Re: [AMBER-Developers] I don't want to change these files... really!

From: David Cerutti <dscerutti.gmail.com>
Date: Wed, 20 Sep 2017 14:55:28 -0400

I have a second amber clone on the back burner--hadn't done anything to it,
really, since I checked it out and went into one of the curated branches.
So I did git status:

On branch pmemd_neb
Your branch is ahead of 'origin/pmemd_neb' by 1 commit.
  (use "git push" to publish your local commits)
Untracked files:
  (use "git add <file>..." to include in what will be committed)
        AmberTools/src/build-bot/ambertools-test/
        AmberTools/src/rism/PeanoPort.sh
        benchmarks/cellulose/blah
        bin/
        include/
        src/pmemd/src/cuda/Casegroup4Port.sh
        src/pmemd/src/cuda/PeanoPort.sh
nothing added to commit but untracked files present (use "git add" to track)

Looks fine, no changes to stage for commit. I then did git checkout master:

M AmberTools/src/ambertools-binary-build
M AmberTools/src/parmed
Switched to branch 'master'
Your branch is up-to-date with 'origin/master'.

Aha... that's starting to look like what happened in my other clone.
Checking out master gets weird merges of files I haven't touched. If I now
do git pull, then git status I see...

remote: Counting objects: 584, done.
remote: Compressing objects: 100% (267/267), done.
remote: Total 269 (delta 217), reused 2 (delta 1)
Receiving objects: 100% (269/269), 67.96 KiB | 0 bytes/s, done.
Resolving deltas: 100% (217/217), completed with 128 local objects.
>From git.ambermd.org:/amber
   d5dac88..6ea8827 master -> origin/master
   1a924d0..d3db673 SpeedBoostSM -> origin/SpeedBoostSM
   6cc4cde..3d6a7f1 cmake -> origin/cmake
   5e9bda4..e1a5d1a pmemd_neb -> origin/pmemd_neb
   b47f3a0..816c236 redox-vwdc -> origin/redox-vwdc
Fetching submodule AmberTools/src/cpptraj
>From github.com:/Amber-MD/cpptraj
   e94f2aa..bed670e master -> origin/master
Fetching submodule cmake
>From https://github.com/Amber-MD/cmake-buildscripts
   cb7c08b..b225014 experimental -> origin/experimental
   4a3ee59..be607b5 master -> origin/master
Updating d5dac88..6ea8827
Fast-forward
 AmberTools/src/mpi4py-2.0.0/CMakeLists.txt | 3 +-
 AmberTools/src/sander/cmd_matrix_a1st.F90 | 336 ---
 AmberTools/src/sander/ew_force.F90 | 2 +-
 AmberTools/src/sander/iwrap2.F90 | 2 +-
 AmberTools/test/Makefile | 6 +-
 AmberTools/test/antechamber/acdoctor/Run.acdoctor | 47 +-
 AmberTools/test/antechamber/acdoctor/mpd.mol2.save | 84 +-
 AmberTools/test/antechamber/acdoctor/mpd_mopout.out.save | 42 -
 AmberTools/test/rism1d/spc-kh/spc.gvv_dT | 4005
-------------------------
 cmake | 2 +-
 src/pmemd.midpoint/src/parallel_dat.F90 | 6 +-
 src/pmemd/src/gb_ene_hybrid.F90 | 3 +-
 src/pmemd/src/parallel_dat.F90 | 6 +-
 test/rism3d/phosphate/Run.ph1 | 2 +-
 test/rism3d/phosphate/Run.ph1-nacl | 2 +-
 test/rism3d/phosphate/Run.ph1-nacl.pse1 | 2 +-
 test/rism3d/phosphate/Run.ph1-nacl.pse3 | 2 +-
 test/rism3d/phosphate/Run.ph1.pse1 | 2 +-
 test/rism3d/phosphate/Run.ph1.pse3 | 2 +-
 test/rism3d/phosphate/Run.ph10 | 2 +-
 test/rism3d/phosphate/Run.ph10-nacl | 2 +-
 test/rism3d/phosphate/Run.ph2 | 2 +-
 test/rism3d/phosphate/Run.ph20 | 2 +-
 23 files changed, 94 insertions(+), 4470 deletions(-)
 delete mode 100644 AmberTools/src/sander/cmd_matrix_a1st.F90
 delete mode 100644 AmberTools/test/rism1d/spc-kh/spc.gvv_dT

and...

On branch master
Your branch is up-to-date with 'origin/master'.
Changes not staged for commit:
  (use "git add <file>..." to update what will be committed)
  (use "git checkout -- <file>..." to discard changes in working directory)
        modified: AmberTools/src/ambertools-binary-build (new commits)
        modified: AmberTools/src/parmed (new commits)
        modified: cmake (new commits)
Untracked files:
...

So it seems I've replicated the error, at least...


On Wed, Sep 20, 2017 at 2:49 PM, David Cerutti <dscerutti.gmail.com> wrote:

> 2.9.5...
>
> From the location it looks like some submodule may not be properly
> addressing everything from the root directory. I was in src/pmemd/src/cuda
> when I did "git submodule update --init," which should be just fine, but it
> said it was searching for something in AmberTools/src/... as if it were a
> subdirectory of pmemd/src/cuda. I went back up to the root amber
> directory. Still got the error, albeit now saying
>
> fatal: No url found for submodule path 'AmberTools/src/build-bot/ambertools-test'
> in .gitmodules
>
> If my clone is borked, I don't know how best to get the good changes out
> of it and into a pristine clone before pushing.
>
> Dave
>
>
> On Wed, Sep 20, 2017 at 2:41 PM, Daniel Roe <daniel.r.roe.gmail.com>
> wrote:
>
>> Not sure why you're getting that error - sounds like your GIT tree may
>> be borked in some way. What version of GIT are you using?
>>
>> -Dan
>>
>> On Wed, Sep 20, 2017 at 2:32 PM, David Cerutti <dscerutti.gmail.com>
>> wrote:
>> > I had been doing "git submodule update --remote --merge," but I tried
>> > submodule update --init again, and I'm getting
>> >
>> > fatal: No url found for submodule path
>> > 'src/pmemd/src/cuda/AmberTools/src/build-bot/ambertools-test' in
>> .gitmodules
>> >
>> >
>> > On Wed, Sep 20, 2017 at 2:29 PM, Daniel Roe <daniel.r.roe.gmail.com>
>> wrote:
>> >
>> >> Make sure your submodules are up to date:
>> >>
>> >> git submodule update --init
>> >>
>> >> With some versions of git this is needed to bring them in line.
>> >>
>> >> -Dan
>> >>
>> >> On Wed, Sep 20, 2017 at 2:22 PM, David Cerutti <dscerutti.gmail.com>
>> >> wrote:
>> >> > I'm hoping someone can help me to understand what's going on with my
>> >> merges
>> >> > from SpeedBoostSM into master. I'm preparing to commit my
>> performance
>> >> > enhancements, and the merge goes more or less smoothly except that,
>> >> after I
>> >> > merge from commit d3db6730442cee05bc7c048b0bb89213d9a6000f of
>> >> SpeedBoostSM
>> >> > (the point at which I deemed things in a suitable state--I've made
>> other
>> >> > commits since then to continue progress), I am still getting the
>> >> following:
>> >> >
>> >> > On branch master
>> >> > Your branch is ahead of 'origin/master' by 59 commits.
>> >> > (use "git push" to publish your local commits)
>> >> > Changes not staged for commit:
>> >> > (use "git add <file>..." to update what will be committed)
>> >> > (use "git checkout -- <file>..." to discard changes in working
>> >> directory)
>> >> > modified: ../../../../AmberTools/src/am
>> bertools-binary-build
>> >> (new
>> >> > commits)
>> >> > modified: ../../../../AmberTools/src/cpptraj (new commits)
>> >> >
>> >> > Untracked files:
>> >> > ...
>> >> >
>> >> > It's the business of ambertools-binary-build bearing "changes"
>> (which,
>> >> for
>> >> > all I can tell with git diff, are perhaps one whitespace
>> character)--same
>> >> > with the cpptraj directory. When I checkout these files /
>> directories
>> >> from
>> >> > master I still see "Changes not staged for commit."
>> >> >
>> >> > Hai pointed out a similar issue in one of my previous commits
>> involving
>> >> the
>> >> > test cases. I'm not sure what it is, but something seems to be
>> >> > persistently lacing these files with some miniscule tweak that I
>> can't
>> >> help
>> >> > but make--I'm not doing anything to these files deliberately. Can
>> anyone
>> >> > suggest what might be going on?
>> >> >
>> >> > Cheers,
>> >> >
>> >> > Dave
>> >> > _______________________________________________
>> >> > AMBER-Developers mailing list
>> >> > AMBER-Developers.ambermd.org
>> >> > http://lists.ambermd.org/mailman/listinfo/amber-developers
>> >>
>> >>
>> >>
>> >> --
>> >> -------------------------
>> >> Daniel R. Roe
>> >> Laboratory of Computational Biology
>> >> National Institutes of Health, NHLBI
>> >> 5635 Fishers Ln, Rm T900
>> >> Rockville MD, 20852
>> >> https://www.lobos.nih.gov/lcb
>> >>
>> >> _______________________________________________
>> >> 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
>>
>>
>>
>> --
>> -------------------------
>> Daniel R. Roe
>> Laboratory of Computational Biology
>> National Institutes of Health, NHLBI
>> 5635 Fishers Ln, Rm T900
>> Rockville MD, 20852
>> https://www.lobos.nih.gov/lcb
>>
>> _______________________________________________
>> 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
Received on Wed Sep 20 2017 - 12:00:06 PDT
Custom Search