X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f
X-Recipient: djgpp AT delorie DOT com
Date: Thu, 14 May 2015 18:20:26 +0300
From: "Eli Zaretskii (eliz AT gnu DOT org)" <djgpp AT delorie DOT com>
Subject: Re: bad pragma in dir.h? (and our structrure packing)
In-reply-to: <CAA2C=vAK_T9ixd6YpNrd2LYL80Lau-Dbf+T5vo89AdYXS0WYOw@mail.gmail.com>
X-012-Sender: halo1 AT inter DOT net DOT il
To: djgpp AT delorie DOT com
Message-id: <83vbfvi3t1.fsf@gnu.org>
References: <CAA2C=vCgLHdH3BJxastGzUsJzhiRddytiYwB1MP_aaiiVpC4nA AT mail DOT gmail DOT com> <83k2wcjt8e DOT fsf AT gnu DOT org> <CAA2C=vB-YiGkyx5dJpa=hcBh0O4_NiEKh2tKm5OHyNX3vW7HsQ AT mail DOT gmail DOT com> <83bnhojnwh DOT fsf AT gnu DOT org> <CAA2C=vAEKFUktuYXGN_eYUEY0JuHQgXR_-q-N8xox=7PHPEqqw AT mail DOT gmail DOT com> <838ucsjnbl DOT fsf AT gnu DOT org> <CAA2C=vAK_T9ixd6YpNrd2LYL80Lau-Dbf+T5vo89AdYXS0WYOw AT mail DOT gmail DOT com>
Reply-To: djgpp AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> Date: Wed, 13 May 2015 23:27:39 +0300
> From: "Ozkan Sezer (sezeroz AT gmail DOT com)" <djgpp AT delorie DOT com>
> 
> > If you (or someone else) can verify that the size of the structure and
> > the offsets of each member are the same in both C and C++ programs,
> > then I see no reason to keep the attribute on every member.
> 
> I did the following as a foo.c and foo.cc:
> 
> #define COMPILE_TIME_ASSERT(name, x) \
>   typedef int _djchk_ ## name[(x) * 2 - 1]
> 
> #include "dos.h"
> COMPILE_TIME_ASSERT(find_t,sizeof(struct _find_t)==286);
> 
> #include "dir.h"
> COMPILE_TIME_ASSERT(ffblk,sizeof(struct ffblk)==290);
> COMPILE_TIME_ASSERT(ffblklfn,sizeof(struct ffblklfn)==318);
> 
> #include "coff.h"
> COMPILE_TIME_ASSERT(exln,sizeof(struct external_lineno)==6);
> COMPILE_TIME_ASSERT(exse,sizeof(struct external_syment)==18);
> COMPILE_TIME_ASSERT(exau,sizeof(union external_auxent)==18);
> COMPILE_TIME_ASSERT(exrl,sizeof(struct external_reloc)==10);
> 
> Compiled using existing and modified headers for djgpp using gcc/g++
> v3.4.6 and 5.1.0, no errors.  I also compiled its coff.h-only parts
> for linux (i686 gcc-4.3.0 and clang-3.4.2), osx (ppc gcc-4.0.1 and
> x86_64 gcc-4.2.1), mingw (gcc-3.4.5), mingw-w64 (x86_64 gcc-4.5.4):
> no errors.
> 
> Diff for the modified headers is attached (1.patch)
> Ideas?

LGTM, thanks.

> On 5/13/15, Andris Pavenis (andris DOT pavenis AT iki DOT fi) <djgpp AT delorie DOT com> wrote:
> > I guess it would be a good idea to have small test in build process to
> > verify that offsets of
> > members are correct for structures intended for DOS calls. So build would
> > fail if something is
> > wrong with member packing
> >
> > Andris
> 
> We can add compile-time asserts like the above to the headers: ideas?

I think a test in djtst would be better.

Thanks.