raw/dpaa2_qdma: support route by port in DMA

RBP or route by ports can help in translating the DMA
address over the PCI. Thus adding the RBP support with
long and short formats

Signed-off-by: Minghuan Lian <minghuan.lian@nxp.com>
Signed-off-by: Sachin Saxena <sachin.saxena@nxp.com>
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>

examples/l2fwd-event: fix build on RHEL 7.6

CC l2fwd_event_generic.o
.../l2fwd_event_generic.c: In function
  ‘l2fwd_rx_tx_adapter_setup_generic’:
.../l2fwd_event_generic.c:203:3: error: missing initializer for field
  ‘impl_opaque’ of ‘struct <anonymous>’
  [-Werror=missing-field-initializers]
   }
   ^
In file included from .../l2fwd_event_generic.c:10:0:
.../include/rte_eventdev.h:1057:12: note: ‘impl_opaque’ declared here
    uint8_t impl_opaque;
            ^

  CC l2fwd_event_internal_port.o
.../l2fwd_event_internal_port.c: In function
  ‘l2fwd_rx_tx_adapter_setup_internal_port’:
.../l2fwd_event_internal_port.c:201:3: error: missing initializer for
  field ‘impl_opaque’ of ‘struct <anonymous>’
  [-Werror=missing-field-initializers]
   }
   ^
In file included from .../l2fwd_event_internal_port.c:10:0:
.../include/rte_eventdev.h:1057:12: note: ‘impl_opaque’ declared here
    uint8_t impl_opaque;
            ^

Fixes: 50f05aa6ed9a ("examples/l2fwd-event: setup Rx/Tx adapter")

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Kevin Traynor <ktraynor@redhat.com>

net/bnxt: remove commented out code

This commented out todo and code is old. Remove it.

Fixes: b7435d660a8c ("net/bnxt: add ntuple filtering support")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>

event/opdl: remove commented out code

Some variables are commented out. Remove them.

Fixes: d548ef513cd7 ("event/opdl: add unit tests")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Acked-by: Liang Ma <liang.j.ma@intel.com>

compress/octeontx: remove commented out code

This code is commented out. Remove it.

Fixes: 43e610bb8565 ("compress/octeontx: introduce octeontx zip PMD")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>

net/ipn3ke: remove commented out code

These struct members and variable were commented out. Remove them.

Fixes: c01c748e4ae6 ("net/ipn3ke: add new driver")
Fixes: c820468ac99c ("net/ipn3ke: support TM")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>

net/ipn3ke: remove useless if statement

Coverity complains that this statement is not needed as the goto
label is on the next line anyway. Remove the if statement.

653        ret = ipn3ke_cfg_parse_i40e_pf_ethdev(afu_name, pf_name);
   CID 337930 (#1 of 1): Identical code for different branches
   (IDENTICAL_BRANCHES)identical_branches: The same code is executed
   when the condition ret is true or false, because the code in the
   if-then branch and after the if statement is identical. Should
   the if statement be removed?
654        if (ret)
655                goto end;
   implicit_else: The code from the above if-then branch is identical
   to the code after the if statement.
656end:

Coverity issue: 337930
Fixes: c01c748e4ae6 ("net/ipn3ke: add new driver")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>

net/ipn3ke: fix TM commit check

Coverity is complaining about identical code regardless of which branch
of the if else is taken. Functionally it means an error will always be
returned if this if else is hit. Remove the else branch.

    CID 337928 (#1 of 1): Identical code for different branches
    (IDENTICAL_BRANCHES)identical_branches: The same code is executed
    regardless of whether n->level != IPN3KE_TM_NODE_LEVEL_COS ||
    n->n_children != 0U is true, because the 'then' and 'else' branches
    are identical. Should one of the branches be modified, or the entire
    'if' statement replaced?
1506  if (n->level != IPN3KE_TM_NODE_LEVEL_COS ||
1507          n->n_children != 0) {
1508          return -rte_tm_error_set(error,
1509                  EINVAL,
1510                  RTE_TM_ERROR_TYPE_UNSPECIFIED,
1511                  NULL,
1512                  rte_strerror(EINVAL));
    else_branch: The else branch, identical to the then branch.
1513  } else {
1514          return -rte_tm_error_set(error,
1515                  EINVAL,
1516                  RTE_TM_ERROR_TYPE_UNSPECIFIED,
1517                  NULL,
1518                  rte_strerror(EINVAL));
1519  }

Coverity issue: 337928
Fixes: c820468ac99c ("net/ipn3ke: support TM")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>

common/cpt: fix possible null dereference

Coverity complains that ctrl_flags is set to NULL at the start
of the function and it may not have been set before there is a
jump to fc_success and it is dereferenced.

Check for NULL before dereference.

312fc_success:
   CID 344983 (#1 of 1): Explicit null dereferenced
   (FORWARD_NULL)7. var_deref_op: Dereferencing null pointer ctrl_flags.
313        *ctrl_flags = rte_cpu_to_be_64(*ctrl_flags);

Coverity issue: 344983
Fixes: 6cc54096520d ("crypto/octeontx: add supported sessions")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>

net/pcap: fix argument checks

Previously rx/tx_queues were passed into eth_from_pcaps_common()
as ptrs and were checked for being NULL.

In commit da6ba28f0540 ("net/pcap: use a struct to pass user options")
that changed to pass in a ptr to a pmd_devargs_all which contains
the rx/tx_queues.

The parameter checking was not updated as part of that commit and
coverity caught that there was still a check if rx/tx_queues were
NULL, apparently after they had been dereferenced.

In fact as they are a members of the devargs_all struct, they will
not be NULL so remove those checks.

1231        struct pmd_devargs *rx_queues = &devargs_all->rx_queues;
1232        struct pmd_devargs *tx_queues = &devargs_all->tx_queues;
1233        const unsigned int nb_rx_queues = rx_queues->num_of_queue;
    deref_ptr: Directly dereferencing pointer tx_queues.
1234        const unsigned int nb_tx_queues = tx_queues->num_of_queue;
1235        unsigned int i;
1236
1237        /* do some parameter checking */
    CID 345004: Dereference before null check (REVERSE_INULL)
    [select issue]
1238        if (rx_queues == NULL && nb_rx_queues > 0)
1239                return -1;
    CID 345029 (#1 of 1): Dereference before null check (REVERSE_INULL)
    check_after_deref: Null-checking tx_queues suggests that it may be
    null, but it has already been dereferenced on all paths leading to
    the check.
1240        if (tx_queues == NULL && nb_tx_queues > 0)
1241                return -1;

Coverity issue: 345029
Coverity issue: 345044
Fixes: da6ba28f0540 ("net/pcap: use a struct to pass user options")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Acked-by: Cian Ferriter <cian.ferriter@intel.com>

cmdline: replace FreeBSD ifdef for IP address parsing

The constants like AF_INET are in sys/socket.h in FreeBSD.
The #ifdef macro __FreeBSD__ is replaced with RTE_EXEC_ENV_FREEBSD
in order to be consistent across DPDK files, and allow to grep
for EXEC_ENV among other benefits.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>

examples/l3fwd: fix IP reserved address range

The original patch used incorrect subnet range for testing.

Fixes: 37afe381bde4 ("examples/l3fwd: use reserved IP addresses")
Cc: stable@dpdk.org
Reported-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Reported-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>

net/qede: clean LTO warnings

During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized.  This patch silences these
warnings.

Exemplary compiler warning to suppress (with LTO enabled):
error: ‘transceiver_type’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
  switch (transceiver_type) {

Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>

net/ifc: clean LTO warnings

During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized.  This patch silences these
warnings.

Exemplary compiler warning to suppress (with LTO enabled):
error: ‘features’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
  if (RTE_VHOST_NEED_LOG(features)) {

Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>

net/i40e: clean LTO warnings

During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized.  This patch silences these
warnings.

Exemplary compiler warning to suppress (with LTO enabled):
error: ‘filter_idx’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
  PMD_DRV_LOG(INFO, "Added port %d with AQ command with index %d",

Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>

net/e1000: clean LTO warnings

During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized.  This patch silences these
warnings.

Exemplary compiler warning to suppress (with LTO enabled):
error: ‘link’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
  if (link) {

Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>

net/dpaa2: fix possible use of uninitialized vars

This patch fixes 'maybe-uninitialized' warnings reported by compiler
when using LTO.

Compiler warning pointing to this error (with LTO enabled):
error: ‘kg_cfg.extracts[0].masks[0].mask’ may be used uninitialized in
this function [-Werror=maybe-uninitialized]
    extr->masks[j].mask = cfg->extracts[i].masks[j].mask;

Fixes: 16bbc98a3e63 ("bus/fslmc: update MC to 10.3.x")
Cc: stable@dpdk.org
Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>

test: clean LTO warnings

During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized.  This patch silences these
warnings.

Exemplary compiler warning to suppress (with LTO enabled):
error: ‘stats.greatest_free_size’ may be used uninitialized in this
function [-Werror=maybe-uninitialized]
  return len - overhead;

Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>
Acked-by: Yipeng Wang <yipeng1.wang@intel.com>

event/octeontx2: clean LTO warnings

During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized.  This patch silences these
warnings.

Exemplary compiler warning to suppress (with LTO enabled):
error: ‘chunk’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
   bkt->current_chunk = (uintptr_t)chunk;

Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>

app/eventdev: clean LTO warnings

During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized.  This patch silences these
warnings.

Exemplary compiler warning to suppress (with LTO enabled):
error: ‘service_id’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
    ret = evt_service_setup(service_id);

Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>

eventdev: fix possible use of uninitialized var

Fix the logic for the case of event queue allowing all schedule types.

Compiler warning pointing to this error (with LTO enabled):
error: ‘sched_type’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
  if ((ret < 0 && ret != -EOVERFLOW) ||

Fixes: 6750b21bd6af ("eventdev: add default software timer adapter")
Cc: stable@dpdk.org
Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>

build: add option to enable LTO

This patch adds an option to enable link time optimization.  In addition
to LTO option itself (-flto) fat-lto-objects are being used.  This is
because during the build pmdinfogen scans the generated ELF objects to
find this_pmd_name* symbol in symbol table.  Without fat-lto-objects gcc
produces ELF only with extra symbols for internal use during linking.

Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>

lib: annotate versioned functions

Every implementation of a particular version of given symbol needs to be
marked in its declaration as such (using `__vsym` macro).  This patch
fixes this and also clarifies the documentation about that.

Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>

doc: fix description of versioning macros

This patch fixes documentation of versioning macros so that they are
aligned with their implementation (no underscore is added by macros).

Fixes: f1ef9794f9bd ("doc: add ABI guidelines")
Cc: stable@dpdk.org
Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>

port: fix build dependency

The port library should be built after eventdev library.

Fixes: 5d92c4e592c4 ("port: add eventdev port type")
Cc: stable@dpdk.org
Signed-off-by: Rahul R Shah <rahul.r.shah@intel.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>

test/atomic: add atomic 16/32/64-bit exchanges test

The test works by creating a token comprised of random data
and a CRC8 value, using the rte_atomicXX_exchange to exchange
the new token for a previously generated token, and then
verifying that the exchanged data is intact (i.e. the CRC8
is still correct for the data).

Signed-off-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@redhat.com>

test/rib: speed up rib6 autotests

Split slow part of rib6_autotest into rib6_slow_autotest

Fixes: abab6d2966dc ("test/rib: add IPv6 autotests")

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
Acked-by: Aaron Conole <aconole@redhat.com>

test/rib: speed up rib autotests

Split slow part of rib_autotest into rib_slow_autotest

Fixes: b35df4dd666e ("test/rib: add autotests")

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
Acked-by: Aaron Conole <aconole@redhat.com>

vfio: fix DMA mapping of external heaps

Currently, externally created heaps are supposed to be automatically
mapped for VFIO DMA by EAL, however they only do so if, at the time of
heap creation, VFIO is initialized and has at least one device
available. If no devices are available at the time of heap creation (or
if devices were available, but were since hot-unplugged, thus dropping
all VFIO container mappings), then VFIO mapping code would have skipped
over externally allocated heaps.

The fix is two-fold. First, we allow externally allocated memory
segments to be marked as "heap" segments. This allows us to distinguish
between external memory segments that were created via heap API, from
those that were created via rte_extmem_register() API.

Then, we fix the VFIO code to only skip non-heap external segments.
Also, since external heaps are not guaranteed to have valid IOVA
addresses, we will skip those which have invalid IOVA addresses as well.

Fixes: 0f526d674f8e ("malloc: separate creating memseg list and malloc heap")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Tested-by: Rajesh Ravi <rajesh.ravi@broadcom.com>
Acked-by: David Marchand <david.marchand@redhat.com>

vfio: remove deprecated DMA mapping functions

The rte_vfio_dma_map/unmap API's have been marked as deprecated in
release 19.05. Remove them.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: David Marchand <david.marchand@redhat.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>

vfio: fix DMA mapping with default container

When requesting DMA mapping to default container, we are meant to
supply the RTE_VFIO_DEFAULT_CONTAINER_FD value, however this is
not handled correctly by get_vfio_cfg_by_container_fd(), because
it only looks at actual fd values and does not check for this
special case.

Fix it to return default container if the fd requested is the
special RTE_VFIO_DEFAULT_CONTAINER_FD value.

Fixes: 4106d89a18f8 ("vfio: allow DMA map to the default container")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>

doc: add application usage guide for l2fwd-event

Add documentation for l2fwd-event example.
Update release notes.

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

examples/l2fwd-event: add graceful teardown

Add graceful teardown that addresses both event mode and poll mode.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

examples/l2fwd-event: add eventdev main loop

Add event dev main loop based on enabled l2fwd options and eventdev
capabilities.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

examples/l2fwd-event: setup service core

Setup service cores for eventdev and Rx/Tx adapter when they don't have
internal port capability.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

examples/l2fwd-event: setup Rx/Tx adapter

Add event eth Rx/Tx adapter setup for both generic and internal port
event device pipelines.

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

examples/l2fwd-event: setup event queue and port

Add event device queue and port setup based on event eth Tx adapter
capabilities.

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

examples/l2fwd-event: setup event device

Add event device device setup based on event eth Tx adapter
capabilities.

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

examples/l2fwd-event: add infra to split eventdev framework

Add infra to split eventdev framework based on event Tx adapter
capability.
If event Tx adapter has internal port capability then we use
`rte_event_eth_tx_adapter_enqueue` to transmitting packets else
we use a SINGLE_LINK event queue to enqueue packets to a service
core which is responsible for transmitting packets.

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

examples/l2fwd-event: add infra for eventdev

Add infra to select event device as a mode to process packets through
command line arguments. Also, allow the user to select the schedule type
to be RTE_SCHED_TYPE_ORDERED, RTE_SCHED_TYPE_ATOMIC or
RTE_SCHED_TYPE_PARALLEL.

Usage:

`--mode="eventdev"` or `--mode="poll"`
`--eventq-sched="ordered"`, `--eventq-sched="atomic"` or
`--event-sched=parallel`

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

examples/l2fwd-event: add default poll mode routines

Add the default l2fwd poll mode routines similar to examples/l2fwd.

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

event/octeontx2: remove hot spots by prefetching data

Recent profiling revealed few hotspots in octeontx2 event device
driver add prefetch hints to reduce stalls.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>

event/octeontx2: use WFE while waiting for head

Use WFE to save power while waiting for tag to become head.

SSO signals EVENTI to allow cores to exit from wfe when they
are waiting for specific operations in which one of them is
setting HEAD bit in GWS_TAG.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>

doc: bump supported SDK and firmware version for DPAA2

This patch bump the supported SDK and MC firmware version
to the latest.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>

config: remove redundant DPAA2 build

dpaa and dpaa2 config have evolved to be same. The same binary
can now work across the platforms. So, there is no need to maintain
two different build configs.
The dpaa config shall work for both generation of dpaa platforms.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>

net/bnx2x: fix meson build

Use kernel headers for __le* types to avoid potential conflicts
resulting in redefinition errors for some Linux build environments.
Add check for FreeBSD execution environments.

Without this fix, aarch64 builds can fail with error below:

In file included from ../drivers/net/bnx2x/bnx2x.h:22,
                 from ../drivers/net/bnx2x/bnx2x_ethdev.c:8:
../drivers/net/bnx2x/bnx2x_osal.h:27:17: error:
conflicting types for ‘uint64_t’
 #define __le64  uint64_t
               ^~~~~~~~
In file included from /usr/include/stdint.h:37,
                 from /usr/lib/gcc/aarch64-linux-gnu/8/include/stdint.h:9,
                 from ../lib/librte_eal/common/include/arch/arm/rte_byteorder.h:16,
                 from ../drivers/net/bnx2x/bnx2x.h:17,
                 from ../drivers/net/bnx2x/bnx2x_ethdev.c:8:
/usr/include/aarch64-linux-gnu/bits/stdint-uintn.h:27:20: note:
previous declaration of ‘uint64_t’ was here
 typedef __uint64_t uint64_t;
                    ^~~~~~~~

There is also one minor change mixed in this commit:
Some use of __FreeBSD__ are replaced by RTE_EXEC_ENV_FREEBSD.

Fixes: 38dff79ba736 ("net/bnx2x: update HSI")

Signed-off-by: Rasesh Mody <rmody@marvell.com>
Tested-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Tested-by: Gavin Hu <gavin.hu@arm.com>

test/mempool: fix false positive result

The ret variable, initialized to -1, is changed to 0 during the test,
making the test successful in some cases where it should return a
failure.

Fix this by always using the GOTO_ERR() macro that sets the ret
variable before doing the goto.

Fixes: 923ceaeac140 ("test/mempool: add unit test cases")

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>

mempool: use specific macro for object alignment

For consistency, RTE_MEMPOOL_ALIGN should be used in place of
RTE_CACHE_LINE_SIZE. They have the same value, because the only arch
that was defining a specific value for it has been removed from DPDK.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>

mempool: prevent objects from being across pages

When populating a mempool, ensure that objects are not located across
several pages, except if user did not request IOVA-contiguous objects.

Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>

mempool: introduce helpers for populate and required size

Introduce new functions that can used by mempool drivers to
calculate required memory size and to populate mempool.

For now, these helpers just replace the *_default() functions
without change. They will be enhanced in next commit.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>

mempool: introduce function to get mempool page size

In rte_mempool_populate_default(), we determine the page size,
which is needed for calc_size and allocation of memory.

Move this in a function and export it, it will be used in a next
commit.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>

mempool: remove optimistic IOVA-contiguous allocation

The previous commit reduced the amount of required memory when
populating the mempool with non IOVA-contiguous memory.

Since there is no big advantage to have a fully iova-contiguous mempool
if it is not explicitly asked, remove this code, it simplifies the
populate function.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>

mempool: reduce wasted space on populate

The size returned by rte_mempool_op_calc_mem_size_default() is aligned
to the specified page size. Therefore, with big pages, the returned size
can be much more that what we really need to populate the mempool.

For instance, populating a mempool that requires 1.1GB of memory with
1GB hugepages can result in allocating 2GB of memory.

This problem is hidden most of the time due to the allocation method of
rte_mempool_populate_default(): when try_iova_contig_mempool=true, it
first tries to allocate an iova contiguous area, without the alignment
constraint. If it fails, it fallbacks to an aligned allocation that does
not require to be iova-contiguous. This can also fallback into several
smaller aligned allocations.

This commit changes rte_mempool_op_calc_mem_size_default() to relax the
alignment constraint to a cache line and to return a smaller size.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>

mempool: allow populating with unaligned virtual area

rte_mempool_populate_virt() currently requires that both addr
and length are page-aligned.

Remove this unneeded constraint which can be annoying with big
hugepages (ex: 1GB).

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Nipun Gupta <nipun.gupta@nxp.com>

bus/fslmc: sanitize device name parsing for clarity

remove unnecessary dup and free

Signed-off-by: Shreyansh Jain <shreyansh.jain@nxp.com>
Reviewed-by: Nipun Gupta <nipun.gupta@nxp.com>

bus/dpaa: move QBMAN global init to bus

DPAA SEC shall be able to work independent of DPAA ETH
driver.
This patch moves qbman init to bus, so that any driver
can use them even when no eth resources are present
or none of the eth devices are probed.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>

mempool/dpaa2: report error on endless loop in mbuf release

When BMAN is not able to accept more buffers, it could be that
there are no FBPR's (internal mem provided to bman) left.
Report error in such condition.

Fixes: 5dc43d22b5ad ("mempool/dpaa2: add hardware offloaded mempool")
Cc: stable@dpdk.org
Signed-off-by: Radu Bulie <radu-andrei.bulie@nxp.com>
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>

raw/dpaa2_cmdif: add retry and timeout in packet enqueue API

This patch adds retry in the DPAA2 CMDIF packet enqueue API

Fixes: 53c71586c789 ("raw/dpaa2_cmdif: support enqueue/dequeue operations")
Cc: stable@dpdk.org
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>

raw/dpaa2_qdma: add retry and timeout in packet enqueue API

This patch adds the logic in the DPAA2 QDMA packet enqueue API

Fixes: 4d9a3f2a0159 ("raw/dpaa2_qdma: support RBP mode")
Cc: stable@dpdk.org
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>

net/dpaa2: add retry and timeout in packet enqueue API

In the packet transmit, if the QBMAN is not able to process the
packets, the Tx function loops infinitely to send the packet out.
This patch changes the logic retry for some time (count) and then
return.

Fixes: cd9935cec873 ("net/dpaa2: enable Rx and Tx operations")
Fixes: 16c4a3c46ab7 ("bus/fslmc: add enqueue response read in qbman")
Cc: stable@dpdk.org
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Signed-off-by: Radu Bulie <radu-andrei.bulie@nxp.com>

common/dpaax: fallback to check separate memory node for VM

In Virtual Machine the memory node in the device tree is at
'/proc/device-tree/memory/reg' which is separate from the memory
node path on the host. This patch enables check on both the paths.

Fixes: 2f3d633aa593 ("common/dpaax: add library for PA/VA translation table")
Cc: stable@dpdk.org
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>

net/dpaa: fix LS1043 alignment check

On LS1043, we are good to check 128 byte alignment of offset to
transmit out the packet

Fixes: f8c7a17a48c9 ("net/dpaa: support Tx scatter gather for non-DPAA buffer")
Cc: stable@dpdk.org
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>

net/dpaa: fix supported RSS types

This patch fixes the supported RSS types on DPAA platform

Fixes: 15aa2a1b02e6 ("net/dpaa: update RSS offload types")
Cc: stable@dpdk.org
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>

test/fib: add IPv6 performance autotests

Performance IPv6 tests for the FIB library.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

test/fib: add performance autotests

Performance tests for the new FIB library.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

test/fib: add IPv6 autotests

Functional IPv6 tests for the FIB library.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

test/fib: add autotests

Functional tests for the new FIB library.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

fib: add dataplane algorithm for IPv6

Add fib implementation for ipv6 using modified DIR24_8 algorithm.
Implementation is similar to current LPM6 implementation but has
few enhancements:
faster control plane operations
more bits for userdata in table entries
configurable userdata size

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

fib: add DIR24-8 dataplane algorithm

Add fib implementation for DIR24_8 algorithm for IPv4.
Implementation is similar to current LPM implementation but has
few enhancements:
faster control plane operations
more bits for userdata in table entries
configurable userdata size

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

fib: support IPv6

Add FIB library support for IPv6.
It implements a dataplane structures and algorithms designed for
fast IPv6 longest prefix match.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

fib: add FIB library

Add FIB (Forwarding Information Base) library. This library
implements a dataplane structures and algorithms designed for
fast longest prefix match.
Internally it consists of two parts - RIB (control plane ops) and
implementation for the dataplane tasks.
Initial version provides two implementations for both IPv4 and IPv6:
dummy (uses RIB as a dataplane) and DIR24_8 (same as current LPM)
Due to proposed design it allows to extend FIB with new algorithms
in future (for example DXR, poptrie, etc).

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

test/rib: add IPv6 autotests

Functional tests for RIB6.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

rib: support IPv6

Extend RIB library with IPv6 support.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

test/rib: add autotests

Functional tests for the new RIB library.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

rib: add RIB library

Add RIB (Routing Information Base) library. This library
implements an IPv4 routing table optimized for control plane
operations. It implements a control plane struct containing routes
in a tree and provides fast add/del operations for routes.
Also it allows to perform fast subtree traversals
(i.e. retrieve existing subroutes for a given prefix).
This structure will be used as a control plane helper structure
for FIB implementation. Also it might be used standalone in other
different places such as bitmaps for example.
Internal implementation is level compressed binary trie.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

crypto/dpaa2_sec: fix build with clang 3.4.2

Building with clang 3.4.2 caused the following error:

  CC dpaa2_sec_dpseci.o
In file included from /dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c:40:
In file included from /dpdk/drivers/common/dpaax/caamflib/desc/ipsec.h:11:
In file included from /dpdk/drivers/common/dpaax/caamflib/rta.h:11:
In file included from .../common/dpaax/caamflib/rta/sec_run_time_asm.h:14:
/dpdk/drivers/common/dpaax/compat.h:108:18:
error: redefinition of typedef 'dma_addr_t' is a C11 feature
[-Werror,-Wtypedef-redefinition]
typedef uint64_t        dma_addr_t;
                        ^
/dpdk/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c:37:18:
note: previous definition is here
typedef uint64_t        dma_addr_t;
                        ^

The dma_addr_t was defined in the common compat.h file, and does not
need to be redefined.

The meson build include directories list was updated to enable the
definition in the common compat.h file be used.

Fixes: 8c83f28cc8a4 ("common/dpaax: move OF library from DPAA bus")

Signed-off-by: Ciara Power <ciara.power@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>

config: disable IEEE1588 on DPAA2

This patch disables the 1588 timer support by default on DPAA2
platform due to the performance impact.
By default it is disabled in config/common_base, so remove the entry
from DPAA2.

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>

bus/fslmc: fix resource leak

1 issue caught by coverity (issue 344967).
Leaked_storage: Variable sep going out of scope leaks the
storage it points to.

When 'sep' is not null and sep_exist is 0, 'sep' is freed
before going out of scope of the function irrespective of
'addr' exists or not.

Coverity issue: 344967
Fixes: e67a61614d0b ("bus/fslmc: support device iteration")
Cc: stable@dpdk.org
Signed-off-by: Agalya Babu RadhaKrishnan <agalyax.babu.radhakrishnan@intel.com>
Acked-by: Sachin Saxena <sachin.saxena@nxp.com>

doc: fix description of links to EAL options pages

Documentation includes separate pages of EAL command-line options for
Linux and for FreeBSD.
Links to these pages use the same text 'EAL parameters', so it is not
clear which link to use for which environment.

This patch adds the text '(Linux)' and '(FreeBSD)' where relevant, to
clearly identify the links.

Fixes: 3ee567cfec37 ("doc: document all EAL parameters in one place")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>

raw/ioat: support xstats reset

The rawdev xstats_reset function is now enabled.  It is called when the
ioat autotest completes, to reset all xstat values after they have been
modified during testing.

Signed-off-by: Ciara Power <ciara.power@intel.com>
Reviewed-by: Bruce Richardson <bruce.richardson@intel.com>

version: 19.11-rc1

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

rename private header files

Some of the internal header files have 'rte_' prefix
and some don't.
Remove 'rte_' prefix from all internal header files.

Suggested-by: Thomas Monjalon <thomas@monjalon.net>
Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>

usertools: fix telemetry client with python 3

When running the dpdk-telemetry-client.py with python 3, we get the
following syntax errors:

  File "usertools/dpdk-telemetry-client.py", line 70
      print "\nResponse: \n", str(data)
                           ^
  SyntaxError: invalid syntax

  File "usertools/dpdk-telemetry-client.py", line 93
      print "\nResponse: \n", str(data)
                           ^
  SyntaxError: invalid syntax

  File "usertools/dpdk-telemetry-client.py", line 111
      file_path = sys.argv[1]
                            ^
  TabError: inconsistent use of tabs and spaces in indentation

Import print_function from __future__ and add parentheses where missing.
Also, use spaces for indentation everywhere.

Fixes: d1b94da4a4e0 ("usertools: add client script for telemetry")
Fixes: 53f293c9a783 ("usertools: replace unsafe input function")
Fixes: 4080e46c8078 ("telemetry: support global metrics")
Cc: stable@dpdk.org
Signed-off-by: Robin Jarry <robin.jarry@6wind.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>

usertools: fix pmdinfo with python 3 and pyelftools>=0.24

Running dpdk-pmdinfo.py on Ubuntu 18.04 (bionic) with python 3 and
pyelftools installed produces no output but no error is reported
neither:

  ~$ python3 usertools/dpdk-pmdinfo.py -r build/app/testpmd
  ~$ echo $?
  0

While with python 2, it works:

  ~# python2 usertools/dpdk-pmdinfo.py -r build/app/testpmd
  {"pci_ids": [], "name": "dpio"}
  {"pci_ids": [], "name": "dpbp"}
  {"pci_ids": [], "name": "dpaa2_qdma"}
  .....

On Ubuntu 18.04, pyelftools is version 0.24. The change log of
pyelftools v0.24 says:

 - Symbol/section names are strings internally now, not bytestrings
   (this may affect API usage in Python 3) (#76).

We cannot guess which version of pyelftools is actually being used. The
elftools.__version__ symbol is not consistent with each distro's package
version. For example, on Ubuntu 16.04 (xenial), the .deb package version
is '0.23-2' but elftools.__version__ contains '0.25'. This is certainly
due to partial backports.

To have a more consistent behaviour of this script across all versions
of python, add the unicode_literals future import so that literal
strings are now always "unicode".

Add 2 utility functions to force a string into bytes or bytes into an
unicode string.

Force pyelftools return values to unicode strings (will do nothing with
recent version of pyelftools).

If elffile.get_section_by_name returns None with a unicode section name,
try with the same one encoded as bytes.

Also, replace all open() calls by io.open() which behaves like the
builtin open in python 3. The only non-binary opened file is
/usr/share/hwdata/pci.ids which is UTF-8 encoded text. Explicitly
specify that encoding.

Link: https://github.com/eliben/pyelftools/blob/v0.24/CHANGES#L7
Link: https://github.com/eliben/pyelftools/commit/108eaea9e75a8b5a
Fixes: 54ca545dce4b ("make python scripts python2/3 compliant")
Cc: stable@dpdk.org
Signed-off-by: Robin Jarry <robin.jarry@6wind.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>

examples/vm_power: send capabilities request from guest

Send request to power manager for core id provided
by user to get related capabilities.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

examples/vm_power: send CPU capabilities on VM request

Send capabilities for requested cores.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

power: add packet type for capabilities

Add new packet type and commands for capabilities query.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

examples/vm_power: add command to query CPU frequency

Add command and related logic to query CPU frequencies
either for specified CPU or all cores.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

examples/vm_power: add mechanism to disable queries

Add new command which gives possibility to enable/disable queries
form VM guest.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

examples/vm_power: process CPU frequency query

On query received from VM guest send CPUs frequencies.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

power: extend guest channel for frequency query

Extend incoming packet reading API with new packet
type which carries CPU frequencies.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

examples/vm_power: fix build without i40e

channel_monitor.c was dependent on i40e driver being available.
This is only necessary for the TRAFFIC policy, so use #ifdef's
to not call these when i40e not available.

Fixes: f14791a8126e ("examples/vm_power_mgr: add policy to channels")
Cc: stable@dpdk.org
Signed-off-by: David Hunt <david.hunt@intel.com>
Tested-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>

examples/vm_power: send confirmation cmd to guest

Use new guest channel API to send confirmation
message for received power command.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

examples/vm_power: process incoming confirmation cmds

Extend vm_power_guest to check incoming confirmations
of messages previously sent to host.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

power: extend guest channel API for reading

Added new experimental API rte_power_guest_channel_receive_msg
which gives possibility to receive messages send to guest.

Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Tested-by: David Hunt <david.hunt@intel.com>
Acked-by: Lee Daly <lee.daly@intel.com>

power: fix socket indicator value

Currently 0 is being used for not connected slot indication.
This is not consistent with linux doc which identifies 0 as valid
(connected) slot, thus modification was done to change it.

Fixes: cd0d5547 ("power: vm communication channels in guest")
Cc: stable@dpdk.org
Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>

examples/l3fwd-power: fix Rx interrupt disabling

Interrupt will not be received when disabling RX interrupt without
synchronization mechanism sometimes which leads to wake up issue.
Add spinlock to fix it.

Fixes: b736d64787 ("examples/l3fwd-power: disable Rx interrupt when waking up")
Cc: stable@dpdk.org
Signed-off-by: Xiao Zhang <xiao.zhang@intel.com>
Reviewed-by: Marvin Liu <yong.liu@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>

examples/l3fwd: prefetch the content of the next packet

The cache-misses problem is very serious when the function
lpm_cb_parse_ptype is called to read the content of packets. That is
because the contents of packages previously stored in the cache are
overwritten by the following instructions or variables.
Thus the prefetch order can be used to prefetch the next packet into
the cache to avoid CPU spending too much time on it.

On Octeon TX platform with built-in NIC, 12% performance gain was
measured by running RFC2544 NDR test with l3fwd. Furthermore, the
cache-misses event of the function lpm_cb_parse_ptype was reduced by
20%, and the CPU task-clock of it dropped from 16.49% to 11.3%, based
on the forwarding test for one minute with the 64B packet.
On the dpaa2 platform, no performance improvement nor drop were seen
with this patch by running RFC2544 NDR test with l3fwd.
On the x86 platform, 15.7% performance gain was measured by running
RFC2544 NDR test with l3fwd.

Signed-off-by: Feifei Wang <feifei.wang@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>

net: remove ethernet packing and set two-byte alignment

The ether header does not need to be packed since that makes no sense for
structures with only bytes in them, but it should be aligned to a two-byte
boundary to simplify access to it from code. Other packed structures that
use this also need to be updated to take account of the change, either by
removing packing - where it is clearly unneeded - or by explicitly giving
those structures 2-byte alignment also.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>

doc: add IOAT sample app guide

Added guide for IOAT sample app usage and code description.

Signed-off-by: Marcin Baran <marcinx.baran@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>