pytorch/torch/cuda/nvtx.py
Peter Bell 5fc5cf6571 Stop using ctypes to interface with CUDA libraries. (#33678)
Summary:
Fixes https://github.com/pytorch/pytorch/issues/33016, Continuation of https://github.com/pytorch/pytorch/issues/31160
Pull Request resolved: https://github.com/pytorch/pytorch/pull/33678

Differential Revision: D20249187

Pulled By: ezyang

fbshipit-source-id: 172ce4a0fee7fbe01436a421d1af22ef6173b6ed
2020-03-11 07:22:46 -07:00

45 lines
1.0 KiB
Python

try:
from torch._C import _nvtx
except ImportError:
class _NVTXStub(object):
@staticmethod
def _fail(*args, **kwargs):
raise RuntimeError("NVTX functions not installed. Are you sure you have a CUDA build?")
rangePushA = _fail
rangePop = _fail
markA = _fail
_nvtx = _NVTXStub()
__all__ = ['range_push', 'range_pop', 'mark']
def range_push(msg):
"""
Pushes a range onto a stack of nested range span. Returns zero-based
depth of the range that is started.
Arguments:
msg (string): ASCII message to associate with range
"""
return _nvtx.rangePushA(msg)
def range_pop():
"""
Pops a range off of a stack of nested range spans. Returns the
zero-based depth of the range that is ended.
"""
return _nvtx.rangePop()
def mark(msg):
"""
Describe an instantaneous event that occurred at some point.
Arguments:
msg (string): ASCII message to associate with the event.
"""
return _nvtx.markA(msg)