| Age | Commit message (Collapse) | Author |
|---|
|
This work enables generic transfer of metadata from XDP into skb. The
basic idea is that we can make use of the fact that the resulting skb
must be linear and already comes with a larger headroom for supporting
bpf_xdp_adjust_head(), which mangles xdp->data. Here, we base our work
on a similar principle and introduce a small helper bpf_xdp_adjust_meta()
for adjusting a new pointer called xdp->data_meta. Thus, the packet has
a flexible and programmable room for meta data, followed by the actual
packet data. struct xdp_buff is therefore laid out that we first point
to data_hard_start, then data_meta directly prepended to data followed
by data_end marking the end of packet. bpf_xdp_adjust_head() takes into
account whether we have meta data already prepended and if so, memmove()s
this along with the given offset provided there's enough room.
xdp->data_meta is optional and programs are not required to use it. The
rationale is that when we process the packet in XDP (e.g. as DoS filter),
we can push further meta data along with it for the XDP_PASS case, and
give the guarantee that a clsact ingress BPF program on the same device
can pick this up for further post-processing. Since we work with skb
there, we can also set skb->mark, skb->priority or other skb meta data
out of BPF, thus having this scratch space generic and programmable
allows for more flexibility than defining a direct 1:1 transfer of
potentially new XDP members into skb (it's also more efficient as we
don't need to initialize/handle each of such new members). The facility
also works together with GRO aggregation. The scratch space at the head
of the packet can be multiple of 4 byte up to 32 byte large. Drivers not
yet supporting xdp->data_meta can simply be set up with xdp->data_meta
as xdp->data + 1 as bpf_xdp_adjust_meta() will detect this and bail out,
such that the subsequent match against xdp->data for later access is
guaranteed to fail.
The verifier treats xdp->data_meta/xdp->data the same way as we treat
xdp->data/xdp->data_end pointer comparisons. The requirement for doing
the compare against xdp->data is that it hasn't been modified from it's
original address we got from ctx access. It may have a range marking
already from prior successful xdp->data/xdp->data_end pointer comparisons
though.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Just do the rename into bpf_compute_data_pointers() as we'll add
one more pointer here to recompute.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
IS_ERR() already implies unlikely(), so it can be omitted.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
IS_ERR() already implies unlikely(), so it can be omitted.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
IS_ERR() already implies unlikely(), so it can be omitted.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Key length can't be negative.
Leave comparisons against nla_len() signed just in case truncated attribute
can sneak in there.
Space savings:
add/remove: 0/0 grow/shrink: 0/7 up/down: 0/-7 (-7)
function old new delta
pneigh_delete 273 272 -1
mlx5e_rep_netevent_event 1415 1414 -1
mlx5e_create_encap_header_ipv6 1194 1193 -1
mlx5e_create_encap_header_ipv4 1071 1070 -1
cxgb4_l2t_get 1104 1103 -1
__pneigh_lookup 69 68 -1
__neigh_create 2452 2451 -1
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
As measured in my prior patch ("sch_netem: faster rb tree removal"),
rbtree_postorder_for_each_entry_safe() is nice looking but much slower
than using rb_next() directly, except when tree is small enough
to fit in CPU caches (then the cost is the same)
Also note that there is not even an increase of text size :
$ size net/core/skbuff.o.before net/core/skbuff.o
text data bss dec hex filename
40711 1298 0 42009 a419 net/core/skbuff.o.before
40711 1298 0 42009 a419 net/core/skbuff.o
From: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
While running TCP tests involving netem storing millions of packets,
I had the idea to speed up tfifo_reset() and did experiments.
I tried the rbtree_postorder_for_each_entry_safe() method that is
used in skb_rbtree_purge() but discovered it was slower than the
current tfifo_reset() method.
I measured time taken to release skbs with three occupation levels :
10^4, 10^5 and 10^6 skbs with three methods :
1) (current 'naive' method)
while ((p = rb_first(&q->t_root))) {
struct sk_buff *skb = netem_rb_to_skb(p);
rb_erase(p, &q->t_root);
rtnl_kfree_skbs(skb, skb);
}
2) Use rb_next() instead of rb_first() in the loop :
p = rb_first(&q->t_root);
while (p) {
struct sk_buff *skb = netem_rb_to_skb(p);
p = rb_next(p);
rb_erase(&skb->rbnode, &q->t_root);
rtnl_kfree_skbs(skb, skb);
}
3) "optimized" method using rbtree_postorder_for_each_entry_safe()
struct sk_buff *skb, *next;
rbtree_postorder_for_each_entry_safe(skb, next,
&q->t_root, rbnode) {
rtnl_kfree_skbs(skb, skb);
}
q->t_root = RB_ROOT;
Results :
method_1:while (rb_first()) rb_erase() 10000 skbs in 690378 ns (69 ns per skb)
method_2:rb_first; while (p) { p = rb_next(p); ...} 10000 skbs in 541846 ns (54 ns per skb)
method_3:rbtree_postorder_for_each_entry_safe() 10000 skbs in 868307 ns (86 ns per skb)
method_1:while (rb_first()) rb_erase() 99996 skbs in 7804021 ns (78 ns per skb)
method_2:rb_first; while (p) { p = rb_next(p); ...} 100000 skbs in 5942456 ns (59 ns per skb)
method_3:rbtree_postorder_for_each_entry_safe() 100000 skbs in 11584940 ns (115 ns per skb)
method_1:while (rb_first()) rb_erase() 1000000 skbs in 108577838 ns (108 ns per skb)
method_2:rb_first; while (p) { p = rb_next(p); ...} 1000000 skbs in 82619635 ns (82 ns per skb)
method_3:rbtree_postorder_for_each_entry_safe() 1000000 skbs in 127328743 ns (127 ns per skb)
Method 2) is simply faster, probably because it maintains a smaller
working size set.
Note that this is the method we use in tcp_ofo_queue() already.
I will also change skb_rbtree_purge() in a second patch.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Provide dsa_port_enable and dsa_port_disable helpers to respectively
enable and disable a switch port. This makes the dsa_port_set_state_now
helper static.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The DSA slave open function configures the unicast MAC addresses on the
master device, enable the switch port, change its STP state, then start
the PHY device.
Make the close function symmetric, by first stopping the PHY device,
then changing the STP state, disabling the switch port and restore the
master device.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use setup_timer function instead of initializing timer with the
function and data fields.
Signed-off-by: Allen Pais <allen.lkml@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use setup_timer function instead of initializing timer with the
function and data fields.
Signed-off-by: Allen Pais <allen.lkml@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use setup_timer function instead of initializing timer with the
function and data fields.
Signed-off-by: Allen Pais <allen.lkml@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use setup_timer function instead of initializing timer with the
function and data fields.
Signed-off-by: Allen Pais <allen.lkml@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
|
|
Pull networking fixes from David Miller:
1) Fix NAPI poll list corruption in enic driver, from Christian
Lamparter.
2) Fix route use after free, from Eric Dumazet.
3) Fix regression in reuseaddr handling, from Josef Bacik.
4) Assert the size of control messages in compat handling since we copy
it in from userspace twice. From Meng Xu.
5) SMC layer bug fixes (missing RCU locking, bad refcounting, etc.)
from Ursula Braun.
6) Fix races in AF_PACKET fanout handling, from Willem de Bruijn.
7) Don't use ARRAY_SIZE on spinlock array which might have zero
entries, from Geert Uytterhoeven.
8) Fix miscomputation of checksum in ipv6 udp code, from Subash Abhinov
Kasiviswanathan.
9) Push the ipv6 header properly in ipv6 GRE tunnel driver, from Xin
Long.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (75 commits)
inet: fix improper empty comparison
net: use inet6_rcv_saddr to compare sockets
net: set tb->fast_sk_family
net: orphan frags on stand-alone ptype in dev_queue_xmit_nit
MAINTAINERS: update git tree locations for ieee802154 subsystem
net: prevent dst uses after free
net: phy: Fix truncation of large IRQ numbers in phy_attached_print()
net/smc: no close wait in case of process shut down
net/smc: introduce a delay
net/smc: terminate link group if out-of-sync is received
net/smc: longer delay for client link group removal
net/smc: adapt send request completion notification
net/smc: adjust net_device refcount
net/smc: take RCU read lock for routing cache lookup
net/smc: add receive timeout check
net/smc: add missing dev_put
net: stmmac: Cocci spatch "of_table"
lan78xx: Use default values loaded from EEPROM/OTP after reset
lan78xx: Allow EEPROM write for less than MAX_EEPROM_SIZE
lan78xx: Fix for eeprom read/write when device auto suspend
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan-next
Stefan Schmidt says:
====================
pull-request: ieee802154 2017-09-20
Here comes a pull request for ieee802154 changes I have queued up for
this merge window.
Normally these have been coming through the bluetooth tree but as this
three have been falling through the cracks so far and I have to review
and ack all of them anyway I think it makes sense if I save the
bluetooth people some work and handle them directly.
Its the first pull request I send to you so please let me know if I did
something wrong or if you prefer a different format.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Private part of allocation is never big enough to warrant size_t.
Space savings:
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-10 (-10)
function old new delta
alloc_netdev_mqs 1120 1110 -10
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
There is no one which would invokes the function skb_header_release.
So just remove it now.
Signed-off-by: Gao Feng <gfree.wind@vip.163.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
When doing my reuseport rework I screwed up and changed a
if (hlist_empty(&tb->owners))
to
if (!hlist_empty(&tb->owners))
This is obviously bad as all of the reuseport/reuse logic was reversed,
which caused weird problems like allowing an ipv4 bind conflict if we
opened an ipv4 only socket on a port followed by an ipv6 only socket on
the same port.
Fixes: b9470c27607b ("inet: kill smallest_size and smallest_port")
Reported-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
In ipv6_rcv_saddr_equal() we need to use inet6_rcv_saddr(sk) for the
ipv6 compare with the fast socket information to make sure we're doing
the proper comparisons.
Fixes: 637bc8bbe6c0 ("inet: reset tb->fastreuseport when adding a reuseport sk")
Reported-and-tested-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
We need to set the tb->fast_sk_family properly so we can use the proper
comparison function for all subsequent reuseport bind requests.
Fixes: 637bc8bbe6c0 ("inet: reset tb->fastreuseport when adding a reuseport sk")
Reported-and-tested-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Zerocopy skbs frags are copied when the skb is looped to a local sock.
Commit 1080e512d44d ("net: orphan frags on receive") introduced calls
to skb_orphan_frags to deliver_skb and __netif_receive_skb for this.
With msg_zerocopy, these skbs can also exist in the tx path and thus
loop from dev_queue_xmit_nit. This already calls deliver_skb in its
loop. But it does not orphan before a separate pt_prev->func().
Add the missing skb_orphan_frags_rx.
Changes
v1->v2: handle skb_orphan_frags_rx failure
Fixes: 1f8b977ab32d ("sock: enable MSG_ZEROCOPY")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use setup_timer function instead of initializing timer with the
function and data fields.
Signed-off-by: Allen Pais <allen.lkml@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Currently, there is a difference in netlink events received when an
interface is modified through bridge ioctl() or through netlink. This
patch generates additional events when an interface is added to or
removed from a bridge via ioctl().
When adding then removing an interface from a bridge with netlink, we
get:
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue master bridge0 state UNKNOWN group default
link/ether 9e:da:60:ee:cf:c8 brd ff:ff:ff:ff:ff:ff
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue master bridge0 state UNKNOWN group default
link/ether 9e:da:60:ee:cf:c8 brd ff:ff:ff:ff:ff:ff
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue master bridge0 state UNKNOWN group default
link/ether 9e:da:60:ee:cf:c8 brd ff:ff:ff:ff:ff:ff
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
Deleted 5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default
link/ether 9e:da:60:ee:cf:c8 brd ff:ff:ff:ff:ff:ff
When using ioctl():
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue master bridge0 state UNKNOWN group default
link/ether 9e:da:60:ee:cf:c8 brd ff:ff:ff:ff:ff:ff
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue master bridge0 state UNKNOWN group default
link/ether 9e:da:60:ee:cf:c8 brd ff:ff:ff:ff:ff:ff
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue master bridge0 state UNKNOWN group default
link/ether 9e:da:60:ee:cf:c8 brd ff:ff:ff:ff:ff:ff
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
Deleted 5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 9e:da:60:ee:cf:c8
5: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default
link/ether 9e:da:60:ee:cf:c8 brd ff:ff:ff:ff:ff:ff
Without this patch, the last netlink notification is not sent.
Signed-off-by: Vincent Bernat <vincent@bernat.im>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use the smc_connection as first parameter with smc_cdc_get_free_slot().
This is just a small code cleanup, no functional change.
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Usually socket closing is delayed if there is still data available in
the send buffer to be transmitted. If a process is killed, the delay
should be avoided.
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The number of outstanding work requests is limited. If all work
requests are in use, tx processing is postponed to another scheduling
of the tx worker. Switch to a delayed worker to have a gap for tx
completion queue events before the next retry.
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
An out-of-sync condition can just be detected by the client.
If the server receives a CLC DECLINE message indicating an out-of-sync
condition for the link groups, the server must clean up the out-of-sync
link group.
There is no need for an extra third parameter in smc_clc_send_decline().
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Client link group creation always follows the server linkgroup creation.
If peer creates a new server link group, client has to create a new
client link group. If peer reuses a server link group for a new
connection, client has to reuse its client link group as well. This
patch introduces a longer delay for client link group removal to make
sure this link group still exists, once the peer decides to reuse a
server link group. This avoids out-of-sync conditions for link groups.
If already scheduled, modify the delay.
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The solicited flag is meaningful for the receive completion queue.
Ask for next work completion of any type on the send queue.
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
smc_pnet_fill_entry() uses dev_get_by_name() adding a refcount to ndev.
The following smc_pnet_enter() has to reduce the refcount if the entry
to be added exists already in the pnet table.
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
smc_netinfo_by_tcpsk() looks up the routing cache. Such a lookup requires
protection by an RCU read lock.
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The SMC receive function currently lacks a timeout check under the
condition that no data were received and no data are available. This
patch adds such a check.
Signed-off-by: Hans Wippel <hwippel@linux.vnet.ibm.com>
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
In the infiniband part, SMC currently uses get_netdev which calls
dev_hold on the returned net device. However, the SMC code never calls
dev_put on that net device resulting in a wrong reference count.
This patch adds a dev_put after the usage of the net device to fix the
issue.
Signed-off-by: Hans Wippel <hwippel@linux.vnet.ibm.com>
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Dumping a DSA port's FDB entries is not specific to a DSA slave, so add
a dsa_port_fdb_dump function, similarly to dsa_port_fdb_add and
dsa_port_fdb_del.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
A few DSA slave functions take a dsa_slave_priv pointer as first
argument, whereas the scope of the slave.c functions is the slave
net_device structure. Fix this and rename dsa_netpoll_send_skb to
dsa_slave_netpoll_send_skb.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Commit 3f1ac7a700d0 ("net: ethtool: add new ETHTOOL_xLINKSETTINGS API")
deprecated the ethtool_cmd::transceiver field, which was fine in
premise, except that the PHY library was actually using it to report the
type of transceiver: internal or external.
Use the first word of the reserved field to put this __u8 transceiver
field back in. It is made read-only, and we don't expect the
ETHTOOL_xLINKSETTINGS API to be doing anything with this anyway, so this
is mostly for the legacy path where we do:
ethtool_get_settings()
-> dev->ethtool_ops->get_link_ksettings()
-> convert_link_ksettings_to_legacy_settings()
to have no information loss compared to the legacy get_settings API.
Fixes: 3f1ac7a700d0 ("net: ethtool: add new ETHTOOL_xLINKSETTINGS API")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Since commit 1dced6a85482 ("ipv4: Restore accept_local behaviour
in fib_validate_source()") a full fib lookup is needed even if
the rp_filter is disabled, if accept_local is false - which is
the default.
What we really need in the above scenario is just checking
that the source IP address is not local, and in most case we
can do that is a cheaper way looking up the ifaddr hash table.
This commit adds a helper for such lookup, and uses it to
validate the src address when rp_filter is disabled and no
'local' routes are created by the user space in the relevant
namespace.
A new ipv4 netns flag is added to account for such routes.
We need that to preserve the same behavior we had before this
patch.
It also drops the checks to bail early from __fib_validate_source,
added by the commit 1dced6a85482 ("ipv4: Restore accept_local
behaviour in fib_validate_source()") they do not give any
measurable performance improvement: if we do the lookup with are
on a slower path.
This improves UDP performances for unconnected sockets
when rp_filter is disabled by 5% and also gives small but
measurable performance improvement for TCP flood scenarios.
v1 -> v2:
- use the ifaddr lookup helper in __ip_dev_find(), as suggested
by Eric
- fall-back to full lookup if custom local routes are present
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Fixes: c15ab236d69d ("net/sched: Change cls_flower to use IDR")
Cc: Chris Mi <chrism@mellanox.com>
Cc: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
If real-time or fair-share curves are enabled in hfsc_change_class()
class isn't inserted into rb-trees yet. Thus init_ed() and init_vf()
must be called in place of update_ed() and update_vf().
Remove isn't required because for now curves cannot be disabled.
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
SKB stored in qdisc->gso_skb also counted into backlog.
Some qdiscs don't reset backlog to zero in ->reset(),
for example sfq just dequeue and free all queued skb.
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Fixes: 2ccccf5fb43f ("net_sched: update hierarchical backlog too")
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
The following patchset contains two Netfilter fixes for your net tree,
they are:
1) Fix NAt compilation with UP, from Geert Uytterhoeven.
2) Fix incorrect number of entries when dumping a set, from
Vishwanath Pai.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Instead of open coding the check.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Since XDP's view of the packet includes the MAC header, moving the start-
of-packet with bpf_xdp_adjust_head needs to also update the offset of the
MAC header (which is relative to skb->head, not to the skb->data that was
changed).
Without this, tcpdump sees packets starting from the old MAC header rather
than the new one, at least in my tests on the loopback device.
Fixes: b5cdae3291f7 ("net: Generic XDP")
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This reverts commit 00ba4cb36da682c68dc87d1703a8aaffe2b4e9c5.
Discussion with David Ahern determined that this change is
actually not needed.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The actual length of cmsg fetched in during the second loop
(i.e., kcmsg - kcmsg_base) could be different from what we
get from the first loop (i.e., kcmlen).
The main reason is that the two get_user() calls in the two
loops (i.e., get_user(ucmlen, &ucmsg->cmsg_len) and
__get_user(ucmlen, &ucmsg->cmsg_len)) could cause ucmlen
to have different values even they fetch from the same userspace
address, as user can race to change the memory content in
&ucmsg->cmsg_len across fetches.
Although in the second loop, the sanity check
if ((char *)kcmsg_base + kcmlen - (char *)kcmsg < CMSG_ALIGN(tmp))
is inplace, it only ensures that the cmsg fetched in during the
second loop does not exceed the length of kcmlen, but not
necessarily equal to kcmlen. But indicated by the assignment
kmsg->msg_controllen = kcmlen, we should enforce that.
This patch adds this additional sanity check and ensures that
what is recorded in kmsg->msg_controllen is the actual cmsg length.
Signed-off-by: Meng Xu <mengxu.gatech@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The commit 6b229cf77d68 ("udp: add batching to udp_rmem_release()")
reduced greatly the cacheline contention between the BH and the US
reader batching the rmem updates in most scenarios.
Such optimization is explicitly avoided if the US reader is faster
then BH processing.
My fault, I initially suggested this kind of behavior due to concerns
of possible regressions with small sk_rcvbuf values. Tests showed
such concerns are misplaced, so this commit relaxes the condition
for rmem bulk updates, obtaining small but measurable performance
gain in the scenario described above.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Currently, when an interface is released from a bridge via
ioctl(), we get a RTM_DELLINK event through netlink:
Deleted 2: dummy0: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 6e:23:c2:54:3a:b3
Userspace has to interpret that as a removal from the bridge, not as a
complete removal of the interface. When an bridged interface is
completely removed, we get two events:
Deleted 2: dummy0: <BROADCAST,NOARP> mtu 1500 master bridge0 state DOWN
link/ether 6e:23:c2:54:3a:b3
Deleted 2: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN group default
link/ether 6e:23:c2:54:3a:b3 brd ff:ff:ff:ff:ff:ff
In constrast, when an interface is released from a bond, we get a
RTM_NEWLINK with only the new characteristics (no master):
3: dummy1: <BROADCAST,NOARP,SLAVE,UP,LOWER_UP> mtu 1500 qdisc noqueue master bond0 state UNKNOWN group default
link/ether ae:dc:7a:8c:9a:3c brd ff:ff:ff:ff:ff:ff
3: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default
link/ether ae:dc:7a:8c:9a:3c brd ff:ff:ff:ff:ff:ff
4: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default
link/ether ae:dc:7a:8c:9a:3c brd ff:ff:ff:ff:ff:ff
3: dummy1: <BROADCAST,NOARP> mtu 1500 qdisc noqueue state DOWN group default
link/ether ae:dc:7a:8c:9a:3c brd ff:ff:ff:ff:ff:ff
3: dummy1: <BROADCAST,NOARP> mtu 1500 qdisc noqueue state DOWN group default
link/ether ca:c8:7b:66:f8:25 brd ff:ff:ff:ff:ff:ff
4: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default
link/ether ae:dc:7a:8c:9a:3c brd ff:ff:ff:ff:ff:ff
Userland may be confused by the fact we say a link is deleted while
its characteristics are only modified. A first solution would have
been to turn the RTM_DELLINK event in del_nbp() into a RTM_NEWLINK
event. However, maybe some piece of userland is relying on this
RTM_DELLINK to detect when a bridged interface is released. Instead,
we also emit a RTM_NEWLINK event once the interface is
released (without master info).
Deleted 2: dummy0: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 master bridge0 state UNKNOWN
link/ether 8a:bb:e7:94:b1:f8
2: dummy0: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default
link/ether 8a:bb:e7:94:b1:f8 brd ff:ff:ff:ff:ff:ff
This is done only when using ioctl(). When using Netlink, such an
event is already automatically emitted in do_setlink().
Signed-off-by: Vincent Bernat <vincent@bernat.im>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Packet socket bind operations must hold the po->bind_lock. This keeps
po->running consistent with whether the socket is actually on a ptype
list to receive packets.
fanout_add unbinds a socket and its packet_rcv/tpacket_rcv call, then
binds the fanout object to receive through packet_rcv_fanout.
Make it hold the po->bind_lock when testing po->running and rebinding.
Else, it can race with other rebind operations, such as that in
packet_set_ring from packet_rcv to tpacket_rcv. Concurrent updates
can result in a socket being added to a fanout group twice, causing
use-after-free KASAN bug reports, among others.
Reported independently by both trinity and syzkaller.
Verified that the syzkaller reproducer passes after this patch.
Fixes: dc99f600698d ("packet: Add fanout support.")
Reported-by: nixioaming <nixiaoming@huawei.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
