ROCm 7.0.0 release notes

2025-09-16

96 min read time

Applies to Linux

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

• Release highlights

• Operating system, hardware, and virtualization support changes

• User space, driver, and firmware dependent changes

• ROCm components versioning

• Detailed component changes

• ROCm known issues

• ROCm resolved issues

• ROCm upcoming changes

Note

If you’re using AMD Radeon™ PRO or Radeon GPUs in a workstation setting with a display connected, see the Use ROCm on Radeon GPUs documentation to verify compatibility and system requirements.

Release highlights

The following are notable new features and improvements in ROCm 7.0.0. For changes to individual components, see Detailed component changes.

Operating system, hardware, and virtualization support changes

ROCm 7.0.0 adds support for AMD Instinct MI355X and MI350X. For details, see the full list of Supported GPUs (Linux).

ROCm 7.0.0 adds support for the following operating systems and kernel versions:

• Ubuntu 24.04.3 (kernel: 6.8 [GA], 6.14 [HWE])

• Rocky Linux 9 (kernel: 5.14.0-570)

ROCm 7.0.0 marks the end of support (EoS) for Ubuntu 24.04.2 (kernel: 6.8 [GA], 6.11 [HWE]) and SLES 15 SP6.

For more information about supported operating systems, see Supported operating systems and install instructions.

See the Compatibility matrix for more information about operating system and hardware compatibility.

Virtualization support

ROCm 7.0.0 introduces support for KVM Passthrough for AMD Instinct MI350X and MI355X GPUs.

All KVM-based SR-IOV supported configurations require the GIM SR-IOV driver version 8.4.0.K. Refer to GIM Release note for more details. In addition, support for VMware ESXi 8 has been introduced for AMD Instinct MI300X GPUs. For more information, see Virtualization Support.

Deep learning and AI framework updates

ROCm provides a comprehensive ecosystem for deep learning development. For more information, see Deep learning frameworks for ROCm and the Compatibility matrix for the complete list of Deep learning and AI framework versions tested for compatibility with ROCm.

Updated framework support

ROCm 7.0.0 introduces several newly supported versions of Deep learning and AI frameworks:

PyTorch

ROCm 7.0.0 enables the following PyTorch features:

• Support for PyTorch 2.7.

• Integrated Fused Rope kernels in APEX.

• Compilation of Python C++ extensions using amdclang++.

• Support for channels-last NHWC format for convolutions via MIOpen.

JAX

ROCm 7.0.0 enables support for JAX 0.6.0.

Megatron-LM

Megatron-LM for ROCm now supports:

• Fused Gradient Accumulation via APEX.

• Fused Rope Kernel in APEX.

• Fused_bias_swiglu kernel.

TensorFlow

ROCm 7.0.0 enables support for TensorFlow 2.19.1.

ONNX Runtime

ROCm 7.0.0 enables support for ONNX Runtime 1.22.0.

vLLM

• Support for Open Compute Project (OCP) FP8 data type.

• FP4 precision for Llama 3.1 405B.

Triton

ROCm 7.0.0 enables support for Triton 3.3.0.

New frameworks

AMD ROCm has officially added support for the following Deep learning and AI frameworks:

• Ray is a unified framework for scaling AI and Python applications from your laptop to a full cluster, without changing your code. Ray consists of a core distributed runtime and a set of AI libraries for simplifying machine learning computations. It is currently supported on ROCm 6.4.1. For more information, see Ray compatibility.

• llama.cpp is an open-source framework for Large Language Model (LLM) inference that runs on both central processing units (CPUs) and graphics processing units (GPUs). It is written in plain C/C++, providing a simple, dependency-free setup. It is currently supported on ROCm 6.4.0. For more information, see llama.cpp compatibility.

AMD GPU Driver/ROCm packaging separation

The AMD GPU Driver (amdgpu) is now distributed separately from the ROCm software stack and is stored under in its own location /amdgpu/ in the package repository at repo.radeon.com. The first release is designated as AMD GPU Driver (amdgpu) version 30.10. See the User and kernel-space support matrix for more information.

AMD SMI continues to stay with the ROCm software stack under the ROCm organization repository.

Consolidation of ROCm library repositories

The following ROCm library repositories are migrating from multiple repositories under ROCm to a single repository under rocm-libraries in the ROCm organization GitHub: hipBLAS, hipBLASLt , hipCUB, hipFFT, hipRAND, hipSPARSE, hipSPARSELt, MIOpen, rocBLAS, rocFFT, rocPRIM, rocRAND, rocSPARSE, rocThrust, and Tensile.

Use the new ROCm Libraries repository to access source code, clone projects, and contribute to the code base and documentation.The change helps to streamline development, CI, and integration. For more information about working with the ROCm Libraries repository, see Contributing to the ROCm Libraries in GitHub.

Other ROCm libraries are also in the process of migration along with ROCm tools to rocm-systems. For latest status information, see the README file. The official completion of migration will be communicated in a future ROCm release.

HIP API compatibility improvements

To improve code portability between AMD ROCm and other programming models, HIP API has been updated in ROCm 7.0.0 to simplify cross-platform programming. These changes are incompatible with prior ROCm releases and might require recompiling existing HIP applications for use with ROCm 7.0.0. For more information, see the HIP API 7.0.0 changes and the HIP changelog below.

HIP runtime updates

The HIP runtime now includes support for:

• Open Compute Project (OCP) MX floating-point FP4, FP6, and FP8 data types and APIs.

• Improved logging by adding more precise pointer information and launch arguments for better tracking and debugging in dispatch methods.

• constexpr operators for FP16 and BF16.

• __syncwarp operation.

• The _sync() version of crosslane builtins such as shfl_sync() are enabled by default. These can be disabled by setting the preprocessor macro HIP_DISABLE_WARP_SYNC_BUILTINS.

• Added warp level primitives: __syncwarp and reduce intrinsics (for example, __reduce_add_sync()).

• Support for the flags in APIs as following, now allows uncached memory allocation.

  • hipExtHostRegisterUncached, used in hipHostRegister.

  • hipHostMallocUncached and hipHostAllocUncached, used in hipHostMalloc and hipHostAlloc.

• A new attribute in HIP runtime was implemented which exposes a new device capability of how many compute dies (chiplets, xcc) are available on a given GPU. Developers can get this attribute via the API hipDeviceGetAttribute, to make use of the best cache locality in a kernel, and optimize the Kernel launch grid layout, for performance improvement.

Additionally, the HIP runtime includes functional improvements, which improve functionality, runtime performance, and the user experience. For more information, see HIP changelog below.

Compiler changes and improvements

ROCm 7.0.0 introduces the AMD Next-Gen Fortran compiler. llvm-flang (sometimes called new-flang or flang-18) is a re-implementation of the Fortran frontend. It is a strategic replacement for classic-flang and is developed in LLVM’s upstream repo at llvm/llvm-project.

Key compiler enhancements include:

• Compiler:

  • Improved memory load and store instructions.

  • Updated clang/llvm to AMD clang version 20.0.0git (equivalent to LLVM 20.0.0 with additional out-of-tree patches).

  • Support added for separate debug file generation for device code.

  • llvm-strip now supports AMD GPU device code objects (EM_AMDGPU).

• Comgr:

  • Added support for an in-memory virtual file system (VFS) for storing temporary files generated during intermediate compilation steps. This is designed to improve performance by reducing on-disk file I/O. Currently, VFS is supported only for the device library link step, with plans for expanded support in future releases.

• SPIR-V:

  • Improved target-specific extensions:

    • Added a new target-specific builtin __builtin_amdgcn_processor_is for late or deferred queries of the current target processor.

    • Added a new target-specific builtin __builtin_amdgcn_is_invocable, enabling fine-grained, per-builtin feature availability.

• The compiler driver now uses parallel code generation by default when compiling using full LTO (including when using the -fgpu-rdc option) for HIP. This divides the optimized LLVM IR module into roughly equal partitions before instruction selection and lowering, which can help improve build times.

  Each kernel in the linked LTO module can be put in a separate partition, and any non-inlined function it depends on can be copied alongside it. Thus, while parallel code generation can improve build time, it can duplicate non-inlined, non-kernel functions across multiple partitions, potentially increasing the binary size of the final object file.

  • Compiler option -flto-partitions=<num> is equivalent to the --lto-partitions=<num> LLD option. Controls the number of partitions used for parallel code generation when using full LTO (including when using -fgpu-rdc). The number of partitions must be greater than 0, and a value of 1 turns off the feature. The default value is 8.

  Developers are encouraged to experiment with different numbers of partitions using the -flto-partitions Clang command line option. For experimentation, recommended values are 1 to 16 partitions, with especially large projects containing many kernels potentially benefiting from up to 64 partitions. It is not recommended to use a value greater than the number of threads on the machine. Smaller projects, or those containing only a few kernels, might not benefit at all from partitioning and might even experience a slight increase in build time due to the small overhead of analyzing and partitioning the modules.

• HIPIFY now supports CUDA 12.9.1 APIs:

  • Added support for all new device and host APIs, including FP4, FP6, and FP128– including support for the corresponding ROCm HIP equivalents.

• The HIPCC Perl scripts (hipcc.pl and hipconfig.pl) have been removed in this release.

Library changes and improvements

New data type support

MX-compliant data types bring microscaling support to ROCm. For more information, see the OCP Microscaling (MX) Formats Specification. ROCm 7.0.0 enables functional support for MX data types FP4, FP6, and FP8 on AMD Instinct MI350 Series GPUs in these ROCm libraries:

• Composable Kernel (FP4, FP6, and FP8 only)

• hipBLASLt

The following libraries are updated to support the Open Compute Project (OCP) floating-point FP8 format on MI350 Series GPUs instead of the NANOO FP8 format:

• Composable Kernel

• hipBLASLt

• hipSPARSELt

• MIGraphX

• rocWMMA

For more information about data types, see Data types and precision support.

hipBLASLt improvement

GEMM performance has been improved for FP8, FP16, BF16, and FP32 data types.

For more information about hipBLASLt changes, see the hipBLASLt changelog below.

MIGraphX improvements

• Support for OCP FP8 on AMD Instinct MI350X and MI355X GPUs.

• Support for PyTorch 2.7 via Torch-MIGraphX.

• Improved performance of Generative AI models.

• Added additional MSFT Contrib Operators for improved ONNX Runtime Experience.

For more information about MIGraphX changes, see the MIGraphX changelog below.

rocSHMEM Reverse Offload conduit inter-node support

The rocSHMEM communications library has added the RO (Reverse Offload) inter-node communication backend which enables communication between GPUs on different nodes through a NIC, using a host-based CPU proxy to forward communication orders to and from the GPU. Inter-node communication requires MPI, and is tested with Open MPI and CX7 IB NICs. For more information, see available network backends for installing rocSHMEM.

See the rocSHMEM changelog for more details.

Tool changes and improvements

AMD SMI

Key enhancements to AMD SMI include the ability to reload the AMD GPU driver from the CLI or API. The amd-smi command-line interface gains a new default view, amd-smi topology support in guest environments, and performance optimizations. Additionally, AMD SMI library APIs have been refined for improved usability. See the AMD SMI changelog for more details.

ROCgdb

ROCgdb now supports FP4, FP6, and FP8 micro-scaling (MX) data types with AMD Instinct MI350 Series GPUs.

See the ROCgdb changelog for more details.

ROCm Compute Profiler

ROCm Compute Profiler includes the following key changes:

• Interactive command line with a Textual User Interface (TUI) has been added to analyze mode. For more details, see TUI analysis.

• Support added for advanced data types: FP4 and FP6

• Support for AMD Instinct MI355X and MI350X with addition of performance counters: CPC, SPI, SQ, TA/TD/TCP, and TCC.

• Roofline enhancement added for AMD Instinct MI350 Series.

• Improved support for Selective Kernel profiling.

• Program Counter (PC) sampling (Software-based) feature has been enabled for AMD Instinct MI200, MI300X, MI350X, and MI355X GPUs. This feature helps in GPU profiling to understand code execution patterns and hotspots during GPU kernel execution. For more details, see Using PC sampling in ROCm Compute Profiler.

• Program Counter (PC) sampling (Hardware-based, Stochastic) feature has been enabled for AMD Instinct MI300X, MI350, and MI355X GPUs.

• Docker files has been added to package the application and dependencies into a single portable and executable standalone binary file.

See the ROCm Compute Profiler changelog for more details.

ROCm Data Center (RDC) improvements

The ROCm Data Center tool (RDC) streamlines the administration of AMD GPUs in cluster data center environments. ROCm 7.0.0 introduces new data center management and monitoring tools for system administrators. For more information, see ROCm Data Center (RDC) tool documentation.

ROCm Systems Profiler

ROCm Systems Profiler includes the following key changes:

• Improved profiling support for Computer Vision workloads through rocDecode and rocJPEG API tracing and engine activity sampling.

• Network profiling support has been added to AMD Instinct MI300X, MI350X, and MI355X.

• Improved profiling of the communication layer with RCCL and MPI API tracing.

See the ROCm Systems Profiler changelog for more details.

ROCm Validation Suite

In ROCm 7.0.0, ROCm Validation Suite includes support for the AMD Instinct MI355X and MI350X GPUs in the IET (Integrated Execution Test), GST (GPU Stress Test), and Babel (memory bandwidth test) modules.

See the ROCm Validation Suite changelog for more details.

ROCprofiler-SDK

Core SDK enhancements

• ROCprofiler-SDK is now compatible with the HIP 7.0.0 API.

• ROCprofiler-SDK adds support for AMD Instinct MI350X and MI355X GPUs.

• The stochastic and host-trap PC sampling support has been added for all AMD Instinct MI300 and MI350 Series GPUs, which provides information particularly useful for understanding stalls during kernel execution.

• The added support for tracing events surfaced by AMD’s Kernel Fusion Driver (KFD) captures low-level driver routines involved in mapping, invalidation, and migration of data between CPU and GPU memories. Such events are central to the support for Unified Memory on AMD systems. Tracing of KFD events helps to detect performance problems arising from excessive data migration.

• New APIs are added for profiling applications using thread traces (beta) which facilitates profiling wavefronts at the instruction timing level.

rocpd

The ROCm Profiling Data (rocpd) is now the default output format for rocprofv3. A subproject of the ROCprofiler-SDK, rocpd enables saving profiling results to a SQLite3 database, providing a structured and efficient foundation for analysis and post-processing.

rocprofv3 CLI tool enhancements

• Added stochastic and host-trap PC sampling support for all AMD Instinct MI300 and MI350 Series GPUs.

• HIP streams translate to Queues in Time Traces in Perfetto output.

• Support for thread trace service.

See the ROCprofiler-SDK changelog for more details.

ROCm Offline Installer Creator updates

The ROCm Offline Installer Creator 7.0.0 includes the following features and improvements:

• Added support for Rocky Linux 9.6.

• Added support for the new graphics repo structure for graphics/Mesa related packages.

• Improvements to kernel header version matching for AMDGPU driver installation.

• Added support for creating an offline installer when the kernel version of the target operating system differs from the operating system of the host creating the installer (for Ubuntu 22.04 and 24.04 only).

See ROCm Offline Installer Creator for more information.

ROCm Runfile Installer updates

The ROCm Runfile Installer 7.0.0 adds the following features and improvements:

• Added support for Rocky Linux 9.6.

• Added untar mode for the .run file to allow extraction of ROCm to a given directory, similar to a normal tarball.

• Added an RVS test script.

• Fixes to the rocm-examples test script.

• Fixes for clinfo and OpenCL use after installation.

For more information, see ROCm Runfile Installer.

ROCm documentation updates

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

• The ROCm AI training and inference benchmarking guides have been updated with expanded model coverage and optimized Docker environments. Highlights include:

  • The Training a model with Primus and Megatron benchmarking guide now leverages the unified AMD Primus framework with the Megatron backend. See Primus: A Lightweight, Unified Training Framework for Large Models on AMD GPUs for an introduction to Primus.

  • The Training a model with PyTorch benchmarking guide now includes fine-tuning for OpenAI GPT OSS and Qwen models. It also includes a multi-node training example.

  • The Training a model with JAX MaxText benchmarking guide now supports MAD-integrated benchmarking. The MaxText training environment now uses JAX 0.6.0 or 0.5.0. FP8 quantized training is supported with JAX 0.5.0.

  • The vLLM inference performance testing documentation now features clearer serving and throughput benchmarking commands – for improved transparency of model benchmarking configurations. The vLLM inference environment now uses vLLM 0.10.1 and includes improved default configurations.

  These training and inference resources will continue to grow with ongoing improvements and expanded model coverage. For a searchable view of supported frameworks and models, see AMD Infinity Hub.

• Tutorials for AI developers have been expanded with the following new inference tutorial: PD disaggregation with SGLang

  In addition, the AI agent with MCPs using vLLM and PydanticAI tutorial has been updated. For more information about the changes, see Changelog for the AI Developer Hub.

• Documentation for rocCV, an efficient GPU-accelerated library for image pre- and post-processing, has been added. rocCV is in an early access state, and using it on production workloads is not recommended.

• ROCm Math libraries support a wide range of data types, enabling optimized performance across various precision requirements. The following Math libraries are now updated with new precision content. For more information, click the Math library’s link:

  • hipBLAS

  • hipBLASLt

  • hipSPARSE

  • rocSPARSE

  • Tensile

• ROCm offers a comprehensive ecosystem for deep learning development, featuring libraries optimized for deep learning operations and ROCm-aware versions of popular deep learning frameworks and libraries. The following deep learning frameworks’ content now includes release notes and known issues:

  • PyTorch

  • JAX

• ROCm components support a wide range of environment variables that can be used for testing, logging, debugging, experimental features, and more. The following components have been updated with new environment variable content. For more information, click the component’s link:

  • hipBLASLt

  • hipSPARSELt

  • ROCm Performance Primitives (RPP)

  • rocSOLVER

  • rocSPARSE

  • Tensile

• Modern computing tasks often require balancing numerical precision against hardware resources and processing speed. Low precision floating point number formats in HIP include FP4 (4-bit) and FP6 (6-bit), which reduce memory and bandwidth requirements. For more information, see the updated Low precision floating point types topic.

User space, driver, and firmware dependent changes

The software for AMD Datacenter GPU products requires maintaining a hardware and software stack with interdependencies between the GPU and baseboard firmware, AMD GPU drivers, and the ROCm user space software.

As of the ROCm 7.0.0 release, these interdependencies are publicly documented. Note that while AMD publishes drivers and ROCm user space, your server or infrastructure provider publishes the GPU and baseboard firmware by bundling AMD’s firmware releases via AMD’s Platform Level Data Model (PLDM) bundle, which includes Integrated Firmware Image (IFWI).

GPU and baseboard firmware versioning might differ across GPU families. With the ROCm 7.0.0 release, the AMD GPU driver (amdgpu) is now versioned separately from ROCm. See AMD GPU Driver/ROCm packaging separation.

ROCm Version	GPU	PLDM Bundle (Firmware)	AMD GPU Driver (amdgpu)	AMD GPU Virtualization Driver (GIM)
ROCm 7.0.0	MI355X	01.25.13.04 (or later) 01.25.11.02	30.10	8.4.0.K
	MI350X	01.25.13.04 (or later) 01.25.11.02	30.10
	MI325X	01.25.04.00 (or later) 01.25.03.03	30.10 6.4.z where z (0-3) 6.3.y where y (1-3)
	MI300X	01.25.03.12 (or later) 01.25.02.04	30.10 6.4.z where z (0–3) 6.3.y where y (0–3) 6.2.x where x (1–4)	8.4.0.K
	MI300A	26 (or later)		Not Applicable
	MI250X	IFWI 47 (or later)
	MI250	MU5 w/ IFWI 75 (or later)
	MI210	MU5 w/ IFWI 75 (or later)		8.4.0.K
	MI100	VBIOS D3430401-037		Not Applicable

New feature details

AMD SMI changes dependent on PLDM bundles (firmware)

New APIs introduced in AMD SMI for ROCm 7.0.0 provide additional data for the AMD Instinct products. To support these features, the following firmware for each GPUs are required:

• AMD Instinct MI355X - PLDM bundle 01.25.13.04

• AMD Instinct MI350X - PLDM bundle 01.25.13.04

• AMD Instinct MI325X - PLDM bundle 01.25.04.00

• AMD Instinct MI300X - PLDM bundle 01.25.03.12

If ROCm 7.0.0 is applied on system with prior version of PLDM bundles (firmware), the new APIs will return N/A to indicate lack of support for these items.

Enhanced temperature telemetry introduced in AMD SMI for MI355X and MI350X GPUs

AMD SMI in ROCm 7.0.0 provides support for enhanced temperature metrics and temperature anomaly detection for AMD Instinct MI350X and MI355X GPUs when paired with: PLDM bundle 01.25.13.04.

For more information on these features, see AMD SMI changelog.

KVM SR-IOV virtualization changes dependent on open source AMD GPU Virtualization Driver (GIM)

KVM SR-IOV support for all Instinct GPUs require the open source AMD GPU Virtualization Driver (GIM) 8.4.0.K. For detailed support information, see virtualization support and GIM Release Note.

GPU partitioning support for AMD Instinct MI355X and MI350X GPUs

NPS2 and DPX partitioning on bare metal is enabled on AMD Instinct MI355X and MI350X GPUs on ROCm 7.0.0 when paired with: PLDM bundle 01.25.13.04.

ROCm components

The following table lists the versions of ROCm components for ROCm 7.0.0, including any version changes from 6.4.3 to 7.0.0. Click the component’s updated version to go to a list of its changes.

Click to go to the component’s source code on GitHub.

Category	Group	Name	Version
Libraries	Machine learning and computer vision	Composable Kernel	1.1.0 ⇒ 1.1.0
		MIGraphX	2.12.0 ⇒ 2.13.0
		MIOpen	3.4.0 ⇒ 3.5.0
		MIVisionX	3.2.0 ⇒ 3.3.0
		rocAL	2.2.0 ⇒ 2.3.0
		rocDecode	0.10.0 ⇒ 1.0.0
		rocJPEG	0.8.0 ⇒ 1.1.0
		rocPyDecode	0.3.1 ⇒ 0.6.0
		RPP	1.9.10 ⇒ 2.0.0
	Communication	RCCL	2.22.3 ⇒ 2.26.6
		rocSHMEM	2.0.1 ⇒ 3.0.0
	Math	hipBLAS	2.4.0 ⇒ 3.0.0
		hipBLASLt	0.12.1 ⇒ 1.0.0
		hipFFT	1.0.18 ⇒ 1.0.20
		hipfort	0.6.0 ⇒ 0.7.0
		hipRAND	2.12.0 ⇒ 3.0.0
		hipSOLVER	2.4.0 ⇒ 3.0.0
		hipSPARSE	3.2.0 ⇒ 4.0.1
		hipSPARSELt	0.2.3 ⇒ 0.2.4
		rocALUTION	3.2.3 ⇒ 4.0.0
		rocBLAS	4.4.1 ⇒ 5.0.0
		rocFFT	1.0.32 ⇒ 1.0.34
		rocRAND	3.3.0 ⇒ 4.0.0
		rocSOLVER	3.28.2 ⇒ 3.30.0
		rocSPARSE	3.4.0 ⇒ 4.0.2
		rocWMMA	1.7.0 ⇒ 2.0.0
		Tensile	4.43.0 ⇒ 4.44.0
	Primitives	hipCUB	3.4.0 ⇒ 4.0.0
		hipTensor	1.5.0 ⇒ 2.0.0
		rocPRIM	3.4.1 ⇒ 4.0.0
		rocThrust	3.3.0 ⇒ 4.0.0
Tools	System management	AMD SMI	25.5.1 ⇒ 26.0.0
		ROCm Data Center Tool	0.3.0 ⇒ 1.1.0
		rocminfo	1.0.0
		ROCm SMI	7.7.0 ⇒ 7.8.0
		ROCm Validation Suite	1.1.0 ⇒ 1.2.0
			Performance	ROCm Bandwidth Test	1.4.0 ⇒ 2.6.0
				ROCm Compute Profiler	3.1.1 ⇒ 3.2.3
ROCm Systems Profiler	1.0.2 ⇒ 1.1.0
ROCProfiler	2.0.0
ROCprofiler-SDK	0.6.0 ⇒ 1.0.0
ROCTracer	4.1.0
	Development	HIPIFY	19.0.0 ⇒ 20.0.0
		ROCdbgapi	0.77.2 ⇒ 0.77.3
		ROCm CMake	0.14.0
		ROCm Debugger (ROCgdb)	15.2 ⇒ 16.3
		ROCr Debug Agent	2.0.4 ⇒ 2.1.0
Compilers		HIPCC	1.1.1
		llvm-project	19.0.0 ⇒ 20.0.0
Runtimes		HIP	6.4.3 ⇒ 7.0.0
		ROCr Runtime	1.15.0 ⇒ 1.18.0

Detailed component changes

The following sections describe key changes to ROCm components.

Note

For a historical overview of ROCm component updates, see the ROCm consolidated changelog.

AMD SMI (26.0.0)

Added

• Ability to restart the AMD GPU driver from the CLI and API.

  • amdsmi_gpu_driver_reload() API and amd-smi reset --reload-driver or amd-smi reset -r CLI options.

  • Driver reload functionality is now separated from memory partition functions; memory partition change requests should now be followed by a driver reload.

  • Driver reload requires all GPU activity on all devices to be stopped.

• Default command:

  A default view has been added. The default view provides a snapshot of commonly requested information such as bdf, current partition mode, version information, and more. Users can access that information by simply typing amd-smi with no additional commands or arguments. Users may also obtain this information through alternate output formats such as json or csv by using the default command with the respective output format: amd-smi default --json or amd-smi default --csv.

• Support for GPU metrics 1.8:

  • Added new fields for amdsmi_gpu_xcp_metrics_t including:

    • Metrics to allow new calculations for violation status:

      • Per XCP metrics gfx_below_host_limit_ppt_acc[XCP][MAX_XCC] - GFX Clock Host limit Package Power Tracking violation counts

      • Per XCP metrics gfx_below_host_limit_thm_acc[XCP][MAX_XCC] - GFX Clock Host limit Thermal (TVIOL) violation counts

      • Per XCP metrics gfx_low_utilization_acc[XCP][MAX_XCC] - violation counts for how did low utilization caused the GPU to be below application clocks.

      • Per XCP metrics gfx_below_host_limit_total_acc[XCP][MAX_XCC]- violation counts for how long GPU was held below application clocks any limiter (see above new violation metrics).

  • Increased available JPEG engines to 40. Current ASICs might not support all 40. These are indicated as UINT16_MAX or N/A in CLI.

• Bad page threshold count.

  • Added amdsmi_get_gpu_bad_page_threshold to Python API and CLI; root/sudo permissions are required to display the count.

• CPU model name for RDC.

  • Added new C and Python API amdsmi_get_cpu_model_name.

  • Not sourced from esmi library.

• New API amdsmi_get_cpu_affinity_with_scope().

• socket power to amdsmi_get_power_info

  • Previously, the C API had the value in the amdsmi_power_info structure, but was unused.

  • The value is representative of the socket’s power agnostic of the the GPU version.

• New event notification types to amdsmi_evt_notification_type_t. The following values were added to the amdsmi_evt_notification_type_t enum:

  • AMDSMI_EVT_NOTIF_EVENT_MIGRATE_START

  • AMDSMI_EVT_NOTIF_EVENT_MIGRATE_END

  • AMDSMI_EVT_NOTIF_EVENT_PAGE_FAULT_START

  • AMDSMI_EVT_NOTIF_EVENT_PAGE_FAULT_END

  • AMDSMI_EVT_NOTIF_EVENT_QUEUE_EVICTION

  • AMDSMI_EVT_NOTIF_EVENT_QUEUE_RESTORE

  • AMDSMI_EVT_NOTIF_EVENT_UNMAP_FROM_GPU

  • AMDSMI_EVT_NOTIF_PROCESS_START

  • AMDSMI_EVT_NOTIF_PROCESS_END

• Power cap to amd-smi monitor.

  • amd-smi monitor -p will display the power cap along with power.

Changed

• Separated driver reload functionality from amdsmi_set_gpu_memory_partition() and amdsmi_set_gpu_memory_partition_mode() APIs – and from the CLI amd-smi set -M <NPS mode>.

• Disabled amd-smi monitor --violation on guests. Modified amd-smi metric -T/--throttle to alias to amd-smi metric -v/--violation.

• Updated amdsmi_get_clock_info in amdsmi_interface.py.

  • The clk_deep_sleep field now returns the sleep integer value.

• The amd-smi topology command has been enabled for guest environments.

  • This includes full functionality so users can use the command just as they would in bare metal environments.

• Expanded violation status tracking for GPU metrics 1.8.

  • The driver will no longer be supporting existing single-value GFX clock below host limit fields (acc_gfx_clk_below_host_limit, per_gfx_clk_below_host_limit, active_gfx_clk_below_host_limit), they are now changed in favor of new per-XCP/XCC arrays.

  • Added new fields to amdsmi_violation_status_t and related interfaces for enhanced violation breakdown:

    • Per-XCP/XCC accumulators and status for:

      • GFX clock below host limit (power, thermal, and total)

      • Low utilization

    • Added 2D arrays to track per-XCP/XCC accumulators, percentage, and active status:

      • acc_gfx_clk_below_host_limit_pwr, acc_gfx_clk_below_host_limit_thm, acc_gfx_clk_below_host_limit_total

      • per_gfx_clk_below_host_limit_pwr, per_gfx_clk_below_host_limit_thm, per_gfx_clk_below_host_limit_total

      • active_gfx_clk_below_host_limit_pwr, active_gfx_clk_below_host_limit_thm, active_gfx_clk_below_host_limit_total

      • acc_low_utilization, per_low_utilization, active_low_utilization

  • Python API and CLI now report these expanded fields.

• The char arrays in the following structures have been changed.

  • amdsmi_vbios_info_t member build_date changed from AMDSMI_MAX_DATE_LENGTH to AMDSMI_MAX_STRING_LENGTH.

  • amdsmi_dpm_policy_entry_t member policy_description changed from AMDSMI_MAX_NAME to AMDSMI_MAX_STRING_LENGTH.

  • amdsmi_name_value_t member name changed from AMDSMI_MAX_NAME to AMDSMI_MAX_STRING_LENGTH.

• For backwards compatibility, updated amdsmi_bdf_t union to have an identical unnamed struct.

• Updated amdsmi_get_temp_metric and amdsmi_temperature_type_t with new values.

  • Added new values to amdsmi_temperature_type_t representing various baseboard and GPU board temperature measures.

  • Updated amdsmi_get_temp_metric API to be able to take in and return the respective values for the new temperature types.

Removed

• Unnecessary API, amdsmi_free_name_value_pairs()

  • This API is only used internally to free up memory from the Python interface and does not need to be exposed to the user.

• Unused definitions:

  • AMDSMI_MAX_NAME, AMDSMI_256_LENGTH, AMDSMI_MAX_DATE_LENGTH, MAX_AMDSMI_NAME_LENGTH, AMDSMI_LIB_VERSION_YEAR, AMDSMI_DEFAULT_VARIANT, AMDSMI_MAX_NUM_POWER_PROFILES, AMDSMI_MAX_DRIVER_VERSION_LENGTH.

• Unused member year in struct amdsmi_version_t.

• amdsmi_io_link_type_t has been replaced with amdsmi_link_type_t.

  • amdsmi_io_link_type_t is no longer needed as amdsmi_link_type_t is sufficient.

  • amdsmi_link_type_t enum has changed; primarily, the ordering of the PCI and XGMI types.

  • This change will also affect amdsmi_link_metrics_t, where the link_type field changes from amdsmi_io_link_type_t to amdsmi_link_type_t.

• amdsmi_get_power_info_v2().

  • The amdsmi_get_power_info() has been unified and the v2 function is no longer needed or used.

• AMDSMI_EVT_NOTIF_RING_HANG event notification type in amdsmi_evt_notification_type_t.

• The amdsmi_get_gpu_vram_info now provides vendor names as a string.

  • amdsmi_vram_vendor_type_t enum structure is removed.

  • amdsmi_vram_info_t member named amdsmi_vram_vendor_type_t is changed to a character string.

  • amdsmi_get_gpu_vram_info now no longer requires decoding the vendor name as an enum.

• Backwards compatibility for amdsmi_get_gpu_metrics_info()’s,jpeg_activityand vcn_activity fields. Alternatively use xcp_stats.jpeg_busy or xcp_stats.vcn_busy.

  • Backwards compatibility is removed for jpeg_activity and vcn_activity fields, if the jpeg_busy or vcn_busy field is available.

  • Providing both vcn_activity/jpeg_activity and XCP (partition) stats vcn_busy/jpeg_busy caused confusion about which field to use. By removing backward compatibility, it is easier to identify the relevant field.

  • The jpeg_busy field increased in size (for supported ASICs), making backward compatibility unable to fully copy the structure into jpeg_activity.

Optimized

• Reduced amd-smi CLI API calls needed to be called before reading or (re)setting GPU features. This improves overall runtime performance of the CLI.

• Removed partition information from the default amd-smi static CLI command.

  • Users can still retrieve the same data by calling amd-smi, amd-smi static -p, or amd-smi partition -c -m/sudo amd-smi partition -a.

  • Reading current_compute_partition may momentarily wake the GPU up. This is due to reading XCD registers, which is expected behavior. Changing partitions is not a trivial operation, current_compute_partition SYSFS controls this action.

• Optimized CLI command amd-smi topology in partition mode.

  • Reduced the number of amdsmi_topo_get_p2p_status API calls to one fourth.

Resolved issues

• Removed duplicated GPU IDs when receiving events using the amd-smi event command.

• Fixed amd-smi monitor decoder utilization (DEC%) not showing up on MI300 Series ASICs.

Known issues

• amd-smi monitor on Linux Guest systems triggers an attribute error.

Note

See the full AMD SMI changelog for details, examples, and in-depth descriptions.

Composable Kernel (1.1.0)

Added

• Support for BF16, F32, and F16 for 2D and 3D NGCHW grouped convolution backward data.

• Fully asynchronous HOST (CPU) arguments copy flow for CK grouped GEMM kernels.

• Support GKCYX for layout for grouped convolution forward (NGCHW/GKCYX/NGKHW, number of instances in instance factory for NGCHW/GKYXC/NGKHW has been reduced).

• Support for GKCYX layout for grouped convolution forward (NGCHW/GKCYX/NGKHW).

• Support for GKCYX layout for grouped convolution backward weight (NGCHW/GKCYX/NGKHW).

• Support for GKCYX layout for grouped convolution backward data (NGCHW/GKCYX/NGKHW).

• Support for Stream-K version of mixed FP8 / BF16 GEMM.

• Support for Multiple D GEMM.

• GEMM pipeline for microscaling (MX) FP8 / FP6 / FP4 data types.

• Support for FP16 2:4 structured sparsity to universal GEMM.

• Support for Split K for grouped convolution backward data.

• Logit soft-capping support for fMHA forward kernels.

• Support for hdim as a multiple of 32 for FMHA (fwd/fwd_splitkv).

• Benchmarking support for tile engine GEMM.

• Ping-pong scheduler support for GEMM operation along the K dimension.

• Rotating buffer feature for CK_Tile GEMM.

• int8 support for CK_TILE GEMM.

• Vectorize Transpose optimization for CK Tile.

• Asynchronous copy for gfx950.

Changed

• Replaced the raw buffer load/store intrinsics with Clang20 built-ins.

• DL and DPP kernels are now enabled by default.

• Number of instances in instance factory for grouped convolution forward NGCHW/GKYXC/NGKHW has been reduced.

• Number of instances in instance factory for grouped convolution backward weight NGCHW/GKYXC/NGKHW has been reduced.

• Number of instances in instance factory for grouped convolution backward data NGCHW/GKYXC/NGKHW has been reduced.

Removed

• Removed support for gfx940 and gfx941 targets.

Optimized

• Optimized the GEMM multiply preshuffle and lds bypass with Pack of KGroup and better instruction layout.

HIP 7.0.0

Added

• New HIP APIs

  • hipLaunchKernelEx dispatches the provided kernel with the given launch configuration and forwards the kernel arguments.

  • hipLaunchKernelExC launches a HIP kernel using a generic function pointer and the specified configuration.

  • hipDrvLaunchKernelEx dispatches the device kernel represented by a HIP function object.

  • hipMemGetHandleForAddressRange gets a handle for the address range requested.

  • __reduce_add_sync, __reduce_min_sync, and __reduce_max_sync functions added for aritimetic reduction across lanes of a warp, and __reduce_and_sync, __reduce_or_sync, and __reduce_xor_sync functions added for logical reduction. For details, see Warp cross-lane functions.

• New support for Open Compute Project (OCP) floating-point FP4/FP6/FP8 as follows. For details, see Low precision floating point document.

  • Data types for FP4/FP6/FP8.

  • HIP APIs for FP4/FP6/FP8, which are compatible with corresponding CUDA APIs.

  • HIP Extensions APIs for microscaling formats, which are supported on AMD GPUs.

• New wptr and rptr values in ClPrint, for better logging in dispatch barrier methods.

• The _sync() version of crosslane builtins such as shfl_sync() are enabled by default. These can be disabled by setting the preprocessor macro HIP_DISABLE_WARP_SYNC_BUILTINS.

• Added constexpr operators for fp16/bf16.

• Added warp level primitives: __syncwarp and reduce intrinsics (e.g. __reduce_add_sync()).

• Support for the flags in APIs as following, now allows uncached memory allocation.

  • hipExtHostRegisterUncached, used in hipHostRegister.

  • hipHostMallocUncached and hipHostAllocUncached, used in hipHostMalloc and hipHostAlloc.

• num_threads total number of threads in the group. The legacy API size is alias.

• Added PCI CHIP ID information as the device attribute.

• Added new tests applications for OCP data types FP4/FP6/FP8.

• A new attribute in HIP runtime was implemented which exposes a new device capability of how many compute dies (chiplets, xcc) are available on a given GPU. Developers can get this attribute via the API hipDeviceGetAttribute, to make use of the best cache locality in a kernel, and optimize the Kernel launch grid layout, for performance improvement.

Changed

• Some unsupported GPUs such as gfx9, gfx8 and gfx7 are deprecated on Microsoft Windows.

• Removal of beta warnings in HIP Graph APIs. All Beta warnings in usage of HIP Graph APIs are removed, they are now officially and fully supported.

• warpSize has changed. In order to match the CUDA specification, the warpSize variable is no longer constexpr. In general, this should be a transparent change; however, if an application was using warpSize as a compile-time constant, it will have to be updated to handle the new definition. For more information, see the discussion of warpSize within the HIP C++ language extensions.

• Behavior changes

  • hipGetLastError now returns the error code which is the last actual error caught in the current thread during the application execution.

  • Cooperative groups in hipLaunchCooperativeKernelMultiDevice and hipLaunchCooperativeKernel functions, additional input parameter validation checks are added.

  • hipPointerGetAttributes returns hipSuccess instead of an error with invalid value hipErrorInvalidValue, in case NULL host or attribute pointer is passed as input parameter. It now matches the functionality of cudaPointerGetAttributes which changed with CUDA 11 and above releases.

  • hipFree previously there was an implicit wait which was applicable for all memory allocations, for synchronization purpose. This wait is now disabled for allocations made with hipMallocAsync and hipMallocFromPoolAsync, to match the behavior of CUDA API cudaFree.

  • hipFreeAsync now returns hipSuccess when the input pointer is NULL, instead of hipErrorInvalidValue , to be consistent with hipFree.

  • Exceptions occurring during a kernel execution will not abort the process anymore but will return an error unless core dump is enabled.

• Changes in hipRTC.

  • Removal of hipRTC symbols from HIP Runtime Library. Any application using hipRTC APIs should link explicitly with the hipRTC library. This makes the usage of hipRTC library on Linux the same as on Windows and matches the behavior of CUDA nvRTC.

  • hipRTC compilation The device code compilation now uses namespace __hip_internal, instead of the standard headers std, to avoid namespace collision.

  • Changes of datatypes from hipRTC. Datatype definitions such as int64_t, uint64_t, int32_t, and uint32_t, etc. are removed to avoid any potential conflicts in some applications. HIP now uses internal datatypes instead, prefixed with __hip, for example, __hip_int64_t.

• HIP header clean up

  • Usage of STD headers, HIP header files only include necessary STL headers.

  • Deprecated structure HIP_MEMSET_NODE_PARAMS is removed. Developers can use the definition hipMemsetParams instead.

• API signature/struct changes

  • API signatures are adjusted in some APIs to match corresponding CUDA APIs. Impacted APIs are as folloing:

    • hiprtcCreateProgram

    • hiprtcCompileProgram

    • hipMemcpyHtoD

    • hipCtxGetApiVersion

  • HIP struct change in hipMemsetParams, it is updated and compatible with CUDA.

  • HIP vector constructor change in hipComplex initialization now generates correct values. The affected constructors will be small vector types such as float2, int4, etc.

• Stream Capture updates

  • Restricted stream capture mode, it is made in HIP APIs via adding the macro CHECK_STREAM_CAPTURE_SUPPORTED (). In the previous HIP enumeration hipStreamCaptureMode, three capture modes were defined. With checking in the macro, the only supported stream capture mode is now hipStreamCaptureModeRelaxed. The rest are not supported, and the macro will return hipErrorStreamCaptureUnsupported. This update involves the following APIs, which is allowed only in relaxed stream capture mode:

    • hipMallocManaged

    • hipMemAdvise

  • Checks stream capture mode, the following APIs check the stream capture mode and return error codes to match the behavior of CUDA.

    • hipLaunchCooperativeKernelMultiDevice

    • hipEventQuery

    • hipStreamAddCallback

  • Returns error during stream capture. The following HIP APIs now returns specific error hipErrorStreamCaptureUnsupported on the AMD platform, but not always hipSuccess, to match behavior with CUDA:

    • hipDeviceSetMemPool

    • hipMemPoolCreate

    • hipMemPoolDestroy

    • hipDeviceSetSharedMemConfig

    • hipDeviceSetCacheConfig

    • hipMemcpyWithStream

• Error code update Returned error/value codes are updated in the following HIP APIs to match the corresponding CUDA APIs.

  • Module Management Related APIs:

    • hipModuleLaunchKernel

    • hipExtModuleLaunchKernel

    • hipExtLaunchKernel

    • hipDrvLaunchKernelEx

    • hipLaunchKernel

    • hipLaunchKernelExC

    • hipModuleLaunchCooperativeKernel

    • hipModuleLoad

  • Texture Management Related APIs: The following APIs update the return codes to match the behavior with CUDA:

    • hipTexObjectCreate, supports zero width and height for 2D image. If either is zero, will not return false.

    • hipBindTexture2D, adds extra check, if pointer for texture reference or device is NULL, returns hipErrorNotFound.

    • hipBindTextureToArray, if any NULL pointer is input for texture object, resource descriptor, or texture descriptor, returns error hipErrorInvalidChannelDescriptor, instead of hipErrorInvalidValue.

    • hipGetTextureAlignmentOffset, adds a return code hipErrorInvalidTexture when the texture reference pointer is NULL.

  • Cooperative Group Related APIs, more calidations are added in the following API implementation:

    • hipLaunchCooperativeKernelMultiDevice

    • hipLaunchCooperativeKernel

• Invalid stream input parameter handling In order to match the CUDA runtime behavior more closely, HIP APIs with streams passed as input parameters no longer check the stream validity. Previously, the HIP runtime returned an error code hipErrorContextIsDestroyed if the stream was invalid. In CUDA version 12 and later, the equivalent behavior is to raise a segmentation fault. HIP runtime now matches the CUDA by causing a segmentation fault. The list of APIs impacted by this change are as follows:

  • Stream Management Related APIs

    • hipStreamGetCaptureInfo

    • hipStreamGetPriority

    • hipStreamGetFlags

    • hipStreamDestroy

    • hipStreamAddCallback

    • hipStreamQuery

    • hipLaunchHostFunc

  • Graph Management Related APIs

    • hipGraphUpload

    • hipGraphLaunch

    • hipStreamBeginCaptureToGraph

    • hipStreamBeginCapture

    • hipStreamIsCapturing

    • hipStreamGetCaptureInfo

    • hipGraphInstantiateWithParams

  • Memory Management Related APIs

    • hipMemcpyPeerAsync

    • hipMemcpy2DValidateParams

    • hipMallocFromPoolAsync

    • hipFreeAsync

    • hipMallocAsync

    • hipMemcpyAsync

    • hipMemcpyToSymbolAsync

    • hipStreamAttachMemAsync

    • hipMemPrefetchAsync

    • hipDrvMemcpy3D

    • hipDrvMemcpy3DAsync

    • hipDrvMemcpy2DUnaligned

    • hipMemcpyParam2D

    • hipMemcpyParam2DAsync

    • hipMemcpy2DArrayToArray

    • hipMemcpy2D

    • hipMemcpy2DAsync

    • hipDrvMemcpy2DUnaligned

    • hipMemcpy3D

  • Event Management Related APIs

    • hipEventRecord

    • hipEventRecordWithFlags

Optimized

HIP runtime has the following functional improvements which improves runtime performance and user experience:

• Reduced usage of the lock scope in events and kernel handling.

  • Switches to shared_mutex for event validation, uses std::unique_lock in HIP runtime to create/destroy event, instead of scopedLock.

  • Reduces the scopedLock in handling of kernel execution. HIP runtime now calls scopedLock during kernel binary creation/initialization, doesn’t call it again during kernel vector iteration before launch.

• Implementation of unifying managed buffer and kernel argument buffer so HIP runtime doesn’t need to create/load a separate kernel argument buffer.

• Refactored memory validation, creates a unique function to validate a variety of memory copy operations.

• Improved kernel logging using demangling shader names.

• Advanced support for SPIRV, now kernel compilation caching is enabled by default. This feature is controlled by the environment variable AMD_COMGR_CACHE, for details, see hip_rtc document.

• Programmatic support for scratch limits on the AMD Instinct MI300 and MI350 Series up GPU devices. More enumeration values were added in hipLimit_t as following:

  • hipExtLimitScratchMin, minimum allowed value in bytes for scratch limit on the device.

  • hipExtLimitScratchMax, maximum allowed value in bytes for scratch limit on the device.

  • hipExtLimitScratchCurrent, current scratch limit threshold in bytes on the device. Must be between the value hipExtLimitScratchMin and hipExtLimitScratchMax. Developers can now use the environment variable HSA_SCRATCH_SINGLE_LIMIT_ASYNC to change the default allocation size with expected scratch limit in ROCR runtime. On top of it, this value can also be overwritten programmatically in the application using the HIP API hipDeviceSetLimit(hipExtLimitScratchCurrent, value) to reset the scratch limit value.

• HIP runtime now enables peer-to-peer (P2P) memory copies to utilize all available SDMA engines, rather than being limited to a single engine. It also selects the best engine first to give optimal bandwidth.

• Improved launch latency for D2D copies and memset on MI300 Series.

• Introduced a threshold to handle the command submission patch to the GPU device(s), considering the synchronization with CPU, for performance improvement.

Resolved issues

• Error of “unable to find modules” in HIP clean up for code object module.

• The issue of incorrect return error hipErrorNoDevice, when a crash occurred on GPU device due to illegal operation or memory violation. HIP runtime now handles the failure on the GPU side properly and reports the precise error code based on the last error seen on the GPU.

• Failures in some framework test applications, HIP runtime fixed the bug in retrieving a memory object from the IPC memory handle.

• A crash in TensorFlow related application. HIP runtime now combines multiple definitions of callbackQueue into a single function, in case of an exception, passes its handler to the application and provides corresponding error code.

• Fixed issue of handling the kernel parameters for the graph launch.

• Failures in roc-obj tools. HIP runtime now makes DEPRECATED message in roc-obj tools as STDERR.

• Support of hipDeviceMallocContiguous flags in hipExtMallocWithFlags(). It now enables HSA_AMD_MEMORY_POOL_CONTIGUOUS_FLAG in the memory pool allocation on GPU device.

• Compilation failure, HIP runtime refactored the vector type alignment with __hip_vec_align_v.

• A numerical error/corruption found in Pytorch during graph replay. HIP runtime fixed the input sizes of kernel launch dimensions in hipExtModuleLaunchKernel for the execution of hipGraph capture.

• A crash during kernel execution in a customer application. The structure of kernel arguments was updated via adding the size of kernel arguments, and HIP runtime does validation before launch kernel with the structured arguments.

• Compilation error when using bfloat16 functions. HIP runtime removed the anonymous namespace from FP16 functions to resolve this issue.

Known issues

• hipLaunchHostFunc returns an error during stream capture. Any application using hipLaunchHostFunc might fail to capture graphs during stream capture, instead, it returns hipErrorStreamCaptureUnsupported.

• Compilation failure in kernels via hiprtc when using option std=c++11.

hipBLAS (3.0.0)

Added

• Added the hipblasSetWorkspace() API.

• Support for codecoverage tests.

Changed

• HIPBLAS_V2 API is the only available API using the hipComplex and hipDatatype types.

• Documentation updates.

• Verbose compilation for hipblas.cpp.

Removed

• hipblasDatatype_t type.

• hipComplex and hipDoubleComplex types.

• Support code for non-production gfx targets.

Resolved issues

• The build time CMake configuration for the dependency on hipBLAS-common is fixed.

• Compiler warnings for unhandled enumerations have been resolved.

hipBLASLt (1.0.0)

Added

• Stream-K GEMM support has been enabled for the FP32, FP16, BF16, FP8, and BF8 data types on the Instinct MI300A APU. To activate this feature, set the TENSILE_SOLUTION_SELECTION_METHOD environment variable to 2, for example, export TENSILE_SOLUTION_SELECTION_METHOD=2.

• Fused Swish/SiLU GEMM (enabled by HIPBLASLT_EPILOGUE_SWISH_EXT and HIPBLASLT_EPILOGUE_SWISH_BIAS_EXT).

• Support for HIPBLASLT_EPILOGUE_GELU_AUX_BIAS for gfx942.

• HIPBLASLT_TUNING_USER_MAX_WORKSPACE to constrain the maximum workspace size for user offline tuning.

• HIPBLASLT_ORDER_COL16_4R16 and HIPBLASLT_ORDER_COL16_4R8 to hipblasLtOrder_t to support FP16/BF16 swizzle GEMM and FP8 / BF8 swizzle GEMM respectively.

• TF32 emulation on gfx950.

• Support for FP6, BF6, and FP4 on gfx950.

• Support for block scaling by setting HIPBLASLT_MATMUL_DESC_A_SCALE_MODE and HIPBLASLT_MATMUL_DESC_B_SCALE_MODE to HIPBLASLT_MATMUL_MATRIX_SCALE_VEC32_UE8M0.

Changed

• The non-V2 APIs (GemmPreference, GemmProblemType, GemmEpilogue, GemmTuning, GemmInputs) in the cpp header are now the same as the V2 APIs (GemmPreferenceV2, GemmProblemTypeV2, GemmEpilogueV2, GemmTuningV2, GemmInputsV2). The original non-V2 APIs are removed.

Removed

• HIPBLASLT_MATMUL_DESC_A_SCALE_POINTER_VEC_EXT and HIPBLASLT_MATMUL_DESC_B_SCALE_POINTER_VEC_EXT are removed. Use the HIPBLASLT_MATMUL_DESC_A_SCALE_MODE and HIPBLASLT_MATMUL_DESC_B_SCALE_MODE attributes to set scalar (HIPBLASLT_MATMUL_MATRIX_SCALE_SCALAR_32F) or vector (HIPBLASLT_MATMUL_MATRIX_SCALE_OUTER_VEC_32F) attributes.

• The hipblasltExtAMaxWithScale API is removed.

Optimized

• Improved performance for 8-bit (FP8 / BF8 / I8) NN/NT cases by adding s_delay_alu to reduce stalls from dependent ALU operations on gfx12+.

• Improved performance for 8-bit and 16-bit (FP16 / BF16) TN cases by enabling software dependency checks (Expert Scheduling Mode) under certain restrictions to reduce redundant hardware dependency checks on gfx12+.

• Improved performance for 8-bit, 16-bit, and 32-bit batched GEMM with a better heuristic search algorithm for gfx942.

Upcoming changes

• V2 APIs (GemmPreferenceV2, GemmProblemTypeV2, GemmEpilogueV2, GemmTuningV2, GemmInputsV2) are deprecated.

hipCUB (4.0.0)

Added

• A new cmake option, BUILD_OFFLOAD_COMPRESS. When hipCUB is built with this option enabled, the --offload-compress switch is passed to the compiler. This causes the compiler to compress the binary that it generates. Compression can be useful in cases where you are compiling for a large number of targets, since this often results in a large binary. Without compression, in some cases, the generated binary may become so large that symbols are placed out of range, resulting in linking errors. The new BUILD_OFFLOAD_COMPRESS option is set to ON by default.

• Single pass operators in agent/single_pass_scan_operators.hpp which contains the following API:

  • BlockScanRunningPrefixOp

  • ScanTileStatus

  • ScanTileState

  • ReduceByKeyScanTileState

  • TilePrefixCallbackOp

• Support for gfx950.

• An overload of BlockScan::InclusiveScan that accepts an initial value to seed the scan.

• An overload of WarpScan::InclusiveScan that accepts an initial value to seed the scan.

• UnrolledThreadLoad, UnrolledCopy, and ThreadLoadVolatilePointer were added to align hipCUB with CUB.

• ThreadStoreVolatilePtr and the IterateThreadStore struct were added to align hipCUB with CUB.

• hipcub::InclusiveScanInit for CUB parity.

Changed

• The CUDA backend now requires CUB, Thrust, and libcu++ 2.7.0. If they aren’t found, they will be downloaded from the CUDA CCCL repository.

• Updated thread_load and thread_store to align hipCUB with CUB.

• All kernels now have hidden symbol visibility. All symbols now have inline namespaces that include the library version, (for example, hipcub::HIPCUB_300400_NS::symbol instead of hipcub::symbol), letting the user link multiple libraries built with different versions of hipCUB.

• Modified the broadcast kernel in warp scan benchmarks. The reported performance may be different to previous versions.

• The hipcub::detail::accumulator_t in rocPRIM backend has been changed to utilise rocprim::accumulator_t.

• The usage of rocprim::invoke_result_binary_op_t has been replaced with rocprim::accumulator_t.

Removed

• The AMD GPU targets gfx803 and gfx900 are no longer built by default. If you want to build for these architectures, specify them explicitly in the AMDGPU_TARGETS cmake option.

• Deprecated hipcub::AsmThreadLoad is removed, use hipcub::ThreadLoad instead.

• Deprecated hipcub::AsmThreadStore is removed, use hipcub::ThreadStore instead.

• Deprecated BlockAdjacentDifference::FlagHeads, BlockAdjacentDifference::FlagTails and BlockAdjacentDifference::FlagHeadsAndTails have been removed.

• This release removes support for custom builds on gfx940 and gfx941.

• Removed C++14 support. Only C++17 is supported.

Resolved issues

• Fixed an issue where Sort(keys, compare_op, valid_items, oob_default) in block_merge_sort.hpp would not fill in elements that are out of range (items after valid_items) with oob_default.

• Fixed an issue where ScatterToStripedFlagged in block_exhange.hpp was calling the wrong function.

Known issues

• BlockAdjacentDifference::FlagHeads, BlockAdjacentDifference::FlagTails and BlockAdjacentDifference::FlagHeadsAndTails have been removed from hipCUB’s CUB backend. They were already deprecated as of version 2.12.0 of hipCUB and they were removed from CCCL (CUB) as of CCCL’s 2.6.0 release.

• BlockScan::InclusiveScan for the CUDA backend does not compute the block aggregate correctly when passing an initial value parameter. This behavior is not matched by the AMD backend.

Upcoming changes

• BlockAdjacentDifference::FlagHeads, BlockAdjacentDifference::FlagTails and BlockAdjacentDifference::FlagHeadsAndTails were deprecated as of version 2.12.0 of hipCUB, and will be removed from the rocPRIM backend in a future release for the next ROCm major version (ROCm 7.0.0).

hipFFT (1.0.20)

Added

• Support for gfx950.

Removed

• Removed hipfft-rider legacy compatibility from clients.

• Removed support for the gfx940 and gfx941 targets from the client programs.

• Removed backward compatibility symlink for include directories.

hipfort (0.7.0)

Added

• Documentation clarifying how hipfort is built for the CUDA platform.

Changed

• Updated and reorganized documentation for clarity and consistency.

HIPIFY (20.0.0)

Added

• CUDA 12.9.1 support.

• cuDNN 9.11.0 support.

• cuTENSOR 2.2.0.0 support.

• LLVM 20.1.8 support.

Resolved issues

• hipDNN support is removed by default.

• #1859[hipify-perl] Fix warnings on unsupported Driver or Runtime APIs which were erroneously not reported.

• #1930 Revise JIT API.

• #1962 Support for cuda-samples helper headers.

• #2035 Removed const_cast<const char**>; in hiprtcCreateProgram and hiprtcCompileProgram.

hipRAND (3.0.0)

Added

• Support for gfx950.

Changed

• Deprecated the hipRAND Fortran API in favor of hipfort.

Removed

• Removed C++14 support, so only C++17 is supported.

hipSOLVER (3.0.0)

Added

• Added compatibility-only functions:

  • csrlsvqr

    • hipsolverSpCcsrlsvqr, hipsolverSpZcsrlsvqr

Resolved issues

• Corrected the value of lwork returned by various bufferSize functions to be consistent with CUDA cuSOLVER. The following functions now return lwork so that the workspace size (in bytes) is sizeof(T) * lwork, rather than lwork. To restore the original behavior, set the environment variable HIPSOLVER_BUFFERSIZE_RETURN_BYTES.

  • hipsolverXorgbr_bufferSize, hipsolverXorgqr_bufferSize, hipsolverXorgtr_bufferSize, hipsolverXormqr_bufferSize, hipsolverXormtr_bufferSize, hipsolverXgesvd_bufferSize, hipsolverXgesvdj_bufferSize, hipsolverXgesvdBatched_bufferSize, hipsolverXgesvdaStridedBatched_bufferSize, hipsolverXsyevd_bufferSize, hipsolverXsyevdx_bufferSize, hipsolverXsyevj_bufferSize, hipsolverXsyevjBatched_bufferSize, hipsolverXsygvd_bufferSize, hipsolverXsygvdx_bufferSize, hipsolverXsygvj_bufferSize, hipsolverXsytrd_bufferSize, hipsolverXsytrf_bufferSize.

hipSPARSE (4.0.1)

Added

• int8, int32, and float16 data types to hipDataTypeToHCCDataType so that sparse matrix descriptors can be used with them.

• Half float mixed precision to hipsparseAxpby where X and Y use float16 and the result and compute type use float.

• Half float mixed precision to hipsparseSpVV where X and Y use float16 and the result and compute type use float.

• Half float mixed precision to hipsparseSpMM where A and B use float16 and C and the compute type use float.

• Half float mixed precision to hipsparseSDDMM where A and B use float16 and C and the compute type use float.

• Half float uniform precision to the hipsparseScatter and hipsparseGather routines.

• Half float uniform precision to the hipsparseSDDMM routine.

• int8 precision to the hipsparseCsr2cscEx2 routine.

• The almalinux operating system name to correct the GFortran dependency.

Changed

• Switched to defaulting to C++17 when building hipSPARSE from source. Previously hipSPARSE was using C++14 by default.

Resolved issues

• Fixed a compilation issue related to using std::filesystem and C++14.

• Fixed an issue where the clients-common package was empty by moving the hipsparse_clientmatrices.cmake and hipsparse_mtx2csr files to it.

Known issues

• In hipsparseSpSM_solve(), the external buffer is passed as a parameter. This does not match the CUDA cuSPARSE API. This extra external buffer parameter will be removed in a future release. For now, this extra parameter can be ignored and nullptr passed in because it is unused internally.

hipSPARSELt (0.2.4)

Added

• Support for the LLVM target gfx950.

• Support for the following data type combinations for the LLVM target gfx950:

  • FP8(E4M3) inputs, F32 output, and F32 Matrix Core accumulation.

  • BF8(E5M2) inputs, F32 output, and F32 Matrix Core accumulation.

• Support for ROC-TX if HIPSPARSELT_ENABLE_MARKER=1 is set.

• Support for the cuSPARSELt v0.6.3 backend.

Removed

• Support for LLVM targets gfx940 and gfx941 has been removed.

• hipsparseLtDatatype_t has been removed.

Optimized

• Improved the library loading time.

• Provided more kernels for the FP16 data type.

hipTensor (2.0.0)

Added

• Element-wise binary operation support.

• Element-wise trinary operation support.

• Support for GPU target gfx950.

• Dynamic unary and binary operator support for element-wise operations and permutation.

• CMake check for f8 datatype availability.

• hiptensorDestroyOperationDescriptor to free all resources related to the provided descriptor.

• hiptensorOperationDescriptorSetAttribute to set attribute of a hiptensorOperationDescriptor_t object.

• hiptensorOperationDescriptorGetAttribute to retrieve an attribute of the provided hiptensorOperationDescriptor_t object.

• hiptensorCreatePlanPreference to allocate the hiptensorPlanPreference_t and enabled users to limit the applicable kernels for a given plan or operation.

• hiptensorDestroyPlanPreference to free all resources related to the provided preference.

• hiptensorPlanPreferenceSetAttribute to set attribute of a hiptensorPlanPreference_t object.

• hiptensorPlanGetAttribute to retrieve information about an already-created plan.

• hiptensorEstimateWorkspaceSize to determine the required workspace size for the given operation.

• hiptensorCreatePlan to allocate a hiptensorPlan_t object, select an appropriate kernel for a given operation and prepare a plan that encodes the execution.

• hiptensorDestroyPlan to free all resources related to the provided plan.

Changed

• Removed architecture support for gfx940 and gfx941.

• Generalized opaque buffer for any descriptor.

• Replaced hipDataType with hiptensorDataType_t for all supported types, for example, HIP_R_32F to HIPTENSOR_R_32F.

• Replaced hiptensorComputeType_t with hiptensorComputeDescriptor_t for all supported types.

• Replaced hiptensorInitTensorDescriptor with hiptensorCreateTensorDescriptor.

• Changed handle type and API usage from *handle to handle.

• Replaced hiptensorContractionDescriptor_t with hipTensorOperationDescriptor_t.

• Replaced hiptensorInitContractionDescriptor with hiptensorCreateContraction.

• Replaced hiptensorContractionFind_t with hiptensorPlanPreference_t.

• Replaced hiptensorInitContractionFind with hiptensorCreatePlanPreference.

• Replaced hiptensorContractionGetWorkspaceSize with hiptensorEstimateWorkspaceSize.

• Replaced HIPTENSOR_WORKSPACE_RECOMMENDED with HIPTENSOR_WORKSPACE_DEFAULT.

• Replaced hiptensorContractionPlan_t with hiptensorPlan_t.

• Replaced hiptensorInitContractionPlan with hiptensorCreatePlan.

• Replaced hiptensorContraction with hiptensorContract.

• Replaced hiptensorPermutation with hiptensorPermute.

• Replaced hiptensorReduction with hiptensorReduce.

• Replaced hiptensorElementwiseBinary with hiptensorElementwiseBinaryExecute.

• Replaced hiptensorElementwiseTrinary with hiptensorElementwiseTrinaryExecute.

• Removed function hiptensorReductionGetWorkspaceSize.

llvm-project (20.0.0)

Added

• The compiler -gsplit-dwarf option to enable the generation of separate debug information file at compile time. When used, separate debug information files are generated for host and for each offload architecture. For additional information, see DebugFission.

• llvm-flang, AMD’s next-generation Fortran compiler. It’s a re-implementation of the Fortran frontend that can be found at llvm/llvm-project/flang on GitHub.

• Comgr support for an in-memory virtual file system (VFS) for storing temporary files generated during intermediate compilation steps to improve performance in the device library link step.

• Compiler support of a new target-specific builtin __builtin_amdgcn_processor_is for late or deferred queries of the current target processor, and __builtin_amdgcn_is_invocable to determine the current target processor ability to invoke a particular builtin.

• HIPIFY support for CUDA 12.9.1 APIs. Added support for all new device and host APIs, including FP4, FP6, and FP128, and support for the corresponding ROCm HIP equivalents.

Changed

• Updated clang/llvm to AMD clang version 20.0.0 (equivalent to LLVM 20.0.0 with additional out-of-tree patches).

• HIPCC Perl scripts (hipcc.pl and hipconfig.pl) have been removed from this release.

Optimized

• Improved compiler memory load and store instructions.

Upcoming changes

• __AMDGCN_WAVEFRONT_SIZE__ macro and HIP’s warpSize variable as constexpr are deprecated and will be disabled in a future release. Users are encouraged to update their code if needed to ensure future compatibility. For more information, see AMDGCN_WAVEFRONT_SIZE deprecation.

• The roc-obj-ls and roc-obj-extract tools are deprecated. To extract all Clang offload bundles into separate code objects use llvm-objdump --offloading <file>. For more information, see Changes to ROCm Object Tooling.

MIGraphX (2.13.0)

Added

• Support for OCP FP8 on AMD Instinct MI350X GPUs.

• Support for PyTorch 2.7 via Torch-MIGraphX.

• Support for the Microsoft ONNX Contrib Operators (Self) Attention, RotaryEmbedding, QuickGelu, BiasAdd, BiasSplitGelu, SkipLayerNorm.

• Support for Sigmoid and AddN TensorFlow operators.

• GroupQuery Attention support for LLMs.

• Support for edge mode in the ONNX Pad operator.

• ONNX runtime Python driver.

• FLUX e2e example.

• C++ and Python APIs to save arguments to a graph as a msgpack file, and then read the file back.

• rocMLIR fusion for kv-cache attention.

• Introduced a check for file-write errors.

Changed

• quantize_bf16 for quantizing the model to BF16 has been made visible in the MIGraphX user API.

• Print additional kernel/module information in the event of compile failure.

• Use hipBLASLt instead of rocBLAS on newer GPUs.

• 1x1 convolutions are now rewritten to GEMMs.

• BF16::max is now represented by its encoding rather than its expected value.

• Direct warnings now go to cout rather cerr.

• FP8 uses hipBLASLt rather than rocBLAS.

• ONNX models are now topologically sorted when nodes are unordered.

• Improved layout of Graphviz output.

• Enhanced debugging for migraphx-driver: consumed environment variables are printed, timestamps and duration are added to the summary.

• Add a trim size flag to the verify option for migraphx-driver.

• Node names are printed to track parsing within the ONNX graph when using the MIGRAPHX_TRACE_ONNX_PARSER flag.

• Update accuracy checker to output test data with the --show-test-data flag.

• The MIGRAPHX_TRACE_BENCHMARKING option now allows the problem cache file to be updated after finding the best solution.

Removed

• ROCM_USE_FLOAT8 macro.

• The BF16 GEMM test was removed for Navi21, as it is unsupported by rocBLAS and hipBLASLt on that platform.

Optimized

• Use common average in compile_ops to reduce run-to-run variations when tuning.

• Improved the performance of the TopK operator.

• Conform to a single layout (NHWC or NCHW) during compilation rather than combining two.

• Slice Channels Conv Optimization (slice output fusion).

• Horizontal fusion optimization after pointwise operations.

• Reduced the number of literals used in GridSample linear sampler.

• Fuse multiple outputs for pointwise operations.

• Fuse reshapes on pointwise inputs for MLIR output fusion.

• MUL operation not folded into the GEMM when the GEMM is used more than once.

• Broadcast not fused after convolution or GEMM MLIR kernels.

• Avoid reduction fusion when operator data-types mismatch.

Resolved issues

• Compilation workaround ICE in clang 20 when using views::transform.

• Fix bug with reshape_lazy in MLIR.

• Quantizelinear fixed for Nearbyint operation.

• Check for empty strings in ONNX node inputs for operations like Resize.

• Parse Resize fix: only check keep_aspect_ratio_policy attribute for sizes input.

• Nonmaxsuppression: fixed issue where identical boxes/scores not ordered correctly.

• Fixed a bug where events were created on the wrong device in a multi-gpu scenario.

• Fixed out of order keys in value for comparisons and hashes when caching best kernels.

• Fixed Controlnet MUL types do not match error.

• Fixed check for scales if ROI input is present in Resize operation.

• Einsum: Fixed a crash on empty squeeze operations.

MIOpen (3.5.0)

Added

• [Conv] Misa kernels for gfx950.

• [Conv] Enabled Split-K support for CK backward data solvers (2D).

• [Conv] Enabled CK wrw solver on gfx950 for the BF16 data type.

• [BatchNorm] Enabled NHWC in OpenCL.

• Grouped convolution + activation fusion.

• Grouped convolution + bias + activation fusion.

• Composable Kernel (CK) can now be built inline as part of MIOpen.

Changed

• Changed to using the median value with outliers removed when deciding on the best solution to run.

• [Conv] Updated the igemm asm solver.

Optimized

• [BatchNorm] Optimized NHWC OpenCL kernels and improved heuristics.

• [RNN] Dynamic algorithm optimization.

• [Conv] Eliminated redundant clearing of output buffers.

• [RNN] Updated selection heuristics.

• Updated tuning for the AMD Instinct MI300 Series.

Resolved issues

• Fixed a segmentation fault when the user specified a smaller workspace than what was required.

• Fixed a layout calculation logic error that returned incorrect results and enabled less restrictive layout selection.

• Fixed memory access faults in misa kernels due to out-of-bounds memory usage.

• Fixed a performance drop on the gfx950 due to transpose kernel use.

• Fixed a memory access fault caused by not allocating enough workspace.

• Fixed a name typo that caused kernel mismatches and long startup times.

MIVisionX (3.3.0)

Added

• Support to enable/disable BatchPD code in VX_RPP extensions by checking the RPP_LEGACY_SUPPORT flag.

Changed

• VX_RPP extension: Version 3.1.0 release.

• Update the parameters and kernel API of Blur, Fog, Jitter, LensCorrection, Rain, Pixelate, Vignette and ResizeCrop wrt tensor kernels replacing the legacy BatchPD API calls in VX_RPP extensions.

Known issues

• Installation on RHEL and SLES requires the manual installation of the FFMPEG and OpenCV dev packages.

Upcoming changes

• Optimized audio augmentations support for VX_RPP.

RCCL (2.26.6)

Added

• Support for the extended fine-grained system memory pool.

• Support for gfx950.

• Support for unroll=1 in device-code generation to improve performance.

• Set a default of 112 channels for a single node with 8 * gfx950.

• Enabled LL128 protocol on the gfx950.

• The ability to choose the unroll factor at runtime using RCCL_UNROLL_FACTOR. This can be set at runtime to 1, 2, or 4. This change currently increases compilation and linking time because it triples the number of kernels generated.

• Added MSCCL support for AllGather multinode on the gfx942 and gfx950 (for instance, 16 and 32 GPUs). To enable this feature, set the environment variable RCCL_MSCCL_FORCE_ENABLE=1. The maximum message size for MSCCL AllGather usage is 12292 * sizeof(datatype) * nGPUs.

• Thread thresholds for LL/LL128 are selected in Tuning Models for the AMD Instinct MI300X. This impacts the number of channels used for AllGather and ReduceScatter. The channel tuning model is bypassed if NCCL_THREAD_THRESHOLDS, NCCL_MIN_NCHANNELS, or NCCL_MAX_NCHANNELS are set.

• Multi-node tuning for AllGather, AllReduce, and ReduceScatter that leverages LL/LL64/LL128 protocols to use nontemporal vector load/store for tunable message size ranges.

• LL/LL128 usage ranges for AllReduce, AllGather, and ReduceScatter are part of the tuning models, which enable architecture-specific tuning in conjunction with the existing Rome Models scheme in RCCL.

• Two new APIs are exposed as part of an initiative to separate RCCL code. These APIs are rcclGetAlgoInfo and rcclFuncMaxSendRecvCount. However, user-level invocation requires that RCCL be built with RCCL_EXPOSE_STATIC enabled.

Changed

• Compatibility with NCCL 2.23.4.

• Compatibility with NCCL 2.24.3.

• Compatibility with NCCL 2.25.1.

• Compatibility with NCCL 2.26.6.

Resolved issues

• Resolved an issue when using more than 64 channels when multiple collectives are used in the same ncclGroup() call.

• Fixed unit test failures in tests ending with the ManagedMem and ManagedMemGraph suffixes.

• Fixed a suboptimal algorithmic switching point for AllReduce on the AMD Instinct MI300X.

• Fixed the known issue “When splitting a communicator using ncclCommSplit in some GPU configurations, MSCCL initialization can cause a segmentation fault” with a design change to use comm instead of rank for mscclStatus. The global map for comm to mscclStatus is still not thread safe but should be explicitly handled by mutexes for read-write operations. This is tested for correctness, but there is a plan to use a thread-safe map data structure in an upcoming release.

rocAL (2.3.0)

Added

• Extended support to rocAL’s video decoder to use rocDecode hardware decoder.

• Setup - installs rocdecode dev packages for Ubuntu, RedHat, and SLES.

• Setup - installs turbojpeg dev package for Ubuntu and Redhat.

• rocAL’s image decoder has been extended to support the rocJPEG hardware decoder.

• Numpy reader support for reading npy files in rocAL.

• Test case for numpy reader in C++ and python tests.

Resolved issues

• TurboJPEG no longer needs to be installed manually. It is now installed by the package installer.

• Hardware decode no longer requires that ROCm be installed with the graphics usecase.

Known issues

• Package installation on SLES requires manually installing TurboJPEG.

• Package installation on RHEL and SLES requires manually installing the FFMPEG Dev package.

Upcoming changes

• rocJPEG support for JPEG decode.

rocALUTION (4.0.0)

Added

• Support for gfx950.

Changed

• Switch to defaulting to C++17 when building rocALUTION from source. Previously rocALUTION was using C++14 by default.

Optimized

• Improved the user documentation.

Resolved issues

• Fix for GPU hashing algorithm when not compiling with -O2/O3.

rocBLAS (5.0.0)

Added

• Support for gfx950.

• Internal API logging for gemm debugging using ROCBLAS_LAYER = 8.

• Support for the AOCL 5.0 gcc build as a client reference library.

• The use of PkgConfig for client reference library fallback detection.

Changed

• CMAKE_CXX_COMPILER is now passed on during compilation for a Tensile build.

• The default atomics mode is changed from allowed to not allowed.

Removed

• Support code for non-production gfx targets.

• rocblas_hgemm_kernel_name, rocblas_sgemm_kernel_name, and rocblas_dgemm_kernel_name API functions.

• The use of warpSize as a constexpr.

• The use of deprecated behavior of hipPeekLastError.

• rocblas_float8.h and rocblas_hip_f8_impl.h files.

• rocblas_gemm_ex3, rocblas_gemm_batched_ex3, and rocblas_gemm_strided_batched_ex3 API functions.

Optimized

• Optimized gemm by using gemv kernels when applicable.

• Optimized gemv for small m and n with a large batch count on gfx942.

• Improved the performance of Level 1 dot for all precisions and variants when N > 100000000 on gfx942.

• Improved the performance of Level 1 asum and nrm2 for all precisions and variants on gfx942.

• Improved the performance of Level 2 sger (single precision) on gfx942.

• Improved the performance of Level 3 dgmm for all precisions and variants on gfx942.

Resolved issues

• Fixed environment variable path-based logging to append multiple handle outputs to the same file.

• Support numerics when trsm is running with rocblas_status_perf_degraded.

• Fixed the build dependency installation of joblib on some operating systems.

• Return rocblas_status_internal_error when rocblas_[set,get]_ [matrix,vector] is called with a host pointer in place of a device pointer.

• Reduced the default verbosity level for internal GEMM backend information.

• Updated from the deprecated rocm-cmake to ROCmCMakeBuildTools.

• Corrected AlmaLinux GFortran package dependencies.

Upcoming changes

• Deprecated the use of negative indices to indicate the default solution is being used for gemm_ex with rocblas_gemm_algo_solution_index.

ROCdbgapi (0.77.3)

Added

• Support for the gfx950 architectures.

Removed

• Support for the gfx940 and gfx941 architectures.

rocDecode (1.0.0)

Added

• VP9 IVF container file parsing support in bitstream reader.

• CTest for VP9 decode on bitstream reader.

• HEVC/AVC/AV1/VP9 stream syntax error handling.

• HEVC stream bit depth change handling and DPB buffer size change handling through decoder reconfiguration.

• AVC stream DPB buffer size change handling through decoder reconfiguration.

• A new avcodec-based decoder built as a separate rocdecode-host library.

Changed

• rocDecode now uses the Cmake CMAKE_PREFIX_PATH directive.

• Changed asserts in query API calls in RocVideoDecoder utility class to error reports, to avoid hard stop during query in case error occurs and to let the caller decide actions.

• libdrm_amdgpu is now explicitly linked with rocdecode.

Removed

• GetStream() interface call from RocVideoDecoder utility class.

Optimized

• Decode session starts latency reduction.

• Bitstream type detection optimization in bitstream reader.

Resolved issues

• Fixed a bug in the videoDecodePicFiles picture files sample that can results in incorrect output frame count.

• Fixed a decoded frame output issue in video size change cases.

• Removed incorrect asserts of bitdepth_minus_8 in GetBitDepth() and num_chroma_planes in GetNumChromaPlanes() API calls in the RocVideoDecoder utility class.

rocFFT (1.0.34)

Added

• Support for gfx950.

Removed

• Removed rocfft-rider legacy compatibility from clients.

• Removed support for the gfx940 and gfx941 targets from the client programs.

• Removed backward compatibility symlink for include directories.

Optimized

• Removed unnecessary HIP event/stream allocation and synchronization during MPI transforms.

• Implemented single-precision 1D kernels for lengths:

  • 4704

  • 5488

  • 6144

  • 6561

  • 8192

• Implemented single-kernel plans for some large 1D problem sizes, on devices with at least 160KiB of LDS.

Resolved issues

• Fixed kernel faults on multi-device transforms that gather to a single device, when the input/output bricks are not contiguous.

ROCgdb (16.3)

Added

• Support for the gfx950 architectures.

Removed

• Support for the gfx940 and gfx941 architectures.

rocJPEG (1.1.0)

Added

• cmake config files.

• CTEST - New tests were introduced for JPEG batch decoding using various output formats, such as yuv_planar, y, rgb, and rgb_planar, both with and without region-of-interest (ROI).

Changed

• Readme - cleanup and updates to pre-reqs.

• The decode_params argument of the rocJpegDecodeBatched API is now an array of RocJpegDecodeParams structs representing the decode parameters for the batch of JPEG images.

• libdrm_amdgpu is now explicitly linked with rocjpeg.

Removed

• Dev Package - No longer installs pkg-config.

Resolved issues

• Fixed a bug that prevented copying the decoded image into the output buffer when the output buffer is larger than the input image.

• Resolved an issue with resizing the internal memory pool by utilizing the explicit constructor of the vector’s type during the resizing process.

• Addressed and resolved CMake configuration warnings.

ROCm Bandwidth Test (2.6.0)

Added

• Plugin architecture:

  • rocm_bandwidth_test is now the framework for individual plugins and features. The framework is available at: /opt/rocm/bin/

  • Individual plugins: The plugins (shared libraries) are available at: /opt/rocm/lib/rocm_bandwidth_test/plugins/

Note

Review the README file for details about the new options and outputs.

Changed

• The CLI and options/parameters have changed due to the new plugin architecture, where the plugin parameters are parsed by the plugin.

Removed

• The old CLI, parameters, and switches.

ROCm Compute Profiler (3.2.3)

Added

CDNA4 (AMD Instinct MI350/MI355) support

• Support for AMD Instinct MI350 Series GPUs with the addition of the following counters:

  • VALU co-issue (Two VALUs are issued instructions) efficiency

  • Stream Processor Instruction (SPI) Wave Occupancy

  • Scheduler-Pipe Wave Utilization

  • Scheduler FIFO Full Rate

  • CPC ADC Utilization

  • F6F4 data type metrics

  • Update formula for total FLOPs while taking into account F6F4 ops

  • LDS STORE, LDS LOAD, LDS ATOMIC instruction count metrics

  • LDS STORE, LDS LOAD, LDS ATOMIC bandwidth metrics

  • LDS FIFO full rate

  • Sequencer -> TA ADDR Stall rates

  • Sequencer -> TA CMD Stall rates

  • Sequencer -> TA DATA Stall rates

  • L1 latencies

  • L2 latencies

  • L2 to EA stalls

  • L2 to EA stalls per channel

• Roofline support for AMD Instinct MI350 Series GPUs.

Textual User Interface (TUI) (beta version)

• Text User Interface (TUI) support for analyze mode

  • A command line based user interface to support interactive single-run analysis.

  • To launch, use --tui option in analyze mode. For example, rocprof-compute analyze --tui.

PC Sampling (beta version)

• Stochastic (hardware-based) PC sampling has been enabled for AMD Instinct MI300X Series and later GPUs.

• Host-trap PC Sampling has been enabled for AMD Instinct MI200 Series and later GPUs.

• Support for sorting of PC sampling by type: offset or count.

• PC Sampling Support on CLI and TUI analysis.

Roofline

• Support for Roofline plot on CLI (single run) analysis.

• FP4 and FP6 data types have been added for roofline profiling on AMD Instinct MI350 Series.

rocprofv3 support

• rocprofv3 is supported as the default backend for profiling.

• Support to obtain performance information for all channels for TCC counters.

• Support for profiling on AMD Instinct MI 100 using rocprofv3.

• Deprecation warning for rocprofv3 interface in favor of the ROCprofiler-SDK interface, which directly accesses rocprofv3 C++ tool.

Others

• Docker files to package the application and dependencies into a single portable and executable standalone binary file.

• Analysis report based filtering

  • -b option in profile mode now also accepts metric id(s) for analysis report based filtering.

  • -b option in profile mode also accepts hardware IP block for filtering; however, this filter support will be deprecated soon.

  • --list-metrics option added in profile mode to list possible metric id(s), similar to analyze mode.

• Support MEM chart on CLI (single run).

• --specs-correction option to provide missing system specifications for analysis.

Changed

• Changed the default rocprof version to rocprofv3. This is used when environment variable ROCPROF is not set.

• Changed normal_unit default to per_kernel.

• Decreased profiling time by not collecting unused counters in post-analysis.

• Updated Dash to >=3.0.0 (for web UI).

• Changed the condition when Roofline PDFs are generated during general profiling and --roof-only profiling (skip only when --no-roof option is present).

• Updated Roofline binaries:

  • Rebuild using latest ROCm stack.

  • Minimum OS distribution support minimum for roofline feature is now Ubuntu 22.04, RHEL 8, and SLES15 SP6.

Removed

• Roofline support for Ubuntu 20.04 and SLES below 15.6.

• Removed support for AMD Instinct MI50 and MI60.

Optimized

• ROCm Compute Profiler CLI has been improved to better display the GPU architecture analytics.

Resolved issues

• Fixed kernel name and kernel dispatch filtering when using rocprofv3.

• Fixed an issue of TCC channel counters collection in rocprofv3.

• Fixed peak FLOPS of F8, I8, F16, and BF16 on AMD Instinct MI300.

• Fixed not detecting memory clock issue when using amd-smi.

• Fixed standalone GUI crashing.

• Fixed L2 read/write/atomic bandwidths on AMD Instinct MI350 Series.

Known issues

• On AMD Instinct MI100, accumulation counters are not collected, resulting in the following metrics failing to show up in the analysis: Instruction Fetch Latency, Wavefront Occupancy, LDS Latency.

  • As a workaround, use the environment variable ROCPROF=rocprof, to use rocprof v1 for profiling on AMD Instinct MI100.

• GPU id filtering is not supported when using rocprofv3.

• Analysis of previously collected workload data will not work due to sysinfo.csv schema change.

  • As a workaround, re-run the profiling operation for the workload and interrupt the process after 10 seconds. Followed by copying the sysinfo.csv file from the new data folder to the old one. This assumes your system specification hasn’t changed since the creation of the previous workload data.

• Analysis of new workloads might require providing shader/memory clock speed using --specs-correction operation if amd-smi or rocminfo does not provide clock speeds.

• Memory chart on ROCm Compute Profiler CLI might look corrupted if the CLI width is too narrow.

• Roofline feature is currently not functional on Azure Linux 3.0 and Debian 12.

Upcoming changes

• rocprof v1/v2/v3 interfaces will be removed in favor of the ROCprofiler-SDK interface, which directly accesses rocprofv3 C++ tool. Using rocprof v1/v2/v3 interfaces will trigger a deprecation warning.

  • To use ROCprofiler-SDK interface, set environment variable ROCPROF=rocprofiler-sdk and optionally provide profile mode option --rocprofiler-sdk-library-path /path/to/librocprofiler-sdk.so. Add --rocprofiler-sdk-library-path runtime option to choose the path to ROCprofiler-SDK library to be used.

• Hardware IP block based filtering using -b option in profile mode will be removed in favor of analysis report block based filtering using -b option in profile mode.

• MongoDB database support will be removed, and a deprecation warning has been added to the application interface.

• Usage of rocm-smi is deprecated in favor of amd-smi, and a deprecation warning has been added to the application interface.

ROCm Data Center Tool (1.1.0)

Added

• More profiling and monitoring metrics, especially for AMD Instinct MI300 and newer GPUs.

• Advanced logging and debugging options, including new log levels and troubleshooting guidance.

Changed

• Completed migration from legacy ROCProfiler to ROCprofiler-SDK.

• Reorganized the configuration files internally and improved README/installation instructions.

• Updated metrics and monitoring support for the latest AMD data center GPUs.

Optimized

• Integration with ROCprofiler-SDK for performance metrics collection.

• Standalone and embedded operating modes, including streamlined authentication and configuration options.

• Support and documentation for diagnostic commands and GPU group management.

• RVS test integration and reporting.

ROCm SMI (7.8.0)

Added

• Support for GPU metrics 1.8.

  • Added new fields for rsmi_gpu_metrics_t including:

    • Adding the following metrics to allow new calculations for violation status:

    • Per XCP metrics gfx_below_host_limit_ppt_acc[XCP][MAX_XCC] - GFX Clock Host limit Package Power Tracking violation counts.

    • Per XCP metrics gfx_below_host_limit_thm_acc[XCP][MAX_XCC] - GFX Clock Host limit Thermal (TVIOL) violation counts.

    • Per XCP metrics gfx_low_utilization_acc[XCP][MAX_XCC] - violation counts for how did low utilization caused the GPU to be below application clocks.

    • Per XCP metrics gfx_below_host_limit_total_acc[XCP][MAX_XCC]- violation counts for how long GPU was held below application clocks any limiter (see above new violation metrics).

  • Increasing available JPEG engines to 40.
    Current ASICs may not support all 40. These will be indicated as UINT16_MAX or N/A in CLI.
    

Removed

• Removed backwards compatibility for rsmi_dev_gpu_metrics_info_get()’s jpeg_activity and vcn_activity fields. Alternatively use xcp_stats.jpeg_busy and xcp_stats.vcn_busy.

  • Backwards compatibility is removed for jpeg_activity and vcn_activity fields, if the jpeg_busy or vcn_busy field is available.

    • Providing both vcn_activity/jpeg_activity and XCP (partition) stats vcn_busy/jpeg_busy caused confusion for users about which field to use. By removing backward compatibility, it is easier to identify the relevant field.

    • The jpeg_busy field increased in size (for supported ASICs), making backward compatibility unable to fully copy the structure into jpeg_activity.

Note

See the full ROCm SMI changelog for details, examples, and in-depth descriptions.

ROCm Systems Profiler (1.1.0)

Added

• Profiling and metric collection capabilities for VCN engine activity, JPEG engine activity, and API tracing for rocDecode, rocJPEG, and VA-APIs.

• How-to document for VCN and JPEG activity sampling and tracing.

• Support for tracing Fortran applications.

• Support for tracing MPI API in Fortran.

Changed

• Replaced ROCm SMI backend with AMD SMI backend for collecting GPU metrics.

• ROCprofiler-SDK is now used to trace RCCL API and collect communication counters.

  • Use the setting ROCPROFSYS_USE_RCCLP = ON to enable profiling and tracing of RCCL application data.

• Updated the Dyninst submodule to v13.0.

• Set the default value of ROCPROFSYS_SAMPLING_CPUS to none.

Resolved issues

• Fixed GPU metric collection settings with ROCPROFSYS_AMD_SMI_METRICS.

• Fixed a build issue with CMake 4.

• Fixed incorrect kernel names shown for kernel dispatch tracks in Perfetto.

• Fixed formatting of some output logs.

ROCm Validation Suite (1.2.0)

Added

• Support for AMD Instinct MI350X and MI355X GPUs.

• Introduced rotating buffer mechanism for GEMM operations.

• Support for read and write tests in Babel.

• Support for AMD Radeon RX9070 and RX9070GRE graphics cards.

Changed

• Migrated SMI API usage from rocm-smi to amd-smi.

• Updated FP8 GEMM operations to use hipBLASLt instead of rocBLAS.

rocPRIM (4.0.0)

Added

• Support for gfx950.

• rocprim::accumulator_t to ensure parity with CCCL.

• Test for rocprim::accumulator_t.

• rocprim::invoke_result_r to ensure parity with CCCL.

• Function is_build_in into rocprim::traits::get.

• Virtual shared memory as a fallback option in rocprim::device_merge when it exceeds shared memory capacity, similar to rocprim::device_select, rocprim::device_partition, and rocprim::device_merge_sort, which already include this feature.

• Initial value support to device level inclusive scans.

• New optimization to the backend for device_transform when the input and output are pointers.

• LoadType to transform_config, which is used for the device_transform when the input and output are pointers.

• rocprim:device_transform for n-ary transform operations API with as input n number of iterators inside a rocprim::tuple.

• rocprim::key_value_pair::operator==.

• The rocprim::unrolled_copy thread function to copy multiple items inside a thread.

• The rocprim::unrolled_thread_load function to load multiple items inside a thread using rocprim::thread_load.

• rocprim::int128_t and rocprim::uint128_t to benchmarks for improved performance evaluation on 128-bit integers.

• rocprim::int128_t to the supported autotuning types to improve performance for 128-bit integers.

• The rocprim::merge_inplace function for merging in-place.

• Initial value support for warp- and block-level inclusive scan.

• Support for building tests with device-side random data generation, making them finish faster. This requires rocRAND, and is enabled with the WITH_ROCRAND=ON build flag.

• Tests and documentation to lookback_scan_state. It is still in the detail namespace.

Changed

• Changed the parameters long_radix_bits and LongRadixBits from segmented_radix_sort to radix_bits and RadixBits, respectively.

• Marked the initialisation constructor of rocprim::reverse_iterator<Iter> explicit, use rocprim::make_reverse_iterator.

• Merged radix_key_codec into type_traits system.

• Renamed type_traits_interface.hpp to type_traits.hpp, rename the original type_traits.hpp to type_traits_functions.hpp.

• The default scan accumulator types for device-level scan algorithms have changed. This is a breaking change. The previous default accumulator types could lead to situations in which unexpected overflow occurred, such as when the input or initial type was smaller than the output type. This is a complete list of affected functions and how their default accumulator types are changing:

  • rocprim::inclusive_scan

    • Previous default: class AccType = typename std::iterator_traits<InputIterator>::value_type>

    • Current default: class AccType = rocprim::accumulator_t<BinaryFunction, typename std::iterator_traits<InputIterator>::value_type>

  • rocprim::deterministic_inclusive_scan

    • Previous default: class AccType = typename std::iterator_traits<InputIterator>::value_type>

    • Current default: class AccType = rocprim::accumulator_t<BinaryFunction, typename std::iterator_traits<InputIterator>::value_type>

  • rocprim::exclusive_scan

    • Previous default: class AccType = detail::input_type_t<InitValueType>>

    • Current default: class AccType = rocprim::accumulator_t<BinaryFunction, rocprim::detail::input_type_t<InitValueType>>

  • rocprim::deterministic_exclusive_scan

    • Previous default: class AccType = detail::input_type_t<InitValueType>>

    • Current default: class AccType = rocprim::accumulator_t<BinaryFunction, rocprim::detail::input_type_t<InitValueType>>

• Undeprecated internal detail::raw_storage.

• A new version of rocprim::thread_load and rocprim::thread_store replaces the deprecated rocprim::thread_load and rocprim::thread_store functions. The versions avoid inline assembly where possible, and don’t hinder the optimizer as much as a result.

• Renamed rocprim::load_cs to rocprim::load_nontemporal and rocprim::store_cs to rocprim::store_nontemporal to express the intent of these load and store methods better.

• All kernels now have hidden symbol visibility. All symbols now have inline namespaces that include the library version, for example, rocprim::ROCPRIM_300400_NS::symbol instead of rocPRIM::symbol, letting the user link multiple libraries built with different versions of rocPRIM.

Removed

• rocprim::detail::float_bit_mask and relative tests, use rocprim::traits::float_bit_mask instead.

• rocprim::traits::is_fundamental, use rocprim::traits::get<T>::is_fundamental() directly.

• The deprecated parameters short_radix_bits and ShortRadixBits from the segmented_radix_sort config. They were unused, it is only an API change.

• The deprecated operator<< from the iterators.

• The deprecated TwiddleIn and TwiddleOut. Use radix_key_codec instead.

• The deprecated flags API of block_adjacent_difference. Use subtract_left() or block_discontinuity::flag_heads() instead.

• The deprecated to_exclusive functions in the warp scans.

• The rocprim::load_cs from the cache_load_modifier enum. Use rocprim::load_nontemporal instead.

• The rocprim::store_cs from the cache_store_modifier enum. Use rocprim::store_nontemporal instead.

• The deprecated header file rocprim/detail/match_result_type.hpp. Include rocprim/type_traits.hpp instead. This header included:

  • rocprim::detail::invoke_result. Use rocprim::invoke_result instead.

  • rocprim::detail::invoke_result_binary_op. Use rocprim::invoke_result_binary_op instead.

  • rocprim::detail::match_result_type. Use rocprim::invoke_result_binary_op_t instead.

• The deprecated rocprim::detail::radix_key_codec function. Use rocprim::radix_key_codec instead.

• Removed rocprim/detail/radix_sort.hpp, functionality can now be found in rocprim/thread/radix_key_codec.hpp.

• Removed C++14 support. Only C++17 is supported.

• Due to the removal of __AMDGCN_WAVEFRONT_SIZE in the compiler, the following deprecated warp size-related symbols have been removed:

  • rocprim::device_warp_size()

    • For compile-time constants, this is replaced with rocprim::arch::wavefront::min_size() and rocprim::arch::wavefront::max_size(). Use this when allocating global or shared memory.

    • For run-time constants, this is replaced with rocprim::arch::wavefront::size().

  • rocprim::warp_size()

    • Use rocprim::host_warp_size(), rocprim::arch::wavefront::min_size() or rocprim::arch::wavefront::max_size() instead.

  • ROCPRIM_WAVEFRONT_SIZE

    • Use rocprim::arch::wavefront::min_size() or rocprim::arch::wavefront::max_size() instead.

  • __AMDGCN_WAVEFRONT_SIZE

    • This was a fallback define for the compiler’s removed symbol, having the same name.

• This release removes support for custom builds on gfx940 and gfx941.

Optimized

• Improved performance of rocprim::device_select and rocprim::device_partition when using multiple streams on the AMD Instinct MI300 Series.

Resolved issues

• Fixed an issue where device_batch_memcpy reported benchmarking throughput being 2x lower than it was in reality.

• Fixed an issue where device_segmented_reduce reported autotuning throughput being 5x lower than it was in reality.

• Fixed device radix sort not returning the correct required temporary storage when a double buffer contains nullptr.

• Fixed constness of equality operators (== and !=) in rocprim::key_value_pair.

• Fixed an issue for the comparison operators in arg_index_iterator and texture_cache_iterator, where < and > comparators were swapped.

• Fixed an issue for the rocprim::thread_reduce not working correctly with a prefix value.

Known issues

• When using rocprim::deterministic_inclusive_scan_by_key and rocprim::deterministic_exclusive_scan_by_key the intermediate values can change order on Navi3x. However, if a commutative scan operator is used then the final scan value (output array) will still always be consistent between runs.

Upcoming changes

• rocprim::invoke_result_binary_op and rocprim::invoke_result_binary_op_t are deprecated. Use rocprim::accumulator_t instead.

ROCprofiler-SDK (1.0.0)

Added

• Support for rocJPEG API Tracing.

• Support for AMD Instinct MI350X and MI355X GPUs.

• rocprofiler_create_counter to facilitate adding custom derived counters at runtime.

• Support in rocprofv3 for iteration based counter multiplexing.

• Perfetto support for counter collection.

• Support for negating rocprofv3 tracing options when using aggregate options such as --sys-trace --hsa-trace=no.

• --agent-index option in rocprofv3 to specify the agent naming convention in the output:

  • absolute == node_id

  • relative == logical_node_id

  • type-relative == logical_node_type_id

• MI300 and MI350 stochastic (hardware-based) PC sampling support in ROCProfiler-SDK and rocprofv3.

• Python bindings for rocprofiler-sdk-roctx.

• SQLite3 output support for rocprofv3 using --output-format rocpd.

• rocprofiler-sdk-rocpd package:

  • Public API in include/rocprofiler-sdk-rocpd/rocpd.h.

  • Library implementation in librocprofiler-sdk-rocpd.so.

  • Support for find_package(rocprofiler-sdk-rocpd).

  • rocprofiler-sdk-rocpd DEB and RPM packages.

• --version option in rocprofv3.

• rocpd Python package.

• Thread trace as experimental API.

• ROCprof Trace Decoder as experimental API:

  • Requires ROCprof Trace Decoder plugin.

• Thread trace option in the rocprofv3 tool under the --att parameters:

  • See using thread trace with rocprofv3

  • Requires ROCprof Trace Decoder plugin.

• rocpd output format documentation:

  • Requires ROCprof Trace Decoder plugin.

• Perfetto support for scratch memory.

• Support in the rocprofv3 avail tool for command-line arguments.

• Documentation for rocprofv3 advanced options.

• AQLprofile is now available as open source.

Changed

• SDK to NOT to create a background thread when every tool returns a nullptr from rocprofiler_configure.

• vaddr-to-file-offset mapping in disassembly.hpp to use the dedicated comgr API.

• rocprofiler_uuid_t ABI to hold 128 bit value.

• rocprofv3 shorthand argument for --collection-period to -P (upper-case) while -p (lower-case) is reserved for later use.

• Default output format for rocprofv3 to rocpd (SQLite3 database).

• rocprofv3 avail tool to be renamed from rocprofv3_avail to rocprofv3-avail tool.

• rocprofv3 tool to facilitate thread trace and PC sampling on the same agent.

Removed

• Support for compilation of gfx940 and gfx941 targets.

Resolved issues

• Fixed missing callbacks around internal thread creation within counter collection service.

• Fixed potential data race in the ROCprofiler-SDK double buffering scheme.

• Fixed usage of std::regex in the core ROCprofiler-SDK library that caused segfaults or exceptions when used under dual ABI.

• Fixed Perfetto counter collection by introducing accumulation per dispatch.

• Fixed code object disassembly for missing function inlining information.

• Fixed queue preemption error and HSA_STATUS_ERROR_INVALID_PACKET_FORMAT error for stochastic PC-sampling in MI300X, leading to stabler runs.

• Fixed the system hang issue for host-trap PC-sampling on AMD Instinct MI300X.

• Fixed rocpd counter collection issue when counter collection alone is enabled. rocpd_kernel_dispatch table is updated to be populated by counters data instead of kernel_dispatch data.

• Fixed rocprofiler_*_id_t structs for inconsistency related to a “null” handle:

  • The correct definition for a null handle is .handle = 0 while some definitions previously used UINT64_MAX.

• Fixed kernel trace csv output generated by rocpd.

rocPyDecode (0.6.0)

Added

• rocpyjpegdecode package.

• src/rocjpeg source new subfolder.

• samples/rocjpeg new subfolder.

Changed

• Minimum version for rocdecode and rocjpeg updated to V1.0.0.

rocRAND (4.0.0)

Added

• Support for gfx950.

• Additional unit tests for test_log_normal_distribution.cpp, test_normal_distribution.cpp, test_rocrand_mtgp32_prng.cpp, test_rocrand_scrambled_sobol32_qrng.cpp, test_rocrand_scrambled_sobol64_qrng.cpp, test_rocrand_sobol32_qrng.cpp, test_rocrand_sobol64_qrng.cpp, test_rocrand_threefry2x32_20_prng.cpp, test_rocrand_threefry2x64_20_prng.cpp, test_rocrand_threefry4x32_20_prng.cpp, test_rocrand_threefry4x64_20_prng.cpp, and test_uniform_distribution.cpp.

• New unit tests for include/rocrand/rocrand_discrete.h in test_rocrand_discrete.cpp, include/rocrand/rocrand_mrg31k3p.h in test_rocrand_mrg31k3p_prng.cpp, include/rocrand/rocrand_mrg32k3a.h in test_rocrand_mrg32k3a_prng.cpp, and include/rocrand/rocrand_poisson.h in test_rocrand_poisson.cpp.

Changed

• Changed the return type for rocrand_generate_poisson for the SOBOL64 and SCRAMBLED_SOBOL64 engines.

• Changed the unnecessarily large 64-bit data type for constants used for skipping in MRG32K3A to the 32-bit data type.

• Updated several gfx942 auto tuning parameters.

• Modified error handling and expanded the error information for the case of double-deallocation of the (scrambled) sobol32 and sobol64 constants and direction vectors.

Removed

• Removed inline assembly and the ENABLE_INLINE_ASM CMake option. Inline assembly was used to optimize multiplication in the Mrg32k3a and Philox 4x32-10 generators. It is no longer needed because the current HIP compiler is able to produce code with the same or better performance.

• Removed instances of the deprecated clang definition __AMDGCN_WAVEFRONT_SIZE.

• Removed C++14 support. Beginning with this release, only C++17 is supported.

• Directly accessing the (scrambled) sobol32 and sobol64 constants and direction vectors is no longer supported. For:

  • h_scrambled_sobol32_constants, use rocrand_get_scramble_constants32 instead.

  • h_scrambled_sobol64_constants, use rocrand_get_scramble_constants64 instead.

  • rocrand_h_sobol32_direction_vectors, use rocrand_get_direction_vectors32 instead.

  • rocrand_h_sobol64_direction_vectors, use rocrand_get_direction_vectors64 instead.

  • rocrand_h_scrambled_sobol32_direction_vectors, use rocrand_get_direction_vectors32 instead.

  • rocrand_h_scrambled_sobol64_direction_vectors, use rocrand_get_direction_vectors64 instead.

Resolved issues

• Fixed an issue where mt19937.hpp would cause kernel errors during auto tuning.

Upcoming changes

• Deprecated the rocRAND Fortran API in favor of hipfort.

ROCr Debug Agent (2.1.0)

Added

• The -e and --precise-alu-exceptions flags to enable precise ALU exceptions reporting on supported configurations.

ROCr Runtime (1.18.0)

Added

• New API hsa_amd_memory_get_preferred_copy_engine to get preferred copy engine that can be used to when calling hsa_amd_memory_async_copy_on_engine.

• New API hsa_amd_portable_export_dmabuf_v2 extension of existing hsa_amd_portable_export_dmabuf API to support new flags parameter. This allows specifying the new HSA_AMD_DMABUF_MAPPING_TYPE_PCIE flag when exporting dma-bufs.

• New flag HSA_AMD_VMEM_ADDRESS_NO_REGISTER adds support for new HSA_AMD_VMEM_ADDRESS_NO_REGISTER when calling hsa_amd_vmem_address_reserve API. This allows virtual address range reservations for SVM allocations to be tracked when running in ASAN mode.

• New sub query HSA_AMD_AGENT_INFO_CLOCK_COUNTERS returns a snapshot of the underlying driver’s clock counters that can be used for profiling.

rocSHMEM (3.0.0)

Added

• Reverse Offload conduit.

• New APIs: rocshmem_ctx_barrier, rocshmem_ctx_barrier_wave, rocshmem_ctx_barrier_wg, rocshmem_barrier_all, rocshmem_barrier_all_wave, rocshmem_barrier_all_wg, rocshmem_ctx_sync, rocshmem_ctx_sync_wave, rocshmem_ctx_sync_wg, rocshmem_sync_all, rocshmem_sync_all_wave, rocshmem_sync_all_wg, rocshmem_init_attr, rocshmem_get_uniqueid, and rocshmem_set_attr_uniqueid_args.

• dlmalloc based allocator.

• XNACK support.

• Support for initialization with MPI communicators other than MPI_COMM_WORLD.

Changed

• Changed collective APIs to use _wg suffix rather than _wg_ infix.

Resolved issues

• Resolved segfault in rocshmem_wg_ctx_create, now provides nullptr if ctx cannot be created.

rocSOLVER (3.30.0)

Added

• Hybrid computation support for existing routines: STEQR

Optimized

• Improved the performance of BDSQR and downstream functions, such as GESVD.

• Improved the performance of STEQR and downstream functions, such as SYEV/HEEV.

• Improved the performance of LARFT and downstream functions, such as GEQR2 and GEQRF.

Resolved issues

• Fixed corner cases that can produce NaNs in SYEVD for valid input matrices.

rocSPARSE (4.0.2)

Added

• The SpGEAM generic routine for computing sparse matrix addition in CSR format.

• The v2_SpMV generic routine for computing sparse matrix vector multiplication. As opposed to the deprecated rocsparse_spmv routine, this routine does not use a fallback algorithm if a non-implemented configuration is encountered and will return an error in such a case. For the deprecated rocsparse_spmv routine, the user can enable warning messages in situations where a fallback algorithm is used by either calling the rocsparse_enable_debug routine upfront or exporting the variable ROCSPARSE_DEBUG (with the shell command export ROCSPARSE_DEBUG=1).

• Half float mixed precision to rocsparse_axpby where X and Y use float16 and the result and compute type use float.

• Half float mixed precision to rocsparse_spvv where X and Y use float16 and the result and compute type use float.

• Half float mixed precision to rocsparse_spmv where A and X use float16 and Y and the compute type use float.

• Half float mixed precision to rocsparse_spmm where A and B use float16 and C and the compute type use float.

• Half float mixed precision to rocsparse_sddmm where A and B use float16 and C and the compute type use float.

• Half float uniform precision to the rocsparse_scatter and rocsparse_gather routines.

• Half float uniform precision to the rocsparse_sddmm routine.

• The rocsparse_spmv_alg_csr_rowsplit algorithm.

• Support for gfx950.

• ROC-TX instrumentation support in rocSPARSE (not available on Windows or in the static library version on Linux).

• The almalinux operating system name to correct the GFortran dependency.

Changed

• Switch to defaulting to C++17 when building rocSPARSE from source. Previously rocSPARSE was using C++14 by default.

Removed

• The deprecated rocsparse_spmv_ex routine.

• The deprecated rocsparse_sbsrmv_ex, rocsparse_dbsrmv_ex, rocsparse_cbsrmv_ex, and rocsparse_zbsrmv_ex routines.

• The deprecated rocsparse_sbsrmv_ex_analysis, rocsparse_dbsrmv_ex_analysis, rocsparse_cbsrmv_ex_analysis, and rocsparse_zbsrmv_ex_analysis routines.

Optimized

• Reduced the number of template instantiations in the library to further reduce the shared library binary size and improve compile times.

• Allow SpGEMM routines to use more shared memory when available. This can speed up performance for matrices with a large number of intermediate products.

• Use of the rocsparse_spmv_alg_csr_adaptive or rocsparse_spmv_alg_csr_default algorithms in rocsparse_spmv to perform transposed sparse matrix multiplication (C=alpha*A^T*x+beta*y) resulted in unnecessary analysis on A and needless slowdown during the analysis phase. This has been improved by skipping the analysis when performing the transposed sparse matrix multiplication.

• Improved the user documentation.

Resolved issues

• Fixed an issue in the public headers where extern "C" was not wrapped by #ifdef __cplusplus, which caused failures when building C programs with rocSPARSE.

• Fixed a memory access fault in the rocsparse_Xbsrilu0 routines.

• Fixed failures that could occur in rocsparse_Xbsrsm_solve or rocsparse_spsm with BSR format when using host pointer mode.

• Fixed ASAN compilation failures.

• Fixed a failure that occurred when using const descriptor rocsparse_create_const_csr_descr with the generic routine rocsparse_sparse_to_sparse. The issue was not observed when using non-const descriptor rocsparse_create_csr_descr with rocsparse_sparse_to_sparse.

• Fixed a memory leak in the rocSPARSE handle.

Upcoming changes

• Deprecated the rocsparse_spmv routine. Use the rocsparse_v2_spmv routine instead.

• Deprecated the rocsparse_spmv_alg_csr_stream algorithm. Use the rocsparse_spmv_alg_csr_rowsplit algorithm instead.

• Deprecated the rocsparse_itilu0_alg_sync_split_fusion algorithm. Use one of rocsparse_itilu0_alg_async_inplace, rocsparse_itilu0_alg_async_split, or rocsparse_itilu0_alg_sync_split instead.

rocThrust (4.0.0)

Added

• Additional unit tests for: binary_search, complex, c99math, catrig, ccosh, cexp, clog, csin, csqrt, and ctan.

• test_param_fixtures.hpp to store all the parameters for typed test suites.

• test_real_assertions.hpp to handle unit test assertions for real numbers.

• test_imag_assertions.hpp to handle unit test assertions for imaginary numbers.

• clang++ is now used to compile google benchmarks on Windows.

• Support for gfx950.

• Merged changes from upstream CCCL/thrust 2.6.0.

Changed

• Updated the required version of Google Benchmark from 1.8.0 to 1.9.0.

• Renamed cpp14_required.h to cpp_version_check.h.

• Refactored test_header.hpp into test_param_fixtures.hpp, test_real_assertions.hpp, test_imag_assertions.hpp, and test_utils.hpp. This is done to prevent unit tests from having access to modules that they’re not testing. This will improve the accuracy of code coverage reports.

Removed

• device_malloc_allocator.h has been removed. This header file was unused and should not impact users.

• Removed C++14 support. Only C++17 is now supported.

• test_header.hpp has been removed. The HIP_CHECK function, as well as the test and inter_run_bwr namespaces, have been moved to test_utils.hpp.

• test_assertions.hpp has been split into test_real_assertions.hpp and test_imag_assertions.hpp.

Resolved issues

• Fixed an issue with internal calls to unqualified distance() which would be ambiguous due to the visible implementation through ADL.

Known issues

• The order of the values being compared by thrust::exclusive_scan_by_key and thrust::inclusive_scan_by_key can change between runs when integers are being compared. This can cause incorrect output when a non-commutative operator such as division is being used.

Upcoming changes

• thrust::device_malloc_allocator is deprecated as of this version. It will be removed in an upcoming version.

rocWMMA (2.0.0)

Added

• Internal register layout transforms to support interleaved MMA layouts.

• Support for the gfx950 target.

• Mixed input BF8/FP8 types for MMA support.

• Fragment scheduler API objects to embed thread block cooperation properties in fragments.

Changed

• Augmented load/store/MMA internals with static loop unrolling.

• Updated linkage of rocwmma::synchronize_workgroup to inline.

• rocWMMA mma_sync API now supports wave tile fragment sizes.

• rocWMMA cooperative fragments are now expressed with fragment scheduler template arguments.

• rocWMMA cooperative fragments now use the same base API as non-cooperative fragments.

• rocWMMA cooperative fragments register usage footprint has been reduced.

• rocWMMA fragments now support partial tile sizes with padding.

Removed

• Support for the gfx940 and gfx941 targets.

• The rocWMMA cooperative API.

• Wave count template parameters from transforms APIs.

Optimized

• Added internal flow control barriers to improve assembly code generation and overall performance.

• Enabled interleaved layouts by default in MMA to improve overall performance.

Resolved issues

• Fixed a validation issue for small precision compute types < B32 on gfx9.

• Fixed CMake validation of compiler support for BF8/FP8 types.

RPP (2.0.0)

Added

• Bitwise NOT, Bitwise AND, and Bitwise OR augmentations on HOST (CPU) and HIP backends.

• Tensor Concat augmentation on HOST (CPU) and HIP backends.

• JPEG Compression Distortion augmentation on HIP backend.

• log1p, defined as log (1 + x), tensor augmentation support on HOST (CPU) and HIP backends.

• JPEG Compression Distortion augmentation on HOST (CPU) backend.

Changed

• Handle creation and destruction APIs have been consolidated. Use rppCreate() for handle initialization and rppDestroy() for handle destruction.

• The logical_operations function category has been renamed to bitwise_operations.

• TurboJPEG package installation enabled for RPP Test Suite with sudo apt-get install libturbojpeg0-dev. Instructions have been updated in utilities/test_suite/README.md.

• The swap_channels augmentation has been changed to channel_permute. channel_permute now also accepts a new argument, permutationTensor (pointer to an unsigned int tensor), that provides the permutation order to swap the RGB channels of each input image in the batch in any order:

  RppStatus rppt_swap_channels_host(RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, rppHandle_t rppHandle);

  changed to:

  RppStatus rppt_channel_permute_host(RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32u *permutationTensor , rppHandle_t rppHandle);

Removed

• Older versions of RPP handle creation inlcuding rppCreateWithBatchSize(), rppCreateWithStream(), and rppCreateWithStreamAndBatchSize(). These have been replaced with rppCreate().

• Older versions of RPP handle destruction API including rppDestroyGPU() and rppDestroyHost(). These have been replaced with rppDestroy().

Resolved issues

• Test package - Debian packages will install required dependencies.

Tensile (4.44.0)

Added

• Support for gfx950.

• Code object compression via bundling.

• Support for non-default HIP SDK installations on Windows.

• Master solution library documentation.

• Compiler version-dependent assembler and architecture capabilities.

• Documentation from GitHub Wiki to ROCm docs.

Changed

• Loosened check for CLI compiler choices.

• Introduced 4-tuple targets for bundler invocations.

• Introduced PATHEXT extensions on Windows when searching for toolchain components.

• Enabled passing fully qualified paths to toolchain components.

• Enabled environment variable overrides when searching for a ROCm stack.

• Improved default toolchain configuration.

• Ignored f824 flake errors.

Removed

• Support for the gfx940 and gfx941 targets.

• Unused tuning files.

• Disabled tests.

Resolved issues

• Fixed configure time path not being invoked at build.

• Fixed find_package for msgpack to work with versions 5 and 6.

• Fixed RHEL 9 testing.

• Fixed gfx908 builds.

• Fixed the ‘argument list too long’ error.

• Fixed version typo in 6.3 changelog.

• Fixed improper use of aliases as nested namespace specifiers.

ROCm known issues

ROCm known issues are noted on GitHub. For known issues related to individual components, review the Detailed component changes.

A memory error in the kernel might lead to applications using the ROCr library becoming unresponsive

Applications using the ROCr library might become unresponsive if a memory error occurs in the launched kernel when the queue from which it was launched is destroyed. The application is unable to receive further signal, resulting in the stall condition. The issue will be fixed in a future ROCm release.

Applications using stream capture APIs might fail during stream capture

Applications using hipLaunchHostFunc with stream capture APIs might fail to capture graphs during stream capture, and return hipErrorStreamCaptureUnsupported. This issue resulted from an update in hipStreamAddCallback. This issue will be fixed in a future ROCm release.

Compilation failure via hipRTC when compiling with std=c++11

Applications compiling kernels using hipRTC might fail while passing the std=c++11 compiler option. This issue will be fixed in a future ROCm release.

Compilation failure when referencing std::array if _GLIBCXX_ASSERTIONS is defined

Compiling from a device kernel or function results in failure when attempting to reference std::array if _GLIBCXX_ASSERTIONS is defined. The issue occurs because there’s no device definition for std::__glibcxx_asert_fail(). This issue will be resolved in a future ROCm release with the implementation of std::__glibcxx_assert_fail().

Segmentation fault in ROCprofiler-SDK due to ABI mismatch affecting std::regex

Starting with GCC 5.1, GNU libstdc++ introduced a dual Application Binary Interface (ABI) to adopt C++11, primarily affecting the std::string and its dependencies, including std::regex. If your code is compiled against headers expecting one ABI but linked or run with the other, it can cause problems with std::string and std::regex, leading to a segmentation fault in ROCprofiler-SDK, which uses std::regex. This issue is resolved in the ROCm Systems develop branch and will be part of a future ROCm release.

Decline in performance of batched GEMM operation for applications using hipBLASLT kernels

Default batched General Matrix Multiplications (GEMM) operations for rocBLAS and hipBLAS on gfx1200 and gfx1201 may have a decline in performance in comparison with non-batched and strided_batched GEMM operations. By default, the batched GEMM uses hipBLASLT kernels, and switching to the Tensile kernel resolves the performance decline issue. The issue will be fixed in a future ROCm release. As a workaround, you can set the environment variable ROCBLAS_USE_HIPBLASLT=0 before the batched GEMM operation is performed on gfx1200 and gfx1201. After completing the batched operation, reset the variable to ROCBLAS_USE_HIPBLASLT=1 before calling non-batched or strided_batched operations.

Failure to declare out-of-bound CPERs for bad memory page

Exceeding bad memory page threshold fails to declare Out-Of-Band Common Platform Error Records (CPERs). This issue affects all AMD Instinct MI300 Series and MI350 Series GPUs, and will be fixed in a future AMD GPU Driver release.

ROCm resolved issues

The following are previously known issues resolved in this release. For resolved issues related to individual components, review the Detailed component changes.

Failure when using a generic target with compression and vice versa

An issue where compiling a generic target resulted in compression failing has been resolved in this release. This issue prevented you from compiling a generic target and using compression simultaneously. See GitHub issue #4602.

Limited support for Sparse API and Pallas functionality in JAX

An issue where due to limited support for Sparse API in JAX, some of the functionality of the Pallas extension were restricted has been resolved. See GitHub issue #4608.

Failure to use –kokkos-trace option in ROCm Compute Profiler

An issue where using of the --kokkos-trace option resulted in a difference between the output of the --kokkos-trace and the counter_collection.csv output file has been resolved. Due to this issue, the program exited with a warning message if the -kokkos-trace option was detected in the ROCm Compute Profiler. This issue was due to the partial implementation of --kokkos-trace in rocprofv3 tool. See GitHub issue #4604.

ROCm upcoming changes

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

ROCm SMI deprecation

ROCm SMI will be phased out in an upcoming ROCm release and will enter maintenance mode. After this transition, only critical bug fixes will be addressed and no further feature development will take place.

It’s strongly recommended to transition your projects to AMD SMI, the successor to ROCm SMI. AMD SMI includes all the features of the ROCm SMI and will continue to receive regular updates, new functionality, and ongoing support. For more information on AMD SMI, see the AMD SMI documentation.

ROCTracer, ROCProfiler, rocprof, and rocprofv2 deprecation

Development and support for ROCTracer, ROCProfiler, rocprof, and rocprofv2 are being phased out in favor of ROCprofiler-SDK in upcoming ROCm releases. Starting with ROCm 6.4, only critical defect fixes will be addressed for older versions of the profiling tools and libraries. All users are encouraged to upgrade to the latest version of the ROCprofiler-SDK library and the (rocprofv3) tool to ensure continued support and access to new features. ROCprofiler-SDK is still in beta today and will be production-ready in a future ROCm release.

It’s anticipated that ROCTracer, ROCProfiler, rocprof, and rocprofv2 will reach end-of-life by future releases, aligning with Q1 of 2026.

AMDGPU wavefront size compiler macro deprecation

Access to the wavefront size as a compile-time constant via the __AMDGCN_WAVEFRONT_SIZE and __AMDGCN_WAVEFRONT_SIZE__ macros are deprecated and will be disabled in a future release. In ROCm 7.0.0 warpSize is only available as a non-constextpr variable. You’re encouraged to update your code if needed to ensure future compatibility.

• The __AMDGCN_WAVEFRONT_SIZE__ macro and __AMDGCN_WAVEFRONT_SIZE alias will be removed in an upcoming release. It is recommended to remove any use of this macro. For more information, see AMDGPU support.

• warpSize is only available as a non-constexpr variable. Where required, the wavefront size should be queried via the warpSize variable in device code, or via hipGetDeviceProperties in host code. Neither of these will result in a compile-time constant. For more information, see warpSize.

• For cases where compile-time evaluation of the wavefront size cannot be avoided, uses of __AMDGCN_WAVEFRONT_SIZE, __AMDGCN_WAVEFRONT_SIZE__, or warpSize can be replaced with a user-defined macro or constexpr variable with the wavefront size(s) for the target hardware. For example:

   #if defined(__GFX9__)
   #define MY_MACRO_FOR_WAVEFRONT_SIZE 64
   #else
   #define MY_MACRO_FOR_WAVEFRONT_SIZE 32
   #endif

Changes to ROCm Object Tooling

ROCm Object Tooling tools roc-obj-ls, roc-obj-extract, and roc-obj were deprecated in ROCm 6.4, and will be removed in a future release. Functionality has been added to the llvm-objdump --offloading tool option to extract all clang-offload-bundles into individual code objects found within the objects or executables passed as input. The llvm-objdump --offloading tool option also supports the --arch-name option, and only extracts code objects found with the specified target architecture. See llvm-objdump for more information.