> [ seebs reports digging out of libc/symlink ]
> [ trouble after a busted install=/ build    ]
>
> Seems to me that perhaps, during an install like this, symlinks should only
> be updated if the new target already exists?
It's been too long since I had my head jammed inside
the *.mk files, so take all this w/ a grain of salt.
Maybe 5 grains.

Do you think the patch in PR 14378 addresses this problem?
That patch tried to fix problems when doing a "make -j."
The idea behind it was to enforce a "links get built only
after the target gets installed" rule, which sounds like
what you're after.

Hm.  And checking the CVS logs of bsd.links.mk,
it looks like the parallel make problem I reported
in PR 14378 has been fixed.  So, somebody should close
that PR, regardless.

> -s
Chris <jepeway@blasted-heath.com>.