dpdk.git
8 years agoixgbe: offload VxLAN and NVGRE Rx checksum on X550
Wenzhuo Lu [Thu, 10 Mar 2016 02:42:13 +0000 (10:42 +0800)]
ixgbe: offload VxLAN and NVGRE Rx checksum on X550

X550 will do VxLAN & NVGRE RX checksum off-load automatically.
This patch exposes the result of the checksum off-load.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
8 years agoixgbe: configure UDP tunnel port
Wenzhuo Lu [Thu, 10 Mar 2016 02:42:12 +0000 (10:42 +0800)]
ixgbe: configure UDP tunnel port

Add UDP tunnel port add/del support on ixgbe. Now only
support VxLAN port configuration.
Although according to the specification the VxLAN port has
a default value 4789, it can be changed. We support VxLAN
port configuration to meet the change.
Note, the default value of VxLAN port in ixgbe NICs is 0. So
please set it when using VxLAN off-load.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
8 years agoethdev: rename UDP tunnel port functions
Wenzhuo Lu [Thu, 10 Mar 2016 02:42:10 +0000 (10:42 +0800)]
ethdev: rename UDP tunnel port functions

The names of function for tunnel port configuration are not
accurate. They're tunnel_add/del, better change them to
tunnel_port_add/del.
The old functions are directly replaced because the API and ABI
compatibility of ethdev are already broken in 16.04.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
8 years agoapp/testpmd: add commands for E-tag operation
Wenzhuo Lu [Fri, 11 Mar 2016 01:10:11 +0000 (09:10 +0800)]
app/testpmd: add commands for E-tag operation

Add the CLIs to support the E-tag operation.
1, Offloading of E-tag insertion and stripping.
2, Forwarding the E-tag packets to pools based on the GRP and E-CID_base.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Shaopeng He <shaopeng.he@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Tested-by: Yong Liu <yong.liu@intel.com>
8 years agoapp/testpmd: add commands for L2 tunnel config
Wenzhuo Lu [Fri, 11 Mar 2016 01:10:10 +0000 (09:10 +0800)]
app/testpmd: add commands for L2 tunnel config

Add CLIs to config ether type of l2 tunnel, and to enable/disable
a type of l2 tunnel.
Now only e-tag tunnel is supported.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Shaopeng He <shaopeng.he@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Tested-by: Yong Liu <yong.liu@intel.com>
8 years agoixgbe: support L2 tunnel operations
Wenzhuo Lu [Fri, 11 Mar 2016 01:10:09 +0000 (09:10 +0800)]
ixgbe: support L2 tunnel operations

Add support of l2 tunnel configuration and operations.
1, Support modifying ether type of a type of l2 tunnel.
2, Support enabling and disabling the support of a type of l2 tunnel.
3, Support enabling/disabling l2 tunnel tag insertion/stripping.
4, Support enabling/disabling l2 tunnel packets forwarding.
5, Support adding/deleting forwarding rules for l2 tunnel packets.
Only support E-tag now.

Also update the release note.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Shaopeng He <shaopeng.he@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Tested-by: Yong Liu <yong.liu@intel.com>
8 years agoethdev: support L2 tunnel operations
Wenzhuo Lu [Fri, 11 Mar 2016 01:10:08 +0000 (09:10 +0800)]
ethdev: support L2 tunnel operations

Add functions to support l2 tunnel configuration and operations.
1, L2 tunnel ether type modification.
   It means modifying the ether type of a specific type of tunnel.
   So the packet with this ether type will be parsed as this type
   of tunnel.
2, Enabling/disabling l2 tunnel support.
   It means enabling/disabling the ability of parsing the specific
   type of tunnel. This ability should be enabled before we enable
   filtering, forwarding, offloading for this specific type of
   tunnel.
3, Insertion and stripping for l2 tunnel tag.
4, Forwarding the packets to a pool based on l2 tunnel tag.

Only support e-tag tunnel now.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Shaopeng He <shaopeng.he@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Tested-by: Yong Liu <yong.liu@intel.com>
8 years agoixgbe: select pool by MAC when using double VLAN
Wenzhuo Lu [Fri, 11 Mar 2016 01:10:07 +0000 (09:10 +0800)]
ixgbe: select pool by MAC when using double VLAN

On X550, as required by datasheet, E-tag packets are not expected
when double VLAN are used. So modify the register PFVTCTL after
enabling double VLAN to select pool by MAC but not MAC or E-tag.

An introduction of E-tag:
It's defined in IEEE802.1br. Please reference this website,
http://www.ieee802.org/1/pages/802.1br.html.

A brief description.
E-tag means external tag, and it's a kind of l2 tunnel. It means a
tag will be inserted in the l2 header. Like below,
   |31            24|23           16|15         8|7           0|
  0|                   Destination MAC address                 |
  4|     Dest MAC address(cont.)    |     Src MAC address      |
  8|                  Source MAC address(cont.)                |
 12| E-tag Etherenet type (0x893f)  |      E-tag header        |
 16|                    E-tag header(cont.)                    |
 20|   VLAN Ethertype(optional)     |   VLAN header(optional)  |
 24|         Original type          |         ......           |
...|                              ......                       |
The E-tag format is like below,
   |0                    15|16   18|19 |20                   31|
   |   Ethertype - 0x893f  | E-PCP |DEI|   Ingress E-CID_base  |

   |32  33|34 35|36      47|48         55    |56             63|
   |  RSV | GRP |E-CID_base|Ingress_E-CID_ext|    E-CID_ext    |

The Ingess_E-CID_ext and E-CID_ext are always zero for endpoints
and are effectively reserved.

The more details of E-tag is in IEEE 802.1BR. 802.1BR is used to
replace 802.1Qbh. 802.1BR is a standard for Bridge Port Extension.
It specifies the operation of Bridge Port Extenders, including
management, protocols, and algorithms. Bridge Port Extenders
operate in support of the MAC Service by Extended Bridges.
The E-tag is added to l2 header to identify the VM channel and
the virtual port.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Shaopeng He <shaopeng.he@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Tested-by: Yong Liu <yong.liu@intel.com>
8 years agoi40e: fix overflow
Helin Zhang [Fri, 11 Mar 2016 16:50:58 +0000 (00:50 +0800)]
i40e: fix overflow

The array 'ptype_table' was defined in depth of 'UINT8_MAX' which
is 255, while the querying index could be from 0 to 255. The issue
can be fixed with expanding the array to one more element.

Fixes: 9571ea028489 ("i40e: replace some offload flags with unified packet type")

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
8 years agoethdev: add vlan type when setting ether type
Helin Zhang [Fri, 11 Mar 2016 16:50:57 +0000 (00:50 +0800)]
ethdev: add vlan type when setting ether type

In order to set ether type of VLAN for single VLAN, inner
and outer VLAN, the VLAN type as an input parameter is added
to 'rte_eth_dev_set_vlan_ether_type()'.
In addition, corresponding changes in e1000, ixgbe and i40e
are also added.

It is an ABI break but ethdev library is already bumped for 16.04.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
8 years agoexamples: use buffered Tx
Tomasz Kulasek [Thu, 10 Mar 2016 17:19:35 +0000 (18:19 +0100)]
examples: use buffered Tx

The internal buffering of packets for TX in sample apps is no longer
needed, so this patchset also replaces this code with calls to the new
rte_eth_tx_buffer* APIs in:

* l2fwd-jobstats
* l2fwd-keepalive
* l2fwd
* l3fwd-acl
* l3fwd-power
* link_status_interrupt
* client_server_mp
* l2fwd_fork
* packet_ordering
* qos_meter

Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
8 years agoethdev: add buffered Tx
Tomasz Kulasek [Thu, 10 Mar 2016 17:19:34 +0000 (18:19 +0100)]
ethdev: add buffered Tx

Many sample apps include internal buffering for single-packet-at-a-time
operation. Since this is such a common paradigm, this functionality is
better suited to being implemented in the ethdev API.

The new APIs in the ethdev library are:
* rte_eth_tx_buffer_init - initialize buffer
* rte_eth_tx_buffer - buffer up a single packet for future transmission
* rte_eth_tx_buffer_flush - flush any unsent buffered packets
* rte_eth_tx_buffer_set_err_callback - set up a callback to be called in
  case transmitting a buffered burst fails. By default, we just free the
  unsent packets.

As well as these, an additional reference callbacks are provided, which
frees the packets:

* rte_eth_tx_buffer_drop_callback - silently drop packets (default
  behavior)
* rte_eth_tx_buffer_count_callback - drop and update user-provided counter
  to track the number of dropped packets

Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
8 years agovhost: fix queue pair reallocation
Yuanhan Liu [Thu, 10 Mar 2016 04:20:01 +0000 (12:20 +0800)]
vhost: fix queue pair reallocation

vq is allocated on pairs, hence we should do pair reallocation
at numa_realloc() as well, otherwise an error like following
occurs while do numa reallocation:

    VHOST_CONFIG: reallocate vq from 0 to 1 node
    PANIC in rte_free():
    Fatal error: Invalid memory

The reason we don't catch it is because numa_realloc() will
not take effect when RTE_LIBRTE_VHOST_NUMA is not enabled,
which is the default case.

Fixes: e049ca6d10e0 ("vhost-user: prepare multiple queue setup")

Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Tested-by: Ciara Loftus <ciara.loftus@intel.com>
8 years agovhost: simplify numa reallocation
Yuanhan Liu [Thu, 10 Mar 2016 04:20:00 +0000 (12:20 +0800)]
vhost: simplify numa reallocation

We could first check if we need realloc vq or not, if so,
reallocate it. We then do similar to vhost dev realloc.

This could get rid of the tons of repeated "if (realloc_dev)"
and "if (realloc_vq)" statements, therefore, makes code
a bit more readable.

Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
8 years agovhost: get rid of linked list for devices
Yuanhan Liu [Thu, 10 Mar 2016 04:19:59 +0000 (12:19 +0800)]
vhost: get rid of linked list for devices

While we use a single linked list to maintain all devices, we could
use a static array to achieve the same goal, just like what we did
to maintain the eth devices with rte_eth_devices array. This could
simplifies the code a bit.

Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
8 years agovhost: fix build with kernel < 3.5
Yuanhan Liu [Tue, 8 Mar 2016 08:51:21 +0000 (16:51 +0800)]
vhost: fix build with kernel < 3.5

VIRTIO_NET_F_GUEST_ANNOUNCE is a new feature introduced since kernel
v3.5. For older kernels (or more precisely, old distributions), we
could simply define it manually, to fix the "macro not defined" error.

Fixes: d293dac8f30e ("vhost: claim support of guest announce")

Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
8 years agoexamples: fix build dependencies
Sergio Gonzalez Monroy [Fri, 11 Mar 2016 15:32:30 +0000 (15:32 +0000)]
examples: fix build dependencies

Building examples fails with CONFIG_RTE_LIBRTE_LPM=n

The error is caused by the new app ipsec-secgw that gets build
without checking for configuration dependencies.

Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample application")

The patch also reorders a couple entries to maintain alphabetic order.

Reported-by: Jan Viktorin <viktorin@rehivetech.com>
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
8 years agoexamples/l3fwd: fix ARM build
Maciej Czekaj [Thu, 10 Mar 2016 16:06:22 +0000 (17:06 +0100)]
examples/l3fwd: fix ARM build

Enable NEON support in exact match mode.
l3fwd example did not compile on ARM due to SSE2 instrincics used
in generic part.
Some instrinsins were used to initialize data structures and those were
replaced by ordinary structure initalization.
All SSE2 intrinsics used in forwarding, i.e. masking the IP/TCP header
are moved to single inline function and made arch-specific.

Signed-off-by: Maciej Czekaj <maciej.czekaj@caviumnetworks.com>
8 years agomaintainers: claim responsibility for arm64 files of hash
Jerin Jacob [Fri, 11 Mar 2016 03:52:59 +0000 (09:22 +0530)]
maintainers: claim responsibility for arm64 files of hash

Fixes: f123e3d2ca92 ("hash: replace libc memcmp with optimized functions for arm64")

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
8 years agolpm/arm: support NEON
Jerin Jacob [Fri, 11 Mar 2016 03:52:58 +0000 (09:22 +0530)]
lpm/arm: support NEON

Enabled CONFIG_RTE_LIBRTE_LPM, CONFIG_RTE_LIBRTE_TABLE,
CONFIG_RTE_LIBRTE_PIPELINE libraries for arm and arm64

TABLE, PIPELINE libraries were disabled due to LPM library dependency.

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Signed-off-by: Jianbo Liu <jianbo.liu@linaro.org>
8 years agolpm/x86: move SSE implementation to be architecture agnostic
Jerin Jacob [Fri, 11 Mar 2016 03:52:57 +0000 (09:22 +0530)]
lpm/x86: move SSE implementation to be architecture agnostic

-Used architecture agnostic xmm_t to represent 128 bit SIMD variable

-Introduced vect_* API abstraction in app/test to test rte_lpm_lookupx4
API in  architecture agnostic way

-Moved rte_lpm_lookupx4 SSE implementation to architecture specific
rte_lpm_sse.h file to accommodate new rte_lpm_lookupx4 implementation
for a different architecture.

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
8 years agomk: fix static build without crypto
Olivier Matz [Fri, 11 Mar 2016 13:29:40 +0000 (14:29 +0100)]
mk: fix static build without crypto

If the experimental CONFIG_RTE_LIBRTE_CRYPTODEV is disabled, build of
any crypto pmds will fail because of the missing dependency.  The commit
94288d645 fixes the issue when compiled with shared libraries but there
is still an issue at link time with static libs:

     LD test
     /usr/bin/ld: cannot find -lrte_pmd_null_crypto
     collect2: error: ld returned 1 exit status

Only add the -l linker flags related to crypto PMDs if CRYPTODEV is
enabled.

Fixes: 94288d645 ("mk: fix build without crypto")

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
8 years agoexamples/l2fwd-crypto: discover capabilities
Pablo de Lara [Fri, 11 Mar 2016 00:02:51 +0000 (00:02 +0000)]
examples/l2fwd-crypto: discover capabilities

Crypto devices now have information about
which crypto operations they are capable of provide.

This patch makes the app use this information,
removing all hardcoded values.

User now needs to create the virtual crypto devices
or bind the HW crypto devices, and the app will use
the ones capable of performing the crypto op specified
(user can select between HW/SW through command line)

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Min Cao <min.cao@intel.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
8 years agoexamples/l2fwd-crypto: add cipher/hash only cases
Pablo de Lara [Fri, 11 Mar 2016 00:02:50 +0000 (00:02 +0000)]
examples/l2fwd-crypto: add cipher/hash only cases

Added cipher-only, hash-only operation cases,
which will be supported in the future.
Also, only sets authentication and ciphering parameters
when needed.

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Min Cao <min.cao@intel.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
8 years agoexamples/l2fwd-crypto: parse AAD parameter
Pablo de Lara [Fri, 11 Mar 2016 00:02:49 +0000 (00:02 +0000)]
examples/l2fwd-crypto: parse AAD parameter

So far, L2fwd crypto app could parse cipher, auth keys
and IV, but not AAD (additional authentication data).

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Min Cao <min.cao@intel.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
8 years agoexamples/l2fwd-crypto: parse key parameters
Pablo de Lara [Fri, 11 Mar 2016 00:02:48 +0000 (00:02 +0000)]
examples/l2fwd-crypto: parse key parameters

Implement key parsing functionality, so user can provide
auth and cipher keys, plus IV, from the command line.

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Min Cao <min.cao@intel.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
8 years agoexamples/l2fwd-crypto: update auth algo list
Pablo de Lara [Fri, 11 Mar 2016 00:02:47 +0000 (00:02 +0000)]
examples/l2fwd-crypto: update auth algo list

Updated authentication algorithm list:
- Added MD5_HMAC and SHA384_HMAC
- Removed SHA1, SHA224, SHA256

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Min Cao <min.cao@intel.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
8 years agoexamples/l2fwd-crypto: clean up
Pablo de Lara [Fri, 11 Mar 2016 00:02:46 +0000 (00:02 +0000)]
examples/l2fwd-crypto: clean up

- Removed unnecessary blank lines
- Changed some variable types (longer)
- Removed commented code

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Min Cao <min.cao@intel.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
8 years agoexamples/ipsec-secgw: add IPsec sample application
Sergio Gonzalez Monroy [Fri, 11 Mar 2016 02:12:40 +0000 (02:12 +0000)]
examples/ipsec-secgw: add IPsec sample application

Sample app implementing an IPsec Security Geteway.
The main goal of this app is to show the use of cryptodev framework
in a "real world" application.

Currently only supported static IPv4 ESP IPsec tunnels for the following
algorithms:
- Cipher: AES-CBC, NULL
- Authentication: HMAC-SHA1, NULL

Not supported:
- SA auto negotiation (No IKE implementation)
- chained mbufs

Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agoaesni_mb: remove parameters from config file
Pablo de Lara [Thu, 10 Mar 2016 17:14:09 +0000 (17:14 +0000)]
aesni_mb: remove parameters from config file

Parse the device parameters from rte_eal_vdev_init,
instead of the config file, so user can change the parameters
at runtime.

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agocryptodev: add capabilities discovery
Declan Doherty [Fri, 11 Mar 2016 01:36:54 +0000 (01:36 +0000)]
cryptodev: add capabilities discovery

This patch add a mechanism for discovery of crypto device features and supported
crypto operations and algorithms. It also provides a method for a crypto PMD to
publish any data range limitations it may have for the operations and algorithms
it supports.

The parameter feature_flags added to rte_cryptodev struct is used to capture
features such as operations supported (symmetric crypto, operation chaining etc)
as well parameter such as whether the device is hardware accelerated or uses
SIMD instructions.

The capabilities parameter allows a PMD to define an array of supported operations
with any limitation which that implementation may have.

Finally the rte_cryptodev_info struct has been extended to allow retrieval of
these parameter using the existing rte_cryptodev_info_get() API.

Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
8 years agomk: fix build without crypto
Panu Matilainen [Fri, 11 Mar 2016 09:13:48 +0000 (11:13 +0200)]
mk: fix build without crypto

If the experimental CONFIG_RTE_LIBRTE_CRYPTODEV is disabled,
build of any crypto pmds will fail because of the missing dependency.
This has been present for a while now but hidden until the addition
of null_crypto since all the other crypto pmds have been disabled
by default.

Conditionalize the entire drivers/crypto directory on
CONFIG_RTE_LIBRTE_CRYPTODEV to fix.

Fixes: 1703e94ac5ce ("qat: add driver for QuickAssist devices")

Signed-off-by: Panu Matilainen <pmatilai@redhat.com>
8 years agonull_crypto: add driver for null crypto operations
Declan Doherty [Fri, 11 Mar 2016 01:04:10 +0000 (01:04 +0000)]
null_crypto: add driver for null crypto operations

This patch provides the implementation of a NULL crypto PMD, which supports
NULL cipher and NULL authentication operations, which can be chained together
as follows:

- Authentication Only
- Cipher Only
- Authentication then Cipher
- Cipher then Authentication

As this is a NULL operation device the crypto operations which are submitted for
processing are not actually modified and are stored in a queue pairs processed
packets ring ready for collection when rte_cryptodev_burst_dequeue() is called.

The patch also contains the related unit tests function to test the PMDs
supported operations.

Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
8 years agomaintainers: add doc for crypto devices
Thomas Monjalon [Thu, 10 Mar 2016 22:35:37 +0000 (23:35 +0100)]
maintainers: add doc for crypto devices

Fixes: 1703e94ac5ce ("qat: add driver for QuickAssist devices")
Fixes: 924e84f87306 ("aesni_mb: add driver for multi buffer based crypto")

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
8 years agomaintainers: claim responsibility for Intel QuickAssist PMD
Fiona Trahe [Fri, 5 Feb 2016 16:36:01 +0000 (16:36 +0000)]
maintainers: claim responsibility for Intel QuickAssist PMD

Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: John Griffin <john.griffin@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
8 years agodoc: update qat guide
Fiona Trahe [Thu, 25 Feb 2016 17:23:55 +0000 (17:23 +0000)]
doc: update qat guide

Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
8 years agoapp/test: add AES-GCM tests for QAT
John Griffin [Tue, 8 Mar 2016 16:22:16 +0000 (16:22 +0000)]
app/test: add AES-GCM tests for QAT

Signed-off-by: John Griffin <john.griffin@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agoqat: fix AES-GCM decryption
John Griffin [Tue, 8 Mar 2016 16:22:15 +0000 (16:22 +0000)]
qat: fix AES-GCM decryption

AES GCM on the cryptodev API was giving invalid results
in some cases, due to an incorrect IV setting.

Added AES GCM in the QAT supported algorithms,
as encryption/decryption is fully functional.

Fixes: 1703e94ac5ce ("qat: add driver for QuickAssist devices")

Signed-off-by: John Griffin <john.griffin@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agoqat: remove AES-GMAC
John Griffin [Tue, 8 Mar 2016 16:22:17 +0000 (16:22 +0000)]
qat: remove AES-GMAC

Remove support for AES GMAC support for which was added to
the code in error.  AES GMAC will be added in a subsequent release
when testing completes.

Fixes: 1703e94ac5ce ("qat: add driver for QuickAssist devices")

Signed-off-by: John Griffin <john.griffin@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agoaesni_gcm: add driver for AES-GCM crypto operations
Declan Doherty [Thu, 10 Mar 2016 16:41:46 +0000 (16:41 +0000)]
aesni_gcm: add driver for AES-GCM crypto operations

This patch provides the implementation of an AES-NI accelerated crypto PMD
which is dependent on Intel's multi-buffer library, see the white paper
"Fast Multi-buffer IPsec Implementations on Intel®  Architecture  Processors"

This PMD supports AES_GCM authenticated encryption and authenticated
decryption using 128-bit AES keys

The patch also contains the related unit tests functions

Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: John Griffin <john.griffin@intel.com>
8 years agocryptodev: add bit-wise handling for SNOW 3G
Deepak Kumar Jain [Thu, 10 Mar 2016 17:44:10 +0000 (17:44 +0000)]
cryptodev: add bit-wise handling for SNOW 3G

Wireless algorithms like Snow3G needs input in bits.
In this patch, changes have been made to incorporate this requirement
in both QAT and SW PMD.

Signed-off-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agosnow3g: add driver for SNOW 3G library
Pablo de Lara [Thu, 10 Mar 2016 16:33:12 +0000 (16:33 +0000)]
snow3g: add driver for SNOW 3G library

Added new SW PMD which makes use of the libsso SW library,
which provides wireless algorithms SNOW 3G UEA2 and UIA2
in software.

This PMD supports cipher-only, hash-only and chained operations
("cipher then hash" and "hash then cipher") of the following
algorithms:
- RTE_CRYPTO_SYM_CIPHER_SNOW3G_UEA2
- RTE_CRYPTO_SYM_AUTH_SNOW3G_UIA2

The SNOW 3G hash and cipher algorithms, which are enabled
by this crypto PMD are implemented by Intel's libsso software
library. For library download and build instructions,
see the documentation included (doc/guides/cryptodevs/snow3g.rst)

The patch also contains the related unit tests function to test the PMD
supported operations.

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
8 years agoapp/test: add SNOW 3G
Deepak Kumar Jain [Thu, 10 Mar 2016 17:12:45 +0000 (17:12 +0000)]
app/test: add SNOW 3G

Signed-off-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agoqat: add SNOW 3G
Deepak Kumar Jain [Thu, 10 Mar 2016 17:12:44 +0000 (17:12 +0000)]
qat: add SNOW 3G

Signed-off-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agoqat: add cipher/auth only
Deepak Kumar Jain [Thu, 10 Mar 2016 17:12:43 +0000 (17:12 +0000)]
qat: add cipher/auth only

Refactored the existing functionality into
modular form to support the cipher/auth only
functionalities.

Signed-off-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agoapp/test: fix qat autotest failure
Deepak Kumar Jain [Tue, 1 Mar 2016 16:55:16 +0000 (16:55 +0000)]
app/test: fix qat autotest failure

This patch fix the QAT autotest failure when run for multiple times.
it was caused as mbuf was not freed.

Fixes: 202d375c60b (app/test: add cryptodev unit and performance tests)

Signed-off-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
8 years agombuf_offload: remove library
Declan Doherty [Thu, 10 Mar 2016 13:43:01 +0000 (13:43 +0000)]
mbuf_offload: remove library

As cryptodev library does not depend on mbuf_offload library
any longer, this patch removes it.

Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
8 years agocryptodev: change burst API to be crypto op oriented
Declan Doherty [Thu, 10 Mar 2016 15:41:12 +0000 (15:41 +0000)]
cryptodev: change burst API to be crypto op oriented

This patch modifies the crypto burst enqueue/dequeue APIs to operate on bursts
rte_crypto_op's rather than the current implementation which operates on
rte_mbuf bursts, this simplifies the burst processing in the crypto PMDs and the
use of crypto operations in general, including new functions for managing
rte_crypto_op pools.

These changes continues the separation of the symmetric operation parameters
from the more general operation parameters, which will simplify the integration
of asymmetric crypto operations in the future.

PMDs, unit tests and sample applications are also modified to work with the
modified and new API.

Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
8 years agocryptodev: remove unused field
Fiona Trahe [Thu, 10 Mar 2016 15:41:11 +0000 (15:41 +0000)]
cryptodev: remove unused field

Remove unused phys_addr field from key in crypto_xform,
simplify struct and fix knock-on impacts in l2fwd-crypto app

Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
8 years agocryptodev: extract symmetric operations
Fiona Trahe [Thu, 10 Mar 2016 15:41:10 +0000 (15:41 +0000)]
cryptodev: extract symmetric operations

This patch splits symmetric specific definitions and
functions away from the common crypto APIs to facilitate the future extension
and expansion of the cryptodev framework, in order to allow asymmetric
crypto operations to be introduced at a later date, as well as to clean the
logical structure of the public includes. The patch also introduces the _sym
prefix to symmetric specific structure and functions to improve clarity in
the API.

Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
8 years agocryptodev: clean up
Fiona Trahe [Thu, 10 Mar 2016 15:41:09 +0000 (15:41 +0000)]
cryptodev: clean up

- Fixed >80char lines in test file
- Removed unused elements from stats struct
- Removed unused objects in rte_cryptodev_pmd.h
- Renamed variables
- Replaced leading spaces with tabs
- Improved performance results display in test

Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
8 years agodoc: fill nics features matrix for cxgbe
Rahul Lakkireddy [Thu, 10 Mar 2016 13:38:21 +0000 (19:08 +0530)]
doc: fill nics features matrix for cxgbe

Fill in the supported features matrix for CXGBE PMD.

Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Signed-off-by: Kumar Sanghvi <kumaras@chelsio.com>
8 years agopipeline: add new functions for action handlers
Jasvinder Singh [Tue, 8 Mar 2016 18:07:49 +0000 (18:07 +0000)]
pipeline: add new functions for action handlers

Two new pipeline API functions have been added to the library. The packet
hijack API function can be called by any input/output port or table action
handler to remove selected packets from the burst of packets read from one
of the pipeline input ports and then either send these packets out through
any pipeline output port or drop them.

Another packet drop API function can be used by the pipeline action
handlers (port in/out, table) to drop the packets selected using packet
mask. This function updates the drop statistics counters correctly.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
8 years agopipeline: support packet redirection at action handlers
Jasvinder Singh [Tue, 8 Mar 2016 18:07:48 +0000 (18:07 +0000)]
pipeline: support packet redirection at action handlers

Currently, there is no mechanism that allows the pipeline ports (in/out)
and table action handlers to override the default forwarding decision
(as previously configured per input port or in the table entry). The port
(in/out) and table action handler prototypes have been changed to allow
pipeline action handlers (port in/out, table) to remove the selected
packets from the further pipeline processing and to take full ownership
for these packets. This feature will be helpful to implement functions
such as exception handling (e.g. TTL =0), load balancing etc.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
8 years agovirtio: skip error when probing kernel managed device
Huawei Xie [Tue, 8 Mar 2016 15:33:43 +0000 (23:33 +0800)]
virtio: skip error when probing kernel managed device

virtio PMD could use IO port to configure the virtio device without
using UIO/VFIO driver in legacy mode.

There are two issues with previous implementation:
1) virtio PMD will take over the virtio device(s) blindly even if not
intended for DPDK.
2) driver conflict between virtio PMD and virtio-net kernel driver.

This patch checks if there is kernel driver other than UIO/VFIO managing
the virtio device before using port IO.

If legacy_virtio_resource_init fails and kernel driver other than
VFIO/UIO is managing the device, return 1 to tell the upper layer we
don't take over this device.
For all other IO port mapping errors, return -1.

Note than if VFIO/UIO fails, now we don't fall back to port IO.

Fixes: da978dfdc43b ("virtio: use port IO to get PCI resource")

Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
8 years agopci: ignore devices already managed in Linux when mapping x86 ioport
Huawei Xie [Tue, 8 Mar 2016 15:33:42 +0000 (23:33 +0800)]
pci: ignore devices already managed in Linux when mapping x86 ioport

call pci_ioport_map (on x86) only if the pci device is not bound
to a kernel driver.

Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
8 years agopci: rework ioport map error handling
Huawei Xie [Tue, 8 Mar 2016 15:33:41 +0000 (23:33 +0800)]
pci: rework ioport map error handling

Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
8 years agopci: identify devices not managed by any kernel driver
Huawei Xie [Tue, 8 Mar 2016 15:33:39 +0000 (23:33 +0800)]
pci: identify devices not managed by any kernel driver

Use RTE_KDRV_NONE to indicate that kernel driver (other than VFIO/UIO) isn't
managing the device.

Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
8 years agopci: fix error code comment
Huawei Xie [Tue, 8 Mar 2016 15:33:38 +0000 (23:33 +0800)]
pci: fix error code comment

positive return of devinit of pci driver means the driver doesn't support
this device.

Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
8 years agopci: use new compiler flag for x86
Huawei Xie [Tue, 8 Mar 2016 15:33:40 +0000 (23:33 +0800)]
pci: use new compiler flag for x86

Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
8 years agolpm: add a new config structure for IPv4
Michal Kobylinski [Wed, 9 Mar 2016 16:57:16 +0000 (17:57 +0100)]
lpm: add a new config structure for IPv4

A new rte_lpm_config structure is used so LPM library will allocate
exactly the amount of memory which is necessary to hold application’s
rules.

Signed-off-by: Michal Kobylinski <michalx.kobylinski@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
8 years agolpm: extend IPv4 next hop field
Michal Kobylinski [Wed, 9 Mar 2016 16:57:15 +0000 (17:57 +0100)]
lpm: extend IPv4 next hop field

This patch extend next_hop field from 8-bits to 24-bits in LPM library
for IPv4.

Added versioning symbols to functions and updated
library and applications that have a dependency on LPM library.

Signed-off-by: Michal Kobylinski <michalx.kobylinski@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
8 years agodoc: add vector fm10k guide
Chen Jing D(Mark) [Fri, 26 Feb 2016 05:56:41 +0000 (13:56 +0800)]
doc: add vector fm10k guide

Add introductions on how to enable Vector FM10K Rx/Tx functions,
the preconditions and assumptions on Rx/Tx configuration parameters.
The new content also lists the limitations of vector, so app/customer
can do better to select best Rx/Tx functions.

Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
8 years agodoc: add statistics read frequency to fm10k guide
Harry van Haaren [Tue, 8 Mar 2016 17:16:35 +0000 (17:16 +0000)]
doc: add statistics read frequency to fm10k guide

This patch documents that the statistics of fm10k based NICs must be
read regularly in order to avoid an undetected 32 bit integer-overflow.

Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
8 years agodoc: add statistics read frequency to ixgbe guide
Harry van Haaren [Tue, 8 Mar 2016 14:29:39 +0000 (14:29 +0000)]
doc: add statistics read frequency to ixgbe guide

This patch adds a note to the ixgbe PMD guide, stating
the minimum time that statistics must be polled from
the hardware in order to avoid register values becoming
saturated and "sticking" to the max value.

Reported-by: Jerry Zhang <jerry.zhang@intel.com>
Tested-by: Marcin Kerlin <marcinx.kerlin@intel.com>
Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Marcin Kerlin <marcinx.kerlin@intel.com>
8 years agodoc: fill nics features matrix for szedata2
Matej Vido [Tue, 8 Mar 2016 15:54:14 +0000 (16:54 +0100)]
doc: fill nics features matrix for szedata2

Signed-off-by: Matej Vido <vido@cesnet.cz>
8 years agodoc: add i40e guide
Jingjing Wu [Wed, 9 Mar 2016 07:28:39 +0000 (15:28 +0800)]
doc: add i40e guide

A new doc is added to introduce i40e pmd driver.

Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
8 years agodoc: fix API change in release notes
Jingjing Wu [Wed, 9 Mar 2016 11:59:03 +0000 (19:59 +0800)]
doc: fix API change in release notes

Move the structure ``rte_eth_fdir_masks`` change announcement from ABI
to API in release notes.

Fixes: 1409f127d7f1 (ethdev: fix byte order consistency of flow director)

Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
8 years agocmdline: bump library version
Thomas Monjalon [Wed, 9 Mar 2016 14:11:16 +0000 (15:11 +0100)]
cmdline: bump library version

There was an ABI change in the release 16.04.

Fixes: fb76dd26a31d ("cmdline: increase command line buffer")

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
8 years agoethdev: bump library version
Thomas Monjalon [Wed, 9 Mar 2016 14:09:50 +0000 (15:09 +0100)]
ethdev: bump library version

There was an ABI change and more are coming in the release 16.04.

Fixes: a9963a86b2e1 ("ethdev: increase RETA entry size")

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
8 years agoeal: check if primary process is alive
Harry van Haaren [Wed, 9 Mar 2016 13:37:25 +0000 (13:37 +0000)]
eal: check if primary process is alive

This patch adds a new function to the EAL API:
int rte_eal_primary_proc_alive(const char *path);

The function indicates if a primary process is alive right now.
This functionality is implemented by testing for a write-
lock on the config file, and the function tests for a lock.

The use case for this functionality is that a secondary
process can wait until a primary process starts by polling
the function and waiting. When the primary is running, the
secondary continues to poll to detect if the primary process
has quit unexpectedly, the secondary process can detect this.

Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Maryam Tahhan <maryam.tahhan@intel.com>
8 years agoeal: fix race condition in multi-process startup
Harry van Haaren [Wed, 9 Mar 2016 13:37:24 +0000 (13:37 +0000)]
eal: fix race condition in multi-process startup

This patch fixes a race-condition when a primary and
secondary process simultaneously probe PCI devices.

This is implemented by moving the rte_eal_mcfg_complete()
function call in rte_eal_init() until after rte_eal_pci_probe().
The memory mapping of PCI device in the secondary process *must*
happen after the primary has finished doing the mapping as it
relies on information written by the primary.

The end result is that the secondary process waits longer,
until the primary has completed its PCI probing, and then
notifies the secondary process.

This race-condition became visible during the development of
a function that allows a secondary process to be polling until
a primary process exists. The secondary would then probe PCI
devices at the same time, causing an error during rte_eal_init()

Linux EAL:
Fixes: 916e4f4f4e45 ("memory: fix for multi process support")

BSD EAL:
Fixes: 764bf26873b9 ("add FreeBSD support")

Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
8 years agoigb_uio: deprecate extended tag
Helin Zhang [Mon, 22 Feb 2016 03:59:45 +0000 (11:59 +0800)]
igb_uio: deprecate extended tag

It deprecates sys files of 'extended_tag' and
'max_read_request_size' which was not documented.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
8 years agopci: remove config of extended tag
Helin Zhang [Mon, 22 Feb 2016 03:59:44 +0000 (11:59 +0800)]
pci: remove config of extended tag

Remove pci configuration of 'extended tag' and 'max read request
size', as they are not required by all devices and it lets PMD to
configure them if necessary.
In addition, 'pci_config_space_set()' is deprecated.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
8 years agoi40e: enable extended tag
Helin Zhang [Mon, 22 Feb 2016 03:59:43 +0000 (11:59 +0800)]
i40e: enable extended tag

PCIe feature of 'Extended Tag' is important for 40G performance.
It adds its enabling during each port initialization, to ensure
the high performance.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
8 years agokeepalive: fix spacing
Harry van Haaren [Tue, 8 Mar 2016 10:50:40 +0000 (10:50 +0000)]
keepalive: fix spacing

This patch removes double newlines between functions
in keepalive.[hc] aligning it with the rest of DPDK.

Fixes: 75583b0d1efd ("eal: add keep alive monitoring")

Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
8 years agokeepalive: set timestamp on core registration
Harry van Haaren [Tue, 8 Mar 2016 10:50:39 +0000 (10:50 +0000)]
keepalive: set timestamp on core registration

This patch sets a timestamp on each lcore when it is registered
for keepalive. This causes the first values read by the monitor
to show time since the core was registered, instead of the delta
between 0 and the timestamp counter.

Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Remy Horton <remy.horton@intel.com>
8 years agodoc: fix keepalive sample app guide
Harry van Haaren [Tue, 8 Mar 2016 10:50:38 +0000 (10:50 +0000)]
doc: fix keepalive sample app guide

This patch fixes some mismatches between the keepalive code
and the docs. Struct names, and descriptions are not in line
with the codebase.

Fixes: e64833f2273a ("examples/l2fwd-keepalive: add sample application")

Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
8 years agomaintainers: claim responsibility for igb_uio
Ferruh Yigit [Fri, 4 Mar 2016 15:07:11 +0000 (15:07 +0000)]
maintainers: claim responsibility for igb_uio

igb_iuo has no maintainer, claim responsibility for igb_uio

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
8 years agoigb_uio: cast private data to correct struct type
Ferruh Yigit [Thu, 3 Mar 2016 17:08:19 +0000 (17:08 +0000)]
igb_uio: cast private data to correct struct type

This was working fine because addresses of two structs are same:

struct A {
struct B b;
} a;

As above sample "a" and "b" has same address.

Now casting private data back to the correct struct type, to the one
stored.

Fixes: af75078fece3 ("first public release")

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
8 years agoigb_uio: use macros for array size calculation
Ferruh Yigit [Fri, 4 Mar 2016 11:21:16 +0000 (11:21 +0000)]
igb_uio: use macros for array size calculation

Minor code cleanup.
Remove array size calculations and remove unnecessary assignment.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
8 years agodoc: fix number of supported bonding modes
Ferruh Yigit [Mon, 7 Mar 2016 10:25:36 +0000 (10:25 +0000)]
doc: fix number of supported bonding modes

Document mentions from 4 bonding mode but there are more modes.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
8 years agodoc: add known clang compilation issue
Pablo de Lara [Thu, 18 Feb 2016 15:21:09 +0000 (15:21 +0000)]
doc: add known clang compilation issue

Add known issue about DPDK not compiling on some CPUs
with clang versions older than 3.7.0.

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
8 years agodoc: comment unsupported ixgbe malicious driver detection
Wenzhuo Lu [Mon, 7 Mar 2016 05:28:10 +0000 (13:28 +0800)]
doc: comment unsupported ixgbe malicious driver detection

Announce that Malicious Driver Detection is not supported.

Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Shaopeng He <shaopeng.he@intel.com>
8 years agodoc: fix references in sample apps guide
Mauricio Vasquez B [Thu, 25 Feb 2016 17:02:27 +0000 (18:02 +0100)]
doc: fix references in sample apps guide

MANY references in the sample applications user guide are wrong because
they are hard-coded and section numbers have changed over the time.
This patch changes thoses references to dynamic ones, in this way if
section numbers change the reference get updated automatically.

Signed-off-by: Mauricio Vasquez B <mauricio.vasquezbernal@studenti.polito.it>
8 years agoexamples/ip_pipeline: add link identification
Fan Zhang [Tue, 1 Mar 2016 10:35:52 +0000 (10:35 +0000)]
examples/ip_pipeline: add link identification

This patch adds link identification feature to packet framework. To
identify a link, user can use both existing port-mask option, or specify
PCI device in every LINK section in the configuration file.

Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
8 years agoexamples/ip_pipeline: measure CPU utilization
Fan Zhang [Mon, 22 Feb 2016 14:07:04 +0000 (14:07 +0000)]
examples/ip_pipeline: measure CPU utilization

This patch adds CPU utilization measurement and idle cycle rate
computation to packet framework. The measurement is done by measuring
the cycles spent while a thread pulls zero packet from RX queue. These
cycles are treated as idle cycles (or headroom). A CLI command is added
to display idle cycle rate of specific thread. The CLI command format is
shown as following:

t <thread_id> headroom

Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
8 years agoexamples/ip_pipeline: clean config parser
Fan Zhang [Wed, 17 Feb 2016 11:14:11 +0000 (11:14 +0000)]
examples/ip_pipeline: clean config parser

This patch updates the pipelne configuration file parser, cleans up nesting
if/else conditions, and add clearer error message display.

Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
8 years agoexamples/ip_pipeline: fix CPU socket id
Jasvinder Singh [Wed, 27 Jan 2016 11:47:09 +0000 (11:47 +0000)]
examples/ip_pipeline: fix CPU socket id

This patch fixes the socket-id error in ip_pipeline sample
application running over uni-processor systems.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
8 years agoport: fix crash for ring writer nodrop
Jasvinder Singh [Wed, 2 Mar 2016 21:19:58 +0000 (21:19 +0000)]
port: fix crash for ring writer nodrop

Error log:
 [APP] Initializing PIPELINE0 ...
 pipeline> [APP] Initializing PIPELINE1 ...
 [PIPELINE1] Pass-through
 [APP] Initializing PIPELINE2 ...
 [PIPELINE2] Pass-through
 Segmentation fault (core dumped)

Fixes: 5f4cd47309d6 ("port: add ring writer nodrop")
Fixes: d58f69c54172 ("port: add ring multi reader or writer")

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
8 years agoport: fix crash for ethdev writer nodrop
Jasvinder Singh [Wed, 2 Mar 2016 21:19:29 +0000 (21:19 +0000)]
port: fix crash for ethdev writer nodrop

Error log:
 [APP] Initializing PIPELINE0 ...
 pipeline> [APP] Initializing PIPELINE1 ...
 [PIPELINE1] Pass-through
 Segmentation fault (core dumped)

Fixes: 304c8091e90a ("port: add ethdev writer nodrop")

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
8 years agodoc: add gcc-multilib as linux package hint
Harry van Haaren [Tue, 16 Feb 2016 13:40:03 +0000 (13:40 +0000)]
doc: add gcc-multilib as linux package hint

When compiling for i686 targets compilation could fail
if the 32bit libc6-dev package is not installed. The
gcc-multilib packages is a meta-package that will pull
in the necessary dependencies, making setup easier for
beginners.

Reported-by: Weichun Chen <weichunx.chen@intel.com>
Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
8 years agoapp/testpmd: fix numa socket detection
Stephen Hurd [Wed, 13 Jan 2016 22:23:36 +0000 (14:23 -0800)]
app/testpmd: fix numa socket detection

Previously, max_socket was set to the highest numbered socket with
an enabled lcore.  The intent is to set it to the highest socket
regardless of it being enabled.

Fixes: 7acf894d07d1 ("app/testpmd: detect numa socket count")

Signed-off-by: Stephen Hurd <shurd@broadcom.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agoapp/testpmd: fix error message when setting Tx VLAN
Wang Xiao W [Fri, 5 Feb 2016 04:50:23 +0000 (12:50 +0800)]
app/testpmd: fix error message when setting Tx VLAN

When using testpmd, sometimes we forget the right order of port_id and
vid in "tx_vlan set (port_id) vlan_id[, vlan_id_outer]\n" command, and
input "tx_vlan set 51 0", we'll get a strange prompt saying "Error, as
QinQ has been enabled.".

In cmd_tx_vlan_set_parsed function, the first thing we do is checking
the port's vlan_offload capability, rather than checking validity of the
port_id, therefore if it's an invalid port_id we'll get the above wrong
message. We should always make sure that we get a valid port_id before
we do other things.

It's the similar issue for cmd_tx_vlan_set_qinq_parsed function.

Fixes: 92ebda07ee58 ("app/testpmd: add qinq stripping and insertion")

Signed-off-by: Wang Xiao W <xiao.w.wang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
8 years agoeal/arm: check support of armv8.1 atomics
Jerin Jacob [Wed, 2 Mar 2016 13:20:59 +0000 (18:50 +0530)]
eal/arm: check support of armv8.1 atomics

armv8.1 adds support for new atomic instructions.
Linux kernel v4.3 onwards, the presence of atomic instruction
support can detect through HWCAP_ATOMICS

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Reviewed-by: Jan Viktorin <viktorin@rehivetech.com>
8 years agoconfig: remove EAL flags for OS environment
Thomas Monjalon [Fri, 4 Mar 2016 23:02:02 +0000 (00:02 +0100)]
config: remove EAL flags for OS environment

CONFIG_RTE_LIBRTE_EAL_*APP can be replaced by CONFIG_RTE_EXEC_ENV_*APP.

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
8 years agoconfig: remove duplicate information
Keith Wiles [Fri, 4 Mar 2016 18:11:12 +0000 (12:11 -0600)]
config: remove duplicate information

In order to cleanup the configuration files some and reduce
the number of duplicate configuration information. Add a new
file called common_base which contains just about all of the
configuration lines in one place. Then have the common_bsdapp,
common_linuxapp files include this one file. Then in those OS
specific files add the delta configuration lines.

Signed-off-by: Keith Wiles <keith.wiles@intel.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
8 years agoconfig: fix missing 64-bit flag on FreeBSD
Keith Wiles [Fri, 4 Mar 2016 17:03:26 +0000 (11:03 -0600)]
config: fix missing 64-bit flag on FreeBSD

Until now, the generic 64-bit flag was used only for ARM or Linux,
and was not defined for BSD environment.

Fixes: d05e7115f466 ("mem: support layout of IBM Power")

Signed-off-by: Keith Wiles <keith.wiles@intel.com>
8 years agodoc: tidy sections in release notes
Thomas Monjalon [Fri, 4 Mar 2016 15:44:26 +0000 (16:44 +0100)]
doc: tidy sections in release notes

Fixes: 5499c1fc9baa ("examples/vhost: fix mbuf allocation")

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
8 years agoethdev: fix byte order consistency of flow director
Jingjing Wu [Mon, 1 Feb 2016 02:48:21 +0000 (10:48 +0800)]
ethdev: fix byte order consistency of flow director

Fixed issue of byte order in ethdev library that the structure
for setting fdir's mask and flow entry is inconsist and made
inputs of mask be in big endian.

Fixes: 2d4c1a9ea2ac ("ethdev: add new flow director masks")
Fixes: 76c6f89e80d4 ("ixgbe: support new flow director masks")

Reported-by: Yaacov Hazan <yaacovh@mellanox.com>
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Zhe Tao <zhe.tao@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>