On an experiment of the FreeBSD EFI implementation. I installed a copy of releng/12 from install media. Which left me with: # gpart show ada0 => 40 312581728 ada0 GPT (149G) 40 409600 1 efi (200M) 409640 31047680 2 freebsd-ufs (15G) 31457320 7680000 3 freebsd-swap (3.7G) 74788904 237792864 - free - (141G) On this Intel based system, I can stab the F12 key to pick my UEFI bootable OS, or let it boot according to the order I setup in the BIOS. So far, so good. I needed a copy of releng/13 to also work with. Installed a copy from install media. Which left me with: # gpart show ada0 => 40 312581728 ada0 GPT (149G) 40 409600 1 efi (200M) 409640 31047680 2 freebsd-ufs (15G) 31457320 7680000 3 freebsd-swap (3.7G) 39137320 532480 4 efi (260M) 39669800 35119104 5 freebsd-ufs (17G) 74788904 237792864 - free - (113G) I *assumed* that the install would activate the new install, and I would boot straight into it. But no. I am still on the previous install, and worse, I can't get into the new install -- even if picking it via stabbing the F12 key. I *still* end up in the previous install. So looking at what might be causing it. I found the following: # releng/12 # mount -t msdosfs /dev/ada0p1 /mnt/ # ls /mnt/efi/boot/ BOOTx64.efi startup.nsh # cat /mnt/efi/boot/startup.nsh BOOTx64.efi # umount /mnt/ releng/13 # mount -t msdosfs /dev/ada0p4 /mnt/ # ls /mnt/EFI/freebsd/ loader.efi Why the difference? When will FreeBSD (u)EFI work as expected? Thanks in advance for any insights! --ChrisReceived on Fri Mar 27 2020 - 20:49:16 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:41:23 UTC