for  identification  purposes  only  and  may  be  trademarks  of  their respective  companies.
+©[2021]Advanced Micro Devices, Inc.All rights reserved.
+
+## Third-party Disclaimer
+Third-party content is licensed to you directly by the third party that owns the content and is not licensed to you by AMD.  ALL LINKED THIRD-PARTY CONTENT IS PROVIDED “AS IS” WITHOUT A WARRANTY OF ANY KIND.  USE OF SUCH THIRD-PARTY CONTENT IS DONE AT YOUR SOLE DISCRETION AND UNDER NO CIRCUMSTANCES WILL AMD BE LIABLE TO YOU FOR ANY THIRD-PARTY CONTENT.  YOU ASSUME ALL RISK AND ARE SOLELY RESPONSIBLE FOR ANY DAMAGES THAT MAY ARISE FROM YOUR USE OF THIRD-PARTY CONTENT. 
+
+
+
 
-## Machine Learning and High Performance Computing Software Stack for AMD GPU
 
-For an updated version of the software stack for AMD GPU, see
 
-https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html#machine-learning-and-high-performance-computing-software-stack-for-amd-gpu-v3-5-0

default.xml

@@ -1,27 +1,26 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <manifest>
     <remote name="roc-github"
-            fetch="http://github.com/RadeonOpenCompute/" />
+fetch="https://github.com/RadeonOpenCompute/" />
     <remote name="rocm-devtools"
-            fetch="https://github.com/ROCm-Developer-Tools/" />
+fetch="https://github.com/ROCm-Developer-Tools/" />
     <remote name="rocm-swplat"
-            fetch="https://github.com/ROCmSoftwarePlatform/" />
+fetch="https://github.com/ROCmSoftwarePlatform/" />
     <remote name="gpuopen-libs"
-            fetch="https://github.com/GPUOpen-ProfessionalCompute-Libraries/" />
+fetch="https://github.com/GPUOpen-ProfessionalCompute-Libraries/" />
     <remote name="gpuopen-tools"
-            fetch="https://github.com/GPUOpen-Tools/" />
+fetch="https://github.com/GPUOpen-Tools/" />
     <remote name="KhronosGroup"
-            fetch="https://github.com/KhronosGroup/" />
-    <default revision="refs/tags/rocm-3.7.0"
-             remote="roc-github"
-             sync-c="true"
-             sync-j="4" />
-    <!--list of projects for ROCM-->
+fetch="https://github.com/KhronosGroup/" />
+    <default revision="refs/tags/rocm-4.5.2"
+     remote="roc-github"
+     sync-c="true"
+     sync-j="4" />
+<!--list of projects for ROCM-->
     <project name="ROCK-Kernel-Driver" />
     <project name="ROCT-Thunk-Interface" />
     <project name="ROCR-Runtime" />
-    <project name="ROC-smi" />
-    <project name="rocm_smi_lib" remote="roc-github" />
+    <project name="rocm_smi_lib" />
     <project name="rocm-cmake" />
     <project name="rocminfo" />
     <project name="rocprofiler" remote="rocm-devtools" />
@@ -29,51 +28,50 @@
     <project name="ROCm-OpenCL-Runtime" />
     <project path="ROCm-OpenCL-Runtime/api/opencl/khronos/icd" name="OpenCL-ICD-Loader" remote="KhronosGroup" revision="6c03f8b58fafd9dd693eaac826749a5cfad515f8" />
     <project name="clang-ocl" />
-    <!--HIP Projects-->
+<!--HIP Projects-->
     <project name="HIP" remote="rocm-devtools" />
+    <project name="hipamd" remote="rocm-devtools" />
     <project name="HIP-Examples" remote="rocm-devtools" />
     <project name="ROCclr" remote="rocm-devtools" />
     <project name="HIPIFY" remote="rocm-devtools" />
-    <!-- The following projects are all associated with the AMDGPU LLVM compiler -->
-    <project name="llvm-project" path="llvm_amd-stg-open" />
+<!-- The following projects are all associated with the AMDGPU LLVM compiler -->
+    <project name="llvm-project" />
     <project name="ROCm-Device-Libs" />
     <project name="atmi" />
     <project name="ROCm-CompilerSupport" />
-    <project name="rocr_debug_agent" remote="rocm-devtools" revision="refs/tags/roc-3.7.0" />
+    <project name="rocr_debug_agent" remote="rocm-devtools" />
     <project name="rocm_bandwidth_test" />
+    <project name="half" remote="rocm-swplat" revision="37742ce15b76b44e4b271c1e66d13d2fa7bd003e" />
     <project name="RCP" remote="gpuopen-tools" revision="3a49405a1500067c49d181844ec90aea606055bb" />
-    <!-- gdb projects -->
+<!-- gdb projects -->
     <project name="ROCgdb" remote="rocm-devtools" />
     <project name="ROCdbgapi" remote="rocm-devtools" />
-    <!-- ROCm Libraries -->
+<!-- ROCm Libraries -->
+    <project name="rdc" />
     <project name="rocBLAS" remote="rocm-swplat" />
+    <project name="Tensile" remote="rocm-swplat" />
     <project name="hipBLAS" remote="rocm-swplat" />
     <project name="rocFFT" remote="rocm-swplat" />
+    <project name="hipFFT" remote="rocm-swplat" />
     <project name="rocRAND" remote="rocm-swplat" />
     <project name="rocSPARSE" remote="rocm-swplat" />
     <project name="rocSOLVER" remote="rocm-swplat" />
+    <project name="hipSOLVER" remote="rocm-swplat" />
     <project name="hipSPARSE" remote="rocm-swplat" />
     <project name="rocALUTION" remote="rocm-swplat" />
     <project name="MIOpenGEMM" remote="rocm-swplat" />
     <project name="MIOpen" remote="rocm-swplat" />
+    <project name="MIOpenTensile" remote="rocm-swplat" />
     <project name="rccl" remote="rocm-swplat" />
     <project name="MIVisionX" remote="gpuopen-libs" />
     <project name="rocThrust" remote="rocm-swplat" />
     <project name="hipCUB" remote="rocm-swplat" />
     <project name="rocPRIM" remote="rocm-swplat" />
-    <project name="AMDMIGraphX" remote="rocm-swplat" revision="e66968a25f9342a28af1157b06cbdbf8579c5519" />
+    <project name="hipfort" remote="rocm-swplat" />
+    <project name="AMDMIGraphX" remote="rocm-swplat" />
     <project name="ROCmValidationSuite" remote="rocm-devtools" />
-    <!-- Projects for AOMP -->
-    <project name="ROCT-Thunk-Interface" path="aomp/roct-thunk-interface" remote="roc-github" />
-    <project name="ROCR-Runtime" path="aomp/rocr-runtime" remote="roc-github" />
-    <project name="ROCm-Device-Libs" path="aomp/rocm-device-libs" remote="roc-github" />
-    <project name="ROCm-CompilerSupport" path="aomp/rocm-compilersupport" remote="roc-github" />
-    <project name="rocminfo" path="aomp/rocminfo" remote="roc-github" />
-    <project name="HIP" path="aomp/hip-on-vdi" remote="rocm-devtools" />
-    <project name="aomp" path="aomp/aomp" remote="rocm-devtools" />
-    <project name="aomp-extras" path="aomp/aomp-extras" remote="rocm-devtools" />
-    <project name="flang" path="aomp/flang" remote="rocm-devtools" />
-    <project name="amd-llvm-project" path="aomp/amd-llvm-project" remote="rocm-devtools" />
-    <project name="ROCclr" path="aomp/vdi" remote="rocm-devtools" />
-    <project name="ROCm-OpenCL-Runtime" path="aomp/opencl-on-vdi" remote="roc-github" />
+<!-- Projects for OpenMP-Extras -->
+    <project name="aomp" path="openmp-extras/aomp" remote="rocm-devtools" />
+    <project name="aomp-extras" path="openmp-extras/aomp-extras" remote="rocm-devtools" />
+    <project name="flang" path="openmp-extras/flang" remote="rocm-devtools" />
 </manifest>

images/test.rst

@@ -0,0 +1 @@
+

version_history.md

@@ -1,503 +0,0 @@
-## ROCm Version History
-This file contains archived version history information for the [ROCm project](https://github.com/RadeonOpenCompute/ROCm)
-
-### Current ROCm Version: 3.3
-- [New features and enhancements in ROCm v3.1](#new-features-and-enhancements-in-rocm-v31)
-- [New features and enhancements in ROCm v3.0](#new-features-and-enhancements-in-rocm-v30)
-- [New features and enhancements in ROCm v2.10](#new-features-and-enhancements-in-rocm-v210)
-- [New features and enhancements in ROCm 2.9](#new-features-and-enhancements-in-rocm-29)
-- [New features and enhancements in ROCm 2.8](#new-features-and-enhancements-in-rocm-28)
-- [New features and enhancements in ROCm 2.7.2](#new-features-and-enhancements-in-rocm-272)
-- [New features and enhancements in ROCm 2.7](#new-features-and-enhancements-in-rocm-27)
-- [New features and enhancements in ROCm 2.6](#new-features-and-enhancements-in-rocm-26)
-- [New features and enhancements in ROCm 2.5](#new-features-and-enhancements-in-rocm-25)
-- [New features and enhancements in ROCm 2.4](#new-features-and-enhancements-in-rocm-24)
-- [New features and enhancements in ROCm 2.3](#new-features-and-enhancements-in-rocm-23)
-- [New features and enhancements in ROCm 2.2](#new-features-and-enhancements-in-rocm-22)
-- [New features and enhancements in ROCm 2.1](#new-features-and-enhancements-in-rocm-21)
-- [New features and enhancements in ROCm 2.0](#new-features-and-enhancements-in-rocm-20)
-- [New features and enhancements in ROCm 1.9.2](#new-features-and-enhancements-in-rocm-192)
-- [New features and enhancements in ROCm 1.9.2](#new-features-and-enhancements-in-rocm-192-1)
-- [New features and enhancements in ROCm 1.9.1](#new-features-and-enhancements-in-rocm-191)
-- [New features and enhancements in ROCm 1.9.0](#new-features-and-enhancements-in-rocm-190)
-- [New features as of ROCm 1.8.3](#new-features-as-of-rocm-183)
-- [New features as of ROCm 1.8](#new-features-as-of-rocm-18)
-- [New Features as of ROCm 1.7](#new-features-as-of-rocm-17)
-- [New Features as of ROCm 1.5](#new-features-as-of-rocm-15)
-
-
-## New features and enhancements in ROCm v3.2
-The AMD ROCm v3.2 release was not productized.
-
-## New features and enhancements in ROCm v3.1
-### Change in ROCm Installation Directory Structure
-
-A fresh installation of the ROCm toolkit installs the packages in the /opt/rocm-<version> folder. Previously, ROCm toolkit packages were installed in the /opt/rocm folder.
-
-### Reliability, Accessibility, and Serviceability Support for Vega 7nm
-
-The Reliability, Accessibility, and Serviceability (RAS) support for Vega7nm is now available.
-
-### SLURM Support for AMD GPU
-
-SLURM (Simple Linux Utility for Resource Management) is an open source, fault-tolerant, and highly scalable cluster management and job scheduling system for large and small Linux clusters.
-
-
-## New features and enhancements in ROCm v3.0
-### Support for CentOS RHEL v7.7 <a id="centos-anchor"></a> 
-Support is extended for CentOS/RHEL v7.7 in the ROCm v3.0 release. For more information about the CentOS/RHEL v7.7 release, see:
-
-[CentOS/RHEL](https://centos.org/forums/viewtopic.php?t=71657)
-
-
-### Initial distribution of AOMP 0.7-5 in ROCm v3.0 <a id="aomp-anchor"></a> 
-The code base for this release of AOMP is the Clang/LLVM 9.0 sources as of October 8th, 2019. The LLVM-project branch used to build this release is AOMP-191008. It is now locked. With this release, an artifact tarball of the entire source tree is created. This tree includes a Makefile in the root directory used to build AOMP from the release tarball. You can use Spack to build AOMP from this source tarball or build manually without Spack.
-
-For more information about AOMP 0.7-5, see: [AOMP](https://github.com/ROCm-Developer-Tools/aomp/tree/roc-3.0.0)
-
-
-### Fast Fourier Transform Updates
-The Fast Fourier Transform (FFT) is an efficient algorithm for computing the Discrete Fourier Transform. Fast Fourier transforms are used in signal processing, image processing, and many other areas. The following real FFT performance change is made in the ROCm v3.0 release:
-
-•	Implement efficient real/complex 2D transforms for even lengths.
-
-Other improvements:
-
-•	More 2D test coverage sizes.
-
-•	Fix buffer allocation error for large 1D transforms.
-
-•	C++ compatibility improvements.
-
-### MemCopy Enhancement for rocProf
-In the v3.0 release, the rocProf tool is enhanced with an additional capability to dump asynchronous GPU memcopy information into a .csv file. You can use the '-hsa-trace' option to create the results_mcopy.csv file.
-Future enhancements will include column labels.
-
-
-### New features and enhancements in ROCm v2.10
-#### rocBLAS Support for Complex GEMM
-The rocBLAS library is a gpu-accelerated implementation of the standard Basic Linear Algebra Subroutines (BLAS). rocBLAS is designed to enable you to develop algorithms, including high performance computing, image analysis, and machine learning.
-
-In the AMD ROCm release v2.10, support is extended to the General Matrix Multiply (GEMM) routine for multiple small matrices processed simultaneously for rocBLAS in AMD Radeon Instinct MI50. Both single and double precision, CGEMM and ZGEMM, are now supported in rocBLAS.
-
-#### Support for SLES 15 SP1
-In the AMD ROCm v2.10 release, support is added for SUSE Linux® Enterprise Server (SLES) 15 SP1. SLES is a modular operating system for both multimodal and traditional IT.
-
-#### Code Marker Support for rocProfiler and rocTracer Libraries
-Code markers provide the external correlation ID for the calling thread. This function indicates that the calling thread is entering and leaving an external API region.
-
-### New features and enhancements in ROCm 2.9
-
-#### Initial release for Radeon Augmentation Library(RALI)
-The AMD Radeon Augmentation Library (RALI) is designed to efficiently decode and process images from a variety of storage formats and modify them through a processing graph programmable by the user. RALI currently provides C API.
-
-#### Quantization in MIGraphX v0.4
-MIGraphX 0.4 introduces support for fp16 and int8 quantization. For additional details, as well as other new MIGraphX features, see [MIGraphX documentation](https://github.com/ROCmSoftwarePlatform/AMDMIGraphX/wiki/Getting-started:-using-the-new-features-of-MIGraphX-0.4).
-
-#### rocSparse csrgemm
-csrgemm enables the user to perform matrix-matrix multiplication with two sparse matrices in CSR format.
-
-#### Singularity Support
-ROCm 2.9 adds support for Singularity container version 2.5.2.
-
-#### Initial release of rocTX
-ROCm 2.9 introduces rocTX, which provides a C API for code markup for performance profiling.  This initial release of rocTX supports annotation of code ranges and ASCII markers.  For an example, see this [code](https://github.com/ROCm-Developer-Tools/roctracer/blob/amd-master/test/MatrixTranspose_test/MatrixTranspose.cpp).
-
-#### Added support for Ubuntu 18.04.3
-Ubuntu 18.04.3 is now supported in ROCm 2.9.
-
-
-
-### New features and enhancements in ROCm 2.8
-
-#### Support for NCCL2.4.8 API
-Implements ncclCommAbort() and ncclCommGetAsyncError() to match the NCCL 2.4.x API
-
-### New features and enhancements in ROCm 2.7.2
-
-This release is a hotfix for ROCm release 2.7.
-
-#### Issues fixed in ROCm 2.7.2
-
-##### A defect in upgrades from older ROCm releases has been fixed.
-
-##### rocprofiler --hiptrace and --hsatrace fails to load roctracer library
-In ROCm 2.7.2, rocprofiler --hiptrace and --hsatrace fails to load roctracer library defect has been fixed.  
-To generate traces, please provide directory path also using the parameter: -d <$directoryPath> for example:
-```shell
-/opt/rocm/bin/rocprof  --hsa-trace -d $PWD/traces /opt/rocm/hip/samples/0_Intro/bit_extract/bit_extract
-  ```
-All traces and results will be saved under $PWD/traces path
-
-#### Upgrading from ROCm 2.7 to 2.7.2
-
-To upgrade, please remove 2.7 completely as specified [for ubuntu](#how-to-uninstall-from-ubuntu-1604-or-Ubuntu-1804) or [for centos/rhel](#how-to-uninstall-rocm-from-centosrhel-76), and install 2.7.2 as per instructions [install instructions](#installing-from-amd-rocm-repositories)
-
-#### Other notes
-
-  To use rocprofiler features, the following steps need to be completed before using rocprofiler:
-
-  ##### Step-1: Install roctracer
-
-###### Ubuntu 16.04 or Ubuntu 18.04:
-
-  ```shell
-  sudo apt install roctracer-dev
-  ```
-
-######  CentOS/RHEL 7.6:
-
-  ```shell
-  sudo yum install roctracer-dev
-  ```
-  ##### Step-2: Add /opt/rocm/roctracer/lib to LD_LIBRARY_PATH
-  
-### New features and enhancements in ROCm 2.7
-
-#### [rocFFT] Real FFT Functional
-Improved real/complex 1D even-length transforms of unit stride. Performance improvements of up to 4.5x are observed. Large problem sizes should see approximately 2x.
-
-#### rocRand Enhancements and Optimizations
-- Added support for new datatypes: uchar, ushort, half.
-- Improved performance on "Vega 7nm" chips, such as on the Radeon Instinct MI50
-- mtgp32 uniform double performance changes due generation algorithm standardization. Better quality random numbers now generated with 30% decrease in performance
-- Up to 5% performance improvements for other algorithms
-
-#### RAS
-Added support for RAS on Radeon Instinct MI50, including:
-- Memory error detection
-- Memory error detection counter
-
-#### ROCm-SMI enhancements
-Added ROCm-SMI CLI and LIB support for FW version, compute running processes, utilization rates, utilization counter, link error counter, and unique ID.
-
-### New features and enhancements in ROCm 2.6
-
-#### ROCmInfo enhancements
-ROCmInfo was extended to do the following:
-For ROCr API call errors including initialization determine if the error could be explained by:
-- ROCk (driver) is not loaded / available
-- User does not have membership in appropriate group - "video"
-- If not above print the error string that is mapped to the returned error code
-- If no error string is available, print the error code in hex
-
-#### Thrust - Functional Support on Vega20
-ROCm2.6 contains the first official release of rocThrust and hipCUB. rocThrust is a port of thrust, a parallel algorithm library. hipCUB is a port of CUB, a reusable software component library. Thrust/CUB has been ported to the HIP/ROCm platform to use the rocPRIM library. The HIP ported library works on HIP/ROCm platforms.
-
-Note: rocThrust and hipCUB library replaces https://github.com/ROCmSoftwarePlatform/thrust (hip-thrust), i.e. hip-thrust has been separated into two libraries, rocThrust and hipCUB. Existing hip-thrust users are encouraged to port their code to rocThrust and/or hipCUB. Hip-thrust will be removed from official distribution later this year.
-
-#### MIGraphX v0.3
-MIGraphX optimizer adds support to read models frozen from Tensorflow framework. Further details and an example usage at https://github.com/ROCmSoftwarePlatform/AMDMIGraphX/wiki/Getting-started:-using-the-new-features-of-MIGraphX-0.3
-
-#### MIOpen 2.0
-- This release contains several new features including an immediate mode for selecting convolutions, bfloat16 support, new layers, modes, and algorithms.
-- MIOpenDriver, a tool for benchmarking and developing kernels is now shipped with MIOpen.
-BFloat16 now supported in HIP requires an updated rocBLAS as a GEMM backend.
-- Immediate mode API now provides the ability to quickly obtain a convolution kernel.
-- MIOpen now contains HIP source kernels and implements the ImplicitGEMM kernels. This is a new feature and is currently disabled by default. Use the environmental variable "MIOPEN_DEBUG_CONV_IMPLICIT_GEMM=1" to activation this feature. ImplicitGEMM requires an up to date HIP version of at least 1.5.9211.
-- A new "loss" catagory of layers has been added, of which, CTC loss is the first. See the API reference for more details.
-2.0 is the last release of active support for gfx803 architectures. In future releases, MIOpen will not actively debug and develop new features specifically for gfx803.
-- System Find-Db in memory cache is disabled by default. Please see build instructions to enable this feature.
-Additional documentation can be found here: https://rocmsoftwareplatform.github.io/MIOpen/doc/html/
-
-#### Bloat16 software support in rocBLAS/Tensile
-Added mixed precision bfloat16/IEEE f32 to gemm_ex. The input and output matrices are bfloat16. All arithmetic is in IEEE f32.
-
-#### AMD Infinity Fabric™ Link enablement
-The ability to connect four Radeon Instinct MI60 or Radeon Instinct MI50 boards in two hives or two Radeon Instinct MI60 or Radeon Instinct MI50 boards in four hives via AMD Infinity Fabric™ Link GPU interconnect technology has been added.
-
-#### ROCm-smi features and bug fixes
-  - mGPU & Vendor check
-  - Fix clock printout if DPM is disabled
-  - Fix finding marketing info on CentOS
-  - Clarify some error messages
-
-#### ROCm-smi-lib enhancements
-- Documentation updates
-- Improvements to *name_get functions
-
-#### RCCL2 Enablement
-RCCL2 supports collectives intranode communication using PCIe, Infinity Fabric™, and pinned host memory, as well as internode communication using Ethernet (TCP/IP sockets) and Infiniband/RoCE (Infiniband Verbs).  Note: For Infiniband/RoCE, RDMA is not currently supported.
-
-#### rocFFT enhancements
-- Added: Debian package with FFT test, benchmark, and sample programs
-- Improved: hipFFT interfaces
-- Improved: rocFFT CPU reference code, plan generation code and logging code
-
-### New features and enhancements in ROCm 2.5
-
-#### UCX 1.6 support
-Support for UCX version 1.6 has been added.
-
-#### BFloat16 GEMM in rocBLAS/Tensile
-Software support for BFloat16 on Radeon Instinct MI50, MI60 has been added.  This includes:
-- Mixed precision GEMM with BFloat16 input and output matrices, and all arithmetic in IEEE32 bit
-- Input matrix values are converted from BFloat16 to IEEE32 bit, all arithmetic and accumulation is IEEE32 bit. Output values are rounded from IEEE32 bit to BFloat16
-- Accuracy should be correct to 0.5 ULP
-
-#### ROCm-SMI enhancements
-CLI support for querying the memory size, driver version, and firmware version has been added to ROCm-smi.
-
-#### [PyTorch] multi-GPU functional support (CPU aggregation/Data Parallel)
-Multi-GPU support is enabled in PyTorch using Dataparallel path for versions of PyTorch built using the 06c8aa7a3bbd91cda2fd6255ec82aad21fa1c0d5 commit or later.
-
-#### rocSparse optimization on Radeon Instinct MI50 and MI60
-This release includes performance optimizations for csrsv routines in the rocSparse library.
-
-#### [Thrust] Preview
-Preview release for early adopters. rocThrust is a port of thrust, a parallel algorithm library. Thrust has been ported to the HIP/ROCm platform to use the rocPRIM library. The HIP ported library works on HIP/ROCm platforms.
-
-Note: This library will replace https://github.com/ROCmSoftwarePlatform/thrust in a future release. The package for rocThrust (this library) currently conflicts with version 2.5 package of thrust. They should not be installed together.
-
-#### Support overlapping kernel execution in same HIP stream
-HIP API has been enhanced to allow independent kernels to run in parallel on the same stream.
-
-#### AMD Infinity Fabric&#x2122; Link enablement
-The ability to connect four Radeon Instinct MI60 or Radeon Instinct MI50 boards in one hive via AMD Infinity Fabric™ Link GPU interconnect technology has been added.
-### New features and enhancements in ROCm 2.4
-
-#### TensorFlow 2.0 support
-ROCm 2.4 includes the enhanced compilation toolchain and a set of bug fixes to support TensorFlow 2.0 features natively
-
-#### AMD Infinity Fabric&#x2122; Link enablement
-ROCm 2.4 adds support to connect two Radeon Instinct MI60 or Radeon Instinct MI50 boards via AMD Infinity Fabric&#x2122; Link GPU interconnect technology.
-
-### New features and enhancements in ROCm 2.3
-
-#### Mem usage per GPU
-Per GPU memory usage is added to rocm-smi.
-Display information regarding used/total bytes for VRAM, visible VRAM and GTT, via the --showmeminfo flag
-
-#### MIVisionX, v1.1 - ONNX
-ONNX parser changes to adjust to new file formats
-
-#### MIGraphX, v0.2
-MIGraphX 0.2 supports the following new features:
-* New Python API
-* Support for additional ONNX operators and fixes that now enable a large set of Imagenet models
-* Support for RNN Operators
-* Support for multi-stream Execution
-* [Experimental] Support for Tensorflow frozen protobuf files
-
-See: [Getting-started:-using-the-new-features-of-MIGraphX-0.2](https://github.com/ROCmSoftwarePlatform/AMDMIGraphX/wiki/Getting-started:-using-the-new-features-of-MIGraphX-0.2) for more details
-
-#### MIOpen, v1.8 - 3d convolutions and int8
-* This release contains full 3-D convolution support and int8 support for inference.
-* Additionally, there are major updates in the performance database for major models including those found in Torchvision.
-
-See: [MIOpen releases](https://github.com/ROCmSoftwarePlatform/MIOpen/releases)
-
-#### Caffe2 -  mGPU support
-Multi-gpu support is enabled for Caffe2.
-
-#### rocTracer library, ROCm tracing API for collecting runtimes API and asynchronous GPU activity traces
-HIP/HCC domains support is introduced in rocTracer library.
-
-#### BLAS -  Int8 GEMM performance, Int8 functional and performance
-Introduces support and performance optimizations for Int8 GEMM, implements TRSV support, and includes improvements and optimizations with Tensile.
-
-#### Prioritized L1/L2/L3 BLAS (functional)
-Functional implementation of BLAS L1/L2/L3 functions
-
-#### BLAS - tensile optimization
-Improvements and optimizations with tensile
-
-#### MIOpen Int8 support
-Support for int8
-
-### New features and enhancements in ROCm 2.2
-
-#### rocSparse Optimization on Vega20
-Cache usage optimizations for csrsv (sparse triangular solve), coomv
-(SpMV in COO format) and ellmv (SpMV in ELL format) are available.
-
-#### DGEMM and DTRSM Optimization
-Improved DGEMM performance for reduced matrix sizes (k=384, k=256)
-
-#### Caffe2
-Added support for multi-GPU training
-
-### New features and enhancements in ROCm 2.1
-
-#### RocTracer v1.0 preview release – 'rocprof' HSA runtime tracing and statistics support -
-Supports HSA API tracing and HSA asynchronous GPU activity including kernels execution and memory copy
-
-#### Improvements to ROCM-SMI tool -
-Added support to show real-time PCIe bandwidth usage via the -b/--showbw flag
-
-#### DGEMM Optimizations -
-Improved DGEMM performance for large square and reduced matrix sizes (k=384, k=256)
-
-### New features and enhancements in ROCm 2.0
-
-#### Adds support for RHEL 7.6 / CentOS 7.6 and Ubuntu 18.04.1
-
-#### Adds support for Vega 7nm, Polaris 12 GPUs
-
-#### Introduces MIVisionX
-* A comprehensive computer vision and machine intelligence libraries, utilities and applications bundled into a single toolkit.
-
-#### Improvements to ROCm Libraries
-* rocSPARSE & hipSPARSE
-* rocBLAS with improved DGEMM efficiency on Vega 7nm
-
-#### MIOpen
-* This release contains general bug fixes and an updated performance database
-* Group convolutions backwards weights performance has been improved
-* RNNs now support fp16
-
-#### Tensorflow multi-gpu and Tensorflow FP16 support for Vega 7nm
-* TensorFlow v1.12 is enabled with fp16 support
-
-#### PyTorch/Caffe2 with Vega 7nm Support
-* fp16 support is enabled
-* Several bug fixes and performance enhancements
-* Known Issue: breaking changes are introduced in ROCm 2.0 which are not addressed upstream yet. Meanwhile, please continue to use ROCm fork at https://github.com/ROCmSoftwarePlatform/pytorch
-
-#### Improvements to ROCProfiler tool
-* Support for Vega 7nm
-
-#### Support for hipStreamCreateWithPriority
-* Creates a stream with the specified priority. It creates a stream on which enqueued kernels have a different priority for execution compared to kernels enqueued on normal priority streams. The priority could be higher or lower than normal priority streams.
-
-#### OpenCL 2.0 support
-* ROCm 2.0 introduces full support for kernels written in the OpenCL 2.0 C language on certain devices and systems.  Applications can detect this support by calling the “clGetDeviceInfo” query function with “parame_name” argument set to “CL_DEVICE_OPENCL_C_VERSION”.  In order to make use of OpenCL 2.0 C language features, the application must include the option “-cl-std=CL2.0” in options passed to the runtime API calls responsible for compiling or building device programs.  The complete specification for the OpenCL 2.0 C language can be obtained using the following link: https://www.khronos.org/registry/OpenCL/specs/opencl-2.0-openclc.pdf
-
-#### Improved Virtual Addressing (48 bit VA) management for Vega 10 and later GPUs
-* Fixes Clang AddressSanitizer and potentially other 3rd-party memory debugging tools with ROCm
-* Small performance improvement on workloads that do a lot of memory management
-* Removes virtual address space limitations on systems with more VRAM than system memory
-
-#### Kubernetes support
-
-### New features and enhancements in ROCm 1.9.2
-#### RDMA(MPI) support on Vega 7nm
-* Support ROCnRDMA based on Mellanox InfiniBand
-
-#### Improvements to HCC
-* Improved link time optimization
-
-#### Improvements to ROCProfiler tool
-* General bug fixes and implemented versioning APIs
-
-### New features and enhancements in ROCm 1.9.2
-#### RDMA(MPI) support on Vega 7nm
-* Support ROCnRDMA based on Mellanox InfiniBand
-
-#### Improvements to HCC
-* Improved link time optimization
-
-#### Improvements to ROCProfiler tool
-* General bug fixes and implemented versioning APIs
-
-#### Critical bug fixes
-
-### New features and enhancements in ROCm 1.9.1
-#### Added DPM support to Vega 7nm
-* Dynamic Power Management feature is enabled on Vega 7nm.
-
-#### Fix for 'ROCm profiling' that used to fail with a “Version mismatch between HSA runtime and libhsa-runtime-tools64.so.1” error
-
-### New features and enhancements in ROCm 1.9.0
-
-#### Preview for Vega 7nm
-* Enables developer preview support for Vega 7nm
-
-#### System Management Interface
-* Adds support for the ROCm SMI (System Management Interface) library, which provides monitoring and management capabilities for AMD GPUs.
-
-#### Improvements to HIP/HCC
-* Support for gfx906
-* Added deprecation warning for C++AMP.  This will be the last version of HCC supporting C++AMP.
-* Improved optimization for global address space pointers passing into a GPU kernel
-* Fixed several race conditions in the HCC runtime
-* Performance tuning to the unpinned copy engine
-* Several codegen enhancement fixes in the compiler backend
-
-#### Preview for rocprof Profiling Tool
-Developer preview (alpha) of profiling tool rocProfiler. It includes a command-line front-end, `rpl_run.sh`, which enables:
-* Cmd-line tool for dumping public per kernel perf-counters/metrics and kernel timestamps
-* Input file with counters list and kernels selecting parameters
-* Multiple counters groups and app runs supported
-* Output results in CSV format
-
-The tool can be installed from the `rocprofiler-dev` package. It will be installed into: `/opt/rocm/bin/rpl_run.sh`
-
-#### Preview for rocr Debug Agent rocr_debug_agent
-The ROCr Debug Agent is a library that can be loaded by ROCm Platform Runtime to provide the following functionality:
-* Print the state for wavefronts that report memory violation or upon executing a "s_trap 2" instruction.
-* Allows SIGINT (`ctrl c`) or SIGTERM (`kill -15`) to print wavefront state of aborted GPU dispatches.
-* It is enabled on Vega10 GPUs on ROCm1.9.
-
-The ROCm1.9 release will install the ROCr Debug Agent library at `/opt/rocm/lib/librocr_debug_agent64.so`
-
-
-#### New distribution support
-
-* Binary package support for Ubuntu 18.04
-
-#### ROCm 1.9 is ABI compatible with KFD in upstream Linux kernels.
-Upstream Linux kernels support the following GPUs in these releases:
-4.17: Fiji, Polaris 10, Polaris 11
-4.18: Fiji, Polaris 10, Polaris 11, Vega10
-
-Some ROCm features are not available in the upstream KFD:
-* More system memory available to ROCm applications
-* Interoperability between graphics and compute
-* RDMA
-* IPC
-
-To try ROCm with an upstream kernel, install ROCm as normal, but do not install the rock-dkms package. Also add a udev rule to control `/dev/kfd` permissions:
-
-```
-    echo 'SUBSYSTEM=="kfd", KERNEL=="kfd", TAG+="uaccess", GROUP="video"' | sudo tee /etc/udev/rules.d/70-kfd.rules
-```
-
-### New features as of ROCm 1.8.3
-
-* ROCm 1.8.3 is a minor update meant to fix compatibility issues on Ubuntu releases running kernel 4.15.0-33
-
-### New features as of ROCm 1.8
-
-#### DKMS driver installation
-
- * Debian packages are provided for DKMS on Ubuntu
- * RPM packages are provided for CentOS/RHEL 7.4 and 7.5 support
- * See the [ROCT-Thunk-Interface](https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/tree/roc-1.8.x) and [ROCK-Kernel-Driver](https://github.com/RadeonOpenCompute/ROCK-Kernel-Driver/tree/roc-1.8.x) for additional documentation on driver setup
-
-#### New distribution support
-
- * Binary package support for Ubuntu 16.04 and 18.04
- * Binary package support for CentOS 7.4 and 7.5
- * Binary package support for RHEL 7.4 and 7.5
-
-#### Improved OpenMPI via UCX support
-
- * UCX support for OpenMPI
- * ROCm RDMA
-
-### New Features as of ROCm 1.7
-
-#### DKMS driver installation
-
- * New driver installation uses Dynamic Kernel Module Support (DKMS)
- * Only amdkfd and amdgpu kernel modules are installed to support AMD hardware
- * Currently only Debian packages are provided for DKMS (no Fedora suport available)
- * See the [ROCT-Thunk-Interface](https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/tree/roc-1.7.x) and [ROCK-Kernel-Driver](https://github.com/RadeonOpenCompute/ROCK-Kernel-Driver/tree/roc-1.7.x) for additional documentation on driver setup
-
-### New Features as of ROCm 1.5
-
-#### Developer preview of the new OpenCL 1.2 compatible language runtime and compiler
-
- * OpenCL 2.0 compatible kernel language support with OpenCL 1.2 compatible
-   runtime
- * Supports offline ahead of time compilation today;
-   during the Beta phase we will add in-process/in-memory compilation.
-
-#### Binary Package support for Ubuntu 16.04
-
-#### Binary Package support for Fedora 24 is not currently available
-
-#### Dropping binary package support for Ubuntu 14.04, Fedora 23
-
-#### IPC support