Now that mlx5_pci PMD checks for enabled classes and performs
probe(), remove() of associated classes, individual class driver
does not need to check if other driver is enabled.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Migrate mlx5 net, vdpa and regex PMD to start using mlx5 common class
driver.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Add generic mlx5 PCI PMD layer as part of existing common_mlx5
module. This enables multiple classes (net, regex, vdpa) PMDs
to be supported at same time.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
mlx5 PCI Device supports multiple classes of devices such as net, vdpa,
and/or regex.
To support these multiple classes, change mlx5_class to a
bitmap values so that if users asks to enable multiple of them, all
supported classes can be parsed.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
mlx5_common is shared library between mlx5 net, VDPA and regex PMD.
It is better to use common initialization helper instead of using
RTE_PRIORITY_CLASS priority.
Suggested-by: David Marchand <david.marchand@redhat.com>
Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
When fail to initialize the device, avoid segmentation fault while
accessing uninitialized priv.
Fixes: cfc672a90b ("regex/mlx5: support probing")
Signed-off-by: Parav Pandit <parav@mellanox.com>
Following two errors are reported when compiled with
gcc (GCC) 8.3.1 20191121 (Red Hat 8.3.1-5).
drivers/common/mlx5/linux/mlx5_glue.h:188:2:
error: function declaration isn't a prototype [-Werror=strict-prototypes]
drivers/common/mlx5/linux/mlx5_glue.h:188:2:
error: function declaration isn't a prototype [-Werror=strict-prototypes]
Fix them by adding void data type in empty argument list.
Fixes: 34fa7c0268 ("net/mlx5: add drop action to Direct Verbs E-Switch")
Fixes: 400d985eb5 ("net/mlx5: add VLAN push/pop DR commands to glue")
Cc: stable@dpdk.org
Signed-off-by: Parav Pandit <parav@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Drivers dependencies are evaluated in the order defined per their parent
directory (also called class). This strict ordering prevents from
having pairs of drivers from two classes with different dependency
ordering. For example, if the mlx5 common code depends on the pci bus
driver, while the dpaax common code is itself a dependency of the dpaa
bus driver. Having a strict ordering between directories bus and common
is too restrictive, as processing either common drivers or bus drivers
first leads us to missing dependencies in this scenario.
This patch makes it possible to have a more fine-grain directory list,
adding a specific driver sub-directory in the top-level drivers
subdirectory list. In this case, the isolated driver must also be removed
from its class list, and the per-class variables must be duplicated in
the isolated driver, because the call "subdir(class)" is skipped in the
isolated driver case.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Define each sub-directory on its own line ended with a comma,
and use a simple indent.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
There are several drivers which duplicate bit generation macro.
Introduce a generic bit macros so that such drivers avoid redefining
same in multiple drivers.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Acked-by: Morten Brørup <mb@smartsharesystems.com>
test_delay_us_sleep registers sleep based delay for testing.
This changes the default delay function of testing environment.
It is not expected.
Restore default delay function after the test to fix the issue.
Fixes: a51639cc72 ("eal: add nanosleep based delay function")
Cc: stable@dpdk.org
Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
The function rte_zmalloc_socket() could return NULL, the return
value need to be checked.
Fixes: 5915699153 ("hash: fix scaling by reducing contention")
Cc: stable@dpdk.org
Reported-by: Bin Huang <brian.huangbin@huawei.com>
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Acked-by: Yipeng Wang <yipeng1.wang@intel.com>
While uninitalizing the driver, the comment was mentioning
initialized instead of uninitalized in description of parameter.
Fixes: 0fe11ec592 ("eal: add vdev init and uninit")
Cc: stable@dpdk.org
Signed-off-by: Muhammad Bilal <m.bilal@emumba.com>
fixed typing error in doc/guides/tools/testbbdev.rst
Fixes: f714a18885 ("app/testbbdev: add test application for bbdev")
Cc: stable@dpdk.org
Signed-off-by: Sarosh Arif <sarosh.arif@emumba.com>
Acked-by: Nicolas Chautru <nicolas.chautru@intel.com>
The display was not proper due to the missing space. Changed
arm64 to aarch64.
Fixes: 2eb7c526b9 ("doc: clarify IOMMU disabling for uio_pci_generic")
Cc: stable@dpdk.org
Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
When perf-config option is specified, we are calling into the power
library even though it may not necessarily be enabled. It is
questionable whether perf-config option is even applicable to non-power
library modes, but for now, fix it just by avoiding calling into the
power library if it wasn't initialized, and assume that every lcore is
high performance core.
Fixes: e0194feb32 ("examples/l3fwd-power: add interrupt-only mode")
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
Anything coming from sysfs has a newline at the end. Cut it off before
comparing the strings.
Fixes: 20ab67608a ("power: add environment capability probing")
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
Tested-by: Lihong Ma <lihongx.ma@intel.com>
Reviewed-by: Bruce Richardson <bruce.richardson@intel.com>
If allocation is successful on the first attempt, typically
there is no problem since we allocated everything required and
we'll terminate the loop (if memory chunk is really sufficient
to populate required number of mempool elements).
If the first attempt fails, we try to allocate half
of mem_size and it succeed, we'll have one more iteration of
the for-loop to allocate memory for remaining elements and
should not try the next time with quarter of the mem_size.
It is wrong that max_alloc_size is divided by 2 in the
case of successful allocation as well, or invalid memory
can be allocated, and leads to population failure, then errno
other than ENOMEM may be returned.
Fixes: 3a3d0c75b4 ("mempool: fix slow allocation of large pools")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Signed-off-by: Zhike Wang <wangzhike@jd.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
This node classifies pkts based on packet type and
sends them to appropriate next node. This is node
helps in distribution of packets from ethdev_rx node
to different next node with a constant overhead for
all packet types.
Currently all except non fragmented IPV4 packets are marked
to be sent to "pkt_drop" node.
Performance difference on ARM64 Octeontx2 is -4.9% due to
addition of new node in the path.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Add fecmode setting on NIOS SPI primary initialization.
this SPI is shared by NIOS core inside FPGA, NIOS will
use this SPI primary to do some one-time initialization
after power up, and then release the control to DPDK.
Fix the timeout initialization for polling the
NIOS_INIT_DONE.
Fixes: bc44402f ("raw/ifpga/base: configure FEC mode")
Cc: stable@dpdk.org
Signed-off-by: Tianfei Zhang <tianfei.zhang@intel.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>
0x4a means idle status on physical layer. when encounter
0x4a on raw data, it need insert a ESCAPE character for
indication.
Fixes: 96ebfcf8 ("raw/ifpga/base: add SPI and MAX10 device driver")
Cc: stable@dpdk.org
Signed-off-by: Tianfei Zhang <tianfei.zhang@intel.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>
The commit below changed the readiness condition of vhost device to fix
multi-queues issues showed with QEMU versions.
Now, the vhost device is ready when the first queue-pair is ready.
When more queues are being ready, the queue state callback will be
triggered to notify the vhost manager.
In case of Rx interrupt configuration, the vhost driver set the
kickfd queue file descriptor in order to be notified on Rx traffic.
So, when queue becomes ready, the kickfd may be changed and should be
updated in the Rx interrupt structure.
Update kickfd when the queue state callback is invoked.
Also update event notification when it is enabled by the user.
Fixes: d0fcc38f5f ("vhost: improve device readiness notifications")
Suggested-by: Marvin Liu <yong.liu@intel.com>
Signed-off-by: Matan Azrad <matan@mellanox.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Rx RSS hash offload should be controlled by the user and should
not be enforced by RSS multi-queue Rx mode.
Fixes: 8b945a7f7d ("drivers/net: update Rx RSS hash offload capabilities")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Defines and functions from rte_mbuf_ptype.h are not used.
Only libefx types and EF10 register definitions are used.
Native datapaths should be independent from main libefx interface.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
efx_types.h uses defines from efx_annote.h, but does not include the
header. As the result if efx_types.h is included by a driver first,
build fails.
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
when trying to compile rte_mpls with pedantic enabled,
on old compilers like 4.8 it will complain about bit field definition.
error: type of bit-field 'bs' is a GCC extension [-Werror=pedantic]
error: type of bit-field 'tc' is a GCC extension [-Werror=pedantic]
error: type of bit-field 'tag_lsb' is a GCC extension [-Werror=pedantic]
This fixes the compilation error by adding extension to the header
definition.
Fixes: e480cf487a ("net: add MPLS header structure")
Cc: stable@dpdk.org
Signed-off-by: Raslan Darawsheh <rasland@mellanox.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
zmbufs should be set to NULL when getting freed to avoid double free on
the same buffer pointer
Fixes: b0a985d1f3 ("vhost: add dequeue zero copy")
Cc: stable@dpdk.org
Signed-off-by: Patrick Fu <patrick.fu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
In async enqueue copy, a packet could be split into multiple copy
segments. When polling the copy completion status, current async data
path assumes the async device callbacks are aware of the packet
boundary and return completed segments only if all segments belonging
to the same packet are done. Such assumption are not generic to common
async devices and may degrade the copy performance if async callbacks
have to implement it in software manner.
This patch adds tracking of the completed copy segments at vhost side.
If async copy device reports partial completion of a packets, only
vhost internal record is updated and vring status keeps unchanged
until remaining segments of the packet are also finished. The async
copy device is no longer necessary to care about the packet boundary.
Fixes: cd6760da10 ("vhost: introduce async enqueue for split ring")
Signed-off-by: Patrick Fu <patrick.fu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Vring should not be touched if vq is disabled. This patch adds the vq
status check in async enqueue polling to avoid accessing to a disabled
queue.
Fixes: cd6760da10 ("vhost: introduce async enqueue for split ring")
Signed-off-by: Patrick Fu <patrick.fu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
This patch adds the check of dev pointer in vhost async enqueue
completion poll. If a NULL dev pointer detected, the poll function
returns immediately.
Coverity issue: 360839
Fixes: cd6760da10 ("vhost: introduce async enqueue for split ring")
Signed-off-by: Patrick Fu <patrick.fu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Free CQ ring memzone on Rx queue release. This prevents CQ using
incorrect memory size when ring size is reconfigured.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
PMD flow priority is different from application flow priority. Flow
rules with higher match granularity assigned higher PMD priority. Also
PMD splits internally RSS flows according to flow RSS layer.
Final PMD flow rule priority derived from the last match item network
level, after PMD adjusts flow rule, where L4 match gets the highest
priority and L2 the lowest.
The patch adjusts tunnels flow rule priority calculation for PMDs
running verb API.
Introduce MLX5_TUNNEL_PRIO_GET macro.
Fixes: 4a78c88e3b ("net/mlx5: fix Verbs flow tunnel")
Cc: stable@dpdk.org
Signed-off-by: Gregory Etelson <getelson@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
Push VLAN action is allowed on Tx only, same as encap action.
Flow rules for hairpin queue are created on Rx, and split
by PMD to Rx and Tx rules, according to the above limitation.
In current implementation the encap action is split to Tx rule.
This patch adds the same handling for push-vlan action, as well as
its complementing actions set-vlan-vid and set-vlan-pcp.
Fixes: d85c7b5ea5 ("net/mlx5: split hairpin flows")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
The combination of decap action followed by pop VLAN action is not
fully validated in existing code.
This patch updates the validation function of pop vlan action.
Pop VLAN with preceding Decap requires inner header with VLAN.
Pop VLAN without preceding Decap requires outer header with VLAN.
Fixes: b41e47da25 ("net/mlx5: support pop flow action on VLAN header")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
When debugging performance of a DPDK application the user may
need to view the different statistics of DPDK (for example out_of_buffer)
This can be enabled by using testpmd command 'show port xstats
<port_id>' for example.
The current implementation assumes legacy mode in which the counters
are at <ibdev_path>/<port_id>/hw_counters/<file_name>.
In switchdev mode the counters file is located right after the device
name, hence resides at <ibdev_path>/hw_counters.
The fix tries to open the path in the second location after a failure
to open the file from the first location.
Fixes: 9c0a9eed37 ("net/mlx5: switch to the names in the shared IB context")
Cc: stable@dpdk.org
Signed-off-by: Shy Shyman <shys@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
The mlx5 PMD did not support queue_start and queue_stop eth_dev API
routines, queue could not be suspended and resumed during device
operation.
There is the use case when this feature is crucial for applications:
- there is the secondary process handling the queue
- secondary process crashed/aborted
- some mbufs were allocated or used by secondary application
- some mbufs were allocated by Rx queues to receive packets
- some mbufs were placed to send queue
- queue goes to undefined state
In this case there was no reliable way to recovery queue handling
by restarted secondary process but reset queue to initial state
freeing all involved resources, including buffers involved in queue
operations, reset the mbuf pools, and then reinitialize queue
to working state:
- reset mbuf pool, allocate all mbuf to initialize pool into
safe state after the crush and allow safe mbuf free calls
- stop queue, free all potentially involved mbufs
- reset mbuf pool again
- start queue, reallocate mbufs needed
This patch introduces the queue start/stop feature with some
limitations:
- hairpin queues are not supported
- it is application responsibility to synchronize start/stop
with datapath routines, rx/tx_burst must be suspended during
the queue_start/queue_stop calls
- it is application responsibility to track queue usage and
provide coordinated queue_start/queue_stop calls from
secondary and primary processes.
- Rx queues with vectorized Rx routine and engaged CQE
compression are not supported by this patch currently
Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
This patch continues the work to use DevX API for different objects
creation and management.
On Rx control path, the RQ, RQT, and TIR objects can already be
created using DevX API.
This patch adds the support to create CQ for RxQ using DevX API.
The corresponding event channel is also created and utilized using
DevX API.
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Several source files include Verbs header files as in (1). These source
files will not compile under non-Linux operating systems. This commit
removes this inclusion in two cases:
Case 1: There is no usage of ibv_* or mlx5dv_* symbols in the source
file so the inclusion in (1) can be safely removed.
Case 2: Verbs symbols are used. Please note the inclusion in (1) already
appears in file linux/mlx5_glue.h (which represents the interface
to the rdma-core library). Therefore, replace (1) in the source file
with (2). Under non-Linux operating systems - file mlx5_glue.h will not
include (1).
(1)
#include <infiniband/verbs.h>
#include <infiniband/mlx5dv.h>
(2)
#include <mlx5_glue.h>
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
1. The shared data communication between the primary and the secondary
processes is implemented using Linux API. Move the Linux API code under
linux directory (file linux/mlx5_os.c).
2. File net/mlx5/mlx5_mp.c handles requests to the primary and secondary
processes (e.g. start_rxtx, stop_rxtx). It is Linux based so it is moved
under linux (new file linux/mlx5_mp_os.c).
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
The cleanup refers to header file mlx5.h.
1. Remove unused prototypes.
2. Move prototypes under their correct title.
3. Change functions to static and remove their prototye from the header
file.
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
This commit eliminates Linux dependencies in shared file mlx5.h.
1. All functions using 'struct ifreq' are moved to file
linux/mlx5_ethdev_os.c such that this struct can be removed from mlx5.h.
2. Function mlx5_set_flags() that uses Linux flags (e.g. IFF_UP) is
changed to static and its prototype is removed from mlx5.h.
3. Remove redundant member verbs_action from 'struct mlx5_priv'.
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
This commit adds Linux implementation of routines mlx5_os_set_promisc()
and mlx5_os_set_promisc(). The routines call netlink APIs.
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Move OS specific MAC operations add, remove, modify VF into file
linux/mlx5_os.c.
Remove unused function mlx5_get_mac().
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
The following Linux calls are replaced by their matching rte APIs.
mmap ==> rte_mem_map()
munmap == >rte_mem_unmap()
sysconf(_SC_PAGESIZE) ==> rte_mem_page_size()
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Function calls mlx5_flow_adjust_priority() and
mlx5_flow_discover_priorities() are Verbs based. Move them from file
mlx5_flow.c to file mlx5_flow_verbs.c
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
There are some limitations on some NICs (at least on ConnectX-6 Dx
and BlueField 2) with supporting FCS (frame checksum) scattering for
the tunnel decapsulated packets.
For the case only one of the features can be supported in the same time,
and the new devarg "decap_en" is introduced to provide the choice to the
users.
If FCS scattering feature is not supposed to be engaged by application,
this new devarg should be specified as "decap_en=0", forcing the FCS
feature enable and rejecting tunnel decap actions in the rte_flow engine.
If FCS scatter is not needed and application supposes to use tunnel
decapsulation in rte_flow, the devarg can be omitted or set to non-zero
value (this is default settings).
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
As scatter FCS might be not supported for decapsulated tunnel
packets in some NIC HW, a new capability bit which indicates
if scatter FCS works with decap is added.
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
