With proliferation of bit fields in sk_buff, __copy_skb_header() became
quite expensive, showing as the most expensive function in a GSO
workload.
__copy_skb_header() performance is also critical for non GSO TCP
operations, as it is used from skb_clone()
This patch carefully moves all the fields that were not copied in a
separate zone : cloned, nohdr, fclone, peeked, head_frag, xmit_more
Then I moved all other fields and all other copied fields in a section
delimited by headers_start[0]/headers_end[0] section so that we
can use a single memcpy() call, inlined by compiler using long
word load/stores.
I also tried to make all copies in the natural orders of sk_buff,
to help hardware prefetching.
I made sure sk_buff size did not change.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cache skb_shinfo(skb) in a variable to avoid computing it multiple
times.
Reorganize the tests to remove one indentation level.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
csum_partial() is a generic function which is not optimised for small fixed
length calculations, and its use requires to store "from" and "to" values in
memory while we already have them available in registers. This also has impact,
especially on RISC processors. In the same spirit as the change done by
Eric Dumazet on csum_replace2(), this patch rewrites inet_proto_csum_replace4()
taking into account RFC1624.
I spotted during a NATted tcp transfert that csum_partial() is one of top 5
consuming functions (around 8%), and the second user of csum_partial() is
inet_proto_csum_replace4().
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While profiling TCP stack, I noticed one useless atomic operation
in tcp_sendmsg(), caused by skb_header_release().
It turns out all current skb_header_release() users have a fresh skb,
that no other user can see, so we can avoid one atomic operation.
Introduce __skb_header_release() to clearly document this.
This gave me a 1.5 % improvement on TCP_RR workload.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
No caller or macro uses the return value so make all
the functions return void.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The send_check logic was only interesting in cases of TCP offload and
UDP UFO where the checksum needed to be initialized to the pseudo
header checksum. Now we've moved that logic into the related
gso_segment functions so gso_send_check is no longer needed.
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
arch/mips/net/bpf_jit.c
drivers/net/can/flexcan.c
Both the flexcan and MIPS bpf_jit conflicts were cases of simple
overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit ce93718fb7 ("net: Don't keep
around original SKB when we software segment GSO frames") frees the
original skb after software GSO even for dodgy gso skbs. This breaks
the stream throughput from untrusted sources, since only header
checking was done during software GSO instead of a true
segmentation. This patch fixes this by freeing the original gso skb
only when it was really segmented by software.
Fixes ce93718fb7 ("net: Don't keep
around original SKB when we software segment GSO frames.")
Cc: David S. Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extract from sock_alloc_send_pskb() code building skb with frags,
so that we can reuse this in other contexts.
Intent is to use it from tcp_send_rcvq(), tcp_collapse(), ...
We also want to replace some skb_linearize() calls to a more reliable
strategy in pathological cases where we need to reduce number of frags.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can allow a segment with FIN to be aggregated,
if we take care to add tcp flags,
and if skb_try_coalesce() takes care of zero sized skbs.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
__netdev_adjacent_dev_insert may add adjust device of different net
namespace, without proper check it leads to emergence of broken
sysfs links from/to devices in another namespace.
Fix: rewrite netdev_adjacent_is_neigh_list macro as a function,
move net_eq check into netdev_adjacent_is_neigh_list.
(thanks David)
related to: 4c75431ac3
Signed-off-by: Alexander Fomichev <git.user@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit "net: bpf: make eBPF interpreter images read-only" has changed bpf_prog
to be vmalloc()ed but never handled some of the errors paths of the old code.
On error within sk_attach_filter (which userspace can easily trigger), we'd
kfree() the vmalloc()ed memory, and leak the internal bpf_work_struct.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently we have 2 pkt_type_offset functions doing the same thing and
spread across the architecture files. Remove those and replace them
with a PKT_TYPE_OFFSET macro helper which gets the constant value from a
zero sized sk_buff member right in front of the bitfield with offsetof.
This new offset marker does not change size of struct sk_buff.
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Markos Chandras <markos.chandras@imgtec.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Daniel Borkmann <dborkman@redhat.com>
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Signed-off-by: Denis Kirjanov <kda@linux-powerpc.org>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
These code is now protected by rtnl lock, rcu read lock
is useless now.
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add __rcu notation to qdisc handling by doing this we can make
smatch output more legible. And anyways some of the cases should
be using rcu_dereference() see qdisc_all_tx_empty(),
qdisc_tx_chainging(), and so on.
Also *wake_queue() API is commonly called from driver timer routines
without rcu lock or rtnl lock. So I added rcu_read_lock() blocks
around netif_wake_subqueue and netif_tx_wake_queue.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a possible issue with the use, or lack thereof of sk_refcnt and
sk_wmem_alloc in the wifi ack status functionality.
Specifically if a socket were to request acknowledgements, and the socket
were to have sk_refcnt drop to 0 resulting in it waiting on sk_wmem_alloc
to reach 0 it would be possible to have sock_queue_err_skb orphan the last
buffer, resulting in __sk_free being called on the socket. After this the
buffer is enqueued on sk_error_queue, however the queue has already been
flushed resulting in at least a memory leak, if not a data corruption.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This change adds some documentation to the call skb_clone_sk. This is
meant to help clarify the purpose of the function for other developers.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/core/net_namespace.c:227:18: warning: incorrect type in argument 1
(different address spaces)
net/core/net_namespace.c:227:18: expected void const *<noident>
net/core/net_namespace.c:227:18: got struct net_generic [noderef]
<asn:4>*gen
We can use rcu_access_pointer() here as read-side access to the pointer
was removed at least one grace period ago.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reported by Mikulas Patocka, kmemcheck currently barks out a
false positive since we don't have special kmemcheck annotation
for bitfields used in bpf_prog structure.
We currently have jited:1, len:31 and thus when accessing len
while CONFIG_KMEMCHECK enabled, kmemcheck throws a warning that
we're reading uninitialized memory.
As we don't need the whole bit universe for pages member, we
can just split it to u16 and use a bool flag for jited instead
of a bitfield.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit d9b2938aab ("net: attempt a single high order allocation)
I forgot to update kerneldoc, as @prio parameter was renamed to @gfp
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch updates some of the flow_dissector api so that it can be used to
parse the length of ethernet buffers stored in fragments. Most of the
changes needed were to __skb_get_poff as it needed to be updated to support
sending a linear buffer instead of a skb.
I have split __skb_get_poff into two functions, the first is skb_get_poff
and it retains the functionality of the original __skb_get_poff. The other
function is __skb_get_poff which now works much like __skb_flow_dissect in
relation to skb_flow_dissect in that it provides the same functionality but
works with just a data buffer and hlen instead of needing an skb.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since sock_efree and sock_demux are essentially the same code for non-TCP
sockets and the case where CONFIG_INET is not defined we can combine the
code or replace the call to sock_edemux in several spots. As a result we
can avoid a bit of unnecessary code or code duplication.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The phy timestamping takes a different path than the regular timestamping
does in that it will create a clone first so that the packets needing to be
timestamped can be placed in a queue, or the context block could be used.
In order to support these use cases I am pulling the core of the code out
so it can be used in other drivers beyond just phy devices.
In addition I have added a destructor named sock_efree which is meant to
provide a simple way for dropping the reference to skb exceptions that
aren't part of either the receive or send windows for the socket, and I
have removed some duplication in spots where this destructor could be used
in place of sock_edemux.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This change merges the shared bits that exist between skb_tx_tstamp and
skb_complete_tx_timestamp. By doing this we can avoid the two diverging as
there were already changes pushed into skb_tx_tstamp that hadn't made it
into the other function.
In addition this resolves issues with the fact that
skb_complete_tx_timestamp was included in linux/skbuff.h even though it was
only compiled in if phy timestamping was enabled.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fix spelling typo found in DocBook/networking.xml.
It is because the neworking.xml is generated from comments
in the source, I have to fix typo in comments within the source.
Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds new ethtool cmd, ETHTOOL_GTUNABLE & ETHTOOL_STUNABLE for getting
tunable values from driver.
Add get_tunable and set_tunable to ethtool_ops. Driver implements these
functions for getting/setting tunable value.
Signed-off-by: Govindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Marcel reported to see the following message when autoloading
is being triggered when adding nlmon device:
Loading kernel module for a network device with
CAP_SYS_MODULE (deprecated). Use CAP_NET_ADMIN and alias
netdev-nlmon instead.
This false-positive happens despite with having correct
capabilities set, e.g. through issuing `ip link del dev nlmon`
more than once on a valid device with name nlmon, but Marcel
has also seen it on creation time when no nlmon module is
previously compiled-in or loaded as module and the device
name equals a link type name (e.g. nlmon, vxlan, team).
Stephen says:
The netdev module alias is a hold over from the past. For
normal devices, people used to create a alias eth0 to and
point it to the type of network device used, that was back
in the bad old ISA days before real discovery.
Also, the tunnels create module alias for the control device
and ip used to use this to autoload the tunnel device.
The message is bogus and should just be removed, I also see
it in a couple of other cases where tap devices are renamed
for other usese.
As mentioned in 8909c9ad8f ("net: don't allow CAP_NET_ADMIN
to load non-netdev kernel modules"), we nevertheless still
might want to leave the old autoloading behaviour in place
as it could break old scripts, so for now, lets just remove
the log message as Stephen suggests.
Reference: http://thread.gmane.org/gmane.linux.kernel/1105168
Reported-by: Marcel Holtmann <marcel@holtmann.org>
Suggested-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Vasiliy Kulikov <segoon@openwall.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
With eBPF getting more extended and exposure to user space is on it's way,
hardening the memory range the interpreter uses to steer its command flow
seems appropriate. This patch moves the to be interpreted bytecode to
read-only pages.
In case we execute a corrupted BPF interpreter image for some reason e.g.
caused by an attacker which got past a verifier stage, it would not only
provide arbitrary read/write memory access but arbitrary function calls
as well. After setting up the BPF interpreter image, its contents do not
change until destruction time, thus we can setup the image on immutable
made pages in order to mitigate modifications to that code. The idea
is derived from commit 314beb9bca ("x86: bpf_jit_comp: secure bpf jit
against spraying attacks").
This is possible because bpf_prog is not part of sk_filter anymore.
After setup bpf_prog cannot be altered during its life-time. This prevents
any modifications to the entire bpf_prog structure (incl. function/JIT
image pointer).
Every eBPF program (including classic BPF that are migrated) have to call
bpf_prog_select_runtime() to select either interpreter or a JIT image
as a last setup step, and they all are being freed via bpf_prog_free(),
including non-JIT. Therefore, we can easily integrate this into the
eBPF life-time, plus since we directly allocate a bpf_prog, we have no
performance penalty.
Tested with seccomp and test_bpf testsuite in JIT/non-JIT mode and manual
inspection of kernel_page_tables. Brad Spengler proposed the same idea
via Twitter during development of this patch.
Joint work with Hannes Frederic Sowa.
Suggested-by: Brad Spengler <spender@grsecurity.net>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Alexei Starovoitov <ast@plumgrid.com>
Cc: Kees Cook <keescook@chromium.org>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit 50cbe9ab5f ("net: Validate xmit SKBs right when we
pull them out of the qdisc") the validation code was moved out of
dev_hard_start_xmit and into dequeue_skb.
However this overlooked the fact that we do not always enqueue
the skb onto a qdisc. First situation is if qdisc have flag
TCQ_F_CAN_BYPASS and qdisc is empty. Second situation is if
there is no qdisc on the device, which is a common case for
software devices.
Originally spotted and inital patch by Alexander Duyck.
As a result Alex was seeing issues trying to connect to a
vhost_net interface after commit 50cbe9ab5f was applied.
Added a call to validate_xmit_skb() in __dev_xmit_skb(), in the
code path for qdiscs with TCQ_F_CAN_BYPASS flag, and in
__dev_queue_xmit() when no qdisc.
Also handle the error situation where dev_hard_start_xmit() could
return a skb list, and does not return dev_xmit_complete(rc) and
falls through to the kfree_skb(), in that situation it should
call kfree_skb_list().
Fixes: 50cbe9ab5f ("net: Validate xmit SKBs right when we pull them out of the qdisc")
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Depending on which parameters were updated, the changes were not propagated via
the notifier chain and netlink.
The new flag has been set only when the change did not cause a call to the
notifier chain and/or to the netlink notification functions.
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no functional changes with this commit, it only prepares the next one.
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The only effect of this patch is to print a warning if IFLA_LINKMODE is updated
and a following change fails.
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The only effect of this patch is to print a warning if IFLA_TXQLEN is updated
and a following change fails.
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sk->sk_error_queue is dequeued in four locations. All share the
exact same logic. Deduplicate.
Also collapse the two critical sections for dequeue (at the top of
the recv handler) and signal (at the bottom).
This moves signal generation for the next packet forward, which should
be harmless.
It also changes the behavior if the recv handler exits early with an
error. Previously, a signal for follow-up packets on the errqueue
would then not be scheduled. The new behavior, to always signal, is
arguably a bug fix.
For rxrpc, the change causes the same function to be called repeatedly
for each queued packet (because the recv handler == sk_error_report).
It is likely that all packets will fail for the same reason (e.g.,
memory exhaustion).
This code runs without sk_lock held, so it is not safe to trust that
sk->sk_err is immutable inbetween releasing q->lock and the subsequent
test. Introduce int err just to avoid this potential race.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This flag indicates that an invalid checksum was detected in the
packet. __skb_mark_checksum_bad helper function was added to set this.
Checksums can be marked bad from a driver or the GRO path (the latter
is implemented in this patch). csum_bad is checked in
__skb_checksum_validate_complete (i.e. calling that when ip_summed ==
CHECKSUM_NONE).
csum_bad works in conjunction with ip_summed value. In the case that
ip_summed is CHECKSUM_NONE and csum_bad is set, this implies that the
first (or next) checksum encountered in the packet is bad. When
ip_summed is CHECKSUM_UNNECESSARY, the first checksum after the last
one validated is bad. For example, if ip_summed == CHECKSUM_UNNECESSARY,
csum_level == 1, and csum_bad is set-- then the third checksum in the
packet is bad. In the normal path, the packet will be dropped when
processing the protocol layer of the bad checksum:
__skb_decr_checksum_unnecessary called twice for the good checksums
changing ip_summed to CHECKSUM_NONE so that
__skb_checksum_validate_complete is called to validate the third
checksum and that will fail since csum_bad is set.
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Then testing the TX limits of the stack, then it is useful to
be-able to disable the do_gettimeofday() timetamping on every packet.
This implements a pktgen flag NO_TIMESTAMP which will disable this
call to do_gettimeofday().
The performance change on (my system E5-2695) with skb_clone=0, goes
from TX 2,423,751 pps to 2,567,165 pps with flag NO_TIMESTAMP. Thus,
the cost of do_gettimeofday() or saving is approx 23 nanosec.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Just maintain the list properly by returning the head of the remaining
SKB list from dev_hard_start_xmit().
Signed-off-by: David S. Miller <davem@davemloft.net>
dev_hard_start_xmit() does two things, it first validates and
canonicalizes the SKB, then it actually sends it.
Make a set of helper functions for doing the first part.
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a slight policy change happening here as well.
The previous code would drop the entire rest of the GSO skb if any of
them got, for example, a congestion notification.
That makes no sense, anything NET_XMIT_MASK and below is something
like congestion or policing. And in the congestion case it doesn't
even mean the packet was actually dropped.
Just continue until dev_xmit_complete() evaluates to false.
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow GRO path to "consume" checksums provided in CHECKSUM_UNNECESSARY
and to report new checksums verfied for use in fallback to normal
path.
Change GRO checksum path to track csum_level using a csum_cnt field
in NAPI_GRO_CB. On GRO initialization, if ip_summed is
CHECKSUM_UNNECESSARY set NAPI_GRO_CB(skb)->csum_cnt to
skb->csum_level + 1. For each checksum verified, decrement
NAPI_GRO_CB(skb)->csum_cnt while its greater than zero. If a checksum
is verfied and NAPI_GRO_CB(skb)->csum_cnt == 0, we have verified a
deeper checksum than originally indicated in skbuf so increment
csum_level (or initialize to CHECKSUM_UNNECESSARY if ip_summed is
CHECKSUM_NONE or CHECKSUM_COMPLETE).
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit ed98df3361 ("net: use __GFP_NORETRY for high order
allocations") we tried to address one issue caused by order-3
allocations.
We still observe high latencies and system overhead in situations where
compaction is not successful.
Instead of trying order-3, order-2, and order-1, do a single order-3
best effort and immediately fallback to plain order-0.
This mimics slub strategy to fallback to slab min order if the high
order allocation used for performance failed.
Order-3 allocations give a performance boost only if they can be done
without recurring and expensive memory scan.
Quoting David :
The page allocator relies on synchronous (sync light) memory compaction
after direct reclaim for allocations that don't retry and deferred
compaction doesn't work with this strategy because the allocation order
is always decreasing from the previous failed attempt.
This means sync light compaction will always be encountered if memory
cannot be defragmented or reclaimed several times during the
skb_page_frag_refill() iteration.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: David Rientjes <rientjes@google.com>
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>
