I have made some minimal updates to allow sander.LES and sander.PIMD to be
compiled with g95.
sander.PIMD has problems with the helium and spcfw test cases using g95.
The
errors involve allocating arrays (maybe already allocated?) G95 is nice
in
pointing out the exact line number where the error occurs, but it is not
obvious to me what has gone wrong. (I suspect that the "first_pme"
variable
is not working as intended under PIMD, but that may be way wrong...)
[Note that these failures do not occur with ifort. Also, that this
testing
so far has been mostly with serial code, not the parallel stuff.]
The qmmm2/pimd* test cases look like they are probably working, but the
output doesn't match the saved one. Wei and/or Ross should see if we can
get these updated.
In principle, Mike should now be able to test his latest changes with
sander.LES and sander.PIMD.
...dac
Received on Wed Apr 05 2006 - 23:49:45 PDT
