Mail Archives: djgpp/1999/09/15/16:25:54
This is a multi-part message in MIME format.
--------------D454512E57769D041CA09C26
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Eli Zaretskii wrote:
> On Thu, 9 Sep 1999, Tony Welsh wrote:
>
> > My only thought is the last command line - is the last backslash the problem
> > as the rest are UNIX style (on MSDOS?) .
>
> No, DJGPP handles both forward and backslashes alike.
>
> > C:\djgpp\src>gcc -c -v chap01_1.c
> > Reading specs from c:/djgpp/lib/specs
> > gcc version 2.95 19990728 (release)
> > c:/djgpp/lib/gcc-lib/djgpp/2.95/cpp.exe -lang-c -v -isystem
> > c:/djgpp/bin/include -D__GNUC__=2 -D__GNUC_MINOR__=95 -Dunix -Di386 -DGO32
> > -DMSDOS -DDJGPP=2 -DDJGPP_MINOR=2 -D__unix__ -D__i386__ -D__GO32__ -D__MSDOS__
> > -D__DJGPP__=2 -D__DJGPP_MINOR__=2 -D__unix -D_i386 -D__GO32 -D__MSDOS
> > -D__DJGPP=2 -D__DJGPP_MINOR=2 chap01_1.c c:/djgpp/tmp\ccnTHF0i.i
>
> This indicates that the program which crashes is cpp.exe. Is that what Windows
> says in the message that it pops up?
No, its a DOS problem, so it only tells you about the DOS fault, only windows
application faults tell you about the item that caused the problem.
> Do you see the temporary file (c:/djgpp/tmp\ccnTHF0i.i in this case) on disk after
> the crash?
Yes, and it is blank, size 0 bytes.
> Please invoke the preprocessor manually, like shown below, and see if it works:
>
> c:\djgpp\lib\gcc-lib\djgpp\2.95\cpp -lang-c -isystem c:/djgpp/bin/include
> chap01_1.c c:/djgpp/tmp\ccnTHF0i.i
>
> Note: this is all one long line!
C:\djgpp>c:\djgpp\lib\gcc-lib\djgpp\2.95\cpp -lang-c -isystem C:/djgpp/include
C:/djgpp/src/chap01_1.c c:/djgpp/tmp\ccnTHF0i.i
(the previous file)
C:\djgpp\src>c:\djgpp\lib\gcc-lib\djgpp\2.95\cpp -lang-c -isystem
C:/djgpp/include
chap01_1.c c:/djgpp/tmp\cc3CiKyf.i
(a file from this morning that crashed as normal with an option 'c')
both execute without errors or for that matter any error messages, it just
returns to the prompt after execution. I notice that after these are
executed, the temporary file mentioned is filled with what looks like the
original source code but
without any non-executable lines... (see cc3CiKyf.i and chap01_1.c attached).
Which is I assume preprocessing - which is a start - now all I need to know is
what's wrong.
> Please also try this command and tell what it produces:
>
> go32-v2 c:/djgpp/bin/gcc.exe -v
C:\djgpp>go32-v2 c:/djgpp/bin/gcc.exe -v
Reading specs from c:/djgpp/lib/specs
gcc version 2.95 19990728 (release)
> > I wouldn't think that a virus scanner would interfere with a
> > compile, I could understand if the program was running and doing
> > some low level commands and then it fails, but not a compile.
>
> Antivirus sits on the Windows program invocation chain and checks every program
> that is invoked. So it could interfere with program invocation, especially since
> DJGPP programs are not real-mode DOS executables.
--
Best Regards - Tony
-
"Mine's a black coffee, with two sugars and milk..."
- Dexter's Lab
--------------D454512E57769D041CA09C26
Content-Type: text/plain; charset=us-ascii;
name="cc3CiKyf.i"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="cc3CiKyf.i"
# 1 "chap01_1.c"
void fn(void)
{
int autoVar;
static int staticVar = 0;
autoVar = 10;
staticVar = 20;
}
int main()
{
fn();
fn();
return 0;
}
--------------D454512E57769D041CA09C26
Content-Type: text/plain; charset=us-ascii;
name="Chap01_1.c"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="Chap01_1.c"
//Chap01_1.c
void fn(void)
{
int autoVar; /*autoVar starts life here*/
static int staticVar = 0; /*value first time fn() is called*/
autoVar = 10;
staticVar = 20;
} /*autoVar ends life here*/
int main()
{
fn();
fn();
return 0;
}
--------------D454512E57769D041CA09C26--
- Raw text -