Forum Message

 

 

We have moved the forum to https://forum.ngsolve.org . This is an archived version of the topics until 05/05/23. All the topics were moved to the new forum and conversations can be continued there. This forum is just kept as legacy to not invalidate old links. If you want to continue a conversation just look for the topic in the new forum.

Notice

The forum is in read only mode.

'SIMD: don't know how I shall evaluate in Assemble BilinearForm' error

More
3 years 8 months ago - 3 years 8 months ago #3042 by dong
I used the solution of a problem to be an input of another problem.

When I tried with an exact solution (a CoefficientFunction), the code worked well.

When I tried with an computed solution (a GridFunctioN), there is an error as follows.
Code:
NgException: GridFunctionCoefficientFunction: SIMD: don't know how I shall evaluatein Assemble BilinearForm 'biform_from_py'

The error was disappeared when I deleted line 248 (But I need this line in the discrete formulation.):
Code:
a += (u*n)*wbar.Trace()*cbar.Trace()*ds

So I guess the error maybe come from u.

Please see the attached file for details.

Can I convert a gridfunction to a similar form of a coefficient function? Could you please tell me how to fix that error?
Thank you so much.
Last edit: 3 years 8 months ago by dong.
More
3 years 8 months ago #3043 by mneunteufel
Hi dong,

u is a VectorL2 GridFunction, which you want to evaluate at the boundary.
L2 (and therefore VectorL2) has no well defined Trace (the L2 space returns a "DummyFE" which seems to cause an exception).

Further, in the definition of the space
Code:
V1 = VectorL2(mesh, order=order, dirichlet="gammaS|gammaD")
the dirichlet flag will be ignored (the number of dofs and number of free dofs coincide).

Best
Michael
The following user(s) said Thank You: dong
Time to create page: 0.159 seconds