Skip to content

Conversation

@oandreeva-nv
Copy link
Contributor

@oandreeva-nv oandreeva-nv commented Aug 5, 2025

Overview:

Besides kv layer order enforcement, here's what I do:

  • Eliminated redundant operations:
  • Replaced list(kv_caches.values())[0] with next(iter(kv_caches.values())) - more efficient for getting the first tensor
  • Removed the unnecessary tensors = list(kv_caches.values()) variable
  • The original code iterated over kv_caches.values() three times, now it only does it once (in the validation check)

Details:

Where should the reviewer start?

Related Issues: (use one of the action keywords Closes / Fixes / Resolves / Relates to)

  • closes GitHub issue: #xxx

@copy-pr-bot
Copy link

copy-pr-bot bot commented Aug 5, 2025

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@ryanolson ryanolson merged commit d52b42e into ryan/connector-dev Aug 6, 2025
6 of 12 checks passed
@ryanolson ryanolson deleted the oandreeva/kvbm/connector/kv_layer_ordering branch August 6, 2025 05:23
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants