examples/l3fwd-vf: update

Signed-off-by: Intel

examples/l2fwd: update

Signed-off-by: Intel

examples/ipv4_multicast: update

Signed-off-by: Intel

examples/ipv4_frag: update

Signed-off-by: Intel

examples/qat: update

Signed-off-by: Intel

examples/qat: rename config files

Signed-off-by: Intel

examples: update link status checks

Signed-off-by: Intel

examples: remove l2fwd-vf

Signed-off-by: Intel

app/testpmd: various updates

Signed-off-by: Intel

app/testpmd: add priority flow control

Signed-off-by: Intel

app/testpmd: add dcb support

Signed-off-by: Intel

app/testpmd: add port configuration settings

Signed-off-by: Intel

app/testpmd: add --total-num-mbufs option

Signed-off-by: Intel

app/testpmd: fix check on mac address

Signed-off-by: Intel

app/testpmd: fix fdir ipv4 filter

Signed-off-by: Intel

app/testpmd: add fdir ipv6 support

Signed-off-by: Intel

app/testpmd: add vlan offload support

Signed-off-by: Intel

app/testpmd: add stats per queue

Signed-off-by: Intel

app/testpmd: fix refcnt in mbuf allocation

test-pmd txonly leaks mbuf from the pool.
The function tx_mbuf_alloc() does not change the refcnt
and the refcnt is 0 when it is first allocated.
However, rte_pktmbuf_free_seg called by the driver's xmit code decrements
reference count to -1. So mbuf never goes back to the pool.
As a result, txonly can't send packets after it exhausts the mempool.

The function tx_mbuf_alloc() was getting mbuf directly from mempool and so
was bypassing mbuf API.
The dedicated function is rte_pktmbuf_alloc() but it is much slower because
it does unnecessary initializations in rte_pktmbuf_reset().
By using the internal API __rte_mbuf_raw_alloc(), refcnt is correctly handled
without adding too much overload.

Signed-off-by: Dongsu Han <dongsuh@cs.cmu.edu>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>

app: various tests update

Signed-off-by: Intel

app: cosmetic

Signed-off-by: Intel

app: update test pci device ids

Signed-off-by: Intel

app: add new tests on eal flags

Signed-off-by: Intel

app: add reentrancy tests

Signed-off-by: Intel

app: change route table for lpm

Signed-off-by: Intel

app: split performance tests

Signed-off-by: Intel

app: rework autotest.py

Signed-off-by: Intel

app: refactor dump_cfg

Signed-off-by: Intel

app: remove chkincs

Signed-off-by: Intel

ixgbe: whitespace changes

Signed-off-by: Intel

ixgbe: various updates

Signed-off-by: Intel

ixgbe: add ipv6 support to fdir

Signed-off-by: Intel

ixgbe: configure CRC stripping behaviour of PF

Signed-off-by: Intel

ixgbe: add an optimised "simple" TX function

Signed-off-by: Intel

ixgbe: DCB / flow control

Signed-off-by: Intel

ixgbe: add vlan offload support

Signed-off-by: Intel

ixgbe: allow rx bulk alloc

Signed-off-by: Intel

ixgbe: update RX/TX queue configuration

Signed-off-by: Intel

ixgbe: stats per queue

Signed-off-by: Intel

ixgbe: DCB in base driver

Signed-off-by: Intel

ixgbe: update base driver

Signed-off-by: Intel

ixgbe: minor changes

Signed-off-by: Intel

ixgbe: update copyright

Signed-off-by: Intel

e1000: whitespace changes

Signed-off-by: Intel

e1000: support EM devices (also known as e1000/e1000e)

Signed-off-by: Intel

e1000: update RX/TX queue configuration

Queues are allocated by rte_ether.

Signed-off-by: Intel

e1000: add vlan offload support

Signed-off-by: Intel

e1000: configure CRC stripping behaviour of PF

Signed-off-by: Intel

e1000: minor changes

Signed-off-by: Intel

igb: various updates

Signed-off-by: Intel

igb: add VF support

Signed-off-by: Intel

igb: more supported devices

Signed-off-by: Intel

e1000: more NICs in base driver

Signed-off-by: Intel

e1000: update base driver

Signed-off-by: Intel

e1000: minor changes in base driver

Signed-off-by: Intel

e1000: update copyright

Signed-off-by: Intel

igb: rename pmd driver as e1000

Signed-off-by: Intel

ethdev: add VF statistics

Signed-off-by: Intel

ethdev: add priority flow control configuration

Signed-off-by: Intel

ethdev: add ipv6 support to fdir

Signed-off-by: Intel

ethdev: add vlan offload support

Signed-off-by: Intel

ethdev: add RX/TX queues configuration

Signed-off-by: Intel

ethdev: add DCB support

Signed-off-by: Intel

ethdev: add stats per queue

Signed-off-by: Intel

ethdev: minor changes

Signed-off-by: Intel

ethdev: init all builtin drivers

Signed-off-by: Intel

pci: make it possible to keep devices bound to uio

Binding code has been refactored as well.

Signed-off-by: Intel

pci: fix uio driver dependency

Allow to run without igb_uio if probed devices don't require it.

The condition for exiting with an error is:
 - Tested driver is needed by a probed device
 AND
 - Tested driver needs igb_uio

Signed-off-by: Damien Millescamps <damien.millescamps@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>

pci: reference driver structure for each device

Add a driver reference (if available) to every PCI devices,
even when blacklisted. This information is made available in the global
device_list variable so users know which network devices are managed or
ignored.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Ivan Boule <ivan.boule@6wind.com>
Acked-by: Damien Millescamps <damien.millescamps@6wind.com>

igb_uio: allow to configure interrupt mode

Signed-off-by: Intel

igb_uio: various updates

Signed-off-by: Intel

igb_uio: GPL copyright

Signed-off-by: Intel

mbuf: group fields in vlan_macip union

Signed-off-by: Intel

mbuf: fix build when scatter gather is disabled

Signed-off-by: Intel

memory: stats for malloc

Signed-off-by: Intel

memory: malloc now supports multi process

Signed-off-by: Intel

memory: add numa-awareness to malloc

Signed-off-by: Intel

memory: fix for multi process support

Signed-off-by: Intel

memory: add --socket-mem option

On NUMA systems, --socket-mem makes it possible to select the node where
allocations will go.

Signed-off-by: Intel

mempool: cache optimisations

Signed-off-by: Intel

ring: add burst API

Signed-off-by: Intel

hash: fix check when creating fbk hash

Signed-off-by: Intel

hash: allocate memory on right numa socket

Signed-off-by: Intel

hash: don't use memzone for allocations

Signed-off-by: Intel

hash: remove comments

Signed-off-by: Intel

hash: use intrinsic

Signed-off-by: Intel

hash: select default hash by looking at SSE flags

Signed-off-by: Intel

lpm: rework lpm lookup + add bulk function

Signed-off-by: Intel

lpm: allocate memory on right numa socket

Signed-off-by: Intel

lpm: don't use memzone for allocations

Signed-off-by: Intel

lpm: minor changes

Signed-off-by: Intel

eal: rework cpu information discovery

Signed-off-by: Intel

eal: minor changes

Signed-off-by: Intel

eal: always check CPU support at runtime

Signed-off-by: Intel

eal: GPL copyright for PCI ids

Signed-off-by: Intel

cmdline: tests

Signed-off-by: Intel

cmdline: various updates

- Support of float number is removed.
- C++ compatibility is done
- More checks

Signed-off-by: Intel

log: fix multi process support

Signed-off-by: Intel

log: better fix on history buffer overflow

Signed-off-by: Intel

log: fix history buffer overflow

The '\0' is written outside the bounds of the log buffer, which can
result in memory corruption or display issues with log messages.

Use a new constant LOG_BUF_SIZE to store the effective size of the
buffer in struct log_history.

Signed-off-by: Dongsu Han <dongsuh@cs.cmu.edu>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>