cfi_cmdset_0002.c: do_xxlock_oneblock: overflow subtract causing oopsen on iPAQ h1900

[Joshua Wise]

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

> Yep, it's an incredibly ugly thing - I can see why it blows things out
> of the water.  I knew it was a bad thing to do - hence the comments.  At
> the time that it was added, there wasn't an unlock function and so
> adding, even a broken one, was a starting place.
Ok, I remember in the past that one of the flash chips that I was using did 
not have an unlock function. It was probably the one in the 1900.

> In the end, please fix up the code so that it works for the general
> case.  I'll look at things later and see what I can do to either black
> list the SST parts or deal with them in a sane way that isn't dangerous
> and ugly.
Hrm ... you make it sound like there are different unlock codes for every 
chip! Is there a general case here?

As to dealing with the SST parts, I believe that if there is not another way 
to detect the parts (ie, doing something htat would cause one to reset, but 
not another), you should probably make the user specify what type of part it 
is ... dangerous hacks like having to read/write 4MiB below the address space 
just to unlock does not seem like a good answer for all chips of the sort!

> SST, PLEASE FIX YOUR CHIPS AND USE UNIQUE IDS FOR NEW PARTS!!!
>
> Yes, I do mean to yell.
You have caps lock stuck... oh wait, you realized that ;)

joshua

- -- 
Joshua Wise | www.joshuawise.com
GPG Key     | 0xEA80E0B3
Quote       | <RockShox> charge up a 50v 1000uf cap and toss it to someone
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)

iD8DBQFAKvo4Pn9tWOqA4LMRAgAqAKCmmEsCDYpH38ZLcjnukqohExbKqgCgrP+s
/YIrhwTsNxFsO25fTd2/3A4=
=adYn
-----END PGP SIGNATURE-----