Rename parameter "softmax_sink" to "sinks" for flash_attn_with_kvcache kernel #30
+22
−24
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
deepvars
approved these changes
Nov 3, 2025
airMeng
approved these changes
Nov 4, 2025
|
"benchmark/bench_flash_attn.py" and "python/sgl_kernel/flash_attn.py" also need rename sink~ |
Hi @sunjiweiswift I already did that. In case I missed anything, can you point to the specific line of code in these files? |
…e kernel, as "sinks" is used in the sglang framework
sunjiweiswift
approved these changes
Nov 4, 2025
ck-intel
force-pushed
the
rename_sfotmax_sink
branch
from
d0646ad to
f468bbf
Compare
Sorry. You've already corrected it. I made a mistake.
Sorry. You've already corrected it. I made a mistake. |
deepvars
approved these changes
Nov 4, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Rename the parameter "softmax_sink" to "sinks" for flash_attn_with_kvcache kernel, as "sinks" is used in the sglang framework.
refer file: sglang/python/sglang/srt/layers/attention/xpu_backend.py
Run the tests: