[Bug target/84521] [8 Regression] aarch64: Frame-pointer corruption with __builtin_setjmp/__builtin_longjmp and -fomit-frame-pointer

["ebotcazou at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84521

--- Comment #23 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
> Where exactly is this documented? I can't find no references to
> __builtin_setjmp or __builtin_longjmp in
> https://gcc.gnu.org/onlinedocs/gcc-7.3.0/gcc.pdf or
> https://gcc.gnu.org/onlinedocs/gccint.pdf.

__builtin_setjmp/__builtin_longjmp is not documented at all.  You can probably
find a discussion in the archives or bugzilla about this.

> Which documentation?

The internal manual, section Machine Description.

> Like you say, AArch64 doesn't need anything special, it's the mid-end 
> implementation that is broken here. I bet most of my examples fail on other
> targets too.

Well, the SJLJ exception handling scheme is piggybacked on this support and has
worked for the past 2 decades.  And, until very recently, the Ada compiler was
using another exception handling scheme even more directly piggybacked on it
and which had worked for the past 2 decades too.