Support the L modifier for floating-point values as an extension.

When L is omitted, double precision is used, so printf(1) gives
reproducable results.  When L is specified, long double precision is
used, which may improve precision, depending on the machine.

1 files changed, 10 insertions, 0 deletions

diff --git a/usr.bin/printf/printf.1 b/usr.bin/printf/printf.1
index 445c7aa..01360a4 100644
--- a/usr.bin/printf/printf.1
+++ b/usr.bin/printf/printf.1
@@ -197,6 +197,11 @@ A character which indicates the type of format to use (one of
 .Cm diouxXfFeEgGaAcsb ) .
 The uppercase formats differ from their lowercase counterparts only in
 that the output of the former is entirely in uppercase.
+The floating-point format specifiers
+.Cm ( fFeEgGaA )
+may be prefixed by an
+.Cm L
+to request that additional precision be used, if available.
 .El
 .Pp
 A field width or precision may be
@@ -326,6 +331,11 @@ Since the floating point numbers are translated from
 .Tn ASCII
 to floating-point and
 then back again, floating-point precision may be lost.
+(By default, the number is translated to an IEEE-754 double-precision
+value before being printed.
+The
+.Cm L
+modifier may produce additional precision, depending on the hardware platform.)
 .Pp
 .Tn ANSI
 hexadecimal character constants were deliberately not provided.