[petsc-users] Parallel assembly of matrix using OpenMP

[Clemens Domanig]

Hi everyone!

I'm writing a FEM-program and at the moment I try to parallize the 
assembly of the stiffness-matrix. The idea would be if I for example use 
4 threads I create 4 matrices and 4 rhs-vectors. Each thread fills its 
matrix and rhs-vector. At the end I add all matrices and vectors.
But the Petsc-manual says that Petsc is not thread-safe.
Is that the reason why it doesn't work?
Usually I get those error-code (after assembling 4-100 elements - 
everytime different):
I.)
[0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, 
probably memory access out of range

or

II.)
*** glibc detected *** ../icona: corrupted double-linked list: 
0x00000000166ff730 ***
======= Backtrace: =========
/lib64/libc.so.6[0x3a9fc72405]
/lib64/libc.so.6(cfree+0x4b)[0x3a9fc7276b]
../icona[0x426cae]
../icona[0x42df99]
../icona[0x41b829]
../icona[0x416203]
../icona[0x417341]
../icona[0x408954]
/lib64/libc.so.6(__libc_start_main+0xf4)[0x3a9fc1d994]
../icona(__gxx_personality_v0+0x3e9)[0x4078f9]
======= Memory map: ========
00400000-00fe1000 r-xp 00000000 00:1b 143920732 
  /home/c7031006/icona2.zip_FILES/icona2/icona
011e1000-011e8000 rw-p 00be1000 00:1b 143920732 
  /home/c7031006/icona2.zip_FILES/icona2/icona
011e8000-0131b000 rw-p 011e8000 00:00 0
16397000-16724000 rw-p 16397000 00:00 0 
  [heap]

If someone could explain - Thank you.
Yours sincerely, Clemens Domanig