Current Release Notes¶
New features and enhancements in ROCm 3.0¶
Support for CentOS RHEL v7.7¶
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:
Initial distribution of AOMP 0.7-5 in ROCm v3.0¶
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
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.
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 2.10¶
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.
Note: The SUSE Linux® Enterprise Server is a licensed platform. Ensure you have registered and have a license key prior to installation. Use the following SUSE command line to apply your license: SUSEConnect -r < Key>
SLES 15 SP1
The following section tells you how to perform an install and uninstall ROCm on SLES 15 SP 1. Run the following commands once for a fresh install on the operating system:
sudo usermod -a -G video $LOGNAME sudo usermod -a -G sudo $LOGNAME sudo reboot
Install the “dkms” package.
sudo SUSEConnect --product PackageHub/15.1/x86_64 sudo zypper install dkms
Add the ROCm repo.
sudo zypper clean --all sudo zypper addrepo --no-gpgcheck http://repo.radeon.com/rocm/zyp/zypper/ rocm sudo zypper ref zypper install rocm-dkms sudo zypper install rocm-dkms sudo reboot
#Run the following command once
cat <<EOF | sudo tee /etc/modprobe.d/10-unsupported-modules.conf allow_unsupported_modules 1 EOF sudo modprobe amdgpu
Verify the ROCm installation.
Run /opt/rocm/bin/rocminfo and /opt/rocm/opencl/bin/x86_64/clinfo commands to list the GPUs and verify that the ROCm installation is successful.
To uninstall, use the following command:
sudo zypper remove rocm-dkms rock-dkms
#Ensure all other installed packages/components are removed
Note: Ensure all the content in the /opt/rocm directory is completely removed.
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.
The rocProfiler library enables you to profile performance counters and derived metrics. This library supports GFX8/GFX9 and provides a hardware-specific low-level performance analysis interface for profiling of GPU compute applications. The profiling includes hardware performance counters with complex performance metrics.
The rocTracer library provides a specific runtime profiler to trace API and asynchronous activity. The API provides functionality for registering the runtimes API callbacks and the asynchronous activity records pool support.
rocTX provides a C API for code markup for performance profiling and supports annotation of code ranges and ASCII markers.
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.
csrgemm enables the user to perform matrix-matrix multiplication with two sparse matrices in CSR format.
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.
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
Hotfix release ROCm 2.7.2¶
This release is a hotfix for ROCm release 2.7.
Defect fixed in ROCm 2.7.2
A defect in upgrades from older ROCm releases has been fixed.
Hotfix release ROCm 2.7.1¶
This release is a hotfix release for ROCm release 2.7.1, and addresses the defect mentioned below. The features and enhancements as mentioned in ROCm 2.7 remain relevant to ROCm release 2.7.1 as well.
Defect fixed in ROCm 2.7.1¶
rocprofiler –hiptrace and –hsatrace fails to load roctracer library
In ROCm 2.7.1, 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 ex:
/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 2.7.1¶
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:
sudo apt install roctracer-dev CentOS/RHEL 7.6: 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
Added support for RAS on Radeon Instinct MI50, including:
Memory error detection
Memory error detection counter
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 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 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 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
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
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.
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
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.
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
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.
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 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™ 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.
Features and enhancements introduced in previous versions of ROCm can be found in version_history.md
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™ Link enablement¶
ROCm 2.4 adds support to connect two Radeon Instinct MI60 or Radeon Instinct MI50 boards via AMD Infinity Fabric™ 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 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 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
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)
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¶
Features and enhancements introduced in previous versions of ROCm can be found in version_history.md
Adds support for RHEL 7.6 / CentOS 7.6 and Ubuntu 18.04.1¶
Adds support for Vega 7nm, Polaris 12 GPUs¶
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
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
HCC: removed support for C++AMP
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’ “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 ‘rpl_run.sh’, cmd-line front-end for rocProfiler, 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 location is: /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