Menu
▾
▴
Re: [mesa-users] PGPSTAR, problem in pgstar_controls.dek during compilation with gfortran
|
From: Bill P. <pa...@ki...> - 2011-02-24 20:49:24
|
Hi, Damn compilers anyway! It seems that gfortran 4.5.1 (but not 4.6) has a nonstandard rule for deciding between local vs. external definitions of a symbol. Everyone else uses the local, but gfortran 4.5.1 uses the external one. In this case, we have a local definition of show_convective_section as a logical function (that's the one we want), but gfortran is instead using a definition of show_convective_section as a subroutine in a module that is being "use"d in pgstar_trho_profile. I should have caught this before the release. Sorry for missing it. There will be a new release very soon (I hope!). And it will be fixed in that. If you are feeling lucky and don't want to wait, you can update to 3039. I think it is okay, but it has been put through the entire test_suite. -Bill On Feb 24, 2011, at 11:44 AM, Mike Zingale wrote: > Hi Bill/Maxine, I am also having trouble compiling the latest MESA with pgstar. I am using gfortran 4.5.1 on Fedora 14, and I get past the pgstar_ctrls_io.f compilation, but die a bit later with: > > gfortran -fno-range-check -fopenmp -I../public -I../private -I../../include -Wunused-value -Werror -W -fimplicit-none -O2 -c -ffree-form ../private/pgstar_trho_profile.f > ../private/pgstar_trho_profile.f:698.49: > > has_convection = show_convective_section() > 1 > Error: Unexpected use of subroutine name 'show_convective_section' at (1) > ../private/pgstar_trho_profile.f:699.47: > > has_overshoot = show_overshoot_section() > 1 > Error: Unexpected use of subroutine name 'show_overshoot_section' at (1) > ../private/pgstar_trho_profile.f:700.57: > > has_semiconvection = show_semiconvective_section() > 1 > Error: Unexpected use of subroutine name 'show_semiconvective_section' at (1) > ../private/pgstar_trho_profile.f:701.55: > > has_thermo_haline = show_thermo_haline_section() > 1 > Error: Unexpected use of subroutine name 'show_thermo_haline_section' at (1) > ../private/pgstar_trho_profile.f:763.57: > > show_convective_section = show_mixing_section(convective_mixing, cl > 1 > Error: Unexpected use of subroutine name 'show_mixing_section' at (1) > ../private/pgstar_trho_profile.f:769.61: > > show_semiconvective_section = show_mixing_section(semiconvective_mi > 1 > Error: Unexpected use of subroutine name 'show_mixing_section' at (1) > ../private/pgstar_trho_profile.f:775.60: > > show_thermo_haline_section = show_mixing_section(thermo_haline_mixi > 1 > Error: Unexpected use of subroutine name 'show_mixing_section' at (1) > ../private/pgstar_trho_profile.f:781.56: > > show_overshoot_section = show_mixing_section(overshoot_mixing, clr_ > 1 > Error: Unexpected use of subroutine name 'show_mixing_section' at (1) > make: *** [pgstar_trho_profile.o] Error 1 > > /home/zingale/development/mesa/star/make > FAILED > > > > > > > > On Thu, 24 Feb 2011, Bill Paxton wrote: > >> Hi Maxime, >> >> I suggest upgrading to 4.5.1 -- you can do that easily at this website: >> >> http://gcc.gnu.org/wiki/GFortranBinaries >> >> Good luck! >> >> -Bill >> >> >> >> >> On Feb 24, 2011, at 10:29 AM, Maxime Viallet wrote: >> >>> >>> Hello Mesa-people, >>> >>> I have successfully compiled and run mesa (rev. 2987) using gfortran 4.5.0 on mac os X 10.6.6 >>> Now I am trying to get the PGPLOT interface working and I am running into this puzzling error while compiling the "star" module: >>> >>> gfortran -fno-range-check -fopenmp -I../public -I../private -I../../include -Wunused-value -Werror -W -fimplicit-none -O2 -c -ffree-form ../private/pgstar_ctrls_io.f >>> pgstar_controls.dek:5.74: >>> Included at ../private/pgstar_ctrls_io.f:33: >>> >>> type (pgstar_win_file_data) :: pgstar_win_file_ptr(num_pgstar_plots) >>> 1 >>> Error: Object 'pgstar_win_file_ptr' at (1) must have the SAVE attribute for default initialization of a component >>> make: *** [pgstar_ctrls_io.o] Error 1 >>> >>> /Users/viallet/postdoc/mesa/star/make >>> FAILED >>> >>> >>> Of course I naively tried to add by hand the "save" statement but then the file star/public/star_def.f would not compile anymore. >>> >>> Has somebody already encountered this problem ? >>> >>> Thanks ! >>> Maxime >>> ------------------------------------------------------------------------------ >>> Free Software Download: Index, Search & Analyze Logs and other IT data in >>> Real-Time with Splunk. Collect, index and harness all the fast moving IT data >>> generated by your applications, servers and devices whether physical, virtual >>> or in the cloud. Deliver compliance at lower cost and gain new business >>> insights. http://p.sf.net/sfu/splunk-dev2dev >>> _______________________________________________ >>> mesa-users mailing list >>> mes...@li... >>> https://lists.sourceforge.net/lists/listinfo/mesa-users >> >> >> ------------------------------------------------------------------------------ >> Free Software Download: Index, Search & Analyze Logs and other IT data in >> Real-Time with Splunk. Collect, index and harness all the fast moving IT data >> generated by your applications, servers and devices whether physical, virtual >> or in the cloud. Deliver compliance at lower cost and gain new business >> insights. http://p.sf.net/sfu/splunk-dev2dev >> _______________________________________________ >> mesa-users mailing list >> mes...@li... >> https://lists.sourceforge.net/lists/listinfo/mesa-users >> > > > ---------------------------------------------------------------------------- > Michael Zingale (mzi...@ma...) > Assistant Professor > > Dept. of Physics and Astronomy office: ESS 440 > Stony Brook University phone: 631-632-8225 > Stony Brook, NY 11794-3800 web: http://www.astro.sunysb.edu/mzingale > ---------------------------------------------------------------------------- > |
