Kernel 2.6.11.y - is arch going to follow?

Hi
Just wondering if ArchLinux is going to follow the new 2.6.x.y 'stable' scheme?
some info: http://kerneltrap.org/node/4803
thanks

apparently, it's in testing

Similar Messages

Cannot use kernel newer than 2.6.16-ARCH

I just got myself a new laptop (Toshiba U205-S5002) and was excited as hell to get Arch installed on it. The install went well.. until I did a pacman -Syu.. my system would freeze upon reboot and the fans would turn on full blast. I did a reinstall and upgraded everything besides kernel26 and it boots up fine. I tried installing kernel26beyond and it does the same thing as before: freezes.
I thought it was an ACPI error, since that's the last bit of text I'd see before the system would freeze, so I did some searching and found a handful of grub commands to try in order to get the system to boot with a newer kernel. I tried acpi=force, acpi=no, nolapci, and some others, but to no avail; the system just froze at different stages. It usually freezes around the ACPI: IRQ statements... other times it freeze at SMP PREEMPT.. and sometimes it says SMP Motherboard not deteched.
I also read that it could be a gcc error. I did a `uname -a` to see which version the kernel was running the the space where it should be is blank. Not just missing, but there is whitespace where it should go. I did a dmesg | grep gcc and found that the kernel is running gcc 4.0, but the installed version (via pacman -Qi gcc) is 4.1 ...
Does anybody have any ideas? This is driving me crazy.
EDIT: more searching uncovered the "nosmp" option. I thought this was going to be my saviour, but, alas, it wasn't. It still hangs, but without all the SMP errors. :x

phrakture wrote:As far as I know, "nosmp" is crap. Try recompiling without SMP support and see if that helps... if it does, this is a bigger fish.
I compiled a custom beyond kernel with SMP disabled and it still managed to hang at the "ACPI: Host Bridge" something or another. Thinking that SMP wasn't actually the problem, and after doing some more forum and wiki searching, I compiled the kernel a second time with "ACPI Processor" disabled and it still hung at system startup.
I'm compiling it again with both SMP and ACPI Processor disabled.. hopefully it works this time, otherwise I don't know what else to do.

EFI/refind boot- how to get kernel vmlinuz in /boot/efi/EFI/arch?

I have a MacBook Pro Retina (10,2) running arch. It EFI boots using rEFInd, booting from /boot/efi/EFI/arch. This problem has been happening since I first installed the system in July - when I use pacman to update to the latest kernel, the mkinitcpio hooks generate for the current kernel, not the one that was just installed. If I don't manually run mkinitcpio after the update, I'm left with an unbootable system (the ususal "Unable to find root device" error, and dumped to recovery shell) because of a kernel/initrd mismatch (I think).
So I think the problem isn't mkinitcpio, but rather that the "linux" package is installing the kernel at /boot/vmlinuz-linux (which I later manually copy to /boot/efi/EFI/arch/vmlinuz-linux), but mkinitcpio is looking in the correct path (/boot/efi/EFI/arch) for the kernel... which is the last kernel, since I haven't copied it there yet.
As an example (yeah, I know I went a LONG time between updates, mainly due to nvidia issues), yesterday I did a full update that bumped me from 3.9.8-1 to 3.10.10-1. But here's the mkinitcpio output during the pacman run:
upgrading linux...
>>> Updating module dependencies. Please wait ...
>>> Generating initial ramdisk, using mkinitcpio. Please wait...
==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'default'
-> -k /boot/efi/EFI/arch/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/efi/EFI/arch/initramfs-linux.img
==> Starting build: 3.9.8-1-ARCH
-> Running build hook: [base]
-> Running build hook: [udev]
-> Running build hook: [autodetect]
-> Running build hook: [modconf]
-> Running build hook: [block]
-> Running build hook: [filesystems]
-> Running build hook: [keyboard]
==> ERROR: module not found: `usbhid'
-> Running build hook: [fsck]
-> Running build hook: [resume]
==> WARNING: No modules were added to the image. This is probably not what you want.
==> Creating gzip initcpio image: /boot/efi/EFI/arch/initramfs-linux.img
==> WARNING: errors were encountered during the build. The image may not be complete.
==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'fallback'
-> -k /boot/efi/EFI/arch/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/efi/EFI/arch/initramfs-linux-fallback.img -S autodetect
==> Starting build: 3.9.8-1-ARCH
-> Running build hook: [base]
-> Running build hook: [udev]
-> Running build hook: [modconf]
-> Running build hook: [block]
-> Running build hook: [filesystems]
-> Running build hook: [keyboard]
==> ERROR: module not found: `usbhid'
-> Running build hook: [fsck]
-> Running build hook: [resume]
==> WARNING: No modules were added to the image. This is probably not what you want.
==> Creating gzip initcpio image: /boot/efi/EFI/arch/initramfs-linux-fallback.img
==> WARNING: errors were encountered during the build. The image may not be complete.
error: command failed to execute correctly
Is there any solution to this problem? Because of the nature of rEFInd, I can't (don't think I can? I remember trying it and having problems) just symlink /boot/efi/EFI/arch/vmlinuz-linux to /boot/vmlinuz-linux.
Any ideas/help would be greatly appreciated.
Here's the uncommented lines in /etc/mkinitcpio.conf:
MODULES=""
BINARIES=""
FILES=""
HOOKS="base udev autodetect modconf block filesystems keyboard fsck resume"
And from /etc/mkinitcpio.d/linux.preset, which I manually specify on the command line (-p linux) when I re-run mkinitcpio after copying vmlinuz into place:
ALL_config="/etc/mkinitcpio.conf"
ALL_kver="/boot/efi/EFI/arch/vmlinuz-linux"
PRESETS=('default' 'fallback')
default_image="/boot/efi/EFI/arch/initramfs-linux.img"
fallback_image="/boot/efi/EFI/arch/initramfs-linux-fallback.img"
fallback_options="-S autodetect"
Is the right solution just changing ALL_kver to /boot/vmlinuz-linux and then copying that into /boot/efi/EFI/arch after the install, or is there some nicer solution so that this can actually be an automatic process?
Thanks for any advice/tips/suggestions.
PS - I'm running arch on the Retina with KDE. All of my configuration after the initial working install is handled by Puppet, and the module I use (along with the initial instructions and a list of what is and isn't working so far) is available at https://github.com/jantman/puppet-archl … bookretina

Thanks, Rod! I agree, a simple mount -L esp /boot configuration is definitely ideal for a single installation system. That was what I first tried when I first installed my UEFI board. But I keep multiple installations on disk, and multiple configs per install, so I would often run into issues with installations interfering with one another's /boot files.
I also tried unique file names per installation - like prepending installation identifiers to filenames - but that required changing defaults at install, /boot monitor scripts and services and of course I would occasionally forget to implement it for an installation and be right back at square one. It was a headache.
Eventually I settled on the solution I described above because it was simply implemented, portable without modification, provided me a sane /boot tree, seemed to play nicely with default options across the board, and, well, because it was the best I could come up with. I figured that UEFI had already imposed the mandatory FAT32 partition on me so I might as well use it for all it was worth. I mean, even Windows fits in right along with the rest. This way I could interface with UEFI and linux bootloaders in the pre-UEFI customary way with simple configuration edits in /etc/fstab and in /boot, thanks mostly, of course, to rEFInd and what I learned reading rodsbooks.
I realize that I'm probably in violation of the GNU filesystem standard with the root-level /esp mount, but, to me, it seemed the intuitive place to put it. And frankly, I really don't know any better. Perhaps it belongs in /var/run or something? I'm fully confident GNU will have resolved a standard mount point for it once and for all sometime in the next decade or so.
Still, as you say, esp as /boot is ideal. Creshal very elegantly described it on the UEFI Bootloaders wiki Talk page like this:
efibootmgr -c -l /vmlinuz-linux -u "$(cat /proc/cmdline)"
Super simple, assuming an esp at /dev/sda1. He doesn't mention the necessary /boot entry in /etc/fstab or the various possible issues with efibootmgr/UEFI NVRAM though, which is of course perfectly acceptable in Talk. Still, I mentioned these things in reply to him there and described my solution like this:
$ mkdir -p /esp; mount -L {,/}esp
$ mkdir -p /esp/EFI/boot/<NEW_SYSTEM>/; mv /boot/* $_; mount --bind $_ /boot
$ cat <<'EOF' >/etc/fstab
LABEL=<NEW_ROOTFS> / <FSTYPE> defaults 0 0
LABEL=esp /esp vfat defaults 0 0
/esp/EFI/boot/<NEW_SYSTEM> /boot none defaults,bind 0 0
EOF
$ sed -ri 's/root=[^ ]*/root=LABEL=<NEW_ROOTFS>/g' /boot/refind_linux.conf
Probably that code works, though it's entirely possible I've stumbled on syntax somewhere as I haven't tested it as written. It is admittedly less elegant than Creshal's proffered solution, but it includes an operable /etc/fstab, skeleton that it is, and edits your refind_linux.conf to boot. Assuming gpt partition labels and a rEFInd install it should do the job with very few required modifications if any aside from replacing the <VAR> strings and adding /etc/fstab entries as appropriate.
I am troubled, though, by the possible incompatibilities you mention with a vfat filesystem mounted at /boot and default symlinking behavior there. Which distributions behave this way? I don't think I've encountered it, though my default install is arch so I suppose it's possible I've just overlooked it.
I ask mainly because I've also added this to the wiki as a Sync EFIStub Kernel option and would like to highlight any inherent incompatibilities there or remove it if necessary, especially considering that I've already so boldly stated: "The following method should work similarly with any distribution or any combination thereof." Guess I jumped the gun on that one, huh?
Anyway, thanks again for rEFInd and your considerable work on rodsbooks.
-Mike
Last edited by mikeserv (2013-09-30 23:04:51)

[cisco VPN] Can't build kernel module with 2.6.9-ARCH

I need to setup a vpn tunnel to my university in order to gain acces to their resources and be able to surf when I am on the campus. With 2.6.8.1 I used the Cisco VPN client 4.0.5 k9. After my upgrade to 2.6.9 I had to rebuild the module, but now it fails to build. Anyone knows how to solve this? Or does anyone know another vpn client that is compatible with Cisco. This piece of software is essential to me. Please help. Here is the output:
Cisco Systems VPN Client Version 4.0.5 (Rel) Linux Installer
Copyright (C) 1998-2001 Cisco Systems, Inc. All Rights Reserved.
By installing this product you agree that you have read the
license.txt file (The VPN Client license) and will comply with
its terms.
Directory where binaries will be installed [/usr/local/bin] /usr/bin
Automatically start the VPN service at boot time [yes] no
In order to build the VPN kernel module, you must have the
kernel headers for the version of the kernel you are running.
For RedHat 6.x users these files are installed in /usr/src/linux by default
For RedHat 7.x users these files are installed in /usr/src/linux-2.4 by default
For Suse 7.3 users these files are installed in /usr/src/linux-2.4.10.SuSE by default
Directory containing linux kernel source code [/lib/modules/2.6.9-ARCH/build]
* Binaries will be installed in "/usr/bin".
* Modules will be installed in "/lib/modules/2.6.9-ARCH/CiscoVPN".
* The VPN service will *NOT* be started automatically at boot time.
* Kernel source from "/lib/modules/2.6.9-ARCH/build" will be used to build the module.
Is the above correct [y] y
Making module
make -C /lib/modules/2.6.9-ARCH/build SUBDIRS=/home/luk/sources/vpnclient modules
make[1]: Entering directory `/usr/src/linux-2.6.9-ARCH'
CC [M] /home/luk/sources/vpnclient/interceptor.o
/home/luk/sources/vpnclient/interceptor.c: In function `add_netdev':
/home/luk/sources/vpnclient/interceptor.c:59: sorry, unimplemented: inlining failed in call to 'supported_device': function body not available
/home/luk/sources/vpnclient/interceptor.c:245: sorry, unimplemented: called from here
make[2]: *** [/home/luk/sources/vpnclient/interceptor.o] Error 1
make[1]: *** [_module_/home/luk/sources/vpnclient] Error 2
make[1]: Leaving directory `/usr/src/linux-2.6.9-ARCH'
make: *** [default] Error 2
Failed to make module "cisco_ipsec.ko".

I modified the pkgbuild posted here by someone (thank you!) so it includes all relevant files (meaning also vpnc-connect and vpnc-disconnect and vpnc.conf).
pkgname=vpnc
pkgver=0.2
pkgrel=1
pkgdesc="Client for Cisco3000 VPN Concentrator"
url="http://www.unix-ag.uni-kl.de/~massar/vpnc/"
license="GPL"
depends=(libgcrypt)
source=(http://www.unix-ag.uni-kl.de/~massar/vpnc/vpnc-0.2-rm+zomb.1.tar.gz)
md5sums=(ded67de747874c4245ed8405146dc94a)
build() {
cd $startdir/src/vpnc-0.2-rm+zomb.1
# We want the CFLAGS specified in makepkg.conf to be used
mv Makefile Makefile.old
sed -e 's/-W -Wall -O -g/$(MYCFLAGS)/g' -e 's/LDFLAGS=-g /LDFLAGS=/g' Makefile.old > Makefile
export MYCFLAGS=$CFLAGS
make
install -d $startdir/pkg/usr/sbin
install vpnc $startdir/pkg/usr/sbin
install vpnc-connect $startdir/pkg/usr/sbin
install vpnc-disconnect $startdir/pkg/usr/sbin
install -d $startdir/pkg/etc
install vpnc.conf $startdir/pkg/etc
Guess what, it works
I can reproduce my steps.
- makepkg
- pacman -A vpnc-xxxxxx.tar.gz
- add tun to the daemons array in rc.conf
- Modify /etc/vpnc.conf
- vpnc-connect

Random kernel panics in 2.6.21-ARCH

Hi, I've installed archlinux duke yesterday. I'm getting random lockups, when the lockup happens the "caps lock" and "scroll lock" lights are flashing, I reboot and check the messages log and I can't find anything about the lock
I know this is a panic, but I don't know why, I've searched in google but can't find anything Can anyone help me?
My hardware:
pentium 4 processor, 2.8ghz
VIA Motherboard
nvidia graphics card, geforce 6200 with latest drivers in pacman, I use NVAGP
WD 80gb Hard disk.
uname -a shows :
Linux arch 2.6.21-ARCH #1 SMP PREEMPT Mon Jun 11 23:47:51 CEST 2007 i686 Intel(R) Pentium(R) 4 CPU 2.80GHz GenuineIntel GNU/Linux
thanks.
ps. sorry for my english
Last edited by omarsx (2007-06-25 22:02:44)

The problem with panics is that the kernel stops and thus cannot log what went wrong. If you are lucky, you may find something that went wrong before the lockup.
How long does it take until the panic happens? If you can somehow provoke it, try doing this while you are on a text console. Then you will see some message about the panics, they may or may not help.
It is possible that 2.6.22 fixes the problem, so you can enable the [kernel]-repository and upgrade (if you have enough time) and see if the problem is gone:
for i686, add this as the first repo to pacman.conf
[kernel]
Server = http://www.archlinux.org/~tpowa/rc-kernel/
for x86_64, use this:
[kernel]
Server = http://www.archlinux.org/~thomas/rc-kernel-x86_64

Compiling kernel: what are the cons against arch kernel?

As the title

Procyon wrote:
broch wrote:so total time is ~25 min
I don't see drawbacks, only advantages.
Don't forget recompiling things that need the kernel like nvidia. That's what's been keeping me from trying it out.
It only takes a few minutes for me to install nvidia drivers to any new kernel that I make. I don't use the Arch nvidia driver, I use either the stable or beta driver offered from this page: http://nvnews.net/vbulletin/showthread.php?t=122606
Basically, after building my new parallel kernel I restart/reboot and then log into the new kernel from my edited GRUB list, then of course I will be greeted by my console login prompt. I log into my username, change directory to my downloaded nvidia driver and run this command:
sudo sh NVIDIA-Linux-x86_64-180.06-pkg2.run
Then click yes and ok a few times and it builds my kernel module and also the 32libs, then it finishes in 1 minutes, then "sudo reboot -n" and now my nvidia work for my new kernel.
..... and I always have my stock -ARCH default kernel as a backup that can either use my downloaded nvidia driver or the Arch one.
What took the most time for me was learning which way to use for building a seperate kernel to install along side of my -ARCH default one. Perfecting the PKGBUILD and preset/install scripts took a bit of figuring out.
The actual compile time isn't that long, 15-25 minutes..... maybe even less? I use the "-j3" option for my MAKEFLAGS in /etc/makepkg.conf file.
The "make menuconfig" way can take anywhere from less than 1 minute to an hour..... if you read about every single thing available (which is a good learning experience even if you don't have that hardware or experience).
Last edited by methuselah (2008-12-07 22:13:05)

Compiling kernel modules against 2.6.15-ARCH

I recently migrated to arch from Gentoo and Debian for a new webserver, and couldn't be happier. The system has worked great so far, and RAID card aside, I haven't had a single problem with Arch. The base system is easy to configure, and there is actually a list of config files in the install docs. Pacman is an absolute dream... Debian's apt is willfully a year out of date, and Gentoo's portage is a convulted and time-consuming nightmare.
In fact, Gentoo's unprompted and unwarranted removal of libstdc++ was the final straw.
However, one thing that confuses me about arch is the kernel build system.
I have an Areca ARC-1210 PCIe to SATAII hardware RAID card, but I can't figure out how to build the driver as a kernel module. If I compile it on its own it fails with literally hundreds of errors, almost like there's a header missing.
The instructions that came with the source want me to add a
config SCSI_ARCMSR
tristate "ARECA SATA RAID HOST Controller"
depends on PCI && SCSI
help
block to /usr/src/linux/drivers/scsi/Kconfig (which I did: it shows up in menuconfig), and a
obj-$(CONFIG_SCSI_ARCMSR) += arcmsr/
line to /usr/src/linux/drivers/scsi/Makefile.
However, arch doesn't seem to have the recursive makefile kernel build system that I and the good people at Areca used to.
How can I build this driver, preferably as a module, against my 2.6.15 kernel?
There are also precompiled x86 modules availible for Red Hat, SuSE, and TurboLinux. Assuming one of them is built against a reasonably current kernel, could they be used safely? I normally wouldn't try such a thing, but time is of the essence, and the RAID card seems to have such simple dependencies, I imagine it unlikely that all three major distros would have SCSI and PCI drivers patched to incompatability.

having said all that ...do u think compiling the kernel myself would provide any improvment for me???? ( I am haveing freezing and instablility problems..ive used 3 of the arch kernels....check out my thread titled Arch Instable.. (please)
p.s. btw Welcome

Can't not found Kernel Headers 3.3.8-1-ARCH

I was update my system use pacman, but after update, vmplayer prompt that it can not found kerenel header 3.3.8-1, and i found that my kernel update to 3.4.2....how to solve this problem?

Your question is not clear.
Try this
http://catb.org/esr/faqs/smart-questions.html

Error event id 2 - Kernel Event Tracing - Session "" failed to start with the following error: 0xC000000D

I keep getting this strange error logged as Error event id 2. Note session string is blank "". Any ideas?
Log Name:      Microsoft-Windows-Kernel-EventTracing/Admin
Source:        Microsoft-Windows-Kernel-EventTracing
Date:          24/12/2009 09:08:20
Event ID:      2
Task Category: Session
Level:         Error
Keywords:      Session
User:          SYSTEM
Computer:      George-XXL
Description:
Session "" failed to start with the following error: 0xC000000D
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
    <Provider Name="Microsoft-Windows-Kernel-EventTracing" Guid="{B675EC37-BDB6-4648-BC92-F3FDC74D3CA2}" />
    <EventID>2</EventID>
    <Version>0</Version>
    <Level>2</Level>
    <Task>2</Task>
    <Opcode>12</Opcode>
    <Keywords>0x8000000000000010</Keywords>
    <TimeCreated SystemTime="2009-12-24T09:08:20.835299100Z" />
    <EventRecordID>36</EventRecordID>
    <Correlation />
    <Execution ProcessID="1944" ThreadID="1948" />
    <Channel>Microsoft-Windows-Kernel-EventTracing/Admin</Channel>
    <Computer>George-XXL</Computer>
    <Security UserID="S-1-5-18" />
</System>
<EventData>
    <Data Name="SessionName">
    </Data>
    <Data Name="FileName">
    </Data>
    <Data Name="ErrorCode">3221225485</Data>
    <Data Name="LoggingMode">5</Data>
</EventData>
</Event>

Can you let us know if you get any Blue Screen crush error while using your comptuer?
Generally we can safely ignore the errors in the Event log. Aas we don't if you have get other issues in your computer, you can test your computer in the Clean Boot mode. This method will help us determine if this issue is caused by a loading program or service.
You can try the the following steps:
1. Click the Start Button type "msconfig" (without quotation marks) in the Start Search box, and then press Enter.
2. Click the "Services" tab, check the "Hide All Microsoft Services" box and click "Disable All" (if it is not gray).
3. Click the "Startup" tab, click "Disable All" and click "OK".
Then, restart the computer. When the "System Configuration Utility" window appears, please check the "Don't show this message or launch the System Configuration Utility when Windows starts" box and click OK.
Please test this issue in this Clean Boot environment, if the issue disappears in the Clean Boot environment, you can use the dichtomy to quickly narrow down which entry is causing the issue. You can refer to the
KB929135.
Good luck!
John
This is a very poor answer. This guy doesn't know what he's talking about because if you tell someone its ok to "safely ignore the errors in the event log" as a blanket instruction ESPECIALLY because as he stated so himself
he didn't know if there were other issues with the OP's pc then he's giving really bad advice. If you ignore errors in the event log then over time your system will degrade and you will have other issues.. Every error on your pc should NOT be ignored.
It should be decided on a case by case situation.
Furthermore this is an error involving the Windows Session which it states has failed. Definitely should NOT be ignored.
I've got the same error which has suddenly just showed up. No BSOD, Do not have WHS installed and all my media center connectors have NEVER been connected. I don't believe this is a service issue. I think it has something
to do with kernel event tracing.
Of course, it's just my opinion....I could be wrong!

Will Arch Linux going to change the default compiler to Clang?

Recently read a news about building the kernel with LLVM Clang.
I would like to ask will Arch Linux adopt this also soon, or in a plan, or no decision yet?

Teho wrote:Would you mind sharing some sources? There seems to be quite a bit of work going on making upstream Clang/LLVM to compile Linux kernel
Same as you (I suppose) I've been following these attempts by looking at overall progress, mailing lists and some of the bug reports, like I said this is not the first attempt to get llvm/clang to compile Linux (afaik this latest one is a consolidation of previous efforts together with some new blood(?) which makes it more likely that it will atleast be able to compile a stable kernel, albeit requiring patches).
Teho wrote:Debian also seems to have plans to compile their entire repository with Clang. Not to mention that LLVM itself is very popular nowadays being installed on almost every Linux system.
Are you referring to that one Debian developer who has compiled the repository with Clang to see how well it went? Is that the basis of your 'Debian also seems to have plans to compile their entire repository with Clang' claim?. AFAIK developers of distros have tried pretty much all the compilers under the sun on their repos and have reported the results, doesn't mean they plan to use that compiler for their official builds.
Teho wrote:To my understanding the patches applied to Linux are also relatively small and non-intrusive.
I don't know what the state is now but the previous patches I recall where not unintrusive, of course there are two targets of problems here, you can pretty much leave Linux alone if you can have clang-llvm accept lots of patches, and vice versa. If the current crop of patches needed were relatively small and non-intrusive they would likely have made in into some testing branches in the trees of their upstream targets by know, the changes needed for FreeBSD to compile using Clang/LLVM were quite substantial as I recall and it was like a 4-5 year period necessary for the transition despite being a very focused effort which will finally reach it's culmination with the upcoming FreeBSD 10 release.
Teho wrote:There's obviously a lot of community intrest considering that it was worth setting up the LLVMLinux project under Linux Foundation (to reduce dublicated work);
Well I don't judge a project's success or community interest by a webpage, I judge it by progress. Currently it seems to be where it was last I checked, patches needing to be applied to both Linux and Clang/LLVM with a unstable kernel as a result.
Teho wrote:Not to mention that LLVM itself is very popular nowadays being installed on almost every Linux system.
Well yes LLVM is pretty much ubiquitous as a JIT-framework which is used in lots of projects but LLVM as a an actual installed package?
Like I said, I would be surprised (as in I don't think it will happen) if upstream linux and clang/llvm took in the patches necessary for compiling linux 'out-of-the-box' based upon past and current progress. I guess we'll have to see.

[Solved] Basic research for arch beginner - kernel

Hello,
if i want to compile my own new kernel so i create a directory to
"/usr/src/" like "linux-2.6.17.7". So i have in some cases more like one kernel directory.
Other distributions set a symlink to the actual kernel directory with
the name "linux", because in the system is "linux" the reference.
So it easy to change the link to the actualy kernel-directory.
I have seen, that Archlinux don't use a symlink to the kernel-directory.
How can i change the reference in the system for a new kernel-directory,
because i think if i compile some modules, the system will got every time
the "old" modules directory and not the new one ?
Thank you and
best regards
bademeister

brain0 wrote:
What does "the actual kernel-directory" mean? There is absolutely no part of the system that looks for /usr/src/linux.
You only need the source (or at least part of it) to compile external modules. The necessary headers are located via the /lib/modules/$(uname -r)/build/ directory.
As said above, you should use abs for kernel compilation so that pacman can keep track of the files.
The best solution for 99% of the users is using the stock kernel anyway, as it is slim and modular, supports all you need and there are tons of precompiled kernel modules for it in the repos.
Hi brain0,
"the actual kernel-directory" means, which directory is now use by the system.
For example, if i download the kernel and place it to /usr/src/linux-2.6.17.7, so i have two directories. With the old kernel before (linux-2.6.17-ARCH) and my new one.
"uname -r" show me the reference directory, now with "2.6.17-ARCH", also
all modules will loaded from: /lib/modules/2.6.17-ARCH/.
Now i compile the new kernel (linux-2.6.17.7-NEW) and during the compilation, all modules will be copied to the new module directory:
/lib/modules/2.6.17-NEW/.
If i look to the reference directory with "uname -r", so i get
2.6.17-ARCH again. All kind of modules will be loaded from the old directory.
So i'm looking for a solution to switch the reference directory, that
all modules will be loaded from "/lib/modules/2.6.17-NEW/".
Regards
bademeister

Arch Kernel Release Cycle

Hi guys,
I recently started compiling my own kernels, but I still use the arch packages as a guideline. For example I use the patches provided by Arch Linux, and at the moment I stay with the version which is in the arch repositories.
So for example if I want to compile 2.6.35.7, the PKGBUILD I use just downloads the source for 2.6.35, and the patch for .7 from Arch. Would it be 'possible' to use the stock kernel patches? What is different in the Arch packages (I know I can just diff, but why are there Arch specific patches)?
Arch gernerally updates the kernel rather fast. But some versions are skipped. Is there any policy on that? For example at the moment 2.6.36 is out, yet Arch packages 2.6.35.8 as the latest.
Thanks and kind regards
Julian

2.6.36 is in [testing]. Kernels usually do not make it to [core] until the around .2 version.
The Arch specific patches are rather small... http://projects.archlinux.org/linux-2.6 … ee/patches AUFS2 and some logo replacements.

[Solved] Installed Arch, I have a kernel problem.

Hi guys, I'm new here, but I use only linux since 2003. I have a problem at boot... When I choose arch from the bootloader I get this:
ERROR: Failed to parse block device name for '/dev/sda6'
ERROR: root fs cannot be detected. Try using the rootfstype= kernel parameter
Or something very very similar. I see that I'm not the only one experiencing this problem so I try solving by myself following hints I've found on this forum... but nothing changed.
I need help
This is what you have to know:
/etc/mkinitcpio.conf:
MODULES=""
HOOKS="base udev autodetect pata keymap filesystems"
I also tried with ide...
fstab
# /etc/fstab: static file system information
# <file system> <dir> <type> <options> <dump> <pass>
none /dev/pts devpts defaults 0 0
none /dev/shm tmpfs defaults 0 0
/dev/cdrom /mnt/cdrom iso9660 ro,user,noauto,unhide 0 0
/dev/cdrom1 /mnt/cdrom1 iso9660 ro,user,noauto,unhide 0 0
/dev/dvd /mnt/dvd udf ro,user,noauto,unhide 0 0
/dev/fd0 /mnt/fd0 vfat user,noauto 0 0
/dev/sda2 swap swap defaults 0 0
/dev/sda6 / ext3 defaults 0 1
grub entry:
#Archlinux
title ArchLinux (on /dev/sda6)
root (hd0,5)
kernel /boot/vmlinuz26 root=/dev/sda6
initrd /boot/kernel26.img
savedefault
boot
Last edited by Raffo (2007-07-14 16:34:42)

That's very curious. I'm just having this problem after installing the lowlatency kernel from this repo:
Server = http://arch.madfire.net/proaudio/i686
So I've just changed the initrd /boot/kernel26.img to /boot/kernel26rt.img and I'm getting exactly the same error. I've also tried changing sda with hda in the grub's menu.lst but nothing changed...

Arch Kernel Questions

(This is a really low-priority set of questions.)
I just wanted to know what some of the differences between the main stable 2.6 kernel and Arch's 2.6 kernel are. I'm not a fan of compiling kernels, so I'd like to know what decisions are made for the default kernel.
Let me be more specific;
1. linux-2.6.26.1/Documentation/java.txt: apparently, native support for java binaries is planned for the kernel. Will this be reflected in Arch's kernel as well?
2. Besides x86, what architectures are supported by the Arch kernel (linux-2.6.26.1/arch)?
3. Does the Arch kernel support virtualization by default?
4. Which drivers (if any) are excluded from Arch's kernel (linux-2.6.26.1/drivers)? (If any, why?)
5. Which filesystems are supported by the Arch kernel (linux-2.6.26.1/fs)? In the install, there are only 5 filesystems available so I'm guessing it's just those 5?
--- And any other major differences that are worth noting.
Thank you for reading!
- vsk

there is a bug with xen in kernels newer then 2.6.18.
arch's kernel is as vanilla as the stable release is from what I gather. Other then a few patches and some config changes. Arch tries to stay as close to vanilla as possible with all it's packages. just do a diff between the two configs.

New FaunOS Live Arch-Based DVD/USB Distro Now Available

Hello fellow archers,
We have just released the latest version of FaunOS. Since this version of FaunOS is the answer to the most important question in the universe, we have named it "fortytwo."
FaunOS-fortytwo contains many improvements over previous releases and more packages (575 total). Improvements, in no particular order, include:
- Improved wireless support, both wlassistant and knetworkmanager are now included.
- Improved automatic X configuration with 3D acceleration.
- Better boot splash support. We have moved to fbsplash which seems to work pretty consistently and has lead to a faster boot time.
- A new boot option for quickly booting into a console-based system with no X windows.
- A new boot option for machines that hang when booting from DVDs using piix modules.
- Ability to save FaunOS session back to a multi-session DVD-+R or DVD-+RW.
- Permanent Arch Linux Installer using archin.
- Permanent FaunOS Installer for creating additional FaunOS USB drives from within FaunOS and for performing a "frugal" install to hard drive. Frugal install (not to be confused with Frugal Linux) installs the compressed FaunOS files to a hard drive permanently leading to an "unbreakable desktop" meaning the end user can totally ruin everything at runtime. To revert back to the starting point all the user has to do is reboot the machine and everything goes back to normal since all changes happen in ram, unless the FaunOS session is saved at boot time.
- FaunOS is probably the only live distro that can perform two kinds of installations.
- New added packages include: wlassitant, madwifi, firmware packages for ipw2100 and ipw2200, ndiswrapper, lynx for text-based web browsing, and kmobiletools accessing phone books and other information on mobile phones.
- All packages have been upgraded to the latest from the Arch repos as of today, 8/27/2007.
- And, most importantly, a new logo.
We have also decided to stay away from the evil 2.6.22 kernel which seems to have created more bugs and headaches than introducing new features, no flame war intended. The latest 2.6.21-faunos kernel is based on the stock arch kernel with the addition of the fbsplash patch from genpatches.
There are those people that may think FaunOS is bloated, but we like to think that it is a muscular and strong live Linux distribution, thanks to Arch. Of course keeping with Arch's KISS principal, we have tried to make FaunOS simple and easy enough for anyone to use.
You can get the latest version at:
http://www.faunos.com/downloads/faunos-fortytwo
A new wiki has been created at http://wiki.faunos.com. Feel free to contribute and change the wiki as you wish.
Our support and discussion forums can still be found at http://forum.faunos.com
We can definitely use some mirrors for our downloads. We will also be creating a torrent shortly and can use some seeders. Please let us know either here or in the FaunOS forums if you can help in any way.
We'll see ya at the FaunOS forums!
So long and thanks for all the fish,
Raymano

We have simply used larch but with some additions and changes:
- FaunOS uses a slightly modified version of larch with the added ability to build a live system from a preinstalled partition on the harddrive.
- Added fbsplash capabilities to the initial initrd images and kernel.
- Applications on FaunOS are preconfigured and integrated to work out of the box.
- Some kernel specific modules/packages have been compiled for the FaunOS kernel, e.g. ndiswrapper.
- All services are pre-configured to execute out of the box.
- All available codecs are included.
- 3D desktop using beryl is preconfigured to run on intel based graphics cards and some ATI cards with a click of a button.
- Network management is simplified with use of knetworkmanager and wlassistant.
- Storage are is included on the FaunOS device to share files with other platforms.
- R/W access Windows parititions by clicking on an icon in KDE.
- Common grub booting on USB and DVD. With multiple boot options for graphical and non-graphical system.
- FaunOS's own mkxorgconf utility was added (using hwd) for better automatic configuration of graphics cards. This includes setting up AIGLX.
- Easily performs frugal install of the live FaunOS system to hard drive, including saved overlays. This was not part of the original larch scripts.
- Use of sudo to allow the user to perform certain system wide configuration tasks.
- Integrated flash java and mplayer plugins for Firefox.
- Ability to put a computer running FaunOS into sleep (suspend-to-ram), remove the FaunOS media, boot and work on other computers with the FaunOS device, return to the original computer, insert the FaunOS device, turn on the computer and continue working where you left off. This in not possible with latest Arch Linux because of the kernel.
- And we believe, most importantly, you can hand a FaunOS DVD to a user with no knowledge of Linux and say go try this out.
Last edited by raymano (2007-10-09 17:38:40)

Kernel 2.6.11.y - is arch going to follow?

Similar Messages

Maybe you are looking for