The release notes provide a summary of notable changes since the previous ROCm release.
If you’re using Radeon™ PRO or Radeon GPUs in a workstation setting with a display connected, see the [Use ROCm on Radeon GPUs](https://rocm.docs.amd.com/projects/radeon/en/latest/docs/compatibility/native_linux/native_linux_compatibility.html)
documentation to verify compatibility and system requirements.
The following are notable new features and improvements in ROCm 6.3.3. For changes to individual components, see Detailed component changes.
The ROCm Offline Installer Creator 6.3.3 adds a new Post-Install Options menu, which includes a new udev option for adding GPU resources access for all users. It also moves the user-specific GPU access option (for the video,render group) from the Driver Options menu to the Post-Install Options menu. See the ROCm Offline Installer Creator documentation for more information.
ROCm documentation continues to be updated to provide clearer and more comprehensive guidance for a wider variety of user needs and use cases.
-
Tutorials for AI developers have been added. These tutorials are Jupyter notebook-based, easy-to-follow documents. They are ideal for AI developers who want to learn about specific topics, including inference, fine-tuning, and training.
-
The LLM inference performance validation guide for AMD Instinct MI300X now includes additional models for performance benchmarking. The accompanying ROCm vLLM Docker has been upgraded to ROCm 6.3.1.
-
The HIP documentation has been updated with new resources for developers. To learn more about concurrency, parallelism, and stream management on devices and multiple GPUs, see Asynchronous concurrent execution
-
The following HIP documentation topics have been updated:
Operating system and hardware support remain unchanged in this release.
See the Compatibility matrix for more information about operating system and hardware compatibility.
The following table lists the versions of ROCm components for ROCm 6.3.3, including any version
changes from 6.3.2 to 6.3.3. Click the component's updated version to go to a list of its changes.
Click {fab}github 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 | |
| MIGraphX | 2.11.0 | |||
| MIOpen | 3.3.0 | |||
| MIVisionX | 3.1.0 | |||
| rocAL | 2.1.0 | |||
| rocDecode | 0.8.0 | |||
| rocJPEG | 0.6.0 | |||
| rocPyDecode | 0.2.0 | |||
| RPP | 1.9.1 | |||
| Communication | RCCL | 2.21.5 | ||
| Math | hipBLAS | 2.3.0 | ||
| hipBLASLt | 0.10.0 | |||
| hipFFT | 1.0.17 | |||
| hipfort | 0.5.1 | |||
| hipRAND | 2.11.1 | |||
| hipSOLVER | 2.3.0 | |||
| hipSPARSE | 3.1.2 | |||
| hipSPARSELt | 0.2.2 | |||
| rocALUTION | 3.2.1 | |||
| rocBLAS | 4.3.0 | |||
| rocFFT | 1.0.31 | |||
| rocRAND | 3.2.0 | |||
| rocSOLVER | 3.27.0 | |||
| rocSPARSE | 3.3.0 | |||
| rocWMMA | 1.6.0 | |||
| Tensile | 4.42.0 | |||
| Primitives | hipCUB | 3.3.0 | ||
| hipTensor | 1.4.0 | |||
| rocPRIM | 3.3.0 | |||
| rocThrust | 3.3.0 | |||
| Tools | System management | AMD SMI | 24.7.1 | |
| ROCm Data Center Tool | 0.3.0 | |||
| rocminfo | 1.0.0 | |||
| ROCm SMI | 7.4.0 | |||
| ROCmValidationSuite | 1.1.0 | |||
| Performance | ROCm Bandwidth Test | 1.4.0 | ||
| ROCm Compute Profiler | 3.0.0 | |||
| ROCm Systems Profiler | 0.1.1 ⇒ 0.1.2 | |||
| ROCProfiler | 2.0.0 | |||
| ROCprofiler-SDK | 0.5.0 | |||
| ROCTracer | 4.1.0 | |||
| Development | HIPIFY | 18.0.0 | ||
| ROCdbgapi | 0.77.0 | |||
| ROCm CMake | 0.14.0 | |||
| ROCm Debugger (ROCgdb) | 15.2 | |||
| ROCr Debug Agent | 2.0.3 | |||
| Compilers | HIPCC | 1.1.1 | ||
| llvm-project | 18.0.0 | |||
| Runtimes | HIP | 6.3.2 | ||
| ROCr Runtime | 1.14.0 | |||
The following sections describe key changes to ROCm components.
- Fixed an error that prevented GPU hardware activity from being presented in certain workloads.
ROCm known issues are noted on {fab}github GitHub. For known
issues related to individual components, review the Detailed component changes.
The ROCTx markers are standalone markers within the ROCProfiler-SDK library. Each marker reports only a single timestamp, which is recorded as the start_timestamp and end_timestamp. As a result, the value for aggregated statistics presented in TotalDurationNs, maxNs, and minNs, is zero. The zero value indicates that the actual execution time is not associated with the markers, which is an expected behavior. See GitHub issue #4396.
The following changes to the ROCm software stack are anticipated for future releases.
Development and support for ROCTracer and ROCProfiler (rocprof and rocprofv2) will phase out in favor of ROCprofiler-SDK (rocprofv3) in upcoming ROCm releases. Going forward, only critical defect fixes will be addressed for older versions of profiling tools and libraries. Upgrade to the latest version of ROCprofiler-SDK (rocprofv3) library to ensure continued support and access to new features.
The __AMDGCN_WAVEFRONT_SIZE__ macro will be deprecated in an upcoming
release. It is recommended to remove any use of this macro. For more information, see AMDGPU
support.
The HIPCC Perl scripts (hipcc.pl and hipconfig.pl) will be removed in an upcoming release.