Re: [AMBER-Developers] MPI Amber build is broken

From: Jamie Smith <>
Date: Wed, 8 Nov 2017 00:23:25 +0000

I just re-ran the Amber Jenkins build, and it completed fine. It looks like the compile error is fixed, thanks!

-----Original Message-----
From: Yinglong Miao []
Sent: Tuesday, November 7, 2017 3:18 PM
To: David A Case <>; AMBER Developers Mailing List <>
Subject: Re: [AMBER-Developers] MPI Amber build is broken

To keep you updated, I have added the missing argument my_atm_lst, which caused the problem as found by Jamie. I have also re-complied and tested the serial CPU, parallel CPU, serial GPU and parallel GPU versions with files in $AMBERHOME/test. They appeared to run normal with most comparisons PASSED.

With that, I have just committed the new change. I will appreciate it if anyone else could double check, and see whether that fixes the build issue ...


On Tue, Nov 7, 2017 at 7:23 AM, David A Case <> wrote:

> On Tue, Nov 07, 2017, Jamie Smith wrote:
> >
> > I’m not sure if anyone else noticed, but since Yinglong Miao’s merge
> > of the new GaMD code, MPI PMEMD has been failing to build.
> I would say that anyone who knows how should revert the last four
> commits, and let Yinglong work on the MPI problem. Commits to master
> that break the build are real no-no's.
> One problem is that commit e8e399d5af7e349dda6ffe is a merge, and
> reverting a merge commit requires more git experience than I have
> (plus, I'm in Germany now, with minimal time to look at this.)
> I suppose if "my_atm_list" is not used/needed for GaMD, someone could
> try to add in the argument. But this seems pretty dangerous: we need
> to know for sure that GaMD works in pmemd.MPI
> ....dac
> _______________________________________________
> AMBER-Developers mailing list
AMBER-Developers mailing list
AMBER-Developers mailing list
Received on Tue Nov 07 2017 - 16:30:03 PST
Custom Search