Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-9424

add gfortran to OSX build deps?

    Details

      Description

      gfortran is required to build oorb on osx, which is a dep of lsst_sims. Brew has removed the gfortran package and the front end is now provided by the brew gcc package/bottle.

      This is a sims deps but I believe gfortran is technically also a pipelines dep.

        Attachments

          Issue Links

            Activity

            Hide
            jsick Jonathan Sick added a comment - - edited

            Okay if we can establish that gfortran is a DM Stack dep then we should certainly add it. Otherwise I think the agreement is that Sims would prefer to own their own installation docs, which are here: https://confluence.lsstcorp.org/display/SIM/Catalogs+and+MAF

            Show
            jsick Jonathan Sick added a comment - - edited Okay if we can establish that gfortran is a DM Stack dep then we should certainly add it. Otherwise I think the agreement is that Sims would prefer to own their own installation docs, which are here: https://confluence.lsstcorp.org/display/SIM/Catalogs+and+MAF
            Hide
            frossie Frossie Economou added a comment -

            Assigning to Tim Jenness as the TCT chair to call the gfrortran issue - please reassign back to me with the recommended implementation.

            Show
            frossie Frossie Economou added a comment - Assigning to Tim Jenness as the TCT chair to call the gfrortran issue - please reassign back to me with the recommended implementation.
            Hide
            tjenness Tim Jenness added a comment -

            I have not previously heard that gfortran was a DM dependency. Which package? SQuaRE needs to worry about it regardless because SQuaRE also builds Sims but I don't think this is a DM issue for now (it will be in the future when MOPS becomes a DM thing).

            This has all been true for a year. What's the problem with just brew installing gfortran?

            Show
            tjenness Tim Jenness added a comment - I have not previously heard that gfortran was a DM dependency. Which package? SQuaRE needs to worry about it regardless because SQuaRE also builds Sims but I don't think this is a DM issue for now (it will be in the future when MOPS becomes a DM thing). This has all been true for a year. What's the problem with just brew installing gfortran?
            Hide
            mtpatter Maria Patterson [X] (Inactive) added a comment -

            I don't understand how jointcal quite fits in, but I just want to note that I ran into issues having to build oorb for jointcal so there is a dependency there.

            Show
            mtpatter Maria Patterson [X] (Inactive) added a comment - I don't understand how jointcal quite fits in, but I just want to note that I ran into issues having to build oorb for jointcal so there is a dependency there.
            Hide
            jsick Jonathan Sick added a comment -

            Oh, sorry, I was thinking of this ticket only in the context of https://pipelines.lsst.io/install/newinstall.html

            That's a separate issue from defining dependencies for CI environments that test lsst_sims.

            Show
            jsick Jonathan Sick added a comment - Oh, sorry, I was thinking of this ticket only in the context of https://pipelines.lsst.io/install/newinstall.html That's a separate issue from defining dependencies for CI environments that test lsst_sims .
            Hide
            jhoblitt Joshua Hoblitt added a comment -

            I believe that gfortran was needed at build time by a DM package at some point in the past. I also have vague memories of a conda package requiring a gfortran .so on linux. In any event, it has been installed on the jenkins build nodes since at least March 30th, 2015: https://github.com/lsst-sqre/puppet-lsststack/commit/2b7d21bf3008332004c326b436386d8500c96188 It also appears in the developer.lsst.io package lists for linux.

            Show
            jhoblitt Joshua Hoblitt added a comment - I believe that gfortran was needed at build time by a DM package at some point in the past. I also have vague memories of a conda package requiring a gfortran .so on linux. In any event, it has been installed on the jenkins build nodes since at least March 30th, 2015: https://github.com/lsst-sqre/puppet-lsststack/commit/2b7d21bf3008332004c326b436386d8500c96188 It also appears in the developer.lsst.io package lists for linux.
            Hide
            tjenness Tim Jenness added a comment -

            I don't think DM depends on gfortran. If you know otherwise please let me know.

            Show
            tjenness Tim Jenness added a comment - I don't think DM depends on gfortran. If you know otherwise please let me know.
            Hide
            jhoblitt Joshua Hoblitt added a comment -

            It might be worth creating an EUPS stub package for gfortran so the dep is at least tracked in some way. Currently, oorb has a check in the eupspkg prep step for it.

            Show
            jhoblitt Joshua Hoblitt added a comment - It might be worth creating an EUPS stub package for gfortran so the dep is at least tracked in some way. Currently, oorb has a check in the eupspkg prep step for it.
            Hide
            tjenness Tim Jenness added a comment -

            How does jointcal trigger an oorb dependency? jointcal doesn't list it as a dependency and doesn't list any Sims packages.

            Show
            tjenness Tim Jenness added a comment - How does jointcal trigger an oorb dependency? jointcal doesn't list it as a dependency and doesn't list any Sims packages.
            Hide
            tjenness Tim Jenness added a comment - - edited

            To be more concrete about jointcal:

            $ rebuild jointcal
                        jointcal:  ok (0.5 sec).
                           scons:  ok (0.5 sec).
                          python:  ok (0.4 sec).
                      sconsUtils:  ok (1.0 sec).
                         doxygen:  ok (0.4 sec).
                           eigen:  ok (0.4 sec).
                           boost:  ok (0.5 sec).
                             afw:  ok (2.1 sec).
                        daf_base:  ok (1.0 sec).
                           utils:  ok (0.9 sec).
                            base:  ok (0.9 sec).
                            swig:  ok (0.5 sec).
                   python_future:  ok (0.5 sec).
                  pex_exceptions:  ok (0.8 sec).
                           numpy:  ok (0.4 sec).
                   python_psutil:  ok (0.5 sec).
                 daf_persistence:  ok (1.5 sec).
                             log:  ok (0.9 sec).
                         log4cxx:  ok (0.5 sec).
                             apr:  ok (0.4 sec).
                        apr_util:  ok (0.5 sec).
                   mariadbclient:  ok (0.5 sec).
                      pex_policy:  ok (0.8 sec).
                         astropy:  ok (0.4 sec).
                          pyyaml:  ok (0.4 sec).
                      pex_config:  ok (0.9 sec).
                         ndarray:  ok (0.6 sec).
                            fftw:  ok (0.4 sec).
                         cfitsio:  ok (0.5 sec).
                          wcslib:  ok (0.5 sec).
                         minuit2:  ok (0.5 sec).
                             gsl:  ok (0.4 sec).
                          pyfits:  ok (0.5 sec).
                    python_d2to1:  ok (0.4 sec).
                 stsci_distutils:  ok (0.5 sec).
                      matplotlib:  ok (0.5 sec).
                         afwdata:  ok (0.8 sec).
                        obs_base:  ok (1.1 sec).
                          skypix:  ok (0.7 sec).
                            geom:  ok (0.7 sec).
                 meas_algorithms:  ok (1.7 sec).
                          esutil:  ok (0.5 sec).
                       meas_base:  ok (1.5 sec).
                     coadd_utils:  ok (0.8 sec).
                       pipe_base:  ok (0.9 sec).
                        obs_test:  ok (0.9 sec).
                      pipe_tasks:  ok (1.5 sec).
                     meas_astrom:  ok (1.0 sec).
                  astrometry_net:  ok (0.4 sec).
                  meas_deblender:  ok (1.1 sec).
             astrometry_net_data:  ok (0.4 sec).
                          ip_isr:  ok (0.9 sec).
                       ip_diffim:  ok (1.0 sec).
                           lmfit:  ok (0.4 sec).
                           scipy:  ok (0.4 sec).
                coadd_chisquared:  ok (0.7 sec).
                          skymap:  ok (0.7 sec).
                          healpy:  ok (0.6 sec).
                     pykg_config:  ok (0.4 sec).
                jointcal_cholmod:  ok (0.4 sec).
               testdata_jointcal:  ok (120.2 sec).
                     obs_lsstSim:  ok (1.2 sec).
              python_mysqlclient:  ok (0.5 sec).
                        obs_cfht:  ok (0.9 sec).
                   testdata_cfht:  ok (0.7 sec).
                       obs_decam:  ok (0.9 sec).
                  testdata_decam:  ok (0.6 sec).
                      obs_subaru:  ok (1.0 sec).
                   meas_modelfit:  ok (1.0 sec).
                        shapelet:  ok (0.8 sec).
            meas_extensions_psfex:  ok (0.8 sec).
                           psfex:  ok (0.7 sec).
            meas_extensions_shapeHSM:  ok (0.7 sec).
                          galsim:  ok (0.5 sec).
                             tmv:  ok (0.5 sec).
                 testdata_subaru:  ok (0.7 sec).
            

            Show
            tjenness Tim Jenness added a comment - - edited To be more concrete about jointcal : $ rebuild jointcal jointcal: ok (0.5 sec). scons: ok (0.5 sec). python: ok (0.4 sec). sconsUtils: ok (1.0 sec). doxygen: ok (0.4 sec). eigen: ok (0.4 sec). boost: ok (0.5 sec). afw: ok (2.1 sec). daf_base: ok (1.0 sec). utils: ok (0.9 sec). base: ok (0.9 sec). swig: ok (0.5 sec). python_future: ok (0.5 sec). pex_exceptions: ok (0.8 sec). numpy: ok (0.4 sec). python_psutil: ok (0.5 sec). daf_persistence: ok (1.5 sec). log: ok (0.9 sec). log4cxx: ok (0.5 sec). apr: ok (0.4 sec). apr_util: ok (0.5 sec). mariadbclient: ok (0.5 sec). pex_policy: ok (0.8 sec). astropy: ok (0.4 sec). pyyaml: ok (0.4 sec). pex_config: ok (0.9 sec). ndarray: ok (0.6 sec). fftw: ok (0.4 sec). cfitsio: ok (0.5 sec). wcslib: ok (0.5 sec). minuit2: ok (0.5 sec). gsl: ok (0.4 sec). pyfits: ok (0.5 sec). python_d2to1: ok (0.4 sec). stsci_distutils: ok (0.5 sec). matplotlib: ok (0.5 sec). afwdata: ok (0.8 sec). obs_base: ok (1.1 sec). skypix: ok (0.7 sec). geom: ok (0.7 sec). meas_algorithms: ok (1.7 sec). esutil: ok (0.5 sec). meas_base: ok (1.5 sec). coadd_utils: ok (0.8 sec). pipe_base: ok (0.9 sec). obs_test: ok (0.9 sec). pipe_tasks: ok (1.5 sec). meas_astrom: ok (1.0 sec). astrometry_net: ok (0.4 sec). meas_deblender: ok (1.1 sec). astrometry_net_data: ok (0.4 sec). ip_isr: ok (0.9 sec). ip_diffim: ok (1.0 sec). lmfit: ok (0.4 sec). scipy: ok (0.4 sec). coadd_chisquared: ok (0.7 sec). skymap: ok (0.7 sec). healpy: ok (0.6 sec). pykg_config: ok (0.4 sec). jointcal_cholmod: ok (0.4 sec). testdata_jointcal: ok (120.2 sec). obs_lsstSim: ok (1.2 sec). python_mysqlclient: ok (0.5 sec). obs_cfht: ok (0.9 sec). testdata_cfht: ok (0.7 sec). obs_decam: ok (0.9 sec). testdata_decam: ok (0.6 sec). obs_subaru: ok (1.0 sec). meas_modelfit: ok (1.0 sec). shapelet: ok (0.8 sec). meas_extensions_psfex: ok (0.8 sec). psfex: ok (0.7 sec). meas_extensions_shapeHSM: ok (0.7 sec). galsim: ok (0.5 sec). tmv: ok (0.5 sec). testdata_subaru: ok (0.7 sec).
            Hide
            mtpatter Maria Patterson [X] (Inactive) added a comment -

            I am trying to rebuild for jointcal ticket DM-8552, with rebuild -r tickets/DM-8552 jointcal. Is there another dependency introduced for that branch John Parejko or should I rebuild with a different command?

            Show
            mtpatter Maria Patterson [X] (Inactive) added a comment - I am trying to rebuild for jointcal ticket DM-8552 , with rebuild -r tickets/ DM-8552 jointcal . Is there another dependency introduced for that branch John Parejko or should I rebuild with a different command?
            Hide
            tjenness Tim Jenness added a comment -

            It doesn't seem to me at first glance that the DM-8552 ticket branch has any changes to the table file.

            Show
            tjenness Tim Jenness added a comment - It doesn't seem to me at first glance that the DM-8552 ticket branch has any changes to the table file.
            Hide
            Parejkoj John Parejko added a comment - - edited

            Jointcal setupOptional depends on testdata_jointcal for unittests, which depends on obs_lsstSim. It's been that way for quite a while.

            Show
            Parejkoj John Parejko added a comment - - edited Jointcal setupOptional depends on testdata_jointcal for unittests, which depends on obs_lsstSim . It's been that way for quite a while.
            Hide
            tjenness Tim Jenness added a comment -

            Right, but obs_lsstSim is not Sims as such and doesn't depend on oorb.

            Show
            tjenness Tim Jenness added a comment - Right, but obs_lsstSim is not Sims as such and doesn't depend on oorb .
            Hide
            Parejkoj John Parejko added a comment -

            Ah. That's all I've got. Looking at my own stack, I don't even have oorb. I don't know where Maria Patterson [X] would be pulling it in from.

            Show
            Parejkoj John Parejko added a comment - Ah. That's all I've got. Looking at my own stack, I don't even have oorb. I don't know where Maria Patterson [X] would be pulling it in from.

              People

              • Assignee:
                frossie Frossie Economou
                Reporter:
                jhoblitt Joshua Hoblitt
                Watchers:
                Frossie Economou, John Parejko, John Swinbank, Jonathan Sick, Joshua Hoblitt, Tim Jenness
              • Votes:
                0 Vote for this issue
                Watchers:
                6 Start watching this issue

                Dates

                • Created:
                  Updated:

                  Summary Panel