Re: [PATCH 1/1] mtd/intel-spi: Support cmdline-based partition

[ron minnich <rminnich@xxxxxxxxx>]

Mika, I have to agree with your comment on parsing the name correctly,
but the pci syntax is sufficiently flexible that I expect it would
create an explosion in code size and complexity and that is a concern.

OTOH, I can think of a few options: we could require, for example,
that PCI names ALWAYS be in the form AAAA:BB:CC.D, and then the test
is a bit simpler: it's a PCI name if id[4] == ':' && id [7] == ':' &&
id[10] == '.', and we can break the id from the parts at id[12]; it it
is not a PCI id, we do a strchr as is done now.

That would mean I could just dump this admittedly ugly change. I admit
the test above is a bit nasty, but it's pretty reliable. WDYT?

Miquel, your patience with this patch as I learn how to contribute is
greatly appreciated. Thanks for the tips. My last real contribution to
Linux was over 15 years ago, with 9p, and the real upstreaming work on
that was done by others; the kernel community is all new to me.

I'm going to drop this patch and start over and try not to make such a
mess of it.

thanks again everyone

ron

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/