Re: [PATCH] mtd_blkdevs: Convert to use the kthread API

[Christoph Hellwig]

On Sun, Apr 22, 2007 at 02:23:00PM +0100, David Woodhouse wrote:
> On Sun, 2007-04-22 at 13:24 +0100, Christoph Hellwig wrote:
> > This is the full conversion I sent to Dave in April 2006, but never
> > got any feedback to:
> 
> Sorry about that; I need prodding sometimes. I'll provide some now... 
> 
> Can you show me why the thread won't now miss a wakeup if it goes to
> sleep just as a new request is added to its queue?

Exactly the same thing that happened before.  If you look at
wake_up_process it's just a tiny wrapper around try_to_wake_up.

And wake_up expands to __wake_up expaneds to __wake_up_common
which just walks the list of threads attached to the waitqueue
and then calls curr->func, which expands to try_to_wake_up.

So when your thread still is in running state nothing changes.
If your thread is not in running state it'll get woken by both
variants.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/