Ori Kam [Mon, 24 Sep 2018 23:17:35 +0000 (23:17 +0000)]
net/mlx5: split flow validation to dedicated function
In current implementation the validation logic reside in the same
function that calculates the size of the verbs spec and also create the
verbs spec.
This approach results in hard to maintain code which can't be shared.
also in current logic there is a use of parser entity that holds the
information between function calls. The main problem with this parser is
that it assumes the connection between different functions. For example
it assumes that the validation function was called and relevant values
were set.
This may result in an issue if and when we only call the validation
function, or call the apply function without the validation (Currently
according to RTE flow we must call validation before creating flow, but
if we want to change that to save time during flow creation, for example
the user validated some rule and just want to change the IP there is no
true reason the validate the rule again).
This commit address both of those issues by extracting the validation
logic into detected functions and remove the use of the parser object.
The side effect of those changes is that in some cases there will be a
need to traverse the item list again.
Signed-off-by: Ori Kam <orika@mellanox.com> Acked-by: Yongseok Koh <yskoh@mellanox.com>
Hyong Youb Kim [Wed, 3 Oct 2018 20:09:28 +0000 (13:09 -0700)]
net/enic: add AVX2 based vectorized Rx handler
Add the vectorized version of the no-scatter Rx handler. It aims to
process 8 descriptors per loop using AVX2 SIMD instructions. This
handler is in its own file enic_rxtx_vec_avx2.c, and makefile and
meson.build are modified to compile it when the compiler supports
AVX2. Under ideal conditions, the vectorized handler reduces
cycles/packet by more than 30%, when compared against the no-scatter
Rx handler. Most implementation ideas come from i40e's AVX2 based
handler, so credit goes to its authors.
At this point, the new handler is meant for field trials, and is not
selected by default. So add a new devarg enable-avx2-rx to allow the
user to request the use of the new handler. When enable-avx2-rx=1, the
driver will consider using the new handler.
Also update the guide doc and introduce the vectorized handler.
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com> Reviewed-by: John Daley <johndale@cisco.com>
Xiaoyun Li [Sat, 29 Sep 2018 02:12:04 +0000 (10:12 +0800)]
ethdev: get Rx queue interrupt fd
Some users want to use their own epoll instances to control both
DPDK rxq interrupt fds and their own other fds. So added a function
to get rxq interrupt fd based on port id and queue id.
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Adrien Mazarguil [Fri, 31 Aug 2018 09:01:11 +0000 (11:01 +0200)]
ethdev: add missing items/actions to flow object converter
Several pattern items and actions were never handled by rte_flow_copy()
because their descriptions were missing. rte_flow_conv() inherited this
deficiency.
This patch adds them and reorders others to match rte_flow.h. It doesn't
pose as a fix because so far no one has complained about it and
rte_flow_conv() would have to be backported as well: this function is
the only sane approach to handle VXLAN and NVGRE encap definitions.
As a matter of fact, it's the last missing piece to finally allow
testpmd users to request the creation of VXLAN/NVGRE encap/decap flow
rules without getting rejected outright.
Adrien Mazarguil [Fri, 31 Aug 2018 09:01:00 +0000 (11:01 +0200)]
ethdev: add flow API object converter
rte_flow_copy() is bound to duplicate flow rule descriptions
(attributes, pattern and list of actions, all at once), however
applications sometimes need more flexibility, for instance the ability
to duplicate only one of the underlying objects (a single pattern item
or action) or retrieve other properties such as their names.
Instead of adding dedicated functions to handle each possible use case,
this patch introduces rte_flow_conv(), which supports any number of
object conversion operations in an extensible manner.
This patch re-implements rte_flow_copy() as a wrapper to
rte_flow_conv().
Igor Romanov [Wed, 3 Oct 2018 09:03:56 +0000 (10:03 +0100)]
net/sfc: support Rx descriptor status on EF10 datapath
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru> Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com> Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
net/sfc: avoid usage of prepared packets number in EF10 Rx
Number of prepared packets is good when one Rx descriptor is one packet.
Introduce pending Rx descriptor pointer which points to the first not
processed Rx descriptors. Rx descriptors from completed to pending have
buffers ready to be passed to application.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com> Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Xiaolong Ye [Fri, 28 Sep 2018 21:47:47 +0000 (05:47 +0800)]
examples/vdpa: introduce a new sample for vDPA
The vdpa sample application creates vhost-user sockets by using the
vDPA backend. vDPA stands for vhost Data Path Acceleration which utilizes
virtio ring compatible devices to serve virtio driver directly to enable
datapath acceleration. As vDPA driver can help to set up vhost datapath,
this application doesn't need to launch dedicated worker threads for vhost
enqueue/dequeue operations.
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com> Acked-by: Xiao Wang <xiao.w.wang@intel.com> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Xiaolong Ye [Fri, 28 Sep 2018 21:47:46 +0000 (05:47 +0800)]
vhost: introduce API to get vDPA device number
It's used to get number of available registered vDPA devices.
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com> Acked-by: Xiao Wang <xiao.w.wang@intel.com> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Gagandeep Singh [Wed, 3 Oct 2018 13:36:08 +0000 (19:06 +0530)]
doc: add guide for ENETC PMD
Add enetc usage document to compile and run the
DPDK application on enetc supported platform.
This document introduces the enetc driver, supported
platforms and supported features.
net/qede/base: add APIs for dscp priority map configuration
Add APIs for dscp priority map configuration. APIs added are
ecore_dcbx_get_dscp_priority(), ecore_dcbx_set_dscp_priority().
These base driver APIs can be used for dscp-map query/config.
Configure the doorbell queue (DORQ) to use vlan-id/priority.
- Get engine affinity from the management FW and configure accordingly
- Add an LLH filter with the primary MAC address in QPAR/NPAR
- Move some of the LLH APIs around
- Add PPFID APIs
- Update all allocated ppfids with the same value for the
following PORT_PF registers:
NIG_REG_DSCP_TO_TC_MAP_ENABLE
- Add port_id, src_pfid and dst_pfid to DMA engine params
Request management FW for OEM values, which are negotiated prior to
the driver load by sending the GET_OEM_UPDATES command after both
engines are initialized.
net/qede/base: workaround to indicate SHMEM data ready
The driver can notify that there was an MCP reset and read the SHMEM
values before the management FW has completed initializing them.
As a temporary solution, the "sup_msgs" field is used as a SHMEM data
ready indication. This should be replaced with an actual indication
when it is provided by the management FW.
Add DPC sync after stopping the physical funciton to allow clean up of
asyncronous events. Post this the driver don't expect the FW to send
async events.
When the firmware version and the driver HWRM version do not match,
we are logging some messages. These messages unnecessarily clutter
the logs and can add to the noise. We are logging the HWRM version
and the firmware version anyway. The difference in version numbers
can be gleaned from that. Removing the remaining log messages.
Bei Sun [Sat, 29 Sep 2018 02:00:04 +0000 (19:00 -0700)]
net/bnxt: set VLAN strip mode before default VNIC cfg
Firmware sets pf pair in default VNIC cfg. If the VLAN strip
setting is not available at this time, it will not be
configured correctly in the CFA.
Set the desired VLAN strip mode before default VNIC configuration.
Fixes: 7fe5668d2ea3 ("net/bnxt: support VLAN filter and strip") Cc: stable@dpdk.org Signed-off-by: Bei Sun <bei.sun@broadcom.com> Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
net/bnxt: fix registration of VF async event completion ring
While registering vf's event completion ring to FW, use the wrong
macro, FW doesn't set up the event completion ring successfully,
VF can't receive any async event.
In the current model, VFs are not trusted.
So it is not allowed to send many HWRM commands.
Newer firmware has added support to allow VF to be trusted.
Now the VF queries if it is a trusted entity and based on that
it can send HWRM commands to the firmware.
For hardware having multiple embedded management processors the firmware
has added support to indicate if the comm channel to the processor has
been enabled. If the channel is enabled, switch the CFA NTUPLE and EM
filtering commands to use the kong channel.
bnxt_start_xmit() was attempting to access an uninitialized ptr - txbd1
which would lead to segmentation fault.
Fix to initialize ptr to NULL and check for the same before access.
John Daley [Fri, 28 Sep 2018 03:08:37 +0000 (20:08 -0700)]
net/enic: support flow counter action
Support counter action for 1400 series adapters.
The adapter API for allocating and freeing counters is independent of
the adapter match/action API. If the filter action is requested, a
counter is first allocated and then assigned to the filter, and when
the filter is deleted, the counter must also be deleted.
Counters are DMAd to pre-allocated consistent memory periodically,
controlled by the define VNIC_FLOW_COUNTER_UPDATE_MSECS. The default is
100 milliseconds.
Signed-off-by: John Daley <johndale@cisco.com> Reviewed-by: Hyong Youb Kim <hyonkim@cisco.com>
John Daley [Fri, 28 Sep 2018 03:08:36 +0000 (20:08 -0700)]
net/enic: fix flow API memory leak
rte_flow structures were not being freed when destroyed or flushed.
Fixes: 6ced137607d0 ("net/enic: flow API for NICs with advanced filters enabled") Cc: stable@dpdk.org Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com> Signed-off-by: John Daley <johndale@cisco.com>
This patch adds phy_lock, acquire/release the lock when performing
PHY transactions. Without this fix driver can run into synchronization
issues with management FW when modifying PHY settings.
net/bnx2x: fix call to link handling periodic function
If link handling periodic function is allowed to be called in interrupt
context, the periodic function can get called too frequently and exhaust
the retry credits to check link status.
This change makes sure link handling periodic function is not called in
interrupt context.
Igor Romanov [Wed, 26 Sep 2018 06:38:04 +0000 (07:38 +0100)]
net/failsafe: report actual device capabilities
Failsafe device capabilities depend on supported by the failsafe
itself plus capabilities supported by sub-devices.
Make fs_dev_infos_get() take failsafe device capabilities into account.
Fixes: cac923cfea47 ("ethdev: support runtime queue setup") Cc: stable@dpdk.org Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru> Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com> Acked-by: Gaetan Rivet <gaetan.rivet@6wind.com>
Reopening vNIC does not automatically disable overlay offload. If it
is previously enabled, it remains enabled even when the user restarts
DPDK and requests overlay offload to be disabled via devarg
disable-overlay=1. So explicitly disable overlay offload when
requested.
Fixes: 93fb21fdbe23 ("net/enic: enable overlay offload for VXLAN and GENEVE") Cc: stable@dpdk.org Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com> Reviewed-by: John Daley <johndale@cisco.com>
net/enic: add VLAN and csum offloads to simple Tx handler
Currently the simple Tx handler supports no offloads, which makes it
usable only for a small number of benchmarks. Add vlan and checksum
offloads to the handler, as cycles/packet increases only by about 3
cycles, and applications commonly use those offloads.
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com> Reviewed-by: John Daley <johndale@cisco.com>
net/enic: set Rx VLAN offload flag for non-stripped packets
The NIC indicates VLAN TCI to the driver even when VLAN stripping is
disabled. The driver sets mbuf's vlan_tci but not PKT_RX_VLAN. Set
PKT_RX_VLAN to indicate that vlan_tci is valid.
Fixes: c6f455507411 ("net/enic: add ethernet VLAN packet type") Cc: stable@dpdk.org Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com> Reviewed-by: John Daley <johndale@cisco.com>
Cisco VIC models support RTE_IOVA_VA, so enable it. This change allows
the driver to work properly when --no-huge is used, in combination
with vfio and iommu.
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com> Reviewed-by: John Daley <johndale@cisco.com>
Bugzilla ID: 39 Fixes: 9913fbb91df0 ("enic/base: common code") Fixes: 322b355f2183 ("net/enic/base: bring NIC interface functions up to date") Cc: stable@dpdk.org Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com> Reviewed-by: John Daley <johndale@cisco.com>
Xiaoyun Li [Sun, 30 Sep 2018 02:29:50 +0000 (10:29 +0800)]
net/i40e: config flow director automatically
I40e driver needed users to config exact fdir mode to create rte_flow
rules but it shouldn't. This patch allows users to create rte_flow rules
without configuring fdir mode and let the driver config fdir automatically.
And remove the workaround in flow filtering example.
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com> Acked-by: Beilei Xing <beilei.xing@intel.com>
Matthew Smith [Mon, 16 Jul 2018 16:36:04 +0000 (11:36 -0500)]
net/ixgbe: wait longer for link after fiber MAC setup
After setting up the link on a fiber port, the maximum wait time for
the link to come up is 500 ms in ixgbe_setup_mac_link_multispeed_fiber().
On an x550 SFP+ port, this is often not sufficiently long for the link
to come up. This can result in never being able to retrieve accurate
link status for the port using rte_eth_link_get_nowait().
Increase the maximum wait time in ixgbe_setup_mac_link_multispeed_fiber()
to 1 s.
Bugzilla ID: 69 Fixes: f3430431abaf ("ixgbe/base: add SFP+ dual-speed support") Cc: stable@dpdk.org Signed-off-by: Matthew Smith <mgsmith@netgate.com> Acked-by: Qi Zhang <qi.z.zhang@intel.com>
net/i40e/base: add new TR bits used for cloud filters
There is a new set of TR bits that can be used when replacing
the cloud filters so add them in. Also added a check to make
sure that the replace cloud filters AQ command doesn't get
executed on an X722 since it is not supported there.
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com> Signed-off-by: Andrey Chilikin <andrey.chilikin@intel.com> Signed-off-by: Kirill Rybalchenko <kirill.rybalchenko@intel.com> Acked-by: Beilei Xing <beilei.xing@intel.com>
net/i40e/base: add capability flag for stopping FW LLDP
Add HW capability flag to indicate that firmware supports stopping
LLDP agent. This feature has been added in FW API 1.7 for XL710
devices and 1.6 for X722. Also raise expected minor version number
for X722 FW API to 6.
These two functions are currently only used in the LED get/set
functions, which are not apart of the VF driver. So the
i40e_aq_set/get_phy_register functions should be wrapped so they
can be removed from the VF driver.
This was brought up in the Linux community that these functions in
the VF driver had no callers in the tree, so they should be removed.
net/i40e/base: add FC threshold parameter for set MAC
This patch adds the default value for Flow Control Refresh Threshold
to set_mac_config AdminQ command. Previously, calling this AdminQ
command would overwrite the default value with 0.
net/i40e/base: add AQ command for rearrange NVM structure
During switching between old NVM structure approach (called
structured NVM) to new one (called flat NVM) or backward flash
needs to be rearranged to required NVM structure.
This is a part of transition from one NVM structure to another.
The function is introduced to command firmware to start
rearrangement process.
Add new field - command flags with only one flag for now. Added flag
tells FW that it shouldn't change page while accessing QSFP module,
as it was set manually.
git.droids-corp.org / dpdk.git / log