Re: setjmp(3) clarification

[David Holland <dholland-tech%netbsd.org@localhost>]

On Sun, Oct 25, 2026 at 11:54:00AM -0400, Greg Troxel wrote:
 > > I would leave the text as-is.
 > 
 > https://pubs.opengroup.org/onlinepubs/9799919799/functions/setjmp.html
 > 
 > POSIX says, with a note that really ISO C defines it:
 > 
 >   A call to setjmp() shall save the calling environment in its env
 >   argument for later use by longjmp().
 > 
 > I think the man page is fine as is, and that we should not wordsmith
 > text from ISO C and POSIX.   The explanatory text about setjmp/longjmp
 > seems reasonably clear, given that this is a fairly advanced programming
 > concept.   So I see nothing that needs fixing in NetBSD.
 > 
 > If someone thinks the ISO C text should change, <humor>I'm sure they
 > have a public bug-tracker someplace</>.

I don't agree; standards text is routinely confusing or misleading to
those not steeped in the assumptions and context of the standard
involved.

The man page already discusses almost all the things necessary to
explain in full, just not in the right order for someone who doesn't
already know exactly what the words mean. My inclination would be to
go rewrite the whole page rather than trying to patch it.

(although if someone really wanted to patch it, I think "execution
environment" might work better than the things already proposed)

-- 
David A. Holland
dholland%netbsd.org@localhost