net/mlx5: support modify header using Direct Verbs

This patch implements the set of actions to support offload
of packet header modifications to MLX5 NIC.

Implementation is based on RFC [1].

[1] http://mails.dpdk.org/archives/dev/2018-November/119971.html

Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>

net/ice: fix CRC strip

Fixes: 50370662b727 ("net/ice: support device and queue ops")

Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>

net/i40e: fix get RSS conf

rte_eth_dev_rss_hash_conf_get API doesn't force 'rss_conf.rss_key'
to be not NULL, so rss_key = NULL should be alllowed in i40e
driver.

Fixes: 16321de09396 ("ethdev: allow to get RSS hash functions and key")
Cc: stable@dpdk.org
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>

version: 19.02-rc1

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>

app/testpmd: expand RED queue thresholds to 64 bits

There's very commonly that more than 4G DDR memory in NIC for HQoS,
so right now the queue threshold size of RED needs to expand to
uint64_t. For struct rte_tm_red_params, it has been fixed, but for
test-pmd TM configuration, it hasn't been fixed. This patch fixes it.

Fixes: bddc2f40b594 ("app/testpmd: add commands for shaper and wred profiles")
Cc: stable@dpdk.org
Signed-off-by: Rosen Xu <rosen.xu@intel.com>
Reviewed-by: Jasvinder Singh <jasvinder.singh@intel.com>

sched: fix memory leak on init failure

In some case, we may create sched port dynamically,
if err when creating so memory will leak.

Fixes: de3cfa2c9823 ("sched: initial import")
Cc: stable@dpdk.org
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>

mbuf: implement generic format for sched field

This patch implements the changes proposed in the deprecation
notes [1][2].

librte_mbuf changes:
The mbuf->hash.sched field is updated to support generic
definition in line with the ethdev traffic manager and meter APIs.
The new generic format contains: queue ID, traffic class, color.

Added public APIs to set and get these new fields to and from mbuf.

librte_sched changes:
In addtion, following API functions of the sched library have
been modified with an additional parameter of type struct
rte_sched_port to accommodate the changes made to mbuf sched field.
(i)rte_sched_port_pkt_write()
(ii) rte_sched_port_pkt_read_tree_path()

librte_pipeline, qos_sched UT, qos_sched app are updated
to make use of new changes.

Also mbuf->hash.txadapter has been added for eventdev txq,
rte_event_eth_tx_adapter_txq_set and rte_event_eth_tx_adapter_txq_get()
are updated to use mbuf->hash.txadapter.txq.

doc:
Release notes updated.
Removed deprecation notice for mbuf->hash.sched and sched API.

[1] http://mails.dpdk.org/archives/dev/2018-February/090651.html
[2] https://mails.dpdk.org/archives/dev/2018-November/119051.html

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Tested-by: Nikhil Rao <nikhil.rao@intel.com>
Reviewed-by: Nikhil Rao <nikhil.rao@intel.com>

meter: unify packet color definition

Added new rte_color definition in librte_meter to
consolidate color definition which is currently replicated
in various places such as rte_meter.h, rte_tm.h and rte_mtr.h

Created aliases for rte_tm_color, rte_mtr_color and rte_meter_color
to use new rte_color values.

The definitions of rte_tm_color, rte_mtr_color and rte_meter_color
will be deprecated in future.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>

telemetry: fix using ports of different types

Different NIC ports can have different numbers of xstats on them, which
means that we can't just use the xstats list from the first port registered
in the telemetry library. Instead, we need to check the type of each port -
by checking its ops structure pointer - and register each port type once
with the metrics lib.

Fixes: fdbdb3f9ce46 ("telemetry: add initial connection socket")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Kevin Laatz <kevin.laatz@intel.com>

net/sfc: pass HW Tx queue index on creation

Software indexes are PMD internal and should not be passed outside.
Right now SW and HW indexes of the Tx queue match, so it is just
a cosmetic fix.

Fixes: dbdc82416b72 ("net/sfc: factor out libefx-based Tx datapath")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>

net/failsafe: add default Tx mbuf fast free capability

This capability is reported when supported by the current emitting
sub-device. Failsafe PMD itself does not exercise fast free logic.

Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Gaetan Rivet <gaetan.rivet@6wind.com>

app/testpmd: add boundary check in flow command

There is need to add boundary for input number from commandline,
If it beyond the definition, code will return error.

Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/avf/base: fix comment referencing internal data

DCR is Intel internal information, no need to be in public code.
Related commit:
1a0833efde70 ("net/i40e/base: fix comment referencing internal data")

Fixes: e5b2a9e957e7 ("net/avf/base: add base code for avf PMD")
Cc: stable@dpdk.org
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/tap: fix possible uninitialized variable access

Fixes: 7c25284e30c2 ("net/tap: add netlink back-end for flow API")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/af_packet: fix setting MTU decrements sockaddr twice

When setting the MTU, eth_dev_mtu_set() is called to validate the
provided MTU. As part of that, it calculates the useful area to store
data and compares it against the MTU, to guarantee that there's enough
space to store the data. It calculates that as:
    "tp_frame_size - TPACKET2_HDRLEN - sizeof(struct sockaddr_ll)"

However, the TPACKET2_HDRLEN macro already increments sizeof(struct
sockaddr_ll) internally, meaning the useuful area of data above will
have sizeof(struct sockaddr_ll) decremented twice.

Instead, the useful area of data should be calculated as:
    "tp_frame_size - TPACKET2_HDRLEN"

This makes sure that there's enough useful area to fit the provided MTU
after excluding tpacket2_hdr and sockaddr_ll.

Fixes: cc68ac4847bc ("net/af_packet: support MTU change")
Cc: stable@dpdk.org
Signed-off-by: Tiago Lam <tiago.lam@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/virtio: improve batching in mergeable path

This patch improves both descriptors dequeue and refill,
by using the same batching strategy as done in in-order path.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Tested-by: Jens Freimann <jfreimann@redhat.com>
Reviewed-by: Jens Freimann <jfreimann@redhat.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>

net/virtio: add non-mergeable support to in-order path

This patch adds support for in-order path when meargeable buffers
feature hasn't been negotiated.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>

net/virtio: inline refill and offload helpers

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Jens Freimann <jfreimann@redhat.com>

vhost: batch used descs chains write-back with packed ring

Instead of writing back descriptors chains in order, let's
write the first chain flags last in order to improve batching.

Also, move the write barrier in logging cache sync, so that it
is done only when logging is enabled. It means there is now
one more barrier for split ring when logging is enabled.

With Kernel's pktgen benchmark, ~3% performance gain is measured.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>

net/virtio: avoid double accounting of bytes

Accounting of bytes was moved to a common function, so at the moment we do
it twice. This patches fixes it for sending packets with packed virtqueues.

Signed-off-by: Jens Freimann <jfreimann@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

vhost: remove useless prefetch for packed ring descriptor

This prefetch does not show any performance improvement.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Tiwei Bie <tiwei.bie@intel.com>

vhost: prefetch descriptor after the read barrier

This patch moves the prefetch after the available index
is read to avoid prefetching a descriptor not available yet.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Tiwei Bie <tiwei.bie@intel.com>

vhost: enforce desc flags and content read ordering

A read barrier is required to ensure that the ordering between
descriptor's flags and content reads is enforced.

1. read flags = desc->flags
if (flags & AVAIL_BIT)
2.   read desc->id

There is a control dependency between steps 1 and step 2.
2 could be speculatively executed before 1, which could result
in 'id' to not be updated yet.

Fixes: 2f3225a7d69b ("vhost: add vector filling support for packed ring")
Cc: stable@dpdk.org
Reported-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Tiwei Bie <tiwei.bie@intel.com>

vhost: enforce avail index and desc read ordering

A read barrier is required to ensure the ordering between
available index and the descriptor reads is enforced.

1. read avail_head = avail->idx
2. read cur_idx = last_avail_idx
if (cur_idx != avail_head) {
    3. read idx = avail->ring[cur_idx]
    4. read desc[idx]
}

There is a control dependency between step 1 and steps 3 & 4,
3 could be speculatively executed before 1, which could result
in 'idx' to not being updated yet.

Fixes: 4796ad63ba1f ("examples/vhost: import userspace vhost application")
Cc: stable@dpdk.org
Reported-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Tiwei Bie <tiwei.bie@intel.com>

net/vhost: fix double free of MAC address

The common data freeing has been moved to rte_eth_dev_release_port(),
so freeing mac_addrs like this in eth_dev_close() is unnecessary and
will cause double free.

Fixes: e16adf08e54d ("ethdev: free all common data when releasing port")
Cc: stable@dpdk.org
Signed-off-by: Hideyuki Yamashita <yamashita.hideyuki@po.ntt-tx.co.jp>
Signed-off-by: Yasufumi Ogawa <ogawa.yasufumi@lab.ntt.co.jp>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>

net/cxgbe: implement reset hit counters for offloaded flows

Implement logic to reset hit counters for offloaded flows.

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>

net/failsafe: remove unnecessary debug messages

The failsafe driver spews messages every time it is polling
making debugging other things more difficult. Just remove it.

Example:
net_failsafe: Hot-plug mutex was locked by thread 140182066431744 by the hot-plug alarm
net_failsafe: Hot-plug mutex was locked by thread 140182066431744 by the hot-plug alarm
...

Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/tap: add buffer overflow checks before checksum

The checksum calculation APIs take only the packet headers pointers as
parameters, so they assume that the lengths reported in those headers
are correct. However, a malicious packet could claim to be far larger
than it is, so we need to check the header lengths in the driver before
calling the checksum API.

A better fix would be to allow the lengths to be passed into the API
function, but that would be an API break, so fixing in TAP driver for
now.

Fixes: 8ae3023387e9 ("net/tap: add Rx/Tx checksum offload support")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>

net: fix underflow for checksum of invalid IPv4 packets

If we receive a packet with an invalid IP header, where the total packet
length is reported as less than the IP header length, we would end up
getting an underflow in the length subtraction.

This could cause us to checksum e.g. 4GB of data in the case where the
result of the subtraction was -1.

We fix this by having the function return 0 - an invalid sum - when
the length is less than the header length.

Fixes: af75078fece3 ("first public release")
Fixes: 6006818cfb26 ("net: new checksum functions")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>

net/mlx5: fix Multi-Packet RQ mempool free

When MPRQ mempool is freed, the pointer stored in priv structure must be
reset to null. Otherwise, the mempool can be freed again if the port is
restarted.

Fixes: 7d6bf6b866b8 ("net/mlx5: add Multi-Packet Rx support")
Cc: stable@dpdk.org
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>

doc: update queue number per VF for i40e

Updated the doc and release notes on the support for requesting more
queues.

Signed-off-by: Zhirun Yan <zhirun.yan@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>

net/i40e: support PF respond VF request more queues

This patch respond the VIRTCHNL_OP_REQUEST_QUEUES msg from VF, and
process to allocated more queues for the requested VF. If successful,
PF will notify VF to reset. If unsuccessful, PF will send message to
inform VF.

Signed-off-by: Zhirun Yan <zhirun.yan@intel.com>
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>

net/i40e: support VF request more queues

Before this patch, VF gets a default number of queues from the PF.
This patch enables VF to request a different number. When VF configures
more queues, it will send VIRTCHNL_OP_REQUEST_QUEUES to PF to request
more queues, if success, PF will reset the VF.

User can run "port stop all", "port config port_id rxq/txq queue_num"
and "port start all" to reconfigure queue number.

Signed-off-by: Zhirun Yan <zhirun.yan@intel.com>
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>

net/i40e: fix config name in comment

This patch fixes I40E RxTx module to use the proper config setting,
CONFIG_RTE_LIBRTE_I40E_INC_VECTOR.

Fixes: 9ed94e5bb04e ("i40e: add vector Rx")
Cc: stable@dpdk.org
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>

net/cxgbe: fix other misc build issues for Windows

Fix following build errors reported by Intel C++ compiler in Windows
build.

C:\> t4_hw.c(5105): warning #147: declaration is incompatible with
"int t4_bar2_sge_qregs(struct adapter *, unsigned int, unsigned int,
u64={uint64_t={unsigned __int64}} *, unsigned int *)"
(declared at line 524 of "..\..\..\..\drivers\net\cxgbe\base\common.h")
    int t4_bar2_sge_qregs(struct adapter *adapter, unsigned int qid,
        ^

C:\> cxgbe_filter.c(42): error : expected an expression
        n_user_filters = mult_frac(adap->tids.nftids,
                         ^

C:\> sge.c(400): error : expression must be a pointer to a complete
object type
                  (uint16_t)(RTE_PTR_ALIGN((char *)mbuf->buf_addr +
                             ^

Build Environment:
1. Target OS: Microsoft Windows Server 2016
2. Compiler: Intel C++ Compiler from Intel Parallel Studio XE 2019 [1]
3. Development Tools:
   3.1 Microsoft Visual Studio 2017 Professional
   3.2 Windows Software Development Kit (SDK) v10.0.17763
   3.3 Windows Driver Kit (WDK) v10.0.17763

[1] https://software.intel.com/en-us/parallel-studio-xe

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>

net/cxgbe: define symbols only when not available

Define symbols only when they are not available.

This fixes following types of issues reported by Intel C++ compiler
in Windows build.

C:\> cxgbe_compat.h(154): warning #47: incompatible redefinition of
macro "min"
        #define min(a, b) RTE_MIN(a, b)
                ^

C:\> t4_hw.c(338): warning #266: function "bzero" declared implicitly
            bzero(p, 0, size);
            ^

C:\> t4_hw.c(5337): warning #266: function "htonl" declared implicitly
            rvc.op_to_viid = htonl(V_FW_CMD_OP(FW_RSS_VI_CONFIG_CMD) |
                             ^

C:\> sge.c(361): error : expected an expression
        struct sge_eth_rxq *rxq = container_of(q, struct sge_eth_rxq, fl);
                                  ^

C:\> sge.c(1350): error : identifier "caddr_t" is undefined
  static void inline_tx_mbuf(const struct sge_txq *q, caddr_t from,
                                                      ^
[...]

Build Environment:
1. Target OS: Microsoft Windows Server 2016
2. Compiler: Intel C++ Compiler from Intel Parallel Studio XE 2019 [1]
3. Development Tools:
   3.1 Microsoft Visual Studio 2017 Professional
   3.2 Windows Software Development Kit (SDK) v10.0.17763
   3.3 Windows Driver Kit (WDK) v10.0.17763

[1] https://software.intel.com/en-us/parallel-studio-xe

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>

net/cxgbe: fix macros related to logs for Windows

Replace "args..." with "fmt, ..." and directly use __VA_ARGS__.

This fixes following errors reported by Intel C++ compiler in Windows
build.

C:\> cxgbe_compat.h(28): error : expected a ")"
        #define dev_printf(level, fmt, args...) \
                                       ^

C:\> cxgbe_compat.h(31): error : expected a ")"
        #define dev_err(x, args...) dev_printf(ERR, args)
                           ^
[...]

Build Environment:
1. Target OS: Microsoft Windows Server 2016
2. Compiler: Intel C++ Compiler from Intel Parallel Studio XE 2019 [1]
3. Development Tools:
   3.1 Microsoft Visual Studio 2017 Professional
   3.2 Windows Software Development Kit (SDK) v10.0.17763
   3.3 Windows Driver Kit (WDK) v10.0.17763

[1] https://software.intel.com/en-us/parallel-studio-xe

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>

net/cxgbe: use relative paths for includes

The Intel C++ compiler is not able to locate the header files without
relative paths in Windows build. Following errors are seen for these
header files.

C:\> base\common.h(9): error : cannot open source file "cxgbe_compat.h"
        #include "cxgbe_compat.h"
                                 ^
[...]

Fix by explicitly stating header file location using relative paths.
Also, remove automatically including header files for Linux, to keep
it consistent across both OS.

Build Environment:
1. Target OS: Microsoft Windows Server 2016
2. Compiler: Intel C++ Compiler from Intel Parallel Studio XE 2019 [1]
3. Development Tools:
   3.1 Microsoft Visual Studio 2017 Professional
   3.2 Windows Software Development Kit (SDK) v10.0.17763
   3.3 Windows Driver Kit (WDK) v10.0.17763

[1] https://software.intel.com/en-us/parallel-studio-xe

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>

net/ice: support descriptor ops

Add below ops,
rx_descriptor_status
tx_descriptor_status

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support statistics

Add below ops,
stats_get
stats_reset
xstats_get
xstats_get_names
xstats_reset

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Jeff Guo <jia.guo@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support advance Rx/Tx

Add Rx functions, scatter and bulk.
Add Tx function, simple.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support EEPROM information getting

Add below ops,
get_eeprom_length
get_eeprom

Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support FW version getting

Add ops fw_version_get.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support Rx queue interruption

Add below ops,
rx_queue_intr_enable
rx_queue_intr_disable

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support RSS

Add below ops,
reta_update
reta_query
rss_hash_update
rss_hash_conf_get

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support VLAN ops

Add below ops,
ice_vlan_filter_set
ice_vlan_offload_set
ice_vlan_tpid_set
ice_vlan_pvid_set

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support MAC ops

Add below ops,
mac_addr_set
mac_addr_add
mac_addr_remove

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support MTU setting

Add ops mtu_set.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice: support basic Rx/Tx

Add basic RX & TX support.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Qi Zhang <qi.z.zhang@intel.com>

net/ice: support packet type getting

Add ops dev_supported_ptypes_get.

Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Qi Zhang <qi.z.zhang@intel.com>

net/ice: support queue information getting

Add below ops,
rxq_info_get
txq_info_get
rx_queue_count

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Qi Zhang <qi.z.zhang@intel.com>

net/ice: support link update

Add ops link_update.
LSC interrupt is also enabled in this patch.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Qi Zhang <qi.z.zhang@intel.com>

net/ice: support getting device information

Add ops dev_infos_get.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Qi Zhang <qi.z.zhang@intel.com>

net/ice: support device and queue ops

Normally when starting/stopping the device the queue
should be started and stopped. Support them both in
this patch.

Below ops are added,
dev_configure
dev_start
dev_stop
dev_close
dev_reset
rx_queue_start
rx_queue_stop
tx_queue_start
tx_queue_stop
rx_queue_setup
rx_queue_release
tx_queue_setup
tx_queue_release

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Qi Zhang <qi.z.zhang@intel.com>

net/ice: support device initialization

Update the documents too.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Qi Zhang <qi.z.zhang@intel.com>

net/ice/base: add OS specific implementation

Add some MACRO defination and small functions which
are specific for DPDK.
Add readme too.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add structures for Rx/Tx queues

Add the structures that define how the Rx/Tx queues
are used.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add protocol structures and defines

Add the structures and defines that define what
protocols the NIC can handle.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add various headers

Add various headers that define status codes and
basic defines for use in the code.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add common functions

Add code that multiple other features use.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add code to work with the NVM

Add code to read/write/query the NVM image.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add virtual switch code

Add code to handle the virtual switch within the NIC.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add basic Tx scheduler

Add code for the basic TX scheduler.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add control queue information

Add the structures for the control queues.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add device IDs for Intel E800 Series NICs

Add all the device IDs that represent the NIC.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add sideband queue info

Add the commands, error codes, and structures
for the sideband queue.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add admin queue structures and commands

Add the commands, error codes, and structures for
the admin queue.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add basic structures

Add the structures required by the NIC.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: add registers for Intel E800 Series NIC

Add the registers that comprise the Intel(R) 800
Series NIC. There is no functionality in this patch.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/cxgbe: skip parsing match items with no spec

Skip parsing pattern match items that have no spec. This fixes NULL
dereference when accessing their non-existent spec.

Fixes: ee61f5113b17 ("net/cxgbe: parse and validate flows")
Cc: stable@dpdk.org
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>

net/cxgbe: fix overlapping regions in TID table

Location of filter TID table should be after active TID table memory,
and not from the beginning of TID table memory. This fixes memory
corruption due to overlapping regions.

Fixes: 3a381a4116ed ("net/cxgbe: query firmware for HASH filter resources")
Cc: stable@dpdk.org
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>

net/ena: update version to 2.0.0

The ENAv2 is introducing many new features, mainly the LLQ feature
(Low Latency Queue) which allows the device to process packets faster
and as a result, the latency is noticeably lower.

The second major feature is configurable depth of hw queues where Rx
and Tx can be reconfigured independently and maximum depth of Rx queue
is 8k.

The release also includes many bug fixes and minor new features, like
improved statistics counters and extended statistics.

Driver is still compatible with ENAv1 device.

Signed-off-by: Michal Krawczyk <mk@semihalf.com>

docs: update ENA PMD documentation and features

Previously, the features list was indicating unsupported ENA PMD
features and were missing few ones, that were actually supported.

The features file was updated, so it is now reflecting current driver
state.

The documentation was updated with the more actual example and features,
especially ones which are ENA and not listed in the features file.

Signed-off-by: Michal Krawczyk <mk@semihalf.com>

net/ena: update completion queue after cleanup

After Rx or Tx cleanup update completion queue head by calling
ena_com_update_dev_comp_head().

Fixes: 1daff5260ff8 ("net/ena: use unmasked head and tail")
Cc: stable@dpdk.org
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: add new way of getting Rx drops

The Rx drops cannot be acquired using the older API. Now, it must be
read in keep alive message.

Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: expose extended stats

ENA PMD is having it's own custom statistics counters. They are exposed
to the application by using the xstats DPDK API.

The deprecated and unused statistics are removed, together with old API.

Signed-off-by: Solganik Alexander <sashas@lightbitslabs.com>
Signed-off-by: Michal Krawczyk <mk@semihalf.com>

net/ena: add per-queue software counters stats

Those counters provide information regards sent/received bytes and
packets per queue.

Signed-off-by: Solganik Alexander <sashas@lightbitslabs.com>
Signed-off-by: Michal Krawczyk <mk@semihalf.com>

net/ena: fix cleanup for out of order packets

When wrong req_id is detected some previous mbufs could be used for
receiving different segments of received packets. In such cases chained
mbufs will be twice returned to pool.

To prevent it chained mbuf is now freed just after error detection.

To simplify cleaning, pointers taken for Rx ring are set to NULL.

As after ena_rx_queue_release_bufs and ena_tx_queue_release_bufs queues
are not used updating of next_to_clean pointer is not necessary.

Fixes: c2034976673d ("net/ena: add Rx out of order completion")
Cc: stable@dpdk.org
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: fix invalid reference to variable in union

Use empty_rx_reqs instead of empty_tx_reqs.
As those two variables are part of union this not cause
any failure, but for consistency should be changed.

Fixes: c2034976673d ("net/ena: add Rx out of order completion")
Cc: stable@dpdk.org
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: add supported RSS offloads types

The PMD was not passing RSS offloads values although it was supporting
the RSS. To allow application to probe the PMD for RSS support, the
missing information was added.

Fixes: 1173fca25af9 ("ena: add polling-mode driver")
Cc: stable@dpdk.org
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: adjust new line in log messages

Only PMD_*_LOG is adding new line character to the log message.
All printouts were adjusted for consistency.

Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: do not reconfigure queues on reset

Reset function should return the port to initial state, in which no Tx
and Rx queues are setup. Then application should reconfigure the queues.

According to DPDK documentation the rte_eth_dev_reset() itself is a
generic function which only does some hardware reset operations through
calling dev_unint() and dev_init().

ena_com_dev_reset which perform NIC registers reset should be called
during stop.

Fixes: 2081d5e2e92d ("net/ena: add reset routine")
Cc: stable@dpdk.org
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: destroy queues if start failed

If start function fails, previously created queues have to be removed.

ena_queue_restart_all() and ena_queue_restart() are renamed to
ena_queue_start_all() and ena_queue_start().

ena_free_io_queues_all() is renamed to ena_queue_stop_all().

Fixes: df238f84c0a2 ("net/ena: recreate HW IO rings on start and stop")
Cc: stable@dpdk.org
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: call additional doorbells if needed

Before sending next packet, check if calling doorbell is needed.

Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: increase maximum Rx ring size

Some of ENA devices supports 8k Rx rings. Maximum supported size is
received upon device initialization.
As ENA_DEFAULT_RING_SIZE_RX macro is upper limit, it needs to be
adjusted.

Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: support LLQv2

LLQ (Low Latency Queue) is the feature that allows pushing header
directly to the device through PCI before even DMA is triggered.
It reduces latency, because device can start preparing packet before
payload is sent through DMA.

Signed-off-by: Michal Krawczyk <mk@semihalf.com>

net/ena: skip packet with wrong request id

When invalid req_id is received, the reset should be handled by the
application, as it is indicating invalid rings state, so further Rx
is not making any sense.

Fixes: c2034976673d ("net/ena: add Rx out of order completion")
Cc: stable@dpdk.org
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: add HW queues depth setup

The device now allows driver to reconfigure Tx and Rx queues depth
independently. Moreover, maximum size for Tx and Rx can be different.
Those maximum values are received from the device.

After reset, previous ring configuration is restored.

If number of descriptor is set to RTE_ETH_DEV_FALLBACK_RX_RINGSIZE
or RTE_ETH_DEV_FALLBACK_TX_RINGSIZE, the maximum value is restored.

Remove checks, if provided number is not too big, as this is done in
generic functions (rte_eth_rx_queue_setup and rte_eth_tx_queue_setup).

Maximum number of segments is being set for Rx packets and provided to
ena_com_rx_pkt() for validation.

Unused definitions were removed.

Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: add reset reason in Rx error

Whenever the driver will receive too many descriptors from the device,
it should trigger the device reset with reset reason set to
ENA_REGS_RESET_TOO_MANY_RX_DESCS.

Fixes: 241da076b1f7 ("net/ena: adjust error checking and cleaning")
Cc: stable@dpdk.org
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena: pass number of CPUs to the host info structure

The new ena_com allows the number of CPUs to be passed to the device in
the host info structure.

Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena/base: update communication layer for the ENAv2

ena_com is the communication layer provided by the vendor and common to
all ENA drivers.
This patch updates it to version from 2018.09.26.

It adds support for ENAv2 device together with LLQ feature, adds
doorbell optimization and reconfiguration of HW queues depth
independently.

The driver was adjusted to the new changes in the HAL.

Signed-off-by: Rafal Kozik <rk@semihalf.com>
Signed-off-by: Michal Krawczyk <mk@semihalf.com>

app/testpmd: batch MAC swap for performance on x86

Do four packets macswap in same loop iterate to squeeze more
CPU cycles.

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>

app/testpmd: improve MAC swap performance for x86

The patch optimizes the mac swap operation by taking advantage
of SSE instructions, it only impacts x86 platform.

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>

app/testpmd: move MAC swap functions

Move macswap workload to dedicate function, so we can further enable
platform specific optimized version.

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>

doc: update ifc NIC document

Add the SW assisted VDPA live migration feature into NIC doc.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

net/ifc: support SW assisted VDPA live migration

In SW assisted live migration mode, driver will stop the device and
setup a mediated virtio ring to relay the communication between the
virtio driver and the VDPA device.

This data path intervention will allow SW to help on guest dirty page
logging for live migration.

This SW fallback is event driven relay thread, so when the network
throughput is low, this SW fallback will take little CPU resource, but
when the throughput goes up, the relay thread's CPU usage will goes up
accordingly.

User needs to take all the factors including CPU usage, guest perf
degradation, etc. into consideration when selecting the live migration
support mode.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

net/ifc: use vhost lib function for used ring logging

Vhost lib has already provided a helper for used ring logging, driver
could use it to reduce code.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

net/ifc: add LM mode parameter

This patch series enables a new method for live migration, i.e. software
assisted live migration. This patch provides a device argument for user
to choose the methold.

When "sw-live-migration=1", driver/device will do live migration with a
relay thread dealing with dirty page logging. Without this parameter,
device will do dirty page logging and there's no relay thread consuming
CPU resource.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

net/ifc: detect if VDPA mode is specified

If user wants the VF to be used in VDPA (vhost data path acceleration)
mode, then the user can add a "vdpa=1" parameter for the device.

So if driver does not find this option, it should quit and let the bus
continue the probe.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

net/ifc: store only registered device instance

If driver fails to register ifc VF device into vhost lib, then this
device should not be stored.

Fixes: a3f8150eac6d ("net/ifcvf: add ifcvf vDPA driver")
Cc: stable@dpdk.org
Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>