klibc 1.5.14-1, klibc-extras 2.5-1, klibc-udev 128-1, klibc-kbd 1.15.20080312-7, klibc-module-init-tools 3.4-2, v86d 0.1.7-1
 

I updated klibc from 1.5 to the Testing 1.5.14. We had to patch 1.5 so
heavily to even work and it had evil bugs, so going with 1.5.14 is better.


First of all, it won't upgrade smoothly due to a pacman bug. Please do
rm /usr/lib/klibc/include/asm
to remove a symlink that causes lots of false file conflicts
(http://archlinux.org/pipermail/pacman-dev/2008-September/012971.html).
After that you can update.


If you want to test these, DO NOT OVERWRITE YOUR CURRENT INITRAMFS AND
DO NOT WAIT FOR A KERNEL UPDATE TO DO SO. I only ran x86_64 on one
machine, i686 is completely untested. Please do the following:

mkinitcpio -k 2.6.26-ARCH -g /boot/kernel26.img.newklibc
Then boot the kernel26.img.newklibc instead of kernel26.img (that can be
done directly from grub).


Here is the news:
- Removed mknod from klibc-extras, using the klibc one instead
- Update to klibc 1.5.14, fixing a memory corruption and probably more
- Add resolve-modalias to klibc-extras
- Update klibc-udev, it should now be in the same state as our udev
package. Module blacklisting with disablemodules= or load_modules=off
from the kernel commandline should now behave exactly like in the real
system.
- Updated v86d, I hope the segfault bug on i686 is now fixed due to the
klibc update.


Please test thoroughly and sign off.

klibc 1.5.14-1, klibc-extras 2.5-1, klibc-udev 128-1, klibc-kbd 1.15.20080312-7, klibc-module-init-tools 3.4-2, v86d 0.1.7-1
 

On Thu, 18 Sep 2008 19:39:31 +0200
Pierre Schmitz <pierre@archlinux.de> wrote:

> Am Donnerstag 18 September 2008 18:48:08 schrieb Thomas Bächler:
> > Please test thoroughly and sign off.
>
> works fine on i686 and x86_64 (but I don't use any special setup)
>

Works here on x86_64. But same as Pierre, I don't use any special setup.

klibc 1.5.14-1, klibc-extras 2.5-1, klibc-udev 128-1, klibc-kbd 1.15.20080312-7, klibc-module-init-tools 3.4-2, v86d 0.1.7-1
 

On Thu, 18 Sep 2008, Tobias Powalowski wrote:


Am Donnerstag 18 September 2008 schrieb Daniel Isenmann:

On Thu, 18 Sep 2008 19:39:31 +0200

Pierre Schmitz <pierre@archlinux.de> wrote:

Am Donnerstag 18 September 2008 18:48:08 schrieb Thomas Bächler:

Please test thoroughly and sign off.


works fine on i686 and x86_64 (but I don't use any special setup)


Works here on x86_64. But same as Pierre, I don't use any special setup.

works fine on raid 1 setup here
signoff x86_64
greetings
tpowa




signoffs: i686 - IDE
x86_64 - SATA with RAID & LVM

Eric
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

klibc 1.5.14-1, klibc-extras 2.5-1, klibc-udev 128-1, klibc-kbd 1.15.20080312-7, klibc-module-init-tools 3.4-2, v86d 0.1.7-1
 

Dan McGee wrote:

On Fri, Sep 19, 2008 at 9:42 PM, Eric Belanger
<belanger@astro.umontreal.ca> wrote:


On Thu, 18 Sep 2008, Tobias Powalowski wrote:



Am Donnerstag 18 September 2008 schrieb Daniel Isenmann:


On Thu, 18 Sep 2008 19:39:31 +0200

Pierre Schmitz <pierre@archlinux.de> wrote:


Am Donnerstag 18 September 2008 18:48:08 schrieb Thomas Bächler:


Please test thoroughly and sign off.


works fine on i686 and x86_64 (but I don't use any special setup)


Works here on x86_64. But same as Pierre, I don't use any special setup.


works fine on raid 1 setup here
signoff x86_64
greetings
tpowa




signoffs: i686 - IDE
x86_64 - SATA with RAID & LVM



i686 with IDE setup is working fine. I even took the plunge and didn't
create a backup initrd. :)

Make sure we get a news item since this upgrade path is a bit funky
due to the file conflicts.



Is this waiting on something to be moved from [testing]?

Allan