Re: Question on conversion from initrd to initramfs (1/2027 initrd support drop from kernel)
Didier Kryn via busybox <[email protected]>
| Newsgroups | gmane.linux.busybox |
|---|---|
| Message-ID | <[email protected]> |
Le 05/03/2026 à 00:35, Jody Bruchon via busybox a écrit : > 10+ years ago, I wanted to switch from initrd to initramfs. I had an > extremely difficult time getting it all correct because switch_root > has a lot of extremely specific requirements that you must meet before > calling it, and any failure immediately after calling switch_root can > be notoriously hard to debug. It's also critical that you use "cpio -o > -H newc" to make the initramfs, then any compression format that your > kernel supports can be used to compress it. More than 20 years ago I had to craft my own initramfs because there was no other choice. I was building kernel+initramfs for Powerpc-based diskless SBCs. AFAIR, the requirements of switch_root was the existence of an executable init file at the expected path in the new rootfs, plus mounted /sys, /proc, /dev, /dev/pts. That said, it is possible that the init sequence of the new rootfs expects more, eg a tmpfs /run. The init program of the initramfs was a script. In order to debug it more easily, when switch_root failed it entered an interactive shell in which I could correct things from the command-line and then exit, or even halt/reboot. When exiting, it retried switch_root, etc... From this shell, one could also try chroot to observe the result. Hope this helps. -- Didier _______________________________________________ busybox mailing list [email protected] https://lists.busybox.net/mailman/listinfo/busybox