Differential D29515

tcp: Use jenkins_hash32() in hostcache
ClosedPublic
Actions

Authored by rscheff on Mar 31 2021, 4:19 PM.

Tags

None

Referenced Files

	Unknown Object (File)
	Thu, Jul 23, 11:46 PM

	Unknown Object (File)
	Thu, Jul 23, 11:46 PM

	Unknown Object (File)
	Thu, Jul 23, 12:03 PM

	Unknown Object (File)
	Sat, Jul 18, 4:17 AM

	Unknown Object (File)
	Mon, Jul 13, 4:57 AM

	Unknown Object (File)
	Mon, Jul 6, 7:27 AM

	Unknown Object (File)
	Sat, Jul 4, 12:05 PM

	Unknown Object (File)
	Fri, Jul 3, 11:52 AM

Subscribers

Details

Reviewers

jhb
tuexen
ae
bcr

Group Reviewers

transport
manpages

Commits

rG139576030ff3: tcp: Use jenkins_hash32() in hostcache
rGb878ec024bbe: tcp: Use jenkins_hash32() in hostcache

Summary

While investigating the reported hostcache
issues and adding a histogram to look at the
distribution of used hashbuckets, and
bucketdepth, I found that the exteremly
fast XOR "hash" would result in a rather
skewed utilization of the hostcache, for
randomly selected remote IP addresses.

As other parts of the base tcp stack (eg.
tcp fastopen) already use jenkins_hash32,
and the properties appear reasonably good,
switching to use that.

Test Plan

Anecdotal evidence (single trial, using
approximately 1000 randomly selected hosts,
overall, but ~400 for XOR and ~600 for Jenkins):

Unused Hash buckets: XOR ~51%, Jenkins ~35%
Average Bucket Depth: XOR ~2.44, Jenkins ~1.54

Diff Detail

Repository

rS FreeBSD src repository - subversion

Lint

Lint Passed

Unit

No Test Coverage

Build Status

Buildable 38218
Build 35107: arc lint + arc unit

Event Timeline

rscheff created this revision.Mar 31 2021, 4:19 PM

Herald added a reviewer: transport. · View Herald TranscriptMar 31 2021, 4:19 PM

Herald added subscribers: melifaro, imp. · View Herald Transcript

rscheff requested review of this revision.Mar 31 2021, 4:19 PM

Harbormaster completed remote builds in B38218: Diff 86626.Mar 31 2021, 4:19 PM

ae added a subscriber: ae.Apr 1 2021, 7:31 AM

ae added inline comments.

sys/netinet/tcp_hostcache.c
187	Maybe would it be better to use full IPv6 address to compute hash?

ae accepted this revision as: ae.Apr 1 2021, 7:32 AM

ae added inline comments.

sys/netinet/tcp_hostcache.c
187	ah, sorry for the noise :)

rscheff marked 2 inline comments as done.Apr 1 2021, 7:33 AM

transport call: initialize salt randomly, and change it when purging, when triggereing purge with "2" instead of "1".

initialize the hash salt with a random number
reseed the salt when purging all entries with "2"
adding man entries for all hostcache related sysctls

Herald added a reviewer: manpages. · View Herald TranscriptApr 8 2021, 4:46 PM

Harbormaster completed remote builds in B38424: Diff 87051.Apr 8 2021, 4:46 PM

rscheff added a reviewer: bcr.Apr 8 2021, 4:47 PM

tuexen accepted this revision.Apr 8 2021, 5:58 PM

This revision is now accepted and ready to land.Apr 8 2021, 5:58 PM

rebase to main

This revision now requires review to proceed.Apr 8 2021, 6:25 PM

Harbormaster completed remote builds in B38425: Diff 87056.Apr 8 2021, 6:25 PM

tuexen accepted this revision.Apr 8 2021, 6:31 PM

This revision is now accepted and ready to land.Apr 8 2021, 6:31 PM

Closed by commit rGb878ec024bbe: tcp: Use jenkins_hash32() in hostcache (authored by rscheff). · Explain WhyApr 8 2021, 6:39 PM

This revision was automatically updated to reflect the committed changes.

rscheff added a commit: rGb878ec024bbe: tcp: Use jenkins_hash32() in hostcache.

rscheff added a commit: rG139576030ff3: tcp: Use jenkins_hash32() in hostcache.Apr 16 2021, 8:47 PM

Revision Contents
Changeset List

Path

Size

sys/

netinet/

tcp_hostcache.c

14 lines

Diff 86626

sys/netinet/tcp_hostcache.c

Loading...