Closed ftynse closed 9 years ago
This is what I expect to be the scattering after two stripmines, an it makes the code iterating over N points.
9 11 7 1 0 1
# e/i| c1 c2 c3 c4 c5 c6 c7 | i | N | 1
0 -1 0 0 0 0 0 0 0 0 0 ## c1 == 0
1 0 -4 0 0 0 1 0 0 0 0 ## -4*c2+c6 >= 0
1 0 4 0 0 0 -1 0 0 0 3 ## 4*c2-c6+3 >= 0
0 0 0 -1 0 0 0 0 0 0 0 ## c3 == 0
1 0 0 0 -3 0 1 0 0 0 0 ## -3*c4+c6 >= 0
1 0 0 0 3 0 -1 0 0 0 2 ## 3*c4-c6+2 >= 0
0 0 0 0 0 -1 0 0 0 0 0 ## c5 == 0
0 0 0 0 0 0 -1 0 1 0 0 ## c6 == i
0 0 0 0 0 0 0 -1 0 0 0 ## c7 == 0
Applying stripmine to this code as
stripmine([0],1,4,0)
works as expected and generates the code iterating over N points. Trying to apply the subsequentstripmine([0,0],2,3,0)
on the dimension created by the former stripmine results in the code iterating over more than N points.The resulting scattering
is missing one of the lines created by the former stripmine, but even introducing this line results in the wrong (even larger) number of points.