Re: [petsc-users] Error to TaoSolve() call

2014-08-12 Thread Jed Brown
Khai Hong Pham khaip...@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

Re: [petsc-users] Error to TaoSolve() call

2014-08-12 Thread Khai Hong Pham
Jed, Thank you for your advise. I'm solving multi-field problem which were solved by static-splitting schema. I tried to reused my code, that's why I tried to use the nested matrix and vector. I'm trying to use the contiguous vector and matrix storage now. I appreciate if you could give me any

[petsc-users] Error to TaoSolve() call

2014-08-11 Thread Khai Hong Pham
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

Re: [petsc-users] Error to TaoSolve() call

2014-08-11 Thread Jason Sarich
Hello Khai The likely culprit is in petsc/src/vec/vec/utils/projection.c, at the end of the VecMedian() there are three calls to VecRestore(), the last of which is wrong and needs to be corrected and recompiled: @@ -852,7 +852,7 @@ PetscErrorCode VecMedian(Vec Vec1, Vec Vec2, Vec Vec3, Vec

Re: [petsc-users] Error to TaoSolve() call

2014-08-11 Thread Khai Hong Pham
Hi Jason, Thank you for your help. I corrected ,recompiled the code and the error has gone. Khai On Mon, Aug 11, 2014 at 10:17 AM, Jason Sarich jason.sar...@gmail.com wrote: Hello Khai The likely culprit is in petsc/src/vec/vec/utils/projection.c, at the end of the VecMedian() there are