p = 107
R = Integers(p**2)
Rpoly = PolynomialRing(R, 2, "z")
d = 64;
M0 = Matrix(R, [[R.random_element() for _ in range(d)] for _ in range(d)])
G = vector(Rpoly, [R.random_element() for _ in range(d)])
for k in range(p):
G = M0*G;
on Sage 7.5.1 it will run much slower than on previous versions, e.g., 7.3.
I suspect that the underlying library stop taking advantage of some flag.
Let me know if there is anything else I can do to help out to figure out this.
I also ran the same code on my desktop, with a 3.3GHz Intel Core i5-4590, and there the running times from Sage 6.9 to 7.5 (built or not built from source) are all around the 16min mark.
On a machine with a 2.3 GHz Intel Xeon E5 v3 with sage compiled from the source
If you run the following code
on Sage 7.5.1 it will run much slower than on previous versions, e.g., 7.3.
Sage 7.3:
Sage 7.5:
Here are the flags
I suspect that the underlying library stop taking advantage of some flag. Let me know if there is anything else I can do to help out to figure out this.
I also ran the same code on my desktop, with a 3.3GHz Intel Core i5-4590, and there the running times from Sage 6.9 to 7.5 (built or not built from source) are all around the 16min mark.
The CPU flags in that case are:
Component: build: configure
Keywords: speed, matrix
Issue created by migration from https://trac.sagemath.org/ticket/22330