Jerin Jacob [Thu, 12 Jan 2017 07:46:54 +0000 (13:16 +0530)]
app/testpmd: fix static build link ordering
By introducing explicit -lrte_pmd_ixgbe link request in
testpmd Makefile,"-Wl,-lrte_pmd_ixgbe" provided twice, and linker
removes the duplication by keeping only first occurrence.
This moves "-Wl,-lrte_pmd_ixgbe" out of "-Wl,--whole-archive" flag
and makes symbol generation totally different than previous version
in case of static build.
This patch fixes the static build linking order by introducing
-lrte_pmd_ixgbe under the shared library config
(CONFIG_RTE_BUILD_SHARED_LIB).
Bruce Richardson [Fri, 13 Jan 2017 13:02:17 +0000 (13:02 +0000)]
devtools: skip capitalization check for commit prefixes
The prefix in the commit title must be a valid component name and is
checked in separate checks. For capitalization, just check the part after
the colon. This is already done for most capitalization checks, just make
the remainder consistent with this.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Wenfeng Liu [Wed, 11 Jan 2017 02:25:28 +0000 (02:25 +0000)]
mempool: use cache in single producer or consumer mode
Currently we will check mempool flags when we put/get objects from
mempool. However, this makes cache useless when mempool is SC|SP,
SC|MP, MC|SP cases.
This patch makes cache available in above cases and improves performance.
Signed-off-by: Wenfeng Liu <liuwf@arraynetworks.com.cn> Acked-by: Olivier Matz <olivier.matz@6wind.com>
Ben Walker [Wed, 11 Jan 2017 17:10:11 +0000 (10:10 -0700)]
pci: separate detaching ethernet ports from PCI devices
Attaching and detaching ethernet ports from an application
is not the same thing as physically removing a PCI device,
so clarify the flags indicating support. All PCI devices
are assumed to be physically removable, so no flag is
necessary in the PCI layer.
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Adrien Mazarguil [Tue, 10 Jan 2017 13:08:30 +0000 (14:08 +0100)]
ethdev: define default item masks in flow API
Leaving default pattern item mask values up for interpretation by PMDs is
an undefined behavior that applications might find difficult to use in the
wild. It also needlessly complicates PMD implementation.
This commit addresses this by defining consistent default masks for each
item type.
Adrien Mazarguil [Tue, 10 Jan 2017 13:08:29 +0000 (14:08 +0100)]
ethdev: clarify RSS action in flow API
Contrary to the current description, mbuf RSS hash result storage does not
overlap with the returned MARK value (hash.fdir.lo vs. hash.fdir.hi), and
both may be combined.
Reflect this change by allowing testpmd to display both values
simultaneously.
Adrien Mazarguil [Tue, 10 Jan 2017 13:08:28 +0000 (14:08 +0100)]
ethdev: clarify MARK and FLAG actions in flow API
Both actions share the PKT_RX_FDIR mbuf flag, as a result there is no way
to tell them apart. Moreover, the maximum allowed value for the MARK action
may not necessarily cover the entire 32-bit space.
Neil Horman [Thu, 5 Jan 2017 19:22:41 +0000 (14:22 -0500)]
pmdinfogen: fix null dereference
Coverity reports a forward null dereference from a for loop
that works with a variable previously tested for null that had no error
handling or condition to prevent it. Pretty obvious fix below.
Coverity issue: 139593 Fixes: 98b0fdb0ffc6 ("pmdinfogen: add buildtools and pmdinfogen utility") Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Tomasz Kulasek [Fri, 23 Dec 2016 18:40:54 +0000 (19:40 +0100)]
app/testpmd: use Tx preparation in checksum engine
Since all current drivers supports Tx preparation API, it is used
in csum forwarding engine by default for all drivers.
Adding additional step to the csum engine costs about 3-4% of performance
drop, on my setup with ixgbe driver. It's caused mostly by the need
of reaccessing and modification of packet data.
Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
uint16_t nb_seg_max;
/**< Max number of segments per whole packet. */
uint16_t nb_mtu_seg_max;
/**< Max number of segments per one MTU */
These fields can be used to create valid packets according to the
following rules:
* For non-TSO packet, a single transmit packet may span up to
"nb_mtu_seg_max" buffers.
* For TSO packet the total number of data descriptors is "nb_seg_max",
and each segment within the TSO may span up to "nb_mtu_seg_max".
Added functions:
int
rte_validate_tx_offload(struct rte_mbuf *m)
to validate general requirements for tx offload set in mbuf of packet
such a flag completness. In current implementation this function is
called optionaly when RTE_LIBRTE_ETHDEV_DEBUG is enabled.
int rte_net_intel_cksum_prepare(struct rte_mbuf *m)
to prepare pseudo header checksum for TSO and non-TSO tcp/udp packets
before hardware tx checksum offload.
- for non-TSO tcp/udp packets full pseudo-header checksum is
counted and set.
- for TSO the IP payload length is not included.
int
rte_net_intel_cksum_flags_prepare(struct rte_mbuf *m, uint64_t ol_flags)
this function uses same logic as rte_net_intel_cksum_prepare, but
allows application to choose which offloads should be taken into
account, if full preparation is not required.
PERFORMANCE TESTS
-----------------
This feature was tested with modified csum engine from test-pmd.
The packet checksum preparation was moved from application to Tx
preparation step placed before burst.
We may expect some overhead costs caused by:
1) using additional callback before burst,
2) rescanning burst,
3) additional condition checking (packet validation),
4) worse optimization (e.g. packet data access, etc.)
We tested it using ixgbe Tx preparation implementation with some parts
disabled to have comparable information about the impact of different
parts of implementation.
IMPACT:
1) For unimplemented Tx preparation callback the performance impact is
negligible,
2) For packet condition check without checksum modifications (nb_segs,
available offloads, etc.) is 14626628/14252168 (~2.62% drop),
3) Full support in ixgbe driver (point 2 + packet checksum
initialization) is 14060924/13588094 (~3.48% drop)
Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Olivier Matz [Fri, 16 Dec 2016 09:44:13 +0000 (10:44 +0100)]
ethdev: fix extended statistics name index
The function rte_eth_xstats_get() return an array of tuples (id,
value). The value is the statistic counter, while the id references a
name in the array returned by rte_eth_xstats_get_name().
Today, each 'id' returned by rte_eth_xstats_get() is equal to the index
in the returned array, making this value useless. It also prevents a
driver from having different indexes for names and value, like in the
example below:
This patch fixes the drivers to set the 'id' in their ethdev->xstats_get()
(except e1000 which was already doing it), and fixes ethdev by not setting
the 'id' field to the index of the table for pmd-specific stats: instead,
they should just be shifted by the max number of generic statistics.
Jan Blunck [Fri, 23 Dec 2016 15:58:09 +0000 (16:58 +0100)]
ethdev: decouple interrupt handling from PCI device
The struct rte_intr_handle is an abstraction layer for different types of
interrupt mechanisms. It is embedded in the low-level device (e.g. PCI).
On allocation of a struct rte_eth_dev a reference to the intr_handle
should be stored for devices supporting interrupts.
Signed-off-by: Jan Blunck <jblunck@infradead.org> Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
Jan Blunck [Fri, 23 Dec 2016 15:58:02 +0000 (16:58 +0100)]
net/virtio: do not depend on PCI device of ethdev
We don't need to depend on rte_eth_dev->pci_dev to differentiate between
the virtio_user and the virtio_pci case. Instead we can use the private
virtio_hw struct to get that information.
Signed-off-by: Jan Blunck <jblunck@infradead.org> Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
Jan Blunck [Fri, 23 Dec 2016 15:58:01 +0000 (16:58 +0100)]
net/virtio: add helper to get interrrupt handle
This adds a helper to get the rte_intr_handle from the virtio_hw. This is
safe to do since the usage of the helper is guarded by RTE_ETH_DEV_INTR_LSC
which is only set if we found a PCI device during initialization.
Signed-off-by: Jan Blunck <jblunck@infradead.org> Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
drivers: remove useless reset of PCI device pointer
Since rte_eth_dev_info_get does memset() on dev_info before
calling device specific code, the explicit assignment of NULL
in all these virtual drivers has no effect.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Jan Blunck <jblunck@infradead.org>
Jan Blunck [Fri, 23 Dec 2016 15:57:52 +0000 (16:57 +0100)]
eal: define container_of macro
This macro is based on Jan Viktorin's original patch but also checks the
type of the passed pointer against the type of the member.
Signed-off-by: Jan Viktorin <viktorin@rehivetech.com> Signed-off-by: Shreyansh Jain <shreyansh.jain@nxp.com>
[jblunck@infradead.org: add type checking and __extension__] Signed-off-by: Jan Blunck <jblunck@infradead.org>
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:42 +0000 (15:51 +0100)]
app/testpmd: add protocol fields to flow command
This commit exposes the following item fields through the flow command:
- VLAN priority code point, drop eligible indicator and VLAN identifier
(all part of TCI).
- IPv4 type of service, time to live and protocol.
- IPv6 traffic class, flow label, next header and hop limit.
- SCTP tag and checksum.
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:39 +0000 (15:51 +0100)]
app/testpmd: add various actions to flow command
- MARK: attach 32 bit value to packets.
- FLAG: flag packets.
- DROP: drop packets.
- COUNT: enable counters for a rule.
- PF: redirect packets to physical device function.
- VF: redirect packets to virtual device function.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com> Acked-by: Olga Shern <olgas@mellanox.com>
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:35 +0000 (15:51 +0100)]
app/testpmd: add item raw to flow command
Matches arbitrary byte strings with properties:
- relative: look for pattern after the previous item.
- search: search pattern from offset (see also limit).
- offset: absolute or relative offset for pattern.
- limit: search area limit for start of pattern.
- length: pattern length.
- pattern: byte string to look for.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com> Acked-by: Olga Shern <olgas@mellanox.com>
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:34 +0000 (15:51 +0100)]
app/testpmd: add various items to flow command
- PF: match packets addressed to the physical function.
- VF: match packets addressed to a virtual function ID.
- PORT: device-specific physical port index to use.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com> Acked-by: Olga Shern <olgas@mellanox.com>
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:30 +0000 (15:51 +0100)]
app/testpmd: add flow item spec handler
Add parser code to fully set individual fields of pattern item
specification structures, using the following operators:
- fix: sets field and applies full bit-mask for perfect matching.
- spec: sets field without modifying its bit-mask.
- last: sets upper value of the spec => last range.
- mask: sets bit-mask affecting both spec and last from arbitrary value.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com> Acked-by: Olga Shern <olgas@mellanox.com>
Either check the validity of a flow rule or create it. Any number of
pattern items and actions can be provided in any order. Completion is
available for convenience.
This commit only adds support for the most basic item and action types,
namely:
- END: terminates pattern items and actions lists.
- VOID: item/action filler, no operation.
- INVERT: inverted pattern matching, process packets that do not match.
- PASSTHRU: action that leaves packets up for additional processing by
subsequent flow rules.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com> Acked-by: Olga Shern <olgas@mellanox.com>
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:23 +0000 (15:51 +0100)]
app/testpmd: add flow command
Managing generic flow API functions from command line requires the use of
dynamic tokens for convenience as flow rules are not fixed and cannot be
defined statically.
This commit adds specific flexible parser code and object for a new "flow"
command in separate file.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com> Acked-by: Olga Shern <olgas@mellanox.com>
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:22 +0000 (15:51 +0100)]
app/testpmd: implement basic support for flow API
Add basic management functions for the generic flow API (validate, create,
destroy, flush, query and list). Flow rule objects and properties are
arranged in lists associated with each port.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com> Acked-by: Olga Shern <olgas@mellanox.com>
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:20 +0000 (15:51 +0100)]
cmdline: support dynamic tokens
Considering tokens must be hard-coded in a list part of the instruction
structure, context-dependent tokens cannot be expressed.
This commit adds support for building dynamic token lists through a
user-provided function, which is called when the static token list is empty
(a single NULL entry).
Because no structures are modified (existing fields are reused), this
commit has no impact on the current ABI.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com> Acked-by: Olga Shern <olgas@mellanox.com>
Adrien Mazarguil [Wed, 21 Dec 2016 14:51:17 +0000 (15:51 +0100)]
ethdev: introduce generic flow API
This new API supersedes all the legacy filter types described in
rte_eth_ctrl.h. It is slightly higher level and as a result relies more on
PMDs to process and validate flow rules.
Benefits:
- A unified API is easier to program for, applications do not have to be
written for a specific filter type which may or may not be supported by
the underlying device.
- The behavior of a flow rule is the same regardless of the underlying
device, applications do not need to be aware of hardware quirks.
- Extensible by design, API/ABI breakage should rarely occur if at all.
- Documentation is self-standing, no need to look up elsewhere.
Existing filter types will be deprecated and removed in the near future.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com> Acked-by: Olga Shern <olgas@mellanox.com>
Ferruh Yigit [Thu, 22 Dec 2016 13:10:13 +0000 (13:10 +0000)]
ethdev: cleanup device ops struct whitespace
- Grouped related items using empty lines
- Aligned arguments to same column
- All item comments that doesn't fit same line are placed blow the item
itself
- Moved some comments to same line if overall line < 100 chars
It is necessary to reconfigure all queues every time because configuration
can be changed.
For example, if we're reconfiguring bonding device with new memory pool,
already configured queues will still use the old one. And if the old
mempool be freed, application likely will panic in attempt to use
freed mempool.
This happens when we use the bonding device with OVS 2.6 while MTU
reconfiguration:
Jan Blunck [Thu, 24 Nov 2016 11:26:48 +0000 (12:26 +0100)]
net/bonding: force reconfiguration of removed slaves
After a slave interface is removed from a bond group it still has the
configuration of the bond interface. Lets enforce that the slave interface
is reconfigured after removal by resetting it.
Jan Blunck [Thu, 17 Nov 2016 17:16:12 +0000 (18:16 +0100)]
ethdev: initialize more fields on allocation
This moves the non-PCI related initialization of the link state interrupt
callback list and the setting of the default MTU to rte_eth_dev_allocate()
so that drivers only need to set non-default values.
Signed-off-by: Jan Blunck <jblunck@infradead.org> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Jerin Jacob [Sun, 20 Nov 2016 08:00:50 +0000 (13:30 +0530)]
eal: postpone vdev initialization
Some platform like octeontx may use pci and
vdev based combined device to represent a logical
dpdk functional device.In such case, postponing the
vdev initialization after pci device
initialization will provide the better view of
the pci device resources in the system in
vdev's probe function, and it allows better
functional subsystem registration in vdev probe
function.
As a bonus, This patch fixes a bond device
initialization use case.
example command to reproduce the issue:
./testpmd -c 0x2 --vdev 'eth_bond0,mode=0,
slave=0000:02:00.0,slave=0000:03:00.0' --
--port-topology=chained
root cause:
In existing case(vdev initialization and then pci
initialization), creates three Ethernet ports with
following port ids
0 - Bond device
1 - PCI device 0
2 - PCI devive 1
Since testpmd, calls the configure/start on all the ports on
start up,it will translate to following illegal setup sequence
Jianfeng Tan [Fri, 2 Dec 2016 17:48:56 +0000 (17:48 +0000)]
eal: restrict cores auto detection
This patch uses pthread_getaffinity_np() to narrow down used
cores when none of below options is specified:
* coremask (-c)
* corelist (-l)
* and coremap (--lcores)
The purpose of this patch is to leave out these core related options
when DPDK applications are deployed under container env, so that
users do not need decide the core related parameters when developing
applications. Instead, when applications are deployed in containers,
use cpu-set to constrain which cores can be used inside this container
instance. And DPDK application inside containers just rely on this
auto detect mechanism to start polling threads.
Note: previously, some users are using isolated CPUs, which could
be excluded by default. Please add commands like taskset to use
those cores.
Test example:
$ taskset 0xc0000 ./examples/helloworld/build/helloworld -m 1024
Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Aaron Conole [Thu, 8 Dec 2016 15:56:45 +0000 (10:56 -0500)]
eal: clarify the argc and argv documentation
It's been a source of confusion in the past, and even with this update
may continue to be a source of confusion. However, the original
language seems to imply that the DPDK EAL will take ownership of the
array passed in. Loosening the language up a bit might give a better
understanding for what is actually happening.
Qiming Yang [Tue, 22 Nov 2016 01:41:24 +0000 (09:41 +0800)]
examples/ethtool: fix driver information
Function pcmd_drvinfo_callback uses struct info to get
the ethtool information of each port. Struct info will
store the information of previous port until this
information be updated. This patch fixes this issue.
Remy Horton [Wed, 30 Nov 2016 02:47:29 +0000 (10:47 +0800)]
examples/ethtool: fix querying non-PCI devices
Doing a device information query on a non-PCI device such as
vhost was resulting in the dereferencing of a NULL pointer
(the absent PCI data), causing a segmentation fault.
git.droids-corp.org / dpdk.git / log