Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!think!harvard!seismo!mcvax!cernvax!jmg
From: jmg@cernvax.UUCP (jmg)
Newsgroups: net.micro.cbm
Subject: C-Power (version 2.8) problems
Message-ID: <264@cernvax.UUCP>
Date: Mon, 27-Jan-86 06:24:58 EST
Article-I.D.: cernvax.264
Posted: Mon Jan 27 06:24:58 1986
Date-Received: Sat, 1-Feb-86 06:36:07 EST
Reply-To: jmg@cernvax.UUCP (Mike Gerard, DD Division, CERN.)
Organization: CERN, Geneva/Switzerland
Lines: 36

Please excuse my not going via the bulletin boards for C-Power: I have
no access to them. I have some odd problems which might be known already,
and might even be "fixed in next rev.".

1. Code sequence
       if (...)
       { int i;
	 .
	 .
       }
       else
       { int i;
	 .
	 .
       }
   The second definition of i is not liked by the compiler (sometimes!).
   I think it may be when there are no more register variables: I
   successively cut down my code, and it eventually was accepted.

2. I have various typedefs defined in a header file. In a procedure which
   I want to change to call sys, I tried to define a structure including
   these typedefs. Was not accepted by the compiler: I had to revert
   to standard types (char, int etc.). The typedefs are correct, and
   accepted for normal variables.

3. When I did call sys as defined in the manual, the x and y values
   were not correct (the address of the structure), but were 1c beyond.

4. How does one return a 16-bit result? The example in the manual simply
   puts an eight-bit answer in a, which seems odd!

5. if x is a doubly-dimensioned array, to get a pointer to a particular
   row, I must use x[n], not *(x+n). The latter is OK on BSD4.2


p.s. I think the product is excellent, really, but not (yet) perfect.