Mats Erik Andersson wrote:
I have made two observations concerning
1) Romcc is generated in duplicate in fallback/ as well as normal/.
Is this necessary? Both executable have identical md5-hashes.
I do at least modify by hand the target 'normal/clean' to leave
romcc for the next round of compilations. I would have expected
the python generator to produce such alternatives by itself.
The python tool generates a lot of stuff twice. (The static device tree,
too). Not sure if it's worth fixing. It's not elegant but it does not
cause problems either. If you have a slick idea, or even a patch, let's
talk about it.
2) The dependencies of most targets are so shallow,
that it is almost
inevitable to perform a full clean in order to assemble a new
normal.rom or fallback.rom, even after the smallest of changes.
Is this intentional?
No, it's a bug. Patches and fixes are highly welcome!
coresystems GmbH • Brahmsstr. 16 • D-79104 Freiburg i. Br.
Tel.: +49 761 7668825 • Fax: +49 761 7664613
Email: info(a)coresystems.de • http://www.coresystems.de/
Registergericht: Amtsgericht Freiburg • HRB 7656
Geschäftsführer: Stefan Reinauer • Ust-IdNr.: DE245674866