All of lore.kernel.org
 help / color / mirror / Atom feed
From: Madhu Challa <challa@noironetworks.com>
To: Alexander Duyck <alexander.h.duyck@redhat.com>
Cc: netdev@vger.kernel.org, stephen@networkplumber.org,
	jiri@resnulli.us, sfeldma@gmail.com,
	David Miller <davem@davemloft.net>
Subject: Re: [net-next PATCH] ipv4: FIB Local/MAIN table collapse
Date: Thu, 12 Mar 2015 12:20:15 -0700	[thread overview]
Message-ID: <CAN_zqfNtYutMCuCuxTLy8bD8gUsPyLF4nb9_=TJx35JPcrs5Tg@mail.gmail.com> (raw)
In-Reply-To: <20150306213830.1139.16932.stgit@ahduyck-vm-fedora20>

Alex,

I see a null pointer deference in fib_trie_unmerge on boot with latest
net-next and thought it might be related. Pl let me know if you need
any additional info.

Thanks.

[  131.289254] BUG: unable to handle kernel NULL pointer dereference
at 0000000000000030
[  131.298052] IP: [<ffffffff8171d75d>] fib_trie_unmerge+0x1d/0x2f0
[  131.304788] PGD 0
[  131.307045] Oops: 0000 [#1] SMP
[  131.310674] Modules linked in: iptable_mangle(+) xt_tcpudp
ip6table_filter ip6_tables ebtable_nat ebtables ipmi_devintf
xt_addrtype xt_conntrack ipt_MASQUERADE nf_nat_masquerade_ipv4
iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat
nf_conntrack bridge stp llc dm_thin_pool iptable_filter ip_tables
x_tables bnep rfcomm bluetooth x86_pkg_temp_thermal intel_powerclamp
coretemp kvm_intel kvm crc32_pclmul ghash_clmulni_intel aesni_intel
joydev aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd
microcode sb_edac edac_core ipmi_si lpc_ich ipmi_msghandler tpm_tis
wmi acpi_power_meter mac_hid parport_pc ppdev lp parport ixgbe
hid_generic igb vxlan ip6_udp_tunnel i2c_algo_bit udp_tunnel usbhid
dca hid ptp mdio pps_core
[  131.383538] CPU: 8 PID: 242 Comm: kworker/u48:1 Not tainted 4.0.0-rc3+ #2
[  131.391130] Hardware name: Cisco Systems Inc
UCSC-C220-M3S/UCSC-C220-M3S, BIOS C220M3.1.5.4f.0.111320130449
11/13/2013
[  131.403090] Workqueue: netns cleanup_net
[  131.407480] task: ffff88380213cb30 ti: ffff8838027e4000 task.ti:
ffff8838027e4000
[  131.415846] RIP: 0010:[<ffffffff8171d75d>]  [<ffffffff8171d75d>]
fib_trie_unmerge+0x1d/0x2f0
[  131.425297] RSP: 0018:ffff8838027e7c38  EFLAGS: 00010292
[  131.431233] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000038
[  131.439211] RDX: ffff88380cae8138 RSI: 00000000000000ff RDI: 0000000000000000
[  131.447191] RBP: ffff8838027e7c88 R08: ffff883810ca2f80 R09: 0000000180190008
[  131.455167] R10: ffffffff81682c43 R11: ffffea00e0432800 R12: ffff88380cae8000
[  131.463139] R13: ffff881fe27efa40 R14: ffff881fe27efac8 R15: ffff88380cae8008
[  131.471117] FS:  0000000000000000(0000) GS:ffff88387fc40000(0000)
knlGS:0000000000000000
[  131.480163] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  131.486583] CR2: 0000000000000030 CR3: 0000000001c0c000 CR4: 00000000001407e0
[  131.494562] Stack:
[  131.496839]  ffff8838027e7c68 ffffffff811c441a ffff883810ca3038
ffff883810ca2fb0
[  131.505278]  ffff883810ca3038 0000000000000000 ffff88380cae8000
ffff881fe27efa40
[  131.513714]  ffff881fe27efac8 ffff88380cae8008 ffff8838027e7ca8
ffffffff81717204
[  131.522147] Call Trace:
[  131.524911]  [<ffffffff811c441a>] ? kmem_cache_free+0xfa/0x160
[  131.531455]  [<ffffffff81717204>] fib_unmerge+0x24/0xd0
[  131.537324]  [<ffffffff8172282f>] fib4_rule_delete+0x1f/0x60
[  131.543674]  [<ffffffff816bb239>] fib_rules_unregister+0xb9/0xf0
[  131.550382]  [<ffffffff81722bb5>] fib4_rules_exit+0x15/0x20
[  131.556609]  [<ffffffff81716643>] ip_fib_net_exit+0x23/0x130
[  131.562933]  [<ffffffff81716785>] fib_net_exit+0x35/0x40
[  131.568871]  [<ffffffff8169409d>] ops_exit_list.isra.7+0x4d/0x70
[  131.575589]  [<ffffffff81694e00>] cleanup_net+0x1b0/0x250
[  131.581628]  [<ffffffff81087a4d>] process_one_work+0x22d/0x400
[  131.588145]  [<ffffffff810882ed>] worker_thread+0x2fd/0x550
[  131.594375]  [<ffffffff81087ff0>] ? rescuer_thread+0x3d0/0x3d0
[  131.600904]  [<ffffffff8108d603>] kthread+0xe3/0xf0
[  131.606355]  [<ffffffff8108d520>] ? kthread_stop+0xf0/0xf0
[  131.612500]  [<ffffffff817a4f58>] ret_from_fork+0x58/0x90
[  131.618538]  [<ffffffff8108d520>] ? kthread_stop+0xf0/0xf0
[  131.624675] Code: c3 66 66 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f
44 00 00 55 48 8d 4f 38 48 89 f8 48 89 e5 41 57 41 56 41 55 41 54 53
48 83 ec 28 <48> 8b 57 30 48 39 ca 48 89 55 c8 0f 84 aa 02 00 00 31 f6
bf ff
[  131.647012] RIP  [<ffffffff8171d75d>] fib_trie_unmerge+0x1d/0x2f0
[  131.653850]  RSP <ffff8838027e7c38>
[  131.657755] CR2: 0000000000000030
[  131.661468] ---[ end trace 8db31cc50a0eb505 ]---
[  131.748516] BUG: unable to handle kernel paging request at ffffffffffffffd8
[  131.756344] IP: [<ffffffff8108daa0>] kthread_data+0x10/0x20
[  131.762599] PGD 1c0f067 PUD 1c11067 PMD 0
[  131.767240] Oops: 0000 [#2] SMP

On Fri, Mar 6, 2015 at 1:47 PM, Alexander Duyck
<alexander.h.duyck@redhat.com> wrote:
> This patch is meant to collapse local and main into one by converting
> tb_data from an array to a pointer.  Doing this allows us to point the
> local table into the main while maintaining the same variables in the
> table.
>
> As such the tb_data was converted from an array to a pointer, and a new
> array called data is added in order to still provide an object for tb_data
> to point to.
>
> In order to track the origin of the fib aliases a tb_id value was added in
> a hole that existed on 64b systems.  Using this we can also reverse the
> merge in the event that custom FIB rules are enabled.
>
> With this patch I am seeing an improvement of 20ns to 30ns for routing
> lookups as long as custom rules are not enabled, with custom rules enabled
> we fall back to split tables and the original behavior.
>
> Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com>
> ---
>
> Changes since the RFC:
>   Added tb_id value so I could split main and local for custom rules
>   Added functionality to split tables if custom rules were enabled
>   Added table replacement and unmerge functions
>
> I have done some testing on this to verify performance gains and that I can
> split the tables correctly when I enable custom rules, but this patch is
> what I would consider to be high risk since I am certain there are things I
> have not considered.
>
> If this gets pulled into someone's switchdev tree instead of into net-next I
> would be perfectly fine with that as I am sure this can use some additional
> testing.
>
>  include/net/fib_rules.h |    2 -
>  include/net/ip_fib.h    |   26 ++-----
>  net/core/fib_rules.c    |    8 ++
>  net/ipv4/fib_frontend.c |   59 ++++++++++++++--
>  net/ipv4/fib_lookup.h   |    1
>  net/ipv4/fib_rules.c    |   20 ++++-
>  net/ipv4/fib_trie.c     |  172 +++++++++++++++++++++++++++++++++++++++++++++--
>  7 files changed, 250 insertions(+), 38 deletions(-)
>
> diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
> index e584de1..88d2ae5 100644
> --- a/include/net/fib_rules.h
> +++ b/include/net/fib_rules.h
> @@ -58,7 +58,7 @@ struct fib_rules_ops {
>                                              struct sk_buff *,
>                                              struct fib_rule_hdr *,
>                                              struct nlattr **);
> -       void                    (*delete)(struct fib_rule *);
> +       int                     (*delete)(struct fib_rule *);
>         int                     (*compare)(struct fib_rule *,
>                                            struct fib_rule_hdr *,
>                                            struct nlattr **);
> diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
> index 1657604..54271ed 100644
> --- a/include/net/ip_fib.h
> +++ b/include/net/ip_fib.h
> @@ -186,7 +186,8 @@ struct fib_table {
>         int                     tb_default;
>         int                     tb_num_default;
>         struct rcu_head         rcu;
> -       unsigned long           tb_data[0];
> +       unsigned long           *tb_data;
> +       unsigned long           __data[0];
>  };
>
>  int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp,
> @@ -196,11 +197,10 @@ int fib_table_delete(struct fib_table *, struct fib_config *);
>  int fib_table_dump(struct fib_table *table, struct sk_buff *skb,
>                    struct netlink_callback *cb);
>  int fib_table_flush(struct fib_table *table);
> +struct fib_table *fib_trie_unmerge(struct fib_table *main_tb);
>  void fib_table_flush_external(struct fib_table *table);
>  void fib_free_table(struct fib_table *tb);
>
> -
> -
>  #ifndef CONFIG_IP_MULTIPLE_TABLES
>
>  #define TABLE_LOCAL_INDEX      (RT_TABLE_LOCAL & (FIB_TABLE_HASHSZ - 1))
> @@ -229,18 +229,13 @@ static inline int fib_lookup(struct net *net, const struct flowi4 *flp,
>                              struct fib_result *res)
>  {
>         struct fib_table *tb;
> -       int err;
> +       int err = -ENETUNREACH;
>
>         rcu_read_lock();
>
> -       for (err = 0; !err; err = -ENETUNREACH) {
> -               tb = fib_get_table(net, RT_TABLE_LOCAL);
> -               if (tb && !fib_table_lookup(tb, flp, res, FIB_LOOKUP_NOREF))
> -                       break;
> -               tb = fib_get_table(net, RT_TABLE_MAIN);
> -               if (tb && !fib_table_lookup(tb, flp, res, FIB_LOOKUP_NOREF))
> -                       break;
> -       }
> +       tb = fib_get_table(net, RT_TABLE_MAIN);
> +       if (tb && !fib_table_lookup(tb, flp, res, FIB_LOOKUP_NOREF))
> +               err = 0;
>
>         rcu_read_unlock();
>
> @@ -270,10 +265,6 @@ static inline int fib_lookup(struct net *net, struct flowi4 *flp,
>         res->tclassid = 0;
>
>         for (err = 0; !err; err = -ENETUNREACH) {
> -               tb = rcu_dereference_rtnl(net->ipv4.fib_local);
> -               if (tb && !fib_table_lookup(tb, flp, res, FIB_LOOKUP_NOREF))
> -                       break;
> -
>                 tb = rcu_dereference_rtnl(net->ipv4.fib_main);
>                 if (tb && !fib_table_lookup(tb, flp, res, FIB_LOOKUP_NOREF))
>                         break;
> @@ -309,6 +300,7 @@ static inline int fib_num_tclassid_users(struct net *net)
>         return 0;
>  }
>  #endif
> +int fib_unmerge(struct net *net);
>  void fib_flush_external(struct net *net);
>
>  /* Exported by fib_semantics.c */
> @@ -320,7 +312,7 @@ void fib_select_multipath(struct fib_result *res);
>
>  /* Exported by fib_trie.c */
>  void fib_trie_init(void);
> -struct fib_table *fib_trie_table(u32 id);
> +struct fib_table *fib_trie_table(u32 id, struct fib_table *alias);
>
>  static inline void fib_combine_itag(u32 *itag, const struct fib_result *res)
>  {
> diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
> index 44706e8..b55677f 100644
> --- a/net/core/fib_rules.c
> +++ b/net/core/fib_rules.c
> @@ -492,6 +492,12 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh)
>                         goto errout;
>                 }
>
> +               if (ops->delete) {
> +                       err = ops->delete(rule);
> +                       if (err)
> +                               goto errout;
> +               }
> +
>                 list_del_rcu(&rule->list);
>
>                 if (rule->action == FR_ACT_GOTO) {
> @@ -517,8 +523,6 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh)
>
>                 notify_rule_change(RTM_DELRULE, rule, ops, nlh,
>                                    NETLINK_CB(skb).portid);
> -               if (ops->delete)
> -                       ops->delete(rule);
>                 fib_rule_put(rule);
>                 flush_route_cache(ops);
>                 rules_ops_put(ops);
> diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
> index e067770..7cda3b0 100644
> --- a/net/ipv4/fib_frontend.c
> +++ b/net/ipv4/fib_frontend.c
> @@ -52,14 +52,14 @@ static int __net_init fib4_rules_init(struct net *net)
>  {
>         struct fib_table *local_table, *main_table;
>
> -       local_table = fib_trie_table(RT_TABLE_LOCAL);
> -       if (local_table == NULL)
> -               return -ENOMEM;
> -
> -       main_table  = fib_trie_table(RT_TABLE_MAIN);
> +       main_table  = fib_trie_table(RT_TABLE_MAIN, NULL);
>         if (main_table == NULL)
>                 goto fail;
>
> +       local_table = fib_trie_table(RT_TABLE_LOCAL, main_table);
> +       if (local_table == NULL)
> +               return -ENOMEM;
> +
>         hlist_add_head_rcu(&local_table->tb_hlist,
>                                 &net->ipv4.fib_table_hash[TABLE_LOCAL_INDEX]);
>         hlist_add_head_rcu(&main_table->tb_hlist,
> @@ -74,7 +74,7 @@ fail:
>
>  struct fib_table *fib_new_table(struct net *net, u32 id)
>  {
> -       struct fib_table *tb;
> +       struct fib_table *tb, *alias = NULL;
>         unsigned int h;
>
>         if (id == 0)
> @@ -83,7 +83,10 @@ struct fib_table *fib_new_table(struct net *net, u32 id)
>         if (tb)
>                 return tb;
>
> -       tb = fib_trie_table(id);
> +       if (id == RT_TABLE_LOCAL)
> +               alias = fib_new_table(net, RT_TABLE_MAIN);
> +
> +       tb = fib_trie_table(id, alias);
>         if (!tb)
>                 return NULL;
>
> @@ -126,6 +129,48 @@ struct fib_table *fib_get_table(struct net *net, u32 id)
>  }
>  #endif /* CONFIG_IP_MULTIPLE_TABLES */
>
> +static void fib_replace_table(struct net *net, struct fib_table *old,
> +                             struct fib_table *new)
> +{
> +#ifdef CONFIG_IP_MULTIPLE_TABLES
> +       switch (new->tb_id) {
> +       case RT_TABLE_LOCAL:
> +               rcu_assign_pointer(net->ipv4.fib_local, new);
> +               break;
> +       case RT_TABLE_MAIN:
> +               rcu_assign_pointer(net->ipv4.fib_main, new);
> +               break;
> +       case RT_TABLE_DEFAULT:
> +               rcu_assign_pointer(net->ipv4.fib_default, new);
> +               break;
> +       default:
> +               break;
> +       }
> +
> +#endif
> +       /* replace the old table in the hlist */
> +       hlist_replace_rcu(&old->tb_hlist, &new->tb_hlist);
> +}
> +
> +int fib_unmerge(struct net *net)
> +{
> +       struct fib_table *old, *new;
> +
> +       old = fib_get_table(net, RT_TABLE_LOCAL);
> +       new = fib_trie_unmerge(old);
> +
> +       if (!new)
> +               return -ENOMEM;
> +
> +       /* replace merged table with clean table */
> +       if (new != old) {
> +               fib_replace_table(net, old, new);
> +               fib_free_table(old);
> +       }
> +
> +       return 0;
> +}
> +
>  static void fib_flush(struct net *net)
>  {
>         int flushed = 0;
> diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h
> index ae2e6ee..c6211ed 100644
> --- a/net/ipv4/fib_lookup.h
> +++ b/net/ipv4/fib_lookup.h
> @@ -12,6 +12,7 @@ struct fib_alias {
>         u8                      fa_type;
>         u8                      fa_state;
>         u8                      fa_slen;
> +       u32                     tb_id;
>         struct rcu_head         rcu;
>  };
>
> diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
> index 190d0d0..e9bc5e4 100644
> --- a/net/ipv4/fib_rules.c
> +++ b/net/ipv4/fib_rules.c
> @@ -174,6 +174,11 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
>         if (frh->tos & ~IPTOS_TOS_MASK)
>                 goto errout;
>
> +       /* split local/main if they are not already split */
> +       err = fib_unmerge(net);
> +       if (err)
> +               goto errout;
> +
>         if (rule->table == RT_TABLE_UNSPEC) {
>                 if (rule->action == FR_ACT_TO_TBL) {
>                         struct fib_table *table;
> @@ -216,17 +221,24 @@ errout:
>         return err;
>  }
>
> -static void fib4_rule_delete(struct fib_rule *rule)
> +static int fib4_rule_delete(struct fib_rule *rule)
>  {
>         struct net *net = rule->fr_net;
> -#ifdef CONFIG_IP_ROUTE_CLASSID
> -       struct fib4_rule *rule4 = (struct fib4_rule *) rule;
> +       int err;
>
> -       if (rule4->tclassid)
> +       /* split local/main if they are not already split */
> +       err = fib_unmerge(net);
> +       if (err)
> +               goto errout;
> +
> +#ifdef CONFIG_IP_ROUTE_CLASSID
> +       if (((struct fib4_rule *)rule)->tclassid)
>                 net->ipv4.fib_num_tclassid_users--;
>  #endif
>         net->ipv4.fib_has_custom_rules = true;
>         fib_flush_external(rule->fr_net);
> +errout:
> +       return err;
>  }
>
>  static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
> diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
> index 9095545..1fa862b 100644
> --- a/net/ipv4/fib_trie.c
> +++ b/net/ipv4/fib_trie.c
> @@ -1122,6 +1122,9 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg)
>                                 break;
>                         if (fa->fa_info->fib_priority != fi->fib_priority)
>                                 break;
> +                       /* duplicate entry from another table */
> +                       if (WARN_ON(fa->tb_id != tb->tb_id))
> +                               continue;
>                         if (fa->fa_type == cfg->fc_type &&
>                             fa->fa_info == fi) {
>                                 fa_match = fa;
> @@ -1198,6 +1201,7 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg)
>         new_fa->fa_type = cfg->fc_type;
>         new_fa->fa_state = 0;
>         new_fa->fa_slen = slen;
> +       new_fa->tb_id = tb->tb_id;
>
>         /* (Optionally) offload fib entry to switch hardware. */
>         err = netdev_switch_fib_ipv4_add(key, plen, fi, tos,
> @@ -1216,7 +1220,7 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg)
>                 tb->tb_num_default++;
>
>         rt_cache_flush(cfg->fc_nlinfo.nl_net);
> -       rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, tb->tb_id,
> +       rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, new_fa->tb_id,
>                   &cfg->fc_nlinfo, 0);
>  succeeded:
>         return 0;
> @@ -1242,7 +1246,7 @@ static inline t_key prefix_mismatch(t_key key, struct key_vector *n)
>  int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp,
>                      struct fib_result *res, int fib_flags)
>  {
> -       struct trie *t = (struct trie *)tb->tb_data;
> +       struct trie *t = (struct trie *) tb->tb_data;
>  #ifdef CONFIG_IP_FIB_TRIE_STATS
>         struct trie_use_stats __percpu *stats = t->stats;
>  #endif
> @@ -1482,6 +1486,9 @@ int fib_table_delete(struct fib_table *tb, struct fib_config *cfg)
>                 if ((fa->fa_slen != slen) || (fa->fa_tos != tos))
>                         break;
>
> +               if (fa->tb_id != tb->tb_id)
> +                       continue;
> +
>                 if ((!cfg->fc_type || fa->fa_type == cfg->fc_type) &&
>                     (cfg->fc_scope == RT_SCOPE_NOWHERE ||
>                      fa->fa_info->fib_scope == cfg->fc_scope) &&
> @@ -1575,6 +1582,120 @@ found:
>         return n;
>  }
>
> +static void fib_trie_free(struct fib_table *tb)
> +{
> +       struct trie *t = (struct trie *)tb->tb_data;
> +       struct key_vector *pn = t->kv;
> +       unsigned long cindex = 1;
> +       struct hlist_node *tmp;
> +       struct fib_alias *fa;
> +
> +       /* walk trie in reverse order and free everything */
> +       for (;;) {
> +               struct key_vector *n;
> +
> +               if (!(cindex--)) {
> +                       t_key pkey = pn->key;
> +
> +                       if (IS_TRIE(pn))
> +                               break;
> +
> +                       n = pn;
> +                       pn = node_parent(pn);
> +
> +                       /* drop emptied tnode */
> +                       put_child_root(pn, n->key, NULL);
> +                       node_free(n);
> +
> +                       cindex = get_index(pkey, pn);
> +
> +                       continue;
> +               }
> +
> +               /* grab the next available node */
> +               n = get_child(pn, cindex);
> +               if (!n)
> +                       continue;
> +
> +               if (IS_TNODE(n)) {
> +                       /* record pn and cindex for leaf walking */
> +                       pn = n;
> +                       cindex = 1ul << n->bits;
> +
> +                       continue;
> +               }
> +
> +               hlist_for_each_entry_safe(fa, tmp, &n->leaf, fa_list) {
> +                       hlist_del_rcu(&fa->fa_list);
> +                       alias_free_mem_rcu(fa);
> +               }
> +
> +               put_child_root(pn, n->key, NULL);
> +               node_free(n);
> +       }
> +
> +#ifdef CONFIG_IP_FIB_TRIE_STATS
> +       free_percpu(t->stats);
> +#endif
> +       kfree(tb);
> +}
> +
> +struct fib_table *fib_trie_unmerge(struct fib_table *oldtb)
> +{
> +       struct trie *ot = (struct trie *)oldtb->tb_data;
> +       struct key_vector *l, *tp = ot->kv;
> +       struct fib_table *local_tb;
> +       struct fib_alias *fa;
> +       struct trie *lt;
> +       t_key key = 0;
> +
> +       if (oldtb->tb_data == oldtb->__data)
> +               return oldtb;
> +
> +       local_tb = fib_trie_table(RT_TABLE_LOCAL, NULL);
> +       if (!local_tb)
> +               return NULL;
> +
> +       lt = (struct trie *)local_tb->tb_data;
> +
> +       while ((l = leaf_walk_rcu(&tp, key)) != NULL) {
> +               struct key_vector *local_l = NULL, *local_tp;
> +
> +               hlist_for_each_entry_rcu(fa, &l->leaf, fa_list) {
> +                       struct fib_alias *new_fa;
> +
> +                       if (local_tb->tb_id != fa->tb_id)
> +                               continue;
> +
> +                       /* clone fa for new local table */
> +                       new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL);
> +                       if (!new_fa)
> +                               goto out;
> +
> +                       memcpy(new_fa, fa, sizeof(*fa));
> +
> +                       /* insert clone into table */
> +                       if (!local_l)
> +                               local_l = fib_find_node(lt, &local_tp, l->key);
> +
> +                       if (fib_insert_alias(lt, local_tp, local_l, new_fa,
> +                                            NULL, l->key))
> +                               goto out;
> +               }
> +
> +               /* stop loop if key wrapped back to 0 */
> +               key = l->key + 1;
> +               if (key < l->key)
> +                       break;
> +       }
> +
> +       return local_tb;
> +out:
> +       fib_trie_free(local_tb);
> +
> +       return NULL;
> +}
> +
>  /* Caller must hold RTNL */
>  void fib_table_flush_external(struct fib_table *tb)
>  {
> @@ -1586,6 +1707,7 @@ void fib_table_flush_external(struct fib_table *tb)
>
>         /* walk trie in reverse order */
>         for (;;) {
> +               unsigned char slen = 0;
>                 struct key_vector *n;
>
>                 if (!(cindex--)) {
> @@ -1595,8 +1717,8 @@ void fib_table_flush_external(struct fib_table *tb)
>                         if (IS_TRIE(pn))
>                                 break;
>
> -                       /* no need to resize like in flush below */
> -                       pn = node_parent(pn);
> +                       /* resize completed node */
> +                       pn = resize(t, pn);
>                         cindex = get_index(pkey, pn);
>
>                         continue;
> @@ -1618,6 +1740,18 @@ void fib_table_flush_external(struct fib_table *tb)
>                 hlist_for_each_entry_safe(fa, tmp, &n->leaf, fa_list) {
>                         struct fib_info *fi = fa->fa_info;
>
> +                       /* if alias was cloned to local then we just
> +                        * need to remove the local copy from main
> +                        */
> +                       if (tb->tb_id != fa->tb_id) {
> +                               hlist_del_rcu(&fa->fa_list);
> +                               alias_free_mem_rcu(fa);
> +                               continue;
> +                       }
> +
> +                       /* record local slen */
> +                       slen = fa->fa_slen;
> +
>                         if (!fi || !(fi->fib_flags & RTNH_F_EXTERNAL))
>                                 continue;
>
> @@ -1626,6 +1760,16 @@ void fib_table_flush_external(struct fib_table *tb)
>                                                    fi, fa->fa_tos,
>                                                    fa->fa_type, tb->tb_id);
>                 }
> +
> +               /* update leaf slen */
> +               n->slen = slen;
> +
> +               if (hlist_empty(&n->leaf)) {
> +                       put_child_root(pn, n->key, NULL);
> +                       node_free(n);
> +               } else {
> +                       leaf_pull_suffix(pn, n);
> +               }
>         }
>  }
>
> @@ -1710,7 +1854,8 @@ static void __trie_free_rcu(struct rcu_head *head)
>  #ifdef CONFIG_IP_FIB_TRIE_STATS
>         struct trie *t = (struct trie *)tb->tb_data;
>
> -       free_percpu(t->stats);
> +       if (tb->tb_data == tb->__data)
> +               free_percpu(t->stats);
>  #endif /* CONFIG_IP_FIB_TRIE_STATS */
>         kfree(tb);
>  }
> @@ -1737,6 +1882,11 @@ static int fn_trie_dump_leaf(struct key_vector *l, struct fib_table *tb,
>                         continue;
>                 }
>
> +               if (tb->tb_id != fa->tb_id) {
> +                       i++;
> +                       continue;
> +               }
> +
>                 if (fib_dump_info(skb, NETLINK_CB(cb->skb).portid,
>                                   cb->nlh->nlmsg_seq,
>                                   RTM_NEWROUTE,
> @@ -1803,18 +1953,26 @@ void __init fib_trie_init(void)
>                                            0, SLAB_PANIC, NULL);
>  }
>
> -struct fib_table *fib_trie_table(u32 id)
> +struct fib_table *fib_trie_table(u32 id, struct fib_table *alias)
>  {
>         struct fib_table *tb;
>         struct trie *t;
> +       size_t sz = sizeof(*tb);
> +
> +       if (!alias)
> +               sz += sizeof(struct trie);
>
> -       tb = kzalloc(sizeof(*tb) + sizeof(struct trie), GFP_KERNEL);
> +       tb = kzalloc(sz, GFP_KERNEL);
>         if (tb == NULL)
>                 return NULL;
>
>         tb->tb_id = id;
>         tb->tb_default = -1;
>         tb->tb_num_default = 0;
> +       tb->tb_data = (alias ? alias->__data : tb->__data);
> +
> +       if (alias)
> +               return tb;
>
>         t = (struct trie *) tb->tb_data;
>         t->kv[0].pos = KEYLENGTH;
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

  parent reply	other threads:[~2015-03-12 19:20 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-06 21:47 [net-next PATCH] ipv4: FIB Local/MAIN table collapse Alexander Duyck
2015-03-11 14:58 ` Alexander Duyck
2015-03-11 16:03   ` Dave Taht
2015-03-11 16:23     ` Alexander Duyck
2015-03-11 17:13       ` Dave Taht
2015-03-11 20:28 ` David Miller
2015-03-12 19:20 ` Madhu Challa [this message]
2015-03-12 20:08   ` Alexander Duyck

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='CAN_zqfNtYutMCuCuxTLy8bD8gUsPyLF4nb9_=TJx35JPcrs5Tg@mail.gmail.com' \
    --to=challa@noironetworks.com \
    --cc=alexander.h.duyck@redhat.com \
    --cc=davem@davemloft.net \
    --cc=jiri@resnulli.us \
    --cc=netdev@vger.kernel.org \
    --cc=sfeldma@gmail.com \
    --cc=stephen@networkplumber.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.