Skip to content

K
Kernel

openeuler / Kernel
接近 2 年 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 25 7月, 2020 1 次提交
  3. 20 7月, 2020 1 次提交
  5. 21 6月, 2020 1 次提交
  7. 02 6月, 2020 1 次提交
  9. 29 5月, 2020 1 次提交
    • E
      tcp: ipv6: support RFC 6069 (TCP-LD) · d2924569
      由 Eric Dumazet 提交于 
      Make tcp_ld_RTO_revert() helper available to IPv6, and
implement RFC 6069 :

Quoting this RFC :

3. Connectivity Disruption Indication

   For Internet Protocol version 6 (IPv6) [RFC2460], the counterpart of
   the ICMP destination unreachable message of code 0 (net unreachable)
   and of code 1 (host unreachable) is the ICMPv6 destination
   unreachable message of code 0 (no route to destination) [RFC4443].
   As with IPv4, a router should generate an ICMPv6 destination
   unreachable message of code 0 in response to a packet that cannot be
   delivered to its destination address because it lacks a matching
   entry in its routing table.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NYuchung Cheng <ycheng@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d2924569
  11. 26 5月, 2020 1 次提交
    • E
      tcp: allow traceroute -Mtcp for unpriv users · 45af29ca
      由 Eric Dumazet 提交于 
      Unpriv users can use traceroute over plain UDP sockets, but not TCP ones.

$ traceroute -Mtcp 8.8.8.8
You do not have enough privileges to use this traceroute method.

$ traceroute -n -Mudp 8.8.8.8
traceroute to 8.8.8.8 (8.8.8.8), 30 hops max, 60 byte packets
 1  192.168.86.1  3.631 ms  3.512 ms  3.405 ms
 2  10.1.10.1  4.183 ms  4.125 ms  4.072 ms
 3  96.120.88.125  20.621 ms  19.462 ms  20.553 ms
 4  96.110.177.65  24.271 ms  25.351 ms  25.250 ms
 5  69.139.199.197  44.492 ms  43.075 ms  44.346 ms
 6  68.86.143.93  27.969 ms  25.184 ms  25.092 ms
 7  96.112.146.18  25.323 ms 96.112.146.22  25.583 ms 96.112.146.26  24.502 ms
 8  72.14.239.204  24.405 ms 74.125.37.224  16.326 ms  17.194 ms
 9  209.85.251.9  18.154 ms 209.85.247.55  14.449 ms 209.85.251.9  26.296 ms^C

We can easily support traceroute over TCP, by queueing an error message
into socket error queue.

Note that applications need to set IP_RECVERR/IPV6_RECVERR option to
enable this feature, and that the error message is only queued
while in SYN_SNT state.

socket(AF_INET6, SOCK_STREAM, IPPROTO_IP) = 3
setsockopt(3, SOL_IPV6, IPV6_RECVERR, [1], 4) = 0
setsockopt(3, SOL_SOCKET, SO_TIMESTAMP_OLD, [1], 4) = 0
setsockopt(3, SOL_IPV6, IPV6_UNICAST_HOPS, [5], 4) = 0
connect(3, {sa_family=AF_INET6, sin6_port=htons(8787), sin6_flowinfo=htonl(0),
        inet_pton(AF_INET6, "2002:a05:6608:297::", &sin6_addr), sin6_scope_id=0}, 28) = -1 EHOSTUNREACH (No route to host)
recvmsg(3, {msg_name={sa_family=AF_INET6, sin6_port=htons(8787), sin6_flowinfo=htonl(0),
        inet_pton(AF_INET6, "2002:a05:6608:297::", &sin6_addr), sin6_scope_id=0},
        msg_namelen=1024->28, msg_iov=[{iov_base="`\r\337\320\0004\6\1&\7\370\260\200\231\16\27\0\0\0\0\0\0\0\0 \2\n\5f\10\2\227"..., iov_len=1024}],
        msg_iovlen=1, msg_control=[{cmsg_len=32, cmsg_level=SOL_SOCKET, cmsg_type=SO_TIMESTAMP_OLD, cmsg_data={tv_sec=1590340680, tv_usec=272424}},
                                   {cmsg_len=60, cmsg_level=SOL_IPV6, cmsg_type=IPV6_RECVERR}],
        msg_controllen=96, msg_flags=MSG_ERRQUEUE}, MSG_ERRQUEUE) = 144

Suggested-by: Maciej Żenczykowski <maze@google.com
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Reviewed-by: NMaciej Żenczykowski <maze@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      45af29ca
  13. 13 3月, 2020 1 次提交
  15. 30 1月, 2020 1 次提交
    • G
      mptcp: Fix undefined mptcp_handle_ipv6_mapped for modular IPV6 · 31484d56
      由 Geert Uytterhoeven 提交于 
      If CONFIG_MPTCP=y, CONFIG_MPTCP_IPV6=n, and CONFIG_IPV6=m:

    ERROR: "mptcp_handle_ipv6_mapped" [net/ipv6/ipv6.ko] undefined!

This does not happen if CONFIG_MPTCP_IPV6=y, as CONFIG_MPTCP_IPV6
selects CONFIG_IPV6, and thus forces CONFIG_IPV6 builtin.

As exporting a symbol for an empty function would be a bit wasteful, fix
this by providing a dummy version of mptcp_handle_ipv6_mapped() for the
CONFIG_MPTCP_IPV6=n case.

Rename mptcp_handle_ipv6_mapped() to mptcpv6_handle_mapped(), to make it
clear this is a pure-IPV6 function, just like mptcpv6_init().

Fixes: cec37a6e ("mptcp: Handle MP_CAPABLE options for outgoing connections")
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      31484d56
  17. 24 1月, 2020 2 次提交
  19. 10 1月, 2020 1 次提交
  21. 03 1月, 2020 3 次提交
    • D
      net: Add device index to tcp_md5sig · 6b102db5
      由 David Ahern 提交于 
      Add support for userspace to specify a device index to limit the scope
of an entry via the TCP_MD5SIG_EXT setsockopt. The existing __tcpm_pad
is renamed to tcpm_ifindex and the new field is only checked if the new
TCP_MD5SIG_FLAG_IFINDEX is set in tcpm_flags. For now, the device index
must point to an L3 master device (e.g., VRF). The API and error
handling are setup to allow the constraint to be relaxed in the future
to any device index.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      6b102db5
    • D
      tcp: Add l3index to tcp_md5sig_key and md5 functions · dea53bb8
      由 David Ahern 提交于 
      Add l3index to tcp_md5sig_key to represent the L3 domain of a key, and
add l3index to tcp_md5_do_add and tcp_md5_do_del to fill in the key.

With the key now based on an l3index, add the new parameter to the
lookup functions and consider the l3index when looking for a match.

The l3index comes from the skb when processing ingress packets leveraging
the helpers created for socket lookups, tcp_v4_sdif and inet_iif (and the
v6 variants). When the sdif index is set it means the packet ingressed a
device that is part of an L3 domain and inet_iif points to the VRF device.
For egress, the L3 domain is determined from the socket binding and
sk_bound_dev_if.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      dea53bb8
    • D
      ipv6/tcp: Pass dif and sdif to tcp_v6_inbound_md5_hash · d14c77e0
      由 David Ahern 提交于 
      The original ingress device index is saved to the cb space of the skb
and the cb is moved during tcp processing. Since tcp_v6_inbound_md5_hash
can be called before and after the cb move, pass dif and sdif to it so
the caller can save both prior to the cb move. Both are used by a later
patch.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d14c77e0
  23. 05 12月, 2019 1 次提交
  25. 07 11月, 2019 1 次提交
  27. 14 10月, 2019 4 次提交
    • E
      tcp: annotate tp->write_seq lockless reads · 0f317464
      由 Eric Dumazet 提交于 
      There are few places where we fetch tp->write_seq while
this field can change from IRQ or other cpu.

We need to add READ_ONCE() annotations, and also make
sure write sides use corresponding WRITE_ONCE() to avoid
store-tearing.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      0f317464
    • E
      tcp: annotate tp->copied_seq lockless reads · 7db48e98
      由 Eric Dumazet 提交于 
      There are few places where we fetch tp->copied_seq while
this field can change from IRQ or other cpu.

We need to add READ_ONCE() annotations, and also make
sure write sides use corresponding WRITE_ONCE() to avoid
store-tearing.

Note that tcp_inq_hint() was already using READ_ONCE(tp->copied_seq)
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      7db48e98
    • E
      tcp: annotate tp->rcv_nxt lockless reads · dba7d9b8
      由 Eric Dumazet 提交于 
      There are few places where we fetch tp->rcv_nxt while
this field can change from IRQ or other cpu.

We need to add READ_ONCE() annotations, and also make
sure write sides use corresponding WRITE_ONCE() to avoid
store-tearing.

Note that tcp_inq_hint() was already using READ_ONCE(tp->rcv_nxt)

syzbot reported :

BUG: KCSAN: data-race in tcp_poll / tcp_queue_rcv

write to 0xffff888120425770 of 4 bytes by interrupt on cpu 0:
 tcp_rcv_nxt_update net/ipv4/tcp_input.c:3365 [inline]
 tcp_queue_rcv+0x180/0x380 net/ipv4/tcp_input.c:4638
 tcp_rcv_established+0xbf1/0xf50 net/ipv4/tcp_input.c:5616
 tcp_v4_do_rcv+0x381/0x4e0 net/ipv4/tcp_ipv4.c:1542
 tcp_v4_rcv+0x1a03/0x1bf0 net/ipv4/tcp_ipv4.c:1923
 ip_protocol_deliver_rcu+0x51/0x470 net/ipv4/ip_input.c:204
 ip_local_deliver_finish+0x110/0x140 net/ipv4/ip_input.c:231
 NF_HOOK include/linux/netfilter.h:305 [inline]
 NF_HOOK include/linux/netfilter.h:299 [inline]
 ip_local_deliver+0x133/0x210 net/ipv4/ip_input.c:252
 dst_input include/net/dst.h:442 [inline]
 ip_rcv_finish+0x121/0x160 net/ipv4/ip_input.c:413
 NF_HOOK include/linux/netfilter.h:305 [inline]
 NF_HOOK include/linux/netfilter.h:299 [inline]
 ip_rcv+0x18f/0x1a0 net/ipv4/ip_input.c:523
 __netif_receive_skb_one_core+0xa7/0xe0 net/core/dev.c:5004
 __netif_receive_skb+0x37/0xf0 net/core/dev.c:5118
 netif_receive_skb_internal+0x59/0x190 net/core/dev.c:5208
 napi_skb_finish net/core/dev.c:5671 [inline]
 napi_gro_receive+0x28f/0x330 net/core/dev.c:5704
 receive_buf+0x284/0x30b0 drivers/net/virtio_net.c:1061

read to 0xffff888120425770 of 4 bytes by task 7254 on cpu 1:
 tcp_stream_is_readable net/ipv4/tcp.c:480 [inline]
 tcp_poll+0x204/0x6b0 net/ipv4/tcp.c:554
 sock_poll+0xed/0x250 net/socket.c:1256
 vfs_poll include/linux/poll.h:90 [inline]
 ep_item_poll.isra.0+0x90/0x190 fs/eventpoll.c:892
 ep_send_events_proc+0x113/0x5c0 fs/eventpoll.c:1749
 ep_scan_ready_list.constprop.0+0x189/0x500 fs/eventpoll.c:704
 ep_send_events fs/eventpoll.c:1793 [inline]
 ep_poll+0xe3/0x900 fs/eventpoll.c:1930
 do_epoll_wait+0x162/0x180 fs/eventpoll.c:2294
 __do_sys_epoll_pwait fs/eventpoll.c:2325 [inline]
 __se_sys_epoll_pwait fs/eventpoll.c:2311 [inline]
 __x64_sys_epoll_pwait+0xcd/0x170 fs/eventpoll.c:2311
 do_syscall_64+0xcf/0x2f0 arch/x86/entry/common.c:296
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

Reported by Kernel Concurrency Sanitizer on:
CPU: 1 PID: 7254 Comm: syz-fuzzer Not tainted 5.3.0+ #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reported-by: Nsyzbot <syzkaller@googlegroups.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      dba7d9b8
    • E
      tcp: add rcu protection around tp->fastopen_rsk · d983ea6f
      由 Eric Dumazet 提交于 
      Both tcp_v4_err() and tcp_v6_err() do the following operations
while they do not own the socket lock :

	fastopen = tp->fastopen_rsk;
 	snd_una = fastopen ? tcp_rsk(fastopen)->snt_isn : tp->snd_una;

The problem is that without appropriate barrier, the compiler
might reload tp->fastopen_rsk and trigger a NULL deref.

request sockets are protected by RCU, we can simply add
the missing annotations and barriers to solve the issue.

Fixes: 168a8f58 ("tcp: TCP Fast Open Server - main code path")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d983ea6f
  29. 27 9月, 2019 3 次提交
  31. 31 7月, 2019 1 次提交
  33. 12 7月, 2019 1 次提交
  35. 09 7月, 2019 1 次提交
    • W
      ipv6: elide flowlabel check if no exclusive leases exist · 59c820b2
      由 Willem de Bruijn 提交于 
      Processes can request ipv6 flowlabels with cmsg IPV6_FLOWINFO.
If not set, by default an autogenerated flowlabel is selected.

Explicit flowlabels require a control operation per label plus a
datapath check on every connection (every datagram if unconnected).
This is particularly expensive on unconnected sockets multiplexing
many flows, such as QUIC.

In the common case, where no lease is exclusive, the check can be
safely elided, as both lease request and check trivially succeed.
Indeed, autoflowlabel does the same even with exclusive leases.

Elide the check if no process has requested an exclusive lease.

fl6_sock_lookup previously returns either a reference to a lease or
NULL to denote failure. Modify to return a real error and update
all callers. On return NULL, they can use the label and will elide
the atomic_dec in fl6_sock_release.

This is an optimization. Robust applications still have to revert to
requesting leases if the fast path fails due to an exclusive lease.

Changes RFC->v1:
  - use static_key_false_deferred to rate limit jump label operations
    - call static_key_deferred_flush to stop timers on exit
  - move decrement out of RCU context
  - defer optimization also if opt data is associated with a lease
  - updated all fp6_sock_lookup callers, not just udp
Signed-off-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      59c820b2
  37. 02 7月, 2019 1 次提交
  39. 15 6月, 2019 1 次提交
  41. 13 6月, 2019 1 次提交
    • E
      tcp: add optional per socket transmit delay · a842fe14
      由 Eric Dumazet 提交于 
      Adding delays to TCP flows is crucial for studying behavior
of TCP stacks, including congestion control modules.

Linux offers netem module, but it has unpractical constraints :
- Need root access to change qdisc
- Hard to setup on egress if combined with non trivial qdisc like FQ
- Single delay for all flows.

EDT (Earliest Departure Time) adoption in TCP stack allows us
to enable a per socket delay at a very small cost.

Networking tools can now establish thousands of flows, each of them
with a different delay, simulating real world conditions.

This requires FQ packet scheduler or a EDT-enabled NIC.

This patchs adds TCP_TX_DELAY socket option, to set a delay in
usec units.

  unsigned int tx_delay = 10000; /* 10 msec */

  setsockopt(fd, SOL_TCP, TCP_TX_DELAY, &tx_delay, sizeof(tx_delay));

Note that FQ packet scheduler limits might need some tweaking :

man tc-fq

PARAMETERS
   limit
       Hard  limit  on  the  real  queue  size. When this limit is
       reached, new packets are dropped. If the value is  lowered,
       packets  are  dropped so that the new limit is met. Default
       is 10000 packets.

   flow_limit
       Hard limit on the maximum  number  of  packets  queued  per
       flow.  Default value is 100.

Use of TCP_TX_DELAY option will increase number of skbs in FQ qdisc,
so packets would be dropped if any of the previous limit is hit.

Use of a jump label makes this support runtime-free, for hosts
never using the option.

Also note that TSQ (TCP Small Queues) limits are slightly changed
with this patch : we need to account that skbs artificially delayed
wont stop us providind more skbs to feed the pipe (netem uses
skb_orphan_partial() for this purpose, but FQ can not use this trick)

Because of that, using big delays might very well trigger
old bugs in TSO auto defer logic and/or sndbuf limited detection.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      a842fe14
  43. 10 6月, 2019 2 次提交
  45. 06 6月, 2019 2 次提交
  47. 31 5月, 2019 1 次提交
  49. 06 5月, 2019 2 次提交
  51. 02 4月, 2019 1 次提交
    • E
      tcp: fix tcp_inet6_sk() for 32bit kernels · f5d54767
      由 Eric Dumazet 提交于 
      It turns out that struct ipv6_pinfo is not located as we think.

inet6_sk_generic() and tcp_inet6_sk() disagree on 32bit kernels by 4-bytes,
because struct tcp_sock has 8-bytes alignment,
but ipv6_pinfo size is not a multiple of 8.

sizeof(struct ipv6_pinfo): 116 (not padded to 8)

I actually first coded tcp_inet6_sk() as this patch does, but thought
that "container_of(tcp_sk(sk), struct tcp6_sock, tcp)" was cleaner.

As Julian told me : Nobody should use tcp6_sock.inet6
directly, it should be accessed via tcp_inet6_sk() or inet6_sk().

This happened when we added the first u64 field in struct tcp_sock.

Fixes: 93a77c11 ("tcp: add tcp_inet6_sk() helper")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Bisected-by: NJulian Anastasov <ja@ssi.bg>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      f5d54767
  53. 24 3月, 2019 1 次提交
    • E
      tcp: add one skb cache for rx · 8b27dae5
      由 Eric Dumazet 提交于 
      Often times, recvmsg() system calls and BH handling for a particular
TCP socket are done on different cpus.

This means the incoming skb had to be allocated on a cpu,
but freed on another.

This incurs a high spinlock contention in slab layer for small rpc,
but also a high number of cache line ping pongs for larger packets.

A full size GRO packet might use 45 page fragments, meaning
that up to 45 put_page() can be involved.

More over performing the __kfree_skb() in the recvmsg() context
adds a latency for user applications, and increase probability
of trapping them in backlog processing, since the BH handler
might found the socket owned by the user.

This patch, combined with the prior one increases the rpc
performance by about 10 % on servers with large number of cores.

(tcp_rr workload with 10,000 flows and 112 threads reach 9 Mpps
 instead of 8 Mpps)

This also increases single bulk flow performance on 40Gbit+ links,
since in this case there are often two cpus working in tandem :

 - CPU handling the NIC rx interrupts, feeding the receive queue,
  and (after this patch) freeing the skbs that were consumed.

 - CPU in recvmsg() system call, essentially 100 % busy copying out
  data to user space.

Having at most one skb in a per-socket cache has very little risk
of memory exhaustion, and since it is protected by socket lock,
its management is essentially free.

Note that if rps/rfs is used, we do not enable this feature, because
there is high chance that the same cpu is handling both the recvmsg()
system call and the TCP rx path, but that another cpu did the skb
allocations in the device driver right before the RPS/RFS logic.

To properly handle this case, it seems we would need to record
on which cpu skb was allocated, and use a different channel
to give skbs back to this cpu.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NSoheil Hassas Yeganeh <soheil@google.com>
Acked-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      8b27dae5
  55. 20 3月, 2019 2 次提交