The Netronome's Network Flow Processor chip is highly programmable
with the goal of processing packets at high speed. Processing units
and other chip components are available from the host through the
PCIe CPP(Command Push Pull bus) interface. The NFP PF PMD configures
a CPP handler for setting up and working with vNICs, perform actions
like link up or down, or accessing extended stats from the MAC component.
There exist NFP host tools which access the NFP components for
programming and debugging but they require the CPP interface. When the
PMD is bound to the PF, the DPDK app owns the CPP interface, so these
host tools can not access the NFP through other means like NFP kernel
drivers.
This patch adds a CPP bridge using the rte_service API which can be
enabled by a DPDK app. Interestingly, DPDK clients like OVS will not
enable specific service cores, but this can be performed with a
secondary process specifically enabling this CPP bridge service and
therefore giving access to the NFP to those host tools.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Previous commit adding multiprocess support broke VF support.
When VFs, the PMD does not set the link up or down.
Fixes: ef28aa96e5 ("net/nfp: support multiprocess")
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
This patch introduces changes for supporting multiprocess support.
This is trivial for VFs but comes with some limitations for the PF.
Due to restrictions when using NFP CPP interface, just one secondary
process is supported by now for the PF.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Current name rte_eal_check_dma_mask does not follow the naming
used in the rest of the file.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
NFP can handle IOVA as VA. It requires to check those IOVAs
being in the supported range what is done during initialization.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
NFP devices can not handle DMA addresses requiring more than
40 bits. This patch uses rte_dev_check_dma_mask with 40 bits
and avoids device initialization if memory out of NFP range.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Three problems are fixed in this patch:
- RSS capabilities not advertised properly
- RSS configuration just done for some RSS types
- RSS hash match reported for just some RSS types
Fixes: 934e4c60fb ("nfp: add RSS")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Removed DEV_RX_OFFLOAD_CRC_STRIP offload flag.
Without any specific Rx offload flag, default behavior by PMDs is to
strip CRC.
PMDs that support keeping CRC should advertise DEV_RX_OFFLOAD_KEEP_CRC
Rx offload capability.
Applications that require keeping CRC should check PMD capability first
and if it is supported can enable this feature by setting
DEV_RX_OFFLOAD_KEEP_CRC in Rx offload flag in rte_eth_dev_configure()
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Tomasz Duszynski <tdu@semihalf.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Jan Remes <remes@netcope.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
Some NFP firmwares support live changes to the MAC address, but
this is not always true and the firmware advertises it accordingly.
This patch checks if firmware does not support live changes and
sets RTE_ETH_DEV_NOLIVE_MAC_ADDR in that case.
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
A constructor is usually declared with RTE_INIT* macros.
As it is a static function, no need to declare before its definition.
The macro is used directly in the function definition.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
In DPDK 17.11, the ethdev offloads API has changed:
commit cba7f53b71 ("ethdev: introduce Tx queue offloads API")
commit ce17eddefc ("ethdev: introduce Rx queue offloads API")
The new API is documented in the programmer's guide:
http://doc.dpdk.org/guides/prog_guide/poll_mode_drv.html#hardware-offload
For reminder, the main concepts in the new API were:
- All offloads are disabled by default
- Distinction between per port and per queue offloads.
The transition bits are now removed:
- Translation of the old API in ethdev
- rte_eth_conf.rxmode.ignore_offload_bitfield
- ETH_TXQ_FLAGS_IGNORE
The old API bits are now removed:
- Rx per-port rte_eth_conf.rxmode.[bit-fields]
- Tx per-queue rte_eth_txconf.txq_flags
- ETH_TXQ_FLAGS_NO*
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Shahaf Shuler <shahafs@mellanox.com>
DEV_RX_OFFLOAD_KEEP_CRC offload flag is added. PMDs that support
keeping CRC should advertise this offload capability.
DEV_RX_OFFLOAD_CRC_STRIP flag will remain one more release
default behavior in PMDs are to keep the CRC until this flag removed
Until DEV_RX_OFFLOAD_CRC_STRIP flag is removed:
- Setting both KEEP_CRC & CRC_STRIP is INVALID
- Setting only CRC_STRIP PMD should strip the CRC
- Setting only KEEP_CRC PMD should keep the CRC
- Not setting both PMD should keep the CRC
A helper function rte_eth_dev_is_keep_crc() has been added to be able to
change the no flag behavior with minimal changes in PMDs.
The PMDs that doesn't report the DEV_RX_OFFLOAD_KEEP_CRC offload can
remove rte_eth_dev_is_keep_crc() checks next release, related code
commented to help the maintenance task.
And DEV_RX_OFFLOAD_CRC_STRIP has been added to virtual drivers since
they don't use CRC at all, when an application requires this offload
virtual PMDs should not return error.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Allain Legacy <allain.legacy@windriver.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
This patch avoids direct access to device config sysfs file using
rte_pci_read_config instead.
Apart from replicating code, it turns out this direct access does
not always work if non-root users execute DPDK apps. In those cases
it is mandatory to go through VFIO specific function for reading pci
config space.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
TX descriptor eop_offset field is not initialized and it could
contain garbage. This patch fixes the potential problem setting
EOP as the only subfield. The other subfield, data offset, is
not used by now.
Fixes: b812daadad ("nfp: add Rx and Tx")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
DPDK apps can be executed as non-root users but current NFP lock
file for avoiding concurrent accesses to CPP interface is precluding
this option or requires to modify system file permissions.
When the NFP device is bound to VFIO, this driver does not allow this
concurrent access, so the lock file is not required at all.
OVS-DPDK as executed in RedHat distributions is the main NFP user
needing this fix.
Fixes: c7e9729da6 ("net/nfp: support CPP")
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
This was missed in a previous patch for using new offload API.
It turns out the line can be removed because that internal PMD
crc_len data is not being used.
Fixes: a922c6a789 ("net/nfp: support new HW offloads API")
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
drivers/net/nfp/nfp_net.c:669:2: error:
‘memcpy’ forming offset [5, 6] is out of the bounds [0, 4]
of object ‘tmp’ with type ‘uint32_t’ {aka ‘unsigned int’}
[-Werror=array-bounds]
memcpy(&hw->mac_addr[0], &tmp, sizeof(struct ether_addr));
Fixes: e6decee382 ("net/nfp: use random MAC address if not configured")
Cc: stable@dpdk.org
Signed-off-by: Andy Green <andy@warmcat.com>
Acked-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Tested-by: Alejandro Lucero <alejandro.lucero@netronome.com>
drivers/net/nfp/nfp_net.c: In function ‘nfp_pf_pci_probe’:
drivers/net/nfp/nfp_net.c:3160: 23: error:
‘%s’ directive writing up to 99 bytes into a region of size 76
[-Werror=format-overflow=]
sprintf(fw_name, "%s/%s.nffw", DEFAULT_FW_PATH, serial);
Note fw_buf still has to increase somewhat even after
restricting serial[], since otherwise:
drivers/net/nfp/nfp_net.c: In function ‘nfp_pf_pci_probe’:
drivers/net/nfp/nfp_net.c:3176:23:
error: ‘%s’ directive writing up to 99 bytes into a region of size 76
[-Werror=format-overflow=]
sprintf(fw_name, "%s/%s", DEFAULT_FW_PATH, card);
^~
drivers/net/nfp/nfp_net.c:3262:32:
err = nfp_fw_upload(dev, nsp, card_desc);
~~~~~~~~~
drivers/net/nfp/nfp_net.c:3176:2:
note: ‘sprintf’ output between 25 and 124 bytes into a destination of size 100
sprintf(fw_name, "%s/%s", DEFAULT_FW_PATH, card);
Fixes: 896c265ef9 ("net/nfp: use new CPP interface")
Signed-off-by: Andy Green <andy@warmcat.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
A new hook function is added and called inside the PMDs at the end
of the device probing:
- in primary process, after allocating, init and config
- in secondary process, after attaching and local init
This new function is almost empty for now.
It will be used later to add some post-initialization processing.
For the PMDs calling the helpers rte_eth_dev_create() or
rte_eth_dev_pci_generic_probe(), the hook rte_eth_dev_probing_finish()
is called from here, and not in the PMD itself.
Note that the helper rte_eth_dev_create() could be used more,
especially for vdevs, avoiding some code duplication in PMDs.
Cc: stable@dpdk.org
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
This patch check if a input requested offloading is valid or not.
Any reuqested offloading must be supported in the device capabilities.
Any offloading is disabled by default if it is not set in the parameter
dev_conf->[rt]xmode.offloads to rte_eth_dev_configure() and
[rt]x_conf->offloads to rte_eth_[rt]x_queue_setup().
If any offloading is enabled in rte_eth_dev_configure() by application,
it is enabled on all queues no matter whether it is per-queue or
per-port type and no matter whether it is set or cleared in
[rt]x_conf->offloads to rte_eth_[rt]x_queue_setup().
If a per-queue offloading hasn't be enabled in rte_eth_dev_configure(),
it can be enabled or disabled for individual queue in
ret_eth_[rt]x_queue_setup().
A new added offloading is the one which hasn't been enabled in
rte_eth_dev_configure() and is reuqested to be enabled in
rte_eth_[rt]x_queue_setup(), it must be per-queue type,
otherwise trigger an error log.
The underlying PMD must be aware that the requested offloadings
to PMD specific queue_setup() function only carries those
new added offloadings of per-queue type.
This patch can make above such checking in a common way in rte_ethdev
layer to avoid same checking in underlying PMD.
This patch assumes that all PMDs in 18.05-rc2 have already
converted to offload API defined in 17.11 . It also assumes
that all PMDs can return correct offloading capabilities
in rte_eth_dev_infos_get().
In the beginning of [rt]x_queue_setup() of underlying PMD,
add offloads = [rt]xconf->offloads |
dev->data->dev_conf.[rt]xmode.offloads; to keep same as offload API
defined in 17.11 to avoid upper application broken due to offload
API change.
PMD can use the info that input [rt]xconf->offloads only carry
the new added per-queue offloads to do some optimization or some
code change on base of this patch.
Signed-off-by: Wei Dai <wei.dai@intel.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Drivers should only log with their assigned logtype, not with the
generic PMD log type.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Be consistent with usage in other drivers.
No need for snowflake drivers.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Alejandro Lucero <alejandro.lucero@netronome.com>
The PMD_INIT_LOG macro always adds a newline, and other drivers version
of PMD_DRV_LOG always adds a newline. Therefore change nfp driver
to be consitent with others.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Alejandro Lucero <alejandro.lucero@netronome.com>
PMDs have the responsibility of releasing mbufs sent through xmit burst
function. NFP PMD attaches those sent mbufs to the TX ring structure,
and it is at the next time a specific ring descriptor is going to be
used when the previous linked mbuf, already transmitted at that point,
is released. Those mbufs belonging to a chained mbuf got its own link
to a ring descriptor, and they are released independently of the mbuf
head of that chain.
The problem is how those mbufs are released when the PMD is stopped or
closed. Instead of releasing those mbufs as the xmit functions does,
this is independently of being in a mbuf chain, the code calls
rte_pktmbuf_free which will release not just the mbuf head in that
chain but all the chained mbufs. The loop will try to release those
mbufs which have already been released again when chained mbufs exist.
This patch fixes the problem using rte_pktmbuf_free_seg instead.
Fixes: b812daadad ("nfp: add Rx and Tx")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
This patch allows to use another MAC address than the one coming
with the NIC by default.
The change requires to tell the vNIC after writing into the port
BAR space. The change will fail if the port is enabled and the
vNIC does not support a live address change.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Chained metadata instead of prepend metadata was added in
firmware version 4. However, it could be old firmwares evolving
but not supporting chained metadata.
This patch adds support for an old firmware being updated and
getting a firmware version number higher than 4, but it still not
implementing chained metadata.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
This new LSO offload version facilitates how firmware implements
this functionality and helps improving the performance.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Public struct rte_eth_dev_info has a "struct rte_pci_device" field in it
although it is common for all ethdev in all buses.
Replacing pci specific struct with generic device struct and updating
places that are using pci device in a way to get this information from
generic device.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: David Marchand <david.marchand@6wind.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
In next 18.05 the old hw offload API will be removed. This patch adds
support for just the new hw offload API.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
PF PMD support was based on NSPU interface. This patch changes the
PMD for using the new CPP user space interface which gives more
flexibility for adding new functionalities.
This change just affects initialization with the datapath being the
same than before.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Use new rte_eth_linkstatus_get/set helper functions to handle link
status update.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Mixing numeric macros with bit shifts macros is not a good idea.
Fixes: 011411586e ("net/nfp: extend speed capabilities advertised")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
The barrier needs to be after reading the DD bit. It has not been
a problem because the potential reads which can not happen before
reading the DD bit seem to be far enough, so the compiler is not
rescheduling them. However, a refactoring could make this problem
to arise.
Fixes: b812daadad ("nfp: add Rx and Tx")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Although this can be done by the app, because other PMDs are doing it,
apps expect this behaviour from the PMD.
Fixes: b812daadad ("nfp: add Rx and Tx")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
This commit aligns the names for dynamic logging with
the newly defined logging format.
Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Create a rte_ethdev_driver.h file and move PMD specific APIs here.
Drivers updated to include this new header file.
There is no update in header content and since ethdev.h included by
ethdev_driver.h, nothing changed from driver point of view, only
logically grouping of APIs. From applications point of view they can't
access to driver specific APIs anymore and they shouldn't.
More PMD specific data structures still remain in ethdev.h because of
inline functions in header use them. Those will be handled separately.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
The pointer to the user parameter of the callback registration is
automatically pass to the callback function.
There is no point to allow changing this user parameter by a caller.
That's why this parameter is always set to NULL by PMDs and set only
in ethdev layer before calling the callback function.
The history is that the user parameter was initially used
by the callback implementation to pass some information
between the application and the driver:
c1ceaf3ad0 ("ethdev: add an argument to internal callback function")
Then a new parameter has been added to leave the user parameter
to its standard usage of context given at registration:
d6af1a13d7 ("ethdev: add return values to callback process API")
The NULL parameter in the internal callback processing function
is now removed. It makes clear that the callback parameter is user
managed and opaque from a DPDK point of view.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Some apps can enable RSS but not update the reta table nor the hash.
This patch adds a default reta table setup based on total number of
configured rx queues. The hash key is dependent on how the app
configures the rx_conf struct.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
NFP does CRC strip by default and it is not configurable. But, even
if an app requests not to do it, that should not be a reason for PMD
configuration failure.
Fixes: defb9a5dd1 ("nfp: introduce driver initialization")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
When jumbo frames is configured, the hardware mtu needs to be updated to
the specified max_rx_pkt_len. Also, changing mtu should be avoided once
the PMD port started.
Fixes: defb9a5dd1 ("nfp: introduce driver initialization")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
The wrong mtu length was used for configuring the hardware. The
max_rx_pktlen reported was also wrong.
Fixes: defb9a5dd1 ("nfp: introduce driver initialization")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Not all struct fields will be written and random data could
confuse readers.
Coverity issue: 140755
Fixes: 92aa491b88 ("nfp: add statistics")
Cc: stable@dpdk.org
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
PF multiport support has some limitations like not allowing plugging or
unplugging of single ports as they all belong to same PCI device.
A previous patch for removing detachable flag introduced a wrong check.
Fixes: 00a3d8104a ("ethdev: remove detachable device flag")
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
The following inline functions and macros have been renamed to be
consistent with the IOVA wording:
rte_mbuf_data_dma_addr -> rte_mbuf_data_iova
rte_mbuf_data_dma_addr_default -> rte_mbuf_data_iova_default
rte_pktmbuf_mtophys -> rte_pktmbuf_iova
rte_pktmbuf_mtophys_offset -> rte_pktmbuf_iova_offset
The deprecated functions and macros are kept to avoid breaking the API.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Rename buf_physaddr to buf_iova.
Keep the deprecated name in an anonymous union to avoid breaking
the API.
Signed-off-by: Santosh Shukla <santosh.shukla@caviumnetworks.com>
Reviewed-by: Anatoly Burakov <anatoly.burakov@intel.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
The struct rte_memzone field .phys_addr is renamed to .iova.
The deprecated name is kept in an anonymous union to avoid breaking
the API.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Santosh Shukla <santosh.shukla@caviumnetworks.com>
Some devices may not support or fail setting VLAN offload
configuration based on dynamic circumstances so the
vlan_offload_set_t vector is modified to return an int so
the caller can determine success or not.
rte_eth_dev_set_vlan_offload is updated to return the
value provided by the vector when called along with restoring
the original offload configs on failure.
Existing vlan_offload_set_t vectors are modified to return
an int. Majority of cases return 0 but a few that actually
can fail now return their failure codes.
Finally, a vlan_offload_set_t vector is added to virtio
to facilitate dynamically turning VLAN strip on or off.
Signed-off-by: David Harton <dharton@cisco.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
