From owner-svn-src-all@FreeBSD.ORG Wed Apr 1 22:26:47 2015 Return-Path: Delivered-To: svn-src-all@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 499C4AEC; Wed, 1 Apr 2015 22:26:47 +0000 (UTC) Received: from svn.freebsd.org (svn.freebsd.org [IPv6:2001:1900:2254:2068::e6a:0]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 32595388; Wed, 1 Apr 2015 22:26:47 +0000 (UTC) Received: from svn.freebsd.org ([127.0.1.70]) by svn.freebsd.org (8.14.9/8.14.9) with ESMTP id t31MQlmP044475; Wed, 1 Apr 2015 22:26:47 GMT (envelope-from glebius@FreeBSD.org) Received: (from glebius@localhost) by svn.freebsd.org (8.14.9/8.14.9/Submit) id t31MQedN044443; Wed, 1 Apr 2015 22:26:40 GMT (envelope-from glebius@FreeBSD.org) Message-Id: <201504012226.t31MQedN044443@svn.freebsd.org> X-Authentication-Warning: svn.freebsd.org: glebius set sender to glebius@FreeBSD.org using -f From: Gleb Smirnoff Date: Wed, 1 Apr 2015 22:26:40 +0000 (UTC) To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r280971 - in head: contrib/ipfilter/tools share/man/man4 sys/contrib/ipfilter/netinet sys/netinet sys/netipsec sys/netpfil/pf X-SVN-Group: head MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-BeenThere: svn-src-all@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: "SVN commit messages for the entire src tree \(except for " user" and " projects" \)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 01 Apr 2015 22:26:47 -0000 Author: glebius Date: Wed Apr 1 22:26:39 2015 New Revision: 280971 URL: https://svnweb.freebsd.org/changeset/base/280971 Log: o Use new function ip_fillid() in all places throughout the kernel, where we want to create a new IP datagram. o Add support for RFC6864, which allows to set IP ID for atomic IP datagrams to any value, to improve performance. The behaviour is controlled by net.inet.ip.rfc6864 sysctl knob, which is enabled by default. o In case if we generate IP ID, use counter(9) to improve performance. o Gather all code related to IP ID into ip_id.c. Differential Revision: https://reviews.freebsd.org/D2177 Reviewed by: adrian, cy, rpaulo Tested by: Emeric POUPON Sponsored by: Netflix Sponsored by: Nginx, Inc. Relnotes: yes Modified: head/contrib/ipfilter/tools/ipftest.c head/share/man/man4/inet.4 head/sys/contrib/ipfilter/netinet/fil.c head/sys/contrib/ipfilter/netinet/ip_fil.h head/sys/contrib/ipfilter/netinet/ip_fil_freebsd.c head/sys/contrib/ipfilter/netinet/ip_nat.c head/sys/netinet/ip_carp.c head/sys/netinet/ip_gre.c head/sys/netinet/ip_id.c head/sys/netinet/ip_input.c head/sys/netinet/ip_mroute.c head/sys/netinet/ip_output.c head/sys/netinet/ip_var.h head/sys/netinet/raw_ip.c head/sys/netinet/sctp_output.c head/sys/netipsec/xform_ipip.c head/sys/netpfil/pf/if_pfsync.c head/sys/netpfil/pf/pf_norm.c Modified: head/contrib/ipfilter/tools/ipftest.c ============================================================================== --- head/contrib/ipfilter/tools/ipftest.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/contrib/ipfilter/tools/ipftest.c Wed Apr 1 22:26:39 2015 (r280971) @@ -864,3 +864,11 @@ void fixv4sums(m, ip) *(u_short *)csump = fr_cksum(&tmp, ip, p, hdr); } } + +void +ip_fillid(struct ip *ip) +{ + static uint16_t ip_id; + + ip->ip_id = ip_id++; +} Modified: head/share/man/man4/inet.4 ============================================================================== --- head/share/man/man4/inet.4 Wed Apr 1 21:48:54 2015 (r280970) +++ head/share/man/man4/inet.4 Wed Apr 1 22:26:39 2015 (r280971) @@ -28,7 +28,7 @@ .\" From: @(#)inet.4 8.1 (Berkeley) 6/5/93 .\" $FreeBSD$ .\" -.Dd January 26, 2012 +.Dd April 2, 2015 .Dt INET 4 .Os .Sh NAME @@ -226,12 +226,24 @@ This .Xr sysctl 8 variable affects packets destined for a local host as well as packets forwarded to some other host. +.It Va ip.rfc6864 +Boolean: control IP IDs generation behaviour. +True value enables RFC6864 support, which specifies that IP ID field of +.Em atomic +datagrams can be set to any value. +The +.Fx implementation sets it to zero. +Enabled by default. .It Va ip.random_id Boolean: control IP IDs generation behaviour. Setting this .Xr sysctl 8 -to non-zero causes the ID field in IP packets to be randomized instead of -incremented by 1 with each packet generated. +to 1 causes the ID field in +.Em non-atomic +IP datagrams (or all IP datagrams, if +.Va ip.rfc6864 +is disabled) to be randomized instead of incremented by 1 with each packet +generated. This closes a minor information leak which allows remote observers to determine the rate of packet generation on the machine by watching the counter. Modified: head/sys/contrib/ipfilter/netinet/fil.c ============================================================================== --- head/sys/contrib/ipfilter/netinet/fil.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/contrib/ipfilter/netinet/fil.c Wed Apr 1 22:26:39 2015 (r280971) @@ -6086,23 +6086,24 @@ ipf_updateipid(fin) u_32_t sumd, sum; ip_t *ip; + ip = fin->fin_ip; + ido = ntohs(ip->ip_id); if (fin->fin_off != 0) { sum = ipf_frag_ipidknown(fin); if (sum == 0xffffffff) return -1; sum &= 0xffff; id = (u_short)sum; + ip->ip_id = htons(id); } else { - id = ipf_nextipid(fin); - if (fin->fin_off == 0 && (fin->fin_flx & FI_FRAG) != 0) + ip_fillid(ip); + id = ntohs(ip->ip_id); + if ((fin->fin_flx & FI_FRAG) != 0) (void) ipf_frag_ipidnew(fin, (u_32_t)id); } - ip = fin->fin_ip; - ido = ntohs(ip->ip_id); if (id == ido) return 0; - ip->ip_id = htons(id); CALC_SUMD(ido, id, sumd); /* DESTRUCTIVE MACRO! id,ido change */ sum = (~ntohs(ip->ip_sum)) & 0xffff; sum += sumd; Modified: head/sys/contrib/ipfilter/netinet/ip_fil.h ============================================================================== --- head/sys/contrib/ipfilter/netinet/ip_fil.h Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/contrib/ipfilter/netinet/ip_fil.h Wed Apr 1 22:26:39 2015 (r280971) @@ -1718,6 +1718,7 @@ extern int ipfioctl __P((ipf_main_softc_ extern void m_freem __P((mb_t *)); extern size_t msgdsize __P((mb_t *)); extern int bcopywrap __P((void *, void *, size_t)); +extern void ip_fillid(struct ip *); #else /* #ifndef _KERNEL */ # if defined(__NetBSD__) && defined(PFIL_HOOKS) extern void ipfilterattach __P((int)); @@ -1932,7 +1933,6 @@ extern int ipf_matchtag __P((ipftag_t * extern int ipf_matchicmpqueryreply __P((int, icmpinfo_t *, struct icmp *, int)); extern u_32_t ipf_newisn __P((fr_info_t *)); -extern u_short ipf_nextipid __P((fr_info_t *)); extern u_int ipf_pcksum __P((fr_info_t *, int, u_int)); extern void ipf_rule_expire __P((ipf_main_softc_t *)); extern int ipf_scanlist __P((fr_info_t *, u_32_t)); Modified: head/sys/contrib/ipfilter/netinet/ip_fil_freebsd.c ============================================================================== --- head/sys/contrib/ipfilter/netinet/ip_fil_freebsd.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/contrib/ipfilter/netinet/ip_fil_freebsd.c Wed Apr 1 22:26:39 2015 (r280971) @@ -97,7 +97,6 @@ MALLOC_DEFINE(M_IPFILTER, "ipfilter", "I # endif -static u_short ipid = 0; static int (*ipf_savep) __P((void *, ip_t *, int, void *, int, struct mbuf **)); static int ipf_send_ip __P((fr_info_t *, mb_t *)); static void ipf_timer_func __P((void *arg)); @@ -231,8 +230,6 @@ ipfattach(softc) if (softc->ipf_control_forwarding & 1) V_ipforwarding = 1; - ipid = 0; - SPL_X(s); #if 0 softc->ipf_slow_ch = timeout(ipf_timer_func, softc, @@ -1074,31 +1071,6 @@ ipf_newisn(fin) } -/* ------------------------------------------------------------------------ */ -/* Function: ipf_nextipid */ -/* Returns: int - 0 == success, -1 == error (packet should be droppped) */ -/* Parameters: fin(I) - pointer to packet information */ -/* */ -/* Returns the next IPv4 ID to use for this packet. */ -/* ------------------------------------------------------------------------ */ -u_short -ipf_nextipid(fin) - fr_info_t *fin; -{ - u_short id; - -#ifndef RANDOM_IP_ID - MUTEX_ENTER(&ipfmain.ipf_rw); - id = ipid++; - MUTEX_EXIT(&ipfmain.ipf_rw); -#else - id = ip_randomid(); -#endif - - return id; -} - - INLINE int ipf_checkv4sum(fin) fr_info_t *fin; Modified: head/sys/contrib/ipfilter/netinet/ip_nat.c ============================================================================== --- head/sys/contrib/ipfilter/netinet/ip_nat.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/contrib/ipfilter/netinet/ip_nat.c Wed Apr 1 22:26:39 2015 (r280971) @@ -5221,7 +5221,7 @@ ipf_nat_out(fin, nat, natadd, nflags) } ip = MTOD(m, ip_t *); - ip->ip_id = htons(ipf_nextipid(fin)); + ip_fillid(ip); s2 = ntohs(ip->ip_id); s1 = ip->ip_len; @@ -5666,7 +5666,7 @@ ipf_nat_in(fin, nat, natadd, nflags) } ip = MTOD(m, ip_t *); - ip->ip_id = htons(ipf_nextipid(fin)); + ip_fillid(ip); sum1 = ntohs(ip->ip_len); ip->ip_len = ntohs(ip->ip_len); ip->ip_len += fin->fin_plen; Modified: head/sys/netinet/ip_carp.c ============================================================================== --- head/sys/netinet/ip_carp.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netinet/ip_carp.c Wed Apr 1 22:26:39 2015 (r280971) @@ -838,11 +838,11 @@ carp_send_ad_locked(struct carp_softc *s ip->ip_hl = sizeof(*ip) >> 2; ip->ip_tos = IPTOS_LOWDELAY; ip->ip_len = htons(len); - ip->ip_id = ip_newid(); ip->ip_off = htons(IP_DF); ip->ip_ttl = CARP_DFLTTL; ip->ip_p = IPPROTO_CARP; ip->ip_sum = 0; + ip_fillid(ip); bzero(&sa, sizeof(sa)); sa.sa_family = AF_INET; Modified: head/sys/netinet/ip_gre.c ============================================================================== --- head/sys/netinet/ip_gre.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netinet/ip_gre.c Wed Apr 1 22:26:39 2015 (r280971) @@ -145,7 +145,7 @@ in_gre_output(struct mbuf *m, int af, in #ifdef INET6 case AF_INET6: gi->gi_ip.ip_tos = 0; /* XXX */ - gi->gi_ip.ip_id = ip_newid(); + ip_fillid(&gi->gi_ip); break; #endif } Modified: head/sys/netinet/ip_id.c ============================================================================== --- head/sys/netinet/ip_id.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netinet/ip_id.c Wed Apr 1 22:26:39 2015 (r280971) @@ -74,26 +74,37 @@ __FBSDID("$FreeBSD$"); * enabled. */ -#include -#include #include -#include -#include -#include +#include +#include +#include #include #include #include -#include +#include #include #include #include #include +#include #include -static MALLOC_DEFINE(M_IPID, "ipid", "randomized ip id state"); +/* + * By default we generate IP ID only for non-atomic datagrams, as + * suggested by RFC6864. We use per-CPU counter for that, or if + * user wants to, we can turn on random ID generation. + */ +static VNET_DEFINE(int, ip_rfc6864) = 1; +static VNET_DEFINE(int, ip_do_randomid) = 0; +#define V_ip_rfc6864 VNET(ip_rfc6864) +#define V_ip_do_randomid VNET(ip_do_randomid) +/* + * Random ID state engine. + */ +static MALLOC_DEFINE(M_IPID, "ipid", "randomized ip id state"); static VNET_DEFINE(uint16_t *, id_array); static VNET_DEFINE(bitstr_t *, id_bits); static VNET_DEFINE(int, array_ptr); @@ -109,12 +120,27 @@ static VNET_DEFINE(struct mtx, ip_id_mtx #define V_random_id_total VNET(random_id_total) #define V_ip_id_mtx VNET(ip_id_mtx) -static void ip_initid(int); +/* + * Non-random ID state engine is simply a per-cpu counter. + */ +static VNET_DEFINE(counter_u64_t, ip_id); +#define V_ip_id VNET(ip_id) + +static int sysctl_ip_randomid(SYSCTL_HANDLER_ARGS); static int sysctl_ip_id_change(SYSCTL_HANDLER_ARGS); +static void ip_initid(int); +static uint16_t ip_randomid(void); static void ipid_sysinit(void); static void ipid_sysuninit(void); SYSCTL_DECL(_net_inet_ip); +SYSCTL_PROC(_net_inet_ip, OID_AUTO, random_id, + CTLTYPE_INT | CTLFLAG_VNET | CTLFLAG_RW, + &VNET_NAME(ip_do_randomid), 0, sysctl_ip_randomid, "IU", + "Assign random ip_id values"); +SYSCTL_INT(_net_inet_ip, OID_AUTO, rfc6864, CTLFLAG_VNET | CTLFLAG_RW, + &VNET_NAME(ip_rfc6864), 0, + "Use constant IP ID for atomic datagrams"); SYSCTL_PROC(_net_inet_ip, OID_AUTO, random_id_period, CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_VNET, &VNET_NAME(array_size), 0, sysctl_ip_id_change, "IU", "IP ID Array size"); @@ -125,6 +151,26 @@ SYSCTL_INT(_net_inet_ip, OID_AUTO, rando &VNET_NAME(random_id_total), 0, "Count of IP IDs created"); static int +sysctl_ip_randomid(SYSCTL_HANDLER_ARGS) +{ + int error, new; + + new = V_ip_do_randomid; + error = sysctl_handle_int(oidp, &new, 0, req); + if (error || req->newptr == NULL) + return (error); + if (new != 0 && new != 1) + return (EINVAL); + if (new == V_ip_do_randomid) + return (0); + if (new == 1 && V_ip_do_randomid == 0) + ip_initid(8192); + /* We don't free memory when turning random ID off, due to race. */ + V_ip_do_randomid = new; + return (0); +} + +static int sysctl_ip_id_change(SYSCTL_HANDLER_ARGS) { int error, new; @@ -164,7 +210,7 @@ ip_initid(int new_size) mtx_unlock(&V_ip_id_mtx); } -uint16_t +static uint16_t ip_randomid(void) { uint16_t new_id; @@ -191,12 +237,34 @@ ip_randomid(void) return (new_id); } +void +ip_fillid(struct ip *ip) +{ + + /* + * Per RFC6864 Section 4 + * + * o Atomic datagrams: (DF==1) && (MF==0) && (frag_offset==0) + * o Non-atomic datagrams: (DF==0) || (MF==1) || (frag_offset>0) + */ + if (V_ip_rfc6864 && (ip->ip_off & htons(IP_DF)) == htons(IP_DF)) + ip->ip_id = 0; + else if (V_ip_do_randomid) + ip->ip_id = ip_randomid(); + else { + counter_u64_add(V_ip_id, 1); + ip->ip_id = htons((*(uint64_t *)zpcpu_get(V_ip_id)) & 0xffff); + } +} + static void ipid_sysinit(void) { mtx_init(&V_ip_id_mtx, "ip_id_mtx", NULL, MTX_DEF); - ip_initid(8192); + V_ip_id = counter_u64_alloc(M_WAITOK); + for (int i = 0; i < mp_ncpus; i++) + arc4rand(zpcpu_get_cpu(V_ip_id, i), sizeof(uint64_t), 0); } VNET_SYSINIT(ip_id, SI_SUB_PROTO_DOMAIN, SI_ORDER_ANY, ipid_sysinit, NULL); @@ -205,7 +273,10 @@ ipid_sysuninit(void) { mtx_destroy(&V_ip_id_mtx); - free(V_id_array, M_IPID); - free(V_id_bits, M_IPID); + if (V_id_array != NULL) { + free(V_id_array, M_IPID); + free(V_id_bits, M_IPID); + } + counter_u64_free(V_ip_id); } VNET_SYSUNINIT(ip_id, SI_SUB_PROTO_DOMAIN, SI_ORDER_ANY, ipid_sysuninit, NULL); Modified: head/sys/netinet/ip_input.c ============================================================================== --- head/sys/netinet/ip_input.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netinet/ip_input.c Wed Apr 1 22:26:39 2015 (r280971) @@ -105,11 +105,6 @@ SYSCTL_INT(_net_inet_ip, IPCTL_SENDREDIR &VNET_NAME(ipsendredirects), 0, "Enable sending IP redirects"); -VNET_DEFINE(int, ip_do_randomid); -SYSCTL_INT(_net_inet_ip, OID_AUTO, random_id, CTLFLAG_VNET | CTLFLAG_RW, - &VNET_NAME(ip_do_randomid), 0, - "Assign random ip_id values"); - /* * XXX - Setting ip_checkinterface mostly implements the receive side of * the Strong ES model described in RFC 1122, but since the routing table @@ -331,8 +326,6 @@ ip_init(void) struct protosw *pr; int i; - V_ip_id = time_second & 0xffff; - TAILQ_INIT(&V_in_ifaddrhead); V_in_ifaddrhashtbl = hashinit(INADDR_NHASH, M_IFADDR, &V_in_ifaddrhmask); Modified: head/sys/netinet/ip_mroute.c ============================================================================== --- head/sys/netinet/ip_mroute.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netinet/ip_mroute.c Wed Apr 1 22:26:39 2015 (r280971) @@ -2501,7 +2501,6 @@ pim_register_send_rp(struct ip *ip, stru */ ip_outer = mtod(mb_first, struct ip *); *ip_outer = pim_encap_iphdr; - ip_outer->ip_id = ip_newid(); ip_outer->ip_len = htons(len + sizeof(pim_encap_iphdr) + sizeof(pim_encap_pimhdr)); ip_outer->ip_src = V_viftable[vifi].v_lcl_addr; @@ -2513,6 +2512,7 @@ pim_register_send_rp(struct ip *ip, stru ip_outer->ip_tos = ip->ip_tos; if (ip->ip_off & htons(IP_DF)) ip_outer->ip_off |= htons(IP_DF); + ip_fillid(ip_outer); pimhdr = (struct pim_encap_pimhdr *)((caddr_t)ip_outer + sizeof(pim_encap_iphdr)); *pimhdr = pim_encap_pimhdr; Modified: head/sys/netinet/ip_output.c ============================================================================== --- head/sys/netinet/ip_output.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netinet/ip_output.c Wed Apr 1 22:26:39 2015 (r280971) @@ -91,8 +91,6 @@ __FBSDID("$FreeBSD$"); #include -VNET_DEFINE(uint32_t, ip_id); - #ifdef MBUF_STRESS_TEST static int mbuf_frag_size = 0; SYSCTL_INT(_net_inet_ip, OID_AUTO, mbuf_frag_size, CTLFLAG_RW, @@ -174,21 +172,10 @@ ip_output(struct mbuf *m, struct mbuf *o ip_len = ntohs(ip->ip_len); ip_off = ntohs(ip->ip_off); - /* - * Fill in IP header. If we are not allowing fragmentation, - * then the ip_id field is meaningless, but we don't set it - * to zero. Doing so causes various problems when devices along - * the path (routers, load balancers, firewalls, etc.) illegally - * disable DF on our packet. Note that a 16-bit counter - * will wrap around in less than 10 seconds at 100 Mbit/s on a - * medium with MTU 1500. See Steven M. Bellovin, "A Technique - * for Counting NATted Hosts", Proc. IMW'02, available at - * . - */ if ((flags & (IP_FORWARDING|IP_RAWOUTPUT)) == 0) { ip->ip_v = IPVERSION; ip->ip_hl = hlen >> 2; - ip->ip_id = ip_newid(); + ip_fillid(ip); IPSTAT_INC(ips_localout); } else { /* Header already set, fetch hlen from there */ Modified: head/sys/netinet/ip_var.h ============================================================================== --- head/sys/netinet/ip_var.h Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netinet/ip_var.h Wed Apr 1 22:26:39 2015 (r280971) @@ -174,7 +174,6 @@ struct inpcb; struct route; struct sockopt; -VNET_DECLARE(uint32_t, ip_id); /* ip packet ctr, for ids */ VNET_DECLARE(int, ip_defttl); /* default IP ttl */ VNET_DECLARE(int, ipforwarding); /* ip forwarding */ #ifdef IPSTEALTH @@ -228,7 +227,7 @@ struct in_ifaddr * void ip_savecontrol(struct inpcb *, struct mbuf **, struct ip *, struct mbuf *); void ip_slowtimo(void); -uint16_t ip_randomid(void); +void ip_fillid(struct ip *); int rip_ctloutput(struct socket *, struct sockopt *); void rip_ctlinput(int, struct sockaddr *, void *); void rip_init(void); @@ -302,22 +301,6 @@ extern int (*ng_ipfw_input_p)(struct mbu extern int (*ip_dn_ctl_ptr)(struct sockopt *); extern int (*ip_dn_io_ptr)(struct mbuf **, int, struct ip_fw_args *); - -VNET_DECLARE(int, ip_do_randomid); -#define V_ip_do_randomid VNET(ip_do_randomid) -static __inline uint16_t -ip_newid(void) -{ - uint16_t res; - - if (V_ip_do_randomid != 0) - return (ip_randomid()); - else { - res = atomic_fetchadd_32(&V_ip_id, 1) & 0xFFFF; - return (htons(res)); - } -} - #endif /* _KERNEL */ #endif /* !_NETINET_IP_VAR_H_ */ Modified: head/sys/netinet/raw_ip.c ============================================================================== --- head/sys/netinet/raw_ip.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netinet/raw_ip.c Wed Apr 1 22:26:39 2015 (r280971) @@ -505,8 +505,12 @@ rip_output(struct mbuf *m, struct socket m_freem(m); return (EINVAL); } + /* + * This doesn't allow application to specify ID of zero, + * but we got this limitation from the beginning of history. + */ if (ip->ip_id == 0) - ip->ip_id = ip_newid(); + ip_fillid(ip); /* * XXX prevent ip_output from overwriting header fields. Modified: head/sys/netinet/sctp_output.c ============================================================================== --- head/sys/netinet/sctp_output.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netinet/sctp_output.c Wed Apr 1 22:26:39 2015 (r280971) @@ -4106,7 +4106,7 @@ sctp_lowlevel_chunk_output(struct sctp_i ip->ip_off = htons(0); } /* FreeBSD has a function for ip_id's */ - ip->ip_id = ip_newid(); + ip_fillid(ip); ip->ip_ttl = inp->ip_inp.inp.inp_ip_ttl; ip->ip_len = htons(packet_length); @@ -10949,8 +10949,8 @@ sctp_send_resp_msg(struct sockaddr *src, ip->ip_v = IPVERSION; ip->ip_hl = (sizeof(struct ip) >> 2); ip->ip_tos = 0; - ip->ip_id = ip_newid(); ip->ip_off = 0; + ip_fillid(ip); ip->ip_ttl = MODULE_GLOBAL(ip_defttl); if (port) { ip->ip_p = IPPROTO_UDP; Modified: head/sys/netipsec/xform_ipip.c ============================================================================== --- head/sys/netipsec/xform_ipip.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netipsec/xform_ipip.c Wed Apr 1 22:26:39 2015 (r280971) @@ -136,9 +136,6 @@ ipip_output(struct mbuf *m, struct ipsec ipo->ip_sum = 0; ipo->ip_src = saidx->src.sin.sin_addr; ipo->ip_dst = saidx->dst.sin.sin_addr; - - ipo->ip_id = ip_newid(); - /* If the inner protocol is IP... */ switch (tp) { case IPVERSION: @@ -178,6 +175,7 @@ ipip_output(struct mbuf *m, struct ipsec default: goto nofamily; } + ip_fillid(ipo); otos = 0; ip_ecn_ingress(ECN_ALLOWED, &otos, &itos); Modified: head/sys/netpfil/pf/if_pfsync.c ============================================================================== --- head/sys/netpfil/pf/if_pfsync.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netpfil/pf/if_pfsync.c Wed Apr 1 22:26:39 2015 (r280971) @@ -1538,7 +1538,7 @@ pfsync_sendout(int schedswi) offset = sizeof(*ip); ip->ip_len = htons(m->m_pkthdr.len); - ip->ip_id = htons(ip_randomid()); + ip_fillid(ip); /* build the pfsync header */ ph = (struct pfsync_header *)(m->m_data + offset); Modified: head/sys/netpfil/pf/pf_norm.c ============================================================================== --- head/sys/netpfil/pf/pf_norm.c Wed Apr 1 21:48:54 2015 (r280970) +++ head/sys/netpfil/pf/pf_norm.c Wed Apr 1 22:26:39 2015 (r280971) @@ -2271,9 +2271,9 @@ pf_scrub_ip(struct mbuf **m0, u_int32_t /* random-id, but not for fragments */ if (flags & PFRULE_RANDOMID && !(h->ip_off & ~htons(IP_DF))) { - u_int16_t ip_id = h->ip_id; + uint16_t ip_id = h->ip_id; - h->ip_id = ip_randomid(); + ip_fillid(h); h->ip_sum = pf_cksum_fixup(h->ip_sum, ip_id, h->ip_id, 0); } }