1. 17 May, 2017 1 commit
  2. 16 May, 2017 4 commits
  3. 15 May, 2017 1 commit
  4. 27 Apr, 2017 2 commits
  5. 26 Apr, 2017 1 commit
    • Lorenz Huedepohl's avatar
      Workaround for PGI compiler · f4aece79
      Lorenz Huedepohl authored
      PGI's compiler wants
      
        -module
      
      as the flag to specify the module output directory. Sadly, this is also
      a libtool argument, resulting in much sorrow.
      
      Thus, we have to escape -module by prefixing it with -Xcompiler. Sadly
      libtool is not used everywhere by automake, thus we have to remove it
      again in those cases were it would be passed to the compiler and cause
      sorrow there.
      
      Sad story. But with a happy ending?
      f4aece79
  6. 24 Apr, 2017 1 commit
  7. 21 Apr, 2017 2 commits
  8. 18 Apr, 2017 1 commit
  9. 06 Apr, 2017 1 commit
  10. 16 Mar, 2017 1 commit
  11. 10 Mar, 2017 1 commit
  12. 13 Feb, 2017 2 commits
    • Lorenz Huedepohl's avatar
      Work around Intel compiler oddities · 7e1dea02
      Lorenz Huedepohl authored
      For some reason, and only in certain cases, the built-in
      preprocessor of ifort cannot handle constructs such as
      
        #define FOO bar
      
        print *, "foo&
        &FOO&
        &baz"
      
      or more specifically
      
        #define DATATYPE double
      
        print *, some_&
        &DATATYPE&
        &_routine
      
      which, however, are quite handy. GCC seems to handle those as
      expected.
      7e1dea02
    • Andreas Marek's avatar
      Make blocking in trans_ev_band_to_full default · f3ce5bd3
      Andreas Marek authored
      This gives a speed-up of up to a factor 2 in the complex version
      f3ce5bd3
  13. 12 Feb, 2017 3 commits
  14. 08 Feb, 2017 1 commit
    • Andreas Marek's avatar
      Introduce configure flag for CUDA GPU compute capability · 4126e87e
      Andreas Marek authored
      Now the CUDA compute capability is not hard-coded anymore.
      Setting
      
      --with-gpu-compute-capability=value
      
      set the nvcc compiler option "-arch value".
      
      A simple sanity checking is done to ensure that "value" is not
      completely unreasonable
      4126e87e
  15. 25 Nov, 2016 1 commit
    • Andreas Marek's avatar
      Update of ELPA version number to 2016.11.001.pre · a1da48c1
      Andreas Marek authored
      - the version number has been updated
      - a warning that this is a pre-release has been added to the configure
        script
      - the changelog has been updated
      - the test_project versioning is consistent with the ELPA versioning
      a1da48c1
  16. 23 Nov, 2016 1 commit
  17. 11 Nov, 2016 1 commit
  18. 09 Nov, 2016 1 commit
  19. 27 Oct, 2016 1 commit
  20. 26 Oct, 2016 1 commit
  21. 25 Oct, 2016 1 commit
  22. 18 Oct, 2016 1 commit
  23. 12 Oct, 2016 1 commit
  24. 29 Sep, 2016 1 commit
  25. 28 Sep, 2016 2 commits
  26. 23 Sep, 2016 1 commit
  27. 02 Sep, 2016 3 commits
  28. 22 Aug, 2016 2 commits
    • Andreas Marek's avatar
      Better control of AVX kernels · d4ab9550
      Andreas Marek authored
      Sometimes it might be desirable that only one of the specifc AVX
      kernels (AVX, AVX2, and in the future AVX512) are build, even though the processor
      supports all (or several) of these instructions.
      
      The flag --disable-[avx | avx2 | avx512] allows to define which kernels
      should NOT be build. The default is: what can be build will be build
      d4ab9550
    • Andreas Marek's avatar
      Better control of AVX kernels · 5a1efab3
      Andreas Marek authored
      Sometimes it might be desirable that only one of the specifc AVX
      kernels (AVX, AVX2, AVX512) are build, even though the processor
      supports all (or several) of these instructions.
      
      The flag --disable-[avx | avx2 | avx512] allows to define which kernels
      should NOT be build. The default is: what can be build will be build
      5a1efab3