Re: kern/60810: /usr/src/sys/conf/osrelease.h -s behavior does not match description

[Rob Whitlock <rwhitlock22%gmail.com@localhost>]

> On Sep 29, 2026, at 2:10 PM, Robert Elz <kre%munnari.OZ.AU@localhost> wrote:
> 
>    Date:        Tue, 29 Sep 2026 01:29:56 -0400
>    From:        Rob Whitlock <rwhitlock22%gmail.com@localhost>
>    Message-ID:  <AB6D15AA-921B-4E1B-9F62-5B54693C95B0%gmail.com@localhost>
> 
>  | but if there were a script that already
>  | did that parsing for you, then why wouldn't you use it?
> 
> But it doesn't, you only believe it does as you've apparently
> only tested in on HEAD.   On the -10 branch currently, osrelease.sh -s
> produces 102_STABLE  on the -11 branch it gives 110_STABLE (that one I haven't
> cvs up'd recently, it might have been updated beyond that since I last did).
> 
> On a 10.2 release version it says 102  (I don't have 11.0 release kernel
> sources handy to check it).   On the final -9 branch release it produces 95

Yes, that is my point. If you just look at the comments which describe
the output as MMmmpp, then you can easily get the wrong idea.

>  | (This is, of course, assuming that this person wouldn't be aware of
>  | corner case differences between the two formats)
> 
> It is more than corner case differences.   The script has a purpose to
> exist, and making version numbers for pkgsrc is not it.
> 
> You'd be better starting with uname -r from the system you want to
> build for (the same as pkgsrc does for the native os version) and
> convert that (which is trivial) to the correct format.

Sure, but when cross-building you can't run uname -r on the target
machine. Running uname -r on the build machine would get you the
OS information for the build machine, not the target machine. This
is why CROSS_OPSYS_VERSION must be derived from the sources that
you have for the target machine. Either that or run the target
machine under qemu or something, but that's a whole other can of
worms and way too heavyweight to just get the output of uname -r.

>  | Very well, I can drop the request to change it. However I still
>  | think it would be a good idea to more precisely document the
>  | given format in the comments,
> 
> I can do that I suppose, including a comment telling people who want
> to use it for purposes other than that for which it is designed, to
> go away and leave that script in peace.

Excellent, thank you!

> kre
>