On 1/31/26 12:14 AM, David Lechner wrote:
On 1/28/26 9:53 PM, Marek Vasut wrote:
Unless specified otherwise using the mkimage -B n option, align
DTs in fitImage external data to 8 Bytes, and retain alignment
of everything else to 4 Bytes. This should fulfill the DTspec
requirement, that DTs must be placed at 8 Byte aligned addresses,
even for DTs that are part of fitImage with external data. For
fitImage with embedded data, there is nothing we can do, as the
embedded data are aligned to 4 Bytes, just like any other DT
property.
Could we fail with an error that tells the user they must use the
-E option for dtbs? Or at least emitting a warning would be better
than nothing.
Or make -E the default outright and drop support for embedded data ?
Not for this patch but for broader discussion, it sounds like a
`/align/ 8` could be useful in the devicetree syntax similar to
how we have `/bits/ 8|16|32|64` already. This could (eventually)
solve the "nothing we can do for embedded data" problem.
This would have to be DTC syntax sugar. How would that be translated
into the DTB encoding ? In any case, this directive is something to
discuss with the DT compiler people.
