Hurd Traffic #32 For 18 Jan 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 46 posts in 128K.

There were 18 different contributors. 9 posted more than once. 8 posted last week too.

The top posters of the week were:


1. GPG Partial Success
6 Jan 2000 - 9 Jan 2000 (6 posts) Subject: "gnupg compiled"
Topics: Compiling Packages
People: Daniel WagnerMarcus Brinkmann

Daniel Wagner reported success compiling GNU Privacy Guard, but had some runtime problems: since GPG wanted to disable core dumping, it used getrlimit and setrlimit to do that. But glibc0.2 did not implement setrlimit. Daniel suggested either disabling the call with an #ifndef preprocessor condition, or else actually implementing the function. Marcus Brinkmann replied that the Hurd didn't support core dumping at the moment anyway, so this was not such an important problem. He gave a code snippet that had worked for him in 'exim', and suggested Daniel try including it in GPG. His code was a simple conditional, that would allow core dumping to magically start working if the Hurd ever started supporting it.


2. The Hurd On Slashdot
9 Jan 2000 (2 posts) Subject: "Hurd on slashdot"
People: David MadoreMarcus Brinkmann

David Madore gave a pointer to a discussion of the Hurd on Slashdot ( , about commercial availability of the Hurd. He added that although he'd been posting to the discussion, he wasn't sure his comments were 100% right. Marcus Brinkmann replied that he felt David's comments in the thread were fine, and offered to join in if he saw anyplace he could contribute.


3. Tetex Package And 'klibtool'
9 Jan 2000 - 10 Jan 2000 (6 posts) Subject: "tetex and klibtool"
Topics: Compiling Packages
People: Chris LingardMarcus Brinkmann

Marcus Brinkmann reported that 'klibtool' in the tetex package did not recognize i386-gnu as explicitly supported, and would fall back to its default behavior. He asked that the folks who compiled tetex make sure 'klibtool' recognized this option, and use the same configuration as Linux for libs. Chris Lingard replied that he and Kapil H Paranjape had been the fellows involved with compiling tetex (November 19). Chris reported that he had added code to fix Marcus' problem at the time, but explained, "I have since deleted the source directory because Debian Org flamed us when it was suggested that changes were needed. I am not sure that it could be build using an empty directory and the ../configure method. It would probably be easiest to port Kapil's (or my) binaries; and produce a Hurd (working) version of Tex/tetex in the future."

Regarding Chris' changes to tetex, Marcus was surprised, and replied, "Ah, you should have let us know. I had the impression no modifications were necessary. As a general note, it is important to tell us every single bit you needed to change, because when the time has come, we will automatically build packages unattented." Regarding Chris and Kapil getting flamed for their changes, Marcus was surprised again, saying, "I hope not! If you mean we said that you should not convert to libtool without consulting upstream, that's right. We try to miniimize our changes."

Under the Subject: tetex installed () , Marcus reported that a fixed version of tetex had been uploaded, and added:

I did not fix klibtool, so there might be "issues", but as shared libs are not build, they should be internal or small. However, we should still fix this.

Now that we have a sane tex again, I can rebuild all those silly packages which absolutely need texinfo to build. There were quite a few of them.


4. Hurd RAM Disk Translator
11 Jan 2000 - 12 Jan 2000 (2 posts) Subject: "RAM Disk"
People: Marcus BrinkmannNeal H Walfield

Just for fun, Neal H Walfield started writing a simple translator, which became a RAM disk. He gave a pointer to his code ( , and added that he hadn't been able to find any other RAM disk projects for the Hurd. A happy Marcus Brinkmann replied:

Hey cool. Even if it doesn't make its way into the Hurd source, it will be packaged for inclusion in the distribution. I will take care of that.

After all, that's the whole idea. It doesn't matter if it is in the core or not, or if the system administrator installs it or not.

When can we start the first translator contest? ;)


5. URLs For Newcomers
12 Jan 2000 - 13 Jan 2000 (2 posts) Subject: "URLs for new developers"
People: Matthew VernonJim Franklin

Jim Franklin posted a number of URLs for the benefit of Debian Hurd newcomers. He gave links to

Matthew Vernon added a link to The Easy Guide To Installing Hurd On A Linux Box ( . End Of Thread.


6. Progress Of The Hurd
13 Jan 2000 - 15 Jan 2000 (10 posts) Subject: "??? downloading from the right place?"
People: Marcus BrinkmannJim Franklin

Seth Aaron Nickell was trying to download Debian Hurd from, but there seemed to be far too many packages in that directory, given his understanding of the Hurd's progress to date. He asked if he was in the right place.

Marcus Brinkmann replied, "This only means that you underestimated our progress. :)" It also came out that a lot of the files in that directory were part of the "binary-all" group of Debian Linux packages: theoretically installable on any platform, including the Hurd. Jim Franklin added, "Unfortunately, there is no seperation between such Hurd and such Linux packages currently, so it's a huge mess. So, all files that are symlinked to the binary-all part of the tree are potentially only useful on linux platforms. But this is not true for all of them."


7. 'cross-install' Requires ncurses-base
13 Jan 2000 - 14 Jan 2000 (7 posts) Subject: "Missing ncurses-base?"
People: Marcus Brinkmann

Seth Aaron Nickell reported that 'cross-install' would complain about a missing ncurses-base, and refuse to continue. He installed libncurses4 and 5, as well as ncurses-bin_5.0, with no luck. Marcus Brinkmann replied that 'cross-install' was not uptodate. He said he'd upload a new version the following week.







We Hope You Enjoy Hurd Traffic

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.