rocm-docs-core experiment

.wordlist.txt

@@ -1,3 +1,18 @@
+Datacenter
+GST
+IET
+LTO
+MX
+Microscaling
+NANOO
+ROCprof
+affinitization
+amdclang
+benefitting
+demangled
+inlined
+microscaling
+roofline
 AAC
 ABI
 ACE

docs/conf.py

@@ -34,69 +34,86 @@ project = "ROCm Documentation"
 project_path = os.path.abspath(".").replace("\\", "/")
 author = "Advanced Micro Devices, Inc."
 copyright = "Copyright (c) 2025 Advanced Micro Devices, Inc. All rights reserved."
-version = "6.4.1"
+version = "7.0 Alpha"
-release = "6.4.1"
+release = "7.0 Alpha"
 setting_all_article_info = True
 all_article_info_os = ["linux", "windows"]
 all_article_info_author = ""
 
 # pages with specific settings
 article_pages = [
-    {"file": "about/release-notes", "os": ["linux"], "date": "2025-05-07"},
+    {"file": "preview/index", "os": ["linux"],},
-    {"file": "release/changelog", "os": ["linux"],},
+    {"file": "preview/release", "os": ["linux"],},
-    {"file": "compatibility/compatibility-matrix", "os": ["linux"]},
+    {"file": "preview/install/index", "os": ["linux"],},
-    {"file": "compatibility/ml-compatibility/pytorch-compatibility", "os": ["linux"]},
+    {"file": "preview/install/instinct-driver", "os": ["linux"],},
-    {"file": "compatibility/ml-compatibility/tensorflow-compatibility", "os": ["linux"]},
+    {"file": "preview/install/rocm", "os": ["linux"],},
-    {"file": "compatibility/ml-compatibility/jax-compatibility", "os": ["linux"]},
+    {"file": "preview/benchmark-docker/index", "os": ["linux"],},
-    {"file": "how-to/deep-learning-rocm", "os": ["linux"]},
+    {"file": "preview/benchmark-docker/training", "os": ["linux"],},
+    {"file": "preview/benchmark-docker/pre-training-megatron-lm-llama-3-8b", "os": ["linux"],},
+    {"file": "preview/benchmark-docker/pre-training-torchtitan-llama-3-70b", "os": ["linux"],},
+    {"file": "preview/benchmark-docker/fine-tuning-lora-llama-2-70b", "os": ["linux"],},
+    {"file": "preview/benchmark-docker/inference", "os": ["linux"],},
+    {"file": "preview/benchmark-docker/inference-vllm-llama-3.1-405b-fp4", "os": ["linux"],},
+    {"file": "preview/benchmark-docker/inference-sglang-deepseek-r1-fp4", "os": ["linux"],},
 
-    {"file": "how-to/rocm-for-ai/index", "os": ["linux"]},
+    # {"file": "about/release-notes", "os": ["linux"], "date": "2025-06-26"},
-    {"file": "how-to/rocm-for-ai/install", "os": ["linux"]},
+    # {"file": "release/changelog", "os": ["linux"],},
-    {"file": "how-to/rocm-for-ai/system-health-check", "os": ["linux"]},
+    # {"file": "compatibility/compatibility-matrix", "os": ["linux"]},
-
+    # {"file": "compatibility/ml-compatibility/pytorch-compatibility", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/index", "os": ["linux"]},
+    # {"file": "compatibility/ml-compatibility/tensorflow-compatibility", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/train-a-model", "os": ["linux"]},
+    # {"file": "compatibility/ml-compatibility/jax-compatibility", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/prerequisite-system-validation", "os": ["linux"]},
+    # {"file": "how-to/deep-learning-rocm", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/benchmark-docker/megatron-lm", "os": ["linux"]},
+    #
-    {"file": "how-to/rocm-for-ai/training/benchmark-docker/pytorch-training", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/index", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/install", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/training/scale-model-training", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/system-health-check", "os": ["linux"]},
-
+    #
-    {"file": "how-to/rocm-for-ai/fine-tuning/index", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/training/index", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/fine-tuning/overview", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/training/train-a-model", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/training/prerequisite-system-validation", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/fine-tuning/single-gpu-fine-tuning-and-inference", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/training/benchmark-docker/megatron-lm", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/fine-tuning/multi-gpu-fine-tuning-and-inference", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/training/benchmark-docker/pytorch-training", "os": ["linux"]},
-
+    # {"file": "how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference/index", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/training/scale-model-training", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference/hugging-face-models", "os": ["linux"]},
+    #
-    {"file": "how-to/rocm-for-ai/inference/llm-inference-frameworks", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/fine-tuning/index", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/vllm", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/fine-tuning/overview", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.8.5-20250513", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference/benchmark-docker/pytorch-inference", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/fine-tuning/single-gpu-fine-tuning-and-inference", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference/deploy-your-model", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/fine-tuning/multi-gpu-fine-tuning-and-inference", "os": ["linux"]},
-
+    #
-    {"file": "how-to/rocm-for-ai/inference-optimization/index", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference/index", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference-optimization/model-quantization", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference/hugging-face-models", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference-optimization/model-acceleration-libraries", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference/llm-inference-frameworks", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference-optimization/optimizing-with-composable-kernel", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference/benchmark-docker/vllm", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference-optimization/optimizing-triton-kernel", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference/benchmark-docker/previous-versions/vllm-0.8.5-20250513", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference-optimization/profiling-and-debugging", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference/benchmark-docker/pytorch-inference", "os": ["linux"]},
-    {"file": "how-to/rocm-for-ai/inference-optimization/workload", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference/deploy-your-model", "os": ["linux"]},
-
+    #
-    {"file": "how-to/system-optimization/index", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference-optimization/index", "os": ["linux"]},
-    {"file": "how-to/system-optimization/mi300x", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference-optimization/model-quantization", "os": ["linux"]},
-    {"file": "how-to/system-optimization/mi200", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference-optimization/model-acceleration-libraries", "os": ["linux"]},
-    {"file": "how-to/system-optimization/mi100", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference-optimization/optimizing-with-composable-kernel", "os": ["linux"]},
-    {"file": "how-to/system-optimization/w6000-v620", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference-optimization/optimizing-triton-kernel", "os": ["linux"]},
-    {"file": "how-to/tuning-guides/mi300x/index", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference-optimization/profiling-and-debugging", "os": ["linux"]},
-    {"file": "how-to/tuning-guides/mi300x/system", "os": ["linux"]},
+    # {"file": "how-to/rocm-for-ai/inference-optimization/workload", "os": ["linux"]},
-    {"file": "how-to/tuning-guides/mi300x/workload", "os": ["linux"]},
+    #
-    {"file": "how-to/system-debugging", "os": ["linux"]},
+    # {"file": "how-to/system-optimization/index", "os": ["linux"]},
-    {"file": "how-to/gpu-enabled-mpi", "os": ["linux"]},
+    # {"file": "how-to/system-optimization/mi300x", "os": ["linux"]},
+    # {"file": "how-to/system-optimization/mi200", "os": ["linux"]},
+    # {"file": "how-to/system-optimization/mi100", "os": ["linux"]},
+    # {"file": "how-to/system-optimization/w6000-v620", "os": ["linux"]},
+    # {"file": "how-to/tuning-guides/mi300x/index", "os": ["linux"]},
+    # {"file": "how-to/tuning-guides/mi300x/system", "os": ["linux"]},
+    # {"file": "how-to/tuning-guides/mi300x/workload", "os": ["linux"]},
+    # {"file": "how-to/system-debugging", "os": ["linux"]},
+    # {"file": "how-to/gpu-enabled-mpi", "os": ["linux"]},
 ]
 
 external_toc_path = "./sphinx/_toc.yml"
+# Options to improve documentation build time for preview release documentation
+external_toc_exclude_missing = True # don't build files that aren't in the TOC
+external_projects_remote_repository = "" # don't fetch data to resolve intersphinx xrefs
 
 # Add the _extensions directory to Python's search path
 sys.path.append(str(Path(__file__).parent / 'extension'))
@@ -122,7 +139,7 @@ html_static_path = ["sphinx/static/css", "extension/how-to/rocm-for-ai/inference
 html_css_files = ["rocm_custom.css", "rocm_rn.css", "vllm-benchmark.css"]
 html_js_files = ["vllm-benchmark.js"]
 
-html_title = "ROCm Documentation"
+html_title = "ROCm 7.0 Alpha documentation"
 
 html_theme_options = {"link_main_doc": False}
 

docs/how-to/rocm-for-ai/install.rst

@@ -1,5 +1,5 @@
 .. meta::
-   :description: How to install ROCm and popular machine learning frameworks.
+   :description: How to install the ROCm 7.0 preview
    :keywords: ROCm, AI, LLM, train, fine-tune, FSDP, DeepSpeed, LLaMA, tutorial
 
 .. _rocm-for-ai-install:

docs/preview/index.md

@@ -0,0 +1,26 @@
+---
+myst:
+  html_meta:
+    "description": "AMD ROCm 7.0 Alpha documentation"
+    "keywords": "Radeon, open, compute, platform, install, how, conceptual, reference, home, docs"
+---
+
+# AMD ROCm 7.0 Alpha documentation
+
+AMD ROCm is an open-source software platform optimized to extract HPC and AI
+workload performance from AMD Instinct™ accelerators while maintaining
+compatibility with industry software frameworks.
+
+This documentation is intended to provide early access information about the ROCm
+software Alpha release. The preview release provides early access to new
+features under development for testing for users to provide feedback.
+It is not recommended for production use.
+
+```{note}
+See [ROCm documentation](https://rocm.docs.amd.com/en/latest/) for the latest stable release for use in production.
+```
+
+The documentation includes:
+
+- [ROCm 7.0 Alpha release notes](release.rst) with feature details and support matrix
+- [Installation instructions](install/index.rst) for the ROCm 7.0 Alpha and the Instinct Driver

docs/preview/install/index.rst

@@ -0,0 +1,28 @@
+.. meta::
+  :description: Installation via native package manager
+  :keywords: ROCm install, installation instructions, package manager, native package manager, AMD,
+    ROCm
+
+****************************************
+ROCm 7.0 Alpha installation instructions
+****************************************
+
+The ROCm 7.0 Alpha must be installed using your Linux distribution's native
+package manager. This release supports specific hardware and software
+configurations -- before installing, see the :ref:`supported OSes and hardware
+<alpha-system-requirements>` outlined in the Alpha release notes.
+
+.. important::
+
+   Upgrades and downgrades are not supported. You must install any existing
+   ROCm installation before installing the Alpha build.
+
+.. grid:: 2
+
+   .. grid-item-card:: Install ROCm
+
+      See :doc:`Install the ROCm 7.0 Alpha via package manager <rocm>`.
+
+   .. grid-item-card:: Install Instinct Driver
+
+      See :doc:`Install the Instinct Driver via package manager <instinct-driver>`.

docs/preview/install/instinct-driver.rst

@@ -0,0 +1,212 @@
+***********************************************
+Install the Instinct Driver via package manager
+***********************************************
+
+This section describes how to install the Instinct Driver using ``apt`` on
+Ubuntu 22.04 or 24.04, or ``dnf`` on Red Hat Enterprise Linux 9.6.
+
+.. important::
+
+   Upgrades and downgrades are not supported. You must uninstall any existing
+   ROCm installation before installing the preview build.
+
+Prerequisites
+=============
+
+Before installing, complete the following prerequisites.
+
+.. tab-set::
+
+   .. tab-item:: Ubuntu 22.04
+      :sync: ubuntu-22
+
+      Install kernel headers.
+
+      .. code-block:: shell
+
+         sudo apt install "linux-headers-$(uname -r)" "linux-modules-extra-$(uname -r)" 
+
+   .. tab-item:: Ubuntu 24.04
+      :sync: ubuntu-24
+
+      Install kernel headers.
+
+      .. code-block:: shell
+
+         sudo apt install "linux-headers-$(uname -r)" "linux-modules-extra-$(uname -r)" 
+
+   .. tab-item:: RHEL 9.6
+      :sync: rhel-96
+
+      1. Register your Enterprise Linux.
+
+         .. code-block:: shell
+
+            subscription-manager register --username <username> --password <password>
+            subscription-manager attach --auto
+
+      2. Update your Enterprise Linux.
+
+         .. code-block:: shell
+
+            sudo dnf update --releasever=9.6 --exclude=\*release\*
+
+      3. Install kernel headers.
+
+         .. code-block:: shell
+
+            sudo dnf install "kernel-headers-$(uname -r)" "kernel-devel-$(uname -r)" "kernel-devel-matched-$(uname -r)"
+
+Register ROCm repositories
+==========================
+
+.. tab-set::
+
+   .. tab-item:: Ubuntu 22.04
+      :sync: ubuntu-22
+
+      1. Add the package signing key.
+
+         .. code-block:: shell
+
+            # Make the directory if it doesn't exist yet.
+            # This location is recommended by the distribution maintainers.
+            sudo mkdir --parents --mode=0755 /etc/apt/keyrings 
+            # Download the key, convert the signing-key to a full
+            # keyring required by apt and store in the keyring directory.
+            wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | \
+              gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null 
+
+      2. Register the kernel mode driver.
+
+         .. code-block:: shell
+
+            echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/30.10_alpha/ubuntu jammy main" \
+              | sudo tee /etc/apt/sources.list.d/amdgpu.list
+            sudo apt update 
+
+   .. tab-item:: Ubuntu 24.04
+      :sync: ubuntu-24
+
+      1. Add the package signing key.
+
+         .. code-block:: shell
+
+            # Make the directory if it doesn't exist yet.
+            # This location is recommended by the distribution maintainers.
+            sudo mkdir --parents --mode=0755 /etc/apt/keyrings 
+            # Download the key, convert the signing-key to a full
+            # keyring required by apt and store in the keyring directory.
+            wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | \
+              gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null 
+
+      2. Register the kernel mode driver.
+
+         .. code-block:: shell
+
+            echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/amdgpu/30.10_alpha/ubuntu noble main" \
+              | sudo tee /etc/apt/sources.list.d/amdgpu.list
+            sudo apt update 
+
+   .. tab-item:: RHEL 9.6
+      :sync: rhel-96
+
+      .. code-block:: shell
+
+         sudo tee /etc/yum.repos.d/amdgpu.repo <<EOF
+         [amdgpu]
+         name=amdgpu
+         baseurl=https://repo.radeon.com/amdgpu/30.10_alpha/rhel/9.6/main/x86_64/
+         enabled=1
+         priority=50
+         gpgcheck=1
+         gpgkey=https://repo.radeon.com/rocm/rocm.gpg.key
+         EOF
+         sudo dnf clean all
+
+Install the kernel driver
+=========================
+
+.. tab-set::
+
+   .. tab-item:: Ubuntu 22.04
+      :sync: ubuntu-22
+
+      .. code-block:: shell
+
+         sudo apt install amdgpu-dkms
+
+   .. tab-item:: Ubuntu 24.04
+      :sync: ubuntu-24
+
+      .. code-block:: shell
+
+         sudo apt install amdgpu-dkms
+
+   .. tab-item:: RHEL 9.6
+      :sync: rhel-96
+
+      .. code-block:: shell
+
+         sudo dnf install amdgpu-dkms
+
+Uninstalling
+============
+
+.. tab-set::
+
+   .. tab-item:: Ubuntu 22.04
+      :sync: ubuntu-22
+
+      1. Uninstall the kernel mode driver.
+
+         .. code-block:: shell
+
+            sudo apt autoremove amdgpu-dkms
+
+      2. Remove AMDGPU repositories.
+
+         .. code-block:: shell
+
+            sudo rm /etc/apt/sources.list.d/amdgpu.list
+            # Clear the cache and clean the system
+            sudo rm -rf /var/cache/apt/*
+            sudo apt clean all
+            sudo apt update
+
+   .. tab-item:: Ubuntu 24.04
+      :sync: ubuntu-24
+
+      1. Uninstall the kernel mode driver.
+
+         .. code-block:: shell
+
+            sudo apt autoremove amdgpu-dkms
+
+      2. Remove AMDGPU repositories.
+
+         .. code-block:: shell
+
+            sudo rm /etc/apt/sources.list.d/amdgpu.list
+            # Clear the cache and clean the system
+            sudo rm -rf /var/cache/apt/*
+            sudo apt clean all
+            sudo apt update
+
+   .. tab-item:: RHEL 9.6
+      :sync: rhel-96
+
+      1. Uninstall the kernel mode driver.
+
+         .. code-block:: shell
+
+            sudo dnf remove amdgpu-dkms
+
+      2. Remove AMDGPU repositories.
+
+         .. code-block:: shell
+
+            sudo rm /etc/yum.repos.d/amdgpu.repo
+            # Clear the cache and clean the system
+            sudo rm -rf /var/cache/dnf
+            sudo dnf clean all

docs/preview/install/rocm.rst

@@ -0,0 +1,288 @@
+**********************************************
+Install the ROCm 7.0 Alpha via package manager
+**********************************************
+
+This page describes how to install the ROCm 7.0 Alpha build using ``apt`` on
+Ubuntu 22.04 or 24.04, or ``dnf`` on Red Hat Enterprise Linux 9.6.
+
+.. important::
+
+   Upgrades and downgrades are not supported. You must uninstall any existing
+   ROCm installation before installing the preview build.
+
+Prerequisites
+=============
+
+Before installing, complete the following prerequisites.
+
+.. tab-set::
+
+   .. tab-item:: Ubuntu 22.04
+      :sync: ubuntu-22
+
+      1. Install development packages.
+
+         .. code-block:: shell
+
+            sudo apt install python3-setuptools python3-wheel
+
+      2. Configure user permissions for GPU access.
+
+         .. code-block:: shell
+
+            sudo usermod -a -G render,video $LOGNAME
+
+   .. tab-item:: Ubuntu 24.04
+      :sync: ubuntu-24
+
+      1. Install development packages.
+
+         .. code-block:: shell
+
+            sudo apt install python3-setuptools python3-wheel
+
+      2. Configure user permissions for GPU access.
+
+         .. code-block:: shell
+
+            sudo usermod -a -G render,video $LOGNAME
+
+   .. tab-item:: RHEL 9.6
+      :sync: rhel-96
+
+      1. Register your Enterprise Linux.
+
+         .. code-block:: shell
+
+            subscription-manager register --username <username> --password <password>
+            subscription-manager attach --auto
+
+      2. Update your Enterprise Linux.
+
+         .. code-block:: shell
+
+            sudo dnf update --releasever=9.6 --exclude=\*release\*
+
+      3. Install additional package repositories.
+
+         Add the EPEL repository:
+
+         .. code-block:: shell
+
+            wget https://dl.fedoraproject.org/pub/epel/epel-release-latest-9.noarch.rpm
+            sudo rpm -ivh epel-release-latest-9.noarch.rpm
+
+         Enable the CodeReady Linux Build (CRB) repository.
+
+         .. code-block:: shell
+
+            sudo dnf install dnf-plugin-config-manager
+            sudo crb enable
+
+      4. Install development packages.
+
+         .. code-block:: shell
+
+            sudo dnf install python3-setuptools python3-wheel
+
+      5. Configure user permissions for GPU access.
+
+         .. code-block:: shell
+
+            sudo usermod -a -G render,video $LOGNAME
+
+Register ROCm repositories
+==========================
+
+.. tab-set::
+
+   .. tab-item:: Ubuntu 22.04
+      :sync: ubuntu-22
+
+      1. Add the package signing key.
+
+         .. code-block:: shell
+
+            # Make the directory if it doesn't exist yet.
+            # This location is recommended by the distribution maintainers.
+            sudo mkdir --parents --mode=0755 /etc/apt/keyrings 
+            # Download the key, convert the signing-key to a full
+            # keyring required by apt and store in the keyring directory.
+            wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | \
+              gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null 
+
+      2. Register ROCm packages.
+
+         .. code-block:: shell
+
+            echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/7.0_alpha jammy main" \
+              | sudo tee /etc/apt/sources.list.d/rocm.list
+
+            echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/graphics/7.0_alpha/ubuntu jammy main" \ 
+              | sudo tee /etc/apt/sources.list.d/rocm-graphics.list
+
+            echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' \
+              | sudo tee /etc/apt/preferences.d/rocm-pin-600
+            sudo apt update
+
+   .. tab-item:: Ubuntu 24.04
+      :sync: ubuntu-24
+
+      1. Add the package signing key.
+
+         .. code-block:: shell
+
+            # Make the directory if it doesn't exist yet.
+            # This location is recommended by the distribution maintainers.
+            sudo mkdir --parents --mode=0755 /etc/apt/keyrings 
+            # Download the key, convert the signing-key to a full
+            # keyring required by apt and store in the keyring directory.
+            wget https://repo.radeon.com/rocm/rocm.gpg.key -O - | \
+              gpg --dearmor | sudo tee /etc/apt/keyrings/rocm.gpg > /dev/null 
+
+      2. Register ROCm packages.
+
+         .. code-block:: shell
+
+            echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/7.0_alpha noble main" \
+              | sudo tee /etc/apt/sources.list.d/rocm.list
+
+            echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/graphics/7.0_alpha/ubuntu noble main" \
+              | sudo tee /etc/apt/sources.list.d/rocm-graphics.list
+
+            echo -e 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' \
+              | sudo tee /etc/apt/preferences.d/rocm-pin-600
+            sudo apt update 
+
+   .. tab-item:: RHEL 9.6
+      :sync: rhel-96
+
+      .. code-block:: shell
+
+         sudo tee /etc/yum.repos.d/rocm.repo <<EOF
+         [ROCm-7.0.0]
+         name=ROCm7.0.0
+         baseurl=https://repo.radeon.com/rocm/el9/7.0_alpha/main
+         enabled=1
+         priority=50
+         gpgcheck=1
+         gpgkey=https://repo.radeon.com/rocm/rocm.gpg.key
+         EOF
+
+         sudo tee /etc/yum.repos.d/rocm-graphics.repo <<EOF
+         [ROCm-7.0.0-Graphics]
+         name=ROCm7.0.0-Graphics
+         baseurl=https://repo.radeon.com/graphics/7.0_alpha/rhel/9/main/x86_64/
+         enabled=1
+         priority=50
+         gpgcheck=1
+         gpgkey=https://repo.radeon.com/rocm/rocm.gpg.key
+         EOF
+         sudo dnf clean all
+
+Install ROCm
+============
+
+.. tab-set::
+
+   .. tab-item:: Ubuntu 22.04
+      :sync: ubuntu-22
+
+      .. code-block:: shell
+
+         sudo apt install rocm
+
+   .. tab-item:: Ubuntu 24.04
+      :sync: ubuntu-24
+
+      .. code-block:: shell
+
+         sudo apt install rocm
+
+   .. tab-item:: RHEL 9.6
+      :sync: rhel-96
+
+      .. code-block:: shell
+
+         sudo dnf install rocm
+
+.. _uninstall-rocm:
+
+Uninstalling
+============
+
+.. tab-set::
+
+   .. tab-item:: Ubuntu 22.04
+      :sync: ubuntu-22
+
+      1. Uninstall specific meta packages.
+
+         .. code-block:: shell
+
+            sudo apt autoremove rocm
+
+      2. Uninstall ROCm packages.
+
+         .. code-block:: shell
+
+            sudo apt autoremove rocm-core
+
+      3. Remove ROCm repositories.
+
+         .. code-block:: shell
+
+            sudo rm /etc/apt/sources.list.d/rocm*.list
+            # Clear the cache and clean the system
+            sudo rm -rf /var/cache/apt/*
+            sudo apt clean all
+            sudo apt update
+
+   .. tab-item:: Ubuntu 24.04
+      :sync: ubuntu-24
+
+      1. Uninstall specific meta packages.
+
+         .. code-block:: shell
+
+            sudo apt autoremove rocm
+
+      2. Uninstall ROCm packages.
+
+         .. code-block:: shell
+
+            sudo apt autoremove rocm-core
+
+      3. Remove ROCm repositories.
+
+         .. code-block:: shell
+
+            sudo rm /etc/apt/sources.list.d/rocm*.list
+            # Clear the cache and clean the system
+            sudo rm -rf /var/cache/apt/*
+            sudo apt clean all
+            sudo apt update
+
+   .. tab-item:: RHEL 9.6
+      :sync: rhel-96
+
+      1. Uninstall specific meta packages.
+
+         .. code-block:: shell
+
+            sudo dnf remove rocm
+
+      2. Uninstall ROCm packages.
+
+         .. code-block:: shell
+
+            sudo dnf remove rocm-core amdgpu-core
+
+      3. Remove ROCm repositories.
+
+         .. code-block:: shell
+
+            sudo rm /etc/yum.repos.d/rocm*.repo*
+            # Clear the cache and clean the system
+            sudo rm -rf /var/cache/dnf
+            sudo dnf clean all

docs/preview/release.rst

@@ -0,0 +1,270 @@
+****************************
+ROCm 7.0 Alpha release notes
+****************************
+
+The ROCm 7.0 Alpha is an early look into the upcoming ROCm 7.0 major release,
+which introduces functional support for AMD Instinct™ MI355X and MI350X
+on bare metal, single node systems. It also includes new features for current-generation
+MI300X, MI200, and MI100 series accelerators. This is an alpha-quality release;
+expect issues and limitations that will be addressed in upcoming previews.
+
+.. important::
+
+   This Alpha release is not intended for performance evaluation.
+   For the latest stable release for production-level functionality,
+   see `ROCm documentation <https://rocm.docs.amd.com/en/latest/>`_.
+
+This page provides a high-level summary of supported systems, key changes to the ROCm software
+stack, developments related to AI frameworks, current known limitations, and installation
+information.
+
+.. _alpha-system-requirements:
+
+Operating system and hardware support
+=====================================
+
+Only the accelerators and operating systems listed here are supported. Multi-node systems,
+virtualized environments, and GPU partitioning are not supported in this Alpha.
+
+* AMD accelerator: Instinct MI355X, MI350X, MI325X [#mi325x]_, MI300X, MI300A, MI250X, MI250, MI210, MI100
+* Operating system: Ubuntu 22.04, Ubuntu 24.04, or RHEL 9.6
+* System type: Bare metal, single node only
+* Partitioning: Not supported
+
+.. [#mi325x] MI325X is only supported with Ubuntu 22.04.
+
+.. _alpha-highlights:
+
+Alpha release highlights
+========================
+
+This section highlights key features enabled in the ROCm 7.0 Alpha.
+
+AI frameworks
+-------------
+
+PyTorch
+~~~~~~~
+
+The ROCm 7.0 Alpha 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
+
+TensorFlow
+~~~~~~~~~~
+
+This Alpha enables support for TensorFlow 2.19.
+
+vLLM
+~~~~
+
+* Support for Open Compute Project (OCP) ``FP8`` data type
+
+* ``FP4`` precision for Llama 3.1 405B
+
+Libraries
+---------
+
+.. _alpha-new-data-type-support:
+
+New data type support
+~~~~~~~~~~~~~~~~~~~~~
+
+MX-compliant data types bring microscaling support to ROCm. For more information, see the `OCP
+Microscaling (MX) Formats Specification
+<https://www.opencompute.org/documents/ocp-microscaling-formats-mx-v1-0-spec-final-pdf>`_. The ROCm
+7.0 Alpha enables functional support for MX data types ``FP4``, ``FP6``, and ``FP8`` on MI355X
+systems in these ROCm libraries:
+
+* Composable Kernel (``FP4`` and ``FP8`` only)
+
+* hipBLASLt
+
+* MIGraphX (``FP4`` only)
+
+The following libraries are updated to support the Open Compute Project (OCP) floating-point ``FP8``
+format on MI355X instead of the NANOO ``FP8`` format:
+
+* Composable Kernel
+
+* hipBLASLt
+
+* hipSPARSELt
+
+* MIGraphX
+
+* rocWMMA
+
+MIGraphX now also supports ``BF16``.
+
+RCCL support
+~~~~~~~~~~~~
+
+RCCL is supported for single-node functional usage only. Multi-node communication capabilities will
+be supported in future preview releases.
+
+MIGraphX
+~~~~~~~~
+
+* Support for OCP ``FP8`` and MX ``FP4`` data types on MI355X
+
+* Support for ``BF16`` on all hardware
+
+* Support for PyTorch 2.7 via Torch-MIGraphX
+
+Tools
+-----
+
+AMD SMI
+~~~~~~~
+
+* The default output of the ``amd-smi`` CLI now displays a simple table view.
+
+* New APIs: CPU affinity shows GPUs' affinitization to each CPU in a system.
+
+ROCgdb
+~~~~~~
+
+* MX data types support: ``FP4``, ``FP6``, and ``FP8``
+
+ROCprof Compute Viewer
+~~~~~~~~~~~~~~~~~~~~~~
+
+* Initial release: ``rocprof-compute-viewer`` allows the visualization of ``rocprofv3``'s thread
+  trace output
+
+ROCprof Trace Decoder
+~~~~~~~~~~~~~~~~~~~~~
+
+* Initial release: ``rocprof-trace-decoder`` a plugin API for decoding thread traces
+
+ROCm Compute Profiler
+~~~~~~~~~~~~~~~~~~~~~
+
+* MX data types support: ``FP4``, ``FP6``, and ``FP8``
+
+* MI355X and MI350X performance counters: CPC, SPI, SQ, TA/TD/TCP, and TCC
+
+* Enhanced roofline analysis with support for ``INT8``, ``INT32``, ``FP8``, ``FP16``, and ``BF16``
+  data types
+
+* Roofline distinction for ``FP32`` and ``FP64`` data types
+
+* Selective kernel profiling
+
+ROCm Systems Profiler
+~~~~~~~~~~~~~~~~~~~~~
+
+* Trace support for computer vision APIs: H264, H265, AV1, VP9, and JPEG
+
+* Trace support for computer vision engine activity
+
+* OpenMP for C++ language and kernel activity support
+
+ROCm Validation Suite
+~~~~~~~~~~~~~~~~~~~~~
+
+* MI355X and MI350X accelerator support in the IET (Integrated Execution Test), GST (GPU Stress Test), and Babel (memory bandwidth test) modules.
+
+ROCprofiler-SDK
+~~~~~~~~~~~~~~~
+
+* Program counter (PC) sampling (host trap-based)
+
+* API for profiling applications using thread traces (beta)
+
+* Support in ``rocprofv3`` CLI tool for thread trace service
+
+HIP
+---
+
+The HIP runtime 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
+
+In addition, the HIP runtime includes the following functional improvements which improve runtime
+performance and user experience:
+
+* Optimized HIP runtime lock contention in some events and kernel handling APIs. Event processing
+  and memory object look-ups now use the shared mutex implementation. Kernel object look-up during
+  C++ kernel launch can now avoid a global lock. These changes improve performance in certain
+  applications with high usage, particularly for multiple GPUs, multiple threads, and HIP streams
+  per GPU.
+
+* Programmatic support for scratch buffer limit on GPU device. Developers can now change the default
+  allocation size with the expected scratch limit.
+
+* Unified managed buffer and kernel argument buffers so the HIP runtime no longer needs to create
+  and load a separate kernel argument buffer.
+
+* Refactored memory validation to create a unique function to validate a variety of memory copy
+  operations.
+
+* Shader names are now demangled for more readable kernel logs
+
+See :ref:`HIP compatibility <hip-known-limitation>`.
+
+Compilers
+---------
+
+* 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 may be put in a separate partition, and any non-inlined
+  function it depends on may 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>``.
+
+    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 disables 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. Recommended values are 1 to 16 partitions, with
+    especially large projects containing many kernels potentially benefitting 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 projects that contain only a few kernels may also not benefit at
+    all from partitioning and may even see a slight increase in build time due to the small overhead
+    of analyzing and partitioning the modules.
+
+* HIPIFY now supports NVIDIA CUDA 12.8.0 APIs. See
+  `<https://github.com/ROCm/HIPIFY/blob/amd-develop/docs/reference/supported_apis.md>`_ for more
+  information.
+
+Instinct Driver / ROCm packaging separation
+-------------------------------------------
+
+The Instinct Driver is now distributed separately from the ROCm software stack -- it is now stored
+in its own location in the package repository at `<repo.radeon.com>`_ under ``/amdgpu/``.
+The first release is designated as Instinct Driver version 30.10 See `ROCm Gets Modular: Meet the
+Instinct Datacenter GPU Driver
+<https://rocm.blogs.amd.com/ecosystems-and-partners/instinct-gpu-driver/README.html>`_ for more
+information.
+
+Forward and backward compatibility between the Instinct Driver and ROCm are not supported in this
+Alpha release. See the :doc:`installation instructions <install/index>`.
+
+Known limitations
+=================
+
+.. _hip-known-limitation:
+
+HIP compatibility
+-----------------
+
+HIP runtime APIs in the ROCm 7.0 Alpha do not include backward-incompatible changes. See `HIP 7.0 Is
+Coming: What You Need to Know to Stay Ahead
+<https://rocm.blogs.amd.com/ecosystems-and-partners/transition-to-hip-7.0:-guidance-on-upcoming-compatibility-changes/README.html>`_ for more information.

docs/sphinx/_toc.yml.in

@@ -3,195 +3,206 @@
 defaults:
   numbered: False
   maxdepth: 6
-root: index
+root: preview/index
 subtrees:
 - entries:
-  - file: what-is-rocm.rst
+  - file: preview/release.rst
-  - file: about/release-notes.md
+    title: Alpha release notes
-    title: Release notes
+  - file: preview/install/index.rst
-  - file: compatibility/compatibility-matrix.rst
+    title: Installation
-    title: Compatibility matrix
-    entries:
-    - url: https://rocm.docs.amd.com/projects/install-on-linux/en/latest/reference/system-requirements.html
-      title: Linux system requirements
-    - url: https://rocm.docs.amd.com/projects/install-on-windows/en/${branch}/reference/system-requirements.html
-      title: Windows system requirements
-
-- caption: Install
-  entries:
-  - url: https://rocm.docs.amd.com/projects/install-on-linux/en/latest/
-    title: ROCm on Linux
-  - url: https://rocm.docs.amd.com/projects/install-on-windows/en/${branch}/
-    title: HIP SDK on Windows
-  - url: https://rocm.docs.amd.com/projects/radeon/en/latest/index.html
-    title: ROCm on Radeon GPUs
-  - file: how-to/deep-learning-rocm.md
-    title: Deep learning frameworks
-  - file: how-to/build-rocm.rst
-    title: Build ROCm from source
-
-- caption: How to
-  entries:
-  - file: how-to/rocm-for-ai/index.rst
-    title: Use ROCm for AI
     subtrees:
     - entries:
-      - file: how-to/rocm-for-ai/install.rst
+        - file: preview/install/rocm
-        title: Installation
+          title: Install ROCm
-      - file: how-to/rocm-for-ai/system-health-check.rst
+        - file: preview/install/instinct-driver
-        title: System health benchmarks
+          title: Install Instinct Driver
-      - file: how-to/rocm-for-ai/training/index.rst
+# - entries:
-        title: Training
+#   - file: what-is-rocm.rst
-        subtrees:
+#   - file: about/release-notes.md
-        - entries:
+#     title: Release notes
-          - file: how-to/rocm-for-ai/training/benchmark-docker/megatron-lm.rst
+#   - file: compatibility/compatibility-matrix.rst
-            title: Train a model with Megatron-LM
+#     title: Compatibility matrix
-          - file: how-to/rocm-for-ai/training/benchmark-docker/pytorch-training.rst
+#     entries:
-            title: Train a model with PyTorch
+#     - url: https://rocm.docs.amd.com/projects/install-on-linux-internal/en/latest/reference/system-requirements.html
-          - file: how-to/rocm-for-ai/training/benchmark-docker/jax-maxtext.rst
+#       title: Linux system requirements
-            title: Train a model with JAX MaxText
+#     - url: https://rocm.docs.amd.com/projects/install-on-windows/en/${branch}/reference/system-requirements.html
-          - file: how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry
+#       title: Windows system requirements
-            title: Train a model with LLM Foundry
+#
-          - file: how-to/rocm-for-ai/training/scale-model-training.rst
+# - caption: Install
-            title: Scale model training
+#   entries:
-
+#   - url: https://rocm.docs.amd.com/projects/install-on-linux-internal/en/latest/
-      - file: how-to/rocm-for-ai/fine-tuning/index.rst
+#     title: ROCm on Linux
-        title: Fine-tuning LLMs
+#   - url: https://rocm.docs.amd.com/projects/install-on-windows/en/${branch}/
-        subtrees:
+#     title: HIP SDK on Windows
-        - entries:
+#   - url: https://rocm.docs.amd.com/projects/radeon/en/latest/index.html
-          - file: how-to/rocm-for-ai/fine-tuning/overview.rst
+#     title: ROCm on Radeon GPUs
-            title: Conceptual overview
+#   - file: how-to/deep-learning-rocm.md
-          - file: how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference.rst
+#     title: Deep learning frameworks
-            title: Fine-tuning
+#   - file: how-to/build-rocm.rst
-            subtrees:
+#     title: Build ROCm from source
-            - entries:
+#
-              - file: how-to/rocm-for-ai/fine-tuning/single-gpu-fine-tuning-and-inference.rst
+# - caption: How to
-                title: Use a single accelerator
+#   entries:
-              - file: how-to/rocm-for-ai/fine-tuning/multi-gpu-fine-tuning-and-inference.rst
+#   - file: how-to/rocm-for-ai/index.rst
-                title: Use multiple accelerators
+#     title: Use ROCm for AI
-
+#     subtrees:
-      - file: how-to/rocm-for-ai/inference/index.rst
+#     - entries:
-        title: Inference
+#       - file: how-to/rocm-for-ai/install.rst
-        subtrees:
+#         title: Installation
-        - entries:
+#       - file: how-to/rocm-for-ai/system-health-check.rst
-          - file: how-to/rocm-for-ai/inference/hugging-face-models.rst
+#         title: System health benchmarks
-            title: Run models from Hugging Face
+#       - file: how-to/rocm-for-ai/training/index.rst
-          - file: how-to/rocm-for-ai/inference/llm-inference-frameworks.rst
+#         title: Training
-            title: LLM inference frameworks
+#         subtrees:
-          - file: how-to/rocm-for-ai/inference/benchmark-docker/vllm.rst
+#         - entries:
-            title: vLLM inference performance testing
+#           - file: how-to/rocm-for-ai/training/benchmark-docker/megatron-lm.rst
-          - file: how-to/rocm-for-ai/inference/benchmark-docker/pytorch-inference.rst
+#             title: Train a model with Megatron-LM
-            title: PyTorch inference performance testing
+#           - file: how-to/rocm-for-ai/training/benchmark-docker/pytorch-training.rst
-          - file: how-to/rocm-for-ai/inference/deploy-your-model.rst
+#             title: Train a model with PyTorch
-            title: Deploy your model
+#           - file: how-to/rocm-for-ai/training/benchmark-docker/jax-maxtext.rst
-
+#             title: Train a model with JAX MaxText
-      - file: how-to/rocm-for-ai/inference-optimization/index.rst
+#           - file: how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry
-        title: Inference optimization
+#             title: Train a model with LLM Foundry
-        subtrees:
+#           - file: how-to/rocm-for-ai/training/scale-model-training.rst
-        - entries:
+#             title: Scale model training
-          - file: how-to/rocm-for-ai/inference-optimization/model-quantization.rst
+#
-          - file: how-to/rocm-for-ai/inference-optimization/model-acceleration-libraries.rst
+#       - file: how-to/rocm-for-ai/fine-tuning/index.rst
-          - file: how-to/rocm-for-ai/inference-optimization/optimizing-with-composable-kernel.md
+#         title: Fine-tuning LLMs
-            title: Optimize with Composable Kernel
+#         subtrees:
-          - file: how-to/rocm-for-ai/inference-optimization/optimizing-triton-kernel.rst
+#         - entries:
-            title: Optimize Triton kernels
+#           - file: how-to/rocm-for-ai/fine-tuning/overview.rst
-          - file: how-to/rocm-for-ai/inference-optimization/profiling-and-debugging.rst
+#             title: Conceptual overview
-            title: Profile and debug
+#           - file: how-to/rocm-for-ai/fine-tuning/fine-tuning-and-inference.rst
-          - file: how-to/rocm-for-ai/inference-optimization/workload.rst
+#             title: Fine-tuning
-            title: Workload optimization
+#             subtrees:
-
+#             - entries:
-      - url: https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/
+#               - file: how-to/rocm-for-ai/fine-tuning/single-gpu-fine-tuning-and-inference.rst
-        title: AI tutorials
+#                 title: Use a single accelerator
-
+#               - file: how-to/rocm-for-ai/fine-tuning/multi-gpu-fine-tuning-and-inference.rst
-  - file: how-to/rocm-for-hpc/index.rst
+#                 title: Use multiple accelerators
-    title: Use ROCm for HPC
+#
-  - file: how-to/system-optimization/index.rst
+#       - file: how-to/rocm-for-ai/inference/index.rst
-    title: System optimization
+#         title: Inference
-  - file: how-to/gpu-performance/mi300x.rst
+#         subtrees:
-    title: AMD Instinct MI300X performance guides
+#         - entries:
-  - file: how-to/system-debugging.md
+#           - file: how-to/rocm-for-ai/inference/hugging-face-models.rst
-  - file: conceptual/compiler-topics.md
+#             title: Run models from Hugging Face
-    title: Use advanced compiler features
+#           - file: how-to/rocm-for-ai/inference/llm-inference-frameworks.rst
-    subtrees:
+#             title: LLM inference frameworks
-    - entries:
+#           - file: how-to/rocm-for-ai/inference/benchmark-docker/vllm.rst
-      - url: https://rocm.docs.amd.com/projects/llvm-project/en/latest/index.html
+#             title: vLLM inference performance testing
-        title: ROCm compiler infrastructure
+#           - file: how-to/rocm-for-ai/inference/benchmark-docker/pytorch-inference.rst
-      - url: https://rocm.docs.amd.com/projects/llvm-project/en/latest/conceptual/using-gpu-sanitizer.html
+#             title: PyTorch inference performance testing
-        title: Use AddressSanitizer
+#           - file: how-to/rocm-for-ai/inference/deploy-your-model.rst
-      - url: https://rocm.docs.amd.com/projects/llvm-project/en/latest/conceptual/openmp.html
+#             title: Deploy your model
-        title: OpenMP support
+#
-  - file: how-to/setting-cus
+#       - file: how-to/rocm-for-ai/inference-optimization/index.rst
-    title: Set the number of CUs
+#         title: Inference optimization
-  - file: how-to/Bar-Memory.rst
+#         subtrees:
-    title: Troubleshoot BAR access limitation  
+#         - entries:
-  - url: https://github.com/amd/rocm-examples
+#           - file: how-to/rocm-for-ai/inference-optimization/model-quantization.rst
-    title: ROCm examples
+#           - file: how-to/rocm-for-ai/inference-optimization/model-acceleration-libraries.rst
-
+#           - file: how-to/rocm-for-ai/inference-optimization/optimizing-with-composable-kernel.md
-
+#             title: Optimize with Composable Kernel
-- caption: Conceptual
+#           - file: how-to/rocm-for-ai/inference-optimization/optimizing-triton-kernel.rst
-  entries:
+#             title: Optimize Triton kernels
-  - file: conceptual/gpu-arch.md
+#           - file: how-to/rocm-for-ai/inference-optimization/profiling-and-debugging.rst
-    title: GPU architecture overview
+#             title: Profile and debug
-    subtrees:
+#           - file: how-to/rocm-for-ai/inference-optimization/workload.rst
-    - entries:
+#             title: Workload optimization
-      - file: conceptual/gpu-arch/mi300.md
+#
-        title: MI300 microarchitecture
+#       - url: https://rocm.docs.amd.com/projects/ai-developer-hub/en/latest/
-        subtrees:
+#         title: AI tutorials
-        - entries:
+#
-          - url: https://www.amd.com/content/dam/amd/en/documents/instinct-tech-docs/instruction-set-architectures/amd-instinct-mi300-cdna3-instruction-set-architecture.pdf
+#   - file: how-to/rocm-for-hpc/index.rst
-            title: AMD Instinct MI300/CDNA3 ISA
+#     title: Use ROCm for HPC
-          - url: https://www.amd.com/content/dam/amd/en/documents/instinct-tech-docs/white-papers/amd-cdna-3-white-paper.pdf
+#   - file: how-to/system-optimization/index.rst
-            title: White paper
+#     title: System optimization
-          - file: conceptual/gpu-arch/mi300-mi200-performance-counters.rst
+#   - file: how-to/gpu-performance/mi300x.rst
-            title: MI300 and MI200 Performance counter
+#     title: AMD Instinct MI300X performance guides
-      - file: conceptual/gpu-arch/mi250.md
+#   - file: how-to/system-debugging.md
-        title: MI250 microarchitecture
+#   - file: conceptual/compiler-topics.md
-        subtrees:
+#     title: Use advanced compiler features
-        - entries:
+#     subtrees:
-          - url: https://www.amd.com/system/files/TechDocs/instinct-mi200-cdna2-instruction-set-architecture.pdf
+#     - entries:
-            title: AMD Instinct MI200/CDNA2 ISA
+#       - url: https://rocm.docs.amd.com/projects/llvm-project/en/latest/index.html
-          - url: https://www.amd.com/content/dam/amd/en/documents/instinct-business-docs/white-papers/amd-cdna2-white-paper.pdf
+#         title: ROCm compiler infrastructure
-            title: White paper
+#       - url: https://rocm.docs.amd.com/projects/llvm-project/en/latest/conceptual/using-gpu-sanitizer.html
-      - file: conceptual/gpu-arch/mi100.md
+#         title: Use AddressSanitizer
-        title: MI100 microarchitecture
+#       - url: https://rocm.docs.amd.com/projects/llvm-project/en/latest/conceptual/openmp.html
-        subtrees:
+#         title: OpenMP support
-        - entries:
+#   - file: how-to/setting-cus
-          - url: https://www.amd.com/system/files/TechDocs/instinct-mi100-cdna1-shader-instruction-set-architecture%C2%A0.pdf
+#     title: Set the number of CUs
-            title: AMD Instinct MI100/CDNA1 ISA
+#   - file: how-to/Bar-Memory.rst
-          - url: https://www.amd.com/content/dam/amd/en/documents/instinct-business-docs/white-papers/amd-cdna-white-paper.pdf
+#     title: Troubleshoot BAR access limitation  
-            title: White paper
+#   - url: https://github.com/amd/rocm-examples
-  - file: conceptual/file-reorg.md
+#     title: ROCm examples
-    title: File structure (Linux FHS)
+#
-  - file: conceptual/gpu-isolation.md
+#
-    title: GPU isolation techniques
+# - caption: Conceptual
-  - file: conceptual/cmake-packages.rst
+#   entries:
-    title: Using CMake
+#   - file: conceptual/gpu-arch.md
-  - file: conceptual/ai-pytorch-inception.md
+#     title: GPU architecture overview
-    title: Inception v3 with PyTorch
+#     subtrees:
-
+#     - entries:
-- caption: Reference
+#       - file: conceptual/gpu-arch/mi300.md
-  entries:
+#         title: MI300 microarchitecture
-  - file: reference/api-libraries.md
+#         subtrees:
-    title: ROCm libraries
+#         - entries:
-  - file: reference/rocm-tools.md
+#           - url: https://www.amd.com/content/dam/amd/en/documents/instinct-tech-docs/instruction-set-architectures/amd-instinct-mi300-cdna3-instruction-set-architecture.pdf
-    title: ROCm tools, compilers, and runtimes
+#             title: AMD Instinct MI300/CDNA3 ISA
-  - file: reference/gpu-arch-specs.rst
+#           - url: https://www.amd.com/content/dam/amd/en/documents/instinct-tech-docs/white-papers/amd-cdna-3-white-paper.pdf
-  - file: reference/gpu-atomics-operation.rst
+#             title: White paper
-  - file: reference/precision-support.rst
+#           - file: conceptual/gpu-arch/mi300-mi200-performance-counters.rst
-    title: Precision support
+#             title: MI300 and MI200 Performance counter
-  - file: reference/graph-safe-support.rst
+#       - file: conceptual/gpu-arch/mi250.md
-    title: Graph safe support
+#         title: MI250 microarchitecture
-
+#         subtrees:
-- caption: Contribute
+#         - entries:
-  entries:
+#           - url: https://www.amd.com/system/files/TechDocs/instinct-mi200-cdna2-instruction-set-architecture.pdf
-  - file: contribute/contributing.md
+#             title: AMD Instinct MI200/CDNA2 ISA
-    title: Contributing to the ROCm documentation
+#           - url: https://www.amd.com/content/dam/amd/en/documents/instinct-business-docs/white-papers/amd-cdna2-white-paper.pdf
-    subtrees:
+#             title: White paper
-    - entries:
+#       - file: conceptual/gpu-arch/mi100.md
-      - file: contribute/toolchain.md
+#         title: MI100 microarchitecture
-        title: ROCm documentation toolchain
+#         subtrees:
-      - file: contribute/building.md
+#         - entries:
-  - file: contribute/feedback.md
+#           - url: https://www.amd.com/system/files/TechDocs/instinct-mi100-cdna1-shader-instruction-set-architecture%C2%A0.pdf
-    title: Providing feedback about the ROCm documentation
+#             title: AMD Instinct MI100/CDNA1 ISA
-  - file: about/license.md
+#           - url: https://www.amd.com/content/dam/amd/en/documents/instinct-business-docs/white-papers/amd-cdna-white-paper.pdf
-    title: ROCm licenses
+#             title: White paper
+#   - file: conceptual/file-reorg.md
+#     title: File structure (Linux FHS)
+#   - file: conceptual/gpu-isolation.md
+#     title: GPU isolation techniques
+#   - file: conceptual/cmake-packages.rst
+#     title: Using CMake
+#   - file: conceptual/ai-pytorch-inception.md
+#     title: Inception v3 with PyTorch
+#
+# - caption: Reference
+#   entries:
+#   - file: reference/api-libraries.md
+#     title: ROCm libraries
+#   - file: reference/rocm-tools.md
+#     title: ROCm tools, compilers, and runtimes
+#   - file: reference/gpu-arch-specs.rst
+#   - file: reference/gpu-atomics-operation.rst
+#   - file: reference/precision-support.rst
+#     title: Precision support
+#   - file: reference/graph-safe-support.rst
+#     title: Graph safe support
+#
+# - caption: Contribute
+#   entries:
+#   - file: contribute/contributing.md
+#     title: Contributing to the ROCm documentation
+#     subtrees:
+#     - entries:
+#       - file: contribute/toolchain.md
+#         title: ROCm documentation toolchain
+#       - file: contribute/building.md
+#   - file: contribute/feedback.md
+#     title: Providing feedback about the ROCm documentation
+#   - file: about/license.md
+#     title: ROCm licenses

docs/sphinx/requirements.in

@@ -1,4 +1,4 @@
-rocm-docs-core==1.20.1
 sphinx-reredirects
 sphinx-sitemap
 sphinxcontrib.datatemplates==0.11.0
+git+https://github.com/ROCm/rocm-docs-core.git@alexxu12/header-cap-space#egg=rocm-docs-core

docs/sphinx/requirements.txt

@@ -21,9 +21,11 @@ babel==2.17.0
     #   sphinx
 beautifulsoup4==4.13.4
     # via pydata-sphinx-theme
+blinker==1.9.0
+    # via flask
 breathe==4.36.0
     # via rocm-docs-core
-certifi==2025.4.26
+certifi==2025.6.15
     # via requests
 cffi==1.17.1
     # via
@@ -33,11 +35,12 @@ charset-normalizer==3.4.2
     # via requests
 click==8.2.1
     # via
+    #   flask
     #   jupyter-cache
     #   sphinx-external-toc
 comm==0.2.2
     # via ipykernel
-cryptography==45.0.3
+cryptography==45.0.4
     # via pyjwt
 debugpy==1.8.14
     # via ipykernel
@@ -60,6 +63,8 @@ fastjsonschema==2.21.1
     # via
     #   nbformat
     #   rocm-docs-core
+flask==3.1.1
+    # via sphinx-sitemap
 gitdb==4.0.12
     # via gitpython
 gitpython==3.1.44
@@ -80,10 +85,13 @@ ipython==8.37.0
     # via
     #   ipykernel
     #   myst-nb
+itsdangerous==2.2.0
+    # via flask
 jedi==0.19.2
     # via ipython
 jinja2==3.1.6
     # via
+    #   flask
     #   myst-parser
     #   sphinx
 jsonschema==4.24.0
@@ -107,7 +115,10 @@ markdown-it-py==3.0.0
     #   mdit-py-plugins
     #   myst-parser
 markupsafe==3.0.2
-    # via jinja2
+    # via
+    #   flask
+    #   jinja2
+    #   werkzeug
 matplotlib-inline==0.1.7
     # via
     #   ipykernel
@@ -134,7 +145,6 @@ nest-asyncio==1.6.0
 packaging==25.0
     # via
     #   ipykernel
-    #   pydata-sphinx-theme
     #   sphinx
 parso==0.8.4
     # via jedi
@@ -152,13 +162,13 @@ pure-eval==0.2.3
     # via stack-data
 pycparser==2.22
     # via cffi
-pydata-sphinx-theme==0.15.4
+pydata-sphinx-theme==0.16.1
     # via
     #   rocm-docs-core
     #   sphinx-book-theme
 pygithub==2.6.1
     # via rocm-docs-core
-pygments==2.19.1
+pygments==2.19.2
     # via
     #   accessible-pygments
     #   ipython
@@ -178,7 +188,7 @@ pyyaml==6.0.2
     #   rocm-docs-core
     #   sphinx-external-toc
     #   sphinxcontrib-datatemplates
-pyzmq==26.4.0
+pyzmq==27.0.0
     # via
     #   ipykernel
     #   jupyter-client
@@ -190,7 +200,8 @@ requests==2.32.4
     # via
     #   pygithub
     #   sphinx
-rocm-docs-core==1.20.1
+    #   sphinx-sitemap
+rocm-docs-core @ git+https://github.com/ROCm/rocm-docs-core.git@alexxu12/header-cap-space
     # via -r requirements.in
 rpds-py==0.25.1
     # via
@@ -215,12 +226,12 @@ sphinx==8.1.3
     #   sphinx-copybutton
     #   sphinx-design
     #   sphinx-external-toc
+    #   sphinx-last-updated-by-git
     #   sphinx-notfound-page
     #   sphinx-reredirects
-    #   sphinx-sitemap
     #   sphinxcontrib-datatemplates
     #   sphinxcontrib-runcmd
-sphinx-book-theme==1.1.4
+sphinx-book-theme==1.1.3
     # via rocm-docs-core
 sphinx-copybutton==0.5.2
     # via rocm-docs-core
@@ -228,11 +239,13 @@ sphinx-design==0.6.1
     # via rocm-docs-core
 sphinx-external-toc==1.0.1
     # via rocm-docs-core
+sphinx-last-updated-by-git==0.3.8
+    # via sphinx-sitemap
 sphinx-notfound-page==1.1.0
     # via rocm-docs-core
 sphinx-reredirects==0.1.6
     # via -r requirements.in
-sphinx-sitemap==2.6.0
+sphinx-sitemap==2.7.1
     # via -r requirements.in
 sphinxcontrib-applehelp==2.0.0
     # via sphinx
@@ -288,6 +301,8 @@ urllib3==2.5.0
     #   requests
 wcwidth==0.2.13
     # via prompt-toolkit
+werkzeug==3.1.3
+    # via flask
 wrapt==1.17.2
     # via deprecated
 zipp==3.23.0