This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Patch to enable -Werror [take 2]
- From: "Kaveh R. Ghazi" <ghazi at caip dot rutgers dot edu>
- To: dje at watson dot ibm dot com
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Sun, 19 Jan 2003 19:05:52 -0500 (EST)
- Subject: Re: Patch to enable -Werror [take 2]
- References: <200301192338.SAA30184@makai.watson.ibm.com>
> From: David Edelsohn <dje@watson.ibm.com>
>
> >>>>> Kaveh R Ghazi writes:
>
> Kaveh> PS: I got your list of AIX warnings. There are only a few minor host
> Kaveh> related issues not covered by existing bypasses. Prototypes for
> Kaveh> strsignal, atoll and ldgetname, and there's the "uchar" conflict
> Kaveh> between cpphash.h and sys/types.h. These are all easy for you to fix
> Kaveh> or add AIX-specific bypasses for in rs6000/t-aix43.
>
> Any suggestion on where to declare atoll? The only use of the
> function is in read-rtl.c. It is defined in read-rtl.c if neither atoll
> nor atoq exist, but it is not static scope. I am surprised that it is not
> provided by libiberty.
> David
The problem with putting it in libiberty, is that your stage1 compiler
may not have 'long long' available, whereas by three-staging
read-rtl.c you get 'long long' in stage2. (I've long thought we
should three-stage libiberty.)
I would suggest simply adding a prototype right above the function
definition, inside the #if. Since it's not used anywhere else, you
might change it to static, but that's not strictly necessary.
--Kaveh
--
Kaveh R. Ghazi ghazi@caip.rutgers.edu