On Tue, Jul 24, 2018 at 01:20:58PM -0400, Eric Sunshine wrote:

On Tue, Jul 24, 2018 at 5:26 AM Jeff King [off-list ref] wrote:

quoted

  1. We'll only trigger with -Wimplicit-function-declaration
     (and only stop compilation with -Werror). These are
     generally enabled by DEVELOPER=1. If you _don't_ have
     that set, we'll still catch the problem, but only at
     link-time, with a slightly less useful message:

     If instead we convert this to a reference to an
     undefined variable, that always dies immediately. But
     gcc seems to print the set of errors twice, which
     clutters things up.

The above does a pretty good job of convincing me that this ought to
be implemented via an undefined variable rather than undefined
function, exactly because it is the newcomer or casual contributor who
is most likely to trip over a banned function, and almost certainly
won't have DEVELOPER=1 set. The gcc clutter seems a minor point
against the benefit this provides to that audience.

OK. I was on the fence, but it should be pretty trivial to switch. Let
me see if I can just make a replacement for patch 1, or if the whole
thing needs to be rebased on top.

-Peff