Skip to content

fix: duplicate pad_between_seqs skips and wrong architecture direction #3390

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
andrewwhitecdw wants to merge 3 commits into
base: main
Choose a base branch
from
+1 −9
Open

fix: duplicate pad_between_seqs skips and wrong architecture direction #3390

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
d6893e9
fix: duplicate pad_between_seqs skips and wrong architecture direction
andrewwhitecdw Aug 13, 2026
f14fef3
Merge branch 'main' into bugfix/test-attention-with-cp-duplicate-pad-…
andrewwhitecdw Aug 18, 2026
8b4412a
Merge branch 'main' into bugfix/test-attention-with-cp-duplicate-pad-…
andrewwhitecdw Aug 19, 2026
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
10 changes: 1 addition & 9 deletions tests/pytorch/attention/test_attention_with_cp.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -316,19 +316,11 @@ def test_cp_with_flash_attention(cp_pool, dtype, model, qkv_format, cp_comm_type
if pad_between_seqs:
if qkv_format != "thd":
pytest.skip("pad_between_seqs only applies to THD format!")
if not FlashAttentionUtils.v3_is_installed or get_device_compute_capability() > (9, 0):
if not FlashAttentionUtils.v3_is_installed or get_device_compute_capability() != (9, 0):
pytest.skip("pad_between_seqs with CP requires Flash Attention v3 on Hopper (sm90)!")
if cp_comm_type == "a2a+p2p":
pytest.skip("pad_between_seqs is not yet supported with A2A+P2P CP comm type!")

if pad_between_seqs:
if qkv_format != "thd":
pytest.skip("pad_between_seqs only applies to THD format!")
if not FlashAttentionUtils.v3_is_installed:
pytest.skip("pad_between_seqs with CP requires Flash Attention v3!")
if cp_comm_type == "a2a+p2p":
pytest.skip("pad_between_seqs is not yet supported with A2A+P2P CP comm type!")

config = model_configs_flash_attn[model]
config.context_parallel = True
config.cp_comm_type = cp_comm_type
Expand Down