On Tue, 19 Sep 2000, Tim Waugh wrote:
>
> On Tue, Sep 19, 2000 at 06:38:03PM +0200, Benno Senoner wrote:
>
> > And I registered a two 6msec peaks on this kernel :-(
> > Perhaps it is due to the polled printer driver ?
>
> It could be. Was it using interrupts (what does dmesg say)? We do a
> couple of fast polls, but I don't think that they should be anything
> like 6msec: more like 25usec. Perhaps the logic there is broken?
what do you mean with broken logic ?
BTW: my PC is a PII and mounts an Asus P2B (BX) motherboard
this is the dmesg output:
parport0: PC-style at 0x378 (0x778) [SPP,ECP,ECPEPP,ECPPS2]
parport0: detected irq 7; use procfs to enable interrupt-driven operation.
parport_probe: succeeded
parport0: Printer, Hewlett-Packard HP LaserJet 1100
lp0: using parport0 (polling).
Do I need to do echo "7" >/proc/.... in order to enable IRQs ?
will this improve the situation in your opinion ?
>
> There is support for using DMA on ECP-enabled chips in 2.4.0-test, but
> not in 2.2. Enable it using a kernel config option.
Interesting,I will try it when Ingo finishes his low latency patches for 2.4.
Again, in your opinion, do you think that this will help to lower latencies ?
many thanks for your infos Tim.
cheers,
Benno.
