We have observed meters working unexpected if traffic is 3+Gbit/s
with multiple connections.
now_ms is not pretected by meter->lock, we may get a negative
long_delta_ms when another cpu updated meter->used, then:
delta_ms = (u32)long_delta_ms;
which will be a large value.
band->bucket += delta_ms * band->rate;
then we get a wrong band->bucket.
OpenVswitch userspace datapath has fixed the same issue[1] some
time ago, and we port the implementation to kernel datapath.
[1] https://patchwork.ozlabs.org/project/openvswitch/patch/20191025114436.9746-1-i.maximets@ovn.org/
Fixes: 96fbc13d7e ("openvswitch: Add meter infrastructure")
Signed-off-by: Tao Liu <thomas.liu@ucloud.cn>
Suggested-by: Ilya Maximets <i.maximets@ovn.org>
Reviewed-by: Ilya Maximets <i.maximets@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implementation of meters supposed to be a classic token bucket with 2
typical parameters: rate and burst size.
Burst size in this schema is the maximum number of bytes/packets that
could pass without being rate limited.
Recent changes to userspace datapath made meter implementation to be
in line with the kernel one, and this uncovered several issues.
The main problem is that maximum bucket size for unknown reason
accounts not only burst size, but also the numerical value of rate.
This creates a lot of confusion around behavior of meters.
For example, if rate is configured as 1000 pps and burst size set to 1,
this should mean that meter will tolerate bursts of 1 packet at most,
i.e. not a single packet above the rate should pass the meter.
However, current implementation calculates maximum bucket size as
(rate + burst size), so the effective bucket size will be 1001. This
means that first 1000 packets will not be rate limited and average
rate might be twice as high as the configured rate. This also makes
it practically impossible to configure meter that will have burst size
lower than the rate, which might be a desirable configuration if the
rate is high.
Inability to configure low values of a burst size and overall inability
for a user to predict what will be a maximum and average rate from the
configured parameters of a meter without looking at the OVS and kernel
code might be also classified as a security issue, because drop meters
are frequently used as a way of protection from DoS attacks.
This change removes rate from the calculation of a bucket size, making
it in line with the classic token bucket algorithm and essentially
making the rate and burst tolerance being predictable from a users'
perspective.
Same change proposed for the userspace implementation.
Fixes: 96fbc13d7e ("openvswitch: Add meter infrastructure")
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
MAINTAINERS
- keep Chandrasekar
drivers/net/ethernet/mellanox/mlx5/core/en_main.c
- simple fix + trust the code re-added to param.c in -next is fine
include/linux/bpf.h
- trivial
include/linux/ethtool.h
- trivial, fix kdoc while at it
include/linux/skmsg.h
- move to relevant place in tcp.c, comment re-wrapped
net/core/skmsg.c
- add the sk = sk // sk = NULL around calls
net/tipc/crypto.c
- trivial
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
'struct ovs_zone_limit' has more members than initialized in
ovs_ct_limit_get_default_limit(). The rest of the memory is a random
kernel stack content that ends up being sent to userspace.
Fix that by using designated initializer that will clear all
non-specified fields.
Fixes: 11efd5cb04 ("openvswitch: Support conntrack zone limit")
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Acked-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
'skb_push()'/'skb_postpush_rcsum()' can be replaced by an equivalent
'skb_push_rcsum()' which is less verbose.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
s/subsytem/subsystem/
Signed-off-by: Bhaskar Chowdhury <unixbhaskar@gmail.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is not unusual to have the bridge port down. Sometimes
it has the old MTU, which is fine since it's not being used.
However, the kernel spams the log with a warning message
when a packet is going to be sent over such port. Fix that
by warning only if the interface is UP.
Signed-off-by: Flavio Leitner <fbl@sysclose.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
When openvswitch conntrack offload with act_ct action. The first rule
do conntrack in the act_ct in tc subsystem. And miss the next rule in
the tc and fallback to the ovs datapath but miss set post_ct flag
which will lead the ct_state_key with -trk flag.
Fixes: 7baf2429a1 ("net/sched: cls_flower add CT_FLAGS_INVALID flag support")
Signed-off-by: wenxu <wenxu@ucloud.cn>
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As requested by upstream OVS, added some error messages in the
validate_and_copy_dec_ttl function.
Includes a small cleanup, which removes an unnecessary parameter
from the dec_ttl_exception_handler() function.
Reported-by: Flavio Leitner <fbl@sysclose.org>
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Acked-by: Flavio Leitner <fbl@sysclose.org>
Link: https://lore.kernel.org/r/161054576573.26637.18396634650212670580.stgit@ebuild
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently, the exception actions are not processed correctly as the wrong
dataset is passed. This change fixes this, including the misleading
comment.
In addition, a check was added to make sure we work on an IPv4 packet,
and not just assume if it's not IPv6 it's IPv4.
This was all tested using OVS with patch,
https://patchwork.ozlabs.org/project/openvswitch/list/?series=21639,
applied and sending packets with a TTL of 1 (and 0), both with IPv4
and IPv6.
Fixes: 69929d4c49 ("net: openvswitch: fix TTL decrement action netlink message format")
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Link: https://lore.kernel.org/r/160733569860.3007.12938188180387116741.stgit@wsfd-netdev64.ntdv.lab.eng.bos.redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
xdp_return_frame_bulk() needs to pass a xdp_buff
to __xdp_return().
strlcpy got converted to strscpy but here it makes no
functional difference, so just keep the right code.
Conflicts:
net/netfilter/nf_tables_api.c
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Simplify the return expression.
Signed-off-by: Zheng Yongjun <zhengyongjun3@huawei.com>
Reviewed-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix to return a negative error code from the error handling
case instead of 0, as done elsewhere in this function.
Changing 'return start' to 'return action_start' can fix this bug.
Fixes: 69929d4c49 ("net: openvswitch: fix TTL decrement action netlink message format")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wang Hai <wanghai38@huawei.com>
Reviewed-by: Eelco Chaudron <echaudro@redhat.com>
Link: https://lore.kernel.org/r/20201204114314.1596-1-wanghai38@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
when openvswitch is configured to mangle the LSE, the current value is
read from the packet dereferencing 4 bytes at mpls_hdr(): ensure that
the label is contained in the skb "linear" area.
Found by code inspection.
Fixes: d27cf5c59a ("net: core: add MPLS update core helper and use in OvS")
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Link: https://lore.kernel.org/r/aa099f245d93218b84b5c056b67b6058ccf81a66.1606987185.git.dcaratti@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Trivial conflict in CAN, keep the net-next + the byteswap wrapper.
Conflicts:
drivers/net/can/usb/gs_usb.c
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently, the openvswitch module is not accepting the correctly formated
netlink message for the TTL decrement action. For both setting and getting
the dec_ttl action, the actions should be nested in the
OVS_DEC_TTL_ATTR_ACTION attribute as mentioned in the openvswitch.h uapi.
When the original patch was sent, it was tested with a private OVS userspace
implementation. This implementation was unfortunately not upstreamed and
reviewed, hence an erroneous version of this patch was sent out.
Leaving the patch as-is would cause problems as the kernel module could
interpret additional attributes as actions and vice-versa, due to the
actions not being encapsulated/nested within the actual attribute, but
being concatinated after it.
Fixes: 744676e777 ("openvswitch: add TTL decrement action")
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Link: https://lore.kernel.org/r/160622121495.27296.888010441924340582.stgit@wsfd-netdev64.ntdv.lab.eng.bos.redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There is no easy way to distinguish if a conntracked tcp packet is
marked invalid because of tcp_in_window() check error or because
it doesn't belong to an existing connection. With this patch,
openvswitch sets liberal tcp flag for the established sessions so
that out of window packets are not marked invalid.
A helper function - nf_ct_set_tcp_be_liberal(nf_conn) is added which
sets this flag for both the directions of the nf_conn.
Suggested-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Numan Siddique <nusiddiq@redhat.com>
Acked-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20201116130126.3065077-1-nusiddiq@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit d3fd65484c ("net: core: add dev_sw_netstats_tx_add") has added
function "dev_sw_netstats_tx_add()" to update net device per-cpu TX
stats.
Use this function instead of own code.
While on it, remove internal_get_stats() and replace it
with dev_get_tstats64().
Signed-off-by: Lev Stipakov <lev@openvpn.net>
Reviewed-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://lore.kernel.org/r/20201113215336.145998-1-lev@openvpn.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Fix smatch warning:
net/openvswitch/meter.c:427 ovs_meter_cmd_set() warn: passing zero to 'PTR_ERR'
dp_meter_create() never returns NULL, use IS_ERR
instead of IS_ERR_OR_NULL to fix this.
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Link: https://lore.kernel.org/r/20201031060153.39912-1-yuehaibing@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net/openvswitch/flow.c:303: warning: Function parameter or member 'key_vh' not described in 'parse_vlan_tag'
net/openvswitch/flow.c:303: warning: Function parameter or member 'skb' not described in 'parse_vlan_tag'
net/openvswitch/flow.c:303: warning: Function parameter or member 'untag_vlan' not described in 'parse_vlan_tag'
net/openvswitch/vport.c:122: warning: Function parameter or member 'parms' not described in 'ovs_vport_alloc'
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20201028004849.930094-1-andrew@lunn.ch
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The flow_lookup() function uses per CPU variables, which must be called
with BH disabled. However, this is fine in the general NAPI use case
where the local BH is disabled. But, it's also called from the netlink
context. The below patch makes sure that even in the netlink path, the
BH is disabled.
In addition, u64_stats_update_begin() requires a lock to ensure one writer
which is not ensured here. Making it per-CPU and disabling NAPI (softirq)
ensures that there is always only one writer.
Fixes: eac87c413b ("net: openvswitch: reorder masks array based on usage")
Reported-by: Juri Lelli <jlelli@redhat.com>
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Link: https://lore.kernel.org/r/160295903253.7789.826736662555102345.stgit@ebuild
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Small conflict around locking in rxrpc_process_event() -
channel_lock moved to bundle in next, while state lock
needs _bh() from net.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
With multiple DNAT rules it's possible that after destination
translation the resulting tuples collide.
For example, two openvswitch flows:
nw_dst=10.0.0.10,tp_dst=10, actions=ct(commit,table=2,nat(dst=20.0.0.1:20))
nw_dst=10.0.0.20,tp_dst=10, actions=ct(commit,table=2,nat(dst=20.0.0.1:20))
Assuming two TCP clients initiating the following connections:
10.0.0.10:5000->10.0.0.10:10
10.0.0.10:5000->10.0.0.20:10
Both tuples would translate to 10.0.0.10:5000->20.0.0.1:20 causing
nf_conntrack_confirm() to fail because of tuple collision.
Netfilter handles this case by allocating a null binding for SNAT at
egress by default. Perform the same operation in openvswitch for DNAT
if no explicit SNAT is requested by the user and allocate a null binding
for SNAT for packets in the "original" direction.
Reported-at: https://bugzilla.redhat.com/1877128
Suggested-by: Florian Westphal <fw@strlen.de>
Fixes: 05752523e5 ("openvswitch: Interface with NAT.")
Signed-off-by: Dumitru Ceara <dceara@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The only usage of these is to assign their address to the small_ops field
in the genl_family struct, which is a const pointer, and applying
ARRAY_SIZE() on them. Make them const to allow the compiler to put them
in read-only memory.
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement TCA_VLAN_ACT_POP_ETH and TCA_VLAN_ACT_PUSH_ETH, to
respectively pop and push a base Ethernet header at the beginning of a
frame.
POP_ETH is just a matter of pulling ETH_HLEN bytes. VLAN tags, if any,
must be stripped before calling POP_ETH.
PUSH_ETH is restricted to skbs with no mac_header, and only the MAC
addresses can be configured. The Ethertype is automatically set from
skb->protocol. These restrictions ensure that all skb's fields remain
consistent, so that this action can't confuse other part of the
networking stack (like GSO).
Since openvswitch already had these actions, consolidate the code in
skbuff.c (like for vlan and mpls push/pop).
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Bulk of the genetlink users can use smaller ops, move them.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the function ovs_ct_limit_exit, there is already a helper vaibale
which could be reused to improve the readability, so i fix it in this
patch.
Signed-off-by: Zeng Tao <prime.zeng@hisilicon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We got slightly different patches removing a double word
in a comment in net/ipv4/raw.c - picked the version from net.
Simple conflict in drivers/net/ethernet/ibm/ibmvnic.c. Use cached
values instead of VNIC login response buffer (following what
commit 507ebe6444 ("ibmvnic: Fix use-after-free of VNIC login
response buffer") did).
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
If nf_conncount_init fails currently the dispatched work is not canceled,
causing problems when the timer fires. This change fixes this by not
scheduling the work until all initialization is successful.
Fixes: a65878d6f0 ("net: openvswitch: fixes potential deadlock in dp cleanup code")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Reviewed-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
keep_flows was introduced by [1], which used as flag to delete flows or not.
When rehashing or expanding the table instance, we will not flush the flows.
Now don't use it anymore, remove it.
[1] - acd051f176
Cc: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Decrease table->count and ufid_count unconditionally,
because we only don't use count or ufid_count to count
when flushing the flows. To simplify the codes, we
remove the "count" argument of table_instance_flow_free.
To avoid a bug when deleting flows in the future, add
WARN_ON in flush flows function.
Cc: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Not change the logic, just improve the coding style.
Cc: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using ipv6_stub->ipv6_fragment to avoid the netfilter dependency
Signed-off-by: wenxu <wenxu@ucloud.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
To avoid some issues, for example RCU usage warning and double free,
we should flush the flows under ovs_lock. This patch refactors
table_instance_destroy and introduces table_instance_flow_flush
which can be invoked by __dp_destroy or ovs_flow_tbl_flush.
Fixes: 50b0e61b32 ("net: openvswitch: fix possible memleak on destroy flow-table")
Reported-by: Johan Knöös <jknoos@google.com>
Reported-at: https://mail.openvswitch.org/pipermail/ovs-discuss/2020-August/050489.html
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ovs_flow_tbl_destroy always is called from RCU callback
or error path. It is no need to check if rcu_read_lock
or lockdep_ovsl_is_held was held.
ovs_dp_cmd_fill_info always is called with ovs_mutex,
So use the rcu_dereference_ovsl instead of rcu_dereference
in ovs_flow_tbl_masks_cache_size.
Fixes: 9bf24f594c ("net: openvswitch: make masks cache size configurable")
Cc: Eelco Chaudron <echaudro@redhat.com>
Reported-by: syzbot+c0eb9e7cdde04e4eb4be@syzkaller.appspotmail.com
Reported-by: syzbot+f612c02823acb02ff9bc@syzkaller.appspotmail.com
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch makes the masks cache size configurable, or with
a size of 0, disable it.
Reviewed-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a counter that counts the number of masks cache hits, and
export it through the megaflow netlink statistics.
Reviewed-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ovs_ct_put_key() is potentially copying uninitialized kernel stack memory
into socket buffers, since the compiler may leave a 3-byte hole at the end
of `struct ovs_key_ct_tuple_ipv4` and `struct ovs_key_ct_tuple_ipv6`. Fix
it by initializing `orig` with memset().
Fixes: 9dd7f8907c ("openvswitch: Add original direction conntrack tuple to sw_flow_key.")
Suggested-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Peilin Ye <yepeilin.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When openvswitch conntrack offload with act_ct action. Fragment packets
defrag in the ingress tc act_ct action and miss the next chain. Then the
packet pass to the openvswitch datapath without the mru. The over
mtu packet will be dropped in output action in openvswitch for over mtu.
"kernel: net2: dropped over-mtu packet: 1528 > 1500"
This patch add mru in the tc_skb_ext for adefrag and miss next chain
situation. And also add mru in the qdisc_skb_cb. The act_ct set the mru
to the qdisc_skb_cb when the packet defrag. And When the chain miss,
The mru is set to tc_skb_ext which can be got by ovs datapath.
Fixes: b57dc7c13e ("net/sched: Introduce action ct")
Signed-off-by: wenxu <wenxu@ucloud.cn>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The previous patch introduced a deadlock, this patch fixes it by making
sure the work is canceled without holding the global ovs lock. This is
done by moving the reorder processing one layer up to the netns level.
Fixes: eac87c413b ("net: openvswitch: reorder masks array based on usage")
Reported-by: syzbot+2c4ff3614695f75ce26c@syzkaller.appspotmail.com
Reported-by: syzbot+bad6507e5db05017b008@syzkaller.appspotmail.com
Reviewed-by: Paolo <pabeni@redhat.com>
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch reorders the masks array every 4 seconds based on their
usage count. This greatly reduces the masks per packet hit, and
hence the overall performance. Especially in the OVS/OVN case for
OpenShift.
Here are some results from the OVS/OVN OpenShift test, which use
8 pods, each pod having 512 uperf connections, each connection
sends a 64-byte request and gets a 1024-byte response (TCP).
All uperf clients are on 1 worker node while all uperf servers are
on the other worker node.
Kernel without this patch : 7.71 Gbps
Kernel with this patch applied: 14.52 Gbps
We also run some tests to verify the rebalance activity does not
lower the flow insertion rate, which does not.
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Tested-by: Andrew Theurer <atheurer@redhat.com>
Reviewed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
