WITH_CLEANUP_EXPR oddity

[Richard Henderson]

On Sun, Jun 20, 2004 at 05:26:47PM -0400, Richard Kenner wrote:
> I believe it's the case that if you gimplify something that's already
> been gimplified that nothing is supposed to happen the second time.
>
> But that's not true if the first gimplification makes a WITH_CLEANUP_EXPR.
> Gimplifying that will ICE.

Yep.

I looked into fixing this once, but instead found so much redundant
scanning being done as to be criminal.  Which lead me to rearrange
how the gimplifier was structured, and afterward the problem naturally
went away.

I actually kinda prefer this solution, as it punishes sloppiness.

In what context are you finding yourself re-gimplifying?

> What's supposed to happen here?

Gimplify once, and a surrounding cleanup_point_expr converts them to
try_finally_exprs.


r~