P2y stop with segmentation fault

[sdwang]

Dear developers,
I am using Yambo-4.4.0 to generate SAVE file, but at the end of the process, it stops with 'segmentation fault'. And I tried to do yamb -i, and it works without any error(I used enable-uspp configuration).
<---> DBs path set to .
<---> detected QE data format: qexsd
<---> == PWscf v.6.x generated data (QEXSD fmt) ==
<---> Header/K-points/Energies... done
<---> Cell data...
<01s> [WARNING]Two or more crystal strcutures fit the given cell done
<01s> Atomic data... done
<01s> Symmetries...[SI no]......[-I no]...[TR yes]
<01s> XC functional...Slater exchange(X)+Perdew & Zunger(C)
<01s> EXX fraction... 0.000000
<01s> EXX screening... 0.000000
<01s> K-points mesh... done
<01s> RL vectors... done
<01s> IGK arrays... done
<01s> Energies... done
<02s> :: Electrons : 52.00000
<02s> :: Temperature [ev]: 0.000000
<02s> :: Lattice factors [a.u.]: 6.03420 5.22577 47.85371
<02s> :: K-points : 102
<02s> :: Bands : 600
<02s> :: Spinor components : 2
<02s> :: Spin polarizations : 1
<02s> :: Spin orbit coupling : yes
<02s> :: Symmetries [spatial]: 6
<02s> :: [T-rev]: yes
<02s> :: Max WF components : 10463
<02s> :: RL vectors (WF): 13265
<02s> :: RL vectors (CHARGE): 328905
<02s> :: XC potential : Slater exchange(X)+Perdew & Zunger(C)
<02s> :: Atomic species : 3
<02s> :: Max atoms/species : 2
<02s> == DB1 (Gvecs and more) ...
<02s> ... Database done
<02s> == DB2 (wavefunctions) ...
<18m-59s> [p2y] wfc_io |########################################| [100%] 18m-57s(E) 18m-57s(X) done ==
<18m-59s> == DB3 (PseudoPotential) ...[wsd2:02686] *** Process received signal ***
[wsd2:02686] Signal: Segmentation fault (11)
[wsd2:02686] Signal code: Address not mapped (1)
[wsd2:02686] Failing at address: 0x10
[wsd2:02686] [ 0] /lib64/libpthread.so.0() [0x313900f7e0]
[wsd2:02686] [ 1] /public/software/yambo-4.3.2-sp/bin/p2y(init_us_1_+0x4d5b) [0x58a03b]
[wsd2:02686] [ 2] /public/software/yambo-4.3.2-sp/bin/p2y(pp_pwscf_driver_+0x1955) [0x4ce845]
[wsd2:02686] [ 3] /public/software/yambo-4.3.2-sp/bin/p2y(p2y_i_+0x1061) [0x4d5da1]
[wsd2:02686] [ 4] /public/software/yambo-4.3.2-sp/bin/p2y(main+0x1055) [0x481345]
[wsd2:02686] [ 5] /lib64/libc.so.6(__libc_start_main+0xfd) [0x3138c1ed1d]
[wsd2:02686] [ 6] /public/software/yambo-4.3.2-sp/bin/p2y() [0x480229]
[wsd2:02686] *** End of error message ***
Segmentation fault (core dumped)
Thanks!

SD

[andrea.ferretti]

Hi Shudong,

thanks for reporting.
Two main comments:
- USPP are currently supported as a preliminary feature, only for HF and with limited symmetry support;
this means USPP are not yet production-level, but mostly usable for testing
- Concerning the problem you are experiencing, it would be very useful to have an input to reproduce (and possibly solve)
the problem

take care
Andrea

[sdwang]

Dear Andrea,
I removed uspp option and now it works.
Thank you!

Best


Shudong

[Jerry]

Dear dxevelopers,
I am using Yambo-5.2.0 to generate SAVE file, but at the end of the process, it stops with 'segmentation fault'.
It seems i get some problem with IGK arrays.

Code: Select all

 <---> DBs path set to : .
 <---> detected QE data format : qexsd
 <---> == PWscf v.6.x generated data (QEXSD fmt) ==
 <---> Header/K-points/Energies... done
 <02s> Cell data...
 <02s> [WARNING] Two or more crystal structures fit the given cell done
 <02s> Atomic data... done
 <03s> Symmetries...[SI no]...[I yes]...[-I no]...[TR yes]
 <03s> XC functional...Slater exchange(X)+Perdew & Zunger(C)
 <03s> EXX fraction...:  0.000000
 <03s> EXX screening...:  0.000000
 <03s> K-points mesh... done
 <04s> RL vectors... done
 <04s> IGK arrays...zsh: segmentation fault (core dumped)  p2y

I don't know how to solve it.
Thanks!
JM

[Daniele Varsano]

Dear Jerry,

can you please post your nscf input file?
Other suggestion is to try to use a more recent release of QE.

Best,
Daniele

[Jerry]

Dear Daniele,
thank you for your reply.
I am using qe-7.2. and here is my nscf input file.

Code: Select all

&control
    calculation='nscf',
    prefix='hBN',
    pseudo_dir = './Pseudos'
    nstep = 200
    verbosity = 'high'
    wf_collect=.true.
/
&system
    ibrav = 4,
    celldm(1) = 4.716
    celldm(3) = 2.582
    nat= 4,
    ntyp= 2,
    force_symmorphic=.true.
    ecutwfc = 40,nbnd = 200
/
&electrons
   diago_thr_init=5.0e-6,
   diago_full_acc=.true.
   electron_maxstep = 100,
   diagonalization='david'
   mixing_mode = 'plain'
   mixing_beta = 0.6
 conv_thr =  1.0e-8
/
ATOMIC_SPECIES
 B 10 B.pz-vbc.UPF
 N 14 N.pz-vbc.UPF
ATOMIC_POSITIONS {crystal}
B  0.6666667  0.3333333  0.0000000
N -0.6666667 -0.3333333  0.0000000
B -0.6666667 -0.3333333  0.5000000
N  0.6666667  0.3333333  0.5000000  
K_POINTS {automatic}
36  36  2 0 0 0 

Best,
Jerry

[andrea.ferretti]

Dear Jerry,

it seems your input files runs correctly with me using qe-7.2 + yambo-5.2.1 both compiled using gfortran.

From your input file: you have quite a lot of kpts, which can make the G-vector grid and maps quite large.
Have you checked the limits of your shell (zsh if I get it right), e.g. using ulimit or similar ?

best
Andrea

[Jerry]

Dear Ferretti,

I changed my ecutrho in qe computation and solve the problem.

Thanks!
JM