Heikki Linnakangas <hlinnaka@iki.fi> writes:
> On 03/17/2017 07:57 PM, Tom Lane wrote:
>> This seems to have broken narwhal:
> It's not very nice to change the requirements in a minor version, but I
> don't think this would be a real problem for anyone. Not many people
> build PostgreSQL using MinGW, let alone with an ancient version of it.
> But if people don't agree, we could instead revert this patch and apply
> the smaller V2 patch [2] instead, in the back-branches.
> Thoughts? Any objections to requiring a newer version of MinGW? Any
> objections to do so in the next minor release?
Hm. I'm +1 for doing that in HEAD, but less so for the back branches.
Can we get some fix on when the functions in question were added
to MinGW? If we knew how new a toolchain we'd be requiring here,
that would help make this decision.
regards, tom lane