[PATCH 3/4] fix-nand-partitions.patch

[Andy Green]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Somebody in the thread at some point said:

|> -	return -1;
|> +	return -EBADMSG;
|
| This looks like a regression. The expected return value is -1,
| see the call to ecc.correct in nand_read_subpage in nand_base.c

These patches came out of a diff from master to andy-tracking, so they
represent our current status affecting these files.  If it's busted,
it's busted in our normal stuff.

I intend this patchset to go after master and before
mokopatches-tracking in our tree, in a new branch called "proposed", so
we too will rebase ourselves around what we're offering upstream.  If
this extra layer doesn't cause the branch system to blow a gasket during
the rebase, we should end up inheriting fixes from here up the rest of
the branches.

| The rest looks good, but parsing the complete device-independent
| partitions in the driver may draw some objections. We'll see.

We all - you too - had a year and more to like or not the look of this
and fix this if we didn't like it... this has just come straight out of
mokopatches.

If it's not good enough we'll fix it / move it around until it is, since
we are in "async cap-in-hand mode" with this to get it upstream.

- -Andy
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmeiCkACgkQOjLpvpq7dMqEmQCfQcp3LuR5i932OIUqyMWn5IgE
q6kAnRAJfnURGbZyApx9jF3sJGgb4zik
=gQl4
-----END PGP SIGNATURE-----