Re: [RFA] H8/300sx port for gas, sim, and opcodes (1/5)

[Michael Snyder <msnyder at redhat dot com>]

Nick Clifton wrote:
> 
> Hi Michael,
> 
> > This is the first of several posts containing a submission of a port
> > extending the h8/300 tools to cover the new h8300sx architecture.
> 
> > The files are so heavily changed that I'm just going to submit them
> > inline, rather than as diffs.
> 
> In the future please do not do this.  Please submit the diffs
> instead.  You can zip/compress/gzip/bzip2 them if necessary, but in
> order to check these patches I had to save out each file, run diff by
> hand and then check the output to see what was happening.
> 
> Also - was it really necessary to make such sweeping changes, just in
> order to make a new architecture variant ? 

Yep -- the h8sx is pretty radically different from its predecessors.
For instance, the opcode table goes from about 350 entries to 8,400.
It adds about 14 new addressing modes to each instruction.

> I realise that you
> probably were not the one who made all of these changes, but they do
> seem quite extensive and an explanation of why they were necessary
> would have been nice.

Sorry.  Big change in architecture.

> > 2003-05-14  Michael Snyder  <msnyder@redhat.com>
> >       and Bernd Schmidt   <bernds@redhat.com>
> >       and Alexandre Oliva <aoliva@redhat.com>
> >       * opcode/h8300.h: Add h8300sx insns, addressing modes, etc.
> 
> Approved - please apply.
> 
> Note - please also add an entry in the gas/NEWS file mentioning the
> support for this new architecture.

OK.

> Will there be a further contribution with an extension to the gas
> testsuite for cover the new instructions and addressing modes
> supported by the h8300sx ?

Yes.  I thought I'd contribute the architecture changes first, 
but I do have a rather large set of gas tests to contribute too.

Michael