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 #22 For 3 Nov 1999

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 51 posts in 139K.

There were 22 different contributors. 14 posted more than once. 7 posted last week too.

The top posters of the week were:


1. Hurd Documentation
24 Oct 1999 - 25 Oct 1999 (5 posts) Archive Link: "Hurd Documentation"
Topics: Bootloaders
People: OKUJI YoshinoriBill WhiteBrent Fulgham

Umut Gokbayrak was anxious to help the Hurd project, and not being a coder, he wanted to write some documentation. Brent Fulgham welcomed him aboard, and Bill White said he (Bill) was also working on documentation. He and OKUJI Yoshinori had been working together on the GRUB manual, and Bill planned on filling out the Hurd Reference Manual when he'd finished with GRUB, or perhaps starting a Hurd tutorial. He also felt that a System Administration Manual would be a good idea. Mark Andreas Meyer suggested a "GNU/Hurd Network Administrator's Guide" along the lines of the Linux Documentation Project. Bill felt it was still a bit early in the Hurd's history for such an ambitious project, and felt that planning should be held off until the Hurd became less of a moving target.


2. Will The Hurd Replace Linux?
27 Oct 1999 - 28 Oct 1999 (8 posts) Archive Link: "[ Should I continue to study linux-based GNU?]"
People: John TobeyChristopher BrowneNorbert Nemec

Dion E. Viglione heard that the Hurd would eventually be replacing Linux, and asked about this on the list. John Tobey replied, "It is not at all certain that the Hurd will replace Unix and Linux. The Hurd has a very different underlying design from Linux, so they will both be better at some things and worse at others. Therefore, I expect Linux or something similar to live long after the Hurd becomes useful." Christopher Browne gave his opinion:

Right now it is not at all clear what is likely.

Right now, Hurd is not far enough done for it to be clear whether it will either:

  1. Flourish, or
  2. Wither.

It's not sufficiently functional for it to be "more useful than Linux" for practical purposes Right Now. At this point, while there may be a few things you can do with Hurd that you can't with Linux, there are a whole lot *more* applications Linux supports that Hurd doesn't.

That needs to be couched with the caveat "Yet."

Two issues then arise:

  1. Is Hurd going to develop to the point where it can support a system relatively comparable in functionality to systems based on Linux?
  2. Is there some compelling purpose that Hurd can serve that Linux inherently *cannot?*

The latter is the critical question; if there is a compelling application, then people will come. If there isn't, then it is reasonably likely that a main merit of Hurd may be in helping Debian to become more portable and more supportive of "kernels other than Linux."

Which is of value; if Hurd lays the path to help it to be possible for there to be a successor OS to Linux, that *is* of merit.

If Hurd provides an experimenting ground that spurs on work on other improvements, that's valuable. There are few other multiserver microkernel systems, and probably fewer instances deployed than there are of Hurd.

Norbert Nemec's opinion was that the only way the Hurd could have a future was if it could become a drop-in replacement for Linux, i.e. if one could take a Linux system, drop in the Hurd kernel and a few core libraries, and have a Hurd system. He added that the learning curve for switching between them had to be zero as well.


3. Using Local Time In The Hurd
27 Oct 1999 - 29 Oct 1999 (5 posts) Archive Link: "Timezone question"
People: Marcus Brinkmann

Roman Belenov asked how to tell the Hurd that his system was set to local time, and Marcus Brinkmann replied that the Hurd didn't support local time yet. He added that if the Hurd could handle it, this would normally be handled by sysvinit, which was not yet complete in the Hurd. Elsewhere, he went on to say that he had had a port ready, but the maintainer had had objections (which Marcus agreed with), so the port had to be somewhat reworked. But he reiterated that even with a working sysvinit, the Hurd was not yet able to support local time.


4. The Hurd On UFS
28 Oct 1999 (2 posts) Archive Link: "Hurd on UFS"
Topics: FS: ext2
People: Roland McGrathIgor Khavkine

Igor Khavkine had experienced filesystem corruption with the Hurd on ext2fs, so he tried installing it on a UFS partition that he'd created with a FreeBSD fixme disk. The installation went fine, but after installation the system would hang at boot. He asked if UFS could be the problem, or if he should try ext2fs again, or if there were some other way to get a stable system running.

Roland McGrath replied:

We certainly believe ufs to work fine, but the majority of people have been using ext2fs for a while and so that is what has received most attention.

Also, a couple of months ago I updated the ext2fs server to handle newer Linux ext2fs filesystem format features, and have made various other cleanups and changes; we believe that some of these changes are to blame for the filesystem corruption problems, since they were not seen before this time.

Contrarily, the ufs server's filesystem format support code has not changed at all in a long time. That probably means it is quite stable. On the other hand, there may well be new ufs format features that are used by current BSD versions that we are not compatible with. I don't know the specific details about ufs format compatibility issues that might have changed since the time our ufs code was written.







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.