Phil Yang [Fri, 18 Oct 2019 11:21:30 +0000 (19:21 +0800)]
stack: enable lock-free implementation for aarch64
Enable both C11 atomic and non C11 atomic lock-free stack for aarch64.
Introduced a new header to reduce the ifdef clutter across generic and C11
files. The rte_stack_lf_stubs.h contains stub implementations of
__rte_stack_lf_count, __rte_stack_lf_push_elems and
__rte_stack_lf_pop_elems.
Suggested-by: Gage Eads <gage.eads@intel.com> Suggested-by: Jerin Jacob <jerinj@marvell.com> Signed-off-by: Phil Yang <phil.yang@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Tested-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Phil Yang [Fri, 18 Oct 2019 11:21:29 +0000 (19:21 +0800)]
test/atomic: add 128-bit atomic compare exchange test
Add 128-bit atomic compare and swap test for aarch64 and x86_64.
Extend the test iteration from 10 thousand to 1 million times to test
the stability of the atomic APIs.
Signed-off-by: Phil Yang <phil.yang@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Acked-by: Gage Eads <gage.eads@intel.com> Acked-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Jerin Jacob <jerinj@marvell.com>
Phil Yang [Fri, 18 Oct 2019 11:21:28 +0000 (19:21 +0800)]
eal/arm64: add 128-bit atomic compare exchange
This patch adds the implementation of the 128-bit atomic compare
exchange API on aarch64. Using 64-bit 'ldxp/stxp' instructions
can perform this operation. Moreover, on the LSE atomic extension
accelerated platforms, it is implemented by 'casp' instructions for
better performance.
Since the '__ARM_FEATURE_ATOMICS' flag only supports GCC-9, this
patch adds a new config flag 'RTE_ARM_FEATURE_ATOMICS' to enable
the 'cas' version on older version compilers.
For octeontx2, we make sure that the lse (and other) extensions are
enabled even if the compiler does not know of the octeontx2 target
cpu.
Since direct x0 register used in the code and cas_op_name() and
rte_atomic128_cmp_exchange() is inline function, based on parent
function load, it may corrupt x0 register aka break aarch64 ABI.
Define CAS operations as rte_noinline functions to avoid an ABI
break [1].
Suggested-by: Jerin Jacob <jerinj@marvell.com> Signed-off-by: Phil Yang <phil.yang@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Tested-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com> Reviewed-by: David Marchand <david.marchand@redhat.com>
Jim Harris [Mon, 7 Oct 2019 15:28:21 +0000 (08:28 -0700)]
eal: calibrate TSC only in primary process
This ensures secondary processes never have to calculate the TSC rate
themselves, which can be noticeable in VMs that don't have access to
arch-specific detection mechanism (such as CPUID leaf 0x15 or MSR 0xCE
on x86).
Since rte_mem_config is now internal to the EAL library, we can add
tsc_hz without ABI breakage concerns.
Reduces rte_eal_init() execution time in a secondary process from 165ms
to 66ms on my test system.
Signed-off-by: Jim Harris <james.r.harris@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Ruifeng Wang [Mon, 9 Sep 2019 13:51:42 +0000 (21:51 +0800)]
rcu: fix spurious thread unregister
Thread unregister returns success while unregister not been performed.
This is due to incorrect thread registration status check.
Fix this issue by correcting bitmap check.
Fixes: 64994b56cfd7 ("rcu: add RCU library supporting QSBR mechanism") Cc: stable@dpdk.org Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com> Reviewed-by: Gavin Hu <gavin.hu@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Reviewed-by: David Marchand <david.marchand@redhat.com>
For a valid service, the core mask of the service
is checked against the current core and the corresponding
entry in the active_on_lcore array is set or reset.
Upto 8 cores share the same cache line for their
service active_on_lcore array entries since each entry is a uint8_t.
Some number of these entries also share the cache line with
the internal_flags member of struct rte_service_spec_impl,
hence this false sharing also makes the service_valid() check
expensive.
Eliminate false sharing by moving the active_on_lcore array to
a per-core data structure. The array is now indexed by service id.
Signed-off-by: Nikhil Rao <nikhil.rao@intel.com> Acked-by: Gage Eads <gage.eads@intel.com> Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
Jim Harris [Mon, 7 Oct 2019 15:40:05 +0000 (08:40 -0700)]
timer: remove useless check on x86 TSC reliability
This code was added 7+ years ago in
commit fb022b85bae4 ("timer: check TSC reliability")
presumably when variant TSCs were still somewhat common.
But this code doesn't do anything except print a warning,
and the warning doesn't give any kind of advice to the user,
so let's just remove it.
While the warning has no functional meaning, the /proc/cpuinfo
parsing consumes a non-trivial amount of time which is especially
noticeable in secondary processes.
On my test system, it consumes 21ms out of the 66ms total execution
time for rte_eal_init() in a secondary process.
Signed-off-by: Jim Harris <james.r.harris@intel.com> Acked-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
The rte_atomic64_exchange operation for ppc_64 incorrectly linked
back to a 32 bit generic operation (__atomic_exchange_4) rather than
the 64 bit generic operation (__atomic_exchange_8). As a result,
applications that used rte_eth_link_get_nowait() would only receive
the link speed, they would not receive the link state, link duplex,
or link autoneg properties.
Fixes: ff2863570fcc ("eal: introduce atomic exchange operation") Cc: stable@dpdk.org Signed-off-by: David Christensen <drc@linux.vnet.ibm.com> Reviewed-by: David Marchand <david.marchand@redhat.com>
This is a commonly used operation that surprisingly the
DPDK has not supported. The new rte_pktmbuf_copy does a
deep copy of packet. This is a complete copy including
meta-data.
It handles the case where the source mbuf comes from a pool
with larger data area than the destination pool. The routine
also has options for skipping data, or truncating at a fixed
length.
This patch also introduces internal inline to copy the
metadata fields of mbuf.
Add a test for this new function, based of the clone tests.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Olivier Matz <olivier.matz@6wind.com>
This copy part of this function is too big to be put inline.
The places it is used are only in special exception paths
where a highly fragmented mbuf arrives at a device that can't handle it.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Andrew Rybchenko <arybchenko@solarflare.com> Acked-by: Olivier Matz <olivier.matz@6wind.com>
The test for cloning changed mbuf would generate an mbuf whose length
and segments count were invalid.
This would cause a crash if test was run with mbuf debugging enabled.
EAL should always use rte_log instead of putting errors to
stderr (which maybe redirected to /dev/null in a daemon).
Also checks for null before rte_free are unnecessary.
Minor code consistency improvements.
Fixes: 21698354c832 ("service: introduce service cores concept") Cc: stable@dpdk.org Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com> Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
Jim Harris [Fri, 16 Aug 2019 12:13:42 +0000 (05:13 -0700)]
vfio: fix leak with multiprocess
The code checks both rte_mp_request_sync() return code and that the number
of messages in the reply equals 1. If rte_mp_request_sync() succeeds but
there was more than one message, those messages would get leaked.
Found via code review by Anatoly Burakov of patches that used the vhost
code as a template for using rte_mp_request_sync().
A while ago telemetry was added in 57ae0ec6 and it also added as-needed
to config/meson.build. This seems no more needed these days as due to other
build changes the ordering in buildlogs is:
[...] -lrte_telemetry [...] -Wl,--no-as-needed [...]
Which means telemetry no more benefits from --no-as-needed anyway.
Overlinking problems get triggered by the meson generated pkgconfig which
will have:
[...] -Wl,--no-as-needed <somelibsusedbydpdk>
This will overlink <somelibs> and in addition anything that follows
as it also doesn't wrap back to --as-needed. So if a projects includes
dpdk libs + <other> it will also consider <other> with --no-as-needed.
Ubuntu bug: https://bugs.launchpad.net/ubuntu/+source/dpdk/+bug/1841759 Fixes: 57ae0ec62620 ("build: add dependency on telemetry to apps with meson") Cc: stable@dpdk.org Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com> Acked-by: Luca Boccassi <bluca@debian.org>
Jerin Jacob [Wed, 14 Aug 2019 13:09:53 +0000 (18:39 +0530)]
bpf: hide internal program argument type
RTE_BPF_ARG_PTR_STACK is used as internal program
arg type. Rename to RTE_BPF_ARG_RESERVED to
avoid exposing internal program type.
Signed-off-by: Jerin Jacob <jerinj@marvell.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Jerin Jacob [Tue, 3 Sep 2019 10:59:37 +0000 (16:29 +0530)]
bpf/arm: add atomic-exchange-and-add operation
Implement XADD eBPF instruction using STADD arm64 instruction.
If the given platform does not have atomics support,
use LDXR and STXR pair for critical section instead of STADD.
Signed-off-by: Jerin Jacob <jerinj@marvell.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
app/eventdev: add options for mbuf and packet sizes
Add options to set mbuf size and max packet size which allow the user to
enable jumbo frames and Rx/Tx scatter gather.
Arrange `struct evt_options` based on ascending order of data type to
make it more readable.
Packet mbuf size can be modified by using `--mbuf_sz=N`.
Max packet size can be modified by using `--max_pkt_sz=N`.
These options are only applicable `pipeline_atq` and `pipeline_queue`
tests.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
The existing code uses NULL as the cipher algo
for testing crypto event adapter.
DPAA1/DPAA2 do not support NULL algo. Hence changing
it to the most common algo AES-CBC, which is supported
by all crypto drivers implementing event crypto adapter.
Pavan Nikhilesh [Mon, 19 Aug 2019 10:46:51 +0000 (16:16 +0530)]
event/octeontx2: fix Rx adapter capabilities
Octeontx2 SSO co-processor allows multiple ethernet device Rx queues
connected to a single Event device queue.
Fix the Rx adapter capabilities to allow application to configure
Rx queueus in n:1 ratio to event queues by adding
`RTE_EVENT_ETH_RX_ADAPTER_CAP_MULTI_EVENTQ` as a capability.
Gage Eads [Tue, 27 Aug 2019 21:34:59 +0000 (16:34 -0500)]
event/sw: fix xstats reset value
The sw PMD implements xstats reset by having the xstat get operations
return a value to the statistic's value at the last reset. The value at the
last reset is maintained in the per-xstat reset_value field, but the PMD
was setting reset_value = current - reset_value instead of reset_value =
current.
Fixes: c1ad03df7ad5 ("event/sw: support xstats") Cc: stable@dpdk.org Signed-off-by: Gage Eads <gage.eads@intel.com> Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
In LAG configuration the devices in the same switch domain
might be spawned on the base of different PCI devices, so
we should check all devices backed by mlx5 PMD whether they
belong to specified switch domain. When the new devices are
being created it is not possible to detect whether the
sibling devices created in the current probe() loop belong
to the driver, driver field is not filled yet (it will be
done on returned success of current probe()). This patch
updates the device scanning, allowing extra match on
current backing PCI device, is being used to create siblings.
net/mlx5: adjust inline setting for large Tx queue sizes
The hardware may have limitations on maximal amount of
supported Tx descriptors building blocks (WQEBB). Application
requires the Tx queue must accept the specified amount of packets.
If inline data feature is engaged the packet may require more WQEBBs
and overall amount of blocks may exceed the hardware capabilities.
Application has to make a trade-off between Tx queue size and maximal
data inline size.
In case if the inline settings are not requested explicitly with
devarg keys the default values are used. This patch adjusts the
applied default values if large Tx queue size is requested and
default inline settings can not be satisfied due to hardware
limitations.
The explicitly requested inline setting may be aligned (enlarging
only) by configurations routines to provide better WQEBB filling,
this implicit alignment is the subject for adjustment either.
The warning message is emitted to the log if adjustment happens.
When rules are being inserted from multiple cores, there are several
race conditions during rte_flow operations.
For example, when inserting rules from 2 cores simultaneously, both
the cores try to fetch a free available filter entry and they both
end up fetching the same entry. Both of them start overwriting the
same filter entry before sending to firmware, which results in wrong
rule being inserted to hardware.
Fix the races by adding spinlock to serialize the rte_flow operations.
Use first receive queue assigned to VNIC as the default receive queue
when configuring Thor VNICs. This is necessary e.g. in order for flow
redirection to a specific receive queue to work correctly.
The required number of statistics contexts is computed as the sum
of the number of receive and transmit rings plus one for the async
completion ring. A statistics context is not actually required for
the async completion ring, so remove it from the calculation.
Thor queue scaling is currently limited by the number of NQs that
can be allocated. Fix by using a common NQ for all receive/transmit
rings instead of allocating a separate NQ for each ring.
Venkat Duvvuru [Fri, 4 Oct 2019 03:48:58 +0000 (20:48 -0700)]
net/bnxt: support CoS classification
Class of Service (CoS) is a way to manage multiple types of
traffic over a network to offer different types of services
to applications. CoS classification (priority to cosqueue) is
determined by the user and configured through the PF driver.
DPDK driver queries this configuration and maps the cos queue
ids to different VNICs. This patch adds this support.
net/bnxt: fix ring alignment for Thor-based adapters
When using transmit/receive queue sizes smaller than 256, alignment
requirements are not being met for Thor-based adapters. Fix by
forcing memory addresses used for transmit/receive/aggregation ring
allocations to be on 4K boundaries.
Pavel Belous [Fri, 20 Sep 2019 16:22:07 +0000 (16:22 +0000)]
net/atlantic: add FW mailbox guard mutex
Driver uses the Firmware mailbox to read statistics and configure
some features.
This patch introduces a mutex to provide consistent access to the
FW mailbox to prevent potential data corruption.
Fixes: 86d36773bd42 ("net/atlantic: implement firmware operations") Cc: stable@dpdk.org Signed-off-by: Pavel Belous <pavel.belous@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com>
Pavel Belous [Fri, 20 Sep 2019 16:22:05 +0000 (16:22 +0000)]
net/atlantic: fix reported flow control mode
Driver reports current flow control mode based on internal flow control
settings. Currently this logic works incorrectly.
Fixes: 921eb6b8ce31 ("net/atlantic: fix flow control by sync settings on Rx") Cc: stable@dpdk.org Signed-off-by: Pavel Belous <pavel.belous@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com>
Pavel Belous [Fri, 20 Sep 2019 16:22:02 +0000 (16:22 +0000)]
net/atlantic: exclude MACsec counters from xstats
Currently, driver always return full set of xstats counters, including
MACSEC counters. But this driver also supports AQC100 chips, which
does not have MACSEC feature.
This fix adds checking for MACSEC availability (based on FW capability
bits) and returns xstats without MACSEC counters if MACSEC feature
is not available.
Fixes: 09d4dfa85359 ("net/atlantic: implement MACsec statistics") Cc: stable@dpdk.org Signed-off-by: Pavel Belous <pavel.belous@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com>
Rasesh Mody [Wed, 2 Oct 2019 19:14:56 +0000 (12:14 -0700)]
net/bnx2x: update to latest FW 7.13.11
Use latest firmware 7.13.11.
7.13.11 FW changelog:
- Packets from a VF with pvid configured which were sent with a
different vlan were transmitted instead of being discarded.
- In some multi-function configurations, inter-PF and inter-VF
Tx switching is incorrectly enabled.
- Wrong assert code in FLR final cleanup in case it is sent not
after FLR.
- Chip may stall in very rare cases under heavy traffic with FW GRO
enabled.
- VF malicious notification error fixes.
- Default gre tunnel to IPGRE which allows proper RSS for IPGRE
packets, L2GRE traffic will reach single queue.
- Removes unnecessary internal mem config, latest FW performs this
autonomously.
Rasesh Mody [Wed, 2 Oct 2019 19:14:54 +0000 (12:14 -0700)]
net/bnx2x: update and reorganize HW registers
Update and reorganize HW registers in preparation to update the firmware
to version 7.13.11.
Move HW_INTERRUT_ASSERT_SET_0 out from ecore_reg.h to bnx2x.h.
Allow VFs to enable backpressure for performance reasons.
The backpressure control is with kernel AF driver that will enable
backpressure even if one PF/VF requests it and disable it only
after all the PFs/VFs request for disable.
Kalesh AP [Wed, 2 Oct 2019 23:26:00 +0000 (16:26 -0700)]
net/bnxt: fix multicast filter programming
Fixed multicast filter programming and allmulti programming.
Fixed to skip programming multicast macs if the user requests
allmulti mode.
Also removed a comment in bnxt_hwrm_cfa_l2_set_rx_mask() which is
no longer valid now.
Fixes: d69851df12b2 ("net/bnxt: support multicast filter and set MAC addr") Cc: stable@dpdk.org Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com> Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Venkat Duvvuru [Wed, 2 Oct 2019 23:25:59 +0000 (16:25 -0700)]
net/bnxt: fix VLAN filtering
Currently, when hw-vlan-filter is enabled on testpmd, driver is
receiving all vlan packets. Instead, it should only receive untagged
packets and vlan packets for which the VLAN filter is programmed.
This is because, the default rule to match on MAC is not getting
deleted, when hw-vlan-filter is ON.
This patch fixes the problem, by deleting the default MAC rule and
programming a new rule to receive only untagged packets, when
hw-vlan-filter is enabled & another rule for each vlan, as and when
that vlan is configured on that port.
Venkat Duvvuru [Wed, 2 Oct 2019 23:25:57 +0000 (16:25 -0700)]
net/bnxt: synchronize between flow related functions
Currently, there are four flow related functions, namely
bnxt_flow_create, bnxt_flow_destroy, bnxt_flow_validate,
bnxt_flow_flush. All these functions are not multi-thread safe.
This patch fixes it by synchronizing these functions with a lock.
Somnath Kotur [Wed, 2 Oct 2019 23:25:54 +0000 (16:25 -0700)]
net/bnxt: check for VNIC ID in rollback
If driver init/probe fails as part of cleanup/rollback, we may end
up invoking this HWRM cmd even on an invalid vNIC which will
unnecessarily log an error message as the cmd will fail.
Check for invalid ID before issuing the HWRM cmd
Ajit Khaparde [Wed, 2 Oct 2019 23:25:53 +0000 (16:25 -0700)]
net/bnxt: check if device is started before flow creation
Check device is started before flow creation.
Since the vnic data structures aren't created until device start,
the driver dereferences NULL vnic if flow creation is attempted before
device is started,
Ajit Khaparde [Wed, 2 Oct 2019 23:25:49 +0000 (16:25 -0700)]
net/bnxt: cleanup VNIC after flow validate
When an application issues flow validate, we free the temporary
filter that is created. But the vnic is not freed up. This can
potentially interfere with subsequent flow creation. So free the vnic.
Ajit Khaparde [Wed, 2 Oct 2019 23:25:48 +0000 (16:25 -0700)]
net/bnxt: delete and flush L2 filters cleanly
Once the last filter associated with a VNIC is deleted when using
RSS action or the Queue action free the VNIC. Also free the RSS
context if the VNIC is using it.
Ajit Khaparde [Wed, 2 Oct 2019 23:25:47 +0000 (16:25 -0700)]
net/bnxt: parse priority attribute for flow creation
Parse priority attribute during flow creation.
This information will be used to give a hint to the FW to
place the flow rule accordingly in the CFA tables.
Ajit Khaparde [Wed, 2 Oct 2019 23:25:46 +0000 (16:25 -0700)]
net/bnxt: support RSS action
Add support for RSS action during flow creation.
group id should not be 0 when RSS action is specified. Driver will
return an error for such a flow.
If a group id is used to create a filter with “n” RSS queues, it cannot
be used to create a filter with a different number of RSS queues till
all the flows using that combination are deleted.
While creating a flow if a group id groups a certain Rx queue ids for
RSS, the same group id shall not create a flow with a different group of
Rx queue ids till all the flows belonging to the group ids are deleted.
While creating a flow if a group id groups a certain Rx queue ids for
RSS, the same queue ids shall not be used with a different group id till
all flows created with that group id are deleted.
Ajit Khaparde [Wed, 2 Oct 2019 23:25:45 +0000 (16:25 -0700)]
net/bnxt: support creating SMAC and inner DMAC filters
We are currently creating only outer DMAC filters.
Create SMAC and inner DMAC filters using HWRM_CFA_L2_FILTER_ALLOC.
For this the HWRM_CFA_L2_FILTER_ALLOC has already been updated.
Thomas Monjalon [Sat, 14 Sep 2019 09:37:00 +0000 (11:37 +0200)]
net/szedata2: fix dependency check
The library libsze2 provides a pkg-config file: libsze2.pc.
Looking for this .pc file - with dependency() - is preferred
than looking for the library - with cc.find_library().
If the library is not installed in a standard path,
it can be found thanks to PKG_CONFIG_PATH variable.
The previous solution required to use CFLAGS and LDFLAGS
environment variables.
Fixes: 508cfe6be9f1 ("net/szedata2: add to meson build") Cc: stable@dpdk.org Signed-off-by: Thomas Monjalon <thomas@monjalon.net> Acked-by: Jan Remes <remes@netcope.com>
Thomas Monjalon [Sat, 14 Sep 2019 09:36:59 +0000 (11:36 +0200)]
net/nfb: fix dependency check
The library libnfb is part of netcope-common which provides
a pkg-config file: netcope-common.pc.
Looking for this .pc file - with dependency() - is preferred
than looking for the library - with cc.find_library().
If the library is not installed in a standard path,
it can be found thanks to PKG_CONFIG_PATH variable.
The previous solution required to use CFLAGS and LDFLAGS
environment variables.
Fixes: 6435f9a0ac22 ("net/nfb: add new netcope driver") Cc: stable@dpdk.org Signed-off-by: Thomas Monjalon <thomas@monjalon.net> Acked-by: Jan Remes <remes@netcope.com>
Shreyansh Jain [Thu, 29 Aug 2019 10:27:35 +0000 (15:57 +0530)]
net/dpaa2: support separate MC portal per process
Earlier, there was a single MCP handle which was available across
complete DPAA2 driver as well as part of the dev_private which was
shared by the secondary process.
For secondary, that is not valid and it would require to open its
own handle for the MC. This is eventually used as part of the DPNI
configuration.
By using the process_private member of the rte_eth_dev, it is
possible to keep separate handles per process. Without worry of
overwriting when secondary process accesses the dev_private.
Shreyansh Jain [Thu, 29 Aug 2019 10:27:32 +0000 (15:57 +0530)]
bus/fslmc: restrict address translation to PA mode
The address translation support for PA->VA is required only in
case of PA mode operation of DPDK. This was causing warning to
be reported on running any DPAA2 application in VA mode:
Add: Incorrect entry for PA->VA Table(xxxxxxxxxx)
Add: Lowest address: xxxxxxxxxxxx
This was caused by call to update the DPAAX table when VA mode
was enabled, in which case the VA==IOVA address.
git.droids-corp.org / dpdk.git / log