le Tue, 29 May 2007 21:38:17 -0700 "Wil Reichert" <[EMAIL PROTECTED]> a écrit:
> On 5/29/07, Joshua Hoblitt <[EMAIL PROTECTED]> wrote: > > On Wed, May 30, 2007 at 02:33:18AM +0200, Florian D. wrote: > > > FYI, genkernel is creating an initrd, not an initramfs, which is the > > preferred way nowadays. > > > Information on how to setup an initramfs can be found at: > > > http://lldn.timesys.com/docs/initramfs > > > > Umm, I think you need to check your facts. genkernel creates a gzip'd > > CPIO archive named "initramfs-genkernel-arch-versionstring"... > > So the command 'genkernel initrd' creates a file called > 'initramfs-...' which contains files called etc/initrd.defaults and > etc/initrd.scripts. Poor naming conventions but it looks like an > initrd to me. > I'm "afraid" it isn't. Try zcat initramfs | cpio -t . initramfs are cpio archives. And genkenrel is such a wild beast, that it compiles a static busybox against uclibc _if_ you have an uclibc toolchain available for your arch through crossdev (this feature really impresses me). If I'm not wrong, when you haven't got any such toolchain, it uses a prebuilt version of busybox. Those informations were gathered empirically by using genkernel. Could someone confirm/infirm/precise? -- [EMAIL PROTECTED] mailing list
