From owner-freebsd-current  Sun Mar  5 17:35:17 1995
Return-Path: current-owner
Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id RAA15043 for current-outgoing; Sun, 5 Mar 1995 17:35:17 -0800
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id RAA15030 for <current@FreeBSD.org>; Sun, 5 Mar 1995 17:35:13 -0800
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id LAA09193; Mon, 6 Mar 1995 11:30:58 +1000
Date: Mon, 6 Mar 1995 11:30:58 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199503060130.LAA09193@godzilla.zeta.org.au>
To: gibbs@estienne.CS.Berkeley.EDU, nate@sneezy.sri.com
Subject: Re: Kernel build fails?
Cc: current@FreeBSD.org
Sender: current-owner@FreeBSD.org
Precedence: bulk

>I fixed these two by doing something like this:

>#ifdef  TCPOUTFLAGS
>/*
> * Flags used when sending segments in tcp_output.
> * Basic flags (TH_RST,TH_ACK,TH_SYN,TH_FIN) are totally
> * determined by state, with the proviso that TH_FIN is sent only
> * if all data queued for output is included in the segment.
> */
>u_char  tcp_outflags[TCP_NSTATES] = { 
>    TH_RST|TH_ACK, 0, TH_SYN, TH_SYN|TH_ACK,
>    TH_ACK, TH_ACK,
>    TH_FIN|TH_ACK, TH_ACK, TH_FIN|TH_ACK, TH_ACK, TH_ACK,
>}; 
>#else				<=== Added this section.
>extern u_char   tcp_outflags[];
>#endif

>And then removing TCPOUTFLAGS from tcp_subr.c.  If this is the correct
>approach, there are other places (for example in tcp_fsm.h) that need
>this treatment.

An initializer in a header file is usually an incorrect approach.  I like
the approach of #defining the initializer _list_ in a header file:

#define TCP_OUTFLAGS	TH_RST|TH_ACK, 0, TH_SYN, TH_SYN|TH_ACK, \
			TH_ACK, TH_ACK, \
			TH_FIN|TH_ACK, TH_ACK, TH_FIN|TH_ACK, TH_ACK, TH_ACK,

and using it exactly once, except I don't like extern data being accessed
from many modules (that's why I just made the data static to handle the
problem with ttydefchars in 1.1.5).

Bruce