diff mbox series

[net-next,v5,1/6] net/sched: cls_api: Support hardware miss to tc action

Message ID 20230125153218.7230-2-paulb@nvidia.com (mailing list archive)
State Superseded
Delegated to: Netdev Maintainers
Headers show
Series net/sched: cls_api: Support hardware miss to tc action | expand

Checks

Context Check Description
netdev/tree_selection success Clearly marked for net-next, async
netdev/fixes_present success Fixes tag not required for -next series
netdev/subject_prefix success Link
netdev/cover_letter success Series has a cover letter
netdev/patch_count success Link
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 5334 this patch: 5334
netdev/cc_maintainers warning 4 maintainers not CCed: dsahern@kernel.org pshelar@ovn.org dev@openvswitch.org imagedong@tencent.com
netdev/build_clang success Errors and warnings before: 1085 this patch: 1085
netdev/module_param success Was 0 now: 0
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success No Fixes tag
netdev/build_allmodconfig_warn success Errors and warnings before: 5548 this patch: 5548
netdev/checkpatch warning WARNING: line length of 81 exceeds 80 columns WARNING: line length of 84 exceeds 80 columns WARNING: line length of 85 exceeds 80 columns WARNING: void function return statements are not generally useful
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline fail Was 0 now: 2

Commit Message

Paul Blakey Jan. 25, 2023, 3:32 p.m. UTC
For drivers to support partial offload of a filter's action list,
add support for action miss to specify an action instance to
continue from in sw.

CT action in particular can't be fully offloaded, as new connections
need to be handled in software. This imposes other limitations on
the actions that can be offloaded together with the CT action, such
as packet modifications.

Assign each action on a filter's action list a unique miss_cookie
which drivers can then use to fill action_miss part of the tc skb
extension. On getting back this miss_cookie, find the action
instance with relevant cookie and continue classifying from there.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
---
 include/linux/skbuff.h     |   6 +-
 include/net/flow_offload.h |   1 +
 include/net/pkt_cls.h      |  34 +++---
 include/net/sch_generic.h  |   2 +
 net/openvswitch/flow.c     |   2 +-
 net/sched/act_api.c        |   2 +-
 net/sched/cls_api.c        | 214 +++++++++++++++++++++++++++++++++++--
 7 files changed, 234 insertions(+), 27 deletions(-)

Comments

Jamal Hadi Salim Jan. 28, 2023, 1:56 p.m. UTC | #1
For the cls parts:
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>

cheers,
jamal

On Wed, Jan 25, 2023 at 10:32 AM Paul Blakey <paulb@nvidia.com> wrote:
>
> For drivers to support partial offload of a filter's action list,
> add support for action miss to specify an action instance to
> continue from in sw.
>
> CT action in particular can't be fully offloaded, as new connections
> need to be handled in software. This imposes other limitations on
> the actions that can be offloaded together with the CT action, such
> as packet modifications.
>
> Assign each action on a filter's action list a unique miss_cookie
> which drivers can then use to fill action_miss part of the tc skb
> extension. On getting back this miss_cookie, find the action
> instance with relevant cookie and continue classifying from there.
>
> Signed-off-by: Paul Blakey <paulb@nvidia.com>
> Reviewed-by: Jiri Pirko <jiri@nvidia.com>
> ---
>  include/linux/skbuff.h     |   6 +-
>  include/net/flow_offload.h |   1 +
>  include/net/pkt_cls.h      |  34 +++---
>  include/net/sch_generic.h  |   2 +
>  net/openvswitch/flow.c     |   2 +-
>  net/sched/act_api.c        |   2 +-
>  net/sched/cls_api.c        | 214 +++++++++++++++++++++++++++++++++++--
>  7 files changed, 234 insertions(+), 27 deletions(-)
>
> diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
> index 4c8492401a10..348673dcb6bb 100644
> --- a/include/linux/skbuff.h
> +++ b/include/linux/skbuff.h
> @@ -316,12 +316,16 @@ struct nf_bridge_info {
>   * and read by ovs to recirc_id.
>   */
>  struct tc_skb_ext {
> -       __u32 chain;
> +       union {
> +               u64 act_miss_cookie;
> +               __u32 chain;
> +       };
>         __u16 mru;
>         __u16 zone;
>         u8 post_ct:1;
>         u8 post_ct_snat:1;
>         u8 post_ct_dnat:1;
> +       u8 act_miss:1; /* Set if act_miss_cookie is used */
>  };
>  #endif
>
> diff --git a/include/net/flow_offload.h b/include/net/flow_offload.h
> index 0400a0ac8a29..88db7346eb7a 100644
> --- a/include/net/flow_offload.h
> +++ b/include/net/flow_offload.h
> @@ -228,6 +228,7 @@ void flow_action_cookie_destroy(struct flow_action_cookie *cookie);
>  struct flow_action_entry {
>         enum flow_action_id             id;
>         u32                             hw_index;
> +       u64                             miss_cookie;
>         enum flow_action_hw_stats       hw_stats;
>         action_destr                    destructor;
>         void                            *destructor_priv;
> diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
> index 4cabb32a2ad9..344f2ca02ebd 100644
> --- a/include/net/pkt_cls.h
> +++ b/include/net/pkt_cls.h
> @@ -59,6 +59,8 @@ int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
>  void tcf_block_put(struct tcf_block *block);
>  void tcf_block_put_ext(struct tcf_block *block, struct Qdisc *q,
>                        struct tcf_block_ext_info *ei);
> +int tcf_exts_init_ex(struct tcf_exts *exts, struct net *net, int action, int police,
> +                    struct tcf_proto *tp, u32 handle, bool used_action_miss);
>
>  static inline bool tcf_block_shared(struct tcf_block *block)
>  {
> @@ -229,6 +231,7 @@ struct tcf_exts {
>         struct tc_action **actions;
>         struct net      *net;
>         netns_tracker   ns_tracker;
> +       struct tcf_exts_miss_cookie_node *miss_cookie_node;
>  #endif
>         /* Map to export classifier specific extension TLV types to the
>          * generic extensions API. Unsupported extensions must be set to 0.
> @@ -240,21 +243,11 @@ struct tcf_exts {
>  static inline int tcf_exts_init(struct tcf_exts *exts, struct net *net,
>                                 int action, int police)
>  {
> -#ifdef CONFIG_NET_CLS_ACT
> -       exts->type = 0;
> -       exts->nr_actions = 0;
> -       /* Note: we do not own yet a reference on net.
> -        * This reference might be taken later from tcf_exts_get_net().
> -        */
> -       exts->net = net;
> -       exts->actions = kcalloc(TCA_ACT_MAX_PRIO, sizeof(struct tc_action *),
> -                               GFP_KERNEL);
> -       if (!exts->actions)
> -               return -ENOMEM;
> +#ifdef CONFIG_NET_CLS
> +       return tcf_exts_init_ex(exts, net, action, police, NULL, 0, false);
> +#else
> +       return -EOPNOTSUPP;
>  #endif
> -       exts->action = action;
> -       exts->police = police;
> -       return 0;
>  }
>
>  /* Return false if the netns is being destroyed in cleanup_net(). Callers
> @@ -353,6 +346,18 @@ tcf_exts_exec(struct sk_buff *skb, struct tcf_exts *exts,
>         return TC_ACT_OK;
>  }
>
> +static inline int
> +tcf_exts_exec_ex(struct sk_buff *skb, struct tcf_exts *exts, int act_index,
> +                struct tcf_result *res)
> +{
> +#ifdef CONFIG_NET_CLS_ACT
> +       return tcf_action_exec(skb, exts->actions + act_index,
> +                              exts->nr_actions - act_index, res);
> +#else
> +       return TC_ACT_OK;
> +#endif
> +}
> +
>  int tcf_exts_validate(struct net *net, struct tcf_proto *tp,
>                       struct nlattr **tb, struct nlattr *rate_tlv,
>                       struct tcf_exts *exts, u32 flags,
> @@ -577,6 +582,7 @@ int tc_setup_offload_action(struct flow_action *flow_action,
>  void tc_cleanup_offload_action(struct flow_action *flow_action);
>  int tc_setup_action(struct flow_action *flow_action,
>                     struct tc_action *actions[],
> +                   u32 miss_cookie_base,
>                     struct netlink_ext_ack *extack);
>
>  int tc_setup_cb_call(struct tcf_block *block, enum tc_setup_type type,
> diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
> index af4aa66aaa4e..fab5ba3e61b7 100644
> --- a/include/net/sch_generic.h
> +++ b/include/net/sch_generic.h
> @@ -369,6 +369,8 @@ struct tcf_proto_ops {
>                                                 struct nlattr **tca,
>                                                 struct netlink_ext_ack *extack);
>         void                    (*tmplt_destroy)(void *tmplt_priv);
> +       struct tcf_exts *       (*get_exts)(const struct tcf_proto *tp,
> +                                           u32 handle);
>
>         /* rtnetlink specific */
>         int                     (*dump)(struct net*, struct tcf_proto*, void *,
> diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c
> index e20d1a973417..b1a5eed8d1a9 100644
> --- a/net/openvswitch/flow.c
> +++ b/net/openvswitch/flow.c
> @@ -1038,7 +1038,7 @@ int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
>  #if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
>         if (tc_skb_ext_tc_enabled()) {
>                 tc_ext = skb_ext_find(skb, TC_SKB_EXT);
> -               key->recirc_id = tc_ext ? tc_ext->chain : 0;
> +               key->recirc_id = tc_ext && !tc_ext->act_miss ? tc_ext->chain : 0;
>                 OVS_CB(skb)->mru = tc_ext ? tc_ext->mru : 0;
>                 post_ct = tc_ext ? tc_ext->post_ct : false;
>                 post_ct_snat = post_ct ? tc_ext->post_ct_snat : false;
> diff --git a/net/sched/act_api.c b/net/sched/act_api.c
> index cd09ef49df22..16fd3d30eb12 100644
> --- a/net/sched/act_api.c
> +++ b/net/sched/act_api.c
> @@ -272,7 +272,7 @@ static int tcf_action_offload_add_ex(struct tc_action *action,
>         if (err)
>                 goto fl_err;
>
> -       err = tc_setup_action(&fl_action->action, actions, extack);
> +       err = tc_setup_action(&fl_action->action, actions, 0, extack);
>         if (err) {
>                 NL_SET_ERR_MSG_MOD(extack,
>                                    "Failed to setup tc actions for offload");
> diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
> index 5b4a95e8a1ee..f3576f261289 100644
> --- a/net/sched/cls_api.c
> +++ b/net/sched/cls_api.c
> @@ -22,6 +22,7 @@
>  #include <linux/idr.h>
>  #include <linux/jhash.h>
>  #include <linux/rculist.h>
> +#include <linux/rhashtable.h>
>  #include <net/net_namespace.h>
>  #include <net/sock.h>
>  #include <net/netlink.h>
> @@ -50,6 +51,110 @@ static LIST_HEAD(tcf_proto_base);
>  /* Protects list of registered TC modules. It is pure SMP lock. */
>  static DEFINE_RWLOCK(cls_mod_lock);
>
> +static struct xarray tcf_exts_miss_cookies_xa;
> +struct tcf_exts_miss_cookie_node {
> +       const struct tcf_chain *chain;
> +       const struct tcf_proto *tp;
> +       const struct tcf_exts *exts;
> +       u32 chain_index;
> +       u32 tp_prio;
> +       u32 handle;
> +       u32 miss_cookie_base;
> +       struct rcu_head rcu;
> +};
> +
> +/* Each tc action entry cookie will be comprised of 32bit miss_cookie_base +
> + * action index in the exts tc actions array.
> + */
> +union tcf_exts_miss_cookie {
> +       struct {
> +               u32 miss_cookie_base;
> +               u32 act_index;
> +       };
> +       u64 miss_cookie;
> +};
> +
> +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
> +static int
> +tcf_exts_miss_cookie_base_alloc(struct tcf_exts *exts, struct tcf_proto *tp,
> +                               u32 handle)
> +{
> +       struct tcf_exts_miss_cookie_node *n;
> +       static u32 next;
> +       int err;
> +
> +       if (WARN_ON(!handle || !tp->ops->get_exts))
> +               return -EINVAL;
> +
> +       n = kzalloc(sizeof(*n), GFP_KERNEL);
> +       if (!n)
> +               return -ENOMEM;
> +
> +       n->chain_index = tp->chain->index;
> +       n->chain = tp->chain;
> +       n->tp_prio = tp->prio;
> +       n->tp = tp;
> +       n->exts = exts;
> +       n->handle = handle;
> +
> +       err = xa_alloc_cyclic(&tcf_exts_miss_cookies_xa, &n->miss_cookie_base,
> +                             n, xa_limit_32b, &next, GFP_KERNEL);
> +       if (err)
> +               goto err_xa_alloc;
> +
> +       exts->miss_cookie_node = n;
> +       return 0;
> +
> +err_xa_alloc:
> +       kfree(n);
> +       return err;
> +}
> +
> +static void tcf_exts_miss_cookie_base_destroy(struct tcf_exts *exts)
> +{
> +       struct tcf_exts_miss_cookie_node *n;
> +
> +       if (!exts->miss_cookie_node)
> +               return;
> +
> +       n = exts->miss_cookie_node;
> +       xa_erase(&tcf_exts_miss_cookies_xa, n->miss_cookie_base);
> +       kfree_rcu(n, rcu);
> +}
> +
> +static struct tcf_exts_miss_cookie_node *
> +tcf_exts_miss_cookie_lookup(u64 miss_cookie, int *act_index)
> +{
> +       union tcf_exts_miss_cookie mc = { .miss_cookie = miss_cookie, };
> +
> +       *act_index = mc.act_index;
> +       return xa_load(&tcf_exts_miss_cookies_xa, mc.miss_cookie_base);
> +}
> +#else /* IS_ENABLED(CONFIG_NET_TC_SKB_EXT) */
> +static inline int
> +tcf_exts_miss_cookie_base_alloc(struct tcf_exts *exts, struct tcf_proto *tp,
> +                               u32 handle)
> +{
> +       return 0;
> +}
> +
> +static inline void tcf_exts_miss_cookie_base_destroy(struct tcf_exts *exts)
> +{
> +       return;
> +}
> +#endif /* IS_ENABLED(CONFIG_NET_TC_SKB_EXT) */
> +
> +static u64 tcf_exts_miss_cookie_get(u32 miss_cookie_base, int act_index)
> +{
> +       union tcf_exts_miss_cookie mc = { .act_index = act_index, };
> +
> +       if (!miss_cookie_base)
> +               return 0;
> +
> +       mc.miss_cookie_base = miss_cookie_base;
> +       return mc.miss_cookie;
> +}
> +
>  #ifdef CONFIG_NET_CLS_ACT
>  DEFINE_STATIC_KEY_FALSE(tc_skb_ext_tc);
>  EXPORT_SYMBOL(tc_skb_ext_tc);
> @@ -1549,6 +1654,8 @@ static inline int __tcf_classify(struct sk_buff *skb,
>                                  const struct tcf_proto *orig_tp,
>                                  struct tcf_result *res,
>                                  bool compat_mode,
> +                                struct tcf_exts_miss_cookie_node *n,
> +                                int act_index,
>                                  u32 *last_executed_chain)
>  {
>  #ifdef CONFIG_NET_CLS_ACT
> @@ -1560,13 +1667,36 @@ static inline int __tcf_classify(struct sk_buff *skb,
>  #endif
>         for (; tp; tp = rcu_dereference_bh(tp->next)) {
>                 __be16 protocol = skb_protocol(skb, false);
> -               int err;
> +               int err = 0;
>
> -               if (tp->protocol != protocol &&
> -                   tp->protocol != htons(ETH_P_ALL))
> -                       continue;
> +               if (n) {
> +                       struct tcf_exts *exts;
> +
> +                       if (n->tp_prio != tp->prio)
> +                               continue;
> +
> +                       /* We re-lookup the tp and chain based on index instead
> +                        * of having hard refs and locks to them, so do a sanity
> +                        * check if any of tp,chain,exts was replaced by the
> +                        * time we got here with a cookie from hardware.
> +                        */
> +                       if (unlikely(n->tp != tp || n->tp->chain != n->chain ||
> +                                    !tp->ops->get_exts))
> +                               return TC_ACT_SHOT;
> +
> +                       exts = tp->ops->get_exts(tp, n->handle);
> +                       if (unlikely(!exts || n->exts != exts))
> +                               return TC_ACT_SHOT;
>
> -               err = tc_classify(skb, tp, res);
> +                       n = NULL;
> +                       err = tcf_exts_exec_ex(skb, exts, act_index, res);
> +               } else {
> +                       if (tp->protocol != protocol &&
> +                           tp->protocol != htons(ETH_P_ALL))
> +                               continue;
> +
> +                       err = tc_classify(skb, tp, res);
> +               }
>  #ifdef CONFIG_NET_CLS_ACT
>                 if (unlikely(err == TC_ACT_RECLASSIFY && !compat_mode)) {
>                         first_tp = orig_tp;
> @@ -1582,6 +1712,9 @@ static inline int __tcf_classify(struct sk_buff *skb,
>                         return err;
>         }
>
> +       if (unlikely(n))
> +               return TC_ACT_SHOT;
> +
>         return TC_ACT_UNSPEC; /* signal: continue lookup */
>  #ifdef CONFIG_NET_CLS_ACT
>  reset:
> @@ -1606,21 +1739,33 @@ int tcf_classify(struct sk_buff *skb,
>  #if !IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
>         u32 last_executed_chain = 0;
>
> -       return __tcf_classify(skb, tp, tp, res, compat_mode,
> +       return __tcf_classify(skb, tp, tp, res, compat_mode, NULL, 0,
>                               &last_executed_chain);
>  #else
>         u32 last_executed_chain = tp ? tp->chain->index : 0;
> +       struct tcf_exts_miss_cookie_node *n = NULL;
>         const struct tcf_proto *orig_tp = tp;
>         struct tc_skb_ext *ext;
> +       int act_index = 0;
>         int ret;
>
>         if (block) {
>                 ext = skb_ext_find(skb, TC_SKB_EXT);
>
> -               if (ext && ext->chain) {
> +               if (ext && (ext->chain || ext->act_miss)) {
>                         struct tcf_chain *fchain;
> +                       u32 chain = ext->chain;
>
> -                       fchain = tcf_chain_lookup_rcu(block, ext->chain);
> +                       if (ext->act_miss) {
> +                               n = tcf_exts_miss_cookie_lookup(ext->act_miss_cookie,
> +                                                               &act_index);
> +                               if (!n)
> +                                       return TC_ACT_SHOT;
> +
> +                               chain = n->chain_index;
> +                       }
> +
> +                       fchain = tcf_chain_lookup_rcu(block, chain);
>                         if (!fchain)
>                                 return TC_ACT_SHOT;
>
> @@ -1632,7 +1777,7 @@ int tcf_classify(struct sk_buff *skb,
>                 }
>         }
>
> -       ret = __tcf_classify(skb, tp, orig_tp, res, compat_mode,
> +       ret = __tcf_classify(skb, tp, orig_tp, res, compat_mode, n, act_index,
>                              &last_executed_chain);
>
>         if (tc_skb_ext_tc_enabled()) {
> @@ -3056,9 +3201,48 @@ static int tc_dump_chain(struct sk_buff *skb, struct netlink_callback *cb)
>         return skb->len;
>  }
>
> +int tcf_exts_init_ex(struct tcf_exts *exts, struct net *net, int action,
> +                    int police, struct tcf_proto *tp, u32 handle,
> +                    bool use_action_miss)
> +{
> +       int err = 0;
> +
> +#ifdef CONFIG_NET_CLS_ACT
> +       exts->type = 0;
> +       exts->nr_actions = 0;
> +       /* Note: we do not own yet a reference on net.
> +        * This reference might be taken later from tcf_exts_get_net().
> +        */
> +       exts->net = net;
> +       exts->actions = kcalloc(TCA_ACT_MAX_PRIO, sizeof(struct tc_action *),
> +                               GFP_KERNEL);
> +       if (!exts->actions)
> +               return -ENOMEM;
> +#endif
> +
> +       exts->action = action;
> +       exts->police = police;
> +
> +       if (!use_action_miss)
> +               return 0;
> +
> +       err = tcf_exts_miss_cookie_base_alloc(exts, tp, handle);
> +       if (err)
> +               goto err_miss_alloc;
> +
> +       return 0;
> +
> +err_miss_alloc:
> +       tcf_exts_destroy(exts);
> +       return err;
> +}
> +EXPORT_SYMBOL(tcf_exts_init_ex);
> +
>  void tcf_exts_destroy(struct tcf_exts *exts)
>  {
>  #ifdef CONFIG_NET_CLS_ACT
> +       tcf_exts_miss_cookie_base_destroy(exts);
> +
>         if (exts->actions) {
>                 tcf_action_destroy(exts->actions, TCA_ACT_UNBIND);
>                 kfree(exts->actions);
> @@ -3547,6 +3731,7 @@ static int tc_setup_offload_act(struct tc_action *act,
>
>  int tc_setup_action(struct flow_action *flow_action,
>                     struct tc_action *actions[],
> +                   u32 miss_cookie_base,
>                     struct netlink_ext_ack *extack)
>  {
>         int i, j, k, index, err = 0;
> @@ -3577,6 +3762,8 @@ int tc_setup_action(struct flow_action *flow_action,
>                 for (k = 0; k < index ; k++) {
>                         entry[k].hw_stats = tc_act_hw_stats(act->hw_stats);
>                         entry[k].hw_index = act->tcfa_index;
> +                       entry[k].miss_cookie =
> +                               tcf_exts_miss_cookie_get(miss_cookie_base, i);
>                 }
>
>                 j += index;
> @@ -3599,10 +3786,15 @@ int tc_setup_offload_action(struct flow_action *flow_action,
>                             struct netlink_ext_ack *extack)
>  {
>  #ifdef CONFIG_NET_CLS_ACT
> +       u32 miss_cookie_base;
> +
>         if (!exts)
>                 return 0;
>
> -       return tc_setup_action(flow_action, exts->actions, extack);
> +       miss_cookie_base = exts->miss_cookie_node ?
> +                          exts->miss_cookie_node->miss_cookie_base : 0;
> +       return tc_setup_action(flow_action, exts->actions, miss_cookie_base,
> +                              extack);
>  #else
>         return 0;
>  #endif
> @@ -3770,6 +3962,8 @@ static int __init tc_filter_init(void)
>         if (err)
>                 goto err_register_pernet_subsys;
>
> +       xa_init_flags(&tcf_exts_miss_cookies_xa, XA_FLAGS_ALLOC1);
> +
>         rtnl_register(PF_UNSPEC, RTM_NEWTFILTER, tc_new_tfilter, NULL,
>                       RTNL_FLAG_DOIT_UNLOCKED);
>         rtnl_register(PF_UNSPEC, RTM_DELTFILTER, tc_del_tfilter, NULL,
> --
> 2.30.1
>
Simon Horman Jan. 28, 2023, 4:45 p.m. UTC | #2
On Wed, Jan 25, 2023 at 05:32:13PM +0200, Paul Blakey wrote:
> For drivers to support partial offload of a filter's action list,
> add support for action miss to specify an action instance to
> continue from in sw.
> 
> CT action in particular can't be fully offloaded, as new connections
> need to be handled in software. This imposes other limitations on
> the actions that can be offloaded together with the CT action, such
> as packet modifications.
> 
> Assign each action on a filter's action list a unique miss_cookie
> which drivers can then use to fill action_miss part of the tc skb
> extension. On getting back this miss_cookie, find the action
> instance with relevant cookie and continue classifying from there.
> 
> Signed-off-by: Paul Blakey <paulb@nvidia.com>
> Reviewed-by: Jiri Pirko <jiri@nvidia.com>

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Simon Horman Jan. 28, 2023, 4:52 p.m. UTC | #3
On Wed, Jan 25, 2023 at 05:32:13PM +0200, Paul Blakey wrote:
> For drivers to support partial offload of a filter's action list,
> add support for action miss to specify an action instance to
> continue from in sw.
> 
> CT action in particular can't be fully offloaded, as new connections
> need to be handled in software. This imposes other limitations on
> the actions that can be offloaded together with the CT action, such
> as packet modifications.
> 
> Assign each action on a filter's action list a unique miss_cookie
> which drivers can then use to fill action_miss part of the tc skb
> extension. On getting back this miss_cookie, find the action
> instance with relevant cookie and continue classifying from there.
> 
> Signed-off-by: Paul Blakey <paulb@nvidia.com>
> Reviewed-by: Jiri Pirko <jiri@nvidia.com>

Signed-off-by: Simon Horman <simon.horman@corigine.com>

Sorry if this is a duplicate; I did something weird the first time around.
diff mbox series

Patch

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 4c8492401a10..348673dcb6bb 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -316,12 +316,16 @@  struct nf_bridge_info {
  * and read by ovs to recirc_id.
  */
 struct tc_skb_ext {
-	__u32 chain;
+	union {
+		u64 act_miss_cookie;
+		__u32 chain;
+	};
 	__u16 mru;
 	__u16 zone;
 	u8 post_ct:1;
 	u8 post_ct_snat:1;
 	u8 post_ct_dnat:1;
+	u8 act_miss:1; /* Set if act_miss_cookie is used */
 };
 #endif
 
diff --git a/include/net/flow_offload.h b/include/net/flow_offload.h
index 0400a0ac8a29..88db7346eb7a 100644
--- a/include/net/flow_offload.h
+++ b/include/net/flow_offload.h
@@ -228,6 +228,7 @@  void flow_action_cookie_destroy(struct flow_action_cookie *cookie);
 struct flow_action_entry {
 	enum flow_action_id		id;
 	u32				hw_index;
+	u64				miss_cookie;
 	enum flow_action_hw_stats	hw_stats;
 	action_destr			destructor;
 	void				*destructor_priv;
diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 4cabb32a2ad9..344f2ca02ebd 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -59,6 +59,8 @@  int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
 void tcf_block_put(struct tcf_block *block);
 void tcf_block_put_ext(struct tcf_block *block, struct Qdisc *q,
 		       struct tcf_block_ext_info *ei);
+int tcf_exts_init_ex(struct tcf_exts *exts, struct net *net, int action, int police,
+		     struct tcf_proto *tp, u32 handle, bool used_action_miss);
 
 static inline bool tcf_block_shared(struct tcf_block *block)
 {
@@ -229,6 +231,7 @@  struct tcf_exts {
 	struct tc_action **actions;
 	struct net	*net;
 	netns_tracker	ns_tracker;
+	struct tcf_exts_miss_cookie_node *miss_cookie_node;
 #endif
 	/* Map to export classifier specific extension TLV types to the
 	 * generic extensions API. Unsupported extensions must be set to 0.
@@ -240,21 +243,11 @@  struct tcf_exts {
 static inline int tcf_exts_init(struct tcf_exts *exts, struct net *net,
 				int action, int police)
 {
-#ifdef CONFIG_NET_CLS_ACT
-	exts->type = 0;
-	exts->nr_actions = 0;
-	/* Note: we do not own yet a reference on net.
-	 * This reference might be taken later from tcf_exts_get_net().
-	 */
-	exts->net = net;
-	exts->actions = kcalloc(TCA_ACT_MAX_PRIO, sizeof(struct tc_action *),
-				GFP_KERNEL);
-	if (!exts->actions)
-		return -ENOMEM;
+#ifdef CONFIG_NET_CLS
+	return tcf_exts_init_ex(exts, net, action, police, NULL, 0, false);
+#else
+	return -EOPNOTSUPP;
 #endif
-	exts->action = action;
-	exts->police = police;
-	return 0;
 }
 
 /* Return false if the netns is being destroyed in cleanup_net(). Callers
@@ -353,6 +346,18 @@  tcf_exts_exec(struct sk_buff *skb, struct tcf_exts *exts,
 	return TC_ACT_OK;
 }
 
+static inline int
+tcf_exts_exec_ex(struct sk_buff *skb, struct tcf_exts *exts, int act_index,
+		 struct tcf_result *res)
+{
+#ifdef CONFIG_NET_CLS_ACT
+	return tcf_action_exec(skb, exts->actions + act_index,
+			       exts->nr_actions - act_index, res);
+#else
+	return TC_ACT_OK;
+#endif
+}
+
 int tcf_exts_validate(struct net *net, struct tcf_proto *tp,
 		      struct nlattr **tb, struct nlattr *rate_tlv,
 		      struct tcf_exts *exts, u32 flags,
@@ -577,6 +582,7 @@  int tc_setup_offload_action(struct flow_action *flow_action,
 void tc_cleanup_offload_action(struct flow_action *flow_action);
 int tc_setup_action(struct flow_action *flow_action,
 		    struct tc_action *actions[],
+		    u32 miss_cookie_base,
 		    struct netlink_ext_ack *extack);
 
 int tc_setup_cb_call(struct tcf_block *block, enum tc_setup_type type,
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index af4aa66aaa4e..fab5ba3e61b7 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -369,6 +369,8 @@  struct tcf_proto_ops {
 						struct nlattr **tca,
 						struct netlink_ext_ack *extack);
 	void			(*tmplt_destroy)(void *tmplt_priv);
+	struct tcf_exts *	(*get_exts)(const struct tcf_proto *tp,
+					    u32 handle);
 
 	/* rtnetlink specific */
 	int			(*dump)(struct net*, struct tcf_proto*, void *,
diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c
index e20d1a973417..b1a5eed8d1a9 100644
--- a/net/openvswitch/flow.c
+++ b/net/openvswitch/flow.c
@@ -1038,7 +1038,7 @@  int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
 #if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
 	if (tc_skb_ext_tc_enabled()) {
 		tc_ext = skb_ext_find(skb, TC_SKB_EXT);
-		key->recirc_id = tc_ext ? tc_ext->chain : 0;
+		key->recirc_id = tc_ext && !tc_ext->act_miss ? tc_ext->chain : 0;
 		OVS_CB(skb)->mru = tc_ext ? tc_ext->mru : 0;
 		post_ct = tc_ext ? tc_ext->post_ct : false;
 		post_ct_snat = post_ct ? tc_ext->post_ct_snat : false;
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index cd09ef49df22..16fd3d30eb12 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -272,7 +272,7 @@  static int tcf_action_offload_add_ex(struct tc_action *action,
 	if (err)
 		goto fl_err;
 
-	err = tc_setup_action(&fl_action->action, actions, extack);
+	err = tc_setup_action(&fl_action->action, actions, 0, extack);
 	if (err) {
 		NL_SET_ERR_MSG_MOD(extack,
 				   "Failed to setup tc actions for offload");
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 5b4a95e8a1ee..f3576f261289 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -22,6 +22,7 @@ 
 #include <linux/idr.h>
 #include <linux/jhash.h>
 #include <linux/rculist.h>
+#include <linux/rhashtable.h>
 #include <net/net_namespace.h>
 #include <net/sock.h>
 #include <net/netlink.h>
@@ -50,6 +51,110 @@  static LIST_HEAD(tcf_proto_base);
 /* Protects list of registered TC modules. It is pure SMP lock. */
 static DEFINE_RWLOCK(cls_mod_lock);
 
+static struct xarray tcf_exts_miss_cookies_xa;
+struct tcf_exts_miss_cookie_node {
+	const struct tcf_chain *chain;
+	const struct tcf_proto *tp;
+	const struct tcf_exts *exts;
+	u32 chain_index;
+	u32 tp_prio;
+	u32 handle;
+	u32 miss_cookie_base;
+	struct rcu_head rcu;
+};
+
+/* Each tc action entry cookie will be comprised of 32bit miss_cookie_base +
+ * action index in the exts tc actions array.
+ */
+union tcf_exts_miss_cookie {
+	struct {
+		u32 miss_cookie_base;
+		u32 act_index;
+	};
+	u64 miss_cookie;
+};
+
+#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
+static int
+tcf_exts_miss_cookie_base_alloc(struct tcf_exts *exts, struct tcf_proto *tp,
+				u32 handle)
+{
+	struct tcf_exts_miss_cookie_node *n;
+	static u32 next;
+	int err;
+
+	if (WARN_ON(!handle || !tp->ops->get_exts))
+		return -EINVAL;
+
+	n = kzalloc(sizeof(*n), GFP_KERNEL);
+	if (!n)
+		return -ENOMEM;
+
+	n->chain_index = tp->chain->index;
+	n->chain = tp->chain;
+	n->tp_prio = tp->prio;
+	n->tp = tp;
+	n->exts = exts;
+	n->handle = handle;
+
+	err = xa_alloc_cyclic(&tcf_exts_miss_cookies_xa, &n->miss_cookie_base,
+			      n, xa_limit_32b, &next, GFP_KERNEL);
+	if (err)
+		goto err_xa_alloc;
+
+	exts->miss_cookie_node = n;
+	return 0;
+
+err_xa_alloc:
+	kfree(n);
+	return err;
+}
+
+static void tcf_exts_miss_cookie_base_destroy(struct tcf_exts *exts)
+{
+	struct tcf_exts_miss_cookie_node *n;
+
+	if (!exts->miss_cookie_node)
+		return;
+
+	n = exts->miss_cookie_node;
+	xa_erase(&tcf_exts_miss_cookies_xa, n->miss_cookie_base);
+	kfree_rcu(n, rcu);
+}
+
+static struct tcf_exts_miss_cookie_node *
+tcf_exts_miss_cookie_lookup(u64 miss_cookie, int *act_index)
+{
+	union tcf_exts_miss_cookie mc = { .miss_cookie = miss_cookie, };
+
+	*act_index = mc.act_index;
+	return xa_load(&tcf_exts_miss_cookies_xa, mc.miss_cookie_base);
+}
+#else /* IS_ENABLED(CONFIG_NET_TC_SKB_EXT) */
+static inline int
+tcf_exts_miss_cookie_base_alloc(struct tcf_exts *exts, struct tcf_proto *tp,
+				u32 handle)
+{
+	return 0;
+}
+
+static inline void tcf_exts_miss_cookie_base_destroy(struct tcf_exts *exts)
+{
+	return;
+}
+#endif /* IS_ENABLED(CONFIG_NET_TC_SKB_EXT) */
+
+static u64 tcf_exts_miss_cookie_get(u32 miss_cookie_base, int act_index)
+{
+	union tcf_exts_miss_cookie mc = { .act_index = act_index, };
+
+	if (!miss_cookie_base)
+		return 0;
+
+	mc.miss_cookie_base = miss_cookie_base;
+	return mc.miss_cookie;
+}
+
 #ifdef CONFIG_NET_CLS_ACT
 DEFINE_STATIC_KEY_FALSE(tc_skb_ext_tc);
 EXPORT_SYMBOL(tc_skb_ext_tc);
@@ -1549,6 +1654,8 @@  static inline int __tcf_classify(struct sk_buff *skb,
 				 const struct tcf_proto *orig_tp,
 				 struct tcf_result *res,
 				 bool compat_mode,
+				 struct tcf_exts_miss_cookie_node *n,
+				 int act_index,
 				 u32 *last_executed_chain)
 {
 #ifdef CONFIG_NET_CLS_ACT
@@ -1560,13 +1667,36 @@  static inline int __tcf_classify(struct sk_buff *skb,
 #endif
 	for (; tp; tp = rcu_dereference_bh(tp->next)) {
 		__be16 protocol = skb_protocol(skb, false);
-		int err;
+		int err = 0;
 
-		if (tp->protocol != protocol &&
-		    tp->protocol != htons(ETH_P_ALL))
-			continue;
+		if (n) {
+			struct tcf_exts *exts;
+
+			if (n->tp_prio != tp->prio)
+				continue;
+
+			/* We re-lookup the tp and chain based on index instead
+			 * of having hard refs and locks to them, so do a sanity
+			 * check if any of tp,chain,exts was replaced by the
+			 * time we got here with a cookie from hardware.
+			 */
+			if (unlikely(n->tp != tp || n->tp->chain != n->chain ||
+				     !tp->ops->get_exts))
+				return TC_ACT_SHOT;
+
+			exts = tp->ops->get_exts(tp, n->handle);
+			if (unlikely(!exts || n->exts != exts))
+				return TC_ACT_SHOT;
 
-		err = tc_classify(skb, tp, res);
+			n = NULL;
+			err = tcf_exts_exec_ex(skb, exts, act_index, res);
+		} else {
+			if (tp->protocol != protocol &&
+			    tp->protocol != htons(ETH_P_ALL))
+				continue;
+
+			err = tc_classify(skb, tp, res);
+		}
 #ifdef CONFIG_NET_CLS_ACT
 		if (unlikely(err == TC_ACT_RECLASSIFY && !compat_mode)) {
 			first_tp = orig_tp;
@@ -1582,6 +1712,9 @@  static inline int __tcf_classify(struct sk_buff *skb,
 			return err;
 	}
 
+	if (unlikely(n))
+		return TC_ACT_SHOT;
+
 	return TC_ACT_UNSPEC; /* signal: continue lookup */
 #ifdef CONFIG_NET_CLS_ACT
 reset:
@@ -1606,21 +1739,33 @@  int tcf_classify(struct sk_buff *skb,
 #if !IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
 	u32 last_executed_chain = 0;
 
-	return __tcf_classify(skb, tp, tp, res, compat_mode,
+	return __tcf_classify(skb, tp, tp, res, compat_mode, NULL, 0,
 			      &last_executed_chain);
 #else
 	u32 last_executed_chain = tp ? tp->chain->index : 0;
+	struct tcf_exts_miss_cookie_node *n = NULL;
 	const struct tcf_proto *orig_tp = tp;
 	struct tc_skb_ext *ext;
+	int act_index = 0;
 	int ret;
 
 	if (block) {
 		ext = skb_ext_find(skb, TC_SKB_EXT);
 
-		if (ext && ext->chain) {
+		if (ext && (ext->chain || ext->act_miss)) {
 			struct tcf_chain *fchain;
+			u32 chain = ext->chain;
 
-			fchain = tcf_chain_lookup_rcu(block, ext->chain);
+			if (ext->act_miss) {
+				n = tcf_exts_miss_cookie_lookup(ext->act_miss_cookie,
+								&act_index);
+				if (!n)
+					return TC_ACT_SHOT;
+
+				chain = n->chain_index;
+			}
+
+			fchain = tcf_chain_lookup_rcu(block, chain);
 			if (!fchain)
 				return TC_ACT_SHOT;
 
@@ -1632,7 +1777,7 @@  int tcf_classify(struct sk_buff *skb,
 		}
 	}
 
-	ret = __tcf_classify(skb, tp, orig_tp, res, compat_mode,
+	ret = __tcf_classify(skb, tp, orig_tp, res, compat_mode, n, act_index,
 			     &last_executed_chain);
 
 	if (tc_skb_ext_tc_enabled()) {
@@ -3056,9 +3201,48 @@  static int tc_dump_chain(struct sk_buff *skb, struct netlink_callback *cb)
 	return skb->len;
 }
 
+int tcf_exts_init_ex(struct tcf_exts *exts, struct net *net, int action,
+		     int police, struct tcf_proto *tp, u32 handle,
+		     bool use_action_miss)
+{
+	int err = 0;
+
+#ifdef CONFIG_NET_CLS_ACT
+	exts->type = 0;
+	exts->nr_actions = 0;
+	/* Note: we do not own yet a reference on net.
+	 * This reference might be taken later from tcf_exts_get_net().
+	 */
+	exts->net = net;
+	exts->actions = kcalloc(TCA_ACT_MAX_PRIO, sizeof(struct tc_action *),
+				GFP_KERNEL);
+	if (!exts->actions)
+		return -ENOMEM;
+#endif
+
+	exts->action = action;
+	exts->police = police;
+
+	if (!use_action_miss)
+		return 0;
+
+	err = tcf_exts_miss_cookie_base_alloc(exts, tp, handle);
+	if (err)
+		goto err_miss_alloc;
+
+	return 0;
+
+err_miss_alloc:
+	tcf_exts_destroy(exts);
+	return err;
+}
+EXPORT_SYMBOL(tcf_exts_init_ex);
+
 void tcf_exts_destroy(struct tcf_exts *exts)
 {
 #ifdef CONFIG_NET_CLS_ACT
+	tcf_exts_miss_cookie_base_destroy(exts);
+
 	if (exts->actions) {
 		tcf_action_destroy(exts->actions, TCA_ACT_UNBIND);
 		kfree(exts->actions);
@@ -3547,6 +3731,7 @@  static int tc_setup_offload_act(struct tc_action *act,
 
 int tc_setup_action(struct flow_action *flow_action,
 		    struct tc_action *actions[],
+		    u32 miss_cookie_base,
 		    struct netlink_ext_ack *extack)
 {
 	int i, j, k, index, err = 0;
@@ -3577,6 +3762,8 @@  int tc_setup_action(struct flow_action *flow_action,
 		for (k = 0; k < index ; k++) {
 			entry[k].hw_stats = tc_act_hw_stats(act->hw_stats);
 			entry[k].hw_index = act->tcfa_index;
+			entry[k].miss_cookie =
+				tcf_exts_miss_cookie_get(miss_cookie_base, i);
 		}
 
 		j += index;
@@ -3599,10 +3786,15 @@  int tc_setup_offload_action(struct flow_action *flow_action,
 			    struct netlink_ext_ack *extack)
 {
 #ifdef CONFIG_NET_CLS_ACT
+	u32 miss_cookie_base;
+
 	if (!exts)
 		return 0;
 
-	return tc_setup_action(flow_action, exts->actions, extack);
+	miss_cookie_base = exts->miss_cookie_node ?
+			   exts->miss_cookie_node->miss_cookie_base : 0;
+	return tc_setup_action(flow_action, exts->actions, miss_cookie_base,
+			       extack);
 #else
 	return 0;
 #endif
@@ -3770,6 +3962,8 @@  static int __init tc_filter_init(void)
 	if (err)
 		goto err_register_pernet_subsys;
 
+	xa_init_flags(&tcf_exts_miss_cookies_xa, XA_FLAGS_ALLOC1);
+
 	rtnl_register(PF_UNSPEC, RTM_NEWTFILTER, tc_new_tfilter, NULL,
 		      RTNL_FLAG_DOIT_UNLOCKED);
 	rtnl_register(PF_UNSPEC, RTM_DELTFILTER, tc_del_tfilter, NULL,