Paolo Bonzini <bonzini <at> gnu.org> writes: > > I'm also basing the idea that newer macros should unilaterally use newer > > semantics from have this comment on AC_CHECK_TYPES: > > > > # TYPES is an m4 list. There are no ambiguities here, we mean the newer > > # AC_CHECK_TYPE. > > It's a bit different. The old AC_CHECK_TYPE needed two arguments, > AC_CHECK_TYPES_ONCE cannot provide the second.
Yes, I agree that AC_CHECK_TYPES_ONCE is a bit different than AC_CHECK_HEADERS_ONCE for what semantics they can expose. However, I don't see any counterargument in your mail why I should not apply the patch, and yours was the only reply in 24 hours, so I'm assuming that everyone is okay with the overall idea (not to mention the faster ./configure runs). As such, I pushed the patch. -- Eric Blake
