Removed mirrorlist file from overlay and modified Makefile to fetch
the file dynamically on iso creation (Thanks, Dan!)
Signed-off-by: Gerhard Brauer <gerbra@archlinux.de>
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
* Added ntp to packages list (Both ARCHes)
* Modified Makefile to parse the CARCH related packages.$ARCH
We need different lists cause not all packages exist for both arches.
And it's bad to edit in the packages.list each build.
Signed-off-by: Gerhard Brauer <gerbra@archlinux.de>
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
Added a script to overlay which call aif's interactiv procedure when
/arch/setup is started. Also removed old arch-installer from
packages.list
Signed-off-by: Gerhard Brauer <gerbra@archlinux.de>
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
We had a few typos in overlay/etc/pacman.d/mirrorlist.
Also upated to the last mirrorlist data file.
This will fix FS#13970.
Signed-off-by: Gerhard Brauer <gerbra@archlinux.de>
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
This should give us *just a bit* more delay before actually trying to do
things with the usb devices. Annoying, but will probably be effective
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
This is used currently for usbdelay related settings, but could be used for
other things in the future
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
USB boot delay is now handled with the usbdelay kernel param (default=0)
We use the built in delay of usb-storage to control this. by setting the
module's delay param. If your USB device doesn't work on first boot, try
setting usbdelay=10 or so
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
The x86_64 images were installing i686 systems. This
was due to the fact that the mirrorlist was in the overlay.
A new devtools release (with the -n arg) will fix this when
running mkarchroot. To be removed later
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
This is crappy and hacky but enables us to build ALL
images faster, allowing for faster iteration
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
It seems udev trigger/settle doesn't do what we expect,
so we'll add functionality for the rootdelay param back
in here.
This should help fix FS#12896
Original-idea: Gerhard Brauer <gerbra@archlinux.de>
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
If testing is enabled, -Sp will give us testing URLs.
Use the $repo/$pkgname format for packages so we get
the right files
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
