Re: [Gnu-arch-users] hackerlab, tla, and `char' signedness

[Andy Tai]

I have reservations about this...  while these warnings (also seen
on x86) from gcc 4 are indications of possible problems, I am not sure
in eacn case the intention of people when using char mean they are
unsigned.

This may have other unseen consequences.  Maybe it is better to go to the code and fix the warnings.

Andy

On 1/12/06, Ludovic Courtès <address@hidden> wrote:

Hi,

Hackerlab and tla extensively use pointers to `t_uchar' to represent
strings.  However, `t_uchar' is different from `char' and unfortunately
C strings are of type pointer to `char' (not `signed char' nor `unsigned
char').  Therefore, we may get warnings such as "pointer targets differ
in signedness" (I do get them on PPC with GCC 4).

>From the GCC 4 manual:

  `-funsigned-char'
       Let the type `char' be unsigned, like `unsigned char'.


So I believe we should explicitly compile with `-funsigned-char'.

Thanks,
Ludovic.