Re: Efficient read of binary data from REL records

From: André Fachat <afachat_at_gmx.de>
Date: Sat, 10 Oct 2020 16:10:32 +0200
Message-ID: <17512d94bc0.27ff.b4d1f2b66006003a6acd9b1a7b71c3b1_at_gmx.de>
Am 10. Oktober 2020 15:58:55 schrieb vtgearhead <snhirsch_at_gmail.com>:

> Since it terminates on various characters, the input# BASIC function is not
> usable in the case where a REL record contains random binary data.  To speed
> things up, is there a way to do a bulk read of given size from assembly
> language that does not incur a TALK/UNTALK cycle for every byte?

Absolutely. Just call TALK/SECTALK, then read as many files from IEEE as
you need, then call UNTALK.

The C64 has IIRC kernal jump table entries for these routines. OS9 seems to
imply PET, you have to call the kernel IEEE488 routines directly. Just look
at the code to load a file for an example.

André

> In my
> application it's guaranteed that all records will be completely filled
> (Super OS-9 container file w/ 129-byte records).
>
>
>
> --
> Sent from: http://cbm-hackers.2304266.n4.nabble.com/
Received on 2020-10-10 17:00:03

Archive generated by hypermail 2.3.0.