iperf_sctp: disable delayed-SACK by default to fix slow single-flow SCTP throughput#2042
Open
nshopik wants to merge 1 commit into
Open
iperf_sctp: disable delayed-SACK by default to fix slow single-flow SCTP throughput#2042nshopik wants to merge 1 commit into
nshopik wants to merge 1 commit into
Conversation
Contributor
|
Thanks for the PR! |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Single-flow SCTP throughput collapses to a few Mbits/sec on Linux (~300× slower than TCP on the same loopback) because the receiver's delayed-SACK timer (sack_delay=200 ms, sack_freq=2) waits for a second DATA chunk that the blocked sender will not produce — each iperf3 64 KB block lands as one DATA chunk and stalls 200 ms per I/O cycle. Reproducible on both arm64 and x86_64; not architecture-specific.
iperf3 is a throughput benchmark, so disable delayed-SACK by setting SCTP_DELAYED_SACK with sack_delay=0, sack_freq=1
on all three SCTP socket entry points:
Also propagate SCTP_NODELAY to both the server's listen and accepted sockets when -N is given, closing a pre-existing asymmetry with iperf_tcp_listen's handling of TCP_NODELAY.
configure gains an AC_CHECK_TYPES probe for struct sctp_sack_info; new code is guarded by HAVE_STRUCT_SCTP_SACK_INFO and ENOPROTOOPT is treated as non-fatal, so platforms without support are silently unaffected.
master and fix performance
PLEASE NOTE the following text from the iperf3 license. Submitting a
pull request to the iperf3 repository constitutes "[making]
Enhancements available...publicly":
The complete iperf3 license is available in the
LICENSEfile in thetop directory of the iperf3 source tree.
Version of iperf3 (or development branch, such as
masteror3.1-STABLE) to which this pull request applies:Issues fixed (if any): slow sctp performance on arm64 #1815
Brief description of code changes (suitable for use as a commit message): disable delayed-SACK by default to fix slow single-flow SCTP throughput (slow sctp performance on arm64 #1815)