Pablo de Lara [Thu, 16 Apr 2020 17:24:24 +0000 (18:24 +0100)]
test/crypto: do not check for internal PMD information
Now that capabilities are checked to see if an algorithm
is supported by a device, there is no need to check
for a specific version of a library used in a PMD.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Adam Dybkowski <adamx.dybkowski@intel.com> Tested-by: Ruifeng Wang <ruifeng.wang@arm.com> Acked-by: Fiona Trahe <fiona.trahe@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Pablo de Lara [Thu, 16 Apr 2020 17:24:23 +0000 (18:24 +0100)]
test/crypto: add capability check
Check if test case is supported by the crypto device,
including algorithm and some of its parameter, such as key length,
IV length, etc, using the capabilities API.
If it is not supported, test case is skipped.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Adam Dybkowski <adamx.dybkowski@intel.com> Tested-by: Ruifeng Wang <ruifeng.wang@arm.com> Acked-by: Fiona Trahe <fiona.trahe@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Adding optional capability to support loopback preloading
and check of the extern HARQ memory.
This function is required to run the HARQ bit exact test successfully.
Signed-off-by: Nicolas Chautru <nicolas.chautru@intel.com> Acked-by: Dave Burley <dave.burley@accelercomm.com> Acked-by: Niall Power <niall.power@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Nicolas Chautru [Sat, 18 Apr 2020 22:46:40 +0000 (15:46 -0700)]
baseband/fpga_5gnr_fec: add info get function
Add in the "info_get" function to the driver, to allow us to query the
device.
No capability are available yet.
Linking bbdev-test to support the PMD with null capability.
Signed-off-by: Nicolas Chautru <nicolas.chautru@intel.com> Acked-by: Dave Burley <dave.burley@accelercomm.com> Acked-by: Niall Power <niall.power@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Adam Dybkowski [Fri, 17 Apr 2020 15:44:04 +0000 (17:44 +0200)]
test/compress: check IM buffer too small
This patch adds new tests for verification of the "internal
QAT IM buffer too small" case handling. These unit tests aren't
specific to the QAT PMD only - they pass or skip on other PMDs like
ISAL and ZLIB (depending on particular PMD capabilities).
Signed-off-by: Adam Dybkowski <adamx.dybkowski@intel.com> Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Adam Dybkowski [Fri, 17 Apr 2020 15:44:03 +0000 (17:44 +0200)]
compress/qat: support IM buffer too small operation
This patch implements a special way of buffer handling when internal
QAT IM buffer is too small for Huffman dynamic compression operation.
Instead of falling back to fixed compression, the operation is now
split into multiple smaller dynamic compression requests (possible to
execute on QAT) and their results are then combined and copied into
the output buffer. This is not possible if any checksum calculation
was requested - in such case the code falls back to fixed compression
as before.
Signed-off-by: Adam Dybkowski <adamx.dybkowski@intel.com> Acked-by: Fiona Trahe <fiona.trahe@intel.com>
This patch introduces set of unit tests of librte_security API functions.
Tests are added to dpdk-test application and can be run with
"security_autotest" runtime command.
This is the first patch in the series of patches as adding all test cases
for all API functions in a single patch would make it unreadable.
This patch defines structure of the file and necessary test framework
initialization. It also contains first subset of unit tests for
rte_security_session_create API function.
Structure of the tests file is following:
- macros for making tests more readable;
- mockup structures and functions for rte_security_ops;
- test suite and test cases setup and teardown functions;
- tests functions;
- declaration of testcases.
Fix session counter to be decreased in rte_security_session_destroy
only when session was successfully destroyed.
Formerly session counter was decreased prior session destroying
and returning session object to mempool. It remained decreased even
if session was not destroyed and mempool object released making counter
invalid.
Fixes: c261d1431bd8 ("security: introduce security API and framework") Cc: stable@dpdk.org Signed-off-by: Lukasz Wojciechowski <l.wojciechow@partner.samsung.com> Acked-by: Anoob Joseph <anoobj@marvell.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
This patch adds verification of the parameters to the ret_security API
functions. All required parameters are checked if they are not NULL.
Checks verify full chain of pointers, e.g. in case of verification of
"instance->ops->session_XXX", they check also "instance"
and "instance->ops".
Fixes: c261d1431bd8 ("security: introduce security API and framework") Fixes: 1a08c379b9b5 ("security: support user data retrieval") Cc: stable@dpdk.org Signed-off-by: Lukasz Wojciechowski <l.wojciechow@partner.samsung.com> Acked-by: Anoob Joseph <anoobj@marvell.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Pablo de Lara [Tue, 14 Apr 2020 17:25:55 +0000 (18:25 +0100)]
crypto/openssl: fix out-of-place encryption
When authenticating after encrypting, if the operation
is out-of-place, the destination buffer is the one
that will get authenticated.
If the cipher offset is higher than the authentication
offset, it means that part of the text to authenticate
will be plaintext, so this needs to get copied to the
destination buffer, or the result will be incorrect.
Fixes: d61f70b4c918 ("crypto/libcrypto: add driver for OpenSSL library") Cc: stable@dpdk.org Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
This patch adds ability to run unit tests in cpu crypto mode
for AESNI MB cryptodev.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Support load distribution in security gateway application using
NIC load distribution feature (Flow Director).
Flow Director is used to redirect the specified inbound ipsec flow
to a specified queue. This is achieved by extending the SA rule syntax
to support specification by adding new action_type of <flow-direction>
to a specified <port_id> <queue_id>.
drivers/crypto: fix log type variables for -fno-common
The four crypto drivers: kasumi, mvsam, snow3g and zuc define logtype
variables in their header file. As the header files are included
in more than one compilation unit, it might cause appearance
of multiple instances of the variable and a linker error.
Such situation can occur, when no common section is allowed
by the compiler settings and tentative definitions are placed
in BSS section.
Bruce Richardson [Fri, 10 Apr 2020 17:27:56 +0000 (18:27 +0100)]
drivers: fix log type variables for -fno-common
The zlib compression driver, as well as the aesni-gcm, aesni-mb and openssl
crypto drivers all defined the logtype variable in the header file
directly. This gives errors with gcc 10, due to -fno-common being the
default, so we need to apply the same fix in all cases:
* move the variable definition to a suitable .c file
* mark the forward declaration of the variable in the header as "extern"
The goal of rte_os.h is to mitigate OS differences for EAL users.
In Windows EAL, rte_os.h did excessive things:
1. It included platform SDK headers (windows.h, etc). Those files are
huge, require specific inclusion order, and are generally unused by
the code including rte_os.h. Declarations from platform SDK may
break otherwise platform-independent code, e.g. min, max, ERROR.
2. It included pthread.h, which is clearly not always required.
3. It defined functions private to Windows EAL.
Reorganize Windows EAL includes in the following way:
1. Create rte_windows.h to properly import Windows-specific facilities.
Primary users are bus drivers, tests, and external applications.
2. Remove platform SDK includes from rte_os.h to prevent breaking
otherwise portable code by including rte_os.h on Windows.
Copy necessary definitions to avoid including those headers.
3. Remove pthread.h include from rte_os.h.
4. Move declarations private to Windows EAL into eal_windows.h.
Fixes: 428eb983f5f7 ("eal: add OS specific header file") Signed-off-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
This fix treats a 0 return value from vfio_open_group_fd
in vfio_get_group_fd as the intended error condition instead
of putting an incorrect 0 file descriptor in the vfio_group table.
Sometimes, the creation of device files in sysfs is not
instantaneously causing vfio_open_groupfd to return 0.
This has been observed when hot removing/adding multiple
NVMe devices (>=4).
Fixes: 340b7bb8d583 ("vfio: extend data structure for multi container") Cc: stable@dpdk.org Signed-off-by: Michael Haeuptle <michael.haeuptle@hpe.com> Acked-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com> Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
When rte_rand_init is invoked, and the kernel running dpdk does not
support *getentropy, at the same time, the cpu does not support rdseed,
then rte_rand_init invoked rte_get_timer_cycles.
If HPET was enabled in the DPDK build (CONFIG_RTE_LIBEAL_USE_HPET=y) and
the system, rte_get_timer_cycles will invoke rte_get_hpet_cycles while
*eal_hpet is not available.
To fix that, use rte_get_tsc_cycles instead of rte_get_timer_cycles.
Fixes: 3f002f069612 ("eal: replace libc-based random generation with LFSR") Cc: stable@dpdk.org Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com> Acked-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
Thinh Tran [Wed, 25 Mar 2020 00:13:56 +0000 (20:13 -0400)]
eal/ppc64: use glibc for cpu cycles count
__ppc_get_timebase() reads and returns the current value of the Time
Base Register. It's more efficient as it uses the processor’s time
base facility directly.
DPDK on FreeBSD currently is not supported on Powerpc64, it should
be safe to include the sys/platform/ppc.h.
Signed-off-by: Thinh Tran <thinhtr@linux.vnet.ibm.com> Reviewed-by: David Christensen <drc@linux.vnet.ibm.com>
Due to the fact that the rte_realloc() test depends on the layout of
underlying memory, it can sometimes fail due to fragmentation of the
memory. To address this, make it so that the realloc autotests are run
using a newly created external memory heap instead of main memory.
Bugzilla ID: 424
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Tested-by: David Christensen <drc@linux.vnet.ibm.com>
For rings with producer/consumer in RTE_RING_SYNC_ST, RTE_RING_SYNC_MT_HTS
mode, provide an ability to split enqueue/dequeue operation
into two phases:
- enqueue/dequeue start
- enqueue/dequeue finish
That allows user to inspect objects in the ring without removing
them from it (aka MT safe peek).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Introduce head/tail sync mode for MT ring synchronization.
In that mode enqueue/dequeue operation is fully serialized:
only one thread at a time is allowed to perform given op.
Suppose to reduce stall times in case when ring is used on
overcommitted cpus (multiple active threads on the same cpu).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Introduce relaxed tail sync (RTS) mode for MT ring synchronization.
Aim to reduce stall times in case when ring is used on
overcommited cpus (multiple active threads on the same cpu).
The main difference from original MP/MC algorithm is that
tail value is increased not by every thread that finished enqueue/dequeue,
but only by the last one.
That allows threads to avoid spinning on ring tail value,
leaving actual tail value change to the last thread in the update queue.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
To make these preparations two main things are done:
- Change from *single* to *sync_type* to allow different
synchronisation schemes to be applied.
Mark *single* as deprecated in comments.
Add new functions to allow user to query ring sync types.
Replace direct access to *single* with appropriate function call.
- Move actual rte_ring and related structures definitions into a
separate file: <rte_ring_core.h>. It allows to refer contents
of <rte_ring_elem.h> from <rte_ring.h> without introducing a
circular dependency.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Introduce stress test for ring enqueue/dequeue operations.
Performs the following pattern on each slave worker:
dequeue/read-write data from the dequeued objects/enqueue.
Serves as both functional and performance test of ring
enqueue/dequeue operations under high contention
(for both over committed and non-over committed scenarios).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
David Marchand [Fri, 17 Apr 2020 14:56:11 +0000 (16:56 +0200)]
eal/ppc: fix build
Compilation is broken on ppc:
CC otx2_rx.o
In file included from .../drivers/net/octeontx2/otx2_rx.c:5:0:
.../builds/ppc_64-power8-linux-gcc/include/rte_vect.h:29:17:
error: expected declaration specifiers or ‘...’ before numeric constant
} __rte_aligned(16) rte_xmm_t;
^~
compilation terminated due to -Wfatal-errors.
Fixes: f35e5b3e07b2 ("replace alignment attributes") Signed-off-by: David Marchand <david.marchand@redhat.com> Tested-by: Thomas Monjalon <thomas@monjalon.net>
Thomas Monjalon [Wed, 11 Mar 2020 23:01:35 +0000 (00:01 +0100)]
doc: remove empty columns from NIC features matrix
The virtual PMDs bonding, KNI, null, ring, softnic and vdev_netvsc
have no real feature to advertise so they can be removed
from the (too) big matrix of ethdev features.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Thomas Monjalon [Wed, 11 Mar 2020 23:01:34 +0000 (00:01 +0100)]
doc: fix matrix CSS for recent sphinx
It seems sphinx >= 2.0 is inserting a <p> tag in each table cell.
The feature table (matrix) style needs to be updated to avoid
cells being too big.
The margin, padding and line height are overridden.
The font size in percentage is replaced with an equivalent pixel size.
The border is explicit because it disappeared for th.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Thomas Monjalon [Mon, 10 Feb 2020 16:00:32 +0000 (17:00 +0100)]
devtools: check use of compiler attributes
The keyword __attribute__ will emit a warning,
because it is preferred to use or define a common __rte macro.
The centralized macros may help to control or workaround some compilers.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Currently, in the case to use bitmap as resource allocator, after
bitmap creation, all the bitmap bits should be set to indicate the
bit available. Every time when allocate one bit, search for the set
bits and clear it to make it in use.
Add a new rte_bitmap_init_with_all_set() function to have a quick
fill up the bitmap bits.
Comparing with the case create the bitmap as empty and set the bitmap
one by one, the new function costs less cycles.
Thomas Monjalon [Thu, 9 Apr 2020 16:41:49 +0000 (18:41 +0200)]
test: remove meson dependency on /proc file
Meson is detecting the path /proc/sys/vm/nr_hugepages in the call to cat
in app/test/meson.build and then adding it as a build dependency.
This causes build loop if the timestamp of this file keeps changing.
It is fixed by hiding hugepage check in a shell script.
Fixes: 77784ef0fba8 ("test: allow no-huge mode for fast-tests") Signed-off-by: Thomas Monjalon <thomas@monjalon.net> Tested-by: Lukasz Wojciechowski <l.wojciechow@partner.samsung.com> Reviewed-by: Lukasz Wojciechowski <l.wojciechow@partner.samsung.com> Acked-by: Aaron Conole <aconole@redhat.com> Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Pavan Nikhilesh [Tue, 31 Mar 2020 12:53:15 +0000 (18:23 +0530)]
examples/l2fwd-event: add option to configure port pairs
Current l2fwd-event application statically configures adjacent ports as
destination ports for forwarding the traffic.
Add a config option to pass the forwarding port pair mapping which allows
the user to configure forwarding port mapping.
If no config argument is specified, destination port map is not
changed and traffic gets forwarded with existing mapping.
To align port/queue configuration of each lcore with destination port
map, port/queue configuration of each lcore gets modified when config
option is specified.
With above config option, traffic received from portid = 0 gets forwarded
to port = 3 and vice versa, similarly traffic gets forwarded on other port
pairs (1,4) and (2,5).
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Reviewed-by: Andrzej Ostruszka <aostruszka@marvell.com>
DSW keeps an internal port load estimate, used by the load balancing
mechanism. As a side effect, it keeps track of the total number of
busy cycles since startup. This metric is indirectly exposed in the
form of DSW xstats' "port_<n>_event_proc_latency", which is the total
number of busy cycles divided by the total number of events processed
on a particular port.
An external application can take (event_latency * dequeued) to go back
to busy_cycles. One reason for doing this is to measure the port's
load during a longer time period, without resorting to sampling
"port_<n>_load". However, as the number dequeued events grows, a
rounding error in event_latency renders the application-calculated
busy_cycles inaccurate.
Thus, it makes sense to directly expose the number of busy cycles as a
DSW xstats, even though it might seem redundant.
DSW limits the rate of migrations on a per-port basis. Hence, as the
number of cores grows, so does the total migration capacity.
In high core-count systems, this allows for a situation where flows
are migrated to a lightly loaded port which recently already received
a number of new flows (from other ports). The processing load
generated by these new flows may not yet be reflected in the lightly
loaded port's load estimate. The result is that the previously lightly
loaded port is now overloaded.
This patch adds a rough estimate of the size of the inbound migrations
to a particular port, which can be factored into the migration logic,
avoiding the above problem.
Allowing moving multiple flows in one migration transaction, to
rebalance load more quickly.
Introduce a threshold to avoid migrating flows between ports with very
similar load.
Simplify logic for selecting which flow to migrate. The aim is now to
move flows in such a way that the receiving port is as lightly-loaded
as possible (after receiving the flow), while still migrating enough
flows from the source port to reduce its load. This is essentially how
legacy strategy work as well, but the code is more readable.
To allow visualization of migrations, track the number flow
immigrations in "port_<N>_immigrations". The "port_<N>_migrations"
retains legacy semantics, but is renamed "port_<N>_emigrations".
Expose the number of events currently undergoing processing
(i.e. pending releases) at a particular port.
git.droids-corp.org / dpdk.git / log