Re: [PATCH 2/2] inpt_queue changed from CIRCLEQ to TAILQ in OpenBSD 5.6

On Fri, 4 May 2018 17:01:05 +0100 Stuart wrote:
SH> On 2018/05/04 10:37, Robert Story wrote:
SH> Personally I think it's a disservice to users to enable them to
SH> run with such an old version of OpenBSD - there's a 2-release
SH> cutoff for important fixes in the base OS and 6 months at the
SH> most for security fixes in ports - we [OpenBSD] don't have the
SH> ecosystem where an OS vendor provides many years of support and
SH> backports for an older major version as is the case with some
SH> Linux distros.

I understand that. But some folks are stuck with legacy systems
that they can't upgrade. That's up to them, and we don't want to
abandon them.

SH> But I'd very much like to cut down the amount of patching we
SH> have to do in the port, so if you really need that I'll take a
SH> look :-)
SH> 
SH> I'm not good enough at autoconf wrangling to come up with a
SH> feature test to figure out whether it's using CIRCLEQ or TAILQ.
SH> If it's necessary to support this then the simplest way to cope
SH> is probably to condition on OpenBSD >= 201411 (defined in
SH> sys/param.h), would that be acceptable?

I suppose it's a good start.

SH> (The current mechanism used for version selection in net-snmp
SH> doesn't fit OpenBSD's release numbering; there are no "major
SH> releases", a change in the first digit of a release indicates
SH> only the passing of time, but that doesn't seem like something
SH> good to poke at during your rc cycle).

Yep. I think we'd be willing to consider patches for that for the
master branch (i.e. 5.9) after 5.8 is out and the V5-8-patches
branch is created.

Robert

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot