In commit f84af32cbc ("net: ip_queue_rcv_skb() helper")
I dropped the skb dst in tcp_data_queue().
This only dealt with so-called TCP input slow path.
When fast path is taken, tcp_rcv_established() calls
tcp_queue_rcv() while skb still has a dst.
This was mostly fine, because most dsts at this point
are not refcounted (thanks to early demux)
However, TCP packets sent over loopback have refcounted dst.
Then commit 68822bdf76 ("net: generalize skb freeing
deferral to per-cpu lists") came and had the effect
of delaying skb freeing for an arbitrary time.
If during this time the involved netns is dismantled, cleanup_net()
frees the struct net with embedded net->ipv6.ip6_dst_ops.
Then when eventually dst_destroy_rcu() is called,
if (dst->ops->destroy) ... triggers an use-after-free.
It is not clear if ip6_route_net_exit() lacks a rcu_barrier()
as syzbot reported similar issues before the blamed commit.
( https://groups.google.com/g/syzkaller-bugs/c/CofzW4eeA9A/m/009WjumTAAAJ )
Fixes: 68822bdf76 ("net: generalize skb freeing deferral to per-cpu lists")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Throw neigh checks in ip6_finish_output2() under a single slow path if,
so we don't have the overhead in the hot path.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are two callers of __ip6_finish_output(), both are in
ip6_finish_output(). We can combine the call sites into one and handle
return code after, that will inline __ip6_finish_output().
Note, error handling under NET_XMIT_CN will only return 0 if
__ip6_finish_output() succeded, and in this case it return 0.
Considering that NET_XMIT_SUCCESS is 0, it'll be returning exactly the
same result for it as before.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Inline dev_queue_xmit() and dev_queue_xmit_accel(), they both are small
proxy functions doing nothing but redirecting the control flow to
__dev_queue_xmit().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_zerocopy_iter_dgram() is a small proxy function, inline it. For
that, move __zerocopy_sg_from_iter into linux/skbuff.h
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sock_alloc_send_skb() is simple and just proxying to another function,
so we can inline it and cut associated overhead.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Merge net branch with the required patch for supporting the io_uring
feature that passes back whether we had more data in the socket or not.
* 'tcp-pass-back-data-left-in-socket-after-receive' of git://git.kernel.org/pub/scm/linux/kernel/git/kuba/linux:
tcp: pass back data left in socket after receive
* for-5.19/io_uring-socket: (73 commits)
io_uring: use the text representation of ops in trace
io_uring: rename op -> opcode
io_uring: add io_uring_get_opcode
io_uring: add type to op enum
io_uring: add socket(2) support
net: add __sys_socket_file()
io_uring: fix trace for reduced sqe padding
io_uring: add fgetxattr and getxattr support
io_uring: add fsetxattr and setxattr support
fs: split off do_getxattr from getxattr
fs: split off setxattr_copy and do_setxattr function from setxattr
io_uring: return an error when cqe is dropped
io_uring: use constants for cq_overflow bitfield
io_uring: rework io_uring_enter to simplify return value
io_uring: trace cqe overflows
io_uring: add trace support for CQE overflow
io_uring: allow re-poll if we made progress
io_uring: support MSG_WAITALL for IORING_OP_SEND(MSG)
io_uring: add support for IORING_ASYNC_CANCEL_ANY
io_uring: allow IORING_OP_ASYNC_CANCEL with 'fd' key
...
This is currently done for CMSG_INQ, add an ability to do so via struct
msghdr as well and have CMSG_INQ use that too. If the caller sets
msghdr->msg_get_inq, then we'll pass back the hint in msghdr->msg_inq.
Rearrange struct msghdr a bit so we can add this member while shrinking
it at the same time. On a 64-bit build, it was 96 bytes before this
change and 88 bytes afterwards.
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/650c22ca-cffc-0255-9a05-2413a1e20826@kernel.dk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This reverts commit 7073ea8799.
We must not try to connect the socket while the transport is under
construction, because the mechanisms to safely tear it down are not in
place. As the code stands, we end up leaking the sockets on a connection
error.
Reported-by: wanghai (M) <wanghai38@huawei.com>
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
The new net.mptcp.pm_type sysctl determines which path manager will be
used by each newly-created MPTCP socket.
v2: Handle builds without CONFIG_SYSCTL
v3: Clarify logic for type-specific PM init (Geliang Tang and Paolo Abeni)
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Userspace-managed sockets should not have their subflows or
advertisements changed by the kernel path manager.
v3: Use helper function for PM mode (Paolo Abeni)
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When a MPTCP connection is managed by a userspace PM, bypass the kernel
PM for incoming advertisements and subflow events. Netlink events are
still sent to userspace.
v2: Remove unneeded check in mptcp_pm_rm_addr_received() (Kishen Maloor)
v3: Add and use helper function for PM mode (Paolo Abeni)
Acked-by: Paolo Abeni <pabeni@redhat.com>
Co-developed-by: Kishen Maloor <kishen.maloor@intel.com>
Signed-off-by: Kishen Maloor <kishen.maloor@intel.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When adding support for netlink path management commands, the kernel
needs to know whether paths are being controlled by the in-kernel path
manager or a userspace PM.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A few members of the mptcp_pm_data struct were assigned to hard-coded
values in mptcp_pm_data_reset(), and then immediately changed in
mptcp_pm_nl_data_init().
Instead, flatten all the assignments in to mptcp_pm_data_reset().
v2: Resolve conflicts due to rename of mptcp_pm_data_reset()
v4: Resolve conflict in mptcp_pm_data_reset()
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pull ceph client fixes from Ilya Dryomov:
"A fix for a NULL dereference that turns out to be easily triggerable
by fsync (marked for stable) and a false positive WARN and snap_rwsem
locking fixups"
* tag 'ceph-for-5.18-rc5' of https://github.com/ceph/ceph-client:
ceph: fix possible NULL pointer dereference for req->r_session
ceph: remove incorrect session state check
ceph: get snap_rwsem read lock in handle_cap_export for ceph_add_cap
libceph: disambiguate cluster/pool full log message
For reasons best known to the author, gss-proxy does not implement a
NULL procedure, and returns RPC_PROC_UNAVAIL. However we still want to
ensure that we connect to the service at setup time.
So add a quirk-flag specially for this case.
Fixes: 1d658336b0 ("SUNRPC: Add RPC based upcall mechanism for RPCGSS auth")
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
When the rpcbind server closes the socket, we need to ensure that the
socket is closed by the kernel as soon as feasible, so add a
sk_state_change callback to trigger this close.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
As a carry over from the CAN_RAW socket (which allows to change the CAN
interface while mantaining the filter setup) the re-binding of the
CAN_ISOTP socket needs to take care about CAN ID address information and
subscriptions. It turned out that this feature is so limited (e.g. the
sockopts remain fix) that it finally has never been needed/used.
In opposite to the stateless CAN_RAW socket the switching of the CAN ID
subscriptions might additionally lead to an interrupted ongoing PDU
reception. So better remove this unneeded complexity.
Fixes: e057dd3fc2 ("can: add ISO 15765-2:2016 transport protocol")
Link: https://lore.kernel.org/all/20220422082337.1676-1-socketcan@hartkopp.net
Cc: stable@vger.kernel.org
Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
1) Fix incorrect TCP connection tracking window reset for non-syn
packets, from Florian Westphal.
2) Incorrect dependency on CONFIG_NFT_FLOW_OFFLOAD, from Volodymyr Mytnyk.
3) Fix nft_socket from the output path, from Florian Westphal.
* git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
netfilter: nft_socket: only do sk lookups when indev is available
netfilter: conntrack: fix udp offload timeout sysctl
netfilter: nf_conntrack_tcp: re-init for syn packets only
====================
Link: https://lore.kernel.org/r/20220428142109.38726-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
If there is still a closed socket associated with the transport, then we
need to trigger an autoclose before we can set up a new connection.
Reported-by: wanghai (M) <wanghai38@huawei.com>
Fixes: f00432063d ("SUNRPC: Ensure we flush any closed sockets before xs_xprt_free()")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
For additional robustness in the face of Hyper-V errors or malicious
behavior, validate all values that originate from packets that Hyper-V
has sent to the guest in the host-to-guest ring buffer. Ensure that
invalid values cannot cause data being copied out of the bounds of the
source buffer in hvs_stream_dequeue().
Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Link: https://lore.kernel.org/r/20220428145107.7878-4-parri.andrea@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Pointers to VMbus packets sent by Hyper-V are used by the hv_sock driver
within the guest VM. Hyper-V can send packets with erroneous values or
modify packet fields after they are processed by the guest. To defend
against these scenarios, copy the incoming packet after validating its
length and offset fields using hv_pkt_iter_{first,next}(). Use
HVS_PKT_LEN(HVS_MTU_SIZE) to initialize the buffer which holds the
copies of the incoming packets. In this way, the packet can no longer
be modified by the host.
Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Link: https://lore.kernel.org/r/20220428145107.7878-3-parri.andrea@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Check if the incoming interface is available and NFT_BREAK
in case neither skb->sk nor input device are set.
Because nf_sk_lookup_slow*() assume packet headers are in the
'in' direction, use in postrouting is not going to yield a meaningful
result. Same is true for the forward chain, so restrict the use
to prerouting, input and output.
Use in output work if a socket is already attached to the skb.
Fixes: 554ced0a6e ("netfilter: nf_tables: add support for native socket matching")
Reported-and-tested-by: Topi Miettinen <toiwoton@gmail.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Luiz Augusto von Dentz says:
====================
bluetooth pull request for net:
- Fix regression causing some HCI events to be discarded when they
shouldn't.
* tag 'for-net-2022-04-27' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth:
Bluetooth: hci_sync: Cleanup hci_conn if it cannot be aborted
Bluetooth: hci_event: Fix creating hci_conn object on error status
Bluetooth: hci_event: Fix checking for invalid handle on error status
====================
Link: https://lore.kernel.org/r/20220427234031.1257281-1-luiz.dentz@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Daniel Borkmann says:
====================
pull-request: bpf-next 2022-04-27
We've added 85 non-merge commits during the last 18 day(s) which contain
a total of 163 files changed, 4499 insertions(+), 1521 deletions(-).
The main changes are:
1) Teach libbpf to enhance BPF verifier log with human-readable and relevant
information about failed CO-RE relocations, from Andrii Nakryiko.
2) Add typed pointer support in BPF maps and enable it for unreferenced pointers
(via probe read) and referenced ones that can be passed to in-kernel helpers,
from Kumar Kartikeya Dwivedi.
3) Improve xsk to break NAPI loop when rx queue gets full to allow for forward
progress to consume descriptors, from Maciej Fijalkowski & Björn Töpel.
4) Fix a small RCU read-side race in BPF_PROG_RUN routines which dereferenced
the effective prog array before the rcu_read_lock, from Stanislav Fomichev.
5) Implement BPF atomic operations for RV64 JIT, and add libbpf parsing logic
for USDT arguments under riscv{32,64}, from Pu Lehui.
6) Implement libbpf parsing of USDT arguments under aarch64, from Alan Maguire.
7) Enable bpftool build for musl and remove nftw with FTW_ACTIONRETVAL usage
so it can be shipped under Alpine which is musl-based, from Dominique Martinet.
8) Clean up {sk,task,inode} local storage trace RCU handling as they do not
need to use call_rcu_tasks_trace() barrier, from KP Singh.
9) Improve libbpf API documentation and fix error return handling of various
API functions, from Grant Seltzer.
10) Enlarge offset check for bpf_skb_{load,store}_bytes() helpers given data
length of frags + frag_list may surpass old offset limit, from Liu Jian.
11) Various improvements to prog_tests in area of logging, test execution
and by-name subtest selection, from Mykola Lysenko.
12) Simplify map_btf_id generation for all map types by moving this process
to build time with help of resolve_btfids infra, from Menglong Dong.
13) Fix a libbpf bug in probing when falling back to legacy bpf_probe_read*()
helpers; the probing caused always to use old helpers, from Runqing Yang.
14) Add support for ARCompact and ARCv2 platforms for libbpf's PT_REGS
tracing macros, from Vladimir Isaev.
15) Cleanup BPF selftests to remove old & unneeded rlimit code given kernel
switched to memcg-based memory accouting a while ago, from Yafang Shao.
16) Refactor of BPF sysctl handlers to move them to BPF core, from Yan Zhu.
17) Fix BPF selftests in two occasions to work around regressions caused by latest
LLVM to unblock CI until their fixes are worked out, from Yonghong Song.
18) Misc cleanups all over the place, from various others.
* https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (85 commits)
selftests/bpf: Add libbpf's log fixup logic selftests
libbpf: Fix up verifier log for unguarded failed CO-RE relos
libbpf: Simplify bpf_core_parse_spec() signature
libbpf: Refactor CO-RE relo human description formatting routine
libbpf: Record subprog-resolved CO-RE relocations unconditionally
selftests/bpf: Add CO-RE relos and SEC("?...") to linked_funcs selftests
libbpf: Avoid joining .BTF.ext data with BPF programs by section name
libbpf: Fix logic for finding matching program for CO-RE relocation
libbpf: Drop unhelpful "program too large" guess
libbpf: Fix anonymous type check in CO-RE logic
bpf: Compute map_btf_id during build time
selftests/bpf: Add test for strict BTF type check
selftests/bpf: Add verifier tests for kptr
selftests/bpf: Add C tests for kptr
libbpf: Add kptr type tag macros to bpf_helpers.h
bpf: Make BTF type match stricter for release arguments
bpf: Teach verifier about kptr_get kfunc helpers
bpf: Wire up freeing of referenced kptr
bpf: Populate pairs of btf_id and destructor kfunc in btf
bpf: Adapt copy_map_value for multiple offset case
...
====================
Link: https://lore.kernel.org/r/20220427224758.20976-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Calling tls_append_frag when max_open_record_len == record->len might
add an empty fragment to the TLS record if the call happens to be on the
page boundary. Normally tls_append_frag coalesces the zero-sized
fragment to the previous one, but not if it's on page boundary.
If a resync happens then, the mlx5 driver posts dump WQEs in
tx_post_resync_dump, and the empty fragment may become a data segment
with byte_count == 0, which will confuse the NIC and lead to a CQE
error.
This commit fixes the described issue by skipping tls_append_frag on
zero size to avoid adding empty fragments. The fix is not in the driver,
because an empty fragment is hardly the desired behavior.
Fixes: e8f6979981 ("net/tls: Add generic NIC offload infrastructure")
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20220426154949.159055-1-maximmi@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Daniel Borkmann says:
====================
pull-request: bpf 2022-04-27
We've added 5 non-merge commits during the last 20 day(s) which contain
a total of 6 files changed, 34 insertions(+), 12 deletions(-).
The main changes are:
1) Fix xsk sockets when rx and tx are separately bound to the same umem, also
fix xsk copy mode combined with busy poll, from Maciej Fijalkowski.
2) Fix BPF tunnel/collect_md helpers with bpf_xmit lwt hook usage which triggered
a crash due to invalid metadata_dst access, from Eyal Birger.
3) Fix release of page pool in XDP live packet mode, from Toke Høiland-Jørgensen.
4) Fix potential NULL pointer dereference in kretprobes, from Adam Zabrocki.
(Masami & Steven preferred this small fix to be routed via bpf tree given it's
follow-up fix to Masami's rethook work that went via bpf earlier, too.)
* https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
xsk: Fix possible crash when multiple sockets are created
kprobes: Fix KRETPROBES when CONFIG_KRETPROBE_ON_RETHOOK is set
bpf, lwt: Fix crash when using bpf_skb_set_tunnel_key() from bpf_xmit lwt hook
bpf: Fix release of page_pool in BPF_PROG_RUN in test runner
xsk: Fix l2fwd for copy mode + busy poll combo
====================
Link: https://lore.kernel.org/r/20220427212748.9576-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
`nf_flowtable_udp_timeout` sysctl option is available only
if CONFIG_NFT_FLOW_OFFLOAD enabled. But infra for this flow
offload UDP timeout was added under CONFIG_NF_FLOW_TABLE
config option. So, if you have CONFIG_NFT_FLOW_OFFLOAD
disabled and CONFIG_NF_FLOW_TABLE enabled, the
`nf_flowtable_udp_timeout` is not present in sysfs.
Please note, that TCP flow offload timeout sysctl option
is present even CONFIG_NFT_FLOW_OFFLOAD is disabled.
I suppose it was a typo in commit that adds UDP flow offload
timeout and CONFIG_NF_FLOW_TABLE should be used instead.
Fixes: 975c57504d ("netfilter: conntrack: Introduce udp offload timeout configuration")
Signed-off-by: Volodymyr Mytnyk <volodymyr.mytnyk@plvision.eu>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Jaco Kroon reported tcp problems that Eric Dumazet and Neal Cardwell
pinpointed to nf_conntrack tcp_in_window() bug.
tcp trace shows following sequence:
I > R Flags [S], seq 3451342529, win 62580, options [.. tfo [|tcp]>
R > I Flags [S.], seq 2699962254, ack 3451342530, win 65535, options [..]
R > I Flags [P.], seq 1:89, ack 1, [..]
Note 3rd ACK is from responder to initiator so following branch is taken:
} else if (((state->state == TCP_CONNTRACK_SYN_SENT
&& dir == IP_CT_DIR_ORIGINAL)
|| (state->state == TCP_CONNTRACK_SYN_RECV
&& dir == IP_CT_DIR_REPLY))
&& after(end, sender->td_end)) {
... because state == TCP_CONNTRACK_SYN_RECV and dir is REPLY.
This causes the scaling factor to be reset to 0: window scale option
is only present in syn(ack) packets. This in turn makes nf_conntrack
mark valid packets as out-of-window.
This was always broken, it exists even in original commit where
window tracking was added to ip_conntrack (nf_conntrack predecessor)
in 2.6.9-rc1 kernel.
Restrict to 'tcph->syn', just like the 3rd condtional added in
commit 82b72cb946 ("netfilter: conntrack: re-init state for retransmitted syn-ack").
Upon closer look, those conditionals/branches can be merged:
Because earlier checks prevent syn-ack from showing up in
original direction, the 'dir' checks in the conditional quoted above are
redundant, remove them. Return early for pure syn retransmitted in reply
direction (simultaneous open).
Fixes: 9fb9cbb108 ("[NETFILTER]: Add nf_conntrack subsystem.")
Reported-by: Jaco Kroon <jaco@uls.co.za>
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Jozsef Kadlecsik <kadlec@netfilter.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch adds a new msk->flags bit MPTCP_FAIL_NO_RESPONSE, then reuses
sk_timer to trigger a check if we have not received a response from the
peer after sending MP_FAIL. If the peer doesn't respond properly, reset
the subflow.
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a new struct member mp_fail_response_expect in struct
mptcp_subflow_context to support MP_FAIL response. In the single subflow
with checksum error and contiguous data special case, a MP_FAIL is sent
in response to another MP_FAIL.
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mptcp_data_lock() needs to be held when manipulating the msk
retransmit_timer or the sk sk_timer. This patch adds the data
lock for the both timers.
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The macro dev_core_stats_##FIELD##_inc() disables preemption and invokes
netdev_core_stats_alloc() to return a per-CPU pointer.
netdev_core_stats_alloc() will allocate memory on its first invocation
which breaks on PREEMPT_RT because it requires non-atomic context for
memory allocation.
This can be avoided by enabling preemption in netdev_core_stats_alloc()
assuming the caller always disables preemption.
It might be better to replace local_inc() with this_cpu_inc() now that
dev_core_stats_##FIELD##_inc() gained a preempt-disable section and does
not rely on already disabled preemption. This results in less
instructions on x86-64:
local_inc:
| incl %gs:__preempt_count(%rip) # __preempt_count
| movq 488(%rdi), %rax # _1->core_stats, _22
| testq %rax, %rax # _22
| je .L585 #,
| add %gs:this_cpu_off(%rip), %rax # this_cpu_off, tcp_ptr__
| .L586:
| testq %rax, %rax # _27
| je .L587 #,
| incq (%rax) # _6->a.counter
| .L587:
| decl %gs:__preempt_count(%rip) # __preempt_count
this_cpu_inc(), this patch:
| movq 488(%rdi), %rax # _1->core_stats, _5
| testq %rax, %rax # _5
| je .L591 #,
| .L585:
| incq %gs:(%rax) # _18->rx_dropped
Use unsigned long as type for the counter. Use this_cpu_inc() to
increment the counter. Use a plain read of the counter.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/YmbO0pxgtKpCw4SY@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When NIC takes care of crypto (or the record has already
been decrypted) we forget to update darg->async. ->async
is supposed to mean whether record is async capable on
input and whether record has been queued for async crypto
on output.
Reported-by: Gal Pressman <gal@nvidia.com>
Fixes: 3547a1f9d9 ("tls: rx: use async as an in-out argument")
Tested-by: Gal Pressman <gal@nvidia.com>
Link: https://lore.kernel.org/r/20220425233309.344858-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Logic added in commit f35f821935 ("tcp: defer skb freeing after socket
lock is released") helped bulk TCP flows to move the cost of skbs
frees outside of critical section where socket lock was held.
But for RPC traffic, or hosts with RFS enabled, the solution is far from
being ideal.
For RPC traffic, recvmsg() has to return to user space right after
skb payload has been consumed, meaning that BH handler has no chance
to pick the skb before recvmsg() thread. This issue is more visible
with BIG TCP, as more RPC fit one skb.
For RFS, even if BH handler picks the skbs, they are still picked
from the cpu on which user thread is running.
Ideally, it is better to free the skbs (and associated page frags)
on the cpu that originally allocated them.
This patch removes the per socket anchor (sk->defer_list) and
instead uses a per-cpu list, which will hold more skbs per round.
This new per-cpu list is drained at the end of net_action_rx(),
after incoming packets have been processed, to lower latencies.
In normal conditions, skbs are added to the per-cpu list with
no further action. In the (unlikely) cases where the cpu does not
run net_action_rx() handler fast enough, we use an IPI to raise
NET_RX_SOFTIRQ on the remote cpu.
Also, we do not bother draining the per-cpu list from dev_cpu_dead()
This is because skbs in this list have no requirement on how fast
they should be freed.
Note that we can add in the future a small per-cpu cache
if we see any contention on sd->defer_lock.
Tested on a pair of hosts with 100Gbit NIC, RFS enabled,
and /proc/sys/net/ipv4/tcp_rmem[2] tuned to 16MB to work around
page recycling strategy used by NIC driver (its page pool capacity
being too small compared to number of skbs/pages held in sockets
receive queues)
Note that this tuning was only done to demonstrate worse
conditions for skb freeing for this particular test.
These conditions can happen in more general production workload.
10 runs of one TCP_STREAM flow
Before:
Average throughput: 49685 Mbit.
Kernel profiles on cpu running user thread recvmsg() show high cost for
skb freeing related functions (*)
57.81% [kernel] [k] copy_user_enhanced_fast_string
(*) 12.87% [kernel] [k] skb_release_data
(*) 4.25% [kernel] [k] __free_one_page
(*) 3.57% [kernel] [k] __list_del_entry_valid
1.85% [kernel] [k] __netif_receive_skb_core
1.60% [kernel] [k] __skb_datagram_iter
(*) 1.59% [kernel] [k] free_unref_page_commit
(*) 1.16% [kernel] [k] __slab_free
1.16% [kernel] [k] _copy_to_iter
(*) 1.01% [kernel] [k] kfree
(*) 0.88% [kernel] [k] free_unref_page
0.57% [kernel] [k] ip6_rcv_core
0.55% [kernel] [k] ip6t_do_table
0.54% [kernel] [k] flush_smp_call_function_queue
(*) 0.54% [kernel] [k] free_pcppages_bulk
0.51% [kernel] [k] llist_reverse_order
0.38% [kernel] [k] process_backlog
(*) 0.38% [kernel] [k] free_pcp_prepare
0.37% [kernel] [k] tcp_recvmsg_locked
(*) 0.37% [kernel] [k] __list_add_valid
0.34% [kernel] [k] sock_rfree
0.34% [kernel] [k] _raw_spin_lock_irq
(*) 0.33% [kernel] [k] __page_cache_release
0.33% [kernel] [k] tcp_v6_rcv
(*) 0.33% [kernel] [k] __put_page
(*) 0.29% [kernel] [k] __mod_zone_page_state
0.27% [kernel] [k] _raw_spin_lock
After patch:
Average throughput: 73076 Mbit.
Kernel profiles on cpu running user thread recvmsg() looks better:
81.35% [kernel] [k] copy_user_enhanced_fast_string
1.95% [kernel] [k] _copy_to_iter
1.95% [kernel] [k] __skb_datagram_iter
1.27% [kernel] [k] __netif_receive_skb_core
1.03% [kernel] [k] ip6t_do_table
0.60% [kernel] [k] sock_rfree
0.50% [kernel] [k] tcp_v6_rcv
0.47% [kernel] [k] ip6_rcv_core
0.45% [kernel] [k] read_tsc
0.44% [kernel] [k] _raw_spin_lock_irqsave
0.37% [kernel] [k] _raw_spin_lock
0.37% [kernel] [k] native_irq_return_iret
0.33% [kernel] [k] __inet6_lookup_established
0.31% [kernel] [k] ip6_protocol_deliver_rcu
0.29% [kernel] [k] tcp_rcv_established
0.29% [kernel] [k] llist_reverse_order
v2: kdoc issue (kernel bots)
do not defer if (alloc_cpu == smp_processor_id()) (Paolo)
replace the sk_buff_head with a single-linked list (Jakub)
add a READ_ONCE()/WRITE_ONCE() for the lockless read of sd->defer_list
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Link: https://lore.kernel.org/r/20220422201237.416238-1-eric.dumazet@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
For now, the field 'map_btf_id' in 'struct bpf_map_ops' for all map
types are computed during vmlinux-btf init:
btf_parse_vmlinux() -> btf_vmlinux_map_ids_init()
It will lookup the btf_type according to the 'map_btf_name' field in
'struct bpf_map_ops'. This process can be done during build time,
thanks to Jiri's resolve_btfids.
selftest of map_ptr has passed:
$96 map_ptr:OK
Summary: 1/0 PASSED, 0 SKIPPED, 0 FAILED
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Menglong Dong <imagedong@tencent.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This attempts to cleanup the hci_conn if it cannot be aborted as
otherwise it would likely result in having the controller and host
stack out of sync with respect to connection handle.
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
It is useless to create a hci_conn object if on error status as the
result would be it being freed in the process and anyway it is likely
the result of controller and host stack being out of sync.
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Commit d5ebaa7c5f introduces checks for handle range
(e.g HCI_CONN_HANDLE_MAX) but controllers like Intel AX200 don't seem
to respect the valid range int case of error status:
> HCI Event: Connect Complete (0x03) plen 11
Status: Page Timeout (0x04)
Handle: 65535
Address: 94:DB:56:XX:XX:XX (Sony Home Entertainment&
Sound Products Inc)
Link type: ACL (0x01)
Encryption: Disabled (0x00)
[1644965.827560] Bluetooth: hci0: Ignoring HCI_Connection_Complete for invalid handle
Because of it is impossible to cleanup the connections properly since
the stack would attempt to cancel the connection which is no longer in
progress causing the following trace:
< HCI Command: Create Connection Cancel (0x01|0x0008) plen 6
Address: 94:DB:56:XX:XX:XX (Sony Home Entertainment&
Sound Products Inc)
= bluetoothd: src/profile.c:record_cb() Unable to get Hands-Free Voice
gateway SDP record: Connection timed out
> HCI Event: Command Complete (0x0e) plen 10
Create Connection Cancel (0x01|0x0008) ncmd 1
Status: Unknown Connection Identifier (0x02)
Address: 94:DB:56:XX:XX:XX (Sony Home Entertainment&
Sound Products Inc)
< HCI Command: Create Connection Cancel (0x01|0x0008) plen 6
Address: 94:DB:56:XX:XX:XX (Sony Home Entertainment&
Sound Products Inc)
Fixes: d5ebaa7c5f ("Bluetooth: hci_event: Ignore multiple conn complete events")
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
