While the PMD avoids from creating hash RXQ with no hash fields and
array of queues after the port was already started, it lacks such
protection when re-creating the flows after the port restarts.
This may lead to inconsistent behavior for flows depending if they were
created before or after the port start.
Fixes: 8086cf08b2 ("net/mlx5: handle RSS hash configuration in RSS flow")
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
VLAN id is limited to MLX5_MAX_VLAN_IDS which is not verified by the code
before trying to add a new VLAN filter.
Fixes: 272733b5eb ("net/mlx5: use flow to enable unicast traffic")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Initialise struct mlx5_fdir to avoid passing non initialised
items/actions arrays elements to the generic flow interpreter.
Fixes: 4c3e9bcdd5 ("net/mlx5: support flow director")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Fix a strange behavior from the NIC, when the flow starts with a VXLAN
layer with a VNI equals to zero all the traffic will match within this
rule.
Fixes: 2e709b6aa0 ("net/mlx5: support VXLAN flow item")
Cc: stable@dpdk.org
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
In validate() it sets the value to 1 whereas the expected one is
IBV_FLOW_SPEC_INNER.
Fixes: c2c6ddd51d ("net/mlx5: limit flow API rules to one tunnel")
Cc: stable@dpdk.org
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
The counter set should be destroyed only after the flow was destroyed.
Fixes: 9a761de8ea ("net/mlx5: flow counter support")
Signed-off-by: Ori Kam <orika@mellanox.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
When adding count action to a drop flow rule while the device was not
started, the counter was not saved in the flow.
Fixes: 9a761de8ea ("net/mlx5: flow counter support")
Signed-off-by: Ori Kam <orika@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Flow director API does not provide a layer 2 configuration when the
filter is for layer 3 and 4 causing the translation to generic flow API
to be wrong, as not providing a mask for layer ends by using the
default one.
In this case, the Ethernet mask layer is full whereas it must be empty.
Fixes: 4c3e9bcdd5 ("net/mlx5: support flow director")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
The flow engine in mlx5 searches for the most specific layer in the
pattern in order to set the flow rule priority properly.
Since the RSS can be currently performed only for the outer headers, avoid
updating the layer for the inner headers.
Fixes: 8086cf08b2 ("net/mlx5: handle RSS hash configuration in RSS flow")
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
The PCI lib defines the types and methods allowing to use PCI elements.
The PCI bus implements a bus driver for PCI devices by constructing
rte_bus elements using the PCI lib.
Move the relevant code out of the EAL to its expected place.
Libraries, drivers, unit tests and applications are updated to use the
new rte_bus_pci.h header when necessary.
Signed-off-by: Gaetan Rivet <gaetan.rivet@6wind.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>
LIST macro are not safe when inside a LIST_FOREACH() a LIST_REMOVE() is
called to remove an entry, this behavior is undefined causing some entries
to disappear from the list.
Fixes: 6e78005a9b ("net/mlx5: add reference counter on DPDK Tx queues")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Configuring UAR as IO-mapped makes maximum throughput decline by
noticeable amount. If UAR is configured as write-combining register,
a write memory barrier is needed on ringing a doorbell.
rte_wmb() is mostly effective when the size of a burst is comparatively
small. Revert the register back to write-combining and enforce a write
memory barrier instead, except for vectorized Tx burst routines.
Application can change it by setting MLX5_SHUT_UP_BF under its own
necessity.
Fixes: 9f9bebae55 ("net/mlx5: don't map doorbell register to write combining")
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Drop queue should also adjust their priority according the most specific
layer in the pattern they are matching to avoid dropping all the
traffic.
Fixes: 8086cf08b2 ("net/mlx5: handle RSS hash configuration in RSS flow")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
PKT_RX_VLAN_PKT and PKT_RX_QINQ_PKT are deprecated for a while.
As explained in [1], these flags were kept to let the applications and
PMDs move to the new flag. There is also a need to support Rx vlan
offload without vlan strip (at least for the ixgbe driver).
This patch renames the old flags for this feature, knowing that some
PMDs were using PKT_RX_VLAN_PKT and PKT_RX_QINQ_PKT to indicate that
the vlan tci has been saved in the mbuf structure.
It is likely that some PMDs do not set the proper flags when doing vlan
offload, and it would be worth making a pass on all of them.
Link: [1] http://dpdk.org/ml/archives/dev/2017-June/067712.html
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
This flag is not necessary at the ether layer anymore.
Buses are able to advertise their hotplug support. The ether layer can
rely upon this capability instead of a special flag.
Signed-off-by: Gaetan Rivet <gaetan.rivet@6wind.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Flows are added by priv_flow_create() in the associated list, adding them a
second time corrupts the list causing an infinite loop when parsing it.
Fixes: 4c3e9bcdd5 ("net/mlx5: support flow director")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
When VLAN is enabled in the Rx side, only packets matching this VLAN are
expected, this also includes the broadcast and all multicast packets.
Fixes: 272733b5eb ("net/mlx5: use flow to enable unicast traffic")
Fixes: 6a6b6828fe ("net/mlx5: use flow to enable all multi mode")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Marking a packet which will not be received by the NIC is useless, even
if this action remains possible, it blocks the creation of the flow
counter which embed a mark action to a drop queue to be created.
Fixes: 31ba9997f1 ("net/mlx5: fully convert a flow to verbs in validate")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Flow director drop action as not been brought back with the new
implementation on top of rte flow.
Fixes: 4c3e9bcdd5 ("net/mlx5: support flow director")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Fix the drop queue rule creation when the port is stopped.
Fixes: 8086cf08b2 ("net/mlx5: handle RSS hash configuration in RSS flow")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Indirection table size must be in log to communicate with verbs when the
number of queue is not a power of two, the maximum indirection table
size is use, but not converted to log2. This makes a memory corruption.
Fixes: 4c7a0f5ff8 ("net/mlx5: make indirection tables shareable")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
When ports are stopped, the hash Rx queue should not be created.
Fixes: 8086cf08b2 ("net/mlx5: handle RSS hash configuration in RSS flow")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Considering the PMD supports only Ethernet transport, packet which
arrives without any packet type flags in the completion should be
marked with L2_ETHER flag.
Fixes: ea16068c00 ("net/mlx5: fix L4 packet type support")
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>
Few bugs fixes in both configuration get and hash update where inputs
are not handled as expected by the ethdev layer.
RSS structure may not be totally usable, the PMD should try to take as
most information from it has it can when it is an hash update or it
should try to fill as most as possible in the configuration get.
This means that in the RSS configuration structure, the memory space for
the RSS hash key may not be present, but the PMD should consider the
hash field valid and process/set it.
Fixes: 29c1d8bb3e ("net/mlx5: handle a single RSS hash key for all protocols")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Priority is wrongly configured when the action is queue, using the
Ethernet layer priority instead of the most specific layer found.
Fixes: 8086cf08b2 ("net/mlx5: handle RSS hash configuration in RSS flow")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Interrupt disable can be called when the interrupt vector is not yet
allocated. Such case ends up with segmentation fault.
Fixing it by adding verification for interrupt vector validity.
Fixes: 09cb5b5817 ("net/mlx5: separate DPDK from verbs Rx queue objects")
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
The list of libraries in LDLIBS was generated from the DEPDIRS-xyz
variable. This is valid when the subdirectory name match the library
name, but it's not always the case, especially for PMDs.
The patches removes this feature and explicitly adds the proper
libraries in LDLIBS.
Some DEPDIRS-xyz variables become useless, remove them.
Reported-by: Gage Eads <gage.eads@intel.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: Gage Eads <gage.eads@intel.com>
The size of Rx completion queue should be doubled if compression is enabled
in case of non-vectorized Rx.
Fixes: 523f5a7421 ("net/mlx5: fix configuration of Rx CQE compression")
Cc: stable@dpdk.org
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
The stats_get dev op API doesn't include return value, so PMD cannot
return an error in case of failure at stats getting process time.
Since PCI devices can be removed and there is a time between the
physical removal to the RMV interrupt, the user may get invalid stats
without any indication.
This patch changes the stats_get API return value to be int instead of
void.
All the net PMDs stats_get dev ops are adjusted by this patch.
Signed-off-by: Matan Azrad <matan@mellanox.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Example for setting rule for counting packets with dest
ip = 192.168.3.1 in testpmd:
testpmd: flow create 0 ingress pattern eth / ipv4 dst is 192.168.3.1
/ end actions queue index 0 / count / end
Reading the number of packets and bytes for the rule:
testpmd: flow query 0 0 count
Note: This feature is only supported starting Mellanox OFED 4.2
Signed-off-by: Ori Kam <orika@mellanox.com>
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
When replenishing Rx ring, there're always buffered slots reserved
between consumed entries and HW owned entries. These have to be filled
with fake mbufs to protect from possible overflow rather than
optimistically expecting successful replenishment which can cause
deadlock with small-sized queue.
Fixes: fc048bd52c ("net/mlx5: fix overflow of Rx SW ring")
Cc: stable@dpdk.org
Reported-by: Martin Weiser <martin.weiser@allegro-packets.com>
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Tested-by: Martin Weiser <martin.weiser@allegro-packets.com>
With the upstream rdma-core, to enable Rx CQE compression,
mlx5dv_create_cq() in Direct Verbs has to be used instead of regular
Verbs call (ibv_create_cq()). And if the size of CQE is 128 bytes,
compression is supported only by certain devices. Thus, it has to be
decided by checking the capability bits.
Fixes: 43e9d9794c ("net/mlx5: support upstream rdma-core")
Cc: stable@dpdk.org
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
The size of Rx completion entry should match the size of a cacheline.
This is already reflected in struct mlx5_cqe by adding 64bytes padding
if a cacheline is 128bytes. Some ARM CPUs have 128bytes cacheline.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Considering more architecture (e.g. ARM and PowerPC) will be added for
vectorized Rx/Tx burst, all the shareable functions which don't use any
vector intrinsics need to be separated from architecture-dependent
functions. All the vector functions for x86 SSE are moved to a new
header file - mlx5_rxtx_vec_sse.h. And shareable common functions are
now in mlx5_rxtx_vec.c.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Replace compile-time sanity check with static_assert() as c11 standard
has been set. Add mlx5_rxtx_vec.h and move the sanity checks to the file
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Updating a consumer index to HW doesn't require a memory barrier in case
that there's no updated data to be posted to HW, but a compiler barrier
is sufficient. rte_wmb() is replaced with rte_io_wmb() when it makes
changes visible to HW, not other core.
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Isolated works exclusively with the generic flow API, this patch adds a
new set of operations valid in this mode.
- promiscuous*()
- allmulticast*()
- reta*()
- rss*()
are not supported in this mode as it is fully supported by generic flow
API.
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Support same functionalities as in
commit cf521eaa3c76 ("net/mlx5: remove flow director support")
This implementation is done on top of the generic flow API.
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
