Mark Kettenis <[email protected]> writes: >> From: [email protected] (=?utf-8?Q?J=C3=A9r=C3=A9mie_Courr=C3=A8ges-Anglas?=) >> Date: Mon, 11 Nov 2013 16:31:08 +0100 >> >> Today I heard someone struggling while trying to understand the meaning >> and possible uses of the -d ping(8) flag. Looking at the code it seems >> that SO_DEBUG has no effect on a raw socket (right?), so I thought that >> it would be better to document that. >> >> ok? > > Makes no sense. The man page still says that SO_DEBUG gets set, but > you remove the code that actually does this. You should either change > the man page, and leave the code alone, or you should change the code > and change the manual page to state that -d is accepted for backwards > compatibility only or remove it from the documentation altogether.
Right. The latter would give: Index: ping.8 =================================================================== RCS file: /cvs/src/sbin/ping/ping.8,v retrieving revision 1.48 diff -u -p -r1.48 ping.8 --- ping.8 14 Aug 2013 08:39:30 -0000 1.48 +++ ping.8 11 Nov 2013 16:18:13 -0000 @@ -83,9 +83,7 @@ Set the .Dv Don't Fragment bit. .It Fl d -Set the -.Dv SO_DEBUG -option on the socket being used. +This option has no effect and is kept for compatibility. .It Fl E Emit an audible beep (by sending an ASCII BEL character to the standard error output) when no packet is received before the next > I'd leave the code alone. I can't see why this should be preferred. -- jca | PGP : 0x06A11494 / 61DB D9A0 00A4 67CF 2A90 8961 6191 8FBF 06A1 1494
