Kernel Traffic
Latest | Archives | People | Topics
Latest | Archives | People | Topics
Latest | Archives | People | Topics
Home | News | RSS Feeds | Mailing Lists | Authors Info | Mirrors | Sleeping Cousins

Hurd Traffic #39 For 15 Mar 2000

By Zack Brown

Do you pine for the nice days of minix-1.1, when men were men and wrote their own device drivers?
Are you without a nice project and just dying to cut your teeth on an OS you can try to modify for your needs?
Are you finding it frustrating when everything works on minix? No more all-nighters to get a nifty program working?
Then this post might be just for you :-)
-- Linus Torvalds, 1991

Table Of Contents

Mailing List Stats For This Week

We looked at 62 posts in 220K.

There were 29 different contributors. 11 posted more than once. 9 posted last week too.

The top posters of the week were:

1. Latest Snapshot; Upgrading; Compiling From Sources; Mounting Storage Media

2 Mar 2000 - 7 Mar 2000 (4 posts) Archive Link: "GNU 2000/03/01 uploaded"

Topics: FS: ext2

People: Marcus Brinkmann

Marcus Brinkmann announced the March 1st snapshot of the FTP archive. He added, "I wish I could have included a new glibc and dpkg into this tar file, but those are delayed a bit, sorry." Albert Ollosa replied that he'd installed an older version of the Hurd, and wanted to upgrade to this new tarball. He asked what the procedure was, and Marcus replied that instead of reinstalling from scratch, he should just use 'dselect' with the 'ftp' method, to upgrade the packages individually. If Albert didn't have net connectivity, Marcus suggested getting the deb files some other way, and installing them via 'dpkg'.

In his original post, Albert also how to compile packages from the sources, and Marcus replied:

If you want to compile Debian packages from source, get the orig.tar.gz, *.dsc and *diff.gz if available, and do

dpkg-source -x *dsc
cd <package>-<version>
dpkg-buildpackage -B

YOu need the dpkg-hurd-dev package for this to work.

Albert had also asked how to mount a floppy disk or cdrom, and Marcus explained:

Translators are the right thing, although there is a mount executable.

cd /dev
MAKEDEV hdd # or whatever for your CD

settrans -a /floppy /hurd/ext2fs /dev/fd0
settrans -a /cdrom /hurd/isofs /dev/hdd

(those mounts go away at reboot. For permanent mounts, leave out the -a option).

If your floppy is in MSDOS format, install the mtools package from the Debian archive.

2. PPP/pfinet Saga Continues

4 Mar 2000 - 12 Mar 2000 (7 posts) Archive Link: "Re: i want to help ! how ?"

People: Igor KhavkineNiels M.llerThomas BushnellRoland McGrathMarcus Brinkmann

Continuing from Issue #38, Section #6  (4 Mar 2000: PPP Saga Continues) , in the course of discussion, Igor Khavkine suggested, "Maybe a better idea would be to make pfinet support a general tunneling interface that gives control of the underlying protocol to a userspace daemon like ppp, slip, plip or whatever else one can run in userspace. Or another idea would be to setup each interface as a translator somewhere under /servers/socket/2/. Does this sound totally crazy? I'm just trying to think of a way to support multiple interface types without adding too much code to pfinet, so that if there is a new type of interface that appears it doesn't have to be rewritten." Niels M.ller pointed out, "Note that pfinet is just as "userspace" as any other daemon. I suppose you really mean dividing up the work between several daemons?" And Igor confirmed that he'd been thinking along the lines of division of labor, not user space vs. system space. Niels also suggested a possible division, in which pfinet would keep track of "ip packets, addresses, sockets (including all tcp processing, etc) and interfaces, and possibly some routing info," while ppp and eth would be in charge of "sending and receiving ip packets using the appropriate link-layer mechanisms" Igor replied:

That's almost exactly what I was thinking. To me taking the interface support out of pfinet seems like a good idea, that way support for different kind of interfaces like ppp or VPN tunneling could be added separately. But I don't know the answers to the basically same questions that you are asking:

Igor said he'd try to start coding it, and also asked for comments from folks like Thomas Bushnell, Roland McGrath, and Marcus Brinkmann, but there were no replies.

3. Installation Problem

6 Mar 2000 (3 posts) Archive Link: "Problem during boot"

Topics: FS: ext2

People: Thomas PoindessousIgor KhavkineMarcus Brinkmann

Thomas Poindessous reported, "I have downloaded hurd.deb, gnumach.deb, libc.deb and base_files.deb. After untarred all these files, I have booted with grub," at which point he got an "ext2.static : ../../libdiskfs/boot_start.c .... No such file or directory" error. Igor Khavkine replied, asking if those were the only deb files Thomas had installed. He added, "Get the latest cross-install and other scripts from and make sure you have at least the basic packages listed there." He went on to say that he'd seen this problem quite a few times lately, and asked Marcus Brinkmann if there might be a problem with 'cross-install', or were people just using an old version of the script. Marcus replied, "Well, they use an older script with new packages, and this makes dpkg unhappy. Although the script aborts with an error message, people keep thinking the install is complete and reboot." He added, "I made the error message more clear in the latest version of the script."

4. Philosophy, Cont.

8 Mar 2000 (1 post) Archive Link: "Re: GNU philosophy[was Re: Hurd on VMware ?]"

Topics: Emulators: VMWare

People: Adam Sampson


Continuing from Issue #38, Section #1  (29 Feb 2000: Hurd On VMWare/plex86; Philosophy Of Using Free Software) , Adam Sampson gave a pointer to Jamie Zawinski's take (and mailing list history) on the Emacs code fork.

5. Hierarchical Network Translators

9 Mar 2000 (1 post) Archive Link: "Hierarchical (layered) network translators"

People: Pauli Ojanpera

Pauli Ojanpera made the following suggestion:

How about a translator structure like this?

         /139 (netbios)
         /80 (http)
         /21 (ftp)
         /22 (ssh)
         /23 (telnet)
         /9824 (some outbound connection)
            / (destination)

In each directory there would be at least a file handle for raw data and a config directory. Port numbers could also be mapped to services with /etc/services ..

There was no reply.

6. Politics Of The Hurd

10 Mar 2000 - 12 Mar 2000 (8 posts) Archive Link: "Three silly questions to the Hurd-Core"

People: Marcus BrinkmannLucas C. WagnerChristopher Browne

In the course of discussion, Lucas C. Wagner said that the Hurd was a microkernel replacement for the Linus monolithic kernel, and Marcus Brinkmann came down on him, with:

This is very wrong, please don't make such errornous statements in the public. It only alienates people.

Sorry to pick on this issue, but it's quite important, because of the negative touch it has to Linux supporters.

Lucas replied, "Well, aside from the political correctness, I would say many people view it as a replacement for the Linux kernel even if those are not the words that you would choose to say them in. RMS's dogmatics don't help the issue and many Linux users say worse things about the Hurd because of Torvalds' discussion a while back. I think they can withstand a few comments like these... however, I will not make them in the future, now that you have pointed it out. ;-)" Marcus explained, "I wouldn't object if you say it's your replacement, or anyones in particular. It's also my replacement, but the truth is even better: It does not replace something else, it fills a long-standing huge gap: Nothing like the Hurd was ever attempted before."

Christopher Browne replied to Lucas at length:

It's almost certainly more complex than that, as there have been various positions by various people over the years:

"I am aware of the benefits of a micro kernel approach. However, the fact remains that Linux is here, and GNU isn't --- and people have been working on Hurd for a lot longer than Linus has been working on Linux." -- Ted T'so, 1992.

"Of course 5 years from now that will be different, but 5 years from now everyone will be running free GNU on their 200 MIPS, 64M SPARCstation-5." -- Andrew Tanenbaum, 1992.

"You'll be rid of most of us when BSD-detox or GNU comes out, which should happen in the next few months (yeah, right)." -- Richard Tobin, 1992. [BSD did follow within a year]

And I seem to remember Linus Torvalds suggesting something about Linux being a "stop gap" measure until "GNU" would become available.

More recently, RMS's comments have been rather more indicative of backing away from the notion of Hurd being a *crucial* project. Comments are more like: "There's enough done that it makes sense to put *some* effort into getting it working."

At any rate, *none* of these comments are so strongly doctrinaire that they should cause anyone great offense.

The *truth* is that some people either:

  1. intend to give, or
  2. intend to take

offense, and any old excuse will do.

There are a few people that are so anti-RMS or anti-GNU or anti-FSF (or multiple combinations thereof) that any mention of something like Hurd will cause their heads to explode, virtually speaking...

7. X Saga Continues

11 Mar 2000 - 12 Mar 2000 (2 posts) Archive Link: "Re: further (incomplete) X patches"

People: Marcus BrinkmannJim Franklin

Jim Franklin gave a pointer to the latest XFree86 news, adding that XFree86 4.0 was out. He asked if Marcus Brinkmann would be using this version in his porting efforts. Marcus replied:

I am not sure. My main goal is to get some X packages into the Debian archive, so dependencies on them can be fulfileed without d/l packages from

Because 4.0 is not packaged for Debian yet, this would not work for this version. Because Branden will completely redo the X packaging, it would probably take longer than I want. I will work with Branden on the Hurd issues of packaging as soon as possible, though.


8. pthreads Implementation

12 Mar 2000 (2 posts) Archive Link: "pthreads"

People: Neal H WalfieldMarcus BrinkmannMark Kettenis

Neal H Walfield announced, "I am working on a pthread implementation to run under gnu mach for my OS project this semester; expect to see some bits in the next month or so." He posted some technical problems he was having, and Marcus Brinkmann gave a pointer to Mark Kettenis' work, and replied, "I hope you contacted Mark Kettenis <> before starting to work on it! He already did a lot of work in this area."

9. Changing IP Addresses

12 Mar 2000 (2 posts) Archive Link: "how to change IP address?"

People: Roland McGrath

Gregory Ade successfully configured his network, but wanted to change his IP address. He was used to the ifconfig/route method, and asked how to do this on the Hurd. Roland McGrath replied, "The IP address is kept in the --address option of the pfinet translator. So, like with any translator's options, you can examine and change them at runtime with fsysopts, and you use settrans to change the permanent setting that will be used automatically at boot."







Sharon And Joy

Kernel Traffic is grateful to be developed on a computer donated by Professor Greg Benson and Professor Allan Cruse in the Department of Computer Science at the University of San Francisco. This is the same department that invented FlashMob Computing. Kernel Traffic is hosted by the generous folks at All pages on this site are copyright their original authors, and distributed under the terms of the GNU General Public License version 2.0.