On Mon, 25 Feb 2002, Benjamin LaHaise wrote:
>
> I disagree: requiring aio to execute completion in user context means
> that we can no longer have quick completion directly from an interrupt
> handler to a busy server executing in userland.
Note that teh IO _completion_ can happen in user land. That does not mean
that the IO page tear-down can (or should) happen at the same time.
There are two issues here:
- the data structures supporting the IO, built up and torn down from
process space.
- the IO itself.
One is synchronous, the other is not.
Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Feb 28 2002 - 21:00:17 EST