[U-Boot] sdcard with messed up partition table with FAT signature on header

[U-Boot] sdcard with messed up partition table with FAT signature on header

[Darwin Dingel]

Hi,

I have a sdcard with a messed up partition table with a FAT signature on
the header. This is readable under linux but not when I do a 'fatls' which
results in "Partition 1 not valid on device 0".

Before get_device_and_partition() was introduced this was still working
because reading the 'fatls' will call fat_register_device(). After it fails
to detect a valid partition due to the messed up partition table, it will
then search the header for a FAT signature as a fallback and will continue
to read it as a PBR. 'fatls' will be able to read it afterward.

Unfortunately, this will now fail on the latest code. I have a patch to do
the fallback check but under get_partition_info_extended(). However I am
unsure if there is really a reason for the absence this FAT signature
checking before failing fatls command.

Do anyone have ideas on a reason for this FAT signature checking not being
implemented in get_device_and_partition() (or
get_partition_info_extended())?


Thanks,

Darwin

[U-Boot] sdcard with messed up partition table with FAT signature on header

[Josh Wu]

Hi, Darwin

On 6/5/2014 11:13 AM, Darwin Dingel wrote:
> Hi,
>
> I have a sdcard with a messed up partition table with a FAT signature on
> the header. This is readable under linux but not when I do a 'fatls' which
> results in "Partition 1 not valid on device 0".

Do you define the CONFIG_DOS_PARTITION in your board config head file 
(include/configs/xxxx.h)?
If not defined that macro will cause above error.

Best Regards,
Josh Wu

>
> Before get_device_and_partition() was introduced this was still working
> because reading the 'fatls' will call fat_register_device(). After it fails
> to detect a valid partition due to the messed up partition table, it will
> then search the header for a FAT signature as a fallback and will continue
> to read it as a PBR. 'fatls' will be able to read it afterward.
>
> Unfortunately, this will now fail on the latest code. I have a patch to do
> the fallback check but under get_partition_info_extended(). However I am
> unsure if there is really a reason for the absence this FAT signature
> checking before failing fatls command.
>
> Do anyone have ideas on a reason for this FAT signature checking not being
> implemented in get_device_and_partition() (or
> get_partition_info_extended())?
>
>
> Thanks,
>
> Darwin
>
>
>
> _______________________________________________
> U-Boot mailing list
> U-Boot at lists.denx.de
> http://lists.denx.de/mailman/listinfo/u-boot

[U-Boot] sdcard with messed up partition table with FAT signature on header

[darwind1]

Josh Wu wrote
> Do you define the CONFIG_DOS_PARTITION in your board config head file 
> (include/configs/xxxx.h)?
> If not defined that macro will cause above error.

Yes. CONFIG_DOS_PARTITION is defined.



--
View this message in context: http://u-boot.10912.n7.nabble.com/sdcard-with-messed-up-partition-table-with-FAT-signature-on-header-tp181462p181469.html
Sent from the U-Boot mailing list archive at Nabble.com.