| import torch |
| import math |
| import triton |
| from typing import Optional |
|
|
| |
| if not torch.cuda.is_available(): |
| raise RuntimeError("CUDA is not available. This benchmark requires a CUDA-enabled GPU.") |
| DEVICE = torch.device("cuda:0") |
| torch.cuda.set_device(DEVICE) |
|
|
| def alloc_fn(size: int, align: int, stream: Optional[int]): |
| assert align == 128 |
| assert stream == 0 |
| return torch.empty(size, dtype=torch.int8, device=DEVICE) |
|
|
| triton.set_allocator(alloc_fn) |
| torch.manual_seed(0) |
| try: |
| torch.cuda.manual_seed_all(0) |
| except Exception: |
| pass |
| assert triton.runtime.driver.active.get_current_target().backend == "cuda", "This benchmark only supports CUDA backend." |
|
|
| def _bench_ms(fn): |
| out = triton.testing.do_bench(fn, quantiles=[0.5]) |
| if isinstance(out, (tuple, list)): |
| return float(out[0]) |
| return float(out) |
|
|
| def _is_close(x: torch.Tensor, y: torch.Tensor, rtol=1e-2, atol=5e-3): |
| return torch.allclose(x, y, rtol=rtol, atol=atol) |
|
|
| def _pt_ragged(Q, K, V, row_lens): |
| """ |
| PyTorch baseline for ragged attention. |
| Q:[M,D], K:[N,D], V:[N,Dv], row_lens:[M] -> O:[M,Dv] |
| """ |
| M, D = Q.shape |
| N = K.shape[0] |
| scale = 1.0 / math.sqrt(D) |
|
|
| idx = torch.arange(N, device=Q.device) |
| mask = idx.unsqueeze(0) < row_lens.to(idx.dtype).unsqueeze(1) |
|
|
| scores = (Q @ K.T) * scale |
| scores = scores.masked_fill(~mask, float("-inf")) |
| P = torch.softmax(scores, dim=-1) |
| O = (P @ V).to(torch.float16) |
| return O |
|
|
| def _cpu_ragged(Q, K, V, row_lens): |
| |
| Q_cpu = Q.cpu().float() |
| K_cpu = K.cpu().float() |
| V_cpu = V.cpu().float() |
| row_lens_cpu = row_lens.cpu() |
| result_cpu = _pt_ragged(Q_cpu, K_cpu, V_cpu, row_lens_cpu) |
| return result_cpu.to(DEVICE) |
|
|
| def _bench_pair(M, N, Dq, Dv, len_min_ratio, answer_ragged_attn, baseline_ragged_attn=_pt_ragged): |
| Q = torch.randn(M, Dq, device=DEVICE, dtype=torch.float16) |
| K = torch.randn(N, Dq, device=DEVICE, dtype=torch.float16) |
| V = torch.randn(N, Dv, device=DEVICE, dtype=torch.float16) |
| low = max(1, int(N * len_min_ratio)) |
| row_lens = torch.randint(low=low, high=N+1, size=(M,), device=DEVICE, dtype=torch.int32) |
| |
| Q32 = Q.float() |
| K32 = K.float() |
| V32 = V.float() |
| |
| |
| torch.cuda.synchronize() |
| import time |
| cpu_times = [] |
| for _ in range(10): |
| start = time.perf_counter() |
| _cpu_ragged(Q, K, V, row_lens) |
| torch.cuda.synchronize() |
| cpu_times.append((time.perf_counter() - start) * 1000) |
| cpu_baseline_ms = sorted(cpu_times)[len(cpu_times)//2] |
| |
| |
| gpu_baseline_ms = _bench_ms(lambda: baseline_ragged_attn(Q32, K32, V32, row_lens)) |
| |
| |
| answer_ms = _bench_ms(lambda: answer_ragged_attn(Q, K, V, row_lens)) |
| |
| |
| ref = baseline_ragged_attn(Q32, K32, V32, row_lens) |
| out = answer_ragged_attn(Q, K, V, row_lens) |
| passed = _is_close(out, ref, rtol=1e-2, atol=5e-3) |
| |
| |
| flops = 2.0 * M * N * (Dq + Dv) |
| to_tflops = lambda ms: flops * 1e-12 / (ms * 1e-3) if ms is not None else None |
| |
| return { |
| "M": M, "N": N, "Dq": Dq, "Dv": Dv, |
| "cpu_baseline_ms": cpu_baseline_ms, |
| "gpu_baseline_ms": gpu_baseline_ms, |
| "answer_ms": answer_ms, |
| "baseline_ms": cpu_baseline_ms, |
| "baseline_tflops": to_tflops(gpu_baseline_ms), |
| "answer_tflops": to_tflops(answer_ms), |
| "close_passed": passed, |
| "rtol": 1e-2, "atol": 5e-3, "passed": passed, |
| } |
|
|
| def _warmup_gpu(iters: int = 10): |
| try: |
| M, N, Dq, Dv = 1024, 1024, 64, 64 |
| Q = torch.randn(M, Dq, device=DEVICE, dtype=torch.float16) |
| K = torch.randn(N, Dq, device=DEVICE, dtype=torch.float16) |
| V = torch.randn(N, Dv, device=DEVICE, dtype=torch.float16) |
| row_lens = torch.randint(low=int(N*0.25), high=N+1, size=(M,), device=DEVICE, dtype=torch.int32) |
| for _ in range(max(1, int(iters))): |
| _ = _pt_ragged(Q.float(), K.float(), V.float(), row_lens) |
| torch.cuda.synchronize() |
| except Exception: |
| pass |
|
|
| def summarize_speedup(answer_ragged_attn, baseline_ragged_attn=None, print_output=False, metadata=None): |
| |
| |
| |
| _warmup_gpu(10) |
| |
| |
| if metadata is None: |
| metadata = {} |
| shapes = metadata.get("shapes", None) |
| if shapes is None: |
| M_list = metadata.get("M_list", [512, 1024]) |
| N = metadata.get("N", 1024) |
| Dq = metadata.get("Dq", 64) |
| Dv = metadata.get("Dv", 64) |
| len_min_ratio = metadata.get("len_min_ratio", 0.25) |
| shapes = [(M, N, Dq, Dv, len_min_ratio) for M in M_list] |
| |
| rows = [] |
| for (M, N, Dq, Dv, len_min_ratio) in shapes: |
| r = _bench_pair(M, N, Dq, Dv, len_min_ratio, answer_ragged_attn, _pt_ragged) |
| rows.append(r) |
| |
| if print_output: |
| print("\n=== Answer vs Baseline: Speedup for each shape (based on median time) ===") |
| |
| speedups_cpu = [] |
| speedups_gpu = [] |
| for r in rows: |
| answer_time = r["answer_ms"] |
| cpu_time = r.get("cpu_baseline_ms") |
| gpu_time = r.get("gpu_baseline_ms") |
| |
| if cpu_time is not None and answer_time is not None: |
| sp_cpu = cpu_time / answer_time |
| speedups_cpu.append(sp_cpu) |
| |
| if gpu_time is not None and answer_time is not None: |
| sp_gpu = gpu_time / answer_time |
| speedups_gpu.append(sp_gpu) |
| |
| status = "OK" if r["close_passed"] else "FAIL" |
| if print_output: |
| print( |
| f"M={r['M']:5d} N={r['N']:5d} Dq={r['Dq']:3d} Dv={r['Dv']:3d} " |
| f"CPU={cpu_time:7.3f} ms GPU={gpu_time:7.3f} ms answer={answer_time:7.3f} ms " |
| f"[Passed: {status} " |
| f"rtol={r['rtol']:.1e} atol={r['atol']:.1e}]" |
| ) |
| |
| if speedups_cpu: |
| geo_mean_cpu = math.exp(sum(math.log(s) for s in speedups_cpu) / len(speedups_cpu)) |
| else: |
| geo_mean_cpu = 0.0 |
| |
| if speedups_gpu: |
| geo_mean_gpu = math.exp(sum(math.log(s) for s in speedups_gpu) / len(speedups_gpu)) |
| else: |
| geo_mean_gpu = 0.0 |
| |
| if print_output: |
| print("\n--- Summary ---") |
| print(f"Geometric mean speedup vs CPU: {geo_mean_cpu:.3f}x") |
| print(f"Geometric mean speedup vs GPU: {geo_mean_gpu:.3f}x") |
| |
| return rows, geo_mean_cpu, geo_mean_gpu, geo_mean_gpu |
|
|
| def run_benchmark(answer_ragged_attn, baseline_ragged_attn=None, print_output=False, metadata=None): |
| |
| |
| rows, geo_mean_cpu, geo_mean_gpu, _ = summarize_speedup(answer_ragged_attn, baseline_ragged_attn, print_output=print_output, metadata=metadata) |
| |
| |
| cpu_times = [r["cpu_baseline_ms"] for r in rows if r.get("cpu_baseline_ms") is not None] |
| gpu_times = [r["gpu_baseline_ms"] for r in rows if r.get("gpu_baseline_ms") is not None] |
| answer_times = [r["answer_ms"] for r in rows if r.get("answer_ms") is not None] |
| |
| geo_mean_cpu_time = math.exp(sum(math.log(t) for t in cpu_times) / len(cpu_times)) if cpu_times else 0.0 |
| geo_mean_gpu_time = math.exp(sum(math.log(t) for t in gpu_times) / len(gpu_times)) if gpu_times else 0.0 |
| geo_mean_answer_time = math.exp(sum(math.log(t) for t in answer_times) / len(answer_times)) if answer_times else 0.0 |
| |
| return { |
| "rows": rows, |
| "geometric_mean_speedup_cpu": geo_mean_cpu, |
| "geometric_mean_speedup_gpu": geo_mean_gpu, |
| "geometric_mean_speedup": geo_mean_gpu, |
| "arithmetic_mean_speedup": geo_mean_gpu, |
| "median_speedup": geo_mean_gpu, |
| "geo_mean_cpu_time": geo_mean_cpu_time, |
| "geo_mean_gpu_time": geo_mean_gpu_time, |
| "geo_mean_answer_time": geo_mean_answer_time, |
| "pass_all": all(r["close_passed"] for r in rows), |
| } |
|
|
|
|