EPW Forum

Dear all:
When I run the epw.x, when it comes to the wannier step, it crashes and print out the "segmentation fault", however, when I use wannier90.x separately, nothing wrong happends, please give me some advices on that.

My input file
--
&inputepw
prefix = 'Cs2SnBr6',
outdir = './'
dvscf_dir = '../4-phonon/save'

elph = .true.
epbwrite = .true.
epbread = .false.
epwwrite = .true.
epwread = .false.

wannierize = .true.
nbndsub = 15
num_iter = 300
bands_skipped = '1:23'
dis_win_max = 6
dis_win_min = 0.5
dis_froz_min= 6
dis_froz_max= 0.5
auto_projections = .true.
scdm_proj = .true.
scdm_sigma = 2
scdm_mu = 2
scdm_entanglement = ‘isolated’

nkf1 = 10
nkf2 = 10
nkf3 = 10
nqf1 = 10
nqf2 = 10
nqf3 = 10

nk1 = 5
nk2 = 5
nk3 = 5
nq1 = 5
nq2 = 5
nq3 = 5
/

The following is the epw.x print out:
PseudoPot. # 1 for Br read from file:
../pp/Br.UPF
MD5 check sum: d3ffb7b29f6225aa16fe06858fb2a80b
Pseudo is Ultrasoft + core correction, Zval = 7.0
Generated by new atomic code, or converted to UPF format
Using radial grid of 879 points, 6 beta functions with:
l(1) = 0
l(2) = 0
l(3) = 1
l(4) = 1
l(5) = 2
l(6) = 2
Q(r) pseudized with 8 coefficients, rinner = 1.100 1.100 1.100
1.100 1.100

PseudoPot. # 2 for Cs read from file:
../pp/Cs.UPF
MD5 check sum: 3476d69cb178dfad3ffaa59df4e07ca4
Pseudo is Ultrasoft + core correction, Zval = 9.0
Generated by new atomic code, or converted to UPF format
Using radial grid of 905 points, 6 beta functions with:
l(1) = 0
l(2) = 0
l(3) = 1
l(4) = 1
l(5) = 2
l(6) = 2
Q(r) pseudized with 6 coefficients, rinner = 1.500 1.500 1.500
1.500 1.500

PseudoPot. # 3 for Sn read from file:
../pp/Sn.UPF
MD5 check sum: 4cf58ce39ec5d5d420df3dd08604eb00
Pseudo is Ultrasoft + core correction, Zval = 14.0
Generated by new atomic code, or converted to UPF format
Using radial grid of 891 points, 6 beta functions with:
l(1) = 0
l(2) = 0
l(3) = 1
l(4) = 1
l(5) = 2
l(6) = 2
Q(r) pseudized with 6 coefficients, rinner = 1.200 1.200 1.200
1.200 1.200
EPW : 3.50s CPU 3.92s WALL

EPW : 15.93s CPU 16.39s WALL

-------------------------------------------------------------------
Wannierization on 5 x 5 x 5 electronic grid
-------------------------------------------------------------------
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
libpthread-2.28.s 0000154BDDB71CF0 Unknown Unknown Unknown
epw.x 00000000010FE707 errore_ 10 error_handler.f90
epw.x 0000000000616F4C pw2wan2epw_mp_pw2 60 pw2wan2epw.f90
epw.x 0000000000426D9A wannierization_mp 73 wannierization.f90
epw.x 0000000000409068 MAIN__ 136 epw.f90
epw.x 00000000004083BD Unknown Unknown Unknown
libc-2.28.so 0000154BDD452D85 __libc_start_main Unknown Unknown
epw.x 00000000004082DE Unknown Unknown Unknown
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
libpthread-2.28.s 00001497C5953CF0 Unknown Unknown Unknown
epw.x 00000000010FE707 errore_ 10 error_handler.f90
epw.x 0000000000616F4C pw2wan2epw_mp_pw2 60 pw2wan2epw.f90
epw.x 0000000000426D9A wannierization_mp 73 wannierization.f90
epw.x 0000000000409068 MAIN__ 136 epw.f90
epw.x 00000000004083BD Unknown Unknown Unknown
libc-2.28.so 00001497C5234D85 __libc_start_main Unknown Unknown
epw.x 00000000004082DE Unknown Unknown Unknown

Hi leeru,
Can you please check the following?
dis_froz_min= 6
dis_froz_max= 0.5

I see you have set min=6 and max=0.5. I think it should be reversed.

Thanks!

Gyanu

I have recompiled the epw code by intel oneapi 2021.4, and now the error has gone, but it may also due to the dis_froz_min= 6
dis_froz_max= 0.5 error, anyway, thank you.