[Bug tree-optimization/83238] [8 Regression] [graphite] ICE in graphite_can_represent_scev, at graphite-scop-detection.c:971

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83238

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
OTOH we do have a side-entry into this "SESE" region, the latch edge of loop B.

Thus this isn't really a SESE region(?!).