5543 Commits

Author SHA1 Message Date
Jasvinder Singh
ea96935de2 net/softnic: add traffic manager object
Add traffic manager(tmgr) object to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
2018-07-12 13:50:22 +02:00
Jasvinder Singh
5d37574846 net/softnic: add tap object
Add tap object implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
2018-07-12 13:50:03 +02:00
Jasvinder Singh
0efefe115a net/softnic: add mempool object
Add mempool object implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
2018-07-12 13:49:44 +02:00
Jasvinder Singh
dfcd81838c net/softnic: add link object
Add link object implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
2018-07-12 13:49:24 +02:00
Jasvinder Singh
161b551565 net/softnic: add software queue object
Add swq object implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
2018-07-12 13:49:05 +02:00
Jasvinder Singh
7e68bc20f8 net/softnic: restructure
Rework the softnic implementation to have flexiblity in enabling
more features to its receive and transmit data path.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
2018-07-12 13:48:46 +02:00
Hemant Agrawal
e507498d81 net/dpaa: move the push queue set to global init
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
2018-07-12 14:42:03 +02:00
Hemant Agrawal
2aaffeee7b drivers: support function name in NXP logs
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
2018-07-12 14:41:28 +02:00
Hemant Agrawal
1e06b6dc71 net/dpaa: remove experimental tag from PMD API
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
2018-07-12 14:40:22 +02:00
Hemant Agrawal
8d6fc8b67e net/dpaa: support default queue mode
In case DPAA FMAN configuration tool (FMC) is not available,
the system can still work with default queue(1 queue per port).

This patch also fixes some logs related to FQ ids, which were
idetified while testing this support.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
2018-07-12 14:40:14 +02:00
Nipun Gupta
f40d5a53ec bus/dpaa: make vdqcr configurable
This patch add support for configurable vdqcr exact flag.
This boost the performance, however this can give
side effects for some extra packet fetch. Which has been
taken care in the patch as well.

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
2018-07-12 14:39:54 +02:00
Hemant Agrawal
9abdad12b8 bus/dpaa: optimize fq callback routine
Avoid array of fq as packets are dq only from a single q.

Signed-off-by: Sunil Kumar Kori <sunil.kori@nxp.com>
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
2018-07-12 14:35:33 +02:00
Hemant Agrawal
65a70a98f1 net/dpaa2: fix prefetch Rx to honor number of packets
This patch fixes prefetch rx routine to
set the next prefetch request to the size of nb_pkts.
It assumes that next request would ideally be
of same size.

Fixes: 4bc5ab88dbd6 ("net/dpaa2: fix Tx only mode")
Cc: stable@dpdk.org

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
2018-07-12 14:32:58 +02:00
Hemant Agrawal
6fd3639a82 net/dpaa: fix queue error handling and logs
Fixes: 5e7455931442 ("net/dpaa: support Rx queue configurations with eventdev")
Cc: stable@dpdk.org

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
2018-07-12 14:32:46 +02:00
Bruce Richardson
5cc0ee66e4 net/kni: fix check for meson build
The configuration value indicating that KNI was build was incorrect,
causing the driver to never be built.

Fixes: 3479586fe636 ("net/kni: add to meson build")

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
2018-07-12 12:29:14 +02:00
Bruce Richardson
8a70b112e6 net/avp: fix 32-bit meson builds
When compiling with meson, extra warnings are enabled about casting from
integers to different size pointers. This triggers an error in AVP as the
addition of the offset to the pointer address causes the result to be a
64-bit integer which doesn't fit a 32-bit pointer. The fix here is to
explicitly indicate that the offset is of type "uintptr_t" which prevents
any promotion which would cause errors.

Fixes: c0ad584222b5 ("net/avp: add device initialization")

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Allain Legacy <allain.legacy@windriver.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
2018-07-12 12:29:07 +02:00
Bruce Richardson
526ca5c6cc net/sfc: disable for 32-bit builds
The sfc driver is not supported on 32-bit so disable in meson in
those cases.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
2018-07-12 12:24:37 +02:00
Thomas Monjalon
f8e9989606 remove useless constructor headers
A constructor is usually declared with RTE_INIT* macros.
As it is a static function, no need to declare before its definition.
The macro is used directly in the function definition.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
2018-07-12 00:00:35 +02:00
Chas Williams
69bce06213 net/bonding: do not clear active slave count
When the bond PMD is stopped, the active slave count is reset.
For 802.3ad mode this potentially leaks memory and clears state since
a second sequential activate_slave() will occur when the bond PMD is
restarted and the LSC callback is triggered while the active slave
count is 0. To fix this, don't clear the active slave count when
stopping. Only deactivate_slave() should be used to clear the slaves.

Fixes: 2efb58cbab6e ("bond: new link bonding library")
Cc: stable@dpdk.org

Signed-off-by: Chas Williams <chas3@att.com>
2018-07-05 15:35:19 +02:00
Rahul Lakkireddy
d67692bacf net/cxgbevf: add missing Tx byte counters
Fixes: a0a344a8f728 ("net/cxgbe: add VF port statistics")
Cc: stable@dpdk.org

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:52:22 +02:00
Rahul Lakkireddy
3abe171957 net/cxgbe: fix Rx channel map and queue type
Set the Rx channel map and ingress queue type properly to allow firmware
to manage the internal mapping correctly.

Fixes: 6c2809628cd5 ("net/cxgbe: improve latency for slow traffic")
Cc: stable@dpdk.org

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:52:22 +02:00
Rahul Lakkireddy
87a3ae3e1e net/cxgbe: query firmware for max queues available
Query firmware for max Tx and Rx queues that can be allocated.
Move the code to determine max queues to common place for both
PF and VF.

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:52:21 +02:00
Rahul Lakkireddy
d97aa41544 net/cxgbe: enable more RSS hash functions
Enable RSS on IPv4 fragmented packets and IPv6 packets with extension
headers based on 2-tuple hash.

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:52:21 +02:00
Rahul Lakkireddy
265af08e75 net/cxgbe: add link up and down ops
Add ops to set link up and down for both PF and VF. If wait_to_complete
is set, poll for link update for up to 10 seconds max.

Original work by Surendra Mobiya <surendra@chelsio.com>

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:52:21 +02:00
Shagun Agrawal
281c6e7baf net/cxgbe: support to redirect packets to egress port
Add action to redirect matched packets to specified egress physical
port without sending them to host.

Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:20:41 +02:00
Shagun Agrawal
6b805d23a5 net/cxgbe: support to match on ingress physical port
Add support to match packets based on ingress physical port.

Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:20:41 +02:00
Shagun Agrawal
a4279771c6 net/cxgbe: support to flush flows in HASH region
Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:20:41 +02:00
Shagun Agrawal
ceaea6d8e5 net/cxgbe: query hit counters for flows in HASH region
Add interface to enable hit counters for flows offloaded in HASH
region.

Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:20:41 +02:00
Shagun Agrawal
41dc98b082 net/cxgbe: support to delete flows in HASH region
Add interface to delete offloaded flows in HASH region. Use the
hash index saved during insertion to delete the corresponding flow.

Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:20:41 +02:00
Shagun Agrawal
af44a57798 net/cxgbe: support to offload flows to HASH region
Add interface to offload flows to HASH region. Translate internal
filter specification to requests to offload flows to HASH region.
Save the returned hash index of the offloaded flow for deletion later.

Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:20:41 +02:00
Shagun Agrawal
3f2c1e209c net/cxgbe: add Compressed Local IP region
CLIP region holds destination IPv6 addresses to be matched for
corresponding flows. Query firmware for CLIP resources and allocate
table to manage them. Also update LE-TCAM to use CLIP to reduce
number of slots needed to offload IPv6 flows.

Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:20:41 +02:00
Shagun Agrawal
78192b38a8 net/cxgbe: validate flows offloaded to HASH region
Fetch supported match items in HASH region. Ensure the mask
is all set for all the supported match items to be offloaded
to HASH region. Otherwise, offload them to LE-TCAM region.

Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:07:57 +02:00
Shagun Agrawal
3a381a4116 net/cxgbe: query firmware for HASH filter resources
Fetch available HASH filter resources and allocate table for managing
them. Currently only supported on Chelsio T6 family of NICs.

Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
2018-07-04 22:07:53 +02:00
Ferruh Yigit
ab3ce1e0c1 ethdev: remove old offload API
In DPDK 17.11, the ethdev offloads API has changed:
	commit cba7f53b717d ("ethdev: introduce Tx queue offloads API")
	commit ce17eddefc20 ("ethdev: introduce Rx queue offloads API")
The new API is documented in the programmer's guide:
	http://doc.dpdk.org/guides/prog_guide/poll_mode_drv.html#hardware-offload

For reminder, the main concepts in the new API were:
	- All offloads are disabled by default
	- Distinction between per port and per queue offloads.

The transition bits are now removed:
	- Translation of the old API in ethdev
	- rte_eth_conf.rxmode.ignore_offload_bitfield
	- ETH_TXQ_FLAGS_IGNORE

The old API bits are now removed:
	- Rx per-port rte_eth_conf.rxmode.[bit-fields]
	- Tx per-queue rte_eth_txconf.txq_flags
	- ETH_TXQ_FLAGS_NO*

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Shahaf Shuler <shahafs@mellanox.com>
2018-07-04 21:50:32 +02:00
Thomas Monjalon
adeddc9c18 net/fm10k: remove unused constant
The macro FM10K_SIMPLE_TX_FLAG was used with old Tx queue flags.
It is no longer used and was forgotten when cleaning old Tx flags.

Fixes: 1778ef67e2e7 ("net/fm10k: remove dependence on Tx queue flags")
Cc: stable@dpdk.org

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
2018-07-04 21:50:27 +02:00
Ed Czeck
540914bc7a net/ark: use SPDX tags
Replace the BSD license header with the SPDX tag.

Signed-off-by: Ed Czeck <ed.czeck@atomicrules.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2018-07-04 21:21:31 +02:00
Andrew Rybchenko
886beef83f net/sfc: support Rx descriptor status for EF10 ESSB datapath
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
2018-07-04 21:21:31 +02:00
Andrew Rybchenko
6e98e4d230 net/sfc: support ops to check ready descriptors in ESSB Rx
Implement EF10 ESSB Rx datapath callback to get number of pending
descriptors.

Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
2018-07-04 21:21:31 +02:00
Andrew Rybchenko
20680dae46 net/sfc: fix double-free in EF10 ESSB Rx queue purge
Number of buffers left in completed descriptor may be 0. If so,
all buffers of the descriptor are freed once again.

Fixes: 390f9b8d82c9 ("net/sfc: support equal stride super-buffer Rx mode")
Cc: stable@dpdk.org

Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
2018-07-04 21:03:02 +02:00
Andrew Rybchenko
aeeb5571ee net/sfc: discard packets with bad CRC on EF10 ESSB Rx
Fixes: 390f9b8d82c9 ("net/sfc: support equal stride super-buffer Rx mode")
Cc: stable@dpdk.org

Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
2018-07-04 20:58:40 +02:00
Ido Goshen
53bf484034 net/pcap: capture only ingress packets from Rx iface
Support rx of in direction packets only
Useful for apps that also tx to eth_pcap ports in order to not see them
echoed back in as rx when out direction is also captured

Example:
In case using rx_iface and sending *single* packet to eth1
it will loop forever as the when it is sent to tx_iface=eth1
it will be captured again on the rx_iface=eth1 and so on
  $RTE_TARGET/app/testpmd l 0-3 -n 4 \
	--vdev 'net_pcap0,rx_iface=eth1,tx_iface=eth1'
  …
  ---------------------- Forward statistics for port 0  ------------
  RX-packets: 758            RX-dropped: 0             RX-total: 758
  TX-packets: 758            TX-dropped: 0             TX-total: 758
  ------------------------------------------------------------------
While if using rx_iface_in it will not be captured on the way out and
be forwarded only once
  $RTE_TARGET/app/testpmd l 0-3 -n 4 \
	--vdev 'net_pcap0,rx_iface_in=eth1,tx_iface=eth1'
  …
  ---------------------- Forward statistics for port 0  ------------
  RX-packets: 1              RX-dropped: 0             RX-total: 1
  TX-packets: 1              TX-dropped: 0             TX-total: 1
  ------------------------------------------------------------------

Signed-off-by: Ido Goshen <ido@cgstowernetworks.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
2018-07-04 20:54:05 +02:00
Andrew Rybchenko
8028142107 net/sfc: cut non VLAN ID bits from TCI
TCI may contain PCP or DEI bits. Matching of these bits is not
supported, but the bits still may be set in specification value and
not covered by mask. So, these bits should be ignored.

Fixes: 894080975e1e ("net/sfc: support VLAN in flow API filters")
Cc: stable@dpdk.org

Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Roman Zhukov <roman.zhukov@oktetlabs.ru>
2018-07-04 20:39:15 +02:00
Maxime Coquelin
db8d679059 net/virtio: improve offload check performance
Instead of checking the multiple Virtio features bits for
every packet, let's do the check once at configure time and
store it in virtio_hw struct.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
2018-07-04 17:03:48 +02:00
Maxime Coquelin
2f90113251 net/virtio: do not use simple Rx if TCP LRO or VLAN strip
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
2018-07-04 17:03:48 +02:00
Maxime Coquelin
4174a7b59d net/virtio: improve Tx offload features negotiation
This patch improves the Tx offload features selection depending
on whether the application request for offloads.

When the application doesn't request for Tx offload features,
the corresponding features bits aren't negotiated.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
2018-07-04 17:03:48 +02:00
Maxime Coquelin
57f818963d net/virtio: remove simple Tx path
The simple Tx path does not comply with the Virtio specification.
Now that VIRTIO_F_IN_ORDER feature is supported by the Virtio PMD,
let's use this optimized path instead.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
2018-07-04 17:03:48 +02:00
Matan Azrad
1ff30d182c net/mlx5: activate Verbs cleanup on removal
Starting from rdma-core v19, Mellanox OFED 4.4, the Verbs resources
cleanup is properly activated in plug-out process when setting the
MLX5_DEVICE_FATAL_CLEANUP environment variable to 1.

Set the aforementioned variable to 1.

Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
2018-07-04 16:48:53 +02:00
John Daley
2c06cebeb9 net/enic: cap Rx packet processing to end of desc ring
In the default Rx handler stop processing packets at the end of
the completion ring so that wrapping doesn't have to be checked
in the inner while loop.

Also, check the color bit in the completion without using a conditional.

Signed-off-by: John Daley <johndale@cisco.com>
Reviewed-by: Hyong Youb Kim <hyonkim@cisco.com>
2018-07-03 01:54:27 +02:00
John Daley
35e2cb6a17 net/enic: add simple Rx handler
Add an optimized Rx handler for non-scattered Rx.

Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Signed-off-by: John Daley <johndale@cisco.com>
2018-07-03 01:54:26 +02:00
Hyong Youb Kim
5a12c38740 net/enic: check maximum packet size in Tx prepare handler
The default tx handler checks the maximum packet size. Check it in the
prepare handler too. WQ stops working if the app/driver tries to send
oversized packets, so these checks are unavoidable.

Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
2018-07-03 01:54:25 +02:00