Re: [PATCH v2 5/6] mtd: partitions: pass around 'mtd_partitions' wrapper struct

[Brian Norris <computersforpeace@gmail.com>]

Hi Boris,

On Fri, Dec 04, 2015 at 05:45:11PM -0800, Brian Norris wrote:
> On Sat, Dec 05, 2015 at 01:30:49AM +0100, Boris Brezillon wrote:
> > How about defining a new function to encourage mtd drivers to pass an
> > mtd_partitions structure instead of the parts + nr_parts arguments.

...

> > Hm, you updated the ->parse_fn() prototype to take a const mtd_partition **
> > in patch 2, and it seemed pretty easy.
> > How about updating it again to take an mtd_partitions pointer here?

OK, so I've hacked around at these two suggestions, and I'm not very
happy with the results so far. I mentioned some of my objection to the
first above already, and when I tried to do some kind of compromise, it
doesn't end up much cleaner IMO.

For the second suggestion, I don't really like the circular dependency,
where we have struct mtd_part_parser include pointers to struct
mtd_partitions which includes pointers to mtd_part_parser. This is
partly because of the re-use of the parts of the same struct as both an
input and an output, depending on the context. I don't think that's very
clean.

So, if there are good ways to extend some version of your suggestions,
perhaps they can be tackled after this patch set?

Brian