On Sun, 2005-12-18 at 10:42 -0800, Linus Torvalds wrote: > It's easy enough to add a "might_sleep()" to the up(). Not strictly true, > but conceptually it would make sense to make up/down match in that sense. > We'd have to mark the (few) places that do down_trylock() + up() in > interrupt context with a special "up_in_interrupt()", but that would be ok > even from a documentation standpoint.
Actually, I don't think you want might_sleep(): there are a few cases where we do an up() from under a spinlock, which will spuriously trigger this. I'd suggest WARN_ON(in_interrupt()) instead. James
