The magnetic field looks a bit strange

[sm-liang]

I want apply the Bx = b0* tanh(y/delta), the harris-current configure, and there is the magnetic field code of ProblemGenerator
if (MAGNETIC_FIELDS_ENABLED) {
for (int k=ks; k<=ke; ++k) {
for (int j=js; j<=je; ++j) {
for (int i=is; i<=ie+1; ++i) {
y0 = pcoord->x2v(j);
pfield->b.x1f(k,j,i) = b0 * tanh(y0/width);
if (i>is){
bx0 = 0.5*(pfield->b.x1f(k,j,i)+pfield->b.x1f(k,j,i-1));
phydro->u(IEN,k,j,i-1) += (bx0*bx0)/2.0;
}
}
}
}
for (int k=ks; k<=ke; ++k) {
for (int j=js; j<=je+1; ++j) {
for (int i=is; i<=ie; ++i) {
pfield->b.x2f(k,j,i) = 0.0;
}
}
}
for (int k=ks; k<=ke+1; ++k) {
for (int j=js; j<=je; ++j) {
for (int i=is;i<=ie;++i) {
pfield->b.x3f(k,j,i) = 0.0;
}
}
}
}
where b0=1, but when I imshow the initial magnetic field, it looks like

there are slices of Bx, there are Bx[mid,:,:], Bx[:,mid,:], Bx[:,:,mid].
And I check the By and Bz, they are ok, both them are 0.
Please tell me where the problem is.
The attach file is my input file.
input.txt

[sm-liang]

This is the ProblemGenerator file.
turb_mhd_iso.txt

[c-white]

If you are compiling with --eos isothermal like your input file claims, then your problem generator has a bug. It's modifying phydro->u(IEN,...) when this variable does not exist, so it's overwriting some other part of memory. This part needs to be protected with if (NON_BAROTROPIC_EOS).

[sm-liang]

It works, Thanks, Chris.