The original intent of the API appears to be that the underlying DPIF
implementaion would choose a local meter id. However, neither of the
existing datapath meter implementations (userspace or Linux) implemented
that; they expected a valid meter id to be passed in, otherwise they
returned an error. This commit follows the existing implementations and
makes the API somewhat cleaner.
Signed-off-by: Justin Pettit <jpettit@ovn.org>
Acked-by: Ben Pfaff <blp@ovn.org>
'key.len' should be calculated before inserting to EMC, otherwise
resulting entry will match with any packet with the same hash.
CC: Yipeng Wang <yipeng1.wang@intel.com>
Fixes: 60d8ccae135f ("dpif-netdev: Add SMC cache after EMC cache")
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Yipeng Wang <yipeng1.wang@intel.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
Another dpif provider will soon add support for meters, so move
some of the common sanity checks up into the dpif layer so that each
provider doesn't need to re-implement them.
Signed-off-by: Justin Pettit <jpettit@ovn.org>
Acked-by: Ben Pfaff <blp@ovn.org>
Commit ab15e70eb587 ("dpctl: Expand the flow dump type filter") will be
reverted, which this patch fixed, so it needs to be reverted as well.
This reverts commit b10ac772218afd4f296db866f6b80258e1d1ca8a.
CC: Gavi Teitz <gavi@mellanox.com>
CC: Simon Horman <simon.horman@netronome.com>
CC: Roi Dayan <roid@mellanox.com>
CC: Aaron Conole <aconole@redhat.com>
Signed-off-by: Justin Pettit <jpettit@ovn.org>
Acked-by: Ben Pfaff <blp@ovn.org>
This patch adds a signature match cache (SMC) after exact match
cache (EMC). The difference between SMC and EMC is SMC only stores
a signature of a flow thus it is much more memory efficient. With
same memory space, EMC can store 8k flows while SMC can store 1M
flows. It is generally beneficial to turn on SMC but turn off EMC
when traffic flow count is much larger than EMC size.
SMC cache will map a signature to an dp_netdev_flow index in
flow_table. Thus, we add two new APIs in cmap for lookup key by
index and lookup index by key.
For now, SMC is an experimental feature that it is turned off by
default. One can turn it on using ovsdb options.
Signed-off-by: Yipeng Wang <yipeng1.wang@intel.com>
Co-authored-by: Jan Scheurich <jan.scheurich@ericsson.com>
Signed-off-by: Jan Scheurich <jan.scheurich@ericsson.com>
Acked-by: Billy O'Mahony <billy.o.mahony@intel.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
Currently, the major trigger for hw flow offload is at upcall handling,
which is actually in the datapath. Moreover, the hw offload installation
and modification is not that lightweight. Meaning, if there are so many
flows being added or modified frequently, it could stall the datapath,
which could result to packet loss.
To diminish that, all those flow operations will be recorded and appended
to a list. A thread is then introduced to process this list (to do the
real flow offloading put/del operations). This could leave the datapath
as lightweight as possible.
Signed-off-by: Yuanhan Liu <yliu@fridaylinux.org>
Co-authored-by: Shahaf Shuler <shahafs@mellanox.com>
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
So that we could skip some very costly CPU operations, including but
not limiting to miniflow_extract, emc lookup, dpcls lookup, etc. Thus,
performance could be greatly improved.
A PHY-PHY forwarding with 1000 mega flows (udp,tp_src=1000-1999) and
1 million streams (tp_src=1000-1999, tp_dst=2000-2999) show more that
260% performance boost.
Note that though the heavy miniflow_extract is skipped, we still have
to do per packet checking, due to we have to check the tcp_flags.
Co-authored-by: Finn Christensen <fc@napatech.com>
Signed-off-by: Yuanhan Liu <yliu@fridaylinux.org>
Signed-off-by: Finn Christensen <fc@napatech.com>
Co-authored-by: Shahaf Shuler <shahafs@mellanox.com>
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
Most modern NICs have the ability to bind a flow with a mark, so that
every packet matches such flow will have that mark present in its
descriptor.
The basic idea of doing that is, when we receives packets later, we could
directly get the flow from the mark. That could avoid some very costly
CPU operations, including (but not limiting to) miniflow_extract, emc
lookup, dpcls lookup, etc. Thus, performance could be greatly improved.
Thus, the major work of this patch is to associate a flow with a mark
id (an uint32_t number). The association in netdev datapath is done
by CMAP, while in hardware it's done by the rte_flow MARK action.
One tricky thing in OVS-DPDK is, the flow tables is per-PMD. For the
case there is only one phys port but with 2 queues, there could be 2
PMDs. In other words, even for a single mega flow (i.e. udp,tp_src=1000),
there could be 2 different dp_netdev flows, one for each PMD. That could
results to the same mega flow being offloaded twice in the hardware,
worse, we may get 2 different marks and only the last one will work.
To avoid that, a megaflow_to_mark CMAP is created. An entry will be
added for the first PMD that wants to offload a flow. For later PMDs,
it will see such megaflow is already offloaded, then the flow will not
be offloaded to HW twice.
Meanwhile, the mark to flow mapping becomes to 1:N mapping. That is
what the mark_to_flow CMAP is for. When the first PMD wants to offload
a flow, it allocates a new mark and performs the flow offload by reusing
the ->flow_put method. When it succeeds, a "mark to flow" entry will be
added. For later PMDs, it will get the corresponding mark by above
megaflow_to_mark CMAP. Then, another "mark to flow" entry will be added.
Signed-off-by: Yuanhan Liu <yliu@fridaylinux.org>
Co-authored-by: Finn Christensen <fc@napatech.com>
Signed-off-by: Finn Christensen <fc@napatech.com>
Co-authored-by: Shahaf Shuler <shahafs@mellanox.com>
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
It's always been OVS coding style to use spaces rather than tabs for
indentation, but some tabs have snuck in over time. This commit converts
them to spaces.
Signed-off-by: Ben Pfaff <blp@ovn.org>
Acked-by: Justin Pettit <jpettit@ovn.org>
Several OVS structs contain embedded named unions, like this:
struct {
...
union {
...
} u;
};
C11 standardized a feature that many compilers already implemented
anyway, where an embedded union may be unnamed, like this:
struct {
...
union {
...
};
};
This is more convenient because it allows the programmer to omit "u."
in many places. OVS already used this feature in several places. This
commit embraces it in several others.
Signed-off-by: Ben Pfaff <blp@ovn.org>
Acked-by: Justin Pettit <jpettit@ovn.org>
Tested-by: Alin Gabriel Serdean <aserdean@ovn.org>
Acked-by: Alin Gabriel Serdean <aserdean@ovn.org>
Unconditional return may cause packet leak in case of
'should_steal == true'.
Additionally, removed redundant checking for depth level.
CC: Sugesh Chandran <sugesh.chandran@intel.com>
Fixes: 7c12dfc527a5 ("tunneling: Avoid datapath-recirc by
combining recirc actions at xlate.")
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Ian Stokess <ian.stokes@intel.com>
When initialization of the DPDK PMD driver fails
(dpdk_eth_dev_init()), the reconfigure_datapath() function will remove
the port from dp_netdev, and the port is not used.
Now when bridge_reconfigure() is called again, no changes to the
previous failing netdev configuration are detected and therefore the
ports gets added to dp_netdev and used uninitialized. This is causing
exceptions...
The fix has two parts to it. First in netdev-dpdk.c we remember if the
DPDK port was started or not, and when calling
netdev_dpdk_reconfigure() we also try re-initialization if the port
was not already active. The second part of the change is in
dpif-netdev.c where it makes sure netdev_reconfigure() is called if
the port needs reconfiguration, as netdev_is_reconf_required() is only
true until netdev_reconfigure() is called (even if it fails).
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Tested-by: Ciara Loftus <ciara.loftus@intel.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
This patch enhances dpif-netdev-perf to detect iterations with
suspicious statistics according to the following criteria:
- iteration lasts longer than US_THR microseconds (default 250).
This can be used to capture events where a PMD is blocked or
interrupted for such a period of time that there is a risk for
dropped packets on any of its Rx queues.
- max vhost qlen exceeds a threshold Q_THR (default 128). This can
be used to infer virtio queue overruns and dropped packets inside
a VM, which are not visible in OVS otherwise.
Such suspicious iterations can be logged together with their iteration
statistics to be able to correlate them to packet drop or other events
outside OVS.
A new command is introduced to enable/disable logging at run-time and
to adjust the above thresholds for suspicious iterations:
ovs-appctl dpif-netdev/pmd-perf-log-set on | off
[-b before] [-a after] [-e|-ne] [-us usec] [-q qlen]
Turn logging on or off at run-time (on|off).
-b before: The number of iterations before the suspicious iteration to
be logged (default 5).
-a after: The number of iterations after the suspicious iteration to
be logged (default 5).
-e: Extend logging interval if another suspicious iteration is
detected before logging occurs.
-ne: Do not extend logging interval (default).
-q qlen: Suspicious vhost queue fill level threshold. Increase this
to 512 if the Qemu supports 1024 virtio queue length.
(default 128).
-us usec: change the duration threshold for a suspicious iteration
(default 250 us).
Note: Logging of suspicious iterations itself consumes a considerable
amount of processing cycles of a PMD which may be visible in the iteration
history. In the worst case this can lead OVS to detect another
suspicious iteration caused by logging.
If more than 100 iterations around a suspicious iteration have been
logged once, OVS falls back to the safe default values (-b 5/-a 5/-ne)
to avoid that logging itself causes continuos further logging.
Signed-off-by: Jan Scheurich <jan.scheurich@ericsson.com>
Acked-by: Billy O'Mahony <billy.o.mahony@intel.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
This patch instruments the dpif-netdev datapath to record detailed
statistics of what is happening in every iteration of a PMD thread.
The collection of detailed statistics can be controlled by a new
Open_vSwitch configuration parameter "other_config:pmd-perf-metrics".
By default it is disabled. The run-time overhead, when enabled, is
in the order of 1%.
The covered metrics per iteration are:
- cycles
- packets
- (rx) batches
- packets/batch
- max. vhostuser qlen
- upcalls
- cycles spent in upcalls
This raw recorded data is used threefold:
1. In histograms for each of the following metrics:
- cycles/iteration (log.)
- packets/iteration (log.)
- cycles/packet
- packets/batch
- max. vhostuser qlen (log.)
- upcalls
- cycles/upcall (log)
The histograms bins are divided linear or logarithmic.
2. A cyclic history of the above statistics for 999 iterations
3. A cyclic history of the cummulative/average values per millisecond
wall clock for the last 1000 milliseconds:
- number of iterations
- avg. cycles/iteration
- packets (Kpps)
- avg. packets/batch
- avg. max vhost qlen
- upcalls
- avg. cycles/upcall
The gathered performance metrics can be printed at any time with the
new CLI command
ovs-appctl dpif-netdev/pmd-perf-show [-nh] [-it iter_len] [-ms ms_len]
[-pmd core] [dp]
The options are
-nh: Suppress the histograms
-it iter_len: Display the last iter_len iteration stats
-ms ms_len: Display the last ms_len millisecond stats
-pmd core: Display only the specified PMD
The performance statistics are reset with the existing
dpif-netdev/pmd-stats-clear command.
The output always contains the following global PMD statistics,
similar to the pmd-stats-show command:
Time: 15:24:55.270
Measurement duration: 1.008 s
pmd thread numa_id 0 core_id 1:
Cycles: 2419034712 (2.40 GHz)
Iterations: 572817 (1.76 us/it)
- idle: 486808 (15.9 % cycles)
- busy: 86009 (84.1 % cycles)
Rx packets: 2399607 (2381 Kpps, 848 cycles/pkt)
Datapath passes: 3599415 (1.50 passes/pkt)
- EMC hits: 336472 ( 9.3 %)
- Megaflow hits: 3262943 (90.7 %, 1.00 subtbl lookups/hit)
- Upcalls: 0 ( 0.0 %, 0.0 us/upcall)
- Lost upcalls: 0 ( 0.0 %)
Tx packets: 2399607 (2381 Kpps)
Tx batches: 171400 (14.00 pkts/batch)
Signed-off-by: Jan Scheurich <jan.scheurich@ericsson.com>
Acked-by: Billy O'Mahony <billy.o.mahony@intel.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
If the caller provides a non-NULL qfill pointer and the netdev
implemementation supports reading the rx queue fill level, the rxq_recv()
function returns the remaining number of packets in the rx queue after
reception of the packet burst to the caller. If the implementation does
not support this, it returns -ENOTSUP instead. Reading the remaining queue
fill level should not substantilly slow down the recv() operation.
A first implementation is provided for ethernet and vhostuser DPDK ports
in netdev-dpdk.c.
This output parameter will be used in the upcoming commit for PMD
performance metrics to supervise the rx queue fill level for DPDK
vhostuser ports.
Signed-off-by: Jan Scheurich <jan.scheurich@ericsson.com>
Acked-by: Billy O'Mahony <billy.o.mahony@intel.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
Until mow, this macro has blindly read the passed-in type's size, but
that's unnecessarily risky. This commit changes it to verify that the
passed-in type is the same size as the field and, on GCC and Clang, that
the types are compatible. It also adds a version that does not check,
for the one case where (currently) we deliberately read the wrong size,
and updates a few uses to use more precise field names.
Signed-off-by: Ben Pfaff <blp@ovn.org>
Reviewed-by: Yifeng Sun <pkusunyifeng@gmail.com>
Reviewed-by: Armando Migliaccio <armamig@gmail.com>
This patch adds support of flushing a conntrack entry specified by the
conntrack 5-tuple in dpif-netdev.
Signed-off-by: Yi-Hung Wei <yihung.wei@gmail.com>
Signed-off-by: Ben Pfaff <blp@ovn.org>
Acked-by: Darrell Ball <dlu998@gmail.com>
ofp-util had been far too large and monolithic for a long time. This
commit breaks it up into units that make some logical sense. It also
moves the pieces of ofp-parse that were specific to each unit into the
relevant unit.
Most of this commit is just moving code around.
Signed-off-by: Ben Pfaff <blp@ovn.org>
Reviewed-by: Yifeng Sun <pkusunyifeng@gmail.com>
This supports using the ct_clear action in the kernel datapath. To
preserve compatibility with current ct_clear behavior on old kernels, we
only pass this action down to the datapath if a probe reveals the
datapath actually supports it.
Signed-off-by: Eric Garver <e@erig.me>
Acked-by: William Tu <u9012063@gmail.com>
Acked-by: Flavio Leitner <fbl@sysclose.org>
Signed-off-by: Justin Pettit <jpettit@ovn.org>
It is based on the length of history that is stored about an
rxq (currently 1 min).
$ ovs-appctl dpif-netdev/pmd-rxq-show
pmd thread numa_id 0 core_id 4:
isolated : false
port: dpdkphy1 queue-id: 0 pmd usage: 70 %
port: dpdkvhost0 queue-id: 0 pmd usage: 0 %
pmd thread numa_id 0 core_id 6:
isolated : false
port: dpdkphy0 queue-id: 0 pmd usage: 64 %
port: dpdkvhost1 queue-id: 0 pmd usage: 0 %
These values are what would be used as part of rxq to pmd
assignment due to a reconfiguration event e.g. adding pmds,
adding rxqs or with the command:
ovs-appctl dpif-netdev/pmd-rxq-rebalance
Signed-off-by: Jan Scheurich <jan.scheurich@ericsson.com>
Co-authored-by: Jan Scheurich <jan.scheurich@ericsson.com>
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
An rxq may have processing cycles counted in the current
counter when a reload happens. That could temporarily create
a small skew on the stats for an rxq. Reset the counter after
reload.
Fixes: 4809891b2e01 ("dpif-netdev: Count the rxq processing cycles for an rxq.")
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
This allows to collect packets from more than one RX burst
and send them together with a configurable intervals.
'other_config:tx-flush-interval' can be used to configure
time that a packet can wait in output batch for sending.
'tx-flush-interval' has microsecond resolution.
Tested-by: Jan Scheurich <jan.scheurich@ericsson.com>
Acked-by: Jan Scheurich <jan.scheurich@ericsson.com>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
Upcoming time-based output batching will allow to collect in a single
output batch packets from different RX queues. Lets keep the list of
RX queues for each output packet and collect cycles for them on send.
Tested-by: Jan Scheurich <jan.scheurich@ericsson.com>
Acked-by: Jan Scheurich <jan.scheurich@ericsson.com>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
Simplify the historically grown TSC cycle counting in PMD threads.
Cycles are currently counted for the following purposes:
1. Measure PMD ustilization
PMD utilization is defined as ratio of cycles spent in busy iterations
(at least one packet received or sent) over the total number of cycles.
This is already done in pmd_perf_start_iteration() and
pmd_perf_end_iteration() based on a TSC timestamp saved in current
iteration at start_iteration() and the actual TSC at end_iteration().
No dependency on intermediate cycle accounting.
2. Measure the processing load per RX queue
This comprises cycles spend on polling and processing packets received
from the rx queue and the cycles spent on delayed sending of these packets
to tx queues (with time-based batching).
The previous scheme using cycles_count_start(), cycles_count_intermediate()
and cycles-count_end() originally introduced to simplify cycle counting
and saving calls to rte_get_tsc_cycles() was rather obscuring things.
Replace by a nestable cycle_timer with with start and stop functions to
embrace a code segment to be timed. The timed code may contain arbitrary
nested cycle_timers. The duration of nested timers is excluded from the
outer timer.
The caller must ensure that each call to cycle_timer_start() is
followed by a call to cycle_timer_end(). Failure to do so will lead to
assertion failure or a memory leak.
The new cycle_timer is used to measure the processing cycles per rx queue.
This is not yet strictly necessary but will be made use of in a subsequent
commit.
All cycle count functions and data are relocated to module
dpif-netdev-perf.
Signed-off-by: Jan Scheurich <jan.scheurich@ericsson.com>
Acked-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Billy O'Mahony <billy.o.mahony@intel.com>
Signed-off: Ian Stokes <ian.stokes@intel.com>
Add module dpif-netdev-perf to host all PMD performance-related
data structures and functions in dpif-netdev. Refactor the PMD
stats handling in dpif-netdev and delegate whatever possible into
the new module, using clean interfaces to shield dpif-netdev from
the implementation details. Accordingly, the all PMD statistics
members are moved from the main struct dp_netdev_pmd_thread into
a dedicated member of type struct pmd_perf_stats.
Include Darrel's prior refactoring of PMD stats contained in
[PATCH v5,2/3] dpif-netdev: Refactor some pmd stats:
1. The cycles per packet counts are now based on packets
received rather than packet passes through the datapath.
2. Packet counters are now kept for packets received and
packets recirculated. These are kept as separate counters for
maintainability reasons. The cost of incrementing these counters
is negligible. These new counters are also displayed to the user.
3. A display statistic is added for the average number of
datapath passes per packet. This should be useful for user
debugging and understanding of packet processing.
4. The user visible 'miss' counter is used for successful upcalls,
rather than the sum of sucessful and unsuccessful upcalls. Hence,
this becomes what user historically understands by OVS 'miss upcall'.
The user display is annotated to make this clear as well.
5. The user visible 'lost' counter remains as failed upcalls, but
is annotated to make it clear what the meaning is.
6. The enum pmd_stat_type is annotated to make the usage of the
stats counters clear.
7. The subtable lookup stats is renamed to make it clear that it
relates to masked lookups.
8. The PMD stats test is updated to handle the new user stats of
packets received, packets recirculated and average number of datapath
passes per packet.
On top of that introduce a "-pmd <core>" option to the PMD info
commands to filter the output for a single PMD.
Made the pmd-stats-show output a bit more readable by adding a blank
between colon and value.
Signed-off-by: Jan Scheurich <jan.scheurich@ericsson.com>
Co-authored-by: Darrell Ball <dlu998@gmail.com>
Signed-off-by: Darrell Ball <dlu998@gmail.com>
Acked-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Billy O'Mahony <billy.o.mahony@intel.com>
Signed-off: Ian Stokes <ian.stokes@intel.com>
A get command is added for number of conntrack connections.
This command is only supported in the userspace datapath
at this time.
Signed-off-by: Darrell Ball <dlu998@gmail.com>
Signed-off-by: Antonio Fischetti <antonio.fischetti@intel.com>
Co-authored-by: Antonio Fischetti <antonio.fischetti@intel.com>
Signed-off-by: Ben Pfaff <blp@ovn.org>
Get and set dpctl commands are added for conntrack maxconns.
These commands are only supported in the userspace
datapath at this time.
Signed-off-by: Darrell Ball <dlu998@gmail.com>
Signed-off-by: Antonio Fischetti <antonio.fischetti@intel.com>
Co-authored-by: Antonio Fischetti <antonio.fischetti@intel.com>
Signed-off-by: Ben Pfaff <blp@ovn.org>
This patch changes OVS_KEY_ATTR_NSH
to nested attribute and adds three new NSH sub attribute keys:
OVS_NSH_KEY_ATTR_BASE: for length-fixed NSH base header
OVS_NSH_KEY_ATTR_MD1: for length-fixed MD type 1 context
OVS_NSH_KEY_ATTR_MD2: for length-variable MD type 2 metadata
Its intention is to align to NSH kernel implementation.
NSH match fields, set and PUSH_NSH action all use the below
nested attribute format:
OVS_KEY_ATTR_NSH begin
OVS_NSH_KEY_ATTR_BASE
OVS_NSH_KEY_ATTR_MD1
OVS_KEY_ATTR_NSH end
or
OVS_KEY_ATTR_NSH begin
OVS_NSH_KEY_ATTR_BASE
OVS_NSH_KEY_ATTR_MD2
OVS_KEY_ATTR_NSH end
In addition, NSH encap and decap actions are renamed as push_nsh
and pop_nsh to meet action naming convention.
Signed-off-by: Yi Yang <yi.y.yang@intel.com>
Signed-off-by: Ben Pfaff <blp@ovn.org>
FreeBSD insists that <sys/types.h> be included before <netinet/in.h> and
that <netinet/in.h> be included before <arpa/inet.h>. This adds guards to
the "sparse" headers to yield a warning if this order is violated. This
commit also adjusts the order of many #includes to suit this requirement.
Signed-off-by: Ben Pfaff <blp@ovn.org>
Acked-by: Justin Pettit <jpettit@ovn.org>
New statistics for 'pmd-stats-show' command:
average number of packets per output batch.
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com
Not needed anymore because 'may_steal' already handled on
dpif-netdev layer and always true.
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com
While processing incoming batch of packets they are scattered
across many per-flow batches and sent separately.
This becomes an issue while using more than a few flows.
For example if we have balanced-tcp OvS bonding with 2 ports
there will be 256 datapath internal flows for each dp_hash
pattern. This will lead to scattering of a single recieved
batch across all of that 256 per-flow batches and invoking
send for each packet separately. This behaviour greatly degrades
overall performance of netdev_send because of inability to use
advantages of vectorized transmit functions.
But the half (if 2 ports in bonding) of datapath flows will
have the same output actions. This means that we can collect
them in a single place back and send at once using single call
to netdev_send. This patch introduces per-port packet batch
for output packets for that purpose.
'output_pkts' batch is thread local and located in send port cache.
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com
In current implementation 'now' variable updated once on each
receive cycle and passed through the whole datapath via function
arguments. It'll be better to keep this variable inside PMD
thread structure to be able to get it at any time. Such solution
will save the stack memory and simplify possible modifications
in current logic.
This patch introduces new structure 'dp_netdev_pmd_thread_ctx'
contained by 'struct dp_netdev_pmd_thread' to store any processing
context of this PMD thread. For now, only time and cycles moved to
that structure. Can be extended in the future.
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
Algs can use variable control port numbers for servers.
The main use case is a kind of feeble security measure; the
thinking being by some is that it obscures the alg traffic.
It is really not very effective, but the kernel has this
capability. This patch mimics the capability.
Signed-off-by: Darrell Ball <dlu998@gmail.com>
Signed-off-by: Ben Pfaff <blp@ovn.org>
Acked-by: Aaron Conole <aconole@redhat.com>
"sparse" warns when odp_port_t is used directly in an inequality
comparison. This avoids the warning.
CC: Kevin Traynor <ktraynor@redhat.com>
Fixes: a130f1a89bd8 ("dpif-netdev: Add port/queue tiebreaker to rxq_cycle_sort.")
Signed-off-by: Ben Pfaff <blp@ovn.org>
Acked-by: Kevin Traynor <ktraynor@redhat.com>
Acked-by: Ian Stokes <ian.stokes@intel.com>
This reverts commit a807c15796ddc43ba1ffb2a6b0bd2ad4e2b73941.
Padding and aligning of dp_netdev_pmd_thread structure members
is useless, broken in a several ways and only greatly degrades
maintainability and extensibility of the structure.
Issues:
1. It's not working because all the instances of struct
dp_netdev_pmd_thread allocated only by usual malloc. All the
memory is not aligned to cachelines -> structure almost never
starts at aligned memory address. This means that any further
paddings and alignments inside the structure are completely
useless. Fo example:
Breakpoint 1, pmd_thread_main
(gdb) p pmd
$49 = (struct dp_netdev_pmd_thread *) 0x1b1af20
(gdb) p &pmd->cacheline1
$51 = (OVS_CACHE_LINE_MARKER *) 0x1b1af60
(gdb) p &pmd->cacheline0
$52 = (OVS_CACHE_LINE_MARKER *) 0x1b1af20
(gdb) p &pmd->flow_cache
$53 = (struct emc_cache *) 0x1b1afe0
All of the above addresses shifted from cacheline start by 32B.
Can we fix it properly? NO.
OVS currently doesn't have appropriate API to allocate aligned
memory. The best candidate is 'xmalloc_cacheline()' but it
clearly states that "The memory returned will not be at the
start of a cache line, though, so don't assume such alignment".
And also, this function will never return aligned memory on
Windows or MacOS.
2. CACHE_LINE_SIZE is not constant. Different architectures have
different cache line sizes, but the code assumes that
CACHE_LINE_SIZE is always equal to 64 bytes. All the structure
members are grouped by 64 bytes and padded to CACHE_LINE_SIZE.
This leads to a huge holes in a structures if CACHE_LINE_SIZE
differs from 64. This is opposite to portability. If I want
good performance of cmap I need to have CACHE_LINE_SIZE equal
to the real cache line size, but I will have huge holes in the
structures. If you'll take a look to struct rte_mbuf from DPDK
you'll see that it uses 2 defines: RTE_CACHE_LINE_SIZE and
RTE_CACHE_LINE_MIN_SIZE to avoid holes in mbuf structure.
3. Sizes of system/libc defined types are not constant for all the
systems. For example, sizeof(pthread_mutex_t) == 48 on my
ARMv8 machine, but only 40 on x86. The difference could be
much bigger on Windows or MacOS systems. But the code assumes
that sizeof(struct ovs_mutex) is always 48 bytes. This may lead
to broken alignment/big holes in case of padding/wrong comments
about amount of free pad bytes.
4. Sizes of the many fileds in structure depends on defines like
DP_N_STATS, PMD_N_CYCLES, EM_FLOW_HASH_ENTRIES and so on.
Any change in these defines or any change in any structure
contained by thread should lead to the not so simple
refactoring of the whole dp_netdev_pmd_thread structure. This
greatly reduces maintainability and complicates development of
a new features.
5. There is no reason to align flow_cache member because it's
too big and we usually access random entries by single thread
only.
So, the padding/alignment only creates some visibility of performance
optimization but does nothing useful in reality. It only complicates
maintenance and adds huge holes for non-x86 architectures and non-Linux
systems. Performance improvement stated in a original commit message
should be random and not valuable. I see no performance difference.
Most of the above issues are also true for some other padded/aligned
structures like 'struct netdev_dpdk'. They will be treated separately.
CC: Bhanuprakash Bodireddy <bhanuprakash.bodireddy@intel.com>
CC: Ben Pfaff <blp@ovn.org>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Jan Scheurich <jan.scheurich@ericsson.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
Valgrind complains in test 1019 (dpctl - add-if set-if del-if):
4,850,896 (4,850,240 direct, 656 indirect) bytes in 1 blocks are
definitely lost in loss record 364 of 364
by 0x517062: xcalloc (util.c:103)
by 0x46CBBC: dp_netdev_set_nonpmd (dpif-netdev.c:4498)
by 0x46CBBC: create_dp_netdev (dpif-netdev.c:1299)
by 0x46CBBC: dpif_netdev_open (dpif-netdev.c:1337)
by 0x472CB0: do_open (dpif.c:350)
by 0x472E6F: dpif_create (dpif.c:404)
by 0x472E6F: dpif_create_and_open (dpif.c:417)
by 0x430EBC: open_dpif_backer (ofproto-dpif.c:727)
by 0x430EBC: construct (ofproto-dpif.c:1411)
by 0x41B714: ofproto_create (ofproto.c:539)
by 0x40C84E: bridge_reconfigure (bridge.c:647)
by 0x4104C5: bridge_run (bridge.c:2998)
by 0x406FA4: main (ovs-vswitchd.c:119)
The reference count wasn't released at this earlier return.
This fix passes the test 'make check'.
Signed-off-by: Yifeng Sun <pkusunyifeng@gmail.com>
Tested-by: Greg Rose <gvrose8192@gmail.com>
Reviewed-by: Greg Rose <gvrose8192@gmail.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
compare_rxq_cycles sums the latest cycles from each queue for
comparison with each other. While each comparison correctly
gets the latest cycles, the cycles could change between calls
to compare_rxq_cycle. In order to use consistent values through
each call of compare_rxq_cycles, sum the cycles before qsort is
called.
Requested-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
This function is used for comparison between queues
as part of the sort. It does not do the sort itself.
As such, give it a more appropriate name.
Suggested-by: Billy O'Mahony <billy.o.mahony@intel.com>
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Acked-by: Billy O'Mahony
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
rxq_cycle_sort is used to compare rx queues by their measured number
of cycles. In the event that they are equal, 0 could be returned.
However, it is observed that returning 0 results in a different sort
order on Windows/Linux. This is ok in practice but it causes a unit
test failure for
"1007: PMD - pmd-cpu-mask/distribution of rx queues" when running
on different OS's.
In order to have a consistent sort result across multiple OS's,
introduce a tiebreaker of port/queue.
Fixes: 655856ef39b9 ("dpif-netdev: Change rxq_scheduling to use rxq processing cycles.")
Reported-by: Alin Gabriel Serdean <aserdean@ovn.org>
Tested-by: Alin Gabriel Serdean <aserdean@ovn.org>
Co-authored-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
This patch adds support of flushing a conntrack entry specified by the
conntrack 5-tuple, and provides the implementation in dpif-netlink.
The implementation of dpif-netlink in the linux datapath utilizes the
NFNL_SUBSYS_CTNETLINK netlink subsystem to delete a conntrack entry in
nf_conntrack. Future patches will add support for the userspace and
Windows datapaths.
VMWare-BZ: #1983178
Signed-off-by: Yi-Hung Wei <yihung.wei@gmail.com>
Signed-off-by: Justin Pettit <jpettit@ovn.org>
rxq_interval was added before there was other #defines
and code related to rxq intervals.
Rename to rxq_next_cycles_store in order to make it more intuitive.
Requested-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Acked-by: Antonio Fischetti <antonio.fischetti@intel.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
Commit 38259bd7eb21 (dpif-netdev: Initialize new rxqs in
port_reconfigure().) added a memset for the dp_netdev_rxq of new rxq's
to remove a valgrind warning for an index field in that struct. With
the addition of that memset, it also means there are some existing
resets on other fields in that struct that are no longer needed and
gives the opportunity to simplify by removing them.
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Signed-off-by: Ben Pfaff <blp@ovn.org>
