# ROCm 6.3.3 release notes The release notes provide a summary of notable changes since the previous ROCm release. - [Release highlights](#release-highlights) - [Operating system and hardware support changes](#operating-system-and-hardware-support-changes) - [ROCm components versioning](#rocm-components) - [Detailed component changes](#detailed-component-changes) - [ROCm known issues](#rocm-known-issues) - [ROCm upcoming changes](#rocm-upcoming-changes) ```{note} 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. ``` ## Release highlights The following are notable new features and improvements in ROCm 6.3.3. For changes to individual components, see [Detailed component changes](#detailed-component-changes). ### ROCm Offline Installer Creator updates 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](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/rocm-offline-installer.html#post-install-options-menu) documentation for more information. ### 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. * [Tutorials for AI developers](https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/) 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](https://rocm.docs.amd.com/en/latest/how-to/rocm-for-ai/inference/vllm-benchmark.html) 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](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_runtime_api/asynchronous.html) * The following HIP documentation topics have been updated: - [Virtual memory management](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_runtime_api/memory_management/virtual_memory.html) - [Programming for HIP runtime compiler (RTC)](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_rtc.html) - [HIP porting guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html) - [Porting CUDA driver API](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_driver_api.html) - [CUDA to HIP API function comparison](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/api_syntax.html) ## Operating system and hardware support changes Operating system and hardware support remain unchanged in this release. See the [Compatibility matrix](https://rocm.docs.amd.com/en/docs-6.3.3/compatibility/compatibility-matrix.html) for more information about operating system and hardware compatibility. ## ROCm components 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
## Detailed component changes The following sections describe key changes to ROCm components. ### **ROCm Systems Profiler** (0.1.2) #### Resolved issues * Fixed an error that prevented GPU hardware activity from being presented in certain workloads. ## ROCm known issues ROCm known issues are noted on {fab}`github` [GitHub](https://github.com/ROCm/ROCm/labels/Verified%20Issue). For known issues related to individual components, review the [Detailed component changes](#detailed-component-changes). ### Zero value is displayed in ROCTx aggregated statistics 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](https://github.com/ROCm/ROCm/issues/4396). ## ROCm upcoming changes The following changes to the ROCm software stack are anticipated for future releases. ### ROCTracer and ROCProfiler (rocprof and rocprofv2) deprecation 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. ### AMDGPU wavefront size compiler macro deprecation 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](https://rocm.docs.amd.com/projects/llvm-project/en/docs-6.3.3/LLVM/clang/html/AMDGPUSupport.html). ### HIPCC Perl scripts deprecation The HIPCC Perl scripts (`hipcc.pl` and `hipconfig.pl`) will be removed in an upcoming release.