Re: [patch] lowlatency-2.2.10-N2, 'multimedia latency improvement' patch

Benno Senoner (sbenno@gardena.net)
Mon, 2 Aug 1999 20:34:35 +0200

On Mon, 02 Aug 1999, Ingo Molnar wrote:
> On Sun, 1 Aug 1999, Andrea Arcangeli wrote:
>
> > >this is the next version of the 'low latencies' patch. Changes relative to
> >
> > I can't see why you added many run_task_queue calls. For the record,
>
> oh, it's just a remainder of a bad idea. (i was testing wether some
> latencies are due to us constructing too big disk-IO queues via plugged
> interfaces - but no. And the code remained accidentally.) Thanks for
> pointing this out, i've fixed it in my tree. (together with the
> bh->count++ thinko)
>
> -- mingo

Mingo,
did you find additional spots which cold be RAM-size related like in my case ?
Yesterday I retested latencies with mem=32m (without swap what a pain :-) ),
and got about 6ms on disk writes, better than the 64m,128m and 256m case.
If I get some time I will try out Roger's profiling patch let's see if I can
find this nasty function which ruins your nice flat diagrams!
:-)

Note that sometime I must run the test 2-3times to get greater than 5ms
latencies.

regards,
Benno.

-
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/