Fixing wording in comment. (GH-10425)
Since the n==1 case just returns *max*, it cannot be larger
than the magnitude of the vector entry.
diff --git a/Modules/mathmodule.c b/Modules/mathmodule.c
index e956314..b83befb 100644
--- a/Modules/mathmodule.c
+++ b/Modules/mathmodule.c
@@ -2043,7 +2043,7 @@
max * sqrt(sum((x / max) ** 2 for x in vec))
The value of the *max* variable must be non-negative and
-at least equal to the absolute value of the largest magnitude
+equal to the absolute value of the largest magnitude
entry in the vector. If n==0, then *max* should be 0.0.
If an infinity is present in the vec, *max* should be INF.