re: Some pkgsrc observations

matthew green writes:
> > > Also, does anyone know why net/bind918 fails on VAX when it compiles fine 
> > > on other CPUs?
>
> bind918 might take a while more..

OK, this is because of using older jemalloc that doesn't have
malloc_size(3), which bind detects and falls back to its own,
which is missing the const.

i would patch the bind918 jemalloc_shim.h to add the missing
'const'.  it thinks we don't have jemalloc because of the lack
of jemalloc/jemalloc.h (same for both.)

you can reproduce this on faster platforms if you build a world
with HAVE_JEMALLOC=100.


.mrg.