Takahiro Kambe <taca%back-street.net@localhost> writes: > In message <smu8tdmmhjv.fsf%linuxpal.mit.edu@localhost> > on Thu, 28 Dec 2017 20:47:16 -0500, > Greg Troxel <gdt%lexort.com@localhost> wrote: >> >> Greg Troxel <gdt%lexort.com@localhost> writes: >> >>> So, with no other comments: >>> >>> I propose to adjust glibmm to have GCC_REQD as 4.9 instead of 4.8. >>> >>> on/after 1700Z on 12/28. >>> >>> This is icky, because it does mean programs that build with 4.8 may >>> link against glibmm which built with 4.9. >> >> And, atkmm does not build against glibmm. So I object to my own idea ;-) >> This seems more complicated than I thought... > How broken? I got a screen full of errors instead of a package :-) > Adding "GCC_REQD+= 4.9" to glibmm and atkmm, I succeeded build it on > NetBSD 6_STABLE/amd64 and NetBSD 7_STABLE/amd64. I think it's probably true that if we build everything with 4.9 it would be ok. But that's veering from a point change to a package to sort of implementing the compiler seletion logic in a hacky way, and we decided early on for that discussion that we weren't going to sprinkle GCC_REQD over vast numbers of packages when the basic issue is that a new compiler is needed in general. My conclusion, more or less, is that 4.8 is now too old for general use.
Attachment:
signature.asc
Description: PGP signature