Re: Visual Studio 2012 RC

From: Andrew Dunstan <andrew(at)dunslane(dot)net>
To: james(at)mansionfamily(dot)plus(dot)com
Cc: Craig Ringer <craig(at)2ndQuadrant(dot)com>, Noah Misch <noah(at)leadboat(dot)com>, Brar Piening <brar(at)gmx(dot)de>, Heikki Linnakangas <hlinnakangas(at)vmware(dot)com>, Magnus Hagander <magnus(at)hagander(dot)net>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: Visual Studio 2012 RC
Date: 2013-01-28 00:11:12
Message-ID: 5105C220.8070703@dunslane.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


On 01/27/2013 06:51 PM, james wrote:
>> On the contrary, only a few months ago there was a far from
>> groundless fear that Microsoft would do just that. Following
>> considerable outcry they changed their mind. But this is definitely
>> not just paranoia. As for w64 support, the mingw-64 project exists
>> more or less explicitly to produce 64 bit compilers, including those
>> hosted on mingw/msys.
>
>
> Huh. The only reason we have to use mingw64 or one of the assorted
> personal builds is because 'mingw support' doesn't deliver on its own,
> and last I looked there was a confusing variety of personal builds
> with various strengths and weaknesses. I managed to make some progress
> but we seem to be a ways off having a reference download (and ideally
> one with clang too I guess).
>
> I'd very much like there to be a good reference implementation, but
> the whole mingw/mingw64 thing is indicative of some problems, and
> reminds me of egcs.
>
> You have references to back up your statements, and demonstrate that
> it wasn't primarily FUD? FWIW I think the higher entry prices of
> pay-for VStudio almost guarantees continued availability of a free
> compiler, though it might end up slightly crippled, but I'm not a
> product planner for MS any more than you are.

I blogged about it at the time:
<http://people.planetpostgresql.org/andrew/index.php?/archives/276-Microsoft-throws-large-developer-communities-under-the-bus.html>
and here when they relented a month later:
<http://people.planetpostgresql.org/andrew/index.php?/archives/277-Microsoft-apparently-does-the-right-thing.html>.
That was based on sources I considered credible at the time. But
frankly, I'm not going to spend a lot of time digging up old references.
If you think it's FUD then feel free, but I was and am far from alone in
thinking it wasn't. Given the work I've put in over the years in
supporting PostgreSQL on Windows I don't think any suggestion I have a
bias against Microsoft has much credibility.

As for mingw64, I still use this one:
<http://sourceforge.net/projects/mingw-w64/files/Toolchains%20targetting%20Win64/Automated%20Builds/mingw-w64-bin_i686-mingw_20111220.zip/download>.
The reason there isn't a later more official build is that they have
screwed up their automated build process (which I also blogged about ;-)
<http://adpgtech.blogspot.com/2013/01/mingw64-fail.html>). I'm going to
try to get to the bottom of that when I get a chance. But this compiler
works perfectly well AFAICT. It's been running on one of my buildfarm
members quite happily for quite a while.

cheers

andrew

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Geoghegan 2013-01-28 00:16:24 Re: Re: Doc patch making firm recommendation for setting the value of commit_delay
Previous Message james 2013-01-27 23:51:19 Re: Visual Studio 2012 RC