[petsc-users] Error to TaoSolve() call

[Jed Brown]

Khai Hong Pham <khaipham at utexas.edu> writes:

> Hello,
>
> I call TaosSolve() in my code. The solution vector is a nested and Jacobian
> matrix is nested too. It shows me the error:
>
> [0]PETSC ERROR:
> ------------------------------------------------------------------------
> [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation,
> probably memory access out of range
> [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger
> [0]PETSC ERROR: or see
> http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind[0]PETSC ERROR:
> or try http://valgrind.org on GNU/linux and Apple Mac OS X to find memory
> corruption errors
> [1]PETSC ERROR: [0]PETSC ERROR: likely location of problem given in stack
> below
> [0]PETSC ERROR: ---------------------  Stack Frames
> ------------------------------------
> [0]PETSC ERROR: Note: The EXACT line numbers in the stack are not available,
> [0]PETSC ERROR:       INSTEAD the line number of the start of the function
> [0]PETSC ERROR:       is given.
> [0]PETSC ERROR: [0] VecRestoreArray_Nest line 731
> /home/khp/bin/petsc-3.5.1/src/vec/vec/impls/nest/vecnest.c
> [0]PETSC ERROR: [0] VecRestoreArray line 1656
> /home/khp/bin/petsc-3.5.1/src/vec/vec/interface/rvector.c
> [0]PETSC ERROR: [0] VecMedian line 806
> /home/khp/bin/petsc-3.5.1/src/vec/vec/utils/projection.c
> [0]PETSC ERROR: [0] TaoSolve_SSILS line 52
> /home/khp/bin/petsc-3.5.1/src/tao/complementarity/impls/ssls/ssils.c
> [0]PETSC ERROR: [0] TaoSolve line 171
> /home/khp/bin/petsc-3.5.1/src/tao/interface/taosolver.c
>
> I debugged with valgrind and received the following message:
>
> ==19424== Invalid read of size 8
> ==19424==    at 0x54089E4: VecRestoreArray_Nest (vecnest.c:746)
> ==19424==    by 0x53F5F6F: VecRestoreArray (rvector.c:1666)
> ==19424==    by 0x54BE874: VecMedian (projection.c:855)

Can you try 'next'?  I think this bad restore has been fixed.

I *strongly* recommend *not* using VecNest.  99% of people that think
they should use it definitely should not use it.  Just use contiguous
storage for your vectors.  VecNest creates a lot of problems that are
almost always larger than the perceived concern that people try to avoid
by using VecNest.

> ==19424==    by 0x6247B74: TaoSolve_SSILS (ssils.c:59)
> ==19424==    by 0x620739E: TaoSolve (taosolver.c:188)
> ==19424==    by 0x40491C: main (main.cpp:265)
> ==19424==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
>
> It looks like the bug relates to the nested vector. Could you please give
> me a hint to locate the bug? Thanks.
>
> Khai
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 818 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20140812/47e548c5/attachment-0001.pgp>