cvs-src summary for September 6-13

Mark Johnston mjohnston at
Mon Sep 13 14:37:51 PDT 2004

FreeBSD cvs-src summary for 30/08/04 to 06/09/04
This is a regular weekly summary of FreeBSD's cutting-edge development.
It is intended to help the FreeBSD community keep up with the fast-paced
work going on in FreeBSD-CURRENT by distilling the deluge of data from
the CVS mailing list into a (hopefully) easy-to-read newsletter.  This
newsletter is marked up in reStructuredText_, so any odd punctuation
that you see is likely intended for the reST parser.

.. _reStructuredText:

You can get old summaries, and an HTML version of this one, at  Please send any comments to Mark Johnston
(mark at

For Lukasz Dudek and Szymon Roczniak's Polish translations of these
summaries, which may lag the English ones slightly, please see

.. contents::

New features
VIA VT6122 gigabit chip supported
Bill Paul (wpaul) added the vge driver, for VIA's VT6122 Gigabit Ethernet
chip.  This chip can be found both on PCI cards, such as some ZyXEL cards,
and in motherboard-integrated network adapters. vge has been added to
GENERIC on i386, pc98, and amd64, but it needs testing on sparc and ia64.

Presentation-oriented slides about FreeBSD available
Murray Stokely (murray) added some general slides about FreeBSD in XML
format.  These are useful for FreeBSD-related presentations.  Thanks to
Simon Nielsen (simon) for the heads up on this addition.

New tool to compress filesystem images
Maxim Sobolev (sobomax) added mkuzip, a utility that compresses filesystem
images for use with the geom_uzip driver.  This eliminates the necessity
of using the Linux-oriented cloop utility.  Currently, it has only been
tested on the i386.

Network code benchmarking and testing tools added
Robert Watson (rwatson) added netsend and netreceive, which transmit
and accept UDP packets of a given size at a fixed rate.  This allows
benchmarking of the network code.  He also added tcpstream, which
generates a simple TCP stream containing a pseudo-random sequence that it
uses to detect data corruption.

Notable changes
Default scheduler changed to 4BSD
Scott Long (scottl) changed the default scheduler from the new ULE one
to the old 4BSD for 5.3.  Some bugs still exist in ULE that affect the
system's stability and performance.  The plan is to switch back to ULE
once 5.3 is released.  This also spawned the thread summarized as `ULE
scheduler bugs preventing it from being used in the release`.

Discussion topics
ULE scheduler bugs preventing it from being used in the release
As mentioned in `Default scheduler changed to 4BSD`, Scott Long (scottl)
changed the default scheduler from the new ULE to the old 4BSD for the 5.3

Jeff Roberson (jeff) replied to Scott's commit, asking, "What's not
working well enough with ULE to run with it?"

Poul-Henning Kamp (phk) offered, "On thing I have observed on my laptop is
that it can take up to 30 seconds for a process to get a signal counting
from when I type CTRL-C."

Jeff asked, "When did this happen?  What change broke it?", remarking, "I
certainly don't get this behavior."

Poul-Henning answered, "It's been that way for some time.  It's
particularly a problem with very cpu intensitive processes, but I have not
tried to collect details."

Don Lewis (truckman) also reported, "There is also the problem that CPU
bound processes that are 'nice' get the same or slightly more CPU time
than CPU bound processes that are not 'nice' [ . . . ]."

Scott responded to Jeff too, saying, "The package build machines were
switched over to 4BSD not long ago and have since been reported to be more
stable.  I can also lock up or panic my test machine within seconds with
ULE, especially if preemption is enabled."

Jeff replied, "I wasn't aware there were any more panics after julian's
preemption change."

Julian Elischer (julian) responded, "There appear to be two problems.  One
that is still related to preemption and one that is not."  He specified,
"the preemption problem is hard to describe, but seems to result in
either a null reference of some kind (last I looked) and the one that is
not driven by preemption finds that a kse is already in a run queue and

Important bug fixes
Bug in ipfw when using NOT or OR with uid, gid, or jail-based rules fixed
Christian S. J. Peron (csjp) fixed a bug in ipfw that caused it to
generate incorrect code when using NOT or OR operations with rules based
on uid, gid, or jail id.  This would result in NOT and OR being discarded
from those rules.  This was reported in `PR 63691`_.

.. _`PR 63691`:

Other bug fixes
Tim Robbins (tjr) fixed two bugs with corrupt MS-DOS (FAT) filesystems.
One resulted in a panic, and the other in possible data corruption.  The
fixes were from NetBSD, and were submitted by Xin LI.

Jun Kuriyama (kuriyama) fixed a crash in portsdb -u.

More information about the freebsd-current mailing list