Skip to content

Conversation

merrymercy
Copy link
Contributor

No description provided.

@merrymercy merrymercy changed the title Disable two stream overlap on amd Disable all two stream overlap on amd May 21, 2025
@merrymercy merrymercy force-pushed the lianmin/sync-code branch from 887151f to 4bae19b Compare May 21, 2025 01:29
@merrymercy merrymercy merged commit 0388691 into main May 21, 2025
1 of 19 checks passed
@merrymercy merrymercy deleted the lianmin/sync-code branch May 21, 2025 02:07
@@ -392,7 +398,7 @@ def set_kv_buffer(
cache_k = cache_k.view(self.store_dtype)
cache_v = cache_v.view(self.store_dtype)

if self.capture_mode and cache_k.shape[0] < 4:
if self.capture_mode and self.alt_stream is not None:
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is it expected to remove the batch size check and do the overlap for all batch sizes ? in that case we should remove the comment on the next line

Layssy pushed a commit to Layssy/sglang-iaas that referenced this pull request Jun 9, 2025
@sogalin sogalin mentioned this pull request Jun 11, 2025
6 tasks
xwu-intel pushed a commit to xwu-intel/sglang that referenced this pull request Jun 17, 2025
Yuechguo pushed a commit to Yuechguo/sglang that referenced this pull request Jul 29, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants