On Mon, Jul 02, 2012 at 12:41:59PM -0700, H. Peter Anvin wrote:

On 07/02/2012 11:33 AM, Mark Brown wrote:

quoted

quoted

Last I saw I saw a patch to that effect, asked what the benefit
was, and got no answer.

quoted

Are you positive about that?  I don't recall you replying any of
the times I sent out the patch and my mail archive isn't
contradicting me either.

I said last time I saw a patch to that effect; it might not have been
from you.  I might not have seen yours for whatever reason (including
losing it on my end.)

I'm kind of surprised anyone else has been sending stuff (unless mine
got resent by someone else, I did include it in some of my serises for
the clock API); I know I've posted mine several times now.  In any case,
I hope the mail you're replying to answers your question about why it's
useful.

In general I'd probably go further and say that (at least when a generic
implementation is available) there should be a very good reason for not
enabling an API on an architecture rather than the other way around.