Add a function rte_rand_max() which generates an uniformly distributed
pseudo-random number less than a user-specified upper bound.
The commonly used pattern rte_rand() % SOME_VALUE creates biased
results (as in some values in the range are more frequently occurring
than others) if SOME_VALUE is not a power of 2.
Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Replace the use of rte_get_timer_cycles() with getentropy() for
seeding the pseudo-random number generator. getentropy() provides a
more truly random value.
getentropy() requires glibc 2.25 and Linux kernel 3.17. In case
getentropy() is not found at compile time, or the relevant syscall
fails in runtime, the rdseed machine instruction will be used as a
fallback.
rdseed is only available on x86 (Broadwell or later). In case it is
not present, rte_get_timer_cycles() will be used as a second fallback.
On non-Meson builds, getentropy() will not be used.
Suggested-by: Bruce Richardson <bruce.richardson@intel.com>
Suggested-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
This commit replaces rte_rand()'s use of lrand48() with a DPDK-native
combined Linear Feedback Shift Register (LFSR) (also known as
Tausworthe) pseudo-random number generator.
This generator is faster and produces better-quality random numbers
than the linear congruential generator (LCG) of lib's lrand48(). The
implementation, as opposed to lrand48(), is multi-thread safe in
regards to concurrent rte_rand() calls from different lcore threads.
A LCG is still used, but only to seed the five per-lcore LFSR
sequences.
In addition, this patch also addresses the issue of the legacy
implementation only producing 62 bits of pseudo randomness, while the
API requires all 64 bits to be random.
This pseudo-random number generator is not cryptographically secure -
just like lrand48().
Bugzilla ID: 114
Bugzilla ID: 276
Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Add new telemetry mode support for l3fwd-power.
This is a standalone mode, in this mode l3fwd-power
does simple l3fwding along with calculating
empty polls, full polls, and busy percentage for
each forwarding core. The aggregation of these
values of all cores is reported as application
level telemetry to metric library for every 500ms from the
master core.
The busy percentage is calculated by recording the poll_count
and when the count reaches a defined value the total
cycles it took is measured and compared with minimum and maximum
reference cycles and busy rate is set according to either 0% or
50% or 100%.
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Build error:
../lib/librte_telemetry/rte_telemetry.c:558:28:
error: comparison of unsigned expression < 0 is always false
[-Werror,-Wtautological-compare]
Build error not observed in default make build because telemetry library
disabled by default but easier to reproduce via meson.
Fixing by converting unsigned variables to signed.
Fixes: 0fe3a37924d4 ("telemetry: format json response when sending stats")
Fixes: 4080e46c8078 ("telemetry: support global metrics")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Kevin Laatz <kevin.laatz@intel.com>
Array ins_chk in lib/librte_bpf/bpf_validate.c has 255 entries.
So the instruction with opcode == 255 will reading beyond array
boundaries.
For more details please refer to:
https://bugs.dpdk.org/show_bug.cgi?id=283
Fixes: 6e12ec4c4d6d ("bpf: add more checks")
Cc: stable@dpdk.org
Reported-by: Michel Machado <michel@digirati.com.br>
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The vmbus scan code can just skip non-network devices.
More importantly, this fixes the bug where some vmbus devices
don't have all the attributes (like monitor_id) and a single
failure would cause the scan to break the loop.
Fixes: 831dba47bd36 ("bus/vmbus: add Hyper-V virtual bus support")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Let's mark as skipped the tests when they are missing some requirements
like a number of used cores or specific hardware availability, like
compress, crypto or eventdev devices.
Signed-off-by: David Marchand <david.marchand@redhat.com>
Running the tests in parallel has two drawbacks:
- the tests are racing on the hugepages allocations,
- the tests are sharing the cores to run their checks which results in
undeterministic execution time,
This results in random failures.
For better reproducibility in CI, start them all in a serialised way.
Signed-off-by: David Marchand <david.marchand@redhat.com>
Based on Michael initial idea of separating the file-prefix subtest in
the eal flags test.
Let's split the biggest tests into their subparts.
It is then easier to have them fit in the 10s timeout we have configured
in Travis.
We also get a better idea of which part fails in the previously big tests
we had.
Those new subtests are called from the meson testsuite.
The autotest tool is left untouched.
Note: we still have an issue with test_hash_readwrite_lf.c, any help from
the original authors would be appreciated.
Signed-off-by: David Marchand <david.marchand@redhat.com>
The eal flags unit test assumes that a certain number of cores are
available (4 and 8 cores), however this may not always be the case.
Individual developers may run the unit test on their local desktop
which typically have 2 to 4 cores, in said case the test is bound
to fail for lacking 4 or 8 cores.
Additionally, as we push forward introducing CI into DPDK we are limited
to the hardware specification of CI services (e.g. Travis CI) that only
have 2 cores on their servers, in which case the test would fail.
To fix this we check available cores before running a subtest. This
applies to subtests that are dedicated to test that the -l and --lcore
flags work correctly. If not enough cores are available the subtest is
simply skipped, otherwise the subtest is run.
Signed-off-by: Michael Santana <msantana@redhat.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Aaron Conole <aconole@redhat.com>
Setting a coremask was mandatory a long time ago but has been optional
for a while.
The checks on PCI whitelist/blacklist, vdev, memory rank, memory channel,
HPET, memory size and other miscs options have no requirement wrt cores.
Let's remove those coremasks so that we only care about it in the
dedicated checks.
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Aaron Conole <aconole@redhat.com>
The -n option is an optimisation configuration option that defaults to 0.
Such a default value makes the mempool library distributes objects as if
there was 4 memory channels, so -n 4 is the same as the default behavior.
This parameter was mandatory a long time ago, but has been optional for
a while. We check that setting this value works fine in its own test.
Remove it everywhere else.
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Aaron Conole <aconole@redhat.com>
We can have up to RTE_MAX_LCORE in a dpdk application.
Remove the limit on 128 cores and tests that are now always false.
Fixes: b87089b0bb19 ("test/rcu: add API and functional tests")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
This patch rectifies slave_id to point to valid core indexes rather than
core ranks in read-write lock-free concurrency test.
It also replaces a 'for' loop with RTE_LCORE_FOREACH API.
Fixes: c7eb0972e74b ("test/hash: add lock-free r/w concurrency")
Cc: stable@dpdk.org
Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Yipeng Wang <yipeng1.wang@intel.com>
This subtest wants to start rwc_core_cnt[n] reader threads, while the
master core is waiting for them to report.
Fixes: c7eb0972e74b ("test/hash: add lock-free r/w concurrency")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Aaron Conole <aconole@redhat.com>
Acked-by: Yipeng Wang <yipeng1.wang@intel.com>
Same treatment than crypto tests, move the eventdev drivers tests in the
driver list.
While at it:
- eventdev_octeontx_autotest has been renamed as
eventdev_selftest_octeontx,
- eventdev_sw_autotest has been renamed as
eventdev_selftest_sw,
Fixes: 50fb749a3972 ("event/octeontx: move test to driver")
Fixes: 85fb515b7318 ("event/sw: move test to driver")
Fixes: 123d67c73b06 ("test/event: register selftests")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Aaron Conole <aconole@redhat.com>
For consistency, put all specific crypto driver tests in the dedicated
list (in alphabetic order).
While at it:
- remove dead reference to cryptodev_sw_mrvl_autotest (renamed as
cryptodev_sw_mvsam_autotest),
- call the crypto scheduler test only when built,
Fixes: 9eabcb682493 ("test: update autotest list")
Fixes: 3d20ffe6ddb1 ("test: reorder test cases in meson")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Aaron Conole <aconole@redhat.com>
Take into account IPv6 fragment extension header when
calculating data size for each fragment.
Fixes: 7a838c8798a9 ("ip_frag: fix IPv6 when MTU sizes not aligned to 8 bytes")
Fixes: 0aa31d7a5929 ("ip_frag: add IPv6 fragmentation support")
Cc: stable@dpdk.org
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
clang raise 'pointer-sign' warnings in __atomic_compare_exchange
when passing 'uint64_t *' to parameter of type 'int64_t *' converts
between pointers to integer types with different sign.
Fixes: 7e6e609939a8 ("stack: add C11 atomic implementation")
Cc: stable@dpdk.org
Suggested-by: Gage Eads <gage.eads@intel.com>
Signed-off-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Acked-by: Gage Eads <gage.eads@intel.com>
Currently kernel modules are installed into /usr/src instead of
/lib/modules when meson build system is used. This patch fixes that.
Cc: stable@dpdk.org
Signed-off-by: Igor Ryzhov <iryzhov@nfware.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Internal changes in the freebsd kernel have meant that additional includes
are now necessary to build the kernel modules for DPDK. Tested with latest
bsd HEAD revision.
Bugzilla ID: 282
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
When adding an alarm, if an error happen when registering
the common alarm callback, it is not considered as a major failure.
The alarm is then inserted in the list.
However it was returning an error code after inserting the alarm.
The error code is not set anymore to be consistent with the behaviour.
Fixes: af75078fece3 ("first public release")
Cc: stable@dpdk.org
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
The Intel compiler is pickier about casts and generates:
otx2_mempool_ops.c(344):
error #191: type qualifier is meaningless on cast type
int64_t * const addr = (int64_t * const)
^
This is because of the nature of const.
In this example, the expression is being cast into a pointer
that can not be modified. This is meaningless because the
expression is already a lvalue.
See https://en.wikipedia.org/wiki/Const_(computer_programming)
Fixes: d7a0da3c0043 ("mempool/octeontx2: add fast path mempool ops")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
This patch changes some void functions to return a value,
so that the init sequence may tear down orderly
instead of calling panic.
Signed-off-by: Arnon Warshavsky <arnon@qwilt.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
When a component uses either XOPEN_SOURCE or POSIX_C_SOURCE macro
explicitly in its build recipe, it restricts visibility of a non POSIX
features subset, such as IANA protocol numbers (IPPROTO_* macros).
Non standard features are enabled by default for DPDK both for Linux
thanks to _GNU_SOURCE and for FreeBSD thanks to __BSD_VISIBLE. However
using XOPEN_SOURCE or POSIX_(C_)SOURCE in a component causes
__BSD_VISIBLE to be defined to 0 for FreeBSD, causing different feature
sets visibility for Linux and FreeBSD. It restricts from using IPPROTO
macros in public headers, such as rte_ip.h, despite the fact they are
already widely used in sources.
Add __BSD_VISIBLE macro specified unconditionally for FreeBSD targets
which enforces feature sets visibility unification between Linux and
FreeBSD.
Add single -D_GNU_SOURCE to config/meson.build as a project argument
instead of adding separate directive for each project subtree.
This patch solves the problem of build breaks for [1] on FreeBSD [2]
following the discussion [3].
[1] https://mails.dpdk.org/archives/dev/2019-May/131885.html
[2] http://mails.dpdk.org/archives/test-report/2019-May/082263.html
[3] https://mails.dpdk.org/archives/dev/2019-May/132110.html
Signed-off-by: Marcin Smoczynski <marcinx.smoczynski@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Claim the maintainership of the sub tree dpdk-next-net-mrvl,
which covers all the Marvell network PMDs.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
The maximum number of mempools per application needs to be configured
on HW during mempool driver initialization. HW can support up to 1M
mempools, Since each mempool costs set of HW resources, the max_pools
devargs parameter is being introduced to configure the number of
mempools required for the application.
For example:
-w 0002:02:00.0,max_pools=512
With the above configuration, the driver will set up only 512 mempools
for the given application to save HW resources.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Harman Kalra <hkalra@marvell.com>
This patch adds an optimized arm64 instruction based routine to leverage
CPU pipeline characteristics of octeontx2. The theme is to fill the
pipeline with CASP operations as much HW can do so that HW can do alloc()
HW ops in full throttle.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
The DPDK mempool free operation frees HW AURA
and POOL reserved in alloc operation. In addition to that it free all
the memory resources allocated in mempool alloc operations.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
The DPDK mempool allocation reserves a single HW AURA
and POOL in 1:1 map mode. Upon reservation, SW programs the slow path
operations such as allocate stack memory for DMA and
bunch HW configurations to respective HW blocks.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Add a helper function to dump aura and pool context for NPA debugging.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Vivek Sharma <viveksharma@marvell.com>
Register and implement NPA IRQ handler for RAS and all type of
error interrupts to get the fatal errors from HW.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Harman Kalra <hkalra@marvell.com>
Implement the low-level NPA HW operations such as
alloc, free memory, etc.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
NPA object needs to initialize memory for queue interrupts context,
pool resource management, etc. This patch adds support for initializing
and finalizing the NPA object.
This patch also updates the otx2_npa_lf definition to meet the init/fini
requirements.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
Add the make and meson based build infrastructure along
with the mempool(NPA) device probe.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Upon receiving FLR request from VF, It is PF responsibly
forward to AF and enable FLR for VFs.
This patch adds support for VF FLR support in PF.
This patch also add otx2_dev_active_vfs() API to find
the number of active VF for given PF.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Signed-off-by: Harman Kalra <hkalra@marvell.com>
The events like PHY link status change by AF or PHY configuration
change by PF would call for the uplink message.
The AF initiated uplink would land it in PF and PF would further
forward to VF(if it is intended for that specific VF)
The PF initiated uplink would be distributed to all active VFs.
This patch adds support for the same.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Signed-off-by: Harman Kalra <hkalra@marvell.com>
PF has additional responsibility being server for VF messages
and forward to AF and once AF process it then forward
the response back to VF.
otx2_vf_pf_mbox_irq() will process the VF mailbox request and
af_pf_wait_msg() will until getting a response back from AF.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Signed-off-by: Krzysztof Kanas <kkanas@marvell.com>
