[Bug target/107920] [13 Regression] ICE in execute_todo, at passes.cc:2140

["rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107920

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1

--- Comment #9 from Richard Biener <rguenth at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #3)
> (In reply to Andrew Pinski from comment #2)
> > It is specifically svld1rq_s8 folding into:
> >   # VUSE <.MEM>
> >   _5 = MEM <vector(16) signed char> [(signed char * {ref-all})x_3(D)];
> > 
> > 
> > Which is done by svld1rq_impl::fold (aarch64-sve-builtins-base.cc:1214).
> 
> Which was added by r13-1055-g494bec025002df .
> 
> The question becomes is gimple folding allowed to add MEM references here or
> not?

Yes, but the folding needs to properly handle virtual operands.  Instead it
does

-  # VUSE <.MEM_2(D)>
-  _4 = svld1rq_s8 (_1, x_3(D));
+  # VUSE <.MEM>
+  _5 = MEM <vector(16) signed char> [(signed char * {ref-all})x_3(D)];
+  _4 = VEC_PERM_EXPR <_5, _5, { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13,
14, 15, ... }>;

see how gimple-fold.cc handles them.