From: David Gibson <david@gibson.dropbear.id.au>
To: Jon Maloy <jmaloy@redhat.com>
Cc: sbrivio@redhat.com, dgibson@redhat.com, passt-dev@passt.top
Subject: Re: [PATCH v3 10/11] ip: Track observed guest IPv6 addresses in unified address array
Date: Tue, 10 Feb 2026 09:30:05 +1100 [thread overview]
Message-ID: <aYpf7TUtPkXafcoD@zatzit> (raw)
In-Reply-To: <20260130214447.2540791-11-jmaloy@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 16558 bytes --]
On Fri, Jan 30, 2026 at 04:44:46PM -0500, Jon Maloy wrote:
> We remove the addr_seen and addr_ll_seen fields in struct ip6_ctx
> and replace them by setting INANY_ADDR_OBSERVED and INANY_ADDR_LINKLOCAL
> flags in the corresponding entry in the unified address array.
> If the seen address is not present in the array we add it first.
>
> This completes the unification of address storage for both IPv4 and
> IPv6, enabling future support for multiple guest addresses per family.
>
> Signed-off-by: Jon Maloy <jmaloy@redhat.com>
A number of the comments on 9/11 will also apply here.
> ---
> v3: - Adapted to previous changes in this series
> ---
> conf.c | 5 ----
> dhcpv6.c | 6 ++---
> dhcpv6.h | 2 +-
> fwd.c | 62 +++++++++++++++++++++++++++++++------------
> migrate.c | 64 +++++++++++++++++++++++++++++++++++++++------
> passt.h | 5 +---
> pasta.c | 15 ++++++++++-
> tap.c | 78 +++++++++++++++++++++++++++++++++++++++++++++----------
> 8 files changed, 185 insertions(+), 52 deletions(-)
>
> diff --git a/conf.c b/conf.c
> index c5818a8..6892cea 100644
> --- a/conf.c
> +++ b/conf.c
> @@ -764,7 +764,6 @@ static void conf_ip4_local(struct ctx *c)
> */
> static unsigned int conf_ip6(unsigned int ifi, struct ctx *c)
> {
> - struct inany_addr_entry *e;
> union inany_addr addr = {0,};
> int prefix_len = 0;
> int rc;
> @@ -800,10 +799,6 @@ static unsigned int conf_ip6(unsigned int ifi, struct ctx *c)
> c->addr_count++;
> }
>
> - e = first_v6(c);
> - if (e)
> - c->ip6.addr_seen = e->addr.a6;
> -
> if (IN6_IS_ADDR_LINKLOCAL(&c->ip6.guest_gw))
> c->ip6.our_tap_ll = c->ip6.guest_gw;
>
> diff --git a/dhcpv6.c b/dhcpv6.c
> index a1e5f15..67307d9 100644
> --- a/dhcpv6.c
> +++ b/dhcpv6.c
> @@ -546,10 +546,11 @@ static size_t dhcpv6_client_fqdn_fill(const struct iov_tail *data,
> * Return: 0 if it's not a DHCPv6 message, 1 if handled, -1 on failure
> */
> int dhcpv6(struct ctx *c, struct iov_tail *data,
> - const struct in6_addr *saddr, const struct in6_addr *daddr)
> + const struct in6_addr *daddr)
> {
> const struct opt_server_id *server_id = NULL;
> const struct opt_hdr *client_id = NULL;
> +
Unrelated whitespace change.
> /* The _storage variables can't be local to the blocks they're used in,
> * because IOV_*_HEADER() may return pointers to them which are
> * dereferenced afterwards. Since we don't have Rust-like lifetime
> @@ -587,8 +588,6 @@ int dhcpv6(struct ctx *c, struct iov_tail *data,
> if (mlen + sizeof(*uh) != ntohs(uh->len) || mlen < sizeof(*mh))
> return -1;
>
> - c->ip6.addr_ll_seen = *saddr;
> -
It's stored differently, but we still want to update the observed
ip6_ll address here.
> src = &c->ip6.our_tap_ll;
>
> mh = IOV_REMOVE_HEADER(data, mh_storage);
> @@ -679,7 +678,6 @@ int dhcpv6(struct ctx *c, struct iov_tail *data,
>
> tap_udp6_send(c, src, 547, tap_ip6_daddr(c, src), 546,
> mh->xid, &resp, n);
> - c->ip6.addr_seen = first_v6(c)->addr.a6;
>
> return 1;
> }
> diff --git a/dhcpv6.h b/dhcpv6.h
> index c706dfd..8cbc769 100644
> --- a/dhcpv6.h
> +++ b/dhcpv6.h
> @@ -7,7 +7,7 @@
> #define DHCPV6_H
>
> int dhcpv6(struct ctx *c, struct iov_tail *data,
> - struct in6_addr *saddr, struct in6_addr *daddr);
> + struct in6_addr *daddr);
> void dhcpv6_init(const struct ctx *c);
>
> #endif /* DHCPV6_H */
> diff --git a/fwd.c b/fwd.c
> index 5af482d..fc26eba 100644
> --- a/fwd.c
> +++ b/fwd.c
> @@ -515,6 +515,36 @@ static const struct in_addr *fwd_guest_addr4(const struct ctx *c)
> return NULL;
> }
>
> +/**
> + * fwd_guest_addr6() - Get first observed IPv6 guest address
> + * @c: Execution context
> + * @linklocal: If true, find link-local address; if false, find global
> + *
> + * Return: first observed IPv6 address, if any,
> + * or first configured, if any, or NULL
> + */
> +static const struct in6_addr *fwd_guest_addr6(const struct ctx *c,
> + bool linklocal)
> +{
> + const struct inany_addr_entry *e;
> +
> + /* Find first observed address of matching scope */
> + for_each_addr(c, e, AF_INET6) {
> + if (!!(e->flags & INANY_ADDR_LINKLOCAL) != linklocal)
> + continue;
> + if (e->flags & INANY_ADDR_OBSERVED)
> + return &e->addr.a6;
> + }
> +
> + /* Fallback to first address of matching scope */
> + for_each_addr(c, e, AF_INET6) {
> + if (!!(e->flags & INANY_ADDR_LINKLOCAL) == linklocal)
> + return &e->addr.a6;
> + }
> +
> + return NULL;
> +}
> +
> /**
> * fwd_guest_accessible4() - Is IPv4 address guest-accessible
> * @c: Execution context
> @@ -563,19 +593,11 @@ static bool fwd_guest_accessible6(const struct ctx *c,
> if (IN6_IS_ADDR_LOOPBACK(addr))
> return false;
>
> - /* Check against all configured guest addresses */
> + /* Check against all guest addresses (configured, host, or observed) */
> for_each_addr(c, e, AF_INET6)
> if (IN6_ARE_ADDR_EQUAL(addr, &e->addr.a6))
> return false;
>
> - /* For IPv6, addr_seen starts unspecified, because we don't know what LL
> - * address the guest will take until we see it. Only check against it
> - * if it has been set to a real address.
> - */
> - if (!IN6_IS_ADDR_UNSPECIFIED(&c->ip6.addr_seen) &&
> - IN6_ARE_ADDR_EQUAL(addr, &c->ip6.addr_seen))
> - return false;
> -
> return true;
> }
>
> @@ -798,10 +820,16 @@ uint8_t fwd_nat_from_host(const struct ctx *c, uint8_t proto,
> }
> tgt->oaddr = inany_any4;
> } else {
> - if (c->host_lo_to_ns_lo)
> + if (c->host_lo_to_ns_lo) {
> tgt->eaddr = inany_loopback6;
> - else
> - tgt->eaddr.a6 = c->ip6.addr_seen;
> + } else {
> + const struct in6_addr *guest_addr6;
> +
> + guest_addr6 = fwd_guest_addr6(c, false);
> + if (!guest_addr6)
> + return PIF_NONE;
> + tgt->eaddr.a6 = *guest_addr6;
> + }
> tgt->oaddr = inany_any6;
> }
>
> @@ -833,10 +861,12 @@ uint8_t fwd_nat_from_host(const struct ctx *c, uint8_t proto,
> return PIF_NONE;
> tgt->eaddr = inany_from_v4(*guest_addr);
> } else {
> - if (inany_is_linklocal6(&tgt->oaddr))
> - tgt->eaddr.a6 = c->ip6.addr_ll_seen;
> - else
> - tgt->eaddr.a6 = c->ip6.addr_seen;
> + bool linklocal = inany_is_linklocal6(&tgt->oaddr);
> + const struct in6_addr *guest_addr6 = fwd_guest_addr6(c, linklocal);
> +
> + if (!guest_addr6)
> + return PIF_NONE;
> + tgt->eaddr.a6 = *guest_addr6;
> }
>
> return PIF_TAP;
> diff --git a/migrate.c b/migrate.c
> index a84af1c..20da1f7 100644
> --- a/migrate.c
> +++ b/migrate.c
> @@ -66,10 +66,7 @@ struct migrate_seen_addrs_v3 {
> static int seen_addrs_source_v2(struct ctx *c,
> const struct migrate_stage *stage, int fd)
> {
> - struct migrate_seen_addrs_v2 addrs = {
> - .addr6 = c->ip6.addr_seen,
> - .addr6_ll = c->ip6.addr_ll_seen,
> - };
> + struct migrate_seen_addrs_v2 addrs = { 0 };
> const struct inany_addr_entry *e;
>
> (void)stage;
> @@ -82,6 +79,23 @@ static int seen_addrs_source_v2(struct ctx *c,
> }
> }
>
> + /* Find first observed IPv6 addresses (global and link-local) */
> + for_each_addr(c, e, AF_INET6) {
> + struct in6_addr tmp;
> +
> + if (!(e->flags & INANY_ADDR_OBSERVED))
> + continue;
> + if (e->flags & INANY_ADDR_LINKLOCAL) {
> + tmp = addrs.addr6_ll;
> + if (IN6_IS_ADDR_UNSPECIFIED(&tmp))
> + addrs.addr6_ll = e->addr.a6;
> + } else {
> + tmp = addrs.addr6;
> + if (IN6_IS_ADDR_UNSPECIFIED(&tmp))
> + addrs.addr6 = e->addr.a6;
> + }
> + }
You could reuse fwd_guest_addr6() here.
> +
> memcpy(addrs.mac, c->guest_mac, sizeof(addrs.mac));
>
> if (write_all_buf(fd, &addrs, sizeof(addrs)))
> @@ -117,6 +131,37 @@ static void migrate_observed_addr4(struct ctx *c, const struct in_addr *addr)
> }
> }
>
> +/**
> + * migrate_observed_addr6() - Add observed IPv6 address to address array
> + * @c: Execution context
> + * @addr: IPv6 address to add
> + */
> +static void migrate_observed_addr6(struct ctx *c, const struct in6_addr *addr)
> +{
> + uint8_t flags = INANY_ADDR_OBSERVED;
> + struct inany_addr_entry *e;
> +
> + if (IN6_IS_ADDR_UNSPECIFIED(addr))
> + return;
> +
> + if (IN6_IS_ADDR_LINKLOCAL(addr))
> + flags |= INANY_ADDR_LINKLOCAL;
> +
> + for_each_addr(c, e, AF_INET6) {
> + if (IN6_ARE_ADDR_EQUAL(addr, &e->addr.a6)) {
> + e->flags |= flags;
> + return;
> + }
> + }
> +
> + if (c->addr_count < INANY_MAX_ADDRS) {
> + c->addrs[c->addr_count].addr.a6 = *addr;
> + c->addrs[c->addr_count].prefix_len = 0;
> + c->addrs[c->addr_count].flags = flags;
> + c->addr_count++;
> + }
Similar comments to v4:
* Can use a unified function for migration and non-migration updates
* Better to stick to 1 LL and 1 non-LL observed address for now, I think.
> +}
> +
> /**
> * seen_addrs_target_v2() - Receive observed addresses on target
> * @c: Execution context
> @@ -129,6 +174,7 @@ static int seen_addrs_target_v2(struct ctx *c,
> const struct migrate_stage *stage, int fd)
> {
> struct migrate_seen_addrs_v2 addrs;
> + struct in6_addr addr6, addr6_ll;
> struct in_addr addr4;
>
> (void)stage;
> @@ -136,12 +182,14 @@ static int seen_addrs_target_v2(struct ctx *c,
> if (read_all_buf(fd, &addrs, sizeof(addrs)))
> return errno;
>
> - c->ip6.addr_seen = addrs.addr6;
> - c->ip6.addr_ll_seen = addrs.addr6_ll;
> -
> - /* Avoid alignment warning */
> + /* Copy from packed struct to avoid alignment issues */
> addr4 = addrs.addr4;
> + addr6 = addrs.addr6;
> + addr6_ll = addrs.addr6_ll;
> +
> migrate_observed_addr4(c, &addr4);
> + migrate_observed_addr6(c, &addr6);
> + migrate_observed_addr6(c, &addr6_ll);
>
> memcpy(c->guest_mac, addrs.mac, sizeof(c->guest_mac));
>
> diff --git a/passt.h b/passt.h
> index 73084ae..7040781 100644
> --- a/passt.h
> +++ b/passt.h
> @@ -70,6 +70,7 @@ enum passt_modes {
> #define INANY_ADDR_CONFIGURED BIT(0) /* User set via -a */
> #define INANY_ADDR_HOST BIT(1) /* From host interface */
> #define INANY_ADDR_OBSERVED BIT(2) /* Seen in guest traffic */
> +#define INANY_ADDR_LINKLOCAL BIT(3) /* Link-local address */
Isn't this redundant with the contents of the address itself?
> /**
> * for_each_addr() - Iterate over addresses in unified array
> @@ -134,8 +135,6 @@ struct ip4_ctx {
>
> /**
> * struct ip6_ctx - IPv6 execution context
> - * @addr_seen: Latest IPv6 global/site address seen as source from tap
> - * @addr_ll_seen: Latest IPv6 link-local address seen as source from tap
> * @guest_gw: IPv6 gateway as seen by the guest
> * @map_host_loopback: Outbound connections to this address are NATted to the
> * host's [::1]
> @@ -151,8 +150,6 @@ struct ip4_ctx {
> * @no_copy_addrs: Don't copy all addresses when configuring namespace
> */
> struct ip6_ctx {
> - struct in6_addr addr_seen;
> - struct in6_addr addr_ll_seen;
> struct in6_addr guest_gw;
> struct in6_addr map_host_loopback;
> struct in6_addr map_guest_addr;
> diff --git a/pasta.c b/pasta.c
> index 8cb5873..7c2efd8 100644
> --- a/pasta.c
> +++ b/pasta.c
> @@ -350,12 +350,19 @@ static void pasta_ns_conf_ip4(struct ctx *c)
> */
> static void pasta_ns_conf_ip6(struct ctx *c)
> {
> + struct in6_addr addr_ll;
> int rc = 0;
>
> - rc = nl_addr_get_ll(nl_sock_ns, c->pasta_ifi, &c->ip6.addr_ll_seen);
> + rc = nl_addr_get_ll(nl_sock_ns, c->pasta_ifi, &addr_ll);
> if (rc < 0) {
> warn("Can't get LL address from namespace: %s",
> strerror_(-rc));
> + } else if (c->addr_count < INANY_MAX_ADDRS) {
> + c->addrs[c->addr_count].addr.a6 = addr_ll;
> + c->addrs[c->addr_count].prefix_len = 64;
> + c->addrs[c->addr_count].flags =
> + INANY_ADDR_OBSERVED | INANY_ADDR_LINKLOCAL;
> + c->addr_count++;
> }
>
> rc = nl_addr_set_ll_nodad(nl_sock_ns, c->pasta_ifi);
> @@ -373,6 +380,11 @@ static void pasta_ns_conf_ip6(struct ctx *c)
> for_each_addr(c, e, AF_INET6) {
> if (IN6_IS_ADDR_UNSPECIFIED(&e->addr.a6))
> continue;
> +
> + /* Skip link-local - kernel auto-configures */
> + if (e->flags & INANY_ADDR_LINKLOCAL)
> + continue;
> +
I'm not sure we need this. The kernel _will_ auto-configure, but if
the user has specified a specific one, we should probably use it.
Usually I wouldn't expect there to be an LL address in the array at
this point
> rc = nl_addr_set(nl_sock_ns, c->pasta_ifi,
> AF_INET6, &e->addr.a6, e->prefix_len);
> if (rc < 0)
> @@ -440,6 +452,7 @@ void pasta_ns_conf(struct ctx *c)
>
> if (c->ifi6)
> pasta_ns_conf_ip6(c);
> +
Unrelated whitespace change.
> }
>
> proto_update_l2_buf(c->guest_mac);
> diff --git a/tap.c b/tap.c
> index f2512f2..ba60f4b 100644
> --- a/tap.c
> +++ b/tap.c
> @@ -193,6 +193,41 @@ static void tap_check_src_addr4(struct ctx *c, const struct in_addr *addr)
> debug("Added new observed IPv4 address");
> }
>
> +/**
> + * tap_check_src_addr6() - Note an IPv6 address seen in guest traffic
> + * @c: Execution context
> + * @addr: IPv6 address seen as source from guest
> + *
> + * Add the address to addrs[] with OBSERVED flag if not already present.
> + * Link-local addresses are also marked with LINKLOCAL flag.
> + */
> +static void tap_check_src_addr6(struct ctx *c, const struct in6_addr *addr)
> +{
> + uint8_t flags = INANY_ADDR_OBSERVED;
> + struct inany_addr_entry *e;
> +
> + if (IN6_IS_ADDR_LINKLOCAL(addr))
> + flags |= INANY_ADDR_LINKLOCAL;
> +
> + /* Check if already in array */
> + for_each_addr(c, e, AF_INET6) {
> + if (IN6_ARE_ADDR_EQUAL(addr, &e->addr.a6)) {
> + e->flags |= flags;
> + return;
> + }
> + }
> +
> + /* Add new entry if space available */
> + if (c->addr_count < INANY_MAX_ADDRS) {
> + c->addrs[c->addr_count].addr.a6 = *addr;
> + c->addrs[c->addr_count].prefix_len = 0;
> + c->addrs[c->addr_count].flags = flags;
> + c->addr_count++;
> + } else {
> + warn("Address table full, can't add IPv6 address");
> + }
> +}
> +
> /**
> * tap_ip6_daddr() - Normal IPv6 destination address for inbound packets
> * @c: Execution context
> @@ -203,9 +238,33 @@ static void tap_check_src_addr4(struct ctx *c, const struct in_addr *addr)
> const struct in6_addr *tap_ip6_daddr(const struct ctx *c,
> const struct in6_addr *src)
> {
> - if (IN6_IS_ADDR_LINKLOCAL(src))
> - return &c->ip6.addr_ll_seen;
> - return &c->ip6.addr_seen;
> + bool want_ll = IN6_IS_ADDR_LINKLOCAL(src);
> + const struct inany_addr_entry *e;
> +
> + /* Find first observed address of matching scope */
> + for_each_addr(c, e, AF_INET6) {
> + bool is_ll = !!(e->flags & INANY_ADDR_LINKLOCAL);
> +
> + if (is_ll != want_ll)
> + continue;
> + if (e->flags & INANY_ADDR_OBSERVED)
> + return &e->addr.a6;
> + }
> +
> + /* Fallback to first address of matching scope */
> + for_each_addr(c, e, AF_INET6) {
> + bool is_ll = !!(e->flags & INANY_ADDR_LINKLOCAL);
> +
> + if (is_ll == want_ll)
> + return &e->addr.a6;
> + }
> +
> + /* Last resort: return first IPv6 address */
I think we have to match scope - an LL address is not meaningful to a
peer who's not in scope. So I think all you can do is return NULL at
this point.
> + e = first_v6(c);
> + if (e)
> + return &e->addr.a6;
> +
> + return &in6addr_any;
> }
>
> /**
> @@ -976,15 +1035,8 @@ resume:
> continue;
> }
>
> - if (IN6_IS_ADDR_LINKLOCAL(saddr)) {
> - c->ip6.addr_ll_seen = *saddr;
> -
> - if (IN6_IS_ADDR_UNSPECIFIED(&c->ip6.addr_seen)) {
> - c->ip6.addr_seen = *saddr;
> - }
> - } else if (!IN6_IS_ADDR_UNSPECIFIED(saddr)){
> - c->ip6.addr_seen = *saddr;
> - }
> + if (!IN6_IS_ADDR_UNSPECIFIED(saddr))
> + tap_check_src_addr6(c, saddr);
>
> if (proto == IPPROTO_ICMPV6) {
> struct iov_tail ndp_data;
> @@ -1015,7 +1067,7 @@ resume:
> if (proto == IPPROTO_UDP) {
> struct iov_tail uh_data = data;
>
> - if (dhcpv6(c, &uh_data, saddr, daddr))
> + if (dhcpv6(c, &uh_data, daddr))
> continue;
> }
>
> --
> 2.52.0
>
--
David Gibson (he or they) | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you, not the other way
| around.
http://www.ozlabs.org/~dgibson
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2026-02-09 22:30 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-30 21:44 [PATCH v3 00/11] Introduce multiple addresses Jon Maloy
2026-01-30 21:44 ` [PATCH v3 01/11] conf: Support CIDR notation for -a/--address option Jon Maloy
2026-02-04 12:50 ` David Gibson
2026-02-05 0:56 ` Jon Maloy
2026-02-06 3:26 ` David Gibson
2026-01-30 21:44 ` [PATCH v3 02/11] ip: Add IN4_MASK() macro for IPv4 netmask calculation Jon Maloy
2026-02-04 12:52 ` David Gibson
2026-01-30 21:44 ` [PATCH v3 03/11] ip: Introduce unified multi-address data structures Jon Maloy
2026-02-06 8:24 ` David Gibson
2026-01-30 21:44 ` [PATCH v3 04/11] fwd: Check all configured addresses in guest accessibility functions Jon Maloy
2026-02-04 13:16 ` David Gibson
2026-02-05 1:01 ` Jon Maloy
2026-02-06 3:29 ` David Gibson
2026-01-30 21:44 ` [PATCH v3 05/11] arp: Check all configured addresses in ARP filtering Jon Maloy
2026-02-06 8:34 ` David Gibson
2026-01-30 21:44 ` [PATCH v3 06/11] pasta: Extract pasta_ns_conf_ip4/6() to reduce nesting Jon Maloy
2026-02-06 8:40 ` David Gibson
2026-01-30 21:44 ` [PATCH v3 07/11] conf: Allow multiple -a/--address options per address family Jon Maloy
2026-02-06 8:47 ` David Gibson
2026-01-30 21:44 ` [PATCH v3 08/11] migrate: Rename v1 address functions to v2 for clarity Jon Maloy
2026-02-06 8:50 ` David Gibson
2026-01-30 21:44 ` [PATCH v3 09/11] ip: Track observed guest IPv4 addresses in unified address array Jon Maloy
2026-02-09 22:17 ` David Gibson
2026-01-30 21:44 ` [PATCH v3 10/11] ip: Track observed guest IPv6 " Jon Maloy
2026-02-09 22:30 ` David Gibson [this message]
2026-01-30 21:44 ` [PATCH v3 11/11] conf: Select addresses for DHCP and NDP distribution Jon Maloy
2026-02-09 22:46 ` David Gibson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=aYpf7TUtPkXafcoD@zatzit \
--to=david@gibson.dropbear.id.au \
--cc=dgibson@redhat.com \
--cc=jmaloy@redhat.com \
--cc=passt-dev@passt.top \
--cc=sbrivio@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://passt.top/passt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for IMAP folder(s).