document that NaN(2) and Inf(2) raise a fp exception
by default.

Reference: /n/sources/patch/applied/nanman
Date: Sun Dec 13 08:50:53 CET 2009
Signed-off-by: quanstro@quanstro.net

--- /sys/man/2/nan	Sun Dec 13 08:47:40 2009
+++ /sys/man/2/nan	Sun Dec 13 08:47:39 2009
@@ -25,7 +25,10 @@
 and division by zero.
 Also, the library functions sometimes return them when
 the arguments are not in the domain, or the result is
-out of range.
+out of range.  By default, these values
+will cause a floating point exception but
+.IR setfcr (2)
+can be use to change that behavior.
 .PP
 .I NaN
 returns a double that is not-a-number.
@@ -42,3 +45,6 @@
 with the same sign as the second argument.
 .SH SOURCE
 .B /sys/src/libc/port/nan.c
+.SH "SEE ALSO"
+.BR getfcr (2),
+.BR setfcr (2).