[PATCH RFC 07/77] PCI/MSI: Re-design MSI/MSI-X interrupts enablement pattern

[tj@kernel.org (Tejun Heo)]

Hello, Alexander.

On Tue, Oct 08, 2013@09:48:26AM +0200, Alexander Gordeev wrote:
> > If there are many which duplicate the above pattern, it'd probably be
> > worthwhile to provide a helper?  It's usually a good idea to reduce
> > the amount of boilerplate code in drivers.
> 
> I wanted to limit discussion in v1 to as little changes as possible.
> I 'planned' those helper(s) for a separate effort if/when the most
> important change is accepted and soaked a bit.

The thing is doing it this way generates more churns and noises.  Once
the simpler ones live behind a wrapper which can be built on the
existing interface, we can have both reduced cost and more latitude on
the complex cases.

> > If we do things this way, it breaks all drivers using this interface
> > until they're converted, right?
> 
> Right. And the rest of the series does it.

Which breaks bisection which we shouldn't do.

> > Also, it probably isn't the best idea
> > to flip the behavior like this as this can go completely unnoticed (no
> > compiler warning or anything, the same function just behaves
> > differently).  Maybe it'd be a better idea to introduce a simpler
> > interface that most can be converted to?
> 
> Well, an *other* interface is a good idea. What do you mean with the
> simpler here?

I'm still talking about a simpler wrapper for common cases, which is
the important part anyway.

Thanks.

-- 
tejun