Date: Wed, 15 Oct 1997 13:12:32 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: Dag Håkon Myrdal <daghakmy AT login DOT eunet DOT no>
cc: djgpp AT delorie DOT com
Subject: Re: Using inportb()
In-Reply-To: <61vc83$565$1@elle.eunet.no>
Message-ID: <Pine.SUN.3.91.971015131204.17868I-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 14 Oct 1997, Dag Håkon Myrdal wrote:

> #include <pc.h>
> [...]
> outportb(0x22C);             /* write to SW trigger */                 
> inportb(0x225);              /* read ADC value and status */
> 
> What happens is that the compiled program seems not to get in
> contact with the labcard. The status never changed when used as below.

You use `outportb' incorrectly: it requires a second argument--the
value being written to the port.  The way you used it, it writes some
random value (whatever it finds on the stack), which will certainly
not work.

You should use the -Wall compilation switch to cause GCC emit
diagnostics for such cases.

I don't know why the emx version doesn't have two arguments.  It
should know somhow what value to write to that port.

> To me, the only obvious difference here is that I no longer 
> explitly ask for access to the IO bus - is this not necessary with
> djgpp?

No, at least as long as you use it under MS-DOS.