From: Stefano Brivio <sbrivio@redhat.com>
To: David Gibson <david@gibson.dropbear.id.au>
Cc: passt-dev@passt.top, jmaloy@redhat.com
Subject: Re: [PATCH v7 19/27] fwd: Update flow forwarding logic for UDP
Date: Mon, 8 Jul 2024 23:26:55 +0200 [thread overview]
Message-ID: <20240708232655.76b43f59@elisabeth> (raw)
In-Reply-To: <20240705020724.3447719-20-david@gibson.dropbear.id.au>
On Fri, 5 Jul 2024 12:07:16 +1000
David Gibson <david@gibson.dropbear.id.au> wrote:
> Add logic to the fwd_nat_from_*() functions to forwarding UDP packets. The
> logic here doesn't exactly match our current forwarding, since our current
> forwarding has some very strange and buggy edge cases. Instead it's
> attempting to replicate what appears to be the intended logic behind the
> current forwarding.
>
> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
> ---
> fwd.c | 26 ++++++++++++++++++++++----
> 1 file changed, 22 insertions(+), 4 deletions(-)
>
> diff --git a/fwd.c b/fwd.c
> index 5731a536..4377de44 100644
> --- a/fwd.c
> +++ b/fwd.c
> @@ -169,12 +169,15 @@ void fwd_scan_ports_init(struct ctx *c)
> uint8_t fwd_nat_from_tap(const struct ctx *c, uint8_t proto,
> const struct flowside *ini, struct flowside *tgt)
> {
> - (void)proto;
> -
> tgt->eaddr = ini->faddr;
> tgt->eport = ini->fport;
>
> - if (!c->no_map_gw) {
> + if (proto == IPPROTO_UDP && tgt->eport == 53) {
> + if (inany_equals4(&tgt->eaddr, &c->ip4.dns_match))
> + tgt->eaddr = inany_from_v4(c->ip4.dns_host);
> + else if (inany_equals6(&tgt->eaddr, &c->ip6.dns_match))
> + tgt->eaddr.a6 = c->ip6.dns_host;
> + } else if (!c->no_map_gw) {
There's a subtle difference here compared to the logic you dropped in
23/27 (udp_tap_handler()), which doesn't look correct to me.
Earlier, with neither c->ip4.dns_match nor c->ip6.dns_match matching,
we would let UDP traffic directed to port 53 be mapped to the host, if
(!c->no_map_gw). That is, the logic was rather equivalent to this:
if (proto == IPPROTO_UDP && tgt->eport == 53 &&
(inany_equals4(&tgt->eaddr, &c->ip4.dns_match) ||
inany_equals6(&tgt->eaddr, &c->ip6.dns_match)) {
if (inany_equals4(&tgt->eaddr, &c->ip4.dns_match))
tgt->eaddr = inany_from_v4(c->ip4.dns_host);
else if (inany_equals6(&tgt->eaddr, &c->ip6.dns_match))
tgt->eaddr.a6 = c->ip6.dns_host;
} else if (!c->no_map_gw) {
...
and I think we should maintain it, because if dns_match doesn't match,
DNS traffic considerations shouldn't affect NAT decisions at all.
> if (inany_equals4(&tgt->eaddr, &c->ip4.gw))
> tgt->eaddr = inany_loopback4;
> else if (inany_equals6(&tgt->eaddr, &c->ip6.gw))
> @@ -191,6 +194,10 @@ uint8_t fwd_nat_from_tap(const struct ctx *c, uint8_t proto,
>
> /* Let the kernel pick a host side source port */
> tgt->fport = 0;
> + if (proto == IPPROTO_UDP) {
> + /* But for UDP we preserve the source port */
> + tgt->fport = ini->eport;
> + }
>
> return PIF_HOST;
> }
> @@ -232,9 +239,14 @@ uint8_t fwd_nat_from_splice(const struct ctx *c, uint8_t proto,
> tgt->eport = ini->fport;
> if (proto == IPPROTO_TCP)
> tgt->eport += c->tcp.fwd_out.delta[tgt->eport];
> + else if (proto == IPPROTO_UDP)
> + tgt->eport += c->udp.fwd_out.f.delta[tgt->eport];
>
> /* Let the kernel pick a host side source port */
> tgt->fport = 0;
> + if (proto == IPPROTO_UDP)
> + /* But for UDP preserve the source port */
> + tgt->fport = ini->eport;
>
> return PIF_HOST;
> }
> @@ -256,20 +268,26 @@ uint8_t fwd_nat_from_host(const struct ctx *c, uint8_t proto,
> tgt->eport = ini->fport;
> if (proto == IPPROTO_TCP)
> tgt->eport += c->tcp.fwd_in.delta[tgt->eport];
> + else if (proto == IPPROTO_UDP)
> + tgt->eport += c->udp.fwd_in.f.delta[tgt->eport];
>
> if (c->mode == MODE_PASTA && inany_is_loopback(&ini->eaddr) &&
> - proto == IPPROTO_TCP) {
> + (proto == IPPROTO_TCP || proto == IPPROTO_UDP)) {
> /* spliceable */
>
> /* Preserve the specific loopback adddress used, but let the
> * kernel pick a source port on the target side */
> tgt->faddr = ini->eaddr;
> tgt->fport = 0;
> + if (proto == IPPROTO_UDP)
> + /* But for UDP preserve the source port */
> + tgt->fport = ini->eport;
>
> if (inany_v4(&ini->eaddr))
> tgt->eaddr = inany_loopback4;
> else
> tgt->eaddr = inany_loopback6;
> +
> return PIF_SPLICE;
> }
>
--
Stefano
next prev parent reply other threads:[~2024-07-08 21:27 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-05 2:06 [PATCH v7 00/27] Unified flow table David Gibson
2024-07-05 2:06 ` [PATCH v7 01/27] flow: Common address information for initiating side David Gibson
2024-07-05 2:06 ` [PATCH v7 02/27] flow: Common address information for target side David Gibson
2024-07-10 21:30 ` Stefano Brivio
2024-07-11 0:19 ` David Gibson
2024-07-05 2:07 ` [PATCH v7 03/27] tcp, flow: Remove redundant information, repack connection structures David Gibson
2024-07-05 2:07 ` [PATCH v7 04/27] tcp: Obtain guest address from flowside David Gibson
2024-07-05 2:07 ` [PATCH v7 05/27] tcp: Manage outbound address via flow table David Gibson
2024-07-05 2:07 ` [PATCH v7 06/27] tcp: Simplify endpoint validation using flowside information David Gibson
2024-07-05 2:07 ` [PATCH v7 07/27] tcp_splice: Eliminate SPLICE_V6 flag David Gibson
2024-07-05 2:07 ` [PATCH v7 08/27] tcp, flow: Replace TCP specific hash function with general flow hash David Gibson
2024-07-05 2:07 ` [PATCH v7 09/27] flow, tcp: Generalise TCP hash table to general flow hash table David Gibson
2024-07-05 2:07 ` [PATCH v7 10/27] tcp: Re-use flow hash for initial sequence number generation David Gibson
2024-07-05 2:07 ` [PATCH v7 11/27] icmp: Remove redundant id field from flow table entry David Gibson
2024-07-05 2:07 ` [PATCH v7 12/27] icmp: Obtain destination addresses from the flowsides David Gibson
2024-07-05 2:07 ` [PATCH v7 13/27] icmp: Look up ping flows using flow hash David Gibson
2024-07-05 2:07 ` [PATCH v7 14/27] icmp: Eliminate icmp_id_map David Gibson
2024-07-05 2:07 ` [PATCH v7 15/27] flow: Helper to create sockets based on flowside David Gibson
2024-07-10 21:32 ` Stefano Brivio
2024-07-11 0:21 ` David Gibson
2024-07-11 0:27 ` David Gibson
2024-07-05 2:07 ` [PATCH v7 16/27] icmp: Manage outbound socket address via flow table David Gibson
2024-07-05 2:07 ` [PATCH v7 17/27] flow, tcp: Flow based NAT and port forwarding for TCP David Gibson
2024-07-05 2:07 ` [PATCH v7 18/27] flow, icmp: Use general flow forwarding rules for ICMP David Gibson
2024-07-05 2:07 ` [PATCH v7 19/27] fwd: Update flow forwarding logic for UDP David Gibson
2024-07-08 21:26 ` Stefano Brivio [this message]
2024-07-09 0:19 ` David Gibson
2024-07-05 2:07 ` [PATCH v7 20/27] udp: Create flows for datagrams from originating sockets David Gibson
2024-07-09 22:32 ` Stefano Brivio
2024-07-09 23:59 ` David Gibson
2024-07-10 21:35 ` Stefano Brivio
2024-07-11 4:26 ` David Gibson
2024-07-11 8:20 ` Stefano Brivio
2024-07-11 22:58 ` David Gibson
2024-07-12 8:21 ` Stefano Brivio
2024-07-15 4:06 ` David Gibson
2024-07-15 16:37 ` Stefano Brivio
2024-07-17 0:49 ` David Gibson
2024-07-05 2:07 ` [PATCH v7 21/27] udp: Handle "spliced" datagrams with per-flow sockets David Gibson
2024-07-09 22:32 ` Stefano Brivio
2024-07-10 0:23 ` David Gibson
2024-07-10 17:13 ` Stefano Brivio
2024-07-11 1:30 ` David Gibson
2024-07-11 8:23 ` Stefano Brivio
2024-07-11 2:48 ` David Gibson
2024-07-12 13:34 ` Stefano Brivio
2024-07-15 4:32 ` David Gibson
2024-07-05 2:07 ` [PATCH v7 22/27] udp: Remove obsolete splice tracking David Gibson
2024-07-10 21:36 ` Stefano Brivio
2024-07-11 0:43 ` David Gibson
2024-07-05 2:07 ` [PATCH v7 23/27] udp: Find or create flows for datagrams from tap interface David Gibson
2024-07-10 21:36 ` Stefano Brivio
2024-07-11 0:45 ` David Gibson
2024-07-05 2:07 ` [PATCH v7 24/27] udp: Direct datagrams from host to guest via flow table David Gibson
2024-07-10 21:37 ` Stefano Brivio
2024-07-11 0:46 ` David Gibson
2024-07-05 2:07 ` [PATCH v7 25/27] udp: Remove obsolete socket tracking David Gibson
2024-07-05 2:07 ` [PATCH v7 26/27] udp: Remove rdelta port forwarding maps David Gibson
2024-07-05 2:07 ` [PATCH v7 27/27] udp: Rename UDP listening sockets 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=20240708232655.76b43f59@elisabeth \
--to=sbrivio@redhat.com \
--cc=david@gibson.dropbear.id.au \
--cc=jmaloy@redhat.com \
--cc=passt-dev@passt.top \
/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).