Commit 25d5540b authored by aiss's avatar aiss
Browse files

add block_thread arg

parent 1c0bde25
...@@ -29,7 +29,8 @@ class FusedAdamBuilder(CUDAOpBuilder): ...@@ -29,7 +29,8 @@ class FusedAdamBuilder(CUDAOpBuilder):
return args + self.version_dependent_macros() return args + self.version_dependent_macros()
def nvcc_args(self): def nvcc_args(self):
nvcc_flags = ['-O3'] + self.version_dependent_macros() nvcc_flags = super().nvcc_args()
nvcc_flags += ['-O3'] + self.version_dependent_macros()
if not self.is_rocm_pytorch(): if not self.is_rocm_pytorch():
nvcc_flags.extend( nvcc_flags.extend(
['-allow-unsupported-compiler' if sys.platform == "win32" else '', '-lineinfo', '--use_fast_math'] + ['-allow-unsupported-compiler' if sys.platform == "win32" else '', '-lineinfo', '--use_fast_math'] +
......
...@@ -29,7 +29,8 @@ class FusedLambBuilder(CUDAOpBuilder): ...@@ -29,7 +29,8 @@ class FusedLambBuilder(CUDAOpBuilder):
return args + self.version_dependent_macros() return args + self.version_dependent_macros()
def nvcc_args(self): def nvcc_args(self):
nvcc_flags = ['-O3'] + self.version_dependent_macros() nvcc_flags = super().nvcc_args()
nvcc_flags += ['-O3'] + self.version_dependent_macros()
if self.is_rocm_pytorch(): if self.is_rocm_pytorch():
ROCM_MAJOR, ROCM_MINOR = self.installed_rocm_version() ROCM_MAJOR, ROCM_MINOR = self.installed_rocm_version()
nvcc_flags += ['-DROCM_VERSION_MAJOR=%s' % ROCM_MAJOR, '-DROCM_VERSION_MINOR=%s' % ROCM_MINOR] nvcc_flags += ['-DROCM_VERSION_MAJOR=%s' % ROCM_MAJOR, '-DROCM_VERSION_MINOR=%s' % ROCM_MINOR]
......
...@@ -2,4 +2,4 @@ ...@@ -2,4 +2,4 @@
export LIBRARY_PATH=/usr/local/lib:$LIBRARY_PATH export LIBRARY_PATH=/usr/local/lib:$LIBRARY_PATH
#export DS_BUILD_VERSION=dtk22.10.1 #export DS_BUILD_VERSION=dtk22.10.1
export LD_LIBRARY_PATH=/usr/local/lib/python3.7/site-packages/torch/lib:$LD_LIBRARY_PATH export LD_LIBRARY_PATH=/usr/local/lib/python3.7/site-packages/torch/lib:$LD_LIBRARY_PATH
DS_BUILD_STRING=.dtk22.10.1.torch1.10 DS_BUILD_RANDOM_LTD=0 DS_BUILD_QUANTIZER=0 DS_BUILD_TRANSFORMER_INFERENCE=0 DS_BUILD_OPS=1 verbose=1 CXX=hipcc CC=hipcc python3 setup.py install bdist_wheel DS_BUILD_RANDOM_LTD=0 DS_BUILD_QUANTIZER=0 DS_BUILD_TRANSFORMER_INFERENCE=0 DS_BUILD_OPS=1 verbose=1 CXX=hipcc CC=hipcc python3 setup.py install bdist_wheel
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment