> Programs that use the kvm library and such now should be using procfs
> when appropiate.  It is *beyond* silly to have things like kernfs, procfs,
> and FIFOs optional.  It is neither elegant nor impressive when a slight
> kernel change can break ps(1).

This is true, but "what is appropriate"?

There's absolutely nothing that you can do to remove the libkvm
interface.  Most things that use it are designed to operate on dead
kernels as well as live ones, and there's no a way a kernel core dump
is going to provide a "procfs" interface.

If you want to provide wrappers around the kvm interface, or
something, so that dead kernel and non-procfs kernels still work,
great.  but don't bother suggesting that the dead-kernel functionality
should be gotten rid of, and as long as you're providing that, there's
no _need_ to use procfs.


> People *should* be able to more or less track -current kernels without
> having to track -current userland.

but they shouldn't have to use procfs.



chris