RFC: requiring GCC >= 4.7 to build glibc

[Joseph Myers]

On Thu, 15 Oct 2015, Florian Weimer wrote:

> On 10/14/2015 06:33 PM, Joseph Myers wrote:
> > Here is a 14-patch series for requiring GCC 4.7 and making consequent
> > cleanups.  I have't written ChangeLog entries or other write-ups for
> > individual patches, or tested the individual patches on their own, or
> > done non-x86_64 testing where applicable, or done comparisons of
> > installed shared libraries to make sure there aren't unexpected
> > changes, but it should be illustrative of what requiring GCC 4.7 would
> > enable (which I think is more than most such version increments).
> 
> I read through the patches, and they look okay to me.

Thanks.  The question at this point is more whether people think these 
changes provide sufficient justification for the version requirement, 
taking into account the concerns expressed by Mike and David.  (See 
Carlos's definition at <https://sourceware.org/glibc/wiki/Consensus>, 
"Consensus need not imply unanimity.".)  The patches would be posted 
individually for review (except where they fall within areas I maintain - 
patches 3, 4, 6, 7, 8, 9, 13) if consensus on the version requirement is 
reached.

(I've realised that patch 1 is missing a NEWS update.)

-- 
Joseph S. Myers
joseph@codesourcery.com