[Bug fortran/27046] New: gfortran print flush in dll

[Bug fortran/27046] New: gfortran print flush in dll

[mikko dot lyly at csc dot fi]

PRINT* in gfortran 4.2.0 compiled dll's needs CALL FLUSH to perform correctly
when called from gcc (3.4.2, 3.4.5, 4.1) compiled main. Reproducible sample:

$ gfortran -shared -o ftesti.dll ftesti.f90

ftesti.f90:
-----------
subroutine print_from_gfortran(txt)
  implicit none
  character :: txt
  print *,txt
end subroutine print_from_gfortran

$ gcc -o ctesti.exe ctesti.c

ctesti.c:
---------
#include <stdio.h>
#include <windows.h>

void print_from_gcc(char* txt) {
  printf("%s\n",txt);
}

int main(int argc, char** argv) {
  HINSTANCE handle=LoadLibrary("ftesti.dll");
  FARPROC print_from_gfortran=GetProcAddress(handle,"print_from_gfortran_");

  print_from_gcc     ("c");
  print_from_gfortran("f");
  print_from_gcc     ("c");
  print_from_gfortran("f");

  return 0;
}

Now, running ctesti.exe produces

c
c
f
f

Correct output should be

c
f
c
f

Adding CALL FLUSH in ftesti.f90 fixes the case:

ftesti.f90:
-----------
subroutine print_from_gfortran(txt)
  implicit none
  character :: txt
  print *,txt
  call flush
end subroutine print_from_gfortran


-- 
           Summary: gfortran print flush in dll
           Product: gcc
           Version: 4.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: mikko dot lyly at csc dot fi
  GCC host triplet: i686-pc-mingw32


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug fortran/27046] gfortran print flush in dll

[pinskia at gcc dot gnu dot org]

------- Comment #1 from pinskia at gcc dot gnu dot org  2006-04-05 16:02 -------
*** Bug 27047 has been marked as a duplicate of this bug. ***


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug libfortran/27046] gfortran print flush in dll

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2006-04-05 16:08 -------
This was fixed for the non windows case for sure.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|fortran                     |libfortran
   GCC host triplet|i686-pc-mingw32             |
 GCC target triplet|                            |i686-pc-mingw32


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug libfortran/27046] gfortran print flush in dll

[fxcoudert at gcc dot gnu dot org]

------- Comment #3 from fxcoudert at gcc dot gnu dot org  2006-04-25 06:04 -------
(In reply to comment #2)
> This was fixed for the non windows case for sure.

Yes:

$ cat ftesti.f90 
subroutine print_from_gfortran(txt)
  implicit none
  character :: txt
  print *,txt
end subroutine print_from_gfortran
$ ./bin/gfortran -shared -o ftesti.so ftesti.f90 
$ cat ctesti.c #include <stdio.h>
#include <dlfcn.h>

void print_from_gcc(char* txt) {
  printf("%s\n",txt);
}

int main(int argc, char** argv) {
  void * hdl = dlopen("./ftesti.so", RTLD_LAZY);
  void (*print_from_gfortran)(char *) = dlsym(hdl, "print_from_gfortran_");

  print_from_gcc     ("c");
  (*print_from_gfortran)("f");
  print_from_gcc     ("c");
  (*print_from_gfortran)("f");
  dlclose (hdl);

  return 0;
}
$ gcc -o ctesti.exe ctesti.c -ldl
$ ./ctesti.exe 
c
 f
c
 f

And this even works when libgfortran is only available as a static library (as
is the case on mingw32). I'll try to look deeper into this, but it looks very
strange...


-- 

fxcoudert at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |fxcoudert at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug libfortran/27046] [mingw32] gfortran print flush in dll

[fxcoudert at gcc dot gnu dot org]

------- Comment #4 from fxcoudert at gcc dot gnu dot org  2006-05-21 14:03 -------
Confirmed. I don't understand why, though. I'll try it harder when I have some
leisure time.


-- 

fxcoudert at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2006-05-21 14:03:39
               date|                            |
            Summary|gfortran print flush in dll |[mingw32] gfortran print
                   |                            |flush in dll


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug libfortran/27046] [mingw32] mixed C-Fortran I/O doesn't flush

[fxcoudert at gcc dot gnu dot org]

------- Comment #5 from fxcoudert at gcc dot gnu dot org  2006-06-12 16:30 -------
This is not DLL-related, the following code doesn't have the expected behaviour
(although it works fine on i686-linux, even in the static case):

$ cat ctesti.c 
#include <stdio.h>

void print_from_gcc(char* txt) {
  printf("%s\n",txt);
}

int main(int argc, char** argv) {
  print_from_gcc     ("c");
  print_from_gfortran_("f");
  print_from_gcc     ("c");
  print_from_gfortran_("f");

  return 0;
}
$ cat ftesti.f90 
subroutine print_from_gfortran(txt)
  implicit none
  character :: txt
  print *,txt
end subroutine print_from_gfortran
$ ../gfortran/bin/gfortran.exe ftesti.f90 ctesti.c
$ ./a.exe 
c
c
 f
 f


-- 

fxcoudert at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|2006-05-21 14:03:39         |2006-06-12 16:30:13
               date|                            |
            Summary|[mingw32] gfortran print    |[mingw32] mixed C-Fortran
                   |flush in dll                |I/O doesn't flush


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug libfortran/27046] [mingw32] mixed C-Fortran I/O doesn't flush

[dannysmith at users dot sourceforge dot net]

------- Comment #6 from dannysmith at users dot sourceforge dot net  2006-09-13 03:59 -------
(In reply to comment #5)
> This is not DLL-related, the following code doesn't have the expected behaviour
> (although it works fine on i686-linux, even in the static case):


With gcc version 4.2.0 20060911, the test case in Comment #5 works as expected
if the app is run from a cmd.exe shell or from a bash shell.

When the output is redirected to a file, I get the results you report, but with
redirection, one might expect that (on windows at least) in absence of flush
call.

Danny


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug libfortran/27046] [mingw32] mixed C-Fortran I/O doesn't flush

[dannysmith at users dot sourceforge dot net]

------- Comment #7 from dannysmith at users dot sourceforge dot net  2006-09-13 10:10 -------
(In reply to comment #5)
> This is not DLL-related, the following code doesn't have the expected behaviour
> (although it works fine on i686-linux, even in the static case):
> $ cat ctesti.c 
> #include <stdio.h>
> void print_from_gcc(char* txt) {
>   printf("%s\n",txt);
> }
> int main(int argc, char** argv) {
>   print_from_gcc     ("c");
>   print_from_gfortran_("f");
>   print_from_gcc     ("c");
>   print_from_gfortran_("f");
>   return 0;
> }


Changing main() in ctesti.c  to start with:
int main(int argc, char** argv) {
  setvbuf(stdout, NULL, _IOLBF, 0);

fixes the redirection problem.

If you stll think that this is a libgfortran bug (I don't)
you could add 
  setvbuf(stdout, NULL, _IOLBF, 0);
to unix.c:output_stream() so that stdout always is line-buffered even when
!isatty(fileno(stdout)) 

Danny



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug libfortran/27046] [mingw32] mixed C-Fortran I/O doesn't flush

[fxcoudert at gcc dot gnu dot org]

------- Comment #8 from fxcoudert at gcc dot gnu dot org  2006-09-20 07:50 -------
(In reply to comment #7)
> If you stll think that this is a libgfortran bug (I don't)
> you could add 
>   setvbuf(stdout, NULL, _IOLBF, 0);
> to unix.c:output_stream() so that stdout always is line-buffered even when
> !isatty(fileno(stdout)) 

Indeed, I think the bug is in the user's code, but I think there's no serious
drawback to the setvbuf workaround you mention.


-- 

fxcoudert at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|2006-06-12 16:30:13         |2006-09-20 07:50:15
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug libfortran/27046] [mingw32] mixed C-Fortran I/O doesn't flush

[fxcoudert at gcc dot gnu dot org]

-- 

fxcoudert at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |fxcoudert at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2006-09-20 07:50:15         |2006-09-28 14:08:26
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046

[Bug libfortran/27046] [mingw32] mixed C-Fortran I/O doesn't flush

[fxcoudert at gcc dot gnu dot org]

------- Comment #9 from fxcoudert at gcc dot gnu dot org  2006-09-28 22:34 -------
(In reply to comment #7)
> Changing main() in ctesti.c  to start with:
> int main(int argc, char** argv) {
>   setvbuf(stdout, NULL, _IOLBF, 0);
> 
> fixes the redirection problem.

After some more testing, this fix doesn't work in all cases (e.g. in a shell of
MSYS window). I'm closing this PR: after all, this is not a bug in the
compiler, mixed-languages code is supposed to correctly flush both I/O systems
after operations. I would have liked to provide a workaround, only there
doesn't appear to be anythink general and robust enough.


-- 

fxcoudert at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27046