[Cbe-oss-dev] [PATCH 2/2] poll: allow f_op->poll to sleep

[Christoph Hellwig]

On Tue, Aug 26, 2008 at 09:58:14AM -0700, Linus Torvalds wrote:
> I don't really see the point.
> 
> poll() isn't allowed to sleep for many reasons. Some are technical. But 
> the most obvious one is that a sleeping "poll()" is totally against the 
> whole point of polling in the first place!
> 
> So is there some big conceptual reason to change how poll() has always 
> worked?

Just as a little sidenote most files on spufs have a ->poll that sleeps,
and currently we don't have any debugging to catch this, I just noticed
this by accident.  The reason that it sleeps is because it needs to grab
a sleping lock, and as that lock protects against a rather complicate
type of hardware context switch it's not possible to replace it with
a spinlock.