Re: [QE-users] GIPAW Segmentation Fault

[Ben Comer]

Thanks for the input, removing that flag did not fix the problem 
unfortunately. I was using those flags because I had just copied an 
install script from our cluster's support staff. However, I was able to 
get everything working by compiling with gcc (though this may not be 
optimal). I believe there is an issue in the intel libraries on our 
cluster. Thanks for the help!



On 08/09/2018 03:33 PM, Davide Ceresoli wrote:

Dear Ben,
    is your mpif90 using the intel compiler or gfortran?
why are you linking libmkl_gf_lp64 ? that is specific for
gfortran and it might be reason of the crashes, when using
the intel compiler.

Best wishes,
    Davide

On 08/08/2018 06:08 PM, Ben Comer wrote:

Davide,

I am using the intel 17.0 compiler with mkl 11.3. Adding the 
suggested line to the bottom of my make.inc file did not change the 
error unfortunately. I've included a dropbox link containing my build 
script and make.inc. I'm using the QE-GIPAW 6.2.2 release (QE and 
GIPAW bundled).


https://www.dropbox.com/sh/1znj73stc2opkkq/AABVcYoSXKIGlhi293GIB8Zpa?dl=0 





Dear Ben,
    I'm afraid it's a problem with MKL-blas ZDOTC, which must
return a complex(dp) result. Very strange, because if you grep
the source code, we have declared it: complex(dp), external::zdtoc

Can you tell us your compiler and MKL version? can you add
DFLAGS+=-Dzdotc=zdotc_wrapper
to the QE make.inc and recompile both (QE and GIPAW)?

Best wishes,
    Davide

On 08/06/2018 03:19 PM, Ben Comer wrote:
Hello,

I've been trying to do g factor calculations in GIPAW working for a 
few days now. I keep getting a segmentation fault (below) no matter 
how I compile it on our cluster. Does anyone have a good idea of 
what might be causing this?



forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image  PC    Routine Line Source
gipaw.x    00C40604  Unknown Unknown Unknown
libpthread-2.12.s  00328DE0F7E0  Unknown Unknown Unknown
libmkl_avx2.so 2AAAB7DA5CA3  mkl_blas_avx2_zdo Unknown  
Unknown


Thanks,

Ben Comer

Georgia Tech










___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] GIPAW Segmentation Fault

[Davide Ceresoli]

Dear Ben,
is your mpif90 using the intel compiler or gfortran?
why are you linking libmkl_gf_lp64 ? that is specific for
gfortran and it might be reason of the crashes, when using
the intel compiler.

Best wishes,
Davide

On 08/08/2018 06:08 PM, Ben Comer wrote:

Davide,

I am using the intel 17.0 compiler with mkl 11.3. Adding the suggested line to 
the bottom of my make.inc file did not change the error unfortunately. I've 
included a dropbox link containing my build script and make.inc. I'm using the 
QE-GIPAW 6.2.2 release (QE and GIPAW bundled).


https://www.dropbox.com/sh/1znj73stc2opkkq/AABVcYoSXKIGlhi293GIB8Zpa?dl=0



Dear Ben,
    I'm afraid it's a problem with MKL-blas ZDOTC, which must
return a complex(dp) result. Very strange, because if you grep
the source code, we have declared it: complex(dp), external::zdtoc

Can you tell us your compiler and MKL version? can you add
DFLAGS+=-Dzdotc=zdotc_wrapper
to the QE make.inc and recompile both (QE and GIPAW)?

Best wishes,
    Davide

On 08/06/2018 03:19 PM, Ben Comer wrote:
Hello,

I've been trying to do g factor calculations in GIPAW working for a few days 
now. I keep getting a segmentation fault (below) no matter how I compile it 
on our cluster. Does anyone have a good idea of what might be causing this?



forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image  PC    Routine Line Source
gipaw.x    00C40604  Unknown Unknown  Unknown
libpthread-2.12.s  00328DE0F7E0  Unknown Unknown  Unknown
libmkl_avx2.so 2AAAB7DA5CA3  mkl_blas_avx2_zdo Unknown  Unknown

Thanks,

Ben Comer

Georgia Tech








--
+--+
  Davide Ceresoli
  CNR Institute of Molecular Science and Technology (CNR-ISTM)
  c/o University of Milan, via Golgi 19, 20133 Milan, Italy
  Email: davide.ceres...@istm.cnr.it
  Phone: +39-02-50314276, +39-347-1001570 (mobile)
  Skype: dceresoli
+--+
___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] GIPAW Segmentation Fault

[Ben Comer]

Davide,

I am using the intel 17.0 compiler with mkl 11.3. Adding the suggested 
line to the bottom of my make.inc file did not change the error 
unfortunately. I've included a dropbox link containing my build script 
and make.inc. I'm using the QE-GIPAW 6.2.2 release (QE and GIPAW bundled).


https://www.dropbox.com/sh/1znj73stc2opkkq/AABVcYoSXKIGlhi293GIB8Zpa?dl=0



Dear Ben,
    I'm afraid it's a problem with MKL-blas ZDOTC, which must
return a complex(dp) result. Very strange, because if you grep
the source code, we have declared it: complex(dp), external::zdtoc

Can you tell us your compiler and MKL version? can you add
DFLAGS+=-Dzdotc=zdotc_wrapper
to the QE make.inc and recompile both (QE and GIPAW)?

Best wishes,
    Davide

On 08/06/2018 03:19 PM, Ben Comer wrote:
Hello,

I've been trying to do g factor calculations in GIPAW working for a 
few days now. I keep getting a segmentation fault (below) no matter 
how I compile it on our cluster. Does anyone have a good idea of what 
might be causing this?



forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image  PC    Routine Line Source
gipaw.x    00C40604  Unknown Unknown  Unknown
libpthread-2.12.s  00328DE0F7E0  Unknown Unknown  Unknown
libmkl_avx2.so 2AAAB7DA5CA3  mkl_blas_avx2_zdo Unknown  Unknown

Thanks,

Ben Comer

Georgia Tech





___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] GIPAW Segmentation Fault

[Holzwarth, Natalie]

I am not sure what is the output for that particular line, but the last
part of the output looks as given below.For different instances of the
run, the particular iteration changes, but the boxed error message returned
by phonon.f90 is always the same.We have seen the error in both 6.2.1
and 6.3 versions of the QE package.  If this is of interest, I will be glad
to send any additional information.The run is for NaCl.Thanks,
Natalie

- end of phonon output file:
 Representation #  3 mode #   3

 Self-consistent Calculation

  iter #   1 total cpu time :  1573.2 secs   av.it.:   6.9
  thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  1.571E-05

  iter #   2 total cpu time :  1602.0 secs   av.it.:  11.4
  thresh= 3.964E-04 alpha_mix =  0.700 |ddv_scf|^2 =  1.095E-05

  iter #   3 total cpu time :  1629.0 secs   av.it.:  10.6
  thresh= 3.308E-04 alpha_mix =  0.700 |ddv_scf|^2 =  3.191E-08

  iter #   4 total cpu time :  1657.8 secs   av.it.:  11.2
  thresh= 1.786E-05 alpha_mix =  0.700 |ddv_scf|^2 =  2.458E-10

  iter #   5 total cpu time :  1687.6 secs   av.it.:  11.9
  thresh= 1.568E-06 alpha_mix =  0.700 |ddv_scf|^2 =  2.328E-11
---
Primary job  terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.
---


N. A. W. Holzwarth   email:
nata...@wfu.edu
Department of Physics  web:
http://www.wfu.edu/~natalie
Wake Forest University phone:
1-336-758-5510
Winston-Salem, NC 27109 USA office: Rm. 300 Olin
Physical Lab

On Tue, Aug 7, 2018 at 12:00 PM, Pietro Delugas  wrote:

> hello
>
> what is the output of
>
> addr2line -p -e ph.x 004BE229
> and what version of ph are you using ?
>
>
> On 07/08/2018 17:06, Holzwarth, Natalie wrote:
>
> This segmentation fault issue has also appeared for us in another QE
> code.Perhaps it is a totally unrelated problem which we find related to
> the openmpi package compiled with intel-3.1.1-2018 and intel-3.1.0-2018.
>  In our case, compiling with openmpi package compiled with intel-2.1.0-2018
> usually (not always) solves the problem.  Compiling with a much older
> openmpi package solves the problem, but is not viable for the current
> configuration of our cluster. Since no one else mentioned the problem until
> now, we think it may have to do with our internal setup???  The error is
> very intermittent,   occurring at different places in the code for the same
> input.The common aspect of this error to the one in the original
> message is libpthread-2.12.so. Most reliably we see the error in ph.x
> with part of the error message:
>
> forrtl: severe (174): SIGSEGV, segmentation fault occurred
> Image  PCRoutineLine
> Source
> ph.x   00D99A1D  for__signal_handl Unknown  Unknown
> libpthread-2.12.s  003271E0F7E0  Unknown   Unknown  Unknown
> mca_btl_vader.so   2AB74BBB99A7  Unknown   Unknown  Unknown
> libopen-pal.so.40  2AB738AD3A54  opal_progress Unknown  Unknown
> libmpi.so.40.10.1  2AB7384DBC04  ompi_request_defa Unknown  Unknown
> libmpi.so.40.10.1  2AB7385384C5  ompi_coll_base_ba Unknown  Unknown
> libmpi.so.40.10.1  2AB7384F26F1  MPI_Barrier   Unknown  Unknown
> libmpi_mpifh.so.4  2AB73826D013  MPI_Barrier_f08   Unknown  Unknown
> ph.x   00BA9E0E  Unknown   Unknown  Unknown
> ph.x   00B9835B  Unknown   Unknown  Unknown
> ph.x   0057FE26  Unknown   Unknown  Unknown
> ph.x   004BE229  Unknown   Unknown  Unknown
> ph.x   004A0F10  Unknown   Unknown  Unknown
> ph.x   00415A65  Unknown   Unknown  Unknown
> ph.x   0040EE73  Unknown   Unknown  Unknown
> ph.x   0040EDDE  Unknown   Unknown  Unknown
> libc-2.12.so   00327161ED1D  __libc_start_main Unknown
> Unknown
> ph.x   0040ECE9  Unknown   Unknown  Unknown
>
> I am very curious about whether you think this may be related or totally
> unrelated.   Thanks,  Natalie Holzwarth
>
> N. A. W. Holzwarth   email:
> nata...@wfu.edu
> Department of Physics  web:
> http://www.wfu.edu/~natalie
> Wake Forest University phone:
> 1-336-758-5510
> Winston-Salem, NC 27109 USA office: Rm. 300 Olin
> Physical Lab
>
> On Tue, Aug 7, 2018 at 8:18 AM, Davide Ceresoli 
> wrote:
>
>> Dear Ben,
>> I'm afraid it's a problem with 

Re: [QE-users] GIPAW Segmentation Fault

[Pietro Delugas]

hello

what is the output of

addr2line -p -e ph.x 004BE229

and what version of ph are you using ?


On 07/08/2018 17:06, Holzwarth, Natalie wrote:
This segmentation fault issue has also appeared for us in another QE 
code.    Perhaps it is a totally unrelated problem which we find 
related to the openmpi package compiled with intel-3.1.1-2018 and 
intel-3.1.0-2018.   In our case, compiling with openmpi package 
compiled with intel-2.1.0-2018 usually (not always) solves the 
problem.  Compiling with a much older openmpi package solves the 
problem, but is not viable for the current configuration of our 
cluster. Since no one else mentioned the problem until now, we think 
it may have to do with our internal setup???  The error is very 
intermittent,  occurring at different places in the code for the same 
input.   The common aspect of this error to the one in the original 
message is libpthread-2.12.so . Most 
reliably we see the error in ph.x with part of the error message:


forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image              PC                Routine Line        Source
ph.x               00D99A1D  for__signal_handl  Unknown  Unknown
libpthread-2.12.s  003271E0F7E0  Unknown  Unknown  Unknown
mca_btl_vader.so   2AB74BBB99A7  Unknown  Unknown  Unknown
libopen-pal.so.40  2AB738AD3A54  opal_progress  Unknown  Unknown
libmpi.so.40.10.1  2AB7384DBC04  ompi_request_defa  Unknown  Unknown
libmpi.so.40.10.1  2AB7385384C5  ompi_coll_base_ba  Unknown  Unknown
libmpi.so.40.10.1  2AB7384F26F1  MPI_Barrier  Unknown  Unknown
libmpi_mpifh.so.4  2AB73826D013  MPI_Barrier_f08  Unknown  Unknown
ph.x               00BA9E0E  Unknown  Unknown  Unknown
ph.x               00B9835B  Unknown  Unknown  Unknown
ph.x               0057FE26  Unknown  Unknown  Unknown
ph.x               004BE229  Unknown  Unknown  Unknown
ph.x               004A0F10  Unknown  Unknown  Unknown
ph.x               00415A65  Unknown  Unknown  Unknown
ph.x               0040EE73  Unknown  Unknown  Unknown
ph.x               0040EDDE  Unknown  Unknown  Unknown
libc-2.12.so       00327161ED1D  
__libc_start_main     Unknown Unknown

ph.x               0040ECE9  Unknown  Unknown  Unknown

I am very curious about whether you think this may be related or 
totally unrelated.   Thanks,  Natalie Holzwarth


N. A. W. Holzwarth email: nata...@wfu.edu 
Department of Physics    web: http://www.wfu.edu/~natalie 


Wake Forest University     phone: 1-336-758-5510
Winston-Salem, NC 27109 USA office: Rm. 300 Olin Physical Lab

On Tue, Aug 7, 2018 at 8:18 AM, Davide Ceresoli 
mailto:davide.ceres...@cnr.it>> wrote:


Dear Ben,
    I'm afraid it's a problem with MKL-blas ZDOTC, which must
return a complex(dp) result. Very strange, because if you grep
the source code, we have declared it: complex(dp), external::zdtoc

Can you tell us your compiler and MKL version? can you add
DFLAGS+=-Dzdotc=zdotc_wrapper
to the QE make.inc and recompile both (QE and GIPAW)?

Best wishes,
    Davide


On 08/06/2018 03:19 PM, Ben Comer wrote:

Hello,

I've been trying to do g factor calculations in GIPAW working
for a few days now. I keep getting a segmentation fault
(below) no matter how I compile it on our cluster. Does anyone
have a good idea of what might be causing this?


forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image  PC    Routine Line    Source
gipaw.x    00C40604  Unknown Unknown Unknown
libpthread-2.12.s  00328DE0F7E0  Unknown Unknown Unknown
libmkl_avx2.so 2AAAB7DA5CA3  mkl_blas_avx2_zdo
Unknown  Unknown

Thanks,

Ben Comer

Georgia Tech




-- 
+--+

  Davide Ceresoli
  CNR Institute of Molecular Science and Technology (CNR-ISTM)
  c/o University of Milan, via Golgi 19, 20133 Milan, Italy
  Email: davide.ceres...@istm.cnr.it

  Phone: +39-02-50314276, +39-347-1001570 (mobile)
  Skype: dceresoli
+--+
___
users mailing list
users@lists.quantum-espresso.org

https://lists.quantum-espresso.org/mailman/listinfo/users





___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users


___
users mailing list

Re: [QE-users] GIPAW Segmentation Fault

[Holzwarth, Natalie]

This segmentation fault issue has also appeared for us in another QE code.
  Perhaps it is a totally unrelated problem which we find related to the
openmpi package compiled with intel-3.1.1-2018 and intel-3.1.0-2018.   In
our case, compiling with openmpi package compiled with intel-2.1.0-2018
usually (not always) solves the problem.  Compiling with a much older
openmpi package solves the problem, but is not viable for the current
configuration of our cluster. Since no one else mentioned the problem until
now, we think it may have to do with our internal setup???  The error is
very intermittent,   occurring at different places in the code for the same
input.The common aspect of this error to the one in the original
message is libpthread-2.12.so. Most reliably we see the error in ph.x with
part of the error message:

forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image  PCRoutineLineSource

ph.x   00D99A1D  for__signal_handl Unknown  Unknown
libpthread-2.12.s  003271E0F7E0  Unknown   Unknown  Unknown
mca_btl_vader.so   2AB74BBB99A7  Unknown   Unknown  Unknown
libopen-pal.so.40  2AB738AD3A54  opal_progress Unknown  Unknown
libmpi.so.40.10.1  2AB7384DBC04  ompi_request_defa Unknown  Unknown
libmpi.so.40.10.1  2AB7385384C5  ompi_coll_base_ba Unknown  Unknown
libmpi.so.40.10.1  2AB7384F26F1  MPI_Barrier   Unknown  Unknown
libmpi_mpifh.so.4  2AB73826D013  MPI_Barrier_f08   Unknown  Unknown
ph.x   00BA9E0E  Unknown   Unknown  Unknown
ph.x   00B9835B  Unknown   Unknown  Unknown
ph.x   0057FE26  Unknown   Unknown  Unknown
ph.x   004BE229  Unknown   Unknown  Unknown
ph.x   004A0F10  Unknown   Unknown  Unknown
ph.x   00415A65  Unknown   Unknown  Unknown
ph.x   0040EE73  Unknown   Unknown  Unknown
ph.x   0040EDDE  Unknown   Unknown  Unknown
libc-2.12.so   00327161ED1D  __libc_start_main Unknown  Unknown
ph.x   0040ECE9  Unknown   Unknown  Unknown

I am very curious about whether you think this may be related or totally
unrelated.   Thanks,  Natalie Holzwarth

N. A. W. Holzwarth   email:
nata...@wfu.edu
Department of Physics  web:
http://www.wfu.edu/~natalie
Wake Forest University phone:
1-336-758-5510
Winston-Salem, NC 27109 USA office: Rm. 300 Olin
Physical Lab

On Tue, Aug 7, 2018 at 8:18 AM, Davide Ceresoli 
wrote:

> Dear Ben,
> I'm afraid it's a problem with MKL-blas ZDOTC, which must
> return a complex(dp) result. Very strange, because if you grep
> the source code, we have declared it: complex(dp), external::zdtoc
>
> Can you tell us your compiler and MKL version? can you add
> DFLAGS+=-Dzdotc=zdotc_wrapper
> to the QE make.inc and recompile both (QE and GIPAW)?
>
> Best wishes,
> Davide
>
>
> On 08/06/2018 03:19 PM, Ben Comer wrote:
>
>> Hello,
>>
>> I've been trying to do g factor calculations in GIPAW working for a few
>> days now. I keep getting a segmentation fault (below) no matter how I
>> compile it on our cluster. Does anyone have a good idea of what might be
>> causing this?
>>
>>
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> Image  PCRoutine LineSource
>> gipaw.x00C40604  Unknown Unknown  Unknown
>> libpthread-2.12.s  00328DE0F7E0  Unknown Unknown  Unknown
>> libmkl_avx2.so 2AAAB7DA5CA3  mkl_blas_avx2_zdo Unknown  Unknown
>>
>> Thanks,
>>
>> Ben Comer
>>
>> Georgia Tech
>>
>>
>>
>>
> --
> +--+
>   Davide Ceresoli
>   CNR Institute of Molecular Science and Technology (CNR-ISTM)
>   c/o University of Milan, via Golgi 19, 20133 Milan, Italy
>   Email: davide.ceres...@istm.cnr.it
>   Phone: +39-02-50314276, +39-347-1001570 (mobile)
>   Skype: dceresoli
> +--+
> ___
> users mailing list
> users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] GIPAW Segmentation Fault

[Davide Ceresoli]

Dear Ben,
I'm afraid it's a problem with MKL-blas ZDOTC, which must
return a complex(dp) result. Very strange, because if you grep
the source code, we have declared it: complex(dp), external::zdtoc

Can you tell us your compiler and MKL version? can you add
DFLAGS+=-Dzdotc=zdotc_wrapper
to the QE make.inc and recompile both (QE and GIPAW)?

Best wishes,
Davide


On 08/06/2018 03:19 PM, Ben Comer wrote:

Hello,

I've been trying to do g factor calculations in GIPAW working for a few days 
now. I keep getting a segmentation fault (below) no matter how I compile it on 
our cluster. Does anyone have a good idea of what might be causing this?



forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image  PC    Routine Line    Source
gipaw.x    00C40604  Unknown Unknown  Unknown
libpthread-2.12.s  00328DE0F7E0  Unknown Unknown  Unknown
libmkl_avx2.so 2AAAB7DA5CA3  mkl_blas_avx2_zdo Unknown  Unknown

Thanks,

Ben Comer

Georgia Tech





--
+--+
  Davide Ceresoli
  CNR Institute of Molecular Science and Technology (CNR-ISTM)
  c/o University of Milan, via Golgi 19, 20133 Milan, Italy
  Email: davide.ceres...@istm.cnr.it
  Phone: +39-02-50314276, +39-347-1001570 (mobile)
  Skype: dceresoli
+--+
___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

[QE-users] GIPAW Segmentation Fault

[Ben Comer]

Hello,

I've been trying to do g factor calculations in GIPAW working for a few 
days now. I keep getting a segmentation fault (below) no matter how I 
compile it on our cluster. Does anyone have a good idea of what might be 
causing this?



forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image  PC    Routine Line    Source
gipaw.x    00C40604  Unknown Unknown  Unknown
libpthread-2.12.s  00328DE0F7E0  Unknown Unknown  Unknown
libmkl_avx2.so 2AAAB7DA5CA3  mkl_blas_avx2_zdo Unknown  Unknown

Thanks,

Ben Comer

Georgia Tech


___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users