[mesa-users] MESA problems - Segmentation fault when running rlo with ifort

Bill Paxton paxton at kitp.ucsb.edu
Tue Jul 23 13:55:36 EDT 2013


On Jul 23, 2013, at 10:47 AM, Michael Zingale wrote:

> Bill, one magic flag that I've always had to use with ifort is "-mp", this tells the compiler not to do any optimizations that change the answer.  You might want to see if using this + -O2 makes it happy.  Generally, we've found ifort to be flaky.
> 
> Mike
> 

Hi Mike,

Gotta love those magic flags!  
I'm looking for the one that enables intelligence, but so far it eludes me.

There are good arguments for making mesa work on as many compilers
as possible (Frank preaches this sermon to me regularly, but without effect).
But ifort is one windmill I'm not willing to take on. 

If gfortran would just do a backtrace of the call stack after a crash I'd be happy.

Bill










> 
> On Tue, Jul 23, 2013 at 1:43 PM, Bill Paxton <paxton at kitp.ucsb.edu> wrote:
> Hi Jonas,
> 
> We have also had problems with the new ifort.  
> In section B.11 of the new mesa instrument paper we report comparisons of results using ifort and gfortran on mac and linux.  In order to get the tests to run at all with ifort we had to drop down to optimization -O1.   Not a good sign.   So I'd suggest either using gfortran with -O2 or ifort with -O1.
> 
> -Bill
> 
> 
> 
> 
> 
> 
> On Jul 23, 2013, at 10:21 AM, Jonas Goliasch wrote:
> 
>> Hello Bill,
>> I am in the group of Lorne Nelson, and I am having trouble running the MESA rlo module with mass loss. I am using version 4942 but a similar problem seems to also occur with the latest version (5118).
>> The mesa module is installed on a computing cluster using the most recent version ifort
>> $ ifort –version
>> ifort (IFORT) 12.1.3 20120212
>> Copyright (C) 1985-2012 Intel Corporation.  All rights reserved.
>>  
>> And I can run the standard ‘work’ module and most of the other modules I tested. And even with the rlo module I can produce a proper single star track (by setting a very large separation). However, a segmentation fault occurs whenever the star starts losing mass due to roche lobe overflow (please see attached segmentation fault output).
>> I have traced the occurrence of the crash to subroutine set_xa in adjust_mass.f (inside the ‘private’ folder of ‘star’). More specifically, it occurs in the first few lines of that routine when s% xa(j,k) is being updated (see the line ‘s% xa(j,k) = xa_old(j,k)
>> ‘ below)
>>  
>>          if (dbg_adjm) &
>>             write(*,2) 'set_xa: k_const_mass', k_const_mass
>>          if (k_const_mass < nz) then
>>             ! for k >= k_const_mass have m_new(k) = m_old(k),
>>             ! so no change in xa_new(:,k) for k > k_const_mass
>>             do k=k_const_mass+1,nz
>>                do j=1,species              
>>                   s% xa(j,k) = xa_old(j,k)
>>                end do
>>             end do
>>          end if
>> 
>>  
>> By placing a dummy statement inside this loop before the reassignment of s% xa(j,k), I can circumvent the segmentation fault which allows the code to run several hundred models further. But it eventually also gets itself into trouble (see attached output). So far I have not been able to get around this issue, which seems to be related to me using the ifort compiler. As mentioned above, I can successfully produce a complete single star evolutionary track using the rlo module.
>> Any suggestions or help to resolve this issue would be greatly appreciated. For completeness I’ve also attached the output when running the latest version of MESA. Note, I have also attached a pdf version of the code output in case the formatting of the e-mail becomes distorted.
>> 
>> Kind Regards,
>> 
>> Jonas Goliasch
>>  
>> 
>> 
>> 
>> Segmentation fault output:
>> [goliasch at ip10 make_mdot_work]$ ./rn
>> svn: '.' is not a working copy
>> DATE: 2013-07-23
>> TIME: 10:10:10
>>  saved initial &star_job inlist values: star_job_namelist.out
>>                                          version_number        4942
>>  read inlist_test_rlo
>>  
>>                                               maccretor    1.0000000000000000D-02
>>                                                  mdonor    1.0000000000000000D+00
>>                                  initial_period_in_days    1.3999999999999999D+00
>>                                         jdot_multiplier    1.0000000000000000D+00
>>                                                      fr    1.0000000000000001D-01
>>  gr_only F
>>  
>>  net name basic.net
>>                       use small_mtx_decsol bcyclic_dble           8         100
>>  kappa_file_prefix gn93
>>  kappa_lowT_prefix lowT_fa05_gs98
>>    eos_file_prefix mesa
>>                                         OMP_NUM_THREADS           8
>> _________________________________________________________________________________________________________________________________________________
>>  
>>        step    lg_Tcntr    Teff       lg_LH     lg_Lnuc     Mass       H_rich     H_cntr     N_cntr     Y_surf     X_avg     eta_cntr   pts  retry
>>    lg_dt_yr    lg_Dcntr    lg_R       lg_L3a    lg_Lneu     lg_Mdot    H_poor     He_cntr    O_cntr     Z_surf     Y_avg     gam_cntr  iters bckup
>>         age    lg_Pcntr    lg_L       lg_LZ     lg_Psurf    lg_Dsurf   He_poor    C_cntr     Ne_cntr    Z_cntr     Z_avg     v_div_cs     dt_limit
>> __________________________________________________________________________________________________________________________________________________
>>          10   7.136146   5616.538  -0.139799  -0.139799   1.000000   1.000000   0.699287   0.001903   0.280000   0.699880  -1.769666    823      0
>>    5.631147   1.888831  -0.052838 -47.216113  -1.678073 -99.000000   0.000000   0.280585   0.009363   0.020000   0.280109   0.078966      2      0
>>  2.1894E+06  17.154144  -0.154319 -15.954590   4.847007  -6.702862   0.000000   0.002675   0.002100  2.013E-02  2.001E-02 -0.742E-11        lgTeff
>>>>          60   7.133184   5622.820  -0.147634  -0.147634   1.000000   1.000000   0.697882   0.003809   0.280000   0.699705  -1.727568    804      0
>>    4.995777   1.901966  -0.049065 -47.342103  -1.745142 -99.000000   0.000000   0.281718   0.009362   0.020000   0.280269   0.080374      2      0
>>  2.3787E+07  17.163974  -0.144832 -15.653560   4.841842  -6.708605   0.000000   0.001042   0.002100  2.040E-02  2.003E-02  0.415E-12        max_dt
>>  
>>          70   7.133155   5623.302  -0.147276  -0.147276   1.000000   1.000000   0.697830   0.003862   0.280000   0.699699  -1.726333    804      0
>>    4.767965   1.902440  -0.049013 -47.342361  -1.746117 -99.000000   0.000000   0.281762   0.009362   0.020000   0.280275   0.080410      2      0
>>  2.4538E+07  17.164406  -0.144578 -99.000000   4.841707  -6.708778   0.000000   0.000996   0.002100  2.041E-02  2.003E-02  0.406E-12        max_dt
>>  
>> limit_retention_by_mdot_edd   1.00000000000000    
>>  
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> Image              PC                Routine            Line        Source            
>> star_rlo_implicit  000000000055F236  adjust_mass_mp_do         247  adjust_mass.f
>> star_rlo_implicit  0000000000502953  evolve_mp_do_evol         233  evolve.f
>> star_rlo_implicit  0000000000421743  star_lib_mp_star_         370  star_lib.f
>> star_rlo_implicit  000000000042ECE0  run_star_support_        2275  run_star_support.f
>> star_rlo_implicit  0000000000421368  run_star_mp_do_ru          29  run_star.f
>> star_rlo_implicit  0000000000421462  MAIN__                     43  run.f
>> star_rlo_implicit  000000000040412C  Unknown               Unknown  Unknown
>> libc.so.6          000000326001D994  Unknown               Unknown  Unknown
>> star_rlo_implicit  0000000000403FB9  Unknown               Unknown  Unknown
>> DATE: 2013-07-23
>> TIME: 10:10:19
>> [goliasch at ip10 make_mdot_work]$
>> 
>>  
>> Output when circumventing segmentation fault:
>>  
>> [goliasch at ip10 make_mdot_work]$ ./rn
>> svn: '.' is not a working copy
>> DATE: 2013-07-23
>> TIME: 10:23:22
>>  saved initial &star_job inlist values: star_job_namelist.out
>>                                          version_number        4942
>>  read inlist_test_rlo
>>                                                maccretor    1.0000000000000000D-02
>>                                                  mdonor    1.0000000000000000D+00
>>                                  initial_period_in_days    1.3999999999999999D+00
>>                                         jdot_multiplier    1.0000000000000000D+00
>>                                                      fr    1.0000000000000001D-01
>>  gr_only F
>>   net name basic.net
>>                       use small_mtx_decsol bcyclic_dble           8         100
>>  kappa_file_prefix gn93
>>  kappa_lowT_prefix lowT_fa05_gs98
>>    eos_file_prefix mesa
>>                                         OMP_NUM_THREADS           8
>>   evolve_loop i=           1
>> __________________________________________________________________________________________________________________________________________________
>>      step    lg_Tcntr    Teff       lg_LH     lg_Lnuc     Mass       H_rich     H_cntr     N_cntr     Y_surf     X_avg     eta_cntr   pts  retry
>>    lg_dt_yr    lg_Dcntr    lg_R       lg_L3a    lg_Lneu     lg_Mdot    H_poor     He_cntr    O_cntr     Z_surf     Y_avg     gam_cntr  iters bckup
>>         age    lg_Pcntr    lg_L       lg_LZ     lg_Psurf    lg_Dsurf   He_poor    C_cntr     Ne_cntr    Z_cntr     Z_avg     v_div_cs     dt_limit
>>          10   7.136146   5616.538  -0.139799  -0.139799   1.000000   1.000000   0.699287   0.001903   0.280000   0.699880  -1.769666    823      0
>>    5.631147   1.888831  -0.052838 -47.216113  -1.678073 -99.000000   0.000000   0.280585   0.009363   0.020000   0.280109   0.078966      2      0
>>  2.1894E+06  17.154144  -0.154319 -15.954590   4.847007  -6.702862   0.000000   0.002675   0.002100  2.013E-02  2.001E-02 -0.742E-11        lgTeff
>>>>          70   7.133155   5623.302  -0.147276  -0.147276   1.000000   1.000000   0.697830   0.003862   0.280000   0.699699  -1.726333    804      0
>>    4.767965   1.902440  -0.049013 -47.342361  -1.746117 -99.000000   0.000000   0.281762   0.009362   0.020000   0.280275   0.080410      2      0
>>  2.4538E+07  17.164406  -0.144578 -99.000000   4.841707  -6.708778   0.000000   0.000996   0.002100  2.041E-02  2.003E-02  0.406E-12        max_dt
>>>> limit_retention_by_mdot_edd   1.00000000000000  
>>          80   7.133147   5623.581  -0.147068  -0.147068   1.000000   1.000000   0.697799   0.003893   0.280000   0.699695  -1.725684    805      0
>>    4.540227   1.902700  -0.048982 -47.342139  -1.746666 -99.000000   0.000000   0.281789   0.009362   0.020000   0.280278   0.080429      2      0
>>  2.4982E+07  17.164649  -0.144430 -99.000000   4.841628  -6.708879   0.000000   0.000970   0.002100  2.041E-02  2.003E-02  0.410E-12        max_dt
>>>>         200   7.133133   5624.563  -0.146889  -0.146889   0.999998   0.999998   0.697756   0.003936   0.280000   0.699690  -1.724669    804      0
>>    1.704756   1.903102  -0.048986 -47.342260  -1.747449  -7.907379   0.000000   0.281826   0.009362   0.020000   0.280283   0.080458      2      0
>>  2.5609E+07  17.165027  -0.144135 -15.954590   4.842993  -6.709504   0.000000   0.000933   0.002100  2.042E-02  2.003E-02  0.348E-06        max_dt
>>>>         400   7.132610   5623.574  -0.153945  -0.153945   0.997556   0.997556   0.697754   0.003937   0.280000   0.699689  -1.724631    798      0
>>    1.322512   1.902343  -0.051336 -47.370921  -1.752430  -7.000000   0.000000   0.281828   0.009362   0.020000   0.280284   0.080508      2      0
>>  2.5634E+07  17.163740  -0.149140 -99.000000   4.845181  -6.707216   0.000000   0.000932   0.002100  2.042E-02  2.003E-02  0.116E-05        max_dt
>>>>         450   7.132586   5623.382  -0.154267  -0.154267   0.997446   0.997446   0.697754   0.003937   0.280000   0.699689  -1.724626    799      0
>>    1.380211   1.902309  -0.051457 -47.372230  -1.752657  -7.000000   0.000000   0.281828   0.009362   0.020000   0.280284   0.080511      2      0
>>  2.5635E+07  17.163683  -0.149442 -15.954590   4.845320  -6.707062   0.000000   0.000932   0.002100  2.042E-02  2.003E-02  0.105E-05        max_dt
>>  
>>                                  bad angular_momentum_j   -1.6607253567366117D+48
>>         450   7.132587   5623.372  -0.154261  -0.154261   0.997449   0.997449   0.697754   0.003937   0.280000   0.699689  -1.724625    799      1
>>    1.229696   1.902311  -0.051455 -47.372212  -1.752653 -99.000000   0.000000   0.281828   0.009362   0.020000   0.280284   0.080511      2      0
>>  2.5635E+07  17.163687  -0.149440 -99.000000   4.845320  -6.707061   0.000000   0.000932   0.002100  2.042E-02  2.003E-02 -0.186E-09    varcontrol
>>  
>>  limit_retention_by_mdot_edd  3.214841505531817E-003
>>  
>>                                  bad angular_momentum_j   -3.2131373377825206D+51
>>                                  bad angular_momentum_j   -2.2718819925086352D+51
>>                                             num_backups          11
>>                                      max_number_backups          10
>>  
>>  stopping because of convergence problems num_backups > max_number_backups
>>  
>> save LOGS/profile3.data for model 457
>> save photos/x457 for model 457
>>  
>>                   runtime, retries, backups, steps          1.686900          2.000000        12        11       457
>>         457   7.132585   5623.375  -0.154261  -0.154274   0.997449   0.997449   0.697754   0.003937   0.280000   0.699689  0.146E+18    799     12
>>   -1.851937   1.902312  -0.051455 -47.372218  -1.752654 -99.000000   0.000000   0.281828   0.009362   0.020000   0.280284  0.855E+14      4     11
>>  2.5635E+07  17.163682  -0.149439 -99.000000   4.845320   0.000000   0.000000   0.000932   0.002100  2.042E-02  2.003E-02  0.000E+00          hold
>>  
>> DATE: 2013-07-23
>> TIME: 10:28:23
>> [goliasch at ip10 make_mdot_work]$
>> 
>>  
>> Output of segmentation fault when running the latest version of MESA:
>>  
>>  
>> [goliasch at ip10 binary_donor_only]$ ./rn
>> DATE: 2013-07-23
>> TIME: 10:35:15
>>  
>>  
>>                                          version_number        5118
>>  donor_controls
>>  
>>                                               maccretor    8.0000000000000004D-01
>>                                                  mdonor    1.0000000000000000D+00
>>                                  initial_period_in_days    5.0000000000000000D-01
>>                                         jdot_multiplier    1.0000000000000000D+00
>>                                                      fr    1.0000000000000001D-01
>>  gr_only F
>>  
>>  set_net basic.net
>>  net name basic.net
>>  extra_terminal_output_file: log1
>>                       use small_mtx_decsol bcyclic_dble           8         100
>>  kappa_file_prefix gn93
>>  kappa_lowT_prefix lowT_fa05_gs98
>>    eos_file_prefix mesa
>>                                         OMP_NUM_THREADS           8
>>  donor_startup
>>  
>>  
>>                 star               model                 age                mass           (r-rl)/rl          last photo
>>                    1                   0     0.000000000E+00     1.000000000E+00            Infinity       
>> __________________________________________________________________________________________________________________________________________________
>>  
>>        step    lg_Tcntr    Teff       lg_LH     lg_Lnuc     Mass       H_rich     H_cntr     N_cntr     Y_surf     X_avg     eta_cntr   pts  retry
>>    lg_dt_yr    lg_Dcntr    lg_R       lg_L3a    lg_Lneu     lg_Mdot    H_poor     He_cntr    O_cntr     Z_surf     Y_avg     gam_cntr  iters bckup
>>         age    lg_Pcntr    lg_L       lg_LZ     lg_Psurf    lg_Dsurf   He_poor    C_cntr     Ne_cntr    Z_cntr     Z_avg     v_div_cs     dt_limit
>> __________________________________________________________________________________________________________________________________________________
>>  
>> forrtl: severe (408): fort: (7): Attempt to use pointer S_ACCRETOR when it is not associated with a target
>>  
>> Image              PC                Routine            Line        Source            
>> star_binary_donor  0000000000B50F5A  Unknown               Unknown  Unknown
>> star_binary_donor  0000000000B4FA56  Unknown               Unknown  Unknown
>> star_binary_donor  0000000000AF7060  Unknown               Unknown  Unknown
>> star_binary_donor  0000000000A9524F  Unknown               Unknown  Unknown
>> star_binary_donor  0000000000A95771  Unknown               Unknown  Unknown
>> star_binary_donor  000000000042BD5B  run_binarydo_run_         477  run_binary.f
>> star_binary_donor  0000000000429291  run_binary_mp_do_         356  run_binary.f
>> star_binary_donor  000000000043D62E  MAIN__                      4  run.f
>> star_binary_donor  000000000040418C  Unknown               Unknown  Unknown
>> libc.so.6          000000326001D994  Unknown               Unknown  Unknown
>> star_binary_donor  0000000000404019  Unknown               Unknown  Unknown
>> DATE: 2013-07-23
>> TIME: 10:35:22
>> [goliasch at ip10 binary_donor_only]$
>> <code_output.pdf>
> 
> 
> ------------------------------------------------------------------------------
> See everything from the browser to the database with AppDynamics
> Get end-to-end visibility with application monitoring from AppDynamics
> Isolate bottlenecks and diagnose root cause in seconds.
> Start your free trial of AppDynamics Pro today!
> http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk
> _______________________________________________
> mesa-users mailing list
> mesa-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mesa-users
> 
> 
> 
> 
> -- 
> Michael Zingale
> Associate Professor
> 
> Dept. of Physics & Astronomy • Stony Brook University • Stony Brook, NY 11794-3800
> phone:  631-632-8225
> e-mail: Michael.Zingale at stonybrook.edu
> web: http://www.astro.sunysb.edu/mzingale

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20130723/c6509c2a/attachment.html>


More information about the Mesa-users mailing list