Hi all, Kostik,
I was just reading through r227697, and it seems to me that this
commit changes KBI. Shouldn't this have a __FreeBSD_version bump?
My concern is with external file system modules like fuse4bsd
(although on first sight it doesn't seem that fuse4bsd is affected).
http://lists.freebsd.org/pipermail/svn-src-all/2011-November/044647.html
--
Robert Millan
_______________________________________________
freebsd- mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-fs
To unsubscribe, send any mail to "freebsd-fs-"
)
> This is r229703 for stable/9.
> The __FreeBSD_version was already bumped after that change, at least 3 times.
> In fact, the bump to 900501 in r229723 is almost immediately after it.
>
> For HEAD, it makes no sense to bump the revision at all, since you are
> assumed to always run latest HEAD.
Really? This puzzles me. You're not really supposed to make KBI changes
in stable- otherwise stable isn't a stable place for vendor KOBJs.
When you make KBI changes, independent of whether head or not, a
revision bump is useful so that dependent code knows when it has to
change. This change carries forward and backwards for compilation
independent of which branch this is in.
_______________________________________________
freebsd- mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-fs
To unsubscribe, send any mail to "freebsd-fs-"
)
