1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
|
% ********************************************************************
% Copyright (C) 2023 Advanced Micro Devices, Inc. All rights reserved.
% ********************************************************************
function A = spdiag( d, idiag )
% A = spdiag( d, idiag )
%
% Generate a sparse matrix based on the diagonal
% This is similar to "diag" but returns
% a sparse matrix instead of a dense matrix
n = length(d);
m = n + abs(idiag);
if (idiag >= 0),
% ----------------------
% upper triangular part
% ----------------------
i1 = 1;
i2 = min(m,i1 + n - 1);
j1 = min(m,1 + idiag);
j2 = min(m,j1 + n -1 );
else
% -----------------------
% lower triangular part
% -----------------------
i1 = min(m,1 + abs(idiag));
i2 = min(m, i1 + n - 1);
j1 = 1;
j2 = min(m, j1 + n - 1);
end;
A = sparse( i1:i2, j1:j2, d, m,m );
|
