You are not logged in.

#26 2006-05-05 05:18:35

tpowa
Developer
From: Lauingen , Germany
Registered: 2004-04-05
Posts: 2,322

Re: new initramfs, testers welcome

@ baze could you try this:
add break=y to your boot prompt and try to write something with your keyboard.
then add the usbinput hook too and retry.
thanks
greetings
tpowa

Offline

#27 2006-05-05 07:22:48

baze
Member
Registered: 2005-10-30
Posts: 393

Re: new initramfs, testers welcome

tpowa wrote:

@ baze could you try this:
add break=y to your boot prompt and try to write something with your keyboard.
then add the usbinput hook too and retry.
thanks
greetings
tpowa

- added break=y: couldn't type anything at the ramfs prompt. (without break, the keyboard works after uevents, just fyi)
- now added usbinput (and break=y)
HOOKS="base udev autodetect sata usbinput"

now i can type at the ramfs prompt. there is much more output at the boot process and some of the usb stuff is printed after the ramfs prompt, but i guess that's just because it takes time to load and is totally fine.

i use the arch stock kernel btw and usb legacy is enabled in the bios.



btw: you said to use mkinitcpio -k when building the initramfs for another kernel than the currently bootet one. does that invole jumps from 2.6.16.13 to 2.6.16.14 too or is it only for let's say 2.6.16.x to 2.6.17? the 2.6.16-ARCH is kept for 2.6.16.14, so it should work if i upgrade to .16.14 and then run mkinitcpio -g /..., right? or is it even needed?

Offline

#28 2006-05-05 07:26:35

tpowa
Developer
From: Lauingen , Germany
Registered: 2004-04-05
Posts: 2,322

Re: new initramfs, testers welcome

nah, only big kernel versions should break an initrd or total different kernels like beyond <-> stock

Offline

#29 2006-05-05 09:09:27

tomk
Forum Fellow
From: Ireland
Registered: 2004-07-21
Posts: 9,839

Re: new initramfs, testers welcome

I produced an image for the stock kernel, using

HOOKS="base udev autodetect ide filesystems"

but it contains no ide modules, and therefore fails to boot. Here's the filelist:

/proc
/sys
/dev
/dev/null
/dev/zero
/dev/console
/lib
/lib/klibc-NDtVTttYAxNaaNiOsQ6T1EIYUSo.so
/bin
/bin/cat
/bin/chroot
/bin/dd
/bin/false
/bin/fstype
/bin/gunzip
/bin/gzip
/bin/insmod
/bin/ipconfig
/bin/kill
/bin/kinit
/bin/ln
/bin/mdassemble
/bin/minips
/bin/mkdir
/bin/mkfifo
/bin/mknod
/bin/modprobe
/bin/mount
/bin/nfsmount
/bin/nuke
/bin/parseblock
/bin/pidof
/bin/pivot_root
/bin/run-init
/bin/sh
/bin/sleep
/bin/true
/bin/umount
/bin/uname
/bin/zcat
/init
/config
/sbin
/sbin/udevd
/sbin/udevtrigger
/sbin/udevsettle
/etc
/etc/udev
/etc/udev/rules.d
/etc/udev/rules.d/udev.rules
/etc/start_udev
/lib/udev
/lib/udev/modalias_ieee1394
/etc/udev/udev.conf
/hooks
/hooks/udev
/lib/modules
/lib/modules/2.6.16-ARCH
/lib/modules/2.6.16-ARCH/kernel
/lib/modules/2.6.16-ARCH/kernel/fs
/lib/modules/2.6.16-ARCH/kernel/fs/jbd
/lib/modules/2.6.16-ARCH/kernel/fs/jbd/jbd.ko
/lib/modules/2.6.16-ARCH/kernel/fs/ext3
/lib/modules/2.6.16-ARCH/kernel/fs/ext3/ext3.ko
/lib/modules/2.6.16-ARCH/kernel/fs/jfs
/lib/modules/2.6.16-ARCH/kernel/fs/jfs/jfs.ko
/lib/modules/2.6.16-ARCH/kernel/fs/reiserfs
/lib/modules/2.6.16-ARCH/kernel/fs/reiserfs/reiserfs.ko
/lib/modules/2.6.16-ARCH/kernel/fs/xfs
/lib/modules/2.6.16-ARCH/kernel/fs/xfs/xfs.ko
/hooks/filesystems
/lib/modules/2.6.16-ARCH/modules.dep
/lib/modules/2.6.16-ARCH/modules.alias
/lib/modules/2.6.16-ARCH/modules.symbols

I'll add the ide stuff to MODULES and try again, but shouldn't the autodetect pick them up?

Offline

#30 2006-05-05 13:59:56

lilsirecho
Veteran
Registered: 2003-10-24
Posts: 5,000

Re: new initramfs, testers welcome

tpowa;

Installed all elements from .pkg.tar.gz format with pacman.
Generated mkinitcpio.img using.....mkinitcpio -g /boot/mkinitcpio.img....
My present lilo :
=====================================================================#
# /etc/lilo.conf
#

boot=/dev/hda
# This line often fixes L40 errors on bootup
# disk=/dev/hda bios=0x80

default=arch
timeout=50
lba32
prompt

image=/boot/vmlinuz26
        label=arch
        root=/dev/hda3
        initrd=/boot/initrd26-full.img
        read-only
image=/boot/vmlinuz26archck
        label=archck
        root=/dev/hda3
        initrd=/boot/initramfs-2.6.15-archck.img
        read-only
image=/boot/vmlinuz26beyond
        label=2.6.16-beyond
        root=/dev/hda3
        initrd=/boot/initramfs-beyond.img
        read-only
image=/boot/vmlinuz26beyond
       label=cpio
        root=/dev/hda3
        initrd=/boot/mkinitcpio.img
        read-only
image=/boot/vmlinuz26archck
        label=cpio1
        root=/dev/hda3
        initrd=/boot/mkinitcpio.img
        read-only
#other=/dev/hda1
#        label=dos

========================================================================
The mkinitcpio.conf file:
=======================================================================
# vim:set ft=sh
# MODULES
# The following modules are loaded before any boot hooks are
# run.  Advanced users may wish to specify all system modules
# in this array.  For instance:
#     MODULES="piix ide_disk reiserfs"
MODULES=""

# BINARIES
# This setting includes, into the CPIO image, and additional
# binaries a given user may wish.  This is run first, so may
# be used to override the actual binaries used in a given hook.
# (Existing files are NOT overwritten is already added)
# BINARIES are dependancy parsed, so you may safely ignore libraries
BINARIES=""

# FILES
# This setting is similar to BINARIES above, however, files are added
# as-is and are not parsed in anyway.  This is useful for config files.
# Some users may wish to include modprobe.conf for custom module options,
# like so:
#    FILES="/etc/modprobe.conf"
FILES=""

# HOOKS
# This is the most important setting in this file.  The HOOKS control the
# modules and scripts added to the image, and what happens at boot time.
# Order is important, and it is recommended that you do not change the
# order in which HOOKS are added.  Run 'mkinitcpio -H <hook>' for
# help on a given hook.
# 'base' is _required_ unless you know precisely what you are doing.
# 'udev' is _required_ in order to automatically load modules
# 'modload' may be used in place of 'udev', but is not recommended
# 'filesystems' is _required_ unless you specify your fs modules in MODULES
# Examples:
#    This setup specifies all modules in the MODULES setting above.
#    No raid, lvm, or encrypted root is needed.
#    HOOKS="base"
#
#    This setup will autodetect all modules for your system and should
#    work as a sane default
#    HOOKS="base udev ide filesystems autodetect"
#
#    This setup will generate a 'full' image which supports most systems.
#    No autodetection is done.
     HOOKS="base udev ide scsi  usb filesystems"
#
#    This setup assembles an ide raid array with an encrypted root FS.
#    Note: See 'mkinitcpio -H raid' for more information on raid devices.
#    HOOKS="base udev ide filesystems raid encrypt"
#
#    This setup loads an LVM volume group on a usb device.
#    HOOKS="base udev usb filesystems lvm"
#HOOKS="base udev ide filesystems autodetect"
======================================================================
Request for dry run of mkinitcpio not understood...please describe how to...
Usb keyboard and mouse in system.

Question: If mkinitcpio .conf is changed, and mkinitcpio -g /boot/mkinitcpio.img is re-run, does the /boot file get changed or must it be deleted first?

Thanks for reply big_smile


Prediction...This year will be a very odd year!
Hard work does not kill people but why risk it: Charlie Mccarthy
A man is not complete until he is married..then..he is finished.
When ALL is lost, what can be found? Even bytes get lonely for a little bit!     X-ray confirms Iam spineless!

Offline

#31 2006-05-05 18:36:56

phrakture
Arch Overlord
From: behind you
Registered: 2003-10-29
Posts: 7,879
Website

Re: new initramfs, testers welcome

Ok, lvm2, encrypt, and resume hooks are verified in working order for at least one user.  Please test these if you have any of these features on your systems.

New Additions:
-L parameter to mkinitcpio - display all available hooks
(this can be used in conjunction with -H <hookname>)
-M parameter to mkinitcpio - display modules that would be autodetect

These changes can be found in the svn repo (tpowa's tarball may not be updated yet).


Lots of responses follow:

baze wrote:

but i have the same thing waldek_a reported:

kinit tries to mount /dev/root on /root with type ext2
kinit tries to mount /dev/root on /root with type reiserfs

but it works fine.

Yes, that's kinit trying to sniff out your FS type.  Adding the filesystems hook will "fix" that - it will remove the need for kinit to try and scan for the filesystem.

lilsirecho wrote:

HOOKS="base udev ide filesystems autodetect"

The order of the autodetect hook has changed.  It should now be added before the subsystems you want to autodetect.

tomk wrote:

I produced an image for the stock kernel, using

HOOKS="base udev autodetect ide filesystems"

but it contains no ide modules, and therefore fails to boot.

Odd, can you updated svn and post the output of "mkinitcpio -M"?

Offline

#32 2006-05-05 20:25:03

lilsirecho
Veteran
Registered: 2003-10-24
Posts: 5,000

Re: new initramfs, testers welcome

Changed HOOKS:

HOOKS="base udev autodetect ide usb filesystems"

I assume I have understood the change recommended but perhaps the usb is not the latest.

This change produced kernel panic: cannot open root device "303" or unknown-block (3,3)

EDIT:  Attempted to run mkinitcpio -M....."invalid option -M"


Prediction...This year will be a very odd year!
Hard work does not kill people but why risk it: Charlie Mccarthy
A man is not complete until he is married..then..he is finished.
When ALL is lost, what can be found? Even bytes get lonely for a little bit!     X-ray confirms Iam spineless!

Offline

#33 2006-05-05 20:51:36

phrakture
Arch Overlord
From: behind you
Registered: 2003-10-29
Posts: 7,879
Website

Re: new initramfs, testers welcome

lilsirecho wrote:

EDIT:  Attempted to run mkinitcpio -M....."invalid option -M"

Please update from svn and use that version:

svn co http://phraktured.net/initramfs
cd initramfs/mkinitcpio
./mkinitcpio -M

Please copy your /etc/mkinitcpio.conf config file to this directory and re-generate the image

./mkinitcpio -g /boot/my_test_image.img

When I get home tonight, I will update packages to make it easier on all of you.

Offline

#34 2006-05-06 04:12:56

phrakture
Arch Overlord
From: behind you
Registered: 2003-10-29
Posts: 7,879
Website

Re: new initramfs, testers welcome

Ok, I have moved the packages to [testing].  It will make updates easier on all of you people.

You can still use the svn if you'd like, as it may be more up-to-date than the testing packages.

If you'd like to test, please follow these simple steps:

· enable the testing repo
·

pacman -Sy mkinitcpio

· configure your /etc/mkinicpio.conf as appropriate.  See "mkinitcpio -L" and "mkinitcpio -H <hook>" for appropriate information about the HOOKS entries.
· run "mkinitcpio -g /boot/my_image_name"
· add an entry to your bootloader telling it to use this new image as the initrd
NOTE: Do not remove old entries from your bootloader, that is asking for trouble.  Add a new entry called "testing" or something

Enjoy, please report anything you find - this is still beta, so feel free to suggest any sort of feature you'd like to see.

Offline

#35 2006-05-06 10:31:11

Gullible Jones
Member
Registered: 2004-12-29
Posts: 4,863

Re: new initramfs, testers welcome

Initcpio works absolutely fine for me, in fact it seems faster than initrd. However, uevents seems to take a *lot* longer with initcpio than with initrd (about 10 seconds), strangely enough... Is klibc known for very poor performance, or something? :?

Offline

#36 2006-05-06 10:36:13

baze
Member
Registered: 2005-10-30
Posts: 393

Re: new initramfs, testers welcome

i updated the klibc, klibc-extras and klibc-udev packages to those in testing and now i ran into problems.

at first during boot i noticed the message

ATA: abnormal status 0x7F on port 0xEFA7

but the system started.
then I just updated to kernel .14-2 and rebootet.
now I got the same error 3 times in a row at bootup:

scsi0 : ata_piix
ATA: abnormal status 0x7F on port 0xEFA7
scsi1 : ata_piix

with the bootprocess stopping there, doing nothing.
I had to shut off the pc, try again. as I said, this happend 3 times in a row.
after that, I had lots of output, but unfortunately I haven't looked very close at it. It was sth about modules wink. The system hang at bootup this time too.

Then I switched of the pc, started again, this time with the initrd and this worked as usual.

edit: tpowa told me yesterday, that there would be no need to create a new initramfs image unless I have a major release change (e.g. 2.6.16 -> 2.6.17), so I didn't create a new image. I just did and then I could start arch with initramfs again, but the

scsi0 : ata_piix
ATA: abnormal status 0x7F on port 0xEFA7
scsi1 : ata_piix

messages where still present.
IIRC I never had them before I updated the klibc* packages to those from testing.

Offline

#37 2006-05-06 10:39:04

tomk
Forum Fellow
From: Ireland
Registered: 2004-07-21
Posts: 9,839

Re: new initramfs, testers welcome

phrakture wrote:
tomk wrote:

I produced an image for the stock kernel, using

HOOKS="base udev autodetect ide filesystems"

but it contains no ide modules, and therefore fails to boot.

Odd, can you updated svn and post the output of "mkinitcpio -M"?

I'm now using the [testing] packages, and the same HOOKS combination gives me

# mkinitcpio -M
Modules autodetected:
   /lib/modules/2.6.16-ARCH/kernel/fs/reiserfs/reiserfs.ko

I need to add MODULES="piix ide_disk" to get a bootable image.

Offline

#38 2006-05-06 10:48:27

Gullible Jones
Member
Registered: 2004-12-29
Posts: 4,863

Re: new initramfs, testers welcome

Hmm, this is interesting... Early in the boot process I get about five "FATAL: module [very long number] not found" messages, but they don't seem to cause any problems.

(Uevents time seems to be back to normal, no idea what was going on there on the last boot.)

Offline

#39 2006-05-06 17:53:47

Gullible Jones
Member
Registered: 2004-12-29
Posts: 4,863

Re: new initramfs, testers welcome

Ack, that's the same thing that happened with gstreamer-0.10's description when it was in Testing.

Offline

#40 2006-05-06 18:11:02

tomk
Forum Fellow
From: Ireland
Registered: 2004-07-21
Posts: 9,839

Re: new initramfs, testers welcome

In the URL, change cvsroot=Current to cvsroot=Extra i.e.

http://cvs.archlinux.org/cgi-bin/viewcv … ag=TESTING

Offline

#41 2006-05-06 21:41:01

brain0
Developer
From: Aachen - Germany
Registered: 2005-01-03
Posts: 1,382

Re: new initramfs, testers welcome

I have started a wiki page here. Please extend it whereever possible, correct errors and of course read it: http://wiki.archlinux.org/index.php/Con … mkinitcpio

Offline

#42 2006-05-06 22:19:39

tomk
Forum Fellow
From: Ireland
Registered: 2004-07-21
Posts: 9,839

Re: new initramfs, testers welcome

phrakture wrote:

feel free to suggest any sort of feature you'd like to see.

How about a module blacklist, like in rc.conf? If I have "autodetect filesystems" in HOOKS, I get modules for reiserfs and ext2, but I only need reiserfs to boot.

Offline

#43 2006-05-06 22:25:23

Gullible Jones
Member
Registered: 2004-12-29
Posts: 4,863

Re: new initramfs, testers welcome

That's rather weird, here it autodetected reiserfs alone as it was supposed to. :?

Offline

#44 2006-05-06 22:40:00

phrakture
Arch Overlord
From: behind you
Registered: 2003-10-29
Posts: 7,879
Website

Re: new initramfs, testers welcome

Gullible Jones wrote:

Initcpio works absolutely fine for me, in fact it seems faster than initrd. However, uevents seems to take a *lot* longer with initcpio than with initrd (about 10 seconds), strangely enough... Is klibc known for very poor performance, or something? :?

Performance has nothing to do with it if you're talking about uevents - uevents are handled by udev *and* are based on alot of IO.
Do you mean that the uevents from *within* early-userspace are slow, or after it runs into Arch's normal boot process?

baze wrote:
scsi0 : ata_piix
ATA: abnormal status 0x7F on port 0xEFA7
scsi1 : ata_piix

I *think* I had a similar problem, and I think this is related to the whole ata_piix/piix thing.  Instead of using 'ata_piix', try with the 'piix' module (specifying it in the MODULES array.

Also, you should rebuild your inicpio image after ever kernel install - even if the versions don't really change, it's just a good idea.

tomk wrote:
# mkinitcpio -M
Modules autodetected:
   /lib/modules/2.6.16-ARCH/kernel/fs/reiserfs/reiserfs.ko

I need to add MODULES="piix ide_disk" to get a bootable image.

Wait, so it only displays the fs modules? Are you actually running under 2.6.16-ARCH?  Can you try running depmod -ae and trying again?  mkinitcpio -M really does nothing related to the hooks, it's meerly convienance.  I have no clue why it would fail, as it's just walking the sysfs tree.

Gullible Jones wrote:

Hmm, this is interesting... Early in the boot process I get about five "FATAL: module [very long number] not found" messages, but they don't seem to cause any problems.

(Uevents time seems to be back to normal, no idea what was going on there on the last boot.)

You *shouldn't* get any modprobe output like that - what hooks are you using?  udev should not cause output like that (but I'm not 100% sure).

tomk wrote:
phrakture wrote:

feel free to suggest any sort of feature you'd like to see.

How about a module blacklist, like in rc.conf? If I have "autodetect filesystems" in HOOKS, I get modules for reiserfs and ext2, but I only need reiserfs to boot.

Well, you can already boot with disablemodules=a,b,c and disablehooks=udev,filesystems or something on the kernel command line, but ignoring modules at install time sounds like a decent idea.

Offline

#45 2006-05-06 22:52:51

iphitus
Forum Fellow
From: Melbourne, Australia
Registered: 2004-10-09
Posts: 4,927

Re: new initramfs, testers welcome

tomk wrote:
phrakture wrote:

feel free to suggest any sort of feature you'd like to see.

How about a module blacklist, like in rc.conf? If I have "autodetect filesystems" in HOOKS, I get modules for reiserfs and ext2, but I only need reiserfs to boot.

Alternatively, whitelist what you need. Add reiserfs to your MODULES="" line and then remove the filesystems hook.

Offline

#46 2006-05-06 23:06:34

Gullible Jones
Member
Registered: 2004-12-29
Posts: 4,863

Re: new initramfs, testers welcome

Err, whoops... Meant udev within the standard boot sequence ("Loading udev uevents..."). My bad. At any rate it's not a problem.

phrakture wrote:

You *shouldn't* get any modprobe output like that - what hooks are you using? udev should not cause output like that (but I'm not 100% sure).

HOOKS="base udev autodetect ide filesystems"

And the messages seem to have disappeared since I posted that. :? Doesn't really matter though, I suppose, they were harmless enough.

Offline

#47 2006-05-07 08:14:07

tomk
Forum Fellow
From: Ireland
Registered: 2004-07-21
Posts: 9,839

Re: new initramfs, testers welcome

phrakture wrote:
tomk wrote:
# mkinitcpio -M
Modules autodetected:
   /lib/modules/2.6.16-ARCH/kernel/fs/reiserfs/reiserfs.ko

I need to add MODULES="piix ide_disk" to get a bootable image.

Wait, so it only displays the fs modules? Are you actually running under 2.6.16-ARCH?

Yes, but this happens with every kernel on my system.

phrakture wrote:

Can you try running depmod -ae and trying again?

No difference.

Offline

#48 2006-05-07 09:20:32

tomk
Forum Fellow
From: Ireland
Registered: 2004-07-21
Posts: 9,839

Re: new initramfs, testers welcome

Here's an update. The problem I've described seems to be machine-specific. I've set up this laptop and a desktop with the current stock kernel 2.6.16.14-2, mkinitcpio from testing, and identical mkinitcpio.conf. On the laptop I get the response as above i.e.

# mkinitcpio -M
Modules autodetected:
   /lib/modules/2.6.16-ARCH/kernel/fs/reiserfs/reiserfs.ko

On the desktop, I get this:

# mkinitcpio -M
Modules autodetected:
   /lib/modules/2.6.16-ARCH/kernel/drivers/cdrom/cdrom.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/char/agp/agpgart.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/char/agp/intel-agp.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/i2c/busses/i2c-piix4.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/i2c/i2c-core.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/ide/ide-cd.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/ide/ide-core.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/ide/ide-disk.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/ide/pci/generic.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/ide/pci/piix.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/input/gameport/gameport.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/input/serio/serio_raw.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/net/tulip/tulip.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/pci/hotplug/pci_hotplug.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/pci/hotplug/shpchp.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/usb/core/usbcore.ko
   /lib/modules/2.6.16-ARCH/kernel/drivers/usb/host/uhci-hcd.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/core/seq/snd-seq-device.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/core/snd-page-alloc.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/core/snd-pcm.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/core/snd-rawmidi.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/core/snd-timer.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/core/snd.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/pci/ac97/snd-ac97-bus.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/pci/ac97/snd-ac97-codec.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/pci/snd-ens1371.ko
   /lib/modules/2.6.16-ARCH/kernel/sound/soundcore.ko
   /lib/modules/2.6.16-ARCH/kernel/fs/ext3/ext3.ko

So the bottom line is - mkinitcpio works fine, and I have something funky going on with my laptop.  roll  :?

Offline

#49 2006-05-07 10:54:25

baze
Member
Registered: 2005-10-30
Posts: 393

Re: new initramfs, testers welcome

new problems here:
I updated to the new klibc and mkinitcpio packages in testing just now, rebuilt the image (autodetect complains about mdadm missing, but that should be fine, since I dont use raid) and then rebootet to test it.
Although mkinitcpio -M detects vfat and ext3, "Mounting Local Filesystems" complains about unknown filesystems vfat and ext3 and doesn't mount those partitions.
I have reiserfs on my / and ext3 on the other partitions (plus a shared vfat) and only the root partition is mounted.

Offline

#50 2006-05-07 12:23:22

_Gandalf_
Member
Registered: 2006-01-12
Posts: 735

Re: new initramfs, testers welcome

tomk you're not the only one, I've also been in active developement of mkinitcpio for my custom needs since the beginning and always i was like yikes, why phrakture can boot it and not me!! Well i think we have the same problem tomk, look:

[root@nasreddine ../current/base/mkinitramfs]# modprobe --show-depends ext3
insmod /lib/modules/2.6.16-NUX/kernel/fs/jbd/jbd.ko.gz
insmod /lib/modules/2.6.16-NUX/kernel/fs/ext3/ext3.ko.gz

So if i don't add 'piix' (actually just piix is needed in MODULES) ide devices will be missing and i get a kernel panic, I'm not sure why this is happening I haven't tried it on a pure Arch setup yet as i have a very custom one, but I think it's the same issue, Snowman Has the same laptop i have so if he have the same issue then we have some weird PCs here sad

@ phrakture maybe at autodetection, if the module is hmm ext3 for example, echo jbd as well as piix no ? that should solve this *machine specific* issue...

@baze: Indeed i noticed vfat partition refuses to mount too, Try playing a bit with the MODULES section and ide specific modules.. i haven't tried a lot but generic piix haven't changed anything it's somewhere else then

Offline

Board footer

Powered by FluxBB