It is now possible to pass multiple crypt_roots= and crypt_swaps=
parameters (mind the "s") and have multiple devices concurring to
building the final root device and swap areas activated.
The change is backward compatible and crypt_root=, crypt_swap= is
still fully supported.
This change makes possible to support multiple device mapper block
devices grouped together into aggregated software raid devices.
For instance, two individual luks devices can be now used to build
a single raid0 or raid1 md device.
Moreover, genkernel-next initramfs now supports multiple encrypted
swap devices.
tuxonice is almost dead. Last update was in 2011 and they're now
trying to resurrect this ugly and non-upstream friendly thing.
Moreover, the tuxonice genkernel code is the ugliest thing I have
ever seen in it. Well no, unionfs was even worse.
So please, let's forget about tuxonice for now and focus on the
good looking code. I am willing to accept patches that restore
tuxonice support if they're good quality ones.
Burn tuxonice init code, burn in hell.
"quiet" is already used by the kernel to avoid printing messages on the
console unless they are errors or warnings. Genkernel should do the same
wrt its initramfs initialization code.
This has also the advantage of improving the boot speed.
scandelay=<secs> should be always preferred. Moreover, nowadays
automatically sleeping waiting for USB storage to come up in
the way we were used to do it doesn't work as one expects, because
USB is very common and USB storage devices are very likely to be
present on a system.
Furthermore, the initialization code after setup_slowusb is already
waiting for devices to come up.
Nearly every general documentation including the one in the
kernel tree and the scripts for GRUB2 all expects this flag
to work.
Signed-off-by: Peter Hjalmarsson <xake@rymdraket.net>
Clean up documentation and functionality of "part" argument to
initramfs. It is meant to imitate the same option to the kernel.
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
We only support creating initramfs nowdays, so RAM disk support is not needed.
Since initramfs does not need root=/dev/ram0 (and we actually ignore it if it is set),
we should not tell people to use it.
We should also stop tell people to use real_root.
Conflicts:
genkernel
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Somewhere in the mists of time we lost part of the multipath patch, so
we were bundling the userspace, but not the kernel modules, or ever
using it.
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
When using an integrated initramfs (including in kernel during build),
it's better NOT to compress the initramfs, as the kernel build process
will then be re-compressing it. This can provide space savings on disk
as well as potential time & memory savings (one less decompress pass is
needed before the kernel can use the initramfs).
This functionality was previously available in the undocumented
COMPRESS_INITRD variable.
Code-by: Richard Yao <ryao@cs.stonybrook.edu>
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
(commit message rewritten).
Provide real boolean commandline options for options that were
previously only available as booleans in the config, so that they can be
enable in configuration and overridden to disable on the commandline.
Remove negation on options: NOINSTALL, NORAMDISKMODULES,
NO_KERNEL_SOURCES.
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Make it possible to choose a custom mdadm.conf for your ramdisk.
If not set, no mdadm.conf will be copied, and ramdisk will auto-detect.
This also fixes a regression where genkernel always tried to copy
/etc/mdadm.conf to the ramdisk, failing if it was non-existant.
Fabio Erculiani
Vadim A. Misbakh-Soloviov
Peter Hjalmarsson
Robin H. Johnson
Rick Farina
Richard Yao
Sebastian Pipping
Christian Kruse
Ian