You are not logged in.
ok i have a question...by bootsplash do you mean kinda like in Ubuntu how you get that screen that has a coloerful ubuntu image while the comp. loads or does that mean that you have a colorful grub...? and i thought you had to recompile the kernel to get the pretty loading screen (or patch it at least) and final question does adding a boot splash slow loading time at all??? hahah sry for all the questions im just a curious new arch/linux user (arch for a week or two: linux in general since june or july...so ya..)
In this land of the pain the sane lose not knowing they were part of the game.
~LP
Offline
bootsplash is like usplash on ubuntu (the brown one u see when system is booting) though i couldn't get it working till now, the only working one is the usual progress but with a background, that's all and no it doesn't slow the system
about patching the kernel u dont need to do it urself, that's what iphitus actually did (look at the first post specially first line)
[My Blog] | [My Repo] | [My AUR Packages]
Offline
And just so everyone is completely clear - archck uses fbsplash, NOT bootsplash - they are two different projects, although bootsplash seems to be dying off. Someone (phrakture, I think) has looked at getting usplash into Arch, but for the moment fbsplash is the way to go.
Offline
I don't know if this problem is really archck related, but I have the problem since. I installed the new archck kernel yesterday, and I can't reboot/shutdown the system anymore since. BUT I have to mention I installed new ATI drivers too (the one on the aur site are autdated (they still install in /lib/modules/archck-2.6.14, and trying to recompile them myself I found out there were only newer drivers available on the ati site))
So I think this error is caused by archck or ati-drivers, or maybe a combination of both.
This is the problem: I use GDM with gnome, when I choose logout/shutdown/reboot from gnome, they all do the same: complete lockup of my system (even numlock doesn't work anymore) and a black screen (screen doesn't go on standby, just black)
When I type "reboot" in tty1, I get some kind of error message which I don't understand and can't copy/paste, as the system freezes again. (but now numlock keeps working)
resetting my computer and looking at dmesg at a fresh reboot gives this:
------------[ cut here ]------------
kernel BUG at mm/swap.c:49!
invalid operand: 0000 [#1]
PREEMPT SMP
Modules linked in: sbp2 ohci1394 eth1394 ieee1394 ohci_hcd ehci_hcd usbhid usblp usb_storage ppp_generic slhc sata_nv aic7xxx generic snd_ac97_codec snd_ac97_bu s snd_mpu401 snd_mpu401_uart snd_rawmidi snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_pcm snd_timer snd_page_alloc snd_mixer_oss snd sh pchp pci_hotplug serio_raw analog ns558 gameport evdev rtc ppdev lp parport eepr om w83l785ts asb100 hwmon_vid i2c_nforce2 i2c_core nvnet 3c59x mii nvsound sound core fglrx nvidia_agp agpgart usbcore xfs reiserfs jfs ext3 jbd raid6 raid5 xor raid10 raid1 raid0 multipath linear md_mod sata_sil libata sd_mod ide_floppy ide _cd cdrom ide_disk amd74xx ide_core
CPU: 0
EIP: 0060:[<c014bcb7>] Tainted: P B VLI
EFLAGS: 00210256 (2.6.15-archck)
EIP is at put_page+0x4f/0x6b
eax: 00000000 ebx: c1373440 ecx: c1373440 edx: c1373440
esi: c1373440 edi: c7798548 ebp: 40152000 esp: c863feb4
ds: 007b es: 007b ss: 0068
Process xscreensaver-gl (pid: 2959, threadinfo=c863e000 task=c93ce030)
Stack: c014fde4 c1373440 d0dff900 ffffffff 00000000 c110d16c 40152000 d5355400
40153000 d5355400 c014ff81 c1485900 d0a26494 d5355400 40152000 40153000
c863ff34 00000000 d5355400 c1485900 40152000 d0a26494 40153000 c015002b
Call Trace:
[<c014fde4>] zap_pte_range+0x182/0x24b
[<c014ff81>] unmap_page_range+0xd4/0xe3
[<c015002b>] unmap_vmas+0x9b/0x1aa
[<c0153d24>] exit_mmap+0x6e/0xf8
[<c011c031>] mmput+0x21/0x7a
[<c01207be>] do_exit+0x16f/0x372
[<c0120a82>] sys_exit_group+0x0/0x11
[<c0102d93>] sysenter_past_esp+0x54/0x75
Code: 32 ff 31 c0 f0 83 42 04 ff 0f 98 c0 84 c0 74 36 8b 4a 38 89 54 24 04 ff e1 8b 02 89 d1 f6 c4 40 74 03 8b 4a 0c 8b 41 04 40 75 08 <0f> 0b 31 00 32 ff 31 c0 f0 83 42 04 ff 0f 98 c0 84 c0 74 07 89
<1>Fixing recursive fault but reboot is needed!
scheduling while atomic: xscreensaver-gl/0x00000002/2959
[<c0302037>] schedule+0x43/0x6da
[<c014bccb>] put_page+0x63/0x6b
[<c011e4e6>] printk+0x12/0x16
[<c0120708>] do_exit+0xb9/0x372
[<c0103f24>] do_trap+0x0/0xc1
[<c0104197>] do_invalid_op+0x0/0x98
[<c0104223>] do_invalid_op+0x8c/0x98
[<c014bcb7>] put_page+0x4f/0x6b
[<e1dca303>] CAILQueryASICInfo+0x53/0x110 [fglrx]
[<c0117fa7>] activate_task+0x1a4/0x1c5
[<c0117c3e>] __activate_task+0xe/0x3a
[<c011880b>] try_to_wake_up+0x32c/0x336
[<c01038ef>] error_code+0x4f/0x54
[<c014bcb7>] put_page+0x4f/0x6b
[<c014fde4>] zap_pte_range+0x182/0x24b
[<c014ff81>] unmap_page_range+0xd4/0xe3
[<c015002b>] unmap_vmas+0x9b/0x1aa
[<c0153d24>] exit_mmap+0x6e/0xf8
[<c011c031>] mmput+0x21/0x7a
[<c01207be>] do_exit+0x16f/0x372
[<c0120a82>] sys_exit_group+0x0/0x11
[<c0102d93>] sysenter_past_esp+0x54/0x75
I don't know if it is the same error I get when I try to reboot, but I noticed the first 4 lines are the same.
Can anyone help me on this?
EDIT:
I downloaded the old ati rpm from somewhere else and made the pkgbuild, but I still have the same problems. So this problem must be archck related as my ati drivers are now the same I used with the stock kernel and archck-2.6.14 and then I never had problems.
Offline
N.B. The following procedure is now obsolete, as it is handled automatically by the latest version of mkinitramfs - thanks iphitus. I'll leave it here for historical purposes only.
=========Original post follows==========================
Yes, good news for suspend2 fans - I've got it working with initramfs. A bit of tinkering with the image is required, but we are Archers, so we don't mind a bit of tinkering - right?
Briefly, the solution involves editing the init script that runs things in the initramfs image. The problems I reported earlier resulted from putting the additional line in the wrong place in the script. This is what you need to do (after backing up your image, of course):
1. Create a directory, and expand the auto-generated initramfs image into it.
# mkdir my_initramfs
# cd my_initramfs
# gzip -dc < /boot/initramfs-2.6.15-archck.img | cpio -i
2. Open the init script in your favourite editor, and add this line
echo > /proc/suspend2/do_resume
straight after the last 'insmod' line.
3. Recompress the image:
# find . | cpio -o -H newc | gzip -9 > /boot/initramfs-2.6.15-archck.img
And that's it. You can now suspend and resume.
If you're not using the archck kernel, change the .img file above as required. This procedure relates to swap suspension only - I don't use, and therefore haven't tested, file suspension.
Finally, and as suggested earlier, I think this could be made a bit easier by having something like
SUSPEND2=1
in /etc/mkinitramfs.conf, and modifying mkinitramfs accordingly. However, I don't know how to do that, so consider this a feature request, iphitus.
Offline
im releasing a new mkinitramfs now.
edit: uploaded, grab pkgbuild and files, from
http://cvs.archlinux.org/cgi-bin/viewcv … vsroot=AUR
Offline
Finally it's working, Good work tomk
anyway to add "Suspend" to logout menu??
Thx
[My Blog] | [My Repo] | [My AUR Packages]
Offline
im releasing a new mkinitramfs now.
edit: uploaded, grab pkgbuild and files, from
http://cvs.archlinux.org/cgi-bin/viewcv … vsroot=AUR
Thx for the fast release iphitus
just a little question, using archck initramfs image must be created using
mkinitramfs auto
or
mkinitramfs full
??
thx
EDIT: Never Mind my answear was in kernel26.install
/sbin/mkinitramfs kernel_version=2.6.15-archck
[My Blog] | [My Repo] | [My AUR Packages]
Offline
Not that I want to stress anyone (dibble for example;-) ), but when do we see an updated version of the ati-drivers package?
Or am I just using the wrong mirrors?Greez Sigi
I don't maintain it anymore
I'm glad I said "no" to this update (because I wondered why kernel26 should be replaced with kernel26archck)
It isn't replaced - archck is an optional installation - oh, wait, did someone screw up the replaces?
The initscripts-gensplash package found in community is for 0.7 and not 0.7.1.
what does that mean? As far as I know it is up to date with the usual initscripts...ok, it looks like I am about 3 weeks behind! Whoops. I'll try and catch up...
I just tried it, first it does something nasty which is replace rc.conf instead of patching the existing one, it's ok i just add changes myself from rc.conf.initscripts-gensplash to rc.conf
pacman should NEVER straight replace a critical file - if it has you have a poorly configured setup. pacman NEVER patches anything - get a clue
One last thing! Can everyone remember that archck uses vesa-tng (more compatible but worse performance) therefore is you were using vesa standard with fbsplash/gensplash then you will need to change your video setting - see the bootloader samples in /etc/splash
Offline
I just tried it, first it does something nasty which is replace rc.conf instead of patching the existing one, it's ok i just add changes myself from rc.conf.initscripts-gensplash to rc.conf
pacman should NEVER straight replace a critical file - if it has you have a poorly configured setup. pacman NEVER patches anything - get a clue
hmmm that's kinda weird then, well I am 100% sure that it has replaced my rc.conf and moved the old one to rc.conf.pacsave, i am 100% positive because just looking at both files when i installed, i could just know from timezone and keymap vars that its not mine, so i've had to mv switch files back and then add what is important in rc.conf coming from this packages into mine
I don't know why u mention poorly configured but ASAIK i didn't touch (didn't dig into) yet any conf file related to makepkg and/or pacman except that i've enabled community and i'm positive about having etc/rc.conf in noupgrade list of pacman (btw why its etc/ and not /etc/ ? )
[options]
LogFile = /var/log/pacman.log
NoUpgrade = etc/passwd etc/group etc/shadow etc/sudoers
NoUpgrade = etc/fstab etc/raidtab etc/ld.so.conf
NoUpgrade = etc/rc.conf etc/rc.local
NoUpgrade = etc/modprobe.conf etc/modules.conf
NoUpgrade = etc/lilo.conf boot/grub/menu.lst
HoldPkg = pacman glibc
[My Blog] | [My Repo] | [My AUR Packages]
Offline
Could we split this rc.conf stuff? my packages dont go within a thousand miles of it. Check the pkgbuilds if you dont believe me.
Offline
Iphitus: Thx for the new ati-drivers-package! I'll give it a try. Anyway, just hours after you updated the package, version 8.21.7 of ati's fglrx driver has been released. Not that I want to hurry anybody, but it would be nice seeing this package upgraded again
Thx a lot! archck rocks!
Greez Sigi
Haven't been here in a while. Still rocking Arch.
Offline
hmm, I'm back to archck 2.6.14, there seems to be a problem with 2.6.15 kernel and ati-drivers (the same error as I had, I found the error on mailing lists of some distributions)
But what I think was weird is that I couldn't use the stock 2.6.14 kernel anymore, I get a busybox error? I even reinstalled the 2.6.14 stock kernel, but no luck. Is it possible this is a result of archck?
But archck 2.6.14 is now working stable here again
Sigi: can you get a stable system running with kernel 2.6.15 and ati-drivers?
Offline
hmm, I'm back to archck 2.6.14, there seems to be a problem with 2.6.15 kernel and ati-drivers (the same error as I had, I found the error on mailing lists of some distributions)
But what I think was weird is that I couldn't use the stock 2.6.14 kernel anymore, I get a busybox error? I even reinstalled the 2.6.14 stock kernel, but no luck. Is it possible this is a result of archck?But archck 2.6.14 is now working stable here again
Sigi: can you get a stable system running with kernel 2.6.15 and ati-drivers?
What was the error? I released a new ATI package only a few hours ago that definitely works with the 2.6.15 kernel as as been shown by Sigi.
Chances are your mirror hasnt updated, if not, either grab it from the Arch mirror or just wait patiently
Also released was a mkinitramfs update, that will now create suspend resumable initramfs'
Offline
What was the error? I released a new ATI package only a few hours ago that definitely works with the 2.6.15 kernel as as been shown by Sigi.
Chances are your mirror hasnt updated, if not, either grab it from the Arch mirror or just wait patiently
Also released was a mkinitramfs update, that will now create suspend resumable initramfs'
The error: I don't know exactly, but is was something like this:
He started my initrd image (my raid arrays were detected), and then in case the line "booting archlinux 0.7.1" and "starting UDEV" apeared, something like this was shown:
/initrd not found
busybox cannot get to tty1
(It wasn't exactly like this, I didn't write it down)
and then came in a busybox shell. But he did start my initrd, as otherwise my raid array never could have been activated.
And I build the ati-drivers myself by pulling the PKGBUILD out of cvs and modify the _kernel var and the version number (I tried both 8.21.7 and 8.20.8 )
Now I'm using archck-2.6.14 with my own build ati drivers (8.21.7) with the PKGBUILD from aur (and setting the _kernel var correctly) and it works like a charm.
Offline
anyway to add "Suspend" to logout menu?
That would depend entirely on whatever desktop you're using - e.g. I've mapped mine to a keyboard shortcut in xfce4. Personally, I use suspend when I want to shutdown without logging out, so I wouldn't have any use for it on the logout menu.
Offline
I use Gnome Desktop, i know how to assign key on Gnome via Gconf but i also prefer to have a radio button in order to select hibernate/suspend
[My Blog] | [My Repo] | [My AUR Packages]
Offline
iphitus wrote:What was the error? I released a new ATI package only a few hours ago that definitely works with the 2.6.15 kernel as as been shown by Sigi.
Chances are your mirror hasnt updated, if not, either grab it from the Arch mirror or just wait patiently
Also released was a mkinitramfs update, that will now create suspend resumable initramfs'
The error: I don't know exactly, but is was something like this:
He started my initrd image (my raid arrays were detected), and then in case the line "booting archlinux 0.7.1" and "starting UDEV" apeared, something like this was shown:
/initrd not found
busybox cannot get to tty1(It wasn't exactly like this, I didn't write it down)
and then came in a busybox shell. But he did start my initrd, as otherwise my raid array never could have been activated.And I build the ati-drivers myself by pulling the PKGBUILD out of cvs and modify the _kernel var and the version number (I tried both 8.21.7 and 8.20.8 )
Now I'm using archck-2.6.14 with my own build ati drivers (8.21.7) with the PKGBUILD from aur (and setting the _kernel var correctly) and it works like a charm.
well I updated to the new archchk, but issues issue issues
- It messed up my stock kernel again. I get the same message as yu above, even I have a nvidia card
- it didn´t create a initramfs but a initrd-2.6.15-archck.img
- there are a lot of error messages while boot, but its too fast to see them,
Ca the 2.6.14.archck kernel be provided again and this one put on hold until it will really work.
mic64
Offline
- It messed up my stock kernel again.
Post the contents of your /boot, and your grub/menu.lst
- it didn´t create a initramfs but a initrd-2.6.15-archck.img
Install the latest version of mkinitramfs. Please note, however, that it doesn't matter what your image is called, it's the contents that count.
- there are a lot of error messages while boot, but its too fast to see them
Look for useful info in dmesg and your logs.
Ca the 2.6.14.archck kernel be provided again
I have kernel26-2.6.14.3-2 in my cache - If you want it, I'll put it online for you.
and this one put on hold until it will really work.
How about just not installing it?
Offline
thats in my /boot
System.map26 initrd26-full.img kconfig26mm.pacsave
System.map26archck initrd26.img vmlinuz26
diag1.img kconfig26 vmlinuz26archck
grub kconfig26.pacsave
initrd-2.6.15-archck.img kconfig26archck
that my menu.lst
# (0) Arch Linux
title Arch Linux Kernel current
root (hd0,2)
kernel /boot/vmlinuz26 root=/dev/hda3 ro devfs=nomount vga=791
initrd /boot/initrd26.img
# (1) Arch Linux
title Arch Linux Kernel archck
root (hd0,2)
kernel /boot/vmlinuz26archck root=/dev/hda3 ro devfs=nomount vga=788
initrd /boot/initrd-2.6.15-archck.img
it looks right, or not?
Well I wanted to remove it, but that I wouldn´t be able to boot into arch at all.
mic64
Offline
@mic64 try without the option devfs=nomount
here's mine if it helps
# (0) Arch Linux Arckck + Splash
title Arch Linux ArchCk 2 + Splash
root (hd0,4)
kernel /vmlinuz26archck root=/dev/hda6 video=vesafb:ywrap,mtrr,1024x768-32@70 splash=verbose,theme:darch console=tty1 resume2=swap:/dev/hda7
initrd /initramfs-2.6.15-archck.img
# (0) Arch Linux Archck Gaphical mode
title Arch Linux ArchCk 2 Graphical
root (hd0,4)
kernel /vmlinuz26archck root=/dev/hda6 video=vesafb:ywrap,mtrr,1024x768-32@70 quiet console=tty1
initrd /initramfs-2.6.15-archck.img
# (0) Arch Linux Archck Non-vga
title Arch Linux ArchCk 2
root (hd0,4)
kernel /vmlinuz26archck root=/dev/hda6 ro
initrd /initramfs-2.6.15-archck.img
# (0) Arch Linux Orig Kernel + Graphical
title Arch Linux Graphical
root (hd0,4)
kernel /vmlinuz26 root=/dev/hda6 ro vga=0x318 console=tty1 quiet
initrd /initrd26.img
[My Blog] | [My Repo] | [My AUR Packages]
Offline
I think some people are confused here, so I will repeat more clearly what my problems are:
*) 2.6.15-archck boots fine, splash works, but the ati-drivers don't seem to work. (but as I said: other distro are also having problems with this, so I suppose this is just kernel 2.6.15 related)
*) 2.6.14-archck works like a charm (but no splash as splash + initrd doesn't go on 2.6.14)
*) But stock kernel 2.6.14 and/or 2.6.15 doesn't work anymore, even after reinstalling. I think archck (or initscripts) did something to mkinitrd that makes the initrd I make for my stock kernel doesn't work
my /boot/grub/menu.lst
title Arch gensplash
root (hd0,2)
kernel /vmlinuz26archck root=/dev/md0 video=vesafb:ywrap,mtrr,1280x1024-32@70 splash=verbose,theme:darch console=tty1
initrd /initramfs-2.6.15-archck.img
title Arch Linux
root (hd0,2)
kernel /vmlinuz26-2.6.14-archck root=/dev/md0 ro vga=795
initrd /initrd-2.6.14-archck.img
title Arch Linux backup
root (hd0,2)
kernel /vmlinuz26 root=/dev/md0 ro vga=795
initrd /initrd26.img
I created the initrd26.img using:
mkinitrd kernel_version=2.6.14-ARCH
(and off course the vmlinuz26 is the one from 2.6.14 stock kernel
So I think normally the last entry should work? also my /etc/mkinitrd.conf is still the same as before.
also: tuxracer is terribly slow, but that might be a sign I should be studying
Offline
i think i just confirmed that mkinitramfs messes up the default kernel. i could not boot my system after installing it (i used it pretty early so maybe i'm talking about something that has been fixed). i rebooted into the noodle disc probably a million times to mess with lilo and rebuilding the image trying to get it to work. that last post gave me one last idea. uninstall mkinitramfs. after i did and rebuilt the kernel image i can boot fine! let me know if you want more info and please share if you have had this issue. i was bad to hold out until now.
Offline
mkdir /initrd seems to have fix some problems with vanilla.
Please make sure you use the latest mkinitramfs (1.1-3) and the latest ArchCK (2.6.15-archck2-3).
Offline
ok, built all the latestet stuff, everything wokrs except my raid
I tried what you said iphitus but it didnt work, heres the error mdadm gives me:
mdadm: error opening /dev/md/0: No such file or directory
mdadm: error opening /dev/md0: No such file or directory
There is no md0 file in /dev.
Offline