RFA: tuples merge for GIMPLE_MODIFY_STMT

[Aldy Hernandez]

> 1- Why the need to use the PROTECTED_* accessors?  Is it really that 
> slow to do the checking GIMPLE_STMT_P in the unprotected accessors?

PROTECTED_* accessors are for accessing operands of *either* a
common tree (i.e. MODIFY_EXPR, CALL_EXPR, etc) or a GIMPLE_MODIFY_STMT.
That is, if we can't determine at compile time if we have a tuple or
not, then we use PROTECTED_* accessors.

The reason I don't have TREE_OPERAND behave like PROTECTED_* currently
does is two-fold:
	- speed (questionable as you have pointed out)
	- I actually like the idea of ICEing if you try to access a tuple
	  as a tree.  For one, it makes it a hell of a lot easier to
	  spot unconverted trees-- they ICE :).

If you are unconvinced, we could rename all GIMPLE_MODIFY_OPERAND's
back to TREE_OPERAND's and have TREE_OPERAND be all knowing.

Let me know.

> 2- What happened to the idea of increasing tree codes to more bits?  Do 
> we have any left over space in tree.base?

Dunno, I didn't even know we were trying to do that.  What should I do?
Where can I start?  How can I help?

Aldy