When dumping proxy entries the dump request has NTF_PROXY set in
ndm_flags. strict mode checking needs to be updated to allow this
flag.
Fixes: 51183d233b ("net/neighbor: Update neigh_dump_info for strict data checking")
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Michael and Sandipan report:
Commit ede95a63b5 introduced a bpf_jit_limit tuneable to limit BPF
JIT allocations. At compile time it defaults to PAGE_SIZE * 40000,
and is adjusted again at init time if MODULES_VADDR is defined.
For ppc64 kernels, MODULES_VADDR isn't defined, so we're stuck with
the compile-time default at boot-time, which is 0x9c400000 when
using 64K page size. This overflows the signed 32-bit bpf_jit_limit
value:
root@ubuntu:/tmp# cat /proc/sys/net/core/bpf_jit_limit
-1673527296
and can cause various unexpected failures throughout the network
stack. In one case `strace dhclient eth0` reported:
setsockopt(5, SOL_SOCKET, SO_ATTACH_FILTER, {len=11, filter=0x105dd27f8},
16) = -1 ENOTSUPP (Unknown error 524)
and similar failures can be seen with tools like tcpdump. This doesn't
always reproduce however, and I'm not sure why. The more consistent
failure I've seen is an Ubuntu 18.04 KVM guest booted on a POWER9
host would time out on systemd/netplan configuring a virtio-net NIC
with no noticeable errors in the logs.
Given this and also given that in near future some architectures like
arm64 will have a custom area for BPF JIT image allocations we should
get rid of the BPF_JIT_LIMIT_DEFAULT fallback / default entirely. For
4.21, we have an overridable bpf_jit_alloc_exec(), bpf_jit_free_exec()
so therefore add another overridable bpf_jit_alloc_exec_limit() helper
function which returns the possible size of the memory area for deriving
the default heuristic in bpf_jit_charge_init().
Like bpf_jit_alloc_exec() and bpf_jit_free_exec(), the new
bpf_jit_alloc_exec_limit() assumes that module_alloc() is the default
JIT memory provider, and therefore in case archs implement their custom
module_alloc() we use MODULES_{END,_VADDR} for limits and otherwise for
vmalloc_exec() cases like on ppc64 we use VMALLOC_{END,_START}.
Additionally, for archs supporting large page sizes, we should change
the sysctl to be handled as long to not run into sysctl restrictions
in future.
Fixes: ede95a63b5 ("bpf: add bpf_jit_limit knob to restrict unpriv allocations")
Reported-by: Sandipan Das <sandipan@linux.ibm.com>
Reported-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
We want to make sure that the following condition holds:
0 <= nhoff <= thoff <= skb->len
BPF program can set out-of-bounds nhoff and thoff, which is dangerous, see
recent commit d0c081b491 ("flow_dissector: properly cap thoff field")'.
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
We are returning thoff from the flow dissector, not the nhoff. Pass
thoff along with nhoff to the bpf program (initially thoff == nhoff)
and expect flow dissector amend/return thoff, not nhoff.
This avoids confusion, when by the time bpf flow dissector exits,
nhoff == thoff, which doesn't make much sense.
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Alexei Starovoitov says:
====================
pull-request: bpf 2018-12-05
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) fix bpf uapi pointers for 32-bit architectures, from Daniel.
2) improve verifer ability to handle progs with a lot of branches, from Alexei.
3) strict btf checks, from Yonghong.
4) bpf_sk_lookup api cleanup, from Joe.
5) other misc fixes
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
netif_napi_add() could report an error like this below due to it allows
to pass a format string for wildcarding before calling
dev_get_valid_name(),
"netif_napi_add() called with weight 256 on device eth%d"
For example, hns_enet_drv module does this.
hns_nic_try_get_ae
hns_nic_init_ring_data
netif_napi_add
register_netdev
dev_get_valid_name
Hence, make it a bit more human-readable by using netdev_err_once()
instead.
Signed-off-by: Qian Cai <cai@gmx.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern and Nicolas Dichtel report that the handling of the netns id
0 is incorrect for the BPF socket lookup helpers: rather than finding
the netns with id 0, it is resolving to the current netns. This renders
the netns_id 0 inaccessible.
To fix this, adjust the API for the netns to treat all negative s32
values as a lookup in the current netns (including u64 values which when
truncated to s32 become negative), while any values with a positive
value in the signed 32-bit integer space would result in a lookup for a
socket in the netns corresponding to that id. As before, if the netns
with that ID does not exist, no socket will be found. Any netns outside
of these ranges will fail to find a corresponding socket, as those
values are reserved for future usage.
Signed-off-by: Joe Stringer <joe@wand.net.nz>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Acked-by: Joey Pabalinas <joeypabalinas@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Currently, pointer offsets in three BPF context structures are
broken in two scenarios: i) 32 bit compiled applications running
on 64 bit kernels, and ii) LLVM compiled BPF programs running
on 32 bit kernels. The latter is due to BPF target machine being
strictly 64 bit. So in each of the cases the offsets will mismatch
in verifier when checking / rewriting context access. Fix this by
providing a helper macro __bpf_md_ptr() that will enforce padding
up to 64 bit and proper alignment, and for context access a macro
bpf_ctx_range_ptr() which will cover full 64 bit member range on
32 bit archs. For flow_keys, we additionally need to force the
size check to sizeof(__u64) as with other pointer types.
Fixes: d58e468b11 ("flow_dissector: implements flow dissector BPF hook")
Fixes: 4f738adba3 ("bpf: create tcp_bpf_ulp allowing BPF to monitor socket TX/RX data")
Fixes: 2dbb9b9e6d ("bpf: Introduce BPF_PROG_TYPE_SK_REUSEPORT")
Reported-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: David S. Miller <davem@davemloft.net>
Tested-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Commit 04157469b7 ("net: Use static_key for XPS maps") introduced a
static key for XPS, but the increments/decrements don't match.
First, the static key's counter is incremented once for each queue, but
only decremented once for a whole batch of queues, leading to large
unbalances.
Second, the xps_rxqs_needed key is decremented whenever we reset a batch
of queues, whether they had any rxqs mapping or not, so that if we setup
cpu-XPS on em1 and RXQS-XPS on em2, resetting the queues on em1 would
decrement the xps_rxqs_needed key.
This reworks the accounting scheme so that the xps_needed key is
incremented only once for each type of XPS for all the queues on a
device, and the xps_rxqs_needed key is incremented only once for all
queues. This is sufficient to let us retrieve queues via
get_xps_queue().
This patch introduces a new reset_xps_maps(), which reinitializes and
frees the appropriate map (xps_rxqs_map or xps_cpus_map), and drops a
reference to the needed keys:
- both xps_needed and xps_rxqs_needed, in case of rxqs maps,
- only xps_needed, in case of CPU maps.
Now, we also need to call reset_xps_maps() at the end of
__netif_set_xps_queue() when there's no active map left, for example
when writing '00000000,00000000' to all queues' xps_rxqs setting.
Fixes: 04157469b7 ("net: Use static_key for XPS maps")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Before commit 80d19669ec ("net: Refactor XPS for CPUs and Rx queues"),
netif_reset_xps_queues() did netdev_queue_numa_node_write() for all the
queues being reset. Now, this is only done when the "active" variable in
clean_xps_maps() is false, ie when on all the CPUs, there's no active
XPS mapping left.
Fixes: 80d19669ec ("net: Refactor XPS for CPUs and Rx queues")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf 2018-11-25
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) Fix an off-by-one bug when adjusting subprog start offsets after
patching, from Edward.
2) Fix several bugs such as overflow in size allocation in queue /
stack map creation, from Alexei.
3) Fix wrong IPv6 destination port byte order in bpf_sk_lookup_udp
helper, from Andrey.
4) Fix several bugs in bpftool such as preventing an infinite loop
in get_fdinfo, error handling and man page references, from Quentin.
5) Fix a warning in bpf_trace_printk() that wasn't catching an
invalid format string, from Martynas.
6) Fix a bug in BPF cgroup local storage where non-atomic allocation
was used in atomic context, from Roman.
7) Fix a NULL pointer dereference bug in bpftool from reallocarray()
error handling, from Jakub and Wen.
8) Add a copy of pkt_cls.h and tc_bpf.h uapi headers to the tools
include infrastructure so that bpftool compiles on older RHEL7-like
user space which does not ship these headers, from Yonghong.
9) Fix BPF kselftests for user space where to get ping test working
with ping6 and ping -6, from Li.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric noted that with UDP GRO and NAPI timeout, we could keep a single
UDP packet inside the GRO hash forever, if the related NAPI instance
calls napi_gro_complete() at an higher frequency than the NAPI timeout.
Willem noted that even TCP packets could be trapped there, till the
next retransmission.
This patch tries to address the issue, flushing the old packets -
those with a NAPI_GRO_CB age before the current jiffy - before scheduling
the NAPI timeout. The rationale is that such a timeout should be
well below a jiffy and we are not flushing packets eligible for sane GRO.
v1 -> v2:
- clarified the commit message and comment
RFC -> v1:
- added 'Fixes tags', cleaned-up the wording.
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Fixes: 3b47d30396 ("net: gro: add a per device gro flush timer")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a packet is trapped and the corresponding SKB marked as
already-forwarded, it retains this marking even after it is forwarded
across veth links into another bridge. There, since it ingresses the
bridge over veth, which doesn't have offload_fwd_mark, it triggers a
warning in nbp_switchdev_frame_mark().
Then nbp_switchdev_allowed_egress() decides not to allow egress from
this bridge through another veth, because the SKB is already marked, and
the mark (of 0) of course matches. Thus the packet is incorrectly
blocked.
Solve by resetting offload_fwd_mark() in skb_scrub_packet(). That
function is called from tunnels and also from veth, and thus catches the
cases where traffic is forwarded between bridges and transformed in a
way that invalidates the marking.
Fixes: 6bc506b4fb ("bridge: switchdev: Add forward mark support for stacked devices")
Fixes: abf4bb6b63 ("skbuff: Add the offload_mr_fwd_mark field")
Signed-off-by: Petr Machata <petrm@mellanox.com>
Suggested-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
eth_type_trans() assumes initial value for skb->pkt_type
is PACKET_HOST.
This is indeed the value right after a fresh skb allocation.
However, it is possible that GRO merged a packet with a different
value (like PACKET_OTHERHOST in case macvlan is used), so
we need to make sure napi->skb will have pkt_type set back to
PACKET_HOST.
Otherwise, valid packets might be dropped by the stack because
their pkt_type is not PACKET_HOST.
napi_reuse_skb() was added in commit 96e93eab20 ("gro: Add
internal interfaces for VLAN"), but this bug always has
been there.
Fixes: 96e93eab20 ("gro: Add internal interfaces for VLAN")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Only first fragment has the sport/dport information,
not the following ones.
If we want consistent hash for all fragments, we need to
ignore ports even for first fragment.
This bug is visible for IPv6 traffic, if incoming fragments
do not have a flow label, since skb_get_hash() will give
different results for first fragment and following ones.
It is also visible if any routing rule wants dissection
and sport or dport.
See commit 5e5d6fed37 ("ipv6: route: dissect flow
in input path if fib rules need it") for details.
[edumazet] rewrote the changelog completely.
Fixes: 06635a35d1 ("flow_dissect: use programable dissector in skb_flow_dissect and friends")
Signed-off-by: 배석진 <soukjin.bae@samsung.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Lookup functions in sk_lookup have different expectations about byte
order of provided arguments.
Specifically __inet_lookup, __udp4_lib_lookup and __udp6_lib_lookup
expect dport to be in network byte order and do ntohs(dport) internally.
At the same time __inet6_lookup expects dport to be in host byte order
and correspondingly name the argument hnum.
sk_lookup works correctly with __inet_lookup, __udp4_lib_lookup and
__inet6_lookup with regard to dport. But in __udp6_lib_lookup case it
uses host instead of expected network byte order. It makes result
returned by bpf_sk_lookup_udp for IPv6 incorrect.
The patch fixes byte order of dport passed to __udp6_lib_lookup.
Originally sk_lookup properly handled UDPv6, but not TCPv6. 5ef0ae84f0
fixes TCPv6 but breaks UDPv6.
Fixes: 5ef0ae84f0 ("bpf: Fix IPv6 dport byte-order in bpf_sk_lookup")
Signed-off-by: Andrey Ignatov <rdna@fb.com>
Acked-by: Joe Stringer <joe@wand.net.nz>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
IPPROTO_RAW isn't registred as an inet protocol, so
inet_protos[protocol] is always NULL for it.
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Xin Long <lucien.xin@gmail.com>
Fixes: bf2ae2e4bf ("sock_diag: request _diag module only when the family or proto has been registered")
Signed-off-by: Andrei Vagin <avagin@gmail.com>
Reviewed-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no reason to discard using source link local address when
remote netconsole IPv6 address is set to be link local one.
The patch allows administrators to use IPv6 netconsole without
explicitly configuring source address:
netconsole=@/,@fe80::5054:ff:fe2f:6012/
Signed-off-by: Matwey V. Kornilov <matwey@sai.msu.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
For non-zero return from dumpit() we should break the loop
in rtnl_dump_all() and return the result. Otherwise, e.g.,
we could get the memory leak in inet6_dump_fib() [1]. The
pointer to the allocated struct fib6_walker there (saved
in cb->args) can be lost, reset on the next iteration.
Fix it by partially restoring the previous behavior before
commit c63586dc9b ("net: rtnl_dump_all needs to propagate
error from dumpit function"). The returned error from
dumpit() is still passed further.
[1]:
unreferenced object 0xffff88001322a200 (size 96):
comm "sshd", pid 1484, jiffies 4296032768 (age 1432.542s)
hex dump (first 32 bytes):
00 01 00 00 00 00 ad de 00 02 00 00 00 00 ad de ................
18 09 41 36 00 88 ff ff 18 09 41 36 00 88 ff ff ..A6......A6....
backtrace:
[<0000000095846b39>] kmem_cache_alloc_trace+0x151/0x220
[<000000007d12709f>] inet6_dump_fib+0x68d/0x940
[<000000002775a316>] rtnl_dump_all+0x1d9/0x2d0
[<00000000d7cd302b>] netlink_dump+0x945/0x11a0
[<000000002f43485f>] __netlink_dump_start+0x55d/0x800
[<00000000f76bbeec>] rtnetlink_rcv_msg+0x4fa/0xa00
[<000000009b5761f3>] netlink_rcv_skb+0x29c/0x420
[<0000000087a1dae1>] rtnetlink_rcv+0x15/0x20
[<00000000691b703b>] netlink_unicast+0x4e3/0x6c0
[<00000000b5be0204>] netlink_sendmsg+0x7f2/0xba0
[<0000000096d2aa60>] sock_sendmsg+0xba/0xf0
[<000000008c1b786f>] __sys_sendto+0x1e4/0x330
[<0000000019587b3f>] __x64_sys_sendto+0xe1/0x1a0
[<00000000071f4d56>] do_syscall_64+0x9f/0x300
[<000000002737577f>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[<0000000057587684>] 0xffffffffffffffff
Fixes: c63586dc9b ("net: rtnl_dump_all needs to propagate error from dumpit function")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Before calling dev_hard_start_xmit(), upper layers tried
to cook optimal skb list based on BQL budget.
Problem is that GSO packets can end up comsuming more than
the BQL budget.
Breaking the loop is not useful, since requeued packets
are ahead of any packets still in the qdisc.
It is also more expensive, since next TX completion will
push these packets later, while skbs are not in cpu caches.
It is also a behavior difference with TSO packets, that can
break the BQL limit by a large amount.
Note that drivers should use __netdev_tx_sent_queue()
in order to have optimal xmit_more support, and avoid
useless atomic operations as shown in the following patch.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove kernel-doc warning:
net/core/skbuff.c:4953: warning: Function parameter or member 'skb' not described in 'skb_gso_size_check'
Signed-off-by: Mathieu Malaterre <malat@debian.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Vakul Garg
David Ahern
Lorenzo Bianconi
Daniel Borkmann
Stanislav Fomichev
David S. Miller
Edward Cree
Eric Dumazet
Qian Cai
Joe Stringer
Sabrina Dubroca
Paolo Abeni
Petr Machata
배석진
Andrey Ignatov
Andrei Vagin
Matwey V. Kornilov
Alexey Kodanev
Mathieu Malaterre