Brendan Jurd <direvus@gmail.com> writes:
> This whole business of passing around global pointers while switching
> memory contexts seems like an optimal breeding-ground for bugs.
Yeah. If it were to get significantly more complicated than this,
the best solution IMO would be to give up on trying to use a temporary
memory context during function inlining, and just accept that whatever
memory we chew up there is going to be leaked for the duration of
planning.
regards, tom lane