Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uwm.edu!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!ira.uka.de!smurf!ipkama!freesid!mamnix!tjack!fiction!Daniel_Roedding
From: Daniel_Roedding@fiction.ms.sub.org
Newsgroups: comp.lang.c
Subject: Two for one...
Message-ID: <683194@fiction>
Date: 29 Aug 90 23:12:16 GMT
Lines: 67
X-Attributes: type 2, from 100, acclv 0, expires

cchase@ee.cornell.edu (Craig M. Chase) writes:

> For reasons that 
> I won't go into here, we've got a bunch of registers that work as
> pairs.  What I mean is that (unless something goes drastically 
> wrong) these registers will always have the save value.  When I 
> write to one, I really want to write to both.

(...)

> Problem is, if I try:
>    if (fancy_reg == 0xFFFF)
> I get:
>    if (reg1 = reg2 == 0xFFFF)
> and that doesn't do the right thing (== binds tighter than =, right?)

BTW: Do you really want a re-assignment when reaching this if-statement?




I think it's more elegant to access such a hardware register by two
pseudo-functions, e.g.:

#define setreg(x,y) x=(y)
#define getreg(x) (x)

Then you could use:

#define fancy_reg reg1 = reg2

... and then:

Your solution:

> So, if I do
>    fancy_reg = 0xFFFF;

> the preprocessor generates:
>    reg1 = reg2 = 0xFFFF;

And the new one:

setreg(fancy_reg,0xffff);

-> reg1 = reg2 = (0xffff);


When reading from the port:

if (getreg(fancy_reg)==0xffff)

-> if ((reg1 = reg2) == 0xffff)


Probably it's more elegant -- but it's no good when you access the registers
very often (or do you like to hundredfold write rendundant stuff like
"getreg"? - anyway, it's the way how *I* would solve the problem).

Daniel

---
Daniel Roedding                                       #include <disclaimer.h>
Geiststrasse 32  "This golden age of communication
D-4400 Muenster   Means everyone talks at the same time" (NMA)
++49 251 525306
daniel@fiction.ms.sub.org; ..!uunet!mcsun!unido!mcshh!veeble!fiction!daniel