GitLab

_BSD_SOURCE doesn't work on NetBSD, so explicitly request the full native namespace.

Signed-off-by: Michal Suchanek <hramrach@gmail.com>
Signed-off-by: Henrik Nordstrom <henrik@henriknordstrom.net>

When - is specified usb-boot interprets it as kernel so the usage
message is clearly bogus.
Signed-off-by: Michal Suchanek <hramrach@gmail.com>
Signed-off-by: Henrik Nordstrom <henrik@henriknordstrom.net>

The U-Boot code has changed slightly, so adjust the tool to output
proper, working code.
Signed-off-by: Emilio López <emilio@elopez.com.ar>

This rename goes with the previous rename of fel-boot -> fel-sdboot
Signed-off-by: Olliver Schinagl <oliver@schinagl.nl>

Signed-off-by: Patrick Wood <patrickhwood@gmail.com>

Signed-off-by: Patrick Wood <patrickhwood@gmail.com>

Signed-off-by: Patrick Wood <patrickhwood@gmail.com>

Build one nand-part program to handle both A10 and A20 mbr header formats.
Changed -f option to take an "a10" or "a20" argument to specify which mbr
format to force.
Signed-off-by: Patrick Wood <patrickhwood@gmail.com>

They were all built with no version information so it's not trivial
to tell which sources they are built from.

It is very easy to rebuild u-boot which gives these. Not really a need
to have them prebuilt here.

A20 boot1 checks the MBR magic string but not the version, but let's
check it anyway (can always force now with the -f option if the version
isn't 0x200 on the system)

command usage and partition numbering in linux nand driver

offset of partition 1 isn't specified, switch to old argument syntax that retains
partition 1 values

allow setting of partition 1 size/offset
if -f option is used, force writing of MBR header, even if CRC or
header magic/version are incorrect (for recovery only)

Conflicts:
	nand-part.c

Signed-off-by: Michal Suchanek <hramrach@gmail.com>
Signed-off-by: Henrik Nordstrom <henrik@henriknordstrom.net>

Signed-off-by: Michal Suchanek <hramrach@gmail.com>
Signed-off-by: Henrik Nordstrom <henrik@henriknordstrom.net>

Signed-off-by: Alejandro Mery <amery@geeks.cl>

closes #16

The stubs is only needed because we are using objects from a full
SPL build. Even these will go away when felboot is integrated in
u-boot, reducing it to only a link script and config changes.

as expected there was no problem to use sunxi_board_init directly
from u-boot. Failure on A13 was a timing issue in the host side
script (need to wait for DRAM init to finish before trying next
fel operation)

just need to wait a little for the DRAM initialization to finish. Can't
use fel until we know it's finished or things breaks down.