yezhengmao/nccl
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[build] Allow setting CXXFLAGS on the command line

Browse Source
This commit is contained in:
Luke Yeager 2020-01-07 13:29:42 -08:00
parent 3899f6e0f2
commit c7ba70ff90
1 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
makefiles/common.mk
View File

@ -42,9 +42,10 @@ else
endif
#$(info NVCC_GENCODE is ${NVCC_GENCODE})
CXXFLAGS := -DCUDA_MAJOR=$(CUDA_MAJOR) -DCUDA_MINOR=$(CUDA_MINOR) -fPIC -fvisibility=hidden
CXXFLAGS += -Wall -Wno-unused-function -Wno-sign-compare -std=c++11 -Wvla
CXXFLAGS += -I $(CUDA_INC)
CXXFLAGS := -DCUDA_MAJOR=$(CUDA_MAJOR) -DCUDA_MINOR=$(CUDA_MINOR) -fPIC -fvisibility=hidden \
-Wall -Wno-unused-function -Wno-sign-compare -std=c++11 -Wvla \
-I $(CUDA_INC) \
$(CXXFLAGS)
# Maxrregcount needs to be set accordingly to NCCL_MAX_NTHREADS (otherwise it will cause kernel launch errors)
# 512 : 120, 640 : 96, 768 : 80, 1024 : 60
# We would not have to set this if we used __launch_bounds__, but this only works on kernels, not on functions.