| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Also fixed minor issues such as
- prevent using vdev on DPDK <= 17.05 as not supported
- potential segmentation fault after panic (if panic thrown early
in prox initialization)
Change-Id: I49a1962e95a4a44dddadd1fdd906e1b3b232465c
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
- vdev devices now support VLAN.
- kernel tap device can be configured with a netmask (/24 was
always used in previous version).
- when sending a (fake) packet to the kernel, this packet will now
not be routed by the kernel (i.e. it will leave through the
interface configured by PROX). This might change in the futture
when PROX supports multiple VLANs per port. But today it prevents
ARP being sent on management interfaces.
- Log error in case kernel unable to send packet.
- Added support for comments (';') in lua sections.
- Prevent duplication of local_ipv4 - should now be configured in port section
local_ipv4 in core section still supported
Change-Id: I8f9a40fe6ad6f3013ff91b58b44627c7f34081e6
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Change-Id: I45ea9741be1c552d8f13a7de027994c23995df69
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Following messages are now handled by PROX
- router_solicitation
- neighbour_solicitation
- router_advertisement
- neighbour_advertisement
The following parameters are supported (through the PROX config file)
- sub mode=ndp
This will enable handling of router and neighbour solicitation
and advertisement.
- local ipv6=xxxx:xxxx:xxxxx:xxxx:xxxx:xxxx:xxxx:xxxx
This will configure the local IPv6 address of the port.
This parameter is optional. If not specified, the local IPv6
will be calculated from the EUI.
- global ipv6=xxxx:xxxx:xxxxx:xxxx:xxxx:xxxx:xxxx:xxxx
This will configure the global IPv6 address of the port.
This parameter is optional. If not specified, the global IPv6
will be calculated from the EUI and the router prefix received
from the router.
- ipv6 router=yes
This will cause the core to behave as an IPv6 router
i.e. it will generate Router Advertisement messages
This is only useful in back to back cases, when no real
IPv6 router is present in the setup.
- router prefix=xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx
The router prefix usedin the router advertisement
The prefix will be used by the node to build an IPv6 global
address in cases none were configured.
"Unsollicited NA" parameter has been added within the core/task section.
If set to yes (Unsollicited NA=yes), then an unsollicited neighbour
Advertisement is sent at startup
A same core/task cannot support both l3 and ndp mode.
Those messages will be generated or handled when submode
is set to "ndp":
- neighbour sollicitation
- neighbour advertisement
- router sollicitation
- router advertisement
An example configuration is provided: config/ipv6.cfg in which
port 0 / core 1 plays the role of the generator and port 1 /
core 2 plays the role of the swap.
Change-Id: Id0ab32d384448b4cf767fb4a1c486fc023f4f395
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
A virtual device can be created adding the following option on the eal field within
the eal options section:
--vdev=net_ring0
This will cause a new (virtual) dpdk port to be created.
Such a device can then be used in a primary process as other DPDK devices i.e.
a [port x] section must be defined in the configuration file
For a secondary process, there is no need to add --vdev, but, of course
--proc-type=secondary (or --proc-type=auto) must be set within the eal field.
There are two ways to use such a virtual device in a secondary process
1) Using the DPDK port number. Note however that the DPDK port number of virtual devices
in the secondary process might not start from 0 and might hence have to be guessed
e.g. if using 2 virtual devices in primary process (--vdev=net_ring0 --vdev=net_ring1)
the port_id will be 0 and 1 in primary process and 2 & 3 in secondary process.
2) Using the port name, as defined in --vdev in the primary process. In this way,
no [port] section must be defined, and the virtual port names are directly used within
the "rx port" and "tx port" configuration e.g.:
rx port=net_ring0
tx port=net_ring1
Limitations
===========
There seems to be a PROX leak causing DPDK port id to increase when restarting the
secondary process (causing the secondary process configuration through port id more difficult).
As the primary process configures the port (including the number of rx and tx queues)
based on its config file, the secondary process can't setup its own queues.
Simple configurations (such as using gen or lat in primary, and swap in secondary) work as PROX
allocates a RX queue even in gen mode and a TX queue even in lat mode. Better configuration/support
for secondary process should be designed.
The use of dpdk port names is only supported by PROX within "rx port" and "tx port". It is not
supported by other configuration fields such as "tx ports from routing table".
The use of dpdk port names in "rx port" and "tx port" is limited to only 1 port.
Change-Id: Iaa606625da471403713a21df79d3ded4bb91b91e
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
For kernel supported devices, add for vlan tag support
This can be configured through port parameter:
vlan tag=<vlan tag>
If this parameter is set, a vlan tagged interface is created
on top of the tap device
This is only supported for vdev tap devices
When sending (untagged) packet to the tap device (through socket)
the tap should react in sending tagged packet
Note that receiving in L3 mode (w/o tap support) a tagged packet
is not yet supported.
Change-Id: I363fa2f8d2341ac41ef23620222ece1d944bf336
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
This commit adds support for the creation of tap devices.
To create a tap device, and associate it with a DPDK port,
add within the port section:
vdev=<tap_name>
local ipv4=<ip_assigned_to_port>
<tap_name> will appear as a kernel network device, with an IP <ip_assigned_to_port>
ARP packets received from the network on the DPDK interface will be forwarded to the kernel.
1st packet which needs to be generated to a new IP will cause an IP packet
to be sent to the kernel (using socket sendto) forcing the kernel to send an ARP_REQUEST.
In this commit, there is no notification from the kernel about ARP table changes.
This is subject to further commits.
Two config files, gen_tap.cfg and swap_tap.cfg have been created. They work one against each other.
*** Note however that they do not work if running within the same host
(with the port being connected back to back) ***
This is due to the fact that the kernel would bypass the ports and use local interface...
Change-Id: Iadeec0d99e3c693472ea44bdb9163a3bf97df2fa
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Change-Id: I58fa7a0ab7aaa27e9f9383f7995bb71bf0952618
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Change-Id: I090bf8f1cd5a5d2dfa4f336bc66c46cc03d6c5bb
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
This commit prepares PROX for supporting changes from DPDK 19.08
Mainly, the "sed" listed below were run.
In addition,
- some forward definition have been removed,
- prox_compat.h has been updated
- prox_compat.h has been included where necessary.
Hence such a commit is rather easy to review and to reapply
on other branches if necessary.
sed -i 's/struct ether_hdr/prox_rte_ether_hdr/g' *.h
sed -i 's/struct ether_hdr/prox_rte_ether_hdr/g' *.c
sed -i 's/struct ether_addr/prox_rte_ether_addr/g' *.c
sed -i 's/struct ether_addr/prox_rte_ether_addr/g' *.h
sed -i 's/prox_rte_ether_addr_port/struct ether_addr_port/g' *.c
sed -i 's/prox_rte_ether_addr_port/struct ether_addr_port/g' *.h
sed -i 's/struct vlan_hdr/prox_rte_vlan_hdr/g' *.c
sed -i 's/struct vlan_hdr/prox_rte_vlan_hdr/g' *.h
sed -i 's/struct ipv4_hdr/prox_rte_ipv4_hdr/g' *.h
sed -i 's/struct ipv4_hdr/prox_rte_ipv4_hdr/g' *.c
sed -i 's/struct ipv6_hdr/prox_rte_ipv6_hdr/g' *.c
sed -i 's/struct ipv6_hdr/prox_rte_ipv6_hdr/g' *.h
sed -i 's/struct udp_hdr/prox_rte_udp_hdr/g' *.c
sed -i 's/struct udp_hdr/prox_rte_udp_hdr/g' *.h
sed -i 's/struct tcp_hdr/prox_rte_tcp_hdr/g' *.c
sed -i 's/struct tcp_hdr/prox_rte_tcp_hdr/g' *.h
sed -i 's/struct prox_rte_ether_addr_copy/ether_addr_copy/g' *.c
sed -i 's/struct prox_rte_ether_addr_copy/ether_addr_copy/g' *.h
sed -i 's/struct prox_rte_ether_addr_copy/prox_rte_ether_addr_copy/g' *.c
sed -i 's/struct prox_rte_ether_addr_copy/ether_addr_copy/g' *.h
sed -i 's/prox_rte_ether_addr_copy/ether_addr_copy/g' *.h
sed -i 's/prox_rte_ether_addr_copy/ether_addr_copy/g' *.c
sed -i 's/ether_addr_copy/prox_rte_ether_addr_copy/g' *.c
sed -i 's/ether_addr_copy/prox_rte_ether_addr_copy/g' *.h
sed -i 's/prox_rte_ether_hdr_arp/struct ether_hdr_arp/g' *.h
sed -i 's/prox_rte_ether_hdr_arp/struct ether_hdr_arp/g' *.c
sed -i 's/struct vxlan_gpe_hdr/prox_rte_vxlan_gpe_hdr/g' *.h
sed -i 's/struct vxlan_gpe_hdr/prox_rte_vxlan_gpe_hdr/g' *.c
sed -i 's/eth_random_addr/prox_rte_eth_random_addr/g' *.c
sed -i 's/eth_random_addr/prox_rte_eth_random_addr/g' *.h
sed -i 's/ETHER_CRC_LEN/PROX_RTE_ETHER_CRC_LEN/g' *.c
sed -i 's/ETHER_CRC_LEN/PROX_RTE_ETHER_CRC_LEN/g' *.h
sed -i 's/ETHER_HDR_LEN/PROX_RTE_ETHER_HDR_LEN/g' *.c
sed -i 's/ETHER_HDR_LEN/PROX_RTE_ETHER_HDR_LEN/g' *.h
sed -i 's/ETHER_MAX_LEN/PROX_RTE_ETHER_MAX_LEN/g' *.c
sed -i 's/ETHER_MAX_LEN/PROX_RTE_ETHER_MAX_LEN/g' *.h
sed -i 's/ETHER_MIN_LEN/PROX_RTE_ETHER_MIN_LEN/g' *.c
sed -i 's/ETHER_MIN_LEN/PROX_RTE_ETHER_MIN_LEN/g' *.h
sed -i 's/TCP_SYN_FLAG/PROX_RTE_TCP_SYN_FLAG/g' *.c
sed -i 's/TCP_RST_FLAG/PROX_RTE_TCP_RST_FLAG/g' *.c
sed -i 's/TCP_FIN_FLAG/PROX_RTE_TCP_FIN_FLAG/g' *.c
sed -i 's/TCP_ACK_FLAG/PROX_RTE_TCP_ACK_FLAG/g' *.c
Change-Id: I4dfe8be68e618c94dcaae28754579fbeb571bb00
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
|
|
A high number of RX descriptor can be helpful in avoiding
packet loss due to the core being interrupted.
This commit increases the default number of descriptors from 256 to 2K.
Those values are checked versus the minimum and maximum supported by the pmd.
Number of mbufs (mempool size) have been updated n config files to support this
default number of dsescriptors.
The number of descriptors can still be overwritten by the config file. However
those values must always remain within the limits reported by the pmd.
Change-Id: I5f7999eca886dad68f3c0733da0d796bf4c06a56
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Some OVS versions reports that they support UDP offload and no IPv4 offload,
but fails when UDP offload is enabled.
Hence, now, when virtio reports that it supports UDP checksum and no IPV4
checksum offload, we disable UDP checksum offload.
Change-Id: I4f22f6be4d97b0aedf267ebeaa498c0dacef3317
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Change-Id: I7c3e2981692d823c86787b77d61190327d6448e4
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Multicast can be enabled through configuration or through command line
- Through configuration
Add multicast=mcast_address (e.g. multicast=01:00:5e:01:02:03) in the port
section
- Through command line
run enable multicast port_id mcast_address (e.g. enable multicast 1 01:00:5e:01:02:03)
IGMP join message is sent unsollicited through command line:
join igmp core_id task_id ip (e.g. join igmp 1 0 224.1.1.3)
To enable swap answering IGMP Query (w/ IGMP Join)
- Through configuration
Add igmp ipv4=ip_address within the core/task section
- Through command line
join igmp core_id task_id ip (e.g. join igmp 1 0 224.1.1.3)
(this will 1st initiate an unsollicated join, then answer any subsequent query)
UDP/TCP packets received on a multicast address (224.0.0.0 => 239.255.255.255) are discarded
To stop sending responses to IGMP query:
leave igmp core_id task_id
Change-Id: I3808ccabf3b38b5a1e10e1b044db63aa05bcd7b5
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
MAC address can be configured through the [port] section.
However this had only some effect on packets being transmitted.
For instance it was used to set the src mac address by the generator.
It had no effect on packets reception: when not in promiscuous mode
only the packets with the MAC originally set by the HW were received.
Now, when not in promiscuous mode, only packets with MAC as defined
in [port] section will be received.
Change-Id: I064817bf6727d098e89fbbb7a90412ed599f0fb6
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
rx_queue_stats_mapping and tx_queue_stats_mapping were called
unrespective of the number of active queues, and failing
on all unactive queues. this caused lot of errros in prox.log
Change-Id: I15ecb7adfb562caff8efb0eb24c341a3ba354571
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
This is required for instance on gcc (GCC) 8.2.1 20180905
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
Change-Id: Id86de5d39d77c5cbf168cc51434f436f84376a4c
|
|
Different PMD have different requirements for minimum rx buffer size.
vmxnet3 has (in DPDK 18.08) requires minimum 1518 + 128 (RTE_PKTMBUF_HEADROOM).
Change-Id: I327ae5b8dcf7eef8dd71fb4e2f0603b9bab6e1d3
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Improve PROX generator performance by pre-calculating bytes_to_tsc.
This improvement is only implemented for non-pcap
generator, where only few different packet sizes are usually generated.
This change might have a negative performance impact in some cases, if
many different packet sizes are generated, resulting in higher memory usage.
This is the case for instance if random is applied to packet size.
In addition, simplified the rx path, receiving now only MAX_PKT_BURST packets
per handle loop.
Before we were trying to empty the NIC looping on RX packets, ending up
with many rx packets per handle loop. This was used to determine an lower bound
for the time the packet was received.
We now set the lower bound when less than MAX_PKT_BURST has been received.
Change-Id: I1ce813d7e4ac1577ea412c590add5d6f94b36ec7
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
JIRA: SAMPLEVNF-151
link speed is used in gen and lat latency extrapolations.
Using a link_speed value lower than the actual link speed
might result in errors (e.g. negative latencies).
Negotiated link speed might be reported slowly (as reported through irq)
Hence it is better to use the device capability link speed.
In addition, this remove the check for link speed changes in fastpath.
Change-Id: I0f475fe5e139b046012de6cd0b710e4390735078
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
RSS was enabled when rss=yes is set in the configuration or
when multiqueue is being used.
But when RSS was enabled the types of packets on which RSS
hashing was applied might not be supported by the PMD, as
almost every PMD supported a different set of packets.
Change-Id: I1d9516da61df7ef18c1d6953879ee15e33fca077
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Change-Id: Ia9df816d522e5db460d9129109525d5d345d997f
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
|
|
JIRA: SAMPLEVNF-149
VLAN can be enabled on a port by adding "vlan=yes" in the port section.
When VLAN is enabled on a port, then DEV_RX_OFFLOAD_VLAN_STRIP
and DEV_TX_OFFLOAD_VLAN_INSERT are enabled (if device supports it).
This will cause VLAN to be stripped from any packets received with
the proper tag, and VLAN to be added for any packets being transmitted.
The VLAN ID themselves are configured through the physical function
using something like (where ens801f1 isthe PF):
ip link set ens801f1 vf 0 vlan 1111
Change-Id: I945c87b0c18565da479ecaa08e5ffce91232a7ce
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
In some cases, like when using virtual functions, getting device
info return a NULL device, which was potentially causing a crash.
Change-Id: Icc65f35bb54cd7d9bd8d837b3d6ec7ebb48b7c5e
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Improve DPDK 18.05 support introduced by 3e532aca.
Support for DPDK 18.08.
Change-Id: Ide712ee94254b506a0ad88c95a7e01b789f99d48
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
JIRA: SAMPLEVNF-127
PROX was crashing in the VM with vector mode enabled and multiple OVS
queues. This was not really a PROX bug, but a DPDK issue, which has been
fixed by DPDK commit https://git.dpdk.org/dpdk/commit/?id=efc83a1e7fc3
This "net/virtio: fix queue setup consistency" is included in DPDK 17.11
and subsequent versions, and has been backported into:
- DPDK 16.11.4:
https://git.dpdk.org/dpdk-stable/commit/?h=16.11&id=516447a5056c
- DPDK 17.08.1:
https://git.dpdk.org/dpdk-stable/commit/?h=17.08&id=907fe4fc263e
This means the fix is not included into any DPDK 17.02.x used by NSB.
Fortunately, a simple workaround consists of calling
rte_eth_tx_queue_setup() before rte_eth_rx_queue_setup().
This change implements this simple workaround, in order to make PROX
work, even with unfixed DPDK versions.
Change-Id: I818e9bb812babe023c6b7225c8b9769a359d9bec
Signed-off-by: Patrice Buriez <patrice.buriez@intel.com>
|
|
This patch makes PROX compile with DPDK 18.05.
However some features are disabled or will not work as expected
such as:
"mem info" command
"police" mode
Those will need to be fixed later (see DPDK1805 within code).
Change-Id: Ie7d72f6ab3db68bc093c42d60d125fe86f8fbda7
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
Signed-off-by: Patrice Buriez <patrice.buriez@intel.com>
|
|
mbuf size was setup to achieve the best performance i.e.
using the smallest mbuf and not segmenting packets.
However this resulted in complex code, much dependent of the way
the pmd are working e.g. a change(fix) in recent dpdk i40e
implementation caused a 1782 (=1518+8+256) bytes mbuf to be too
small to hold a 1518 bytes packets.
Hence this change simplifies the mbuf size selection at the price
of a potential decreases in performance - as more memory is now used.
Except if jumbo frames are used, the mbuf size will now be the same
for all modes. The packets will not be segmented except if jumbo
frames are enabled.
If jumbo frames are enabled, packets are by default segmented, except
if the mbuf size is configured big enough in the config file.
Change-Id: I222fcac7a65c0d221d5d422f419deb9c0f864172
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
Signed-off-by: Deepak S <deepak.s@linux.intel.com>
|
|
When configuring a port, prox configures queue flags.
For historitical reasons, offloads had to be disabled on
some pmds.
There is now an API to query the ports and gets their
capabilities. This fix sets now the flags based on the capabilities,
and not the driver name.
Change-Id: Ic72d15001992f26910f653c70d31e4381fff8089
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Jumbo frames are now supported through the addition of a "mtu"
parameter of the port in the config file.
Setting the mtu to a value higher than 1500 bytes will enable
the reception of jumbo frames.
In addition, the rte_eth_dev_set_mtu is now set for all pmds.
Finally, setting mbuf_size does not set MEMPOOL_F_NO_SPREAD
anymore. This option was only used for pure debugging.
Big packets can be received using two ways
- Using multiple "small" mbufs, i.e. around 2K. This is the default.
- Using one big mbuf holding the whole packet. This can be enabled
by setting a parameter mbuf_size in the receiving core configuration
Change-Id: Idd60ad31f41c89f9522dff4d894af2696b7a2ea1
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
irq mode can be used to show how a core is interrupted by other tasks.
This mode does not handle packets. It only loops reading tsc.
When the difference between two consecutive calls to rdtsc() is high
then it means the core was interrupted.
This task implementes the display, so that we can see a histogram of
interrupts as well as the maximum, per core.
Command line is also supported, through "show irq buckets" (too show
the intervals of each buckets, in micrcoseconds), and the stats
command line (showing the number of items in each buckets and the max)..
Change-Id: I153cc3deaa7b86ae2776ea44e46ef9ecfd116992
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Change-Id: I5611ead4b61b23d6c1c983852e8c75619e08ecf9
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Change-Id: Ie6d4e7ce22c27967117a446626f5923643397812
Signed-off-by: Patrice Buriez <patrice.buriez@intel.com>
|
|
Change-Id: I65e8b2d1cea2a1edfa738671e460b6a0170d3363
Signed-off-by: Deepak S <deepak.s@linux.intel.com>
|
|
JIRA: SAMPLEVNF-55
PROX is a DPDK-based application implementing Telco use-cases such as
a simplified BRAS/BNG, light-weight AFTR... It also allows configuring
finer grained network functions like QoS, Routing, load-balancing...
(We are moving PROX version v039 to sampleVNF
https://01.org/intel-data-plane-performance-demonstrators/prox-overview)
Change-Id: Ia3cb02cf0e49ac5596e922c197ff7e010293d033
Signed-off-by: Deepak S <deepak.s@linux.intel.com>
|
Xavier Simonart
Deepak S
Patrice Buriez