[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[netCDF #CUL-335078]: Regression error with run_f90_par_test.sh in netCDF-Fortran 4.5.2

Hello Carl,

I would have sworn this was fixed.  Thank you for alerting us to the 
regression, I'll go take care of it.  I've just pushed a new netCDF-C release 
out, perhaps we will do a new Fortran release as well.

Thanks again, have a great afternoon,

-Ward

> I've had a persistent problem with this regression-test when i build
> netCDF-Fortran with the PGI compiler:
> 
> *FAIL: run_f90_par_test.sh*
> ============================================================================
> Testsuite summary for netCDF-Fortran 4.5.2
> ============================================================================
> # TOTAL: 16
> # PASS:  15
> # SKIP:  0
> # XFAIL: 0
> *# FAIL:  1*
> # XPASS: 0
> # ERROR: 0
> ============================================================================
> See nf03_test4/test-suite.log
> Please report to address@hidden
> ============================================================================
> 
> Here's the error-message
> 
> make  check-TESTS
> make[2]: Entering directory
> '/home/demouser/install/017.NetCDF-F/distro/nf03_test4'
> /usr/bin/mpifort -I../fortran -fPIC -m64 -tp=px -noswitcherror -c -o
> f90tst_parallel_fill.o  f90tst_parallel_fill.f90
> *PGF90-S-0034-Syntax error at or near end of line
> (f90tst_parallel_fill.f90: 51)*
> PGF90/x86-64 Linux 19.9-0: compilation completed with severe errors
> Makefile:881: recipe for target 'f90tst_parallel_fill.o' failed
> make[2]: [f90tst_parallel_fill.o] Error 2 (ignored)
> /bin/bash ../libtool  --tag=FC   --mode=link /usr/bin/mpifort
> -I../fortran -fPIC -m64 -tp=px -noswitcherror
> -L/usr/local/cuda-10.1/lib64 -L/usr/lib -L/usr/lib/x86_64-linux-gnu
> -o f90tst_parallel_fill f90tst_parallel_fill.o
> ../fortran/libnetcdff.la -lnetcdf -ldl -lm
> libtool: link: /usr/bin/mpifort -I../fortran -fPIC -m64 -tp=px
> -noswitcherror -o .libs/f90tst_parallel_fill f90tst_parallel_fill.o 
> -L/usr/local/cuda-10.1/lib64 -L/usr/lib -L/usr/lib/x86_64-linux-gnu
> ../fortran/.libs/libnetcdff.so /usr/lib/libnetcdf.so -ldl -lm
> /usr/bin/ld: cannot find f90tst_parallel_fill.o: No such file or
> directory
> Makefile:826: recipe for target 'f90tst_parallel_fill' failed
> make[2]: [f90tst_parallel_fill] Error 2 (ignored)
> 
> The line in question is here
> 
> 50   if (my_rank .eq. 0) then
> *51      print *,*
> 52      print *, '*** Testing fill values with parallel I/O.'
> 53   endif
> 
> Evidently the PGI compiler doesn't like the dangling comma at the end of
> the line.
> I haven't seen this error building with the GCC compiler, but likely the
> PGI is adhering more strictly to the Fortran standard.
> Can you fix this? It's the only case that fails for me.
> 
> -----------------------------------------------------------------------------------
> This email message is for the sole use of the intended recipient(s) and may 
> contain
> confidential information.  Any unauthorized review, use, disclosure or 
> distribution
> is prohibited.  If you are not the intended recipient, please contact the 
> sender by
> reply email and destroy all copies of the original message.
> -----------------------------------------------------------------------------------
> 
> 


Ticket Details
===================
Ticket ID: CUL-335078
Department: Support netCDF
Priority: Normal
Status: Closed
===================
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata 
inquiry tracking system and then made publicly available through the web.  If 
you do not want to have your interactions made available in this way, you must 
let us know in each email you send to us.