svn commit: r264935 - head/release/amd64

[Nathan Whitehorn]

On 04/25/14 15:00, Adrian Chadd wrote:
> On 25 April 2014 14:41, Nathan Whitehorn <nwhitehorn at freebsd.org> wrote:
>> On 04/25/14 14:11, Glen Barber wrote:
>>> Author: gjb
>>> Date: Fri Apr 25 21:11:00 2014
>>> New Revision: 264935
>>> URL: http://svnweb.freebsd.org/changeset/base/264935
>>>
>>> Log:
>>>     Add a separate script to build the memstick.img and the
>>>     mini-memstick.img with UEFI support.
>>>        As the comments in the file suggest, 1) there must
>>>     be existing ${.OBJDIR}/usr/src/release/{release,bootonly};
>>>     2) TARGET/TARGET_ARCH must be amd64; and 3) it must be
>>>     a vt(4)-enabled kernel with vt_efifb (*not* vt_vga).
>>>        This script is not hooked into release/Makefile in any way
>>>     until further testing is complete.
>>>        Sponsored by:     The FreeBSD Foundation
>>>
>>> Added:
>>>     head/release/amd64/make-uefi-memstick.sh   (contents, props changed)
>>>
>>>
>> I understand this is for testing, but is there any reason we can't just have
>> one GPT memstick that boots both BIOS and EFI eventually? Or is that the
>> plan already?
> Because BIOSes don't get enough testing.
>
> It's the same reason why I'm highly tempted to figure out how to teach
> the installer to do MBR installs out of the box as well as GPT
> installs out of the box, because I've found a bunch of BIOSes on
> recent hardware that seems to assume GPTs == EFI booting, and just
> plain don't boot your system. (And yes, it'd be nice to have EFI work
> there, but there's still no i386 EFI support.)
>

The installer *does* do MBR out of the box. Just format the disk with 
MBR and press "Auto". It also does straight bsd labels in the same way.
-Nathan