Da Zheng, le Sat 03 Jan 2009 14:33:38 +, a écrit :
Can we use Java in Hurd now?
Bootstraping a java environment is a pain. I had succeeded to build
some gcj at some point, but there were some bugs and then a new release
came, breaking other things, etc. So it should be possible, but it's
Praveen A, le Sat 03 Jan 2009 17:51:28 -0800, a écrit :
2009/1/3 Da Zheng :
It's bad to hear that.
but are there any other Java compiler we can use? jikes is an alternative,
but it seems not to support all features required by classpath while the
virtual machine, kaffe, seems to require
hurd-i386 got a /proc procfs filesystem and procps now compiles with a
few patch, which thus solves the issue.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Hello,
Praveen A, le Sat 03 Jan 2009 17:51:28 -0800, a écrit :
2009/1/3 Da Zheng :
It's bad to hear that.
but are there any other Java compiler we can use? jikes is an alternative,
but it seems not to support all features required by classpath while the
virtual machine, kaffe, seems to
Thomas Schwinge, le Wed 22 Apr 2009 08:55:03 +0200, a écrit :
... and indeed:
tschwi...@flubber:~ $ ls -l /usr/lib/gcc/i486-gnu/4.3.3/libgcc_s.so
-rw-r--r-- 1 root root 0 Apr 17 01:14
/usr/lib/gcc/i486-gnu/4.3.3/libgcc_s.so
I remember something that Michael or Samuel fixed this
Hello,
Where is wireless-tools-gnumach maintained? Ideally, that package
should provide an implementation of the libiw library, used by
network-manager, used by libproxy, used by libsoup2.4, used by gvfs,
used by libgnome2.4, needed for all the gnome stuff. Else we can ask
for dropping the
mino...@brunitika.ch, le Fri 01 May 2009 08:49:22 -0600, a écrit :
I had exactly the bad magic number error with a partition up to 128
Gb.
2^28 * 2^9 = 2^37 = 2^7 * 2^30, that's the LBA28 limit. To overcome that
we need to implement LBA48 in Mach.
Samuel
--
To UNSUBSCRIBE, email to
Hello,
libasound-dev/libasound2-dev build-deps are painful. Fortunately, OSS
(opensound.com) provides libsalsa, a compatibility layer between OSS
and ALSA: it is ABI-compatible with libasound (actually using alsa's
headers), and just makes OSS ioctls. I've prepared a package that
builds it, and
Arne Babenhauserheide, le Sat 16 May 2009 21:17:18 +0200, a écrit :
On Saturday, 16. May 2009 15:24:04 Samuel Thibault wrote:
libasound-dev/libasound2-dev build-deps are painful. Fortunately, OSS
(opensound.com) provides libsalsa, a compatibility layer between OSS
and ALSA: it is ABI
Cyril Brulebois, le Sat 16 May 2009 16:55:30 +0200, a écrit :
Samuel Thibault samuel.thiba...@ens-lyon.org (16/05/2009):
If people think it's a good idea, maybe we can setup some VCS
somewhere to work on it? (where/in which Alioth project?)
Sounds nice. I for one would use collab-maint
Oh
Package: wnpp
Severity: wishlist
Owner: Samuel Thibault sthiba...@debian.org
* Package name: oss-libsalsa
Version : 4.1-build1052b
Upstream Author : 4Front Technologies www.4front-tech.com
* URL : http://developer.opensound.com/sources/
* License : BSD
Hello,
A lot of packages depend on libusb, and more and more do. It seems to
me illusory to try to make them all include [!hurd-i386] conditionals,
it'd be easier to just manage to compile libusb, even if it's only to
always report that no device is available.
Now the problem is that PATH_MAX
Carl Fredrik Hammar, le Sun 17 May 2009 21:18:57 +0200, a écrit :
Would this field even be used at runtime given that we don't have
any devices?
Not used for now, but the code still needs to be compiled.
In that case we could just remove the PATH_MAX.
That'd prevent from actually providing
Carl Fredrik Hammar, le Sun 17 May 2009 21:48:01 +0200, a écrit :
On Sun, May 17, 2009 at 09:23:38PM +0200, Samuel Thibault wrote:
Carl Fredrik Hammar, le Sun 17 May 2009 21:18:57 +0200, a écrit :
In that case we could just remove the PATH_MAX.
That'd prevent from actually providing USB
Carl Fredrik Hammar, le Mon 18 May 2009 15:07:49 +0200, a écrit :
Hi,
On Sun, May 17, 2009 at 09:57:26PM +0200, Samuel Thibault wrote:
Carl Fredrik Hammar, le Sun 17 May 2009 21:48:01 +0200, a écrit :
On Sun, May 17, 2009 at 09:23:38PM +0200, Samuel Thibault wrote:
Carl Fredrik Hammar
Carl Fredrik Hammar, le Fri 22 May 2009 14:43:39 +0200, a écrit :
On Tue, May 19, 2009 at 12:19:10AM +0200, Samuel Thibault wrote:
Carl Fredrik Hammar, le Mon 18 May 2009 15:07:49 +0200, a écrit :
A similar problem happens when defining LIBUSB_PATH_MAX, but apps will
only break when
Carl Fredrik Hammar, le Fri 22 May 2009 15:55:06 +0200, a écrit :
Well, it seems my two cents are all used up. It will be interesting to
see which solution you finally go for.
I'll go the pointer way, as it permits to avoid recompiling applications
whenever we implement some usb support. We
Hello,
Barry deFreese, le Sat 23 May 2009 00:01:17 -0400, a écrit :
@@ -235,7 +235,7 @@
const char *s;
uid_t uid;
charbuf[256];
-charttybuf[PATH_MAX];
+char*ttybuf;
DBusError error;
dbus_bool_t
Marc-André Lureau, le Tue 26 May 2009 00:19:19 +0300, a écrit :
On Sun, May 17, 2009 at 7:53 PM, Samuel Thibault sthiba...@debian.org wrote:
Description : Libsalsa is a library providing an ALSA interface on top
of OSS
It's unfortunate, but libsalsa is also the name of the tiny ALSA
Barry deFreese, le Mon 25 May 2009 22:47:42 -0400, a écrit :
+#ifdef (__GLIBC__) || _POSIX_VERSION = 200809L
#if defined(__GLIBC__) || _POSIX_VERSION = 200809L
Else, should be fine.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of unsubscribe.
Nathan Gray, le Wed 27 May 2009 14:11:57 -0400, a écrit :
Warning: xrdb command not found; X resources not merged
Mmm, not having the x11-xserver-utils package installed leads to a lot
of kinds of problems, yes.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a
Barry deFreese, le Wed 27 May 2009 23:09:29 -0400, a écrit :
This block of code in Src/cond.c seems to be the initial issue with
zsh-beta:
Yes it is.
#if defined(GET_ST_MTIME_NSEC) defined(GET_ST_ATIME_NSEC)
if (!(st = getstat(left)))
return 1;
Barry deFreese, le Thu 28 May 2009 11:18:03 -0400, a écrit :
_GNU_SOURCE is defined but I don't think that brings in __USE_MISC,
right?
It should, see features.h: _GNU_SOURCE makes _SVID_SOURCE defined to 1,
which makes __USE_MISC defined to 1.
Samuel
--
To UNSUBSCRIBE, email to
Nathan Gray, le Wed 27 May 2009 20:08:40 -0400, a écrit :
On Wed, May 27, 2009 at 11:30:28PM +0200, Samuel Thibault wrote:
Nathan Gray, le Wed 27 May 2009 14:11:57 -0400, a écrit :
Warning: xrdb command not found; X resources not merged
Mmm, not having the x11-xserver-utils package
Nathan Gray, le Sat 30 May 2009 10:46:53 -0400, a écrit :
On Sat, May 30, 2009 at 12:35:32AM +0200, Samuel Thibault wrote:
Do you have the debian-ports repository in your sources. I just realize
that a critical libxcb fix is there and not (yet) in the main archive.
I do have an entry
Hello,
Ok, making the zalloc area bigger needs more virtual memory, the
attached patch should fix it.
Samuel
Index: b/i386/i386at/model_dep.c
===
--- a/i386/i386at/model_dep.c 10 Nov 2008 15:18:47 - 1.9.2.19
+++
Thomas Schwinge, le Tue 16 Jun 2009 14:15:55 +0200, a écrit :
tschwi...@flubber:~ $ ls -l /usr/bin/cmp /usr/bin/diff /usr/bin/diff3
/usr/bin/sdiff
-rw-r--r-- 1 root root 18692 May 30 01:34 /usr/bin/cmp
-rw-r--r-- 1 root root 64284 May 30 01:34 /usr/bin/diff
-rw-r--r-- 1 root
Michael Banck, le Tue 16 Jun 2009 22:51:19 +0200, a écrit :
a member of the Debian Qt/KDE packaging team asked me about the
80_hurd_max_path.diff patch in qt4-x11 today. It looks like it is
obsolete these days due to Alfredo Beaumont's work.
Indeed, the code doesn't use PATH_MAX when
Hello,
Barry deFreese, le Tue 26 May 2009 10:47:47 -0400, a écrit :
I can't seem to get anywhere with openjdk-6 at the moment. But if
someone with lots of memory/swap space wants to try it, attached are the
changes I made to debian/rules.
You probably have also changed autoconf stuff,
Thomas Schwinge, le Fri 19 Jun 2009 18:04:32 +0200, a écrit :
I guess that
this may be a bug with fakeroot, I've seen bugs with it from times to
times.
But not something that is reliably reproducible, I guess?
Well, it happend a second time now, I don't know.
Samuel
--
To
Xueqian Zhao, le Mon 22 Jun 2009 22:28:58 +0800, a écrit :
Which arch should I choose if any?
The simplest (read: oldest) possible, starting from the 386 with 16MB
RAM. PCI could still be a better idea.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of
Package: fakeroot
Version: 1.12.2
Severity: important
Hello,
We're getting FTBFSs on hurd-i386 because of /usr/bin/diff getting
644 instead of 755. This is apparently due to a race in fakeroot-tcp
triggered by install -s, which does the following:
- copy the target file, which is now 600
-
sthiba...@debian.org, le Fri 26 Jun 2009 13:28:45 +0200, a écrit :
-rw-r--r-- 1 root root 64140 Jun 23 21:18 /usr/bin/diff
?!
Pff... I'll see when I have some time.
Ok, you've probably seen my bug against fakeroot. I've added a sleep in
install -s as a hack to fix it. +b3 is OK now (really
Archive Administrator, le Mon 29 Jun 2009 16:32:13 +, a écrit :
Rejected: hurd_20090404-1.dsc: syntax error on line 25.
Rejected: hurd_20090404-1.dsc: no Files: field.
Graaah.
Ok, never mind, let's rebuild from start...
Samuel
--
To UNSUBSCRIBE, email to
Bruno Minotti, le Wed 08 Jul 2009 12:35:11 +0200, a écrit :
To boot with more than 768Mb RAM after install with crosshurd you have to
change the virtual memory of mach. To do this Samuel Thibauld wrote a patch to
apply on gnumach:
This is pending for next upload of gnumach.
Samuel
--
To
olafbuddenha...@gmx.net, le Mon 17 Aug 2009 07:47:19 +0200, a écrit :
As for Wine, I don't see any fundamental reason why it couldn't work
on the Hurd -- probably just needs some fixing here and there...
Yep, IIRC the buildd currently fails on a PATH_MAX issue.
Samuel
--
To UNSUBSCRIBE,
Thomas Schwinge, le Tue 18 Aug 2009 22:20:54 +0200, a écrit :
Or is the glibc code already supposed to handle that correctly?
It does, else kill(getpid(), 0) would fail too. See
_hurd_internal_post_signal's if (signo == 0)
Samuel
--
To UNSUBSCRIBE, email to
Hello,
Yavor Doganov, le Sun 23 Aug 2009 10:57:14 +0300, a écrit :
I prepared a patch, and I kindly ask you to confirm if it is sane.
It looks sane, yes, thanks for taking care of it :)
+OS_CPPFLAGS += -DPATH_MAX=1024 -DMAXPATHLEN=1024
Is this a proper way to solve the problem?
Nope.
Philip Charles, le Sun 20 Sep 2009 13:14:33 +1200, a écrit :
Are these still the entries for menu.lst for booting the Hurd?
Yes, but avoid newlines between lines, else the \ won't manage to join
them.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of
Philip Charles, le Sun 20 Sep 2009 21:25:18 +1200, a écrit :
Would those who have installed the Hurd from scratch in the last few weeks
let me know how they did it.
I usually just use crosshurd.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of
Philip Charles, le Sun 20 Sep 2009 23:16:56 +1200, a écrit :
On Sun, 20 Sep 2009, Samuel Thibault wrote:
Philip Charles, le Sun 20 Sep 2009 21:25:18 +1200, a écrit :
Would those who have installed the Hurd from scratch in the last few
weeks let me know how they did it.
I usually just
Philip Charles, le Sun 20 Sep 2009 23:16:56 +1200, a écrit :
On Sun, 20 Sep 2009, Samuel Thibault wrote:
Philip Charles, le Sun 20 Sep 2009 21:25:18 +1200, a écrit :
Would those who have installed the Hurd from scratch in the last few
weeks let me know how they did it.
I usually just
Yes, deboostrap seems to be working atm.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Michael Banck, le Sun 20 Sep 2009 14:51:00 +0200, a écrit :
Does debootstrap work for the cross-strap use-case (i.e., running it on
GNU/Linux to produce a GNU/Hurd chroot)? I thought #498731 needed to be
fixed first,
Ah, possibly yes. As a matter of fact I only used it for chroots, where
Philip Charles, le Mon 21 Sep 2009 01:36:43 +1200, a écrit :
On Mon, 21 Sep 2009, Samuel Thibault wrote:
Yes, deboostrap seems to be working atm.
Playing catchup and making progress. Problem, what is the root password
for a new installation.
I guess an empty pasword should be fine
Matthias Klose, le Sun 20 Sep 2009 19:58:22 +0200, a écrit :
I'm not sure about Java, which show some regressions compared to 4.3. Otoh
it's not amymore the default java compiler for all architectures except
hurd(?)
I haven't explicitely tested but I don't know any issue with gcj-4.4 on
Bruno Minotti, le Mon 21 Sep 2009 17:52:15 +0200, a écrit :
Philip Charles wrote:
I cannot find a password that will enable me to log into my new Hurd
installation. Installed using crosshurd.
Any ideas?
I had the same problem and I solved mounting the Hurd Partition on
GNU/Linux and
Michael Banck, le Mon 21 Sep 2009 20:59:55 +0200, a écrit :
On Mon, Sep 21, 2009 at 06:08:21PM +0200, Samuel Thibault wrote:
apparently crosshurd sets x in the passwd field and doesn't build a
shadow file indeed.
Well, AFAICT it copies over /usr/share/base-passwd/passwd.master
Philip Charles, le Tue 22 Sep 2009 18:00:25 +1200, a écrit :
Is the /usr . symlink still the standard on Debian GNU/Hurd systems?
Nope, it poses too many issues that we can't necessarily fix.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of
Hello,
Hiroyuki Yamamoto, le Tue 22 Sep 2009 19:23:15 +0900, a écrit :
Your package fails to build from source on hurd.
Attached is a patch which fixes the issue.
Better just disable rdev build in the Makefile, that's what we got
commited upstream some time ago (but apparently it got lost at
Philip Charles, le Tue 22 Sep 2009 23:33:41 +1200, a écrit :
Good progress except for the above. The new tarball build is based on
crosshurd. From start to mini iso ready to burn, 40 seconds.
I will prepare a report in a day or so.
Cool! :)
I am sticking with grub legacy in the
മഹേഷ് മുകുന്ദന് | Mahesh M, le Tue 22 Sep 2009 11:35:39 -0100, a écrit :
What is the status of grub in hurd?
Hasn't changed.
Can anybody please fix it if it has not yet been.
It's in TODO lists since a long time, yes.
I think i was told that only little more has to be done.
But the little
Hiroyuki Yamamoto, le Tue 22 Sep 2009 22:17:57 +0900, a écrit :
Samuel Thibault wrote:
Hiroyuki Yamamoto, le Tue 22 Sep 2009 19:23:15 +0900, a écrit :
Your package fails to build from source on hurd.
Attached is a patch which fixes the issue.
Better just disable rdev build in the Makefile
Hello,
I've just tried the upgrade to grub2, and although os-prober find my
hurd partition, update-grub2 says
GNU/Hurd is not yet supported by grub-mkconfig.
could someone work on it?
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of unsubscribe.
Michael Banck, le Sat 26 Sep 2009 11:55:48 +0200, a écrit :
It now got reuploaded and should be back over the next few days,
hopefully. As db4.* always fails in the testsuite due to locking
problems, manual intervention is needed here.
I'm doing it.
Samuel
--
To UNSUBSCRIBE, email to
Philip Charles, le Sat 26 Sep 2009 12:23:48 +1200, a écrit :
We have libdb4.2, 4.3, 4.5, 4.6, 4.8, in the archive, but no 4.7
Rebuilt and uploaded, should be available in next mirror push.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of unsubscribe.
Hello,
I have found the bug (asm memory clobber...) which was making us use
gcc-4.2, it will be fixed in next upload.
Thanks,
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Hello,
I've just uploaded a fixed version of xorg-server: stupid fix, just
exchaning two function parameters...
It seems to work for me in qemu with the cirrus driver. Could people
check it out? (Note: version 1.6.3.901-1+hurd.1, should be available
within a few hours, not just 1.6.3.901-1)
Hello,
Kurt Roeckx, le Sun 27 Sep 2009 17:07:35 +0200, a écrit :
I think at some point in time there was a request to move hurd
back to buildd.debian.org? What happened to this? Do you still
want to move there?
That'd be a good thing yes, at least to expose build logs more clearly.
If you
Samuel Thibault, le Sun 27 Sep 2009 17:45:12 +0200, a écrit :
I don't know for beethoven (it is down atm)
Michael thinks beethoven can be left out for the moment.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas
Here it is again, signed.
Samuel Thibault, le Sun 27 Sep 2009 17:45:12 +0200, a écrit :
192.33.98.55 for rossini
80.87.129.151 for both bach mozart (they are Xen domains)
ssh-dss
B3NzaC1kc3MAAACBAOp+ROMyuV/cJp5abowxGStdasV5cvi6rAc/ft1AVHkIBxIKEqUB3dWLRX3+wCuCNPNofR6KkEAcoiGL1zUMlzb
Hello,
Philip Charles, le Mon 28 Sep 2009 10:37:48 +1300, a écrit :
Almost complete.
Cool!
Login problem. Bruno Minotti's /usr/share/base-passwd/passwd.master
works. This can be implemented during the build process. This should be
seen as a temporary measure.
Someone needs to check why
Michael Banck, le Mon 28 Sep 2009 00:08:45 +0200, a écrit :
On Sun, Sep 27, 2009 at 11:44:59PM +0200, Samuel Thibault wrote:
netbase configuration awaiting resolution.
Should just be a matter of fixing the init.d headers. Anybody?
That doesn't look Hurd-specific though, yet I do
Hello,
Matthias Klose, le Wed 23 Sep 2009 00:39:52 +0200, a écrit :
On 20.09.2009 22:20, Samuel Thibault wrote:
Matthias Klose, le Sun 20 Sep 2009 19:58:22 +0200, a écrit :
I'm not sure about Java, which show some regressions compared to 4.3. Otoh
it's not amymore the default java compiler
Hello,
Matthias Klose, le Wed 23 Sep 2009 00:39:52 +0200, a écrit :
Then try to build packages natively (ecj, gjdoc),
install the corresponding -gcj packages und make sure that the natively
compiled code is picked up at runtime and things still work.
It seems to be working yes.
finally we
Philip Charles, le Sat 03 Oct 2009 21:19:12 +1300, a écrit :
The Hurd cannot access files toward the end of a DVD.
How far to the end? It may just be the 2GB or 4GB limit.
BTW, the Makedev fix is already in both the upstream source and the
debian repository.
Thanks!
Samuel
--
To
Hello,
mozart and bach are back online, I could generate rsa keys for them,
here they are, could you install them for
buildd_hurd-i...@buildd.debian.org ?
ssh-rsa
Thomas Schwinge, le Tue 13 Oct 2009 10:16:15 +0200, a écrit :
On Tue, Oct 13, 2009 at 08:50:58PM +1300, Philip Charles wrote:
On Tue, 13 Oct 2009, Samuel Thibault wrote:
Philip Charles, le Mon 12 Oct 2009 23:52:32 +1300, a écrit :
1. Does entropy still have to be copied to /dev/urandom
Philip Charles, le Tue 13 Oct 2009 20:50:58 +1300, a écrit :
2. Configuring the network
ten days ago settrans -fgap /servers/socket/2 /hurd/pfinet -i eth0
\ -a 192.168.1.10 -g 192.168.1.4 -m 255.255.255.0 -4
worked nicely, now it is broken. pfinet --usage says that it
should be
Philip Charles, le Tue 13 Oct 2009 23:20:50 +1300, a écrit :
OK I will still provide entropy. Is /dev/random needed?
Yes.
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Philip Charles, le Wed 14 Oct 2009 00:43:56 +1300, a écrit :
Maybe you have the old version (2.14~rc2-0) of the package from
debian-ports? The current unstable version (2.16.1-3) has the
/sbin/fsck file according to
http://packages.debian.org/sid/hurd-i386/util-linux/filelist
fsck
Philip Charles, le Wed 14 Oct 2009 15:29:40 +1300, a écrit :
The only available util-linux is in unreleased, 2.14~rc2-0+hurd.1 dated
2 Oct 2008.
Is there another version hiding somewhere?
It's not hiding:
$ apt-cache policy util-linux
util-linux:
Installed: 2.16.1-4
Candidate:
Philip Charles, le Thu 15 Oct 2009 00:41:33 +1300, a écrit :
New problem when configuring initscripts. 'Can't exec insserv No such
file or directory at /usr/bin/update-rc.d line 387'
insserv has not been configured and is not listed as a dependency of
initscripts - is this the cause of the
Philip Charles, le Thu 15 Oct 2009 10:00:56 +1300, a écrit :
The problem is that insserv is not configured before sysv-rc which means
that update-rc.d cannot find insserv.
Mmm, I don't understand: there's not much to be configured for insserv,
it just needs to be unpacked and /sbin/insserv
Philip Charles, le Thu 15 Oct 2009 10:50:17 +1300, a écrit :
On Thu, 15 Oct 2009, Samuel Thibault wrote:
Philip Charles, le Thu 15 Oct 2009 10:00:56 +1300, a écrit :
The problem is that insserv is not configured before sysv-rc which
means that update-rc.d cannot find insserv.
Mmm, I
Walter Franzini, le Fri 16 Oct 2009 09:48:03 +0200, a écrit :
It seems to me that the build is run under gid 1005 but a matching entry
is missing in /etc/group.
Ah, yes, I had propagated passwd only, not group. I have fixed it and
requeued
Thanks!
Samuel
--
To UNSUBSCRIBE, email to
An updated log is available on buildd.debian.org/aegis
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Walter Franzini, le Fri 16 Oct 2009 17:31:10 +0200, a écrit :
Samuel Thibault sthiba...@debian.org writes:
Walter Franzini, le Fri 16 Oct 2009 09:48:03 +0200, a écrit :
It seems to me that the build is run under gid 1005 but a matching entry
is missing in /etc/group.
Ah, yes, I had
Walter Franzini, le Fri 16 Oct 2009 23:31:35 +0200, a écrit :
The code contains the following fragment:
#ifdef __hurd__
#define WHOLE_FILE_LOCKS_ONLY
#endif
Hello,
Daniel Moerner, le Tue 18 Aug 2009 11:42:20 -0700, a écrit :
Both upstream and I are mystified by this 16392 Resource lost clause.
That will be fixed in hurd_20090404-3. What was happening is that
libpthread doesn't handle the sig==0 case and blindly uses sigaddset,
which thus overflows
Thomas Schwinge, le Sun 18 Oct 2009 12:53:33 +0200, a écrit :
I also enhanced libpthread's test-16.c with respect to that -- does
that look alright?
http://git.savannah.gnu.org/cgit/hurd/libpthread.git/log/?h=master-test
It seems so. Maybe it could be useful to also try pthread_kill on
Hello,
Philip Charles, le Mon 19 Oct 2009 00:23:21 +1300, a écrit :
I want to have a break from the Hurd iso's so I am proposing to upload the
two images as is with some modification of Cook-book. Or would people
prefer to wait two or three weeks when I will have time to polish up the
Michael Banck, le Sun 25 Oct 2009 11:20:20 +0100, a écrit :
The outdated packages now got removed. I am keeping unreleased pinned
over unstable right now, because the last important package from
debian-ports for a debootstrap is dhcp-client, which is an empty
transitional package to
I've submitted patch in various places to fix booting gnumach from
grub2. The git version of gnumach should be bootable by grub 1.97
in all circumstances (even a lot of memory etc.). The current Debian
gnumach should be bootable if given less than 800MiB memory (else grub2
may put the modules
Philip Charles, le Tue 20 Oct 2009 00:13:14 +1300, a écrit :
On Mon, 19 Oct 2009, Samuel Thibault wrote:
I'm fixing a bug that arises with grub2 and which could also happen
with grub on machines with a lot of memory. Maybe we'd rather wait for
the fix to be included in Debian.
I
Hello,
Jonathan Nieder, le Mon 19 Oct 2009 16:09:22 -0500, a écrit :
I am no expert in these things, but it looks as though the assembly
code checks for Mach-O format with __MACH__, but Hurd, which uses ELF,
defines that symbol to convey it uses the Mach microkernel. If this is
right, what
Hello,
Matthijs Kooijman, le Tue 03 Nov 2009 16:57:44 +0100, a écrit :
in the hopes that there would be a hurd buildd for experimental, but
apparently there is none.
Actually there is but with the move to buildd.debian.org and the current
ftp-master mess, it's currently down :/
Can anyone
Matthijs Kooijman, le Tue 03 Nov 2009 18:13:51 +0100, a écrit :
It fails very early:
detecting OS... none detected
D'oh! I've added a patch, but forgot to add it to the 00list file... Would you
be so kind as to try the updated version I've attached?
04-compile-on-hurd doesn't apply due
hari krishnan, le Sun 08 Nov 2009 16:00:30 +0530, a écrit :
Is there any active torrent for downloading the latest debian-hurd.
Well, I don't think such torrent would be very active, but who knows...
I am unable to download from ftp servers due to bandwidth problems(frequent
disconnection).
gordev, le Mon 09 Nov 2009 21:50:58 +0200, a écrit :
http://www.gnu.org/software/grub/manual/html_node/GNU_002fHurd.html#GNU_002fHurd
That's very outdated.
, but in /boot/ directory only two files gnumach.gz and servers.boot.
And thus it can't be used, servers.boot is _not_ the same as
Walter Franzini, le Fri 16 Oct 2009 17:31:10 +0200, a écrit :
The error is now more interesting:
aegis: fcntl(/tmp/5465/lib/lockfile, F_SETLK, {type = F_WRLCK, whence =
SEEK_SET, start = 1, len = 1, pid = 0}): Not
olafbuddenha...@gmx.net, le Sun 15 Nov 2009 00:10:37 +0100, a écrit :
(EE) MGA(0): Unable to map BAR 0. No such device or address (1073741830)
Are you using the debian gnumach? (which has the iopl device)
Samuel
--
To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org
with a subject
Bastian Blank, le Mon 16 Nov 2009 15:27:45 +0100, a écrit :
Hurd uses a login shell by default on the ttys. This is a security
problem as it allows access with other permissions to all ressources.
Note: to do so we simply need to create a login account with default
shell set to /bin/loginpr and
Bastian Blank, le Mon 16 Nov 2009 19:02:12 +0100, a écrit :
hurd should support Unix98 style ptys, aka /dev/ptmx and /dev/pts. This
could be implemented as a special filesystem on /dev/pts and a link from
/dev/pts/ptmx to /dev/ptmx, like current Linux is using in a special
mode.
IIRC a
olafbuddenha...@gmx.net, le Mon 16 Nov 2009 23:52:24 +0100, a écrit :
On Sun, Nov 15, 2009 at 02:21:46AM +0100, Samuel Thibault wrote:
olafbuddenha...@gmx.net, le Sun 15 Nov 2009 00:10:37 +0100, a écrit :
(EE) MGA(0): Unable to map BAR 0. No such device or address
(1073741830
olafbuddenha...@gmx.net, le Tue 17 Nov 2009 01:23:02 +0100, a écrit :
On Mon, Nov 16, 2009 at 03:27:45PM +0100, Bastian Blank wrote:
Hurd uses a login shell by default on the ttys. This is a security
problem as it allows access with other permissions to all ressources.
This is not exactly
olafbuddenha...@gmx.net, le Tue 17 Nov 2009 23:12:47 +0100, a écrit :
hal 0.5.13-4 now compiles and runs again on the Hurd. Thus changing
this tracker for further work on a native backend for the Hurd.
I think it can be safely closed: HAL is deprecated anyways;
Ugh, I didn't know that. Fun
olafbuddenha...@gmx.net, le Thu 19 Nov 2009 00:06:21 +0100, a écrit :
On Wed, Nov 18, 2009 at 01:20:33AM +0100, Samuel Thibault wrote:
olafbuddenha...@gmx.net, le Tue 17 Nov 2009 01:23:02 +0100, a écrit :
On Mon, Nov 16, 2009 at 03:27:45PM +0100, Bastian Blank wrote:
actually the access
no packages.
-- no debconf information
--
Samuel Thibault samuel.thiba...@fnac.net
$ du temp.iso
2,0Ttemp.iso
$ ls temp.iso -l
-r-xr-xr-x1 samy thibault 16E 2003-03-22 14:44 temp.iso*
-+- je vous dirai pas la marque de mon disque dur, na :p -+-
--
To UNSUBSCRIBE, email
for manipulating filesys
crosshurd suggests no packages.
-- no debconf information
--
Samuel Thibault samuel.thiba...@fnac.net
$ du temp.iso
2,0Ttemp.iso
$ ls temp.iso -l
-r-xr-xr-x1 samy thibault 16E 2003-03-22 14:44 temp.iso*
-+- je vous dirai pas la marque de mon disque dur, na :p
1 - 100 of 2552 matches
Mail list logo
