wandboard does not boot with current mainline

[Heiko Schocher]

Hello Fabio,

Am 07.05.2020 um 23:16 schrieb Fabio Estevam:
> Hi Heiko,
> 
> On Thu, May 7, 2020 at 9:12 AM Heiko Schocher <hs at denx.de> wrote:
>>
>> Hello Fabio,
>>
>> I have my wandboard DL in my automated daily build setup example:
>>
>> http://xeidos.ddns.net/ubtestresults/
>>
>> and wondered, why my last result is from may 4th ...
>>
>> (Ok, there is a bug, that I do not see not booting boards in this page,
>>    but this is just a proof of concept page ...)
>>
>>
>> So, what I see is:
>>
>> │   │   ├─UBOOT (wandboard-uboot)
>> │   │   │    <> Connecting to
>> /dev/serial/by-id/usb-Prolific_Technology_Inc._USB-Serial_Controller_D-if00-port0, speed 115200
>> │   │   │    <>  Escape character: Ctrl-\ (ASCII 28, FS): enabled
>> │   │   │    <> Type the escape character followed by C to get back,
>> │   │   │    <> or followed by ? to see other options.
>> │   │   │    <> ----------------------------------------------------
>> │   │   │    <>
>> │   │   │    <> U-Boot SPL 2020.07-rc1-tbot-00298-g425fefa9a3 (May 05 2020 - 04:18:47 +0200)
>> │   │   │    <> Trying to boot from MMC1
>>
>>
>> and no more output ...
>>
>> Do you have any idea, what is wrong with current mainline?
> 
> Thanks for the report.
> 
> This boot failure is caused by:
> 
> commit 20a154f95bfe0a3b5bfba90bea7f001c58217536
> Author: Marek Vasut <marex at denx.de>
> Date:   Fri May 1 17:40:25 2020 +0200
> 
>      mkimage: fit: Do not tail-pad fitImage with external data
> 
>      There is no reason to tail-pad fitImage with external data to 4-bytes,
>      while fitImage without external data does not have any such padding and
>      is often unaligned. DT spec also does not mandate any such padding.
> 
>      Moreover, the tail-pad fills the last few bytes with uninitialized data,
>      which could lead to a potential information leak.
> 
>      $ echo -n xy > /tmp/data ; \
>              ./tools/mkimage -E -f auto -d /tmp/data /tmp/fitImage ; \
>              hexdump -vC /tmp/fitImage | tail -n 3
> 
>      before:
>      00000260  61 2d 6f 66 66 73 65 74  00 64 61 74 61 2d 73 69
> |a-offset.data-si|
>      00000270  7a 65 00 00 78 79 64 64                           |ze..xydd|
>                         ^^       ^^ ^^
>      after:
>      00000260  61 2d 6f 66 66 73 65 74  00 64 61 74 61 2d 73 69
> |a-offset.data-si|
>      00000270  7a 65 00 78 79                                    |ze.xy|
> 
>      Signed-off-by: Marek Vasut <marex at denx.de>
>      Reviewed-by: Simon Glass <sjg at chromium.org>
>      Cc: Heinrich Schuchardt <xypron.glpk at gmx.de>
>      Cc: Tom Rini <trini at konsulko.com>
> 
> which has been reverted by Tom in commit:
> 
> commit 7946a814a31989998120b4b4aa417222ba21b2fa
> Author: Tom Rini <trini at konsulko.com>
> Date:   Wed May 6 11:05:17 2020 -0400
> 
>      Revert "mkimage: fit: Do not tail-pad fitImage with external data"
> 
>      This has been reported to break booting of U-Boot from SPL on a number
>      of platforms due to a lack of alignment of the external data.  The
>      issues this commit is addressing will need to be resolved another way.
> 
>      Re-introduce a data leak in the padding for now.
> 
>      This reverts commit 20a154f95bfe0a3b5bfba90bea7f001c58217536.
> 
>      Reported-by: Alex Kiernan <alex.kiernan at gmail.com>
>      Reported-by: Michael Walle <michael at walle.cc>
>      Tested-by: Jan Kiszka <jan.kiszka at siemens.com>
>      Signed-off-by: Tom Rini <trini at konsulko.com>
> 

Yep, thanks! Board works again, see:

http://xeidos.ddns.net/ubtestresults/result/60

BTW: since commit c693f212c5b0 size of u-boot.img has increased
from 561004 bytes to 562068 bytes for the wandboard

http://xeidos.ddns.net/ubtestresults/stats/wandboard_defconfig/8

bye,
Heiko
-- 
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-52   Fax: +49-8142-66989-80   Email: hs at denx.de