| import functools |
| import gc |
| import logging |
| import os |
| from dataclasses import dataclass, field |
| from typing import Any |
|
|
| import torch |
| import torch.profiler |
|
|
|
|
| logging.basicConfig(level=logging.INFO, format="%(asctime)s %(levelname)s %(name)s: %(message)s") |
| logger = logging.getLogger(__name__) |
|
|
|
|
| def annotate(func, name): |
| """Wrap a function with torch.profiler.record_function for trace annotation.""" |
|
|
| @functools.wraps(func) |
| def wrapper(*args, **kwargs): |
| with torch.profiler.record_function(name): |
| return func(*args, **kwargs) |
|
|
| return wrapper |
|
|
|
|
| def annotate_pipeline(pipe): |
| """Apply profiler annotations to key pipeline methods. |
| |
| Monkey-patches bound methods so they appear as named spans in the trace. |
| Non-invasive — no source modifications required. |
| """ |
| annotations = [ |
| ("transformer", "forward", "transformer_forward"), |
| ("vae", "decode", "vae_decode"), |
| ("vae", "encode", "vae_encode"), |
| ("scheduler", "step", "scheduler_step"), |
| ] |
|
|
| |
| for component_name, method_name, label in annotations: |
| component = getattr(pipe, component_name, None) |
| if component is None: |
| continue |
| method = getattr(component, method_name, None) |
| if method is None: |
| continue |
|
|
| |
| if "LTX2" in pipe.__class__.__name__: |
| func = getattr(method, "__func__", method) |
| wrapped = annotate(func, label) |
| bound_method = wrapped.__get__(component, type(component)) |
| setattr(component, method_name, bound_method) |
| else: |
| |
| setattr(component, method_name, annotate(method, label)) |
|
|
| |
| if hasattr(pipe, "encode_prompt"): |
| pipe.encode_prompt = annotate(pipe.encode_prompt, "encode_prompt") |
|
|
|
|
| def flush(): |
| gc.collect() |
| torch.cuda.empty_cache() |
| torch.cuda.reset_max_memory_allocated() |
| torch.cuda.reset_peak_memory_stats() |
|
|
|
|
| def benchmark_fn(f, *args, num_runs=5, num_warmups=2, **kwargs): |
| """Benchmark a function using CUDA events for accurate GPU timing. |
| |
| Uses CUDA events to measure wall-clock time including GPU execution, |
| without the overhead of torch.profiler. Reports mean and standard deviation |
| over multiple runs. |
| |
| Returns: |
| dict with keys: mean_ms, std_ms, runs_ms (list of individual timings) |
| """ |
| |
| for _ in range(num_warmups): |
| f(*args, **kwargs) |
| torch.cuda.synchronize() |
|
|
| |
| times = [] |
| for _ in range(num_runs): |
| start = torch.cuda.Event(enable_timing=True) |
| end = torch.cuda.Event(enable_timing=True) |
|
|
| start.record() |
| f(*args, **kwargs) |
| end.record() |
|
|
| torch.cuda.synchronize() |
| times.append(start.elapsed_time(end)) |
|
|
| mean_ms = sum(times) / len(times) |
| variance = sum((t - mean_ms) ** 2 for t in times) / len(times) |
| std_ms = variance**0.5 |
|
|
| return {"mean_ms": mean_ms, "std_ms": std_ms, "runs_ms": times} |
|
|
|
|
| @dataclass |
| class PipelineProfilingConfig: |
| name: str |
| pipeline_cls: Any |
| pipeline_init_kwargs: dict[str, Any] |
| pipeline_call_kwargs: dict[str, Any] |
| compile_kwargs: dict[str, Any] | None = field(default=None) |
| compile_regional: bool = False |
|
|
|
|
| class PipelineProfiler: |
| def __init__(self, config: PipelineProfilingConfig, output_dir: str = "profiling_results"): |
| self.config = config |
| self.output_dir = output_dir |
| os.makedirs(output_dir, exist_ok=True) |
|
|
| def setup_pipeline(self, annotate=True): |
| """Load the pipeline from pretrained, optionally compile, and annotate.""" |
| logger.info(f"Loading pipeline: {self.config.name}") |
| pipe = self.config.pipeline_cls.from_pretrained(**self.config.pipeline_init_kwargs) |
| pipe.to("cuda") |
|
|
| if self.config.compile_kwargs: |
| if self.config.compile_regional: |
| logger.info( |
| f"Regional compilation (compile_repeated_blocks) with kwargs: {self.config.compile_kwargs}" |
| ) |
| pipe.transformer.compile_repeated_blocks(**self.config.compile_kwargs) |
| else: |
| logger.info(f"Full compilation with kwargs: {self.config.compile_kwargs}") |
| pipe.transformer.compile(**self.config.compile_kwargs) |
|
|
| |
| pipe.set_progress_bar_config(disable=True) |
|
|
| if annotate: |
| annotate_pipeline(pipe) |
| return pipe |
|
|
| def run(self): |
| """Execute the profiling run: warmup, then profile one pipeline call.""" |
| pipe = self.setup_pipeline() |
| flush() |
|
|
| mode = "compile" if self.config.compile_kwargs else "eager" |
| trace_file = os.path.join(self.output_dir, f"{self.config.name}_{mode}.json") |
|
|
| |
| logger.info("Running warmup...") |
| pipe(**self.config.pipeline_call_kwargs) |
| flush() |
|
|
| |
| logger.info("Running profiled iteration...") |
| activities = [ |
| torch.profiler.ProfilerActivity.CPU, |
| torch.profiler.ProfilerActivity.CUDA, |
| ] |
| with torch.profiler.profile( |
| activities=activities, |
| record_shapes=True, |
| profile_memory=True, |
| with_stack=True, |
| ) as prof: |
| with torch.profiler.record_function("pipeline_call"): |
| pipe(**self.config.pipeline_call_kwargs) |
|
|
| |
| prof.export_chrome_trace(trace_file) |
| logger.info(f"Chrome trace saved to: {trace_file}") |
|
|
| |
| print("\n" + "=" * 80) |
| print(f"Profile summary: {self.config.name} ({mode})") |
| print("=" * 80) |
| print( |
| prof.key_averages().table( |
| sort_by="cuda_time_total", |
| row_limit=20, |
| ) |
| ) |
|
|
| |
| pipe.to("cpu") |
| del pipe |
| flush() |
|
|
| return trace_file |
|
|
| def benchmark(self, num_runs=5, num_warmups=2): |
| """Benchmark pipeline wall-clock time without profiler overhead. |
| |
| Uses CUDA events for accurate GPU-inclusive timing over multiple runs. |
| No annotations are applied to avoid any overhead from record_function wrappers. |
| Reports mean, std, and individual run times. |
| """ |
| pipe = self.setup_pipeline(annotate=False) |
| flush() |
|
|
| mode = "compile" if self.config.compile_kwargs else "eager" |
|
|
| logger.info(f"Benchmarking {self.config.name} ({mode}): {num_warmups} warmup + {num_runs} timed runs...") |
| result = benchmark_fn(pipe, num_runs=num_runs, num_warmups=num_warmups, **self.config.pipeline_call_kwargs) |
|
|
| print("\n" + "=" * 80) |
| print(f"Benchmark: {self.config.name} ({mode})") |
| print("=" * 80) |
| print(f" Runs: {num_runs} (after {num_warmups} warmup)") |
| print(f" Mean: {result['mean_ms']:.1f} ms") |
| print(f" Std: {result['std_ms']:.1f} ms") |
| print(f" Individual: {', '.join(f'{t:.1f}' for t in result['runs_ms'])} ms") |
| print("=" * 80) |
|
|
| |
| pipe.to("cpu") |
| del pipe |
| flush() |
|
|
| return result |
|
|