File: record_function_ops.h

package info (click to toggle)
pytorch-cuda 2.6.0%2Bdfsg-7
  • links: PTS, VCS
  • area: contrib
  • in suites: forky, sid, trixie
  • size: 161,620 kB
  • sloc: python: 1,278,832; cpp: 900,322; ansic: 82,710; asm: 7,754; java: 3,363; sh: 2,811; javascript: 2,443; makefile: 597; ruby: 195; xml: 84; objc: 68
file content (27 lines) | stat: -rw-r--r-- 935 bytes parent folder | download | duplicates (3)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
#pragma once
#include <ATen/record_function.h>
#include <torch/custom_class.h>
#include <optional>

namespace torch::autograd::profiler {

struct PythonRecordFunction : public torch::CustomClassHolder {
  at::RecordFunction record;

  explicit PythonRecordFunction(
      at::RecordScope scope = at::RecordScope::FUNCTION)
      : record(scope) {}
};

// Creates a new profiling scope using RecordFunction and invokes its starting
// callbacks.
TORCH_API c10::intrusive_ptr<PythonRecordFunction> record_function_enter_new(
    const std::string& name,
    const std::optional<std::string>& args = std::nullopt);

// Schedules RecordFunction's end callbacks to be run on completion of a future.
TORCH_API c10::intrusive_ptr<c10::ivalue::Future> _call_end_callbacks_on_fut_new(
    const c10::intrusive_ptr<PythonRecordFunction>& record,
    const c10::intrusive_ptr<c10::ivalue::Future>& fut);

} // namespace torch::autograd::profiler