[Bug middle-end/121661] [13/14/15/16 Regression] miscompilation involving complex and nested functions since r13-1762-gf9d4c3b45c5ed5

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121661

--- Comment #23 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #22)
> (In reply to Richard Biener from comment #21)
> > Andrew, you're still on it?  IMO all force_gimple_* should never resolve
> > DECL_VALUE_EXPRs, possibly achievable with a new gimplify_ctx flag.
> > Getting rid of force_gimple_* is of course best!  Using it for "memory"
> > expressions is questionable anyway.
> 
> This has been high on my come back to and finish up. I will be working on it
> either later today or tomorrow.

I didn't get to it today (as I was finishing up a few other patches) but it
will be tomorrow.