https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117100
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Depends on| |109934
--- Comment #9 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #7)
> (In reply to Sam James from comment #6)
> > Simplified a bit:
>
> Just some debug:
>
> When we create one of the unswitch_predicate for the second switch, we have:
>
> true_range:
> [irange] int [4, 4][6, 6][8, 8] NONZERO 0xf
> false_range:
> [irange] int [4, +INF] NONZERO 0x7fffffff
>
> That false_range looks wrong because there is no .
>
> It was created by doing:
> false_range = true_range;
> if (!false_range.varying_p ()
> && !false_range.undefined_p ())
> false_range.invert ();
>
> The code looks correct.
This is exactly the same debug as what was done in PR 109934 :).
Referenced Bugs:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109934
[Bug 109934] [14 Regression] Wrong code at -O3 on x86_64-linux-gnu
