Hi Harald,
you right, that solution looks pretty obvious to me. Ok by me. Thanks for the
patch and the opportunity to give back a review.
Keep up the good work!
Regards,
Andre
On Thu, 18 Jul 2024 21:27:13 +0200
Harald Anlauf <anl...@gmx.de> wrote:
> Dear all,
>
> here's a quite obvious fix for an ICE when processing an array constructor
> where the first element is of deferred length, and at least four constant
> elements followed, or an iterator with at least four elements. There
> is a code path that then tries to combine these constant elements and
> take the element size of the first (variable length) element in the
> constructor.
>
> (For gcc with checking=release, no ICE occured; wrong code was generated
> instead.)
>
> Obvious fix: if we see that the element size is not constant, falls back
> to the case handling the constructor element-wise.
>
> Regtested on x86_64-pc-linux-gnu. OK for mainline / backports?
>
> Thanks,
> Harald
>
--
Andre Vehreschild * Email: vehre ad gmx dot de
