Re: building deutex 4.4.902 with gcc 4.0.3

  • From: Andre Majorel <amajorel@xxxxxxxxx>
  • To: Jon Dowland <jon@xxxxxxxxxxx>
  • Date: Wed, 15 Mar 2006 15:11:47 +0100

On 2006-03-14 23:16 +0000, Jon Dowland wrote:

> GCC 4 complains about differing signedness of variables.
> 
> There are a few instances of this in the deutex source. I've
> attached the relevant build log (build.log) and a summary of
> the signed-ness of things passed to  "WadRreadBytes"  as the
> second argument (WadRreadBytes-summary.txt) :  char *  (i.e.
> implementation-dependent sign) is a clear favourite compared
> to "unsigned char *" for current usage. I can't see anything
> which would break by changing the prototype to unsigned, and
> all the callers,  but I can't see the vars being interpreted
> as characters either... just data buffers.

> src/ident.c:205: warning: pointer targets in passing argument 2
>   of ???WADRreadBytes??? differ in signedness
> src/ident.c:208: warning: pointer targets in passing argument 1
>   of ???parse_pic_header??? differ in signedness
[...]

Hi Jon. I know about these. If there are any architectures where
you can't swap (char *) and (unsigned char *), I doubt they have a
Debian port. How badly do you want these warnings removed ? Unless
there's a strong reason for it, I plan to leave them alone for the
time being.

-- 
André Majorel <amajorel@xxxxxxxxx>
http://www.teaser.fr/~amajorel/

Other related posts: