home *** CD-ROM | disk | FTP | other *** search
- Date: Thu, 21 Jul 1994 17:18:38 -0400
- From: "Nicholas S Castellano" <entropy@terminator.rs.itd.umich.edu>
- To: rgivan@cix.compulink.co.uk
- In-Reply-To: Roland Givan's message of Thu, 21 Jul 94 00:18 BST-1 <memo.716071@cix.compulink.co.uk>
- Subject: Rsconf problems under MiNT 1.10h6
-
- >Date: Thu, 21 Jul 94 00:18 BST-1
- >From: rgivan@cix.compulink.co.uk (Roland Givan)
- >Reply-To: rgivan@cix.compulink.co.uk
- >
- >Thanks to everybody who pointed me in the right direction for h6 patches.
- >I have now successfully built a h6 MTOS kernel.
- >
- >All seems to work fine except Rsconf().
- >
- >I first noticed this when MODEM.CPX displayed rubbish instead of the baud
- >rate. Removing all ACCs and AUTO programs didn't help, So I tried the
- >following:
- >
- >#include <stdio.h>
- >#include <osbind.h>
- >
- >main()
- >{
- > printf("speed = %d\n",Rsconf(-2,0,0,0,0,0));
- > getchar();
- >}
- >
- >This was compiled under GCC 2.33 MiNT Libs PL44 (as was the kernel).
- >
- >Under h6 'speed' was random (large) numbers. Under h3 and TOS 2.06 'speed'
- >was correct.
- >
- >I also tried MODEM.CPX under h6 on a TOS 1.4 machine as well. It had
- >problems too.
- >
- >Any ideas anybody?
-
- First of all, testing this without any auto folder programs is
- definitely a good idea...i know of at least one TOS patch that breaks
- Rsconf(-2, ...).
-
- Also, i'm not sure if it matters, but you probably want to use -1
- instead of 0 for all the other parameters, as -1 means "do not
- change". The -2 for bps-rate inquire may cause Rsconf() to ignore
- the rest of the args anyway, but I'm not sure in which order the
- parameters are checked.
-
- Next, make sure that both your MiNT kernel and the program calling
- Rsconf() have been compiled with the latest MiNT headers (The library
- version isn't so important here, but the osbind.h you compile with
- should be a recent one, as old ones had a bad Rsconf() binding.)
-
- If none of that helps, let me know what TOS version you're
- running...the MiNT Rsconf() code is very version-dependent. I tested
- Rsconf(-2, -1, -1, -1, -1, -1) on a TT with the latest TT TOS and an
- h6 kernel, and it gave perfectly reasonable results.
-
- -entropy
-
- --
- entropy -- it's not just a good idea, it's the second law.
- Personal mail: entropy@gnu.ai.mit.edu
- MiNT library mail: entropy@terminator.rs.itd.umich.edu
- "what do you have against octal?" -jrb
-
-