profiler.py

import asyncio
import time
from functools import wraps

import torch
import torch.distributed as dist
from loguru import logger

from lightx2v.utils.envs import *


class _ProfilingContext:
    def __init__(self, name):
        self.name = name
        self.rank_info = ""
        if dist.is_initialized():
            self.rank_info = f"Rank {dist.get_rank()}"
        else:
            self.rank_info = "Single GPU"

    def __enter__(self):
        torch.cuda.synchronize()
        if torch.cuda.is_available():
            torch.cuda.reset_peak_memory_stats()
        self.start_time = time.perf_counter()
        return self

    def __exit__(self, exc_type, exc_val, exc_tb):
        torch.cuda.synchronize()
        if torch.cuda.is_available():
            peak_memory = torch.cuda.max_memory_allocated() / (1024**3)  # 转换为GB
            logger.info(f"[Profile] {self.rank_info} - {self.name} Peak Memory: {peak_memory:.2f} GB")
        else:
            logger.info(f"[Profile] {self.rank_info} - {self.name} executed without GPU.")
        elapsed = time.perf_counter() - self.start_time
        logger.info(f"[Profile] {self.rank_info} - {self.name} cost {elapsed:.6f} seconds")
        return False

    async def __aenter__(self):
        torch.cuda.synchronize()
        if torch.cuda.is_available():
            torch.cuda.reset_peak_memory_stats()
        self.start_time = time.perf_counter()
        return self

    async def __aexit__(self, exc_type, exc_val, exc_tb):
        torch.cuda.synchronize()
        if torch.cuda.is_available():
            peak_memory = torch.cuda.max_memory_allocated() / (1024**3)  # 转换为GB
            logger.info(f"[Profile] {self.rank_info} - {self.name} Peak Memory: {peak_memory:.2f} GB")
        else:
            logger.info(f"[Profile] {self.rank_info} - {self.name} executed without GPU.")
        elapsed = time.perf_counter() - self.start_time
        logger.info(f"[Profile] {self.rank_info} - {self.name} cost {elapsed:.6f} seconds")
        return False

    def __call__(self, func):
        if asyncio.iscoroutinefunction(func):

            @wraps(func)
            async def async_wrapper(*args, **kwargs):
                async with self:
                    return await func(*args, **kwargs)

            return async_wrapper
        else:

            @wraps(func)
            def sync_wrapper(*args, **kwargs):
                with self:
                    return func(*args, **kwargs)

            return sync_wrapper


class _NullContext:
    # Context manager without decision branch logic overhead
    def __init__(self, *args, **kwargs):
        pass

    def __enter__(self):
        return self

    def __exit__(self, *args):
        return False

    async def __aenter__(self):
        return self

    async def __aexit__(self, *args):
        return False

    def __call__(self, func):
        return func


ProfilingContext = _ProfilingContext
ProfilingContext4Debug = _ProfilingContext if CHECK_ENABLE_PROFILING_DEBUG() else _NullContext