[Bug target/97822] [13/14/15/16 Regression] internal compiler error: in choose_baseaddr

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97822

--- Comment #31 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
> Does this mean that even current master is broken in this regard?

Could you elaborate?  What is broken exactly, apart from the testcase for this
PR?  If you want correct SEH, you cannot use stack realignment since SEH does
not support it.