EPW fail to calculate the electron-phonon matrix elements in the Wannier representation

[jack]

Dear,
I use the EPW to calculate the electron-phonon matrix elements in the Wannier representation. The input file is list as below,

&inputepw
prefix = 'Cs'
outdir = './'

elph = .true.
epwwrite = .true.

lpolar = .true.
nbndsub = 3
dvscf_dir = './save/'

etf_mem = 0
epw_memdist = .true.

bands_skipped = 'exclude_bands = 1:47,51-57'
wannierize = .true.
num_iter = 500
iprint = 2

proj(1) = 'Sb:p'

wannier_plot = .true.
wannier_plot_supercell = 6 6 6

nk1 = 6
nk2 = 6
nk3 = 6
nq1 = 6
nq2 = 6
nq3 = 6

band_plot = .true.
filkf = './kpoints'
filqf = './kpoints'
/"
The epw broke when doing the calculation "the irreducible q point # 1". The output file is cut at "q( 1 ) = ( 0.0000000 0.0000000 0.0000000 )". And the system prints "forrtl: severe (174): SIGSEGV, segmentation fault occurred". The command I used to submit the epw is "mpirun -np 96 epw.x -npool 96 < epw.in > epw.out". The qe and epw are compiled with the intel oneapi and its MKL.

[stiwari]

Hi Jack,

The step where the error has occurred means something is wrong with the phonon (ph.x) related calculations. Check the files inside the "save" folder which you must have generated using pp.py.

If the error persists then post the .out and .in files which will help us debug.

Best regards,
Sabya.

[jack]

Dear,
Thank you for your response. I make a mistake. The nq in epw.in isn't equal to nq in ph.x. I correct it, and the epw works well.
Bset