Custom Search
-- Ray Luo, Ph.D. Professor Biochemistry, Molecular Biophysics, Chemical Physics, Chemical and Biomedical Engineering University of California, Irvine, CA 92697-3900 On Thu, Feb 18, 2016 at 11:23 AM, Ray Luo <rluo.uci.edu> wrote: > Thanks, I'm downloading your fix ... > > Also, the python-related installation seems to mess up the dependence > issue whether I use single or multi-thread make ... I don't know why > my changes in pbsa and sander have caused this to break. > > 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 Thu, Feb 18, 2016 at 11:08 AM, Gerald Monard > <Gerald.Monard.univ-lorraine.fr> wrote: >> 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 _______________________________________________ AMBER-Developers mailing list AMBER-Developers.ambermd.org http://lists.ambermd.org/mailman/listinfo/amber-developersReceived on Thu Feb 18 2016 - 13:00:03 PST