apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2026-04-23 00:17:25 +08:00
Code Issues Actions 21 Packages Projects Releases Wiki Activity

[Iluvatar] Support wi4a16 group_gemm (#7078)

Browse Source
This commit is contained in:
yzwu
2026-03-30 19:03:51 +08:00
committed by GitHub
parent 18062c55bb
commit 8789329457
13 changed files with 722 additions and 144 deletions
Download Patch File Download Diff File Expand all files Collapse all files
custom_ops/setup_ops.py
+5 -4
View File
@@ -584,14 +584,13 @@ elif paddle.is_compiled_with_cuda():
elif paddle.is_compiled_with_xpu():
assert False, "For XPU, please use setup_ops.py in the xpu_ops directory to compile custom ops."
elif paddle.is_compiled_with_custom_device("iluvatar_gpu"):
_iluvatar_clang_cuda_flags = ["-Wno-non-pod-varargs", "-DPADDLE_DEV", "-DPADDLE_WITH_CUSTOM_DEVICE"]
setup(
name="fastdeploy_ops",
ext_modules=CUDAExtension(
extra_compile_args={
"nvcc": [
"-DPADDLE_DEV",
"-DPADDLE_WITH_CUSTOM_DEVICE",
]
"cxx": _iluvatar_clang_cuda_flags,
"nvcc": _iluvatar_clang_cuda_flags,
},
sources=[
"gpu_ops/save_with_output_msg.cc",
@@ -625,6 +624,8 @@ elif paddle.is_compiled_with_custom_device("iluvatar_gpu"):
"iluvatar_ops/mixed_fused_attn.cu",
"iluvatar_ops/w8a16_group_gemm.cu",
"iluvatar_ops/w8a16_group_gemv.cu",
"iluvatar_ops/wi4a16_group_gemm.cu",
"iluvatar_ops/wi4a16_weight_quantize.cu",
"iluvatar_ops/restore_tokens_per_expert.cu",
"iluvatar_ops/runtime/iluvatar_context.cc",
"iluvatar_ops/cpp_extensions.cc",