Thomas Schwinge wrote:
Since the old Hurd wiki is basically no longer satisfying both content
wise and software wise and both parts are rather unfrugal to get sorted
out easily, I'm about to present an idea of a new wiki system Neal
pointed me to: URL:http://ikiwiki.kitenet.net/, which is in
It's been a long, long, time since I have had a working Hurd
installation. My old Tyan Thunder Pentium II system used an incompatible
SCSI card, and my new Asus A7N8X had simply defied all efforts of
booting any kind of gnumach until last night.
I had been fooling around with Bochs and the
Ok, I have a Classic HP Vectra Pentium 90 over in
the corner collecting dust. I have run hurd in the
past... I'm actually a minor Hurd developer.
Look here... http://cvs.ossp.org/pkg/lib/pth/ChangeLog
Search for Hilton and you will see:
*) Add Autoconf support for GNU Hurd.
[B. Douglas
Most of you probably don't remember, but I gave a good
try to hack SMP support for OSKit/Mach over the last
couple years.
I still have full serial remote debugging capabilities
via my Netwinder which I cross-compiled gdb for Hurd.
I think that I'm Debian material, and I would be honored
if a Hurd
I'm really not suprised though. I have been inactive with Hurd
development for about a year. My old SMP box died, so I rebuilt
it with an Asus A7N8X (nForce2) MoBo and and XP2400+. It runs
great and I like it etc etc...
So I figured I'd try and reinstall Hurd now that I have an IDE
drive (my
Yea, I really need to get my Hurd booting again. I'm having
some small problems with my hardware. I run a Tyan Thunder 100
with AHA-3490 pure UW-SCSI and dual PII-450. Debian's vanilla
GnuMach works ( if I remember correctly ) but OSKit-Mach hangs
and any home-made gnuMach fails also. I used to
I haven't been very active in Hurd development lately since
my latest hardware is a bit too sophisticated for gnuMach.
Anyways, here is a very small bump-mapped version of one of
my earliest Hurd banners. I've learned a few Gimp tricks
lately so expect some really awesome Hurd artwork to follow
Hi. I have had some experience building cross-gcc compilers, and the
gist of it is using the autotools configure program. Now, before
you go any further, I need to confirm that you app is running arm-linux
right? (Not CE). If so, then you will be booting an arm linux kernel
and will be using the
I use an AGP card with Hurd (GeForce2MX), but AFAIK
it just treats it like a PCI board.
- Doug
Ognyan Kulev wrote:
Danalien wrote:
I am wondering if there is now AGP-bus support för HURD?
(I first asked this question a while back)
In GNU/Hurd Hardware Compatibility Guide[1] is the answer to
Danalien wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi!
I use an AGP card with Hurd (GeForce2MX)
What a coincidence, I am also GeForce2MX(200) user :)
So can I just burn the hurd iso images, reboot, start installing?
Xor, is there a trick to get it running?
No trick. Just set XFree86
Marcus Brinkmann wrote:
On Mon, Aug 05, 2002 at 06:35:56PM -0400, B. Douglas Hilton wrote:
AFAIK, L4 just doesn't have this capability. It is faster in a
uniprocessing scenario, but cannot scale in its current state.
The L4/Hurd project is just trying to leverage existing code
from Hurd so
Double-check your grub boot parameters. There has been
some confusing stuff floating around on the net, and
incorrect boot params cause the exact symptoms you
are describing.
This is an example of how I boot my oskit-mach:
# from menu.lst
title Debian GNU/Hurd (oskit-mach)
root (hd0,0)
kernel
No way! Mach still hasn't even hit its stride. It is designed
for massive multiprocessing scalability across thousands and
millions of nodes. Unfortunately Hurd hasn't advanced enough
yet for it to really go online. Theoretically Hurder's could
link their machines into one humongous supercomputer
Jeff, I am still interested in hacking on SMP support for
OSKit-Mach, but I have been busy and otherwise occupied
for quite a while. I run a Tyan Thunder 100 Pro with dual
Pentium II 450's right now. I haven't made the big Hurd
update yet.
I have a massive development network including the SMP
Hi.
Jeff Bailey wrote:
Oskit doesn't currently claim to have SMP support - You've got a steep
curve ahead of you.
True, but I at least have a working SMP box to test things on.
Would perhaps sponsor me and I will start seriously hacking away at
OSKit-Mach again?
Being an advocate for a
Jeff I would love to help out with the websites. What can I
do? Likewise I want to become an official Hurd developer
but I am flabbergasted by the red tape and can't find any
packages to adopt.
- Doug
Jeff Bailey wrote:
On Mon, May 13, 2002 at 05:59:12PM +0200, David Nielsen wrote:
c) I want to
Sucks, but I am moving so all hacking on OSKit-Mach has
ground to a a halt for now. My Hurd is still bootable
via either GnuMach or 1cpu OSKit-Mach but thats all
for now.
My Gnu Pth (pthreads) hack for Hurd is still available
on my netwinder server at http://www.gyrodynamic.net
but I have not
Moritz Schulte wrote:
Bill Bland [EMAIL PROTECTED] writes:
and I'm confused about something: RMS says You can even use GDB to
debug the file system while the system is running--thread-specific
breakpoints allow you to...
That is right. All the Hurd server run entirely in User space which
makes
Hey thanks for the good ideas. I'll try it out the next time I put
on my artist's hat :-)
Yeah, I have layers and such, but got better results by merging
the words / thingy / hurd into a single layer, then plasmafying
it etc. To make the blues I tinkered with the chroma curves
quite a lot
That sounds pretty funny. The levitating gnu is classic, maybe
he could be a prototype. Do some kind of far-side like characture
of a gnu and clump them together in a pack :-)
Mike Warner wrote:
I think I'm going to throw my hat into this ring. My tendency is to go with
something humorous. I like
s guessing!
Cheers!
Timothy Rue wrote:
On 30-Mar-02 21:29:44 Derek L Davies [EMAIL PROTECTED] wrote: DLD This would be so sweet on the back of a t shirt! DLD "B. Douglas Hilton" [EMAIL PROTECTED] writes:
I made the colorful one of which you refer.Yo
Thanks for all the comments,
Yeah, I admit its dark, I fooled around
with the colors for hours and that was
the best I came up with so far. It is
pretty, but I admit it's hard to read.
That might make somebody take a closer
look :-)
No problem making it square or any
other shape, but some scheme
Here's a new version which is brighter, more legible,
and subtly different. It's also bigger, about 20Kb was
as small as I could make it and still have it look nice.
As before, I have just linked to the pic on my own
website here. As you can see there can be quite a
lot of variations of
I made the colorful one of which you refer.
You can get it at http://www.gyrodynamic.net
( I got the debian thingy backwards, need to
fix it someday ... or not. )
Cheers
- Doug
Mike Warner wrote:
Awhile back there was an energetic discussion about a hurd logo. I'm putting
up a new website that
I think it would be cool if they had the preamble to
the GPL on the back side. They used to have a shirt
like that but don't anymore I think.
- Doug
Moritz Schulte wrote:
Hello people,
it seems there are more people interested in Hurd shirts than I
thought - already approx. 20 (most of them came
no way to load the extra modules after the kernel,
hence the need for a grub minikernel. However, the
machine itself is very similar to a DEC Shark, which
has some OSKit support.
Port Grub to arm... yep that's a good start. I'll try it
and see how rough it looks.
- Doug
Niels Möller wrote:
B. Douglas
The reason I'm asking is I've been setting up a
Netwinder (Intel StrongARM) all weekend, and
was thinking about compiling gnumach for it
and see if it does anything.
Except, Netwinder's nettrom seems only able
to boot a linux kernel. No multiboot capability
that I am aware of.
If grub could be
Just tried to compile w/ --enable-smp=2 and here
is the result:
gcc -g -nostdlib -nostartfiles -r -o oskit-kernel.o \
-Wl,-\( kernel.o clib-routines.o -L/usr/local/lib/oskit/.. \
-loskit_clientos -loskit_c -loskit_kern -loskit_lmm -loskit_com
-loskit_smp -loskit_exec -loskit_unsupp
This is a nifty little box! Kind of like a Linux iMac, but
probably not for beginners. An auxiliary server with
nfs and tftp is pretty much a must for these things, with
no floppy or cdrom they are tough nuts to crack!
The way to recover a pre-owned netwinder without
a root password involves
assembler hacks. After all, computers are getting faster and more numerous
all the time, but CPU's have changed radically!
L8r!
Sam Dennis wrote:
Some time around 7 o'clock PM on March 12, a Tuesday, Sammy Mannaert wrote:
B. Douglas Hilton wrote:
It cheerfully complied with no problems, loads up
Unless you have a laptop or something, I would suggest picking
up a cheap ethercard on Ebay or something. 3c509's are easily
gotten for chump change. An EtherExpress would be
the Cadillac.
If you are lucky, you could copy some linux driver files into
the oskit tree and hack this file:
Well, I did some good old empirical research on
this serverboot mystery and by changing
this:-T typed ${root} $(task-create) $(task-resume)
to this:-T typed ${root-device} $(task-create) $(task-resume)
I can then get kernel to boot up to where it hangs at ext2fs.static
with the
Ok, that worked! I got the -T typed from here:
The Hurd Installation Guide
http://web.walfield.org/papers/hurd-installation-guide/english/hurd-install-guide.html#SEC5
Next, the root file system server and the |exec| server must be
loaded. This is done using Grub's boot module capability. The
Hi,
I built a --enable-smp=2 kernel-ide and have determined that
it crashes very quickly ( no real suprise ). It seems to be rebooting
in main : ldt_init() but I am not sure. For some reason my sources
got screwed up and I can't trace the program into the OSKit source.
This last time ( which
I may have just scored on a used netwinder box. If
so this brings my Debian GNU Zoo up to this:
(current)
alphatron: Debian Linux 2.0 ( laptop Compaq Contura 4/25c ) - The
original! [gave to wife]
pavilion: Debian Linux 2.1 ( HP Pavilion 8250 ) - Waaay faster than
alphatron. [on loan to wife]
Ok, I improved my debugging report. This time,
I just loaded the debugger and hit c to let it
run. It always crashed in oskit-mach/ipc/ipc_port.c
at line 1126 in function ipc_port_release_send().
Through the use of breakpoints, I found that it
fails the *third* time it runs through this function,
Hi, mission accomplished. ${root} and $(task-create) are NULL.
I don't fully understand how this works, so the ball is in your
court.
Marcus Brinkmann wrote:
Ok, so there is a bug in the boot script cleanup handler. It should
probably be more helpful. ;)
module 0: /hurd/ext2fs.static
Aha, I have found some discrepancies. My module lines are from
Neil's web page:
http://web.walfield.org/papers/hurd-installation-guide/english/hurd-install-guide.html#SEC5
Wherein he says:
Next, the root file system server and the |exec| server must be
loaded. This is done using Grub's boot
Hi. I recompiled everything and was able to boot the oskit
test kernels. Oskit-mach compiled as normal, I had to disable
unix, examples/unix etc in modules.x86.pc
It is exhibiting the same behavior as last week, and I
have captured the debugging script. Here it is.
Later,
- Doug
snip
Script
In reference to compiling oskit / oskit-mach
Should I perhaps use gcc-3.0 ?? Disable optimization ?
i.e. $ export CC=gcc-3.0
I did in fact build oskit-mach ( kernel-ide )
with gcc-3.0 but it didn't behave any differently
than the one compiled with gcc 2.95.4
I'm not sure how to proceed, so I'll
Right now I'm rebuilding oskit from the Debian sources and
hopefully I'll get it right this time. It is really frustrating that
the monster will not compile without hacks on modules.x86.pc
I would use the Debian oskit package, except then where should
the source be located for the debugging
(thanks to Debian, I didn't do anything)
So I'm too busy drinkin' tonight to do any serious hacking.
Party on dudes...
Marcus Brinkmann wrote:
On Sun, Mar 03, 2002 at 11:38:38PM -0500, B. Douglas Hilton wrote:
Program received signal SIGSEGV, Segmentation Fault.
0x00156173 in trap_from_kernel
I am trying to follow Igor's instruction at this
web page - http://hurd.dyndns.org/oskit-boot.txt
First off, the patch doesn't compile properly. There
are conflicting types between the prototype and the
function.
Here is the patch:
--- gdb-stub.diff ---Index:
Doooh! Ok, I get it. Never mind. I run the program gdb-break
on the running Hurd system.
Daniel Wagner wrote:
When you attach gdb the kernel, set a breakpoint at gdb_break_stub.
Then start you kernel (pressing 'c').
Say you want to debug a function in the kernel where you don't have
break
Well, the install from tarball went fairly smoothly. Here are a few
notes from my experience today.
- native_install did not add root to group root, so a lot of things
got installed as root.1005
- If you don't have a floppy drive, then /dev/fd causes a lot of
programs to segfault, and it is
Hi,
I worked all day attempting to recompile oskit and oskit-mach
and have yet to duplicate my astounding stroke of luck last
weekend where I managed to get oskit-mach to boot.
Sadly, it's been instant reboots all day long. I finally got it
to crash at a trap now though on my latest compile. Here
I noticed my gdb isn't properly finding the oskit sources
when I am stepping through the kernel. How should the
source tree be arranged so that it works as expected?
BTW: I allocated my 640MB IDE drive to oskit-mach
since the problem with aic78xx seems formidable right
now. That should enable me
I just did a cp -a of my /gnu partition onto oskit-mach's
new home on the IDE drive. I was able to boot my
kernel-ide with no problems. It happily fired up
and gave me a Bourne shell with TERM=dumb and
some error about /dev/console is not a console.
I assume this is normal for oskit-mach. For an
sometimes. It usually takes me a
couple tries to get it just right.
- Doug
Moritz Schulte wrote:
B. Douglas Hilton [EMAIL PROTECTED] writes:
I noticed my gdb isn't properly finding the oskit sources when I am
stepping through the kernel. How should the source tree be arranged
so that it works
Another tough question, as usual.
Since I didn't exactly cross-compile oskit-mach
or oskit, do I really need to make a i386-pc-unknown-gnu0.2
cross debugger for oskit-mach? That target really isn't supported
too well on the debian-hppa flavor of gdb.
However, if I could say, use an
You know, I'm not sure I really need a cross-debugger at all.
It seems that gdb includes a whole bunch of targets. I'm
not sure where I got this idea that I needed to do this,
but I think it may be wrong.
- Doug
I wrote:
Since I didn't exactly cross-compile oskit-mach
or oskit, do I really need
Ok, I _do_ need to enable a cross-debugger. Here is
an excerpt from the info page for gdb.
`--target=TARGET'
Configure GDB for cross-debugging programs running on the specified
TARGET. Without this option, GDB is configured to debug programs
that run on the same machine (HOST) as GDB
It worked!
Muahhahah... now armed with my cross-debugger, my
army of evil robots will conquor the world! Er, uh,
nevermind about that. Anyways,
I built puffin's cvs source for gdb 5.0 / hppa and gave it
this rather simple configure line:
../configure --target=i686-linux
It cranked and chugged for
Ok, it actually works! I was able to attach gdb to my kernel
via ttyS0 and step into the code. That is really nifty... well
the work week is about to start again so my hacking time is
about to evaporate, but I had a really productive weekend.
Accomplishments:
1. Compiled OSKit correctly with
Ok, that looked promising for OSKit at least. It appears that the -D__ELF__
is what has been messing me up. Maybe this will work for me now...
Oh, i used --prefix=/usr/i386-gnu so that the libs and headers get installed
for the cross-compiler in the preferred place. I think this is right.
I tried
Just CVS'd the oskit mach branch, and while compiling
make broke because it could not locate string.h in
about six files located in kern/ and oskit/
I changed these as such
/* #include string.h */
#include oskit/c/string.h
Was this the right thing to do?
BTW, sadly I can compile these kernels but
I have attempted to get this OSKit-Mach development
system working for a while now, and all of my home-made
kernels either get User Trap 2 or else just reboot themselves.
However, I was able to boot Marcus's old 1999 test kernel
and it booted up until it didn't find an ide drive. Thus,
I suspect I
When I first run make, it usually crashes, but
then if I make clean and run make a second
time it will compile.
I'm having a terrible time with this, I have
compiled OSKit many times tonight and
many attempts with oskit-mach. The one
time it completed the make the kernel
instantly rebooted my
Well sure, cthreads is ancient. I'm not disputing that, and
pthreads has a lot more functionality. It is imperative that
Hurd get its native pthreads fixed up. I hope you don't
think that now that we have this cheezy Pth implementation
that people will forget about it, rather I hope it will incite
I was trying to get a pthread substitute going and tried
old pth. Apparently this has been done before and the
same problems - tests for sigstack and sigaltstack make
Hurd go out to lunch.
I think it will compile, and I am hacking configure.in
to try and disable these tests and go with the
Ok, I have enacted my changes and have created a
pretty decent patch for pth-1.4.1 which you can
get here:
http://www.geocities.com/doug_hilton/
I decided against posting it to the list as it was
around 8k when gzipped.
I have run minimal tests so far, but the pthread
test appears to be
Hi!
Well, I ran all of the test apps and they appear to work.
For my own test I compiled the web browser Dillo
which relies on pthreads extensively. Amazingly it
compiles and loads, but it appears to malfunction.
It gave an error about no shmem and loaded, but
would not follow any url's and could
. Douglas Hilton [EMAIL PROTECTED] wrote:
Hi!
Well, I ran all of the test apps and they appear to work.
I made a debian package of dmalloc recently, to do so i turned off support for
pthreads.
With it turned on it can debug pthreaded apps, maybe it would be usefull for
testing.
Ah, but the Hurd servers are already built with cthreads
which is wired right into gnumach. Any OS-critical program
should rightly be built using the facilities provided by Mach,
I think, unless that is to be somehow abstracted? My understanding
was that rather than a big monolithic kernel, the
Ah, the agony of defeat.
I'm throwing in the towel on this one. I did manage to compile gdb
for hppa using the parisc-linux.org code, but when I tried to make
the cross debugger for Hurd it wanted Mach headers for some reason.
So I copied my headers over from the Hurd box, and make just
blew up
Ok, I followed your instructions and got the same results.
First, I tried make kernel-ide which compiled ok, but
upon booting said ... Welcome to GNU Mach OSKit...??
and instantly went black and rebooted. I tried to hit the pause
key a few times to read it but I'm not fast enough.
Next,
that until you are full Jedi Kung-Foo
masters, lest you be taken by the dark side...
Thanks list!
- Doug
[1] On 09/07/2001, Matt Taggart wrote to me thusly:
B. Douglas Hilton writes...
To this end, I purchased a 715/80 box and loaded up Woody
on it thinking (perhaps naively) that this would be an easy
I'm just using the oskit in the Hurd deb package. However they
compiled it is how I have it.
- Doug
Jeroen Dekkers wrote:
On Thu, Feb 07, 2002 at 12:38:33AM -0500, B. Douglas Hilton wrote:
I do have a P-II SMP machine with 440GX. I know that SMP
is supposed to be broken, I wonder if it trys
installed and played with Debian on my 386 back in 1992!
- Doug
Jeroen Dekkers wrote:
On Thu, Feb 07, 2002 at 12:35:41PM -0500, B. Douglas Hilton wrote:
I'm just using the oskit in the Hurd deb package. However they
compiled it is how I have it.
No I'm talking about OSKit-Mach, not about OSKit
Several months ago I made a Hurd banner from a real photo which was very
cool, but had to retract it as the photographer wanted a royalty to use it.
So now, I have artistically created my own banner, similar to the other one,
using Gimp. I hereby release it for general use by anybody who
, 2002 at 08:07:22PM -0500, B. Douglas Hilton wrote:
So now, I have artistically created my own banner, similar to the other one,
using Gimp. I hereby release it for general use by anybody who wants to use
it in the spirit of the GPL.
Nice! Can you make the G more like a G and less like a C
M -0500, B. Douglas Hilton wrote:
So now, I have artistically created my own banner, similar to the other one,using Gimp. I hereby release it for general use by anybody who wants to useit in the spirit of the GPL.
Nice! Can you make the G more like a G and less like a C?Tha
kind of fun!
- Doug
James Morrison wrote:
Hi, I'm curious to see this picture, is it up on a website anywhere?
--- B. Douglas Hilton [EMAIL PROTECTED] wrote:
Several months ago I made a Hurd banner from a real photo which was
very
cool, but had to retract it as the photographer wanted a royalty
For those of you with text only email and such, I apologize for
the binary posts. As Mr. Morrison requested, I have put the
new and improved banner online. Feel free to download it.
Grumble... my whole geocites site has vanished without a
trace. Here is a link to my new page which currently has
I finally remembered that I can just use vanilla debian gnumach.gz
to boot since I switched back to the adaptec scsi. For some reason
when I compile gnumach with aic7xxx it results in a kernel panic,
but the debian flavor works fine.
Anyways, I installed grub onto the Hurd partition and chainload
Ok, I got oskit-mach from cvs no problem (in Hurd) with your
command there. I was able to configure and make with no
problem. The only warnings I noticed were this:
warning: __P redefined (stdio.h)
Unfortunately when I booted kernel it rebooted so fast
I didn't even have a chance to read any error
Yes there is some inconsistency there. Here is a
snippet of the problem:
Setting this up isn't hard at all. It always is placed in
/servers/socket/2, because that's where
glibc will look for it. So be sure to install it this way: settrans
/servers/socket/1 /hurd/pfinet --interface=eth0
RT is especially important to me. I have a variety of microcontroller
and industrial automation projects at my workplace just begging for
something like a RT Hurd. IMHO Hurd would be a great platform for this
kind of experimentation and using X11 and the not logged in account
would be ideal for
Well in my cases, the user-land programs do not require any RT awareness
other than being able to periodically empty a FIFO or somesuch. As long
as the client can check the FIFO every second or so all should be well. We
would implement the RT hardware monitoring code as a hack directly to
the
Even better would just be a chipset that integrates gnuMach in hardware.
- Doug
Jason Telfer wrote:
Will the GNU Mach microkernel ever be replaced, as far as the Hurd goes,
with a completely new microkernel written from scratch with the specific aim
of being used for the Hurd?
I have it running on a dual PII450 machine except it can
only run one cpu right now. OSKit-Mach holds promise
for SMP but it calls a trap when I boot it and I haven't been
able to debug it yet.
I suppose I ought to try and build it again now that all these
core packages have been rebuilt. I
What would be exciting would be to boot Hurd on one of those
ultra density Crusoe rackmount server machines with
336 servers... Mach was actually designed to be used on such
systems where it just goes and tries to grab CPU cycles wherever
it can find them. Does anybody know if that code is still
! That is why I want to help be a OSKit-Mach kernel hacker... the
business
potential is enormous for the engineer!
Thimk about it :-)
- Doug
Ryan Golbeck wrote:
On Mon, Oct 01, 2001 at 10:51:35PM -0400, B. Douglas Hilton wrote:
Where exactly is Kingston Ontario? I live in Bradford Pennsylvania and
I can
You're right on track. I started using Debian in 1993 ( v 0.9something )
Lets see, I'm 34 now...
34 - ( 2001-1993 ) = 26
Well, I guess you have a six year lead on me. By the time you are 34
you will be a stupendous Debian wizard if you keep at it! :-)
- Doug
James Morrison wrote:
Occationally
It is tricky. Hurd has dselect of course, but not all of the
options seem to work. If you have ethernet then that is
your best option for installing once you get bootstrapped.
Somehow I read the easy guide and also the Debian Hurd
FAQ's on the translators. One of the FAQ's will cover
how to set
FYI, I believe the correct mouse setup procedure is to use
settrans with the appropriate protocol; e.g. ps/2, but the big
Hurd secret is to set the mouse type in XF86Config to
osmouse. I posted something to this extent about adding
this info to the Debian Hurd setup FAQ a while back.
- Doug
Any ideas about how to port gnat to Hurd? I'm personally
trying to use Ada instead of C++ from now on, and gnat
would be cool to have in Hurd. The GtkAda annex is
absolutely wonderful!
The current gnat is built into gcc-2.8.1 and unfortunately
to build gnat you already need an Ada compiler as
After researching this, and from some responses that I have recieved
on this, I agree that it would be a rather pointless and also monumental
task to accomplish this. My time would be much better spent adding
more comments to the Mach sources and just working with them.
And so that's what I'm
I used to have /dev/sd1s7 mounted as /CD1,
and it worked fine. I had Hurd-F1 disk-1 there.
In Linux I did rm -r dists in that partition,
and then copied a bunch of stuff over. Now,
when booting Hurd, that partition appears empty.
When I reboot to Linux, all is well.
I have enabled the new ext2
Here's where I got the info for the mouse setup.
http://kt.zork.net/debian-hurd/dh2712_54_print.html
Well, I finally got my copy of Programming Under Mach
from an out of print book dealer. Adding this to my shiny
new copy of Debugging with GDB, I am getting dangerously
close to being able
Hi there,
Here's how you use it, first prepend /usr/i386-gnu/bin to your path, then run
it.
example:
$ cd /usr/src/myprogram
$ vi myprog.c
$ export PATH=/usr/i386-gnu/bin:$PATH
$ gcc -o myprog myprog.c
This goes for everything, just issue
$ export PATH=/usr/i386-gnu/bin:$PATH
before running
I managed to get X to start up, but lacking a serial mouse at the moment
I couldn't do much. I think I have a PS/2-serial adapter here somewhere
if I can find it.
Using the nv driver, I notice that each time I start / stop X, my console
gets progressively darker. After the fourth time the
Hi,
Here is the info you requested,
76>(thunder)[/gnu/src/oskit-mach/build]
$ sym-trace -o kernel
001fffba 0013ac4c 0013d23c 0012c4c5
Generating backtrace for object file kernel
0x01fffba in (0x01fff90) panic()
0x013ac4c in (0x013a9e0) user_trap()
0x013d23c in (0x013d236) _take_trap()
0x012c4c5
Hello!
I did another test boot and copied down the error messages.
Here they are:
... skipping GRUB stuff
Welcome to GNUmach 1.2.91-OSKit!
user trap, type 2, code = 14295c
Dump of trap-state at 0x0d6d7e80:
EAX 000e EBX 001 ECX 0dadc000 EDX 0dadc000
ESI 20584145 EDI 0d6d7eb4 EBP 30303030
Ho there,
I have the Hurd F1 set here, and a burner. I'll sell a set for
say $5 US plus any shipping. Go to www.iship.com to
find your best shipping method. My shipping address
is: Bradford, PA 16701, select carrier box @ 2 lbs 0 oz.
I can accept credit card payment through PayPal or
Yahoo! wallet
Hi list.
Sorry, I haven't had much time to Hurd-hack lately. After
coding all day at work I just want to blast some monsters and
read some Slashdot at home right now. I must say that using
the QuakeForge OpenGL client with Quake Team Fortress
and MegaTF is very fun! Who needs Half-Life TFC :-)
But it compiled with no warnings ( native Hurd compile ).
Initially it failed, MiG wanted to use i386-gnu-gcc
I did this:
# cd /usr/bin
# ln -s gcc i386-gnu-gcc
Then back to the build
# cd /src/oskit-mach/build
# ../configure --host=i386-pc-gnu
# make MAKE=make -j10
It built in a few minutes
Well, I'm hooked up with Cable-Modem internet now though... Yippee!
I have been very busy playing QuakeForge Team Fortress, you know, uh,
debugging it and stuff :-)
I have successfully compiled OSkit for Hurd and Linux, but am stumped
with oskit-mach. In both Linux-i386-gnu and native Hurd it
... and the smp sample kernel booted!
It took a grueling 75 minutes, wherein my spiffy UW SCSI
hard drive made incessant grinding crunchy noises the likes
of which I haven't heard since about 1990 with a ST225 MFM
drive! ( I can build oskit natively for Linux in 10 mins )
But it worked! I must
Hi,
For some reason or other, I couldn't build oskit-mach in Linux. It has
been a while since I tried it, so I'm not exactly sure why. I know I couldn't
seem to cross-compile oskit, it choked on memcpy or something, couldn't
link it properly. Maybe I was trying to cross-compile oskit-mach and
1 - 100 of 113 matches
Mail list logo
