Sachin Saxena [Mon, 28 Sep 2020 23:14:17 +0000 (01:14 +0200)]
net/dpaa2: release port upon close
With removal of old close behavior, the private
port resources must be released in the .dev_close callback.
Freeing of port private resources is moved from
the ".remove(device)" to the ".dev_close(port)" operation
Signed-off-by: Sachin Saxena <sachin.saxena@oss.nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Sachin Saxena [Mon, 28 Sep 2020 23:14:16 +0000 (01:14 +0200)]
net/dpaa: release port upon close
With removal of old close behavior, the private
port resources must be released in the .dev_close callback.
Freeing of port private resources is moved from
the ".remove(device)" to the ".dev_close(port)" operation
Signed-off-by: Sachin Saxena <sachin.saxena@oss.nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Thomas Monjalon [Mon, 28 Sep 2020 23:14:15 +0000 (01:14 +0200)]
net/bonding: release port upon close
The flag RTE_ETH_DEV_CLOSE_REMOVE is set so all port resources
can be freed by rte_eth_dev_close().
Freeing of private port resources is moved
from the ".remove(device)" to the ".dev_close(port)" operation.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Rasesh Mody [Mon, 28 Sep 2020 23:14:14 +0000 (01:14 +0200)]
net/bnx2x: release port upon close
Set RTE_ETH_DEV_CLOSE_REMOVE upon probe so all the private resources
for the port can be freed by rte_eth_dev_close(). With this change the
private port resources are released in the .dev_close callback.
Signed-off-by: Rasesh Mody <rmody@marvell.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Thomas Monjalon [Mon, 28 Sep 2020 23:14:13 +0000 (01:14 +0200)]
net/axgbe: release port upon close
The flag RTE_ETH_DEV_CLOSE_REMOVE is set so all port resources
can be freed by rte_eth_dev_close().
Freeing of private port resources is moved
from the ".remove(device)" to the ".dev_close(port)" operation.
The ".dev_close" callback is also called as part of the ".remove" one.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Thomas Monjalon [Mon, 28 Sep 2020 23:14:12 +0000 (01:14 +0200)]
net/atlantic: release port upon close
The flag RTE_ETH_DEV_CLOSE_REMOVE is set so all port resources
can be freed by rte_eth_dev_close().
Freeing of private port resources is moved
from the ".remove(device)" to the ".dev_close(port)" operation.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Igor Russkikh <irusskikh@marvell.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Thomas Monjalon [Mon, 28 Sep 2020 23:14:11 +0000 (01:14 +0200)]
net/af_packet: release port upon close
The flag RTE_ETH_DEV_CLOSE_REMOVE is set so all port resources
can be freed by rte_eth_dev_close().
Freeing of private port resources is moved
from the ".remove(device)" to the ".dev_close(port)" operation.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Thomas Monjalon [Mon, 28 Sep 2020 23:14:10 +0000 (01:14 +0200)]
ethdev: allow drivers to return error on close
The device operation .dev_close was returning void.
This driver interface is changed to return an int.
Note that the API rte_eth_dev_close() is still returning void,
although a deprecation notice is pending to change it as well.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>
Reviewed-by: Sachin Saxena <sachin.saxena@oss.nxp.com>
Reviewed-by: Liron Himi <lironh@marvell.com>
Reviewed-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Thomas Monjalon [Mon, 28 Sep 2020 23:14:09 +0000 (01:14 +0200)]
ethdev: reset device and interrupt pointers on release
The pointers .device and .intr_handle were already reset by the helper
rte_eth_dev_pci_generic_remove().
It is now made part of rte_eth_dev_release_port().
It makes rte_eth_dev_pci_release() meaningless,
so it is replaced with a call to rte_eth_dev_release_port().
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Huisong Li [Fri, 25 Sep 2020 12:47:19 +0000 (20:47 +0800)]
app/testpmd: fix displaying Rx/Tx queues information
Currently, the information of Rx/Tx queues from PMD driver is not
displayed exactly in the rxtx_config_display function. Because
"ports[pid].rx_conf" and "ports[pid].tx_conf" maintained in testpmd
application may be not the value actually used by PMD driver. For
instance, user does not set a field, but PMD driver has to use the
default value.
This patch fixes rxtx_config_display so that the information of Rx/Tx
queues can be really displayed for the PMD driver that implement
.rxq_info_get and .txq_info_get ops callback function.
Fixes:
75c530c1bd53 ("app/testpmd: fix port configuration print")
Fixes:
d44f8a485f5d ("app/testpmd: enable per queue configure")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Chengchang Tang [Fri, 25 Sep 2020 12:47:18 +0000 (20:47 +0800)]
app/testpmd: fix descriptor id check
The number of desc is a per queue configuration. But in the check
function, nb_txd & nb_rxd are used to check whether the desc_id is
valid. nb_txd & nb_rxd are the global configuration of number of desc.
If the queue configuration is changed by cmdline liks: "port config xx
txq xx ring_size xxx", the real value will be changed.
This patch use the real value to check whether the desc_id is valid.
And if these are not configured by user. It will use the default value
to check it, since the rte_eth_rx_queue_setup & rte_eth_tx_queue_setup
will use a default value to configure the queue if nb_rx_desc or
nb_tx_desc is zero.
Fixes:
af75078fece3 ("first public release")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Chengchang Tang [Fri, 25 Sep 2020 12:47:17 +0000 (20:47 +0800)]
app/testpmd: fix packet header in txonly mode
In txonly forward mode, the packet header is fixed by the initial
setting, including the packet length and checksum. So when the packets
varies, this may cause a packet header error. Currently, there are two
methods in txonly mode to randomly change the packets.
1. Set txsplit random and txpkts (x[,y]*), the number of segments
each packets will be a random value between 1 and total number of
segments determined by txpkts settings.
The step as follows:
a) ./testpmd -w xxx -l xx -n 4 -- -i --disable-device-start
b) port config 0 tx_offload multi_segs on
c) set fwd txonly
d) set txsplit rand
e) set txpkts 2048,2048,2048,2048
f) start
The nb_segs of the packets sent by testpmd will be 1~4. The real packet
length will be 2048, 4096, 6144 and 8192. But in fact the packet length
in ip header and udp header will be fixed by 8178 and 8158.
2. Set txonly-multi-flow. the ip address will be varied to generate
multiple flow.
The step as follows:
a) ./testpmd -w xxx -l xx -n 4 -- -i --txonly-multi-flow
b) set fwd txonly
c) start
The ip address of each pkts will change randomly, but since the header
is fixed, the checksum may be a error value.
Therefore, this patch adds a function to update the packet length and
check sum in the pkts header when the txsplit mode is set to rand or
multi-flow is set.
Fixes:
82010ef55e7c ("app/testpmd: make txonly mode generate multiple flows")
Fixes:
79bec05b32b7 ("app/testpmd: add ability to split outgoing packets")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Chengchang Tang [Fri, 25 Sep 2020 12:47:16 +0000 (20:47 +0800)]
app/testpmd: remove restriction on Tx segments set
Currently, if nb_txd is not set, the txpkts is not allowed to be set
because the nb_txd is used to avoid the number of segments exceed the Tx
ring size and the default value of nb_txd is 0. And there is a bug that
nb_txd is the global configuration for Tx ring size and the ring size
could be changed by some command per queue. So these valid check is
unreliable and introduced unnecessary constraints.
This patch adds a valid check function to use the real Tx ring size to
check the validity of txpkts.
Fixes:
af75078fece3 ("first public release")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Chengchang Tang [Fri, 25 Sep 2020 12:47:15 +0000 (20:47 +0800)]
app/testpmd: fix VLAN configuration on failure
When failing to configure VLAN offloads after the port was started, there
is no need to update the port configuration. Currently, when user
configure an unsupported VLAN offloads and fails, and then restart the
port, it will fails since the configuration has been refreshed.
This patch makes the function return directly instead of refreshing the
configuration when execution fails.
Fixes:
384161e00627 ("app/testpmd: adjust on the fly VLAN configuration")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Chengchang Tang [Fri, 25 Sep 2020 12:47:14 +0000 (20:47 +0800)]
app/testpmd: fix port id check in Tx VLAN command
To set Tx vlan offloads, it is required to stop port firstly. But before
checking whether the port is stopped, the port id entered by the user
is not checked for validity. When the port id is illegal, it would lead
to a segmentation fault since it attempts to access a member of
non-existent port.
This patch adds verification of port id in tx vlan offloads and remove
duplicated check.
Fixes:
597f9fafe13b ("app/testpmd: convert to new Tx offloads API")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Andy Moreton [Thu, 24 Sep 2020 12:12:35 +0000 (13:12 +0100)]
common/sfc_efx/base: handle descriptor proxy queue events
The TXQ_DESC and VIRTQ_DESC events are used to pass host descriptors
over an extended width event queue to an application processor for
handling. See SF-122927-TC and SF-122966-SW for details.
Signed-off-by: Andy Moreton <amoreton@xilinx.com>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Andy Moreton [Thu, 24 Sep 2020 12:12:34 +0000 (13:12 +0100)]
common/sfc_efx/base: add option for descriptor proxy queues
EF100 uses descriptor proxy queues to support virtio-blk proxy.
Signed-off-by: Andy Moreton <amoreton@xilinx.com>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Andy Moreton [Thu, 24 Sep 2020 12:12:33 +0000 (13:12 +0100)]
common/sfc_efx/base: handle normal events in extended width
Process the encasulated events as for the normal event loop. The phase
bit in the encapsulated event should be ignored, as the polling loop
uses the phase bit from the extended-width event.
Signed-off-by: Andy Moreton <amoreton@xilinx.com>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Andy Moreton [Thu, 24 Sep 2020 12:12:32 +0000 (13:12 +0100)]
common/sfc_efx/base: poll extended width event queues
Extended width queues use a different layout and so require
a different polling loop.
Signed-off-by: Andy Moreton <amoreton@xilinx.com>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Andy Moreton [Thu, 24 Sep 2020 12:12:31 +0000 (13:12 +0100)]
common/sfc_efx/base: support creation of extended width EvQ
Add a flag to request an extended width event queue, and
check that the supplied buffer is large enough to hold the
event queue descriptors.
Signed-off-by: Andy Moreton <amoreton@xilinx.com>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Andy Moreton [Thu, 24 Sep 2020 12:12:30 +0000 (13:12 +0100)]
common/sfc_efx/base: add 256-bit type
EF100 requires support for extended-width event descriptors
for use with descriptor proxy queues. Extend libefx types
used for hardware access (and endian conversion) to support
a 256bit data type.
Signed-off-by: Andy Moreton <amoreton@xilinx.com>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Andy Moreton [Thu, 24 Sep 2020 12:12:29 +0000 (13:12 +0100)]
common/sfc_efx/base: add option for extended width events
EF100 uses event queues with 256bit extended width events to
support descriptor proxy queues.
Signed-off-by: Andy Moreton <amoreton@xilinx.com>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:28 +0000 (13:12 +0100)]
common/sfc_efx/base: allocate vAdaptor on Riverhead
Riverhead has EVB support similar to EF10 and NIC must allocate
its vAdaptor on init.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:27 +0000 (13:12 +0100)]
common/sfc_efx/base: use EF10 EVB methods for Riverhead
There is no difference yet in EVB support on EF10 and Riverhead.
So, it is better to reuse existing methods.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Igor Romanov [Thu, 24 Sep 2020 12:12:26 +0000 (13:12 +0100)]
common/sfc_efx/base: replace PCI efsys macros with functions
efsys macros that manipulate PCI devices cannot be defined in
common sfc_efx DPDK driver since in DPDK build the bus drivers
that provide required functionality are built after common drivers.
Replace the macros with function callbacks to remove that build
dependency. Drivers now should pass the callbacks directly to
efx function instead of defining implementation in efsys.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Igor Romanov [Thu, 24 Sep 2020 12:12:25 +0000 (13:12 +0100)]
common/sfc_efx/base: support UDP tunnel operations for EF100
EF100 uses VNIC encapsulation rule MCDI for configuring UDP
tunnels in HW individually. Use busy added and removed states
of UDP tunnel table entries for the implementation.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Igor Romanov [Thu, 24 Sep 2020 12:12:24 +0000 (13:12 +0100)]
common/sfc_efx/base: introduce states for UDP tunnel entries
UDP tunnel reconfigure operation takes UDP tunnel table, which contains
entries that need to be applied to HW. This approach does not retain
information about what entries were removed or added, which is required
for Riverhead.
Add states to the table entries to indicate add or remove operations.
On tunnel reconfiguration added and removed entries become busy to
indicate that the entries are currently configured and to prevent add or
remove operations from other threads. After tunnel reconfiguration is
complete, the states are reset - added entries become applied and
removed entries are purged from the table.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Igor Romanov [Thu, 24 Sep 2020 12:12:23 +0000 (13:12 +0100)]
common/sfc_efx/base: introduce UDP tunnel destruct operation
The procedures for destroying UDP tunnels are NIC family specific,
so they should be implemented separately for each of them.
Check for supported UDP encapsulation is removed from generic
operations since it is no longer used by the generic libefx API.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Igor Romanov [Thu, 24 Sep 2020 12:12:22 +0000 (13:12 +0100)]
common/sfc_efx/base: add NIC magic check on BAR lookup
Check that BAR lookup was successful by looking at the content
of signature register.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Igor Romanov [Thu, 24 Sep 2020 12:12:21 +0000 (13:12 +0100)]
common/sfc_efx/base: add Xilinx capabilities table lookup
APIs for searching a capability in a Xilinx capabilities table and
reading table entries are needed for function control window lookup
to get the bar index and offset of the window.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Igor Romanov [Thu, 24 Sep 2020 12:12:20 +0000 (13:12 +0100)]
common/sfc_efx/base: add efsys API to find a memory BAR
Function control window lookup needs memory BARs handles to
search Xilinx capabilities tables.
Define an API to get a memory BAR handle by a PCIe device handle and
BAR index.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Igor Romanov [Thu, 24 Sep 2020 12:12:19 +0000 (13:12 +0100)]
common/sfc_efx/base: add function control window lookup API
Riverhead NIC may provide a locator of function control window
(EF100 resource locator).
The locator may be present in a Xilinx capabilities table which
itself is located by looking into extended PCI capabilities.
PCI capabilities are made possible to access by adding PCI config
space interface to efsys.
APIs are implemented to facilitate function control window lookup:
- API to find an extended PCI capability given a capability ID;
- API to read Xilinx PCI capability;
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Igor Romanov [Thu, 24 Sep 2020 12:12:18 +0000 (13:12 +0100)]
common/sfc_efx/base: add function control window concept
Function control window can be located at a different offset than
other windows on Riverhead. Meaning that the drivers must handle
accesses to the function control window differently in case of EF100.
Add accessor macros for function control window and change
EFX NIC create API to facilitate that accessors.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:17 +0000 (13:12 +0100)]
common/sfc_efx/base: choose smallest Rx prefix on Riverhead
Riverhead supports many Rx prefixes. The Rx prefix may be chosen
based on which information is required.
To have better performance choose the smallest Rx prefix which
meets our requirements.
Right now there is no way to specify requirements on Rx queue
creation, but it could be added in the future.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:16 +0000 (13:12 +0100)]
common/sfc_efx/base: group RxQ parameters into a structure
Make number of efx_mcdi_init_rxq() arguments reasonable before
addition of one more argument.
Non essential parameters not supported in some cases are moved
into helper structure.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:15 +0000 (13:12 +0100)]
common/sfc_efx/base: add API to get Rx prefix information
Define default Siena, EF10 default, packed stream, equal-stride
superbuffer and Riverhead default prefixes in order to make an API
to get Rx prefix layout information generic and usable on all NICs.
Riverhead supports many Rx prefixes. Riverhead FW supports MCDI to
choose Rx prefix based on required Rx prefix fields and allows to
query the prefix layout using MCDI. The patch prepares to introduce
the support in libefx and provides fallback for NICs and FW which
lacks the support.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:14 +0000 (13:12 +0100)]
common/sfc_efx: do not include libefx headers from efsys.h
efsys.h may depends on target system headers only. Otherwise, it could
result in cross-dependency when libefx header depends on efsys.h and
efsys.h depends on the libefx header.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andrew Lee <andrewle@xilinx.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Ivan Malov [Thu, 24 Sep 2020 12:12:13 +0000 (13:12 +0100)]
common/sfc_efx/base: report restrictions for TSO version 3
Riverhead puts a number of restrictions on TSO transactions.
Reflect some of them in the NIC configuration structure.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andrew Lee <andrewle@xilinx.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Ivan Malov [Thu, 24 Sep 2020 12:12:12 +0000 (13:12 +0100)]
common/sfc_efx/base: indicate support for TSO version 3
Riverhead boards support TSO version 3.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andrew Lee <andrewle@xilinx.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:11 +0000 (13:12 +0100)]
common/sfc_efx/base: handle Tx complete on Riverhead
Introduce a new event callback which has the same prototype, but
provides number of completed descriptors instead of the last
completed descriptor index.
When all libefx-based drivers implement the new callback, libefx
may be updated to use it for Siena and EF10 family NICs and
the old one may be removed.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:10 +0000 (13:12 +0100)]
common/sfc_efx/base: handle Rx events for Riverhead
Rx event on Riverhead provides a number of received packets and
no classification/offloads information is available Rx event.
Introduce a new event callback to be implemented by drivers.
The callback provides information about the number of completed
packets. libefx-based drivers should implement the new callback
for Riverhead and keep the old one for Siena and EF10 NICs.
The new callback may be used for Medford2 NO_CONT_EV Rx mode
support.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:09 +0000 (13:12 +0100)]
common/sfc_efx/base: fix Tx descriptor DMA sync on Riverhead
Rx/Tx queue DMA sync should not assume descriptor size to be the same
for all NIC familties since it Tx descritor size is 16 on Riverhead.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:08 +0000 (13:12 +0100)]
common/sfc_efx/base: implement Tx control path for Riverhead
Tx control path on Riverhead is very similar to EF10, but datapath
differs a lot since Tx descriptor size is 16 bytes (vs 8 bytes on EF10).
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:07 +0000 (13:12 +0100)]
common/sfc_efx/base: implement Rx control path for Riverhead
Reuse EF10 RSS-related functions since current version of the RSS
interface is compatible with EF10.
Implement own functions to create and destroy Rx queues which reuse
MCDI wrappers which are shared with EF10.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:06 +0000 (13:12 +0100)]
common/sfc_efx/base: move Rx index check to generic code
Make NIC family specific functions a bit shorter and reduce
code duplication.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:05 +0000 (13:12 +0100)]
common/sfc_efx/base: free Rx queue structure in generic code
Rx queue structure is allocated in generic code, but was freed in NIC
family specific callbacks. Move free to generic function makes NIC
family specific callbacks shorter and reduces code duplication.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:04 +0000 (13:12 +0100)]
common/sfc_efx/base: maintain RxQ counter in generic code
The counter is incremented in generic efx_rx_qcreate(), but was
asserted and decremented in NIC family specific queue create and
destroy callbacks. Move assert and decrement to generic functions
as well to make NIC family specific callbacks shorter.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:03 +0000 (13:12 +0100)]
common/sfc_efx/base: switch TxQ init to extended version
Extended version of MCDI allows up to 64 DMA addresses which are
required for Riverhead.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:02 +0000 (13:12 +0100)]
common/sfc_efx/base: move TxQ init/fini wrappers to generic
TxQ init/fini MCDI is similar on Riverhead and these
functions should be reused to implement TxQ creation and
destruction on Riverhead.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:01 +0000 (13:12 +0100)]
common/sfc_efx/base: move RxQ init/fini wrappers to generic
RxQ init/fini MCDI is similar on Riverhead and these
functions should be reused to implement RxQ creation and
destruction on Riverhead.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:12:00 +0000 (13:12 +0100)]
common/sfc_efx/base: handle MCDI events on Riverhead
EF100 MCDI event layout is same as on EF10 except added QDMA phase
bit which is unused on EF10.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Ivan Malov [Thu, 24 Sep 2020 12:11:59 +0000 (13:11 +0100)]
common/sfc_efx/base: complete EvQ creation on Riverhead
Client drivers relying on interrupts will fail to complete event
queue creation on Riverhead boards as the latter have no support
for INIT_DONE events which means that it's useless to wait until
initialisation callback is triggered by interrupt-driven polling.
Client drivers which avoid interrupt-driven polling still handle
INIT_DONE events by direct polling and will fail to do so.
Solve this problem by adding an extra poll-once method which
will be called by client drivers after queue creation (probably,
with driver-specific lock being held). The method will invoke
initialisation callback on Riverhead and do nothing on the other
boards. Then the drivers will proceed with normal waiting which
will complete immediately in the case of Riverhead.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:58 +0000 (13:11 +0100)]
common/sfc_efx/base: add event queue module for Riverhead
Events are significantly reworked on Riverhead, so it is better
to implement own set of callbacks to simplify future development
and avoid inheritance of legacy code.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:57 +0000 (13:11 +0100)]
common/sfc_efx/base: merge versions of init EvQ wrappers
The decision on which version of the INIT_EVQ command to use may
be done inside the function itself. Caller should just provide
enough information sufficient in both cases. It avoids code
duplication and simplifies maintenance.
If v2 is not supported, keep decision about low-latency hint outside
the MCDI helper function since it will differ on Riverhead (there is
no EVB yet, but still want merging for better throughput).
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:56 +0000 (13:11 +0100)]
common/sfc_efx/base: prepare to merge EvQ init functions
v1 and v2 defines are the same except output length.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:55 +0000 (13:11 +0100)]
common/sfc_efx/base: move EvQ create generic checks
There is no point to duplicate these generic checks in NIC family
specific handlers.
As the side effect it fixes bug with incorrect interrupt moderation
settings silently ignored on event queue create on Siena.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:54 +0000 (13:11 +0100)]
common/sfc_efx/base: move EvQ init/fini wrappers to generic
EvQ init/fini MCDI is similar on Riverhead and these
functions should be reused to implement EvQ creation and
destruction on Riverhead.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:53 +0000 (13:11 +0100)]
common/sfc_efx/base: add interrupts module for Riverhead
Riverhead supports interrupt aggregation rings. Right now there
are no plans to support it in libefx, but it is better to have
own set of callbacks to easier handle EF10 and Riverhead
differences in the future.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:52 +0000 (13:11 +0100)]
common/sfc_efx/base: use EF10 MAC methods for Riverhead
There is no difference yet in MAC support on EF10 and Riverhead.
So, it is better to reuse existing methods.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:51 +0000 (13:11 +0100)]
common/sfc_efx/base: use EF10 filter methods for Riverhead
Filtering MCDI on Riverhead are the same as on EF10.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:50 +0000 (13:11 +0100)]
common/sfc_efx/base: use dummy tunnel ops for Riverhead
Riverhead does not support tunnels yet.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:49 +0000 (13:11 +0100)]
common/sfc_efx/base: add Riverhead support to NIC module
Define basic NIC features and limitations.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:48 +0000 (13:11 +0100)]
common/sfc_efx/base: factor out MCDI entity reset helper
The helper will be reused on Riverhead.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:47 +0000 (13:11 +0100)]
common/sfc_efx/base: set NIC features in generic place
There is no point to duplicate it NIC family specific probe handlers.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:46 +0000 (13:11 +0100)]
common/sfc_efx/base: factor out helper to get board config
The helper will be used on Riverhead.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:45 +0000 (13:11 +0100)]
common/sfc_efx/base: remove PF count get from caps get
ef10_get_datapath_caps() should simply fill in NIC configuration
based MC_CMD_GET_CAPABILITIES output.
Number of PFs is a part of board configuration should be obtained
in corresponding function.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:44 +0000 (13:11 +0100)]
common/sfc_efx/base: move 14b prefix check out of caps get
The function which gets capabilities from MC should simply fill in
NIC configuration. Caller should decide if the configuration is
supported or not.
The change is required to prepare the function to be reused on
Riverhead.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:43 +0000 (13:11 +0100)]
common/sfc_efx/base: use EF10 PHY methods for Riverhead
There is no difference yet in PHY support on EF10 and Riverhead.
So, it is better to reuse existing methods.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:42 +0000 (13:11 +0100)]
common/sfc_efx/base: use EF10 MCDI methods for Riverhead
MCDI on Riverhead is very close to the EF10 and there is no
point to duplicate the code.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:41 +0000 (13:11 +0100)]
common/sfc_efx/base: update registers check for Riverhead
Riverhead family got letter 'G' for registers definition.
Registers used for Riverhead should have the letter in the
specification range.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:40 +0000 (13:11 +0100)]
common/sfc_efx/base: add Riverhead NIC family
Riverhead is the first NIC family of the EF100 architecture.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:39 +0000 (13:11 +0100)]
common/sfc_efx/base: add efsys option for Riverhead
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Vijay Kumar Srivastava <vsrivast@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:38 +0000 (13:11 +0100)]
common/sfc_efx/base: add event queue operation to do polling
Event queue host interface differ on Riverhead and implementation
will be different.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:37 +0000 (13:11 +0100)]
common/sfc_efx/base: update MCDI headers
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Andrew Rybchenko [Thu, 24 Sep 2020 12:11:36 +0000 (13:11 +0100)]
common/sfc_efx/base: add EF100 registers definitions
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Kiran Kumar K [Mon, 24 Aug 2020 07:04:24 +0000 (12:34 +0530)]
net/octeontx2: support VLAN insert and strip actions
Adding support for RTE Flow VLAN insert and strip actions.
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Kiran Kumar K [Fri, 21 Aug 2020 06:59:30 +0000 (12:29 +0530)]
net/octeontx2: set max VTAG insertion size
When TX side VTAG insertion is enabled, SMQ should be configured with
the maximum VTAG insertion size to avoid generating NIX_SQINT_SEND_ERR
interrupt. Since the default value is zero, This patch configures the
VTAG insertion size to the max supported value.
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Manish Chopra [Fri, 25 Sep 2020 11:55:08 +0000 (04:55 -0700)]
net/qede: support VF FLR
This patch adds required bit to handle VF FLR
indication from Management FW (MFW) of the device
With that VFs were able to load in VM (VF attached as PCI
passthrough to the guest VM) followed by FLR successfully
Updated the docs/guides with the feature support
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Igor Russkikh <irusskikh@marvell.com>
Signed-off-by: Rasesh Mody <rmody@marvell.com>
Manish Chopra [Fri, 25 Sep 2020 11:55:07 +0000 (04:55 -0700)]
net/qede: initialize VF MAC and link
This patch configures VFs with random mac if no MAC is
provided by the PF/bulletin. This also adds required bulletin
APIs by PF-PMD driver to communicate LINK properties/changes to
the VFs through bulletin update mechanism.
With these changes, VF-PMD instance is able to run
fastpath over PF-PMD driver instance.
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Igor Russkikh <irusskikh@marvell.com>
Signed-off-by: Rasesh Mody <rmody@marvell.com>
Manish Chopra [Fri, 25 Sep 2020 11:55:06 +0000 (04:55 -0700)]
net/qede: add infrastructure support for VF load
This patch adds necessary infrastructure support (required to handle
messages from VF and sending ramrod on behalf of VF's configuration
request from alarm handler context) to start/load the VF-PMD driver
instance on top of PF-PMD driver instance.
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Igor Russkikh <irusskikh@marvell.com>
Signed-off-by: Rasesh Mody <rmody@marvell.com>
Manish Chopra [Fri, 25 Sep 2020 11:55:05 +0000 (04:55 -0700)]
net/qede: configure VFs on hardware
Based on number of VFs enabled at PCI, PF-PMD driver instance
enables/configures those VFs from hardware perspective, such
that in later patches they could get required HW access to
communicate with PFs for slowpath configuration and run the
fastpath themselves.
This patch also add two new qede IOV files [qede_sriov(.c|.h)]
under qede directory to add non-base driver IOV APIs/contents there.
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Igor Russkikh <irusskikh@marvell.com>
Signed-off-by: Rasesh Mody <rmody@marvell.com>
Manish Chopra [Fri, 25 Sep 2020 11:55:04 +0000 (04:55 -0700)]
net/qede: define PCI config space specific osals
This patch defines various PCI config space access APIs
in order to read and find IOV specific PCI capabilities.
With these definitions implemented, it enables the base
driver to do SR-IOV specific initialization and HW specific
configuration required from PF-PMD driver instance.
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Igor Russkikh <irusskikh@marvell.com>
Signed-off-by: Rasesh Mody <rmody@marvell.com>
Manish Chopra [Fri, 25 Sep 2020 11:55:03 +0000 (04:55 -0700)]
bus/pci: query PCI extended capabilities
By adding generic API, this patch removes individual
functions/defines implemented by drivers to find extended
PCI capabilities.
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Igor Russkikh <irusskikh@marvell.com>
Reviewed-by: Gaetan Rivet <grive@u256.net>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>
Yunjian Wang [Mon, 28 Sep 2020 12:44:42 +0000 (20:44 +0800)]
net/fm10k: fix memory leak when thresh check fails
In fm10k_rx_queue_setup(), we allocate memory for the queue
structure but not released when thresh check fails.
Fixes:
6cfe8969c969 ("fm10k: add Rx queue setup/release")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Lance Richardson [Fri, 25 Sep 2020 21:40:19 +0000 (17:40 -0400)]
net/bnxt: fix vector mode for 32-bit x86
Descriptor valid mask should be 64-bit all ones, use
appropriate initializer type (unsigned long long) to
obtain correct value in 32-bit mode.
Fixes:
deae85145c64 ("net/bnxt: handle multiple packets per loop in vector Rx")
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Kiran Kumar K [Thu, 13 Aug 2020 07:22:49 +0000 (12:52 +0530)]
net/octeontx2: support VLAN based RSS hash
Adding support for VLAN based RSS hash. Two bytes of TCI field
will be considered for hashing.
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Sarosh Arif [Thu, 24 Sep 2020 10:39:32 +0000 (15:39 +0500)]
common/octeontx2: fix memory mapping API usage
Replace mmap() and munmap() APIs with EAL abstraction for the same.
Fixes:
e1d9a02ad8f0 ("common/octeontx2: introduce common device class")
Cc: stable@dpdk.org
Signed-off-by: Sarosh Arif <sarosh.arif@emumba.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Harman Kalra [Mon, 10 Aug 2020 10:55:45 +0000 (16:25 +0530)]
net/octeontx2: fix multi segment mode for jumbo packets
Scatter gather mode should get enabled:
- If mbuf size is less than max received packet length.
- MTU is reconfigured greater than mbuf size.
Fixes:
0e2efd02db58 ("net/octeontx2: add MTU set operation")
Cc: stable@dpdk.org
Signed-off-by: Harman Kalra <hkalra@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Pavan Nikhilesh [Tue, 28 Jul 2020 18:43:46 +0000 (00:13 +0530)]
net/octeontx: fix Tx preparation
When building send command for a given descriptor it expects
it to contain the AURA identifier of the pool that it belongs
to rather than the pool identifier itself.
Fixes:
7f4116bdbb1c ("net/octeontx: add framework for Rx/Tx offloads")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Pavan Nikhilesh [Tue, 28 Jul 2020 18:43:45 +0000 (00:13 +0530)]
mempool/octeontx: fix aura to pool mapping
When sending commands to Kernel FPA PF driver, it expects the
aura to be in the range of 0 to 16 for a given FPA pool.
In OCTEON TX we map a pool,aura pair as single mempool handle,
always set the aura id to 0 for a given FPA pool.
Fixes:
179c7e893f64 ("mempool/octeontx: fix pool to aura mapping")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Wei Hu (Xavier) [Tue, 22 Sep 2020 12:03:29 +0000 (20:03 +0800)]
net/hns3: fix storing RSS info when creating flow action
Currently, when calling the rte_flow_query API function to query the RSS
information, the queue related information is not as expected.
The root cause is that when application call the rte_flow_create API
function to create RSS action, the operation of storing the data whose typs
is struct rte_flow_action_rss is incorrect in the '.create' ops
implementation function named hns3_flow_create.
This patch fixes it by replacing memcpy with hns3_rss_conf_copy function to
store the RSS information in the hns3_flow_create.
Fixes:
c37ca66f2b27 ("net/hns3: support RSS")
Cc: stable@dpdk.org
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Wei Hu (Xavier) [Tue, 22 Sep 2020 12:03:28 +0000 (20:03 +0800)]
net/hns3: fix configuring device with RSS enabled
Currently, when running the following commands in the CLI of testpmd
application, the driver reports an -EINVAL error when performing the No.3
step.
1) flow create 0 ingress pattern end actions rss key <key> func simple_xor
types all end / end
2) flow flush 0
3) port config dcb vt off pfc off
The root cause as below:
In the No.2 step, when RSS rules is flushed, we set the the flag
hw->rss_dis_flag with true to indicate RSS id disabled. And in the No.3
step, calling rte_eth_dev_configure API function, the internal function
named hns3_dev_rss_hash_update check hw->rss_dis_flag is true and return
-EINVAL.
When user calls the rte_eth_dev_configure API function with the input
parameter dev_conf->rxmode.mq_mode having ETH_MQ_RX_RSS_FLAG to enable RSS,
driver should set internal flag hw->rss_dis_flag with false to indicate RSS
is enabled in the '.dev_configure' ops implementation function named
hns3_dev_configure and hns3vf_dev_configure.
Fixes:
5e782bc2570c ("net/hns3: fix configuring RSS hash when rules are flushed")
Cc: stable@dpdk.org
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Lijun Ou [Tue, 22 Sep 2020 12:03:27 +0000 (20:03 +0800)]
net/hns3: fix flushing RSS rule
When user create a flow without RSS by calling rte_flow_create API and then
destroy it by calling rte_flow_flush API, driver should not clear RSS rule.
A reasonable handling method is that when user creates an RSS rule, the
driver should clear the created RSS rule when flushing destroy all flow
rules. Also, hw->rss_info should save the RSS config of the last success
RSS rule. When create n RSS rules, the RSS should not be disabled before
the last RSS rule destroyed.
Fixes:
c37ca66f2b27 ("net/hns3: support RSS")
Cc: stable@dpdk.org
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Wei Hu (Xavier) [Tue, 22 Sep 2020 12:03:26 +0000 (20:03 +0800)]
net/hns3: fix flow RSS queue number 0
When user specifies RSS queue num for 0 in action list by flow create API,
it should create a valid flow rule. The following flow rule should be
success in the command line of the testpmd application:
flow create 0 <pattern> actions rss queues / end
Fixes:
c37ca66f2b27 ("net/hns3: support RSS")
Cc: stable@dpdk.org
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Lijun Ou [Tue, 22 Sep 2020 12:03:25 +0000 (20:03 +0800)]
net/hns3: fix config when creating RSS rule after flush
Currnetly, when user create a flow RSS rule and then flush the rule, driver
will set 0 for the internal structure of hw->rss_info maintained in driver.
And then user create a flow RSS rule without specified RSS hash key, driver
configure a validate RSS hash key into hardware network engine and will
cause an RSS error. The related steps when using testpmd as
follows:
flow 0 <pattern> action rss xx end / end
flow flush 0
flow 0 <pattern> action rss queues 0 1 end / end
To slove the preceding problem, the flow flush processing is modified.
it don't clear all RSS configurations for hw->rss_info. Actually the RSS
key information in the hardware is not cleared, therefore, the
hw->rss_info.key is kept consistent with the value in the hardware. In
addition, because reset the redirection table, we need to set queues NULL
and queue_num for zero that indicate the corresponding parameters in the
hardware is unavailable. Also we set hw->rss_info.func to
RTE_ETH_HASH_FUNCTION_MAX that indicate it is invalid.
Fixes:
c37ca66f2b27 ("net/hns3: support RSS")
Cc: stable@dpdk.org
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Lijun Ou [Tue, 22 Sep 2020 12:03:24 +0000 (20:03 +0800)]
net/hns3: set RSS hash type input configuration
This patch support for hash type input set configuration. That is, upper
level application can call the rte_flow_create or
rte_eth_dev_rss_hash_update API to set the specified tuples by RSS types.
As a result, the hardware can selectively use the tuples to perform hash
calculation based on the specific tuple to enable the fields.
The following uses the RSS flow as an example:
When application calls the rte_flow_create API to select tuple input for
ipv4-tcp l3-src-only, the driver can enable the ipv4-tcp src field of the
hardware register by calling the internal function named
hns3_set_rss_tuple_by_rss_hf according to the RSS types in actions list.
In testpmd application, the command as follows:
testpmd> flow create 0 ingress pattern end actions rss types \
ipv4-tcp l3-src-only end queues 1 3 end / end
The following use the rte_eth_dev_rss_hash_update as an example:
if the user set the select tuple input for ipv4-tcp l3-src-only by
rte_eth_rss_conf structure in rte_eth_dev_rss_hash_update. the next
flow is the same as that of the RSS flow. In testpmd application,
the command as follows:
testpmd> port config all rss ipv4-tcp l3-src-only
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Lijun Ou [Tue, 22 Sep 2020 12:03:23 +0000 (20:03 +0800)]
net/hns3: check input RSS type when creating flow with RSS
This patch adds checking the input RSS type when creating a flow with RSS.
If the input RSS type are not supported based on hns3 network engine, an
error is returned.
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Lijun Ou [Tue, 22 Sep 2020 12:03:22 +0000 (20:03 +0800)]
net/hns3: support querying RSS flow rule
This patch enables to query some RSS configurations of the specified
rule. For example, show RSS hash function and rss types.
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Chengwen Feng [Tue, 22 Sep 2020 12:03:21 +0000 (20:03 +0800)]
net/hns3: support flow action of queue region
Kunpeng 930 hardware support spread packets to region of queues which can
be configured by FDIR rule, it means user can create one FDIR rule which
action is one region of queues, and then RSS use the region info to spread
packets.
As we know, RTE_FLOW_ACTION_TYPE_RSS is used to spread packets among
several queues, user could config such as func/level/types/key/queue
parameter to control RSS function, so we provide this feature under the
RTE_FLOW_ACTION_TYPE_RSS framework.
Consider RSS input tuple don't have eth header, we use the following
rule to distinguish them (whether it's queue region configuration or
rss general configuration):
Case 1: pattern have ETH and action's queue_num > 0, indicate it is
queue region configuration.
Case other: rss general configuration.
So if user want to configure one flow which ipv4=192.168.1.192 spread to
queue region of queue 0/1/2/3, the patter should:
RTE_FLOW_ITEM_TYPE_ETH with spec=last=mask=NULL
RTE_FLOW_ITEM_TYPE_IPV4 with spec=192.168.1.192 last=mask=NULL
RTE_FLOW_ITEM_TYPE_END
the action should:
RTE_FLOW_ACTION_TYPE_RSS with queue_num=4 queue=0/1/2/3
RTE_FLOW_ACTION_TYPE_END
after calling rte_flow_create, one FDIR rule will be created.
Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Hongbo Zheng [Tue, 22 Sep 2020 12:03:20 +0000 (20:03 +0800)]
net/hns3: break loop in adding error stats
This patch solves the redundant operation during traversal. In the internal
function named hns3_error_int_stats_add for adding error statistics,
because only one statistical item will be found in the for loop statement,
a break can be executed after finding the error statistical item without
traversing the remaining table entries.
Signed-off-by: Hongbo Zheng <zhenghongbo3@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>