gfortran says to submit bug report

gfortran says to submit bug report

[Joe Koski]

While attempting to compile the NIST FDS4 code (f90 based) I get the
following error

pcp05454784pcs:~/Codes/Fortran_Compiles/fds_test_compile jakoski$ make
gfortran -c -O3 -ffixed-form mods.f
mods.f:628: internal compiler error: backend decl for module variable
y_f_inlet already exists
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
make: *** [mods.o] Error 1

No "preprocessed source" appeared when I added -save-temps to the compile
line.

The compiler info is

pcp05454784pcs:~/Codes/Fortran_Compiles/fds_test_compile jakoski$ gfortran
-v
Reading specs from /usr/local/lib/gcc/powerpc-apple-darwin7.5.0/4.0.0/specs
Configured with: ../gcc/configure --enable-threads=posix
--enable-languages=f95 --enable-static --disable-shared
Thread model: posix
gcc version 4.0.0 20041009 (experimental)

I'm running on a dual processor G5 Mac under OS X 10.3.5 with Xcode 1.5
installed.

The source code and make file (for xlf Fortran) are available as downloads
from

http://fire.nist.gov/fds/

Hope this helps your efforts. Let me know if you need more info.

Joe Koski
jkoski11@comcast.net

Re: gfortran says to submit bug report

[Andrew Pinski]

On Oct 11, 2004, at 12:38 PM, Joe Koski wrote:

> backend decl for module variable


This is most likely the same as PR 17917.

Thanks,
Andrew Pinski

Re: gfortran says to submit bug report

[Joe Koski]

Andrew,

After reading 17917, I agree. Thanks for your reply.

Joe Koski


on 10/11/04 11:03 AM, Andrew Pinski at pinskia@physics.uc.edu wrote:

> 
> On Oct 11, 2004, at 12:38 PM, Joe Koski wrote:
> 
>> backend decl for module variable
> 
> 
> This is most likely the same as PR 17917.
> 
> Thanks,
> Andrew Pinski
> 

FW: gfortran says to submit bug report

[Joe Koski]

[-- Attachment #1: Type: text/plain, Size: 1621 bytes --]

While attempting to compile the NIST FDS4 code (f90 based) I get the
following error and a request to submit a bug report:

jakoski$ gfortran -c -O3 -ffixed-form -fno-underscoring mods.f
mods.f: In function 'lpcoef':
mods.f:5099: error: unrecognizable insn:
(insn 613 612 614 22 (set (reg:DF 729)
        (subreg:DF (mem/s:DC (plus:SI (reg/f:SI 114 virtual-stack-vars)
                    (const_int 56112 [0xdb30])) [1413 equiv.1.cm+32 S16
A64]) 8)) -1 (nil)
    (nil))
mods.f:5099: internal compiler error: in extract_insn, at recog.c:2082
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.

The compiler that I am using (and can get) is:

Joe-Koskis-Computer:~/Fortran_Compiles/fds_test_compile_gfortran/compiler
error jakoski$ gfortran -v
Using built-in specs.
Target: powerpc-apple-darwin7.9.0
Configured with: ../gcc/configure --enable-languages=c++,f95
Thread model: posix
gcc version 4.1.0 20050517 (experimental)

I'm running on a dual processor G5 Mac under OS X 10.3.9 with Xcode 1.5
installed. A tar of the mods.f file is attached, and the error can be
reproduced by entering

  gfortran -c -O3 -ffixed-form -fno-underscoring mods.f

The source code and make file (for xlf Fortran) are available as downloads
from

http://fire.nist.gov/fds/

Hope this helps your efforts. Let me know if you need more info. If this is
fixed in CVS, just let me know and I'll try gfortran again in a few months.

Sorry for e-mailing, but I tried bugzilla, and find it totally confusing,
especially the "triplets."

Joe Koski
jkoski11@comcast.net


[-- Attachment #2: mods_f.tar.gz --]
[-- Type: application/octet-stream, Size: 53938 bytes --]

Re: gfortran says to submit bug report

[Andrew Pinski]

On Sep 28, 2005, at 4:53 PM, Joe Koski wrote:

> While attempting to compile the NIST FDS4 code (f90 based) I get the
> following error and a request to submit a bug report:
>
> jakoski$ gfortran -c -O3 -ffixed-form -fno-underscoring mods.f
> mods.f: In function 'lpcoef':
> mods.f:5099: error: unrecognizable insn:
> (insn 613 612 614 22 (set (reg:DF 729)
>         (subreg:DF (mem/s:DC (plus:SI (reg/f:SI 114 virtual-stack-vars)
>                     (const_int 56112 [0xdb30])) [1413 equiv.1.cm+32 S16
> A64]) 8)) -1 (nil)
>     (nil))
> mods.f:5099: internal compiler error: in extract_insn, at recog.c:2082
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <URL:http://gcc.gnu.org/bugs.html> for instructions.
>
> The compiler that I am using (and can get) is:
>
> Joe-Koskis-Computer:~/Fortran_Compiles/fds_test_compile_gfortran/ 
> compiler
> error jakoski$ gfortran -v
> Using built-in specs.
> Target: powerpc-apple-darwin7.9.0
> Configured with: ../gcc/configure --enable-languages=c++,f95
> Thread model: posix
> gcc version 4.1.0 20050517 (experimental)


Can you try a new compiler as the one which you are using is more
than 4 months old and I think this has already been fixed.

Thanks,
Andrew Pinski