* TI boards vs. distro boot
@ 2020-05-19 7:35 Jan Kiszka
2020-05-21 17:05 ` Lokesh Vutla
0 siblings, 1 reply; 3+ messages in thread
From: Jan Kiszka @ 2020-05-19 7:35 UTC (permalink / raw)
To: u-boot
Hi Lokesh,
seems like all boards including include/environment/ti/mmc.h will only
search for ${boot_scripts} in the root of the target fs, not in
${boot_prefixes}. Known bug? Used to work fine with your downstream
U-Boot some years ago. Just updated an SD card, only replacing the
bootloader artifacts, and that card now no longer boots.
Jan
--
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 3+ messages in thread
* TI boards vs. distro boot
2020-05-19 7:35 TI boards vs. distro boot Jan Kiszka
@ 2020-05-21 17:05 ` Lokesh Vutla
2020-05-22 6:45 ` Jan Kiszka
0 siblings, 1 reply; 3+ messages in thread
From: Lokesh Vutla @ 2020-05-21 17:05 UTC (permalink / raw)
To: u-boot
+Tom
Hi Jan,
On 19/05/20 1:05 pm, Jan Kiszka wrote:
> Hi Lokesh,
>
> seems like all boards including include/environment/ti/mmc.h will only search
> for ${boot_scripts} in the root of the target fs, not in ${boot_prefixes}. Known
It searches in boot partitions of SD cards and this was always the case from a
long time.
> bug? Used to work fine with your downstream U-Boot some years ago. Just updated
Any chance you re collect which U-Boot version?
> an SD card, only replacing the bootloader artifacts, and that card now no longer
> boots.
We are planning to move to distro boot completely. It should clean all the env
variables.
Thanks and regards,
Lokesh
>
> Jan
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* TI boards vs. distro boot
2020-05-21 17:05 ` Lokesh Vutla
@ 2020-05-22 6:45 ` Jan Kiszka
0 siblings, 0 replies; 3+ messages in thread
From: Jan Kiszka @ 2020-05-22 6:45 UTC (permalink / raw)
To: u-boot
On 21.05.20 19:05, Lokesh Vutla wrote:
> +Tom
>
> Hi Jan,
>
> On 19/05/20 1:05 pm, Jan Kiszka wrote:
>> Hi Lokesh,
>>
>> seems like all boards including include/environment/ti/mmc.h will only search
>> for ${boot_scripts} in the root of the target fs, not in ${boot_prefixes}. Known
>
> It searches in boot partitions of SD cards and this was always the case from a
> long time.
There is no boot partition in our case. So your open-code approach does
not find a /boot/boot.scr on the root partition. It does not evaluate
boot_prefixes for the script because it does not use scan_dev_for_boot &
friends.
>
>> bug? Used to work fine with your downstream U-Boot some years ago. Just updated
>
> Any chance you re collect which U-Boot version?
>
Oh, I found the reason: I fixed up that BSP with an own environment back
then, to emulate what proper distro boot would have found.
>> an SD card, only replacing the bootloader artifacts, and that card now no longer
>> boots.
>
> We are planning to move to distro boot completely. It should clean all the env
> variables.
That will likely help. Anything that complicates this because of
backward compatibility? Otherwise, it should just be about removing own
defines. I just cleaned up our k3-based board this way, simply dropping
all the stuff that piled up during early development.
Jan
--
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-05-22 6:45 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-19 7:35 TI boards vs. distro boot Jan Kiszka
2020-05-21 17:05 ` Lokesh Vutla
2020-05-22 6:45 ` Jan Kiszka
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.