[DIRAC]Program received signal SIGSEGV: Segmentation fault - invalid memory reference

[fadili]

 Dear dirac-users:
please help to come across this erreur

________________________________________
OS:
Fedora17 Linux  3.8.4-102.fc17.i686
________________________________________
Erreur:
pam, stdout info: process ended with nonzero stderr stream - check

 **** dirac-executable stderr console output : ****

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:
#0  0xB75E4C1B
#1  0xB75E5294
#2  0xB77103FF
#3  0x80B3BF6 in __memory_tracker_MOD_track_deletefromlist at memory_tracker.F90:518
#4  0x80B3E10 in __memory_tracker_MOD_track_moveentry at memory_tracker.F90:553
#5  0x80B71D2 in __memory_tracker_MOD_track_registerdeallocation at memory_tracker.F90:132
#6  0x80B38C8 in __allocator_track_if_MOD_allocator_registerdeallocation at memory_track_if.F90:153
#7  0x80C64E2 in __allocator_internal_a_1d_r_MOD_dealloc_a_1d_r at memory_internal_ada_1d.F90:87
#8  0x890FA9D in readin_ at herrdn.F:409
#9  0x8934F43 in herinp_ at herdrv.F:992
#10  0x82CE0EF in paminp_ at dirrdn.F:175
#11  0x80CE252 in dirctl_ at dirac.F:102
#12  0x80B29CB in dirac12_program at main.F90:151

directory: /home/driss/Pers_Data/step_by_step/DIRAC-12.0/build
   inputs: H.mol  &  H.inp
  creating archive file  H_H.tgz
  archived working files: ['MOLECULE.MOL', 'DIRAC.INP']
  content of the (master) scratch directory
  ------------------------------------------------------------------------------
                     name    size (MB)    last accessed
  ------------------------------------------------------------------------------
                  dirac.x     62.535   05/02/2013 01:48:42 PM
                DIRAC.INP      0.000   05/02/2013 01:48:42 PM
             MOLECULE.MOL      0.000   05/02/2013 01:48:42 PM
                  MNF.INP      0.000   05/02/2013 01:48:42 PM
  ------------------------------------------------------------------------------
      Total size of all files :      62.535  MB
      Disk info:  used    available   capacity [GB]
                 27.725     28.594       56.319

  going to delete the scratch directory ... done

  exit date      : 2013-05-02 13:48:44.199255
  elapsed time   : 00h00m02s
  exit code      : -11
  exit           : ABNORMAL (CHECK DIRAC OUTPUT)

****************************************************************************+
OS:FEDORA Linux  3.8.4-102.fc17.i686

[Radovan Bast]

hi,

this is a known problem of gfortran.

for a solution (you need to edit 4 lines of code) see the 6th post in this thread:

https://groups.google.com/forum/?fromgroups=#!searchin/dirac-users/tracking/dirac-users/IC7oZ-8v7mU/HaAWF-AcwxUJ

good luck,

  radovan

--
You received this message because you are subscribed to the Google Groups "dirac-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dirac-users...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[fadili driss]

Hi,

I make all changes in the file "src/gp/memory_track_if.F90", and all is fine

Thank you

fadili

[Ademola Adetunji]

Hello Radovan and Fadil,

I am having a similar problem but I cannot locate "src/gp/memory_track_if.F90" in order to make the suggested corrections. Please, can you tell me the complete path to this file as my /usr/src  has only 2 empty folders. I have unsuccessfully tried "locate memory" and other ways to locate the said file. Your help will be greatly appreciated.

Below is the error message: 

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:

#0  0x3DF9019037
#1  0x3DF9019604
#2  0x3DE6A35C1F
#3  0x42F614 in phasetensor_
#4  0x423BA6 in MAIN__ at mtmap_v3.7.f:?
Segmentation fault (core dumped)

Thanks

[fadili driss]

hi Ademola,
The file is located in Dirac source files
after you extract the archive file DIRAC-12.6-Source.tar.gz:
you can find it in:
DIRAC-12.6-Source/src/gp/memory_track_if.F90

good luck

fadili

[Radovan Bast]

hi,

thanks Fadili.

we will make it easier in the next patch to deactivate these lines

(with a preprocessor statement).

best regards,

  radovan