On Wed, Nov 22, 2017 at 10:28:49PM +0100, Marius Bakke wrote:
> Hello!
> 
> I discovered that 'icu4c' failed to build for x86_64 on 'core-updates'.
> After some investigation, it turns out to be a problem with <math.h> in
> C++ mode, due to its usage of C-only builtins (in the 2.26 release).
> 
> Here are the relevant bug reports I've found so far by digging through
> the "release/2.26/master" branch, aka "2.26 stable"[0]:
> 
> <https://sourceware.org/bugzilla/show_bug.cgi?id=21930>
> <https://sourceware.org/bugzilla/show_bug.cgi?id=22235>
> <https://sourceware.org/bugzilla/show_bug.cgi?id=22146>
> <https://sourceware.org/bugzilla/show_bug.cgi?id=22296>
> 
> The attached patch includes the fixes from those bugs, as well as a
> couple of others that looked important.  However it's still a very small
> subset of the 2.26 post-release fixes.
> 
> I've read through _most_ of the commits and around half of them look
> important enough to pick "unconditionally".  The other half I mainly
> lack the context or skills to assess.
> 
> So I wonder if we should simply pick everything from this branch,
> instead of only the few that fixes immediately visible problems.
> Thoughts?

Based on this discussion [0], I think we should take the whole branch.
It sounds like commits on the release branches are considered important
bug fixes and "stable".

There was talk of a mid-October 2.26.1 release, but that didn't happen,
as we know.

Are you able to prepare a patch, Marius? If not, I can do it later
tonight.

[0]
https://sourceware.org/ml/libc-alpha/2017-09/msg01134.html