Skip to content

Add headdim 256 md5 and Disable Python GC during timed iterations #24

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
baoqiwen wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Add headdim 256 md5 and Disable Python GC during timed iterations #24

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions benchmark_flashmask.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,8 @@
from paddle.nn.functional.flash_attention import flashmask_attention
import random
import os
import gc

from datetime import datetime

np.random.seed(0)
Expand Down Expand Up @@ -89,7 +91,11 @@ def do_bench(fn, warmup=25, rep=100, grad_to_none=None, quantiles=None, fast_flu
# Warm-up
for _ in range(n_warmup):
fn()

# Benchmark
# Disable Python GC during timed iterations
gc.collect()
gc.disable()
for i in range(n_repeat):
# we don't want `fn` to accumulate gradient values
# if it contains a backward pass. So we clear the
Expand All @@ -103,6 +109,8 @@ def do_bench(fn, warmup=25, rep=100, grad_to_none=None, quantiles=None, fast_flu
start_event[i].record()
fn()
end_event[i].record()
gc.enable()

# Record clocks
paddle.device.synchronize()
times = paddle.to_tensor([s.elapsed_time(e) for s, e in zip(start_event, end_event)], dtype=paddle.float32)
Expand Down
Loading