istwfk variable

[Alexander89]

Hi all,

I am using wannier90 with abinit (prtwant 2) and I am getting some errors that I don't know how to fix. What I am doing is to run one of the examples abinit has, but adding spin orbit coupling (with hgh pseudopotential).
When I run this example without SOC everything goes well.

I think one of the problems is the variable istwfk. How can I set properly this variable?
These are the errors I am getting in the log file:

--- !WARNING
src_file: m_hdr.F90
src_line: 4020
message: |
input nkpt=8 not equal disk file nkpt=3
...

!WARNING
src_file: m_hdr.F90
src_line: 4549
message: |
Restart of self-consistent calculation need translated wavefunctions.
...

Indeed, critical differences between current calculation and
restart file have been detected in:
* the number, position, or weight of k-points
* the format of wavefunctions (istwfk)

mlwfovlp: you should give k-point in the full brillouin zone
with explicit k-points (or kptopt=3) and istwfk 1

Program received signal SIGABRT: Process abort signal.

Backtrace for this error:
#0 0x7f72b47ffed7 in ???
#1 0x7f72b47ff10d in ???
#2 0x7f72b3f327ef in ???
#3 0x7f72b3f3277f in ???
#4 0x7f72b3f34379 in ???
#5 0x7f72b3f7608f in ???
#6 0x7f72b3f7fc39 in ???
#7 0x7f72b3f83d2b in ???
#8 0x5648099c2e30 in mlwfovlp_

[ebousquet]

Dear Alexander,
Could you test by setting istwfk*1?
Best wishes,
Eric
PS: In the future I would advise you to use the pseudos from pseudodojo table:
http://www.pseudo-dojo.org/
There is no selection with SOC, but it is easy to generate, I have several of them with SOC already, just ask if you need. They are more optimized and need a lower cutoff than the HGH.