Re: prepare 2.6.13
On Tue, Aug 30, 2005 at 10:40:12AM +0200, Marco Amadori wrote:
> On Tuesday 30 August 2005 07:47, you wrote:
>
> > > > > > > - initrd-tools needs to loose devfs usage.
> > > > > > BTW, isn't it time to switch to initramfs ?
> > >
> > > I saw that ubuntu use "initramfs-tools" and in debian we have "yaird"
> > > that I personally use since it hit sid to boot from lvm on a md array.
> > >
>
> > My personal believe is that something like yaird will produce smaller
> > intram filesystems, and thus better suited to arches and subarches who have
> > actually trouble with bigger bloated initrds.
>
> This is sure but I think will be a really nice feature to have in etch, with
> initramfs-tools concept we could change hardware quite a few and expect our
> beloved debian to boot without problem, recognizing at startup that e.g. we
> changed the mainboard and cpu so the old sata disc is mounted on a sas
> controller and change the module to load without leaving a un-bootable sistem
> like out actual mkinitrd and it seems also yaird.
Well, agreed, both have their place. I think some of our supported arches will
like a yaird-like situation or at least a more user-controlled initrd
situation better over a huge bloated all-encompassing initramfs.
> > I am also a bit curious how initramfs behaves on the kernel wise, i mean i
> > understand and all that it is just a bunch of cpio archives which are
> > copied at the end of the kernel, but, taking the powerpc case for an
> > example, there is either a bootwrapper, like the arch/ppc/boot/openfirmware
> > stuff, or boot loaders, like yaboot, who know how to put the kernel and
> > initrd in ram, and then pass the right info to the kernel to find the
> > ramdisk back (passed as argument in r3/r4 if i remember well.
>
> I remebered booting linux from the destkop gui on in my Amiga times...
Well, yes, but this is not the question, the question is how the kernel knows
about initramfs, beyond the way yaboot, lilo, amiboot or whatever gets this
info from the user.
> > Now, initramfs is nothing more than a file organisation which is a bit
> > different for the initial ramdisk, or is there more to it, and the above
> > code path, for which i have seen no major change recently, will still work
> > ?
>
> The question is : we want to support nice things like EVMS at boot time? A
> tool for generating bootable initrd for evms is needed, it is targeted for
> etch evms support I hope.
>
> It seems that none our 3 tools supports it right now.
But right now is the time to investigate those issues, and fix the tools if
possible, and not 6-12 month down the road, when we will be in late-freeze
situation or something.
Let's maybe list all the things we want for sucha tool. My personal requisite
are :
- allow as well for the creation of generic images, as well as minimal ones.
- allow for hand selected inclusion list as well as exclusion of modules.
- allow to include script as well as mklibs-manipulated binaries and
libraries.
Friendly,
Sven Luther
Reply to: