dpdk.git
6 years agopmdinfogen: fix cross compilation for ARM big endian
Hemant Agrawal [Mon, 18 Dec 2017 07:56:45 +0000 (13:26 +0530)]
pmdinfogen: fix cross compilation for ARM big endian

Cross compiling DPDK for BE mode on ARM results into errors
"PMDINFO portal/dpaa2_hw_dpio.o.pmd.c No drivers registered"

The original code assumes the sh_size to be 32 bit, while it can
be Elf32_Word or Elf64_Xword based on 32bit or 64 bit systems.

This patches replaces the sh_size conversion routines to use ADDR_SIZE

Fixes: 98b0fdb0ffc6 ("pmdinfogen: add buildtools and pmdinfogen utility")
Cc: stable@dpdk.org
Signed-off-by: Jun Yang <jun.yang@nxp.com>
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
6 years agovfio: fix enabled check on error
Anatoly Burakov [Fri, 22 Dec 2017 10:30:36 +0000 (10:30 +0000)]
vfio: fix enabled check on error

rte_eal_check_module() might return -1, which would have been a
"not false" condition for mod_available. Fix that to only report
vfio being enabled if rte_eal_check_module() returns 1.

Fixes: 221f7c220d6b ("vfio: move global config out of PCI files")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomalloc: fix end for bounded elements
Anatoly Burakov [Thu, 21 Dec 2017 16:54:24 +0000 (16:54 +0000)]
malloc: fix end for bounded elements

In cases when alignment is bigger than boundary, we may incorrectly
calculate end of a bounded malloc element.

Consider this: suppose we are allocating a bounded malloc element
that should be of 128 bytes in size, bounded to 128 bytes and
aligned on a 256-byte boundary. Suppose our malloc element ends
at 0x140 - that is, 256 plus one cacheline.

So, right at the start, we are aligning our new_data_start to
include the required element size, and to be aligned on a specified
boundary - so new_data_start becomes 0. This fails the following
bounds check, because our element cannot go above 128 bytes from
the start, and we are at 320. So, we enter the bounds handling
branch.

While we're in there, we are aligning end_pt to our boundedness
requirement of 128 byte, and end up with 0x100 (since 256 is
128-byte aligned). We recalculate new_data_size and it stays at
0, however our end is at 0x100, which is beyond the 128 byte
boundary, and we report inability to reserve a bounded element
when we could have.

This patch adds an end_pt recalculation after new_data_start
adjustment - we already know that size <= bound, so we can do it
safely - and we then correctly report that we can, in fact, try
using this element for bounded malloc allocation.

Fixes: fafcc11985a2 ("mem: rework memzone to be allocated by malloc")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomalloc: protect stats with lock
Anatoly Burakov [Thu, 21 Dec 2017 17:32:04 +0000 (17:32 +0000)]
malloc: protect stats with lock

When we're gathering statistics, we are traversing the freelist,
which may change under our feet in multithreaded scenario. This
is verified by occasional segfaults when running malloc autotest
on a machine with big amount of cores.

This patch protects malloc heap stats call with a lock. It changes
its definition in the process due to locking invalidating the
const-ness, but this isn't a public API, so that's OK.

Fixes: 2a5c356e177d ("memory: stats for malloc")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomemzone: fix leak on allocation error
Anatoly Burakov [Thu, 21 Dec 2017 18:07:07 +0000 (18:07 +0000)]
memzone: fix leak on allocation error

We check if there's space in config after we allocated the memzone,
but if there isn't, we never free it back. This patch adds memzone
free if there's no room in memzone config.

Fixes: ff909fe21f0a ("mem: introduce memzone freeing")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoservice: add attribute for number of invocations
Harry van Haaren [Fri, 12 Jan 2018 10:27:18 +0000 (10:27 +0000)]
service: add attribute for number of invocations

This commit adds a new attribute to the service cores attributes
API, which allows the application to retrieve the number of times
that a service-core called the service to perform its action.

Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
6 years agoservice: add reset all attributes for service
Harry van Haaren [Fri, 12 Jan 2018 10:27:17 +0000 (10:27 +0000)]
service: add reset all attributes for service

This commit introduces a new API, allowing the application to
reset attributes of a service like the cycle count. Given this
functionality is now exposed to the user, remove the resetting
of stats during a dump() call.

Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
6 years agoservice: add attribute get function
Harry van Haaren [Fri, 12 Jan 2018 10:27:16 +0000 (10:27 +0000)]
service: add attribute get function

This commit adds a new function to the service API to allow
the application to retrieve items about each individual service
in the system. A unit test checks the return values of a variety
of invalid and valid calls.

Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
6 years agobus/fslmc: change EQCR stashing threshold to 1
Nipun Gupta [Wed, 3 Jan 2018 15:39:04 +0000 (21:09 +0530)]
bus/fslmc: change EQCR stashing threshold to 1

Changing the EQCR stashing threshold boosts the performance
of l3fwd application on LS2088 by more than 20% as it helps
in burst packet processing at the Tx side. CPU is immediately
informed about the empty EQCR entries once consumed by the
hardware.

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agobus/pci: fix interrupt handler type
Zhiyong Yang [Wed, 10 Jan 2018 02:32:26 +0000 (10:32 +0800)]
bus/pci: fix interrupt handler type

For virtio legacy device, testpmd startup fails when using uio_pci_generic.

The issue is caused by invoking the function pci_ioport_map. The correct
value of intr_handle.type is already set before calling it, we should avoid
overwriting the default value "RTE_INTR_HANDLE_UNKNOWN" in this function.
Besides, the removal has no harm to other cases because it is set to 0 by a
memset on the whole struct during allocation in the function pci_scan_one.

Such assignments are removed in the meanwhile in pci_uio_map_resource(),
pci_vfio_map_resource_primary() and pci_vfio_map_resource_secondary() in
order to keep consistencies and avoid future questions.

Fixes: 756ce64b1ecd ("eal: introduce PCI ioport API")
Cc: stable@dpdk.org
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
Reviewed-by: Thomas Monjalon <thomas@monjalon.net>
6 years agobus/vdev: add custom scan hook
Thomas Monjalon [Fri, 1 Dec 2017 00:27:10 +0000 (01:27 +0100)]
bus/vdev: add custom scan hook

The scan callback allows to spawn a vdev automatically
given some custom scan rules.
It is especially useful to create a TAP device automatically
connected to a netdevice as remote.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Jianfeng Tan <jianfeng.tan@intel.com>
6 years agoeal/x86: get hypervisor name
Thomas Monjalon [Thu, 30 Nov 2017 21:27:01 +0000 (22:27 +0100)]
eal/x86: get hypervisor name

The CPUID instruction is caught by hypervisor which can return
a flag indicating one is running, and its name.

Suggested-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
6 years agoeal: remove use of register keyword
Avi Kivity [Tue, 9 Jan 2018 16:11:08 +0000 (18:11 +0200)]
eal: remove use of register keyword

The 'register' keyword does nothing, and has been removed in C++17.

Remove it for compatibility.

Signed-off-by: Avi Kivity <avi@scylladb.com>
6 years agomem: warn if address hint is not respected
Jonas Pfefferle [Tue, 31 Oct 2017 09:08:29 +0000 (10:08 +0100)]
mem: warn if address hint is not respected

Print a warning if the --base-virtaddr hint is not respected
since this might lead to problems when mapping memory in
the secondary process.

Signed-off-by: Jonas Pfefferle <jpf@zurich.ibm.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agombuf: check sanity of data and packet lengths
Ilya V. Matveychikov [Sat, 9 Dec 2017 21:39:18 +0000 (00:39 +0300)]
mbuf: check sanity of data and packet lengths

Update rte_mbuf_sanity_check() to check sanity of data_len and pkt_len
fields. For segmented packets it is supposed that head's pkt_len field
should be the sum of all segments data_len values.

Signed-off-by: Ilya V. Matveychikov <matvejchikov@gmail.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
6 years agombuf: cleanup function to get last segment
Ilya V. Matveychikov [Sat, 9 Dec 2017 21:08:22 +0000 (00:08 +0300)]
mbuf: cleanup function to get last segment

There is no reason to have local variable m2.

Fixes: af75078fece3 ("first public release")

Signed-off-by: Ilya V. Matveychikov <matvejchikov@gmail.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
6 years agopower: clean KVM files
Marko Kovacevic [Tue, 12 Dec 2017 14:03:27 +0000 (14:03 +0000)]
power: clean KVM files

rename private header file rte_power_kvm_vm.c
to power_kvm_vm.c. This prevents the private
functions from leaking into the documentation.
Change any private functions from
rte_<function_name> to just <function_name>.
Reserve the rte_ for public functions

Signed-off-by: Marko Kovacevic <marko.kovacevic@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
6 years agopower: clean ACPI files
Marko Kovacevic [Tue, 12 Dec 2017 14:03:26 +0000 (14:03 +0000)]
power: clean ACPI files

Rename private header file rte_power_acpi_cpufreq.c
to power_acpi_cpufreq.c.This prevents the private
functions from leaking into the documentation.
Change any private functions from rte_<function_name>
to just <function_name>.Reserve the rte_ for public functions.

Signed-off-by: Marko Kovacevic <marko.kovacevic@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
6 years agopower: clean common header
Marko Kovacevic [Tue, 12 Dec 2017 14:03:25 +0000 (14:03 +0000)]
power: clean common header

Rename private header file rte_power_common.h
to power_common.h to prevent private functions
from leaking into the documentation.

Signed-off-by: Marko Kovacevic <marko.kovacevic@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
6 years agopower: changed unsigned to unsigned int
Marko Kovacevic [Tue, 12 Dec 2017 14:03:24 +0000 (14:03 +0000)]
power: changed unsigned to unsigned int

Since this patch-set attempts to clean up the power library,
and there are many instances of "unsigned" caught by checkpatch,
it was decided to clean these up first rather than have them included
in the later patches in the patch set. And would also minimise this
type of error being caught by checkpatch in the future

Signed-off-by: Marko Kovacevic <marko.kovacevic@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
6 years agoservice: fix service core launch
Harry van Haaren [Tue, 9 Jan 2018 13:37:41 +0000 (13:37 +0000)]
service: fix service core launch

This patch fixes a potential bug, which was not consistently
showing up in the unit tests. The issue was that the service-
lcore being started was not in a "WAIT" state, and hence EAL
would return -EBUSY instead of launching the lcore.

In order to ensure a core is in a launch-ready state, the application
must call rte_eal_wait_lcore, to ensure that the core has completed
its previous task, and that EAL is ready to re-launch it.

The call to rte_eal_wait_lcore() is explicitly not in the
service core function, to make it visible to the application.
Requiring an explicit function call ensures the developer sees
that a lcore could block in the rte_eal_wait_lcore() function
if the core hasn't returned from its previous function.

From a usability perspective, hiding the wait_lcore() inside
service cores would cause confusion.

This patch adds rte_eal_wait_lcore() calls to the unit tests,
to ensure that the lcores for testing functionality are ready
to run the test.

Fixes: 21698354c832 ("service: introduce service cores concept")
Cc: stable@dpdk.org
Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
6 years agoservice: fix lcore role after delete
Harry van Haaren [Tue, 9 Jan 2018 13:37:40 +0000 (13:37 +0000)]
service: fix lcore role after delete

This patch fixes the reset of the service core,
that when rte_service_lcore_del() is called, the
lcore_role is restored to RTE.

This issue was reported as when running the unit tests, an
error was thrown that "failed to allocate lcore". Investigating
revealed that the state of the service-cores after del() was
not allowing a core to be re-used at a later point in time.

Fixes: 21698354c832 ("service: introduce service cores concept")
Cc: stable@dpdk.org
Reported-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
6 years agoservice: fix number mapped cores count
Pavan Nikhilesh [Tue, 9 Jan 2018 09:40:41 +0000 (15:10 +0530)]
service: fix number mapped cores count

When adding service the number of mapped cores should only be incremented
when the core is not already a service core or vice versa.

Fixes: 21698354c832 ("service: introduce service cores concept")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
6 years agogro: support VxLAN GRO
Jiayu Hu [Wed, 10 Jan 2018 14:03:12 +0000 (22:03 +0800)]
gro: support VxLAN GRO

This patch adds a framework that allows GRO on tunneled packets.
Furthermore, it leverages that framework to provide GRO support for
VxLAN-encapsulated packets. Supported VxLAN packets must have an outer
IPv4 header, and contain an inner TCP/IPv4 packet.

VxLAN GRO doesn't check if input packets have correct checksums and
doesn't update checksums for output packets. Additionally, it assumes
the packets are complete (i.e., MF==0 && frag_off==0), when IP
fragmentation is possible (i.e., DF==0).

Signed-off-by: Jiayu Hu <jiayu.hu@intel.com>
Reviewed-by: Junjie Chen <junjie.j.chen@intel.com>
Tested-by: Lei Yao <lei.a.yao@intel.com>
6 years agogro: comply RFC 6864 to process IPv4 ID
Jiayu Hu [Wed, 10 Jan 2018 14:03:11 +0000 (22:03 +0800)]
gro: comply RFC 6864 to process IPv4 ID

This patch complies RFC 6864 to process IPv4 ID fields. Specifically, GRO
ingores IPv4 ID fields for the packets whose DF bit is 1, and checks IPv4
ID fields for the packets whose DF bit is 0.

Signed-off-by: Jiayu Hu <jiayu.hu@intel.com>
Reviewed-by: Junjie Chen <junjie.j.chen@intel.com>
6 years agogro: cleanup
Jiayu Hu [Wed, 10 Jan 2018 14:03:10 +0000 (22:03 +0800)]
gro: cleanup

This patch updates codes as follows:
- change appropriate names for internal structures, variants and functions
- update comments and the content of the gro programmer guide for better
  understanding
- remove needless check and redundant comments

Signed-off-by: Jiayu Hu <jiayu.hu@intel.com>
Reviewed-by: Junjie Chen <junjie.j.chen@intel.com>
6 years agoflow_classify: remove table id parameter from API
Jasvinder Singh [Tue, 19 Dec 2017 14:29:17 +0000 (14:29 +0000)]
flow_classify: remove table id parameter from API

This patch removes table id parameter from all the flow
classify apis to reduce the complexity alongwith some code
cleanup.

The validate api is exposed as public api to allow user
to validate the flow before adding it to the classifier.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
6 years agoigb_uio: convert license header to SPDX tag
Ferruh Yigit [Tue, 19 Dec 2017 00:03:14 +0000 (00:03 +0000)]
igb_uio: convert license header to SPDX tag

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agokni: convert license headers to SPDX tags
Ferruh Yigit [Tue, 9 Jan 2018 15:12:28 +0000 (15:12 +0000)]
kni: convert license headers to SPDX tags

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agokni: fix build dependency
Ferruh Yigit [Tue, 12 Dec 2017 18:39:56 +0000 (18:39 +0000)]
kni: fix build dependency

kni library has a dependency to new PCI library, adding that dependency.

build error:
  CC rte_kni.o
In file included from dpdk/lib/librte_kni/rte_kni.c:48:0:
  dpdk/build/include/rte_kni.h:49:21:
  fatal error: rte_pci.h: No such file or directory
    #include <rte_pci.h>
                       ^

Fixes: c752998b5e2e ("pci: introduce library and driver")
Cc: stable@dpdk.org
Reported-by: Bernard Iremonger <bernard.iremonger@intel.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
6 years agodoc: fix typo in KNI sample app guide
Rami Rosen [Fri, 29 Dec 2017 12:08:13 +0000 (14:08 +0200)]
doc: fix typo in KNI sample app guide

This trivial patch fixes a typo in sample apps guide, in
the kernel nic interface (kni) section.
Using "-config", as it is now today in this doc, will not
work and will emit an error; it should be "--config" instead,
and this patch fixes it accordingly.

Fixes: d0dff9ba445e ("doc: sample application user guide")

Signed-off-by: Rami Rosen <rami.rosen@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agodoc: clarify wording for flow rule actions in prog guide
Roy Franz [Fri, 1 Dec 2017 18:26:39 +0000 (10:26 -0800)]
doc: clarify wording for flow rule actions in prog guide

Current wording regarding actions and flow rules doesn't make sense.

Signed-off-by: Roy Franz <roy.franz@cavium.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
6 years agodoc: fix typo in prog guide
Roy Franz [Fri, 1 Dec 2017 18:26:40 +0000 (10:26 -0800)]
doc: fix typo in prog guide

Fix trivial typo (an -> and) in description of service core masks.

Signed-off-by: Roy Franz <roy.franz@cavium.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
6 years agodoc: fix typo in prog guide
Gong Deli [Mon, 27 Nov 2017 09:02:15 +0000 (17:02 +0800)]
doc: fix typo in prog guide

Fix the mismatch of two table's title and content

Fixes: fc1f2750a3ec ("doc: programmers guide")

Signed-off-by: Gong Deli <gnnnnng@gmail.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
6 years agofix typos
Thierry Herbelot [Fri, 15 Dec 2017 12:34:29 +0000 (13:34 +0100)]
fix typos

Repeated occurrences of 'the'.

The change was obtained using the following command:

  sed -i "s;the the ;the ;" `git grep -l "the "`

Signed-off-by: Thierry Herbelot <thierry.herbelot@6wind.com>
6 years agodoc: add metering and policing in testpmd guide
Jasvinder Singh [Mon, 20 Nov 2017 16:39:20 +0000 (16:39 +0000)]
doc: add metering and policing in testpmd guide

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Marko Kovacevic <marko.kovacevic@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
6 years agoapp/testpmd: add commands for metering and policing
Jasvinder Singh [Mon, 20 Nov 2017 16:39:19 +0000 (16:39 +0000)]
app/testpmd: add commands for metering and policing

Add following CLIs to testpmd application;
- show port metering and policing capabilities.
- enable/disable meter of the MTR object.
- update the MTR object dscp table entries.

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
6 years agoapp/testpmd: clean metering and policing commands
Jasvinder Singh [Mon, 20 Nov 2017 16:39:18 +0000 (16:39 +0000)]
app/testpmd: clean metering and policing commands

This patch updates the metering and policing CLIs as follows:
- change name of set port meter CLI to create port meter and add meter
  enable option, dscp table entries arguments, action mask, policer actions
  and previous meter color option as an input color
- set the right metering algorithm in add meter profile CLIs related to
  srtcm(rfc2697) and trtcm(rfc2698,rfc4115)
- change clear flag type from uint32_t to string in show meter stats CLI

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
6 years agoapp/testpmd: register a specific log type
Olivier Matz [Fri, 8 Dec 2017 13:19:10 +0000 (14:19 +0100)]
app/testpmd: register a specific log type

Instead of using USERx logs, register a dynamic log type, as introduced in
commit c1b5fa94a46f ("eal: support dynamic log types").

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
6 years agoapp/testpmd: dump more port info
Olivier Matz [Thu, 14 Dec 2017 14:33:18 +0000 (15:33 +0100)]
app/testpmd: dump more port info

Display some more useful information from dev_info when
calling "show port info <portid>".

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agoapp/testpmd: fix crash of txonly with multiple segments
Yongseok Koh [Wed, 27 Dec 2017 03:51:38 +0000 (19:51 -0800)]
app/testpmd: fix crash of txonly with multiple segments

Running txonly mode can crash with "--txpkts=20,470". copy_len should be
properly adjusted.

Fixes: af75078fece3 ("first public release")
Cc: stable@dpdk.org
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
6 years agoapp/testpmd: fix port id allocation
Yanglong Wu [Tue, 2 Jan 2018 05:35:42 +0000 (13:35 +0800)]
app/testpmd: fix port id allocation

In the feature of increasing port_id range from 8 bits to 16 bits,
vlan port_id allocation function was forget to substitute UINT8 with
UINT16, so the vlan port_id was allocated as a inccrete number.

Fixes: 28caa76aea71 ("app/testpmd: fix port id type")
Cc: stable@dpdk.org
Signed-off-by: Yanglong Wu <yanglong.wu@intel.com>
Reviewed-by: Zhiyong Yang <zhiyong.yang@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
6 years agomem: fix mmap error check on huge page attach
Michael McConville [Fri, 1 Dec 2017 06:51:13 +0000 (23:51 -0700)]
mem: fix mmap error check on huge page attach

mmap(2) returns MAP_FAILED, not NULL, on failure.

Signed-off-by: Michael McConville <mmcco@mykolab.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agocontigmem: fix build on FreeBSD 12
Kefu Chai [Fri, 1 Dec 2017 11:22:39 +0000 (19:22 +0800)]
contigmem: fix build on FreeBSD 12

include <sys/vmmeter.h> to fix build
otherwise the build fails with FreeBSD 12, like

In file included from contigmem.c:57:
/usr/srcs/head/src/sys/vm/vm_phys.h:122:10: error:
use of undeclared identifier 'vm_cnt'
        return (vm_cnt.v_free_count += adj);
                ^

Signed-off-by: Kefu Chai <tchaikov@gmail.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agodoc: use SPDX tag for Cavium copyright files
Jerin Jacob [Mon, 8 Jan 2018 05:25:18 +0000 (10:55 +0530)]
doc: use SPDX tag for Cavium copyright files

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agotools: use SPDX tag for Cavium copyright files
Jerin Jacob [Mon, 8 Jan 2018 05:25:20 +0000 (10:55 +0530)]
tools: use SPDX tag for Cavium copyright files

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agotest: use SPDX tag for Cavium copyright files
Jerin Jacob [Mon, 8 Jan 2018 05:25:12 +0000 (10:55 +0530)]
test: use SPDX tag for Cavium copyright files

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agoexamples: use SPDX tag for Cavium copyright files
Jerin Jacob [Mon, 8 Jan 2018 05:25:15 +0000 (10:55 +0530)]
examples: use SPDX tag for Cavium copyright files

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agoapp: use SPDX tag for Cavium copyright files
Jerin Jacob [Mon, 8 Jan 2018 05:25:17 +0000 (10:55 +0530)]
app: use SPDX tag for Cavium copyright files

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agodrivers: use SPDX tag for Cavium copyright files
Jerin Jacob [Mon, 8 Jan 2018 05:25:16 +0000 (10:55 +0530)]
drivers: use SPDX tag for Cavium copyright files

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agolib: use SPDX tag for Cavium copyright files
Jerin Jacob [Mon, 8 Jan 2018 05:25:14 +0000 (10:55 +0530)]
lib: use SPDX tag for Cavium copyright files

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agomk: use SPDX tag for Cavium copyright files
Jerin Jacob [Mon, 8 Jan 2018 05:25:13 +0000 (10:55 +0530)]
mk: use SPDX tag for Cavium copyright files

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agoconfig: use SPDX tag for Cavium copyright files
Jerin Jacob [Mon, 8 Jan 2018 05:25:19 +0000 (10:55 +0530)]
config: use SPDX tag for Cavium copyright files

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agomaintainers: claim maintainership of net-intel sub tree
Helin Zhang [Tue, 9 Jan 2018 13:50:07 +0000 (21:50 +0800)]
maintainers: claim maintainership of net-intel sub tree

Claim the maintainership of the sub tree dpdk-next-net-intel,
which covers all the Intel PMDs.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
6 years agomaintainers: update mlx drivers sub tree
Shahaf Shuler [Thu, 4 Jan 2018 20:01:07 +0000 (22:01 +0200)]
maintainers: update mlx drivers sub tree

Mellanox PMD patches should target the next-net-mlx sub tree.

Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
6 years agodevtools: ignore checkpatch warning for maintainers file
Thomas Monjalon [Fri, 5 Jan 2018 11:33:20 +0000 (12:33 +0100)]
devtools: ignore checkpatch warning for maintainers file

The script checkpatch.pl from Linux is enforcing a tab
in the MAINTAINERS file (Linux commit 628f91a28649).
It can be ignored in our wrapper checkpatches.sh.

Suggested-by: Remy Horton <remy.horton@intel.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Remy Horton <remy.horton@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
6 years agobus/pci: do not use kernel version to determine MSIX defines
Stephen Hemminger [Wed, 27 Dec 2017 18:25:40 +0000 (10:25 -0800)]
bus/pci: do not use kernel version to determine MSIX defines

In real life, kernel version is only weakly corolated with presence
or absence of defines in header files. Instead, check directly if
the needed value is defined.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
6 years agoexamples/load_balancer: remove unnecessary parenthesis
Stephen Hemminger [Thu, 14 Dec 2017 23:32:21 +0000 (15:32 -0800)]
examples/load_balancer: remove unnecessary parenthesis

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
6 years agodistributor: remove unnecessary parenthesis
Stephen Hemminger [Thu, 14 Dec 2017 23:32:20 +0000 (15:32 -0800)]
distributor: remove unnecessary parenthesis

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
6 years agoethdev: remove redundant parenthesis
Stephen Hemminger [Thu, 14 Dec 2017 23:32:19 +0000 (15:32 -0800)]
ethdev: remove redundant parenthesis

Obvious extra parenthesis do not help readability

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
6 years agokni: remove redundant parenthesis
Stephen Hemminger [Thu, 14 Dec 2017 23:32:18 +0000 (15:32 -0800)]
kni: remove redundant parenthesis

Don't need parens on simple if expression

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
6 years agonet/bnx2x: remove redundant parenthesis
Stephen Hemminger [Thu, 14 Dec 2017 23:32:17 +0000 (15:32 -0800)]
net/bnx2x: remove redundant parenthesis

No need for extra parentheis around simple if expresssion

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
6 years agonet: update licence for network headers
Olivier Matz [Fri, 22 Dec 2017 15:59:02 +0000 (16:59 +0100)]
net: update licence for network headers

To be compliant with the DPDK licensing guidelines, switch to
BSD-3-Clause. It can be done safely since the BSD headers from which
these files derive also exist as a BSD-3-Clause license in FreeBSD.

Link: https://raw.githubusercontent.com/freebsd/freebsd/78a6b0861813af31e1354fa407c5701e8764b4d6/sys/netinet/ip_icmp.h
Link: https://raw.githubusercontent.com/freebsd/freebsd/78a6b0861813af31e1354fa407c5701e8764b4d6/sys/netinet/ip.h
Link: https://raw.githubusercontent.com/freebsd/freebsd/78a6b0861813af31e1354fa407c5701e8764b4d6/sys/netinet/sctp.h
Link: https://raw.githubusercontent.com/freebsd/freebsd/78a6b0861813af31e1354fa407c5701e8764b4d6/sys/netinet/tcp.h
Link: https://raw.githubusercontent.com/freebsd/freebsd/78a6b0861813af31e1354fa407c5701e8764b4d6/sys/netinet/udp.h
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agodoc: use SPDX tag for Intel copyright file
Bruce Richardson [Tue, 19 Dec 2017 15:49:06 +0000 (15:49 +0000)]
doc: use SPDX tag for Intel copyright file

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agotest: use SPDX tag for Intel copyright files
Bruce Richardson [Tue, 19 Dec 2017 15:49:05 +0000 (15:49 +0000)]
test: use SPDX tag for Intel copyright files

Replace the BSD license header with the SPDX tag for files
with only an Intel copyright on them.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agotools: use SPDX tag for Intel copyright files
Bruce Richardson [Tue, 19 Dec 2017 15:49:00 +0000 (15:49 +0000)]
tools: use SPDX tag for Intel copyright files

Replace the BSD license header with the SPDX tag for
scripting files with only an Intel copyright on them.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agoexamples: use SPDX tag for Intel copyright files
Bruce Richardson [Tue, 19 Dec 2017 15:49:02 +0000 (15:49 +0000)]
examples: use SPDX tag for Intel copyright files

Replace the BSD license header with the SPDX tag for files
with only an Intel copyright on them.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agoapp: use SPDX tag for Intel copyright files
Bruce Richardson [Tue, 19 Dec 2017 15:48:59 +0000 (15:48 +0000)]
app: use SPDX tag for Intel copyright files

Replace the BSD license header with the SPDX tag for files
with only an Intel copyright on them.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agodrivers: use SPDX tag for Intel copyright files
Bruce Richardson [Tue, 19 Dec 2017 15:49:01 +0000 (15:49 +0000)]
drivers: use SPDX tag for Intel copyright files

Replace the BSD license header with the SPDX tag for files
with only an Intel copyright on them.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agolib: use SPDX tag for Intel copyright files
Bruce Richardson [Tue, 19 Dec 2017 15:49:03 +0000 (15:49 +0000)]
lib: use SPDX tag for Intel copyright files

Replace the BSD license header with the SPDX tag for files
with only an Intel copyright on them.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agomk: use SPDX tag for Intel copyright files
Bruce Richardson [Tue, 19 Dec 2017 15:49:04 +0000 (15:49 +0000)]
mk: use SPDX tag for Intel copyright files

Replace the BSD license header with the SPDX tag for Makefiles
with only an Intel copyright on them.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agodrivers: use SPDX tag in NXP dpaa files
Hemant Agrawal [Tue, 19 Dec 2017 10:14:41 +0000 (15:44 +0530)]
drivers: use SPDX tag in NXP dpaa files

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agodrivers: use SPDX tag in NXP dpaa2 files
Hemant Agrawal [Tue, 19 Dec 2017 10:14:40 +0000 (15:44 +0530)]
drivers: use SPDX tag in NXP dpaa2 files

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agolicense: use SPDX tag in root makefile
Hemant Agrawal [Tue, 19 Dec 2017 10:14:39 +0000 (15:44 +0530)]
license: use SPDX tag in root makefile

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agolicense: introduce SPDX identifiers
Hemant Agrawal [Tue, 19 Dec 2017 10:14:38 +0000 (15:44 +0530)]
license: introduce SPDX identifiers

The DPDK uses the Open Source BSD-3-Clause license for the core libraries
and drivers. The kernel components are naturally GPLv2 licensed.

Many of the files in the DPDK source code contain the full text of the
applicable license. For example, most of the BSD-3-Clause files contain a
full copy of the BSD-3-Clause license text.

Including big blocks of License headers in all files blows up the source
code with mostly redundant information.  An additional problem is that even
the same licenses are referred to by a number of slightly varying text
blocks (full, abbreviated, different indentation, line wrapping and/or
white space, with obsolete address information, ...) which makes validation
and automatic processing a nightmare.

To make this easier, DPDK uses of a single line reference to
Unique License Identifiers in source files as defined by the Linux
Foundation's SPDX project https://spdk.org.

Adding license information in this fashion, rather than adding full license
text, can be more efficient for developers; decreases errors; and improves
automated detection of licenses. The current set of valid, predefined SPDX
identifiers is set forth on the SPDX License List at
https://spdx.org/licenses/.

For example, to label a file as subject to the BSD-3-Clause license,
the following text would be used as the top line of the file.

SPDX-License-Identifier: BSD-3-Clause

Note: Any new file contributions in DPDK shall adhere to the above scheme.
It is also recommended to replace or at least amend the existing license
text in the code with SPDX-License-Identifiers.

Any exception to DPDK IP policies shall be approved by DPDK tech board and
DPDK Governing Board. Steps for any exception approval:
1. Mention the appropriate license identifier form SPDX. If the license is
   not listed in SPDX Licenses. It is the submitters responsibiliity to get
   it first listed.
2. Get the required approval from the DPDK Technical Board. Technical board
   may advise the author to check alternate means first. If no other
   alternatives are found and the merit of the contributions are important
   for DPDK's mission, it may decide on such exception with two-thirds vote
   of the members.
3. Technical board then approach Governing board for such limited approval
   for the given contribution only.

Any approvals shall be documented in "licenses/exceptions.txt" with record
dates.

Note: From the legal point of view, this patch is supposed to be only a
change to the textual representation of the license information, but in no
way any change to the actual license terms. With this patch applied, all
files will still be licensed under the same terms they were before.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
6 years agodevtools: add git log checks for SCTP and IOVA
Ferruh Yigit [Sat, 16 Dec 2017 02:22:41 +0000 (02:22 +0000)]
devtools: add git log checks for SCTP and IOVA

Acronyms must be uppercase.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agodevtools: enhance error message in get maintainer script
Olivier Matz [Thu, 14 Dec 2017 14:21:40 +0000 (15:21 +0100)]
devtools: enhance error message in get maintainer script

When the DPDK_GETMAINTAINER_PATH variable was not set, the error
message from get-maintainer.sh was quite cheap:

$ devtools/get-maintainer.sh --help
usage: get-maintainer.sh <patch>
Cannot execute DPDK_GETMAINTAINER_PATH

Add a more detailed explanation about this variable in the help.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agomaintainers: claim responsibility for virtio PMD
Tiwei Bie [Fri, 1 Dec 2017 11:04:35 +0000 (19:04 +0800)]
maintainers: claim responsibility for virtio PMD

Add myself as co-maintainer for virtio driver.

Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Acked-by: Yuanhan Liu <yliu@fridaylinux.org>
Acked-by: Maxime Coquelin <maxime.coquelin@redhat.com>
6 years agomaintainers: update for i40e
Jingjing Wu [Mon, 4 Dec 2017 06:15:51 +0000 (14:15 +0800)]
maintainers: update for i40e

Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
6 years agomaintainers: update mrvl crypto
Tomasz Duszynski [Thu, 14 Dec 2017 09:43:58 +0000 (10:43 +0100)]
maintainers: update mrvl crypto

Fix typo in email address.

Fixes: 8a61c83af2fa ("crypto/mrvl: add mrvl crypto driver")

Signed-off-by: Tomasz Duszynski <tdu@semihalf.com>
6 years agomaintainers: update for testpmd
Jingjing Wu [Mon, 4 Dec 2017 06:18:26 +0000 (14:18 +0800)]
maintainers: update for testpmd

Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
6 years agomaintainers: claim co-maintainership of docs
Marko Kovacevic [Tue, 5 Dec 2017 15:22:07 +0000 (15:22 +0000)]
maintainers: claim co-maintainership of docs

Signed-off-by: Marko Kovacevic <marko.kovacevic@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
6 years agomaintainers: resign from maintenance
Sergio Gonzalez Monroy [Mon, 18 Dec 2017 14:02:06 +0000 (14:02 +0000)]
maintainers: resign from maintenance

I will not be directly working on the DPDK project anymore.

Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agobus/dpaa: fix build when assert enabled
Ferruh Yigit [Fri, 22 Dec 2017 02:13:16 +0000 (02:13 +0000)]
bus/dpaa: fix build when assert enabled

Enabling CONFIG_RTE_ENABLE_ASSERT cause build error because some fields
protected by RTE_LIBRTE_DPAA_HWDEBUG macro accessed within DPAA_ASSERT.

All DPAA_ASSERT macro usage causing the build error wrapped with
RTE_LIBRTE_DPAA_HWDEBUG macro.

Fixes: f38f61e982f8 ("bus/dpaa: add BMAN hardware interfaces")
Fixes: c47ff048b99a ("bus/dpaa: add QMAN driver core routines")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agoeal: update assertion macro
Ilya V. Matveychikov [Sun, 19 Nov 2017 08:18:49 +0000 (12:18 +0400)]
eal: update assertion macro

Update RTE_VERIFY macro to make it possible to use complex expressions
in RTE_ASSERT.

Now it’s possible to have % char inside the expression, for example:
RTE_ASSERT((sizeof(some_struct) % 64) == 0)

Before the patch, “%" sign acts like a conversion specification
beginning character.

Fixes: 148f963fb532 ("xen: core library changes")

Signed-off-by: Ilya V. Matveychikov <matvejchikov@gmail.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agomk: remove TILE-Gx machine type
Ferruh Yigit [Fri, 15 Dec 2017 23:09:49 +0000 (23:09 +0000)]
mk: remove TILE-Gx machine type

This file was forgotten when removing TILE-Gx arch.

Fixes: 31123211bd27 ("remove unmaintained TILE-Gx architecture")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agodoc: update oldest supported kernel in Linux guide
Stephen Hemminger [Tue, 26 Sep 2017 07:17:12 +0000 (08:17 +0100)]
doc: update oldest supported kernel in Linux guide

The DPDK needs to stay up to date with current LTS Linux kernel support.
If the kernel is older than LTS it is likely to be insecure and buggy.
Therefore only require DPDK to work on oldest LTS kernel.

If distribution vendors want to support DPDK on older kernels, that is
their choice. But the upstream source does not need to be cluttered
with support for this.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Marko Kovacevic <marko.kovacevic@intel.com>
6 years agoapp/testpmd: remove xenvirt again
Timothy Redaelli [Fri, 17 Nov 2017 14:40:17 +0000 (15:40 +0100)]
app/testpmd: remove xenvirt again

The xenvirt PMD was removed from 17.11
while this link parameter was added.

Fixes: 5b590fbe09b6 ("app/testpmd: add traffic management forwarding mode")
Cc: stable@dpdk.org
Signed-off-by: Timothy Redaelli <tredaelli@redhat.com>
Acked-by: Luca Boccassi <bluca@debian.org>
Acked-by: Eelco Chaudron <echaudro@redhat.com>
6 years agokni: fix build with kernel 4.15
Ferruh Yigit [Tue, 28 Nov 2017 23:45:53 +0000 (23:45 +0000)]
kni: fix build with kernel 4.15

build error:
.../dpdk/build/build/lib/librte_eal/linuxapp/kni/igb_main.c:2809:2:
 error: implicit declaration of function ‘setup_timer’;
 did you mean ‘sk_stop_timer’? [-Werror=implicit-function-declaration]
  setup_timer(&adapter->watchdog_timer, &igb_watchdog,
  ^~~~~~~~~~~
  sk_stop_timer
cc1: all warnings being treated as errors

error observed whed CONFIG_RTE_KNI_KMOD_ETHTOOL config option enabled.

Because Linux removed setup_timer macros for kernel version >= 4.15
Linux: 513ae785c63c ("timer: Remove setup_*timer() interface")

Replaced setup_timer with timer_setup for new kernel versions.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agoversion: 18.02-rc0
Thomas Monjalon [Thu, 23 Nov 2017 22:33:52 +0000 (23:33 +0100)]
version: 18.02-rc0

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
6 years agodoc: add template release notes for 18.02
John McNamara [Wed, 22 Nov 2017 17:38:21 +0000 (17:38 +0000)]
doc: add template release notes for 18.02

Add template release notes for DPDK 18.02 with inline
comments and explanations of the various sections.

Signed-off-by: John McNamara <john.mcnamara@intel.com>
6 years agoversion: 17.11.0
Thomas Monjalon [Wed, 15 Nov 2017 00:24:53 +0000 (01:24 +0100)]
version: 17.11.0

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
6 years agodoc: fix GSO guide as PDF
Thomas Monjalon [Wed, 15 Nov 2017 17:54:07 +0000 (18:54 +0100)]
doc: fix GSO guide as PDF

The PDF cannot be built because of some images integration being
forced as SVG. They are converted for PDF format, so the extension
must be a wildcard in the RST file.

Fixes: f6010c7655cc ("doc: add GSO programmer's guide")

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
6 years agonet/mlx5: fix flow mark action on port start
Shahaf Shuler [Wed, 15 Nov 2017 07:09:49 +0000 (09:09 +0200)]
net/mlx5: fix flow mark action on port start

Mark action flag was not set on the flow, causing the rx burst function
after port start to ignore the flow mark.

Fixes: 8086cf08b2f0 ("net/mlx5: handle RSS hash configuration in RSS flow")

Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
6 years agomaintainers: move NXP bus drivers in related section
Thomas Monjalon [Wed, 15 Nov 2017 00:42:24 +0000 (01:42 +0100)]
maintainers: move NXP bus drivers in related section

A bus maintainers section has been introduced in 17.11.
Let's move all bus drivers in the same place.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
6 years agodoc: add bus libraries to release notes
Thomas Monjalon [Wed, 15 Nov 2017 00:35:34 +0000 (01:35 +0100)]
doc: add bus libraries to release notes

Fixes: 919eeaccb2ba ("bus/dpaa: introduce NXP DPAA bus driver skeleton")
Fixes: 7e7df6d0a41d ("bus/fslmc: introduce fsl-mc bus driver")
Fixes: c752998b5e2e ("pci: introduce library and driver")

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
6 years agodoc: announce behaviour change to i40e RSS
John McNamara [Fri, 3 Nov 2017 15:24:23 +0000 (15:24 +0000)]
doc: announce behaviour change to i40e RSS

Add deprecation notice to highlight a future change in the
default configuration behavior for flexible payload for RSS.

Signed-off-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Andrey Chilikin <andrey.chilikin@intel.com>
6 years agopdump: fix typo in error log
Maria Lingemark [Tue, 14 Nov 2017 08:04:24 +0000 (09:04 +0100)]
pdump: fix typo in error log

Fixes: 278f945402c5 ("pdump: add new library for packet capture")

Signed-off-by: Maria Lingemark <maria.lingemark@ericsson.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agombuf: fix segments number type increase
Ilya V. Matveychikov [Fri, 10 Nov 2017 13:56:43 +0000 (16:56 +0300)]
mbuf: fix segments number type increase

Update types of variables to correspond to nb_segs type change from
uint8_t to uint16_t.

Fixes: 97cb466d65c9 ("mbuf: use 2 bytes for port and nb segments")
Cc: stable@dpdk.org
Signed-off-by: Ilya V. Matveychikov <matvejchikov@gmail.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
6 years agonet/sfc: avoid Tx queue setup failure if thresholds are set
Andrew Rybchenko [Tue, 14 Nov 2017 11:32:35 +0000 (11:32 +0000)]
net/sfc: avoid Tx queue setup failure if thresholds are set

Tx queue prefetch, host and writeback thresholds are used for
performance fine-tuning and not applicable to Solarflare NICs.
It is safe to just log warning and do not fail Tx queue setup
to be more friendly to DPDK applications which hardcode it.

Fixes: b1b7ad933b39 ("net/sfc: set up and release Tx queues")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>