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 #47 For 10 May 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 55 posts in 171K.

There were 22 different contributors. 8 posted more than once. 5 posted last week too.

The top posters of the week were:


1. Important Step Forward: A Hurd Autobuilder
27 Apr 2000 - 2 May 2000 (5 posts) Subject: "autobuilder web pages with (failed) build logs"
People: Marcus BrinkmannJohn TobeyColin Watson

Marcus Brinkmann announced:

you can find the autobuilder status and the build logs at

I am sorry for the inconvenience german error messages in the logs might cause you. Ask me if you need a translation. I will switch over to english from now on.

Pick up any missing or out-of-date package, check the highest version number (displayed first in the list) and check the log link at the bottom of the version page.

John Tobey extolled:

Excellent work, Marcus. Congratulations and THANK YOU!

Take a look everybody, Marcus has spent the past 7 months rewriting the Debian build daemon from scratch so that it works with the Hurd. He calls it "Turtle, the friendly autobuilder that never catches up."

Though not quite finished yet, Marcus has added quite a number of fabulous features that bid fair to play a central role in the Debian distribution infrastructure for years to come (Linux as well as Hurd-based).

A quick glance at the error logs shows that elvis-tiny needs termio.h and ncurses is too big to compile (leading to farm purchases). Most of the other "failures" appear to be Turtle guessing wrong about the name of the source directory or the changes file. Hopefully we'll get that fixed by the weekend.

Colin Watson reported problems with the server, giving some 'traceroute' output. Marcus replied:

Yes, sorry. I hope they will be resolved tuesday or so (I don't have physical access to this machine, so I can't do anything but wait).

I will post when it works again.

A day later, under the Subject: back, he reported that the page was back online.


2. Hurd Conference In Paris
2 May 2000 (1 post) Subject: "Hurd conference in Paris"
People: Thomas Poindessous

Thomas Poindessous announced:

I'm a student in a french computer science school. With all of the members of the school Lug, we want to organize a Debian GNU/Linux GNU/Hurd conference. We are in touch with Debian members for Debian speeches. We are looking for Hurd developers to talk about Hurd. If they can speak french, it will be greater, it isn't necessary. You can talk about what you want in Hurd life (technical, philosophical, translators, micro-kernel, etc.)

There will be 150-200 people and it's planning for Saturday 17 June.

We can pay for the trip, but we don't have a lot of money so if you aren't europeans, it will be difficult to find all the money.

I hope some of you will be with us.

If you have any questions, please mail me at

There was no reply.


3. Source Diffs; Splitting The Hurd Into Smaller Packages
2 May 2000 - 3 May 2000 (2 posts) Subject: "Hurd sources distribution"
People: Marcus BrinkmannTomasz Wegrzanowski

Tomasz Wegrzanowski had some questions about the Hurd. First, he asked where he could find the diffs to Hurd source packages. Marcus Brinkmann took the blame:

Sorry, my omission. I always try to keep the debian directory uptodate in CVS, and most of the time the rest is unchanged, so I don't package upstream/diff, but only as a single tar file.

The very few cases I apply patches I will try to remember to put them somewhere in debian/. Note that I always mention changes explicitly in the debian changelog, so if it doesn't mention anything, nothing changed from CVS.

Tomasz also asked why the Hurd sources hadn't been split into separate packages. He pointed out that compiling a Linux kernel took about half an hour on his machine, while compiling the Hurd took 10 hours or more. Marcus replied:

Yes, the Hurd is somewhat bigger, and the fact that there are pic libraries created as well doesn't make it better. Furthermore, native compilation is always slower than cross compiling from linux (becuase of the general slowiness of the hurd).

However, what would a split gain you, except that you have to compile several packages? Maybe when all hurd libraries have versioned symbols a split might become useful. This is something that will be decided by the Hurd core when it becomes interesting.

Tomasz pointed out, "dpkg was invented to support many little packages, huge, multi-functional packages aren't very compatible." Marcus replied:

Actually, huge packages which contain related stuff are much easier to handle.

For now, please just accept that it is much more convenient for the developers to keep the whole hurd source in single tree, although it might this is not the Most Perfect Thing To Everyone.

There is a technical reason, too: The Hurd libraries don't carry versioned symbols (actually, the latest libthreads does), and the soname is not kept uptodate in CVS, so you might get obscure problems if you update only parts of the Hurd and are not careful about it. With split packages, I would end up to spent most of the packaging time in figuring out the correct dependencies, and might still fail. This is not a workable solution.

Remember that we are working with development snapshots, as opposed to released version.

A tip: Keep the build tree around. When updating the source code, the Makefiles will only build the changed object files, resulting in a much shorter build time.


4. Hurd Domain Name
3 May 2000 (2 posts) Subject: " domain"
People: Daniel De KokKalle Olavi Niemitalo

Daniel De Kok asked why the Hurd didn't have its own internet domain name. He asked,
It seems is registered by someone in Istanbul, is he somebody of the HURD people?!?
Kalle Olavi Niemitalo replied,
This has been asked before. See the thread at:


5. 'bcopy()' In The Hurd
3 May 2000 (2 posts) Subject: "bcopy"
People: Marcus BrinkmannTomasz Wegrzanowski

Tomasz Wegrzanowski noticed that 'bcopy()' was used in the Hurd instead of 'memcpy()', and asked if there was any reason for this, and if it was OK to send a patch to fix it. As far as the reasoning behind it went, Marcus Brinkmann replied:

Well, I don't know if there is such a reason. Probably Thomas spent too much time in the BSD sources :)

But let me point out that bcopy is equivalent to memmove, not memcpy (which doesn't handle overlapping regions).

As for accepting patches, he added, "I can't answer that for the core developers, however, is there a particular reason beside cosmetic? Of course, if you bother to find out where memcpy is sufficient instead memmove/bcopy, it might be a small performance advantage."


6. FAT Filesystem Translator
3 May 2000 - 5 May 2000 (15 posts) Subject: "Major GNU/Hurd using report"
Topics: FS: FAT
People: Marcus BrinkmannTomasz Wegrzanowski

Tomasz Wegrzanowski had a lot of questions. Among them, he asked if there was a 'fatfs' translator, a question that was first covered in Issue #43, Section #1  (2 Apr 2000: Hurd Success; Accessing FAT Filesystems; Listing Active Translators) . This time, Marcus Brinkmann had some news, and replied:

I am currently writing one! It can already list the root directory and stat some files, however, it is hopelessly buggy and no files can be read (I/O error). However, I did only work for a couple of hours on it, so there is hope.

I will make the source public in a couple of days, so people can join me. (Need to clear it up before this, and attach proper copyright notices).

I only aim at read-only without long file names. Long file names might follow, but I will not be able to implement write support in the near future. However, FAT12/16/32 will be supported.

Tomasz volunteered to test the code, and Marcus replied, "Great. I hope to release it tonight. I got it working for FAT12 so far, and if I didn't do a stupid mistake, it should work on any other format as well."


7. Porting Linux Networking Daemons
4 May 2000 - 5 May 2000 (3 posts) Subject: "Hurd networking"
Topics: POSIX
People: Marcus BrinkmannTomasz Wegrzanowski

Tomasz Wegrzanowski asked if Linux networking daemons would work unchanged under the Hurd, and Marcus Brinkmann pointed out, "Couldn't say they are special "Linux" net daemons..." and added regarding portability:

It depends a lot on the details. If they are fairly unixish, they will run without or with only slight modifications. Beside the usual POSIX incompatibilities you find everywhere, the Hurd lacks the net/* interface ioctls, so you only can listen to all interfaces. (As there is only one interface supported....) This will change of course, but currently that's the way it is.

Porting any of these applications should be pretty straightforward, if they don't compile out of the box. Let us know if you had problems, and report patches to the authors then :)


8. Difficulty Viewing
5 May 2000 (1 post) Subject: ""
People: Sujit Mathew

Sujit Mathew complained, "The site does not render correctly on Netscape 4.72. The left hand corner with the links that say Mailing List..etc. is overlapping over the links that say About this site, HOWTO, etc. I am running the latest netscape. Any ideas?" There was no reply, but by KC Debian Hurd press time, the problem had not yet been fixed.


9. New Hurd Web Page
6 May 2000 (1 post) Subject: "another site for HURD"
People: Ognyan Kulev

Ognyan Kulev gave a pointer to his personal Hurd page, and explained, "My intend is to update it every time I find something that is worth enough for helping others to use and develop HURD." There was no reply.







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.