Re: [PATCH v2] cleanup: adjust scoped_guard() to avoid potential warning

[Przemek Kitszel]

On 10/10/24 22:13, David Lechner wrote:
> On 10/9/24 6:44 AM, Przemek Kitszel wrote:
> > Change scoped_guard() to make reasoning about it easier for static
> > analysis tools (smatch, compiler diagnostics), especially to enable them
> > to tell if the given scoped_guard() is conditional (interruptible-locks,
> > try-locks) or not (like simple mutex_lock()).
> >
> > Add compile-time error if scoped_cond_guard() is used for non-conditional
> > lock class.
> >
> > Beyond easier tooling and a little shrink reported by bloat-o-meter:
> > add/remove: 3/2 grow/shrink: 45/55 up/down: 1573/-2069 (-496)
> > this patch enables developer to write code like:
> >
> > int foo(struct my_drv *adapter)
> > {
> > 	scoped_guard(spinlock, &adapter->some_spinlock)
> > 		return adapter->spinlock_protected_var;
> > }
> > >
> > Current scoped_guard() implementation does not support that,
> > due to compiler complaining:
> > error: control reaches end of non-void function [-Werror=return-type]
>
> I was hoping that this would allow us to do the same with
> scoped_cond_guard() so that we could remove a bunch of
> unreachable(); that we had to add in the IIO subsystem. But
> with this patch we still get compile errors if we remove them.
>
> Is it possible to apply the same if/goto magic to scoped_cond_guard()
> to make it better too?
sure, will do

I will also combine both macros __helper at the same time to reduce
duplication