linux_dsm_epyc7002/include/net
Martin KaFai Lau 2dbb9b9e6d bpf: Introduce BPF_PROG_TYPE_SK_REUSEPORT
This patch adds a BPF_PROG_TYPE_SK_REUSEPORT which can select
a SO_REUSEPORT sk from a BPF_MAP_TYPE_REUSEPORT_ARRAY.  Like other
non SK_FILTER/CGROUP_SKB program, it requires CAP_SYS_ADMIN.

BPF_PROG_TYPE_SK_REUSEPORT introduces "struct sk_reuseport_kern"
to store the bpf context instead of using the skb->cb[48].

At the SO_REUSEPORT sk lookup time, it is in the middle of transiting
from a lower layer (ipv4/ipv6) to a upper layer (udp/tcp).  At this
point,  it is not always clear where the bpf context can be appended
in the skb->cb[48] to avoid saving-and-restoring cb[].  Even putting
aside the difference between ipv4-vs-ipv6 and udp-vs-tcp.  It is not
clear if the lower layer is only ipv4 and ipv6 in the future and
will it not touch the cb[] again before transiting to the upper
layer.

For example, in udp_gro_receive(), it uses the 48 byte NAPI_GRO_CB
instead of IP[6]CB and it may still modify the cb[] after calling
the udp[46]_lib_lookup_skb().  Because of the above reason, if
sk->cb is used for the bpf ctx, saving-and-restoring is needed
and likely the whole 48 bytes cb[] has to be saved and restored.

Instead of saving, setting and restoring the cb[], this patch opts
to create a new "struct sk_reuseport_kern" and setting the needed
values in there.

The new BPF_PROG_TYPE_SK_REUSEPORT and "struct sk_reuseport_(kern|md)"
will serve all ipv4/ipv6 + udp/tcp combinations.  There is no protocol
specific usage at this point and it is also inline with the current
sock_reuseport.c implementation (i.e. no protocol specific requirement).

In "struct sk_reuseport_md", this patch exposes data/data_end/len
with semantic similar to other existing usages.  Together
with "bpf_skb_load_bytes()" and "bpf_skb_load_bytes_relative()",
the bpf prog can peek anywhere in the skb.  The "bind_inany" tells
the bpf prog that the reuseport group is bind-ed to a local
INANY address which cannot be learned from skb.

The new "bind_inany" is added to "struct sock_reuseport" which will be
used when running the new "BPF_PROG_TYPE_SK_REUSEPORT" bpf prog in order
to avoid repeating the "bind INANY" test on
"sk_v6_rcv_saddr/sk->sk_rcv_saddr" every time a bpf prog is run.  It can
only be properly initialized when a "sk->sk_reuseport" enabled sk is
adding to a hashtable (i.e. during "reuseport_alloc()" and
"reuseport_add_sock()").

The new "sk_select_reuseport()" is the main helper that the
bpf prog will use to select a SO_REUSEPORT sk.  It is the only function
that can use the new BPF_MAP_TYPE_REUSEPORT_ARRAY.  As mentioned in
the earlier patch, the validity of a selected sk is checked in
run time in "sk_select_reuseport()".  Doing the check in
verification time is difficult and inflexible (consider the map-in-map
use case).  The runtime check is to compare the selected sk's reuseport_id
with the reuseport_id that we want.  This helper will return -EXXX if the
selected sk cannot serve the incoming request (e.g. reuseport_id
not match).  The bpf prog can decide if it wants to do SK_DROP as its
discretion.

When the bpf prog returns SK_PASS, the kernel will check if a
valid sk has been selected (i.e. "reuse_kern->selected_sk != NULL").
If it does , it will use the selected sk.  If not, the kernel
will select one from "reuse->socks[]" (as before this patch).

The SK_DROP and SK_PASS handling logic will be in the next patch.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-08-11 01:58:46 +02:00
..
9p 9p: Implement show_options 2017-07-11 06:08:58 -04:00
bluetooth Bluetooth: Add new quirk for non-persistent setup settings 2018-07-30 14:00:15 +02:00
caif caif: reduce stack size with KASAN 2018-01-19 14:02:12 -05:00
iucv Revert changes to convert to ->poll_mask() and aio IOCB_CMD_POLL 2018-06-28 10:40:47 -07:00
netfilter netfilter: use kvmalloc_array to allocate memory for hashtable 2018-08-03 18:37:55 +02:00
netns net: ipv4: Control SKB reprioritization after forwarding 2018-08-01 09:52:30 -07:00
nfc NFC: Add nfc_dbg() macro 2017-04-05 10:15:20 +02:00
phonet proc: introduce proc_create_net{,_data} 2018-05-16 07:24:30 +02:00
sctp sctp: add support for dscp and flowlabel per transport 2018-07-04 11:36:54 +09:00
tc_act Merge ra.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux 2018-07-20 21:17:12 -07:00
6lowpan.h 6lowpan: Fix IID format for Bluetooth 2017-04-12 22:02:36 +02:00
act_api.h tc/act: remove unneeded RCU lock in action callback 2018-07-30 09:31:13 -07:00
addrconf.h bpf: Introduce BPF_PROG_TYPE_SK_REUSEPORT 2018-08-11 01:58:46 +02:00
af_ieee802154.h ieee802154: add rx LQI from userspace 2018-07-13 12:18:18 -04:00
af_rxrpc.h rxrpc: Push iov_iter up from rxrpc_kernel_recv_data() to caller 2018-08-03 12:46:20 -07:00
af_unix.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
af_vsock.h vsock: split dwork to avoid reinitializations 2018-08-07 12:39:13 -07:00
ah.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
arp.h ipv4: Make neigh lookup keys for loopback/point-to-point devices be INADDR_ANY 2018-01-15 14:53:43 -05:00
atmclip.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ax25.h proc: introduce proc_create_seq{,_data} 2018-05-16 07:23:35 +02:00
ax88796.h net-next: ax88796: add interrupt status callback to platform data 2018-04-19 16:11:11 -04:00
bond_3ad.h include/net/bond_3ad: Simplify the code by using the ARRAY_SIZE 2018-08-04 13:23:15 -07:00
bond_alb.h
bond_options.h bonding: Prevent duplicate userspace notification 2017-05-27 18:51:41 -04:00
bonding.h net: Add lag.h, net_lag_port_dev_txable() 2018-07-11 23:10:19 -07:00
busy_poll.h net: remove sock_poll_busy_flag 2018-07-30 09:10:25 -07:00
calipso.h net, calipso: convert calipso_doi.refcount from atomic_t to refcount_t 2017-07-04 22:35:16 +01:00
cfg80211-wext.h
cfg80211.h Merge ra.kernel.org:/pub/scm/linux/kernel/git/davem/net 2018-07-24 19:21:58 -07:00
cfg802154.h
checksum.h
cipso_ipv4.h net, ipv4: convert cipso_v4_doi.refcount from atomic_t to refcount_t 2017-07-04 01:29:04 -07:00
cls_cgroup.h
codel_impl.h
codel_qdisc.h
codel.h
compat.h net: remove compat_sys_*() prototypes from net/compat.h 2018-04-02 20:16:17 +02:00
datalink.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
dcbevent.h
dcbnl.h net: dcb: Add priority-to-DSCP map getters 2018-07-27 13:17:50 -07:00
devlink.h devlink: Add generic parameters region_snapshot 2018-07-12 17:37:13 -07:00
dn_dev.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
dn_fib.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
dn_neigh.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
dn_nsp.h net/decnet: Convert timers to use timer_setup() 2017-10-18 12:39:36 +01:00
dn_route.h decnet: Move dn_next into decnet route structure. 2017-11-30 09:54:25 -05:00
dn.h Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-11-04 09:26:51 +09:00
dsa.h net: dsa: Remove VLA usage 2018-07-18 15:08:31 -07:00
dsfield.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
dst_cache.h net: core: dst_cache_set_ip6: Rename 'addr' parameter to 'saddr' for consistency 2018-03-05 12:52:45 -05:00
dst_metadata.h Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-11-04 09:26:51 +09:00
dst_ops.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
dst.h xfrm: Remove xfrmi interface ID from flowi 2018-07-20 10:14:41 +02:00
erspan.h erspan: set bso bit based on mirrored packet's len 2018-05-20 18:31:42 -04:00
esp.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ethoc.h inet: whitespace cleanup 2018-02-28 11:43:28 -05:00
failover.h net: Introduce generic failover module 2018-05-28 22:59:54 -04:00
fib_notifier.h net: Add extack to fib_notifier_info 2017-11-01 11:50:43 +09:00
fib_rules.h net: fib_rules: add extack support 2018-04-23 10:21:24 -04:00
firewire.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
flow_dissector.h flow_dissector: allow dissection of tunnel options from metadata 2018-08-07 12:22:14 -07:00
flow.h xfrm: Remove xfrmi interface ID from flowi 2018-07-20 10:14:41 +02:00
fou.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
fq_impl.h Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-10-30 21:09:24 +09:00
fq.h fq: support filtering a given tin 2017-10-11 09:49:34 +02:00
garp.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
gen_stats.h net: sched: add support for TCQ_F_NOLOCK subqueues to sch_mq 2017-12-08 13:32:26 -05:00
genetlink.h genetlink: fix genlmsg_nlhdr() 2017-11-16 10:49:00 +09:00
geneve.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
gre.h net: GRE: Add is_gretap_dev, is_ip6gretap_dev 2018-02-27 14:46:26 -05:00
gro_cells.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
gtp.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
gue.h fou: fix some member types in guehdr 2017-12-11 14:10:06 -05:00
hwbm.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
icmp.h
ieee80211_radiotap.h radiotap: add structs for HE 2018-06-15 14:04:00 +02:00
ieee802154_netdev.h
if_inet6.h net/ipv6: Add support for specifying metric of connected routes 2018-05-29 10:12:45 -04:00
ife.h net: sched: ife: handle malformed tlv length 2018-04-22 21:12:00 -04:00
ila.h
inet6_connection_sock.h
inet6_hashtables.h net: ipv6: add second dif to inet6 socket lookups 2017-08-07 11:39:22 -07:00
inet_common.h net: Convert GRO SKB handling to list_head. 2018-06-26 11:33:04 +09:00
inet_connection_sock.h net: ipv4: remove define INET_CSK_DEBUG and unnecessary EXPORT_SYMBOL 2018-05-10 17:43:55 -04:00
inet_ecn.h Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-11-04 09:26:51 +09:00
inet_frag.h ip: use rb trees for IP frag queue. 2018-08-05 17:16:46 -07:00
inet_hashtables.h inet: Add a 2nd listener hashtable (port+addr) 2017-12-03 10:18:28 -05:00
inet_sock.h net: add helpers checking if socket can be bound to nonlocal address 2018-08-01 09:50:04 -07:00
inet_timewait_sock.h net-tcp: remove useless tw_timeout field 2018-06-05 10:45:24 -04:00
inetpeer.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ip6_checksum.h
ip6_fib.h ipv6: use fib6_info_hold_safe() when necessary 2018-07-23 11:19:02 -07:00
ip6_route.h net/ipv6: Revert attempt to simplify route replace and append 2018-07-04 15:22:13 +09:00
ip6_tunnel.h ip6_gre: add erspan v2 support 2017-12-15 12:34:00 -05:00
ip_fib.h net/ipv4: Add helper to return path MTU based on fib result 2018-05-22 10:51:09 +02:00
ip_tunnels.h net: check tunnel option type in tunnel flags 2018-06-29 23:50:26 +09:00
ip_vs.h ipvs: add assured state for conn templates 2018-07-18 11:26:40 +02:00
ip.h ip: remove tx_flags from ipcm_cookie and use same logic for v4 and v6 2018-07-07 10:58:49 +09:00
ipcomp.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ipconfig.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ipv6_frag.h ipv6: remove dependency of nf_defrag_ipv6 on ipv6 module 2018-07-18 11:26:53 +02:00
ipv6.h net: don't declare IPv6 non-local bind helper if CONFIG_IPV6 undefined 2018-08-01 13:45:31 -07:00
ipx.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
iw_handler.h net: Spelling s/stucture/structure/ 2018-03-27 09:51:23 +02:00
kcm.h
l3mdev.h
lag.h net: Add lag.h, net_lag_port_dev_txable() 2018-07-11 23:10:19 -07:00
lapb.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
lib80211.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
llc_c_ac.h net: LLC: Convert timers to use timer_setup() 2017-10-25 12:06:25 +09:00
llc_c_ev.h
llc_c_st.h
llc_conn.h llc: delete timers synchronously in llc_sk_free() 2018-04-22 14:55:03 -04:00
llc_if.h
llc_pdu.h
llc_s_ac.h
llc_s_ev.h
llc_s_st.h
llc_sap.h
llc.h llc: use refcount_inc_not_zero() for llc_sap_find() 2018-08-07 15:54:00 -07:00
lwtunnel.h net: Move ipv4 set_lwt_redirect helper to lwtunnel 2018-02-14 14:43:32 -05:00
mac80211.h mac80211: add support for HE 2018-06-18 22:40:32 +02:00
mac802154.h
mip6.h
mld.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
mpls_iptunnel.h net: mpls: Increase max number of labels for lwt encap 2017-04-01 20:21:44 -07:00
mpls.h
mrp.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ncsi.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ndisc.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
neighbour.h neighbour: support for NTF_EXT_LEARNED flag 2018-04-25 13:19:59 -04:00
net_failover.h net: Introduce net_failover driver 2018-05-28 22:59:54 -04:00
net_namespace.h net: create reusable function for getting ownership info of sysfs inodes 2018-07-20 23:44:36 -07:00
net_ratelimit.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
netevent.h net: ipv4: Notify about changes to ip_forward_update_priority 2018-08-01 09:52:30 -07:00
netlabel.h net: convert netlbl_lsm_cache.refcount from atomic_t to refcount_t 2017-07-01 07:39:09 -07:00
netlink.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
netprio_cgroup.h
netrom.h proc: introduce proc_create_seq{,_data} 2018-05-16 07:23:35 +02:00
nexthop.h net: fix rtnh_ok() 2018-04-07 22:32:31 -04:00
nl802154.h
nsh.h openvswitch: enable NSH support 2017-11-08 16:12:33 +09:00
p8022.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
page_pool.h xdp: introduce xdp_return_frame_rx_napi 2018-05-24 18:36:15 -07:00
ping.h ipv{4,6}/ping: simplify proc file creation 2018-05-16 07:23:35 +02:00
pkt_cls.h net: sched: make tcf_chain_{get,put}() static 2018-08-01 10:06:19 -07:00
pkt_sched.h net/sched: Add HW offloading capability to ETF 2018-07-04 22:30:27 +09:00
pptp.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
protocol.h IPv4: early demux can return an error code 2017-10-01 03:55:47 +01:00
psample.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
psnap.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
raw.h ipv{4,6}/raw: simplify ѕeq_file code 2018-05-16 07:23:35 +02:00
rawv6.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
red.h net_sched: red: Avoid illegal values 2017-12-05 14:37:13 -05:00
regulatory.h cfg80211: read wmm rules from regulatory database 2018-03-29 11:11:40 +02:00
request_sock.h tcp: socket option to set TCP fast open key 2017-10-20 13:21:36 +01:00
rose.h proc: introduce proc_create_seq{,_data} 2018-05-16 07:23:35 +02:00
route.h net/ipv4: Add support for specifying metric of connected routes 2018-05-29 10:12:45 -04:00
rsi_91x.h Bluetooth: btrsi: add new rsi bluetooth driver 2018-03-13 18:37:02 +02:00
rtnetlink.h rtnetlink: Remove VLA usage 2018-05-31 22:48:46 -04:00
sch_generic.h net/tc: introduce TC_ACT_REINSERT. 2018-07-30 09:31:14 -07:00
scm.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
secure_seq.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
seg6_hmac.h rhashtable: split rhashtable.h 2018-06-22 13:43:27 +09:00
seg6_local.h bpf: add End.DT6 action to bpf_lwt_seg6_action helper 2018-07-31 09:22:48 +02:00
seg6.h rhashtable: split rhashtable.h 2018-06-22 13:43:27 +09:00
slhc_vj.h slip: Check if rstate is initialized before uncompressing 2018-04-11 10:33:46 -04:00
smc.h net/smc: add pnetid support for SMC-D and ISM 2018-06-30 20:42:25 +09:00
snmp.h
sock_reuseport.h bpf: Introduce BPF_PROG_TYPE_SK_REUSEPORT 2018-08-11 01:58:46 +02:00
sock.h net: remove bogus RCU annotations on socket.wq 2018-07-31 12:40:22 -07:00
Space.h net/mac89x0: Convert to platform_driver 2018-03-01 21:21:36 -05:00
stp.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
strparser.h strparser: Add __strp_unpause and use it in ktls. 2018-06-06 14:07:53 -04:00
switchdev.h switchdev: Add fdb.added_by_user to switchdev notifications 2018-05-03 13:46:47 -04:00
tcp_states.h tcp: remove the hardcode in the definition of TCPF Macro 2018-02-21 15:06:05 -05:00
tcp.h tcp: Avoid TCP syncookie rejected by SO_REUSEPORT socket 2018-08-11 01:58:45 +02:00
timewait_sock.h
tipc.h flow_dissector: do not rely on implicit casts 2018-05-08 00:02:41 -04:00
tls.h tls: Add rx inline crypto offload 2018-07-16 00:13:11 -07:00
transp_v6.h ipv6: fold sockcm_cookie into ipcm6_cookie 2018-07-07 10:58:49 +09:00
tso.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
tun_proto.h vxlan: factor out VXLAN-GPE next protocol 2017-08-29 15:16:52 -07:00
udp_tunnel.h net: Convert GRO SKB handling to list_head. 2018-06-26 11:33:04 +09:00
udp.h Merge ra.kernel.org:/pub/scm/linux/kernel/git/davem/net 2018-07-03 10:29:26 +09:00
udplite.h udplite: fix partial checksum initialization 2018-02-16 15:57:42 -05:00
vsock_addr.h
vxlan.h vxlan: add ttl inherit support 2018-04-17 13:53:13 -04:00
wext.h lift handling of SIOCIW... out of dev_ioctl() 2018-01-24 19:13:45 -05:00
wimax.h
x25.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
x25device.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
xdp_sock.h xsk: fix potential race in SKB TX completion code 2018-07-02 18:37:12 -07:00
xdp.h xdp: Helper function to clear kernel pointers in xdp_frame 2018-08-10 16:12:20 +02:00
xfrm.h xfrm: Remove xfrmi interface ID from flowi 2018-07-20 10:14:41 +02:00