UMR (User-Mode Memory Registration) WQE can present data buffers
scattered within multiple mbufs with single indirect mkey. Take
advantage of the UMR WQE, scattered mbuf in one operation can be
presented to an indirect mkey. The RegEx which only accepts one
mkey can now process the whole scattered mbuf in one operation.
The maximum scattered mbuf can be supported in one UMR WQE is now
defined as 64. The mbufs from multiple operations can be combined
into one UMR WQE as well if there is enough space in the KLM array,
since the operations can address their own mbuf's content by the
mkey's address and length. However, one operation's scattered mbuf's
can't be placed in two different UMR WQE's KLM array, if the UMR
WQE's KLM does not has enough free space for one operation, the
extra UMR WQE will be engaged.
In case the UMR WQE's indirect mkey will be over wrapped by the SQ's
WQE move, the mkey's index used by the UMR WQE should be the index
of last the RegEX WQE in the operations. As one operation consumes
one WQE set, build the RegEx WQE by reverse helps address the mkey
more efficiently. Once the operations in one burst consumes multiple
mkeys, when the mkey KLM array is full, the reverse WQE set index
will always be the last of the new mkey's for the new UMR WQE.
In GGA mode, the SQ WQE's memory layout becomes UMR/NOP and RegEx
WQE by interleave. The UMR and RegEx WQE can be called as WQE set.
The SQ's pi and ci will also be increased as WQE set not as WQE.
For operations don't have scattered mbuf, uses the mbuf's mkey directly,
the WQE set combination is NOP + RegEx.
For operations have scattered mbuf but share the UMR WQE with others,
the WQE set combination is NOP + RegEx.
For operations complete the UMR WQE, the WQE set combination is UMR +
RegEx.
Signed-off-by: John Hurley <jhurley@nvidia.com>
Signed-off-by: Suanming Mou <suanmingm@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
The high priority match request flags means that the
RegEx engine should stop on the first match.
This commit add this flag check to the RegEx engine.
Signed-off-by: Ori Kam <orika@nvidia.com>
In order to know which groups in the RegEx engine
should be used there is a need to check the req_flags.
This commit adds the missing check.
Fixes: 4d4e245ad637 ("regex/mlx5: support enqueue")
Cc: stable@dpdk.org
Signed-off-by: Ori Kam <orika@nvidia.com>
This commit propagate the response flags from the regex engine.
Signed-off-by: Francis Kelly <fkelly@nvidia.com>
Signed-off-by: Ori Kam <orika@nvidia.com>
The constant representing the size of the metadata is defined as a
regular number (32-bit signed), even though all of its uses request an
unsigned int variable.
Similarly the constant representing the maximal output is also defined
as a regular number, even though all of its uses request an unsigned int
variable.
Change the type of the above constants to unsigned.
Fixes: 5f41b66d12cd ("regex/mlx5: setup fast path")
Cc: stable@dpdk.org
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
Currently job data is being copied to pre-registered buffer.
To avoid memcpy on the datapath, use dynamic memory registration.
This change will reduce latency when sending regex jobs. The first few
jobs may have high latency due to registration, but assuming all
following mbufs will arrive from the same mempool/hugepage, there will
be no further memory registration.
Signed-off-by: Yuval Avnery <yuvalav@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
Since the 20.08 release deprecated rte_cio_*mb APIs because these APIs
provide the same functionality as rte_io_*mb APIs on all platforms, so
remove them and use rte_io_*mb instead.
Signed-off-by: Phil Yang <phil.yang@arm.com>
Signed-off-by: Joyce Kong <joyce.kong@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Acked-by: David Marchand <david.marchand@redhat.com>
Added missing code to free Input/Output buffers and memory
registration.
Also added calls to this code in case of error in the qp setup
procedure.
The rollback code itself did not handle rollback properly
and did not check return value from the fastpath setup.
Signed-off-by: Yuval Avnery <yuvalav@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
When enqueueing a buffer the PMD check if there is room
in its send queue (SQ).
The current implementation did not take into account that
queue indices are wrapping around, which may result in
consumer index (sq->ci) can have bigger value than than
the producer index (sq->pi).
Fixes: 4d4e245ad637 ("regex/mlx5: support enqueue")
Signed-off-by: Yuval Avnery <yuvalav@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
Will look for a free SQ to send the job on.
doorbell will be given when sq is full, or no more jobs on the burst.
Signed-off-by: Yuval Avnery <yuvalav@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
