On Jan 10, 2008, at 8:03 AM, Erik Mouw wrote:
Your current kernel will also have new unknown troubles, plus quite
some known troubles. Not upgrading doesn't mean you have no unknown
troubles, only different unknown troubles.
True. It means you have exchanged some known troubles for unknown
ones. If one is fielding a well-tested system that's a negative due to
the amount of regression testing needed to get back to a similar level
of confidence.
Before I'll upgrade to new version I'd like to try finding the
reason of this behavior.
Tracing already fixed bugs is a waste of time. The best way to avoid
that waste is to upgrade your kernel and see if that fixes the
problem.
True, but spending time to try the new kernel only to find that it
doesn't help is also a waste of time. It can be a large waste if
there's a significant porting effort (like for a custom board).
To avoid both of those forms of waste the original poster (I think)
was asking if this issue was familiar to anyone. Something like, "Yes,
I saw that behavior prior to 2.6.23 but not since" would make it much
more likely to help and makes committing the resources needed to "try
it" easier to support.
Doing so when no one knows recognizes the problem and no release notes
mention it is moving in the direction of wishful thinking.
-Mike
--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
Please read the FAQ at http://kernelnewbies.org/FAQ
