[true-async] bump base image to v0.7.0-rc.6 (restore HTTP/1+H2 perf)#776
Closed
EdmondDantes wants to merge 1 commit into
Closed
[true-async] bump base image to v0.7.0-rc.6 (restore HTTP/1+H2 perf)#776EdmondDantes wants to merge 1 commit into
EdmondDantes wants to merge 1 commit into
Conversation
rc.6 reverts the experimental 1ms reactor-poll throttle back to the original CLOCK_MONOTONIC_COARSE jiffy (~10ms), which is what HTTP/1 (pipelined/baseline) and HTTP/2 want for batching. The H3 stream-credit fix (ngtcp2_conn_extend_max_streams_bidi) stays in (server v0.7.1). Supersedes MDA2AV#775 (rc.5 adaptive throttle regressed h2c at scale).
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
Bumps
frameworks/true-async-serverbase image0.7.0-rc.4→0.7.0-rc.6-php8.6.Why
rc.6 reverts the experimental 1ms reactor-poll throttle back to the original
CLOCK_MONOTONIC_COARSEjiffy (~10ms). The fine 1ms tick (introduced for an H3 PTO fix) regressed HTTP/1 (pipelined/baseline) and HTTP/2, which benefit from coarse batching (fewer reactor polls = fewer syscalls = more throughput). This restores H1/H2 to their prior numbers.The H3 stream-credit fix stays in (server v0.7.1 —
ngtcp2_conn_extend_max_streams_bidion stream close, the thing that took baseline-h3/static-h3 c=64 from ~1277 req/s back to normal).Supersedes #775 (rc.5 used an adaptive throttle that shifted the regression onto h2c at scale — rejected).
Image
trueasync/php-true-async:0.7.0-rc.6-php8.6(on Docker Hub, php 8.6).