Peter Maydell <peter.mayd...@linaro.org> writes: > On Tue, 24 Aug 2021 at 09:14, Markus Armbruster <arm...@redhat.com> wrote: >> In functions with an Error **errp parameter, use of &error_fatal is >> almost always wrong. > > What are the cases where it is not wrong?
I can't think of a use that isn't wrong. Doesn't mean no such use could exist. Most rules have exceptions... > My guess is "in board > code and other places where the error handling would have been > 'print a message and call exit()' anyway". When you know that all callers handle errors like &error_fatal does, use of &error_fatal doesn't produce wrong behavior. It's still kind of wrong, because relying on such a non-local argument without a genuine need is.
