On Tue, Mar 07, 2000, Vojtech Pavlik <vojtech@suse.cz> wrote:
> On Tue, Mar 07, 2000 at 01:25:18AM +0100, Jamie Lokier wrote:
>
> > > Wet string? Could you describe the problem better? I'd like to know if
> > > there's anything wrong with ACM ... Pavel has reported massive system
> > > slowdown with using JE's UHCI (due to bandwith reclamation), is it
> > > that?
> >
> > Bandwidth reclamation is turned off, but I am using JE's UHCI. My
> > system seems ok, it's just the modem connection that's slow.
Ummm. You can't turn it off for my UHCI driver, unless you hacked the
driver? The AFS driver has a compile option tho.
> > The TCP session is very conditional. Press a key, wait 5 seconds for
> > display update. Here are some ping times. It could be my ISP today, or
> > maybe the LAN at the far end is overloaded today.
> >
> > I only noticed since I went from 2.3.40 to 2.3.49 recently.
> >
> > 64 bytes from xx.xx.xx.xx: icmp_seq=10 ttl=241 time=270.0 ms
> > 64 bytes from xx.xx.xx.xx: icmp_seq=11 ttl=241 time=730.0 ms
> > 64 bytes from xx.xx.xx.xx: icmp_seq=12 ttl=241 time=300.0 ms
> > 64 bytes from xx.xx.xx.xx: icmp_seq=13 ttl=241 time=320.0 ms
> > 64 bytes from xx.xx.xx.xx: icmp_seq=14 ttl=241 time=760.0 ms
> > 64 bytes from xx.xx.xx.xx: icmp_seq=15 ttl=241 time=540.0 ms
>
> Well, this looks like a lightly loaded modem connection. Could you
> please make sure the modem isn't used for anything but the ping
> and ping the opposite point of the link only?
>
> Btw, there were no important changes between 2.3.40 and 2.3.49 in acm.c
> that could be causing this (except to the wakeup stuff now, which could
> delay data by up to 10 ms).
JE
-
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 : Tue Mar 07 2000 - 21:00:22 EST