>   imho, that depends on whether or not the grand design of being able to
>   run "foo -M netbsd.0.core -N netbsd.0" on dead kernels has now been
>   completely abandoned.
>   
>   say it ain't so?
>
>why does it depend?

it just seemed to me that everything is being converted to methods
that use executable code in the kernel to extract the needed values,
rather than just treating the kernel (and memory) as the files they
once were.

i'm concerned that it will "just work" using sysctls and the older
methods will die an unused dusty death.  pcvt was removed from the
kernel last weekend, and while i couldn't come up with any sound
reasons for it not to be (or problems with it being removed), i will
miss it.  it fell into disuse, was unmaintained, and was excised.  if
the kmem grovelling routines are not used, they'll start to be
unmaintained (since the sysctl stuff "works") and i'm afraid they'll
eventually die.

or am i just being pointelessly paranoid?

>	- kvm grovelling of dead kernels will always be supported in
>	  NetBSD as far as i can tell.

good.

>	- permenent removal set-id bits will always be supported in
>	  NetBSD as far as i can tell.

better.

>(of course, quality is always an issue).

yep.

-- 
|-----< "CODE WARRIOR" >-----|
codewarrior@daemon.org             * "ah!  i see you have the internet
twofsonet@graffiti.com (Andrew Brown)                that goes *ping*!"
andrew@crossbar.com       * "information is power -- share the wealth."