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>
Introduce an helper for PMD to expand easily flows items list with RSS
action into multiple flow items lists with priority information.
For instance a user items list being "eth / end" with rss action types
"ipv4-udp ipv6-udp end" needs to be expanded into three items lists:
- eth
- eth / ipv4 / udp
- eth / ipv6 / udp
to match the user request. Some drivers are unable to reach such
request without this expansion, this API is there to help those.
Only PMD should use such API for their internal cooking, the application
will still handle a single flow.
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
- We don't support QinQ offload, so removing it now.
- Fix incorrect offload flags in default rxconf
Since qede PMD does not support per queue rx offload, it
should not set default_rxconf.offload flags in .dev_infos_get().
Although these offloads are enabled by default, they are per port
and not per queue.
Fixes: 946dfd18a4 ("net/qede: convert to new Rx/Tx offloads API")
Cc: stable@dpdk.org
Signed-off-by: Shahed Shaikh <shahed.shaikh@cavium.com>
This patch disables extended VLAN offload by default as PMD does not
support it.
Fixes: d87246a437 ("net/qede: enable and disable VLAN filtering")
Cc: stable@dpdk.org
Signed-off-by: Rasesh Mody <rasesh.mody@cavium.com>
The optimal values of several transmission & reception related
parameters, such as burst sizes, descriptor ring sizes, and number
of queues, varies between different network interface devices. This
patch adds the values for the ixgbe PMD.
Signed-off-by: Remy Horton <remy.horton@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Since its return type is unsigned, if_nametoindex() returns 0 in case of
error, never -1.
Fixes: ccdcba53a3 ("net/mlx5: use Netlink to add/remove MAC addresses")
Cc: stable@dpdk.org
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
If WQE ID is used in CQE for Multi-Packet RQ, the ratio of CQE compression
drops a little bit. In order to reach to 100Gbps with 64B traffic, it is
needed to further save PCIe bandwidth by increasing the number of strides
in a WQE. It is now 64 by default but adjustable by a PMD parameter -
mprq_log_stride_num.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
Multi-Packet Receive Queue is to receive multiple packets on a single large
buffer. The number of consumed strides in CQE is accumulated to keep track
of the current stride index. However, it is safer to directly use stride
index in CQE to avoid out-of-order situation which can possibly be caused
by introducing LRO in the future.
If Rx CQE compression is enabled, HW can be configured to store the stride
index in a mini-CQE but this will need newer version of library/driver.
Therefore, since this change, MPRQ is only supported with the newer
library/driver and Rx hash result is not supported if MPRQ is enabled along
with Rx CQE compression.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
If Multi-Packet RQ is enabled but not supported by device or
kernel/library, print out a warning message.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Stride index is added to mlx5_mini_cqe8 structure and WQE ID is added to
mlx5_cqe structure.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
mlx5_rx_poll_len() returns Rx hash result extracted from either mini CQE or
regular CQE. As mini CQE may not have the hash result if configured
otherwise, it shouldn't assume the first DWORD of mini CQE is always hash
result. mlx5_rx_poll_len() is changed to return pointer to the mini CQE if
compressed.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
The threshold of buffer replenishment for vectorized Rx burst is a constant
value (64). If the size of Rx queue is comparatively small, device could
run out of buffers. For example, if the size of Rx queue is 128, buffers
are replenished only twice per a wraparound. This can cause jitter in
receiving packets and the jitter can cause unnecessary retransmission for
TCP connections.
Fixes: 6cb559d67b ("net/mlx5: add vectorized Rx/Tx burst for x86")
Fixes: 570acdb1da ("net/mlx5: add vectorized Rx/Tx burst for ARM")
Cc: stable@dpdk.org
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
NFP CPP interface dinamically configures NFP CPP BARs for accessing
any NFP chip component from user space. This requires to map PCI BAR
regions specifically. However, this does not require to do such map
over the usual map done by VFIO or UIO drivers with the device PCI
BARs.
This patch avoids this remapping and therefore also avoids to access
the device sysfs resource0 file for doing that remapping.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Getting the bar size is required for NFP CPP interface configuration.
However, this information can be obtained from the VFIO or UIO driver
instead of accessing the sysfs resource file.
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
COUNT action has been modified and has several fields not addressable
though testpmd. In addition, as those fields are not definable testpmd
is providing an empty configuration which is undefined.
Fixes: fb8fd96d42 ("ethdev: add shared counter to flow API")
Cc: stable@dpdk.org
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
When the number of rx queues is 0 (what can be when application does
not receive) failed with SIGFPE.
It happens when the application is also requesting ETH_MQ_RX_RSS_FLAG
in the rte_dev->data->dev_conf.rxmode.mq_mode.
Fixed adding zero rx queues check.
Signed-off-by: Daria Kolistratova <daria.kolistratova@intel.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>
This commit implements TCP segmentation offload in TAP.
librte_gso library is used to segment large TCP payloads (e.g. packets
of 64K bytes size) into smaller MTU size buffers.
By supporting TSO offload capability in software a TAP device can be used
as a failsafe sub device and be paired with another PCI device which
supports TSO capability in HW.
For more details on librte_gso implementation please refer to dpdk
documentation.
The number of newly generated TCP TSO segments is limited to 64.
Reviewed-by: Raslan Darawsheh <rasland@mellanox.com>
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
Prior to this commit IP/UDP/TCP checksum offload calculations
were skipped in case of a multi segments packet.
This commit enables TAP checksum calculations for multi segments
packets.
The only restriction is that the first segment must contain
headers of layers 3 (IP) and 4 (UDP or TCP)
Reviewed-by: Raslan Darawsheh <rasland@mellanox.com>
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
This was added to dev_stop when set MTU requested vport restart.
We are not requiring vport restart any more with set MTU as it just
needs vport to be inactive and doesn't need the port reconfigured.
Fixes: d121a6b5f7 ("net/qede: fix VF MTU update")
Cc: stable@dpdk.org
Signed-off-by: Rasesh Mody <rasesh.mody@cavium.com>
qede pmd does not have support for legacy interrupt mode.
This causes slow path completion failure with uio_pci_generic module,
since it uses legacy interrupt (INTx) mode.
Fix this issue by installing legacy interrupt handler.
Fixes: ec94dbc573 ("qede: add base driver")
Cc: stable@dpdk.org
Signed-off-by: Shahed Shaikh <shahed.shaikh@cavium.com>
We did not register unicast mac configuration handlers
for VF causing failure in bonding of VFs.
Also, mac_addr_set operation requires mac_remove followed
by mac_add.
Fixes: 86a2265e59 ("qede: add SRIOV support")
Cc: stable@dpdk.org
Signed-off-by: Shahed Shaikh <shahed.shaikh@cavium.com>
build error gcc version 6.3.1 20161221 (Red Hat 6.3.1-1),
with EXTRA_CFLAGS="-O3":
.../drivers/net/thunderx/nicvf_ethdev.c:907:9:
error: ‘txq’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
if (txq->pool_free == nicvf_single_pool_free_xmited_buffers)
~~~^~~~~~~~~~~
.../drivers/net/thunderx/nicvf_ethdev.c:886:20:
note: ‘txq’ was declared here
struct nicvf_txq *txq;
^~~
Same error on function 'nicvf_eth_dev_init' and 'nicvf_dev_start', it
seems 'nicvf_set_tx_function' inlined when optimization enabled.
Initialize the txq and add NULL check before using it to fix.
Fixes: 7413feee66 ("net/thunderx: add device start/stop and close")
Cc: stable@dpdk.org
Reported-by: Richard Walsh <richard.walsh@intel.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Add functions to add/remove MAC addresses
Signed-off-by: Alex Kiselev <alex@therouter.net>
Acked-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Chas Williams <chas3@att.com>
static logging macro RTE_PMD_DEBUG_TRACE is enabled with a few DEBUG
config options, including RTE_LIBRTE_ETHDEV_DEBUG
RTE_LIBRTE_ETHDEV_DEBUG is still used for data path logging, but all
ethdev logging switched to dynamic logging, so no need to enable static
logging macro for ethdev.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Replace RTE_PMD_DEBUG_TRACE with RTE_ETHDEV_LOG.
RTE_PMD_DEBUG_TRACE is using hardcoded PMD logtype and ERR log level,
controlled by compile time flags.
RTE_ETHDEV_LOG is using dynamic ethdev_logtype.
Also a few minor cleanups, like
- use %u for unsigned values like port_id which is uint16_t
- use PRIx64 for owner_id
- Join some log lines
- Unify to not have a "." at the end of the log
- Unify log start with uppercase
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Macro moved to header to be able to convert logging usage in header.
And since it has been moved to public header changed naming and added
RTE prefix, ethdev_log -> RTE_ETHDEV_LOG
Also need to add logtype variable to map file since logging macro used
from other libraries.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Change open_rx/tx_pcap/iface functions to open only a single pcap/dumper
and not loop num_of_queue times
The num_of_queue loop is already achieved by the
caller rte_kvargs_process
Fixing:
1. Opens N requested pcaps/dumpers instead of N^2
2. Leak of pcap/dumper's which are being overwritten by
the sequential calls to open_rx/tx_pcap/iface functions
3. Use the filename/iface args per queue and not just the last one
that overwrites the previous names
Fixes: 4c173302c3 ("pcap: add new driver")
Cc: stable@dpdk.org
Signed-off-by: Ido Goshen <ido@cgstowernetworks.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
This feature is used to create a hole between HEADROOM
and actual data.Size of hole is specified in bytes as
module param to pmd
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@caviumnetworks.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
The generic tunnel TSO was depended in the regular one capabilities to
be enabled.
Cc: stable@dpdk.org
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
The log_cache_nb_elem was never incremented, resulting
in all dirty pages to be missed during live migration.
Fixes: c16915b871 ("vhost: improve dirty pages logging performance")
Cc: stable@dpdk.org
Reported-by: Peng He <xnhp0320@icloud.com>
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Ilya Maximets <i.maximets@samsung.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
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>
At this point valid data is already set by rte_eth_get_device_info.
device field becomes zero and consumer is not able to retrieve pci data.
Fixes: 4861cde461 ("i40e: new poll mode driver")
Cc: stable@dpdk.org
Signed-off-by: Damjan Marion <damarion@cisco.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Some illegal packets will lead to TX/RX hang and
can't recover automatically. This patch check those
illegal packets and protect TX/RX from hanging.
Signed-off-by: Yanglong Wu <yanglong.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The GL_SWR_PM_UP_THR value is not impacted from the link speed, its
value is set according to the total number of ports for a better
pipe-monitor configuration.
All bellowing relevant device IDs are considered (NICs, LOMs, Mezz
and Backplane):
Device-ID Value Comments
0x1572 0x03030303 10G SFI
0x1581 0x03030303 10G Backplane
0x1586 0x03030303 10G BaseT
0x1589 0x03030303 10G BaseT (FortPond)
0x1580 0x06060606 40G Backplane
0x1583 0x06060606 2x40G QSFP
0x1584 0x06060606 1x40G QSFP
0x1587 0x06060606 20G Backplane (HP)
0x1588 0x06060606 20G KR2 (HP)
0x158A 0x06060606 25G Backplane
0x158B 0x06060606 25G SFP28
Fixes: c9223a2bf5 ("i40e: workaround for XL710 performance")
Fixes: 75d133dd32 ("net/i40e: enable 25G device")
Cc: stable@dpdk.org
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The vlan tci mask should be set to 0xEFFF, not 0x0,
the wrong mask will cause mask error for register set.
Fixes: d9d5e6f2f0 ("app/testpmd: set default flow director mask")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
The summarized global register change info will be logged
no matter if there's real global register change. Since
only real changes are logged now, there's no need to
summarize global register change info, otherwise will
cause misunderstanding.
Signed-off-by: Beilei Xing <beilei.xing@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Currently no matter if there's global change, the global
configuration will be always logged. But there's no value
to log the info if the configuration is not changed.
This patch prints only real global changes.
Also, change log level from DEBUG to WARNING.
Signed-off-by: Beilei Xing <beilei.xing@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Cppcheck reports following error,
(error) Shifting 32-bit value by 36 bits is undefined behaviour
According to datasheet, there's PHY type and PHY type extension
in setting PHY config command, should exclude PHY type extension
when setting PHY type.
Fixes: 1bb8f66116 ("net/i40e: fix link down and negotiation")
Cc: stable@dpdk.org
Signed-off-by: Beilei Xing <beilei.xing@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
MAC address bits in mask registers should be set to zero
when the is mac mask is 0xFF, otherwise if it is 0x0
these bits should be to 0x3F.
Fixes: 82fb702077 ("ixgbe: support new flow director modes for X550")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Tunnel type format should be translated to ixgbe required format
before register set in FDIR cloud mode, Ans also some register
not useful in cloud mode but only useful in IP mode should be set
to zero as datasheet request.
Fixes: 82fb702077 ("ixgbe: support new flow director modes for X550")
Fixes: 11777435c7 ("net/ixgbe: parse flow director filter")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
In cloud mode for FDIR, tunnel id should be set as protocol
request, the lower 8 bits should be set as reserved.
Fixes: 82fb702077 ("ixgbe: support new flow director modes for X550")
Fixes: 11777435c7 ("net/ixgbe: parse flow director filter")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
In IP mode FDIR, X550 can support not only 4 tuple parameters
but also vlan tci in protocol, so add this feature to flow parser.
Fixes: 11777435c7 ("net/ixgbe: parse flow director filter")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
There are many registeres in x550 support stats of
flow director filters, for example the number of added
or removed rules and the number match or miss match packet
count for this for port, all these important information
can be read form registeres in x550 and display with command
xstats.
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
When detaching a port bound to ixgbe PMD, if the port
does not have any VFs, *vfinfo is not set and there is
a NULL dereference attempt, when calling
rte_eth_switch_domain_free(), which expects VFs to be used,
causing a segmentation fault.
Steps to reproduce:
./testpmd -- -i
testpmd> port stop all
testpmd> port close all
testpmd> port detach 0
Bugzilla ID: 57
Fixes: cf80ba6e20 ("net/ixgbe: add support for representor ports")
Cc: stable@dpdk.org
Reported-by: Anatoly Burakov <anatoly.burakov@intel.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Remy Horton <remy.horton@intel.com>
rte_errno should be saved only if error has occurred because rte_errno
could have garbage value.
Fixes: a6d83b6a92 ("net/mlx5: standardize on negative errno values")
Cc: stable@dpdk.org
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Split maintainers logs from user logs.
A lot of debug logs are present providing internal information on how
the PMD works to users. Such logs should not be available for them and
thus should remain available only when the PMD is compiled in debug
mode.
This commits removes some useless debug logs, move the Maintainers ones
under DEBUG and also move dump into debug mode only.
Cc: stable@dpdk.org
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Write combining (WC) increases NIC performance by making better
utilization of PCI bus. ENA PMD may make usage of this feature.
To enable it load igb_uio driver with wc_activate set to 1.
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
