Compilation of Abinit-5.8.4 on Opteron with pgf95

[YKURITA]

Dear Moderators
I am studying physics in Tokyo University of Science.
For study of many body effects,I want to use your program "Abinit-5.8.4".
Our university center machines have complex system program configuration,so Abinit's auto-configurator
often falls in many troubles.
I guess the most easiest way to build executable is on the SMP machines.
But I am still struggling in trouble fixing after 10 months fighting.
Would you mind helping me with your kind suggestions?
I attatch the error file of "make multi".
The system I am using is IBM x 3755 ,Dual core AMD Opeteron 28GHz(totally 8 cores).
Compilers: GCC and pgf95
Your kind assistance would be very much appreciated.
Thank you very much and best regards
Y.Kurita(YKURITA)
PGF90-W-0435-Array declared with zero size (denfgr.F90: 100)
0 inform, 1 warnings, 0 severes, 0 fatal for denfgr
pgf95-Fatal-/opt/pgi/linux86-64/7.1-6/bin/pgf902 TERMINATED by signal 11
Arguments to /opt/pgi/linux86-64/7.1-6/bin/pgf902
/opt/pgi/linux86-64/7.1-6/bin/pgf902 /tmp/pgf95C0RupwHMAkz.ilm -fn m_electrons.F90 -debug -x 120 0x200 -opt 2 -terse 1 -inform warn -x 51 0x20 -x 119 0xa10000 -x 122 0x40 -x 123 0x1000 -x 127 4 -x 127 17 -x 19 0x400000 -x 28 0x40000 -x 70 0x8000 -x 122 1 -quad -x 59 4 -x 59 4 -tp k8-64e -x 124 0x1400 -y 15 2 -x 57 0x3b0000 -x 58 0x48000000 -x 49 0x100 -x 120 0x200 -astype 0 -x 124 1 -cmdline '+pgf95 m_electrons.F90 -DHAVE_CONFIG_H -I. -I../.. -I../../src/61_ionetcdf -I../../src/61_ionetcdf -I../../src/32_util -I../../src/32_util -I../../src/67_common -I../../src/67_common -I../../src/28_numeric_noabirule -I../../src/28_numeric_noabirule -I../../src/51_manage_mpi -I../../src/51_manage_mpi -I../../src/65_nonlocal -I../../src/65_nonlocal -I../../src/16_hideleave -I../../src/16_hideleave -I../../src/interfaces -I../../src/interfaces -I../../src/59_io_mpi -I../../src/59_io_mpi -I../../src/62_occeig -I../../src/62_occeig -I../../src/56_recipspace -I../../src/56_recipspace -I../../src/56_xc -I../../src/56_xc -I../../src/42_geometry -I../../src/42_geometry -I../../src/53_ffts -I../../src/53_ffts -I../../src/66_wfs -I../../src/66_wfs -I../../src/62_iowfdenpot -I../../src/62_iowfdenpot -I../../src/18_timing -I../../src/18_timing -I../../src/12_hide_mpi -I../../src/12_hide_mpi -I../../src/10_defs -I../../src/10_defs -I../../src/14_hidewrite -I../../src/14_hidewrite -I../../src/66_paw -I../../src/66_paw -I../../src/57_iovars -I../../src/57_iovars -I../../src/50_abitypes_defs -I../../src/50_abitypes_defs -I../../src/27_toolbox_oop -I../../src/27_toolbox_oop -O2 -g -module /home/prof/hamada/abinit-5.8.4/tmp-modules -I/home/prof/hamada/abinit-5.8.4/tmp-modules -Mextend -c -o m_electrons.o' -asm /tmp/pgf9580R0jiQjFtB.s
make[4]: *** [m_electrons.o] Error 127
make[4]: *** Waiting for unfinished jobs....
make[3]: *** [all-recursive] Error 1
make[2]: *** [all-recursive] Error 1
make[1]: *** [all] Error 2
make: *** [multi] Error 2

[gmatteo]

PGI is not officially supported. For a list of supported compilers see:
viewtopic.php?f=3&t=40

[mverstra]

It looks like you could just add a flag in FCFLAGS_EXTRA for the acceptance of 0 size arrays, providing pgf has this.

However, Matteo is right, the pgi compilers have been unsatisfactory and impossible to support for several years now. gfortran is quite good, and ifort is the standard.

Matthieu