Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 19 Jun 2025 13:39:20 +0000
From:      bugzilla-noreply@freebsd.org
To:        net@FreeBSD.org
Subject:   [Bug 287229] IP reassembly issue in FreeBSD 14.1
Message-ID:  <bug-287229-7501-eQiDPfZ4IC@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-287229-7501@https.bugs.freebsd.org/bugzilla/>
References:  <bug-287229-7501@https.bugs.freebsd.org/bugzilla/>

next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D287229

--- Comment #21 from Michael Tuexen <tuexen@freebsd.org> ---
(In reply to Michael Tuexen from comment #20)
Answering myself: I guess we want to limit
* the length of the queue based on the CPU capabilities
* the total number of entries based on the available memory
I guess we use also the memory for estimating the CPU capabilities. You don=
't
use a lot of memory on slow systems...
So using a constant queue length does not seem very good unless we use a li=
mit
small enough for slow CPUs...

@Lucas
If you put on the FreeBSD machine with 200MB kernel space
net.inet.ip.maxfragbucketsize=3D100
in /etc/sysctl.conf and reboot it and run again your tests.
Are the results the same? If you still observe differences, then you are
hitting the limit of 400 entries in the hash table.

--=20
You are receiving this mail because:
You are on the CC list for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-287229-7501-eQiDPfZ4IC>