| Message ID | 20230124140207.3975283-7-vladbu@nvidia.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | Allow offloading of UDP NEW connections via act_ct | expand |
On Tue, 24 Jan 2023 15:02:06 +0100 Vlad Buslov wrote: > Modify the offload algorithm of UDP connections to the following: > > - Offload NEW connection as unidirectional. > > - When connection state changes to ESTABLISHED also update the hardware > flow. However, in order to prevent act_ct from spamming offload add wq for > every packet coming in reply direction in this state verify whether > connection has already been updated to ESTABLISHED in the drivers. If that > it the case, then skip flow_table and let conntrack handle such packets > which will also allow conntrack to potentially promote the connection to > ASSURED. > > - When connection state changes to ASSURED set the flow_table flow > NF_FLOW_HW_BIDIRECTIONAL flag which will cause refresh mechanism to offload > the reply direction. > > All other protocols have their offload algorithm preserved and are always > offloaded as bidirectional. > > Note that this change tries to minimize the load on flow_table add > workqueue. First, it tracks the last ctinfo that was offloaded by using new > flow 'ext_data' field and doesn't schedule the refresh for reply direction > packets when the offloads have already been updated with current ctinfo. > Second, when 'add' task executes on workqueue it always update the offload > with current flow state (by checking 'bidirectional' flow flag and > obtaining actual ctinfo/cookie through meta action instead of caching any > of these from the moment of scheduling the 'add' work) preventing the need > from scheduling more updates if state changed concurrently while the 'add' > work was pending on workqueue. Clang is not happy: net/sched/act_ct.c:677:12: warning: cast to smaller integer type 'enum ip_conntrack_info' from 'typeof (_Generic((flow->ext_data), char: (char)0, unsigned char: (unsigned char)0, signed char: (signed char)0, unsigned short: (unsigned short)0, short: (short)0, unsigned int: (unsigned int)0, int: (int)0, unsigned long: (unsigned long)0, long: (long)0, unsigned long long: (unsigned long long)0, long long: (long long)0, default: (flow->ext_data)))' (aka 'void *') [-Wvoid-pointer-to-enum-cast] else if ((enum ip_conntrack_info)READ_ONCE(flow->ext_data) == ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
On Tue 24 Jan 2023 at 21:42, Jakub Kicinski <kuba@kernel.org> wrote: > On Tue, 24 Jan 2023 15:02:06 +0100 Vlad Buslov wrote: >> Modify the offload algorithm of UDP connections to the following: >> >> - Offload NEW connection as unidirectional. >> >> - When connection state changes to ESTABLISHED also update the hardware >> flow. However, in order to prevent act_ct from spamming offload add wq for >> every packet coming in reply direction in this state verify whether >> connection has already been updated to ESTABLISHED in the drivers. If that >> it the case, then skip flow_table and let conntrack handle such packets >> which will also allow conntrack to potentially promote the connection to >> ASSURED. >> >> - When connection state changes to ASSURED set the flow_table flow >> NF_FLOW_HW_BIDIRECTIONAL flag which will cause refresh mechanism to offload >> the reply direction. >> >> All other protocols have their offload algorithm preserved and are always >> offloaded as bidirectional. >> >> Note that this change tries to minimize the load on flow_table add >> workqueue. First, it tracks the last ctinfo that was offloaded by using new >> flow 'ext_data' field and doesn't schedule the refresh for reply direction >> packets when the offloads have already been updated with current ctinfo. >> Second, when 'add' task executes on workqueue it always update the offload >> with current flow state (by checking 'bidirectional' flow flag and >> obtaining actual ctinfo/cookie through meta action instead of caching any >> of these from the moment of scheduling the 'add' work) preventing the need >> from scheduling more updates if state changed concurrently while the 'add' >> work was pending on workqueue. > > Clang is not happy: > > net/sched/act_ct.c:677:12: warning: cast to smaller integer type 'enum ip_conntrack_info' from 'typeof (_Generic((flow->ext_data), char: (char)0, unsigned char: (unsigned char)0, signed char: (signed char)0, unsigned short: (unsigned short)0, short: (short)0, unsigned int: (unsigned int)0, int: (int)0, unsigned long: (unsigned long)0, long: (long)0, unsigned long long: (unsigned long long)0, long long: (long long)0, default: (flow->ext_data)))' (aka 'void *') [-Wvoid-pointer-to-enum-cast] > else if ((enum ip_conntrack_info)READ_ONCE(flow->ext_data) == > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Thanks, will fix it in v5.
diff --git a/net/sched/act_ct.c b/net/sched/act_ct.c index 2b81a7898662..355ae48bd371 100644 --- a/net/sched/act_ct.c +++ b/net/sched/act_ct.c @@ -401,7 +401,7 @@ static void tcf_ct_flow_tc_ifidx(struct flow_offload *entry, static void tcf_ct_flow_table_add(struct tcf_ct_flow_table *ct_ft, struct nf_conn *ct, - bool tcp) + bool tcp, bool bidirectional) { struct nf_conn_act_ct_ext *act_ct_ext; struct flow_offload *entry; @@ -420,6 +420,8 @@ static void tcf_ct_flow_table_add(struct tcf_ct_flow_table *ct_ft, ct->proto.tcp.seen[0].flags |= IP_CT_TCP_FLAG_BE_LIBERAL; ct->proto.tcp.seen[1].flags |= IP_CT_TCP_FLAG_BE_LIBERAL; } + if (bidirectional) + __set_bit(NF_FLOW_HW_BIDIRECTIONAL, &entry->flags); act_ct_ext = nf_conn_act_ct_ext_find(ct); if (act_ct_ext) { @@ -443,26 +445,34 @@ static void tcf_ct_flow_table_process_conn(struct tcf_ct_flow_table *ct_ft, struct nf_conn *ct, enum ip_conntrack_info ctinfo) { - bool tcp = false; - - if ((ctinfo != IP_CT_ESTABLISHED && ctinfo != IP_CT_ESTABLISHED_REPLY) || - !test_bit(IPS_ASSURED_BIT, &ct->status)) - return; + bool tcp = false, bidirectional = true; switch (nf_ct_protonum(ct)) { case IPPROTO_TCP: - tcp = true; - if (ct->proto.tcp.state != TCP_CONNTRACK_ESTABLISHED) + if ((ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED_REPLY) || + !test_bit(IPS_ASSURED_BIT, &ct->status) || + ct->proto.tcp.state != TCP_CONNTRACK_ESTABLISHED) return; + + tcp = true; break; case IPPROTO_UDP: + if (!nf_ct_is_confirmed(ct)) + return; + if (!test_bit(IPS_ASSURED_BIT, &ct->status)) + bidirectional = false; break; #ifdef CONFIG_NF_CT_PROTO_GRE case IPPROTO_GRE: { struct nf_conntrack_tuple *tuple; - if (ct->status & IPS_NAT_MASK) + if ((ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED_REPLY) || + !test_bit(IPS_ASSURED_BIT, &ct->status) || + ct->status & IPS_NAT_MASK) return; + tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; /* No support for GRE v1 */ if (tuple->src.u.gre.key || tuple->dst.u.gre.key) @@ -478,7 +488,7 @@ static void tcf_ct_flow_table_process_conn(struct tcf_ct_flow_table *ct_ft, ct->status & IPS_SEQ_ADJUST) return; - tcf_ct_flow_table_add(ct_ft, ct, tcp); + tcf_ct_flow_table_add(ct_ft, ct, tcp, bidirectional); } static bool @@ -657,13 +667,31 @@ static bool tcf_ct_flow_table_lookup(struct tcf_ct_params *p, flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]); ct = flow->ct; + if (dir == FLOW_OFFLOAD_DIR_REPLY && + !test_bit(NF_FLOW_HW_BIDIRECTIONAL, &flow->flags)) { + /* Only offload reply direction after connection became + * assured. + */ + if (test_bit(IPS_ASSURED_BIT, &ct->status)) + set_bit(NF_FLOW_HW_BIDIRECTIONAL, &flow->flags); + else if ((enum ip_conntrack_info)READ_ONCE(flow->ext_data) == + IP_CT_ESTABLISHED) + /* If flow_table flow has already been updated to the + * established state, then don't refresh. + */ + return false; + } + if (tcph && (unlikely(tcph->fin || tcph->rst))) { flow_offload_teardown(nf_ft, flow); return false; } - ctinfo = dir == FLOW_OFFLOAD_DIR_ORIGINAL ? IP_CT_ESTABLISHED : - IP_CT_ESTABLISHED_REPLY; + if (dir == FLOW_OFFLOAD_DIR_ORIGINAL) + ctinfo = test_bit(IPS_SEEN_REPLY_BIT, &ct->status) ? + IP_CT_ESTABLISHED : IP_CT_NEW; + else + ctinfo = IP_CT_ESTABLISHED_REPLY; flow_offload_refresh(nf_ft, flow); nf_conntrack_get(&ct->ct_general);
Modify the offload algorithm of UDP connections to the following: - Offload NEW connection as unidirectional. - When connection state changes to ESTABLISHED also update the hardware flow. However, in order to prevent act_ct from spamming offload add wq for every packet coming in reply direction in this state verify whether connection has already been updated to ESTABLISHED in the drivers. If that it the case, then skip flow_table and let conntrack handle such packets which will also allow conntrack to potentially promote the connection to ASSURED. - When connection state changes to ASSURED set the flow_table flow NF_FLOW_HW_BIDIRECTIONAL flag which will cause refresh mechanism to offload the reply direction. All other protocols have their offload algorithm preserved and are always offloaded as bidirectional. Note that this change tries to minimize the load on flow_table add workqueue. First, it tracks the last ctinfo that was offloaded by using new flow 'ext_data' field and doesn't schedule the refresh for reply direction packets when the offloads have already been updated with current ctinfo. Second, when 'add' task executes on workqueue it always update the offload with current flow state (by checking 'bidirectional' flow flag and obtaining actual ctinfo/cookie through meta action instead of caching any of these from the moment of scheduling the 'add' work) preventing the need from scheduling more updates if state changed concurrently while the 'add' work was pending on workqueue. Signed-off-by: Vlad Buslov <vladbu@nvidia.com> --- Notes: Changes V3 -> V4: - Refactor the patch to leverage the refresh code and new flow 'ext_data' field in order to change the offload state instead of relying on async gc update. net/sched/act_ct.c | 52 +++++++++++++++++++++++++++++++++++----------- 1 file changed, 40 insertions(+), 12 deletions(-)