| Message ID | 20240311183231.37253-1-i.maximets@ovn.org |
|---|---|
| State | Accepted |
| Commit | 7df30c86ce12833cf7f9bfc71c166c34692ceff4 |
| Headers | show |
| Series | [ovs-dev] netdev-dpdk: Clean up all marker flags if no offloads requested. | expand |
| Context | Check | Description |
|---|---|---|
| ovsrobot/apply-robot | success | apply and check: success |
| ovsrobot/github-robot-_Build_and_Test | success | github build: passed |
| ovsrobot/intel-ovs-compilation | success | test: success |
On Mon, Mar 11, 2024 at 2:31 PM Ilya Maximets <i.maximets@ovn.org> wrote: > > Some drivers (primarily, Intel ones) do not expect any marking flags > being set if no offloads are requested. If these flags are present, > driver will fail Tx preparation or behave abnormally. > > For example, ixgbe driver will refuse to process the packet with > only RTE_MBUF_F_TX_TUNNEL_GENEVE and RTE_MBUF_F_TX_OUTER_IPV4 set. > This pretty much breaks Geneve tunnels on these cards. > > An extra check is added to make sure we don't have any unexpected > Tx offload flags set. > > Fixes: 084c8087292c ("userspace: Support VXLAN and GENEVE TSO.") > Reported-at: https://github.com/openvswitch/ovs-issues/issues/321 > Signed-off-by: Ilya Maximets <i.maximets@ovn.org> Acked-by: Mike Pattrick <mkp@redhat.com>
On 3/12/24 14:36, Mike Pattrick wrote: > On Mon, Mar 11, 2024 at 2:31 PM Ilya Maximets <i.maximets@ovn.org> wrote: >> >> Some drivers (primarily, Intel ones) do not expect any marking flags >> being set if no offloads are requested. If these flags are present, >> driver will fail Tx preparation or behave abnormally. >> >> For example, ixgbe driver will refuse to process the packet with >> only RTE_MBUF_F_TX_TUNNEL_GENEVE and RTE_MBUF_F_TX_OUTER_IPV4 set. >> This pretty much breaks Geneve tunnels on these cards. >> >> An extra check is added to make sure we don't have any unexpected >> Tx offload flags set. >> >> Fixes: 084c8087292c ("userspace: Support VXLAN and GENEVE TSO.") >> Reported-at: https://github.com/openvswitch/ovs-issues/issues/321 >> Signed-off-by: Ilya Maximets <i.maximets@ovn.org> > > Acked-by: Mike Pattrick <mkp@redhat.com> > Thanks! Applied and backported to 3.3. Best regards, Ilya Maximets.
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c index 9444c53b1..8c52accff 100644 --- a/lib/netdev-dpdk.c +++ b/lib/netdev-dpdk.c @@ -607,6 +607,9 @@ int netdev_dpdk_get_vid(const struct netdev_dpdk *dev); struct ingress_policer * netdev_dpdk_get_ingress_policer(const struct netdev_dpdk *dev); +static void netdev_dpdk_mbuf_dump(const char *prefix, const char *message, + const struct rte_mbuf *); + static bool is_dpdk_class(const struct netdev_class *class) { @@ -2569,9 +2572,29 @@ netdev_dpdk_prep_hwol_packet(struct netdev_dpdk *dev, struct rte_mbuf *mbuf) struct dp_packet *pkt = CONTAINER_OF(mbuf, struct dp_packet, mbuf); struct tcp_header *th; - if (!(mbuf->ol_flags & (RTE_MBUF_F_TX_IP_CKSUM | RTE_MBUF_F_TX_L4_MASK - | RTE_MBUF_F_TX_TCP_SEG))) { - mbuf->ol_flags &= ~(RTE_MBUF_F_TX_IPV4 | RTE_MBUF_F_TX_IPV6); + const uint64_t all_requests = (RTE_MBUF_F_TX_IP_CKSUM | + RTE_MBUF_F_TX_L4_MASK | + RTE_MBUF_F_TX_OUTER_IP_CKSUM | + RTE_MBUF_F_TX_OUTER_UDP_CKSUM | + RTE_MBUF_F_TX_TCP_SEG); + const uint64_t all_marks = (RTE_MBUF_F_TX_IPV4 | + RTE_MBUF_F_TX_IPV6 | + RTE_MBUF_F_TX_OUTER_IPV4 | + RTE_MBUF_F_TX_OUTER_IPV6 | + RTE_MBUF_F_TX_TUNNEL_MASK); + + if (!(mbuf->ol_flags & all_requests)) { + /* No offloads requested, no marks should be set. */ + mbuf->ol_flags &= ~all_marks; + + uint64_t unexpected = mbuf->ol_flags & RTE_MBUF_F_TX_OFFLOAD_MASK; + if (OVS_UNLIKELY(unexpected)) { + VLOG_WARN_RL(&rl, "%s: Unexpected Tx offload flags: %#"PRIx64, + netdev_get_name(&dev->up), unexpected); + netdev_dpdk_mbuf_dump(netdev_get_name(&dev->up), + "Packet with unexpected ol_flags", mbuf); + return false; + } return true; }
Some drivers (primarily, Intel ones) do not expect any marking flags being set if no offloads are requested. If these flags are present, driver will fail Tx preparation or behave abnormally. For example, ixgbe driver will refuse to process the packet with only RTE_MBUF_F_TX_TUNNEL_GENEVE and RTE_MBUF_F_TX_OUTER_IPV4 set. This pretty much breaks Geneve tunnels on these cards. An extra check is added to make sure we don't have any unexpected Tx offload flags set. Fixes: 084c8087292c ("userspace: Support VXLAN and GENEVE TSO.") Reported-at: https://github.com/openvswitch/ovs-issues/issues/321 Signed-off-by: Ilya Maximets <i.maximets@ovn.org> --- lib/netdev-dpdk.c | 29 ++++++++++++++++++++++++++--- 1 file changed, 26 insertions(+), 3 deletions(-)