Re: Severe regression in autoconf 2.61

Bruce Momjian <bruce(at)momjian(dot)us> writes:
> Have you see these lines lower in configure.in?

> if test $ac_cv_func_fseeko = yes; then
> AC_SYS_LARGEFILE
> fi

> Is this broken too?

Yeah, I thought so at first, but looking closer I think it's not too
relevant to the problem. This is for testing whether a couple of *other*
macros need to be defined, it's not about _LARGEFILE_SOURCE.

> It just seemed more straight-forward when the defined HAVE_FSEEKO based
> on ac_cv_func_fseeko rather than ac_cv_sys_largefile_source.

Well, I think 2.61's treatment of fseeko is simply broken. I'm tempted
to propose fixing this by defining PGAC_FUNC_SEEKO the same way 2.59
defined AC_FUNC_SEEKO, and then we wouldn't need the changes you've been
making.

But someone ought to kick this upstream and ask what they were thinking.

regards, tom lane