From 34c0266496a5485b077a23483fab49f574bc7ae8 Mon Sep 17 00:00:00 2001 From: Peter Park Date: Wed, 6 Nov 2024 12:42:55 -0500 Subject: [PATCH] 6.2.4 release notes: add known/fixed issues (#193) * add "for compute workloads" wording for clarity * add AMDSMI resolved issue * add dlm known issue intro text wording * update wording rm bullet point update wording * fix spellcheck due to spacing * rm s * rm gfx1151 * remove dlm known issue * update list of updated docs; note for Radeon users fmt * update GA date for 6.2.4 * fix rdc version --- RELEASE.md | 73 ++++++++----------- docs/conf.py | 2 +- .../templates/extra_components/6.2.4.md | 9 +++ tools/autotag/templates/highlights/6.2.4.md | 27 ++++--- 4 files changed, 60 insertions(+), 51 deletions(-) diff --git a/RELEASE.md b/RELEASE.md index 935d6997b..726a599d4 100644 --- a/RELEASE.md +++ b/RELEASE.md @@ -15,8 +15,8 @@ The release notes provide a summary of notable changes since the previous ROCm r - [ROCm upcoming changes](#rocm-upcoming-changes) ```{note} -If you’re using Radeon™ PRO or Radeon GPUs for graphics workloads, -continue to use ROCm 6.2.3. See the [Use ROCm on Radeon +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. ``` @@ -33,7 +33,6 @@ 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](https://rocm.docs.amd.com/projects/gpu-cluster-networking/en/latest/index.html). - 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. @@ -46,9 +45,17 @@ a wider variety of user needs and use cases. * Updated the [Porting CUDA driver API](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_driver_api.html) section. +* Updated the [Post-installation instructions](https://rocm.docs.amd.com/projects/install-on-linux/en/docs-6.2.4/install/post-install.html) + 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](https://rocm.docs.amd.com/en/docs-6.2.4/how-to/performance-validation/mi300x/vllm-benchmark.html) + documentation with more detailed guidance, new models, and the `float8` data type. + ## Operating system and hardware support changes -ROCm 6.2.4 adds support for the [AMD Radeon PRO V710](https://www.amd.com/en/products/accelerators/radeon-pro/amd-radeon-pro-v710.html) GPU. See +ROCm 6.2.4 adds support for the [AMD Radeon PRO V710](https://www.amd.com/en/products/accelerators/radeon-pro/amd-radeon-pro-v710.html) GPU for compute workloads. See [Supported GPUs](https://advanced-micro-devices-demo--287.com.readthedocs.build/projects/install-on-linux-internal/en/287/reference/system-requirements.html) for more information. @@ -154,7 +161,7 @@ Click the component's updated version to go to a detailed list of its changes. C hipFFT - 1.0.15 ⇒ 1.0.16 + 1.0.15 ⇒ 1.0.16 @@ -166,7 +173,7 @@ Click the component's updated version to go to a detailed list of its changes. C hipRAND - 2.11.0 ⇒ 2.11.1 + 2.11.0 ⇒ 2.11.1 @@ -190,13 +197,13 @@ Click the component's updated version to go to a detailed list of its changes. C rocALUTION - 3.2.0 ⇒ 3.2.1 + 3.2.0 ⇒ 3.2.1 rocBLAS - 4.2.1 ⇒ 4.2.4 + 4.2.1 ⇒ 4.2.4 @@ -208,7 +215,7 @@ Click the component's updated version to go to a detailed list of its changes. C rocRAND - 3.1.0 ⇒ 3.1.1 + 3.1.0 ⇒ 3.1.1 @@ -220,7 +227,7 @@ Click the component's updated version to go to a detailed list of its changes. C rocSPARSE - 3.2.0 ⇒ 3.2.1 + 3.2.0 ⇒ 3.2.1 @@ -242,7 +249,7 @@ Click the component's updated version to go to a detailed list of its changes. C Primitives hipCUB - 3.2.0 ⇒ 3.2.1 + 3.2.0 ⇒ 3.2.1 @@ -254,13 +261,13 @@ Click the component's updated version to go to a detailed list of its changes. C rocPRIM - 3.2.1 ⇒ 3.2.2 + 3.2.1 ⇒ 3.2.2 rocThrust - 3.1.0 ⇒ 3.1.1 + 3.1.0 ⇒ 3.1.1 @@ -270,7 +277,7 @@ Click the component's updated version to go to a detailed list of its changes. C Tools System management AMD SMI - 24.6.3 + 24.6.3 ⇒ 24.6.3 @@ -282,7 +289,7 @@ Click the component's updated version to go to a detailed list of its changes. C ROCm Data Center Tool - 1.0.0 + 0.3.0 @@ -413,34 +420,16 @@ Click the component's updated version to go to a detailed list of its changes. C The following sections describe key changes to ROCm components. -### Hardware architecture support updates +### **AMD SMI** (24.6.3) -Updated the following math and primitives libraries to pre-enable support for -an upcoming hardware architecture. +#### Resolved issues -* hipCUB (3.2.1) +* Fixed support for the API calls `amdsmi_get_gpu_process_isolation` and + `amdsmi_clean_gpu_local_data`, along with the `amd-smi set + --process-isolation <0 or 1>` command. See issue + [#3500](https://github.com/ROCm/ROCm/issues/3500) on GitHub. -* hipFFT (1.0.16) - -* hipRAND (2.11.1) - -* rocALUTION (3.2.1) - -* rocBLAS (4.2.4) - -* rocFFT (1.0.30) - -* rocPRIM (3.2.2) - -* rocRAND (3.1.1) - -* rocSOLVER (3.26.2) - -* rocSPARSE (3.2.1) - -* rocThrust (3.1.1) - -### **rocFFT** (1.0.30)[*](#hardware-architecture-support-updates) +### **rocFFT** (1.0.30) #### Optimized @@ -450,7 +439,7 @@ an upcoming hardware architecture. * Fixed plan creation failure on some even-length real-complex transforms that use Bluestein's algorithm. -### **rocSOLVER** (3.26.2)[*](#hardware-architecture-support-updates) +### **rocSOLVER** (3.26.2) #### Resolved issues @@ -459,6 +448,8 @@ an upcoming hardware architecture. ## ROCm known issues ROCm known issues are tracked on [GitHub](https://github.com/ROCm/ROCm/labels/Verified%20Issue). +Known issues related to individual components are listed in the [Detailed component changes](#detailed-component-changes) +section. ## ROCm upcoming changes diff --git a/docs/conf.py b/docs/conf.py index cca8f0ad3..fe3f22199 100644 --- a/docs/conf.py +++ b/docs/conf.py @@ -38,7 +38,7 @@ all_article_info_author = "" # pages with specific settings article_pages = [ - {"file": "about/release-notes", "os": ["linux", "windows"], "date": "2024-10-18"}, + {"file": "about/release-notes", "os": ["linux", "windows"], "date": "2024-11-06"}, {"file": "how-to/deep-learning-rocm", "os": ["linux"]}, {"file": "how-to/rocm-for-ai/index", "os": ["linux"]}, {"file": "how-to/rocm-for-ai/install", "os": ["linux"]}, diff --git a/tools/autotag/templates/extra_components/6.2.4.md b/tools/autotag/templates/extra_components/6.2.4.md index 4dee22b54..37f6d2962 100644 --- a/tools/autotag/templates/extra_components/6.2.4.md +++ b/tools/autotag/templates/extra_components/6.2.4.md @@ -24,3 +24,12 @@ an upcoming hardware architecture. * rocSPARSE (3.2.1) * rocThrust (3.1.1) + +### **AMD SMI** (24.6.3) + +#### Resolved issues + +* Fixed support for the API calls `amdsmi_get_gpu_process_isolation` and + `amdsmi_clean_gpu_local_data`, along with the + `amd-smi set --process-isolation <0 or 1>` command. See issue + [#3500](https://github.com/ROCm/ROCm/issues/3500) on GitHub. diff --git a/tools/autotag/templates/highlights/6.2.4.md b/tools/autotag/templates/highlights/6.2.4.md index da45a17d3..4d8af7762 100644 --- a/tools/autotag/templates/highlights/6.2.4.md +++ b/tools/autotag/templates/highlights/6.2.4.md @@ -1,4 +1,6 @@ -These release notes provide a summary of notable changes since the previous ROCm release. +# ROCm 6.2.4 release notes + +The release notes provide a summary of notable changes since the previous ROCm release. - [Release highlights](#release-highlights) @@ -13,9 +15,8 @@ These release notes provide a summary of notable changes since the previous ROCm - [ROCm upcoming changes](#rocm-upcoming-changes) ```{note} -ROCm 6.2.3 is supported on systems using AMD Radeon™ or Radeon PRO workstation -GPUs for graphics workloads. If you’re using ROCm in this context, use ROCm -version 6.2.3. See the [Use ROCm on Radeon +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. ``` @@ -30,8 +31,8 @@ The following are notable new features and improvements in ROCm 6.2.4. For chang 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 topic. See - [Cluster network performance validation for AMD Instinct accelerators](https://rocm.docs.amd.com/projects/gpu-cluster-networking/en/docs-6.2.4/index.html). +* Added a new GPU cluster networking guide. See + [Cluster network performance validation for AMD Instinct accelerators](https://rocm.docs.amd.com/projects/gpu-cluster-networking/en/latest/index.html). This documentation provides guidelines on validating network configurations in single-node and multi-node environments to attain optimal speed and bandwidth @@ -39,8 +40,16 @@ a wider variety of user needs and use cases. * Updated the HIP runtime documentation. - * Added a new topic on how to use [HIP graphs](https://rocm.docs.amd.com/projects/HIP/en/docs-6.2.4/how-to/hipgraph.html). + * Added a new section on how to use [HIP graphs](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hipgraph.html). - * Added a new topic about the [Stream ordered memory allocator (SOMA)](https://rocm.docs.amd.com/projects/HIP/en/docs-6.2.4/how-to/stream_ordered_allocator.html). + * Added a new section about the [Stream ordered memory allocator (SOMA)](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/stream_ordered_allocator.html). - * Updated the [Porting CUDA driver API](https://rocm.docs.amd.com/projects/HIP/en/docs-6.2.4/how-to/hip_porting_driver_api.html) topic. + * Updated the [Porting CUDA driver API](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_driver_api.html) section. + +* Updated the [Post-installation instructions](https://rocm.docs.amd.com/projects/install-on-linux/en/docs-6.2.4/install/post-install.html) + with guidance on using the `update-alternatives` utility and environment modules to help you manage multiple ROCm + versions and streamline PATH configuration. + +* Updated [LLM inference performance validation on AMD Instinct + MI300X](https://rocm.docs.amd.com/en/docs-6.2.4/how-to/performance-validation/mi300x/vllm-benchmark.html) + documentation with more detailed guidance, new models, and the `float8` data type.