Better tty hardware? (WAS Re: some uucp transfer rates . . .)
Kris A. Kugel
kak at hico2.UUCP
Sat Jun 8 02:39:04 AEST 1991
In article <1991Jun2.231810.7514 at ingres.Ingres.COM>,
rog at ingres.com (Roger Taranto) writes:
> In article <1991May29.224256.24529 at ceilidh.beartrack.com>
> dnichols at ceilidh.beartrack.com (DoN Nichols) writes:
> >In article <1789 at hico2.UUCP> kak at hico2.westmark.com writes:
> >>I suspect that a 3b1 can send faster than it can receive.
> >
> > That, I believe, has been the general concensus of the discussion.
>
> That seems to make sense. It seems to me that terminal drivers and
> interrupt routines were designed to be able to send lots of data, but
> receive comparatively low amounts. (How many users do you know who
> can type at 19200 bps?)
>
> We have a microvax II . . .
> With nothing else going on and just one UUCP connection
> (incoming), the CPU is pegged at ~93-95% trying to handle the single
> character interrupts for that uucico process.
>
> -Roger
> {mtxinu,pacbell,amdahl,sun,hoptoad}!rtech!rog rog at ingres.com
A couple of years ago, I remember seeing some kind of discussion
about producing a serial board that could buffer several characters,
so that there was no longer one interrupt per character.
This would allow for a driver that could collect several characters
at a time.
Unfortunatly, I think this was in relationship to 286/386 machines,
but if the interrupt handling is our preformance problem,
perhaps such a thing could be added to the Mondo Combo board
(How IS that coming, by the way?)
Kris A. Kugel
( 908 ) 842-2707
uunet!tsdiag.ccur.com!hico2!kak (maybe)
{daver,ditka,zorch}!hico2!kak
internet: kak at hico2.westmark.com
More information about the Comp.sys.3b1
mailing list