svn commit: r221972 - head/sys/geom/part

[Marcel Moolenaar]

On May 22, 2011, at 9:26 AM, Warner Losh wrote:

> 
> On May 22, 2011, at 7:03 AM, Adrian Chadd wrote:
> 
>> This also bit me on embedded platform stuff.
>> 
>> Is it possible to disable this by default for now and have it just warn loudly?
>> And/or hide the default value behind a kernel configuration variable
>> so we can disable it
>> (but still get the warnings) for now?
> 
> Or just delete it entirely as a bad idea?  We had this with Marcel's warning for a long time that turned out to be utterly bogus so we removed it.

Really?

The warning wasn't bogus. It was there to help us understand what was
going on and we found over time that it was a harmless condition.
That's why we removed the warning. The warning was good to raise
awareness.

This is in no way similar to what we have now. Here we have to deal with
broken and fundamentally invalid MBRs that has caused us harm before and
we're trying to do something sensible. As it turns out, a whole bunch of
people have invalid MBRs, probably caused by crappy tools. Now what do
you suggest we should do? Accept it silently and suffer the consequences
later,  should we raise awareness so that administrators can try and fix
things or should we reject the MBR out of pedanticism?

Rather than just calling it a bad idea, why not come up with something
constructive?

-- 
Marcel Moolenaar
marcel at xcllnt.net