Repository navigation
Conversation
Re-arm trimming and setsockopt dedup were dropped: dev's RX-epoll path (epoll_register, epoll_fd_state) already issues EPOLL_CTL_MOD directly, and each accept path sets TCP_NODELAY exactly once. The fcntl pair stays: cwist_http_pool_submit_async is a public API that blocking fds reach.
gg582
force-pushed
the
perf/epoll-syscall-trim
branch
from
October 13, 2026 13:27
cffe976 to
fe131b6
Compare
Member
Author
|
Closing: CLOCK_MONOTONIC is already vDSO-served on Linux, so COARSE saves tens of nanoseconds per poll slice, and the classic blocking pool this touches is not exercised by the benchmark harness (default async/io_uring). No measurable effect; the re-arm optimization this PR originally proposed is already covered by dev's RX-epoll path (epoll_register, 0fd8c21). |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The classic pool's header-read loop calls clock_gettime every poll slice; CLOCK_MONOTONIC_COARSE (vDSO, ~4 ms granularity) removes that syscall against a 120 s budget.
Scope was reduced during rebase: dev's RX-epoll path (epoll_register with fd-state MOD-first hints) already trims the re-arm syscall, and no per-accept setsockopt duplication exists. Measured end-to-end delta is below noise; the win is one fewer syscall per idle poll slice.