Files
FreeToken/python/freetoken/kernel/__main__.py
0d652e73a4 feat(rocm): add RDNA3 and RDNA4 runtime foundation (#132)
* feat: add simple ROCm GPU support for RDNA3 (gfx1100-1103)

- Add hip_compat.h shim mapping CUDA runtime API to HIP equivalents
- Update pinned_tensor.cpp to compile under both nvcc and hipcc
- Add ROCm detection in arch.py (is_rocm, get_rocm_gfx_arch, is_gfx11xx_family)
- Guard NVIDIA arch checks to return None on ROCm
- Skip nvcc version check in _toolchain.py when on ROCm
- Add ROCm build path in setup.py (ROCM_HOME, amdhip64, --offload-arch)
- Add _hip_cflags() in kernel/utils.py for JIT compilation on ROCm
- Add is_rocm() and driver_hip_version() in backend.py
- Add rocm-smi fallback in __main__.py for clangd generation
- Add TODO(ROCm) for NCCL->RCCL, flashinfer/sgl_kernel ROCm builds,
  Triton autotune RDNA3 tuning, PDL equivalent, hiprtc JIT cache
- Add AMD ROCm classifier in pyproject.toml

* fix(rocm): fall back for single-bank fast index copy

* test(rocm): cover fast index copy fallback

* test(rocm): use explicit fast copy module import

* fix(rocm): complete RDNA3 runtime path

* fix(rocm): preserve CPU MoE implementation

* build(rocm): support RDNA4 discovery and modular SDKs

* fix(rocm): complete modular SDK runtime support

* test(rocm): harden foundation platform contracts

* fix(rocm): pin jit target and split install guide

* perf(kernel): restore ROCm non-temporal index copy

---------

Co-authored-by: bouclem <bouclement2012@gmail.com>
Co-authored-by: Yuu <206304251+nekomario28@users.noreply.github.com>
Co-authored-by: LZ-QWQ <lz1830837889@163.com>
2026-09-25 17:49:45 -07:00

53 lines
1.8 KiB
Python

assert __name__ == "__main__"
def generate_clangd():
import os
import subprocess
from freetoken.kernel.utils import DEFAULT_INCLUDE
from freetoken.utils import get_rocm_gfx_arch, init_logger, is_rocm
from tvm_ffi.libinfo import find_dlpack_include_path, find_include_path
logger = init_logger(__name__)
logger.info("Generating .clangd file...")
include_paths = [find_include_path(), find_dlpack_include_path()] + DEFAULT_INCLUDE
# TODO(ROCm): hiprtc JIT cache should be separate from nvcc JIT cache to avoid stale binaries.
if is_rocm():
arch_flags = ["-xhip", f"--offload-arch={get_rocm_gfx_arch() or 'gfx1201'}"]
else:
try:
status = subprocess.run(
args=["nvidia-smi", "--query-gpu=compute_cap", "--format=csv,noheader"],
capture_output=True,
check=True,
)
compute_cap = status.stdout.decode("utf-8").strip().split("\n")[0]
major, minor = compute_cap.split(".")
except (subprocess.CalledProcessError, FileNotFoundError, ValueError):
import torch
major, minor = torch.cuda.get_device_capability()
arch_flags = ["-xcuda", f"--cuda-gpu-arch=sm_{major}{minor}"]
compile_flags = ",\n ".join(
arch_flags + ["-std=c++20", "-Wall", "-Wextra"]
+ [f"-isystem{path}" for path in include_paths]
)
clangd_content = f"""
CompileFlags:
Add: [
{compile_flags}
]
"""
if os.path.exists(".clangd"):
logger.warning(".clangd file already exists, nothing done.")
logger.warning(f"suggested content: {clangd_content}")
else:
with open(".clangd", "w") as f:
f.write(clangd_content)
logger.info(".clangd file generated.")
generate_clangd()