[ Upstream commit e44699d2c2 ]
Recently I started seeing warnings about pages with refcount -1. The
problem was traced to packets being reused after their head was merged into
a GRO packet by skb_gro_receive(). While bisecting the issue pointed to
commit c21b48cc1b ("net: adjust skb->truesize in ___pskb_trim()") and
I have never seen it on a kernel with it reverted, I believe the real
problem appeared earlier when the option to merge head frag in GRO was
implemented.
Handling NAPI_GRO_FREE_STOLEN_HEAD state was only added to GRO_MERGED_FREE
branch of napi_skb_finish() so that if the driver uses napi_gro_frags()
and head is merged (which in my case happens after the skb_condense()
call added by the commit mentioned above), the skb is reused including the
head that has been merged. As a result, we release the page reference
twice and eventually end up with negative page refcount.
To fix the problem, handle NAPI_GRO_FREE_STOLEN_HEAD in napi_frags_finish()
the same way it's done in napi_skb_finish().
Fixes: d7e8883cfc ("net: make GRO aware of skb->head_frag")
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 6f64ec7451 ]
Similar to the fix provided by Dominik Heidler in commit
9b3dc0a17d ("l2tp: cast l2tp traffic counter to unsigned")
we need to take care of 32bit kernels in dev_get_stats().
When using atomic_long_read(), we add a 'long' to u64 and
might misinterpret high order bit, unless we cast to unsigned.
Fixes: caf586e5f2 ("net: add a core netdev->rx_dropped counter")
Fixes: 015f0688f5 ("net: net: add a core netdev->tx_dropped counter")
Fixes: 6e7333d315 ("net: add rx_nohandler stat counter")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jarod Wilson <jarod@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit db833d40ad ]
Network interface groups support added while ago, however
there is no IFLA_GROUP attribute description in policy
and netlink message size calculations until now.
Add IFLA_GROUP attribute to the policy.
Fixes: cbda10fa97 ("net_device: add support for network device groups")
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit f186ce61bb ]
It looks like this:
Message from syslogd@flamingo at Apr 26 00:45:00 ...
kernel:unregister_netdevice: waiting for lo to become free. Usage count = 4
They seem to coincide with net namespace teardown.
The message is emitted by netdev_wait_allrefs().
Forced a kdump in netdev_run_todo, but found that the refcount on the lo
device was already 0 at the time we got to the panic.
Used bcc to check the blocking in netdev_run_todo. The only places
where we're off cpu there are in the rcu_barrier() and msleep() calls.
That behavior is expected. The msleep time coincides with the amount of
time we spend waiting for the refcount to reach zero; the rcu_barrier()
wait times are not excessive.
After looking through the list of callbacks that the netdevice notifiers
invoke in this path, it appears that the dst_dev_event is the most
interesting. The dst_ifdown path places a hold on the loopback_dev as
part of releasing the dev associated with the original dst cache entry.
Most of our notifier callbacks are straight-forward, but this one a)
looks complex, and b) places a hold on the network interface in
question.
I constructed a new bcc script that watches various events in the
liftime of a dst cache entry. Note that dst_ifdown will take a hold on
the loopback device until the invalidated dst entry gets freed.
[ __dst_free] on DST: ffff883ccabb7900 IF tap1008300eth0 invoked at 1282115677036183
__dst_free
rcu_nocb_kthread
kthread
ret_from_fork
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit cf124db566 ]
Network devices can allocate reasources and private memory using
netdev_ops->ndo_init(). However, the release of these resources
can occur in one of two different places.
Either netdev_ops->ndo_uninit() or netdev->destructor().
The decision of which operation frees the resources depends upon
whether it is necessary for all netdev refs to be released before it
is safe to perform the freeing.
netdev_ops->ndo_uninit() presumably can occur right after the
NETDEV_UNREGISTER notifier completes and the unicast and multicast
address lists are flushed.
netdev->destructor(), on the other hand, does not run until the
netdev references all go away.
Further complicating the situation is that netdev->destructor()
almost universally does also a free_netdev().
This creates a problem for the logic in register_netdevice().
Because all callers of register_netdevice() manage the freeing
of the netdev, and invoke free_netdev(dev) if register_netdevice()
fails.
If netdev_ops->ndo_init() succeeds, but something else fails inside
of register_netdevice(), it does call ndo_ops->ndo_uninit(). But
it is not able to invoke netdev->destructor().
This is because netdev->destructor() will do a free_netdev() and
then the caller of register_netdevice() will do the same.
However, this means that the resources that would normally be released
by netdev->destructor() will not be.
Over the years drivers have added local hacks to deal with this, by
invoking their destructor parts by hand when register_netdevice()
fails.
Many drivers do not try to deal with this, and instead we have leaks.
Let's close this hole by formalizing the distinction between what
private things need to be freed up by netdev->destructor() and whether
the driver needs unregister_netdevice() to perform the free_netdev().
netdev->priv_destructor() performs all actions to free up the private
resources that used to be freed by netdev->destructor(), except for
free_netdev().
netdev->needs_free_netdev is a boolean that indicates whether
free_netdev() should be done at the end of unregister_netdevice().
Now, register_netdevice() can sanely release all resources after
ndo_ops->ndo_init() succeeds, by invoking both ndo_ops->ndo_uninit()
and netdev->priv_destructor().
And at the end of unregister_netdevice(), we invoke
netdev->priv_destructor() and optionally call free_netdev().
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit c28294b941 ]
KMSAN reported a use of uninitialized memory in dev_set_alias(),
which was caused by calling strlcpy() (which in turn called strlen())
on the user-supplied non-terminated string.
Signed-off-by: Alexander Potapenko <glider@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 38b257938a ]
Prior to f5f99309fa (sock: do not set sk_err in
sock_dequeue_err_skb), sk_err was reset to the error of
the skb on the head of the error queue.
Applications, most notably ping, are relying on this
behavior to reset sk_err for ICMP packets.
Set sk_err to the ICMP error when there is an ICMP packet
at the head of the error queue.
Fixes: f5f99309fa (sock: do not set sk_err in sock_dequeue_err_skb)
Reported-by: Cyril Hrubis <chrubis@suse.cz>
Tested-by: Cyril Hrubis <chrubis@suse.cz>
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 41703a7310 ]
The bpf_clone_redirect() still needs to be listed in
bpf_helper_changes_pkt_data() since we call into
bpf_try_make_head_writable() from there, thus we need
to invalidate prior pkt regs as well.
Fixes: 36bbef52c7 ("bpf: direct packet write and access for helpers for clsact progs")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 3fb07daff8 ]
Andrey Konovalov reported crashes in ipv4_mtu()
I could reproduce the issue with KASAN kernels, between
10.246.7.151 and 10.246.7.152 :
1) 20 concurrent netperf -t TCP_RR -H 10.246.7.152 -l 1000 &
2) At the same time run following loop :
while :
do
ip ro add 10.246.7.152 dev eth0 src 10.246.7.151 mtu 1500
ip ro del 10.246.7.152 dev eth0 src 10.246.7.151 mtu 1500
done
Cong Wang attempted to add back rt->fi in commit
82486aa6f1 ("ipv4: restore rt->fi for reference counting")
but this proved to add some issues that were complex to solve.
Instead, I suggested to add a refcount to the metrics themselves,
being a standalone object (in particular, no reference to other objects)
I tried to make this patch as small as possible to ease its backport,
instead of being super clean. Note that we believe that only ipv4 dst
need to take care of the metric refcount. But if this is wrong,
this patch adds the basic infrastructure to extend this to other
families.
Many thanks to Julian Anastasov for reviewing this patch, and Cong Wang
for his efforts on this problem.
Fixes: 2860583fe8 ("ipv4: Kill rt->fi")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Andrey Konovalov <andreyknvl@google.com>
Reviewed-by: Julian Anastasov <ja@ssi.bg>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit f6c5775ff0 ]
In general, rtnetlink dumps do not anticipate failure to dump a single
object (e.g., link or route) on a single pass. As both route and link
objects have grown via more attributes, that is no longer a given.
netlink dumps can handle a failure if the dump function returns an
error; specifically, netlink_dump adds the return code to the response
if it is <= 0 so userspace is notified of the failure. The missing
piece is the rtnetlink dump functions returning the error.
Fix route and link dump functions to return the errors if no object is
added to an skb (detected by skb->len != 0). IPv6 route dumps
(rt6_dump_route) already return the error; this patch updates IPv4 and
link dumps. Other dump functions may need to be ajusted as well.
Reported-by: Jan Moskyto Matejka <mq@ucw.cz>
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 9142e9007f ]
If CONFIG_INET is not set, net/core/sock.c can not compile :
net/core/sock.c: In function ‘skb_orphan_partial’:
net/core/sock.c:1810:2: error: implicit declaration of function
‘skb_is_tcp_pure_ack’ [-Werror=implicit-function-declaration]
if (skb_is_tcp_pure_ack(skb))
^
Fix this by always including <net/tcp.h>
Fixes: f6ba8d33cf ("netem: fix skb_orphan_partial()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit f6ba8d33cf ]
I should have known that lowering skb->truesize was dangerous :/
In case packets are not leaving the host via a standard Ethernet device,
but looped back to local sockets, bad things can happen, as reported
by Michael Madsen ( https://bugzilla.kernel.org/show_bug.cgi?id=195713 )
So instead of tweaking skb->truesize, lets change skb->destructor
and keep a reference on the owner socket via its sk_refcnt.
Fixes: f2f872f927 ("netem: Introduce skb_orphan_partial() helper")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Michael Madsen <mkm@nabto.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a6a5993243 upstream.
The patch 3278682123 (make skb_copy_datagram_msg() et.al. preserve
->msg_iter on error) will revert the iov buffer if copy to iter
failed, but it didn't copy any datagram if the skb_checksum_complete
error, so no need to revert any data at this place.
v2: Sabrina notice that return -EFAULT when checksum error is not correct
here, it would confuse the caller about the return value, so fix it.
Fixes: 3278682123 ("make skb_copy_datagram_msg() et.al. preserve->msg_iter on error")
Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 84b114b984 ]
Whole point of randomization was to hide server uptime, but an attacker
can simply start a syn flood and TCP generates 'old style' timestamps,
directly revealing server jiffies value.
Also, TSval sent by the server to a particular remote address vary
depending on syncookies being sent or not, potentially triggering PAWS
drops for innocent clients.
Lets implement proper randomization, including for SYNcookies.
Also we do not need to export sysctl_tcp_timestamps, since it is not
used from a module.
In v2, I added Florian feedback and contribution, adding tsoff to
tcp_get_cookie_sock().
v3 removed one unused variable in tcp_v4_connect() as Florian spotted.
Fixes: 95a22caee3 ("tcp: randomize tcp timestamp offsets for each connection")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Florian Westphal <fw@strlen.de>
Tested-by: Florian Westphal <fw@strlen.de>
Cc: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 77ef033b68 ]
IFLA_PHYS_PORT_NAME is a string attribute, so terminate it with \0.
Otherwise libnl3 fails to validate netlink messages with this attribute.
"ip -detail a" assumes too that the attribute is NUL-terminated when
printing it. It often was, due to padding.
I noticed this as libvirtd failing to start on a system with sfc driver
after upgrading it to Linux 4.11, i.e. when sfc added support for
phys_port_name.
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Andrey found a way to trigger the WARN_ON_ONCE(delta < len) in
skb_try_coalesce() using syzkaller and a filter attached to a TCP
socket.
As we did recently in commit 158f323b98 ("net: adjust skb->truesize in
pskb_expand_head()") we can adjust skb->truesize from ___pskb_trim(),
via a call to skb_condense().
If all frags were freed, then skb->truesize can be recomputed.
This call can be done if skb is not yet owned, or destructor is
sock_edemux().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Andrey Konovalov <andreyknvl@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 07b26c9454 ("gso: Support partial splitting at the frag_list
pointer") assumes that all SKBs in a frag_list (except maybe the last
one) contain the same amount of GSO payload.
This assumption is not always correct, resulting in the following
warning message in the log:
skb_segment: too many frags
For example, mlx5 driver in Striding RQ mode creates some RX SKBs with
one frag, and some with 2 frags.
After GRO, the frag_list SKBs end up having different amounts of payload.
If this frag_list SKB is then forwarded, the aforementioned assumption
is violated.
Validate the assumption, and fall back to software GSO if it not true.
Change-Id: Ia03983f4a47b6534dd987d7a2aad96d54d46d212
Fixes: 07b26c9454 ("gso: Support partial splitting at the frag_list pointer")
Signed-off-by: Ilan Tayari <ilant@mellanox.com>
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Syzkaller reported a use-after-free in ip_recv_error at line
info->ipi_ifindex = skb->dev->ifindex;
This function is called on dequeue from the error queue, at which
point the device pointer may no longer be valid.
Save ifindex on enqueue in __skb_complete_tx_timestamp, when the
pointer is valid or NULL. Store it in temporary storage skb->cb.
It is safe to reference skb->dev here, as called from device drivers
or dev_queue_xmit. The exception is when called from tcp_ack_tstamp;
in that case it is NULL and ifindex is set to 0 (invalid).
Do not return a pktinfo cmsg if ifindex is 0. This maintains the
current behavior of not returning a cmsg if skb->dev was NULL.
On dequeue, the ipv4 path will cast from sock_exterr_skb to
in_pktinfo. Both have ifindex as their first element, so no explicit
conversion is needed. This is by design, introduced in commit
0b922b7a82 ("net: original ingress device index in PKTINFO"). For
ipv6 ip6_datagram_support_cmsg converts to in6_pktinfo.
Fixes: 829ae9d611 ("net-timestamp: allow reading recv cmsg on errqueue with origin tstamp")
Reported-by: Andrey Konovalov <andreyknvl@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since dev_change_xdp_fd() is only used in rtnetlink, which must
be built-in, there's no reason to export dev_change_xdp_fd().
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The last argument to __skb_header_pointer() should be a buffer large
enough to store struct arphdr. This can be a pointer to a struct arphdr
structure. The code was previously using a pointer to a pointer to
struct arphdr.
By my counting the storage available both before and after is 8 bytes on
x86_64.
Fixes: 55733350e5 ("flow disector: ARP support")
Reported-by: Nicolas Iooss <nicolas.iooss_linux@m4x.org>
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes the mess observed in e.g. rsync over a noisy link we'd been
seeing since last Summer. What happens is that we copy part of
a datagram before noticing a checksum mismatch. Datagram will be
resent, all right, but we want the next try go into the same place,
not after it...
All this family of primitives (copy/checksum and copy a datagram
into destination) is "all or nothing" sort of interface - either
we get 0 (meaning that copy had been successful) or we get an
error (and no way to tell how much had been copied before we ran
into whatever error it had been). Make all of them leave iterator
unadvanced in case of errors - all callers must be able to cope
with that (an error might've been caught before the iterator had
been advanced), it costs very little to arrange, it's safer for
callers and actually fixes at least one bug in said callers.
Cc: stable@vger.kernel.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Unfortunately too many devices (not under our control) use tcp_tw_recycle=1,
which depends on timestamps being identical of the same saddr.
Although tcp_tw_recycle got removed in net-next we can't make
such end hosts disappear so downgrade to per-host timestamp offsets.
Cc: Soheil Hassas Yeganeh <soheil@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Reported-by: Yvan Vanrossomme <yvan@vanrossomme.net>
Fixes: 95a22caee3 ("tcp: randomize tcp timestamp offsets for each connection")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
This change basically codifies what I think was already the limitations on
the busy_poll and busy_read sysctl interfaces. We weren't checking the
lower bounds and as such could input negative values. The behavior when
that was used was dependent on the architecture. In order to prevent any
issues with that I am just disabling support for values less than 0 since
this way we don't have to worry about any odd behaviors.
By limiting the sysctl values this way it also makes it consistent with how
we handle the SO_BUSY_POLL socket option since the value appears to be
reported as a signed integer value and negative values are rejected.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Dmitry posted a nice reproducer of a bug triggering in neigh_probe()
when dereferencing a NULL neigh->ops->solicit method.
This can happen for arp_direct_ops/ndisc_direct_ops and similar,
which can be used for NUD_NOARP neighbours (created when dev->header_ops
is NULL). Admin can then force changing nud_state to some other state
that would fire neigh timer.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In sk_clone_lock(), we create a new socket and inherit most of the
parent's members via sock_copy() which memcpy()'s various sections.
Now, in case the parent socket had a BPF socket filter attached,
then newsk->sk_filter points to the same instance as the original
sk->sk_filter.
sk_filter_charge() is then called on the newsk->sk_filter to take a
reference and should that fail due to hitting max optmem, we bail
out and release the newsk instance.
The issue is that commit 278571baca ("net: filter: simplify socket
charging") wrongly combined the dismantle path with the failure path
of xfrm_sk_clone_policy(). This means, even when charging failed, we
call sk_free_unlock_clone() on the newsk, which then still points to
the same sk_filter as the original sk.
Thus, sk_free_unlock_clone() calls into __sk_destruct() eventually
where it tests for present sk_filter and calls sk_filter_uncharge()
on it, which potentially lets sk_omem_alloc wrap around and releases
the eBPF prog and sk_filter structure from the (still intact) parent.
Fix it by making sure that when sk_filter_charge() failed, we reset
newsk->sk_filter back to NULL before passing to sk_free_unlock_clone(),
so that we don't mess with the parents sk_filter.
Only if xfrm_sk_clone_policy() fails, we did reach the point where
either the parent's filter was NULL and as a result newsk's as well
or where we previously had a successful sk_filter_charge(), thus for
that case, we do need sk_filter_uncharge() to release the prior taken
reference on sk_filter.
Fixes: 278571baca ("net: filter: simplify socket charging")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The net_cls controller controls the classid field of each socket which
is associated with the cgroup. Because the classid is per-socket
attribute, when a task migrates to another cgroup or the configured
classid of the cgroup changes, the controller needs to walk all
sockets and update the classid value, which was implemented by
3b13758f51 ("cgroups: Allow dynamically changing net_classid").
While the approach is not scalable, migrating tasks which have a lot
of fds attached to them is rare and the cost is born by the ones
initiating the operations. However, for simplicity, both the
migration and classid config change paths call update_classid() which
scans all fds of all tasks in the target css. This is an overkill for
the migration path which only needs to cover a much smaller subset of
tasks which are actually getting migrated in.
On cgroup v1, this can lead to unexpected scalability issues when one
tries to migrate a task or process into a net_cls cgroup which already
contains a lot of fds. Even if the migration traget doesn't have many
to get scanned, update_classid() ends up scanning all fds in the
target cgroup which can be extremely numerous.
Unfortunately, on cgroup v2 which doesn't use net_cls, the problem is
even worse. Before bfc2cf6f61 ("cgroup: call subsys->*attach() only
for subsystems which are actually affected by migration"), cgroup core
would call the ->css_attach callback even for controllers which don't
see actual migration to a different css.
As net_cls is always disabled but still mounted on cgroup v2, whenever
a process is migrated on the cgroup v2 hierarchy, net_cls sees
identity migration from root to root and cgroup core used to call
->css_attach callback for those. The net_cls ->css_attach ends up
calling update_classid() on the root net_cls css to which all
processes on the system belong to as the controller isn't used. This
makes any cgroup v2 migration O(total_number_of_fds_on_the_system)
which is horrible and easily leads to noticeable stalls triggering RCU
stall warnings and so on.
The worst symptom is already fixed in upstream by bfc2cf6f61
("cgroup: call subsys->*attach() only for subsystems which are
actually affected by migration"); however, backporting that commit is
too invasive and we want to avoid other cases too.
This patch updates net_cls's cgrp_attach() to iterate fds of only the
processes which are actually getting migrated. This removes the
surprising migration cost which is dependent on the total number of
fds in the target cgroup. As this leaves write_classid() the only
user of update_classid(), open-code the helper into write_classid().
Reported-by: David Goode <dgoode@fb.com>
Fixes: 3b13758f51 ("cgroups: Allow dynamically changing net_classid")
Cc: stable@vger.kernel.org # v4.4+
Cc: Nina Schiff <ninasc@fb.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
SOF_TIMESTAMPING_OPT_STATS can be enabled and disabled
while packets are collected on the error queue.
So, checking SOF_TIMESTAMPING_OPT_STATS in sk->sk_tsflags
is not enough to safely assume that the skb contains
OPT_STATS data.
Add a bit in sock_exterr_skb to indicate whether the
skb contains opt_stats data.
Fixes: 1c885808e4 ("tcp: SOF_TIMESTAMPING_OPT_STATS option for SO_TIMESTAMPING")
Reported-by: JongHwan Kim <zzoru007@gmail.com>
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
__sock_recv_timestamp can be called for both normal skbs (for
receive timestamps) and for skbs on the error queue (for transmit
timestamps).
Commit 1c885808e4
(tcp: SOF_TIMESTAMPING_OPT_STATS option for SO_TIMESTAMPING)
assumes any skb passed to __sock_recv_timestamp are from
the error queue, containing OPT_STATS in the content of the skb.
This results in accessing invalid memory or generating junk
data.
To fix this, set skb->pkt_type to PACKET_OUTGOING for packets
on the error queue. This is safe because on the receive path
on local sockets skb->pkt_type is never set to PACKET_OUTGOING.
With that, copy OPT_STATS from a packet, only if its pkt_type
is PACKET_OUTGOING.
Fixes: 1c885808e4 ("tcp: SOF_TIMESTAMPING_OPT_STATS option for SO_TIMESTAMPING")
Reported-by: JongHwan Kim <zzoru007@gmail.com>
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I mistakenly added the code to release sk->sk_frag in
sk_common_release() instead of sk_destruct()
TCP sockets using sk->sk_allocation == GFP_ATOMIC do no call
sk_common_release() at close time, thus leaking one (order-3) page.
iSCSI is using such sockets.
Fixes: 5640f76858 ("net: use a per task frag allocator")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When we notify peers of potential changes, it's also good to update
IGMP memberships. For example, during VM migration, updating IGMP
memberships will redirect existing multicast streams to the VM at the
new location.
Signed-off-by: Vladislav Yasevich <vyasevic@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The previous idea was to check whether a net namespace is in
net_exit_list or not. It doesn't work, because net->exit_list is used in
__register_pernet_operations and __unregister_pernet_operations where
all namespaces are added to a temporary list to make cleanup in a error
case, so list_empty(&net->exit_list) always returns false.
Reported-by: Mantas Mikulėnas <grawity@gmail.com>
Fixes: 002d8a1a6c ("net: skip genenerating uevents for network namespaces that are exiting")
Signed-off-by: Andrei Vagin <avagin@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Lockdep issues a circular dependency warning when AFS issues an operation
through AF_RXRPC from a context in which the VFS/VM holds the mmap_sem.
The theory lockdep comes up with is as follows:
(1) If the pagefault handler decides it needs to read pages from AFS, it
calls AFS with mmap_sem held and AFS begins an AF_RXRPC call, but
creating a call requires the socket lock:
mmap_sem must be taken before sk_lock-AF_RXRPC
(2) afs_open_socket() opens an AF_RXRPC socket and binds it. rxrpc_bind()
binds the underlying UDP socket whilst holding its socket lock.
inet_bind() takes its own socket lock:
sk_lock-AF_RXRPC must be taken before sk_lock-AF_INET
(3) Reading from a TCP socket into a userspace buffer might cause a fault
and thus cause the kernel to take the mmap_sem, but the TCP socket is
locked whilst doing this:
sk_lock-AF_INET must be taken before mmap_sem
However, lockdep's theory is wrong in this instance because it deals only
with lock classes and not individual locks. The AF_INET lock in (2) isn't
really equivalent to the AF_INET lock in (3) as the former deals with a
socket entirely internal to the kernel that never sees userspace. This is
a limitation in the design of lockdep.
Fix the general case by:
(1) Double up all the locking keys used in sockets so that one set are
used if the socket is created by userspace and the other set is used
if the socket is created by the kernel.
(2) Store the kern parameter passed to sk_alloc() in a variable in the
sock struct (sk_kern_sock). This informs sock_lock_init(),
sock_init_data() and sk_clone_lock() as to the lock keys to be used.
Note that the child created by sk_clone_lock() inherits the parent's
kern setting.
(3) Add a 'kern' parameter to ->accept() that is analogous to the one
passed in to ->create() that distinguishes whether kernel_accept() or
sys_accept4() was the caller and can be passed to sk_alloc().
Note that a lot of accept functions merely dequeue an already
allocated socket. I haven't touched these as the new socket already
exists before we get the parameter.
Note also that there are a couple of places where I've made the accepted
socket unconditionally kernel-based:
irda_accept()
rds_rcp_accept_one()
tcp_accept_from_sock()
because they follow a sock_create_kern() and accept off of that.
Whilst creating this, I noticed that lustre and ocfs don't create sockets
through sock_create_kern() and thus they aren't marked as for-kernel,
though they appear to be internal. I wonder if these should do that so
that they use the new set of lock keys.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TX skbs do not necessarily hold a reference on skb->sk->sk_refcnt
By the time TX completion happens, sk_refcnt might be already 0.
sock_hold()/sock_put() would then corrupt critical state, like
sk_wmem_alloc and lead to leaks or use after free.
Fixes: 62bccb8cdb ("net-timestamp: Make the clone operation stand-alone from phy timestamping")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: Johannes Berg <johannes@sipsolutions.net>
Cc: Soheil Hassas Yeganeh <soheil@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TX skbs do not necessarily hold a reference on skb->sk->sk_refcnt
By the time TX completion happens, sk_refcnt might be already 0.
sock_hold()/sock_put() would then corrupt critical state, like
sk_wmem_alloc.
Fixes: bf7fa551e0 ("mac80211: Resolve sk_refcnt/sk_wmem_alloc issue in wifi ack path")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: Johannes Berg <johannes@sipsolutions.net>
Cc: Soheil Hassas Yeganeh <soheil@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking fixes from David Miller:
1) Fix double-free in batman-adv, from Sven Eckelmann.
2) Fix packet stats for fast-RX path, from Joannes Berg.
3) Netfilter's ip_route_me_harder() doesn't handle request sockets
properly, fix from Florian Westphal.
4) Fix sendmsg deadlock in rxrpc, from David Howells.
5) Add missing RCU locking to transport hashtable scan, from Xin Long.
6) Fix potential packet loss in mlxsw driver, from Ido Schimmel.
7) Fix race in NAPI handling between poll handlers and busy polling,
from Eric Dumazet.
8) TX path in vxlan and geneve need proper RCU locking, from Jakub
Kicinski.
9) SYN processing in DCCP and TCP need to disable BH, from Eric
Dumazet.
10) Properly handle net_enable_timestamp() being invoked from IRQ
context, also from Eric Dumazet.
11) Fix crash on device-tree systems in xgene driver, from Alban Bedel.
12) Do not call sk_free() on a locked socket, from Arnaldo Carvalho de
Melo.
13) Fix use-after-free in netvsc driver, from Dexuan Cui.
14) Fix max MTU setting in bonding driver, from WANG Cong.
15) xen-netback hash table can be allocated from softirq context, so use
GFP_ATOMIC. From Anoob Soman.
16) Fix MAC address change bug in bgmac driver, from Hari Vyas.
17) strparser needs to destroy strp_wq on module exit, from WANG Cong.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (69 commits)
strparser: destroy workqueue on module exit
sfc: fix IPID endianness in TSOv2
sfc: avoid max() in array size
rds: remove unnecessary returned value check
rxrpc: Fix potential NULL-pointer exception
nfp: correct DMA direction in XDP DMA sync
nfp: don't tell FW about the reserved buffer space
net: ethernet: bgmac: mac address change bug
net: ethernet: bgmac: init sequence bug
xen-netback: don't vfree() queues under spinlock
xen-netback: keep a local pointer for vif in backend_disconnect()
netfilter: nf_tables: don't call nfnetlink_set_err() if nfnetlink_send() fails
netfilter: nft_set_rbtree: incorrect assumption on lower interval lookups
netfilter: nf_conntrack_sip: fix wrong memory initialisation
can: flexcan: fix typo in comment
can: usb_8dev: Fix memory leak of priv->cmd_msg_buffer
can: gs_usb: fix coding style
can: gs_usb: Don't use stack memory for USB transfers
ixgbe: Limit use of 2K buffers on architectures with 256B or larger cache lines
ixgbe: update the rss key on h/w, when ethtool ask for it
...
We are going to split <linux/sched/user.h> out of <linux/sched.h>, which
will have to be picked up from other headers and a couple of .c files.
Create a trivial placeholder <linux/sched/user.h> file that just
maps to <linux/sched.h> to make this patch obviously correct and
bisectable.
Include the new header in the files that are going to need it.
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We are going to split <linux/sched/signal.h> out of <linux/sched.h>, which
will have to be picked up from other headers and a couple of .c files.
Create a trivial placeholder <linux/sched/signal.h> file that just
maps to <linux/sched.h> to make this patch obviously correct and
bisectable.
Include the new header in the files that are going to need it.
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
It is now very clear that silly TCP listeners might play with
enabling/disabling timestamping while new children are added
to their accept queue.
Meaning net_enable_timestamp() can be called from BH context
while current state of the static key is not enabled.
Lets play safe and allow all contexts.
The work queue is scheduled only under the problematic cases,
which are the static key enable/disable transition, to not slow down
critical paths.
This extends and improves what we did in commit 5fa8bbda38 ("net: use
a work queue to defer net_disable_timestamp() work")
Fixes: b90e5794c5 ("net: dont call jump_label_dec from irq context")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While playing with mlx4 hardware timestamping of RX packets, I found
that some packets were received by TCP stack with a ~200 ms delay...
Since the timestamp was provided by the NIC, and my probe was added
in tcp_v4_rcv() while in BH handler, I was confident it was not
a sender issue, or a drop in the network.
This would happen with a very low probability, but hurting RPC
workloads.
A NAPI driver normally arms the IRQ after the napi_complete_done(),
after NAPI_STATE_SCHED is cleared, so that the hard irq handler can grab
it.
Problem is that if another point in the stack grabs NAPI_STATE_SCHED bit
while IRQ are not disabled, we might have later an IRQ firing and
finding this bit set, right before napi_complete_done() clears it.
This can happen with busy polling users, or if gro_flush_timeout is
used. But some other uses of napi_schedule() in drivers can cause this
as well.
thread 1 thread 2 (could be on same cpu, or not)
// busy polling or napi_watchdog()
napi_schedule();
...
napi->poll()
device polling:
read 2 packets from ring buffer
Additional 3rd packet is
available.
device hard irq
// does nothing because
NAPI_STATE_SCHED bit is owned by thread 1
napi_schedule();
napi_complete_done(napi, 2);
rearm_irq();
Note that rearm_irq() will not force the device to send an additional
IRQ for the packet it already signaled (3rd packet in my example)
This patch adds a new NAPI_STATE_MISSED bit, that napi_schedule_prep()
can set if it could not grab NAPI_STATE_SCHED
Then napi_complete_done() properly reschedules the napi to make sure
we do not miss something.
Since we manipulate multiple bits at once, use cmpxchg() like in
sk_busy_loop() to provide proper transactions.
In v2, I changed napi_watchdog() to use a relaxed variant of
napi_schedule_prep() : No need to set NAPI_STATE_MISSED from this point.
In v3, I added more details in the changelog and clears
NAPI_STATE_MISSED in busy_poll_stop()
In v4, I added the ideas given by Alexander Duyck in v3 review
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Alexander Duyck <alexander.duyck@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
hrtimer handlers run with masked hard IRQ, we can therefore
use napi_schedule_irqoff()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The USEC_PER_SEC is used once in sock_set_timeout as the max value of
tv_usec. But there are other similar codes which use the literal
1000000 in this file.
It is minor cleanup to keep consitent.
Signed-off-by: Gao Feng <fgao@ikuai8.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
