> Linus is right, of course. But if the problem really is memory block
> operations...?
>
> > Agreed, but the question is: is some kernel hacker motivated to do
> > implement this before 2.4 ?
We can hope, or try!
I believe this sort of thing can be fixed even after feature freeze, after
all it is acknowledged as a bug, not a "missing" feature.
> > > That would give the same low latency without hurting disk performance under
> > > load.
> >
> > I'm not 100%sure about this ..
> > Nothing comes at zero cost..
>
The "good" solution would be the better one. According to Linus,
the bad latency is a bug, because something is using more time than
necessary. Fixing that will be a win-only scenario, less time used
will improve all cases, both better audio latencies and even better
disk performance under cpu load.
Maybe it won't improve audio latency *enough*, in that case
insert some of the reschedule calls anyway. But you won't need
as many of them if the underlying bug is fixed.
Helge Hafting
-
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/