Document for reenabling Gfx110X iGPU support

README.md

@@ -142,7 +142,7 @@ In case you don't have an existing ROCm/HIP installation from which you can run
 
 You can install the `rocm` Python package for any architecture inside a venv and run `offload-arch` from there:
 
-1. `python build_tools/setup_venv.py --index-name nightly --index-subdir gfx110X-dgpu --packages rocm .tmpvenv`
+1. `python build_tools/setup_venv.py --index-name nightly --index-subdir gfx110X-all --packages rocm .tmpvenv`
 1. `.tmpvenv/bin/offload-arch` on Linux, `.tmpvenv\Scripts\offload-arch` on Windows
 1. `rm -rf .tmpvenv`
 
@@ -223,7 +223,7 @@ See instructions in the next section for [Linux](#ccache-usage-on-linux) and [Wi
 Otherwise, ROCm/HIP can be configured and build with just the following commands:
 
 ```bash
-cmake -B build -GNinja . -DTHEROCK_AMDGPU_FAMILIES=gfx110X-dgpu
+cmake -B build -GNinja . -DTHEROCK_AMDGPU_FAMILIES=gfx110X-all
 cmake --build build
 ```
 
@@ -249,7 +249,7 @@ Example:
 # Any shell used to build must eval setup_ccache.py to set environment
 # variables.
 eval "$(./build_tools/setup_ccache.py)"
-cmake -B build -GNinja -DTHEROCK_AMDGPU_FAMILIES=gfx110X-dgpu \
+cmake -B build -GNinja -DTHEROCK_AMDGPU_FAMILIES=gfx110X-all \
   -DCMAKE_C_COMPILER_LAUNCHER=ccache \
   -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
   .

RELEASES.md

@@ -76,9 +76,10 @@ project layouts.**
 | ---------------------------------- | ---------- | ------------ | ------------------------------------------------------------------ |
 | MI300A/MI300X                      | gfx942     | gfx94X-dcgpu | [rocm](#rocm-for-gfx94X-dcgpu) // [torch](#torch-for-gfx94X-dcgpu) |
 | MI350X/MI355X                      | gfx950     | gfx950-dcgpu | [rocm](#rocm-for-gfx950-dcgpu) // [torch](#torch-for-gfx950-dcgpu) |
-| AMD RX 7900 XTX                    | gfx1100    | gfx110X-dgpu | [rocm](#rocm-for-gfx110X-dgpu) // [torch](#torch-for-gfx110X-dgpu) |
-| AMD RX 7800 XT                     | gfx1101    | gfx110X-dgpu | [rocm](#rocm-for-gfx110X-dgpu) // [torch](#torch-for-gfx110X-dgpu) |
-| AMD RX 7700S / Framework Laptop 16 | gfx1102    | gfx110X-dgpu | [rocm](#rocm-for-gfx110X-dgpu) // [torch](#torch-for-gfx110X-dgpu) |
+| AMD RX 7900 XTX                    | gfx1100    | gfx110X-all  | [rocm](#rocm-for-gfx110X-all) // [torch](#torch-for-gfx110X-all)   |
+| AMD RX 7800 XT                     | gfx1101    | gfx110X-all  | [rocm](#rocm-for-gfx110X-all) // [torch](#torch-for-gfx110X-all)   |
+| AMD RX 7700S / Framework Laptop 16 | gfx1102    | gfx110X-all  | [rocm](#rocm-for-gfx110X-all) // [torch](#torch-for-gfx110X-all)   |
+| AMD Radeon 780M Laptop iGPU        | gfx1103    | gfx110X-all  | [rocm](#rocm-for-gfx110X-all) // [torch](#torch-for-gfx110X-all)   |
 | AMD Strix Halo iGPU                | gfx1151    | gfx1151      | [rocm](#rocm-for-gfx1151) // [torch](#torch-for-gfx1151)           |
 | AMD RX 9060 / XT                   | gfx1200    | gfx120X-all  | [rocm](#rocm-for-gfx120X-all) // [torch](#torch-for-gfx120X-all)   |
 | AMD RX 9070 / XT                   | gfx1201    | gfx120X-all  | [rocm](#rocm-for-gfx120X-all) // [torch](#torch-for-gfx120X-all)   |
@@ -132,7 +133,7 @@ python -m pip install \
   "rocm[libraries,devel]"
 ```
 
-#### rocm for gfx110X-dgpu
+#### rocm for gfx110X-all
 
 Supported devices in this family:
 
@@ -141,12 +142,13 @@ Supported devices in this family:
 | AMD RX 7900 XTX                    | gfx1100    |
 | AMD RX 7800 XT                     | gfx1101    |
 | AMD RX 7700S / Framework Laptop 16 | gfx1102    |
+| AMD Radeon 780M Laptop iGPU        | gfx1103    |
 
 Install instructions:
 
 ```bash
 python -m pip install \
-  --index-url https://rocm.nightlies.amd.com/v2/gfx110X-dgpu/ \
+  --index-url https://rocm.nightlies.amd.com/v2/gfx110X-all/ \
   "rocm[libraries,devel]"
 ```
 
@@ -193,7 +195,7 @@ pip freeze | grep rocm
 # rocm==6.5.0rc20250610
 # rocm-sdk-core==6.5.0rc20250610
 # rocm-sdk-devel==6.5.0rc20250610
-# rocm-sdk-libraries-gfx110X-dgpu==6.5.0rc20250610
+# rocm-sdk-libraries-gfx110X-all==6.5.0rc20250610
 ```
 
 You should also see various tools on your `PATH` and in the `bin` directory:
@@ -312,7 +314,7 @@ python -m pip install \
   --pre torch torchaudio torchvision
 ```
 
-#### torch for gfx110X-dgpu
+#### torch for gfx110X-all
 
 Supported devices in this family:
 
@@ -321,10 +323,11 @@ Supported devices in this family:
 | AMD RX 7900 XTX                    | gfx1100    |
 | AMD RX 7800 XT                     | gfx1101    |
 | AMD RX 7700S / Framework Laptop 16 | gfx1102    |
+| AMD Radeon 780M Laptop iGPU        | gfx1103    |
 
 ```bash
 python -m pip install \
-  --index-url https://rocm.nightlies.amd.com/v2/gfx110X-dgpu/ \
+  --index-url https://rocm.nightlies.amd.com/v2/gfx110X-all/ \
   --pre torch torchaudio torchvision
 ```
 
@@ -396,7 +399,7 @@ After downloading, simply extract the release tarball into place:
 ```bash
 mkdir therock-tarball && cd therock-tarball
 # For example...
-wget https://therock-nightly-tarball.s3.us-east-2.amazonaws.com/therock-dist-linux-gfx110X-dgpu-6.5.0rc20250610.tar.gz
+wget https://therock-nightly-tarball.s3.us-east-2.amazonaws.com/therock-dist-linux-gfx110X-all-6.5.0rc20250610.tar.gz
 
 mkdir install
 tar -xf *.tar.gz -C install
@@ -473,13 +476,13 @@ Examples:
 - Downloads the version `6.4.0rc20250516` gfx110X artifacts from GitHub release tag `nightly-tarball` to the specified output directory `build`:
 
   ```bash
-  python build_tools/install_rocm_from_artifacts.py --release 6.4.0rc20250516 --amdgpu-family gfx110X-dgpu --output-dir build
+  python build_tools/install_rocm_from_artifacts.py --release 6.4.0rc20250516 --amdgpu-family gfx110X-all --output-dir build
   ```
 
 - Downloads the version `6.4.0.dev0+e015c807437eaf32dac6c14a9c4f752770c51b14` gfx110X artifacts from GitHub release tag `dev-tarball` to the default output directory `therock-build`:
 
   ```bash
-  python build_tools/install_rocm_from_artifacts.py --release 6.4.0.dev0+e015c807437eaf32dac6c14a9c4f752770c51b14 --amdgpu-family gfx110X-dgpu
+  python build_tools/install_rocm_from_artifacts.py --release 6.4.0.dev0+e015c807437eaf32dac6c14a9c4f752770c51b14 --amdgpu-family gfx110X-all
   ```
 
 Select your AMD GPU family from this file [therock_amdgpu_targets.cmake](https://github.com/ROCm/TheRock/blob/59c324a759e8ccdfe5a56e0ebe72a13ffbc04c1f/cmake/therock_amdgpu_targets.cmake#L44-L81)

ROADMAP.md

@@ -38,6 +38,7 @@ See also the [ROCm Device Support Wishlist GitHub Discussion](https://github.com
 | **RDNA4**    | **gfx1200** | ✅            | ✅            | ✅            |
 | **RDNA3.5**  | **gfx1151** | ✅            | ✅            |               |
 | **RDNA3.5**  | **gfx1150** | ✅            | ✅            |               |
+| **RDNA3**    | **gfx1103** | ✅            | ✅            |               |
 | **RDNA3**    | **gfx1102** | ✅            | ✅            |               |
 | **RDNA3**    | **gfx1101** | ✅            | ✅            |               |
 | **RDNA3**    | **gfx1100** | ✅            | ✅            |               |
@@ -62,6 +63,7 @@ Check [windows_support.md](docs/development/windows_support.md) on current statu
 | **RDNA4**    | **gfx1200** | ✅            |               |               |
 | **RDNA3.5**  | **gfx1151** | ✅            | ✅            | ✅            |
 | **RDNA3.5**  | **gfx1150** | ✅            |               |               |
+| **RDNA3**    | **gfx1103** | ✅            |               |               |
 | **RDNA3**    | **gfx1102** | ✅            |               |               |
 | **RDNA3**    | **gfx1101** | ✅            |               |               |
 | **RDNA3**    | **gfx1100** | ✅            |               |               |

docs/development/development_guide.md

@@ -315,7 +315,7 @@ Settings for CMake builds can be specified in `.vscode/settings.json` or a
     "-DCMAKE_C_COMPILER_LAUNCHER=ccache",
     "-DCMAKE_CXX_COMPILER_LAUNCHER=ccache",
     "-DPython3_EXECUTABLE=${workspaceFolder}/.venv/Scripts/python",
-    "-DTHEROCK_AMDGPU_FAMILIES=gfx110X-dgpu",  // Set to your GPU target family.
+    "-DTHEROCK_AMDGPU_FAMILIES=gfx110X-all",  // Set to your GPU target family.
     //
     // You can include both option settings for easy toggling.
     "-DBUILD_TESTING=ON",

docs/development/github_actions_debugging.md

@@ -93,7 +93,7 @@ cd _work/TheRock/TheRock
 export TEATIME_FORCE_INTERACTIVE=1
 
 # Copy the configure command from the "Configure Projects" step
-cmake -B "B:/build" -GNinja . -DTHEROCK_AMDGPU_FAMILIES=gfx110X-dgpu -DCMAKE_C_COMPILER_LAUNCHER=ccache -DCMAKE_CXX_COMPILER_LAUNCHER=ccache -DTHEROCK_VERBOSE=ON -DBUILD_TESTING=ON -DCMAKE_C_COMPILER="C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Tools/MSVC/14.44.35207/bin/Hostx64/x64/cl.exe" -DCMAKE_CXX_COMPILER="C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Tools/MSVC/14.44.35207/bin/Hostx64/x64/cl.exe" -DCMAKE_LINKER="C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Tools/MSVC/14.44.35207/bin/Hostx64/x64/link.exe" -DTHEROCK_BACKGROUND_BUILD_JOBS=4
+cmake -B "B:/build" -GNinja . -DTHEROCK_AMDGPU_FAMILIES=gfx110X-all -DCMAKE_C_COMPILER_LAUNCHER=ccache -DCMAKE_CXX_COMPILER_LAUNCHER=ccache -DTHEROCK_VERBOSE=ON -DBUILD_TESTING=ON -DCMAKE_C_COMPILER="C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Tools/MSVC/14.44.35207/bin/Hostx64/x64/cl.exe" -DCMAKE_CXX_COMPILER="C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Tools/MSVC/14.44.35207/bin/Hostx64/x64/cl.exe" -DCMAKE_LINKER="C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Tools/MSVC/14.44.35207/bin/Hostx64/x64/link.exe" -DTHEROCK_BACKGROUND_BUILD_JOBS=4
 
 # Build CMake targets
 # You could also run buildctl.py here to enable/disable specific subprojects

docs/development/installing_artifacts.md

@@ -48,14 +48,14 @@ Nightly tarballs are built daily and follow the naming pattern: `MAJOR.MINOR.PAT
 
 1. Visit the [nightly tarball S3 bucket](https://therock-nightly-tarball.s3.amazonaws.com/)
 1. Look for files matching your GPU family. Files are named: `therock-dist-linux-{GPU_FAMILY}-{VERSION}.tar.gz`
-   - Example: `therock-dist-linux-gfx110X-dgpu-6.4.0rc20250514.tar.gz`
+   - Example: `therock-dist-linux-gfx110X-all-6.4.0rc20250514.tar.gz`
 1. Extract the version from the filename (the part after the last hyphen, before `.tar.gz`)
    - In the example above, the version is: `6.4.0rc20250514`
 1. Use this version string with `--release`:
    ```bash
    python build_tools/install_rocm_from_artifacts.py \
        --release 6.4.0rc20250514 \
-       --amdgpu-family gfx110X-dgpu
+       --amdgpu-family gfx110X-all
    ```
 
 **Version format:** `X.Y.ZrcYYYYMMDD`
@@ -99,7 +99,7 @@ Dev tarballs are built from specific commits and follow the naming pattern: `MAJ
 ```bash
 python build_tools/install_rocm_from_artifacts.py \
     --run-id 15575624591 \
-    --amdgpu-family gfx110X-dgpu \
+    --amdgpu-family gfx110X-all \
     --blas --tests
 ```
 
@@ -214,7 +214,7 @@ Test that artifacts can be fetched with your new flag:
 # Test with a CI run
 python build_tools/install_rocm_from_artifacts.py \
     --run-id YOUR_RUN_ID \
-    --amdgpu-family gfx110X-dgpu \
+    --amdgpu-family gfx110X-all \
     --newcomponent --tests
 ```
 

docs/development/test_environment_reproduction.md

@@ -28,7 +28,7 @@ $ python build_tools/github_actions/test_executable_scripts/test_rocblas.py
 `install_rocm_from_artifacts.py` parameters
 
 - CI_RUN_ID is sourced from the CI run (ex: https://github.com/ROCm/TheRock/actions/runs/16948046392 -> CI_RUN_ID = 16948046392)
-- GPU_FAMILY is the LLVM target name (ex: gfx94X-dcgpu, gfx1151, gfx110X-dgpu)
+- GPU_FAMILY is the LLVM target name (ex: gfx94X-dcgpu, gfx1151, gfx110X-all)
 - GITHUB_REPO is the GitHub repository that this CI run was executed. (ex: ROCm/rocm-libraries, ROCm/rccl)
 
 To view which python test wrappers we have, please checkout [`test_executable_scripts/`](https://github.com/ROCm/TheRock/tree/main/build_tools/github_actions/test_executable_scripts)

docs/development/windows_support.md

@@ -235,7 +235,7 @@ the [instructions in the root README](../../README.md#configuration) for other
 options you may want to set.
 
 ```bash
-cmake -B build -GNinja . -DTHEROCK_AMDGPU_FAMILIES=gfx110X-dgpu
+cmake -B build -GNinja . -DTHEROCK_AMDGPU_FAMILIES=gfx110X-all
 
 # If iterating and wishing to cache, add these:
 #  -DCMAKE_C_COMPILER_LAUNCHER=ccache \

external-builds/pytorch/README.md

@@ -156,15 +156,15 @@ mix/match build steps.
 
   ```bash
   python build_prod_wheels.py build \
-    --install-rocm --index-url https://rocm.nightlies.amd.com/v2/gfx110X-dgpu/ \
+    --install-rocm --index-url https://rocm.nightlies.amd.com/v2/gfx110X-all/ \
     --output-dir $HOME/tmp/pyout
   ```
 
 - On Windows:
 
   ```batch
   python build_prod_wheels.py build ^
-    --install-rocm --index-url https://rocm.nightlies.amd.com/v2/gfx110X-dgpu/ ^
+    --install-rocm --index-url https://rocm.nightlies.amd.com/v2/gfx110X-all/ ^
     --pytorch-dir C:/b/pytorch ^
     --pytorch-audio-dir C:/b/audio ^
     --pytorch-vision-dir C:/b/vision ^
@@ -222,7 +222,7 @@ The `rocm[libraries,devel]` packages can be installed in multiple ways:
 
   ```bash
   build_prod_wheels.py
-      --index-url https://rocm.nightlies.amd.com/v2/gfx110X-dgpu/ \
+      --index-url https://rocm.nightlies.amd.com/v2/gfx110X-all/ \
       install-rocm
   ```
 
@@ -231,12 +231,12 @@ The `rocm[libraries,devel]` packages can be installed in multiple ways:
   ```bash
   # From therock-nightly-python
   python -m pip install \
-    --index-url https://rocm.nightlies.amd.com/v2/gfx110X-dgpu/ \
+    --index-url https://rocm.nightlies.amd.com/v2/gfx110X-all/ \
     rocm[libraries,devel]
 
   # OR from therock-dev-python
   python -m pip install \
-    --index-url https://rocm.devreleases.amd.com/v2/gfx110X-dgpu/ \
+    --index-url https://rocm.devreleases.amd.com/v2/gfx110X-all/ \
     rocm[libraries,devel]
   ```
 
@@ -250,7 +250,7 @@ The `rocm[libraries,devel]` packages can be installed in multiple ways:
   mkdir $HOME/.therock/17123441166/artifacts
   python ./build_tools/fetch_artifacts.py \
     --run-id=17123441166 \
-    --target=gfx110X-dgpu \
+    --target=gfx110X-all \
     --output-dir=$HOME/.therock/17123441166/artifacts
 
   python ./build_tools/build_python_packages.py \