Re: Patch: add prefix to condition args in opcodes

[Jeffrey A Law <law@cygnus.com>]

  In message <14239.3863.98454.352088@gargle.gargle.HOWL>you write:
  > Sorry.  I was thinking they were close enough, but you're right.  They are
  > independent.
It happens.  Often I can extract the pieces, but my time is very limited right
now, so I have to ask others to extract the independent pieces.


  > I think we should also do the completers and float regs once you're happy
  > with this change.
I think so too.  Hopefully that'll give us enough letters to finish the PA2.0
work and maybe, just maybe clean up the code a little in the process.

One of the big issues at hand was whether or not to completely revamp the
entire PA assembler when we added PA2.0 support.  I wanted to do that, but
simply didn't have the time.  So, we're stuck trying to make the current
mess of spaghetti code work and trying to clean up pieces as we go along.

  >  > The flow control (at least to me) is easier to follow with that kind of 
  > style
  >  > (it's still not ideal, but I'm not sure what else we can do to clean thi
  > s up
  >  > without starting over with a complete redesign).
  > 
  > I saw that construction and I had thought what I did was preferable
  > because it avoided a redundant test.  If you think it's better to write
  > it that way, I'll change it.
I prefer code that's easier to read/follow over code that avoids a comparison.
Particularly in this case since assembly time is usually dwarfed by the compile
time.

  > Welcome back.  Glad to see you now have a chance to breathe :-)
Breathe?  No way, I barely get time to blink :-)

jeff