Lines Matching +full:- +full:- +full:package

43         i_inout = (i <= n_in) ? i : i - n_in
44         dims = (i <= n_in) ? dims_in[i] : dims_out[i-n_in]
98                 "/gpu/cuda/gen is not compatible with user Q-functions defined with ",
121                     "User Q-functions with CUDA backends require the CUDA.jl package to be ",
123                     "Please ensure that the CUDA.jl package is installed and loaded.",
146     for a ∈ args[1:end-1]
154             ndim = length(a.args) - 3
194 Creates a user-defined interior (volumetric) Q-function, and assigns it to a variable named
195 `name`. The definition of the Q-function is given as:
207 The definitions of form `const=val` are used for definitions which will be compile-time
208 constants in the Q-function. For example, if `dim` is a variable set to the dimension of the
209 problem, then `dim=dim` will make `dim` available in the body of the Q-function as a
210 compile-time constant.
212 If the user wants to provide a context struct to the Q-function, that can be achieved by
214 struct, and `ctx` is the name to which is will be bound in the body of the Q-function.
220 consists of scalars (one number per Q-point) then `dims` should be omitted.
224 - Q-function to compute the "Q-data" for the mass operator, which is given by the quadrature
226   points) and the quadrature weights are given as input arrays, and the Q-data is the output
227   array. `dim` is given as a compile-time constant, and so the array `J` is statically
258             ndim = length(a.args) - 3