Re: Must check what?

[Helge Hafting <helge.hafting@aitel.hist.no>]

Vadim Lobanov wrote:
> On Wednesday 04 October 2006 12:43, Andrew Morton wrote:
>   
>> I like assertions personally.  If we had something like:
>>
>> void foo(args)
>> {
>> 	locals;
>>
>> 	assert_irqs_enabled();
>> 	assert_spin_locked(some_lock);
>> 	assert_in_atomic();
>> 	assert_mutex_locked(some_mutex);
>>
>> then we get documentation which is (optionally) checked at runtime - best
>> of both worlds.  Better than doing it in kernel-doc.  Automatically
>> self-updating (otherwise kernels go BUG).
>>     
>
> Uhoh! How much is that going to hurt runtime? :) It actually seems to me like 
> this should be doable by static code analysis tools without terribly much 
> pain (in the relative sense of the term). Or am I wrong on this thought?
>   
Surely, any debugging that hurts will only really be there
if you enable CONFIG_DEBUG_something

The kind of stuff you ask people to turn on when they report
strange crashes.

Helge Hafting