Re: Why NetBSD uses monolithic kernel?

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

> In all references that I have read, monolithic kernels are known for
> being cumbersome, difficult to maintain, difficult to debug, coding
> in kernel is very challenging, a faulty part of kernel can bring down
> the whole system and so on ... .

Yes, if you focus on just the downsides.

> On the other hand, hybrid kernels and micro-kernels have solved these
> problems.  I wonder why NetBSD uses monolithic kernel?

My guess would be that the major reason is history.  NetBSD can trace
its genesis all the way back to U C Berkeley, as I understand it, to a
codebase that existed before microkernels.  A minor reason is probably
all the people who already have their heads around the existing
paradigm.

I'd also point out that microkernels have their own issues, such as a
broken part just locking the system up instead of crashing it; I don't
know about you, but I'd prefer that my system crash and reboot rather
than just wedge with no filesystem, or no pager, or no network stack,
or whatever.

There are plenty of good microkernel-based systems out there.  If you
want one, I suggest you look at them rather than expecting NetBSD to be
something it's not.

> View this message in context: http://netbsd.2816.n7.nabble.com/Why-NetBSD-uses-monolithic-kernel-tp332072.html

Um, no.  I subscribe to a mailing list.  I have no interest in
subjecting myself to Web crap.  If you can't be bothered to quote
enough context for your message to make sense, I certainly can't be
bothered to patch up the resulting damage, especially if it involves
the Web.

> Sent from the tech-misc mailing list archive at Nabble.com.

Please don't include such ads in list mail.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B