[Bug tree-optimization/126592] [13/14/15/16/17 Regression] Wrong code with compex mul vectorisation and is_linear_load_p

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D126592

Tamar Christina <tnfchris at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2026-08-03
           Assignee|unassigned at gcc dot gnu.org      |tnfchris at gcc dot=
 gnu.org
             Status|UNCONFIRMED                 |ASSIGNED

--- Comment #1 from Tamar Christina <tnfchris at gcc dot gnu.org> ---
(In reply to ktkachov from comment #0)
> /* PR: candidate tree-vect-slp-patterns.cc:158 - wrong code.
>=20
>    is_linear_load_p decides PERM_EVENEVEN from the parity of each load
>    permutation entry alone, so {0,2,2,2} is accepted exactly like the
>    intended {0,0,2,2}.

{0,0,2,2} is invalid too as the for linearity there can't be gaps.

Testing a patch.=