dpdk.git
5 years agonet/bnxt: fix filter type for non-ntuple flows
Rahul Gupta [Thu, 24 Oct 2019 07:44:24 +0000 (13:14 +0530)]
net/bnxt: fix filter type for non-ntuple flows

Use HWRM_CFA_L2_FILTER filter_type, as its checked during SRC MAC
filter creation in bnxt_create_l2_filter().

Fixes: afef822b2e1b ("net/bnxt: support creating SMAC and inner DMAC filters")

Signed-off-by: Rahul Gupta <rahul.gupta@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
5 years agonet/bnxt: fix race between start and interrupt handler
Kalesh AP [Thu, 24 Oct 2019 07:44:23 +0000 (13:14 +0530)]
net/bnxt: fix race between start and interrupt handler

During port start, driver enables interrupts. At the end of port start
driver schedules the alarm for fw health check. This can cause race
between interrupt handler and schedule fw health check alarm thread
which falsely detects FW error.

Fixes: 9d0cbaecc91a ("net/bnxt: support periodic FW health monitoring")

Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
5 years agonet/bnxt: fix default MAC check
Venkat Duvvuru [Thu, 24 Oct 2019 07:44:22 +0000 (13:14 +0530)]
net/bnxt: fix default MAC check

bnxt_set_hwrm_vnic_filters programs default mac addr and the
same default mac is added by mac_add_addr_op routine as well.
This redundant mac add is avoided by checking if the default
mac is already added.

However, that check is wrong. The check should consider the
mac index as well to determine the default mac. This patch
fixes it by using mac index to determine the default mac.

Fixes: 68f589f2c728 ("net/bnxt: fix setting primary MAC address")
Cc: stable@dpdk.org
Signed-off-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
5 years agonet/bnxt: update HWRM API to version 1.10.1.6
Ajit Khaparde [Thu, 24 Oct 2019 07:44:21 +0000 (13:14 +0530)]
net/bnxt: update HWRM API to version 1.10.1.6

Update HWRM API to version 1.10.1.6

Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
5 years agonet/bnxt: fix RSS configuration for Thor controller
Santoshkumar Karanappa Rastapur [Thu, 24 Oct 2019 07:44:20 +0000 (13:14 +0530)]
net/bnxt: fix RSS configuration for Thor controller

rss_table structure stores receive, completion ring IDs for Thor based
controllers and receive queue id for others. Below commit accidentally
left the old code and hence was always overwriting receive ring id with
queue id for Thor. This causes subsequent VNIC_RSS_CFG command to fail.

Fixes: 38412304b50a ("net/bnxt: enable RSS for thor-based controllers")
Cc: stable@dpdk.org
Signed-off-by: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
5 years agonet/bnxt: fix writing MTU to FW
Ajit Khaparde [Thu, 24 Oct 2019 07:44:19 +0000 (13:14 +0530)]
net/bnxt: fix writing MTU to FW

We are currently writing the MAX supported MTU size even though
the MTU passed by the application is a different value.
Configure the application specified MTU instead.

Fixes: b7778e8a1c00 ("net/bnxt: refactor to properly allocate resources for PF/VF")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
5 years agonet/bnxt: fix setting max RSS contexts
Ajit Khaparde [Thu, 24 Oct 2019 07:44:18 +0000 (13:14 +0530)]
net/bnxt: fix setting max RSS contexts

Configure max RSS contexts in case the firmware indicates a maximal
resource reservation strategy.

Fixes: 6d8109bcb398 ("net/bnxt: check VF resources if resource manager is enabled")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
5 years agonet/bnxt: fix ping with MTU change
Santoshkumar Karanappa Rastapur [Thu, 24 Oct 2019 07:44:17 +0000 (13:14 +0530)]
net/bnxt: fix ping with MTU change

Driver was setting maximum receive unit differently in bnxt_mtu_set_op
and bnxt_hwrm_vnic_alloc. Moreover firmware adds the 4 bytes for FCS.
Fixed it by setting correct maximum receive unit value.

Fixes: daef48efe5e5 ("net/bnxt: support set MTU")
Cc: stable@dpdk.org
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
5 years agonet/bnxt: fix check while freeing filter
Kalesh AP [Thu, 24 Oct 2019 07:44:16 +0000 (13:14 +0530)]
net/bnxt: fix check while freeing filter

This patch fixes the coverity UNUSED_VALUE issue.

Coverity issue: 349938
Fixes: 49d0709b257f ("net/bnxt: delete and flush L2 filters cleanly")

Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
5 years agoraw/ifpga/base: fix dereference before null check
Tianfei Zhang [Tue, 17 Sep 2019 22:24:59 +0000 (06:24 +0800)]
raw/ifpga/base: fix dereference before null check

Add pointer null check before dereference.

Coverity issue: 344976
Fixes: 12f92a513a13 ("raw/ifpga/base: fix retimer link status")
Cc: stable@dpdk.org
Signed-off-by: Tianfei Zhang <tianfei.zhang@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agoapp/testpmd: support multiple raw encap/decap
Xiaoyu Min [Mon, 16 Sep 2019 09:21:02 +0000 (17:21 +0800)]
app/testpmd: support multiple raw encap/decap

In some scenarios, the raw_encap/raw_decap actions could be multiple in
one single flow (e,g. hirepin flow):

  ... actions raw_decap / raw_encap / raw_decap / raw_encap / ...

This requires the testpmd supports multiple raw_encap/raw_decap data
settings as well.

With the multiple raw_encap/raw_decap settings, the testpmd commands –
set raw_encap / set raw_decap will become:

  set raw_encap <index> <item pattern>
  set raw_decap <index> <item pattern>

And the actions – raw_encap/raw_decap also could optionally choose which
global raw_encap/raw_decap confs to be used by index:

  ... actions raw_decap index 1 / raw_encap index 2 / ...

If there is no `index` specified, the default index is 0:

  set raw_encap <item pattern>
  ... actions raw_decap / raw_encap / ...

which will use raw_encap index 0.

In addition to the set raw_encap/raw_decap commands,

  show <raw_encap/raw_decap> <index>
  show <raw_encap/raw_decap> all

are also introduced into in order to check which index is set and to
what.

Signed-off-by: Xiaoyu Min <jackmin@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
5 years agoapp/testpmd: fix Tx checksum when TSO enabled
Peng Huang [Thu, 24 Oct 2019 09:37:34 +0000 (09:37 +0000)]
app/testpmd: fix Tx checksum when TSO enabled

This patch fixed the TX checksum value problem when enabled TSO in
tunnel packets, because outer UDP checksum calculation depend on
the TSO configuration.

Fixes: 0f62d63593ed ("app/testpmd: support tunneled TSO in checksum engine")
Cc: stable@dpdk.org
Signed-off-by: Peng Huang <peng.huang@intel.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
5 years agonet/mlx: fix debug build with icc
Raslan Darawsheh [Wed, 23 Oct 2019 11:31:45 +0000 (11:31 +0000)]
net/mlx: fix debug build with icc

Trying to compile mlx5 pmd in debug mode with icc
will lead to compilation failures due to the fact that
icc doesn't have support for the pragma of pedantic.

Cc: stable@dpdk.org
Signed-off-by: Raslan Darawsheh <rasland@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
5 years agovhost: fix IPv4 checksum
Flavio Leitner [Thu, 24 Oct 2019 16:29:19 +0000 (13:29 -0300)]
vhost: fix IPv4 checksum

Currently the IPv4 header checksum is calculated including its
current value, which can be a valid checksum or just garbage.
In any case, if the original value is not zero, then the result
is always wrong.

The IPv4 checksum is defined in RFC791, page 14 says:
  Header Checksum:  16 bits

  The checksum algorithm is:
  The checksum field is the 16 bit one's complement of the one's
  complement sum of all 16 bit words in the header.  For purposes of
  computing the checksum, the value of the checksum field is zero.

Thus force the csum field to always be zero.

Fixes: b08b8cfeb2ae ("vhost: fix IP checksum")
Cc: stable@dpdk.org
Signed-off-by: Flavio Leitner <fbl@sysclose.org>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agoexamples: delete vhost SCSI example
Jin Yu [Thu, 24 Oct 2019 10:21:59 +0000 (18:21 +0800)]
examples: delete vhost SCSI example

This example is too old and SPDK will not maintain this example
anymore. Also SPDK has submitted a new vhost example vhost-blk.
We will keep on maintaining vhost-blk and It shows the packed
ring and live recovery support.

Signed-off-by: Jin Yu <jin.yu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agonet/virtio: reject unsupported Tx multi-queue modes
Andrew Rybchenko [Wed, 9 Oct 2019 12:32:08 +0000 (13:32 +0100)]
net/virtio: reject unsupported Tx multi-queue modes

This driver supports none of DCB or VMDQ modes, therefore must
check and return error if configured incorrectly.

Fixes: c1f86306a026 ("virtio: add new driver")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agonet/virtio: reject unsupported Rx multi-queue modes
Dilshod Urazov [Wed, 9 Oct 2019 12:32:07 +0000 (13:32 +0100)]
net/virtio: reject unsupported Rx multi-queue modes

This driver supports none of DCB, RSS or VMDQ modes, therefore must
check and return error if configured incorrectly.

Virtio can distribute Rx packets across multi-queue, but there is
no controls (algorithm, redirection table, hash function) except
number of Rx queues and ETH_MQ_RX_NONE is the best fit meaning
no method is enforced on how to route packets to MQs.

Fixes: c1f86306a026 ("virtio: add new driver")
Cc: stable@dpdk.org
Signed-off-by: Dilshod Urazov <dilshod.urazov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agonet/virtio: reject deferred Tx start
Dilshod Urazov [Wed, 9 Oct 2019 12:32:06 +0000 (13:32 +0100)]
net/virtio: reject deferred Tx start

Deferred start Tx queue is not supported by the driver.

Fixes: 0748be2cf9a2 ("ethdev: queue start and stop")
Cc: stable@dpdk.org
Signed-off-by: Dilshod Urazov <dilshod.urazov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agonet/virtio: reject deferred Rx start
Dilshod Urazov [Wed, 9 Oct 2019 12:32:05 +0000 (13:32 +0100)]
net/virtio: reject deferred Rx start

Deferred start Rx queue is not supported by the driver.

Fixes: 0748be2cf9a2 ("ethdev: queue start and stop")
Cc: stable@dpdk.org
Signed-off-by: Dilshod Urazov <dilshod.urazov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: disable host TSO for linear buffers without extbuf
Ilya Maximets [Thu, 17 Oct 2019 15:00:19 +0000 (17:00 +0200)]
vhost: disable host TSO for linear buffers without extbuf

If linear buffers requested and external buffers are not, vhost
will not be able to receive any buffer that doesn't fit in a
single mbuf.  Moreover, if such a buffer will appear in a vring
it will never be dequeued and the whole vring will become dead
breaking the network connection.

Disable segmentation offloading from the host side to avoid
having such a big buffers.

Fixes: c3ff0ac70acb ("vhost: improve performance by supporting large buffer")

Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: return error message for mbuf allocation failure
Ilya Maximets [Wed, 16 Oct 2019 14:31:07 +0000 (16:31 +0200)]
vhost: return error message for mbuf allocation failure

mbuf allocation failure is a hard failure that highlights some
significant issues with memory pool size or a mbuf leak.

We still have the message for subsequent chained mbufs, but not
for the first one.  It was removed while introducing extbuf
support for large buffers.  But it was useful for catching
mempool issues and needs to be returned back.

Fixes: c3ff0ac70acb ("vhost: improve performance by supporting large buffer")

Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Reviewed-by: Flavio Leitner <fbl@sysclose.org>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
5 years agovhost: optimize packed ring dequeue when in-order
Marvin Liu [Thu, 24 Oct 2019 16:08:32 +0000 (00:08 +0800)]
vhost: optimize packed ring dequeue when in-order

When VIRTIO_F_IN_ORDER feature is negotiated, vhost can optimize dequeue
function by only update first used descriptor.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: optimize packed ring dequeue
Marvin Liu [Thu, 24 Oct 2019 16:08:31 +0000 (00:08 +0800)]
vhost: optimize packed ring dequeue

Optimize vhost device packed ring dequeue function by splitting batch
and single functions. No-chained and direct descriptors will be handled
by batch and other will be handled by single as before.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: add packed ring zcopy batch and single dequeue
Marvin Liu [Thu, 24 Oct 2019 16:08:30 +0000 (00:08 +0800)]
vhost: add packed ring zcopy batch and single dequeue

Add vhost packed ring zero copy batch and single dequeue functions like
normal dequeue path.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: optimize packed ring enqueue
Marvin Liu [Thu, 24 Oct 2019 16:08:29 +0000 (00:08 +0800)]
vhost: optimize packed ring enqueue

Optimize vhost device packed ring enqueue function by splitting batch
and single functions. Packets can be filled into one desc will be
handled by batch and others will be handled by single as before.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: update packed ring dequeue
Marvin Liu [Thu, 24 Oct 2019 16:08:28 +0000 (00:08 +0800)]
vhost: update packed ring dequeue

Buffer used ring updates as many as possible in vhost dequeue function
for coordinating with virtio driver. For supporting buffer, shadow used
ring element should contain descriptor's flags. First shadowed ring
index was recorded for calculating buffered number.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: flush batched enqueue descs directly
Marvin Liu [Thu, 24 Oct 2019 16:08:27 +0000 (00:08 +0800)]
vhost: flush batched enqueue descs directly

Flush used elements when batched enqueue function is finished.
Descriptor's flags are pre-calculated as they will be reset by vhost.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: flush enqueue updates by cacheline
Marvin Liu [Thu, 24 Oct 2019 16:08:26 +0000 (00:08 +0800)]
vhost: flush enqueue updates by cacheline

Buffer vhost packed ring enqueue updates, flush ring descs if buffered
content filled up one cacheline. Thus virtio can receive packets at a
faster frequency.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: add packed ring batch dequeue
Marvin Liu [Thu, 24 Oct 2019 16:08:25 +0000 (00:08 +0800)]
vhost: add packed ring batch dequeue

Add batch dequeue function like enqueue function for packed ring, batch
dequeue function will not support chained descriptors, single packet
dequeue function will handle it.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: add packed ring single dequeue
Marvin Liu [Thu, 24 Oct 2019 16:08:24 +0000 (00:08 +0800)]
vhost: add packed ring single dequeue

Add vhost single packet dequeue function for packed ring and meanwhile
left space for shadow used ring update function.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: add packed ring batch enqueue
Marvin Liu [Thu, 24 Oct 2019 16:08:23 +0000 (00:08 +0800)]
vhost: add packed ring batch enqueue

Batch enqueue function will first check whether descriptors are cache
aligned. It will also check prerequisites in the beginning. Batch
enqueue function do not support chained mbufs, single packet enqueue
function will handle it.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: try to unroll for each loop
Marvin Liu [Thu, 24 Oct 2019 16:08:22 +0000 (00:08 +0800)]
vhost: try to unroll for each loop

Create macro for adding unroll pragma before for each loop. Batch
functions will be contained of several small loops which can be
optimized by compilers' loop unrolling pragma.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: add packed ring single enqueue
Marvin Liu [Thu, 24 Oct 2019 16:08:21 +0000 (00:08 +0800)]
vhost: add packed ring single enqueue

Add vhost enqueue function for single packet and meanwhile left space
for flush used ring function.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agovhost: add packed ring indexes increasing function
Marvin Liu [Thu, 24 Oct 2019 16:08:20 +0000 (00:08 +0800)]
vhost: add packed ring indexes increasing function

When enqueuing or dequeuing, the virtqueue's local available and used
indexes are increased.

Signed-off-by: Marvin Liu <yong.liu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
5 years agonet/e1000: fix zeroing of RSS config
Congwen Zhang [Wed, 7 Aug 2019 08:31:45 +0000 (16:31 +0800)]
net/e1000: fix zeroing of RSS config

The type of rss_conf is struct igb_rte_flow_rss_conf *,
 not struct rte_eth_rss_conf *.

Fixes: 424ae915baf0 ("net/e1000: move RSS to flow API")
Cc: stable@dpdk.org
Signed-off-by: Congwen Zhang <zhang.congwen@zte.com.cn>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agonet/ixgbe: fix zeroing of RSS config
Congwen Zhang [Thu, 8 Aug 2019 01:42:07 +0000 (09:42 +0800)]
net/ixgbe: fix zeroing of RSS config

Correct the rss_conf type in sizeof.

Fixes: 518cc3927b13 ("net/ixgbe: move RSS to flow API")
Cc: stable@dpdk.org
Signed-off-by: Congwen Zhang <zhang.congwen@zte.com.cn>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agonet/i40e: fix exception with multi-driver
Alvin Zhang [Tue, 8 Oct 2019 10:52:31 +0000 (18:52 +0800)]
net/i40e: fix exception with multi-driver

If support-multi-driver is enabled, the global registers should not
be configured. But with the current code base, if creating a flow
with rte_flow API, the global register GLQF_FD_MSK may be changed.

Fixes: cfdfca493cae ("net/i40e: fix multiple driver support")
Cc: stable@dpdk.org
Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agonet/ice: fix IPv6 UDP port matching
Yahui Cao [Thu, 24 Oct 2019 12:56:37 +0000 (20:56 +0800)]
net/ice: fix IPv6 UDP port matching

Reverse the src and dest notion, since the HW expects them to be from Tx
perspective where as the input from user is from Rx filter view.

Fixes: f5cafa961fae ("net/ice: add flow director create and destroy")

Signed-off-by: Yahui Cao <yahui.cao@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agonet/ice: fix flow director for IPv6 dest MAC
Yahui Cao [Wed, 23 Oct 2019 10:58:14 +0000 (18:58 +0800)]
net/ice: fix flow director for IPv6 dest MAC

Fixes: f5cafa961fae ("net/ice: add flow director create and destroy")

Signed-off-by: Yahui Cao <yahui.cao@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agonet: add new header file for VXLAN
Flavia Musatescu [Wed, 23 Oct 2019 13:19:27 +0000 (14:19 +0100)]
net: add new header file for VXLAN

The VXLAN related definitions and structures are moved from
rte_ether.h to a new header file: rte_xvlan.h.

Also introducing a new define macro for VXLAN default port id:
RTE_VXLAN_DEFAULT_PORT

Signed-off-by: Flavia Musatescu <flavia.musatescu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Tested-by: Raslan Darawsheh <rasland@mellanox.com>
5 years agonet/bnxt: fix mbuf free when clearing Tx queue
Lance Richardson [Wed, 23 Oct 2019 14:27:35 +0000 (10:27 -0400)]
net/bnxt: fix mbuf free when clearing Tx queue

When freeing pending transmit mbufs, use rte_pktmbuf_free_seg()
instead of rte_pktmbuf_free(), otherwise linked mbufs may be freed
more than once.

Fixes: 6eb3cc2294fd ("net/bnxt: add initial Tx code")
Cc: stable@dpdk.org
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
5 years agonet/mlx5: add Altivec Rx
David Christensen [Fri, 16 Aug 2019 17:56:04 +0000 (12:56 -0500)]
net/mlx5: add Altivec Rx

Added mlx5_rxtx_vec_altivec.h which supports vectorized RX
using Altivec vector code.  Modified associated build files
to use the new code.

Signed-off-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Tested-by: Raslan Darawsheh <rasland@mellanox.com>
5 years agonet: hide internal CRC defines
David Marchand [Wed, 23 Oct 2019 13:40:01 +0000 (15:40 +0200)]
net: hide internal CRC defines

No need to let those (non RTE_ prefixed) defines public.
Hide them where we use them.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet: add rte prefix to MPLS structure
David Marchand [Wed, 23 Oct 2019 13:40:00 +0000 (15:40 +0200)]
net: add rte prefix to MPLS structure

Add 'rte_' prefix to structures:
- rename struct mpls_hdr as struct rte_mpls_hdr.

Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet: add missing rte prefix on PPPoE defines
David Marchand [Wed, 23 Oct 2019 13:39:59 +0000 (15:39 +0200)]
net: add missing rte prefix on PPPoE defines

Those two defines have been missed.

Fixes: 35b2d13fd6fd ("net: add rte prefix to ether defines")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agodoc: add note on prefixing ESP tail structure
David Marchand [Wed, 23 Oct 2019 13:39:58 +0000 (15:39 +0200)]
doc: add note on prefixing ESP tail structure

Fixes: 7dde68cf0e13 ("net: add missing rte prefix for ESP tail")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoethdev: fix include of ethernet header file
Ciara Power [Wed, 23 Oct 2019 12:53:00 +0000 (13:53 +0100)]
ethdev: fix include of ethernet header file

The include for rte_ether.h in each of these files should not use
quotes, as the header file is not in the librte_ethdev directory.

These are now updated to use <> symbols, to search directories
pre-designated by the compiler.

Fixes: 57668ed7bc08 ("net: move ethernet definitions to the net library")
Cc: stable@dpdk.org
Signed-off-by: Ciara Power <ciara.power@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
5 years agoapp/testpmd: fix missing GENEVE item in raw encap
Wisam Jaddo [Thu, 17 Oct 2019 08:13:52 +0000 (08:13 +0000)]
app/testpmd: fix missing GENEVE item in raw encap

Fixes: 0f4203fe9d18 ("app/testpmd: support GENEVE pattern item in flow rules")
Cc: stable@dpdk.org
Signed-off-by: Wisam Jaddo <wisamm@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
5 years agonet/memif: support multiple Tx
Anand Sunkad [Thu, 17 Oct 2019 13:20:50 +0000 (13:20 +0000)]
net/memif: support multiple Tx

When Multiple slave/master Memif's interfaces are created in single
process data transmission over second connection is not successful.

Issue is because of "mq->in_port" is not initialized with
"dev->data->port_id" in memif_tx_queue_setup() function, and while
transmitting packets over second connection in eth_memif_tx function
it refer "mq->in_port" which is always zero, which leads to data
transmission always in 0th port.

To mitigate the issue,"mq->in_port" is initialized with
"dev->data->port_id" in memif_tx_queue_setup() function.

Signed-off-by: Anand Sunkad <anand.sunkad@benisontech.com>
Reviewed-by: Jakub Grajciar <jgrajcia@cisco.com>
5 years agonet/memif: fix Unix domain address length
Jakub Grajciar [Wed, 23 Oct 2019 08:07:32 +0000 (10:07 +0200)]
net/memif: fix Unix domain address length

Define MEMIF_SOCKET_UN_SIZE to size of unix domain socket address.
Report error in case of longer path.

Fixes: b923866c6974 ("net/memif: allow for full key size in socket name")

Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoapp/testpmd: add GTP parsing and Tx checksum offload
Ting Xu [Tue, 22 Oct 2019 16:26:36 +0000 (16:26 +0000)]
app/testpmd: add GTP parsing and Tx checksum offload

Enable testpmd to forward GTP packet in csum fwd mode.

A GTP header structure (without optional fields and extension header)
is defined in new rte_gtp.h.
A parser function in testpmd is added.  GTPU and GTPC packets are both
supported, with respective UDP destination port and GTP message type.

Signed-off-by: Ting Xu <ting.xu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agombuf: support dynamic fields and flags
Olivier Matz [Sat, 26 Oct 2019 12:39:01 +0000 (14:39 +0200)]
mbuf: support dynamic fields and flags

Many features require to store data inside the mbuf. As the room in mbuf
structure is limited, it is not possible to have a field for each
feature. Also, changing fields in the mbuf structure can break the API
or ABI.

This commit addresses these issues, by enabling the dynamic registration
of fields or flags:

- a dynamic field is a named area in the rte_mbuf structure, with a
  given size (>= 1 byte) and alignment constraint.
- a dynamic flag is a named bit in the rte_mbuf structure.

The typical use case is a PMD that registers space for an offload
feature, when the application requests to enable this feature.  As
the space in mbuf is limited, the space should only be reserved if it
is going to be used (i.e when the application explicitly asks for it).

The registration can be done at any moment, but it is not possible
to unregister fields or flags.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
5 years agoeal: fix memory config allocation for multi-process
Anatoly Burakov [Thu, 24 Oct 2019 12:36:50 +0000 (13:36 +0100)]
eal: fix memory config allocation for multi-process

Currently, mem config will be mapped without using the virtual
area reservation infrastructure, which means it will be mapped
at an arbitrary location. This may cause failures to map the
shared config in secondary process due to things like PCI
whitelist arguments allocating memory in a space where the
primary has allocated the shared mem config.

Fix this by using virtual area reservation to reserve space for
the mem config, thereby avoiding the problem and reserving the
shared config (hopefully) far away from any normal memory
allocations.

Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
5 years agomem: make base address hint OS specific
Anatoly Burakov [Thu, 24 Oct 2019 12:36:49 +0000 (13:36 +0100)]
mem: make base address hint OS specific

Not all OS's follow Linux's memory layout, which may lead to
problems following the suggested common address hint absent
of a base-virtaddr flag. Make this address hint OS-specific.

Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
5 years agovfio: fix truncated BAR offset for 32-bit
Michal Krawczyk [Thu, 24 Oct 2019 12:10:46 +0000 (14:10 +0200)]
vfio: fix truncated BAR offset for 32-bit

When 32-bit application is built on 64-bit system it is possible that
the offset of the resource is outside of the 32-bit value.

The problem with the unsigned long is, that it is 32-bit and not 64-bit
when using armhf compiler. Although the system is returning u64 value,
we are losing it's value if it's higher than 32-bit in the conversion
process. It can further cause mmap to fail due to offset being 0 or to
map not intended memory region.

To make it more portable, the uint64_t value is now being used for
storing offset instead of unsigned long. The size of being 32-bit seems
to be fine as the 32-bit application won't be able to access bigger
memory and it is further converted to size_t anyway. But for better
readability and to be consistent, it's type was changed to size_t as
well.

Fixes: 0205f873557c ("vfio: fix overflow of BAR region offset and size")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
5 years agoeal: move CPU operations to OS specific headers
Pallavi Kadam [Mon, 30 Sep 2019 17:09:34 +0000 (10:09 -0700)]
eal: move CPU operations to OS specific headers

Moving RTE_CPU* definitions from the common code to the Linux and
FreeBSD rte_os.h file to avoid #ifdef clutter.

Signed-off-by: Pallavi Kadam <pallavi.kadam@intel.com>
Signed-off-by: Antara Ganesh Kolar <antara.ganesh.kolar@intel.com>
Reviewed-by: Ranjit Menon <ranjit.menon@intel.com>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
5 years agoreciprocal: fix off-by-one with 32-bit divisor
Pavan Nikhilesh [Tue, 3 Sep 2019 19:16:45 +0000 (00:46 +0530)]
reciprocal: fix off-by-one with 32-bit divisor

Fix off-by-one error in 64bit reciprocal division when divisor is 32bit.

Caught with the unit test:

RTE>>reciprocal_division
Validating unsigned 32bit division.
Validating unsigned 64bit division.
Validating unsigned 64bit division with 32bit divisor.
Division failed, 16983222950483802557/819 = expected 20736535959076681
result 20736535959076682
Validating division by power of 2.
Test Failed

Fixes: 6d45659eacb8 ("eal: add u64-bit variant for reciprocal divide")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
5 years agotest: fix global variable multiple definitions
Ferruh Yigit [Thu, 5 Sep 2019 14:53:15 +0000 (15:53 +0100)]
test: fix global variable multiple definitions

Multiple global variable are defined in multiple unit test files with
same name, but all unit test files are linked into single executable,
which means those variables share same storage which is not the
intention, fixed by making global variables 'static'.

Issue has been detected by '-fno-common' gcc flag.

Fixes: fdeb30fa7102 ("test/bitrate: add unit tests for bitrate library")
Fixes: c3eabff124e6 ("distributor: add unit tests")
Fixes: 0e925aef2779 ("app/test: add EFD functional and perf tests")
Fixes: 359e17bf081f ("app/test: improve hash unit tests")
Fixes: c7eb0972e74b ("test/hash: add lock-free r/w concurrency")
Fixes: 1e3676a06e4c ("test/latency: add unit tests for latencystats library")
Fixes: 0cc67a96e486 ("test/member: add functional and perf tests")
Fixes: e6a14121f4ae ("test/rcu: remove arbitrary limit on max core count")
Fixes: 104dbec2081a ("test/rcu: increase size of core numbers")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoapp/testpmd: fix global variable multiple definitions
Ferruh Yigit [Thu, 5 Sep 2019 14:53:13 +0000 (15:53 +0100)]
app/testpmd: fix global variable multiple definitions

Some flow config related global variables are defined in a header file
which was causing multiple definitions of the variables, fixed it by
moving them to the .c file.

Issue has been detected by '-fno-common' gcc flag.

Also while being there,
removed duplicated 'ACTION_RAW_ENCAP_MAX_DAT definition,
moved 'vxlan_encap_conf' & 'nvgre_encap_conf' initialization to
  'cmdline_flow.c' which is better location than 'testpmd.c'
relocated 'action_raw_encap_data' & 'action_raw_decap_data' struct
   definitions slightly within the file

Fixes: 1960be7d32f8 ("app/testpmd: add VXLAN encap/decap")
Fixes: dcd962fc6b4e ("app/testpmd: add NVGRE encap/decap")
Fixes: a1191d39cb57 ("app/testpmd: add MPLSoUDP encapsulation")
Fixes: 3e77031be855 ("app/testpmd: add MPLSoGRE encapsulation")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agocompress/octeontx: fix global variable multiple definitions
Ferruh Yigit [Thu, 5 Sep 2019 14:53:12 +0000 (15:53 +0100)]
compress/octeontx: fix global variable multiple definitions

'octtx_zip_logtype_driver' global variable is defined in a header file
which was causing multiple definitions of the variable, fixed it by
moving it to the .c file.

Issue has been detected by '-fno-common' gcc flag.

Fixes: 43e610bb8565 ("compress/octeontx: introduce octeontx zip PMD")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Ashish Gupta <ashishg@marvell.com>
5 years agocrypto/virtio: fix global variable multiple definitions
Ferruh Yigit [Thu, 5 Sep 2019 14:53:11 +0000 (15:53 +0100)]
crypto/virtio: fix global variable multiple definitions

'virtio_hw_internal' global variable is defined in both 'crypto/virtio'
and 'net/virtio' PMDs. This means they share same storage when
application linked with static DPDK library, which is not the intention.

Fixing by adding 'crypto_' prefix to the 'crypto/virtio' driver.

Issue has been detected by '-fno-common' gcc flag.

Fixes: 25500d4b8076 ("crypto/virtio: support device init")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agocrypto/null: fix global variable multiple definitions
Ferruh Yigit [Thu, 5 Sep 2019 14:53:08 +0000 (15:53 +0100)]
crypto/null: fix global variable multiple definitions

'null_logtype_driver' global variable is defined in a header file which
was causing multiple definitions of the variable, fixed it by moving it
to the .c file.

Issue has been detected by '-fno-common' gcc flag.

Fixes: 735b783d8c2b ("crypto/null: add dynamic logging")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/igb: fix global variable multiple definitions
Ferruh Yigit [Thu, 5 Sep 2019 14:53:07 +0000 (15:53 +0100)]
net/igb: fix global variable multiple definitions

Filtering related global variables are defined in a header file which
was causing multiple definitions of the variables, fixed it by moving
them to the .c file.

Issue has been detected by '-fno-common' gcc flag.

Fixes: 22bb13410cb2 ("net/igb: create consistent filter")
Fixes: 424ae915baf0 ("net/e1000: move RSS to flow API")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agobus/fslmc: fix global variable multiple definitions
Ferruh Yigit [Thu, 5 Sep 2019 14:53:06 +0000 (15:53 +0100)]
bus/fslmc: fix global variable multiple definitions

'qman_version' global variable is defined in a header file which was
causing multiple definitions of the variable, fixed it by moving it to
the .c file.

Issue has been detected by '-fno-common' gcc flag.

Fixes: 293c0ca94c36 ("bus/fslmc: support memory backed portals with QBMAN 5.0")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Sachin Saxena <sachin.saxena@nxp.com>
5 years agoexamples/bpf: remove duplicate mbuf definition
Konstantin Ananyev [Fri, 27 Sep 2019 13:50:54 +0000 (14:50 +0100)]
examples/bpf: remove duplicate mbuf definition

Get rid of duplicate definitions for rte_mbuf and related macros.
Include rte_mbuf_core.h instead.

Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Michel Machado <michel@digirati.com.br>
5 years agombuf: move definitions into a separate file
Konstantin Ananyev [Fri, 27 Sep 2019 13:50:53 +0000 (14:50 +0100)]
mbuf: move definitions into a separate file

Right now inclusion of rte_mbuf.h header can cause inclusion of
some arch/os specific headers.
That prevents it to be included directly by some
non-DPDK (but related) entities: KNI, BPF programs, etc.
To overcome that problem usually a separate definitions of rte_mbuf
structure is created within these entities.
That aproach has a lot of drawbacks: code duplication, error prone, etc.
This patch moves rte_mbuf structure definition (and some related macros)
into a separate file that can be included by both rte_mbuf.h and
other non-DPDK entities.

Note that it doesn't introduce any change for current DPDK code.

Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Michel Machado <michel@digirati.com.br>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
5 years agoeal: move cache line and IOVA related definitions
Konstantin Ananyev [Fri, 27 Sep 2019 13:50:52 +0000 (14:50 +0100)]
eal: move cache line and IOVA related definitions

Right now RTE_CACHE_ and IOVA definitions are located inside rte_memory.h
That might cause an unwanted inclusions of arch/os specific header files.
See [1] for particular problem example.
Probably the simplest way to deal with such problems -
move these definitions into rte_commmon.h

Note that this move doesn't introduce any change in functionality.

[1] https://bugs.dpdk.org/show_bug.cgi?id=321

Suggested-by: Vipin Varghese <vipin.varghese@intel.com>
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Michel Machado <michel@digirati.com.br>
5 years agoport: add eventdev port type
Rahul Shah [Tue, 1 Oct 2019 18:05:26 +0000 (11:05 -0700)]
port: add eventdev port type

Adding a new port type called eventdev to the
rte_port library.

Signed-off-by: Rahul Shah <rahul.r.shah@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
5 years agosched: support 64-bit values
Jasvinder Singh [Wed, 16 Oct 2019 14:17:29 +0000 (15:17 +0100)]
sched: support 64-bit values

Modify internal structure and functions to support 64-bit
values for rates and stats parameters.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
5 years agosched: add 64-bit values
Jasvinder Singh [Wed, 16 Oct 2019 14:17:28 +0000 (15:17 +0100)]
sched: add 64-bit values

To support high bandwidth network interfaces, all rates (port,
subport level token bucket and traffic class rates, pipe level
token bucket and traffic class rates) and stats counters defined
in public data structures (rte_sched.h) are modified to support
64 bit counters.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
5 years agosched: remove redundant code
Jasvinder Singh [Fri, 25 Oct 2019 10:51:24 +0000 (11:51 +0100)]
sched: remove redundant code

Remove redundant data structure fields from port level data
structures and update the release notes.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agoexamples/qos_sched: add subport config flexibility
Jasvinder Singh [Fri, 25 Oct 2019 10:51:23 +0000 (11:51 +0100)]
examples/qos_sched: add subport config flexibility

Modify qos sample app to allow different subports of the same port
to have different configuration in terms of number of pipes, pipe
queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agoexamples/ip_pipeline: add subport config flexibility to TM
Jasvinder Singh [Fri, 25 Oct 2019 10:51:22 +0000 (11:51 +0100)]
examples/ip_pipeline: add subport config flexibility to TM

Modify ip pipeline traffic management function to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agonet/softnic: add subport config flexibility to TM
Jasvinder Singh [Fri, 25 Oct 2019 10:51:21 +0000 (11:51 +0100)]
net/softnic: add subport config flexibility to TM

Modify softnic traffic management function to allow different
subports of the same port to have different configuration in
terms of number of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agotest/sched: add subport config flexibility
Jasvinder Singh [Fri, 25 Oct 2019 10:51:20 +0000 (11:51 +0100)]
test/sched: add subport config flexibility

Modify tests function to allow different subports of the same port
to have different configuration in terms of number of pipes, pipe
queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: update queue stats read for config flexibility
Jasvinder Singh [Fri, 25 Oct 2019 10:51:19 +0000 (11:51 +0100)]
sched: update queue stats read for config flexibility

Modify pipe queue stats read function to allow different subports
of the same port to have different configuration in terms of number
of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: update pkt dequeue for flexible config
Jasvinder Singh [Fri, 25 Oct 2019 10:51:18 +0000 (11:51 +0100)]
sched: update pkt dequeue for flexible config

Modify scheduler packet dequeue operation to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: update grinder functions for config flexibility
Jasvinder Singh [Fri, 25 Oct 2019 10:51:17 +0000 (11:51 +0100)]
sched: update grinder functions for config flexibility

Modify packet grinder functions of the schedule to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: update memory compute to support flexiblity
Jasvinder Singh [Fri, 25 Oct 2019 10:51:16 +0000 (11:51 +0100)]
sched: update memory compute to support flexiblity

Update memory footprint compute function for allowing subports of
the same port to have different configuration in terms of number of
pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: modify pkt enqueue for config flexibility
Jasvinder Singh [Fri, 25 Oct 2019 10:51:15 +0000 (11:51 +0100)]
sched: modify pkt enqueue for config flexibility

Modify scheduler packet enqueue operation of the scheduler to allow
different subports of the same port to have different configuration
in terms of number of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: modify pipe functions for config flexibility
Jasvinder Singh [Fri, 25 Oct 2019 10:51:14 +0000 (11:51 +0100)]
sched: modify pipe functions for config flexibility

Modify pipe level functions to allow different subports of the same
port to have different configuration in terms of number of pipes,
pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: add pipe config to subport level
Jasvinder Singh [Fri, 25 Oct 2019 10:51:13 +0000 (11:51 +0100)]
sched: add pipe config to subport level

Add pipes configuration from the port level to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: remove pipe params config from port level
Jasvinder Singh [Fri, 25 Oct 2019 10:51:12 +0000 (11:51 +0100)]
sched: remove pipe params config from port level

Remove pipes configuration from the port level to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: modify internal structs for config flexibility
Jasvinder Singh [Fri, 25 Oct 2019 10:51:11 +0000 (11:51 +0100)]
sched: modify internal structs for config flexibility

Update internal structures related to port and subport to allow
different subports of the same port to have different configuration
in terms of number of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agosched: add pipe config params to subport struct
Jasvinder Singh [Fri, 25 Oct 2019 10:51:10 +0000 (11:51 +0100)]
sched: add pipe config params to subport struct

Add pipe configuration parameters to subport level structure to
allow different subports of the same port to have different
configuration in terms of number of pipes, pipe queue sizes, etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com>
5 years agonet/ice: add GTP tunnel type
Ting Xu [Mon, 21 Oct 2019 10:00:01 +0000 (10:00 +0000)]
net/ice: add GTP tunnel type

Add GTP tunnel type in ice pmd tunneling parameters parse
to enable Tx checksum offload.

Signed-off-by: Ting Xu <ting.xu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agombuf: add GTP tunnel type
Ting Xu [Mon, 21 Oct 2019 10:00:00 +0000 (10:00 +0000)]
mbuf: add GTP tunnel type

Add GTP tunnel type flag in mbuf for future use in GTP
Tx checksum offload.

Signed-off-by: Ting Xu <ting.xu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agoapp/testpmd: increase RSS max queue size
Yahui Cao [Wed, 23 Oct 2019 14:00:30 +0000 (22:00 +0800)]
app/testpmd: increase RSS max queue size

Max queue indices ACTION_RSS_QUEUE_NUM , which limits testpmd actions
rss queue size, has worked for a long time.

As more powered PMD is merged, more RSS queues are supported(e.g. Intel
ice driver FDIR supports up to 128 RSS queue numbers).

The original max queue indices number doesn't satisfy PMD's requirement
and it needs to become bigger.

Signed-off-by: Yahui Cao <yahui.cao@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/enetc: enable dpaax library
Gagandeep Singh [Wed, 23 Oct 2019 06:06:02 +0000 (11:36 +0530)]
net/enetc: enable dpaax library

dpaaX is used to maintain a local copy of PA->VA translations.

Using the rte_mem_virt2iova or rte_mem_virt2phy is expensive. This
library is an attempt to reduce the overall cost associated with
this translation.

This patch enables this dpaaX library by populating a dpaaX's
table for PA to VA translation. This change will also help the
caam JR driver in PA to VA translation when used with enetc driver
over enetc supported SoCs.

Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
5 years agonet/enetc: print MAC address logs at notice level
Gagandeep Singh [Wed, 23 Oct 2019 06:06:01 +0000 (11:36 +0530)]
net/enetc: print MAC address logs at notice level

To display random MAC address as notice,
a log level NOTICE is added.

Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
5 years agonet/enetc: set random MAC in case no MAC for SI
Gagandeep Singh [Wed, 23 Oct 2019 06:06:00 +0000 (11:36 +0530)]
net/enetc: set random MAC in case no MAC for SI

for SGMII interfaces, there can be 0 value
written on MAC registers.
This patch set the random MAC address for those
interfaces.

Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
5 years agonet/enetc: fix BD ring alignment
Gagandeep Singh [Wed, 23 Oct 2019 06:05:59 +0000 (11:35 +0530)]
net/enetc: fix BD ring alignment

enetc BD rings should be aligned to 128
instead of RTE_CACHE_LINE_SIZE.

Fixes: 469c6111a799 ("net/enetc: enable Rx and Tx")
Cc: stable@dpdk.org
Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
5 years agonet/mlx5: fix DevX event registration timing
Matan Azrad [Tue, 22 Oct 2019 07:33:35 +0000 (07:33 +0000)]
net/mlx5: fix DevX event registration timing

The DevX counter management triggers an asynchronous event to get back
the new counters values from the HW.

The counter management doesn't trigger 2 parallel events for the same
pool, hence, the pool cannot be updated again in the event waiting time.

When the port is stopped, the DevX event mechanism wrongly was
destroyed what remained all the waiting pools in waiting state forever.

As a result, the counters of the stuck pools were never updated again.

Separate the DevX interrupt installation from the dev installation and
remove the DevX interrupt unregistration\registration from the
stop\start operations.

Now, the DevX interrupt should be installed in probe and uninstalled in
close.

Cc: stable@dpdk.org
Fixes: f15db67df09c ("net/mlx5: accelerate DV flow counter query")

Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
5 years agonet/ice: fix queue index for switch filter
Wei Zhao [Tue, 22 Oct 2019 06:39:26 +0000 (14:39 +0800)]
net/ice: fix queue index for switch filter

When calculate queue index for queue action in
switch filter, vsi base queue should be included.

Fixes: 57c4f2693588 ("net/ice: enable switch filter")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agonet/af_xdp: support unaligned umem chunks
Ciara Loftus [Mon, 30 Sep 2019 16:42:05 +0000 (16:42 +0000)]
net/af_xdp: support unaligned umem chunks

This patch enables the unaligned chunks feature for AF_XDP which allows
chunks to be placed at arbitrary places in the umem, as opposed to them
being required to be aligned to 2k. This allows for DPDK application
mempools to be mapped directly into the umem and in turn enable zero copy
transfer between umem and the PMD.

This patch replaces the zero copy via external mbuf mechanism introduced
in commit e9ff8bb71943 ("net/af_xdp: enable zero copy by external mbuf").
The pmd_zero copy vdev argument is also removed as now the PMD will
auto-detect presence of the unaligned chunks feature and enable it if so
and otherwise fall back to copy mode if not detected.

Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Signed-off-by: Kevin Laatz <kevin.laatz@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
5 years agoethdev: add HIGIG2 key field to flow API
Kiran Kumar K [Tue, 22 Oct 2019 04:16:48 +0000 (09:46 +0530)]
ethdev: add HIGIG2 key field to flow API

Add new rte_flow_item_higig2_hdr in order to match higig2 header.
It is a layer 2.5 protocol and used in Broadcom switches.
Header format is based on the following document.
http://read.pudn.com/downloads558/doc/comm/2301468/HiGig_protocol.pdf

Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
5 years agodrivers/net: enable promiscuous and multicast by default
Ciara Power [Mon, 21 Oct 2019 12:22:38 +0000 (13:22 +0100)]
drivers/net: enable promiscuous and multicast by default

The promiscuous and multicast fields are now initialised as enabled for
some virtual PMDs. This allows the devices to be used when running
applications that attempt to enable promiscuous or multicast mode.

Signed-off-by: Ciara Power <ciara.power@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
5 years agoethdev: check device promiscuous state
Ciara Power [Mon, 21 Oct 2019 12:22:37 +0000 (13:22 +0100)]
ethdev: check device promiscuous state

The promiscuous enable and disable functions now check the
promiscuous state of the device before checking if the dev_ops
function exists for the device.

This change is necessary to allow sample applications run on
virtual PMDs, as previously -ENOTSUP returned when the promiscuous
enable function was called. This caused the sample application to
fail unnecessarily.

Signed-off-by: Ciara Power <ciara.power@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
5 years agonet/af_packet: improve Tx statistics accuracy
Flavia Musatescu [Fri, 18 Oct 2019 14:24:31 +0000 (15:24 +0100)]
net/af_packet: improve Tx statistics accuracy

When sendto call fails and ENOBUFS/EAGAIN error is being set
some of the packets are actually successfully transmitted.
There is no available count of those packets, so in order to
make the statistics more accurate, all the previously enqueued
packets will be considered successful, even though this is not
entirely correct.

Statistics numbers before this update:

Pktgen:
   Total Rx Pkts:               1360084
         Tx Pkts:               2000000
testpmd:
   RX-packets: 1408346  RX-missed: 0       RX-bytes:  84503418
   TX-packets: 526486   TX-errors: 881851  TX-bytes:  31589724

Statistics numbers after this update:

Pktgen:
   Total Rx Pkts:               1329872
         Tx Pkts:               2000000
testpmd:
   RX-packets: 1389156  RX-missed: 0       RX-bytes:  83349360
   TX-packets: 1389156  TX-errors: 0       TX-bytes:  83349360

Fixes: 74b7fc0a0ff1 ("net/af_packet: fix packet bytes counting")
Cc: stable@dpdk.org
Signed-off-by: Flavia Musatescu <flavia.musatescu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
5 years agonet/mlx5: fix crash on hash Rx queue handling for drop
Xiaoyu Min [Thu, 17 Oct 2019 10:49:46 +0000 (13:49 +0300)]
net/mlx5: fix crash on hash Rx queue handling for drop

When to create hrxq for the drop, it could fail on creating qp and goto
the error handle which will release created ind_table by calling drop
release function, which takes rte_ethdev as the only parameter and uses
the priv->drop_queue.hrxq as input to release.

Unfortunately, at this point, the hrxq is not allocated and
priv->drop_queue.hrxq is still NULL, which leads to a segfault.

This patch fixes the above by allocating the hrxq at first place and
when the error happens, hrxq is released as the last one.

This patch also release other allocated resources by the correct order,
which is missing previously.

Fixes: 78be885295b8 ("net/mlx5: handle drop queues as regular queues")
Cc: stable@dpdk.org
Reported-by: Zengmo Gao <gaozengmo@jd.com>
Signed-off-by: Xiaoyu Min <jackmin@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>