Re: I3 sched brakes "realtime" task

Dieter =?ISO-8859-15?q?N=FCtzel?= (Dieter.Nuetzel@hamburg.de)
Thu, 17 Jan 2002 10:17:12 +0100


On Thursday, 17. January 2002 00:09, you wrote:
> On Wed, 16 Jan 2002, Dieter [iso-8859-15] Nützel wrote:
> > running I3 on top of 2.4.18-pre3-I3-VM-22-preempt-lock ;-) This
> > combination rouks!
>
> could you try the -J0 patch i've just uploaded? It has a number of fixes
> and improvements.

Somewhat better but a reniced artsd (-16 or -19) stutters a little during
dbench 32. Maybe there is one show stopper slipped into preempt or lock-break
during my 10_vm-22 merge. Few stalls during heavy IO. But I have the buffer.c
fix applied.

2.4.18-pre4-J0-VM-22-preempt-lock

32 clients started
.......................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
.........................................................+..+...................................................................+..+.+.+.++++.++.+...........++...........++...+.+++.+++.........+...+.++.....+...++....+********************************
Throughput 51.711 MB/sec (NB=64.6388 MB/sec 517.11 MBit/sec)
14.260u 49.860s 1:22.71 77.5% 0+0k 0+0io 939pf+0w

More to come after some sleep...

-Dieter

BTW I'll prepare a unified patch for wider testing if anybody want.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/