Re: fortran/4398: float c subroutine returns bad value to FORTRAN main

Re: fortran/4398: float c subroutine returns bad value to FORTRAN main

[toon]

Synopsis: float c subroutine returns bad value to FORTRAN main

State-Changed-From-To: feedback->closed
State-Changed-By: toon
State-Changed-When: Mon Oct 29 23:38:28 2001
State-Changed-Why:
    Feedback convinced me to reopen this as a target/SPARC
    Problem Report, number 4731.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=4398&database=gcc

Re: fortran/4398: float c subroutine returns bad value to FORTRAN main

[toon]

Synopsis: float c subroutine returns bad value to FORTRAN main

Responsible-Changed-From-To: unassigned->toon
Responsible-Changed-By: toon
Responsible-Changed-When: Thu Sep 27 00:53:13 2001
Responsible-Changed-Why:
    My job
State-Changed-From-To: open->feedback
State-Changed-By: toon
State-Changed-When: Thu Sep 27 00:53:13 2001
State-Changed-Why:
    Feedback requested on warning messages and exact description
    of wrong output.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=4398&database=gcc

Re: fortran/4398: float c subroutine returns bad value to FORTRAN main

[Toon Moene]

The following reply was made to PR fortran/4398; it has been noted by GNATS.

From: Toon Moene <toon@moene.indiv.nluug.nl>
To: Sabba.Hillel@verizon.net
Cc: gcc-gnats@gcc.gnu.org
Subject: Re: fortran/4398: float c subroutine returns bad value to FORTRAN main
Date: Wed, 26 Sep 2001 21:18:42 +0200

 Sabba.Hillel@verizon.net wrote:
 
 > Solaris 5.6, 5.7, 5.8
 > compiled with g77-3.0.1 and gcc-3.0.1
 > used -fno-second-underscore
 
 > When compiling a FORTRAN main program calling a C subroutine which returns a floating point value,
 > the result printed from the FORTRAN main routin is incorrect.
 > When compiling using Solaris Workshop 6, f77, the result is correct.
 > When using double returns, the results are correct.
 > 
 > I am unable to connect the system with the gcc to the net so I am typing in the test program
 > 
 > test-gcc.f
 > 
 >       PROGRAM TEST_G77
 > 
 >       REAL*4 PRINT_A
 >       REAL*4 TEST_GCC
 >       REAL*4 S, V, T
 >       REAL*8 PRINT_DA
 >       REAL*8 TEST_DGCC
 >       REAL*8 DS, DV, DT
 > 
 >       S = TEST_GCC(T)
 >       WRITE (*, '(A, F16.5, 3X, Z16)') "S = ", S, S
 >       WRITE (*, '(A, F16.5, 3X, Z16)') "T = ", T, T
 >       V = PRINT_A(S)
 >       V = PRINT_A(T)
 > 
 >       DS = TEST_DGCC(DT)
 >       WRITE (*, '(A, F16.5, 3X, Z16)') "DS = ", DS, DS
 >       WRITE (*, '(A, F16.5, 3X, Z16)') "DT = ", DT, DT
 >       DV = PRINT_A(DS)
 >       DV = PRINT_A(DT)
 >       END
 
 Hmmm, exactly which values turn up wrong ?  You might want to study the
 warning messages g77 issues:
 
 testf.f: In program `test_g77':
 testf.f:13: warning:
          V = PRINT_A(S)
              1
 testf.f:19: (continued):
          DV = PRINT_A(DS)
               2
 Argument #1 of `print_a' is one precision at (2) but is some other
 precision at (1) [info -f g77 M GLOBALS]
 testf.f:13: warning:
          V = PRINT_A(S)
              1
 testf.f:20: (continued):
          DV = PRINT_A(DT)
               2
 Argument #1 of `print_a' is one precision at (2) but is some other
 precision at (1) [info -f g77 M GLOBALS]
 
 -- 
 Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
 Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
 Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
 Join GNU Fortran 95: http://g95.sourceforge.net/ (under construction)

fortran/4398: float c subroutine returns bad value to FORTRAN main

[Sabba.Hillel]

>Number:         4398
>Category:       fortran
>Synopsis:       float c subroutine returns bad value to FORTRAN main
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Sep 25 15:46:01 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     Sabba.Hillel@verizon.net
>Release:        gcc-3.0.1 g77-3.0.1
>Organization:
>Environment:
Solaris 5.6, 5.7, 5.8
compiled with g77-3.0.1 and gcc-3.0.1

used -fno-second-underscore
>Description:
When compiling a FORTRAN main program calling a C subroutine which returns a floating point value,
the result printed from the FORTRAN main routin is incorrect.
When compiling using Solaris Workshop 6, f77, the result is correct.
When using double returns, the results are correct.

I am unable to connect the system with the gcc to the net so I am typing in the test program


test-gcc.f

      PROGRAM TEST_G77

      REAL*4 PRINT_A
      REAL*4 TEST_GCC
      REAL*4 S, V, T
      REAL*8 PRINT_DA
      REAL*8 TEST_DGCC
      REAL*8 DS, DV, DT

      S = TEST_GCC(T)
      WRITE (*, '(A, F16.5, 3X, Z16)') "S = ", S, S
      WRITE (*, '(A, F16.5, 3X, Z16)') "T = ", T, T
      V = PRINT_A(S)
      V = PRINT_A(T)

      DS = TEST_DGCC(DT)
      WRITE (*, '(A, F16.5, 3X, Z16)') "DS = ", DS, DS
      WRITE (*, '(A, F16.5, 3X, Z16)') "DT = ", DT, DT
      DV = PRINT_A(DS)
      DV = PRINT_A(DT)
      END



C language functions test-gcc.c


float test_gcc_(float *b)
{
  float a;

  *b = 10.0;
  a = *b;
  printf("a=%f (%#x)\n", a, a);
  printf("b=%f (%#x)\n", b, b);
  return(a);
}

float print_a_(float *a)
{
  printf("A = %f (%#x)\n", *a, *a);
  return (0);
}

double test_dgcc_(double *db)
{
  double da;

  *db = 10.0;
  da = *db;
  printf("da=%f (%#x)\n", da, da);
  printf("db=%f (%#x)\n", db, db);
  return(da);
}

double print_da_(float *da)
{
  printf("DA = %f (%#x)\n", *da, *da);
  return (0);
}


-- 
Said the fox to the fish, "Join me ashore."
The fish are the Jews, Torah is our water.

Hillel (Sabba) Markowitz - sabbahem@bcpl.net
>How-To-Repeat:
See test program in description
>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted: