[Bug rtl-optimization/80357] [7 Regression] ICE in model_update_limit_points_in_group, at haifa-sched.c:1982 on powerpc64le-linux-gnu

[segher at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80357

--- Comment #11 from Segher Boessenkool <segher at gcc dot gnu.org> ---
The pressure becomes quite negative here...  that isn't supposed to
happen as far as I can see.  Allowing all negative pressures in the
asserts makes everything compile fine, but there really shouldn't
be negative pressures at all?