[Met_help] [rt.rap.ucar.edu #70455] History for met v5 compilation aborted for met_stats.cc (code 2)
John Halley Gotway via RT
met_help at ucar.edu
Tue Jan 27 09:33:02 MST 2015
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Dear
I am trying to install metv5 with intel fortran compiler on mac 10.9.5
Maverick using netcdf4 and all other recommended libraries. during the make
process I got the following error
met_stats.cc(2280): error: "rank" is ambiguous
ties = rank(data, data_rank, n);
^
compilation aborted for met_stats.cc (code 2)
make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
make[3]: *** [all-recursive] Error 1
make[2]: *** [all-recursive] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2
the configuration log file "met.log" is attached for your reference.
Thanks
--
Regards,
Shahzad Sultan
PhD student
Chinese University of Hong Kong
+852 56420557
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: John Halley Gotway
Time: Wed Jan 21 11:00:12 2015
Shazad,
This is a new compilation error message we haven't encountered before.
The
function "rank" is defined in the internal MET library named
"vx_math".
The compilation error says that rank is ambiguous... Perhaps "rank" is
also
defined somewhere else with a different signature.
I notice that you're compiling MET using the Intel compilers. And I
see in
met.log, that you're using the icpc compiler. In all of our previous
compilation and testing of MET with Intel, we've used the "icc"
compiler.
That may, or may not, be the source of the problem. Can you please
try
re-running the configure script and recompiling MET using "icc"
instead?
Try this:
make clean
setenv CXX `which icc`
setenv CC `which icc`
setenv F77 `which ifort`
./configure (using whatever options you'd like)
make install >& make_met.log
Please let me know how that goes. If that doesn't fix the problem,
please
send me the files "make_met.log" and "config.log".
Thanks,
John Halley Gotway
met_help at ucar.edu
On Wed, Jan 21, 2015 at 7:09 AM, shahzad.sultan at gmail.com via RT <
met_help at ucar.edu> wrote:
>
> Wed Jan 21 07:09:47 2015: Request 70455 was acted upon.
> Transaction: Ticket created by shahzad.sultan at gmail.com
> Queue: met_help
> Subject: met v5 compilation aborted for met_stats.cc (code 2)
> Owner: Nobody
> Requestors: shahzad.sultan at gmail.com
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
>
>
> Dear
> I am trying to install metv5 with intel fortran compiler on mac
10.9.5
> Maverick using netcdf4 and all other recommended libraries. during
the make
> process I got the following error
>
> met_stats.cc(2280): error: "rank" is ambiguous
>
> ties = rank(data, data_rank, n);
>
> ^
>
>
> compilation aborted for met_stats.cc (code 2)
>
> make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
>
> make[3]: *** [all-recursive] Error 1
>
> make[2]: *** [all-recursive] Error 1
>
> make[1]: *** [all-recursive] Error 1
>
> make: *** [all] Error 2
>
> the configuration log file "met.log" is attached for your reference.
>
> Thanks
> --
> Regards,
> Shahzad Sultan
> PhD student
> Chinese University of Hong Kong
> +852 56420557
>
>
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: shahzad.sultan at gmail.com
Time: Wed Jan 21 21:43:10 2015
Dear John
Thanks for your reply. I follow the directions you have said and got
little
different error, I think I should try some previous version of met?.
the
log file is attached.
Thanks again
icc -DHAVE_CONFIG_H -I. -I../../.. -I../../../src/basic/vx_cal
-I../../../src/$
num_array.cc(574): error: "rank" is ambiguous
ties = rank(data, data_rank, n);
^
compilation aborted for num_array.cc (code 2)
make[3]: *** [libvx_util_a-num_array.o] Error 2
make[2]: *** [install-recursive] Error 1
make[1]: *** [install-recursive] Error 1
make: *** [install-recursive] Error 1
On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:
> Shazad,
>
> This is a new compilation error message we haven't encountered
before. The
> function "rank" is defined in the internal MET library named
"vx_math".
> The compilation error says that rank is ambiguous... Perhaps "rank"
is also
> defined somewhere else with a different signature.
>
> I notice that you're compiling MET using the Intel compilers. And I
see in
> met.log, that you're using the icpc compiler. In all of our
previous
> compilation and testing of MET with Intel, we've used the "icc"
compiler.
> That may, or may not, be the source of the problem. Can you please
try
> re-running the configure script and recompiling MET using "icc"
instead?
>
> Try this:
> make clean
> setenv CXX `which icc`
> setenv CC `which icc`
> setenv F77 `which ifort`
> ./configure (using whatever options you'd like)
> make install >& make_met.log
>
> Please let me know how that goes. If that doesn't fix the problem,
please
> send me the files "make_met.log" and "config.log".
>
> Thanks,
> John Halley Gotway
> met_help at ucar.edu
>
>
> On Wed, Jan 21, 2015 at 7:09 AM, shahzad.sultan at gmail.com via RT <
> met_help at ucar.edu> wrote:
>
> >
> > Wed Jan 21 07:09:47 2015: Request 70455 was acted upon.
> > Transaction: Ticket created by shahzad.sultan at gmail.com
> > Queue: met_help
> > Subject: met v5 compilation aborted for met_stats.cc (code 2)
> > Owner: Nobody
> > Requestors: shahzad.sultan at gmail.com
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
> >
> >
> > Dear
> > I am trying to install metv5 with intel fortran compiler on mac
10.9.5
> > Maverick using netcdf4 and all other recommended libraries. during
the
> make
> > process I got the following error
> >
> > met_stats.cc(2280): error: "rank" is ambiguous
> >
> > ties = rank(data, data_rank, n);
> >
> > ^
> >
> >
> > compilation aborted for met_stats.cc (code 2)
> >
> > make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
> >
> > make[3]: *** [all-recursive] Error 1
> >
> > make[2]: *** [all-recursive] Error 1
> >
> > make[1]: *** [all-recursive] Error 1
> >
> > make: *** [all] Error 2
> >
> > the configuration log file "met.log" is attached for your
reference.
> >
> > Thanks
> > --
> > Regards,
> > Shahzad Sultan
> > PhD student
> > Chinese University of Hong Kong
> > +852 56420557
> >
> >
>
>
--
Regards,
Shahzad Sultan
PhD student
Chinese University of Hong Kong
+852 56420557
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: John Halley Gotway
Time: Thu Jan 22 09:40:20 2015
Shazad,
I think we should be able to find a solution for this one way of the
other. And I doubt using a previous version of MET would provide
better
success. Could you please send me the file "met-5.0/config.log"?
That'll
tell me the version of the compilers you're using and will help me
investigate the issue further.
Thanks,
John
On Wed, Jan 21, 2015 at 9:43 PM, shahzad.sultan at gmail.com via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
>
> Dear John
> Thanks for your reply. I follow the directions you have said and got
little
> different error, I think I should try some previous version of met?.
the
> log file is attached.
>
> Thanks again
>
>
> icc -DHAVE_CONFIG_H -I. -I../../.. -I../../../src/basic/vx_cal
> -I../../../src/$
>
> num_array.cc(574): error: "rank" is ambiguous
>
> ties = rank(data, data_rank, n);
>
> ^
>
>
> compilation aborted for num_array.cc (code 2)
>
> make[3]: *** [libvx_util_a-num_array.o] Error 2
>
> make[2]: *** [install-recursive] Error 1
>
> make[1]: *** [install-recursive] Error 1
>
> make: *** [install-recursive] Error 1
>
> On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
> > Shazad,
> >
> > This is a new compilation error message we haven't encountered
before.
> The
> > function "rank" is defined in the internal MET library named
"vx_math".
> > The compilation error says that rank is ambiguous... Perhaps
"rank" is
> also
> > defined somewhere else with a different signature.
> >
> > I notice that you're compiling MET using the Intel compilers. And
I see
> in
> > met.log, that you're using the icpc compiler. In all of our
previous
> > compilation and testing of MET with Intel, we've used the "icc"
compiler.
> > That may, or may not, be the source of the problem. Can you
please try
> > re-running the configure script and recompiling MET using "icc"
instead?
> >
> > Try this:
> > make clean
> > setenv CXX `which icc`
> > setenv CC `which icc`
> > setenv F77 `which ifort`
> > ./configure (using whatever options you'd like)
> > make install >& make_met.log
> >
> > Please let me know how that goes. If that doesn't fix the
problem,
> please
> > send me the files "make_met.log" and "config.log".
> >
> > Thanks,
> > John Halley Gotway
> > met_help at ucar.edu
> >
> >
> > On Wed, Jan 21, 2015 at 7:09 AM, shahzad.sultan at gmail.com via RT <
> > met_help at ucar.edu> wrote:
> >
> > >
> > > Wed Jan 21 07:09:47 2015: Request 70455 was acted upon.
> > > Transaction: Ticket created by shahzad.sultan at gmail.com
> > > Queue: met_help
> > > Subject: met v5 compilation aborted for met_stats.cc (code
2)
> > > Owner: Nobody
> > > Requestors: shahzad.sultan at gmail.com
> > > Status: new
> > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
> >
> > >
> > >
> > > Dear
> > > I am trying to install metv5 with intel fortran compiler on mac
10.9.5
> > > Maverick using netcdf4 and all other recommended libraries.
during the
> > make
> > > process I got the following error
> > >
> > > met_stats.cc(2280): error: "rank" is ambiguous
> > >
> > > ties = rank(data, data_rank, n);
> > >
> > > ^
> > >
> > >
> > > compilation aborted for met_stats.cc (code 2)
> > >
> > > make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
> > >
> > > make[3]: *** [all-recursive] Error 1
> > >
> > > make[2]: *** [all-recursive] Error 1
> > >
> > > make[1]: *** [all-recursive] Error 1
> > >
> > > make: *** [all] Error 2
> > >
> > > the configuration log file "met.log" is attached for your
reference.
> > >
> > > Thanks
> > > --
> > > Regards,
> > > Shahzad Sultan
> > > PhD student
> > > Chinese University of Hong Kong
> > > +852 56420557
> > >
> > >
> >
> >
>
>
> --
> Regards,
> Shahzad Sultan
> PhD student
> Chinese University of Hong Kong
> +852 56420557
>
>
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: shahzad.sultan at gmail.com
Time: Thu Jan 22 17:11:18 2015
Thanks for your reply
I have also tried the MET v4.1 but error remain same. So I guess the
issue
is either dependency libraries like netcdf. I have also tried to
disable
many module but still error remain same.
The config.log is attached for your reference for METv5.0.
On Fri, Jan 23, 2015 at 12:40 AM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:
> Shazad,
>
> I think we should be able to find a solution for this one way of the
> other. And I doubt using a previous version of MET would provide
better
> success. Could you please send me the file "met-5.0/config.log"?
That'll
> tell me the version of the compilers you're using and will help me
> investigate the issue further.
>
> Thanks,
> John
>
> On Wed, Jan 21, 2015 at 9:43 PM, shahzad.sultan at gmail.com via RT <
> met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
> >
> > Dear John
> > Thanks for your reply. I follow the directions you have said and
got
> little
> > different error, I think I should try some previous version of
met?. the
> > log file is attached.
> >
> > Thanks again
> >
> >
> > icc -DHAVE_CONFIG_H -I. -I../../.. -I../../../src/basic/vx_cal
> > -I../../../src/$
> >
> > num_array.cc(574): error: "rank" is ambiguous
> >
> > ties = rank(data, data_rank, n);
> >
> > ^
> >
> >
> > compilation aborted for num_array.cc (code 2)
> >
> > make[3]: *** [libvx_util_a-num_array.o] Error 2
> >
> > make[2]: *** [install-recursive] Error 1
> >
> > make[1]: *** [install-recursive] Error 1
> >
> > make: *** [install-recursive] Error 1
> >
> > On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via RT <
> > met_help at ucar.edu> wrote:
> >
> > > Shazad,
> > >
> > > This is a new compilation error message we haven't encountered
before.
> > The
> > > function "rank" is defined in the internal MET library named
"vx_math".
> > > The compilation error says that rank is ambiguous... Perhaps
"rank" is
> > also
> > > defined somewhere else with a different signature.
> > >
> > > I notice that you're compiling MET using the Intel compilers.
And I
> see
> > in
> > > met.log, that you're using the icpc compiler. In all of our
previous
> > > compilation and testing of MET with Intel, we've used the "icc"
> compiler.
> > > That may, or may not, be the source of the problem. Can you
please try
> > > re-running the configure script and recompiling MET using "icc"
> instead?
> > >
> > > Try this:
> > > make clean
> > > setenv CXX `which icc`
> > > setenv CC `which icc`
> > > setenv F77 `which ifort`
> > > ./configure (using whatever options you'd like)
> > > make install >& make_met.log
> > >
> > > Please let me know how that goes. If that doesn't fix the
problem,
> > please
> > > send me the files "make_met.log" and "config.log".
> > >
> > > Thanks,
> > > John Halley Gotway
> > > met_help at ucar.edu
> > >
> > >
> > > On Wed, Jan 21, 2015 at 7:09 AM, shahzad.sultan at gmail.com via RT
<
> > > met_help at ucar.edu> wrote:
> > >
> > > >
> > > > Wed Jan 21 07:09:47 2015: Request 70455 was acted upon.
> > > > Transaction: Ticket created by shahzad.sultan at gmail.com
> > > > Queue: met_help
> > > > Subject: met v5 compilation aborted for met_stats.cc
(code 2)
> > > > Owner: Nobody
> > > > Requestors: shahzad.sultan at gmail.com
> > > > Status: new
> > > > Ticket <URL:
> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
> > >
> > > >
> > > >
> > > > Dear
> > > > I am trying to install metv5 with intel fortran compiler on
mac
> 10.9.5
> > > > Maverick using netcdf4 and all other recommended libraries.
during
> the
> > > make
> > > > process I got the following error
> > > >
> > > > met_stats.cc(2280): error: "rank" is ambiguous
> > > >
> > > > ties = rank(data, data_rank, n);
> > > >
> > > > ^
> > > >
> > > >
> > > > compilation aborted for met_stats.cc (code 2)
> > > >
> > > > make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
> > > >
> > > > make[3]: *** [all-recursive] Error 1
> > > >
> > > > make[2]: *** [all-recursive] Error 1
> > > >
> > > > make[1]: *** [all-recursive] Error 1
> > > >
> > > > make: *** [all] Error 2
> > > >
> > > > the configuration log file "met.log" is attached for your
reference.
> > > >
> > > > Thanks
> > > > --
> > > > Regards,
> > > > Shahzad Sultan
> > > > PhD student
> > > > Chinese University of Hong Kong
> > > > +852 56420557
> > > >
> > > >
> > >
> > >
> >
> >
> > --
> > Regards,
> > Shahzad Sultan
> > PhD student
> > Chinese University of Hong Kong
> > +852 56420557
> >
> >
>
>
--
Regards,
Shahzad Sultan
PhD student
Chinese University of Hong Kong
+852 56420557
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: John Halley Gotway
Time: Fri Jan 23 10:58:27 2015
Shazad,
I see that you're using Intel compiler version 14. I was able to
locate
that version of the Intel compiler and test with it. Unfortunately,
I'm
not able to replicate the compilation error you're seeing. I'm
testing on
Linux, not on a Mac.
So let's try this next...
I believe the problem is a conflict in the name of the "rank()"
function.
I suspect that's already been defined in a slightly different way on
your
system. I found two versions of the "rank()" function in the MET code
tree. I modified the code to rename one of them "do_rank()" and the
other
"get_rank()".
I was able to compile and run the modified version of the code on my
machine. So hopefully, I made the changes in all the right spots.
Please download and attempt to compile this version of met-5.0:
ftp://ftp.rap.ucar.edu/incoming/irap/met_help/sultan_data/met-
5.0_shazad.sultan.tar.gz
And please let me know if that solves the compilation problem. If so,
I'll
make these changes in the MET code repository and post them as an
official
bugfix for met-5.0.
Thanks,
John
On Thu, Jan 22, 2015 at 5:11 PM, shahzad.sultan at gmail.com via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
>
> Thanks for your reply
>
> I have also tried the MET v4.1 but error remain same. So I guess the
issue
> is either dependency libraries like netcdf. I have also tried to
disable
> many module but still error remain same.
>
> The config.log is attached for your reference for METv5.0.
>
> On Fri, Jan 23, 2015 at 12:40 AM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
> > Shazad,
> >
> > I think we should be able to find a solution for this one way of
the
> > other. And I doubt using a previous version of MET would provide
better
> > success. Could you please send me the file "met-5.0/config.log"?
> That'll
> > tell me the version of the compilers you're using and will help me
> > investigate the issue further.
> >
> > Thanks,
> > John
> >
> > On Wed, Jan 21, 2015 at 9:43 PM, shahzad.sultan at gmail.com via RT <
> > met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
> > >
> > > Dear John
> > > Thanks for your reply. I follow the directions you have said and
got
> > little
> > > different error, I think I should try some previous version of
met?.
> the
> > > log file is attached.
> > >
> > > Thanks again
> > >
> > >
> > > icc -DHAVE_CONFIG_H -I. -I../../.. -I../../../src/basic/vx_cal
> > > -I../../../src/$
> > >
> > > num_array.cc(574): error: "rank" is ambiguous
> > >
> > > ties = rank(data, data_rank, n);
> > >
> > > ^
> > >
> > >
> > > compilation aborted for num_array.cc (code 2)
> > >
> > > make[3]: *** [libvx_util_a-num_array.o] Error 2
> > >
> > > make[2]: *** [install-recursive] Error 1
> > >
> > > make[1]: *** [install-recursive] Error 1
> > >
> > > make: *** [install-recursive] Error 1
> > >
> > > On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via RT <
> > > met_help at ucar.edu> wrote:
> > >
> > > > Shazad,
> > > >
> > > > This is a new compilation error message we haven't encountered
> before.
> > > The
> > > > function "rank" is defined in the internal MET library named
> "vx_math".
> > > > The compilation error says that rank is ambiguous... Perhaps
"rank"
> is
> > > also
> > > > defined somewhere else with a different signature.
> > > >
> > > > I notice that you're compiling MET using the Intel compilers.
And I
> > see
> > > in
> > > > met.log, that you're using the icpc compiler. In all of our
previous
> > > > compilation and testing of MET with Intel, we've used the
"icc"
> > compiler.
> > > > That may, or may not, be the source of the problem. Can you
please
> try
> > > > re-running the configure script and recompiling MET using
"icc"
> > instead?
> > > >
> > > > Try this:
> > > > make clean
> > > > setenv CXX `which icc`
> > > > setenv CC `which icc`
> > > > setenv F77 `which ifort`
> > > > ./configure (using whatever options you'd like)
> > > > make install >& make_met.log
> > > >
> > > > Please let me know how that goes. If that doesn't fix the
problem,
> > > please
> > > > send me the files "make_met.log" and "config.log".
> > > >
> > > > Thanks,
> > > > John Halley Gotway
> > > > met_help at ucar.edu
> > > >
> > > >
> > > > On Wed, Jan 21, 2015 at 7:09 AM, shahzad.sultan at gmail.com via
RT <
> > > > met_help at ucar.edu> wrote:
> > > >
> > > > >
> > > > > Wed Jan 21 07:09:47 2015: Request 70455 was acted upon.
> > > > > Transaction: Ticket created by shahzad.sultan at gmail.com
> > > > > Queue: met_help
> > > > > Subject: met v5 compilation aborted for met_stats.cc
(code 2)
> > > > > Owner: Nobody
> > > > > Requestors: shahzad.sultan at gmail.com
> > > > > Status: new
> > > > > Ticket <URL:
> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
> > > >
> > > > >
> > > > >
> > > > > Dear
> > > > > I am trying to install metv5 with intel fortran compiler on
mac
> > 10.9.5
> > > > > Maverick using netcdf4 and all other recommended libraries.
during
> > the
> > > > make
> > > > > process I got the following error
> > > > >
> > > > > met_stats.cc(2280): error: "rank" is ambiguous
> > > > >
> > > > > ties = rank(data, data_rank, n);
> > > > >
> > > > > ^
> > > > >
> > > > >
> > > > > compilation aborted for met_stats.cc (code 2)
> > > > >
> > > > > make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
> > > > >
> > > > > make[3]: *** [all-recursive] Error 1
> > > > >
> > > > > make[2]: *** [all-recursive] Error 1
> > > > >
> > > > > make[1]: *** [all-recursive] Error 1
> > > > >
> > > > > make: *** [all] Error 2
> > > > >
> > > > > the configuration log file "met.log" is attached for your
> reference.
> > > > >
> > > > > Thanks
> > > > > --
> > > > > Regards,
> > > > > Shahzad Sultan
> > > > > PhD student
> > > > > Chinese University of Hong Kong
> > > > > +852 56420557
> > > > >
> > > > >
> > > >
> > > >
> > >
> > >
> > > --
> > > Regards,
> > > Shahzad Sultan
> > > PhD student
> > > Chinese University of Hong Kong
> > > +852 56420557
> > >
> > >
> >
> >
>
>
> --
> Regards,
> Shahzad Sultan
> PhD student
> Chinese University of Hong Kong
> +852 56420557
>
>
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: shahzad.sultan at gmail.com
Time: Fri Jan 23 17:35:37 2015
Dear John
Thanks
I am really thankful to you and finally it installed successfully :).
I have used the following configuration
./configure --prefix=/Users/shahzad/wrf/met5/ FC=ifort F9X=ifort
CXX=icc
CC=icc F77=ifort --disable-pb2nc
I also would like to ask that, should I apply the patches also? if so
then
i will do after receiving your reply.
Thanks
Since pb2nc was giving some error during make, therefore I try with
disabling it. the list of the installed tools are as follow
ascii2nc
ensemble_stat
gen_circle_mask
gen_poly_mask
grid_stat
madis2nc
mode
mode_analysis
pcp_combine
plot_data_plane
plot_point_obs
point_stat
series_analysis
stat_analysis
tc_dland
tc_pairs
tc_stat
wavelet_stat
wwmca_plot
wwmca_regrid
On Sat, Jan 24, 2015 at 1:58 AM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:
> Shazad,
>
> I see that you're using Intel compiler version 14. I was able to
locate
> that version of the Intel compiler and test with it. Unfortunately,
I'm
> not able to replicate the compilation error you're seeing. I'm
testing on
> Linux, not on a Mac.
>
> So let's try this next...
>
> I believe the problem is a conflict in the name of the "rank()"
function.
> I suspect that's already been defined in a slightly different way on
your
> system. I found two versions of the "rank()" function in the MET
code
> tree. I modified the code to rename one of them "do_rank()" and the
other
> "get_rank()".
>
> I was able to compile and run the modified version of the code on my
> machine. So hopefully, I made the changes in all the right spots.
>
> Please download and attempt to compile this version of met-5.0:
>
>
> ftp://ftp.rap.ucar.edu/incoming/irap/met_help/sultan_data/met-
5.0_shazad.sultan.tar.gz
>
> And please let me know if that solves the compilation problem. If
so, I'll
> make these changes in the MET code repository and post them as an
official
> bugfix for met-5.0.
>
> Thanks,
> John
>
> On Thu, Jan 22, 2015 at 5:11 PM, shahzad.sultan at gmail.com via RT <
> met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
> >
> > Thanks for your reply
> >
> > I have also tried the MET v4.1 but error remain same. So I guess
the
> issue
> > is either dependency libraries like netcdf. I have also tried to
disable
> > many module but still error remain same.
> >
> > The config.log is attached for your reference for METv5.0.
> >
> > On Fri, Jan 23, 2015 at 12:40 AM, John Halley Gotway via RT <
> > met_help at ucar.edu> wrote:
> >
> > > Shazad,
> > >
> > > I think we should be able to find a solution for this one way of
the
> > > other. And I doubt using a previous version of MET would
provide
> better
> > > success. Could you please send me the file "met-
5.0/config.log"?
> > That'll
> > > tell me the version of the compilers you're using and will help
me
> > > investigate the issue further.
> > >
> > > Thanks,
> > > John
> > >
> > > On Wed, Jan 21, 2015 at 9:43 PM, shahzad.sultan at gmail.com via RT
<
> > > met_help at ucar.edu> wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
>
> > > >
> > > > Dear John
> > > > Thanks for your reply. I follow the directions you have said
and got
> > > little
> > > > different error, I think I should try some previous version of
met?.
> > the
> > > > log file is attached.
> > > >
> > > > Thanks again
> > > >
> > > >
> > > > icc -DHAVE_CONFIG_H -I. -I../../..
-I../../../src/basic/vx_cal
> > > > -I../../../src/$
> > > >
> > > > num_array.cc(574): error: "rank" is ambiguous
> > > >
> > > > ties = rank(data, data_rank, n);
> > > >
> > > > ^
> > > >
> > > >
> > > > compilation aborted for num_array.cc (code 2)
> > > >
> > > > make[3]: *** [libvx_util_a-num_array.o] Error 2
> > > >
> > > > make[2]: *** [install-recursive] Error 1
> > > >
> > > > make[1]: *** [install-recursive] Error 1
> > > >
> > > > make: *** [install-recursive] Error 1
> > > >
> > > > On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via RT <
> > > > met_help at ucar.edu> wrote:
> > > >
> > > > > Shazad,
> > > > >
> > > > > This is a new compilation error message we haven't
encountered
> > before.
> > > > The
> > > > > function "rank" is defined in the internal MET library named
> > "vx_math".
> > > > > The compilation error says that rank is ambiguous... Perhaps
"rank"
> > is
> > > > also
> > > > > defined somewhere else with a different signature.
> > > > >
> > > > > I notice that you're compiling MET using the Intel
compilers. And
> I
> > > see
> > > > in
> > > > > met.log, that you're using the icpc compiler. In all of our
> previous
> > > > > compilation and testing of MET with Intel, we've used the
"icc"
> > > compiler.
> > > > > That may, or may not, be the source of the problem. Can you
please
> > try
> > > > > re-running the configure script and recompiling MET using
"icc"
> > > instead?
> > > > >
> > > > > Try this:
> > > > > make clean
> > > > > setenv CXX `which icc`
> > > > > setenv CC `which icc`
> > > > > setenv F77 `which ifort`
> > > > > ./configure (using whatever options you'd like)
> > > > > make install >& make_met.log
> > > > >
> > > > > Please let me know how that goes. If that doesn't fix the
problem,
> > > > please
> > > > > send me the files "make_met.log" and "config.log".
> > > > >
> > > > > Thanks,
> > > > > John Halley Gotway
> > > > > met_help at ucar.edu
> > > > >
> > > > >
> > > > > On Wed, Jan 21, 2015 at 7:09 AM, shahzad.sultan at gmail.com
via RT <
> > > > > met_help at ucar.edu> wrote:
> > > > >
> > > > > >
> > > > > > Wed Jan 21 07:09:47 2015: Request 70455 was acted upon.
> > > > > > Transaction: Ticket created by shahzad.sultan at gmail.com
> > > > > > Queue: met_help
> > > > > > Subject: met v5 compilation aborted for met_stats.cc
(code
> 2)
> > > > > > Owner: Nobody
> > > > > > Requestors: shahzad.sultan at gmail.com
> > > > > > Status: new
> > > > > > Ticket <URL:
> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
> > > > >
> > > > > >
> > > > > >
> > > > > > Dear
> > > > > > I am trying to install metv5 with intel fortran compiler
on mac
> > > 10.9.5
> > > > > > Maverick using netcdf4 and all other recommended
libraries.
> during
> > > the
> > > > > make
> > > > > > process I got the following error
> > > > > >
> > > > > > met_stats.cc(2280): error: "rank" is ambiguous
> > > > > >
> > > > > > ties = rank(data, data_rank, n);
> > > > > >
> > > > > > ^
> > > > > >
> > > > > >
> > > > > > compilation aborted for met_stats.cc (code 2)
> > > > > >
> > > > > > make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
> > > > > >
> > > > > > make[3]: *** [all-recursive] Error 1
> > > > > >
> > > > > > make[2]: *** [all-recursive] Error 1
> > > > > >
> > > > > > make[1]: *** [all-recursive] Error 1
> > > > > >
> > > > > > make: *** [all] Error 2
> > > > > >
> > > > > > the configuration log file "met.log" is attached for your
> > reference.
> > > > > >
> > > > > > Thanks
> > > > > > --
> > > > > > Regards,
> > > > > > Shahzad Sultan
> > > > > > PhD student
> > > > > > Chinese University of Hong Kong
> > > > > > +852 56420557
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > Regards,
> > > > Shahzad Sultan
> > > > PhD student
> > > > Chinese University of Hong Kong
> > > > +852 56420557
> > > >
> > > >
> > >
> > >
> >
> >
> > --
> > Regards,
> > Shahzad Sultan
> > PhD student
> > Chinese University of Hong Kong
> > +852 56420557
> >
> >
>
>
--
Regards,
Shahzad Sultan
PhD student
Chinese University of Hong Kong
+852 56420557
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: shahzad.sultan at gmail.com
Time: Fri Jan 23 21:07:55 2015
Dear John
I have also tried after applying patches but it does not work and
bring the
same error. I have also tried to remove the *-DBLOCK4 from makefile
but
still got error of pb2nc (library not found for -lburf). of course if
we
disable pb2nc then it works fine.*
*Thanks*
On Sat, Jan 24, 2015 at 8:35 AM, Shahzad Sultan
<shahzad.sultan at gmail.com>
wrote:
> Dear John
> Thanks
> I am really thankful to you and finally it installed successfully
:).
>
> I have used the following configuration
>
> ./configure --prefix=/Users/shahzad/wrf/met5/ FC=ifort F9X=ifort
CXX=icc
> CC=icc F77=ifort --disable-pb2nc
>
> I also would like to ask that, should I apply the patches also? if
so then
> i will do after receiving your reply.
>
>
> Thanks
>
> Since pb2nc was giving some error during make, therefore I try with
> disabling it. the list of the installed tools are as follow
>
> ascii2nc
>
> ensemble_stat
>
> gen_circle_mask
>
> gen_poly_mask
>
> grid_stat
>
> madis2nc
>
> mode
>
> mode_analysis
>
> pcp_combine
>
> plot_data_plane
>
> plot_point_obs
>
> point_stat
>
> series_analysis
>
> stat_analysis
>
> tc_dland
>
> tc_pairs
>
> tc_stat
>
> wavelet_stat
>
> wwmca_plot
>
> wwmca_regrid
>
> On Sat, Jan 24, 2015 at 1:58 AM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
>> Shazad,
>>
>> I see that you're using Intel compiler version 14. I was able to
locate
>> that version of the Intel compiler and test with it.
Unfortunately, I'm
>> not able to replicate the compilation error you're seeing. I'm
testing on
>> Linux, not on a Mac.
>>
>> So let's try this next...
>>
>> I believe the problem is a conflict in the name of the "rank()"
function.
>> I suspect that's already been defined in a slightly different way
on your
>> system. I found two versions of the "rank()" function in the MET
code
>> tree. I modified the code to rename one of them "do_rank()" and
the other
>> "get_rank()".
>>
>> I was able to compile and run the modified version of the code on
my
>> machine. So hopefully, I made the changes in all the right spots.
>>
>> Please download and attempt to compile this version of met-5.0:
>>
>>
>> ftp://ftp.rap.ucar.edu/incoming/irap/met_help/sultan_data/met-
5.0_shazad.sultan.tar.gz
>>
>> And please let me know if that solves the compilation problem. If
so,
>> I'll
>> make these changes in the MET code repository and post them as an
official
>> bugfix for met-5.0.
>>
>> Thanks,
>> John
>>
>> On Thu, Jan 22, 2015 at 5:11 PM, shahzad.sultan at gmail.com via RT <
>> met_help at ucar.edu> wrote:
>>
>> >
>> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
>> >
>> > Thanks for your reply
>> >
>> > I have also tried the MET v4.1 but error remain same. So I guess
the
>> issue
>> > is either dependency libraries like netcdf. I have also tried to
disable
>> > many module but still error remain same.
>> >
>> > The config.log is attached for your reference for METv5.0.
>> >
>> > On Fri, Jan 23, 2015 at 12:40 AM, John Halley Gotway via RT <
>> > met_help at ucar.edu> wrote:
>> >
>> > > Shazad,
>> > >
>> > > I think we should be able to find a solution for this one way
of the
>> > > other. And I doubt using a previous version of MET would
provide
>> better
>> > > success. Could you please send me the file "met-
5.0/config.log"?
>> > That'll
>> > > tell me the version of the compilers you're using and will help
me
>> > > investigate the issue further.
>> > >
>> > > Thanks,
>> > > John
>> > >
>> > > On Wed, Jan 21, 2015 at 9:43 PM, shahzad.sultan at gmail.com via
RT <
>> > > met_help at ucar.edu> wrote:
>> > >
>> > > >
>> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
>
>> > > >
>> > > > Dear John
>> > > > Thanks for your reply. I follow the directions you have said
and got
>> > > little
>> > > > different error, I think I should try some previous version
of met?.
>> > the
>> > > > log file is attached.
>> > > >
>> > > > Thanks again
>> > > >
>> > > >
>> > > > icc -DHAVE_CONFIG_H -I. -I../../..
-I../../../src/basic/vx_cal
>> > > > -I../../../src/$
>> > > >
>> > > > num_array.cc(574): error: "rank" is ambiguous
>> > > >
>> > > > ties = rank(data, data_rank, n);
>> > > >
>> > > > ^
>> > > >
>> > > >
>> > > > compilation aborted for num_array.cc (code 2)
>> > > >
>> > > > make[3]: *** [libvx_util_a-num_array.o] Error 2
>> > > >
>> > > > make[2]: *** [install-recursive] Error 1
>> > > >
>> > > > make[1]: *** [install-recursive] Error 1
>> > > >
>> > > > make: *** [install-recursive] Error 1
>> > > >
>> > > > On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via RT <
>> > > > met_help at ucar.edu> wrote:
>> > > >
>> > > > > Shazad,
>> > > > >
>> > > > > This is a new compilation error message we haven't
encountered
>> > before.
>> > > > The
>> > > > > function "rank" is defined in the internal MET library
named
>> > "vx_math".
>> > > > > The compilation error says that rank is ambiguous...
Perhaps
>> "rank"
>> > is
>> > > > also
>> > > > > defined somewhere else with a different signature.
>> > > > >
>> > > > > I notice that you're compiling MET using the Intel
compilers.
>> And I
>> > > see
>> > > > in
>> > > > > met.log, that you're using the icpc compiler. In all of
our
>> previous
>> > > > > compilation and testing of MET with Intel, we've used the
"icc"
>> > > compiler.
>> > > > > That may, or may not, be the source of the problem. Can
you
>> please
>> > try
>> > > > > re-running the configure script and recompiling MET using
"icc"
>> > > instead?
>> > > > >
>> > > > > Try this:
>> > > > > make clean
>> > > > > setenv CXX `which icc`
>> > > > > setenv CC `which icc`
>> > > > > setenv F77 `which ifort`
>> > > > > ./configure (using whatever options you'd like)
>> > > > > make install >& make_met.log
>> > > > >
>> > > > > Please let me know how that goes. If that doesn't fix the
>> problem,
>> > > > please
>> > > > > send me the files "make_met.log" and "config.log".
>> > > > >
>> > > > > Thanks,
>> > > > > John Halley Gotway
>> > > > > met_help at ucar.edu
>> > > > >
>> > > > >
>> > > > > On Wed, Jan 21, 2015 at 7:09 AM, shahzad.sultan at gmail.com
via RT
>> <
>> > > > > met_help at ucar.edu> wrote:
>> > > > >
>> > > > > >
>> > > > > > Wed Jan 21 07:09:47 2015: Request 70455 was acted upon.
>> > > > > > Transaction: Ticket created by shahzad.sultan at gmail.com
>> > > > > > Queue: met_help
>> > > > > > Subject: met v5 compilation aborted for met_stats.cc
(code
>> 2)
>> > > > > > Owner: Nobody
>> > > > > > Requestors: shahzad.sultan at gmail.com
>> > > > > > Status: new
>> > > > > > Ticket <URL:
>> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
>> > > > >
>> > > > > >
>> > > > > >
>> > > > > > Dear
>> > > > > > I am trying to install metv5 with intel fortran compiler
on mac
>> > > 10.9.5
>> > > > > > Maverick using netcdf4 and all other recommended
libraries.
>> during
>> > > the
>> > > > > make
>> > > > > > process I got the following error
>> > > > > >
>> > > > > > met_stats.cc(2280): error: "rank" is ambiguous
>> > > > > >
>> > > > > > ties = rank(data, data_rank, n);
>> > > > > >
>> > > > > > ^
>> > > > > >
>> > > > > >
>> > > > > > compilation aborted for met_stats.cc (code 2)
>> > > > > >
>> > > > > > make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
>> > > > > >
>> > > > > > make[3]: *** [all-recursive] Error 1
>> > > > > >
>> > > > > > make[2]: *** [all-recursive] Error 1
>> > > > > >
>> > > > > > make[1]: *** [all-recursive] Error 1
>> > > > > >
>> > > > > > make: *** [all] Error 2
>> > > > > >
>> > > > > > the configuration log file "met.log" is attached for your
>> > reference.
>> > > > > >
>> > > > > > Thanks
>> > > > > > --
>> > > > > > Regards,
>> > > > > > Shahzad Sultan
>> > > > > > PhD student
>> > > > > > Chinese University of Hong Kong
>> > > > > > +852 56420557
>> > > > > >
>> > > > > >
>> > > > >
>> > > > >
>> > > >
>> > > >
>> > > > --
>> > > > Regards,
>> > > > Shahzad Sultan
>> > > > PhD student
>> > > > Chinese University of Hong Kong
>> > > > +852 56420557
>> > > >
>> > > >
>> > >
>> > >
>> >
>> >
>> > --
>> > Regards,
>> > Shahzad Sultan
>> > PhD student
>> > Chinese University of Hong Kong
>> > +852 56420557
>> >
>> >
>>
>>
>
>
> --
> Regards,
> Shahzad Sultan
> PhD student
> Chinese University of Hong Kong
> +852 56420557
>
>
--
Regards,
Shahzad Sultan
PhD student
Chinese University of Hong Kong
+852 56420557
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: John Halley Gotway
Time: Sun Jan 25 08:21:44 2015
Shazad,
That build I sent you already did contain the latest set of patches.
Now
that we know the changes work, I'll make those changes in the code
repository on Monday and post an updated set of patches. At that
point
please do try rebuilding using MET version 5 plus the latest set of
patches. I'd like you to test with the publicly posted code.
As for BUFR, please locate the file names libbufr.a on your machine.
Set
the MET_BUFRLIB environment variable to that directory. Then rerun
configure and try compiling again. If that doesn't work, please send
me
config.log and make_met.log.
Thanks
Johm
On Friday, January 23, 2015, shahzad.sultan at gmail.com via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
>
> Dear John
> I have also tried after applying patches but it does not work and
bring the
> same error. I have also tried to remove the *-DBLOCK4 from makefile
but
> still got error of pb2nc (library not found for -lburf). of course
if we
> disable pb2nc then it works fine.*
>
> *Thanks*
>
> On Sat, Jan 24, 2015 at 8:35 AM, Shahzad Sultan
<shahzad.sultan at gmail.com
> <javascript:;>>
> wrote:
>
> > Dear John
> > Thanks
> > I am really thankful to you and finally it installed successfully
:).
> >
> > I have used the following configuration
> >
> > ./configure --prefix=/Users/shahzad/wrf/met5/ FC=ifort F9X=ifort
CXX=icc
> > CC=icc F77=ifort --disable-pb2nc
> >
> > I also would like to ask that, should I apply the patches also? if
so
> then
> > i will do after receiving your reply.
> >
> >
> > Thanks
> >
> > Since pb2nc was giving some error during make, therefore I try
with
> > disabling it. the list of the installed tools are as follow
> >
> > ascii2nc
> >
> > ensemble_stat
> >
> > gen_circle_mask
> >
> > gen_poly_mask
> >
> > grid_stat
> >
> > madis2nc
> >
> > mode
> >
> > mode_analysis
> >
> > pcp_combine
> >
> > plot_data_plane
> >
> > plot_point_obs
> >
> > point_stat
> >
> > series_analysis
> >
> > stat_analysis
> >
> > tc_dland
> >
> > tc_pairs
> >
> > tc_stat
> >
> > wavelet_stat
> >
> > wwmca_plot
> >
> > wwmca_regrid
> >
> > On Sat, Jan 24, 2015 at 1:58 AM, John Halley Gotway via RT <
> > met_help at ucar.edu <javascript:;>> wrote:
> >
> >> Shazad,
> >>
> >> I see that you're using Intel compiler version 14. I was able to
locate
> >> that version of the Intel compiler and test with it.
Unfortunately, I'm
> >> not able to replicate the compilation error you're seeing. I'm
testing
> on
> >> Linux, not on a Mac.
> >>
> >> So let's try this next...
> >>
> >> I believe the problem is a conflict in the name of the "rank()"
> function.
> >> I suspect that's already been defined in a slightly different way
on
> your
> >> system. I found two versions of the "rank()" function in the MET
code
> >> tree. I modified the code to rename one of them "do_rank()" and
the
> other
> >> "get_rank()".
> >>
> >> I was able to compile and run the modified version of the code on
my
> >> machine. So hopefully, I made the changes in all the right
spots.
> >>
> >> Please download and attempt to compile this version of met-5.0:
> >>
> >>
> >>
> ftp://ftp.rap.ucar.edu/incoming/irap/met_help/sultan_data/met-
5.0_shazad.sultan.tar.gz
> >>
> >> And please let me know if that solves the compilation problem.
If so,
> >> I'll
> >> make these changes in the MET code repository and post them as an
> official
> >> bugfix for met-5.0.
> >>
> >> Thanks,
> >> John
> >>
> >> On Thu, Jan 22, 2015 at 5:11 PM, shahzad.sultan at gmail.com
> <javascript:;> via RT <
> >> met_help at ucar.edu <javascript:;>> wrote:
> >>
> >> >
> >> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
> >> >
> >> > Thanks for your reply
> >> >
> >> > I have also tried the MET v4.1 but error remain same. So I
guess the
> >> issue
> >> > is either dependency libraries like netcdf. I have also tried
to
> disable
> >> > many module but still error remain same.
> >> >
> >> > The config.log is attached for your reference for METv5.0.
> >> >
> >> > On Fri, Jan 23, 2015 at 12:40 AM, John Halley Gotway via RT <
> >> > met_help at ucar.edu <javascript:;>> wrote:
> >> >
> >> > > Shazad,
> >> > >
> >> > > I think we should be able to find a solution for this one way
of the
> >> > > other. And I doubt using a previous version of MET would
provide
> >> better
> >> > > success. Could you please send me the file "met-
5.0/config.log"?
> >> > That'll
> >> > > tell me the version of the compilers you're using and will
help me
> >> > > investigate the issue further.
> >> > >
> >> > > Thanks,
> >> > > John
> >> > >
> >> > > On Wed, Jan 21, 2015 at 9:43 PM, shahzad.sultan at gmail.com
> <javascript:;> via RT <
> >> > > met_help at ucar.edu <javascript:;>> wrote:
> >> > >
> >> > > >
> >> > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
> >> > > >
> >> > > > Dear John
> >> > > > Thanks for your reply. I follow the directions you have
said and
> got
> >> > > little
> >> > > > different error, I think I should try some previous version
of
> met?.
> >> > the
> >> > > > log file is attached.
> >> > > >
> >> > > > Thanks again
> >> > > >
> >> > > >
> >> > > > icc -DHAVE_CONFIG_H -I. -I../../..
-I../../../src/basic/vx_cal
> >> > > > -I../../../src/$
> >> > > >
> >> > > > num_array.cc(574): error: "rank" is ambiguous
> >> > > >
> >> > > > ties = rank(data, data_rank, n);
> >> > > >
> >> > > > ^
> >> > > >
> >> > > >
> >> > > > compilation aborted for num_array.cc (code 2)
> >> > > >
> >> > > > make[3]: *** [libvx_util_a-num_array.o] Error 2
> >> > > >
> >> > > > make[2]: *** [install-recursive] Error 1
> >> > > >
> >> > > > make[1]: *** [install-recursive] Error 1
> >> > > >
> >> > > > make: *** [install-recursive] Error 1
> >> > > >
> >> > > > On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via RT
<
> >> > > > met_help at ucar.edu <javascript:;>> wrote:
> >> > > >
> >> > > > > Shazad,
> >> > > > >
> >> > > > > This is a new compilation error message we haven't
encountered
> >> > before.
> >> > > > The
> >> > > > > function "rank" is defined in the internal MET library
named
> >> > "vx_math".
> >> > > > > The compilation error says that rank is ambiguous...
Perhaps
> >> "rank"
> >> > is
> >> > > > also
> >> > > > > defined somewhere else with a different signature.
> >> > > > >
> >> > > > > I notice that you're compiling MET using the Intel
compilers.
> >> And I
> >> > > see
> >> > > > in
> >> > > > > met.log, that you're using the icpc compiler. In all of
our
> >> previous
> >> > > > > compilation and testing of MET with Intel, we've used the
"icc"
> >> > > compiler.
> >> > > > > That may, or may not, be the source of the problem. Can
you
> >> please
> >> > try
> >> > > > > re-running the configure script and recompiling MET using
"icc"
> >> > > instead?
> >> > > > >
> >> > > > > Try this:
> >> > > > > make clean
> >> > > > > setenv CXX `which icc`
> >> > > > > setenv CC `which icc`
> >> > > > > setenv F77 `which ifort`
> >> > > > > ./configure (using whatever options you'd like)
> >> > > > > make install >& make_met.log
> >> > > > >
> >> > > > > Please let me know how that goes. If that doesn't fix
the
> >> problem,
> >> > > > please
> >> > > > > send me the files "make_met.log" and "config.log".
> >> > > > >
> >> > > > > Thanks,
> >> > > > > John Halley Gotway
> >> > > > > met_help at ucar.edu <javascript:;>
> >> > > > >
> >> > > > >
> >> > > > > On Wed, Jan 21, 2015 at 7:09 AM, shahzad.sultan at gmail.com
> <javascript:;> via RT
> >> <
> >> > > > > met_help at ucar.edu <javascript:;>> wrote:
> >> > > > >
> >> > > > > >
> >> > > > > > Wed Jan 21 07:09:47 2015: Request 70455 was acted upon.
> >> > > > > > Transaction: Ticket created by shahzad.sultan at gmail.com
> <javascript:;>
> >> > > > > > Queue: met_help
> >> > > > > > Subject: met v5 compilation aborted for
met_stats.cc
> (code
> >> 2)
> >> > > > > > Owner: Nobody
> >> > > > > > Requestors: shahzad.sultan at gmail.com <javascript:;>
> >> > > > > > Status: new
> >> > > > > > Ticket <URL:
> >> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
> >> > > > >
> >> > > > > >
> >> > > > > >
> >> > > > > > Dear
> >> > > > > > I am trying to install metv5 with intel fortran
compiler on
> mac
> >> > > 10.9.5
> >> > > > > > Maverick using netcdf4 and all other recommended
libraries.
> >> during
> >> > > the
> >> > > > > make
> >> > > > > > process I got the following error
> >> > > > > >
> >> > > > > > met_stats.cc(2280): error: "rank" is ambiguous
> >> > > > > >
> >> > > > > > ties = rank(data, data_rank, n);
> >> > > > > >
> >> > > > > > ^
> >> > > > > >
> >> > > > > >
> >> > > > > > compilation aborted for met_stats.cc (code 2)
> >> > > > > >
> >> > > > > > make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
> >> > > > > >
> >> > > > > > make[3]: *** [all-recursive] Error 1
> >> > > > > >
> >> > > > > > make[2]: *** [all-recursive] Error 1
> >> > > > > >
> >> > > > > > make[1]: *** [all-recursive] Error 1
> >> > > > > >
> >> > > > > > make: *** [all] Error 2
> >> > > > > >
> >> > > > > > the configuration log file "met.log" is attached for
your
> >> > reference.
> >> > > > > >
> >> > > > > > Thanks
> >> > > > > > --
> >> > > > > > Regards,
> >> > > > > > Shahzad Sultan
> >> > > > > > PhD student
> >> > > > > > Chinese University of Hong Kong
> >> > > > > > +852 56420557
> >> > > > > >
> >> > > > > >
> >> > > > >
> >> > > > >
> >> > > >
> >> > > >
> >> > > > --
> >> > > > Regards,
> >> > > > Shahzad Sultan
> >> > > > PhD student
> >> > > > Chinese University of Hong Kong
> >> > > > +852 56420557
> >> > > >
> >> > > >
> >> > >
> >> > >
> >> >
> >> >
> >> > --
> >> > Regards,
> >> > Shahzad Sultan
> >> > PhD student
> >> > Chinese University of Hong Kong
> >> > +852 56420557
> >> >
> >> >
> >>
> >>
> >
> >
> > --
> > Regards,
> > Shahzad Sultan
> > PhD student
> > Chinese University of Hong Kong
> > +852 56420557
> >
> >
>
>
> --
> Regards,
> Shahzad Sultan
> PhD student
> Chinese University of Hong Kong
> +852 56420557
>
>
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: John Halley Gotway
Time: Mon Jan 26 17:29:03 2015
Shazad,
OK, I just posted a new set of bugfixes to the MET website. They
include
the changes required to get MET to compile on your machine. I'd
really
appreciate it if you grab a clean build of MET version 5.0:
http://www.dtcenter.org/met/users/downloads/MET_releases/met-
5.0.20140905.tar.gz
Then follow the instructions on this page to apply the latest set of
patches:
http://www.dtcenter.org/met/users/support/known_issues/METv5.0/index.php
And make sure MET still compiles OK for you.
Thanks,
John
On Sun, Jan 25, 2015 at 8:21 AM, John Halley Gotway <johnhg at ucar.edu>
wrote:
> Shazad,
>
> That build I sent you already did contain the latest set of patches.
Now
> that we know the changes work, I'll make those changes in the code
> repository on Monday and post an updated set of patches. At that
point
> please do try rebuilding using MET version 5 plus the latest set of
> patches. I'd like you to test with the publicly posted code.
>
> As for BUFR, please locate the file names libbufr.a on your machine.
Set
> the MET_BUFRLIB environment variable to that directory. Then rerun
> configure and try compiling again. If that doesn't work, please
send me
> config.log and make_met.log.
>
> Thanks
> Johm
>
>
>
> On Friday, January 23, 2015, shahzad.sultan at gmail.com via RT <
> met_help at ucar.edu> wrote:
>
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
>>
>> Dear John
>> I have also tried after applying patches but it does not work and
bring
>> the
>> same error. I have also tried to remove the *-DBLOCK4 from makefile
but
>> still got error of pb2nc (library not found for -lburf). of course
if we
>> disable pb2nc then it works fine.*
>>
>> *Thanks*
>>
>> On Sat, Jan 24, 2015 at 8:35 AM, Shahzad Sultan
<shahzad.sultan at gmail.com
>> >
>> wrote:
>>
>> > Dear John
>> > Thanks
>> > I am really thankful to you and finally it installed successfully
:).
>> >
>> > I have used the following configuration
>> >
>> > ./configure --prefix=/Users/shahzad/wrf/met5/ FC=ifort F9X=ifort
CXX=icc
>> > CC=icc F77=ifort --disable-pb2nc
>> >
>> > I also would like to ask that, should I apply the patches also?
if so
>> then
>> > i will do after receiving your reply.
>> >
>> >
>> > Thanks
>> >
>> > Since pb2nc was giving some error during make, therefore I try
with
>> > disabling it. the list of the installed tools are as follow
>> >
>> > ascii2nc
>> >
>> > ensemble_stat
>> >
>> > gen_circle_mask
>> >
>> > gen_poly_mask
>> >
>> > grid_stat
>> >
>> > madis2nc
>> >
>> > mode
>> >
>> > mode_analysis
>> >
>> > pcp_combine
>> >
>> > plot_data_plane
>> >
>> > plot_point_obs
>> >
>> > point_stat
>> >
>> > series_analysis
>> >
>> > stat_analysis
>> >
>> > tc_dland
>> >
>> > tc_pairs
>> >
>> > tc_stat
>> >
>> > wavelet_stat
>> >
>> > wwmca_plot
>> >
>> > wwmca_regrid
>> >
>> > On Sat, Jan 24, 2015 at 1:58 AM, John Halley Gotway via RT <
>> > met_help at ucar.edu> wrote:
>> >
>> >> Shazad,
>> >>
>> >> I see that you're using Intel compiler version 14. I was able
to
>> locate
>> >> that version of the Intel compiler and test with it.
Unfortunately,
>> I'm
>> >> not able to replicate the compilation error you're seeing. I'm
>> testing on
>> >> Linux, not on a Mac.
>> >>
>> >> So let's try this next...
>> >>
>> >> I believe the problem is a conflict in the name of the "rank()"
>> function.
>> >> I suspect that's already been defined in a slightly different
way on
>> your
>> >> system. I found two versions of the "rank()" function in the
MET code
>> >> tree. I modified the code to rename one of them "do_rank()" and
the
>> other
>> >> "get_rank()".
>> >>
>> >> I was able to compile and run the modified version of the code
on my
>> >> machine. So hopefully, I made the changes in all the right
spots.
>> >>
>> >> Please download and attempt to compile this version of met-5.0:
>> >>
>> >>
>> >>
>> ftp://ftp.rap.ucar.edu/incoming/irap/met_help/sultan_data/met-
5.0_shazad.sultan.tar.gz
>> >>
>> >> And please let me know if that solves the compilation problem.
If so,
>> >> I'll
>> >> make these changes in the MET code repository and post them as
an
>> official
>> >> bugfix for met-5.0.
>> >>
>> >> Thanks,
>> >> John
>> >>
>> >> On Thu, Jan 22, 2015 at 5:11 PM, shahzad.sultan at gmail.com via RT
<
>> >> met_help at ucar.edu> wrote:
>> >>
>> >> >
>> >> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
>
>> >> >
>> >> > Thanks for your reply
>> >> >
>> >> > I have also tried the MET v4.1 but error remain same. So I
guess the
>> >> issue
>> >> > is either dependency libraries like netcdf. I have also tried
to
>> disable
>> >> > many module but still error remain same.
>> >> >
>> >> > The config.log is attached for your reference for METv5.0.
>> >> >
>> >> > On Fri, Jan 23, 2015 at 12:40 AM, John Halley Gotway via RT <
>> >> > met_help at ucar.edu> wrote:
>> >> >
>> >> > > Shazad,
>> >> > >
>> >> > > I think we should be able to find a solution for this one
way of
>> the
>> >> > > other. And I doubt using a previous version of MET would
provide
>> >> better
>> >> > > success. Could you please send me the file "met-
5.0/config.log"?
>> >> > That'll
>> >> > > tell me the version of the compilers you're using and will
help me
>> >> > > investigate the issue further.
>> >> > >
>> >> > > Thanks,
>> >> > > John
>> >> > >
>> >> > > On Wed, Jan 21, 2015 at 9:43 PM, shahzad.sultan at gmail.com
via RT <
>> >> > > met_help at ucar.edu> wrote:
>> >> > >
>> >> > > >
>> >> > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
>> >> > > >
>> >> > > > Dear John
>> >> > > > Thanks for your reply. I follow the directions you have
said and
>> got
>> >> > > little
>> >> > > > different error, I think I should try some previous
version of
>> met?.
>> >> > the
>> >> > > > log file is attached.
>> >> > > >
>> >> > > > Thanks again
>> >> > > >
>> >> > > >
>> >> > > > icc -DHAVE_CONFIG_H -I. -I../../..
-I../../../src/basic/vx_cal
>> >> > > > -I../../../src/$
>> >> > > >
>> >> > > > num_array.cc(574): error: "rank" is ambiguous
>> >> > > >
>> >> > > > ties = rank(data, data_rank, n);
>> >> > > >
>> >> > > > ^
>> >> > > >
>> >> > > >
>> >> > > > compilation aborted for num_array.cc (code 2)
>> >> > > >
>> >> > > > make[3]: *** [libvx_util_a-num_array.o] Error 2
>> >> > > >
>> >> > > > make[2]: *** [install-recursive] Error 1
>> >> > > >
>> >> > > > make[1]: *** [install-recursive] Error 1
>> >> > > >
>> >> > > > make: *** [install-recursive] Error 1
>> >> > > >
>> >> > > > On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via RT
<
>> >> > > > met_help at ucar.edu> wrote:
>> >> > > >
>> >> > > > > Shazad,
>> >> > > > >
>> >> > > > > This is a new compilation error message we haven't
encountered
>> >> > before.
>> >> > > > The
>> >> > > > > function "rank" is defined in the internal MET library
named
>> >> > "vx_math".
>> >> > > > > The compilation error says that rank is ambiguous...
Perhaps
>> >> "rank"
>> >> > is
>> >> > > > also
>> >> > > > > defined somewhere else with a different signature.
>> >> > > > >
>> >> > > > > I notice that you're compiling MET using the Intel
compilers.
>> >> And I
>> >> > > see
>> >> > > > in
>> >> > > > > met.log, that you're using the icpc compiler. In all of
our
>> >> previous
>> >> > > > > compilation and testing of MET with Intel, we've used
the "icc"
>> >> > > compiler.
>> >> > > > > That may, or may not, be the source of the problem. Can
you
>> >> please
>> >> > try
>> >> > > > > re-running the configure script and recompiling MET
using "icc"
>> >> > > instead?
>> >> > > > >
>> >> > > > > Try this:
>> >> > > > > make clean
>> >> > > > > setenv CXX `which icc`
>> >> > > > > setenv CC `which icc`
>> >> > > > > setenv F77 `which ifort`
>> >> > > > > ./configure (using whatever options you'd like)
>> >> > > > > make install >& make_met.log
>> >> > > > >
>> >> > > > > Please let me know how that goes. If that doesn't fix
the
>> >> problem,
>> >> > > > please
>> >> > > > > send me the files "make_met.log" and "config.log".
>> >> > > > >
>> >> > > > > Thanks,
>> >> > > > > John Halley Gotway
>> >> > > > > met_help at ucar.edu
>> >> > > > >
>> >> > > > >
>> >> > > > > On Wed, Jan 21, 2015 at 7:09 AM,
shahzad.sultan at gmail.com via
>> RT
>> >> <
>> >> > > > > met_help at ucar.edu> wrote:
>> >> > > > >
>> >> > > > > >
>> >> > > > > > Wed Jan 21 07:09:47 2015: Request 70455 was acted
upon.
>> >> > > > > > Transaction: Ticket created by
shahzad.sultan at gmail.com
>> >> > > > > > Queue: met_help
>> >> > > > > > Subject: met v5 compilation aborted for
met_stats.cc
>> (code
>> >> 2)
>> >> > > > > > Owner: Nobody
>> >> > > > > > Requestors: shahzad.sultan at gmail.com
>> >> > > > > > Status: new
>> >> > > > > > Ticket <URL:
>> >> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
>> >> > > > >
>> >> > > > > >
>> >> > > > > >
>> >> > > > > > Dear
>> >> > > > > > I am trying to install metv5 with intel fortran
compiler on
>> mac
>> >> > > 10.9.5
>> >> > > > > > Maverick using netcdf4 and all other recommended
libraries.
>> >> during
>> >> > > the
>> >> > > > > make
>> >> > > > > > process I got the following error
>> >> > > > > >
>> >> > > > > > met_stats.cc(2280): error: "rank" is ambiguous
>> >> > > > > >
>> >> > > > > > ties = rank(data, data_rank, n);
>> >> > > > > >
>> >> > > > > > ^
>> >> > > > > >
>> >> > > > > >
>> >> > > > > > compilation aborted for met_stats.cc (code 2)
>> >> > > > > >
>> >> > > > > > make[4]: *** [libvx_statistics_a-met_stats.o] Error 2
>> >> > > > > >
>> >> > > > > > make[3]: *** [all-recursive] Error 1
>> >> > > > > >
>> >> > > > > > make[2]: *** [all-recursive] Error 1
>> >> > > > > >
>> >> > > > > > make[1]: *** [all-recursive] Error 1
>> >> > > > > >
>> >> > > > > > make: *** [all] Error 2
>> >> > > > > >
>> >> > > > > > the configuration log file "met.log" is attached for
your
>> >> > reference.
>> >> > > > > >
>> >> > > > > > Thanks
>> >> > > > > > --
>> >> > > > > > Regards,
>> >> > > > > > Shahzad Sultan
>> >> > > > > > PhD student
>> >> > > > > > Chinese University of Hong Kong
>> >> > > > > > +852 56420557
>> >> > > > > >
>> >> > > > > >
>> >> > > > >
>> >> > > > >
>> >> > > >
>> >> > > >
>> >> > > > --
>> >> > > > Regards,
>> >> > > > Shahzad Sultan
>> >> > > > PhD student
>> >> > > > Chinese University of Hong Kong
>> >> > > > +852 56420557
>> >> > > >
>> >> > > >
>> >> > >
>> >> > >
>> >> >
>> >> >
>> >> > --
>> >> > Regards,
>> >> > Shahzad Sultan
>> >> > PhD student
>> >> > Chinese University of Hong Kong
>> >> > +852 56420557
>> >> >
>> >> >
>> >>
>> >>
>> >
>> >
>> > --
>> > Regards,
>> > Shahzad Sultan
>> > PhD student
>> > Chinese University of Hong Kong
>> > +852 56420557
>> >
>> >
>>
>>
>> --
>> Regards,
>> Shahzad Sultan
>> PhD student
>> Chinese University of Hong Kong
>> +852 56420557
>>
>>
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: shahzad.sultan at gmail.com
Time: Mon Jan 26 18:17:06 2015
Dear
Thanks for updates. I will try a fresh installation today with
enabling
HDF4 and EOS as well and will let you know the results.
Thanks again
On Tue, Jan 27, 2015 at 8:29 AM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:
> Shazad,
>
> OK, I just posted a new set of bugfixes to the MET website. They
include
> the changes required to get MET to compile on your machine. I'd
really
> appreciate it if you grab a clean build of MET version 5.0:
>
>
> http://www.dtcenter.org/met/users/downloads/MET_releases/met-
5.0.20140905.tar.gz
>
> Then follow the instructions on this page to apply the latest set of
> patches:
>
>
http://www.dtcenter.org/met/users/support/known_issues/METv5.0/index.php
>
> And make sure MET still compiles OK for you.
>
> Thanks,
> John
>
> On Sun, Jan 25, 2015 at 8:21 AM, John Halley Gotway
<johnhg at ucar.edu>
> wrote:
>
> > Shazad,
> >
> > That build I sent you already did contain the latest set of
patches. Now
> > that we know the changes work, I'll make those changes in the code
> > repository on Monday and post an updated set of patches. At that
point
> > please do try rebuilding using MET version 5 plus the latest set
of
> > patches. I'd like you to test with the publicly posted code.
> >
> > As for BUFR, please locate the file names libbufr.a on your
machine. Set
> > the MET_BUFRLIB environment variable to that directory. Then
rerun
> > configure and try compiling again. If that doesn't work, please
send me
> > config.log and make_met.log.
> >
> > Thanks
> > Johm
> >
> >
> >
> > On Friday, January 23, 2015, shahzad.sultan at gmail.com via RT <
> > met_help at ucar.edu> wrote:
> >
> >>
> >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
> >>
> >> Dear John
> >> I have also tried after applying patches but it does not work and
bring
> >> the
> >> same error. I have also tried to remove the *-DBLOCK4 from
makefile but
> >> still got error of pb2nc (library not found for -lburf). of
course if we
> >> disable pb2nc then it works fine.*
> >>
> >> *Thanks*
> >>
> >> On Sat, Jan 24, 2015 at 8:35 AM, Shahzad Sultan <
> shahzad.sultan at gmail.com
> >> >
> >> wrote:
> >>
> >> > Dear John
> >> > Thanks
> >> > I am really thankful to you and finally it installed
successfully :).
> >> >
> >> > I have used the following configuration
> >> >
> >> > ./configure --prefix=/Users/shahzad/wrf/met5/ FC=ifort
F9X=ifort
> CXX=icc
> >> > CC=icc F77=ifort --disable-pb2nc
> >> >
> >> > I also would like to ask that, should I apply the patches also?
if so
> >> then
> >> > i will do after receiving your reply.
> >> >
> >> >
> >> > Thanks
> >> >
> >> > Since pb2nc was giving some error during make, therefore I try
with
> >> > disabling it. the list of the installed tools are as follow
> >> >
> >> > ascii2nc
> >> >
> >> > ensemble_stat
> >> >
> >> > gen_circle_mask
> >> >
> >> > gen_poly_mask
> >> >
> >> > grid_stat
> >> >
> >> > madis2nc
> >> >
> >> > mode
> >> >
> >> > mode_analysis
> >> >
> >> > pcp_combine
> >> >
> >> > plot_data_plane
> >> >
> >> > plot_point_obs
> >> >
> >> > point_stat
> >> >
> >> > series_analysis
> >> >
> >> > stat_analysis
> >> >
> >> > tc_dland
> >> >
> >> > tc_pairs
> >> >
> >> > tc_stat
> >> >
> >> > wavelet_stat
> >> >
> >> > wwmca_plot
> >> >
> >> > wwmca_regrid
> >> >
> >> > On Sat, Jan 24, 2015 at 1:58 AM, John Halley Gotway via RT <
> >> > met_help at ucar.edu> wrote:
> >> >
> >> >> Shazad,
> >> >>
> >> >> I see that you're using Intel compiler version 14. I was able
to
> >> locate
> >> >> that version of the Intel compiler and test with it.
Unfortunately,
> >> I'm
> >> >> not able to replicate the compilation error you're seeing.
I'm
> >> testing on
> >> >> Linux, not on a Mac.
> >> >>
> >> >> So let's try this next...
> >> >>
> >> >> I believe the problem is a conflict in the name of the
"rank()"
> >> function.
> >> >> I suspect that's already been defined in a slightly different
way on
> >> your
> >> >> system. I found two versions of the "rank()" function in the
MET
> code
> >> >> tree. I modified the code to rename one of them "do_rank()"
and the
> >> other
> >> >> "get_rank()".
> >> >>
> >> >> I was able to compile and run the modified version of the code
on my
> >> >> machine. So hopefully, I made the changes in all the right
spots.
> >> >>
> >> >> Please download and attempt to compile this version of met-
5.0:
> >> >>
> >> >>
> >> >>
> >>
> ftp://ftp.rap.ucar.edu/incoming/irap/met_help/sultan_data/met-
5.0_shazad.sultan.tar.gz
> >> >>
> >> >> And please let me know if that solves the compilation problem.
If
> so,
> >> >> I'll
> >> >> make these changes in the MET code repository and post them as
an
> >> official
> >> >> bugfix for met-5.0.
> >> >>
> >> >> Thanks,
> >> >> John
> >> >>
> >> >> On Thu, Jan 22, 2015 at 5:11 PM, shahzad.sultan at gmail.com via
RT <
> >> >> met_help at ucar.edu> wrote:
> >> >>
> >> >> >
> >> >> > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
> >> >> >
> >> >> > Thanks for your reply
> >> >> >
> >> >> > I have also tried the MET v4.1 but error remain same. So I
guess
> the
> >> >> issue
> >> >> > is either dependency libraries like netcdf. I have also
tried to
> >> disable
> >> >> > many module but still error remain same.
> >> >> >
> >> >> > The config.log is attached for your reference for METv5.0.
> >> >> >
> >> >> > On Fri, Jan 23, 2015 at 12:40 AM, John Halley Gotway via RT
<
> >> >> > met_help at ucar.edu> wrote:
> >> >> >
> >> >> > > Shazad,
> >> >> > >
> >> >> > > I think we should be able to find a solution for this one
way of
> >> the
> >> >> > > other. And I doubt using a previous version of MET would
provide
> >> >> better
> >> >> > > success. Could you please send me the file "met-
5.0/config.log"?
> >> >> > That'll
> >> >> > > tell me the version of the compilers you're using and will
help
> me
> >> >> > > investigate the issue further.
> >> >> > >
> >> >> > > Thanks,
> >> >> > > John
> >> >> > >
> >> >> > > On Wed, Jan 21, 2015 at 9:43 PM, shahzad.sultan at gmail.com
via
> RT <
> >> >> > > met_help at ucar.edu> wrote:
> >> >> > >
> >> >> > > >
> >> >> > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
> >
> >> >> > > >
> >> >> > > > Dear John
> >> >> > > > Thanks for your reply. I follow the directions you have
said
> and
> >> got
> >> >> > > little
> >> >> > > > different error, I think I should try some previous
version of
> >> met?.
> >> >> > the
> >> >> > > > log file is attached.
> >> >> > > >
> >> >> > > > Thanks again
> >> >> > > >
> >> >> > > >
> >> >> > > > icc -DHAVE_CONFIG_H -I. -I../../..
-I../../../src/basic/vx_cal
> >> >> > > > -I../../../src/$
> >> >> > > >
> >> >> > > > num_array.cc(574): error: "rank" is ambiguous
> >> >> > > >
> >> >> > > > ties = rank(data, data_rank, n);
> >> >> > > >
> >> >> > > > ^
> >> >> > > >
> >> >> > > >
> >> >> > > > compilation aborted for num_array.cc (code 2)
> >> >> > > >
> >> >> > > > make[3]: *** [libvx_util_a-num_array.o] Error 2
> >> >> > > >
> >> >> > > > make[2]: *** [install-recursive] Error 1
> >> >> > > >
> >> >> > > > make[1]: *** [install-recursive] Error 1
> >> >> > > >
> >> >> > > > make: *** [install-recursive] Error 1
> >> >> > > >
> >> >> > > > On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via
RT <
> >> >> > > > met_help at ucar.edu> wrote:
> >> >> > > >
> >> >> > > > > Shazad,
> >> >> > > > >
> >> >> > > > > This is a new compilation error message we haven't
> encountered
> >> >> > before.
> >> >> > > > The
> >> >> > > > > function "rank" is defined in the internal MET library
named
> >> >> > "vx_math".
> >> >> > > > > The compilation error says that rank is ambiguous...
Perhaps
> >> >> "rank"
> >> >> > is
> >> >> > > > also
> >> >> > > > > defined somewhere else with a different signature.
> >> >> > > > >
> >> >> > > > > I notice that you're compiling MET using the Intel
compilers.
> >> >> And I
> >> >> > > see
> >> >> > > > in
> >> >> > > > > met.log, that you're using the icpc compiler. In all
of our
> >> >> previous
> >> >> > > > > compilation and testing of MET with Intel, we've used
the
> "icc"
> >> >> > > compiler.
> >> >> > > > > That may, or may not, be the source of the problem.
Can you
> >> >> please
> >> >> > try
> >> >> > > > > re-running the configure script and recompiling MET
using
> "icc"
> >> >> > > instead?
> >> >> > > > >
> >> >> > > > > Try this:
> >> >> > > > > make clean
> >> >> > > > > setenv CXX `which icc`
> >> >> > > > > setenv CC `which icc`
> >> >> > > > > setenv F77 `which ifort`
> >> >> > > > > ./configure (using whatever options you'd like)
> >> >> > > > > make install >& make_met.log
> >> >> > > > >
> >> >> > > > > Please let me know how that goes. If that doesn't fix
the
> >> >> problem,
> >> >> > > > please
> >> >> > > > > send me the files "make_met.log" and "config.log".
> >> >> > > > >
> >> >> > > > > Thanks,
> >> >> > > > > John Halley Gotway
> >> >> > > > > met_help at ucar.edu
> >> >> > > > >
> >> >> > > > >
> >> >> > > > > On Wed, Jan 21, 2015 at 7:09 AM,
shahzad.sultan at gmail.com
> via
> >> RT
> >> >> <
> >> >> > > > > met_help at ucar.edu> wrote:
> >> >> > > > >
> >> >> > > > > >
> >> >> > > > > > Wed Jan 21 07:09:47 2015: Request 70455 was acted
upon.
> >> >> > > > > > Transaction: Ticket created by
shahzad.sultan at gmail.com
> >> >> > > > > > Queue: met_help
> >> >> > > > > > Subject: met v5 compilation aborted for
met_stats.cc
> >> (code
> >> >> 2)
> >> >> > > > > > Owner: Nobody
> >> >> > > > > > Requestors: shahzad.sultan at gmail.com
> >> >> > > > > > Status: new
> >> >> > > > > > Ticket <URL:
> >> >> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
> >> >> > > > >
> >> >> > > > > >
> >> >> > > > > >
> >> >> > > > > > Dear
> >> >> > > > > > I am trying to install metv5 with intel fortran
compiler on
> >> mac
> >> >> > > 10.9.5
> >> >> > > > > > Maverick using netcdf4 and all other recommended
libraries.
> >> >> during
> >> >> > > the
> >> >> > > > > make
> >> >> > > > > > process I got the following error
> >> >> > > > > >
> >> >> > > > > > met_stats.cc(2280): error: "rank" is ambiguous
> >> >> > > > > >
> >> >> > > > > > ties = rank(data, data_rank, n);
> >> >> > > > > >
> >> >> > > > > > ^
> >> >> > > > > >
> >> >> > > > > >
> >> >> > > > > > compilation aborted for met_stats.cc (code 2)
> >> >> > > > > >
> >> >> > > > > > make[4]: *** [libvx_statistics_a-met_stats.o] Error
2
> >> >> > > > > >
> >> >> > > > > > make[3]: *** [all-recursive] Error 1
> >> >> > > > > >
> >> >> > > > > > make[2]: *** [all-recursive] Error 1
> >> >> > > > > >
> >> >> > > > > > make[1]: *** [all-recursive] Error 1
> >> >> > > > > >
> >> >> > > > > > make: *** [all] Error 2
> >> >> > > > > >
> >> >> > > > > > the configuration log file "met.log" is attached for
your
> >> >> > reference.
> >> >> > > > > >
> >> >> > > > > > Thanks
> >> >> > > > > > --
> >> >> > > > > > Regards,
> >> >> > > > > > Shahzad Sultan
> >> >> > > > > > PhD student
> >> >> > > > > > Chinese University of Hong Kong
> >> >> > > > > > +852 56420557
> >> >> > > > > >
> >> >> > > > > >
> >> >> > > > >
> >> >> > > > >
> >> >> > > >
> >> >> > > >
> >> >> > > > --
> >> >> > > > Regards,
> >> >> > > > Shahzad Sultan
> >> >> > > > PhD student
> >> >> > > > Chinese University of Hong Kong
> >> >> > > > +852 56420557
> >> >> > > >
> >> >> > > >
> >> >> > >
> >> >> > >
> >> >> >
> >> >> >
> >> >> > --
> >> >> > Regards,
> >> >> > Shahzad Sultan
> >> >> > PhD student
> >> >> > Chinese University of Hong Kong
> >> >> > +852 56420557
> >> >> >
> >> >> >
> >> >>
> >> >>
> >> >
> >> >
> >> > --
> >> > Regards,
> >> > Shahzad Sultan
> >> > PhD student
> >> > Chinese University of Hong Kong
> >> > +852 56420557
> >> >
> >> >
> >>
> >>
> >> --
> >> Regards,
> >> Shahzad Sultan
> >> PhD student
> >> Chinese University of Hong Kong
> >> +852 56420557
> >>
> >>
>
>
--
Regards,
Shahzad Sultan
PhD student
Chinese University of Hong Kong
+852 56420557
------------------------------------------------
Subject: met v5 compilation aborted for met_stats.cc (code 2)
From: shahzad.sultan at gmail.com
Time: Mon Jan 26 20:26:19 2015
Dear I have downloaded the said version with supplied updates and
installed
successfully using intel-fortran compiler. the config.log file is
attached.
I have tried with grib2, HDF-EOS and HDF4 but did not succeeded. but
all
other things are fine. one thing more on mac one must use CXX=icc as
without this it will not compiled successfully. As I have mentioned in
previous correspondence that user must not use homebrew or macport to
install the dependent libraries on mac.
The configuration command is as follow
./configure --prefix=/Users/shahzad/wrf/MET5_intel/ FC=ifort CC=icc
F9x=ifort F77=ifort FXX=ifort CXX=icc >&config.log
The following operator has been successfully compiled
ascii2nc
ensemble_stat
gen_circle_mask
gen_poly_mask
grid_stat
madis2nc
mode
mode_analysis
pb2nc
pcp_combine
plot_data_plane
plot_point_obs
point_stat
series_analysis
stat_analysis
tc_dland
tc_pairs
tc_stat
wavelet_stat
wwmca_plot
wwmca_regrid
Thanks for your support
On Tue, Jan 27, 2015 at 9:17 AM, Shahzad Sultan
<shahzad.sultan at gmail.com>
wrote:
> Dear
> Thanks for updates. I will try a fresh installation today with
enabling
> HDF4 and EOS as well and will let you know the results.
>
> Thanks again
>
> On Tue, Jan 27, 2015 at 8:29 AM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
>> Shazad,
>>
>> OK, I just posted a new set of bugfixes to the MET website. They
include
>> the changes required to get MET to compile on your machine. I'd
really
>> appreciate it if you grab a clean build of MET version 5.0:
>>
>>
>> http://www.dtcenter.org/met/users/downloads/MET_releases/met-
5.0.20140905.tar.gz
>>
>> Then follow the instructions on this page to apply the latest set
of
>> patches:
>>
>>
http://www.dtcenter.org/met/users/support/known_issues/METv5.0/index.php
>>
>> And make sure MET still compiles OK for you.
>>
>> Thanks,
>> John
>>
>> On Sun, Jan 25, 2015 at 8:21 AM, John Halley Gotway
<johnhg at ucar.edu>
>> wrote:
>>
>> > Shazad,
>> >
>> > That build I sent you already did contain the latest set of
patches.
>> Now
>> > that we know the changes work, I'll make those changes in the
code
>> > repository on Monday and post an updated set of patches. At that
point
>> > please do try rebuilding using MET version 5 plus the latest set
of
>> > patches. I'd like you to test with the publicly posted code.
>> >
>> > As for BUFR, please locate the file names libbufr.a on your
machine.
>> Set
>> > the MET_BUFRLIB environment variable to that directory. Then
rerun
>> > configure and try compiling again. If that doesn't work, please
send me
>> > config.log and make_met.log.
>> >
>> > Thanks
>> > Johm
>> >
>> >
>> >
>> > On Friday, January 23, 2015, shahzad.sultan at gmail.com via RT <
>> > met_help at ucar.edu> wrote:
>> >
>> >>
>> >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
>> >>
>> >> Dear John
>> >> I have also tried after applying patches but it does not work
and bring
>> >> the
>> >> same error. I have also tried to remove the *-DBLOCK4 from
makefile but
>> >> still got error of pb2nc (library not found for -lburf). of
course if
>> we
>> >> disable pb2nc then it works fine.*
>> >>
>> >> *Thanks*
>> >>
>> >> On Sat, Jan 24, 2015 at 8:35 AM, Shahzad Sultan <
>> shahzad.sultan at gmail.com
>> >> >
>> >> wrote:
>> >>
>> >> > Dear John
>> >> > Thanks
>> >> > I am really thankful to you and finally it installed
successfully :).
>> >> >
>> >> > I have used the following configuration
>> >> >
>> >> > ./configure --prefix=/Users/shahzad/wrf/met5/ FC=ifort
F9X=ifort
>> CXX=icc
>> >> > CC=icc F77=ifort --disable-pb2nc
>> >> >
>> >> > I also would like to ask that, should I apply the patches
also? if so
>> >> then
>> >> > i will do after receiving your reply.
>> >> >
>> >> >
>> >> > Thanks
>> >> >
>> >> > Since pb2nc was giving some error during make, therefore I try
with
>> >> > disabling it. the list of the installed tools are as follow
>> >> >
>> >> > ascii2nc
>> >> >
>> >> > ensemble_stat
>> >> >
>> >> > gen_circle_mask
>> >> >
>> >> > gen_poly_mask
>> >> >
>> >> > grid_stat
>> >> >
>> >> > madis2nc
>> >> >
>> >> > mode
>> >> >
>> >> > mode_analysis
>> >> >
>> >> > pcp_combine
>> >> >
>> >> > plot_data_plane
>> >> >
>> >> > plot_point_obs
>> >> >
>> >> > point_stat
>> >> >
>> >> > series_analysis
>> >> >
>> >> > stat_analysis
>> >> >
>> >> > tc_dland
>> >> >
>> >> > tc_pairs
>> >> >
>> >> > tc_stat
>> >> >
>> >> > wavelet_stat
>> >> >
>> >> > wwmca_plot
>> >> >
>> >> > wwmca_regrid
>> >> >
>> >> > On Sat, Jan 24, 2015 at 1:58 AM, John Halley Gotway via RT <
>> >> > met_help at ucar.edu> wrote:
>> >> >
>> >> >> Shazad,
>> >> >>
>> >> >> I see that you're using Intel compiler version 14. I was
able to
>> >> locate
>> >> >> that version of the Intel compiler and test with it.
Unfortunately,
>> >> I'm
>> >> >> not able to replicate the compilation error you're seeing.
I'm
>> >> testing on
>> >> >> Linux, not on a Mac.
>> >> >>
>> >> >> So let's try this next...
>> >> >>
>> >> >> I believe the problem is a conflict in the name of the
"rank()"
>> >> function.
>> >> >> I suspect that's already been defined in a slightly different
way on
>> >> your
>> >> >> system. I found two versions of the "rank()" function in the
MET
>> code
>> >> >> tree. I modified the code to rename one of them "do_rank()"
and the
>> >> other
>> >> >> "get_rank()".
>> >> >>
>> >> >> I was able to compile and run the modified version of the
code on my
>> >> >> machine. So hopefully, I made the changes in all the right
spots.
>> >> >>
>> >> >> Please download and attempt to compile this version of met-
5.0:
>> >> >>
>> >> >>
>> >> >>
>> >>
>> ftp://ftp.rap.ucar.edu/incoming/irap/met_help/sultan_data/met-
5.0_shazad.sultan.tar.gz
>> >> >>
>> >> >> And please let me know if that solves the compilation
problem. If
>> so,
>> >> >> I'll
>> >> >> make these changes in the MET code repository and post them
as an
>> >> official
>> >> >> bugfix for met-5.0.
>> >> >>
>> >> >> Thanks,
>> >> >> John
>> >> >>
>> >> >> On Thu, Jan 22, 2015 at 5:11 PM, shahzad.sultan at gmail.com via
RT <
>> >> >> met_help at ucar.edu> wrote:
>> >> >>
>> >> >> >
>> >> >> > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455 >
>> >> >> >
>> >> >> > Thanks for your reply
>> >> >> >
>> >> >> > I have also tried the MET v4.1 but error remain same. So I
guess
>> the
>> >> >> issue
>> >> >> > is either dependency libraries like netcdf. I have also
tried to
>> >> disable
>> >> >> > many module but still error remain same.
>> >> >> >
>> >> >> > The config.log is attached for your reference for METv5.0.
>> >> >> >
>> >> >> > On Fri, Jan 23, 2015 at 12:40 AM, John Halley Gotway via RT
<
>> >> >> > met_help at ucar.edu> wrote:
>> >> >> >
>> >> >> > > Shazad,
>> >> >> > >
>> >> >> > > I think we should be able to find a solution for this one
way of
>> >> the
>> >> >> > > other. And I doubt using a previous version of MET would
>> provide
>> >> >> better
>> >> >> > > success. Could you please send me the file
>> "met-5.0/config.log"?
>> >> >> > That'll
>> >> >> > > tell me the version of the compilers you're using and
will help
>> me
>> >> >> > > investigate the issue further.
>> >> >> > >
>> >> >> > > Thanks,
>> >> >> > > John
>> >> >> > >
>> >> >> > > On Wed, Jan 21, 2015 at 9:43 PM, shahzad.sultan at gmail.com
via
>> RT <
>> >> >> > > met_help at ucar.edu> wrote:
>> >> >> > >
>> >> >> > > >
>> >> >> > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
>> >
>> >> >> > > >
>> >> >> > > > Dear John
>> >> >> > > > Thanks for your reply. I follow the directions you have
said
>> and
>> >> got
>> >> >> > > little
>> >> >> > > > different error, I think I should try some previous
version of
>> >> met?.
>> >> >> > the
>> >> >> > > > log file is attached.
>> >> >> > > >
>> >> >> > > > Thanks again
>> >> >> > > >
>> >> >> > > >
>> >> >> > > > icc -DHAVE_CONFIG_H -I. -I../../..
>> -I../../../src/basic/vx_cal
>> >> >> > > > -I../../../src/$
>> >> >> > > >
>> >> >> > > > num_array.cc(574): error: "rank" is ambiguous
>> >> >> > > >
>> >> >> > > > ties = rank(data, data_rank, n);
>> >> >> > > >
>> >> >> > > > ^
>> >> >> > > >
>> >> >> > > >
>> >> >> > > > compilation aborted for num_array.cc (code 2)
>> >> >> > > >
>> >> >> > > > make[3]: *** [libvx_util_a-num_array.o] Error 2
>> >> >> > > >
>> >> >> > > > make[2]: *** [install-recursive] Error 1
>> >> >> > > >
>> >> >> > > > make[1]: *** [install-recursive] Error 1
>> >> >> > > >
>> >> >> > > > make: *** [install-recursive] Error 1
>> >> >> > > >
>> >> >> > > > On Thu, Jan 22, 2015 at 2:00 AM, John Halley Gotway via
RT <
>> >> >> > > > met_help at ucar.edu> wrote:
>> >> >> > > >
>> >> >> > > > > Shazad,
>> >> >> > > > >
>> >> >> > > > > This is a new compilation error message we haven't
>> encountered
>> >> >> > before.
>> >> >> > > > The
>> >> >> > > > > function "rank" is defined in the internal MET
library named
>> >> >> > "vx_math".
>> >> >> > > > > The compilation error says that rank is ambiguous...
Perhaps
>> >> >> "rank"
>> >> >> > is
>> >> >> > > > also
>> >> >> > > > > defined somewhere else with a different signature.
>> >> >> > > > >
>> >> >> > > > > I notice that you're compiling MET using the Intel
>> compilers.
>> >> >> And I
>> >> >> > > see
>> >> >> > > > in
>> >> >> > > > > met.log, that you're using the icpc compiler. In all
of our
>> >> >> previous
>> >> >> > > > > compilation and testing of MET with Intel, we've used
the
>> "icc"
>> >> >> > > compiler.
>> >> >> > > > > That may, or may not, be the source of the problem.
Can you
>> >> >> please
>> >> >> > try
>> >> >> > > > > re-running the configure script and recompiling MET
using
>> "icc"
>> >> >> > > instead?
>> >> >> > > > >
>> >> >> > > > > Try this:
>> >> >> > > > > make clean
>> >> >> > > > > setenv CXX `which icc`
>> >> >> > > > > setenv CC `which icc`
>> >> >> > > > > setenv F77 `which ifort`
>> >> >> > > > > ./configure (using whatever options you'd like)
>> >> >> > > > > make install >& make_met.log
>> >> >> > > > >
>> >> >> > > > > Please let me know how that goes. If that doesn't
fix the
>> >> >> problem,
>> >> >> > > > please
>> >> >> > > > > send me the files "make_met.log" and "config.log".
>> >> >> > > > >
>> >> >> > > > > Thanks,
>> >> >> > > > > John Halley Gotway
>> >> >> > > > > met_help at ucar.edu
>> >> >> > > > >
>> >> >> > > > >
>> >> >> > > > > On Wed, Jan 21, 2015 at 7:09 AM,
shahzad.sultan at gmail.com
>> via
>> >> RT
>> >> >> <
>> >> >> > > > > met_help at ucar.edu> wrote:
>> >> >> > > > >
>> >> >> > > > > >
>> >> >> > > > > > Wed Jan 21 07:09:47 2015: Request 70455 was acted
upon.
>> >> >> > > > > > Transaction: Ticket created by
shahzad.sultan at gmail.com
>> >> >> > > > > > Queue: met_help
>> >> >> > > > > > Subject: met v5 compilation aborted for
met_stats.cc
>> >> (code
>> >> >> 2)
>> >> >> > > > > > Owner: Nobody
>> >> >> > > > > > Requestors: shahzad.sultan at gmail.com
>> >> >> > > > > > Status: new
>> >> >> > > > > > Ticket <URL:
>> >> >> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70455
>> >> >> > > > >
>> >> >> > > > > >
>> >> >> > > > > >
>> >> >> > > > > > Dear
>> >> >> > > > > > I am trying to install metv5 with intel fortran
compiler
>> on
>> >> mac
>> >> >> > > 10.9.5
>> >> >> > > > > > Maverick using netcdf4 and all other recommended
>> libraries.
>> >> >> during
>> >> >> > > the
>> >> >> > > > > make
>> >> >> > > > > > process I got the following error
>> >> >> > > > > >
>> >> >> > > > > > met_stats.cc(2280): error: "rank" is ambiguous
>> >> >> > > > > >
>> >> >> > > > > > ties = rank(data, data_rank, n);
>> >> >> > > > > >
>> >> >> > > > > > ^
>> >> >> > > > > >
>> >> >> > > > > >
>> >> >> > > > > > compilation aborted for met_stats.cc (code 2)
>> >> >> > > > > >
>> >> >> > > > > > make[4]: *** [libvx_statistics_a-met_stats.o] Error
2
>> >> >> > > > > >
>> >> >> > > > > > make[3]: *** [all-recursive] Error 1
>> >> >> > > > > >
>> >> >> > > > > > make[2]: *** [all-recursive] Error 1
>> >> >> > > > > >
>> >> >> > > > > > make[1]: *** [all-recursive] Error 1
>> >> >> > > > > >
>> >> >> > > > > > make: *** [all] Error 2
>> >> >> > > > > >
>> >> >> > > > > > the configuration log file "met.log" is attached
for your
>> >> >> > reference.
>> >> >> > > > > >
>> >> >> > > > > > Thanks
>> >> >> > > > > > --
>> >> >> > > > > > Regards,
>> >> >> > > > > > Shahzad Sultan
>> >> >> > > > > > PhD student
>> >> >> > > > > > Chinese University of Hong Kong
>> >> >> > > > > > +852 56420557
>> >> >> > > > > >
>> >> >> > > > > >
>> >> >> > > > >
>> >> >> > > > >
>> >> >> > > >
>> >> >> > > >
>> >> >> > > > --
>> >> >> > > > Regards,
>> >> >> > > > Shahzad Sultan
>> >> >> > > > PhD student
>> >> >> > > > Chinese University of Hong Kong
>> >> >> > > > +852 56420557
>> >> >> > > >
>> >> >> > > >
>> >> >> > >
>> >> >> > >
>> >> >> >
>> >> >> >
>> >> >> > --
>> >> >> > Regards,
>> >> >> > Shahzad Sultan
>> >> >> > PhD student
>> >> >> > Chinese University of Hong Kong
>> >> >> > +852 56420557
>> >> >> >
>> >> >> >
>> >> >>
>> >> >>
>> >> >
>> >> >
>> >> > --
>> >> > Regards,
>> >> > Shahzad Sultan
>> >> > PhD student
>> >> > Chinese University of Hong Kong
>> >> > +852 56420557
>> >> >
>> >> >
>> >>
>> >>
>> >> --
>> >> Regards,
>> >> Shahzad Sultan
>> >> PhD student
>> >> Chinese University of Hong Kong
>> >> +852 56420557
>> >>
>> >>
>>
>>
>
>
> --
> Regards,
> Shahzad Sultan
> PhD student
> Chinese University of Hong Kong
> +852 56420557
>
>
--
Regards,
Shahzad Sultan
PhD student
Chinese University of Hong Kong
+852 56420557
------------------------------------------------
More information about the Met_help
mailing list