On Thu, Feb 24, 2000 at 10:43:31PM +0100, Ingolf Koch wrote:
> To be more precise: 20 cps with /dev/par0.

It turns out that the 20 cps were caused by some printf()s
I had inserted (two per transmitted character). Sorry for
the noise.

Without the printf()s, it's faster but still only about
500 cps.

The code however seems to be unnecessary complicated:
why not use the timo parameter of the tsleep(9) instead
of installing a seperate timeout(9)?

See the attached patch file. This does not speed things
up significantly but improves readability a bit.

    Ingolf
-- 

Ingolf Koch     ICQ#60829470     Beste Kneipe in Jena-Ost
PGP: 0x7B3B5661  213C 828E 0C92 16B5  05D0 4D5B A324 EC04