Compiler warnings in gcc 4.3.2

[Austin English]

I recently upgraded to intrepid, which has gcc 4.3.2. It gave a few
new warnings, and I've got patches ready for all but this one. I
started to write a patch to remove the check, since it's useless, but
wasn't sure if it should instead be converted to a signed integer
instead.

austin(a)austin-desktop:~/wine-git/dlls/dnsapi$ make
ccache gcc -c -I. -I. -I../../include -I../../include  -D__WINESRC__
-D_REENTRANT -fPIC -Wall -pipe -fno-strength-reduce
-fno-strict-aliasing -Wdeclaration-after-statement -Wwrite-strings
-Wtype-limits -Wpointer-arith  -Werror -g  -o ns_parse.o ns_parse.c
cc1: warnings being treated as errors
ns_parse.c: In function 'dns_ns_parserr':
ns_parse.c:139: error: comparison of unsigned expression < 0 is always false
make: *** [ns_parse.o] Error 1

Relevant code:
int
dns_ns_parserr(ns_msg *handle, ns_sect section, int rrnum, ns_rr *rr) {
	int b;

	/* Make section right. */
	if (section < 0 || section >= ns_s_max)
		RETERR(ENODEV);

-- 
-Austin

[James Hawkins]

On Tue, Nov 11, 2008 at 9:17 PM, Austin English <austinenglish(a)gmail.com> wrote:
> I recently upgraded to intrepid, which has gcc 4.3.2. It gave a few
> new warnings, and I've got patches ready for all but this one. I
> started to write a patch to remove the check, since it's useless, but
> wasn't sure if it should instead be converted to a signed integer
> instead.

Change what to signed?  We don't control the type of ns_sect.

-- 
James Hawkins

[Austin English]

On Tue, Nov 11, 2008 at 9:28 PM, James Hawkins <truiken(a)gmail.com> wrote:
> On Tue, Nov 11, 2008 at 9:17 PM, Austin English <austinenglish(a)gmail.com> wrote:
> > I recently upgraded to intrepid, which has gcc 4.3.2. It gave a few
> > new warnings, and I've got patches ready for all but this one. I
> > started to write a patch to remove the check, since it's useless, but
> > wasn't sure if it should instead be converted to a signed integer
> > instead.
>
> Change what to signed?  We don't control the type of ns_sect.
>
> --
> James Hawkins

Sorry, that was worded poorly. We could change section to a signed
int, but that just screams hackish. Removing the check didn't seem
appropriate either, but since it's useless, may be the appropriate
method.

-- 
-Austin