Before da413eec72 ("packet: Fixed TPACKET V3 to signal poll when block is
closed rather than every packet") poll listening for an af_packet socket was
not signaled if there was no packets to process. After the patch poll is
signaled evety time when block retire timer expires. That happens because
af_packet closes the current block on timeout even if the block is empty.
Passing empty blocks to the user not only wastes CPU but also wastes ring
buffer space increasing probability of packets dropping on small timeouts.
Signed-off-by: Alexander Drozdov <al.drozdov@gmail.com>
Cc: Dan Collins <dan@dcollins.co.nz>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Guy Harris <guy@alum.mit.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
Packets defragmentation was introduced for PACKET_FANOUT_HASH only,
see 7736d33f42 ("packet: Add pre-defragmentation support for ipv4
fanouts")
It may be useful to have defragmentation enabled regardless of
fanout type. Without that, the AF_PACKET user may have to:
1. Collect fragments from different rings
2. Defragment by itself
Signed-off-by: Alexander Drozdov <al.drozdov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Contrary to common expectations for an "int" return, these functions
return only a positive value -- if used correctly they cannot even
return 0 because the message header will necessarily be in the skb.
This makes the very common pattern of
if (genlmsg_end(...) < 0) { ... }
be a whole bunch of dead code. Many places also simply do
return nlmsg_end(...);
and the caller is expected to deal with it.
This also commonly (at least for me) causes errors, because it is very
common to write
if (my_function(...))
/* error condition */
and if my_function() does "return nlmsg_end()" this is of course wrong.
Additionally, there's not a single place in the kernel that actually
needs the message length returned, and if anyone needs it later then
it'll be very easy to just use skb->len there.
Remove this, and make the functions void. This removes a bunch of dead
code as described above. The patch adds lines because I did
- return nlmsg_end(...);
+ nlmsg_end(...);
+ return 0;
I could have preserved all the function's return values by returning
skb->len, but instead I've audited all the places calling the affected
functions and found that none cared. A few places actually compared
the return value with <= 0 in dump functionality, but that could just
be changed to < 0 with no change in behaviour, so I opted for the more
efficient version.
One instance of the error I've made numerous times now is also present
in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't
check for <0 or <=0 and thus broke out of the loop every single time.
I've preserved this since it will (I think) have caused the messages to
userspace to be formatted differently with just a single message for
every SKB returned to userspace. It's possible that this isn't needed
for the tools that actually use this, but I don't even know what they
are so couldn't test that changing this behaviour would be acceptable.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/xen-netfront.c
Minor overlapping changes in xen-netfront.c, mostly to do
with some buffer management changes alongside the split
of stats into TX and RX.
Signed-off-by: David S. Miller <davem@davemloft.net>
The same macros are used for rx as well. So rename it.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
The expression in ll_header_truncated() tests less than or equal, but
the warning prints less than. Update the warning.
Reported-by: Jouni Malinen <jkmalinen@gmail.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Due to a misplaced parenthesis, the expression
(unlikely(offset) < 0),
which expands to
(__builtin_expect(!!(offset), 0) < 0),
never evaluates to true. Therefore, when sending packets with
PF_PACKET/SOCK_DGRAM, packet_snd() does not abort as intended
if the creation of the layer 2 header fails.
Spotted by Coverity - CID 1259975 ("Operands don't affect result").
Fixes: 9c7077622d ("packet: make packet_snd fail on len smaller than l2 header")
Signed-off-by: Christoph Jaeger <cj@linux.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make TPACKET_V3 signal poll when block is closed rather than for every
packet. Side effect is that poll will be signaled when block retire
timer expires which didn't previously happen. Issue was visible when
sending packets at a very low frequency such that all blocks are retired
before packets are received by TPACKET_V3. This caused avoidable packet
loss. The fix ensures that the signal is sent when blocks are closed
which covers the normal path where the block is filled as well as the
path where the timer expires. The case where a block is filled without
moving to the next block (ie. all blocks are full) will still cause poll
to be signaled.
Signed-off-by: Dan Collins <dan@dcollins.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking updates from David Miller:
1) New offloading infrastructure and example 'rocker' driver for
offloading of switching and routing to hardware.
This work was done by a large group of dedicated individuals, not
limited to: Scott Feldman, Jiri Pirko, Thomas Graf, John Fastabend,
Jamal Hadi Salim, Andy Gospodarek, Florian Fainelli, Roopa Prabhu
2) Start making the networking operate on IOV iterators instead of
modifying iov objects in-situ during transfers. Thanks to Al Viro
and Herbert Xu.
3) A set of new netlink interfaces for the TIPC stack, from Richard
Alpe.
4) Remove unnecessary looping during ipv6 routing lookups, from Martin
KaFai Lau.
5) Add PAUSE frame generation support to gianfar driver, from Matei
Pavaluca.
6) Allow for larger reordering levels in TCP, which are easily
achievable in the real world right now, from Eric Dumazet.
7) Add a variable of napi_schedule that doesn't need to disable cpu
interrupts, from Eric Dumazet.
8) Use a doubly linked list to optimize neigh_parms_release(), from
Nicolas Dichtel.
9) Various enhancements to the kernel BPF verifier, and allow eBPF
programs to actually be attached to sockets. From Alexei
Starovoitov.
10) Support TSO/LSO in sunvnet driver, from David L Stevens.
11) Allow controlling ECN usage via routing metrics, from Florian
Westphal.
12) Remote checksum offload, from Tom Herbert.
13) Add split-header receive, BQL, and xmit_more support to amd-xgbe
driver, from Thomas Lendacky.
14) Add MPLS support to openvswitch, from Simon Horman.
15) Support wildcard tunnel endpoints in ipv6 tunnels, from Steffen
Klassert.
16) Do gro flushes on a per-device basis using a timer, from Eric
Dumazet. This tries to resolve the conflicting goals between the
desired handling of bulk vs. RPC-like traffic.
17) Allow userspace to ask for the CPU upon what a packet was
received/steered, via SO_INCOMING_CPU. From Eric Dumazet.
18) Limit GSO packets to half the current congestion window, from Eric
Dumazet.
19) Add a generic helper so that all drivers set their RSS keys in a
consistent way, from Eric Dumazet.
20) Add xmit_more support to enic driver, from Govindarajulu
Varadarajan.
21) Add VLAN packet scheduler action, from Jiri Pirko.
22) Support configurable RSS hash functions via ethtool, from Eyal
Perry.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1820 commits)
Fix race condition between vxlan_sock_add and vxlan_sock_release
net/macb: fix compilation warning for print_hex_dump() called with skb->mac_header
net/mlx4: Add support for A0 steering
net/mlx4: Refactor QUERY_PORT
net/mlx4_core: Add explicit error message when rule doesn't meet configuration
net/mlx4: Add A0 hybrid steering
net/mlx4: Add mlx4_bitmap zone allocator
net/mlx4: Add a check if there are too many reserved QPs
net/mlx4: Change QP allocation scheme
net/mlx4_core: Use tasklet for user-space CQ completion events
net/mlx4_core: Mask out host side virtualization features for guests
net/mlx4_en: Set csum level for encapsulated packets
be2net: Export tunnel offloads only when a VxLAN tunnel is created
gianfar: Fix dma check map error when DMA_API_DEBUG is enabled
cxgb4/csiostor: Don't use MASTER_MUST for fw_hello call
net: fec: only enable mdio interrupt before phy device link up
net: fec: clear all interrupt events to support i.MX6SX
net: fec: reset fep link status in suspend function
net: sock: fix access via invalid file descriptor
net: introduce helper macro for_each_cmsghdr
...
Note that the code _using_ ->msg_iter at that point will be very
unhappy with anything other than unshifted iovec-backed iov_iter.
We still need to convert users to proper primitives.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This merely fixes sparse warnings, without actually
adding support for the new APIs.
Still working out the best way to enable the new
functionality.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
af_packet produces lots of these:
net/packet/af_packet.c:384:39: warning: incorrect type in return expression (different modifiers)
net/packet/af_packet.c:384:39: expected struct page [pure] *
net/packet/af_packet.c:384:39: got struct page *
this seems to be because sparse does not realize that _pure
refers to function, not the returned pointer.
Tweak code slightly to avoid the warning.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When sending packets out with PF_PACKET, SOCK_RAW, ensure that the
packet is at least as long as the device's expected link layer header.
This check already exists in tpacket_snd, but not in packet_snd.
Also rate limit the warning in tpacket_snd.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This encapsulates all of the skb_copy_datagram_iovec() callers
with call argument signature "skb, offset, msghdr->msg_iov, length".
When we move to iov_iters in the networking, the iov_iter object will
sit in the msghdr.
Having a helper like this means there will be less places to touch
during that transformation.
Based upon descriptions and patch from Al Viro.
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace occurences of skb_get_queue_mapping() and follow-up
netdev_get_tx_queue() with an actual helper function.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
af_packet can currently overwrite kernel memory by out of bound
accesses, because it assumed a [new] block can always hold one frame.
This is not generally the case, even if most existing tools do it right.
This patch clamps too long frames as API permits, and issue a one time
error on syslog.
[ 394.357639] tpacket_rcv: packet too big, clamped from 5042 to 3966. macoff=82
In this example, packet header tp_snaplen was set to 3966,
and tp_len was set to 5042 (skb->len)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Fixes: f6fb8f100b ("af-packet: TPACKET_V3 flexible buffer implementation.")
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
No device driver will ever return an skb_shared_info structure with
syststamp non-zero, so remove the branch that tests for this and
optionally marks the packet timestamp as TP_STATUS_TS_SYS_HARDWARE.
Do not remove the definition TP_STATUS_TS_SYS_HARDWARE, as processes
may refer to it.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is possible by passing a netlink socket to a more privileged
executable and then to fool that executable into writing to the socket
data that happens to be valid netlink message to do something that
privileged executable did not intend to do.
To keep this from happening replace bare capable and ns_capable calls
with netlink_capable, netlink_net_calls and netlink_ns_capable calls.
Which act the same as the previous calls except they verify that the
opener of the socket had the desired permissions as well.
Reported-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The permission check in sock_diag_put_filterinfo is wrong, and it is so removed
from it's sources it is not clear why it is wrong. Move the computation
into packet_diag_dump and pass a bool of the result into sock_diag_filterinfo.
This does not yet correct the capability check but instead simply moves it to make
it clear what is going on.
Reported-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The caller needs capabilities on the namespace being queried, not on
their own namespace. This is a security bug, although it likely has
only a minor impact.
Cc: stable@vger.kernel.org
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Several spots in the kernel perform a sequence like:
skb_queue_tail(&sk->s_receive_queue, skb);
sk->sk_data_ready(sk, skb->len);
But at the moment we place the SKB onto the socket receive queue it
can be consumed and freed up. So this skb->len access is potentially
to freed up memory.
Furthermore, the skb->len can be modified by the consumer so it is
possible that the value isn't accurate.
And finally, no actual implementation of this callback actually uses
the length argument. And since nobody actually cared about it's
value, lots of call sites pass arbitrary values in such as '0' and
even '1'.
So just remove the length argument from the callback, that way there
is no confusion whatsoever and all of these use-after-free cases get
fixed as a side effect.
Based upon a patch by Eric Dumazet and his suggestion to audit this
issue tree-wide.
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, in packet_direct_xmit() we test the assigned netdevice queue
for netif_xmit_frozen_or_stopped() before doing an ndo_start_xmit().
This can have the side-effect that BQL enabled drivers which make use
of netdev_tx_sent_queue() internally, set __QUEUE_STATE_STACK_XOFF from
within the stack and would not fully fill the device's TX ring from
packet sockets with PACKET_QDISC_BYPASS enabled.
Instead, use a test without BQL bit so that bursts can be absorbed
into the NICs TX ring. Fix and code suggested by Eric Dumazet, thanks!
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since commit 015f0688f5 ("net: net: add a core netdev->tx_dropped
counter"), we can now account for TX drops from within the core
stack instead of drivers.
Therefore, fix packet_direct_xmit() and increase drop count when we
encounter a problem before driver's xmit function was called (we do
not want to doubly account for it).
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Quite often it can be useful to test with dummy or similar
devices as a blackhole sink for skbs. Such devices are only
equipped with a single txq, but marked as NETIF_F_LLTX as
they do not require locking their internal queues on xmit
(or implement locking themselves). Therefore, rather use
HARD_TX_{UN,}LOCK API, so that NETIF_F_LLTX will be respected.
trafgen mmap/TX_RING example against dummy device with config
foo: { fill(0xff, 64) } results in the following performance
improvements for such scenarios on an ordinary Core i7/2.80GHz:
Before:
Performance counter stats for 'trafgen -i foo -o du0 -n100000000' (10 runs):
160,975,944,159 instructions:k # 0.55 insns per cycle ( +- 0.09% )
293,319,390,278 cycles:k # 0.000 GHz ( +- 0.35% )
192,501,104 branch-misses:k ( +- 1.63% )
831 context-switches:k ( +- 9.18% )
7 cpu-migrations:k ( +- 7.40% )
69,382 cache-misses:k # 0.010 % of all cache refs ( +- 2.18% )
671,552,021 cache-references:k ( +- 1.29% )
22.856401569 seconds time elapsed ( +- 0.33% )
After:
Performance counter stats for 'trafgen -i foo -o du0 -n100000000' (10 runs):
133,788,739,692 instructions:k # 0.92 insns per cycle ( +- 0.06% )
145,853,213,256 cycles:k # 0.000 GHz ( +- 0.17% )
59,867,100 branch-misses:k ( +- 4.72% )
384 context-switches:k ( +- 3.76% )
6 cpu-migrations:k ( +- 6.28% )
70,304 cache-misses:k # 0.077 % of all cache refs ( +- 1.73% )
90,879,408 cache-references:k ( +- 1.35% )
11.719372413 seconds time elapsed ( +- 0.24% )
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The packet hash can be considered a property of the packet, not just
on RX path.
This patch changes name of rxhash and l4_rxhash skbuff fields to be
hash and l4_hash respectively. This includes changing uses of the
field in the code which don't call the access functions.
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Mahesh Bandewar <maheshb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 57f89bfa21 ("network: Allow af_packet to transmit +4 bytes
for VLAN packets.") added the possibility for non-mmaped frames to
send extra 4 byte for VLAN header so the MTU increases from 1500 to
1504 byte, for example.
Commit cbd89acb9e ("af_packet: fix for sending VLAN frames via
packet_mmap") attempted to fix that for the mmap part but was
reverted as it caused regressions while using eth_type_trans()
on output path.
Lets just act analogous to 57f89bfa21 and add a similar logic
to TX_RING. We presume size_max as overcharged with +4 bytes and
later on after skb has been built by tpacket_fill_skb() check
for ETH_P_8021Q header on packets larger than normal MTU. Can
be easily reproduced with a slightly modified trafgen in mmap(2)
mode, test cases:
{ fill(0xff, 12) const16(0x8100) fill(0xff, <1504|1505>) }
{ fill(0xff, 12) const16(0x0806) fill(0xff, <1500|1501>) }
Note that we need to do the test right after tpacket_fill_skb()
as sockets can have PACKET_LOSS set where we would not fail but
instead just continue to traverse the ring.
Reported-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Ben Greear <greearb@candelatech.com>
Cc: Phil Sutter <phil@nwl.cc>
Tested-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
At first glance it looks like there is a missing curly brace but
actually the code works the same either way. I have adjusted the
indenting but left the code the same.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Mathias reported that on an AMD Geode LX embedded board (ALiX)
with ath9k driver PACKET_QDISC_BYPASS, introduced in commit
d346a3fae3 ("packet: introduce PACKET_QDISC_BYPASS socket
option"), triggers a WARN_ON() coming from the driver itself
via 066dae93bd ("ath9k: rework tx queue selection and fix
queue stopping/waking").
The reason why this happened is that ndo_select_queue() call
is not invoked from direct xmit path i.e. for ieee80211 subsystem
that sets queue and TID (similar to 802.1d tag) which is being
put into the frame through 802.11e (WMM, QoS). If that is not
set, pending frame counter for e.g. ath9k can get messed up.
So the WARN_ON() in ath9k is absolutely legitimate. Generally,
the hw queue selection in ieee80211 depends on the type of
traffic, and priorities are set according to ieee80211_ac_numbers
mapping; working in a similar way as DiffServ only on a lower
layer, so that the AP can favour frames that have "real-time"
requirements like voice or video data frames.
Therefore, check for presence of ndo_select_queue() in netdev
ops and, if available, invoke it with a fallback handler to
__packet_pick_tx_queue(), so that driver such as bnx2x, ixgbe,
or mlx4 can still select a hw queue for transmission in
relation to the current CPU while e.g. ieee80211 subsystem
can make their own choices.
Reported-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a queue mapping mode to the fanout operation of af_packet
sockets. This allows user space af_packet users to better filter on flows
ingressing and egressing via a specific hardware queue, and avoids the potential
packet reordering that can occur when FANOUT_CPU is being used and irq affinity
varies.
Tested successfully by myself. applies to net-next
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
CC: "David S. Miller" <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
As David Laight suggests, we shouldn't necessarily call this
reciprocal_divide() when users didn't requested a reciprocal_value();
lets keep the basic idea and call it reciprocal_scale(). More
background information on this topic can be found in [1].
Joint work with Hannes Frederic Sowa.
[1] http://homepage.cs.uiowa.edu/~jones/bcd/divide.html
Suggested-by: David Laight <david.laight@aculab.com>
Cc: Jakub Zawadzki <darkjames-ws@darkjames.pl>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Many functions have open coded a function that returns a random
number in range [0,N-1]. Under the assumption that we have a PRNG
such as taus113 with being well distributed in [0, ~0U] space,
we can implement such a function as uword t = (n*m')>>32, where
m' is a random number obtained from PRNG, n the right open interval
border and t our resulting random number, with n,m',t in u32 universe.
Lets go with Joe and simply call it prandom_u32_max(), although
technically we have an right open interval endpoint, but that we
have documented. Other users can further be migrated to the new
prandom_u32_max() function later on; for now, we need to make sure
to migrate reciprocal_divide() users for the reciprocal_divide()
follow-up fixup since their function signatures are going to change.
Joint work with Hannes Frederic Sowa.
Cc: Jakub Zawadzki <darkjames-ws@darkjames.pl>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Doesn't bring much, but also doesn't hurt us to fix 'em:
1) In tpacket_rcv() flush dcache page we can restirct the scope
for start and end and remove one layer of indent.
2) In tpacket_destruct_skb() we can restirct the scope for ph.
3) In alloc_one_pg_vec_page() we can remove the NULL assignment
and change spacing a bit.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is a follow-up patch to f3d3342602 ("net: rework recvmsg
handler msg_name and msg_namelen logic").
DECLARE_SOCKADDR validates that the structure we use for writing the
name information to is not larger than the buffer which is reserved
for msg->msg_name (which is 128 bytes). Also use DECLARE_SOCKADDR
consistently in sendmsg code paths.
Signed-off-by: Steffen Hurrle <steffen@hurrle.net>
Suggested-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
In PF_PACKET's packet mmap(), we can avoid using one atomic_inc()
and one atomic_dec() call in skb destructor and use a percpu
reference count instead in order to determine if packets are
still pending to be sent out. Micro-benchmark with [1] that has
been slightly modified (that is, protcol = 0 in socket(2) and
bind(2)), example on a rather crappy testing machine; I expect
it to scale and have even better results on bigger machines:
./packet_mm_tx -s7000 -m7200 -z700000 em1, avg over 2500 runs:
With patch: 4,022,015 cyc
Without patch: 4,812,994 cyc
time ./packet_mm_tx -s64 -c10000000 em1 > /dev/null, stable:
With patch:
real 1m32.241s
user 0m0.287s
sys 1m29.316s
Without patch:
real 1m38.386s
user 0m0.265s
sys 1m35.572s
In function tpacket_snd(), it is okay to use packet_read_pending()
since in fast-path we short-circuit the condition already with
ph != NULL, since we have next frames to process. In case we have
MSG_DONTWAIT, we also do not execute this path as need_wait is
false here anyway, and in case of _no_ MSG_DONTWAIT flag, it is
okay to call a packet_read_pending(), because when we ever reach
that path, we're done processing outgoing frames anyway and only
look if there are skbs still outstanding to be orphaned. We can
stay lockless in this percpu counter since it's acceptable when we
reach this path for the sum to be imprecise first, but we'll level
out at 0 after all pending frames have reached the skb destructor
eventually through tx reclaim. When people pin a tx process to
particular CPUs, we expect overflows to happen in the reference
counter as on one CPU we expect heavy increase; and distributed
through ksoftirqd on all CPUs a decrease, for example. As
David Laight points out, since the C language doesn't define the
result of signed int overflow (i.e. rather than wrap, it is
allowed to saturate as a possible outcome), we have to use
unsigned int as reference count. The sum over all CPUs when tx
is complete will result in 0 again.
The BUG_ON() in tpacket_destruct_skb() we can remove as well. It
can _only_ be set from inside tpacket_snd() path and we made sure
to increase tx_ring.pending in any case before we called po->xmit(skb).
So testing for tx_ring.pending == 0 is not too useful. Instead, it
would rather have been useful to test if lower layers didn't orphan
the skb so that we're missing ring slots being put back to
TP_STATUS_AVAILABLE. But such a bug will be caught in user space
already as we end up realizing that we do not have any
TP_STATUS_AVAILABLE slots left anymore. Therefore, we're all set.
Btw, in case of RX_RING path, we do not make use of the pending
member, therefore we also don't need to use up any percpu memory
here. Also note that __alloc_percpu() already returns a zero-filled
percpu area, so initialization is done already.
[1] http://wiki.ipxwarzone.com/index.php5?title=Linux_packet_mmap
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In tpacket_snd(), when we've discovered a first frame that is
not in status TP_STATUS_SEND_REQUEST, and return a NULL buffer,
we exit the send routine in case of MSG_DONTWAIT, since we've
finished traversing the mmaped send ring buffer and don't care
about pending frames.
While doing so, we still unconditionally call an expensive
schedule() in the packet_current_frame() "error" path, which
is unnecessary in this case since it's enough to just quit
the function.
Also, in case MSG_DONTWAIT is not set, we should rather test
for need_resched() first and do schedule() only if necessary
since meanwhile pending frames could already have finished
processing and called skb destructor.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Most people acquire PF_PACKET sockets with a protocol argument in
the socket call, e.g. libpcap does so with htons(ETH_P_ALL) for
all its sockets. Most likely, at some point in time a subsequent
bind() call will follow, e.g. in libpcap with ...
memset(&sll, 0, sizeof(sll));
sll.sll_family = AF_PACKET;
sll.sll_ifindex = ifindex;
sll.sll_protocol = htons(ETH_P_ALL);
... as arguments. What happens in the kernel is that already
in socket() syscall, we install a proto hook via register_prot_hook()
if our protocol argument is != 0. Yet, in bind() we're almost
doing the same work by doing a unregister_prot_hook() with an
expensive synchronize_net() call in case during socket() the proto
was != 0, plus follow-up register_prot_hook() with a bound device
to it this time, in order to limit traffic we get.
In the case when the protocol and user supplied device index (== 0)
does not change from socket() to bind(), we can spare us doing
the same work twice. Similarly for re-binding to the same device
and protocol. For these scenarios, we can decrease create/bind
latency from ~7447us (sock-bind-2 case) to ~89us (sock-bind-1 case)
with this patch.
Alternatively, for the first case, if people care, they should
simply create their sockets with proto == 0 argument and define
the protocol during bind() as this saves a call to synchronize_net()
as well (sock-bind-3 case).
In all other cases, we're tied to user space behaviour we must not
change, also since a bind() is not strictly required. Thus, we need
the synchronize_net() to make sure no asynchronous packet processing
paths still refer to the previous elements of po->prot_hook.
In case of mmap()ed sockets, the workflow that includes bind() is
socket() -> setsockopt(<ring>) -> bind(). In that case, a pair of
{__unregister, register}_prot_hook is being called from setsockopt()
in order to install the new protocol receive handler. Thus, when
we call bind and can skip a re-hook, we have already previously
installed the new handler. For fanout, this is handled different
entirely, so we should be good.
Timings on an i7-3520M machine:
* sock-bind-1: 89 us
* sock-bind-2: 7447 us
* sock-bind-3: 75 us
sock-bind-1:
socket(PF_PACKET, SOCK_RAW, htons(ETH_P_IP)) = 3
bind(3, {sa_family=AF_PACKET, proto=htons(ETH_P_IP), if=all(0),
pkttype=PACKET_HOST, addr(0)={0, }, 20) = 0
sock-bind-2:
socket(PF_PACKET, SOCK_RAW, htons(ETH_P_IP)) = 3
bind(3, {sa_family=AF_PACKET, proto=htons(ETH_P_IP), if=lo(1),
pkttype=PACKET_HOST, addr(0)={0, }, 20) = 0
sock-bind-3:
socket(PF_PACKET, SOCK_RAW, 0) = 3
bind(3, {sa_family=AF_PACKET, proto=htons(ETH_P_IP), if=lo(1),
pkttype=PACKET_HOST, addr(0)={0, }, 20) = 0
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cleanup checkpatch errors.Specially,the second changed line
is exactly 80 columns long.
Signed-off-by: Weilong Chen <chenweilong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This enables userspace to get VLAN TPID as well as the VLAN TCI.
Signed-off-by: Atzm Watanabe <atzm@stratosphere.co.jp>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct tpacket{2,3}_hdr is aligned to a multiple of TPACKET_ALIGNMENT.
Explicitly defining and zeroing the gap of this makes additional changes
easier.
Signed-off-by: Atzm Watanabe <atzm@stratosphere.co.jp>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct tpacket{2,3}_hdr is aligned to a multiple of TPACKET_ALIGNMENT.
We may add members to them until current aligned size without forcing
userspace to call getsockopt(..., PACKET_HDRLEN, ...).
Signed-off-by: Atzm Watanabe <atzm@stratosphere.co.jp>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Changing name of function as part of making the hash in skbuff to be
generic property, not just for receive path.
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Alexander Drozdov