[tree-ssa] Inlining vs gimple vs compound expressions

[Jason Merrill]

On Thu, 27 Feb 2003 15:34:30 -0500, Daniel Berlin <dberlin@dberlin.org> wrote:

> That said, it's not that i like CE's particularly, since right now the
> restrictions on them make me unable to easily do SSAPRE insertion (in
> particular, the requirement that the first argument of a CE not be a CE
> itself)

This needn't be a requirement, except in the interests of avoiding
recursion.  In the gimplifier I frequently replace simple statements with
CEs, which usually means putting one in the first argument of another CE;
this is later tidied up with a call to rationalize_compound_expr.  Would a
strategy like that work for you?

Jason