ab13dd segfaults for n=1 (single-state systems) #9
Description
Description
ab13dd crashes with segmentation fault when called with a single-state system (n=1).
Environment
- Platform: macOS (darwin arm64)
- Python: 3.14.2
- slicot version: (installed from this repo)
Minimal Reproducer
import numpy as np
from slicot import ab13dd
# Single-state system: G(s) = 1/(s+1)
n, m, p = 1, 1, 1
A = np.asfortranarray([[-1.0]])
E = np.asfortranarray([[1.0]])
B = np.asfortranarray([[1.0]])
C = np.asfortranarray([[1.0]])
D = np.asfortranarray([[0.0]])
fpeak_in = np.asfortranarray([0.0, 1.0])
# This crashes with segfault
result = ab13dd('C', 'I', 'N', 'Z', n, m, p, fpeak_in, A, E, B, C, D, 0.0)Expected Behavior
Should return (array([1., 1.]), array([0., 1.]), 0) (L-infinity norm of 1).
Actual Behavior
Segmentation fault in mb03xd called from ab13dd:
Fatal Python error: Segmentation fault
...
Binary file "slicot/_slicot.cpython-314-darwin.so", at mb03xd+0x528
Binary file "slicot/_slicot.cpython-314-darwin.so", at ab13dd+0x2b48
Workaround
Using n=2 or higher works correctly:
n, m, p = 2, 1, 1
A = np.asfortranarray([[-1.0, 0.0], [0.0, -2.0]])
E = np.asfortranarray([[1.0, 0.0], [0.0, 1.0]])
B = np.asfortranarray([[1.0], [1.0]])
C = np.asfortranarray([[1.0, 0.0]])
D = np.asfortranarray([[0.0]])
fpeak_in = np.asfortranarray([0.0, 1.0])
# This works
result = ab13dd('C', 'I', 'N', 'Z', n, m, p, fpeak_in, A, E, B, C, D, 0.0)Impact
This issue affects python-control's linfnorm() and system_norm() functions when computing L-infinity norms of first-order systems.