Jarno Rajahalme reported up to 40% performance gain on netperf TCP_CRR with
an earlier version of this patch in combination with a kernel NUMA patch,
together with a reduction in variance:
http://openvswitch.org/pipermail/dev/2014-January/035867.html
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Ethan Jackson <ethan@nicira.com>
This is a first step in making thread safety more granular in dpif-netdev,
to allow for multithreaded forwarding.
Signed-off-by: Ben Pfaff <blp@nicira.com>
This is analogous to the split between rule and rule_actions in
ofproto. As there, it will allow retaining a reference to a rule's
actions, while processing them, without having to retain a reference
to the rule itself.
Signed-off-by: Ben Pfaff <blp@nicira.com>
Normally all the ports have the same mtu anyhow, so there is little
advantage in keeping track of the maximum mtu on a per-bridge basis. In
upcoming commits, tracking mtu will require more locking and present
even less advantage (because the packet buffer will become per-thread, so
that reallocating once per thread becomes essentially a null cost).
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Ethan Jackson <ethan@nicira.com>
The goal is to make it easy to divide the ports into groups for handling
by threads. It seems easy enough to do that by hash value, and a little
harder otherwise.
This commit has the side effect of raising the maximum number of ports from
256 to UINT32_MAX-1. That is why some tests need to be updated:
previously, internally generated port names like "ovs_vxlan_4341" were
ignored because 4341 is bigger than the previous limit of 256.
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Ethan Jackson <ethan@nicira.com>
ovsthread_counter is an abstract interface that could be implemented
different ways. The initial implementation is simple but less than
optimally efficient.
Signed-off-by: Ben Pfaff <blp@nicira.com>
This helps reduce confusion about when a flow is a flow and when it is
just metadata.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
Use one callback instead of many, helps in adding new functionality
later on.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
The skb_priority, pkt_mark and tunl parameters of dp_netdev_port_input()
are always passed as 0, 0 and NULL respectively. So rather than
passing these values to dp_netdev_port_input() just use them directly.
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Ben Pfaff <blp@nicira.com>
The command ovs-dpctl can wrongly output the masks even if the
datapath does not implement mega flows. In this case the output
will be similar to the following:
system@ovs-system:
lookups: hit:14 missed:41 lost:0
flows: 0
masks: hit:18446744073709551615 total:4294967295
hit/pkt:335395346794719104.00
port 0: ovs-system (internal)
port 1: gre_system (gre: df_default=false, ttl=0)
port 2: ots-br0 (internal)
port 3: int0 (internal)
port 4: vnet0
port 5: vnet1
The problem depends on the fact that n_masks stats is stored as a
uint32 in the struct ovs_dp_megaflow_stats and as a uint64 in the
struct dpif_dp_stats. UINT32_MAX instead of UINT64_MAX should be
used to detect if the datapath supports megaflows or not.
Signed-off-by: Francesco Fusco <ffusco@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
Allowing the packet to be modified by execution allows less data
copying for userspace action execution. Some users of the
dpif_execute already expect that the packet may be modified. This
patch makes this behavior uniform and makes the userspace datapath and
the execution helpers modify the packet as it is being executed.
Userspace action now steals the packet if given permission, as the
packet is normally not needed after it. The only exception is the
sample action, and this is accounted for my keeping track of any
actions that could be following the userspace action.
The packet in dpif_upcall is changed from a pointer to a struct,
allowing the packet to be honest about it's headroom. After this
change the packet can safely be pushed on over the precarious 4 byte
limit earlier allowed by the netlink data preceding the packet.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
Normally OVS userspace supplies a mask along with a flow key for each
new data path flow that should be created. OVS also provides an
option to disable the kernel wildcarding, in which case the flows are
created without a mask. When kernel wildcarding is disabled, the
datapath should use exact match, i.e. not wildcard any bits in the
flow key. Currently, what happens with the userspace datapath instead
is that a datapath flow with mostly empty mask is created (i.e., most
fields are wildcarded), as the current code does not examine the given
mask key length to find out that the mask key is actually empty. This
results in the same datapath flow matching on packets of multiple
different flows, wrong actions being processed, and stats being
incorrect.
This patch refactors userspace datapath code to explicitly initialize
a suitable exact match mask when a flow put without a mask is
executed.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
Subtable lookup is performed in ranges defined for struct flow,
starting from metadata (registers, in_port, etc.), then L2 header, L3,
and finally L4 ports. Whenever it is found that there are no matches
in the current subtable, the rest of the subtable can be skipped. The
rationale of this logic is that as many fields as possible can remain
wildcarded.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Userspace action needs the original flow key. This also
matches the kernel datapath behavior.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
Extract the flow key from the packet instead of the execute->key.
This reflects how the kernel datapath behaves.
Also use ofpbuf_clone_with_headroom() instead of open coding the same.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
Instead of an exact match flow table, we introduce a classifier.
This enables mega-flows in userspace datapath.
Signed-off-by: Gurucharan Shetty <gshetty@nicira.com>
[blp@nicira.com tweaked flow lookup]
Signed-off-by: Ben Pfaff <blp@nicira.com>
'struct dp_netdev_flow' is currently being instantiated as 'flow'.
An upcoming commit introduces a classifier to dpif-netdev
which uses 'struct flow' at a few places and that can cause
confusion while reading code.
Signed-off-by: Gurucharan Shetty <gshetty@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
Widen TCP flags handling from 7 bits (uint8_t) to 12 bits (uint16_t).
The kernel interface remains at 8 bits, which makes no functional
difference now, as none of the higher bits is currently of interest
to the userspace.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Jesse Gross <jesse@nicira.com>
The declaration of 'get_max_ports()' to return odp_port_t adds
unwanted complexity to coding. This commit changes it back to
return uint32_t type.
Signed-off-by: Alex Wang <alexw@nicira.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
The skb_mark field is currently only available with the Linux datapath
and is only used internally. However, it is desirable to expose this
through OpenFlow and when it is exposed ideally it would not be system-
specific. In preparation for this, skb_mark is rename to pkt_mark in
internal data structures for consistency.
This does not rename the Linux interfaces because doing so would break
the API. It would not necessarily be desirable to do anyways since in
Linux-specific code it is clearer to use the actual name rather than a
generic one. This can lead to confusion in some places, however, because
we do not always strictly separate generic and platform dependent code
(one example is actions). This seems inevitable though at this point if
the lower and upper layers have different names (as they must given the
above requirements).
Signed-off-by: Jesse Gross <jesse@nicira.com>
Acked-by: Andy Zhou <azhou@nicira.com>
This commit adds annotations for thread safety check. And the
check can be conducted by using -Wthread-safety flag in clang.
Co-authored-by: Alex Wang <alexw@nicira.com>
Signed-off-by: Alex Wang <alexw@nicira.com>
Signed-off-by: Ethan Jackson <ethan@nicira.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
This can happen on ESX.
Also adds a test to make sure this works.
Bug #17634.
Signed-off-by: Ben Pfaff <blp@nicira.com>
Tested-by: Guolin Yang <gyang@vmware.com>
This can be improved later but it is the simple thing to do for now.
I marked a couple of races with XXX. I don't have a really good solution
for these, but I hope to find one. They may be harmless in practice.
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Ethan Jackson <ethan@nicira.com>
Commit 4e022ec09e14 (Create specific types for ofp and odp port) broke
OpenFlow OFPP_PACKET_OUT requests that use in_port=OFPP_CONTROLLER. This
commit fixes the problem and adds a regression test.
CC: Alex Wang <alexw@nicira.com>
Reported-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Signed-off-by: Ben Pfaff <blp@nicira.com>
Until now, datapath ports and openflow ports were both represented by
unsigned integers of various sizes. With implicit conversions, etc., it is
easy to mix them up and use one where the other is expected. This commit
creates two typedefs, ofp_port_t and odp_port_t. Both of these two types
are marked by "__attribute__((bitwise))" so that sparse can be used to
detect any misuse.
Signed-off-by: Alex Wang <alexw@nicira.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
Added support to allow mega flow specified and displayed. ovs-dpctl tool
is mainly used as debugging tool.
This patch also implements the low level user space routines to send
and receive mega flow netlink messages. Those netlink suppor
routines are required for forthcoming user space mega flow patches.
Added a unit test to test parsing and display of mega flows.
Ethan contributed the ovs-dpctl mega flow output function.
Co-authored-by: Ethan Jackson <ethan@nicira.com>
Signed-off-by: Ethan Jackson <ethan@nicira.com>
Signed-off-by: Andy Zhou <azhou@nicira.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
Returning a static data buffer makes code more brittle and definitely
not thread-safe, so this commit switches to using a caller-provided
buffer instead.
Signed-off-by: Ben Pfaff <blp@nicira.com>
The ports that exist within a dpif have already been translated through
netdev_vport_get_dpif_port(), so there is no value to translating them
again in the interfaces that query or dump ports (and possibly a drawback
if somehow the translation could change).
After this change, dpif-netdev translates port names in just one place,
the port_add path, which makes dpif-netdev act the same way as dpif-linux
in this respect.
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Ethan Jackson <ethan@nicira.com>
The motivation for this is to allow such actions to be honoured
if they are encountered; by the user-space datapath before recirculation;
or by internal processing of actions by ovs-vswitchd before recirculation.
Recirculation will be added by a subsequent patch.
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Ben Pfaff <blp@nicira.com>
This moves generic action execution code out of lib/dpif-netedev.c
and into a new file, lib/odp-execute.c.
This is in preparation for using odp_execute_actions()
in lib/odp-util.c to handle recirculation/
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Ben Pfaff <blp@nicira.com>
This is in preparation for making dp_netdev_action_userspace()
more generic and passing dp_netdev_output_userspace() as a callback.
In this case it makes sense to decode userdata in generic code.
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Ben Pfaff <blp@nicira.com>
Separating packet capture from "struct netdev" means that there is no
remaining per-"struct netdev" state, which will allow us to get rid of
"struct netdev_dev" (by renaming it "struct netdev").
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Ethan Jackson <ethan@nicira.com>
This gets rid of the only per-instance data in "struct netdev", which
will make it possible to merge "struct netdev_dev" into "struct netdev" in
a later commit.
Ed Maste wrote the netdev-bsd changes in this commit.
Signed-off-by: Ben Pfaff <blp@nicira.com>
Co-authored-by: Ed Maste <emaste@freebsd.org>
Signed-off-by: Ed Maste <emaste@freebsd.org>
Tested-by: Ed Maste <emaste@freebsd.org>
Since userspace flow based tunneling code is checked in, the kernel
port based tunneling code can be removed.
Patch removes following components:
- tunnel ports hash table and moved tunnel ports list to individual
vports.
- Cleaned per tnl-port config.
- OVS_KEY_ATTR_TUN_ID action is removed.
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Acked-by: Jesse Gross <jesse@nicira.com>
Bug #15078
Until now, the optional OVS_USERSPACE_ATTR_USERDATA attribute had to be
exactly 64 bits long, if it was present. However, 64 bits is not enough
space to associate as much information with a flow as would be convenient
for some userspace features now under development. This commit generalizes
the attribute, allowing it to be any length.
This generalization is backward-compatible: if userspace only uses 64-bit
attributes, then it will not see any change in behavior.
CC: Romain Lenglet <rlenglet@vmware.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Jesse Gross <jesse@nicira.com>
This patch implements use-space datapath and non-datapath code
to match and use the datapath API set out in Leo Alterman's patch
"user-space datapath: Add basic MPLS support to kernel".
The resulting MPLS implementation supports:
* Pushing a single MPLS label
* Poping a single MPLS label
* Modifying an MPLS lable using set-field or load actions
that act on the label value, tc and bos bit.
* There is no support for manipulating the TTL
this is considered future work.
The single-level push pop limitation is implemented by processing
push, pop and set-field/load actions in order and discarding information
that would require multiple levels of push/pop to be supported.
e.g.
push,push -> the first push is discarded
pop,pop -> the first pop is discarded
This patch is based heavily on work by Ravi K.
Cc: Ravi K <rkerur@gmail.com>
Reviewed-by: Isaku Yamahata <yamahata@valinux.co.jp>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Ben Pfaff <blp@nicira.com>
In future patches, a netdev's datapath port name may not
necessarily be the same as its device name. This patch prepares for
this by making the distinction in the netdev and dpif layers.
Signed-off-by: Ethan Jackson <ethan@nicira.com>
