Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 12 Feb 2021 11:16:07 -0600
From:      "Brandon Bergren" <bdragon@FreeBSD.org>
To:        "Mateusz Guzik" <mjguzik@gmail.com>, "Alexander V. Chernikov" <melifaro@freebsd.org>
Cc:        src-committers@freebsd.org, dev-commits-src-all@freebsd.org, dev-commits-src-main@freebsd.org
Subject:   Re: git: 145bf6c0af48 - main - Fix blackhole/reject routes.
Message-ID:  <2a143d84-ee5e-48f6-99a9-366b3f0fdf40@www.fastmail.com>
In-Reply-To: <CAGudoHFyKrj8LofGASdGbDk1%2BUruRYM1_2kgCQduG4%2BaTvu3Lw@mail.gmail.com>
References:  <202102112310.11BNABTx015766@gitrepo.freebsd.org> <CAGudoHHm8puj39FBVH-g7pBtf=duCGRqx7dPyQR4d8QRVkR6eg@mail.gmail.com> <CAGudoHFyKrj8LofGASdGbDk1%2BUruRYM1_2kgCQduG4%2BaTvu3Lw@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
This actually directly breaks the mips and mips64 builds because the MALTA and MALTA64 kernels do not have INET6 enabled.

On Thu, Feb 11, 2021, at 7:59 PM, Mateusz Guzik wrote:
> I mean *without* INET6
> 
> On 2/12/21, Mateusz Guzik <mjguzik@gmail.com> wrote:
> > This breaks building a kernel with INET6 for me, I presume it will
> > also break LINT-NOINET
> >
> > On 2/12/21, Alexander V. Chernikov <melifaro@freebsd.org> wrote:
> >> The branch main has been updated by melifaro:
> >>
> >> URL:
> >> https://cgit.FreeBSD.org/src/commit/?id=145bf6c0af48b89f13465e145f4516de37c31d85
> >>
> >> commit 145bf6c0af48b89f13465e145f4516de37c31d85
> >> Author:     Alexander V. Chernikov <melifaro@FreeBSD.org>
> >> AuthorDate: 2021-02-08 23:29:05 +0000
> >> Commit:     Alexander V. Chernikov <melifaro@FreeBSD.org>
> >> CommitDate: 2021-02-11 23:08:55 +0000
> >>
> >>     Fix blackhole/reject routes.
> >>
> >>     Traditionally *BSD routing stack required to supply some
> >>      interface data for blackhole/reject routes. This lead to
> >>      varieties of hacks in routing daemons when inserting such routes.
> >>     With the recent routeing stack changes, gateway sockaddr without
> >>      RTF_GATEWAY started to be treated differently, purely as link
> >>      identifier.
> >>
> >>     This change broke net/bird, which installs blackhole routes with
> >>      127.0.0.1 gateway without RTF_GATEWAY flags.
> >>
> >>     Fix this by automatically constructing necessary gateway data at
> >>      rtsock level if RTF_REJECT/RTF_BLACKHOLE is set.
> >>
> >>     Reported by:    Marek Zarychta <zarychtam at plan-b.pwste.edu.pl>
> >>     Reviewed by:    donner
> >>     MFC after:      1 week
> >> ---
> >>  sys/net/rtsock.c | 58
> >> ++++++++++++++++++++++++++++++++++++++++++++++++++++++--
> >>  1 file changed, 56 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/sys/net/rtsock.c b/sys/net/rtsock.c
> >> index ba1182d55439..f67252f9fd5f 100644
> >> --- a/sys/net/rtsock.c
> >> +++ b/sys/net/rtsock.c
> >> @@ -562,6 +562,50 @@ rtm_get_jailed(struct rt_addrinfo *info, struct
> >> ifnet
> >> *ifp,
> >>  	return (0);
> >>  }
> >>
> >> +static int
> >> +fill_blackholeinfo(struct rt_addrinfo *info, union sockaddr_union *saun)
> >> +{
> >> +	struct ifaddr *ifa;
> >> +	sa_family_t saf;
> >> +
> >> +	if (V_loif == NULL) {
> >> +		printf("Unable to add blackhole/reject nhop without loopback");
> >> +		return (ENOTSUP);
> >> +	}
> >> +	info->rti_ifp = V_loif;
> >> +
> >> +	saf = info->rti_info[RTAX_DST]->sa_family;
> >> +
> >> +	CK_STAILQ_FOREACH(ifa, &info->rti_ifp->if_addrhead, ifa_link) {
> >> +		if (ifa->ifa_addr->sa_family == saf) {
> >> +			info->rti_ifa = ifa;
> >> +			break;
> >> +		}
> >> +	}
> >> +	if (info->rti_ifa == NULL)
> >> +		return (ENOTSUP);
> >> +
> >> +	bzero(saun, sizeof(union sockaddr_union));
> >> +	switch (saf) {
> >> +	case AF_INET:
> >> +		saun->sin.sin_family = AF_INET;
> >> +		saun->sin.sin_len = sizeof(struct sockaddr_in);
> >> +		saun->sin.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
> >> +		break;
> >> +	case AF_INET6:
> >> +		saun->sin6.sin6_family = AF_INET6;
> >> +		saun->sin6.sin6_len = sizeof(struct sockaddr_in6);
> >> +		saun->sin6.sin6_addr = in6addr_loopback;
> >> +		break;
> >> +	default:
> >> +		return (ENOTSUP);
> >> +	}
> >> +	info->rti_info[RTAX_GATEWAY] = &saun->sa;
> >> +	info->rti_flags |= RTF_GATEWAY;
> >> +
> >> +	return (0);
> >> +}
> >> +
> >>  /*
> >>   * Fills in @info based on userland-provided @rtm message.
> >>   *
> >> @@ -944,7 +988,6 @@ route_output(struct mbuf *m, struct socket *so, ...)
> >>  #endif
> >>  	int alloc_len = 0, len, error = 0, fibnum;
> >>  	sa_family_t saf = AF_UNSPEC;
> >> -	struct walkarg w;
> >>  	struct rib_cmd_info rc;
> >>  	struct nhop_object *nh;
> >>
> >> @@ -972,7 +1015,6 @@ route_output(struct mbuf *m, struct socket *so, ...)
> >>
> >>  	m_copydata(m, 0, len, (caddr_t)rtm);
> >>  	bzero(&info, sizeof(info));
> >> -	bzero(&w, sizeof(w));
> >>  	nh = NULL;
> >>
> >>  	if (rtm->rtm_version != RTM_VERSION) {
> >> @@ -1004,6 +1046,18 @@ route_output(struct mbuf *m, struct socket *so,
> >> ...)
> >>  		goto flush;
> >>  	}
> >>
> >> +	union sockaddr_union gw_saun;
> >> +	int blackhole_flags = rtm->rtm_flags & (RTF_BLACKHOLE|RTF_REJECT);
> >> +	if (blackhole_flags != 0) {
> >> +		if (blackhole_flags != (RTF_BLACKHOLE | RTF_REJECT))
> >> +			error = fill_blackholeinfo(&info, &gw_saun);
> >> +		else
> >> +			error = EINVAL;
> >> +		if (error != 0)
> >> +			senderr(error);
> >> +		/* TODO: rebuild rtm from scratch */
> >> +	}
> >> +
> >>  	switch (rtm->rtm_type) {
> >>  	case RTM_ADD:
> >>  	case RTM_CHANGE:
> >> _______________________________________________
> >> dev-commits-src-all@freebsd.org mailing list
> >> https://lists.freebsd.org/mailman/listinfo/dev-commits-src-all
> >> To unsubscribe, send any mail to
> >> "dev-commits-src-all-unsubscribe@freebsd.org"
> >>
> >
> >
> > --
> > Mateusz Guzik <mjguzik gmail.com>
> >
> 
> 
> -- 
> Mateusz Guzik <mjguzik gmail.com>
>

-- 
  Brandon Bergren
  bdragon@FreeBSD.org



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2a143d84-ee5e-48f6-99a9-366b3f0fdf40>