tools: allow nic status query without igb_uio

Allow the nic bind/unbind script to print out its status messages even
if the igb_uio driver is not loaded. For binding and unbinding NICs, the
behaviour is the same, and the igb_uio driver still needs to be loaded.

Signed-off-by: Bruce richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

tools: unbind ports that were erroneously bound

When binding devices to a generic driver (i.e. one that doesn't have a
PCI ID table, some devices that are not bound to any other driver could
be bound even if no one has asked them to. hence, we check the list of
drivers again, and see if some of the previously-unbound devices were
erroneously bound. if such devices are found, they are unbound back.

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

app/testpmd: allow to set part of flow control parameters

We might want to only change a parameter rather than have to set all possible
parameters, so add "partial" commands.
These commands only change the specified parameter.

To avoid duplicating code all around, a unique parser is kept. This parser uses
the .data parameter to select the right behavior.

Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

app/testpmd: move flow control parser

Move parser after declarations to prepare rework in next commit.

Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

app/testpmd: add missing autoneg field to set flow control

Following commit 2d95b84aaacb3d2d0bd70367c0530d15e0cbb14e, rte_eth_fc_conf
struct contains a autoneg field that must be set by callers.
Add this parameter to testpmd.

Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

eal: fix invalid memory read

Valgrind reports this issue:

==29880== Invalid read of size 1
==29880==    at 0x56FF9A5: cpu_socket_id (eal_lcore.c:101)
==29880==    by 0x56FFAE9: rte_eal_cpu_init (eal_lcore.c:168)
==29880==    by 0x56F944A: rte_eal_init (eal.c:975)

The problem is that endptr points to memory allocated underneath the DIR
handle, which has already been freed. So move the closedir() call lower.

Signed-off-by: Aaron Campbell <aaron@arbor.net>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

eal: deprecate rte_snprintf

The function rte_snprintf serves no useful purpose. It is the
same as snprintf() for all valid inputs. Deprecate it and
replace all uses in current code.

Leave the tests for the deprecated function in place.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

eal: fix usage of printf-like functions

Mark the rte_log, cmdline_printf and rte_snprintf functions as
being printf-style functions. This causes compilation errors
due to mis-matched parameter types, so the parameter types are
fixed where appropriate.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

mk: override HOSTCC when overriding CC

When overriding the CC variable on the commandline, the HOSTCC value
was remaining at "gcc", which caused errors on Free BSD10 when using a gcc
as e.g. gcc48, without a binary just called "gcc". This change overrides
HOSTCC when CC is overridden, which means that setting CC=gcc48, will
now also set HOSTCC=gcc48, allowing the "testhost" app to compile on
BSD10.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

acl: disable for bsd

Using gcc 4.8 on FreeBSD 10, support for SSE4.x is not detected by
the compiler, meaning that the ACL library, which depends on SSE4.2
cannot compile. Disable this library for the native target allows
compiles to succeed on FreeBSD 10 using gcc.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

mem: fix build because of missing stdio.h for FILE

Since commit "add FILE argument to debug functions" (591a9d7985c1230),
application which includes rte_memory.h without stdio.h will be hit
compilation failure:

/path/to/include/rte_memory.h:146:30: error: unknown type name ‘FILE’
 void rte_dump_physmem_layout(FILE *f);

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Reviewed-by: Hayato Momma <h-momma@ce.jp.nec.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

ethdev: add link state interrupt flag

Only some devices support the link state interrupt configuration option.
Link state control does not work in virtual drivers
(virtio, vmxnet3, igbvf, and ixgbevf). Instead of having the application
try and guess whether it will work or not provide a driver flag that
can be checked instead.

Note: if device driver doesn't support link state control, what
would happen previously is that the code would never detect link
transitions. This prevents that.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
[Thomas: rename flag]
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

ethdev: read link state interrupt without link update service

It is now possible to read link status updated by interrupt without
having manual link_update() service provided by the PMD.
Indeed link_update() is useless in interrupt case. So check of this
function pointer must be done in the interrupt case only.

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>

examples/ip_reassembly: reduce memory usage

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

examples/ip_frag: fix socket detection

Make everything NUMA-related depend on lcore sockets, not device
sockets. This is because the init_mem() function allocates all data
structures based on NUMA nodes of the lcores in the coremask. Therefore,
when no cores are on socket 0, but there are devices on socket 0, it may
lead to segmentation faults.

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

examples/ip_frag: check for non-existent ports

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

ip_frag: add config option to enable statistics

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

ip_frag: custom memmove

Some implementations of memmove may make a copy of src before writing to
dst. We avoid that by explicitly writing from src to dst backwards.

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

ip_frag: fix order of key compare arguments

When using key compare function, it uses key length of the first
argument to determine how long should be the keys that are compared.
However, currently we are passing a key from the fragmentation table as
first argument. the problem with this is that this key is potentially
uninitialized (i.e. contains all zeroes, including key length). this
leads to a nasty bug of comparing only the key id's and not keys
themselves.

Of course, a safer way would be to do RTE_MAX between key lengths, but
since this compare is done per-packet, every cycle counts, so we just
use the key whose length is guaranteed to be correct because it comes
from an actual packet.

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

ip_frag: fix debug macros

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

ip_frag: fix comment

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

ip_frag: replace hardcoded value with a macro

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

ip_frag: remove unneeded rte prefixes

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

malloc: fix unit tests

Fix typos and false assumptions in malloc unit tests.

Without enhancements to lib rte_malloc, malloc autotest fails every
second (2nd) run. With enhancements, malloc autotest fails in
function test_multi_alloc_statistics, because we compare the wrong
sets of statistics.

Signed-off-by: Robert Sanford <rsanford2@gmail.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>

malloc: fix linear complexity

Problems with lib rte_malloc:
1. Rte_malloc searches a heap's entire free list looking for the best
   fit, resulting in linear complexity.
2. Heaps store free blocks in a singly-linked list, resulting in
   linear complexity when rte_free needs to remove an adjacent block.
3. The library inserts and removes free blocks with ad hoc, in-line
   code, rather than using linked-list functions or macros.
4. The library wastes potential small blocks of size 64 and 128 bytes
   (plus overhead of 64 bytes) as padding when reusing free blocks or
   resizing allocated blocks.

This patch addresses those problems as follows:
1. Replace single free list with a handful of free lists. Each free
   list contains blocks of a specified size range, for example:
     list[0]: (0   , 2^8]
     list[1]: (2^8 , 2^10]
     list[2]: (2^10, 2^12]
     list[3]: (2^12, 2^14]
     list[4]: (2^14, MAX_SIZE]

   When allocating a block, start at the first list that can contain
   a big enough block. Search subsequent lists, if necessary.
   Terminate the search as soon as we find a block that is big enough.
2. Use doubly-linked lists, so that we can remove free blocks in
   constant time.
3. Use BSD LIST macros, as defined in sys/queue.h and the QUEUE(3)
   man page.
4. Change code to utilize small blocks of data size 64 and 128, when
   splitting larger blocks.

Signed-off-by: Robert Sanford <rsanford2@gmail.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>

app/testpmd: tune default parameters

We usually use testpmd fwd to demonstrate IO forwarding throughput.
For best throughput, it has to assign special parameters to testpmd.
To make it easier to run, now set it as defalut value.
Such parameters are MBUF Mempool Cache and RX/TX threshold registers.
    MBCACHE: 250
    RX threshold registers: pthresh=8 hthresh=8 wthresh=0
    TX threshold registers: pthresh=32 hthresh=0 wthresh=0

Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Yong Liu <yong.liu@intel.com>
Tested-by: Zhaochen Zhan <zhaochen.zhan@intel.com>

app/testpmd: increase default burst size to 32

The vpmd RX don't accept burst size less than 32.
As vPMD is set =y by default, while default testpmd burst size is 16.
Which will cause RX nothing if not assign burst size correctly.

Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Yong Liu <yong.liu@intel.com>
Tested-by: Zhaochen Zhan <zhaochen.zhan@intel.com>

ixgbe: prefer enabling olflags rather than not disabling

RTE_IXGBE_RX_OLFLAGS_ENABLE gives a hint whick keeping packet type
in RX ol_flags or not.
By default it is set to update ol_flags in RX mbuf header.
If unset it, will gain addtional performance, but will lose packet
type information.

Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Yong Liu <yong.liu@intel.com>
Tested-by: Zhaochen Zhan <zhaochen.zhan@intel.com>

ixgbe: fix build with IEEE1588 enabled

Vectorized Rx cannot be enabled if RTE_LIBRTE_IEEE1588=y.
So variable dev is not used in this case.

Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Yong Liu <yong.liu@intel.com>
Tested-by: Zhaochen Zhan <zhaochen.zhan@intel.com>

ixgbe: fix build of standard Rx with vectorized Tx

The vpmd RX routine won't be used if RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC
is not defined or its condition check fail.
The case RTE_LIBRTE_IXGBE_RX_ALLOC_BULK_ALLOC=n and RTE_IXGBE_INC_VECTOR=y
may exist when choose to use standard RX and optimized TX
(ixgbe_xmit_pkts_vec/ixgbe_xmit_pkts_simple).

Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Yong Liu <yong.liu@intel.com>
Tested-by: Zhaochen Zhan <zhaochen.zhan@intel.com>

ixgbe: rework fix of media type for bypass device

This was previously solved in commit 60a70d4e042350ca
(fix link status interrupt of bypass device),
but this alternative fix reverts the previous one and solves the
same issue without modifying the base driver (also named "shared code").

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>

i40e/base: ignore warning

There is a warning in base driver (shared code) on 32-bits RHEL6.3/6.5:
lib/librte_pmd_i40e/i40e/i40e_lan_hmc.c:917:
error: integer constant is too large for ‘long’ type

As we don't modify base driver, this warning must be ignored.

Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>

explicit shared code naming as base driver

Intel PMDs are built on top of the base drivers which are provided by Intel
and shouldn't be modified to allow easy batch upgrade from Intel.

The base driver is a "shared code" between many projects. But in DPDK,
the "base driver" naming makes more sense.

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>

app/testpmd: rework display of Rx descriptors

i40e supports 16 and 32 bytes RX descriptors which can be configured.
It needs to check the driver type and the configuration to determine
if 16 or 32 bytes RX descriptors is being used, for reading and
displaying the different sizes of RX descriptors.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>

i40e: ignore failure when updating default macvlan filter

For NVM4.2.2 or after, the firmware has the correct configurations
and load the macvlan filter as expected. It is not needed to
Update the default macvlan filter which cannot be removed at
all during initialization.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>

i40e: disable double vlan by default

Double vlan should be specifically disabled by default during
port initialization which is expected.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>

i40e: fix updating hash lookup table of PF RSS

The bit shifting were written wrongly in '0x1 < j',
the correct one should be '0x1 << j'.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>

i40e: fix RSS hash copy

It wrongly gets the RSS hash result from the RX descriptor which
has been modified for receiving new packet. The fix is to get the
RSS hash result from the buffer which saves the RX descriptor.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>

i40evf: fix link status

Fix a bug of copying wrong size of link information in the
function i40evf_get_link_status().

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>

i40evf: remove useless function

i40evf_dev_atomic_read_link_status() was defined but not used.
To avoid possible warnings by some compilers, it needs to
delete the whole function.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>

kni: fix build on Oracle Linux 6.4 and RHEL 6.5

The compile errors are copied as follows. The fixes came from
Linux drivers of ixgbe-3.21.2 and igb-5.1.2 with modifications.
The idea is to use self-defined functions no matter they have
already been defined somewhere or not.

* Oracle Linux6.4
lib/librte_eal/linuxapp/kni/ethtool/ixgbe/kcompat.h:3111:
error: redefinition of 'ether_addr_equal'
include/linux/etherdevice.h:180: note: previous definition
of 'ether_addr_equal' was here

* RHEL6.5
lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h:3597:
error: redefinition of 'mmd_eee_cap_to_ethtool_sup_t'
include/linux/mdio.h:387: note: previous definition of
'mmd_eee_cap_to_ethtool_sup_t' was here
lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h:3625:
error: redefinition of 'mmd_eee_adv_to_ethtool_adv_t'
include/linux/mdio.h:415: note: previous definition of
'mmd_eee_adv_to_ethtool_adv_t' was here
lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h:3653:
error: redefinition of 'ethtool_adv_to_mmd_eee_adv_t'
include/linux/mdio.h:443: note: previous definition of
'ethtool_adv_to_mmd_eee_adv_t' was here

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>

i40e: fix build with icc

Cast constant to fix the ICC compilation error
introduced in commit 2b12431b5369ca06d43ea
(add vlan stripping and insertion to VF)

Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

distributor: split get_pkt into request and poll

Take the existing get_pkt API and split out the parts for requesting a new
packet from the part to poll for arrival of a new packet. These individual
functions can then be used independently of the get function, which still acts
as before.

The split functions for request and poll will allow a worker to pull
packets from multiple distributors, or to act as multiple workers with a
single distributor if needed for better performance.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>

i40e: add vlan stripping and insertion to VF

VF driver add support to configure vlan offload and pvid set/clear
when host driver is DPDK. For Linux driver, it's not supported yet
since it doesn't support from host side.

Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>

i40e: add vlan stripping and insertion to PF messaging

PF driver add 2 commands, data structures and corresponding
message handling functions to support VF doing VLAN offload
and set/clear pvid operation.

Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>

i40e: rework vlan stripping and insertion

Change original 2 functions to common ones that can take effect on
both VF and PF VSIs.

Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>

i40e: add sanity checks in PF messaging

Add sanity check in message handling functions on the request
buffer pointer from VF.

Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>

i40e: add permanent mac address into mac list

In old firmware versions, the default mac vlan filter setting is
not the one that we expected, so we'll try to remove it and add
new one to change default setting. After firmware updated, it
change default setting to the one that we expected and don't
allow to remove the setting. In this case, we should add the
perenant mac address into mac list and then return.

Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>

i40e: destroy MSI-X pool on close

Free MSIX pool resource in function i40e_dev_close.

Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>

eal: fix option --base-virtaddr

When parsing EAL option --base-virtaddr
errno was not being set to 0 before calling strtoull,
therefore function might fail unnecesarily.

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Signed-off-by: Aaron Campbell <aaron@arbor.net>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>

version: 1.7.0-rc1

Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>

vfio: more verbose error messages

also, making VFIO code distinguish between actual unexpected values
and ioctl() failures, providing appropriate error messages.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>

vfio: open container at startup rather than during init

Currently, VFIO only checks for being able to access the /dev/vfio
directory when initializing VFIO, deferring actual VFIO container
initialization to VFIO binding code. This doesn't bode well for when
VFIO container cannot be initialized for whatever reason, because
it results in unrecoverable error even if the user didn't set up
VFIO and didn't even want to use it in the first place.

This patch fixes this by moving container initialization into the
code that checks if VFIO is available at runtime. Therefore, any
issues with the container will be known at initialization stage and
VFIO will simply be turned off if container could not be set up.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>

i40e: remove endian.h include

endian.h is not needed for the compilation of i40e_rxtx.c and its
inclusion prevents building on FreeBSD systems.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

app/test-acl: remove unused variable

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

app/testpmd: simplify handling of stats mappings error

Simplifiy the logic in the error checking branch. Rather than having a
single error branch which checked both RX and TX conditions and made
extensive use of the ? operator, move the error checking explicitly into
the RX and TX individual branches.

The original code caused compilation issues when attempting compilation
with clang on BSD, due to type mismatches. This change fixes the issues
while making the code easier to read and maintain overall.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

app/testpmd: fix cmdline variable types

A number of commandline entries in the testpmd commandline were actually
defined as being string type values when in fact they were being
initialized as integer types. Correct this by specifying them as integer
type values in the type definition.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

scripts: fix filtering of config comments on bsd

On BSD 10, the cpp binary behaves a little differently and often leaves
lines starting with a space before the inital '#' character. This change
ensures those lines are filtered out properly.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

tools: fix alignment of cpu layout output

Bug: when "core id" is greater than 9, the cpu_layout.py output doesn't align.

        Socket 0        Socket 1
        ---------       ---------
Core 9  [4, 16]         [10, 22]

Core 10         [5, 17]         [11, 23]

Solution: adjust output format to align based on the maximum length of the "core id" and "processor"

        Socket 0        Socket 1
        --------        --------
Core 9  [4, 16]         [10, 22]

Core 10 [5, 17]         [11, 23]

Signed-off-by: Shannon Zhao <zhaoshenglong@huawei.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>

ixgbe/base: update copyright to 2014

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: various changes

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: vf changes

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: x540 changes

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: filtering changes

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: flow director changes

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: allow to read in sff8472

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: allow access to PHY register without lock

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: rework protected access to autoc register

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: rework locking

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: rework flow control

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: rework link negotiation

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: disable laser control if manageability enabled

igned-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: allow to disable link on D3

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: allow to enable and disable Rx

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: allow to read rtrup2tc register

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: rework PCIe disabling

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: name constants

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: improve error handling

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

ixgbe/base: minor changes

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: update copyright to 2014

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: various changes

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: fix K1 beacon for 82579

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: m88e1512 phy in 82575

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: fix m88e1112 for 82575

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: fix i354

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: fix i217

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: i210 PLL workaround

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: improve error handling

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: minor changes

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

e1000/base: cleanup c10, ulp and nahum6 i218 support

Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>
[Thomas: split code drop]

app/testpmd: command to configure MTU

Take avantage of the .set_mtu ethdev function and make it possible to configure
MTU on devices using testpmd.

Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>

ixgbevf: allow to set MTU

The support of jumbo frames in the ixgbevf Poll Mode Driver of 10GbE
82599 VF functions consists in the following enhancements:

- Implement the mtu_set function in the ixgbevf PMD, using the IXGBE_VF_SET_LPE
  request of the version 1.0 of the VF/PF mailbox API for this purpose.

- Add a detailed explanation on the VF/PF rx max frame len negotiation.

Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>

ethdev: MTU accessors

This patch adds two new functions in ethdev api to retrieve current MTU and
change MTU of a port.
Only .mtu_set function is pmd specific.
pmd should update its max_rx_pkt_len if needed.

This operation has been implemented for rte_em_pmd, rte_igb_pmd and
rte_ixgbe_pmd.

Signed-off-by: Samuel Gauthier <samuel.gauthier@6wind.com>
Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>

ethdev: allow to force Rx scatter mode

We might want to be sure the scatter packets reception handler is selected in a
pmd. This makes it possible to then change mtu later, without the need of
restarting a port.
It is then the pmd duty to tell it enabled the scatter reception handler by
setting dev->data->scattered_rx to 1.

Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>

ethdev: store minimum Rx buffer size

This avoids code duplication in PMD when dealing with mtu changes.

Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>

ethdev: autoneg parameter in flow control accessors

Add autoneg field in flow control parameters.
This makes it easier to understand why changing some parameters does not always
have the expected result.

Changing autoneg is not supported at the moment.

Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>

ethdev: retrieve flow control configuration

This patch adds a new function in ethdev api to retrieve current flow control
configuration.
This operation has been implemented for rte_em_pmd, rte_igb_pmd and
rte_ixgbe_pmd.

Signed-off-by: Zijie Pan <zijie.pan@6wind.com>
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>

kvargs: fix build with gcc 4.5.1

GCC 4.5.1 warns about an uninitialized pointer:
error: 'ctx1' may be used uninitialized in this function

Signed-off-by: Min Cao <min.cao@intel.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>

app/testpmd: add L3 packet type in offload flags

As i40e PMD need to know the L3 packet type for TX checksum offloading,
the packet type has been added in rte_mbuf offload flags.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Signed-off-by: Jing Chen <jing.d.chen@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Heqing Zhu <heqing.zhu@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>

app/testpmd: support displaying 32-byte Rx descriptors

i40e supports both 16 and 32 bytes RX descriptors, while ixgbe and
igb support 16 bytes size only. Code changes have been made in
test-pmd to support both 16 and 32 bytes RX descriptors according
to the configuration in config files.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Signed-off-by: Jing Chen <jing.d.chen@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Heqing Zhu <heqing.zhu@intel.com>
Tested-by: Waterman Cao <waterman.cao@intel.com>