Re: Interrupt latency on 2.0.3x

Harald Koenig (koenig@tat.physik.uni-tuebingen.de)
Sat, 7 Jun 1997 11:38:23 +0200


On Jun 6, Philip Gladstone wrote:

> Does anyone know what the maximum interrupt latencies are
> under 2.0.30? I realise that this depends on the speed of
> the CPU chip, but what sort of numbers should I expect?

the absolute maximum latencies (talking about hard realtime constraints)
can't be given for Linux not being a non-realtime OS.

for my good old 486dx/66 once I got ~12-13 usec interrupt latencies
where timing histograms showed some exceeds over ~100uses, and
very few >~500usec (this was using Linux ~1.0..1.1 I think).
while the 12usec was great, it was not possible to guearantee 1msec...

Harald

--
All SCSI disks will from now on                     ___       _____
be required to send an email notice                0--,|    /OOOOOOO\
24 hours prior to complete hardware failure!      <_/  /  /OOOOOOOOOOO\
                                                    \  \/OOOOOOOOOOOOOOO\
                                                      \ OOOOOOOOOOOOOOOOO|//
Harald Koenig,                                         \/\/\/\/\/\/\/\/\/
Inst.f.Theoret.Astrophysik                              //  /     \\  \
koenig@tat.physik.uni-tuebingen.de                     ^^^^^       ^^^^^