Dan Hollis writes:
> On Wed, 5 Jul 2000, Richard Gooch wrote:
> > Dan Hollis writes:
> > > On Wed, 5 Jul 2000, Richard Gooch wrote:
> > > > We should always fix bugs, and fix stupid drivers. No question. We
> > > > should always seek to reduce latencies (as long as we don't kill
> > > > overall performance).
> > > So is N>=4000 a bug or feature? Should it be fixed?
> > I have no idea. Where is this huge N coming from? Has anyone done the
> > analysis to determine what's contributing to this big N, and whether
> > or not it's fixable?
>
> benno measured N>=4000, not sure where its coming from
OK, so this is a real number. Damn. That is bad.
> I know we cant get it in 100% of cases (say broken hardware) but in
> the case of known *good* hardware configurations is N<=5 a
> reasonable target?
That's my point: who knows?!? Without an audit, it's shrouded in
darkness.
> Right now even with known "good" hardware we are lucky to get N<=200
>
> At the very least we should keep a list of known bad hardware with
> the reasoning behind it (eg ps2 mouse?)
Right. Start an audit.
> I suspect if people start looking there may be clean ways to deal
> with "bad" hardware and still get good latency without convoluted
> kernel code.
I hope so. But we don't know. And without and audit we'll never know.
Regards,
Richard....
Permanent: rgooch@atnf.csiro.au
Current: rgooch@ras.ucalgary.ca
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jul 07 2000 - 21:00:17 EST