[PATCH,fortran] *The* ISO C Binding patch
Andreas Schwab
schwab@suse.de
Sun Jun 24 09:13:00 GMT 2007
Steve Kargl <sgk@troutmask.apl.washington.edu> writes:
> I am interested in technical reviews and bootstrap testing on as many
> targets as possible.
ia64:
configure:10291: /tmp/cvs/gcc-20070624/Build/./gcc/gfortran -B/tmp/cvs/gcc-20070624/Build/./gcc/ -B/tmp/cvs/gcc-20070624/Build/root/ia64-suse-linux/bin/ -B/tmp/cvs/gcc-20070624/Build/root/ia64-suse-linux/lib/ -isystem /tmp/cvs/gcc-20070624/Build/root/ia64-suse-linux/include -isystem /tmp/cvs/gcc-20070624/Build/root/ia64-suse-linux/sys-include -c conftest.F >&5
conftest.F:0: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
configure:10297: $? = 1
configure: failed program was:
| program main
| #ifndef __GNUC__
| choke me
| #endif
|
| end
Program received signal SIGSEGV, Segmentation fault.
0x4000000000199520 in gfc_type_for_mode (mode=RCmode, unsignedp=0)
at ../../gcc/fortran/trans-types.c:2040
2040 if (type && mode == TYPE_MODE (type))
(gdb) p type
$1 = (tree) 0x1ffffff9c
(gdb) p gfc_complex_types
$2 = {0x2000000000479f80, 0x200000000047a040, 0x200000000047a100, 0x0, 0x0,
0x1ffffff9c}
Hardware watchpoint 3: gfc_complex_types[5]
Old value = (tree) 0x0
New value = (tree) 0xffffff9c
gfc_init_kinds () at ../../gcc/fortran/trans-types.c:214
214 c_interop_kinds_table[i].name[0] = '\0';
(gdb) l
209
210 /* init all pointers in the list to NULL */
211 for (i = ISOCBINDING_INVALID; i < ISOCBINDING_NUMBER; i++)
212 {
213 /* Initialize the name and value fields. */
214 c_interop_kinds_table[i].name[0] = '\0';
215 c_interop_kinds_table[i].value = -100;
216 c_interop_kinds_table[i].f90_type = BT_UNKNOWN;
217 }
218
$ grep ISOCBINDING_INVALID *.h
gfortran.h: ISOCBINDING_INVALID = -1,
Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, MaxfeldstraÃe 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
More information about the Fortran
mailing list