What does `weights` argument do?

The `weights` argument of `mmrm()` is described as > an optional vector of weights to be used in the fitting process. Should be NULL or a numeric vector. I can't seem to find any further details in the documentation about how the weights are used in the model fit. Are they precision weights, where $$Var(e_{ij}) = \frac{\sigma_{ij}^2}{w_{ij}}$$ and with off-diagonal terms $$Cov(e_{hj},e_{ij}) = \frac{\sigma_{hij}}{\sqrt{w_{hj}w_{ij}}}$$ ? Or is the fit based on some sort of weighted likelihood?