cvs commit: src/sys/alpha/alpha support.s src/sys/i386/i386 swtch.s src/sys/kern kern_shutdown.c src/sys/sys systm.h

[Marcel Moolenaar]

On Fri, Feb 20, 2004 at 07:32:09PM +1100, Bruce Evans wrote:
> On Wed, 18 Feb 2004, John Baldwin wrote:
> 
> > On Wednesday 18 February 2004 12:40 pm, Marcel Moolenaar wrote:
> > > I like the behaviour of MPASS(), just not its name and the fact it's
> > > defined in lock.h. If we can call it ASSERT and move its definition
> > > to systm.h, then I'm hooked.
> 
> MPASS() has the same problems as line numbers in panic messages (except
> for the implementation bugs (misformatting of the output etc.)): verbose
> messages and breakage of simple regression tests.

Note that however you look at it, failed assertions and panics are
expected to not happen. I hope at least that we're not designing
the kernel based on the frequent occurrence of these. In case they
do happen, I expect that they help in identifying the root cause.

If a failed assertion or a panic breaks a simple regression test, it
means there's a bug and that it needs to be found and fixed. You're
not supposed to complain that the message printed by the assert or
panic logic messes up the testing.

So, verbosity is of lesser concern than having concise, to the point
and informative messages that stand out enough to be noticed and
actually help developers to start triaging the problem and get to
the bottom of it.

-- 
 Marcel Moolenaar	  USPA: A-39004		 marcel at xcllnt.net