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, continue to use ROCm 6.2.3. See the [Use ROCm on Radeon
GPUs](https://rocm.docs.amd.com/projects/radeon/en/latest/index.html)
documentation to verify compatibility and system requirements.
The following are notable new features and improvements in ROCm 6.2.4. For changes to individual components, see Detailed component changes.
ROCm documentation continues to be updated to provide clearer and more comprehensive guidance for a wider variety of user needs and use cases.
-
Added a new GPU cluster networking guide. See Cluster network performance validation for AMD Instinct accelerators. This documentation provides guidelines on validating network configurations in single-node and multi-node environments to attain optimal speed and bandwidth in AMD Instinct-powered clusters.
-
Updated the HIP runtime documentation.
-
Added a new section on how to use HIP graphs.
-
Added a new section about the Stream ordered memory allocator (SOMA).
-
Updated the Porting CUDA driver API section.
-
-
Updated the Post-installation instructions with guidance on using the
update-alternatives
utility and environment modules to help you manage multiple ROCm versions and streamline PATH configuration. -
Updated the LLM inference performance validation on AMD Instinct MI300X documentation with more detailed guidance, new models, and the
float8
data type.
ROCm 6.2.4 adds support for the AMD Radeon PRO V710 GPU for compute workloads. See Supported GPUs for more information.
This release maintains the same operating system support as 6.2.2.
The following table lists the versions of ROCm components for ROCm 6.2.4, including any version changes from 6.2.2 to 6.2.4.
Click the component's updated version to go to a detailed 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 | |
MIGraphX | 2.10 | |||
MIOpen | 3.2.0 | |||
MIVisionX | 3.0.0 | |||
rocAL | 2.0.0 | |||
rocDecode | 0.6.0 | |||
rocPyDecode | 0.1.0 | |||
RPP | 1.8.0 | |||
Communication | RCCL | 2.20.5 | ||
Math | hipBLAS | 2.2.0 | ||
hipBLASLt | 0.8.0 | |||
hipFFT | 1.0.15 ⇒ 1.0.16 | |||
hipfort | 0.4.0 | |||
hipRAND | 2.11.0 ⇒ 2.11.1 | |||
hipSOLVER | 2.2.0 | |||
hipSPARSE | 3.1.1 | |||
hipSPARSELt | 0.2.1 | |||
rocALUTION | 3.2.0 ⇒ 3.2.1 | |||
rocBLAS | 4.2.1 ⇒ 4.2.4 | |||
rocFFT | 1.0.29 ⇒ 1.0.30 | |||
rocRAND | 3.1.0 ⇒ 3.1.1 | |||
rocSOLVER | 3.26.0 ⇒ 3.26.2 | |||
rocSPARSE | 3.2.0 ⇒ 3.2.1 | |||
rocWMMA | 1.5.0 | |||
Tensile | 4.41.0 | |||
Primitives | hipCUB | 3.2.0 ⇒ 3.2.1 | ||
hipTensor | 1.3.0 | |||
rocPRIM | 3.2.1 ⇒ 3.2.2 | |||
rocThrust | 3.1.0 ⇒ 3.1.1 | |||
Tools | System management | AMD SMI | 24.6.3 ⇒ 24.6.3 | |
rocminfo | 1.0.0 | |||
ROCm Data Center Tool | 0.3.0 | |||
ROCm SMI | 7.3.0 | |||
ROCm Validation Suite | 1.0.0 | |||
Performance | Omniperf | 2.0.1 | ||
Omnitrace | 1.11.2 | |||
ROCm Bandwidth Test | 1.4.0 | |||
ROCProfiler | 2.0.0 | |||
ROCprofiler-SDK | 0.4.0 | |||
ROCTracer | 4.1.0 | |||
Development | HIPIFY | 18.0.0 | ||
ROCdbgapi | 0.76.0 | |||
ROCm CMake | 0.13.0 | |||
ROCm Debugger (ROCgdb) | 14.2 | |||
ROCr Debug Agent | 2.0.3 | |||
Compilers | HIPCC | 1.1.1 | ||
llvm-project | 18.0.0 | |||
Runtimes | HIP | 6.2.4 | ||
ROCr Runtime | 1.14.0 |
The following sections describe key changes to ROCm components.
- Fixed support for the API calls
amdsmi_get_gpu_process_isolation
andamdsmi_clean_gpu_local_data
, along with theamd-smi set --process-isolation <0 or 1>
command. See issue #3500 on GitHub.
- Implemented 1D kernels for factorizable sizes greater than 1024 and less than 2048.
- Fixed plan creation failure on some even-length real-complex transforms that use Bluestein's algorithm.
- Fixed synchronization issue in STEIN.
ROCm known issues are tracked on GitHub. Known issues related to individual components are listed in the Detailed component changes section.
The following changes to the ROCm software stack are anticipated for future releases.
The rocm-llvm-alt
package will be removed in an upcoming release. Users relying on the functionality provided by the closed-source compiler should transition to the open-source compiler. Once the rocm-llvm-alt
package is removed, any compilation requesting functionality provided by the closed-source compiler will result in a Clang warning: "[AMD] proprietary optimization compiler has been removed".
The RCCL plugin package, rccl-rdma-sharp-plugins
, will be removed in an upcoming ROCm release.