From: Ben Pfaff <blp@nicira.com>
To: Jarno Rajahalme <jrajahalme@nicira.com>
Cc: Simon Horman <horms@verge.net.au>,
"dev@openvswitch.org" <dev@openvswitch.org>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
Jesse Gross <jesse@nicira.com>, Ravi K <rkerur@gmail.com>
Subject: Re: [ovs-dev] [PATCH v2.50 1/6] ofp-actions: Consider L4 actions after mpls_push as inconsistent
Date: Fri, 15 Nov 2013 09:45:30 -0800 [thread overview]
Message-ID: <20131115174530.GE18635@nicira.com> (raw)
In-Reply-To: <6A09EE11-CDFA-4E49-AB7F-34746ED4765D@nicira.com>
Thanks Simon, thanks Jarno, applied.
On Fri, Nov 15, 2013 at 08:55:42AM -0800, Jarno Rajahalme wrote:
> Looks good to me,
>
> Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
>
> > On Nov 15, 2013, at 1:10 AM, Simon Horman <horms@verge.net.au> wrote:
> >
> > After an mpls_push action the resulting packet is MPLS and
> > the MPLS payload is opaque. Thus nothing can be assumed
> > about the packets network protocol and it is inconsistent
> > to apply L4 actions.
> >
> > With regards to actions that affect the packet at other layers
> > of the protocol stack:
> >
> > * L3: The consistency of L3 actions should already be handled correctly
> > by virtue of the dl_type of the flow being temporarily altered
> > during consistency checking by both push_mpls and pop_mpls actions.
> >
> > * MPLS: The consistency checking of MPLS actions appear to already be
> > handled correctly.
> >
> > * VLAN: At this time Open vSwitch on mpls_push an MPLS LSE is always
> > added after any VLAN tags that follow the ethernet header.
> > That is the tag ordering defined prior to OpenFlow1.3. As such
> > VLAN actions should sill be equally valid before and after mpls_push
> > and mpls_pop actions.
> >
> > * L2 actions are equally valid before and after mpls_push and mpls_pop actions.
> >
> > Signed-off-by: Simon Horman <horms@verge.net.au>
> >
> > ---
> > v2.50
> > * Correct typo in comment. nw_protocol -> nw_proto
> >
> > v2.49
> > * First post
> > ---
> > lib/ofp-actions.c | 9 ++++++++-
> > tests/ofp-actions.at | 12 ++++++++++++
> > 2 files changed, 20 insertions(+), 1 deletion(-)
> >
> > diff --git a/lib/ofp-actions.c b/lib/ofp-actions.c
> > index abc9505..03ad1a4 100644
> > --- a/lib/ofp-actions.c
> > +++ b/lib/ofp-actions.c
> > @@ -1869,7 +1869,8 @@ ofpact_check_output_port(ofp_port_t port, ofp_port_t max_ports)
> > }
> > }
> >
> > -/* May modify flow->dl_type and flow->vlan_tci, caller must restore them.
> > +/* May modify flow->dl_type, flow->nw_proto and flow->vlan_tci,
> > + * caller must restore them.
> > *
> > * Modifies some actions, filling in fields that could not be properly set
> > * without context. */
> > @@ -2056,6 +2057,10 @@ ofpact_check__(struct ofpact *a, struct flow *flow,
> >
> > case OFPACT_PUSH_MPLS:
> > flow->dl_type = ofpact_get_PUSH_MPLS(a)->ethertype;
> > + /* The packet is now MPLS and the MPLS payload is opaque.
> > + * Thus nothing can be assumed about the network protocol.
> > + * Temporarily mark that we have no nw_proto. */
> > + flow->nw_proto = 0;
> > return 0;
> >
> > case OFPACT_POP_MPLS:
> > @@ -2127,6 +2132,7 @@ ofpacts_check(struct ofpact ofpacts[], size_t ofpacts_len,
> > struct ofpact *a;
> > ovs_be16 dl_type = flow->dl_type;
> > ovs_be16 vlan_tci = flow->vlan_tci;
> > + uint8_t nw_proto = flow->nw_proto;
> > enum ofperr error = 0;
> >
> > OFPACT_FOR_EACH (a, ofpacts, ofpacts_len) {
> > @@ -2139,6 +2145,7 @@ ofpacts_check(struct ofpact ofpacts[], size_t ofpacts_len,
> > /* Restore fields that may have been modified. */
> > flow->dl_type = dl_type;
> > flow->vlan_tci = vlan_tci;
> > + flow->nw_proto = nw_proto;
> > return error;
> > }
> >
> > diff --git a/tests/ofp-actions.at b/tests/ofp-actions.at
> > index cdca8ca..08ebccf 100644
> > --- a/tests/ofp-actions.at
> > +++ b/tests/ofp-actions.at
> > @@ -477,3 +477,15 @@ AT_CHECK(
> > [ovs-ofctl '-vPATTERN:console:%c|%p|%m' parse-ofp11-instructions < input.txt],
> > [0], [expout], [experr])
> > AT_CLEANUP
> > +
> > +AT_SETUP([ofp-actions - inconsistent MPLS actions])
> > +OVS_VSWITCHD_START
> > +dnl OK: Use fin_timeout action on TCP flow
> > +AT_CHECK([ovs-ofctl -O OpenFlow11 -vwarn add-flow br0 'tcp actions=fin_timeout(idle_timeout=1)'])
> > +dnl Bad: Use fin_timeout action on TCP flow that has been converted to MPLS
> > +AT_CHECK([ovs-ofctl -O OpenFlow11 -vwarn add-flow br0 'tcp actions=push_mpls:0x8847,fin_timeout(idle_timeout=1)'],
> > + [1], [], [dnl
> > +ovs-ofctl: actions are invalid with specified match (OFPBAC_MATCH_INCONSISTENT)
> > +])
> > +OVS_VSWITCHD_STOP
> > +AT_CLEANUP
> > --
> > 1.8.4
> >
> > _______________________________________________
> > dev mailing list
> > dev@openvswitch.org
> > http://openvswitch.org/mailman/listinfo/dev
next prev parent reply other threads:[~2013-11-15 17:45 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-15 9:10 [PATCH v2.50 0/6] MPLS actions and matches Simon Horman
2013-11-15 9:10 ` [PATCH v2.50 2/6] ofp-actions: Account for OF version when enforcing action consistency Simon Horman
2013-11-15 22:34 ` Ben Pfaff
[not found] ` <1384506623-18503-1-git-send-email-horms-/R6kz+dDXgpPR4JQBCEnsQ@public.gmane.org>
2013-11-15 9:10 ` [PATCH v2.50 1/6] ofp-actions: Consider L4 actions after mpls_push as inconsistent Simon Horman
[not found] ` <1384506623-18503-2-git-send-email-horms-/R6kz+dDXgpPR4JQBCEnsQ@public.gmane.org>
2013-11-15 16:55 ` Jarno Rajahalme
2013-11-15 17:45 ` Ben Pfaff [this message]
2013-11-18 6:38 ` [ovs-dev] " Simon Horman
2013-11-15 9:10 ` [PATCH v2.50 3/6] odp: Allow VLAN actions after MPLS actions Simon Horman
2013-11-15 9:10 ` [PATCH v2.50 4/6] lib: Support pushing of MPLS LSE before or after VLAN tag Simon Horman
2013-11-15 9:10 ` [PATCH v2.50 5/6] datapath: Break out deacceleration portion of vlan_push Simon Horman
2013-11-15 9:10 ` [PATCH v2.50 6/6] datapath: Add basic MPLS support to kernel Simon Horman
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=20131115174530.GE18635@nicira.com \
--to=blp@nicira.com \
--cc=dev@openvswitch.org \
--cc=horms@verge.net.au \
--cc=jesse@nicira.com \
--cc=jrajahalme@nicira.com \
--cc=netdev@vger.kernel.org \
--cc=rkerur@gmail.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.
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 NNTP newsgroup(s).