Pull ceph fix from Ilya Dryomov: "A fix for a nasty use-after-free, marked for stable" * tag 'ceph-for-5.18-rc8' of https://github.com/ceph/ceph-client: libceph: fix misleading ceph_osdc_cancel_request() comment libceph: fix potential use-after-free on linger ping and resends
Pull networking fixes from Paolo Abeni:
"Including fixes from can, xfrm and netfilter subtrees.
Notably this reverts a recent TCP/DCCP netns-related change to address
a possible UaF.
Current release - regressions:
- tcp: revert "tcp/dccp: get rid of inet_twsk_purge()"
- xfrm: set dst dev to blackhole_netdev instead of loopback_dev in
ifdown
Previous releases - regressions:
- netfilter: flowtable: fix TCP flow teardown
- can: revert "can: m_can: pci: use custom bit timings for Elkhart
Lake"
- xfrm: check encryption module availability consistency
- eth: vmxnet3: fix possible use-after-free bugs in
vmxnet3_rq_alloc_rx_buf()
- eth: mlx5: initialize flow steering during driver probe
- eth: ice: fix crash when writing timestamp on RX rings
Previous releases - always broken:
- mptcp: fix checksum byte order
- eth: lan966x: fix assignment of the MAC address
- eth: mlx5: remove HW-GRO from reported features
- eth: ftgmac100: disable hardware checksum on AST2600"
* tag 'net-5.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (50 commits)
net: bridge: Clear offload_fwd_mark when passing frame up bridge interface.
ptp: ocp: change sysfs attr group handling
selftests: forwarding: fix missing backslash
netfilter: nf_tables: disable expression reduction infra
netfilter: flowtable: move dst_check to packet path
netfilter: flowtable: fix TCP flow teardown
net: ftgmac100: Disable hardware checksum on AST2600
igb: skip phy status check where unavailable
nfc: pn533: Fix buggy cleanup order
mptcp: Do TCP fallback on early DSS checksum failure
mptcp: fix checksum byte order
net: af_key: check encryption module availability consistency
net: af_key: add check for pfkey_broadcast in function pfkey_process
net/mlx5: Drain fw_reset when removing device
net/mlx5e: CT: Fix setting flow_source for smfs ct tuples
net/mlx5e: CT: Fix support for GRE tuples
net/mlx5e: Remove HW-GRO from reported features
net/mlx5e: Properly block HW GRO when XDP is enabled
net/mlx5e: Properly block LRO when XDP is enabled
net/mlx5e: Block rx-gro-hw feature in switchdev mode
...
It is possible to stack bridges on top of each other. Consider the
following which makes use of an Ethernet switch:
br1
/ \
/ \
/ \
br0.11 wlan0
|
br0
/ | \
p1 p2 p3
br0 is offloaded to the switch. Above br0 is a vlan interface, for
vlan 11. This vlan interface is then a slave of br1. br1 also has a
wireless interface as a slave. This setup trunks wireless lan traffic
over the copper network inside a VLAN.
A frame received on p1 which is passed up to the bridge has the
skb->offload_fwd_mark flag set to true, indicating that the switch has
dealt with forwarding the frame out ports p2 and p3 as needed. This
flag instructs the software bridge it does not need to pass the frame
back down again. However, the flag is not getting reset when the frame
is passed upwards. As a result br1 sees the flag, wrongly interprets
it, and fails to forward the frame to wlan0.
When passing a frame upwards, clear the flag. This is the Rx
equivalent of br_switchdev_frame_unmark() in br_dev_xmit().
Fixes: f1c2eddf4cb6 ("bridge: switchdev: Use an helper to clear forward mark")
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Tested-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://lore.kernel.org/r/20220518005840.771575-1-andrew@lunn.ch
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Pablo Neira Ayuso says: ==================== Netfilter fixes for net 1) Reduce number of hardware offload retries from flowtable datapath which might hog system with retries, from Felix Fietkau. 2) Skip neighbour lookup for PPPoE device, fill_forward_path() already provides this and set on destination address from fill_forward_path for PPPoE device, also from Felix. 4) When combining PPPoE on top of a VLAN device, set info->outdev to the PPPoE device so software offload works, from Felix. 5) Fix TCP teardown flowtable state, races with conntrack gc might result in resetting the state to ESTABLISHED and the time to one day. Joint work with Oz Shlomo and Sven Auhagen. 6) Call dst_check() from flowtable datapath to check if dst is stale instead of doing it from garbage collector path. 7) Disable register tracking infrastructure, either user-space or kernel need to pre-fetch keys inconditionally, otherwise register tracking assumes data is already available in register that might not well be there, leading to incorrect reductions. * git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf: netfilter: nf_tables: disable expression reduction infra netfilter: flowtable: move dst_check to packet path netfilter: flowtable: fix TCP flow teardown netfilter: nft_flow_offload: fix offload with pppoe + vlan net: fix dev_fill_forward_path with pppoe + bridge netfilter: nft_flow_offload: skip dst neigh lookup for ppp devices netfilter: flowtable: fix excessive hw offload attempts after failure ==================== Link: https://lore.kernel.org/r/20220518213841.359653-1-pablo@netfilter.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
cancel_request() never guaranteed that after its return the OSD client would be completely done with the OSD request. The callback (if specified) can still be invoked and a ref can still be held. Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Reviewed-by: Xiubo Li <xiubli@redhat.com>
request_reinit() is not only ugly as the comment rightfully suggests,
but also unsafe. Even though it is called with osdc->lock held for
write in all cases, resetting the OSD request refcount can still race
with handle_reply() and result in use-after-free. Taking linger ping
as an example:
handle_timeout thread handle_reply thread
down_read(&osdc->lock)
req = lookup_request(...)
...
finish_request(req) # unregisters
up_read(&osdc->lock)
__complete_request(req)
linger_ping_cb(req)
# req->r_kref == 2 because handle_reply still holds its ref
down_write(&osdc->lock)
send_linger_ping(lreq)
req = lreq->ping_req # same req
# cancel_linger_request is NOT
# called - handle_reply already
# unregistered
request_reinit(req)
WARN_ON(req->r_kref != 1) # fires
request_init(req)
kref_init(req->r_kref)
# req->r_kref == 1 after kref_init
ceph_osdc_put_request(req)
kref_put(req->r_kref)
# req->r_kref == 0 after kref_put, req is freed
<further req initialization/use> !!!
This happens because send_linger_ping() always (re)uses the same OSD
request for watch ping requests, relying on cancel_linger_request() to
unregister it from the OSD client and rip its messages out from the
messenger. send_linger() does the same for watch/notify registration
and watch reconnect requests. Unfortunately cancel_request() doesn't
guarantee that after it returns the OSD client would be completely done
with the OSD request -- a ref could still be held and the callback (if
specified) could still be invoked too.
The original motivation for request_reinit() was inability to deal with
allocation failures in send_linger() and send_linger_ping(). Switching
to using osdc->req_mempool (currently only used by CephFS) respects that
and allows us to get rid of request_reinit().
Cc: stable@vger.kernel.org
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Reviewed-by: Xiubo Li <xiubli@redhat.com>
Acked-by: Jeff Layton <jlayton@kernel.org>
Either userspace or kernelspace need to pre-fetch keys inconditionally
before comparisons for this to work. Otherwise, register tracking data
is misleading and it might result in reducing expressions which are not
yet registers.
First expression is also guaranteed to be evaluated always, however,
certain expressions break before writing data to registers, before
comparing the data, leaving the register in undetermined state.
This patch disables this infrastructure by now.
Fixes: b2d306542ff9 ("netfilter: nf_tables: do not reduce read-only expressions")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Fixes sporadic IPv6 packet loss when flow offloading is enabled.
IPv6 route GC and flowtable GC are not synchronized.
When dst_cache becomes stale and a packet passes through the flow before
the flowtable GC teardowns it, the packet can be dropped.
So, it is necessary to check dst every time in packet path.
Fixes: 227e1e4d0d6c ("netfilter: nf_flowtable: skip device lookup from interface index")
Signed-off-by: Ritaro Takenaka <ritarot634@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch addresses three possible problems:
1. ct gc may race to undo the timeout adjustment of the packet path, leaving
the conntrack entry in place with the internal offload timeout (one day).
2. ct gc removes the ct because the IPS_OFFLOAD_BIT is not set and the CLOSE
timeout is reached before the flow offload del.
3. tcp ct is always set to ESTABLISHED with a very long timeout
in flow offload teardown/delete even though the state might be already
CLOSED. Also as a remark we cannot assume that the FIN or RST packet
is hitting flow table teardown as the packet might get bumped to the
slow path in nftables.
This patch resets IPS_OFFLOAD_BIT from flow_offload_teardown(), so
conntrack handles the tcp rst/fin packet which triggers the CLOSE/FIN
state transition.
Moreover, teturn the connection's ownership to conntrack upon teardown
by clearing the offload flag and fixing the established timeout value.
The flow table GC thread will asynchonrnously free the flow table and
hardware offload entries.
Before this patch, the IPS_OFFLOAD_BIT remained set for expired flows on
which is also misleading since the flow is back to classic conntrack
path.
If nf_ct_delete() removes the entry from the conntrack table, then it
calls nf_ct_put() which decrements the refcnt. This is not a problem
because the flowtable holds a reference to the conntrack object from
flow_offload_alloc() path which is released via flow_offload_free().
This patch also updates nft_flow_offload to skip packets in SYN_RECV
state. Since we might miss or bump packets to slow path, we do not know
what will happen there while we are still in SYN_RECV, this patch
postpones offload up to the next packet which also aligns to the
existing behaviour in tc-ct.
flow_offload_teardown() does not reset the existing tcp state from
flow_offload_fixup_tcp() to ESTABLISHED anymore, packets bump to slow
path might have already update the state to CLOSE/FIN.
Joint work with Oz and Sven.
Fixes: 1e5b2471bcc4 ("netfilter: nf_flow_table: teardown flow timeout race")
Signed-off-by: Oz Shlomo <ozsh@nvidia.com>
Signed-off-by: Sven Auhagen <sven.auhagen@voleatech.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
RFC 8684 section 3.7 describes several opportunities for a MPTCP
connection to "fall back" to regular TCP early in the connection
process, before it has been confirmed that MPTCP options can be
successfully propagated on all SYN, SYN/ACK, and data packets. If a peer
acknowledges the first received data packet with a regular TCP header
(no MPTCP options), fallback is allowed.
If the recipient of that first data packet finds a MPTCP DSS checksum
error, this provides an opportunity to fail gracefully with a TCP
fallback rather than resetting the connection (as might happen if a
checksum failure were detected later).
This commit modifies the checksum failure code to attempt fallback on
the initial subflow of a MPTCP connection, only if it's a failure in the
first data mapping. In cases where the peer initiates the connection,
requests checksums, is the first to send data, and the peer is sending
incorrect checksums (see
https://github.com/multipath-tcp/mptcp_net-next/issues/275), this allows
the connection to proceed as TCP rather than reset.
Fixes: dd8bcd1768ff ("mptcp: validate the data checksum")
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>