Hi,
I think that I found a fix. The problem was in the makedepend program in
pbsa/ that had not been updated. That should work now. Can anybody test?
Best,
Gerald.
On 02/18/2016 07:01 PM, Ross Walker wrote:
> I second this - I tried to disable PBSA building for now but of course it is linked into Sander etc so we'd disabling Sander as well.
>
> I'd suggest spending a few hours trying to fix it and if you can't the checkin should be reversed until this is fixed.
>
> Serial builds really aren't a viable option these days - try doing a serial build on a Xeon Phi and you'll understand. :-(
>
> All the best
> Ross
>
>> On Feb 18, 2016, at 9:13 AM, Daniel Roe <daniel.r.roe.gmail.com> wrote:
>>
>> On Wed, Feb 17, 2016 at 6:04 PM, Ray Luo, Ph.D. <ray.luo.uci.edu> wrote:
>>> Ross,
>>>
>>> Please pull again to see whether it fixed the build problem on your
>>> side. Looks like multi-thread make does not work ... Too many
>>> dependence issues ...
>>
>> The multi-threaded build definitely doesn't work for me when I hit pbsa stuff:
>>
>> $ make -j6 install
>> ...
>> make[2]: Entering directory '/home/droe/Amber/GIT/amber/AmberTools/src/pbsa'
>> gcc -c -O3 -mtune=native -fPIC ...
>> ...
>> Makefile:238: recipe for target 'pb_p3m.LIBPBSA.o' failed
>> Makefile:238: recipe for target 'pb_fdfrc.LIBPBSA.o' failed
>>
>> Amber is pretty large nowadays, and not being able to build in
>> parallel is a real drawback. We can currently get around this by
>> adding a '.NOTPARALLEL' target to the pbsa Makefile (or forcing -j1),
>> but ideally I think the multi-threaded build should be fixed.
>>
>> -Dan
>>
>>>
>>> All the best,
>>> Ray
>>> --
>>> Ray Luo, Ph.D.
>>> Professor
>>> Biochemistry, Molecular Biophysics, Chemical Physics,
>>> Chemical and Biomedical Engineering
>>> University of California, Irvine, CA 92697-3900
>>>
>>>
>>> On Wed, Feb 17, 2016 at 12:59 PM, Ray Luo, Ph.D. <ray.luo.uci.edu> wrote:
>>>> Okay, I think I've fixed it. Don't know why the same makefile works in
>>>> amber15 but doesn't work in amber16. I'm testing all the
>>>> sander/pbsa/nab combs to make sure all can build without interruption
>>>> and also pass the tests.
>>>>
>>>> Ray
>>>> --
>>>> Ray Luo, Ph.D.
>>>> Professor
>>>> Biochemistry, Molecular Biophysics, Chemical Physics,
>>>> Chemical and Biomedical Engineering
>>>> University of California, Irvine, CA 92697-3900
>>>>
>>>>
>>>> On Tue, Feb 16, 2016 at 10:34 PM, Ray Luo, Ph.D. <ray.luo.uci.edu> wrote:
>>>>> Finally I can reproduce your problem in building the MPI sander ...
>>>>>
>>>>> Will get it fixed tomorrow morning ...
>>>>>
>>>>> Ray
>>>>> --
>>>>> Ray Luo, Ph.D.
>>>>> Professor
>>>>> Biochemistry, Molecular Biophysics, Chemical Physics,
>>>>> Chemical and Biomedical Engineering
>>>>> University of California, Irvine, CA 92697-3900
>>>>>
>>>>>
>>>>> On Tue, Feb 16, 2016 at 10:17 PM, Ray Luo, Ph.D. <ray.luo.uci.edu> wrote:
>>>>>> Hi Ross,
>>>>>>
>>>>>> Looks like that there is no extra untracked file at least in all the
>>>>>> src folders, though there were indeed two .swp files in the test
>>>>>> cases. So I think lack of "git clean -fdx" is not the problem. I've
>>>>>> finished the serial build and is running "test.sander.BASIC". Looking
>>>>>> good, so far.
>>>>>>
>>>>>> I'll do the MPI build next after another "git clean -fdx" ...
>>>>>>
>>>>>> However, I think one possible reason could be the use of multi-thread
>>>>>> in make. There may be inter-dependent issues involved.
>>>>>>
>>>>>> I'll do some experiment with the MPI build to see whether this is the cause.
>>>>>>
>>>>>> Ray
>>>>>> --
>>>>>> Ray Luo, Ph.D.
>>>>>> Professor
>>>>>> Biochemistry, Molecular Biophysics, Chemical Physics,
>>>>>> Chemical and Biomedical Engineering
>>>>>> University of California, Irvine, CA 92697-3900
>>>>>>
>>>>>>
>>>>>> On Tue, Feb 16, 2016 at 8:24 PM, Ross Walker <ross.rosswalker.co.uk> wrote:
>>>>>>> Hi Ray,
>>>>>>>
>>>>>>> It's not about whether you compiled in that folder - it is whether you maybe edited files in that folder that you never committed / pushed. Try doing:
>>>>>>>
>>>>>>> git clean -f -d -x
>>>>>>> git status
>>>>>>>
>>>>>>> And see what it reports.
>>>>>>>
>>>>>>> All the best
>>>>>>> Ross
>>>>>>>
>>>>>>>> On Feb 16, 2016, at 23:04, Ray Luo <rluo.uci.edu> wrote:
>>>>>>>>
>>>>>>>> Hi Ross,
>>>>>>>>
>>>>>>>> Yes, I think my folder is clean since I have never compiled anything
>>>>>>>> in my amber git folder. Every time after syncing with the master
>>>>>>>> branch, I copy the whole master branch to a new folder for compiling
>>>>>>>> and testing.
>>>>>>>>
>>>>>>>> I'm about to test this evening's synced version, but will do "git
>>>>>>>> clean -fxd" before configuring/compiling. Maybe this is why.
>>>>>>>>
>>>>>>>> By the way, could you add my uci email to the mailing list as well,
>>>>>>>> i.e. "ray.luo.uci.edu"?
>>>>>>>>
>>>>>>>> All the best,
>>>>>>>> Ray
>>>>>>>> --
>>>>>>>> Ray Luo, Ph.D.
>>>>>>>> Professor
>>>>>>>> Biochemistry, Molecular Biophysics, Chemical Physics,
>>>>>>>> Chemical and Biomedical Engineering
>>>>>>>> University of California, Irvine, CA 92697-3900
>>>>>>>>
>>>>>>>>
>>>>>>>> On Tue, Feb 16, 2016 at 7:51 PM, Ross Walker <ross.rosswalker.co.uk> wrote:
>>>>>>>>> Hi Ray,
>>>>>>>>>
>>>>>>>>> If I do this - on my OSX laptop (GCC 4.9.3), MPICH version 3.1.4
>>>>>>>>>
>>>>>>>>> git pull
>>>>>>>>> git clean -f -d -x
>>>>>>>>> git status
>>>>>>>>> On branch master
>>>>>>>>> Your branch is up-to-date with 'origin/master'.
>>>>>>>>> nothing to commit, working directory clean
>>>>>>>>>
>>>>>>>>> ./configure -mpi gnu
>>>>>>>>> make install
>>>>>>>>>
>>>>>>>>> ...
>>>>>>>>> ...
>>>>>>>>> dnrm2.F90:113.24:
>>>>>>>>>
>>>>>>>>> ASSIGN 110 TO NEXT
>>>>>>>>> 1
>>>>>>>>> Warning: Deleted feature: ASSIGN statement at (1)
>>>>>>>>> mpif90 -DBINTRAJ -DEMIL -DMPI -c -O3 -mtune=native -fPIC -ffree-form -I/Users/rcw/amber/amber/include -I/Users/rcw/amber/amber/include -o dcopy.o dcopy.F90
>>>>>>>>> mpif90 -DBINTRAJ -DEMIL -DMPI -c -O3 -mtune=native -fPIC -ffree-form -I/Users/rcw/amber/amber/include -I/Users/rcw/amber/amber/include -o pb_force.o pb_force.F90
>>>>>>>>> pb_force.F90:292.77:
>>>>>>>>>
>>>>>>>>> sa_init, sa_driver, sa_free, sa_free_mb, &
>>>>>>>>> 1
>>>>>>>>> Error: Symbol 'saslave_init' referenced at (1) not found in module 'solvent_accessibility'
>>>>>>>>> make[3]: *** [pb_force.o] Error 1
>>>>>>>>> make[2]: *** [libpbsa] Error 2
>>>>>>>>> make[1]: *** [parallel] Error 2
>>>>>>>>> make: *** [install] Error 2
>>>>>>>>>
>>>>>>>>> Same problem. Are you sure you have the latest tree and it is clean and doesn't have any modified files locally or files you forgot to add?
>>>>>>>>>
>>>>>>>>> All the best
>>>>>>>>> Ross
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> On Feb 16, 2016, at 15:56, Ray Luo, Ph.D. <ray.luo.uci.edu> wrote:
>>>>>>>>>>
>>>>>>>>>> Hi Ross,
>>>>>>>>>>
>>>>>>>>>> So far I can "configure -mpi gnu", "make install", and can also pass
>>>>>>>>>> "make sander.BASIC.MPI". I'm running "make test.parallel.MM". I'm
>>>>>>>>>> using "mpirun -np 4" for the mpi jobs. This is on Rocks 6.1/Centos
>>>>>>>>>> 6.1. Which step was the problem?
>>>>>>>>>>
>>>>>>>>>> However I disabled "pbsa_mbfocus" in AmberTools/test/ since I'm in the
>>>>>>>>>> process of removing this feature to accommodate incoming CUDA. I also
>>>>>>>>>> need to update the test cases for sander/nab/mmpbsa ... which will be
>>>>>>>>>> partially in today for sander/nab. I'm trying to get mmpbsa.py to work
>>>>>>>>>> with the new python environment.
>>>>>>>>>>
>>>>>>>>>> All the best,
>>>>>>>>>> Ray
>>>>>>>>>> --
>>>>>>>>>> Ray Luo, Ph.D.
>>>>>>>>>> Professor
>>>>>>>>>> Biochemistry, Molecular Biophysics, Chemical Physics,
>>>>>>>>>> Chemical and Biomedical Engineering
>>>>>>>>>> University of California, Irvine, CA 92697-3900
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> On Tue, Feb 16, 2016 at 10:21 AM, Ray Luo, Ph.D. <ray.luo.uci.edu> wrote:
>>>>>>>>>>> Hi Ross,
>>>>>>>>>>>
>>>>>>>>>>> I'm looking into this ... There was a major overhaul of the code in
>>>>>>>>>>> addition to new features in the last check in.
>>>>>>>>>>>
>>>>>>>>>>> All the best,
>>>>>>>>>>>
>>>>>>>>>>> Ray
>>>>>>>>>>>
>>>>>>>>>>> --
>>>>>>>>>>> Ray Luo, Ph.D.
>>>>>>>>>>> Professor
>>>>>>>>>>> Biochemistry, Molecular Biophysics, Chemical Physics,
>>>>>>>>>>> Chemical and Biomedical Engineering
>>>>>>>>>>> University of California, Irvine, CA 92697-3900
>>>>>>>>>
>>>>>>>
>>>
>>> _______________________________________________
>>> AMBER-Developers mailing list
>>> AMBER-Developers.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber-developers
>>
>>
>>
>> --
>> -------------------------
>> Daniel R. Roe, PhD
>> Department of Medicinal Chemistry
>> University of Utah
>> 30 South 2000 East, Room 307
>> Salt Lake City, UT 84112-5820
>> http://home.chpc.utah.edu/~cheatham/
>> (801) 587-9652
>> (801) 585-6208 (Fax)
>>
>> _______________________________________________
>> AMBER-Developers mailing list
>> AMBER-Developers.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber-developers
>
>
> _______________________________________________
> AMBER-Developers mailing list
> AMBER-Developers.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber-developers
>
--
____________________________________________________________________________
Prof. Gerald MONARD
SRSMC, Université de Lorraine, CNRS
Boulevard des Aiguillettes B.P. 70239
F-54506 Vandoeuvre-les-Nancy, FRANCE
e-mail : Gerald.Monard.univ-lorraine.fr
tel. : +33 (0)383.684.381
fax : +33 (0)383.684.371
web : http://www.monard.info
____________________________________________________________________________
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Thu Feb 18 2016 - 11:30:05 PST
