re: oddness with libdrm shlib versions

[matthew green <mrg%eterna.com.au@localhost>]

Martin Husemann writes:
> [Updating from 5 to 6 and then 7 ends up with libdrm.so from 6]
> 
> The definitive answer should be in the set lists, see below for various 
> branches:
> 
> src-5/distrib/sets/lists/xbase/shl.mi:./usr/X11R7/lib/libdrm.so.2.4
> src-6-0/distrib/sets/lists/xbase/shl.mi:./usr/X11R7/lib/libdrm.so.3.2
> src-6-1/distrib/sets/lists/xbase/shl.mi:./usr/X11R7/lib/libdrm.so.3.2
> src-6/distrib/sets/lists/xbase/shl.mi:./usr/X11R7/lib/libdrm.so.3.3
> src-7-0/distrib/sets/lists/xbase/shl.mi:./usr/X11R7/lib/libdrm.so.3.2
> src-7-1/distrib/sets/lists/xbase/shl.mi:./usr/X11R7/lib/libdrm.so.3.2
> src-7/distrib/sets/lists/xbase/shl.mi:./usr/X11R7/lib/libdrm.so.3.2
> src/distrib/sets/lists/xbase/shl.mi:./usr/X11R7/lib/libdrm.so.3.4
> 
> So: yes, something looks wrong in -6. It comes from:
> 
>         Update libdrm to 2.4.36.
>         Enable radeon KMS support in xf86-video-ati so that a netbsd-6
>         userland will work with a more recent (netbsd-7 and up) kernel.
>         [mrg, ticket #1326]
> 
> Matt, erroneous bump in that ticket, or some pullup missing for -7?
> We may have to teach postinstall about it, but that would be a pain.

i'm not sure.  i guess missing pullup, but i'd have to check the
whole thing to be sure.


.mrg.