[Cocci] modifying initializers with spatch?

[johannes@sipsolutions.net (Johannes Berg)]

> It seems that the control flow was quite complete in one function,
> and the when any at the end of the rule was taking a long time.

Oh, interesting, but that makes sense.

> But actually you don't need the outer ... and braces at
> all.??Furthermore, from the initial declaration to the first
> following statement, there should be no branches,
> so it is safe to put exists on each rule, ie trace through each
> possible control-flow path individually.??The revised semantic patch
> is attached. It terminates quite quickly.

Good point, thanks!

> You can add an option --timeout 120, or some other number of
> seconds.??It will tell you if there are any timeouts, so you can see
> where there may be problems.

That's also useful, didn't know about that. :)

johannes