It's not clear if there was a bug in the original version and zroot_vdevs should have been alwas set to "$zroot_vdevs $disk$targetpart.eli" in the ZFSBOOT_GELI_ENCRYPTION case?

There was a discussion on this on the freebsd-fs list in June (e.g. http://lists.freebsd.org/pipermail/freebsd-fs/2014-June/019535.html). The consensus seemed to be that forcing the ZFS block size to larger than the stripe size was actually harmful in many cases and that this feature should be removed entirely rather than shifted to use the sysctl.

In D566#5, @nwhitehorn wrote:

There was a discussion on this on the freebsd-fs list in June (e.g. http://lists.freebsd.org/pipermail/freebsd-fs/2014-June/019535.html). The consensus seemed to be that forcing the ZFS block size to larger than the stripe size was actually harmful in many cases and that this feature should be removed entirely rather than shifted to use the sysctl.

As a user of said I'm afraid I would have to disagree with this conclusion for two reason:

1. We can't garanttee that 512b devices will continue to be available, we've had number of cases where we've had to replace disks with newer 4k version, which results in zpool status warning
2. Its often the case that the reported stripe size is wrong, usually a missing quirk

Given this we should definitely not remove this option, which is an administration choice.