Skip to content

Conversation

EvgeniiMekhanik
Copy link
Contributor

According our investigation collapsing skb in socket write queue significantly increases perfomance.

- According our investigation collapsing skb in socket
write queue significantly increases perfomance for small
responses.
- Fix `tcp_can_coalesce_send_queue_head`. We should make
Tempesta FW checks before checking `len <= skb->len`.
- We should also check `SKBTX_SHARED_FRAG` flag because
collapsing skbs with defferent value of this flag is
prohibited.
@EvgeniiMekhanik EvgeniiMekhanik force-pushed the MekhanikEvgenii/collapce-skb branch from b21359f to 2c97b77 Compare July 23, 2025 20:46
@EvgeniiMekhanik EvgeniiMekhanik marked this pull request as draft August 4, 2025 13:30
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.

1 participant