Hurd Traffic #110 For 1 Oct 2001

By Paul Emsley

Mach 4 (http://www.cs.utah.edu/projects/flux/mach4/html/) | Hurd Servers (http://www.gnu.org/software/hurd/hurd.html) | Debian Hurd Home (http://www.gnu.org/software/hurd/debian-gnu-hurd.html) | Debian Hurd FAQ (http://www.debian.org/ports/hurd/hurd-faq) | debian-hurd List Archives (http://lists.debian.org/#debian-hurd) | bug-hurd List Archives (http://mail.gnu.org/pipermail/bug-hurd/) | help-hurd List Archive (http://mail.gnu.org/pipermail/help-hurd/) | Hurd Reference Manual (http://www.gnu.org/software/hurd/reference-manual.html) | Hurd Installation Guide (http://pick.sel.cam.ac.uk/~mcv21/hurd.html) | Cross-Compiling GNUMach (http://pages.hotbot.com/sf/igorkh/gnumach-cross.txt) | Hurd Hardware Compatibility Guide (http://www.urbanophile.com/arenn/hacking/hurd/hurd-hardware.html)

Table Of Contents

1. More On The Naming Of The System

27 Sep 2001 - 28 Sep 2001 (3 posts) Archive Link: "Antwort: Re: Debian GNU or Debian GNU/Hurd?"

People: Richard Stallman

There was some more discussion of the name of the system also continued on help-hurd () , there was a notable contribution from Richard Stallman put an end to it:

GNU and GNU/Hurd are two names for the same system; the latter name emphasizes the fact that the kernel is the Hurd. It is useful to emphasize that when making a contrast with GNU/Linux. In this context, we want to make that emphasis.

Yes [Debian GNU/Hurd] is its official name, and the reason is that this is a place where it is useful to make the contrast with GNU/Linux.

2. Network Block Device Client

27 Sep 2001 (3 posts) Archive Link: "nbd-server"

Topics: FS: NFS

People: Roland McGrathMarcus Brinkmann

Wouter Verhelst realised that his Network Block Device client (nbd-client) had not appeared in the archive, so he wanted to know how to find out what went wrong in the build.

Marcus Brinkmann posted the build-log as it was not exposed via the web (the initial problems are that linux/nbd.c and sys/mount.h are not available).

Roland took a look at the code and said: "it's so simple that I just implemented the client side for the Hurd. That is, I've added an "nbd" store type to libstore. " Roland suggested that Hurd hackers should check out this code and test it, but noted "nbd cannot currently be used for paging on the Hurd. (We only support paging to portions of local disk, including files on disk.) But the issues are well-understood here and we already know what implementation work is required to make it possible to do paging over the network in general. (i.e., if we did the work necessary to e.g. page to NFS files, that would also make it possible to use nbd for paging.)"

3. AMD 400MHz K6 Boot Problems

22 Sep 2001 - 28 Sep 2001 (20 posts) Archive Link: "Kernel Divide error trap"

People: Marcus BrinkmannMark PaulusDaniel Wagner

Piotr Krukowiecki was having problems booting his AMD K6 400-based system. Marcus Brinkmann told him to install the gnumach-dbg 20010918-1 package and then use the /boot/gnumach-dbg.gz kernel. There followed analysis of gnumach's microfind(). There was much puzzlement as the behaviour of the system did not seem to correspond to the code, Marcus thinking "so, may it be that the loop in tenmicrosec was executed faster than the clock could go down? But that would be trifle strange as we have run the Hurd on mach faster machines than an AMD K6 400MHz and not have this bug (and any faster machine should show the same bug). I have personally run the Hurd on a Pentium III with 1GHz, and it worked just fine. "

Mark Paulus provided a clue suggesting "Could it be a compiler optimization issue, where the empty loop is simply optimized away, and therefore never executed?"

Daniel Wagner also contributed with: "that's only one side of the truth. AMD had in those K6 series a very fast loop optimation. It was so fast, that Windows couldn't boot correctly, because they were also counting down a counter. The result was used as a dividend which caused and division by zero. AMD removed this optimation later, since most people thought it was AMD's fault and not MS's... It might be the same here. "

Marcus Brinkmann was pleased with this information "I guess that the code is really broken on such processors heavily optimizing at run time" and posted a patch.

Indeed Microsoft had found (what seems to be) similar problems and have provided a patch (http://www.microsoft.com/windows95/downloads/contents/WURecommended/S_WUServicePacks/AMDPatch/Default.asp) for Window 95.

There was, as yet, no reply from Piotr.

4. Booting Strangeness

22 Sep 2001 - 23 Sep 2001 (3 posts) Archive Link: "I'm a bit pouched"

Topics: FS: ext2

People: James MorrisonNeil WalfieldMarcus Brinkmann

James Morrison was having problems booting after having ugraded gnumach and hurd. After some advice from Marcus Brinkmann, he changed his serverboot.gz to that from the F2 cd and could boot. He then had ext2fs crash: "Again I could not use serverboot.gz as a module. I got the same error as before. However, I was able to use the F2 cd again and fsck the filesystem. Then voila both my original serverboot and the one from the F2 cd worked. My question is why this would happen? serverboot is only used on boot right? So how would it change while the system is running? "

Neil Walfield thought this behaviour strange "All that I can think of is bad ram" , he said.

5. Replacing Serverboot Pager Legacy Interface

27 Sep 2001 (3 posts) Archive Link: "default pager using legacy interfaces"

People: Marcus BrinkmannThomas Bushnell

The default pager in serverboot is currently using the legacy interfaces memory_object_set_attributes() and memory_object_data_write(). Marcus Brinkmann wanted to change them: "the newer documentation suggests that memory_object_ready() and memory_object_data_return() should be used." Marcus posted a patch for consideration.

Thomas Bushnell wrote:

In general I have no problem with this. Switching to memory_object_ready means that you have to commit to using the new interface entirely, and IIRC there are several changes to worry about. I don't think any of them affect the default pager.

But I would feel happier if you compiled this and did some major stress-testing of it first.

 

 

 

 

 

 

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 kernel.org. All pages on this site are copyright their original authors, and distributed under the terms of the GNU General Public License version 2.0.