Reduce TLS write-path overhead#13202
Open
zwoop wants to merge 1 commit into
Open
Conversation
Insert a buffer BIO on the write side so multiple TLS records accumulate without syscalls, flushed once after the write loop. This reduces write() calls from one-per-record to one-per-flush, directly addressing the ~15% of CPU spent in write() syscalls. TCP_CORK/NOPUSH wraps the write loop as a secondary optimization for TCP segment batching. Consolidate SSL ex_data from 9 slots to 1 struct pointer, skip handshake buffer allocation when blind-tunnel is not configured, fix sslTotalBytesSent to update inside the write loop for correct dynamic record sizing ramp, batch metric counter increments to reduce cross-thread cache-line contention, and use the cached NetHandler timestamp instead of calling ink_get_hrtime per write event.
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Insert a buffer BIO on the write side so multiple TLS records accumulate without syscalls, flushed once after the write loop. This reduces write() calls from one-per-record to one-per-flush, directly addressing the ~15% of CPU spent in write() syscalls. TCP_CORK/NOPUSH wraps the write loop as a secondary optimization for TCP segment batching. Consolidate SSL ex_data from 9 slots to 1 struct pointer, skip handshake buffer allocation when blind-tunnel is not configured, fix sslTotalBytesSent to update inside the write loop for correct dynamic record sizing ramp, batch metric counter increments to reduce cross-thread cache-line contention, and use the cached NetHandler timestamp instead of calling ink_get_hrtime per write event.