ABINIT Discussion Forums

Hi all...
This issue has been addressed before but in other versions of Abinit. This thread is somehow related to mine here:
http://forum.abinit.org/viewtopic.php?f=3&t=589

However, I think that in the new release the previous "recipe" does not go so well... I did a configure --with-config-file using this script:



and everything was OK (or so it seemed)... I trusted (maybe I shoudn't) the auto-detect feature of abinit in order to find libraries and includes, that's why I just declared the libraries by name and not by _libs and/or _incs. Apparently, after checking the config.log, those libraries were correctly detected by abinit.

Afterwards, make mj4 ran for a few minutes and then this message, just in the making of wannier90



I tried the recipe found in http://forum.abinit.org/viewtopic.php?f=3&t=589 and did not work, also compiled the wannier90-1.2 and it didn't also... however, wannier90-2.0 did compile smoothly but, when I tried to play dirty (copying the new 2.0 .a and .x files instead of the 1.2) and "touching" the stamps files for wannier90, the same error appeared time after time...

Currently, I know that doing all these with the GNU environment does not make such fuss... but intel libs are worth the pain in the neck.

Below, some info on my platform:
cpuinfo gives:

Intel(R) processor family information utility, Version 4.1 Update 3 Build 20131205
Copyright (C) 2005-2013 Intel Corporation. All rights reserved.

===== Processor composition =====
Processor name : Intel(R) Core(TM) i7 Q 720
Packages(sockets) : 1
Cores : 4
Processors(CPUs) : 8
Cores per package : 4
Threads per core : 2

This is a Dell Precision M6500 running Linux Mint 16 Mate x64 with 32 GB of RAM. All my compilers are "Intel(R) 64 Compiler XE for applications running on Intel(R) 64, Version 14.0.1.106 Build 20131008", the MKL is 11.0 and the MPI Library Runtime Environment for Linux is installed.

Hope it may get solved.
Thanks in advance guys!!

Could you attach ~abinit/<builddir>/fallbacks/source/wannier90-1.2/config.log?

Thanks for yor reply Yann...

Unfortunately, I deleted that config and tried to reconfigure and rebuild. I did it just last night... Now thigs worked with the following build.ac:



The output for the wannier90 config.log is attached below...

Now, I am running the full tests using the runtests.py script and I am seeing the same "returned exit_code: 174" error in BIGDFT t06, t14, FOX t02, LIBXC t01, t81, TUTORIAL tbase2_5, V1 t28, t29, t19, V2 t40, t41... besides, there is a "failed: erroneous lines 32 > 24" in BIGDFT t31...

Maybe I built a buggy abinit?? Have to wait till the end of the Tests_Suite but, up to now, these are the news.

Thanks in advance... any help will be appreciated.

Now all the Tests_Suite have finished... with 30 failed tests, most of them by the infamous 174 return code described before... Below is the txt with reports.
I think I will not do a "make install" until a "convincent" build of abinit has been attained... and it looks too buggy to me. I will explore another workarounds in the meantime...
Any idea?

We need more info to pinpoint the problem.

Each test redirects the standard error to a file named t[test_num].stderr
Do you have some message in these files?

If the stderr is empty, you can try to attach a debugger to Abinit.
Example:

cd tests/Test_suite/v3_t30
# Run abinit under the control of the GNU debugger
gdb path_to_abinit_executable
# wait for the gdb prompt ...
(gdb) run < t30.stdin
# If the test fails, use the backtrace command and post the output
(gdb) backtrace

Hi, thanks again...
Here are the stderr of every single "failed". The HTML auto-report was very helpful and informative also... but the stderrs registered, as you may see, a SIGSEGV segmentation fault for almost every error.
I suspect that is an MKL issue...
Need more info??
Hugs!!

BIGDFT Tests Suite

(01) Standard Error of test t06:
TODO, remove me
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(02) Standard Error of test t14:
TODO, remove me
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(03) Standard Error of test t31:
TODO, remove me
TODO, remove me

FOX Tests Suite

(04) Standard Error of test t02:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

LIBXC Tests Suite

(05) Standard Error of test t01:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(06) Standard Error of test t81:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

TUTORIAL Tests Suite

(07) Standard Error of test tbase2_5:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

V1 Tests Suite

(08) Standard Error of test t19:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(09) Standard Error of test t28:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(10) Standard Error of test t29:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

V2 Tests Suite

(11) Standard Error of test t40:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(12) Standard Error of test t41:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

V3 Tests Suite

(13) Standard Error of test t08:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(14) Standard Error of test t16:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(15) Standard Error of test t18:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(16) Standard Error of test t70:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(17) Standard Error of test t71:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(18) Standard Error of test t72:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(19) Standard Error of test t73:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(20) Standard Error of test t76:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(21) Standard Error of test t82:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

V4 Tests Suite

(22) Standard Error of test t58:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

V5 Tests Suite

(23) Standard Error of test t07:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(24) Standard Error of test t29:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(25) Standard Error of test t48:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(26) Standard Error of test t55:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

V67MBPT Tests Suite

(27) Standard Error of test t03:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

(28) Standard Error of test t14:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

VDWXC Tests Suite

(29) Standard Error of test t10:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

WANNIER90 Tests Suite

(30) Standard Error of test t12:
forrtl: severe (174): SIGSEGV, segmentation fault occurred

And backtracing the error for the BIGDFT t06 gives:

#0 0x000000000181f75b in xcpbe (exci=...,
npts=<error reading variable: Cannot access memory at address 0x40>,
nspden=<error reading variable: Cannot access memory at address 0x40>,
option=0,
order=<error reading variable: Cannot access memory at address 0x8f1>,
rho_updn=..., vxci=..., ndvxci=0, ngr2=1, nd2vxci=0, d2vxci=...,
dvxcdgr=..., dvxci=...,
exexch=<error reading variable: Cannot access memory at address 0x0>,
grho2_updn=...) at ../../../7.4.3/src/56_xc/xcpbe.F90:537
#1 0x00000000017edd9b in drivexc (exc=...,
ixc=<error reading variable: Cannot access memory at address 0x40>,
npts=<error reading variable: Cannot access memory at address 0x40>,
nspden=0,
order=<error reading variable: Cannot access memory at address 0x8f1>,
rho_updn=..., vxcrho=..., ndvxc=0, ngr2=1, nd2vxc=0, nvxcgrho=3, dvxc=...,
d2vxc=..., grho2_updn=..., vxcgrho=...,
exexch=<error reading variable: Cannot access memory at address 0x0>,
lrho_updn=..., vxclrho=..., tau_updn=..., vxctau=..., xc_tb09_c=0)
at ../../../7.4.3/src/56_xc/drivexc.F90:429
#2 0x00000000022312a4 in MODULE_XC::xc_getvxc (
npts=<error reading variable: Cannot access memory at address 0x23c10>,
exc=...,
nspden=<error reading variable: Cannot access memory at address 0x40>,
---Type <return> to continue, or q <return> to quit---
rho=..., vxc=..., grho2=..., vxcgr=..., dvxci=...) at xc.f90:358
#3 0x0000000001f21ec4 in xc_energy_new (
geocode=<error reading variable: Cannot access memory at address 0x23c10>,
m1=<error reading variable: Cannot access memory at address 0x40>,
m3=<error reading variable: Cannot access memory at address 0x40>, nxc=0,
nwb=<error reading variable: Cannot access memory at address 0x8f1>,
nxt=<error reading variable: Cannot access memory at address 0x1>,
nwbl=36, nwbr=-1597734584, nxcl=0, nxcr=0, ixc=1,
hx=4.9406564584124654e-324, hy=2.1219957963999944e-314,
hz=0.15747717774047026, rho=..., gradient=..., vxci=...,
exc=3.0484682560998278e-44, vxc=1.0000000000000004e-20, order=0, ndvxc=0,
dvxci=..., nspden=0,
wbstr=<error reading variable: Cannot access memory at address 0x100>,
.tmp.GEOCODE.len_V$2c60=51690420) at PSolver/xcenergy.f90:875
#4 0x0000000001f1c2e5 in xc_potential (
geocode=<error reading variable: Cannot access memory at address 0x23c10>,
datacode=<error reading variable: Cannot access memory at address 0x40>,
iproc=<error reading variable: Cannot access memory at address 0x40>,
nproc=0,
n01=<error reading variable: Cannot access memory at address 0x8f1>,
n02=<error reading variable: Cannot access memory at address 0x1>,
n03=<error reading variable: Cannot access memory at address 0x24000000f0>, ixc=<error reading variable: Cannot access memory at address 0x1300000078>,
---Type <return> to continue, or q <return> to quit---
hx=<error reading variable: Cannot access memory at address 0x7900000013>,
hy=<error reading variable: Cannot access memory at address 0x53>,
hz=<error reading variable: Cannot access memory at address 0x2400000084>,
rho=...,
exc=<error reading variable: Cannot access memory at address 0x100000024>,
vxc=<error reading variable: Cannot access memory at address 0x1>,
nspin=<error reading variable: Cannot access memory at address 0x100000000>,
rhocore=<error reading variable: Cannot access memory at address 0x2400000024>, potxc=..., xcstr=..., dvxcdrho=..., rhohat=..., .tmp.GEOCODE.len_V$2716=0,
.tmp.DATACODE.len_V$2718=-41584) at PSolver/xcenergy.f90:528
#5 0x000000000253c4fd in updatepotential (
iproc=<error reading variable: Cannot access memory at address 0x23c10>,
nproc=<error reading variable: Cannot access memory at address 0x40>,
geocode=<error reading variable: Cannot access memory at address 0x40>,
ixc=0,
nspin=<error reading variable: Cannot access memory at address 0x8f1>,
hxh=<error reading variable: Cannot access memory at address 0x1>,
hyh=0.17499999999999999, hzh=0.15747717774047026, glr=..., denspot=...,
ehart=0, eexcu=0, vexcu=0, .tmp.GEOCODE.len_V$2796=1)
at linear/energyAndForces.f90:91
#6 0x000000000217b205 in input_wf_diag (
iproc=<error reading variable: Cannot access memory at address 0x23c10>,
---Type <return> to continue, or q <return> to quit---
nproc=<error reading variable: Cannot access memory at address 0x40>,
at=<error reading variable: Cannot access memory at address 0x40>,
denspot=...,
orbs=<error reading variable: Cannot access memory at address 0x8f1>,
nvirt=<error reading variable: Cannot access memory at address 0x1>,
comms=..., lzd=..., energs=..., rxyz=..., nlpspd=..., proj=..., ixc=11,
psi=0x5e6ca30, hpsi=0x0, psit=0x0, g=..., nspin=1, symobj=..., gpu=...,
input=..., onlywf=.FALSE., proj_g=...,
paw=<error reading variable: Cannot access memory at address 0x0>)
at init.f90:1755
#7 0x000000000061d896 in wvl_wfsinp_scratch (
dtset=<error reading variable: Cannot access memory at address 0x23c10>,
mpi_enreg=<error reading variable: Cannot access memory at address 0x40>,
occ=..., rprimd=...,
wvl=<error reading variable: Cannot access memory at address 0x8f1>,
xred=...) at ../../../7.4.3/src/79_seqpar_mpi/wvl_wfsinp_scratch.F90:180
#8 0x00000000005e5a92 in inwffil (
ask_accurate=<error reading variable: Cannot access memory at address 0x23c10>, cg=...,
dtset=<error reading variable: Cannot access memory at address 0x40>,
ecut=0,
ecut_eff=<error reading variable: Cannot access memory at address 0x8f1>,
eigen=..., exchn2n3d=0, formeig=0, gmet=..., hdr=..., ireadwf=0,
---Type <return> to continue, or q <return> to quit---
istwfk=..., kg=..., kptns=..., localrdwf=1, mband=4, mcg=0, mkmem=-1,
mpi_enreg=..., mpw=0, nband=..., ngfft=..., nkpt=1, npwarr=..., nsppol=1,
nsym=1, occ=..., optorth=1, rprimd=..., symafm=..., symrel=..., tnons=...,
unkg=17, wff1=..., wffnow=..., unwff1=1, unwfnow=11, wffnm=...,
wft1nm=..., wvl=..., .tmp.WFFNM.len_V$1058a=264,
.tmp.WFT1NM.len_V$1058c=264)
at ../../../7.4.3/src/79_seqpar_mpi/inwffil.F90:734
#9 0x00000000004fe7f3 in gstate (
acell=<error reading variable: Cannot access memory at address 0x23c10>,
amu=...,
codvsn=<error reading variable: Cannot access memory at address 0x40>,
cpui=0,
dtfil=<error reading variable: Cannot access memory at address 0x8f1>,
dtset=<error reading variable: Cannot access memory at address 0x1>,
iexit=0, initialized=0, mpi_enreg=..., npwtot=..., occ=..., pawang=...,
pawrad=..., pawtab=..., psps=..., results_gs=..., rprim=...,
scf_history=..., vel=..., vel_cell=..., xred=...,
.tmp.CODVSN.len_V$44ac2=6) at ../../../7.4.3/src/95_drive/gstate.F90:647
#10 0x0000000000423167 in gstateimg (acell_img=..., amu_img=...,
codvsn=<error reading variable: Cannot access memory at address 0x40>,
cpui=0,
dtfil=<error reading variable: Cannot access memory at address 0x8f1>,
dtset=<error reading variable: Cannot access memory at address 0x1>,
---Type <return> to continue, or q <return> to quit---
etotal_img=..., fcart_img=..., fred_img=..., iexit=0, mixalch_img=...,
mpi_enreg=..., nimage=1, npwtot=..., occ_img=..., pawang=..., pawrad=...,
pawtab=..., psps=..., rprim_img=..., strten_img=..., vel_cell_img=...,
vel_img=..., xred_img=..., filnam=..., filstat=..., idtset=1, jdtset=...,
ndtset=0, .tmp.CODVSN.len_V$cc6b=6, .tmp.FILSTAT.len_V$ccd2=264)
at ../../../7.4.3/src/95_drive/gstateimg.F90:421
#11 0x0000000000411700 in driver (
codvsn=<error reading variable: Cannot access memory at address 0x23c10>,
cpui=<error reading variable: Cannot access memory at address 0x40>,
dtsets=..., filnam=...,
filstat=<error reading variable: Cannot access memory at address 0x8f1>,
mpi_enregs=..., ndtset=0, ndtset_alloc=1, npsp=2, pspheads=...,
results_out=..., .tmp.CODVSN.len_V$fd04=6, .tmp.FILNAM.len_V$fd0b=264,
.tmp.FILSTAT.len_V$fd0d=264) at ../../../7.4.3/src/95_drive/driver.F90:653
#12 0x0000000000407da5 in abinit ()
at ../../../7.4.3/src/98_main/abinit.F90:440

The Intel debugger, for the same test file, gives out

Reading symbols from /home/amaury/Downloads/abinit/Build.v1_intel14-nompi-mkl/src/98_main/abinit...done.
(idb) run < t06.stdin
Starting program: /home/amaury/Downloads/abinit/Build.v1_intel14-nompi-mkl/src/98_main/abinit
[New Thread 18883 (LWP 18883)]
Program received signal SIGSEGV
xcpbe (exci=(...), npts=<no value>, nspden=<no value>, option=0, order=<no value>, rho_updn=(...), vxci=(...), ndvxci=0, ngr2=1, nd2vxci=0, d2vxci=(...), dvxcdgr=(...), dvxci=(...), exexch=<no value>, grho2_updn=(...)) at /home/amaury/Downloads/abinit/7.4.3/src/56_xc/xcpbe.F90:537
537 if(exexch==1) cycle

And backtracing from idb gives out:

(idb) backtrace
0x000000000181f75b in xcpbe (exci=(...), npts=<no value>, nspden=<no value>, option=0, order=<no value>, rho_updn=(...), vxci=(...), ndvxci=0, ngr2=1, nd2vxci=0, d2vxci=(...), dvxcdgr=(...), dvxci=(...), exexch=<no value>, grho2_updn=(...)) "/home/amaury/Downloads/abinit/7.4.3/src/56_xc/xcpbe.F90":537
0x00000000017edd9b in drivexc (exc=(...), ixc=<no value>, npts=<no value>, nspden=0, order=<no value>, rho_updn=(...), vxcrho=(...), ndvxc=0, ngr2=1, nd2vxc=0, nvxcgrho=3, dvxc=(...), d2vxc=(...), grho2_updn=(...), vxcgrho=(...), exexch=<no value>, lrho_updn=(...), vxclrho=(...), tau_updn=(...), vxctau=(...), xc_tb09_c=0) "/home/amaury/Downloads/abinit/7.4.3/src/56_xc/drivexc.F90":429
0x00000000022312a4 in MODULE_XC::xc_getvxc (npts=<no value>, exc=(...), nspden=<no value>, rho=(...), vxc=(...), grho2=<no value>, vxcgr=(...), dvxci=(...)) "/home/amaury/Downloads/abinit/Build.v1_intel14-nompi-mkl/fallbacks/sources/bigdft-abi-1.0.4/src/modules/xc.f90":358
0x0000000001f21ec4 in xc_energy_new (geocode=<no value>, m1=<no value>, m3=<no value>, nxc=0, nwb=<no value>, nxt=<no value>, nwbl=36, nwbr=-1597734584, nxcl=0, nxcr=0, ixc=1, hx=4.9406564584124654e-324, hy=2.1219957963999944e-314, hz=0.15747717774047026, rho=(...), gradient=(...), vxci=(...), exc=3.0484682560998278e-44, vxc=1.0000000000000004e-20, order=0, ndvxc=0, dvxci=(...), nspden=0, wbstr=<no value>) "/home/amaury/Downloads/abinit/Build.v1_intel14-nompi-mkl/fallbacks/sources/bigdft-abi-1.0.4/src/PSolver/xcenergy.f90":875
0x0000000001f1c2e5 in xc_potential (geocode=<no value>, datacode=<no value>, iproc=<no value>, nproc=0, n01=<no value>, n02=<no value>, n03=<no value>, ixc=<no value>, hx=<no value>, hy=<no value>, hz=<no value>, rho=<no value>, exc=<no value>, vxc=<no value>, nspin=<no value>, rhocore=(...), potxc=<no value>, xcstr=(...), dvxcdrho=(...), rhohat=(...)) "/home/amaury/Downloads/abinit/Build.v1_intel14-nompi-mkl/fallbacks/sources/bigdft-abi-1.0.4/src/PSolver/xcenergy.f90":528

For me, I am starting to get into the "advanced chinese course" territory... In other words, I'm a little bit confused now.

Hope this info could help

The segmentation fault is due to a wrong usage of optional arguments.
Strangely this bug has never triggered a SIGSEGV in our test farm (the same error is present in version 7.7!)

Open 56_xc/drivexc.F90, go to line 429 and call xcpbe without the exexch argument if the dummy variable is not present, i.e.:

=== modified file 'src/41_xc_lowlevel/drivexc.F90'
--- src/41_xc_lowlevel/drivexc.F90 2014-01-17 00:34:33 +0000
+++ src/41_xc_lowlevel/drivexc.F90 2014-01-17 23:28:17 +0000
@@ -426,8 +426,13 @@

else if ((ixc >= 11 .and. ixc <= 15) .or. (ixc>=23 .and. ixc<=24)) then
if (order**2 <= 1) then
- call xcpbe(exc,npts,nspden,optpbe,order,rho_updn,vxcrho,ndvxc,ngr2,nd2vxc,&
-& dvxcdgr=vxcgrho,exexch=exexch,grho2_updn=grho2_updn)
+ if (present(exexch)) then
+ call xcpbe(exc,npts,nspden,optpbe,order,rho_updn,vxcrho,ndvxc,ngr2,nd2vxc,&
+& dvxcdgr=vxcgrho,exexch=exexch,grho2_updn=grho2_updn)
+ else
+ call xcpbe(exc,npts,nspden,optpbe,order,rho_updn,vxcrho,ndvxc,ngr2,nd2vxc,&
+& dvxcdgr=vxcgrho,grho2_updn=grho2_updn)
+ end if
else if (order /=3) then
if(ixc==12 .or. ixc==24)then


Let me know if this solves the problem with the tests!

same errors...

just double-checking: I did what you said in the Bigdft drivexc.F90 file under src/56_xc of the build directory... more explicitly:
these are the lines from 406 to 419...


and the file was located at /home/amaury/Downloads/abinit/Build.v1_intel14-nompi-mkl/fallbacks/sources/bigdft-abi-1.0.4/libABINIT/src/56_xc

is it right??

Sorry, I was posting a patch done on top of my 7.7 version. Your file comes from an old version and is slightly different
The new code should read:

else if ((ixc >= 11 .and. ixc <= 15) .or. (ixc==23)) then
if (order**2 <= 1) then
if (present(exexch)) then
call xcpbe(exc,npts,nspden,optpbe,order,rho_updn,vxcrho,ndvxc,ngr2,nd2vxc,&
& dvxcdgr=vxcgrho,exexch=exexch,grho2_updn=grho2_updn)
else
call xcpbe(exc,npts,nspden,optpbe,order,rho_updn,vxcrho,ndvxc,ngr2,nd2vxc,&
& dvxcdgr=vxcgrho,grho2_updn=grho2_updn)
end if
else if (order /=3) then

Thanks, did it again as you said but the same remains the same

Could you try version 7.6.1? The bug is still there but maybe we can reduce the number of failures and I can try
to fix the problems before the release of 7.6.2. It's very strange that I cannot reproduce your problem with ifort

Could you start with the simplest configuration:

prefix=/opt/abinit/7.4.3/build_no1
enable_64bit_flags="yes"
enable_gw_dpc="yes"
enable_mpi="no"
CC="icc"
CXX="icpc"
FC="ifort"
with_trio_flavor="netcdf+etsf_io+fox"
with_dft_flavor="atompaw+bigdft+libxc+wannier90"

then we'll try to enable MKL and OpenMp.

OK, I'll try...

BTW, I did not know there was a 7.6.1 abinit... just cooked!!

Thank you very very much for your kind attention

Best regards!!

However, one thing is clear:
At least in my platform:
1. mixing GNU and Intel (iCompilers + ATLAS + fftw3, for instance) brings up the very same bug
2. working with pure Intel resources (iCompilers + MKL + fftw3-mkl) brings up the bug
3. working with pure GNU (gCompilers + ATLAS + fftw3) results in the most stable Abinit
But computation times with iCompilers + MKL improve a lot...
Let us try with the new baby and post the results...

I have tested the new abinit with the following config file:



./configure went well

* test AtomPAW : yes
* test BigDFT : yes
* test ETSF_IO : yes
* test FoX : yes
* test GPU : no
* test LibXC : yes
* test MPI : no
* test Wannier90 : yes

* timeout : no (0 seconds)

* MPI_RUNNER =
* MPI_2PROCS =
* MPI runner present : no
* MPI runner working : unknown

Summary of important options:

* C compiler : intel version 14.0
* Fortran compiler: intel version 14.0
* architecture : unknown unknown (64 bits)

* debugging : basic
* optimizations : standard

* OpenMP enabled : no (collapse: ignored)
* MPI enabled : no
* MPI-IO enabled : no
* GPU enabled : no (flavor: none)

* TRIO flavor = netcdf-fallback+etsf_io-fallback+fox-fallback
* TIMER flavor = abinit (libs: ignored)
* LINALG flavor = mkl (libs: auto-detected)
* ALGO flavor = none (libs: ignored)
* FFT flavor = fftw3-mkl (libs: auto-detected)
* MATH flavor = none (libs: ignored)
* DFT flavor = libxc-fallback+atompaw-fallback+bigdft-fallback+wannier90-fallback

then make mj4 also went well

then make check, with the very same old problems...

with returned exit_code: 174

1 [bigdft][t06] fldiff.pl fatal error
2 [bigdft][t14] fldiff.pl fatal error
3 [fox][t02] fldiff.pl fatal error
4 [libxc][t01] fldiff.pl fatal error
5 [libxc][t81] fldiff.pl fatal error
6 [tutorial][tbase2_5] fldiff.pl fatal error
7 [v1][t28] fldiff.pl fatal error
8 [v1][t29] fldiff.pl fatal error
9 [v1][t19] fldiff.pl fatal error
10 [v2][t40] fldiff.pl fatal error
11 [v2][t41] fldiff.pl fatal error
12 [v3][t08] fldiff.pl fatal error
13 [v3][t16] fldiff.pl fatal error
14 [v3][t18] fldiff.pl fatal error
15 [v3][t72] fldiff.pl fatal error
16 [v3][t71] fldiff.pl fatal error:
17 [v3][t70] fldiff.pl fatal error
18 [v3][t73] fldiff.pl fatal error:
19 [v3][t76] fldiff.pl fatal error:
20 [v3][t82] fldiff.pl fatal error:
21 [v4][t58] fldiff.pl fatal error:
22 [v5][t07] fldiff.pl fatal error:
23 [v5][t29] fldiff.pl fatal error:
24 [v5][t48] fldiff.pl fatal error:
25 [v5][t55] fldiff.pl fatal error:
26 [v67mbpt][t03] fldiff.pl fatal error:
27 [v67mbpt][t14] fldiff.pl fatal error
28 [v7][t06] fldiff.pl fatal error
29 [vdwxc][t10] fldiff.pl fatal error:
30 [wannier90][t12] fldiff.pl fatal error

Then I went again with the configuration file (in this case the mkl libs are explicitly declared)



and the result was also the same: good "./configure", good "make mj4" but "make check" returned the 30 errors already reported above.

Up to this very moment, my best results (time & accuracy) are obtained mixing GNU compilers with Intel MKL libs using the following config file:



A pure gnu compilation, as is well known, does not give so much headache and ran very well in my laptop.

DELL, Precision M6500, Intel(R) Core(TM) i7 Q 720, 32 GB RAM
Linux Mint 13, 64 bits, 3.2.0-23-generic kernel
Icompilers 14.0.1.106 Build 20131008
GNUcompilers 4.6.3
MKL 11.0
MPI Library Runtime Environment for Linux

I used this configuration.ac file:



I will be backtracking some of the errors and keeping the eye on the problematic segments of the code... Hope to post something on this subject.

Dear Matteo...
I did this with the new abinit:

gmatteo wrote:Could you try version 7.6.1? The bug is still there but maybe we can reduce the number of failures and I can try
to fix the problems before the release of 7.6.2. It's very strange that I cannot reproduce your problem with ifort

Could you start with the simplest configuration:

prefix=/opt/abinit/7.4.3/build_no1
enable_64bit_flags="yes"
enable_gw_dpc="yes"
enable_mpi="no"
CC="icc"
CXX="icpc"
FC="ifort"
with_trio_flavor="netcdf+etsf_io+fox"
with_dft_flavor="atompaw+bigdft+libxc+wannier90"

then we'll try to enable MKL and OpenMp.

and errors are the same...
sorry!

Hi,

some infos on the post : viewtopic.php?f=2&t=2389#p7569

I hope it can help you...

jmb

Hi... thanks for your attention, jbeuken!!

Unfortunately, the same errors...

Just as you, I have Open MPI compiled with the intel compilers (with the last versions, by the way)...my environment variables are well defined, etc...
I did the same as you said, except for the paths to fftw and mkl that in my case point to /opt/intel/composer_xe_2013_sp1.1.106/mkl/lib/intel64
Another issue: when configuring, the MPI environment was not well understood by config so I just had to declare FC="mpif90", CC="mpicc" and CXX="mpic++"

config went well, make mj4 also went well... but make check showed the same (bad) results in the same tests:
bigdft_t06 is the first one: with exit code 174, segmentation fault... the same errors I have been reporting in this thread... then bigdft_t14 and so on...

I would like to know what is going on with this setup.

Thank you all guys for your kindness and (lot of) patience...

Best regards.

Hello,

can you post the results of :



jmb

@Jean-Michel

Having info on the environment is useful but I'm pretty sure that the problem is in Abinit (there are several routines that do not handle correctly optional arguments)
The strange thing is that intel14 passes all the tests in our test farm!

Here is the info as requested:

~ $ which mpif90
/usr/local/bin/mpif90

~ $ mpif90 -show
ifort -I/usr/local/include -I/usr/local/lib -L/usr/local/lib -lmpi_f90 -lmpi_f77 -lmpi -ldl -lm -Wl,--export-dynamic -lrt -lnsl -lutil

~ $ mpif90 --version
ifort (IFORT) 14.0.1 20131008
Copyright (C) 1985-2013 Intel Corporation. All rights reserved.

~ $ env | sort
COLORTERM=gnome-terminal
CPATH=/opt/intel/composer_xe_2013_sp1.1.106/mkl/include:/opt/intel/composer_xe_2013_sp1.1.106/tbb/include
DBUS_SESSION_BUS_ADDRESS=unix:abstract=/tmp/dbus-aJmidJUws0,guid=32d68fb64c696e4fca36346700000024
DEFAULTS_PATH=/usr/share/gconf/default.default.path
DESKTOP_SESSION=default
DISPLAY=:0.0
GDB_CROSS=/opt/intel/composer_xe_2013_sp1.1.106/debugger/gdb/intel64_mic/py27/bin/gdb-mic
GDBSERVER_MIC=/opt/intel/composer_xe_2013_sp1.1.106/debugger/gdb/target/mic/bin/gdbserver
GJS_DEBUG_OUTPUT=stderr
GJS_DEBUG_TOPICS=JS ERROR;JS LOG
GNOME_DESKTOP_SESSION_ID=this-is-deprecated
GNOME_KEYRING_CONTROL=/tmp/keyring-RhX3Wb
GNOME_KEYRING_PID=1630
GPG_AGENT_INFO=/tmp/keyring-RhX3Wb/gpg:0:1
HOME=/home/amaury
IDB_HOME=/opt/intel/composer_xe_2013_sp1.1.106/bin/intel64
I_MPI_ROOT=/opt/intel/impi/4.1.3.045
INCLUDE=/opt/intel/composer_xe_2013_sp1.1.106/mkl/include
INTEL_LICENSE_FILE=/opt/intel/composer_xe_2013_sp1.1.106/licenses:/opt/intel/licenses:/home/amaury/intel/licenses
IPPROOT=/opt/intel/composer_xe_2013_sp1.1.106/ipp
LANG=en_US.UTF-8
LD_LIBRARY_PATH=/opt/intel/impi/4.1.3.045/intel64/lib:/opt/intel/composer_xe_2013_sp1.1.106/compiler/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/mpirt/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/ipp/../compiler/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/ipp/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/compiler/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/mkl/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/tbb/lib/intel64/gcc4.4:/opt/intel/impi/4.1.3.045/intel64/lib:/usr/local/lib
LIBGL_DRIVERS_PATH=/usr/lib/fglrx/dri:/usr/lib32/fglrx/dri
LIBRARY_PATH=/opt/intel/composer_xe_2013_sp1.1.106/compiler/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/ipp/../compiler/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/ipp/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/compiler/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/mkl/lib/intel64:/opt/intel/composer_xe_2013_sp1.1.106/tbb/lib/intel64/gcc4.4
LOGNAME=amaury
MANDATORY_PATH=/usr/share/gconf/default.mandatory.path
MANPATH=/opt/intel/composer_xe_2013_sp1.1.106/man/en_US:/opt/intel/composer_xe_2013_sp1.1.106/man/en_US:/opt/intel/impi/4.1.3.045/man:/usr/local/man:/usr/local/share/man:/usr/share/man:
MDM_LANG=en_US.UTF-8
MDMSESSION=default
MDM_XSERVER_LOCATION=local
MIC_LD_LIBRARY_PATH=/opt/intel/composer_xe_2013_sp1.1.106/compiler/lib/mic:/opt/intel/composer_xe_2013_sp1.1.106/mpirt/lib/mic:/opt/intel/composer_xe_2013_sp1.1.106/compiler/lib/mic:/opt/intel/composer_xe_2013_sp1.1.106/mkl/lib/mic:/opt/intel/composer_xe_2013_sp1.1.106/tbb/lib/mic
MIC_LIBRARY_PATH=/opt/intel/composer_xe_2013_sp1.1.106/tbb/lib/mic
MKLROOT=/opt/intel/composer_xe_2013_sp1.1.106/mkl
NLSPATH=/opt/intel/composer_xe_2013_sp1.1.106/compiler/lib/intel64/locale/%l_%t/%N:/opt/intel/composer_xe_2013_sp1.1.106/ipp/lib/intel64/locale/%l_%t/%N:/opt/intel/composer_xe_2013_sp1.1.106/mkl/lib/intel64/locale/%l_%t/%N:/opt/intel/composer_xe_2013_sp1.1.106/debugger/gdb/intel64_mic/py27/share/locale/%l_%t/%N:/opt/intel/composer_xe_2013_sp1.1.106/debugger/gdb/intel64/py27/share/locale/%l_%t/%N:/opt/intel/composer_xe_2013_sp1.1.106/debugger/intel64/locale/%l_%t/%N
PATH=/opt/intel/impi/4.1.3.045/intel64/bin:/opt/intel/composer_xe_2013_sp1.1.106/bin/intel64:/opt/intel/composer_xe_2013_sp1.1.106/mpirt/bin/intel64:/opt/intel/composer_xe_2013_sp1.1.106/debugger/gdb/intel64_mic/py27/bin:/opt/intel/composer_xe_2013_sp1.1.106/debugger/gdb/intel64/py27/bin:/opt/intel/composer_xe_2013_sp1.1.106/bin/intel64:/opt/intel/composer_xe_2013_sp1.1.106/bin/intel64_mic:/opt/intel/composer_xe_2013_sp1.1.106/debugger/gui/intel64:/opt/intel/impi/4.1.3.045/intel64/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/lib:/usr/local/lib
PWD=/home/amaury
SESSION_MANAGER=local/Helena-lnx:@/tmp/.ICE-unix/1707,unix/Helena-lnx:/tmp/.ICE-unix/1707
SHELL=/bin/bash
SHLVL=1
SSH_AGENT_PID=1786
SSH_AUTH_SOCK=/tmp/keyring-RhX3Wb/ssh
TBBROOT=/opt/intel/composer_xe_2013_sp1.1.106/tbb
TERM=xterm
USER=amaury
USERNAME=amaury
_=/usr/bin/env
WINDOWID=41943046
WINDOWPATH=8
XAUTHORITY=/home/amaury/.Xauthority
XDG_CONFIG_DIRS=/etc/xdg/xdg-default:/etc/xdg
XDG_CURRENT_DESKTOP=GNOME
XDG_DATA_DIRS=/usr/share/default:/usr/share/gnome:/usr/local/share/:/usr/share/:/usr/share/mdm/
XDG_SESSION_COOKIE=2b7cf3c2581d8204477753710000000a-1390247317.118983-1841456552

and ths was not requested but it may help...

~ $ ompi_info
Package: Open MPI root@Helena-lnx Distribution
Open MPI: 1.6.5
Open MPI SVN revision: r28673
Open MPI release date: Jun 26, 2013
Open RTE: 1.6.5
Open RTE SVN revision: r28673
Open RTE release date: Jun 26, 2013
OPAL: 1.6.5
OPAL SVN revision: r28673
OPAL release date: Jun 26, 2013
MPI API: 2.1
Ident string: 1.6.5
Prefix: /usr/local
Configured architecture: x86_64-unknown-linux-gnu
Configure host: Helena-lnx
Configured by: root
Configured on: Sun Jan 19 21:40:02 CST 2014
Configure host: Helena-lnx
Built by: root
Built on: Sun Jan 19 21:57:07 CST 2014
Built host: Helena-lnx
C bindings: yes
C++ bindings: yes
Fortran77 bindings: yes (all)
Fortran90 bindings: yes
Fortran90 bindings size: small
C compiler: icc
C compiler absolute: /opt/intel/composer_xe_2013_sp1.1.106/bin/intel64/icc
C compiler family name: INTEL
C compiler version: 1400.20131008
C++ compiler: icpc
C++ compiler absolute: /opt/intel/composer_xe_2013_sp1.1.106/bin/intel64/icpc
Fortran77 compiler: ifort
Fortran77 compiler abs: /opt/intel/composer_xe_2013_sp1.1.106/bin/intel64/ifort
Fortran90 compiler: ifort
Fortran90 compiler abs: /opt/intel/composer_xe_2013_sp1.1.106/bin/intel64/ifort
C profiling: yes
C++ profiling: yes
Fortran77 profiling: yes
Fortran90 profiling: yes
C++ exceptions: no
Thread support: posix (MPI_THREAD_MULTIPLE: no, progress: no)
Sparse Groups: no
Internal debug support: no
MPI interface warnings: no
MPI parameter check: runtime
Memory profiling support: no
Memory debugging support: no
libltdl support: yes
Heterogeneous support: no
mpirun default --prefix: no
MPI I/O support: yes
MPI_WTIME support: gettimeofday
Symbol vis. support: yes
Host topology support: yes
MPI extensions: affinity example
FT Checkpoint support: no (checkpoint thread: no)
VampirTrace support: yes
MPI_MAX_PROCESSOR_NAME: 256
MPI_MAX_ERROR_STRING: 256
MPI_MAX_OBJECT_NAME: 64
MPI_MAX_INFO_KEY: 36
MPI_MAX_INFO_VAL: 256
MPI_MAX_PORT_NAME: 1024
MPI_MAX_DATAREP_STRING: 128
MCA backtrace: execinfo (MCA v2.0, API v2.0, Component v1.6.5)
MCA memory: linux (MCA v2.0, API v2.0, Component v1.6.5)
MCA paffinity: hwloc (MCA v2.0, API v2.0, Component v1.6.5)
MCA carto: auto_detect (MCA v2.0, API v2.0, Component v1.6.5)
MCA carto: file (MCA v2.0, API v2.0, Component v1.6.5)
MCA shmem: mmap (MCA v2.0, API v2.0, Component v1.6.5)
MCA shmem: posix (MCA v2.0, API v2.0, Component v1.6.5)
MCA shmem: sysv (MCA v2.0, API v2.0, Component v1.6.5)
MCA maffinity: first_use (MCA v2.0, API v2.0, Component v1.6.5)
MCA maffinity: hwloc (MCA v2.0, API v2.0, Component v1.6.5)
MCA timer: linux (MCA v2.0, API v2.0, Component v1.6.5)
MCA installdirs: env (MCA v2.0, API v2.0, Component v1.6.5)
MCA installdirs: config (MCA v2.0, API v2.0, Component v1.6.5)
MCA sysinfo: linux (MCA v2.0, API v2.0, Component v1.6.5)
MCA hwloc: hwloc132 (MCA v2.0, API v2.0, Component v1.6.5)
MCA dpm: orte (MCA v2.0, API v2.0, Component v1.6.5)
MCA pubsub: orte (MCA v2.0, API v2.0, Component v1.6.5)
MCA allocator: basic (MCA v2.0, API v2.0, Component v1.6.5)
MCA allocator: bucket (MCA v2.0, API v2.0, Component v1.6.5)
MCA coll: basic (MCA v2.0, API v2.0, Component v1.6.5)
MCA coll: hierarch (MCA v2.0, API v2.0, Component v1.6.5)
MCA coll: inter (MCA v2.0, API v2.0, Component v1.6.5)
MCA coll: self (MCA v2.0, API v2.0, Component v1.6.5)
MCA coll: sm (MCA v2.0, API v2.0, Component v1.6.5)
MCA coll: sync (MCA v2.0, API v2.0, Component v1.6.5)
MCA coll: tuned (MCA v2.0, API v2.0, Component v1.6.5)
MCA io: romio (MCA v2.0, API v2.0, Component v1.6.5)
MCA mpool: fake (MCA v2.0, API v2.0, Component v1.6.5)
MCA mpool: rdma (MCA v2.0, API v2.0, Component v1.6.5)
MCA mpool: sm (MCA v2.0, API v2.0, Component v1.6.5)
MCA pml: bfo (MCA v2.0, API v2.0, Component v1.6.5)
MCA pml: csum (MCA v2.0, API v2.0, Component v1.6.5)
MCA pml: ob1 (MCA v2.0, API v2.0, Component v1.6.5)
MCA pml: v (MCA v2.0, API v2.0, Component v1.6.5)
MCA bml: r2 (MCA v2.0, API v2.0, Component v1.6.5)
MCA rcache: vma (MCA v2.0, API v2.0, Component v1.6.5)
MCA btl: ofud (MCA v2.0, API v2.0, Component v1.6.5)
MCA btl: openib (MCA v2.0, API v2.0, Component v1.6.5)
MCA btl: self (MCA v2.0, API v2.0, Component v1.6.5)
MCA btl: sm (MCA v2.0, API v2.0, Component v1.6.5)
MCA btl: tcp (MCA v2.0, API v2.0, Component v1.6.5)
MCA topo: unity (MCA v2.0, API v2.0, Component v1.6.5)
MCA osc: pt2pt (MCA v2.0, API v2.0, Component v1.6.5)
MCA osc: rdma (MCA v2.0, API v2.0, Component v1.6.5)
MCA iof: hnp (MCA v2.0, API v2.0, Component v1.6.5)
MCA iof: orted (MCA v2.0, API v2.0, Component v1.6.5)
MCA iof: tool (MCA v2.0, API v2.0, Component v1.6.5)
MCA oob: tcp (MCA v2.0, API v2.0, Component v1.6.5)
MCA odls: default (MCA v2.0, API v2.0, Component v1.6.5)
MCA ras: cm (MCA v2.0, API v2.0, Component v1.6.5)
MCA ras: loadleveler (MCA v2.0, API v2.0, Component v1.6.5)
MCA ras: slurm (MCA v2.0, API v2.0, Component v1.6.5)
MCA rmaps: load_balance (MCA v2.0, API v2.0, Component v1.6.5)
MCA rmaps: rank_file (MCA v2.0, API v2.0, Component v1.6.5)
MCA rmaps: resilient (MCA v2.0, API v2.0, Component v1.6.5)
MCA rmaps: round_robin (MCA v2.0, API v2.0, Component v1.6.5)
MCA rmaps: seq (MCA v2.0, API v2.0, Component v1.6.5)
MCA rmaps: topo (MCA v2.0, API v2.0, Component v1.6.5)
MCA rml: oob (MCA v2.0, API v2.0, Component v1.6.5)
MCA routed: binomial (MCA v2.0, API v2.0, Component v1.6.5)
MCA routed: cm (MCA v2.0, API v2.0, Component v1.6.5)
MCA routed: direct (MCA v2.0, API v2.0, Component v1.6.5)
MCA routed: linear (MCA v2.0, API v2.0, Component v1.6.5)
MCA routed: radix (MCA v2.0, API v2.0, Component v1.6.5)
MCA routed: slave (MCA v2.0, API v2.0, Component v1.6.5)
MCA plm: rsh (MCA v2.0, API v2.0, Component v1.6.5)
MCA plm: slurm (MCA v2.0, API v2.0, Component v1.6.5)
MCA filem: rsh (MCA v2.0, API v2.0, Component v1.6.5)
MCA errmgr: default (MCA v2.0, API v2.0, Component v1.6.5)
MCA ess: env (MCA v2.0, API v2.0, Component v1.6.5)
MCA ess: hnp (MCA v2.0, API v2.0, Component v1.6.5)
MCA ess: singleton (MCA v2.0, API v2.0, Component v1.6.5)
MCA ess: slave (MCA v2.0, API v2.0, Component v1.6.5)
MCA ess: slurm (MCA v2.0, API v2.0, Component v1.6.5)
MCA ess: slurmd (MCA v2.0, API v2.0, Component v1.6.5)
MCA ess: tool (MCA v2.0, API v2.0, Component v1.6.5)
MCA grpcomm: bad (MCA v2.0, API v2.0, Component v1.6.5)
MCA grpcomm: basic (MCA v2.0, API v2.0, Component v1.6.5)
MCA grpcomm: hier (MCA v2.0, API v2.0, Component v1.6.5)
MCA notifier: command (MCA v2.0, API v1.0, Component v1.6.5)
MCA notifier: syslog (MCA v2.0, API v1.0, Component v1.6.5)

Hi,

…
Another issue: when configuring, the MPI environment was not well understood by config so I just had to declare FC="mpif90", CC="mpicc" and CXX="mpic++"
...

...
LD_LIBRARY_PATH=/opt/intel/impi/4.1.3.045/intel64/lib:<SNIP>…</SNIP>:/usr/local/lib
…
PATH=/opt/intel/impi/4.1.3.045/intel64/bin:/opt/intel/composer_xe_2013_sp1.1.106/bin/intel64:<SNIP>…</SNIP>
:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/lib:/usr/local/lib

the issue with MPI is perhaps due to the mixing of 2 (or 3 ) MPI flavors : impi, OpenMPI/intel ( and OpenMPI/gcc system package ? )

even if it does not solve the problem, it's more clean to write the paths as :



alternatively, you could try with impi :



jmb

Dear jbeuken,

In fact, I have only 2 flavors: Open MPI + GNU and Open MPI + Intel... IMPI is just installed as an environment since I don have a license for Intel MPI compilers.
With OpenMPI/GNU + MKL libs (incl FFTW3-MKL) abinit runs fine here.
With OpenMP/GNU + ATLAS libs (incl FFTW3) abinit also runs fine here but slower.
My pain in the neck comes when trying to compile in the environment OpenMPI/Intel + MKL (incl FFTW3-MKL)... a smoother running configuration for my native platform...