Magnus Hagander <magnus(at)hagander(dot)net> writes:
> Here's what I came up with and will apply as soon as my msvc build
> completes. (the mingw one works with this)

This is still going to need manual adjustment in the future, since
probably when mingw adds the #define, they will put it in <mstcpip.h>,
and this code will not see it.  So at some point we'll need to add an
autoconf test for <mstcpip.h> and replace those #ifdef
WIN32_ONLY_COMPILER checks with #ifdef HAVE_MSTCPIP_H.  But I see no
need to bother until such a version of mingw exists, and I'd just as
soon not be making such changes on the day of a wrap.  So this is just
a note for later.

			regards, tom lane

In response to

• Re: [HACKERS] pgsql: Add support for TCP keepalives on Windows, both for backend and at 2010-07-08 16:06:57 from Magnus Hagander

Responses

• Re: [HACKERS] pgsql: Add support for TCP keepalives on Windows, both for backend and at 2010-07-08 16:42:18 from Magnus Hagander

pgsql-hackers by date

Next:	From: Magnus Hagander	Date: 2010-07-08 16:42:18
	Subject: Re: [HACKERS] pgsql: Add support for TCP keepalives on Windows, both for backend and
Previous:	From: Robert Haas	Date: 2010-07-08 16:29:47
	Subject: Re: Reviewfest 2010-06 Plans and Call for Reviewers

pgsql-committers by date

Next:	From: Andrew Dunstan	Date: 2010-07-08 16:40:02
	Subject: Re: pgsql: Make the Windows tcp keepalive support depend on the existance of
Previous:	From: Tom Lane	Date: 2010-07-08 16:30:13
	Subject: pgsql: Add a cross-reference to precedence information to CREATE