Skip to content

[Bugfix][DP] Add with_prefill_across_dp to AscendMetadata to fix dp #1094

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

Merged
merged 1 commit into from
Jun 6, 2025
Merged

[Bugfix][DP] Add with_prefill_across_dp to AscendMetadata to fix dp #1094

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
30 changes: 19 additions & 11 deletions vllm_ascend/attention/attention_v1.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,6 +132,8 @@ class AscendMetadata:
# For logging.
num_input_tokens: int = 0 # Number of tokens including padding.

with_prefill_across_dp: bool = False


class AscendAttentionMetadataBuilder:

Expand All @@ -142,8 +144,12 @@ def reorder_batch(self, input_batch: "InputBatch",
scheduler_output: "SchedulerOutput") -> bool:
return False

def build(self, num_reqs, num_actual_tokens, max_query_len,
common_prefix_len):
def build(self,
num_reqs,
num_actual_tokens,
max_query_len,
common_prefix_len,
with_prefill_across_dp: bool = False):

block_table = self.runner.input_batch.block_table[0].get_device_tensor(
)
Expand All @@ -160,15 +166,17 @@ def build(self, num_reqs, num_actual_tokens, max_query_len,
query_start_loc = query_start_loc_cpu.to(self.runner.device,
non_blocking=True)

attn_metadata = AscendMetadata(num_actual_tokens=num_actual_tokens,
block_tables=block_table,
query_start_loc=query_start_loc,
query_lens=query_lens,
seq_lens=seq_lens,
max_query_len=max_query_len,
slot_mapping=slot_mapping,
attn_mask=attn_mask,
attn_state=attn_state)
attn_metadata = AscendMetadata(
num_actual_tokens=num_actual_tokens,
block_tables=block_table,
query_start_loc=query_start_loc,
query_lens=query_lens,
seq_lens=seq_lens,
max_query_len=max_query_len,
slot_mapping=slot_mapping,
attn_mask=attn_mask,
attn_state=attn_state,
with_prefill_across_dp=with_prefill_across_dp)
return attn_metadata


Expand Down