[RFQ] make witness panic an option

[Attilio Rao]

On 11/15/12, Adrian Chadd <adrian at freebsd.org> wrote:
> On 15 November 2012 05:27, Giovanni Trematerra
> <giovanni.trematerra at gmail.com> wrote:
>
>> I really do think that is a very bad idea.
>> When a locking assertion fails you have just to stop your mind and
>> think what's wrong,
>> no way to postpone on this.
>
> Not all witness panics are actually fatal. For a developer who is
> sufficiently cluey in their area, they are quite likely able to just
> stare at the code paths for a while to figure out why the
> incorrectness occured.

The problem is that such mechanism can be abused, just like the
BLESSING one and that's why this is disabled by default.

I believe having a mechanism to use printf for witness is not a good idea.

> As I said, I do this primarily so I can sprinkle lots of lock
> owned/unowned assertions around my driver(s) and then use that to
> catch when things aren't being correct. Having to reboot upon _every_
> lock assertion quickly got old.

You can use it as a local patch then. This is not really the usual way
to develop locking policies.

Attilio


-- 
Peace can only be achieved by understanding - A. Einstein