Re: EP9312, Maverick Crunch fixes

[Nick Clifton <nickc at redhat dot com>]

Hi Petko,

> I am not sure who i need to contact so i'm sending the patch to both
> lists.

Both lists are used by the binutils project, so sending to either one
is OK.

> We've fixed some bugs in Maverick Crunch coprocessor opcode
> generation in gas.  These fixes now allow code compiled for ep9312
> (silicon revision D0 and D1, i.e. the latest as of now) to run
> properly on the target.  The patch is generated against CVS code
> from 2004-02-05.

Thanks for submitting this patch.  Is there a document describing the
D0 and D1 revisions ?  (I only have access to revision 1.0 of the
Maverick ISA description and I would like to know if there is a newer
version available).

> Please let me know if the patch doesn't comply with the binutils
> coding standards or something needs to be redone.

Actually the patch is fine although it was missing a couple of things:

  - a set of ChangeLog entries describing the fixes that have been
    applied.

  - a patch to the gas testsuite to update the Maverick instruction
    tests.
        
It is also worth noting that although is covered by the obvious-change
rule and does not need a copyright assignment, if it had been any
bigger it would have needed one.

> Sorry about the group email, but i don't know who's the official
> maintainer(s).

The maintainers are listed in the file binutils/MAINTAINERS.  Since
the Maverick is an ARM coprocessor you could have addressed this email
to either myself or Richard Earnshaw, but in any case, sending to the
list is the correct thing to do, so that everybody is kept informed.

If you can submit a revised patch with a set of ChangeLog entries and
a fix to the gas testsuite, I will be happy to apply it.

Cheers
        Nick