github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-11-17 04:27:31 +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

25
op_builder/fused_optim.py
View File

@@ -1,5 +1,3 @@
import os
from .builder import Builder
from .utils import get_cuda_cc_flag
@@ -10,25 +8,30 @@ class FusedOptimBuilder(Builder):
def __init__(self):
super().__init__(name=FusedOptimBuilder.NAME, prebuilt_import_path=FusedOptimBuilder.PREBUILT_IMPORT_PATH)
def sources_files(self):
ret = [
self.csrc_abs_path(fname) for fname in [
'colossal_C_frontend.cpp', 'multi_tensor_sgd_kernel.cu', 'multi_tensor_scale_kernel.cu',
'multi_tensor_adam.cu', 'multi_tensor_l2norm_kernel.cu', 'multi_tensor_lamb.cu'
self.csrc_abs_path(fname)
for fname in [
"colossal_C_frontend.cpp",
"multi_tensor_sgd_kernel.cu",
"multi_tensor_scale_kernel.cu",
"multi_tensor_adam.cu",
"multi_tensor_l2norm_kernel.cu",
"multi_tensor_lamb.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):
version_dependent_macros = ['-DVERSION_GE_1_1', '-DVERSION_GE_1_3', '-DVERSION_GE_1_5']
return ['-O3'] + version_dependent_macros
version_dependent_macros = ["-DVERSION_GE_1_1", "-DVERSION_GE_1_3", "-DVERSION_GE_1_5"]
return ["-O3"] + version_dependent_macros
def nvcc_flags(self):
extra_cuda_flags = ['-lineinfo']
extra_cuda_flags = ["-lineinfo"]
extra_cuda_flags.extend(get_cuda_cc_flag())
return ['-O3', '--use_fast_math'] + extra_cuda_flags
return ["-O3", "--use_fast_math"] + extra_cuda_flags