Legacy Mac Installation Problem

Problems with Dalton installation? Find answers or ask for help here
topper
Posts: 9
Joined: 12 Sep 2016, 14:48
First name(s): Robert
Middle name(s): Q.
Last name(s): Topper
Affiliation: Cooper Union
Country: United States

Legacy Mac Installation Problem

Post by topper » 12 Sep 2016, 16:54

Hi, I am trying (but failing) to install DALTON 2016 on a Mac running MacOS 10.7.5.
Can anyone please have a look here and see if they can give me a suggestion to overcome the problem I am encountering?
Many thanks for any help.

(I;ve edited this original post and attached the problematic output as a text file.)

Robert
Attachments
error.txt
(11.72 KiB) Downloaded 124 times
Last edited by topper on 15 Sep 2016, 17:20, edited 1 time in total.

taylor
Posts: 525
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Legacy Mac Installation Problem

Post by taylor » 12 Sep 2016, 18:31

I am sorry, I have no immediate suggestions, since I have never tried to build the code for a Mac --- you may get suggestions from other people. But can I please ask you to post actual files, as attachments, as opposed to posting things inline! It is really difficult to read postings in which things are cut-and-pasted inline! In my (browser-based) Email the formatting makes it almost useless. Please post files!

All that said, as I remarked, I do not really understand why there is a problem with your installation (despite being a long-time Mac user it has never occurred to me to build Dalton for the machine --- we have a Linux/x86 compute farm at home we use for calculations...) but maybe somebody with more experience can contribute. I hope someone can help!

Best regards
Pete

janus
Posts: 82
Joined: 27 Aug 2013, 22:05
First name(s): Janus
Middle name(s): Juul
Last name(s): Eriksen
Affiliation: Institut für Physikalische Chemie, Johannes Gutenberg-Universität Mainz
Country: Germany

Re: Legacy Mac Installation Problem

Post by janus » 12 Sep 2016, 19:56

You seem to be using different fortran (in /usr/local/bin/) and C/C++ compilers (in /Applications/Xcode.app/Contents/Developer/usr/bin/), which *might* be the problem. It would seem that why've you made symlinks for the C/C++ compilers, but not for the fortran one, since you're simply invoking them as gfortran, gcc, and g++ (maybe you've done this for a different project, for which you didn't need the fortran compiler, no?).

topper
Posts: 9
Joined: 12 Sep 2016, 14:48
First name(s): Robert
Middle name(s): Q.
Last name(s): Topper
Affiliation: Cooper Union
Country: United States

Re: Legacy Mac Installation Problem

Post by topper » 13 Sep 2016, 02:20

Thanks very much. I now see how to attach files, sorry for the giant post.

I very much appreciate the comments. I'm not quite sure how to address the multiple compilers issue, assuming that *is* the issue, but I'll try to research what combination of compilers is needed and how to get things set up so that they are the ones invoked.

taylor
Posts: 525
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Legacy Mac Installation Problem

Post by taylor » 13 Sep 2016, 03:36

What I am about to say may seem counterintuitive, but I think it may be useful. Assuming you have access to some external resource (departmental computing facility, central university facility, some sort of national high-performance computing setup) can you build the code successfully there? My point is that when one is using an external facility one has to rather carefully learn the details of using that facility: compilers, libraries, etc. And the psychology of this is quite different from building for "your own" machine, which you have been using for years and "know" as a result everything about it. It may well be (as has already been suggested) that you may be inadvertently overlooking something in your local setup. Whereas if you are forced to learn all the ins-and-outs of a new facility, you have a better chance of getting things "right". Any prospects of doing this?

Best regards
Pete

topper
Posts: 9
Joined: 12 Sep 2016, 14:48
First name(s): Robert
Middle name(s): Q.
Last name(s): Topper
Affiliation: Cooper Union
Country: United States

Re: Legacy Mac Installation Problem

Post by topper » 13 Sep 2016, 15:23

Pete, that's a good suggestion.

Unfortunately we have no departmental computing facility, no
access to the central facility (tightly controlled wrt what software is
installed). I do have limited run-only access to a high-performance
computing setup but can only run Gaussian, and I need to carry out an
CCSD(T) optimization - which Gaussian, q-Chem and GAMESS apparently cannot do
(alhough they all handle CCSD opt and CCSD(T) single point).
Plus I am up against a very tight publication deadline. I just have one
very small optimization that I have to run for confirmatory purposes.
If I can get Dalton working for me I'd like to use it a lot more in
the future for lots of things, though.

janus
Posts: 82
Joined: 27 Aug 2013, 22:05
First name(s): Janus
Middle name(s): Juul
Last name(s): Eriksen
Affiliation: Institut für Physikalische Chemie, Johannes Gutenberg-Universität Mainz
Country: Germany

Re: Legacy Mac Installation Problem

Post by janus » 13 Sep 2016, 22:11

To repeat what I said above, could you please try one of the following two options:

1) $> ./setup --fc=/Applications/Xcode.app/Contents/Developer/usr/bin/gfortran --cc=/Applications/Xcode.app/Contents/Developer/usr/bin/gcc --cxx=/Applications/Xcode.app/Contents/Developer/usr/bin/g++ build_4_2

2) $> ./setup --fc=/usr/local/bin/gfortran --cc=/usr/local/bin/gcc --cxx=/usr/local/bin/g++ build_4_8

(Depending on what size of system you need to run the (T) geo opt for, I'm not necessarily sure you want to use Dalton, but give it go by all means...)

topper
Posts: 9
Joined: 12 Sep 2016, 14:48
First name(s): Robert
Middle name(s): Q.
Last name(s): Topper
Affiliation: Cooper Union
Country: United States

Re: Legacy Mac Installation Problem

Post by topper » 14 Sep 2016, 14:44

Janus, thank you very much. I tried both of your suggestions. The first one failed at the setup stage with the error:

Could not find compiler set in environment variable FC:
/Applications/Xcode.app/Contents/Developer/usr/bin/gfortran

The second failed at the make stage:
[ 53%] Linking Fortran static library libgen1int.a
ar: CMakeFiles/gen1int.dir/src/geom/geo: No such file or directory
make[5]: *** [libgen1int.a] Error 1
make[5]: *** Deleting file `libgen1int.a'
make[4]: *** [CMakeFiles/gen1int.dir/all] Error 2
make[3]: *** [all] Error 2
make[2]: *** [external/gen1int-stamp/gen1int-build] Error 2
make[1]: *** [gen1int/CMakeFiles/gen1int.dir/all] Error 2
make: *** [all] Error 2

taylor
Posts: 525
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Legacy Mac Installation Problem

Post by taylor » 14 Sep 2016, 14:51

I am not sure why your gen1int build fails, but I have a pretty strong feeling you can do without building this. At least that was true in older versions. You should look at the options for the "setup" step. There are various things that can be enabled (like MPI, say), and disabled. If you can get even a "skeleton" version of the code built and running it should be possible to gradually sneak up on the other features, one step at a time.

Best regards
Pete

taylor
Posts: 525
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Legacy Mac Installation Problem

Post by taylor » 14 Sep 2016, 15:00

Another thought, although you probably don't want to do this (and maybe you really won't want to do this...). But you have basically an x86 box. There is nothing to prevent you installing a regular OpenBSD, or even Linux, on it: there's plenty of stuff on the web about this. Now, obviously, if you need some Mac software (or Microsoft software) this is not a useful path. But If you have some experience, or want to acquire some, as a sysadmin, it's a possibility.

We are Mac people at home for our lap/desktops and because of our needs we just stay with Mac OSX on these. But we have the advantage of ten other server class computers in what would otherwise be our spare bedroom, and in addition to running a compute farm we also run all sorts of Linux variants in virtual machines --- if you can find even a vaguely recent x86 desktop you should be able to explore the possibilities of building and running Dalton on other platforms. The performance may/will not be great, but at least you will know what works and what doesn't.

By the way, Dalton can be built, and will run at about 20% of the speed of a current desktop, on a Raspberry Pi. They are $35... I posted about this earlier in the year.

Best regards
Pete

olav
Posts: 143
Joined: 28 Aug 2013, 06:20
First name(s): Olav
Last name(s): Vahtras

Re: Legacy Mac Installation Problem

Post by olav » 14 Sep 2016, 16:40

If all else fails, you can try virtualization with Docker
  • Install Docker for your mac, https://docs.docker.com/v1.10/mac/step_one/
  • Start an ubuntu container (described in link above)
  • Install compilers and tools (apt install gfortran g++ liblapack-dev cmake)
  • Get the Dalton source to your virtual ubuntu machine and compile.

I am curious to see if it works!

Regards
Olav

janus
Posts: 82
Joined: 27 Aug 2013, 22:05
First name(s): Janus
Middle name(s): Juul
Last name(s): Eriksen
Affiliation: Institut für Physikalische Chemie, Johannes Gutenberg-Universität Mainz
Country: Germany

Re: Legacy Mac Installation Problem

Post by janus » 14 Sep 2016, 20:46

topper wrote:
topper:DALTON-Source topper$ ./setup
FC=gfortran CC=gcc CXX=g++ cmake -DENABLE_MPI=OFF -DENABLE_SGI_MPT=OFF -DENABLE_OMP=OFF -DENABLE_64BIT_INTEGERS=OFF -DENABLE_GPU=OFF -DENABLE_CUBLAS=OFF -DENABLE_REAL_SP=OFF -DENABLE_STATIC_LINKING=OFF -DENABLE_SCALAPACK=OFF -DCMAKE_BUILD_TYPE=release /Users/topper/Desktop/DALTON-Source

-- System : Darwin
-- Processor type : x86_64
[...]
-- The Fortran compiler identification is GNU 4.8.2
-- The C compiler identification is GNU 4.2.1
-- The CXX compiler identification is GNU 4.2.1
[...]
-- Check for working Fortran compiler: /usr/local/bin/gfortran
-- Check for working Fortran compiler: /usr/local/bin/gfortran -- works
[...]
-- Check for working C compiler: /Applications/Xcode.app/Contents/Developer/usr/bin/gcc
-- Check for working C compiler: /Applications/Xcode.app/Contents/Developer/usr/bin/gcc -- works
[...]
-- Check for working CXX compiler: /Applications/Xcode.app/Contents/Developer/usr/bin/g++
-- Check for working CXX compiler: /Applications/Xcode.app/Contents/Developer/usr/bin/g++ -- works
[...]
-- Build files have been written to: /Users/topper/Desktop/DALTON-Source/build/external/gen1int-build
[ 0%] Performing build step for 'gen1int'
Scanning dependencies of target gen1int
[ 0%] Building Fortran object CMakeFiles/gen1int.dir/src/xkind.F90.o
[ 1%] Building Fortran object CMakeFiles/gen1int.dir/src/london_ao.F90.o
[ 2%] Building Fortran object CMakeFiles/gen1int.dir/src/gen1int_geom.F90.o
[ 3%] Building Fortran object CMakeFiles/gen1int.dir/src/gen1int_carmom.F90.o
[ 4%] Building Fortran object CMakeFiles/gen1int.dir/src/gen1int_nucpot.F90.o
[ 4%] Building Fortran object CMakeFiles/gen1int.dir/src/gen1int_gaupot.F90.o
[ 5%] Building Fortran object CMakeFiles/gen1int.dir/src/gen1int_onehamil.F90.o
[ 6%] Building Fortran object CMakeFiles/gen1int.dir/src/gen1int.F90.o
[ 7%] Building Fortran object CMakeFiles/gen1int.dir/tools/str_decode.F90.o
[ 8%] Building Fortran object CMakeFiles/gen1int.dir/src/dump_info.F90.o
[ 9%] Building Fortran object CMakeFiles/gen1int.dir/src/error_stop.F90.o
[ 10%] Building Fortran object CMakeFiles/gen1int.dir/src/xtimer.F90.o
[ 11%] Building Fortran object CMakeFiles/gen1int.dir/src/tools/norm_contr_cgto.F90.o
[ 12%] Building Fortran object CMakeFiles/gen1int.dir/src/tools/norm_contr_sgto.F90.o
[ 13%] Building Fortran object CMakeFiles/gen1int.dir/src/tools/reorder_ints.F90.o
[ 13%] Building Fortran object CMakeFiles/gen1int.dir/src/tools/trace_ints.F90.o
[ 14%] Building Fortran object CMakeFiles/gen1int.dir/src/tools/get_address_list.F90.o
[ 15%] Building Fortran object CMakeFiles/gen1int.dir/src/basic/binom_coeff.F90.o
[ 16%] Building Fortran object CMakeFiles/gen1int.dir/src/basic/const_contr_ints.F90.o
[ 17%] Building Fortran object CMakeFiles/gen1int.dir/src/basic/hgto_to_cgto.F90.o
[ 18%] Building Fortran object CMakeFiles/gen1int.dir/src/basic/hgto_to_sgto.F90.o
[ 19%] Building Fortran object CMakeFiles/gen1int.dir/src/basic/next_permutation.F90.o
[ 19%] Building Fortran object CMakeFiles/gen1int.dir/src/basic/shell_scatter.F90.o
[ 20%] Building Fortran object CMakeFiles/gen1int.dir/src/basic/sort_cents.F90.o
[ 21%] Building Fortran object CMakeFiles/gen1int.dir/src/auxfun/aux_boys_vec.F90.o
[ 22%] Building Fortran object CMakeFiles/gen1int.dir/src/geom/geom_total.F90.o
[ 23%] Building Fortran object CMakeFiles/gen1int.dir/src/geom/geom_part_zero.F90.o
[ 24%] Building Fortran object CMakeFiles/gen1int.dir/src/geom/geom_part_one.F90.o
[ 25%] Building Fortran object CMakeFiles/gen1int.dir/src/mag/hgto_to_lcgto.F90.o
[ 25%] Building Fortran object CMakeFiles/gen1int.dir/src/mag/london_mom_hgto.F90.o
[ 26%] Building Fortran object CMakeFiles/gen1int.dir/src/carmom/carmom_deriv.F90.o
[ 27%] Building Fortran object CMakeFiles/gen1int.dir/src/carmom/carmom_hbra.F90.o
[ 28%] Building Fortran object CMakeFiles/gen1int.dir/src/carmom/carmom_hrr_ket.F90.o
[ 29%] Building Fortran object CMakeFiles/gen1int.dir/src/carmom/carmom_moment.F90.o
[ 30%] Building Fortran object CMakeFiles/gen1int.dir/src/carmom/prim_hgto_carmom.F90.o
[ 31%] Building Fortran object CMakeFiles/gen1int.dir/src/carmom/contr_cgto_carmom.F90.o
[ 32%] Building Fortran object CMakeFiles/gen1int.dir/src/carmom/contr_sgto_carmom.F90.o
[ 32%] Building Fortran object CMakeFiles/gen1int.dir/src/carmom/contr_csgto_carmom.F90.o
[ 33%] Building Fortran object CMakeFiles/gen1int.dir/src/delta/delta_geom.F90.o
[ 34%] Building Fortran object CMakeFiles/gen1int.dir/src/delta/delta_hket.F90.o
[ 35%] Building Fortran object CMakeFiles/gen1int.dir/src/delta/delta_moment.F90.o
[ 36%] Building Fortran object CMakeFiles/gen1int.dir/src/delta/prim_hgto_delta.F90.o
[ 37%] Building Fortran object CMakeFiles/gen1int.dir/src/delta/contr_cgto_delta.F90.o
[ 38%] Building Fortran object CMakeFiles/gen1int.dir/src/delta/contr_sgto_delta.F90.o
[ 38%] Building Fortran object CMakeFiles/gen1int.dir/src/nucpot/nucpot_geom.F90.o
[ 39%] Building Fortran object CMakeFiles/gen1int.dir/src/nucpot/nucpot_hket.F90.o
[ 40%] Building Fortran object CMakeFiles/gen1int.dir/src/nucpot/nucpot_hbra.F90.o
[ 41%] Building Fortran object CMakeFiles/gen1int.dir/src/nucpot/prim_hgto_nucpot.F90.o
[ 42%] Building Fortran object CMakeFiles/gen1int.dir/src/nucpot/contr_cgto_nucpot.F90.o
[ 43%] Building Fortran object CMakeFiles/gen1int.dir/src/nucpot/contr_sgto_nucpot.F90.o
[ 44%] Building Fortran object CMakeFiles/gen1int.dir/src/gaupot/gaupot_geom.F90.o
[ 44%] Building Fortran object CMakeFiles/gen1int.dir/src/gaupot/prim_hgto_gaupot.F90.o
[ 45%] Building Fortran object CMakeFiles/gen1int.dir/src/gaupot/contr_cgto_gaupot.F90.o
[ 46%] Building Fortran object CMakeFiles/gen1int.dir/src/gaupot/contr_sgto_gaupot.F90.o
[ 47%] Building Fortran object CMakeFiles/gen1int.dir/src/odist/prim_hgto_odist.F90.o
[ 48%] Building Fortran object CMakeFiles/gen1int.dir/src/odist/contr_cgto_odist.F90.o
[ 49%] Building Fortran object CMakeFiles/gen1int.dir/src/odist/contr_sgto_odist.F90.o
[ 50%] Building Fortran object CMakeFiles/gen1int.dir/src/value/prim_hgto_value.F90.o
[ 50%] Building Fortran object CMakeFiles/gen1int.dir/src/value/const_contr_gto.F90.o
[ 51%] Building Fortran object CMakeFiles/gen1int.dir/src/value/contr_cgto_value.F90.o
[ 52%] Building Fortran object CMakeFiles/gen1int.dir/src/value/contr_sgto_value.F90.o
[ 53%] Linking Fortran static library libgen1int.a
ar: CMakeFiles/gen1int.dir/src/geom/geom_part_zero: No such file or directory
make[5]: *** [libgen1int.a] Error 1
make[5]: *** Deleting file `libgen1int.a'
make[4]: *** [CMakeFiles/gen1int.dir/all] Error 2
make[3]: *** [all] Error 2
make[2]: *** [external/gen1int-stamp/gen1int-build] Error 2
make[1]: *** [gen1int/CMakeFiles/gen1int.dir/all] Error 2
make: *** [all] Error 2
Okay, so a number of things...

1 --- Forget about the built-in compilers (in /Applications/Xcode.app/Contents/Developer/usr/bin/). There's no need to pursue this any further (i.e., getting the gfortran compiler in addition to the existing gcc and g++ compilers), as these are anyways very old (4.2).
2 --- It's indeed strange that you're still having problems with the never 4.8 compilers...
3 --- How exactly did you obtain Dalton? (and why do you keep the source on your desktop?)

topper
Posts: 9
Joined: 12 Sep 2016, 14:48
First name(s): Robert
Middle name(s): Q.
Last name(s): Topper
Affiliation: Cooper Union
Country: United States

Re: Legacy Mac Installation Problem

Post by topper » 14 Sep 2016, 21:10

Olav, I may just try that virtualization option that you mentioned, thank you.

Janus, I am not sure how to respond to 1 and 2. All I can say is that I've installed other packages in the past on
this machine successfully. I'm definitely no expert, but I have been working on unix systems on and off since 1990,
writing small codes to get things done, and installing packages when needed. And this is the first time I've
really had any difficulties that I couldn't figure out how to surmount.

3 is easy - I downloaded the installer from http://daltonprogram.org/www/download.html after approval of my license request.

hjaaj
Posts: 282
Joined: 27 Jun 2013, 18:44
First name(s): Hans Jørgen
Middle name(s): Aagaard
Last name(s): Jensen
Affiliation: Universith of Southern Denmark
Country: Denmark

Re: Legacy Mac Installation Problem

Post by hjaaj » 14 Sep 2016, 21:52

Robert, I am regularly compiling dalton on my MacBook without problems. When compiling Dalton I use MacPorts for compilers, cmake, python, openBLAS, openMPI etc.
(If you have an Intel license, you can also use their compilers and mkl - but I do not know if they work with old Tiger.)
Thus I use gcc, gxx/g++, and gfortran from the MacPorts port, and nothing from Apple's XCode. (In your original post one can see you use the C and C++ compilers are from XCode, and gfortran you had installed yourself, I guess. A fortran compiler is not part of XCode, this is why Janus's suggestion no. 1 did not work.)

Alternatively, if you do not need gen1int, you can disable it before compiling with "make".

Code: Select all

cd build
ccmake ..
#in ccmake window:
# - find and set ENABLE_GEN1INT to off
# - find and set ENABLE_PELIB to off
# - configure
# - generate
make
(gen1int is currently used for ECPs and pelib, so if you do not want to use ECPs or pelib, you do not need gen1int.)

-- Hans Jørgen.

gaobin
Posts: 95
Joined: 27 Aug 2013, 17:56
First name(s): Bin
Last name(s): Gao
Affiliation: CTCC
Country: Norway

Re: Legacy Mac Installation Problem

Post by gaobin » 14 Sep 2016, 22:22

Hi,

Although I do not think it is the problem of Gen1Int (I speculate it is more related to the linking problem), I suggest that you disable it and PELib as suggested by Hans Jørgen. This is a quick solution, and because you can still use Dalton's old code for ECPs with some basis sets (sorry I forgot the limitation of Dalton's old code for ECPs), there will be no harm for you to disable these two.

Best regards
Gao

topper
Posts: 9
Joined: 12 Sep 2016, 14:48
First name(s): Robert
Middle name(s): Q.
Last name(s): Topper
Affiliation: Cooper Union
Country: United States

Re: Legacy Mac Installation Problem

Post by topper » 15 Sep 2016, 17:17

Many thanks. I tried Hans' suggestion. The installation proceeded all the way to 99% and then produced a new error as follows:

[ 98%] Building Fortran object CMakeFiles/dalton.dir/DALTON/gp/mpi_dummy.F.o
[ 99%] Building Fortran object CMakeFiles/dalton.dir/binary_info.F90.o
[ 99%] Linking Fortran static library lib/libdalton.a
ar: CMakeFiles/dalton.dir/DALTON/dft/fun-example4.c: No such file or directory
make[2]: *** [lib/libdalton.a] Error 1
make[2]: *** Deleting file `lib/libdalton.a'
make[1]: *** [CMakeFiles/dalton.dir/all] Error 2
make: *** [all] Error 2

Do you possibly have any other suggestions?
best, Robert
Attachments
error2.txt
(68.01 KiB) Downloaded 139 times

rob
Posts: 17
Joined: 15 Oct 2014, 13:43
First name(s): Roberto
Last name(s): Di Remigio
Affiliation: CTCC
Country: Norway

Re: Legacy Mac Installation Problem

Post by rob » 16 Sep 2016, 14:24

Can you try to run a verbose compilation on one process only? Run

Code: Select all

VERBOSE=1 make
and post the section of the output right before the error message.

topper
Posts: 9
Joined: 12 Sep 2016, 14:48
First name(s): Robert
Middle name(s): Q.
Last name(s): Topper
Affiliation: Cooper Union
Country: United States

Re: Legacy Mac Installation Problem

Post by topper » 16 Sep 2016, 14:52

Okay, I will. Right now I have finally (duh!) figured out how to carry out Hans' other suggestion, which I believe would be:

./setup --fc=gfortran --cc=gcc --cxx=g++

(whoops - that didn't work either)

topper
Posts: 9
Joined: 12 Sep 2016, 14:48
First name(s): Robert
Middle name(s): Q.
Last name(s): Topper
Affiliation: Cooper Union
Country: United States

Re: Legacy Mac Installation Problem

Post by topper » 16 Sep 2016, 15:00

Rob, here's the result of VERBOSE=1. I've also attached a complete record of the setup and compile attempt that it's from.

[ 52%] Building Fortran object CMakeFiles/gen1int.dir/src/value/contr_sgto_value.F90.o
/usr/local/bin/gfortran -DBUILD_F90_MODULE -I/Users/topper/Desktop/Dalton/DALTON2016.2-Source/external/gen1int/include -m64 -O3 -J/Users/topper/Desktop/Dalton/DALTON2016.2-Source/build/modules -c /Users/topper/Desktop/Dalton/DALTON2016.2-Source/external/gen1int/src/value/contr_sgto_value.F90 -o CMakeFiles/gen1int.dir/src/value/contr_sgto_value.F90.o
[ 53%] Linking Fortran static library libgen1int.a
/Applications/CMake.app/Contents/bin/cmake -P CMakeFiles/gen1int.dir/cmake_clean_target.cmake
/Applications/CMake.app/Contents/bin/cmake -E cmake_link_script CMakeFiles/gen1int.dir/link.txt --verbose=1
/Applications/Xcode.app/Contents/Developer/usr/bin/ar qc libgen1int.a CMakeFiles/gen1int.dir/src/xkind.F90.o CMakeFiles/gen1int.dir/src/dump_info.F90.o CMakeFiles/gen1int.dir/src/error_stop.F90.o CMakeFiles/gen1int.dir/src/xtimer.F90.o CMakeFiles/gen1int.dir/src/tools/norm_contr_cgto.F90.o CMakeFiles/gen1int.dir/src/tools/norm_contr_sgto.F90.o CMakeFiles/gen1int.dir/src/tools/reorder_ints.F90.o CMakeFiles/gen1int.dir/src/tools/trace_ints.F90.o CMakeFiles/gen1int.dir/src/tools/get_address_list.F90.o CMakeFiles/gen1int.dir/src/basic/binom_coeff.F90.o CMakeFiles/gen1int.dir/src/basic/const_contr_ints.F90.o CMakeFiles/gen1int.dir/src/basic/hgto_to_cgto.F90.o CMakeFiles/gen1int.dir/src/basic/hgto_to_sgto.F90.o CMakeFiles/gen1int.dir/src/basic/next_permutation.F90.o CMakeFiles/gen1int.dir/src/basic/shell_scatter.F90.o CMakeFiles/gen1int.dir/src/basic/sort_cents.F90.o CMakeFiles/gen1int.dir/src/auxfun/aux_boys_vec.F90.o CMakeFiles/gen1int.dir/src/geom/geom_total.F90.o CMakeFiles/gen1int.dir/src/geom/geom_part_zero
ar: CMakeFiles/gen1int.dir/src/geom/geom_part_zero: No such file or directory
make[5]: *** [libgen1int.a] Error 1
make[5]: *** Deleting file `libgen1int.a'
make[4]: *** [CMakeFiles/gen1int.dir/all] Error 2
make[3]: *** [all] Error 2
make[2]: *** [external/gen1int-stamp/gen1int-build] Error 2
make[1]: *** [gen1int/CMakeFiles/gen1int.dir/all] Error 2
make: *** [all] Error 2
topper:build topper$
Attachments
error3.txt
(51.84 KiB) Downloaded 122 times

gaobin
Posts: 95
Joined: 27 Aug 2013, 17:56
First name(s): Bin
Last name(s): Gao
Affiliation: CTCC
Country: Norway

Re: Legacy Mac Installation Problem

Post by gaobin » 16 Sep 2016, 17:06

Hi

Although I do not think the problem is related to Gen1Int, could you disable it and pelib, and make with VERBOSE=1?

hjaaj
Posts: 282
Joined: 27 Jun 2013, 18:44
First name(s): Hans Jørgen
Middle name(s): Aagaard
Last name(s): Jensen
Affiliation: Universith of Southern Denmark
Country: Denmark

Re: Legacy Mac Installation Problem

Post by hjaaj » 16 Sep 2016, 17:26

It seems to be a very strange, unusual cmake error to me. The file name in the line causing the error should be "geom_part_zero.F90.o" instead of "geom_part_zero" - analogously with the previous lines. Very strange. I am not a cmake expert, does any cmake expert have an idea of what might be happening?

-- Hans Jørgen.

hjaaj
Posts: 282
Joined: 27 Jun 2013, 18:44
First name(s): Hans Jørgen
Middle name(s): Aagaard
Last name(s): Jensen
Affiliation: Universith of Southern Denmark
Country: Denmark

Re: Legacy Mac Installation Problem

Post by hjaaj » 16 Sep 2016, 17:35

I now looked at the error3.txt you attached. The problematic line there is:

Code: Select all

/Applications/Xcode.app/Contents/Developer/usr/bin/ar qc libgen1int.a  CMakeFiles/gen1int.dir/src/xkind.F90.o CMakeFiles/gen1int.dir/src/dump_info.F90.o CMakeFiles/gen1int.dir/src/error_stop.F90.o CMakeFiles/gen1int.dir/src/xtimer.F90.o CMakeFiles/gen1int.dir/src/tools/norm_contr_cgto.F90.o CMakeFiles/gen1int.dir/src/tools/norm_contr_sgto.F90.o CMakeFiles/gen1int.dir/src/tools/reorder_ints.F90.o CMakeFiles/gen1int.dir/src/tools/trace_ints.F90.o CMakeFiles/gen1int.dir/src/tools/get_address_list.F90.o CMakeFiles/gen1int.dir/src/basic/binom_coeff.F90.o CMakeFiles/gen1int.dir/src/basic/const_contr_ints.F90.o CMakeFiles/gen1int.dir/src/basic/hgto_to_cgto.F90.o CMakeFiles/gen1int.dir/src/basic/hgto_to_sgto.F90.o CMakeFiles/gen1int.dir/src/basic/next_permutation.F90.o CMakeFiles/gen1int.dir/src/basic/shell_scatter.F90.o CMakeFiles/gen1int.dir/src/basic/sort_cents.F90.o CMakeFiles/gen1int.dir/src/auxfun/aux_boys_vec.F90.o CMakeFiles/gen1int.dir/src/geom/geom_total.F90.o CMakeFiles/gen1int.dir/src/geom/geom_part_zero
which counts to 1023 characters.

My guess is now that your cmake for some strange stupid reason is with a 1024 line length limit (giving 1023 characters plus a newline character). I myself am using the cmake from MacPorts.

-- Hans Jørgen.

topper
Posts: 9
Joined: 12 Sep 2016, 14:48
First name(s): Robert
Middle name(s): Q.
Last name(s): Topper
Affiliation: Cooper Union
Country: United States

Re: Legacy Mac Installation Problem

Post by topper » 16 Sep 2016, 18:46

Thank you, Hans! What a great catch!
I installed MacPorts and used it to bring in cmake, and the installation completed.
Haven't yet run the code, but so far it looks like I have an executable image.

In addition to helping me solve this problem, you've also educated me as to how I can keep my system up to speed
using MacPorts, so that's a terrific help as well.
Looking forward to testing out Dalton.

Thanks again to you and everyone else for your guidance.

taylor
Posts: 525
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Legacy Mac Installation Problem

Post by taylor » 17 Sep 2016, 09:35

I rather wish I had jumped in a bit earlier because I myself am much less of a Macports enthusiast. I started to use it about five years ago and eventually got to a stage where it wanted to update one package, but it couldn't do that without updating another package, and it couldn't update the latter package without updating the former package, and I was, to use a complicated technical term from computer engineering and maintenance, wedged. Or, old-school, "deadly embrace", see PS...

Part of the problem was that I had probably installed a lot more "stuff" than was actually needed, so I can accept that getting Macports confused might be partly my fault, but I could never get it resolved. In the meantime I had mentioned this to my partner and said "don't go there!" and so she went with the then-alternative, which was Fink. This turned out to be even more of a disaster (and I still get crap for it...).

In my view there are two safer alternatives (others may disagree, of course) for the Mac, and I made this transition when I swapped my Macbook Pro for an Air. The first is to use Homebrew. The list of supported/available software may be shorter than Macports, but the reliability and solidity, is in my view incomparably better. The second, and is undoubtedly (again, in my view) the best alternative, but which probably requires paying a bit for the software, is to install a virtual machine (VM) manager and thus use virtualization. I use VMWare on the Mac, although our (Linux-based) servers at home run Centos and this ships, free, with KVM. But even if you have to pay a few bob for VMWare for the Mac, as long as your chipset has hardware VM support things should be fine. In this way you can run almost any flavour of Linux, Windows (not that we do, he said in horrified tones!), OpenBSD, whatever, on your system.

I return briefly to Macports: I do not want to just seem to be bagging on it. But it did end up causing me considerable (perhaps partly self-inflicted, I admit) problems and I think there are (much) better solutions.

Best regards
Pete
P.S. The term "deadly embrace" was used for such Catch-22 situations in the '70s and '80s (yes, I was there, although only a lad...). It was used to refer to the situation where a multiuser/time-sharing system had (say) two tape drives. (By "tape drive" I mean by the way one of those classic movie old-school devices with a big spool of tape going forward and backward that told the viewer that this was some serious shit type of computer and computing going on...)

So user A's job asks for a particular tape to be mounted, and the operator (a human, can you believe it?!) went off, fetched that particular tape, and mounted it. Meanwhile, user B's job has asked for a particular tape to be mounted, and another operator has gone off and done that, using the second tape drive. And no doubt the operators have had a bitch-fest about how if it weren't for these damn users and their jobs they could go have a smoke... And then user A's job asks for a second tape... There isn't a drive free. And then user B's job asks for a second tape... And both jobs are completely deadlocked: neither can complete, and the only solution is to kill one and use the drive for the other job. So whose job do you kill? Or do you kill both? This form of deadlock was what was termed a "deadly embrace"...

taylor
Posts: 525
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Legacy Mac Installation Problem

Post by taylor » 17 Sep 2016, 09:46

And (in terms of my PS) yes, yes, the obvious solution was to have user jobs specify up front, in the (1980s equivalent) of the jobscript, that they needed, say, two tape drives. But unfortunately that functionality/requirement had not been implemented in some (e.g., Control Data's) operating systems...

Best regards
Pete

Post Reply

Who is online

Users browsing this forum: No registered users and 5 guests