From owner-freebsd-bugs@FreeBSD.ORG  Sat Apr 10 16:48:42 2004
Return-Path: <owner-freebsd-bugs@FreeBSD.ORG>
Delivered-To: freebsd-bugs@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id D8A1C16A4CE
	for <freebsd-bugs@freebsd.org>; Sat, 10 Apr 2004 16:48:42 -0700 (PDT)
Received: from mail.iinet.net.au (mail-04.iinet.net.au [203.59.3.36])
	by mx1.FreeBSD.org (Postfix) with SMTP id E50B743D39
	for <freebsd-bugs@freebsd.org>; Sat, 10 Apr 2004 16:48:40 -0700 (PDT)
	(envelope-from darrenr@reed.wattle.id.au)
Received: (qmail 11555 invoked from network); 10 Apr 2004 23:45:45 -0000
Received: from unknown (HELO firewall.reed.wattle.id.au) (203.217.34.157)
  by mail.iinet.net.au with SMTP; 10 Apr 2004 23:45:45 -0000
Received: (from root@localhost)
	by firewall.reed.wattle.id.au (8.11.0/8.11.0) id i3ANjoe05555;
	Sun, 11 Apr 2004 09:45:50 +1000 (EST)
From: Darren Reed <darrenr@reed.wattle.id.au>
Message-Id: <200404102345.JAA02555@avalon.reed.wattle.id.au>
To: andre@freebsd.org, vys@renet.ru
Date: Sun, 11 Apr 2004 09:45:23 +1000 (EST)
X-Mailer: ELM [version 2.4ME+ PL99d (25)]
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=US-ASCII
cc: freebsd-bugs@freebsd.org
Subject: kern/56441 - bpf_tap() used incorrectly in bpf
X-BeenThere: freebsd-bugs@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: Bug reports <freebsd-bugs.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-bugs>,
	<mailto:freebsd-bugs-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-bugs>
List-Post: <mailto:freebsd-bugs@freebsd.org>
List-Help: <mailto:freebsd-bugs-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-bugs>,
	<mailto:freebsd-bugs-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Sat, 10 Apr 2004 23:48:43 -0000

Andre,

The patch below should resolve this PR.  It modifies the behaviour of
bpf_mtap() in a manner that resembles bpf_tap(), providing the speed
"boost", whilst preserving the application of "seesent".

Darren

Index: bpf.c
===================================================================
RCS file: /home/ncvs/src/sys/net/bpf.c,v
retrieving revision 1.125
diff -c -r1.125 bpf.c
*** bpf.c	7 Apr 2004 20:46:11 -0000	1.125
--- bpf.c	10 Apr 2004 23:38:02 -0000
***************
*** 1197,1209 ****
  	struct bpf_if *bp;
  	struct mbuf *m;
  {
  	struct bpf_d *d;
! 	u_int pktlen, slen;
  
  	pktlen = m_length(m, NULL);
  	if (pktlen == m->m_len) {
! 		bpf_tap(bp, mtod(m, u_char *), pktlen);
! 		return;
  	}
  
  	BPFIF_LOCK(bp);
--- 1197,1216 ----
  	struct bpf_if *bp;
  	struct mbuf *m;
  {
+ 	void *(*cpfn) __P((void *, const void *, size_t));
+ 	u_int pktlen, slen, buflen;
  	struct bpf_d *d;
! 	void *marg;
  
  	pktlen = m_length(m, NULL);
  	if (pktlen == m->m_len) {
! 		cpfn = bcopy;
! 		marg = mtod(m, void *);
! 		buflen = pktlen;
! 	} else {
! 		cpfn = bpf_mcopy;
! 		marg = m;
! 		buflen = 0;
  	}
  
  	BPFIF_LOCK(bp);
***************
*** 1212,1224 ****
  			continue;
  		BPFD_LOCK(d);
  		++d->bd_rcount;
! 		slen = bpf_filter(d->bd_filter, (u_char *)m, pktlen, 0);
  		if (slen != 0)
  #ifdef MAC
  			if (mac_check_bpfdesc_receive(d, bp->bif_ifp) == 0)
  #endif
! 				catchpacket(d, (u_char *)m, pktlen, slen,
! 				    bpf_mcopy);
  		BPFD_UNLOCK(d);
  	}
  	BPFIF_UNLOCK(bp);
--- 1219,1230 ----
  			continue;
  		BPFD_LOCK(d);
  		++d->bd_rcount;
! 		slen = bpf_filter(d->bd_filter, marg, pktlen, buflen);
  		if (slen != 0)
  #ifdef MAC
  			if (mac_check_bpfdesc_receive(d, bp->bif_ifp) == 0)
  #endif
! 				catchpacket(d, (u_char *)m, pktlen, slen, cpfn);
  		BPFD_UNLOCK(d);
  	}
  	BPFIF_UNLOCK(bp);