Bug#661620: marked as done (Recent hd-media initrd lacks ext4 support; d-i cannot find iso on such fs)

2012-03-27 Thread Debian Bug Tracking System 
Your message dated Tue, 27 Mar 2012 15:19:28 +
with message-id 
and subject line Bug#661620: fixed in debian-installer 20120327
has caused the Debian Bug report #661620,
regarding Recent hd-media initrd lacks ext4 support; d-i cannot find iso on 
such fs
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)


-- 
661620: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=661620
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Package: debian-installer
Version: testing

I just tried to install Debian the hd-media way, with recent initrd.gz and
vmlinuz from /debian/dists/squeeze/main/installer-i386/current/images/hd-media/

This initrd lacks ext4 support, and d-i is thus unable to find its installation
media, in my case the debian-testing-i386-netinst.iso I had placed on an ext4
partition on the same disk.

I think it would be reasonable to let the inirds contain ext4 support.


--- End Message ---
--- Begin Message ---
Source: debian-installer
Source-Version: 20120327

We believe that the bug you reported is fixed in the latest version of
debian-installer, which is due to be installed in the Debian FTP archive:

debian-installer-images_20120327_amd64.tar.gz byhand
debian-installer_20120327.dsc
  to main/d/debian-installer/debian-installer_20120327.dsc
debian-installer_20120327.tar.gz
  to main/d/debian-installer/debian-installer_20120327.tar.gz
debian-installer_20120327_amd64.deb
  to main/d/debian-installer/debian-installer_20120327_amd64.deb



A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to 661...@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Otavio Salvador  (supplier of updated debian-installer 
package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmas...@debian.org)


-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Format: 1.8
Date: Tue, 27 Mar 2012 09:21:45 -0300
Source: debian-installer
Binary: debian-installer
Architecture: source amd64
Version: 20120327
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team 
Changed-By: Otavio Salvador 
Description: 
 debian-installer - Debian installer
Closes: 596596 600229 609476 610524 610906 617943 618320 621769 622187 624144 
633606 637339 637460 642183 650979 651641 661620
Changes: 
 debian-installer (20120327) unstable; urgency=low
 .
   [ Jurij Smakov ]
   * Add pata-modules to cdrom config for sparc and sparc64.
 Closes: #610906.
   * Use xz compression for initrd during netboot image build for sparc
 to get it down to acceptable size (otherwise it does not boot).
 .
   [ Samuel Thibault ]
   * Add netboot hurd-i386 package list.
   * Do not enable the Hurd console by default.
   * Stick gnumach kernel versioning to Debian style.
   * Add hurd-i386 Xen and gtk netboot image support
   * Hack hurd-i386 netboot image size to have some room for downloaded udebs.
   * Enable netcfg on hurd-i386, now having working DHCP support.
   * Add hurd-i386 cdrom and pxe support, inspired from kfreebsd.
   * Add sub-hurd d-i testing target.
   * Move gtk-common's xserver-xorg-input-evdev-udeb and
 xserver-xorg-video-fbdev-udeb to arch-specific packages. Use
 xserver-xorg-input-kbd-udeb, xserver-xorg-input-mouse-udeb and
 xserver-xorg-video-vesa-udeb for kfreebsd-i386,
 kfreebsd-amd64 and hurd-i386.
   * Add sound-modules and espeakup-udeb to i386 and amd64 gtk images.
   * Add software speech synthesis as last x86 menu choice with an 's' shortcut.
   * Widen x86 menu.
   * Set default desktop to xfce on hurd-i386.
   * Fix "extras" option in build/boot/x86/xen/debian.cfg. Closes: #637460.
 Thanks Gedalya for the report.
   * Rename amdspkgtk.cfg x86 menu item into amdspgtk.cfg to fit 8+3 oldiness.
   * Add usb-serial-modules to armel images.
   * Expand libbogl-dev dependency to !linux-any, as it is now available
 everywhere for the font reduction tool.
   * Include console-setup-pc-ekmap on Linux gtk images only. Include
 console-setup-pc-ekbd and console-setup-freebsd-charmaps-udeb instead on
 kFreeBSD gtk images (Closes: #600229).
 .
   [ Joey Hess ]
   * Initial changes to target wheezy.
 .
   [ Christian Perrier ]
   * Add needed characters for Uyghur
   * Add needed characters for Ossetian
   * Add needed characters for Sindhi
   *

Bug#661620: Recent hd-media initrd lacks ext4 support; d-i cannot find iso on such fs

2012-02-28 Thread Daniel 
Package: debian-installer
Version: testing

I just tried to install Debian the hd-media way, with recent initrd.gz and
vmlinuz from /debian/dists/squeeze/main/installer-i386/current/images/hd-media/

This initrd lacks ext4 support, and d-i is thus unable to find its installation
media, in my case the debian-testing-i386-netinst.iso I had placed on an ext4
partition on the same disk.

I think it would be reasonable to let the inirds contain ext4 support.



-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120228151942.ga10...@kelebek.lublin.se

Re: Results of the second run of research about ext4 support in bootloaders

2009-07-03 Thread Guido Günther 
On Tue, Jun 02, 2009 at 06:58:14PM +0200, Stefano Canepa wrote:
> Hi all,
>   I continued my research about ext4 support in bootloaders and this is
> the result till now:
> ext4 support
> 
> * grub: amd64, i386, powerpc w/ ubuntu patch
> * grub2: amd64, i386, powerpc
> * lilo-installer: amd64 i386 
> 
> Looks without ext4 support
> 
> * aboot: alpha
> * yaboot: powerpc
> * arcboot: mips
Arcboot statically links against e2fslib so this shouldn't be much of an
issue.

> * silo: sparc 
> * delo: mipsel
Same here IIRC.
 -- Guido


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: Results of the second run of research about ext4 support in bootloaders

2009-06-23 Thread Wartan Hachaturow 
2009/6/2 Stefano Canepa :
> Looks without ext4 support
>
>    * yaboot: powerpc

BTW, while I am at it: yaboot's second stage uses libext2fs to access
ext2/ext3 filesystems, so I guess (since modern e2fsprogs seem to
include ext4 support) it should be okay with ext4. I didn't test it
myself yet, though.

-- 
Regards, Wartan.


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: Results of the second run of research about ext4 support in bootloaders

2009-06-23 Thread Wouter Verhelst 
On Mon, Jun 22, 2009 at 11:17:04AM +, Etienne Lorrain wrote:
> 
>  Hello,
> 
>  May I ask what would it take to have the Gujin bootloader added to
>  this list, URL: http://gujin.sourceforge.net.  It is a "newer"
>  bootloader, but there isn't any Debian package for it,

That would be a first step. Find someone with experience in packaging
software for Debian (or figure it out yourself -- it's not that hard)
and try to get it uploaded.

Feel free to ask on the debian-ment...@lists.debian.org list for help,
should you need it.

-- 
The biometric identification system at the gates of the CIA headquarters
works because there's a guard with a large gun making sure no one is
trying to fool the system.
  http://www.schneier.com/blog/archives/2009/01/biometrics.html


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: Results of the second run of research about ext4 support in bootloaders

2009-06-22 Thread Etienne Lorrain 

 Hello,

 May I ask what would it take to have the Gujin bootloader added to this list, 
URL: http://gujin.sourceforge.net.
 It is a "newer" bootloader, but there isn't any Debian package for it,
I have never made a Debian package before. Just fetching the source code and 
typing "make" will produce the "./gujin" executable on ia32 and amd64, and this 
executable is able to make partition, disk, disk image bootable.
 It does support quite a few filesystems, including EXT4 as long as it has 
constant inode size (dealing with one file with a different inode size means 
scanning all inodes at "mount time").

 Different bootloader do things differently, anyone can find good and bad 
points about any of them - let's point to one thing Gujin can do to improve 
Debian:

Problem: a lot of small PC no more have a CD/DVD drive to save space, and a lot 
of them only boot the "superfloppy" format out of USB thumb disks or SD cards 
(i.e. the PC will not boot if there is a partition in the partition table, 
usually meaning C/H/S description is in the boot record).

Tested solution to install Debian on such a PC:
- download the small ISO image: debian-501-i386-businesscard.iso
- download and rebuild Gujin, or download the precompiled (ia32):
http://www.mirrorservice.org/sites/download.sourceforge.net/pub/sourceforge/g/gu/gujin/install-2.6.tar.gz/install/gujin?extract=true
  or (amd64)
http://www.mirrorservice.org/sites/download.sourceforge.net/pub/sourceforge/g/gu/gujin/install-2.6.tar.gz/install/gujin64?extract=true
- type "./gujin debian-501-i386-businesscard.iso", it will modify the first 512 
bytes (previously empty) of the image and add approx 300 Kbytes at the end of 
the image, nothing else.
- type (USB disk as /dev/sdg will be fully erased, replace sdg as needed):
umount /dev/sdg*
cat debian-501-i386-businesscard.iso > /dev/sdg
sync
That USB superfloppy will boot (for instance an EeePc) as El-Torito, start the 
usual installation system - the only trick from there is in the installer 
"Detect and mount CD-ROM":
 -> No common CD-ROM drive was detected
 Load CD-ROM drivers from removeable media: No
 Manually select a CD-ROM module and device: No
 Inatallation step failed.
 Control Alt F2: return
 mount -t iso9660 /dev/sdb /cdrom
 Control Alt F5
 -> Restart install at "load installer compoments from CD"

Obviously this problem could be solved so that Debian would distribute a dual 
superfloppy/ISO image, where user would only need to do:
cat debian-501-i386-businesscard.iso > /dev/sdg ; sync

That would also simplify the docs at:
http://d-i.alioth.debian.org/manual/en.i386/ch04s03.html
mostly because those instructions generates a device with a partition table, 
which will not boot on the kind of PC described (for instance my EeePc).

 Regards,
 Etienne.





--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: Results of the second run of research about ext4 support in bootloaders

2009-06-03 Thread Otavio Salvador 
Hello,

2009/6/2 Stefano Canepa :
[...]
> Looks without ext4 support
>
>    * aboot: alpha
>    * yaboot: powerpc
>    * arcboot: mips
>    * silo: sparc
>    * delo: mipsel
>    * sibyl: mips, mipsel
>    * prep: powerpc
>    * palo: hppa

What people think we ought to do in this case?

I think that if user selected ext4 for the partition that will have
/boot we ought to show a error message so user can react to it. I'm
unsure if it has been dealed by current patches commited by Colin but
I think it is not.

-- 
Otavio Salvador  O.S. Systems
E-mail: ota...@ossystems.com.br  http://www.ossystems.com.br
Mobile: +55 53 9981-7854 http://projetos.ossystems.com.br


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: Results of the second run of research about ext4 support in bootloaders

2009-06-02 Thread Martin Michlmayr 
* Stefano Canepa  [2009-06-02 18:58]:
> Looks without ext4 support

colo (mipsel) is probably missing from that list.

> * delo: mipsel

delo is not used in d-i anymore (the devices supported by delo are no
longer supported by d-i or the Debian kernel).
-- 
Martin Michlmayr
http://www.cyrius.com/


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Results of the second run of research about ext4 support in bootloaders

2009-06-02 Thread Stefano Canepa 
Hi all,
I continued my research about ext4 support in bootloaders and this is
the result till now:
ext4 support

* grub: amd64, i386, powerpc w/ ubuntu patch
* grub2: amd64, i386, powerpc
* lilo-installer: amd64 i386 

Looks without ext4 support

* aboot: alpha
* yaboot: powerpc
* arcboot: mips
* silo: sparc 
* delo: mipsel
* sibyl: mips, mipsel 
* prep: powerpc
* palo: hppa

I based my search on the debian-installer list of packages that install
a boot loader if there is obsolete packages among this list or if I
forgot some please tell me.

I searched through mailing list archives and sometimes into code. Many
of the bootloaders in the second list looks unmaintained.

Bye
Stefano

-- 
Stefano Canepa aka sc: s...@linux.it - http://www.stefanocanepa.it
Three great virtues of a programmer: laziness, impatience and hubris.
Le tre grandi virtù di un programmatore: pigrizia, impazienza e
arroganza. (Larry Wall)


signature.asc
Description: Questa è una parte del messaggio firmata digitalmente

Re: ext4 support

2009-01-21 Thread Colin Watson 
On Sat, Jan 10, 2009 at 02:29:14PM -0200, Otavio Salvador wrote:
> Christian Perrier  writes:
> > If the ext4 support is a quite long term work, shouldn't it be
> > listed in the wiki
> > (http://wiki.debian.org/DebianInstaller/SqueezeGoals...which is
> > currently a copy of the Lenny goals list, with done things removed)
> > and maybe tracked either in the wiki or in the BTS with a meta-bug?
> 
> Yes, it makes sense. 
> 
> I do like the idea of meta bug since it makes really easy to spot what
> is still missing. What people things about it?

I'm more comfortable with the wiki for relatively small things like
this; I added it to DebianInstaller/SqueezeGoals. Feel free to convert
it to a bug with dependencies if you want, though.

-- 
Colin Watson   [cjwat...@debian.org]


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: ext4 support

2009-01-21 Thread Colin Watson 
On Sat, Jan 10, 2009 at 02:11:34AM +0100, Frans Pop wrote:
> On Saturday 10 January 2009, Colin Watson wrote:
> > I realised I'd forgotten about the kernel udebs (we do those
> > differently in Ubuntu so it slipped my mind). Should I just add ext4 to
> > kernel-wedge/modules/ext3-modules, by analogy with the way I added ext4
> > support to partman-ext3? Although again it perhaps feels a little odd,
> > it would involve the smallest source change, would avoid a pass through
> > NEW for kernel udebs on all architectures when we get round to
> > uploading all this, and generally seems as if it'd save hassle.
> 
> It would have been nice to have a bit more time to comment on such a 
> substantial patch, but well.

I didn't think it was actually very complicated or non-obvious, despite
the size, which is why I went ahead.

> I would strongly suggest to at least not actually _show_ ext4 as an option 
> as long as not all blockers are fixed, i.e. not yet list it in 
> valid_filesystems (IIRC). That will allow us to upload partman-ext3 if 
> needed without having to worry about confusing users.

OK; done.

-- 
Colin Watson   [cjwat...@debian.org]


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: ext4 support

2009-01-10 Thread Otavio Salvador 
Christian Perrier  writes:

> Quoting Frans Pop (elen...@planet.nl):
>
>> It would have been nice to have a bit more time to comment on such a 
>> substantial patch, but well.
>> 
>> I would strongly suggest to at least not actually _show_ ext4 as an option 
>> as long as not all blockers are fixed, i.e. not yet list it in 
>> valid_filesystems (IIRC). That will allow us to upload partman-ext3 if 
>> needed without having to worry about confusing users.
>
> If the ext4 support is a quite long term work, shouldn't it be listed
> in the wiki
> (http://wiki.debian.org/DebianInstaller/SqueezeGoals...which is
> currently a copy of the Lenny goals list, with done things removed)
> and maybe tracked either in the wiki or in the BTS with a meta-bug?

Yes, it makes sense. 

I do like the idea of meta bug since it makes really easy to spot what
is still missing. What people things about it?

-- 
O T A V I OS A L V A D O R
-
 E-mail: ota...@debian.org  UIN: 5906116
 GNU/Linux User: 239058 GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
-
"Microsoft sells you Windows ... Linux gives
 you the whole house."


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: ext4 support

2009-01-09 Thread Christian Perrier 
Quoting Frans Pop (elen...@planet.nl):

> It would have been nice to have a bit more time to comment on such a 
> substantial patch, but well.
> 
> I would strongly suggest to at least not actually _show_ ext4 as an option 
> as long as not all blockers are fixed, i.e. not yet list it in 
> valid_filesystems (IIRC). That will allow us to upload partman-ext3 if 
> needed without having to worry about confusing users.

If the ext4 support is a quite long term work, shouldn't it be listed
in the wiki
(http://wiki.debian.org/DebianInstaller/SqueezeGoals...which is
currently a copy of the Lenny goals list, with done things removed)
and maybe tracked either in the wiki or in the BTS with a meta-bug?


-- 





signature.asc
Description: Digital signature

Re: ext4 support

2009-01-09 Thread Frans Pop 
On Saturday 10 January 2009, Colin Watson wrote:
> I realised I'd forgotten about the kernel udebs (we do those
> differently in Ubuntu so it slipped my mind). Should I just add ext4 to
> kernel-wedge/modules/ext3-modules, by analogy with the way I added ext4
> support to partman-ext3? Although again it perhaps feels a little odd,
> it would involve the smallest source change, would avoid a pass through
> NEW for kernel udebs on all architectures when we get round to
> uploading all this, and generally seems as if it'd save hassle.

It would have been nice to have a bit more time to comment on such a 
substantial patch, but well.

I would strongly suggest to at least not actually _show_ ext4 as an option 
as long as not all blockers are fixed, i.e. not yet list it in 
valid_filesystems (IIRC). That will allow us to upload partman-ext3 if 
needed without having to worry about confusing users.


signature.asc
Description: This is a digitally signed message part.

Re: ext4 support

2009-01-09 Thread Colin Watson 
On Thu, Jan 08, 2009 at 04:42:36PM +, Colin Watson wrote:
> As you'll see, I decided to tweak partman-ext3 rather than create a
> new partman-ext4 component: I don't think the differences between ext3
> and ext4 from the installer point of view justify a whole new udeb.
> 
> Given that we won't be looking at d-i uploads aimed at Squeeze for a
> while anyway, I'd expect there to be plenty of time to get the patches
> above applied (and if they aren't then NMUs would probably not be out of
> order). Thus, unless anyone objects, I propose to go ahead and apply the
> attached patch to trunk soon in order to make it easier for interested
> people to experiment.

Otavio said in /query on IRC that he thought the d-i bits of this looked
OK, so I've gone ahead and committed them to trunk.

I realised I'd forgotten about the kernel udebs (we do those differently
in Ubuntu so it slipped my mind). Should I just add ext4 to
kernel-wedge/modules/ext3-modules, by analogy with the way I added ext4
support to partman-ext3? Although again it perhaps feels a little odd,
it would involve the smallest source change, would avoid a pass through
NEW for kernel udebs on all architectures when we get round to uploading
all this, and generally seems as if it'd save hassle.

-- 
Colin Watson   [cjwat...@ubuntu.com]


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: ext4 support

2009-01-09 Thread Colin Watson 
On Thu, Jan 08, 2009 at 04:42:36PM +, Colin Watson wrote:
> I've attached the necessary d-i patches (against current Ubuntu, but
> I'll do any necessary work to merge into Debian) to this mail.

I actually revised the patches to be against d-i trunk before sending
this mail, but forgot to rewrite this paragraph.

I missed a few bits, in iso-scan, os-prober, partconf (not that I care
much, but hey, it's in the codebase, and we still use some of its
utilities here and there), and rescue. Attached.

-- 
Colin Watson   [cjwat...@ubuntu.com]
Index: packages/iso-scan/debian/iso-scan.postinst
===
--- packages/iso-scan/debian/iso-scan.postinst	(revision 56948)
+++ packages/iso-scan/debian/iso-scan.postinst	(working copy)
@@ -103,7 +103,7 @@
 hw-detect iso-scan/detect_progress_title || true
 
 # Load up every filesystem known to man. The drive could have anything.
-FS="ext2 ext3 reiserfs fat vfat xfs iso9660 hfsplus hfs ntfs"
+FS="ext2 ext3 ext4 reiserfs fat vfat xfs jfs iso9660 hfsplus hfs ntfs"
 for fs in $FS; do
 	modprobe $fs >/dev/null 2>&1 || true
 done
Index: packages/os-prober/os-probes/init/common/10filesystems
===
--- packages/os-prober/os-probes/init/common/10filesystems	(revision 56948)
+++ packages/os-prober/os-probes/init/common/10filesystems	(working copy)
@@ -2,7 +2,7 @@
 # Make sure filesystems are available.
 set +e	# ignore errors from modprobe
 
-FILESYSTEMS='ext2 ext3 reiserfs xfs jfs msdos vfat ntfs minix hfs hfsplus qnx4 ufs'
+FILESYSTEMS='ext2 ext3 ext4 reiserfs xfs jfs msdos vfat ntfs minix hfs hfsplus qnx4 ufs'
 # The Ubuntu kernel udebs put a number of filesystem modules in
 # fs-{core,secondary}-modules. It's fairly cheap to check for these too.
 FILESYSTEMS="$FILESYSTEMS fs-core fs-secondary"
Index: packages/partconf/mkfstab.c
===
--- packages/partconf/mkfstab.c	(revision 56948)
+++ packages/partconf/mkfstab.c	(working copy)
@@ -92,7 +92,9 @@
 		dummy->options = strdup(options);
 	} else {
 		if((strcmp(dummy->mountpoint, "/") == 0) &&
-		  ((strcmp(dummy->typ, "ext2") == 0) || (strcmp(dummy->typ, "ext3") == 0))) {
+		  ((strcmp(dummy->typ, "ext2") == 0) ||
+		   (strcmp(dummy->typ, "ext3") == 0) ||
+		   (strcmp(dummy->typ, "ext4") == 0))) {
 			dummy->options = strdup("errors=remount-ro");
 		} else {
 			dummy->options = strdup("defaults");
Index: packages/partconf/mountpoint.c
===
--- packages/partconf/mountpoint.c	(revision 56948)
+++ packages/partconf/mountpoint.c	(working copy)
@@ -14,6 +14,7 @@
 const char *fs_list[] = {
 "ext2",
 "ext3",
+"ext4",
 "fat16",
 "fat32",
 "hfs",
Index: packages/partconf/partconf.c
===
--- packages/partconf/partconf.c	(revision 56948)
+++ packages/partconf/partconf.c	(working copy)
@@ -591,7 +591,7 @@
 };
 
 /* FIXME: How can we tell which file system modules to load?  */
-char *file_system_modules[] = {"ext2", "ext3", "reiserfs", "jfs", "xfs", NULL};
+char *file_system_modules[] = {"ext2", "ext3", "ext4", "reiserfs", "jfs", "xfs", NULL};
 
 debconf = debconfclient_new();
 debconf_capb(debconf, "backup");
Index: packages/rescue/debian/rescue-mode.postinst
===
--- packages/rescue/debian/rescue-mode.postinst	(revision 56948)
+++ packages/rescue/debian/rescue-mode.postinst	(working copy)
@@ -66,6 +66,7 @@
 
 try_load_module ext2
 try_load_module ext3
+try_load_module ext4
 try_load_module jfs
 try_load_module reiserfs
 try_load_module xfs

Re: ext4 support

2009-01-09 Thread Otavio Salvador 
Colin Watson  writes:

> On Thu, Jan 08, 2009 at 09:00:41PM -0200, Otavio Salvador wrote:
>> Colin Watson  writes:
>> > parted:
>> >   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=511224
>> >   (fairly basic support but enough to get the job done)
>> <...>
>> > Comments?
>> 
>> Parted has the detection code support in master branch. I could take
>> a look and port it to stable. That way we could have it as upstream
>> and avoid diverting from it.
>
> The patch committed upstream is strictly inferior to the last one I
> posted to the upstream bug, as far as I can tell.

Let's take it upstream and merge the current support with your
patch. That way we can have it there and avoid diverting.

Thanks by looking at it.

-- 
O T A V I OS A L V A D O R
-
 E-mail: ota...@debian.org  UIN: 5906116
 GNU/Linux User: 239058 GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
-
"Microsoft sells you Windows ... Linux gives
 you the whole house."


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: ext4 support

2009-01-09 Thread Colin Watson 
On Fri, Jan 09, 2009 at 10:54:36AM +, Colin Watson wrote:
> On Thu, Jan 08, 2009 at 09:00:41PM -0200, Otavio Salvador wrote:
> > Parted has the detection code support in master branch. I could take
> > a look and port it to stable. That way we could have it as upstream
> > and avoid diverting from it.
> 
> The patch committed upstream is strictly inferior to the last one I
> posted to the upstream bug, as far as I can tell.

IOW, I've posted to the upstream list suggesting that we merge my patch
there, rather than the other way round. As I posted to the Trac ticket:

  rishi and I took fairly similar approaches, though the arguments I
  used to _ext2_generic_probe differ. The main additions in my patch are
  that I've fleshed out the superblock with reference to current ext4,
  I've added (untested) 64-bit block counts support, and that I've added
  some guards to make sure that bits of libparted that haven't been
  updated for ext4 throw an exception to avoid people inadvertently
  trashing their filesystems.

-- 
Colin Watson   [cjwat...@ubuntu.com]


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: ext4 support

2009-01-09 Thread Colin Watson 
On Thu, Jan 08, 2009 at 09:00:41PM -0200, Otavio Salvador wrote:
> Colin Watson  writes:
> > parted:
> >   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=511224
> >   (fairly basic support but enough to get the job done)
> <...>
> > Comments?
> 
> Parted has the detection code support in master branch. I could take
> a look and port it to stable. That way we could have it as upstream
> and avoid diverting from it.

The patch committed upstream is strictly inferior to the last one I
posted to the upstream bug, as far as I can tell.

-- 
Colin Watson   [cjwat...@debian.org]


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: ext4 support

2009-01-08 Thread Otavio Salvador 
Colin Watson  writes:

> parted:
>   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=511224
>   (fairly basic support but enough to get the job done)
<...>
> Comments?

Parted has the detection code support in master branch. I could take
a look and port it to stable. That way we could have it as upstream
and avoid diverting from it.

-- 
O T A V I OS A L V A D O R
-
 E-mail: ota...@debian.org  UIN: 5906116
 GNU/Linux User: 239058 GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
-
"Microsoft sells you Windows ... Linux gives
 you the whole house."


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

ext4 support

2009-01-08 Thread Colin Watson 
I've been working on adding ext4 support to d-i for Ubuntu. It's
dependent on the following bugs external to d-i:

e2fsprogs:
  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=511207
  (missing mkfs.ext4 symlink in e2fsprogs-udeb)

grub:
  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=511121
  (I understand that grub2 also supports ext4, but applying the patch
  above would decouple ext4 support from the grub2 transition.)

klibc:
  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=510758
  (fstype fails to detect ext4)

parted:
  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=511224
  (fairly basic support but enough to get the job done)

I've attached the necessary d-i patches (against current Ubuntu, but
I'll do any necessary work to merge into Debian) to this mail. As you'll
see, I decided to tweak partman-ext3 rather than create a new
partman-ext4 component: I don't think the differences between ext3 and
ext4 from the installer point of view justify a whole new udeb.

Given that we won't be looking at d-i uploads aimed at Squeeze for a
while anyway, I'd expect there to be plenty of time to get the patches
above applied (and if they aren't then NMUs would probably not be out of
order). Thus, unless anyone objects, I propose to go ahead and apply the
attached patch to trunk soon in order to make it easier for interested
people to experiment.

Comments?

-- 
Colin Watson   [cjwat...@ubuntu.com]
Index: packages/partman/partman-auto/lib/recipes.sh
===
--- packages/partman/partman-auto/lib/recipes.sh	(revision 57133)
+++ packages/partman/partman-auto/lib/recipes.sh	(working copy)
@@ -80,7 +80,7 @@
 max=$min
 			fi
 			case "$4" in # allow only valid file systems
-			ext2|ext3|xfs|reiserfs|jfs|linux-swap|fat16|fat32|hfs)
+			ext2|ext3|ext4|xfs|reiserfs|jfs|linux-swap|fat16|fat32|hfs)
 fs="$4"
 ;;
 			*)
@@ -336,7 +336,7 @@
 		max=$3
 		fs=$4
 		case "$fs" in
-		ext2|ext3|linux-swap|fat16|fat32|hfs)
+		ext2|ext3|ext4|linux-swap|fat16|fat32|hfs)
 			true
 			;;
 		*)
Index: packages/partman/partman-ext3/check.d/nomountpoint_ext3
===
--- packages/partman/partman-ext3/check.d/nomountpoint_ext3	(revision 57133)
+++ packages/partman/partman-ext3/check.d/nomountpoint_ext3	(working copy)
@@ -20,8 +20,11 @@
 		[ -f $id/acting_filesystem ] || continue
 		filesystem=$(cat $id/acting_filesystem)
 		case "$filesystem" in
-		ext3)
+		ext3|ext4)
 			[ ! -f "$id/mountpoint" ] || continue
+			RET=''
+			db_metaget partman/filesystem_short/"$filesystem" description || RET=''
+			[ "$RET" ] || RET="$filesystem"
 			db_subst partman-ext3/no_mount_point PARTITION "$num"
 			db_subst partman-ext3/no_mount_point FILESYSTEM "$filesystem"
 			db_subst partman-ext3/no_mount_point DEVICE $(humandev $(cat device))
Index: packages/partman/partman-ext3/fstab.d/ext3
===
--- packages/partman/partman-ext3/fstab.d/ext3	(revision 57133)
+++ packages/partman/partman-ext3/fstab.d/ext3	(working copy)
@@ -13,7 +13,7 @@
 		method=$(cat $id/method)
 		filesystem=$(cat $id/acting_filesystem)
 		case "$filesystem" in
-		ext3)
+		ext3|ext4)
 			[ -f "$id/mountpoint" ] || continue
 			mountpoint=$(cat $id/mountpoint)
 			# due to #249322, #255135, #258117:
@@ -31,7 +31,7 @@
 			else
 pass=2
 			fi
-			echo "$path" "$mountpoint" ext3 $options 0 $pass
+			echo "$path" "$mountpoint" $filesystem $options 0 $pass
 			;;
 		esac
 	done
Index: packages/partman/partman-ext3/finish.d/aptinstall_ext3
===
--- packages/partman/partman-ext3/finish.d/aptinstall_ext3	(revision 57133)
+++ packages/partman/partman-ext3/finish.d/aptinstall_ext3	(working copy)
@@ -14,7 +14,7 @@
 		[ -f $id/method -a -f $id/acting_filesystem ] || continue
 		filesystem=$(cat $id/acting_filesystem)
 		case $filesystem in
-		ext3)
+		ext3|ext4)
 			ext3=yes
 			;;
 		esac
Index: packages/partman/partman-ext3/debian/partman-ext3.templates
===
--- packages/partman/partman-ext3/debian/partman-ext3.templates	(revision 57133)
+++ packages/partman/partman-ext3/debian/partman-ext3.templates	(working copy)
@@ -1,13 +1,13 @@
 Template: partman-ext3/progress_checking
 Type: text
 # :sl1:
-_Description: Checking the ext3 file system in partition #${PARTITION} of ${DEVICE}...
+_Description: Checking the ${TYPE} file system in partition #${PARTITION} of ${DEVICE}...
 
 Template: partman-ext3/check_failed
 Type: boolean
 # :sl2:
 _Description: Go back to the menu an