On Sat, Mar 24, 2018 at 10:38 AM, David A Case <david.case.rutgers.edu>
wrote:
>
> 3. Does anyone besides Scott have access to Intel 18.0 compilers?
> Basic things like the dhfr or water tests seem to be failing there. If
> this is reproduced, can someone put a trap in the configure2 script?
>
I have Intel 18.0.1
4. The PGI compilers are favorites for some people, and we need a
> second opinion here, vs the srb results on the wiki. If anyone has
> access to other versions, that might be worth a shot as well.
>
I have PGI 17.10
I'll try with these two and report the results on the Wiki along with GCC
6.4.0 (I also have 7.2.0).
A few comments that I've found just trying to configure (with the old
system, not CMake yet):
- I committed some changes that fixed errors I saw, (basically if CC, CXX,
and FC variables point to full paths of compilers, the version check fails)
- The -xHost flag on the Intel compilers actually fails to build a working
executable on my machine (!!). It's an AMD chip, but the Intel compilers
clearly misdetect the available instruction set. I had to do a global
deletion of all -xHost strings in configure2 for the intel compilers to get
past the FFTW configure step. I'm not sure how commonplace my experience
will be, but -xHost doesn't build executables that run on all nodes of a
heterogeneous cluster anyway.
The commits I think should be cherry-picked especially include:
206846039a1f300f1e10f248fcc9b950b4c21fa6
81391a63d64f16075dd22e86b76459e4b212d95d
That's all for now. Hope everyone's well!
Thanks,
Jason
--
Jason M. Swails
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Sun Mar 25 2018 - 18:00:02 PDT
