dpdk.git
7 years agoscripts: move to buildtools
Thomas Monjalon [Thu, 15 Dec 2016 21:46:47 +0000 (22:46 +0100)]
scripts: move to buildtools

There is already a directory buildtools for pmdinfogen used by
the build system. The scripts used in makefiles are moved here.

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
7 years agodoc: add required python versions
John McNamara [Wed, 21 Dec 2016 15:03:49 +0000 (15:03 +0000)]
doc: add required python versions

Add a requirement to support both Python 2 and 3 to the
DPDK Python Coding Standards and Getting started Guide.

Signed-off-by: John McNamara <john.mcnamara@intel.com>
7 years agomake python scripts python2/3 compliant
John McNamara [Wed, 21 Dec 2016 15:03:48 +0000 (15:03 +0000)]
make python scripts python2/3 compliant

Make all the DPDK Python apps work with Python 2 or 3 to
allow them to work with whatever is the system default.

Signed-off-by: John McNamara <john.mcnamara@intel.com>
7 years agomake python scripts PEP8 compliant
John McNamara [Wed, 21 Dec 2016 15:03:47 +0000 (15:03 +0000)]
make python scripts PEP8 compliant

Make all DPDK python application compliant with the PEP8 standard
to allow for consistency checking of patches and to allow further
refactoring.

Signed-off-by: John McNamara <john.mcnamara@intel.com>
7 years agomk: disable icc warning 188
Ferruh Yigit [Tue, 3 Jan 2017 16:15:42 +0000 (16:15 +0000)]
mk: disable icc warning 188

error #188: enumerated type mixed with another type

This is get when an integer assigned to an enum variable.

Since this usage is common and causing many ICC compilation errors, and
other compilers accept this usage. Disabling the warning.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
7 years agoapp/testpmd: use Tx preparation in checksum engine
Tomasz Kulasek [Fri, 23 Dec 2016 18:40:54 +0000 (19:40 +0100)]
app/testpmd: use Tx preparation in checksum engine

Since all current drivers supports Tx preparation API, it is used
in csum forwarding engine by default for all drivers.

Adding additional step to the csum engine costs about 3-4% of performance
drop, on my setup with ixgbe driver. It's caused mostly by the need
of reaccessing and modification of packet data.

Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
7 years agonet/ena: add Tx preparation
Konstantin Ananyev [Fri, 23 Dec 2016 18:40:53 +0000 (19:40 +0100)]
net/ena: add Tx preparation

Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
7 years agonet/vmxnet3: add Tx preparation
Konstantin Ananyev [Fri, 23 Dec 2016 18:40:52 +0000 (19:40 +0100)]
net/vmxnet3: add Tx preparation

Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Yong Wang <yongwang@vmware.com>
7 years agonet/fm10k: add Tx preparation
Tomasz Kulasek [Fri, 23 Dec 2016 18:40:49 +0000 (19:40 +0100)]
net/fm10k: add Tx preparation

Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
7 years agonet/i40e: add Tx preparation
Tomasz Kulasek [Fri, 23 Dec 2016 18:40:50 +0000 (19:40 +0100)]
net/i40e: add Tx preparation

Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
7 years agonet/ixgbe: add Tx preparation
Tomasz Kulasek [Fri, 23 Dec 2016 18:40:51 +0000 (19:40 +0100)]
net/ixgbe: add Tx preparation

Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
7 years agonet/e1000: add Tx preparation
Tomasz Kulasek [Fri, 23 Dec 2016 18:40:48 +0000 (19:40 +0100)]
net/e1000: add Tx preparation

Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
7 years agoethdev: add Tx preparation
Tomasz Kulasek [Fri, 23 Dec 2016 18:40:47 +0000 (19:40 +0100)]
ethdev: add Tx preparation

Added API for `rte_eth_tx_prepare`

uint16_t rte_eth_tx_prepare(uint8_t port_id, uint16_t queue_id,
struct rte_mbuf **tx_pkts, uint16_t nb_pkts)

Added fields to the `struct rte_eth_desc_lim`:

uint16_t nb_seg_max;
/**< Max number of segments per whole packet. */

uint16_t nb_mtu_seg_max;
/**< Max number of segments per one MTU */

These fields can be used to create valid packets according to the
following rules:

 * For non-TSO packet, a single transmit packet may span up to
   "nb_mtu_seg_max" buffers.

 * For TSO packet the total number of data descriptors is "nb_seg_max",
   and each segment within the TSO may span up to "nb_mtu_seg_max".

Added functions:

int
rte_validate_tx_offload(struct rte_mbuf *m)

  to validate general requirements for tx offload set in mbuf of packet
  such a flag completness. In current implementation this function is
  called optionaly when RTE_LIBRTE_ETHDEV_DEBUG is enabled.

int rte_net_intel_cksum_prepare(struct rte_mbuf *m)

  to prepare pseudo header checksum for TSO and non-TSO tcp/udp packets
  before hardware tx checksum offload.
   - for non-TSO tcp/udp packets full pseudo-header checksum is
     counted and set.
   - for TSO the IP payload length is not included.

int
rte_net_intel_cksum_flags_prepare(struct rte_mbuf *m, uint64_t ol_flags)

  this function uses same logic as rte_net_intel_cksum_prepare, but
  allows application to choose which offloads should be taken into
  account, if full preparation is not required.

PERFORMANCE TESTS
-----------------

This feature was tested with modified csum engine from test-pmd.

The packet checksum preparation was moved from application to Tx
preparation step placed before burst.

We may expect some overhead costs caused by:
1) using additional callback before burst,
2) rescanning burst,
3) additional condition checking (packet validation),
4) worse optimization (e.g. packet data access, etc.)

We tested it using ixgbe Tx preparation implementation with some parts
disabled to have comparable information about the impact of different
parts of implementation.

IMPACT:

1) For unimplemented Tx preparation callback the performance impact is
   negligible,
2) For packet condition check without checksum modifications (nb_segs,
   available offloads, etc.) is 14626628/14252168 (~2.62% drop),
3) Full support in ixgbe driver (point 2 + packet checksum
   initialization) is 14060924/13588094 (~3.48% drop)

Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
7 years agoethdev: clarify extended statistics documentation
Olivier Matz [Fri, 23 Dec 2016 20:35:48 +0000 (21:35 +0100)]
ethdev: clarify extended statistics documentation

Reword the API documentation of ethdev xstats.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Remy Horton <remy.horton@intel.com>
7 years agoethdev: fix extended statistics name index
Olivier Matz [Fri, 16 Dec 2016 09:44:13 +0000 (10:44 +0100)]
ethdev: fix extended statistics name index

The function rte_eth_xstats_get() return an array of tuples (id,
value). The value is the statistic counter, while the id references a
name in the array returned by rte_eth_xstats_get_name().

Today, each 'id' returned by rte_eth_xstats_get() is equal to the index
in the returned array, making this value useless. It also prevents a
driver from having different indexes for names and value, like in the
example below:

  rte_eth_xstats_get_name() returns:
    0: "rx0_stat"
    1: "rx1_stat"
    2: ...
    7: "rx7_stat"
    8: "tx0_stat"
    9: "tx1_stat"
    ...
    15: "tx7_stat"

  rte_eth_xstats_get() returns:
    0: id=0, val=<stat>    ("rx0_stat")
    1: id=1, val=<stat>    ("rx1_stat")
    2: id=8, val=<stat>    ("tx0_stat")
    3: id=9, val=<stat>    ("tx1_stat")

This patch fixes the drivers to set the 'id' in their ethdev->xstats_get()
(except e1000 which was already doing it), and fixes ethdev by not setting
the 'id' field to the index of the table for pmd-specific stats: instead,
they should just be shifted by the max number of generic statistics.

Fixes: bd6aa172cf35 ("ethdev: fetch extended statistics with integer ids")

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Remy Horton <remy.horton@intel.com>
7 years agoethdev: decouple from PCI device
Jan Blunck [Fri, 23 Dec 2016 15:58:11 +0000 (16:58 +0100)]
ethdev: decouple from PCI device

This makes struct rte_eth_dev independent of struct rte_pci_device by
replacing it with a pointer to the generic struct rte_device.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
7 years agoethdev: move info filling of PCI into drivers
Jan Blunck [Fri, 23 Dec 2016 15:58:10 +0000 (16:58 +0100)]
ethdev: move info filling of PCI into drivers

Only the drivers itself can decide if it could fill PCI information fields
of dev_info.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
7 years agoethdev: decouple interrupt handling from PCI device
Jan Blunck [Fri, 23 Dec 2016 15:58:09 +0000 (16:58 +0100)]
ethdev: decouple interrupt handling from PCI device

The struct rte_intr_handle is an abstraction layer for different types of
interrupt mechanisms. It is embedded in the low-level device (e.g. PCI).
On allocation of a struct rte_eth_dev a reference to the intr_handle
should be stored for devices supporting interrupts.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
7 years agonet/vmxnet3: use driver name from ethdev
Jan Blunck [Fri, 23 Dec 2016 15:58:08 +0000 (16:58 +0100)]
net/vmxnet3: use driver name from ethdev

Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agonet/szedata2: localize handling of PCI resources
Jan Blunck [Fri, 23 Dec 2016 15:58:06 +0000 (16:58 +0100)]
net/szedata2: localize handling of PCI resources

This changes the driver to handle the PCI resource directly instead
of repeatedly going through eth_dev.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agonet/nfp: localize mapping of ethdev to PCI device
Jan Blunck [Fri, 23 Dec 2016 15:58:07 +0000 (16:58 +0100)]
net/nfp: localize mapping of ethdev to PCI device

This simplifies later changes to ethdev.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agonet/qede: localize mapping of ethdev to PCI device
Jan Blunck [Fri, 23 Dec 2016 15:58:05 +0000 (16:58 +0100)]
net/qede: localize mapping of ethdev to PCI device

This simplifies later changes to ethdev.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Harish Patil <harish.patil@qlogic.com>
7 years agonet/bnx2x: localize mapping of ethdev to PCI device
Jan Blunck [Fri, 23 Dec 2016 15:58:03 +0000 (16:58 +0100)]
net/bnx2x: localize mapping of ethdev to PCI device

Use device private information to minimize the places that assume eth_dev
contains pci_dev.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Harish Patil <harish.patil@qlogic.com>
7 years agonet/fm10k: localize mapping of ethdev to PCI device
Jan Blunck [Fri, 23 Dec 2016 15:58:04 +0000 (16:58 +0100)]
net/fm10k: localize mapping of ethdev to PCI device

This simplifies later changes to ethdev.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agonet/virtio: do not depend on PCI device of ethdev
Jan Blunck [Fri, 23 Dec 2016 15:58:02 +0000 (16:58 +0100)]
net/virtio: do not depend on PCI device of ethdev

We don't need to depend on rte_eth_dev->pci_dev to differentiate between
the virtio_user and the virtio_pci case. Instead we can use the private
virtio_hw struct to get that information.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
7 years agonet/virtio: add helper to get interrrupt handle
Jan Blunck [Fri, 23 Dec 2016 15:58:01 +0000 (16:58 +0100)]
net/virtio: add helper to get interrrupt handle

This adds a helper to get the rte_intr_handle from the virtio_hw. This is
safe to do since the usage of the helper is guarded by RTE_ETH_DEV_INTR_LSC
which is only set if we found a PCI device during initialization.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
7 years agonet/virtio: remove useless driver name copy
Jan Blunck [Fri, 23 Dec 2016 15:58:00 +0000 (16:58 +0100)]
net/virtio: remove useless driver name copy

This is overwritten in rte_eth_dev_info_get().

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Reviewed-by: David Marchand <david.marchand@6wind.com>
7 years agonet/bnxt: localize mapping of ethdev to PCI device
Stephen Hemminger [Fri, 23 Dec 2016 15:57:59 +0000 (16:57 +0100)]
net/bnxt: localize mapping of ethdev to PCI device

Use existing information about pci and interrupt handle to minimize
the number of places that assume eth_dev contains pci_device
information.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Jan Blunck <jblunck@infradead.org>
7 years agonet/i40e: localize mapping of ethdev to PCI device
Stephen Hemminger [Fri, 23 Dec 2016 15:57:58 +0000 (16:57 +0100)]
net/i40e: localize mapping of ethdev to PCI device

Simplify later changes to eth_dev.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Jan Blunck <jblunck@infradead.org>
7 years agonet/ixgbe: localize mapping of ethdev to PCI device
Stephen Hemminger [Fri, 23 Dec 2016 15:57:57 +0000 (16:57 +0100)]
net/ixgbe: localize mapping of ethdev to PCI device

Since later changes will change where PCI information is,
localize mapping in one macro.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Jan Blunck <jblunck@infradead.org>
7 years agonet/e1000: localize mapping of ethdev to PCI device
Stephen Hemminger [Fri, 23 Dec 2016 15:57:56 +0000 (16:57 +0100)]
net/e1000: localize mapping of ethdev to PCI device

Create one macro for where PCI device information is extracted
from ethernet device. Makes later changes easier to review, and test.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Jan Blunck <jblunck@infradead.org>
7 years agodrivers: remove useless reset of PCI device pointer
Stephen Hemminger [Fri, 23 Dec 2016 15:57:55 +0000 (16:57 +0100)]
drivers: remove useless reset of PCI device pointer

Since rte_eth_dev_info_get does memset() on dev_info before
calling device specific code, the explicit assignment of NULL
in all these virtual drivers has no effect.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Jan Blunck <jblunck@infradead.org>
7 years agoeal: make driver pointer const in device struct
Stephen Hemminger [Fri, 23 Dec 2016 15:57:54 +0000 (16:57 +0100)]
eal: make driver pointer const in device struct

The info in rte_device about driver is immutable and
shouldn't change.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
7 years agoeal: allow passing const interrupt handle
Jan Blunck [Fri, 23 Dec 2016 15:57:53 +0000 (16:57 +0100)]
eal: allow passing const interrupt handle

Both register/unregister and enable/disable don't necessarily require the
rte_intr_handle to be modifiable. Therefore lets constify it.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agoeal: define container_of macro
Jan Blunck [Fri, 23 Dec 2016 15:57:52 +0000 (16:57 +0100)]
eal: define container_of macro

This macro is based on Jan Viktorin's original patch but also checks the
type of the passed pointer against the type of the member.

Signed-off-by: Jan Viktorin <viktorin@rehivetech.com>
Signed-off-by: Shreyansh Jain <shreyansh.jain@nxp.com>
[jblunck@infradead.org: add type checking and __extension__]
Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agoapp/testpmd: fix flow command build on FreeBSD
Adrien Mazarguil [Fri, 23 Dec 2016 15:52:56 +0000 (16:52 +0100)]
app/testpmd: fix flow command build on FreeBSD

A missing include causes the following compilation errors:

 error: use of undeclared identifier 'AF_INET'
 error: use of undeclared identifier 'AF_INET6'

Fixes: ef6e38550f07 ("app/testpmd: add items ipv4/ipv6 to flow command")

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
7 years agodoc: update release notes for flow API
Adrien Mazarguil [Fri, 23 Dec 2016 14:00:05 +0000 (15:00 +0100)]
doc: update release notes for flow API

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
7 years agoapp/testpmd: add protocol fields to flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:42 +0000 (15:51 +0100)]
app/testpmd: add protocol fields to flow command

This commit exposes the following item fields through the flow command:

- VLAN priority code point, drop eligible indicator and VLAN identifier
  (all part of TCI).
- IPv4 type of service, time to live and protocol.
- IPv6 traffic class, flow label, next header and hop limit.
- SCTP tag and checksum.

Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
7 years agodoc: describe testpmd flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:41 +0000 (15:51 +0100)]
doc: describe testpmd flow command

Document syntax, interaction with rte_flow and provide usage examples.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agoapp/testpmd: add queue actions to flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:40 +0000 (15:51 +0100)]
app/testpmd: add queue actions to flow command

- QUEUE: assign packets to a given queue index.
- DUP: duplicate packets to a given queue index.
- RSS: spread packets among several queues.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add various actions to flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:39 +0000 (15:51 +0100)]
app/testpmd: add various actions to flow command

- MARK: attach 32 bit value to packets.
- FLAG: flag packets.
- DROP: drop packets.
- COUNT: enable counters for a rule.
- PF: redirect packets to physical device function.
- VF: redirect packets to virtual device function.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add L4 items to flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:38 +0000 (15:51 +0100)]
app/testpmd: add L4 items to flow command

Add the ability to match a few properties of common L4[.5] protocol
headers:

- ICMP: type and code.
- UDP: source and destination ports.
- TCP: source and destination ports.
- SCTP: source and destination ports.
- VXLAN: network identifier.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add items ipv4/ipv6 to flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:37 +0000 (15:51 +0100)]
app/testpmd: add items ipv4/ipv6 to flow command

Add the ability to match basic fields from IPv4 and IPv6 headers (source
and destination addresses only).

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add items eth/vlan to flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:36 +0000 (15:51 +0100)]
app/testpmd: add items eth/vlan to flow command

These pattern items match basic Ethernet headers (source, destination and
type) and related 802.1Q/ad VLAN headers.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add item raw to flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:35 +0000 (15:51 +0100)]
app/testpmd: add item raw to flow command

Matches arbitrary byte strings with properties:

- relative: look for pattern after the previous item.
- search: search pattern from offset (see also limit).
- offset: absolute or relative offset for pattern.
- limit: search area limit for start of pattern.
- length: pattern length.
- pattern: byte string to look for.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add various items to flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:34 +0000 (15:51 +0100)]
app/testpmd: add various items to flow command

- PF: match packets addressed to the physical function.
- VF: match packets addressed to a virtual function ID.
- PORT: device-specific physical port index to use.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add item any to flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:33 +0000 (15:51 +0100)]
app/testpmd: add item any to flow command

This pattern item matches any protocol in place of the current layer and
has two properties:

- min: minimum number of layers covered (0 or more).
- max: maximum number of layers covered (0 means infinity).

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: support flow bit-field
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:32 +0000 (15:51 +0100)]
app/testpmd: support flow bit-field

Several rte_flow structures expose bit-fields that cannot be set in a
generic fashion at byte level. Add bit-mask support to handle them.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add flow item spec prefix length
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:31 +0000 (15:51 +0100)]
app/testpmd: add flow item spec prefix length

Generating bit-masks from prefix lengths is often more convenient than
providing them entirely (e.g. to define IPv4 and IPv6 subnets).

This commit adds the "prefix" operator that assigns generated bit-masks to
any pattern item specification field.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add flow item spec handler
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:30 +0000 (15:51 +0100)]
app/testpmd: add flow item spec handler

Add parser code to fully set individual fields of pattern item
specification structures, using the following operators:

- fix: sets field and applies full bit-mask for perfect matching.
- spec: sets field without modifying its bit-mask.
- last: sets upper value of the spec => last range.
- mask: sets bit-mask affecting both spec and last from arbitrary value.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add flow query command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:29 +0000 (15:51 +0100)]
app/testpmd: add flow query command

Syntax:

 flow query {port_id} {rule_id} {action}

Query a specific action of an existing flow rule.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add flow validate/create commands
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:28 +0000 (15:51 +0100)]
app/testpmd: add flow validate/create commands

Syntax:

 flow (validate|create) {port_id}
    [group {group_id}] [priority {level}] [ingress] [egress]
    pattern {item} [/ {item} [...]] / end
    actions {action} [/ {action} [...]] / end

Either check the validity of a flow rule or create it. Any number of
pattern items and actions can be provided in any order. Completion is
available for convenience.

This commit only adds support for the most basic item and action types,
namely:

- END: terminates pattern items and actions lists.
- VOID: item/action filler, no operation.
- INVERT: inverted pattern matching, process packets that do not match.
- PASSTHRU: action that leaves packets up for additional processing by
  subsequent flow rules.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add flow destroy command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:27 +0000 (15:51 +0100)]
app/testpmd: add flow destroy command

Syntax:

 flow destroy {port_id} rule {rule_id} [...]

Destroy a given set of flow rules associated with a port.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add flow flush command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:26 +0000 (15:51 +0100)]
app/testpmd: add flow flush command

Syntax:

 flow flush {port_id}

Destroy all flow rules on a port.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add flow list command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:25 +0000 (15:51 +0100)]
app/testpmd: add flow list command

Syntax:

 flow list {port_id} [group {group_id}] [...]

List configured flow rules on a port. Output can optionally be limited to a
given set of group identifiers.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: support flow integer
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:24 +0000 (15:51 +0100)]
app/testpmd: support flow integer

Parse all integer types and handle conversion to network byte order in a
single function.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: add flow command
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:23 +0000 (15:51 +0100)]
app/testpmd: add flow command

Managing generic flow API functions from command line requires the use of
dynamic tokens for convenience as flow rules are not fixed and cannot be
defined statically.

This commit adds specific flexible parser code and object for a new "flow"
command in separate file.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoapp/testpmd: implement basic support for flow API
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:22 +0000 (15:51 +0100)]
app/testpmd: implement basic support for flow API

Add basic management functions for the generic flow API (validate, create,
destroy, flush, query and list). Flow rule objects and properties are
arranged in lists associated with each port.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agocmdline: add alignment constraint
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:21 +0000 (15:51 +0100)]
cmdline: add alignment constraint

This prevents sigbus errors on architectures that cannot handle unexpected
unaligned accesses to the output buffer.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agocmdline: support dynamic tokens
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:20 +0000 (15:51 +0100)]
cmdline: support dynamic tokens

Considering tokens must be hard-coded in a list part of the instruction
structure, context-dependent tokens cannot be expressed.

This commit adds support for building dynamic token lists through a
user-provided function, which is called when the static token list is empty
(a single NULL entry).

Because no structures are modified (existing fields are reused), this
commit has no impact on the current ABI.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agodoc: announce deprecation of legacy filter types
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:19 +0000 (15:51 +0100)]
doc: announce deprecation of legacy filter types

They are superseded by the generic flow API (rte_flow). Target release is
not defined yet.

Suggested-by: Kevin Traynor <ktraynor@redhat.com>
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agodoc: add generic flow API guide
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:18 +0000 (15:51 +0100)]
doc: add generic flow API guide

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agoethdev: introduce generic flow API
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:17 +0000 (15:51 +0100)]
ethdev: introduce generic flow API

This new API supersedes all the legacy filter types described in
rte_eth_ctrl.h. It is slightly higher level and as a result relies more on
PMDs to process and validate flow rules.

Benefits:

- A unified API is easier to program for, applications do not have to be
  written for a specific filter type which may or may not be supported by
  the underlying device.

- The behavior of a flow rule is the same regardless of the underlying
  device, applications do not need to be aware of hardware quirks.

- Extensible by design, API/ABI breakage should rarely occur if at all.

- Documentation is self-standing, no need to look up elsewhere.

Existing filter types will be deprecated and removed in the near future.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Olga Shern <olgas@mellanox.com>
7 years agoethdev: cleanup device ops struct whitespace
Ferruh Yigit [Thu, 22 Dec 2016 13:10:13 +0000 (13:10 +0000)]
ethdev: cleanup device ops struct whitespace

- Grouped related items using empty lines
- Aligned arguments to same column
- All item comments that doesn't fit same line are placed blow the item
  itself
- Moved some comments to same line if overall line < 100 chars

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
7 years agoethdev: remove invalid function from version map
Ferruh Yigit [Thu, 22 Dec 2016 12:00:16 +0000 (12:00 +0000)]
ethdev: remove invalid function from version map

Fixes: 9d41beed24b0 ("lib: provide initial versioning")

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
7 years agonet/bonding: reconfigure all slave queues every time
Ilya Maximets [Thu, 24 Nov 2016 11:26:49 +0000 (12:26 +0100)]
net/bonding: reconfigure all slave queues every time

This reverts commit 5b7bb2bda5519b7800f814df64d4e015282140e5.

It is necessary to reconfigure all queues every time because configuration
can be changed.

For example, if we're reconfiguring bonding device with new memory pool,
already configured queues will still use the old one. And if the old
mempool be freed, application likely will panic in attempt to use
freed mempool.

This happens when we use the bonding device with OVS 2.6 while MTU
reconfiguration:

PANIC in rte_mempool_get_ops():
assert "(ops_index >= 0) && (ops_index < RTE_MEMPOOL_MAX_OPS_IDX)" failed

Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Jan Blunck <jblunck@infradead.org>
7 years agonet/bonding: force reconfiguration of removed slaves
Jan Blunck [Thu, 24 Nov 2016 11:26:48 +0000 (12:26 +0100)]
net/bonding: force reconfiguration of removed slaves

After a slave interface is removed from a bond group it still has the
configuration of the bond interface. Lets enforce that the slave interface
is reconfigured after removal by resetting it.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agoethdev: add internal reset function
Jan Blunck [Thu, 24 Nov 2016 11:26:47 +0000 (12:26 +0100)]
ethdev: add internal reset function

This is a helper for DPDK internal users to force a reconfiguration of a
device.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agoethdev: free queue array after releasing all queues
Jan Blunck [Thu, 24 Nov 2016 11:26:46 +0000 (12:26 +0100)]
ethdev: free queue array after releasing all queues

If all queues are released lets also free up the dev->data->rx/tx_queues
to be able to properly reinitialize.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agoethdev: release queue before setting up
Jan Blunck [Thu, 24 Nov 2016 11:26:45 +0000 (12:26 +0100)]
ethdev: release queue before setting up

If a queue has been setup before lets release it before we setup.
Otherwise we might leak resources.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
7 years agoethdev: initialize more fields on allocation
Jan Blunck [Thu, 17 Nov 2016 17:16:12 +0000 (18:16 +0100)]
ethdev: initialize more fields on allocation

This moves the non-PCI related initialization of the link state interrupt
callback list and the setting of the default MTU to rte_eth_dev_allocate()
so that drivers only need to set non-default values.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
7 years agoethdev: clear data when allocating device
Jan Blunck [Thu, 17 Nov 2016 17:16:11 +0000 (18:16 +0100)]
ethdev: clear data when allocating device

Lets clear the eth_dev->data when allocating a new rte_eth_dev so that
drivers only need to set non-zero values.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
7 years agopci: remove unused unbind support
Stephen Hemminger [Wed, 7 Dec 2016 18:04:36 +0000 (10:04 -0800)]
pci: remove unused unbind support

No device driver sets the unbind flag in current public code base.
Therefore it is good time to remove the unused dead code.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
7 years agoeal: postpone vdev initialization
Jerin Jacob [Sun, 20 Nov 2016 08:00:50 +0000 (13:30 +0530)]
eal: postpone vdev initialization

Some platform like octeontx may use pci and
vdev based combined device to represent a logical
dpdk functional device.In such case, postponing the
vdev initialization after pci device
initialization will provide the better view of
the pci device resources in the system in
vdev's probe function, and it allows better
functional subsystem registration in vdev probe
function.

As a bonus, This patch fixes a bond device
initialization use case.

example command to reproduce the issue:
./testpmd -c 0x2  --vdev 'eth_bond0,mode=0,
slave=0000:02:00.0,slave=0000:03:00.0' --
--port-topology=chained

root cause:
In existing case(vdev initialization and then pci
initialization), creates three Ethernet ports with
following port ids
0 - Bond device
1 - PCI device 0
2 - PCI devive 1

Since testpmd, calls the configure/start on all the ports on
start up,it will translate to following illegal setup sequence

1)bond device configure/start
1.1) pci device0 stop/configure/start
1.2) pci device1 stop/configure/start
2)pci device 0 configure(illegal setup case,
as device in start state)

The fix changes the initialization sequence and
allow initialization in following valid setup order
1) pcie device 0 configure/start
2) pcie device 1 configure/start
3) bond device 2 configure/start
3.1) pcie device 0/stop/configure/start
3.2) pcie device 1/stop/configure/start

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
7 years agoeal: restrict cores auto detection
Jianfeng Tan [Fri, 2 Dec 2016 17:48:56 +0000 (17:48 +0000)]
eal: restrict cores auto detection

This patch uses pthread_getaffinity_np() to narrow down used
cores when none of below options is specified:
  * coremask (-c)
  * corelist (-l)
  * and coremap (--lcores)

The purpose of this patch is to leave out these core related options
when DPDK applications are deployed under container env, so that
users do not need decide the core related parameters when developing
applications. Instead, when applications are deployed in containers,
use cpu-set to constrain which cores can be used inside this container
instance. And DPDK application inside containers just rely on this
auto detect mechanism to start polling threads.

Note: previously, some users are using isolated CPUs, which could
be excluded by default. Please add commands like taskset to use
those cores.

Test example:
$ taskset 0xc0000 ./examples/helloworld/build/helloworld -m 1024

Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
7 years agoeal: clarify the argc and argv documentation
Aaron Conole [Thu, 8 Dec 2016 15:56:45 +0000 (10:56 -0500)]
eal: clarify the argc and argv documentation

It's been a source of confusion in the past, and even with this update
may continue to be a source of confusion.  However, the original
language seems to imply that the DPDK EAL will take ownership of the
array passed in.  Loosening the language up a bit might give a better
understanding for what is actually happening.

Signed-off-by: Aaron Conole <aconole@redhat.com>
7 years agodrivers: advertise kmod dependencies in pmdinfo
Olivier Matz [Thu, 15 Dec 2016 13:46:39 +0000 (14:46 +0100)]
drivers: advertise kmod dependencies in pmdinfo

Add a new macro RTE_PMD_REGISTER_KMOD_DEP() that allows a driver to
declare the list of kernel modules required to run properly.

Today, most PCI drivers require uio/vfio.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
7 years agoexamples/ethtool: fix driver information
Qiming Yang [Tue, 22 Nov 2016 01:41:24 +0000 (09:41 +0800)]
examples/ethtool: fix driver information

Function pcmd_drvinfo_callback uses struct info to get
the ethtool information of each port. Struct info will
store the information of previous port until this
information be updated. This patch fixes this issue.

Fixes: bda68ab9d1e7 ("examples/ethtool: add user-space ethtool sample application")

Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Acked-by: Remy Horton <remy.horton@intel.com>
7 years agoexamples/ethtool: fix querying non-PCI devices
Remy Horton [Wed, 30 Nov 2016 02:47:29 +0000 (10:47 +0800)]
examples/ethtool: fix querying non-PCI devices

Doing a device information query on a non-PCI device such as
vhost was resulting in the dereferencing of a NULL pointer
(the absent PCI data), causing a segmentation fault.

Fixes: bda68ab9d1e7 ("examples/ethtool: add user-space ethtool sample application")

Signed-off-by: Remy Horton <remy.horton@intel.com>
7 years agodoc: fix typo in testpmd guide
Rami Rosen [Thu, 15 Dec 2016 21:03:20 +0000 (23:03 +0200)]
doc: fix typo in testpmd guide

This patch fixes a trivial typo in testpmd application user guide.

Signed-off-by: Rami Rosen <rami.rosen@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: fix typo in sample app guide
Baruch Siach [Tue, 13 Dec 2016 13:40:26 +0000 (15:40 +0200)]
doc: fix typo in sample app guide

Fixes: bda68ab9d1e7 ("examples/ethtool: add user-space ethtool sample application")

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Acked-by: Remy Horton <remy.horton@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: fix source extract command in Linux guide
Baruch Siach [Tue, 13 Dec 2016 11:18:55 +0000 (13:18 +0200)]
doc: fix source extract command in Linux guide

DPDK source archives are .tar.xz or .tar.gz, not .zip. Use .tar.xz in the
instructions, since that is what the main download page links to.

Also, correct the archive file and directory name capitalization.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: fix required tools list layout in Linux guide
Baruch Siach [Mon, 19 Dec 2016 19:28:40 +0000 (21:28 +0200)]
doc: fix required tools list layout in Linux guide

The Python requirement should appear in the bullet list.

Also, indent the x32 note, since it is related to the previous bullet.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: fix mistakes in contribution guide
Yong Wang [Fri, 9 Dec 2016 11:52:29 +0000 (06:52 -0500)]
doc: fix mistakes in contribution guide

Signed-off-by: Yong Wang <wang.yong19@zte.com.cn>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: add details of sub-trees and maintainers
John McNamara [Fri, 16 Dec 2016 17:50:34 +0000 (17:50 +0000)]
doc: add details of sub-trees and maintainers

Add a new section to the Code Contributors Guide to outline
the roles of tree and component maintainers and how they
can be added and removed.

Signed-off-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: introduce PVP reference benchmark
Maxime Coquelin [Tue, 6 Dec 2016 12:24:40 +0000 (13:24 +0100)]
doc: introduce PVP reference benchmark

Having reference benchmarks is important in order to obtain
reproducible performance figures.

This patch describes required steps to configure a PVP setup
using testpmd in both host and guest.

Not relying on external vSwitch ease integration in a CI loop by
not being impacted by DPDK API changes.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: Kevin Traynor <ktraynor@redhat.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agomaintainers: update for qede PMD and bnx2x PMD
Rasesh Mody [Wed, 7 Dec 2016 19:22:19 +0000 (11:22 -0800)]
maintainers: update for qede PMD and bnx2x PMD

Following Cavium's acquisition of QLogic we need to update all the
qlogic PMD maintainer's entries to point to our new e-mail addresses.
Update driver's maintainers as they are no longer working for Cavium.

Thanks to Sony Chacko for his support and development of our various
dpdk drivers.

Signed-off-by: Rasesh Mody <rasesh.mody@cavium.com>
7 years agovdev: fix detaching with alias
Anatoly Burakov [Fri, 18 Nov 2016 16:32:49 +0000 (16:32 +0000)]
vdev: fix detaching with alias

Fixes: d63eed6b2dca ("eal: add driver name alias")

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
7 years agoethdev: fix port lookup if none
Anatoly Burakov [Sat, 19 Nov 2016 13:10:11 +0000 (13:10 +0000)]
ethdev: fix port lookup if none

Aside from avoiding doing useless work, this also fixes a segfault
when calling rte_eth_dev_get_port_by_name() whenever no devices
were found yet, and therefore rte_eth_dev_data wasn't yet allocated.

Fixes: 9c5b8d8b9feb ("ethdev: clean port id retrieval when attaching")

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
7 years agokni: avoid using lsb_release script
Robin Jarry [Fri, 25 Nov 2016 12:22:17 +0000 (13:22 +0100)]
kni: avoid using lsb_release script

The lsb_release script is part of an optional package which is not
always installed. On the other hand, /etc/lsb-release is always present
even on minimal Ubuntu installations.

    root@ubuntu1604:~# dpkg -S /etc/lsb-release
    base-files: /etc/lsb-release

Read the file if present and use the variables defined in it.

Signed-off-by: Robin Jarry <robin.jarry@6wind.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
7 years agoconfig: remove insecure warnings
Thomas Monjalon [Sun, 4 Dec 2016 22:08:45 +0000 (23:08 +0100)]
config: remove insecure warnings

There was an option CONFIG_RTE_INSECURE_FUNCTION_WARNING (disabled by
default), which prevents from using some libc functions:
sprintf, snprintf, vsnprintf, strcpy, strncpy, strcat, strncat, sscanf,
strtok, strsep and strlen.

It's all about using them at the right place with the right precautions.
However, it is neither really possible nor a good advice to disable them.

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
7 years agodoc: fix description of attach to indirect mbuf
Baruch Siach [Tue, 6 Dec 2016 08:25:25 +0000 (10:25 +0200)]
doc: fix description of attach to indirect mbuf

Since commit 4a96e8b652181 ("mbuf: allow to clone an indirect mbuf")
attaching to an indirect mbuf creates a clone of that mbuf. Update the mbuf
documentation to match.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: fix mistakes in prog guide
Baruch Siach [Tue, 29 Nov 2016 09:25:43 +0000 (11:25 +0200)]
doc: fix mistakes in prog guide

- Remove wrong document description:
The Programmer’s Guide intro is not the Release Notes.

- Fix heading of section which only deals with Tx queues configuration.

- Fix wrong verbatim text paragraphs:
Reduce the indentation of these paragraphs since they are not part of the
verbatim block.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: fix relative path of NIC table input file
John McNamara [Sun, 4 Dec 2016 16:47:37 +0000 (16:47 +0000)]
doc: fix relative path of NIC table input file

Fix relative path between sphinx conf.py file and Nic table file
to allow automatic build on ReadTheDocs.

Fixes: 9db3f52126fb ("doc: generate NIC overview table from ini files")

Signed-off-by: John McNamara <john.mcnamara@intel.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
7 years agodoc: add pdump library to API doxygen
Reshma Pattan [Thu, 1 Dec 2016 11:02:10 +0000 (11:02 +0000)]
doc: add pdump library to API doxygen

Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: remove repeated words in contribution guide
Yong Wang [Tue, 6 Dec 2016 12:30:49 +0000 (07:30 -0500)]
doc: remove repeated words in contribution guide

Signed-off-by: Yong Wang <wang.yong19@zte.com.cn>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agodoc: fix typos in code comments
Yong Wang [Tue, 29 Nov 2016 22:23:39 +0000 (17:23 -0500)]
doc: fix typos in code comments

Signed-off-by: Yong Wang <wang.yong19@zte.com.cn>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agomempool: fix API documentation
Olivier Matz [Tue, 22 Nov 2016 10:04:58 +0000 (11:04 +0100)]
mempool: fix API documentation

A previous commit changed the local_cache table into a
pointer, reducing the size of the rte_mempool structure.

Fix the API comment of rte_mempool_create() related to
this modification.

Fixes: 213af31e0960 ("mempool: reduce structure size if no cache needed")

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
7 years agomempool: remove a redundant word in comment
Wei Zhao [Mon, 5 Dec 2016 03:40:00 +0000 (11:40 +0800)]
mempool: remove a redundant word in comment

There is a redundant repetition word "for" in comment line of the
file rte_mempool.h after the definition of RTE_MEMPOOL_OPS_NAMESIZE.
The word "for" appears twice in line 359 and 360. One of them is
redundant, so delete it.

Fixes: 449c49b93a6b ("mempool: support handler operations")

Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
7 years agomempool: remove redundant socket id assignment
Wei Zhao [Mon, 5 Dec 2016 03:39:04 +0000 (11:39 +0800)]
mempool: remove redundant socket id assignment

There is a redundant repetition mempool socket_id assignment in the
file rte_mempool.c in function rte_mempool_create_empty. The
statement "mp->socket_id = socket_id;"appear twice in line 821
and 824. One of them is redundant, so delete it.

Fixes: 85226f9c526b ("mempool: introduce a function to create an empty pool")

Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>