Re: including sparse headers in C++ code

[Bernd Petrovitsch <bernd@sysprog.at>]

On Mit, 2010-10-20 at 00:49 +0200, Tomas Klacko wrote: 
> On Tue, Oct 19, 2010 at 11:31 PM, Al Viro <viro@zeniv.linux.org.uk> wrote:
> > On Tue, Oct 19, 2010 at 10:03:42PM +0200, Tomas Klacko wrote:
> 
> >>  static inline int match_op(struct token *token, int op)
> >>  {
> >> -     return token->pos.type == TOKEN_SPECIAL && token->special == op;
> >> +     return token->pos.type == TOKEN_SPECIAL && token->special == (unsigned int)op;
> >
> > What was that one for?
> 
> "warning: comparison between signed and unsigned integer expressions"

Add "-Wsign-compare" to the CFLAGS (for a pure C compile) and you will
find many more of them.
IMHO it needs a decision if the goal is to prefer signed or unsigned int
(where no negative values ever used/assigned) and convert struct members
and/or global variables. Most of them TTBOMK are never negative anyways.

For the "C vs C++" issue: if you have an enum, the rules (or at least
the reality defined by gcc) are subtly different if it's a signed int or
an unsigned int and the values of their members.

Bernd
-- 
mobile: +43 664 4416156              http://www.sysprog.at/
    Linux Software Development, Consulting and Services