Wolfgang Weisselberg wrote:
> Hi!
>
> Trying to kill the keyboard, [EMAIL PROTECTED] produced:
> [tape spooling fore and back during write (if I understood correctly]
> > cat $WHERE | cpio -o -a --io-size=$IO_SIZE --format=crc -O $OUTPUT_DEVICE
>
> > afio -o -v -f -b 10240 -c 800 /dev/qft0
>
> I'd suggest piping the output to 'buffer' and let that program use
> a couple of megs for buffering[1]. Man buffer should enlighten
> on it's parameters.
"man buffer" and "info buffer" say no such thing on my system. I can guess
what it must do, and I would suppose cpio will do that with sufficient
--io-size. With ft_dbc_rate_limit = 1000, I get none of this, so it is not a
problem. I asked the question more to get insight into how ftape handles the
buffering.
> -Wolfgang
>
> [1] Why tune a programm when there is another that was written
> for such tasks? :-)
Because my machine does not have it? How do I tell if I need it? I.e., does
ftape work any better with larger input blocks? I assume once they are larger
than ftape's internal buffers, it makes no sense to deliver it larger blocks.
So, how big are those buffers that ftape allocates itself 3 of? Obviously,
they must be at least 10240 bytes, but are they 30720 bytes or something?
--
Registered Linux User 85642.
Jean-David Beyer
Shrewsbury, New Jersey
S/MIME Cryptographic Signature
