Version_memo.txt

gkvp_f0.62                                              S. Maeyama     Mar 2023
1) equib_type = "ring" is added for ring dipole geometry.

2) Rotating flux-tube model is implemented to treat equilibrium shearflows,
   available for torus: equib_type = "s-alpha", "s-alpha-shift", "analytic",
   "circMHD", "vmec", "igs". (But not available for "slab", "ring")



gkvp_f0.61                                              S. Maeyama     Mar 2021
1) equib_type = "s-alpha-shift" is added. s-alpha model with Shafranov shift.

2) random_seed is applied in lib/gkvp_math_portable.f90.

3) Treat the tracer particle setting (fcs=0).



gkvp_f0.60                                              S. Maeyama     Feb 2021
1) NetCDF4+parallel HDF5 is added for optional output of GKV.



gkvp_f0.59                                              S. Maeyama     Oct 2020
1) NEC SX-Aurora Makefile/shoot/sub.q are added. Tuned source codes are used.

2) Extension of diag (write NetCDF by out_netcdf.f90)



gkvp_f0.58                                              S. Maeyama     Oct 2020
1) Some benchmark data are added.

2) init_randoma, to switch random number for initialization, is added to gkvp_namelist.



gkvp_f0.57                                              S. Maeyama     Oct 2020
1) Version number f0.** is removed from filename. 
   *** Update history should be clarified in each module files too. ***

2) A bug at bias factor of collision, nu, is modified in colliimp module.

3) equib_type = "slab" is added.

4) Minor debugs:
  - Padding region iend_y < my is carefully initalized.
  - menergy(:,:)=0 for beta=0
  - freq module is modified to support nx0=0.



gkvp_f0.55                                              S. Maeyama     Dec 2019
1) freq module is updated for kx /= 0 mode.

2) set module is modified. When s_hat = 0.0, m_j = 0 and kxmin = kymin are enforced.

3) For linear runs, tips_rescale_for_linear_runs are added to avoid divergence



gkvp_f0.54                                              M. Nakata      Dec 2019
1) Mean ExB shearing effect in Eulerian approach is added: "shearflow" module.  

2) igs module is fixed for reading the tokamak geometry. 



gkvp_f0.53                                              M. Nakata      Jan 2019
1) vmecbzx (and set) module is fixed for reading the stellarator geometry data for N_tht>1 cases. 



gkvp_f0.52                                              S. Maeyama     Sep 2018
1) Optimizations for reduced memory usage and efficient commun.-comput. overlap.





gkvp_f0.50                                              S. Maeyama     Sep 2017
1) gkvp_f0.50_colliimp.f90 is added for implicit collision solver.

2) lib/gkvp_f0.50_math_portable.f90 is added.





gkvp_f0.48                                              S. Maeyama     Dec 2016
1) gkvp_f0.48_vmecbzx.f90 is added in stead of gkvp_f0.48_vmecin.f90.

2) Namelist is changed from (r_minor, q_d, n_alp) to (kymin, m_j, del_c).

3) Re-organize output files.





gkvp_f0.47                                              S. Maeyama     Nov 2016

1) Triad transfer analysis subroutine trans_triad is added in trans module.

2) math_random in lib/gkvp_f0.47_math_SSL2.f90 has been extended for double precision.





gkvp_f0.46                                              S. Maeyama     May 2016

1) 0-padding in bndry module is modified, for the case nzb>=3, z_bound=outflow.

2) Remove OpenMP NOWAIT directive in intgrl module, which is the cause of
   a error in the reduction against iv, im.

3) Remove OpenMP NOWAIT directive in colli module, which is the cause of 
   a error in the reduction against is2. (This modification introduces the
   implicit barrier synchronization among OpenMP threads, and thus prevent 
   computation-communication overlap by using OpenMP MASTER thread.
   Improvement is required.)

4) Step-job script "shoot" is added.

5) Transform between ff and hh is contained in fld_ff2hh and fld_hh2ff.

6) z_calc and art_diff is added in namelist.





gkvp_f0.45                                               M. Nakata    July 2015

Modification from previous f0.40 ver. (Nakata July2015) 

1) OMP directives are corrected in colli-module: nowait-directive was missing.

2) Overlap procedures in are corrected in caldlt_full routine in advnc-module: 
The GK field-particle and non-isothermal collision terms are calculated together in GK_colli_CF_DT. 

3) dtc-module is upgraded so as to include the velocity-dependent collision frequency: 
HOWEVER, it is not always good time-step estimation due to the steepness of e-i collision frequency.  

4) Neglect "ic2zero" flag in namelist: this was just for debug. 

5) calc_dh in advnc-module is modified: same as f0.30