Currently, ip_route_me_harder after ip_vs_out_icmp
is called even if packet is not related to IPVS connection.
Move it into handle_response_icmp. Also, force rerouting
if sending to local client because IPv4 stack uses addresses
from the route.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Create new function ip_vs_defrag_user to return correct
IP_DEFRAG_xxx user depending on the hooknum. It will be needed
when we add handlers in LOCAL_OUT.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
The recent change in IP_VS_XMIT_TUNNEL to set
CHECKSUM_NONE is not correct. After adding IPIP header
skb->csum becomes invalid but the CHECKSUM_PARTIAL
case must be supported. So, use skb_forward_csum() which is
most suitable for us to allow local clients to send IPIP
to remote real server.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Delivering locally ICMP from FORWARD hook is not supported.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
This patch is needed to avoid scheduling of
packets from local real server when we add ip_vs_in
in LOCAL_OUT hook to support local client.
Currently, when ip_vs_in can not find existing
connection it tries to create new one by calling ip_vs_schedule.
The default indication from ip_vs_schedule was if
connection was scheduled to real server. If real server is
not available we try to use the bypass forwarding method
or to send ICMP error. But in some cases we do not want to use
the bypass feature. So, add flag 'ignored' to indicate if
the scheduler ignores this packet.
Make sure we do not create new connections from replies.
We can hit this problem for persistent services and local real
server when ip_vs_in is added to LOCAL_OUT hook to handle
local clients.
Also, make sure ip_vs_schedule ignores SYN packets
for Active FTP DATA from local real server. The FTP DATA
connection should be created on SYN+ACK from client to assign
correct connection daddr.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Change skb->ipvs_property semantic. This is preparation
to support ip_vs_out processing in LOCAL_OUT. ipvs_property=1
will be used to avoid expensive lookups for traffic sent by
transmitters. Now when conntrack support is not used we call
ip_vs_notrack method to avoid problems in OUTPUT and
POST_ROUTING hooks instead of exiting POST_ROUTING as before.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Avoid full checksum calculation for apps that can provide
info whether csum was broken after payload mangling. For now only
ip_vs_ftp mangles payload and it updates the csum, so the full
recalculation is avoided for all packets.
Add CHECKSUM_UNNECESSARY for snat_handler (TCP and UDP).
It is needed to support SNAT from local address for the case
when csum is fully recalculated.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Fix CHECKSUM_PARTIAL handling. Tested for IPv4 TCP,
UDP not tested because it needs network card with HW CSUM support.
May be fixes problem where IPVS can not be used in virtual boxes.
Problem appears with DNAT to local address when the local stack
sends reply in CHECKSUM_PARTIAL mode.
Fix tcp_dnat_handler and udp_dnat_handler to provide
vaddr and daddr in right order (old and new IP) when calling
tcp_partial_csum_update/udp_partial_csum_update (CHECKSUM_PARTIAL).
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
If some of the underlying devices support it, enable vlan offload on
transmit for bridge devices. This allows senders to take advantage of the
hardware support, similar to other forms of acceleration.
Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that vlan acceleration is handled consistently regardless of usage,
it is possible to enable and disable it at will. This adds support for
Ethtool operations that change the offloading status for debugging
purposes, similar to other forms of hardware acceleration.
Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently each driver that is capable of vlan hardware acceleration
must be aware of the vlan groups that are configured and then pass
the stripped tag to a specialized receive function. This is
different from other types of hardware offload in that it places a
significant amount of knowledge in the driver itself rather keeping
it in the networking core.
This makes vlan offloading function more similarly to other forms
of offloading (such as checksum offloading or TSO) by doing the
following:
* On receive, stripped vlans are passed directly to the network
core, without attempting to check for vlan groups or reconstructing
the header if no group
* vlans are made less special by folding the logic into the main
receive routines
* On transmit, the device layer will add the vlan header in software
if the hardware doesn't support it, instead of spreading that logic
out in upper layers, such as bonding.
There are a number of advantages to this:
* Fixes all bugs with drivers incorrectly dropping vlan headers at once.
* Avoids having to disable VLAN acceleration when in promiscuous mode
(good for bridging since it always puts devices in promiscuous mode).
* Keeps VLAN tag separate until given to ultimate consumer, which
avoids needing to do header reconstruction as in tg3 unless absolutely
necessary.
* Consolidates common code in core networking.
Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A struct net_device always maps to zero or one vlan groups and we
always know the device when we are looking up a group. We currently
do a hash table lookup on the device to find the group but it is
much simpler to just store a pointer.
Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently users of hardware vlan accleration need to know whether
the device supports it before generating packets. However, vlan
acceleration will soon be available in a more flexible manner so
knowing ahead of time becomes much more difficult. This adds
a software fallback path for vlan packets on devices without the
necessary offloading support, similar to other types of hardware
accleration.
Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
VLAN_GROUP_ARRAY_LEN is simply the number of possible vlan VIDs.
Since vlan groups will soon be more of an implementation detail
for vlan devices, rename the constant to be descriptive of its
actual purpose.
Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
An upcoming commit will allow packets with hardware vlan acceleration
information to be passed though more parts of the network stack, including
packets trunked through the bridge. This adds support for matching and
filtering those packets through ebtables.
Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When CONFIG_NF_CONNTRACK_SECMARK is not set we accidentally attempt to use
the secmark fielf of struct nf_conn. Problem is when that config isn't set
the field doesn't exist. whoops. Wrap the incorrect usage in the config.
Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
The current secmark code exports a secmark= field which just indicates if
there is special labeling on a packet or not. We drop this field as it
isn't particularly useful and instead export a new field secctx= which is
the actual human readable text label.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: James Morris <jmorris@namei.org>
The conntrack code can export the internal secid to userspace. These are
dynamic, can change on lsm changes, and have no meaning in userspace. We
should instead be sending lsm contexts to userspace instead. This patch sends
the secctx (rather than secid) to userspace over the netlink socket. We use a
new field CTA_SECCTX and stop using the the old CTA_SECMARK field since it did
not send particularly useful information.
Signed-off-by: Eric Paris <eparis@redhat.com>
Reviewed-by: Paul Moore <paul.moore@hp.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: James Morris <jmorris@namei.org>
Right now secmark has lots of direct selinux calls. Use all LSM calls and
remove all SELinux specific knowledge. The only SELinux specific knowledge
we leave is the mode. The only point is to make sure that other LSMs at
least test this generic code before they assume it works. (They may also
have to make changes if they do not represent labels as strings)
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Paul Moore <paul.moore@hp.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: James Morris <jmorris@namei.org>
Commit 4a5a5c73 attempted to pass decent error messages back to userspace for
netfilter errors. In xt_SECMARK.c however the patch screwed up and returned
on 0 (aka no error) early and didn't finish setting up secmark. This results
in a kernel BUG if you use SECMARK.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: James Morris <jmorris@namei.org>
This only happened when parse_extra_token was not passed
to ceph_parse_option() (hence, only happened in rbd).
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
These facilitate preallocation of pages so that we can encode into the pagelist
in an atomic context.
Signed-off-by: Greg Farnum <gregf@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
The rados block device (rbd), based on osdblk, creates a block device
that is backed by objects stored in the Ceph distributed object storage
cluster. Each device consists of a single metadata object and data
striped over many data objects.
The rbd driver supports read-only snapshots.
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
This factors out protocol and low-level storage parts of ceph into a
separate libceph module living in net/ceph and include/linux/ceph. This
is mostly a matter of moving files around. However, a few key pieces
of the interface change as well:
- ceph_client becomes ceph_fs_client and ceph_client, where the latter
captures the mon and osd clients, and the fs_client gets the mds client
and file system specific pieces.
- Mount option parsing and debugfs setup is correspondingly broken into
two pieces.
- The mon client gets a generic handler callback for otherwise unknown
messages (mds map, in this case).
- The basic supported/required feature bits can be expanded (and are by
ceph_fs_client).
No functional change, aside from some subtle error handling cases that got
cleaned up in the refactoring process.
Signed-off-by: Sage Weil <sage@newdream.net>
There is no point using RCU for dst we allocate for a very short time
(used once).
Change dst_release() to take DST_NOCACHE into account, but also change
skb_dst_set_noref() to force a refcount increment for such dst.
This is a _huge_ gain, because we dont waste memory to store xx thousand
of dsts. Instead of queueing them to RCU, we can free them instantly.
CPU caches can stay hot, re-using same memory blocks to hold temporary
dsts.
Note : remove unneeded smp_mb__before_atomic_dec(); in dst_release(),
since atomic_dec_return() implies a full memory barrier.
Stress test, 160.000.000 udp frames sent, IP route cache disabled
(DDOS).
Before:
real 0m38.091s
user 0m13.189s
sys 7m53.018s
After:
real 0m29.946s
user 0m12.157s
sys 7m40.605s
For reference, if IP route cache was enabled :
real 0m32.030s
user 0m10.521s
sys 8m15.243s
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch introduces netif_alloc_netdev_queues which is called from
register_device instead of alloc_netdev_mq. This makes TX queue
allocation symmetric with RX allocation. Also, queue locks allocation
is done in netdev_init_one_queue. Change set_real_num_tx_queues to
fail if requested number < 1 or greater than number of allocated
queues.
Signed-off-by: Tom Herbert <therbert@google.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Clean up in RX queue allocation. In netif_set_real_num_rx_queues
return error on attempt to set zero queues, or requested number is
greater than number of allocated queues. In netif_alloc_rx_queues,
do BUG_ON if queue_count is zero.
Signed-off-by: Tom Herbert <therbert@google.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In alloc_netdev_mq fail if requested queue_count < 1.
Signed-off-by: Tom Herbert <therbert@google.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In an erlier patch I modified napi_poll so that devices with IFF_MASTER polled
the per_cpu list instead of the device list for napi. I did this because the
bonding driver has no napi instances to poll, it instead expects to check the
slave devices napi instances, which napi_poll was unaware of. Looking at this
more closely however, I now see this isn't strictly needed. As the bond driver
poll_controller calls the slaves poll_controller via netpoll_poll_dev, which
recursively calls poll_napi on each slave, allowing those napi instances to get
serviced. The earlier patch isn't at all harmfull, its just not needed, so lets
revert it to make the code cleaner. Sorry for the noise,
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Reviewed-by: WANG Cong <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Lists were initialized after the module was registered. Multiple ipvsadm
processes at module load triggered a race condition that resulted in a null
pointer dereference in do_ip_vs_get_ctl(). As a result, __ip_vs_mutex
was left locked preventing all further ipvsadm commands.
Signed-off-by: Eduardo J. Blanco <ejblanco@google.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
Convert inetdev_by_index() to not increment in_dev refcount.
Callers hold RCU or RTNL, and should not decrement in_dev refcount.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We hold RTNL in ip_mc_find_dev(), no need to touch device refcount.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The sunrpc cache_ioctl function does not need the big kernel lock
because it uses its own queue_lock already.
rpc_pipe_ioctl apparently should be using i_lock like the other
operations on the pipe file descriptor do.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
IPv6 encapsulation uses a bad source address for the tunnel.
i.e. VIP will be used as local-addr and encap. dst addr.
Decapsulation will not accept this.
Example
LVS (eth1 2003::2:0:1/96, VIP 2003::2:0:100)
(eth0 2003::1:0:1/96)
RS (ethX 2003::1:0:5/96)
tcpdump
2003::2:0:100 > 2003::1:0:5: IP6 (hlim 63, next-header TCP (6) payload length: 40) 2003::3:0:10.50991 > 2003::2:0:100.http: Flags [S], cksum 0x7312 (correct), seq 3006460279, win 5760, options [mss 1440,sackOK,TS val 1904932 ecr 0,nop,wscale 3], length 0
In Linux IPv6 impl. you can't have a tunnel with an any cast address
receiving packets (I have not tried to interpret RFC 2473)
To have receive capabilities the tunnel must have:
- Local address set as multicast addr or an unicast addr
- Remote address set as an unicast addr.
- Loop back addres or Link local address are not allowed.
This causes us to setup a tunnel in the Real Server with the
LVS as the remote address, here you can't use the VIP address since it's
used inside the tunnel.
Solution
Use outgoing interface IPv6 address (match against the destination).
i.e. use ip6_route_output() to look up the route cache and
then use ipv6_dev_get_saddr(...) to set the source address of the
encapsulated packet.
Additionally, cache the results in new destination
fields: dst_cookie and dst_saddr and properly check the
returned dst from ip6_route_output. We now add xfrm_lookup
call only for the tunneling method where the source address
is a local one.
Signed-off-by:Hans Schillstrom <hans.schillstrom@ericsson.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
This patch allows to listen to events that inform about
expectations destroyed.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Peter Zijlstra found a bug in the way softirq time is accounted in
VIRT_CPU_ACCOUNTING on this thread:
http://lkml.indiana.edu/hypermail//linux/kernel/1009.2/01366.html
The problem is, softirq processing uses local_bh_disable internally. There
is no way, later in the flow, to differentiate between whether softirq is
being processed or is it just that bh has been disabled. So, a hardirq when bh
is disabled results in time being wrongly accounted as softirq.
Looking at the code a bit more, the problem exists in !VIRT_CPU_ACCOUNTING
as well. As account_system_time() in normal tick based accouting also uses
softirq_count, which will be set even when not in softirq with bh disabled.
Peter also suggested solution of using 2*SOFTIRQ_OFFSET as irq count
for local_bh_{disable,enable} and using just SOFTIRQ_OFFSET while softirq
processing. The patch below does that and adds API in_serving_softirq() which
returns whether we are currently processing softirq or not.
Also changes one of the usages of softirq_count in net/sched/cls_cgroup.c
to in_serving_softirq.
Looks like many usages of in_softirq really want in_serving_softirq. Those
changes can be made individually on a case by case basis.
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-2-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Usually the netpoll path, when preforming a napi poll can get away with just
polling all the napi instances of the configured device. Thats not the case for
the bonding driver however, as the napi instances which may wind up getting
flagged as needing polling after the poll_controller call don't belong to the
bonded device, but rather to the slave devices. Fix this by checking the device
in question for the IFF_MASTER flag, if set, we know we need to check the full
poll list for this cpu, rather than just the devices napi instance list.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The bonding driver currently modifies the netpoll structure in its xmit path
while sending frames from netpoll. This is racy, as other cpus can access the
netpoll structure in parallel. Since the bonding driver points np->dev to a
slave device, other cpus can inadvertently attempt to send data directly to
slave devices, leading to improper locking with the bonding master, lost frames,
and deadlocks. This patch fixes that up.
This patch also removes the real_dev pointer from the netpoll structure as that
data is really only used by bonding in the poll_controller, and we can emulate
its behavior by check each slave for IS_UP.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change a few checks against the hardcoded broadcast address,
0xffffffff, to ipv4_is_lbcast(). Remove some existing checks
using ipv4_is_lbcast() that are now obviously superfluous.
Signed-off-by: Andy Walls <awalls@md.metrocast.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The patch below updates broken web addresses in the kernel
Signed-off-by: Justin P. Mattock <justinmattock@gmail.com>
Cc: Maciej W. Rozycki <macro@linux-mips.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Finn Thain <fthain@telegraphics.com.au>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Dimitry Torokhov <dmitry.torokhov@gmail.com>
Cc: Mike Frysinger <vapier.adi@gmail.com>
Acked-by: Ben Pfaff <blp@cs.stanford.edu>
Acked-by: Hans J. Koch <hjk@linutronix.de>
Reviewed-by: Finn Thain <fthain@telegraphics.com.au>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Optimize processing in TIPC's bearer shutdown code, including:
1. Remove an unnecessary check to see if TIPC bearer's can exist.
2. Don't release spinlocks before calling a media-specific disabling
routine, since the routine can't sleep.
3. Make bearer_disable() operate directly on a struct bearer, instead
of needlessly taking a name and then mapping that to the struct.
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Reviewed-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It's completely unused and exporting a static symbol
makes no sense and breaks the build.
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
In error cases when the ACL is insecure or we fail to allocate a new
struct sock, we jump to the "response" label. If so, "sk" will be
null and the kernel crashes.
Signed-off-by: Nathan Holstein <nathan.holstein@gmail.com>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
Get rid of fib_hash_lock rwlock.
The fn_zone hash table resize is the noticeable part of this patch.
I added a seqlock per fn_zone, so that readers can restart their lookup
in the (very rare) case a writer expanded the hash table.
Add rcu heads in fib_alias and fib_node, use call_rcu() to defer their
freeing, and use appropriate _rcu list manipulations.
Stress test (160.000.000 udp frames sent, IP route cache disabled to
mimic DDOS attack, FIB_HASH)
Before:
real 0m41.191s
user 0m13.137s
sys 8m55.241s
After:
real 0m38.091s
user 0m13.189s
sys 7m53.018s
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
First step for RCU conversion of fib_hash :
struct fn_zone are created and never deleted.
Very classic conversion, using rcu_assign_pointer(), rcu_dereference()
and rtnl_dereference() verbs.
__rcu markers on fz_next and fn_zone_list
They are created under RTNL, we dont need fib_hash_lock anymore in
fn_new_zone().
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While looking for false sharing problems, I noticed
sizeof(struct fn_zone) was small (28 bytes) and possibly sharing a cache
line with an often written kernel structure.
Most of the time, fn_zone uses its initial hash table of 16 slots.
We can avoid the false sharing problem by embedding this initial hash
table in fn_zone itself, so that sizeof(fn_zone) > L1_CACHE_BYTES
We did a similar optimization in commit a6501e080c (Reduce memory needs
and speedup lookups)
Add a fz_revorder field to speedup fn_hash() a bit.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As CWR is stronger than CA_Disorder state, we can miscount
SACK/Reno failure into other timeouts. Not a bad problem as
it can happen only due to ECN, FRTO detecting spurious RTO
or xmit error which are the only callers of tcp_enter_cwr.
And even then losses and RTO must still follow thereafter
to actually end up into the relevant code paths.
Compile tested.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
When only fast rexmit should be done, tcp_mark_head_lost marks
L too far. Also, sacked_upto below 1 is perfectly valid number,
the packets == 0 then needs to be trapped elsewhere.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Do some cleanups of TIPC based on make namespacecheck
1. Don't export unused symbols
2. Eliminate dead code
3. Make functions and variables local
4. Rename buf_acquire to tipc_buf_acquire since it is used in several files
Compile tested only.
This make break out of tree kernel modules that depend on TIPC routines.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While doing profile analysis, I found fib_hash_table was sometime in a
cache line shared by a possibly often written kernel structure.
(CONFIG_IP_ROUTE_MULTIPATH || !CONFIG_IPV6_MULTIPLE_TABLES)
It's hard to detect because not easily reproductible.
Make sure we allocate a full cache line to keep this shared in all cpus
caches.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
fib_table_lookup() might use fls() to speedup an open coded loop.
Noticed while doing a profile analysis.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
fib_nl_delrule() calls synchronize_rcu() for no apparent reason,
while rtnl is held.
I suspect it was done to avoid an atomic_inc_not_zero() in
fib_rules_lookup(), which commit 7fa7cb7109 added anyway.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Avoid two atomic ops on found rule in fib6_rule_lookup()
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit b30973f877 (node-aware skb allocation) spread a wrong habit of
allocating net drivers skbs on a given memory node : The one closest to
the NIC hardware. This is wrong because as soon as we try to scale
network stack, we need to use many cpus to handle traffic and hit
slub/slab management on cross-node allocations/frees when these cpus
have to alloc/free skbs bound to a central node.
skb allocated in RX path are ephemeral, they have a very short
lifetime : Extra cost to maintain NUMA affinity is too expensive. What
appeared as a nice idea four years ago is in fact a bad one.
In 2010, NIC hardwares are multiqueue, or we use RPS to spread the load,
and two 10Gb NIC might deliver more than 28 million packets per second,
needing all the available cpus.
Cost of cross-node handling in network and vm stacks outperforms the
small benefit hardware had when doing its DMA transfert in its 'local'
memory node at RX time. Even trying to differentiate the two allocations
done for one skb (the sk_buff on local node, the data part on NIC
hardware node) is not enough to bring good performance.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There's a bug with radiotap vendor namespace
parsing if you don't register for the given
namespace extensions. Fix this by passing
only the unknown vendor namespaces and the
registered data to frontends, but not both.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Don't try to "optimize" rds_page_copy_user() by using kmap_atomic() and
the unsafe atomic user mode accessor functions. It's actually slower
than the straightforward code on any reasonable modern CPU.
Back when the code was written (although probably not by the time it was
actually merged, though), 32-bit x86 may have been the dominant
architecture. And there kmap_atomic() can be a lot faster than kmap()
(unless you have very good locality, in which case the virtual address
caching by kmap() can overcome all the downsides).
But these days, x86-64 may not be more populous, but it's getting there
(and if you care about performance, it's definitely already there -
you'd have upgraded your CPU's already in the last few years). And on
x86-64, the non-kmap_atomic() version is faster, simply because the code
is simpler and doesn't have the "re-try page fault" case.
People with old hardware are not likely to care about RDS anyway, and
the optimization for the 32-bit case is simply buggy, since it doesn't
verify the user addresses properly.
Reported-by: Dan Rosenberg <drosenberg@vsecurity.com>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
All file_operations should get a .llseek operation so we can make
nonseekable_open the default for future file operations without a
.llseek pointer.
The three cases that we can automatically detect are no_llseek, seq_lseek
and default_llseek. For cases where we can we can automatically prove that
the file offset is always ignored, we use noop_llseek, which maintains
the current behavior of not returning an error from a seek.
New drivers should normally not use noop_llseek but instead use no_llseek
and call nonseekable_open at open time. Existing drivers can be converted
to do the same when the maintainer knows for certain that no user code
relies on calling seek on the device file.
The generated code is often incorrectly indented and right now contains
comments that clarify for each added line why a specific variant was
chosen. In the version that gets submitted upstream, the comments will
be gone and I will manually fix the indentation, because there does not
seem to be a way to do that using coccinelle.
Some amount of new code is currently sitting in linux-next that should get
the same modifications, which I will do at the end of the merge window.
Many thanks to Julia Lawall for helping me learn to write a semantic
patch that does all this.
===== begin semantic patch =====
// This adds an llseek= method to all file operations,
// as a preparation for making no_llseek the default.
//
// The rules are
// - use no_llseek explicitly if we do nonseekable_open
// - use seq_lseek for sequential files
// - use default_llseek if we know we access f_pos
// - use noop_llseek if we know we don't access f_pos,
// but we still want to allow users to call lseek
//
@ open1 exists @
identifier nested_open;
@@
nested_open(...)
{
<+...
nonseekable_open(...)
...+>
}
@ open exists@
identifier open_f;
identifier i, f;
identifier open1.nested_open;
@@
int open_f(struct inode *i, struct file *f)
{
<+...
(
nonseekable_open(...)
|
nested_open(...)
)
...+>
}
@ read disable optional_qualifier exists @
identifier read_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
expression E;
identifier func;
@@
ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off)
{
<+...
(
*off = E
|
*off += E
|
func(..., off, ...)
|
E = *off
)
...+>
}
@ read_no_fpos disable optional_qualifier exists @
identifier read_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
@@
ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off)
{
... when != off
}
@ write @
identifier write_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
expression E;
identifier func;
@@
ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off)
{
<+...
(
*off = E
|
*off += E
|
func(..., off, ...)
|
E = *off
)
...+>
}
@ write_no_fpos @
identifier write_f;
identifier f, p, s, off;
type ssize_t, size_t, loff_t;
@@
ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off)
{
... when != off
}
@ fops0 @
identifier fops;
@@
struct file_operations fops = {
...
};
@ has_llseek depends on fops0 @
identifier fops0.fops;
identifier llseek_f;
@@
struct file_operations fops = {
...
.llseek = llseek_f,
...
};
@ has_read depends on fops0 @
identifier fops0.fops;
identifier read_f;
@@
struct file_operations fops = {
...
.read = read_f,
...
};
@ has_write depends on fops0 @
identifier fops0.fops;
identifier write_f;
@@
struct file_operations fops = {
...
.write = write_f,
...
};
@ has_open depends on fops0 @
identifier fops0.fops;
identifier open_f;
@@
struct file_operations fops = {
...
.open = open_f,
...
};
// use no_llseek if we call nonseekable_open
////////////////////////////////////////////
@ nonseekable1 depends on !has_llseek && has_open @
identifier fops0.fops;
identifier nso ~= "nonseekable_open";
@@
struct file_operations fops = {
... .open = nso, ...
+.llseek = no_llseek, /* nonseekable */
};
@ nonseekable2 depends on !has_llseek @
identifier fops0.fops;
identifier open.open_f;
@@
struct file_operations fops = {
... .open = open_f, ...
+.llseek = no_llseek, /* open uses nonseekable */
};
// use seq_lseek for sequential files
/////////////////////////////////////
@ seq depends on !has_llseek @
identifier fops0.fops;
identifier sr ~= "seq_read";
@@
struct file_operations fops = {
... .read = sr, ...
+.llseek = seq_lseek, /* we have seq_read */
};
// use default_llseek if there is a readdir
///////////////////////////////////////////
@ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @
identifier fops0.fops;
identifier readdir_e;
@@
// any other fop is used that changes pos
struct file_operations fops = {
... .readdir = readdir_e, ...
+.llseek = default_llseek, /* readdir is present */
};
// use default_llseek if at least one of read/write touches f_pos
/////////////////////////////////////////////////////////////////
@ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @
identifier fops0.fops;
identifier read.read_f;
@@
// read fops use offset
struct file_operations fops = {
... .read = read_f, ...
+.llseek = default_llseek, /* read accesses f_pos */
};
@ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @
identifier fops0.fops;
identifier write.write_f;
@@
// write fops use offset
struct file_operations fops = {
... .write = write_f, ...
+ .llseek = default_llseek, /* write accesses f_pos */
};
// Use noop_llseek if neither read nor write accesses f_pos
///////////////////////////////////////////////////////////
@ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @
identifier fops0.fops;
identifier read_no_fpos.read_f;
identifier write_no_fpos.write_f;
@@
// write fops use offset
struct file_operations fops = {
...
.write = write_f,
.read = read_f,
...
+.llseek = noop_llseek, /* read and write both use no f_pos */
};
@ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @
identifier fops0.fops;
identifier write_no_fpos.write_f;
@@
struct file_operations fops = {
... .write = write_f, ...
+.llseek = noop_llseek, /* write uses no f_pos */
};
@ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @
identifier fops0.fops;
identifier read_no_fpos.read_f;
@@
struct file_operations fops = {
... .read = read_f, ...
+.llseek = noop_llseek, /* read uses no f_pos */
};
@ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @
identifier fops0.fops;
@@
struct file_operations fops = {
...
+.llseek = noop_llseek, /* no read or write fn */
};
===== End semantic patch =====
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Julia Lawall <julia@diku.dk>
Cc: Christoph Hellwig <hch@infradead.org>
Remove all instances of legacy, or as yet to be implemented code
that is currently living within an #if 0 ... #endif block.
In the rare instance that some of it be needed in the future,
it can still be dragged out of history, but there is no need
for it to sit in mainline.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It looks like I submitted a different patch
than I tested, because clearly the code in
mac80211 is missing actually propagating the
requested SMPS mode. Fix that!
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Using the frame registration notification, we
can see when probe requests are requested and
notify the low-level driver via filtering. The
flag is also set in AP and IBSS modes.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Drivers may need to adjust their filters according
to frame registrations, so notify them about them.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
ip_vs_dbg_callid() and IP_VS_DEBUG_CALLID() are only needed
it CONFIG_IP_VS_DEBUG is defined.
This resolves the following build warning when CONFIG_IP_VS_DEBUG is
not defined.
net/netfilter/ipvs/ip_vs_pe_sip.c:11: warning: 'ip_vs_dbg_callid' defined but not used
Reported-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Many of the used macros are just there for userspace compatibility.
Substitute the in-kernel code to directly use the terminal macro
and stuff the defines into #ifndef __KERNEL__ sections.
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
The index cannot be used to reliably reconstruct a phy
name, so explicitly add the phy name to sysfs so that scripts
can figure out the parent phy device for a particular
wireless interface.
Signed-off-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
We tried very hard to remove all possible dev_hold()/dev_put() pairs in
network stack, using RCU conversions.
There is still an unavoidable device refcount change for every dst we
create/destroy, and this can slow down some workloads (routers or some
app servers, mmap af_packet)
We can switch to a percpu refcount implementation, now dynamic per_cpu
infrastructure is mature. On a 64 cpus machine, this consumes 256 bytes
per device.
On x86, dev_hold(dev) code :
before
lock incl 0x280(%ebx)
after:
movl 0x260(%ebx),%eax
incl fs:(%eax)
Stress bench :
(Sending 160.000.000 UDP frames,
IP route cache disabled, dual E5540 @2.53GHz,
32bit kernel, FIB_TRIE)
Before:
real 1m1.662s
user 0m14.373s
sys 12m55.960s
After:
real 0m51.179s
user 0m15.329s
sys 10m15.942s
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Update conf_state with L2CAP_CONF_REQ_SENT before send config_req out in
l2cap_config_req().
Signed-off-by: Haijun Liu <haijun.liu@atheros.com>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
&err points to the proper error set by bt_skb_send_alloc() when it
fails.
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
The Bluetooth core uses the the BD_ADDR in the opposite order from the
human readable order. So we are changing batostr() to print in the
correct order and then removing some baswap(), as they are not needed
anymore.
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
A value was attributed to 'src', but no one was using.
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
BLUETOOTH SPECIFICATION Version 4.0 [Vol 3] page 36 mentioned
"Note: Start Fragments always begin with the Basic L2CAP header
of a PDU."
Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@nokia.com>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
Current Bluetooth code assembles fragments of big L2CAP packets
in l2cap_recv_acldata and then checks allowed L2CAP size in
assemled L2CAP packet (pi->imtu < skb->len).
The patch moves allowed L2CAP size check to the early stage when
we receive the first fragment of L2CAP packet. We do not need to
reserve and keep L2CAP fragments for bad packets.
Updated version after comments from Mat Martineau <mathewm@codeaurora.org>
and Gustavo Padovan <padovan@profusion.mobi>.
Trace below is received when using stress tools sending big
fragmented L2CAP packets.
...
[ 1712.798492] swapper: page allocation failure. order:4, mode:0x4020
[ 1712.804809] [<c0031870>] (unwind_backtrace+0x0/0xdc) from [<c00a1f70>]
(__alloc_pages_nodemask+0x4)
[ 1712.814666] [<c00a1f70>] (__alloc_pages_nodemask+0x47c/0x4d4) from
[<c00a1fd8>] (__get_free_pages+)
[ 1712.824645] [<c00a1fd8>] (__get_free_pages+0x10/0x3c) from [<c026eb5c>]
(__alloc_skb+0x4c/0xfc)
[ 1712.833465] [<c026eb5c>] (__alloc_skb+0x4c/0xfc) from [<bf28c738>]
(l2cap_recv_acldata+0xf0/0x1f8 )
[ 1712.843322] [<bf28c738>] (l2cap_recv_acldata+0xf0/0x1f8 [l2cap]) from
[<bf0094ac>] (hci_rx_task+0x)
...
Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@nokia.com>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
Clearing the blacklist in hci_dev_do_close() would mean that user space
needs to do extra work to re-block devices after a DEVDOWN-DEVUP cycle.
This patch removes the clearing of the blacklist in this case and
thereby saves user space from the extra work.
Signed-off-by: Johan Hedberg <johan.hedberg@nokia.com>
Acked-by: Ville Tervo <ville.tervo@nokia.com>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
L2CAP ERTM sockets can be opened with the SOCK_STREAM socket type,
which is a mandatory request for ERTM mode.
However, these sockets still have SOCK_SEQPACKET read semantics when
bt_sock_recvmsg() is used to pull data from the receive queue. If the
application is only reading part of a frame, then the unread portion
of the frame is discarded. If the application requests more bytes
than are in the current frame, only the current frame's data is
returned.
This patch utilizes common code derived from RFCOMM's recvmsg()
function to make L2CAP SOCK_STREAM reads behave like RFCOMM reads (and
other SOCK_STREAM sockets in general). The application may read one
byte at a time from the input stream and not lose any data, and may
also read across L2CAP frame boundaries.
Signed-off-by: Mat Martineau <mathewm@codeaurora.org>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
To reduce code duplication, have rfcomm_sock_recvmsg() call
bt_sock_stream_recvmsg(). The common bt_sock_stream_recvmsg()
code is nearly identical, with the RFCOMM-specific functionality
for deferred setup and connection unthrottling left in
rfcomm_sock_recvmsg().
Signed-off-by: Mat Martineau <mathewm@codeaurora.org>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
This commit adds a bt_sock_stream_recvmsg() function for use by any
Bluetooth code that uses SOCK_STREAM sockets. This code is copied
from rfcomm_sock_recvmsg() with minimal modifications to remove
RFCOMM-specific functionality and improve readability.
L2CAP (with the SOCK_STREAM socket type) and RFCOMM have common needs
when it comes to reading data. Proper stream read semantics require
that applications can read from a stream one byte at a time and not
lose any data. The RFCOMM code already operated on and pulled data
from the underlying L2CAP socket, so very few changes were required to
make the code more generic for use with non-RFCOMM data over L2CAP.
Applications that need more awareness of L2CAP frame boundaries are
still free to use SOCK_SEQPACKET sockets, and may verify that they
connection did not fall back to basic mode by calling getsockopt().
Signed-off-by: Mat Martineau <mathewm@codeaurora.org>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
Valid L2CAP PSMs are odd numbers, and the least significant bit of the
most significant byte must be 0.
Signed-off-by: Mat Martineau <mathewm@codeaurora.org>
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
According to the ETSI 3GPP TS 07.10 the default bit rate value for RFCOMM
is 9600 bit/s. Return this bit rate in case of RPN request and accept other
sane bit rates proposed by the sender in RPM command.
Signed-off-by: Yuri Kululin <ext-yuri.kululin@nokia.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
HCI transport drivers may not know what type of radio an AMP device has
so only say whether they're BR/EDR or AMP devices.
Signed-off-by: David Vrabel <david.vrabel@csr.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
This omits the redundant "DCCP:" in warning messages, since DCCP_WARN() already
echoes the function name, avoiding messages like
kernel: [10988.766503] dccp_close: DCCP: ABORT -- 209 bytes unread
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
This schedules an Ack when receiving a timestamp, exploiting the
existing inet_csk_schedule_ack() function, saving one case in the
`dccp_ack_pending()' function.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
This patch generalises the task of determining data loss from RFC 4340, 7.7.1.
Let S_A, S_B be sequence numbers such that S_B is "after" S_A, and let
N_B be the NDP count of packet S_B. Then, using modulo-2^48 arithmetic,
D = S_B - S_A - 1 is an upper bound of the number of lost data packets,
D - N_B is an approximation of the number of lost data packets
(there are cases where this is not exact).
The patch implements this as
dccp_loss_count(S_A, S_B, N_B) := max(S_B - S_A - 1 - N_B, 0)
Signed-off-by: Ivo Calado <ivocalado@embedded.ufcg.edu.br>
Signed-off-by: Erivaldo Xavier <desadoc@gmail.com>
Signed-off-by: Leandro Sales <leandroal@gmail.com>
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
This removes the argument `more' from ccid_hc_tx_packet_sent, since it was
nowhere used in the entire code.
(Btw, this argument was not even used in the original KAME code where the
function initially came from; compare the variable moreToSend in the
freebsd61-dccp-kame-28.08.2006.patch kept by Emmanuel Lochin.)
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
After moving the assignment of GAR/ISS from dccp_connect_init() to
dccp_transmit_skb(), the former function becomes very small, so that
a merger with dccp_connect() suggests itself.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
This fixes a problem and a potential loophole with regard to seqno/ackno
validity: currently the initial adjustments to AWL/SWL are only performed
once at the begin of the connection, during the handshake.
Since the Sequence Window feature is always greater than Wmin=32 (7.5.2),
it is however necessary to perform these adjustments at least for the first
W/W' (variables as per 7.5.1) packets in the lifetime of a connection.
This requirement is complicated by the fact that W/W' can change at any time
during the lifetime of a connection.
Therefore it is better to perform that safety check each time SWL/AWL are
updated, as implemented by the patch.
A second problem solved by this patch is that the remote/local Sequence Window
feature values (which set the bounds for AWL/SWL/SWH) are undefined until the
feature negotiation has completed.
During the initial handshake we have more stringent sequence number protection;
the changes added by this patch effect that {A,S}W{L,H} are within the correct
bounds at the instant that feature negotiation completes (since the SeqWin
feature activation handlers call dccp_update_gsr/gss()).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
struct dst_ops tracks number of allocated dst in an atomic_t field,
subject to high cache line contention in stress workload.
Switch to a percpu_counter, to reduce number of time we need to dirty a
central location. Place it on a separate cache line to avoid dirtying
read only fields.
Stress test :
(Sending 160.000.000 UDP frames,
IP route cache disabled, dual E5540 @2.53GHz,
32bit kernel, FIB_TRIE, SLUB/NUMA)
Before:
real 0m51.179s
user 0m15.329s
sys 10m15.942s
After:
real 0m45.570s
user 0m15.525s
sys 9m56.669s
With a small reordering of struct neighbour fields, subject of a
following patch, (to separate refcnt from other read mostly fields)
real 0m41.841s
user 0m15.261s
sys 8m45.949s
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a seqlock in struct neighbour to protect neigh->ha[], and avoid
dirtying neighbour in stress situation (many different flows / dsts)
Dirtying takes place because of read_lock(&n->lock) and n->used writes.
Switching to a seqlock, and writing n->used only on jiffies changes
permits less dirtying.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
