Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 14 10月, 2019 2 次提交
  3. 10 10月, 2019 4 次提交
    • E
      net: silence KCSAN warnings about sk->sk_backlog.len reads · 70c26558
      由 Eric Dumazet 提交于 
      sk->sk_backlog.len can be written by BH handlers, and read
from process contexts in a lockless way.

Note the write side should also use WRITE_ONCE() or a variant.
We need some agreement about the best way to do this.

syzbot reported :

BUG: KCSAN: data-race in tcp_add_backlog / tcp_grow_window.isra.0

write to 0xffff88812665f32c of 4 bytes by interrupt on cpu 1:
 sk_add_backlog include/net/sock.h:934 [inline]
 tcp_add_backlog+0x4a0/0xcc0 net/ipv4/tcp_ipv4.c:1737
 tcp_v4_rcv+0x1aba/0x1bf0 net/ipv4/tcp_ipv4.c:1925
 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
 virtnet_receive drivers/net/virtio_net.c:1323 [inline]
 virtnet_poll+0x436/0x7d0 drivers/net/virtio_net.c:1428
 napi_poll net/core/dev.c:6352 [inline]
 net_rx_action+0x3ae/0xa50 net/core/dev.c:6418

read to 0xffff88812665f32c of 4 bytes by task 7292 on cpu 0:
 tcp_space include/net/tcp.h:1373 [inline]
 tcp_grow_window.isra.0+0x6b/0x480 net/ipv4/tcp_input.c:413
 tcp_event_data_recv+0x68f/0x990 net/ipv4/tcp_input.c:717
 tcp_rcv_established+0xbfe/0xf50 net/ipv4/tcp_input.c:5618
 tcp_v4_do_rcv+0x381/0x4e0 net/ipv4/tcp_ipv4.c:1542
 sk_backlog_rcv include/net/sock.h:945 [inline]
 __release_sock+0x135/0x1e0 net/core/sock.c:2427
 release_sock+0x61/0x160 net/core/sock.c:2943
 tcp_recvmsg+0x63b/0x1a30 net/ipv4/tcp.c:2181
 inet_recvmsg+0xbb/0x250 net/ipv4/af_inet.c:838
 sock_recvmsg_nosec net/socket.c:871 [inline]
 sock_recvmsg net/socket.c:889 [inline]
 sock_recvmsg+0x92/0xb0 net/socket.c:885
 sock_read_iter+0x15f/0x1e0 net/socket.c:967
 call_read_iter include/linux/fs.h:1864 [inline]
 new_sync_read+0x389/0x4f0 fs/read_write.c:414
 __vfs_read+0xb1/0xc0 fs/read_write.c:427
 vfs_read fs/read_write.c:461 [inline]
 vfs_read+0x143/0x2c0 fs/read_write.c:446

Reported by Kernel Concurrency Sanitizer on:
CPU: 0 PID: 7292 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: NJakub Kicinski <jakub.kicinski@netronome.com>
      70c26558
    • E
      net: annotate sk->sk_rcvlowat lockless reads · eac66402
      由 Eric Dumazet 提交于 
      sock_rcvlowat() or int_sk_rcvlowat() might be called without the socket
lock for example from tcp_poll().

Use READ_ONCE() to document the fact that other cpus might change
sk->sk_rcvlowat under us and avoid KCSAN splats.

Use WRITE_ONCE() on write sides too.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
      eac66402
    • E
      net: silence KCSAN warnings around sk_add_backlog() calls · 8265792b
      由 Eric Dumazet 提交于 
      sk_add_backlog() callers usually read sk->sk_rcvbuf without
owning the socket lock. This means sk_rcvbuf value can
be changed by other cpus, and KCSAN complains.

Add READ_ONCE() annotations to document the lockless nature
of these reads.

Note that writes over sk_rcvbuf should also use WRITE_ONCE(),
but this will be done in separate patches to ease stable
backports (if we decide this is relevant for stable trees).

BUG: KCSAN: data-race in tcp_add_backlog / tcp_recvmsg

write to 0xffff88812ab369f8 of 8 bytes by interrupt on cpu 1:
 __sk_add_backlog include/net/sock.h:902 [inline]
 sk_add_backlog include/net/sock.h:933 [inline]
 tcp_add_backlog+0x45a/0xcc0 net/ipv4/tcp_ipv4.c:1737
 tcp_v4_rcv+0x1aba/0x1bf0 net/ipv4/tcp_ipv4.c:1925
 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
 virtnet_receive drivers/net/virtio_net.c:1323 [inline]
 virtnet_poll+0x436/0x7d0 drivers/net/virtio_net.c:1428
 napi_poll net/core/dev.c:6352 [inline]
 net_rx_action+0x3ae/0xa50 net/core/dev.c:6418

read to 0xffff88812ab369f8 of 8 bytes by task 7271 on cpu 0:
 tcp_recvmsg+0x470/0x1a30 net/ipv4/tcp.c:2047
 inet_recvmsg+0xbb/0x250 net/ipv4/af_inet.c:838
 sock_recvmsg_nosec net/socket.c:871 [inline]
 sock_recvmsg net/socket.c:889 [inline]
 sock_recvmsg+0x92/0xb0 net/socket.c:885
 sock_read_iter+0x15f/0x1e0 net/socket.c:967
 call_read_iter include/linux/fs.h:1864 [inline]
 new_sync_read+0x389/0x4f0 fs/read_write.c:414
 __vfs_read+0xb1/0xc0 fs/read_write.c:427
 vfs_read fs/read_write.c:461 [inline]
 vfs_read+0x143/0x2c0 fs/read_write.c:446
 ksys_read+0xd5/0x1b0 fs/read_write.c:587
 __do_sys_read fs/read_write.c:597 [inline]
 __se_sys_read fs/read_write.c:595 [inline]
 __x64_sys_read+0x4c/0x60 fs/read_write.c:595
 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: 0 PID: 7271 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: NJakub Kicinski <jakub.kicinski@netronome.com>
      8265792b
    • E
      net: avoid possible false sharing in sk_leave_memory_pressure() · 503978ac
      由 Eric Dumazet 提交于 
      As mentioned in https://github.com/google/ktsan/wiki/READ_ONCE-and-WRITE_ONCE#it-may-improve-performance
a C compiler can legally transform :

if (memory_pressure && *memory_pressure)
        *memory_pressure = 0;

to :

if (memory_pressure)
        *memory_pressure = 0;

Fixes: 06044751 ("tcp: add TCPMemoryPressuresChrono counter")
Fixes: 180d8cd9 ("foundations of per-cgroup memory pressure controlling.")
Fixes: 3ab224be ("[NET] CORE: Introducing new memory accounting interface.")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
      503978ac
  5. 05 10月, 2019 1 次提交
  7. 01 10月, 2019 1 次提交
    • M
      net: Unpublish sk from sk_reuseport_cb before call_rcu · 8c7138b3
      由 Martin KaFai Lau 提交于 
      The "reuse->sock[]" array is shared by multiple sockets.  The going away
sk must unpublish itself from "reuse->sock[]" before making call_rcu()
call.  However, this unpublish-action is currently done after a grace
period and it may cause use-after-free.

The fix is to move reuseport_detach_sock() to sk_destruct().
Due to the above reason, any socket with sk_reuseport_cb has
to go through the rcu grace period before freeing it.

It is a rather old bug (~3 yrs).  The Fixes tag is not necessary
the right commit but it is the one that introduced the SOCK_RCU_FREE
logic and this fix is depending on it.

Fixes: a4298e45 ("net: add SOCK_RCU_FREE socket flag")
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Suggested-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      8c7138b3
  9. 25 8月, 2019 1 次提交
  11. 18 8月, 2019 1 次提交
  13. 09 8月, 2019 1 次提交
    • J
      net/tls: prevent skb_orphan() from leaking TLS plain text with offload · 41477662
      由 Jakub Kicinski 提交于 
      sk_validate_xmit_skb() and drivers depend on the sk member of
struct sk_buff to identify segments requiring encryption.
Any operation which removes or does not preserve the original TLS
socket such as skb_orphan() or skb_clone() will cause clear text
leaks.

Make the TCP socket underlying an offloaded TLS connection
mark all skbs as decrypted, if TLS TX is in offload mode.
Then in sk_validate_xmit_skb() catch skbs which have no socket
(or a socket with no validation) and decrypted flag set.

Note that CONFIG_SOCK_VALIDATE_XMIT, CONFIG_TLS_DEVICE and
sk->sk_validate_xmit_skb are slightly interchangeable right now,
they all imply TLS offload. The new checks are guarded by
CONFIG_TLS_DEVICE because that's the option guarding the
sk_buff->decrypted member.

Second, smaller issue with orphaning is that it breaks
the guarantee that packets will be delivered to device
queues in-order. All TLS offload drivers depend on that
scheduling property. This means skb_orphan_partial()'s
trick of preserving partial socket references will cause
issues in the drivers. We need a full orphan, and as a
result netem delay/throttling will cause all TLS offload
skbs to be dropped.

Reusing the sk_buff->decrypted flag also protects from
leaking clear text when incoming, decrypted skb is redirected
(e.g. by TC).

See commit 0608c69c ("bpf: sk_msg, sock{map|hash} redirect
through ULP") for justification why the internal flag is safe.
The only location which could leak the flag in is tcp_bpf_sendmsg(),
which is taken care of by clearing the previously unused bit.

v2:
 - remove superfluous decrypted mark copy (Willem);
 - remove the stale doc entry (Boris);
 - rely entirely on EOR marking to prevent coalescing (Boris);
 - use an internal sendpages flag instead of marking the socket
   (Boris).
v3 (Willem):
 - reorganize the can_skb_orphan_partial() condition;
 - fix the flag leak-in through tcp_bpf_sendmsg.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: NWillem de Bruijn <willemb@google.com>
Reviewed-by: NBoris Pismenny <borisp@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      41477662
  15. 13 7月, 2019 1 次提交
    • A
      mm: security: introduce init_on_alloc=1 and init_on_free=1 boot options · 6471384a
      由 Alexander Potapenko 提交于 
      Patch series "add init_on_alloc/init_on_free boot options", v10.

Provide init_on_alloc and init_on_free boot options.

These are aimed at preventing possible information leaks and making the
control-flow bugs that depend on uninitialized values more deterministic.

Enabling either of the options guarantees that the memory returned by the
page allocator and SL[AU]B is initialized with zeroes.  SLOB allocator
isn't supported at the moment, as its emulation of kmem caches complicates
handling of SLAB_TYPESAFE_BY_RCU caches correctly.

Enabling init_on_free also guarantees that pages and heap objects are
initialized right after they're freed, so it won't be possible to access
stale data by using a dangling pointer.

As suggested by Michal Hocko, right now we don't let the heap users to
disable initialization for certain allocations.  There's not enough
evidence that doing so can speed up real-life cases, and introducing ways
to opt-out may result in things going out of control.

This patch (of 2):

The new options are needed to prevent possible information leaks and make
control-flow bugs that depend on uninitialized values more deterministic.

This is expected to be on-by-default on Android and Chrome OS.  And it
gives the opportunity for anyone else to use it under distros too via the
boot args.  (The init_on_free feature is regularly requested by folks
where memory forensics is included in their threat models.)

init_on_alloc=1 makes the kernel initialize newly allocated pages and heap
objects with zeroes.  Initialization is done at allocation time at the
places where checks for __GFP_ZERO are performed.

init_on_free=1 makes the kernel initialize freed pages and heap objects
with zeroes upon their deletion.  This helps to ensure sensitive data
doesn't leak via use-after-free accesses.

Both init_on_alloc=1 and init_on_free=1 guarantee that the allocator
returns zeroed memory.  The two exceptions are slab caches with
constructors and SLAB_TYPESAFE_BY_RCU flag.  Those are never
zero-initialized to preserve their semantics.

Both init_on_alloc and init_on_free default to zero, but those defaults
can be overridden with CONFIG_INIT_ON_ALLOC_DEFAULT_ON and
CONFIG_INIT_ON_FREE_DEFAULT_ON.

If either SLUB poisoning or page poisoning is enabled, those options take
precedence over init_on_alloc and init_on_free: initialization is only
applied to unpoisoned allocations.

Slowdown for the new features compared to init_on_free=0, init_on_alloc=0:

hackbench, init_on_free=1:  +7.62% sys time (st.err 0.74%)
hackbench, init_on_alloc=1: +7.75% sys time (st.err 2.14%)

Linux build with -j12, init_on_free=1:  +8.38% wall time (st.err 0.39%)
Linux build with -j12, init_on_free=1:  +24.42% sys time (st.err 0.52%)
Linux build with -j12, init_on_alloc=1: -0.13% wall time (st.err 0.42%)
Linux build with -j12, init_on_alloc=1: +0.57% sys time (st.err 0.40%)

The slowdown for init_on_free=0, init_on_alloc=0 compared to the baseline
is within the standard error.

The new features are also going to pave the way for hardware memory
tagging (e.g.  arm64's MTE), which will require both on_alloc and on_free
hooks to set the tags for heap objects.  With MTE, tagging will have the
same cost as memory initialization.

Although init_on_free is rather costly, there are paranoid use-cases where
in-memory data lifetime is desired to be minimized.  There are various
arguments for/against the realism of the associated threat models, but
given that we'll need the infrastructure for MTE anyway, and there are
people who want wipe-on-free behavior no matter what the performance cost,
it seems reasonable to include it in this series.

[glider@google.com: v8]
  Link: http://lkml.kernel.org/r/20190626121943.131390-2-glider@google.com
[glider@google.com: v9]
  Link: http://lkml.kernel.org/r/20190627130316.254309-2-glider@google.com
[glider@google.com: v10]
  Link: http://lkml.kernel.org/r/20190628093131.199499-2-glider@google.com
Link: http://lkml.kernel.org/r/20190617151050.92663-2-glider@google.comSigned-off-by: NAlexander Potapenko <glider@google.com>
Acked-by: NKees Cook <keescook@chromium.org>
Acked-by: Michal Hocko <mhocko@suse.cz>		[page and dmapool parts
Acked-by: James Morris <jamorris@linux.microsoft.com>]
Cc: Christoph Lameter <cl@linux.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Cc: "Serge E. Hallyn" <serge@hallyn.com>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Kostya Serebryany <kcc@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Sandeep Patil <sspatil@android.com>
Cc: Laura Abbott <labbott@redhat.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Jann Horn <jannh@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Marco Elver <elver@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      6471384a
  17. 09 7月, 2019 1 次提交
    • A
      coallocate socket_wq with socket itself · 333f7909
      由 Al Viro 提交于 
      socket->wq is assign-once, set when we are initializing both
struct socket it's in and struct socket_wq it points to.  As the
matter of fact, the only reason for separate allocation was the
ability to RCU-delay freeing of socket_wq.  RCU-delaying the
freeing of socket itself gets rid of that need, so we can just
fold struct socket_wq into the end of struct socket and simplify
the life both for sock_alloc_inode() (one allocation instead of
two) and for tun/tap oddballs, where we used to embed struct socket
and struct socket_wq into the same structure (now - embedding just
the struct socket).

Note that reference to struct socket_wq in struct sock does remain
a reference - that's unchanged.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      333f7909
  19. 19 6月, 2019 1 次提交
  21. 15 6月, 2019 2 次提交
    • E
      net: add high_order_alloc_disable sysctl/static key · ce27ec60
      由 Eric Dumazet 提交于 
      >From linux-3.7, (commit 5640f768 "net: use a per task frag
allocator") TCP sendmsg() has preferred using order-3 allocations.

While it gives good results for most cases, we had reports
that heavy uses of TCP over loopback were hitting a spinlock
contention in page allocations/freeing.

This commits adds a sysctl so that admins can opt-in
for order-0 allocations. Hopefully mm layer might optimize
order-3 allocations in the future since it could give us
a nice boost  (see 8 lines of following benchmark)

The following benchmark shows a win when more than 8 TCP_STREAM
threads are running (56 x86 cores server in my tests)

for thr in {1..30}
do
 sysctl -wq net.core.high_order_alloc_disable=0
 T0=`./super_netperf $thr -H 127.0.0.1 -l 15`
 sysctl -wq net.core.high_order_alloc_disable=1
 T1=`./super_netperf $thr -H 127.0.0.1 -l 15`
 echo $thr:$T0:$T1
done

1: 49979: 37267
2: 98745: 76286
3: 141088: 110051
4: 177414: 144772
5: 197587: 173563
6: 215377: 208448
7: 241061: 234087
8: 267155: 263373
9: 295069: 297402
10: 312393: 335213
11: 340462: 368778
12: 371366: 403954
13: 412344: 443713
14: 426617: 473580
15: 474418: 507861
16: 503261: 538539
17: 522331: 563096
18: 532409: 567084
19: 550824: 605240
20: 525493: 641988
21: 564574: 665843
22: 567349: 690868
23: 583846: 710917
24: 588715: 736306
25: 603212: 763494
26: 604083: 792654
27: 602241: 796450
28: 604291: 797993
29: 611610: 833249
30: 577356: 841062
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      ce27ec60
    • M
      bpf: net: Add SO_DETACH_REUSEPORT_BPF · 99f3a064
      由 Martin KaFai Lau 提交于 
      There is SO_ATTACH_REUSEPORT_[CE]BPF but there is no DETACH.
This patch adds SO_DETACH_REUSEPORT_BPF sockopt.  The same
sockopt can be used to undo both SO_ATTACH_REUSEPORT_[CE]BPF.

reseport_detach_prog() is added and it is mostly a mirror
of the existing reuseport_attach_prog().  The differences are,
it does not call reuseport_alloc() and returns -ENOENT when
there is no old prog.

Cc: Craig Gallek <kraig@google.com>
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Reviewed-by: NStanislav Fomichev <sdf@google.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
      99f3a064
  23. 12 6月, 2019 1 次提交
  25. 31 5月, 2019 1 次提交
  27. 28 4月, 2019 1 次提交
    • M
      bpf: Introduce bpf sk local storage · 6ac99e8f
      由 Martin KaFai Lau 提交于 
      After allowing a bpf prog to
- directly read the skb->sk ptr
- get the fullsock bpf_sock by "bpf_sk_fullsock()"
- get the bpf_tcp_sock by "bpf_tcp_sock()"
- get the listener sock by "bpf_get_listener_sock()"
- avoid duplicating the fields of "(bpf_)sock" and "(bpf_)tcp_sock"
  into different bpf running context.

this patch is another effort to make bpf's network programming
more intuitive to do (together with memory and performance benefit).

When bpf prog needs to store data for a sk, the current practice is to
define a map with the usual 4-tuples (src/dst ip/port) as the key.
If multiple bpf progs require to store different sk data, multiple maps
have to be defined.  Hence, wasting memory to store the duplicated
keys (i.e. 4 tuples here) in each of the bpf map.
[ The smallest key could be the sk pointer itself which requires
  some enhancement in the verifier and it is a separate topic. ]

Also, the bpf prog needs to clean up the elem when sk is freed.
Otherwise, the bpf map will become full and un-usable quickly.
The sk-free tracking currently could be done during sk state
transition (e.g. BPF_SOCK_OPS_STATE_CB).

The size of the map needs to be predefined which then usually ended-up
with an over-provisioned map in production.  Even the map was re-sizable,
while the sk naturally come and go away already, this potential re-size
operation is arguably redundant if the data can be directly connected
to the sk itself instead of proxy-ing through a bpf map.

This patch introduces sk->sk_bpf_storage to provide local storage space
at sk for bpf prog to use.  The space will be allocated when the first bpf
prog has created data for this particular sk.

The design optimizes the bpf prog's lookup (and then optionally followed by
an inline update).  bpf_spin_lock should be used if the inline update needs
to be protected.

BPF_MAP_TYPE_SK_STORAGE:
-----------------------
To define a bpf "sk-local-storage", a BPF_MAP_TYPE_SK_STORAGE map (new in
this patch) needs to be created.  Multiple BPF_MAP_TYPE_SK_STORAGE maps can
be created to fit different bpf progs' needs.  The map enforces
BTF to allow printing the sk-local-storage during a system-wise
sk dump (e.g. "ss -ta") in the future.

The purpose of a BPF_MAP_TYPE_SK_STORAGE map is not for lookup/update/delete
a "sk-local-storage" data from a particular sk.
Think of the map as a meta-data (or "type") of a "sk-local-storage".  This
particular "type" of "sk-local-storage" data can then be stored in any sk.

The main purposes of this map are mostly:
1. Define the size of a "sk-local-storage" type.
2. Provide a similar syscall userspace API as the map (e.g. lookup/update,
   map-id, map-btf...etc.)
3. Keep track of all sk's storages of this "type" and clean them up
   when the map is freed.

sk->sk_bpf_storage:
------------------
The main lookup/update/delete is done on sk->sk_bpf_storage (which
is a "struct bpf_sk_storage").  When doing a lookup,
the "map" pointer is now used as the "key" to search on the
sk_storage->list.  The "map" pointer is actually serving
as the "type" of the "sk-local-storage" that is being
requested.

To allow very fast lookup, it should be as fast as looking up an
array at a stable-offset.  At the same time, it is not ideal to
set a hard limit on the number of sk-local-storage "type" that the
system can have.  Hence, this patch takes a cache approach.
The last search result from sk_storage->list is cached in
sk_storage->cache[] which is a stable sized array.  Each
"sk-local-storage" type has a stable offset to the cache[] array.
In the future, a map's flag could be introduced to do cache
opt-out/enforcement if it became necessary.

The cache size is 16 (i.e. 16 types of "sk-local-storage").
Programs can share map.  On the program side, having a few bpf_progs
running in the networking hotpath is already a lot.  The bpf_prog
should have already consolidated the existing sock-key-ed map usage
to minimize the map lookup penalty.  16 has enough runway to grow.

All sk-local-storage data will be removed from sk->sk_bpf_storage
during sk destruction.

bpf_sk_storage_get() and bpf_sk_storage_delete():
------------------------------------------------
Instead of using bpf_map_(lookup|update|delete)_elem(),
the bpf prog needs to use the new helper bpf_sk_storage_get() and
bpf_sk_storage_delete().  The verifier can then enforce the
ARG_PTR_TO_SOCKET argument.  The bpf_sk_storage_get() also allows to
"create" new elem if one does not exist in the sk.  It is done by
the new BPF_SK_STORAGE_GET_F_CREATE flag.  An optional value can also be
provided as the initial value during BPF_SK_STORAGE_GET_F_CREATE.
The BPF_MAP_TYPE_SK_STORAGE also supports bpf_spin_lock.  Together,
it has eliminated the potential use cases for an equivalent
bpf_map_update_elem() API (for bpf_prog) in this patch.

Misc notes:
----------
1. map_get_next_key is not supported.  From the userspace syscall
   perspective,  the map has the socket fd as the key while the map
   can be shared by pinned-file or map-id.

   Since btf is enforced, the existing "ss" could be enhanced to pretty
   print the local-storage.

   Supporting a kernel defined btf with 4 tuples as the return key could
   be explored later also.

2. The sk->sk_lock cannot be acquired.  Atomic operations is used instead.
   e.g. cmpxchg is done on the sk->sk_bpf_storage ptr.
   Please refer to the source code comments for the details in
   synchronization cases and considerations.

3. The mem is charged to the sk->sk_omem_alloc as the sk filter does.

Benchmark:
---------
Here is the benchmark data collected by turning on
the "kernel.bpf_stats_enabled" sysctl.
Two bpf progs are tested:

One bpf prog with the usual bpf hashmap (max_entries = 8192) with the
sk ptr as the key. (verifier is modified to support sk ptr as the key
That should have shortened the key lookup time.)

Another bpf prog is with the new BPF_MAP_TYPE_SK_STORAGE.

Both are storing a "u32 cnt", do a lookup on "egress_skb/cgroup" for
each egress skb and then bump the cnt.  netperf is used to drive
data with 4096 connected UDP sockets.

BPF_MAP_TYPE_HASH with a modifier verifier (152ns per bpf run)
27: cgroup_skb  name egress_sk_map  tag 74f56e832918070b run_time_ns 58280107540 run_cnt 381347633
    loaded_at 2019-04-15T13:46:39-0700  uid 0
    xlated 344B  jited 258B  memlock 4096B  map_ids 16
    btf_id 5

BPF_MAP_TYPE_SK_STORAGE in this patch (66ns per bpf run)
30: cgroup_skb  name egress_sk_stora  tag d4aa70984cc7bbf6 run_time_ns 25617093319 run_cnt 390989739
    loaded_at 2019-04-15T13:47:54-0700  uid 0
    xlated 168B  jited 156B  memlock 4096B  map_ids 17
    btf_id 6

Here is a high-level picture on how are the objects organized:

       sk
    ┌──────┐
    │      │
    │      │
    │      │
    │*sk_bpf_storage─────▶ bpf_sk_storage
    └──────┘                 ┌───────┐
                 ┌───────────┤ list  │
                 │           │       │
                 │           │       │
                 │           │       │
                 │           └───────┘
                 │
                 │     elem
                 │  ┌────────┐
                 ├─▶│ snode  │
                 │  ├────────┤
                 │  │  data  │          bpf_map
                 │  ├────────┤        ┌─────────┐
                 │  │map_node│◀─┬─────┤  list   │
                 │  └────────┘  │     │         │
                 │              │     │         │
                 │     elem     │     │         │
                 │  ┌────────┐  │     └─────────┘
                 └─▶│ snode  │  │
                    ├────────┤  │
   bpf_map          │  data  │  │
 ┌─────────┐        ├────────┤  │
 │  list   ├───────▶│map_node│  │
 │         │        └────────┘  │
 │         │                    │
 │         │           elem     │
 └─────────┘        ┌────────┐  │
                 ┌─▶│ snode  │  │
                 │  ├────────┤  │
                 │  │  data  │  │
                 │  ├────────┤  │
                 │  │map_node│◀─┘
                 │  └────────┘
                 │
                 │
                 │          ┌───────┐
     sk          └──────────│ list  │
  ┌──────┐                  │       │
  │      │                  │       │
  │      │                  │       │
  │      │                  └───────┘
  │*sk_bpf_storage───────▶bpf_sk_storage
  └──────┘
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
      6ac99e8f
  29. 24 4月, 2019 1 次提交
  31. 20 4月, 2019 1 次提交
  33. 17 4月, 2019 1 次提交
  35. 02 3月, 2019 2 次提交
  37. 25 2月, 2019 1 次提交
  39. 17 2月, 2019 1 次提交
    • G
      sock: consistent handling of extreme SO_SNDBUF/SO_RCVBUF values · 4057765f
      由 Guillaume Nault 提交于 
      SO_SNDBUF and SO_RCVBUF (and their *BUFFORCE version) may overflow or
underflow their input value. This patch aims at providing explicit
handling of these extreme cases, to get a clear behaviour even with
values bigger than INT_MAX / 2 or lower than INT_MIN / 2.

For simplicity, only SO_SNDBUF and SO_SNDBUFFORCE are described here,
but the same explanation and fix apply to SO_RCVBUF and SO_RCVBUFFORCE
(with 'SNDBUF' replaced by 'RCVBUF' and 'wmem_max' by 'rmem_max').

Overflow of positive values

===========================

When handling SO_SNDBUF or SO_SNDBUFFORCE, if 'val' exceeds
INT_MAX / 2, the buffer size is set to its minimum value because
'val * 2' overflows, and max_t() considers that it's smaller than
SOCK_MIN_SNDBUF. For SO_SNDBUF, this can only happen with
net.core.wmem_max > INT_MAX / 2.

SO_SNDBUF and SO_SNDBUFFORCE are actually designed to let users probe
for the maximum buffer size by setting an arbitrary large number that
gets capped to the maximum allowed/possible size. Having the upper
half of the positive integer space to potentially reduce the buffer
size to its minimum value defeats this purpose.

This patch caps the base value to INT_MAX / 2, so that bigger values
don't overflow and keep setting the buffer size to its maximum.

Underflow of negative values
============================

For negative numbers, SO_SNDBUF always considers them bigger than
net.core.wmem_max, which is bounded by [SOCK_MIN_SNDBUF, INT_MAX].
Therefore such values are set to net.core.wmem_max and we're back to
the behaviour of positive integers described above (return maximum
buffer size if wmem_max <= INT_MAX / 2, return SOCK_MIN_SNDBUF
otherwise).

However, SO_SNDBUFFORCE behaves differently. The user value is
directly multiplied by two and compared with SOCK_MIN_SNDBUF. If
'val * 2' doesn't underflow or if it underflows to a value smaller
than SOCK_MIN_SNDBUF then buffer size is set to its minimum value.
Otherwise the buffer size is set to the underflowed value.

This patch treats negative values passed to SO_SNDBUFFORCE as null, to
prevent underflows. Therefore negative values now always set the buffer
size to its minimum value.

Even though SO_SNDBUF behaves inconsistently by setting buffer size to
the maximum value when passed a negative number, no attempt is made to
modify this behaviour. There may exist some programs that rely on using
negative numbers to set the maximum buffer size. Avoiding overflows
because of extreme net.core.wmem_max values is the most we can do here.

Summary of altered behaviours
=============================

val      : user-space value passed to setsockopt()
val_uf   : the underflowed value resulting from doubling val when
           val < INT_MIN / 2
wmem_max : short for net.core.wmem_max
val_cap  : min(val, wmem_max)
min_len  : minimal buffer length (that is, SOCK_MIN_SNDBUF)
max_len  : maximal possible buffer length, regardless of wmem_max (that
           is, INT_MAX - 1)
^^^^     : altered behaviour

SO_SNDBUF:
+-------------------------+-------------+------------+----------------+
|       CONDITION         | OLD RESULT  | NEW RESULT |    COMMENT     |
+-------------------------+-------------+------------+----------------+
| val < 0 &&              |             |            | No overflow,   |
| wmem_max <= INT_MAX/2   | wmem_max*2  | wmem_max*2 | keep original  |
|                         |             |            | behaviour      |
+-------------------------+-------------+------------+----------------+
| val < 0 &&              |             |            | Cap wmem_max   |
| INT_MAX/2 < wmem_max    | min_len     | max_len    | to prevent     |
|                         |             | ^^^^^^^    | overflow       |
+-------------------------+-------------+------------+----------------+
| 0 <= val <= min_len/2   | min_len     | min_len    | Ordinary case  |
+-------------------------+-------------+------------+----------------+
| min_len/2 < val &&      | val_cap*2   | val_cap*2  | Ordinary case  |
| val_cap <= INT_MAX/2    |             |            |                |
+-------------------------+-------------+------------+----------------+
| min_len < val &&        |             |            | Cap val_cap    |
| INT_MAX/2 < val_cap     | min_len     | max_len    | again to       |
| (implies that           |             | ^^^^^^^    | prevent        |
| INT_MAX/2 < wmem_max)   |             |            | overflow       |
+-------------------------+-------------+------------+----------------+

SO_SNDBUFFORCE:
+------------------------------+---------+---------+------------------+
|          CONDITION           | BEFORE  | AFTER   |     COMMENT      |
|                              | PATCH   | PATCH   |                  |
+------------------------------+---------+---------+------------------+
| val < INT_MIN/2 &&           | min_len | min_len | Underflow with   |
| val_uf <= min_len            |         |         | no consequence   |
+------------------------------+---------+---------+------------------+
| val < INT_MIN/2 &&           | val_uf  | min_len | Set val to 0 to  |
| val_uf > min_len             |         | ^^^^^^^ | avoid underflow  |
+------------------------------+---------+---------+------------------+
| INT_MIN/2 <= val < 0         | min_len | min_len | No underflow     |
+------------------------------+---------+---------+------------------+
| 0 <= val <= min_len/2        | min_len | min_len | Ordinary case    |
+------------------------------+---------+---------+------------------+
| min_len/2 < val <= INT_MAX/2 | val*2   | val*2   | Ordinary case    |
+------------------------------+---------+---------+------------------+
| INT_MAX/2 < val              | min_len | max_len | Cap val to       |
|                              |         | ^^^^^^^ | prevent overflow |
+------------------------------+---------+---------+------------------+
Signed-off-by: NGuillaume Nault <gnault@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      4057765f
  41. 14 2月, 2019 1 次提交
  43. 04 2月, 2019 7 次提交
  45. 20 1月, 2019 1 次提交
  47. 18 1月, 2019 1 次提交
    • D
      net: introduce SO_BINDTOIFINDEX sockopt · f5dd3d0c
      由 David Herrmann 提交于 
      This introduces a new generic SOL_SOCKET-level socket option called
SO_BINDTOIFINDEX. It behaves similar to SO_BINDTODEVICE, but takes a
network interface index as argument, rather than the network interface
name.

User-space often refers to network-interfaces via their index, but has
to temporarily resolve it to a name for a call into SO_BINDTODEVICE.
This might pose problems when the network-device is renamed
asynchronously by other parts of the system. When this happens, the
SO_BINDTODEVICE might either fail, or worse, it might bind to the wrong
device.

In most cases user-space only ever operates on devices which they
either manage themselves, or otherwise have a guarantee that the device
name will not change (e.g., devices that are UP cannot be renamed).
However, particularly in libraries this guarantee is non-obvious and it
would be nice if that race-condition would simply not exist. It would
make it easier for those libraries to operate even in situations where
the device-name might change under the hood.

A real use-case that we recently hit is trying to start the network
stack early in the initrd but make it survive into the real system.
Existing distributions rename network-interfaces during the transition
from initrd into the real system. This, obviously, cannot affect
devices that are up and running (unless you also consider moving them
between network-namespaces). However, the network manager now has to
make sure its management engine for dormant devices will not run in
parallel to these renames. Particularly, when you offload operations
like DHCP into separate processes, these might setup their sockets
early, and thus have to resolve the device-name possibly running into
this race-condition.

By avoiding a call to resolve the device-name, we no longer depend on
the name and can run network setup of dormant devices in parallel to
the transition off the initrd. The SO_BINDTOIFINDEX ioctl plugs this
race.
Reviewed-by: NTom Gundersen <teg@jklm.no>
Signed-off-by: NDavid Herrmann <dh.herrmann@gmail.com>
Acked-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      f5dd3d0c
  49. 02 1月, 2019 1 次提交
    • D
      sock: Make sock->sk_stamp thread-safe · 3a0ed3e9
      由 Deepa Dinamani 提交于 
      Al Viro mentioned (Message-ID
<20170626041334.GZ10672@ZenIV.linux.org.uk>)
that there is probably a race condition
lurking in accesses of sk_stamp on 32-bit machines.

sock->sk_stamp is of type ktime_t which is always an s64.
On a 32 bit architecture, we might run into situations of
unsafe access as the access to the field becomes non atomic.

Use seqlocks for synchronization.
This allows us to avoid using spinlocks for readers as
readers do not need mutual exclusion.

Another approach to solve this is to require sk_lock for all
modifications of the timestamps. The current approach allows
for timestamps to have their own lock: sk_stamp_lock.
This allows for the patch to not compete with already
existing critical sections, and side effects are limited
to the paths in the patch.

The addition of the new field maintains the data locality
optimizations from
commit 9115e8cd ("net: reorganize struct sock for better data
locality")

Note that all the instances of the sk_stamp accesses
are either through the ioctl or the syscall recvmsg.
Signed-off-by: NDeepa Dinamani <deepa.kernel@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      3a0ed3e9
  51. 08 12月, 2018 1 次提交
    • Y
      net: call sk_dst_reset when set SO_DONTROUTE · 0fbe82e6
      由 yupeng 提交于 
      after set SO_DONTROUTE to 1, the IP layer should not route packets if
the dest IP address is not in link scope. But if the socket has cached
the dst_entry, such packets would be routed until the sk_dst_cache
expires. So we should clean the sk_dst_cache when a user set
SO_DONTROUTE option. Below are server/client python scripts which
could reprodue this issue:

server side code:

==========================================================================
import socket
import struct
import time

s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.bind(('0.0.0.0', 9000))
s.listen(1)
sock, addr = s.accept()
sock.setsockopt(socket.SOL_SOCKET, socket.SO_DONTROUTE, struct.pack('i', 1))
while True:
    sock.send(b'foo')
    time.sleep(1)
==========================================================================

client side code:
==========================================================================
import socket
import time

s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.connect(('server_address', 9000))
while True:
    data = s.recv(1024)
    print(data)
==========================================================================
Signed-off-by: Nyupeng <yupeng0921@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      0fbe82e6
  53. 04 12月, 2018 1 次提交
    • W
      udp: msg_zerocopy · b5947e5d
      由 Willem de Bruijn 提交于 
      Extend zerocopy to udp sockets. Allow setting sockopt SO_ZEROCOPY and
interpret flag MSG_ZEROCOPY.

This patch was previously part of the zerocopy RFC patchsets. Zerocopy
is not effective at small MTU. With segmentation offload building
larger datagrams, the benefit of page flipping outweights the cost of
generating a completion notification.

tools/testing/selftests/net/msg_zerocopy.sh after applying follow-on
test patch and making skb_orphan_frags_rx same as skb_orphan_frags:

    ipv4 udp -t 1
    tx=191312 (11938 MB) txc=0 zc=n
    rx=191312 (11938 MB)
    ipv4 udp -z -t 1
    tx=304507 (19002 MB) txc=304507 zc=y
    rx=304507 (19002 MB)
    ok
    ipv6 udp -t 1
    tx=174485 (10888 MB) txc=0 zc=n
    rx=174485 (10888 MB)
    ipv6 udp -z -t 1
    tx=294801 (18396 MB) txc=294801 zc=y
    rx=294801 (18396 MB)
    ok

Changes
  v1 -> v2
    - Fixup reverse christmas tree violation
  v2 -> v3
    - Split refcount avoidance optimization into separate patch
      - Fix refcount leak on error in fragmented case
        (thanks to Paolo Abeni for pointing this one out!)
      - Fix refcount inc on zero
      - Test sock_flag SOCK_ZEROCOPY directly in __ip_append_data.
        This is needed since commit 5cf4a853 ("tcp: really ignore
	MSG_ZEROCOPY if no SO_ZEROCOPY") did the same for tcp.
Signed-off-by: NWillem de Bruijn <willemb@google.com>
Acked-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      b5947e5d
  55. 09 11月, 2018 1 次提交