Merge branch 'worktree-gfx1151-torch-docs'

Add gfx1151 Strix Halo PyTorch install notes to QUICKSTART.

Co-authored-by: Cursor <cursoragent@cursor.com>

QUICKSTART.md

@@ -132,7 +132,8 @@ Install **one** torch line into the same env as `meshnet-node`:
 |----------|---------|
 | NVIDIA CUDA | `pip install torch` (default index) |
 | CPU only | `pip install torch --index-url https://download.pytorch.org/whl/cpu` |
-| AMD ROCm | `pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm6.3` |
+| AMD ROCm (discrete, arch in official wheels) | `pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm6.3` |
+| AMD Strix Halo / Ryzen AI Max (gfx1151) | `pip install torch torchvision torchaudio --index-url https://rocm.nightlies.amd.com/v2/gfx1151/` |
 
 On Windows `.venv`, prefix with `.\.venv\Scripts\pip.exe`. Conda users with CUDA
 torch already installed can skip this step.
@@ -181,7 +182,21 @@ python -m pip install -e packages/tracker -e packages/node -e packages/p2p -e pa
 python -m pip install "transformers>=5.12" accelerate safetensors
 python -m pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm6.3
 ```
- <!-- .venv-rocm/bin/pip install --index-url https://rocm.nightlies.amd.com/v2/gfx1151/ --upgrade  -->
+
+**Strix Halo / Ryzen AI Max APUs (gfx1151, e.g. Radeon 8060S):** the official
+`download.pytorch.org` ROCm wheels do NOT ship gfx1151 kernels — every GPU op
+fails with `HIP error: invalid device function` and `HSA_OVERRIDE_GFX_VERSION`
+does not help. Install AMD's gfx1151-native builds instead (TheRock nightlies,
+self-contained, no system ROCm required):
+
+```bash
+python -m pip install torch torchvision torchaudio --index-url https://rocm.nightlies.amd.com/v2/gfx1151/
+```
+
+Check that your arch is actually in the wheel:
+`python -c "import torch; print(torch.cuda.get_arch_list())"` must list your
+GPU's `gcnArchName` (from `torch.cuda.get_device_properties(0)`).
+
 Keep this separate from a known-good CPU `.venv` until ROCm is verified on that
 machine. ROCm wheels are large and host-runtime-sensitive; a failed ROCm install
 should not break the CPU fallback environment.
@@ -241,6 +256,11 @@ HF_HOME=/path/to/models .venv-rocm/bin/meshnet-node start \
   or `libroctx64.so` are host ROCm runtime problems, not meshnet-node problems.
 - Some AMD APUs and consumer GPUs require newer ROCm/Radeon support than server
   Instinct cards. Check AMD's ROCm Radeon/Ryzen support matrix for the exact model.
+- `HIP error: invalid device function` (or `no kernel image is available`) on a
+  working driver means the installed torch wheel has no kernels compiled for
+  your GPU arch. Compare `torch.cuda.get_device_properties(0).gcnArchName`
+  against `torch.cuda.get_arch_list()`; if your arch is missing, install a wheel
+  built for it (see the Strix Halo/gfx1151 note above).
 
 ### Qwen3.5/3.6-MoE notes