github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-11-18 14:46:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

[misc] update pre-commit and run all files (#4752)

Browse Source 
* [misc] update pre-commit

* [misc] run pre-commit

* [misc] remove useless configuration files

* [misc] ignore cuda for clang-format
This commit is contained in:
Hongxin Liu
2023-09-19 14:20:26 +08:00
committed by GitHub
parent 3c6b831c26
commit 079bf3cb26
1268 changed files with 50037 additions and 38444 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
op_builder/layernorm.py
View File

@@ -1,5 +1,3 @@
import os
from .builder import Builder
from .utils import append_nvcc_threads, get_cuda_cc_flag
@@ -12,18 +10,18 @@ class LayerNormBuilder(Builder):
super().__init__(name=LayerNormBuilder.NAME, prebuilt_import_path=LayerNormBuilder.PREBUILT_IMPORT_PATH)
def sources_files(self):
ret = [self.csrc_abs_path(fname) for fname in ['layer_norm_cuda.cpp', 'layer_norm_cuda_kernel.cu']]
ret = [self.csrc_abs_path(fname) for fname in ["layer_norm_cuda.cpp", "layer_norm_cuda_kernel.cu"]]
return ret
def include_dirs(self):
ret = [self.csrc_abs_path('kernels/include'), self.get_cuda_home_include()]
ret = [self.csrc_abs_path("kernels/include"), self.get_cuda_home_include()]
return ret
def cxx_flags(self):
return ['-O3'] + self.version_dependent_macros
return ["-O3"] + self.version_dependent_macros
def nvcc_flags(self):
extra_cuda_flags = ['-maxrregcount=50']
extra_cuda_flags = ["-maxrregcount=50"]
extra_cuda_flags.extend(get_cuda_cc_flag())
ret = ['-O3', '--use_fast_math'] + extra_cuda_flags + self.version_dependent_macros
ret = ["-O3", "--use_fast_math"] + extra_cuda_flags + self.version_dependent_macros
return append_nvcc_threads(ret)