* Uwe Hermann uwe@hermann-uwe.de [070701 02:14]:
OK, I totally misread what the code is supposed to do.
I'm not sure the spin_define() is such a good idea then. It has a nice name and all, but it might be obfuscating things more than necessary. Maybe a simple variable declaration (i.e., drop the #define) would be better?
I don't see why it would be better, there are no obvious advantages, but some obvious disadvantages. Here our principles start clashing. We can't do this for several reasons:
* The declaration is static, so it has to be surrounded by #if defined(CONFIG_SMP) would waste all the places in the code. This is controverse to our goal to make changes as local as possible. If we don't, we end up with warnings in the code for almost every board
* Since some of us started hating typedefs, we don't have anything to declare anymore. You simply can not rely that a spinlock is represented by struct spinlock on all possible systems. So to get this done right, we would have to go back to spinlock_t, in which case I would vote to also go back to device_t and all the others, so we stay consistent.
Stefan