This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: g77 bug (SAVE "skiped")


> I ran your testcase on a variety of systems, including Linux (Red Hat
> (...)
O.K. You asked for it.
-------------------------------------------------------------------
(... everything recompiled with "-g" ...)
[... test1]$ ./test1.csh 
 start HET ... 
Segmentation fault (core dumped)
 (...)
gdb ../../bin/Linux/het.x core
GNU gdb 4.18
 (...)
Program terminated with signal 11, Segmentation fault.
Reading symbols from /lib/libm.so.6...done.
Reading symbols from /lib/libc.so.6...done.
Reading symbols from /lib/ld-linux.so.2...done.
#0  0x80a09ff in selj1_ (xtil=0x81a35cc) at elas.f:46
46         52 SUMA(I) = SUMA(I-1) +A(I)
(gdb) l
41            XM = X(I) + (DELTAX/2.)
42            A(I) = FX(XM)
43         50 X(I+1) = X(I) + DELTAX
44            SUMA(1) = A(1)
45            DO 52 I=2,ISIZE
46         52 SUMA(I) = SUMA(I-1) +A(I)
47      C      NORMALIZE TO 1.
48            DIV = SUMA(ISIZE)
49            DO 75 I=1,ISIZE
50         75 SUMA(I) = SUMA(I)/DIV
(gdb) print ifirst
$1 = 12345
(gdb) print isize
$2 = 1107827787
(gdb) print deltax
$3 = 36.0940666
-------------------------------------------------------------------
As I have already written there is no place in the subroutine where these
values (IFIRST, ISIZE, DELTAX) are modified after the first
initialization. Values of A,SUMA,X seem to be reasonable.
> (...)
> While you're at it, please provide version and platform information.
As written in my first mail. The system is RH6.1/i386 with it's default
compiler g77 version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)
(from FSF-g77 version 0.5.24-19981002).
Best regards,
Jacek.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]