From: Michael Orlitzky <michael@orlitzky.com>
Date: Tue, 7 Apr 2026 18:53:06 +0000 (-0400)
Subject: tests/preconditioned_conjugate_gradient_method_tests.m: output diff
X-Git-Url: https://gitweb.michael.orlitzky.com/?a=commitdiff_plain;h=b9f404c5839d49b3d9934d10f3847bb3f2a7350e;p=octave.d.git

tests/preconditioned_conjugate_gradient_method_tests.m: output diff

Temporarily output the diff between the actual and expected results.
---

diff --git a/tests/preconditioned_conjugate_gradient_method_tests.m b/tests/preconditioned_conjugate_gradient_method_tests.m
index d0189de..7342ca8 100644
--- a/tests/preconditioned_conjugate_gradient_method_tests.m
+++ b/tests/preconditioned_conjugate_gradient_method_tests.m
@@ -89,7 +89,7 @@ for n = [ 5, 10, 25, 50, 100 ]
    						    max_iterations);
 
   diff = norm(o_x - x, 'inf');
-  msg = sprintf("Our PCGM agrees with Octave's, n=%d.", n);
+  msg = sprintf("Our PCGM agrees with Octave's, n=%d, diff=%.12f", n, diff);
   ## There's no good way to choose the tolerance here, since each
   ## individual algorithm terminates based on the (2,infinity)-norm of
   ## the gradient. So we use two orders of magnitude.