linux-toradex.git/net/openvswitch/datapath.c, branch v5.14

openvswitch: add trace points

2021-06-22T17:47:32+00:00

This makes openvswitch module use the event tracing framework
to log the upcall interface and action execution pipeline.  When
using openvswitch as the packet forwarding engine, some types of
debugging are made possible simply by using the ovs-vswitchd's
ofproto/trace command.  However, such a command has some
limitations:

  1. When trying to trace packets that go through the CT action,
     the state of the packet can't be determined, and probably
     would be potentially wrong.

  2. Deducing problem packets can sometimes be difficult as well
     even if many of the flows are known

  3. It's possible to use the openvswitch module even without
     the ovs-vswitchd (although, not common use).

Introduce the event tracing points here to make it possible for
working through these problems in kernel space.  The style is
copied from the mac80211 driver-trace / trace code for
consistency - this creates some checkpatch splats, but the
official 'guide' for adding tracepoints, as well as the existing
examples all add the same splats so it seems acceptable.

Signed-off-by: Aaron Conole 
Signed-off-by: David S. Miller

net: openvswitch: silence suspicious RCU usage warning

2020-11-04T00:57:42+00:00

Silence suspicious RCU usage warning in ovs_flow_tbl_masks_cache_resize()
by replacing rcu_dereference() with rcu_dereference_ovsl().

In addition, when creating a new datapath, make sure it's configured under
the ovs_lock.

Fixes: 9bf24f594c6a ("net: openvswitch: make masks cache size configurable")
Reported-by: syzbot+9a8f8bfcc56e8578016c@syzkaller.appspotmail.com
Signed-off-by: Eelco Chaudron 
Link: https://lore.kernel.org/r/160439190002.56943.1418882726496275961.stgit@ebuild
Signed-off-by: Jakub Kicinski

genetlink: move to smaller ops wherever possible

2020-10-03T02:11:11+00:00

Bulk of the genetlink users can use smaller ops, move them.

Signed-off-by: Jakub Kicinski 
Reviewed-by: Johannes Berg 
Signed-off-by: David S. Miller

net: openvswitch: fixes crash if nf_conncount_init() fails

2020-09-01T20:23:23+00:00

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: a65878d6f00b ("net: openvswitch: fixes potential deadlock in dp cleanup code")
Reported-by: kernel test robot 
Signed-off-by: Eelco Chaudron 
Reviewed-by: Tonghao Zhang 
Signed-off-by: David S. Miller

net: openvswitch: improve the coding style

2020-09-01T18:42:15+00:00

Not change the logic, just improve the coding style.

Cc: Pravin B Shelar 
Signed-off-by: Tonghao Zhang 
Acked-by: Pravin B Shelar 
Signed-off-by: David S. Miller

net: openvswitch: introduce common code for flushing flows

2020-08-13T22:53:30+00:00

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: 50b0e61b32ee ("net: openvswitch: fix possible memleak on destroy flow-table")
Reported-by: Johan Knöös 
Reported-at: https://mail.openvswitch.org/pipermail/ovs-discuss/2020-August/050489.html
Signed-off-by: Tonghao Zhang 
Reviewed-by: Cong Wang 
Signed-off-by: David S. Miller

net: openvswitch: make masks cache size configurable

2020-08-03T22:17:48+00:00

This patch makes the masks cache size configurable, or with
a size of 0, disable it.

Reviewed-by: Paolo Abeni 
Reviewed-by: Tonghao Zhang 
Signed-off-by: Eelco Chaudron 
Signed-off-by: David S. Miller

net: openvswitch: add masks cache hit counter

2020-08-03T22:17:48+00:00

Add a counter that counts the number of masks cache hits, and
export it through the megaflow netlink statistics.

Reviewed-by: Paolo Abeni 
Reviewed-by: Tonghao Zhang 
Signed-off-by: Eelco Chaudron 
Signed-off-by: David S. Miller

net: openvswitch: fixes potential deadlock in dp cleanup code

2020-07-24T23:58:38+00:00

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: eac87c413bf9 ("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 
Signed-off-by: Eelco Chaudron 
Signed-off-by: David S. Miller

net: openvswitch: reorder masks array based on usage

2020-07-17T17:36:50+00:00

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 
Tested-by: Andrew Theurer 
Reviewed-by: Paolo Abeni 
Signed-off-by: David S. Miller