On 04.11.25 18:18, Philippe Mathieu-Daudé wrote: > On 4/11/25 18:00, Jan Kiszka wrote: >> On 04.11.25 17:27, Philippe Mathieu-Daudé wrote: >>> On 4/11/25 15:45, Jan Kiszka wrote: >>>> From: Jan Kiszka <[email protected]> >>>> >>>> As an eMMC block device image may consist of more than just the user >>>> data partition, provide a helper script that can compose the image from >>>> boot partitions, an RPMB partition and the user data image. The script >>>> also does the required size validation and/or rounding. >>>> >>>> Signed-off-by: Jan Kiszka <[email protected]> >>>> --- > > >>> Still: >>> >>> $ dash scripts/mkemmc.sh -b firmware.img -r /dev/zero:2MB os.img >>> emmc.img >> ^^ >> User error ;-). This must be just "...:2M". > > Per the doc: > > +.. code-block:: console > + > + scripts/mkemmc.sh -b firmware.img -r /dev/zero:2MB os.img emmc.img > > (I'll update it) >
Ouch, indeed. >> >>> Invalid value '2MB' specified for /dev/zero image size. >>> scripts/mkemmc.sh: 170: [: Illegal number: >>> Creating eMMC image >>> Boot partition 1 and 2: 128K each >>> scripts/mkemmc.sh: 192: [: Illegal number: >>> User data: 4 bytes >>> >>> $ bash scripts/mkemmc.sh -b firmware.img -r /dev/zero:2MB os.img >>> emmc.img >>> Invalid value '2MB' specified for /dev/zero image size. >>> scripts/mkemmc.sh: line 170: [: : integer expression expected >>> Creating eMMC image >>> Boot partition 1 and 2: 128K each >>> scripts/mkemmc.sh: line 192: [: : integer expression expected >>> User data: 4 bytes >>> >>> But now the file is generated :) >> >> ...which is actually a bug as well: We are not terminating the script >> inside the process_size function when calling "exit" there. Deviates >> from the behavior under Linux. Sigh. > If you have a diff I can test & squash. Someone rather needs to debug. A first remote experiment couldn't explain why we are not terminating. If you could run with -x, that trace may be a first hint. Jan -- Siemens AG, Foundational Technologies Linux Expert Center
