On Thursday 03 January 2008 13:10, Carlos E. R. wrote:
> If the joystick driver is part of the kernel, then the culprit remains
> being the kernel.
Yes, I disagree.
Pluggable modules extent the kernel but are not THE kernel.
(1) kernel runs without module
(2) kernel panics with module plugged in
(3) kernel runs when module unplugged
The culprit is the module... not the kernel.
And all the rhetoric and flapdoodle about pluggable kernel extensions
is just
that. Yes, the module provides services (kernel services) through a standard
kernel interface, and yes, typically only kernel developers write modules,
and yes, the kernel "team" determines which modules get "into" the kernel...
but all of that flap is semantics when it comes time to debug a problem.
Which module is it?
If I come along and write a module replacement (unsanctioned) for
what-ever-device (tainting the kernel) and the silly thing runs fine its my
work as a module writer and nobody knows or cares, but if the silly thing
causes the kernel to panic do you think anybody in kernel land wants a bug
report for it?? No way. There would be nothing wrong with THE kernel except
for the simple fact that MY module caused it to panic. On the other hand if a
sanctioned module is causing the kernel to panic, then I think it would be
helpful to be able to isolate the failing module BEFORE giving a report to
the kernel team.
--
Kind regards,
M Harris <><
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
