dpdk.git
4 years agotest/mbuf: add unit test on mbuf flag names
Pallantla Poornima [Fri, 30 Aug 2019 14:13:15 +0000 (15:13 +0100)]
test/mbuf: add unit test on mbuf flag names

Added UT for the below functions:
- rte_get_rx_ol_flag_list,
- rte_get_tx_ol_flag_list,
- rte_get_rx_ol_flag_name,
- rte_get_tx_ol_flag_name

Signed-off-by: Pallantla Poornima <pallantlax.poornima@intel.com>
Reviewed-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
4 years agotest/mbuf: add unit test cases
Lavanya Govindarajan [Fri, 30 Aug 2019 14:13:14 +0000 (15:13 +0100)]
test/mbuf: add unit test cases

Added new unit test cases to cover the below functions:
- rte_validate_tx_offload,
- rte_pktmbuf_alloc_bulk,
- rte_pktmbuf_read,
- rte_pktmbuf_ext_shinfo_init_helper,
- rte_pktmbuf_attach_extbuf,
- rte_mbuf_ext_refcnt_read,
- rte_mbuf_ext_refcnt_update,
- rte_mbuf_ext_refcnt_set,
- rte_pktmbuf_detach_extbuf

Signed-off-by: Lavanya Govindarajan <lavanyax.govindarajan@intel.com>
Reviewed-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
4 years agotest/mbuf: add case for bulk alloc/free functions
Morten Brørup [Wed, 23 Oct 2019 10:11:11 +0000 (10:11 +0000)]
test/mbuf: add case for bulk alloc/free functions

Add unit test for functions for allocating and freeing a bulk of mbufs.

Signed-off-by: Morten Brørup <mb@smartsharesystems.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
4 years agombuf: add bulk free function
Morten Brørup [Wed, 23 Oct 2019 10:11:10 +0000 (10:11 +0000)]
mbuf: add bulk free function

Add function for freeing a bulk of mbufs.

Signed-off-by: Morten Brørup <mb@smartsharesystems.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
4 years agoconfig: set KNI preempt default with meson
Igor Ryzhov [Mon, 16 Sep 2019 10:08:28 +0000 (13:08 +0300)]
config: set KNI preempt default with meson

Same behavior as in make build system.

Signed-off-by: Igor Ryzhov <iryzhov@nfware.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
4 years agodrivers: remove duplicated compiler flags
Bruce Richardson [Mon, 7 Oct 2019 14:30:13 +0000 (15:30 +0100)]
drivers: remove duplicated compiler flags

Now that -Wextra, and other warning flags are standard in the build, remove
any duplication in driver build specifications.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
4 years agobuild: enable extra warnings with meson
Bruce Richardson [Mon, 7 Oct 2019 14:30:12 +0000 (15:30 +0100)]
build: enable extra warnings with meson

While meson always adds -Wall flag to C compiles, the make build adds extra
warning flags that are not present in the meson build. This addresses that
shortcoming by adding additional warning flags to our builds. The one
omission is the -Wcast-align flag, which though present in make gcc builds,
gives a lot of warnings/errors when used with clang.

The removed warning "-Wunused-parameter" is covered by the "-Wextra"
parameter so is unnecessary.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
4 years agoraw/ifpga: remove unneeded compiler flags
Bruce Richardson [Mon, 7 Oct 2019 14:30:11 +0000 (15:30 +0100)]
raw/ifpga: remove unneeded compiler flags

The warning disable flags for the base driver code were copy-pasted from
another source, and are actually unnecessary for this driver. Therefore
remove them.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
4 years agobuild: allow stricter fallthrough warnings
Bruce Richardson [Mon, 7 Oct 2019 14:30:10 +0000 (15:30 +0100)]
build: allow stricter fallthrough warnings

DPDK currently compiles with implicit-fallthrough=2 warning level. With gcc
-Wextra flag, the default level is 3, so some minor changes are needed to
support this in DPDK.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
4 years agobuild: support disabling drivers with meson
Bruce Richardson [Wed, 25 Sep 2019 14:55:31 +0000 (15:55 +0100)]
build: support disabling drivers with meson

Add support for a new build option to turn off certain drivers. Any other
drivers which depend on the one being disabled will also be disabled with a
suitable debug message.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
Acked-by: Maxime Coquelin <maxime.coquelin@redhat.com>
4 years agobuild: process dependencies before main build check
Bruce Richardson [Wed, 25 Sep 2019 14:55:30 +0000 (15:55 +0100)]
build: process dependencies before main build check

If we want to add support for turning off components because of missing
dependencies, then we need to check for those dependencies before we
make a determination as to whether a component should be built or not,
assuming that the component says it should be built.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
Acked-by: Maxime Coquelin <maxime.coquelin@redhat.com>
4 years agobuild: align variable names between drivers and libs
Bruce Richardson [Wed, 25 Sep 2019 14:55:29 +0000 (15:55 +0100)]
build: align variable names between drivers and libs

The variable names in the library and drivers meson.build files are slighty
different with "static_deps" in one and "static_objs" in the other. Rename
to use "static_deps" in both for consistency.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
Acked-by: Maxime Coquelin <maxime.coquelin@redhat.com>
4 years agobuild: print out dependency names for clarity
Bruce Richardson [Thu, 19 Sep 2019 15:07:55 +0000 (16:07 +0100)]
build: print out dependency names for clarity

To help developers to get the correct dependency name e.g. when creating a
new example that depends on a specific component, print out the dependency
name for each lib/driver as it is processed.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
4 years agoexamples/vm_power: fix build
Ferruh Yigit [Mon, 14 Oct 2019 10:03:50 +0000 (11:03 +0100)]
examples/vm_power: fix build

Fixes: 70febdcfd60f ("examples: check status of getting MAC address")

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Tested-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
4 years agoevent/dpaa2: support Tx adapter
Nipun Gupta [Thu, 17 Oct 2019 08:29:04 +0000 (13:59 +0530)]
event/dpaa2: support Tx adapter

This patch adds the support of Tx adapter for DPAA2 platform

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agoevent/dpaa2: set priority as per DPCON device
Nipun Gupta [Thu, 17 Oct 2019 08:29:03 +0000 (13:59 +0530)]
event/dpaa2: set priority as per DPCON device

This patch sets the priority of the dpcon dev, such that it is
within the supported range of dpcon

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agoevent/dpaa: support Tx adapter
Nipun Gupta [Fri, 11 Oct 2019 13:47:57 +0000 (19:17 +0530)]
event/dpaa: support Tx adapter

This patch adds the support of Tx adapter for DPAA1 platform

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agoevent/dpaa: fix number of supported atomic flows
Nipun Gupta [Fri, 11 Oct 2019 13:47:56 +0000 (19:17 +0530)]
event/dpaa: fix number of supported atomic flows

The number of atomic flows supported was not returned correctly for
DPAA driver. This patch fixes the same.

Fixes: b08dc6430abd ("event/dpaa: add queue config get/set")
Cc: stable@dpdk.org
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agoeventdev: add Tx flag for packets with same destination
Nipun Gupta [Fri, 11 Oct 2019 13:03:06 +0000 (18:33 +0530)]
eventdev: add Tx flag for packets with same destination

This patch introduces a `flag` in the Eth TX adapter enqueue API.
Some drivers may support burst functionality only with the packets
having same destination device and queue.

The flag `RTE_EVENT_ETH_TX_ADAPTER_ENQUEUE_SAME_DEST` can be used
to indicate this so the underlying driver, for drivers to utilize
burst functionality appropriately.

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agorcu: fix reference to offline function
David Marchand [Fri, 18 Oct 2019 09:57:53 +0000 (11:57 +0200)]
rcu: fix reference to offline function

Fixes: 64994b56cfd7 ("rcu: add RCU library supporting QSBR mechanism")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
4 years agorcu: update QS only when there are updates from writer
Honnappa Nagarahalli [Tue, 8 Oct 2019 21:12:20 +0000 (16:12 -0500)]
rcu: update QS only when there are updates from writer

When the writer is checking the quiescent state status, it is not
deleting any entries in the data structure. This means, the readers
do not need to update their quiescent state during that period.
Readers update the quiescent state only when there are updates
available from the writer.

Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
4 years agorcu: add least acknowledged token optimization
Honnappa Nagarahalli [Tue, 8 Oct 2019 21:12:19 +0000 (16:12 -0500)]
rcu: add least acknowledged token optimization

When the rte_rcu_qsbr_check API is called, it is possible to
calculate the least valued token acknowledged by all the readers.
When the API is called next time, the readers' token counters do
not need to be scanned if the value of the token being queried is
less than the last least token acknowledged. This avoids the
cache line bounces between readers and writer.

Fixes: 64994b56cfd7 ("rcu: add RCU library supporting QSBR mechanism")
Cc: stable@dpdk.org
Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
4 years agotest/rcu: enhance log nomenclature
Honnappa Nagarahalli [Tue, 8 Oct 2019 21:12:18 +0000 (16:12 -0500)]
test/rcu: enhance log nomenclature

Use 'quiescent state updates' instead of just 'updates'.

Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
4 years agotest/rcu: fix memory size integer truncate
Honnappa Nagarahalli [Tue, 8 Oct 2019 21:12:17 +0000 (16:12 -0500)]
test/rcu: fix memory size integer truncate

Variables used to store the return value of rte_rcu_qsbr_get_memsize
in variables of type 'int' or 'uint32_t'.
While the former variables are a problem, the latter have been aligned
for consistency.
The variables are of type 'size_t' now.

Fixes: b87089b0bb19 ("test/rcu: add API and functional tests")
Cc: stable@dpdk.org
Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
4 years agodoc: describe token and resource retention in RCU guide
Honnappa Nagarahalli [Tue, 8 Oct 2019 21:12:16 +0000 (16:12 -0500)]
doc: describe token and resource retention in RCU guide

After calling rte_rcu_qsbr_start API, the token and the deleted
resource need to be stored for subsequent query/free.

Fixes: 64994b56cfd7 ("rcu: add RCU library supporting QSBR mechanism")
Cc: stable@dpdk.org
Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
4 years agodoc: remove wrong limitation in RCU guide
Honnappa Nagarahalli [Tue, 8 Oct 2019 21:12:15 +0000 (16:12 -0500)]
doc: remove wrong limitation in RCU guide

There is no limitation of 1024 reader threads.

Fixes: 64994b56cfd7 ("rcu: add RCU library supporting QSBR mechanism")
Cc: stable@dpdk.org
Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
4 years agodoc: proofread RCU guide
Honnappa Nagarahalli [Tue, 8 Oct 2019 21:12:14 +0000 (16:12 -0500)]
doc: proofread RCU guide

Fixes: 64994b56cfd7 ("rcu: add RCU library supporting QSBR mechanism")
Cc: stable@dpdk.org
Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
4 years agoexamples/ntb: fix build
David Marchand [Mon, 21 Oct 2019 15:44:54 +0000 (17:44 +0200)]
examples/ntb: fix build

lcore indexes can't be represented with uint8_t.
Seen on armv8 target:

examples/ntb/ntb_fwd.c: In function ‘cmd_quit_parsed’:
arm64-armv8a-linux-gcc/include/rte_lcore.h:296:8:
error: comparison is always true due to limited range of data type
[-Werror=type-limits]
       i<RTE_MAX_LCORE;      \
        ^
examples/ntb/ntb_fwd.c:164:2: note: in expansion of
macro ‘RTE_LCORE_FOREACH_SLAVE’
  RTE_LCORE_FOREACH_SLAVE(lcore_id) {
  ^~~~~~~~~~~~~~~~~~~~~~~

Fixes: 5194299d6ef5 ("examples/ntb: support more functions")

Signed-off-by: David Marchand <david.marchand@redhat.com>
4 years agodevtools: fix cleanup of checkpatch temporary file
David Marchand [Tue, 13 Aug 2019 06:38:22 +0000 (08:38 +0200)]
devtools: fix cleanup of checkpatch temporary file

The regexp part of the cleanup routine was not updated accordingly when
a common prefix for temp files was introduced.
Set the trap handler only when required.

Fixes: ff37ca5d3773 ("devtools: use a common prefix for temporary files")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
4 years agoclean bare metal support traces
David Marchand [Mon, 12 Aug 2019 07:00:54 +0000 (09:00 +0200)]
clean bare metal support traces

Bare metal support has been gone for quite some time but we still had
some checks on system includes.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
4 years agoexamples/ntb: support more functions
Xiaoyun Li [Thu, 26 Sep 2019 03:20:22 +0000 (11:20 +0800)]
examples/ntb: support more functions

Support to transmit files between two systems.
Support iofwd between one ethdev and NTB device.
Support rxonly and txonly for NTB device.
Support to set forwarding mode as file-trans, txonly,
rxonly or iofwd.
Support to show/clear port stats and throughput.

Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
4 years agoraw/ntb: add enqueue and dequeue functions
Xiaoyun Li [Thu, 26 Sep 2019 03:20:21 +0000 (11:20 +0800)]
raw/ntb: add enqueue and dequeue functions

Introduce enqueue and dequeue functions to support packet based
processing. And enable write-combining for ntb driver since it
can improve the performance a lot.

Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
4 years agoraw/ntb: support xstats
Xiaoyun Li [Thu, 26 Sep 2019 03:20:20 +0000 (11:20 +0800)]
raw/ntb: support xstats

Add xstats support for NTB rawdev.
Support tx-packets, tx-bytes, tx-errors and
rx-packets, rx-bytes, rx-missed.

Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
4 years agoraw/ntb: setup queues
Xiaoyun Li [Thu, 26 Sep 2019 03:20:19 +0000 (11:20 +0800)]
raw/ntb: setup queues

Setup and init NTB txq and rxq. And negotiate queue information
with the peer. If queue size and number of queues are not
consistent on both sides, return error.

Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
4 years agostack: enable lock-free implementation for aarch64
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>
4 years agotest/atomic: add 128-bit atomic compare exchange test
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>
4 years agoeal/arm64: add 128-bit atomic compare exchange
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].

1: https://git.dpdk.org/dpdk/commit/?id=5b40ec6b9662

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>
4 years agoeal: calibrate TSC only in primary process
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>
4 years agorcu: fix spurious thread unregister
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>
4 years agoservice: avoid false sharing on core state
Nikhil Rao [Mon, 16 Sep 2019 10:01:02 +0000 (15:31 +0530)]
service: avoid false sharing on core state

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>
4 years agotimer: remove useless check on x86 TSC reliability
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>
4 years agoeal/ppc: add SPDX license tag
Hemant Agrawal [Fri, 27 Sep 2019 09:04:28 +0000 (14:34 +0530)]
eal/ppc: add SPDX license tag

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: David Christensen <drc@linux.vnet.ibm.com>
4 years agoeal/ppc: fix 64-bit atomic exchange operation
David Christensen [Tue, 15 Oct 2019 21:16:14 +0000 (14:16 -0700)]
eal/ppc: fix 64-bit atomic exchange operation

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>
4 years agombuf: add a copy routine
Stephen Hemminger [Tue, 8 Oct 2019 16:33:49 +0000 (09:33 -0700)]
mbuf: add a copy routine

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>
4 years agombuf: deinline clone function
Stephen Hemminger [Tue, 8 Oct 2019 16:33:48 +0000 (09:33 -0700)]
mbuf: deinline clone function

Cloning mbufs requires allocations and iteration
and therefore should not be an inline.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
4 years agombuf: deinline linearize function
Stephen Hemminger [Tue, 8 Oct 2019 16:33:47 +0000 (09:33 -0700)]
mbuf: deinline linearize function

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>
4 years agotest/mbuf: fix forged mbuf in clone test
Stephen Hemminger [Tue, 8 Oct 2019 16:33:46 +0000 (09:33 -0700)]
test/mbuf: fix forged mbuf in clone test

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.

Fixes: 4ccd2bb3a9e2 ("app/test: enhance mbuf refcnt check")
Fixes: af75078fece3 ("first public release")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
4 years agomempool: clarify default populate function
Olivier Matz [Tue, 8 Oct 2019 09:34:06 +0000 (11:34 +0200)]
mempool: clarify default populate function

No functional change.
Clarify the populate function to make future changes easier to
understand.

Rename the variables:
- to avoid negation in the name
- to have more understandable names

Remove useless variable (no_pageshift is equivalent to pg_sz == 0).

Remove duplicate affectation of "external" variable.

Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
4 years agokni: support allmulticast mode set
Xiaolong Ye [Mon, 12 Aug 2019 03:06:30 +0000 (11:06 +0800)]
kni: support allmulticast mode set

This patch adds support to allow users enable/disable allmulticast mode for
kni interface.

This requirement comes from bugzilla 312, more details can refer to:
https://bugs.dpdk.org/show_bug.cgi?id=312

Bugzilla ID: 312

Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
4 years agoservice: use log for error messages
Stephen Hemminger [Wed, 21 Aug 2019 09:12:52 +0000 (10:12 +0100)]
service: use log for error messages

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>
4 years agoeal: fix mapping leak in secondary process
Arnon Warshavsky [Mon, 19 Aug 2019 13:57:44 +0000 (16:57 +0300)]
eal: fix mapping leak in secondary process

Have rte_eal_config_reattach clean up the mapped address which is a valid
address but not the one intended.

Coverity issue: 343439
Fixes: 4e8854ae89fa ("eal: do not panic on shared memory init")
Fixes: b149a7064261 ("eal/freebsd: add config reattach in secondary process")
Cc: stable@dpdk.org
Signed-off-by: Arnon Warshavsky <arnon@qwilt.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
4 years agovfio: fix leak with multiprocess
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().

Fixes: 83a73c5fef66 ("vfio: use generic multi-process channel")
Cc: stable@dpdk.org
Reported-by: Anatoly Burakov <anatoly.burakov@intel.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
4 years agobuild: avoid overlinking
Christian Ehrhardt [Thu, 29 Aug 2019 15:30:03 +0000 (17:30 +0200)]
build: avoid overlinking

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>
4 years agobpf: hide internal program argument type
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>
4 years agobpf/arm: add branch operation
Jerin Jacob [Tue, 3 Sep 2019 10:59:38 +0000 (16:29 +0530)]
bpf/arm: add branch operation

Add branch and call operations.

jump_offset_* APIs used for finding the relative offset
to jump w.r.t current eBPF program PC.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
4 years agobpf/arm: add atomic-exchange-and-add operation
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>
4 years agobpf/arm: add load and store operations
Jerin Jacob [Tue, 3 Sep 2019 10:59:36 +0000 (16:29 +0530)]
bpf/arm: add load and store operations

Add load and store operations.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
4 years agobpf/arm: add byte swap operations
Jerin Jacob [Tue, 3 Sep 2019 10:59:35 +0000 (16:29 +0530)]
bpf/arm: add byte swap operations

add le16, le32, le64, be16, be32 and be64 operations.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
4 years agobpf/arm: add logical operations
Jerin Jacob [Tue, 3 Sep 2019 10:59:34 +0000 (16:29 +0530)]
bpf/arm: add logical operations

Add OR, AND, NEG, XOR, shift operations for immediate
and source register variants.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
4 years agobpf/arm: add basic arithmetic operations
Jerin Jacob [Tue, 3 Sep 2019 10:59:33 +0000 (16:29 +0530)]
bpf/arm: add basic arithmetic operations

Add mov, add, sub, mul, div and mod arithmetic
operations for immediate and source register variants.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
4 years agobpf/arm: add prologue and epilogue
Jerin Jacob [Tue, 3 Sep 2019 10:59:32 +0000 (16:29 +0530)]
bpf/arm: add prologue and epilogue

Add prologue and epilogue as per arm64 procedure call standard.

As an optimization the generated instructions are
the function of whether eBPF program has stack and/or
CALL class.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
4 years agobpf/arm: add build infrastructure
Jerin Jacob [Tue, 3 Sep 2019 10:59:31 +0000 (16:29 +0530)]
bpf/arm: add build infrastructure

Add build infrastructure and documentation
update for arm64 JIT support.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
4 years agoapp/eventdev: add options for mbuf and packet sizes
Pavan Nikhilesh [Mon, 30 Sep 2019 16:48:41 +0000 (22:18 +0530)]
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>
4 years agotest/event: enable dpaa2 self test
Hemant Agrawal [Mon, 30 Sep 2019 08:32:15 +0000 (14:02 +0530)]
test/event: enable dpaa2 self test

This patch add the support to include dpaa2 event test
from the test framework.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agoevent/dpaa2: add self test
Hemant Agrawal [Mon, 30 Sep 2019 08:32:14 +0000 (14:02 +0530)]
event/dpaa2: add self test

This patch add support for testing dpaa2 eventdev self test
for basic sanity for parallel and atomic queues.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agoevent/dpaa2: add retry break in packet enqueue
Nipun Gupta [Mon, 30 Sep 2019 08:32:13 +0000 (14:02 +0530)]
event/dpaa2: add retry break in packet enqueue

The patch adds the break in the TX function, if it is failing
to send the packets out. Previously the system was trying
infinitely to send packet out.

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
4 years agoevent/dpaa2: support destroy
Hemant Agrawal [Mon, 30 Sep 2019 08:32:12 +0000 (14:02 +0530)]
event/dpaa2: support destroy

This patch add support to destroy the event device

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agoevent/dpaa2: remove conditional compilation
Hemant Agrawal [Mon, 30 Sep 2019 08:32:11 +0000 (14:02 +0530)]
event/dpaa2: remove conditional compilation

This patch removes the conditional compilation for
cryptodev event support from RTE_LIBRTE_SECURITY flag.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agoevent/dpaa2: fix default queue configuration
Hemant Agrawal [Mon, 30 Sep 2019 08:32:10 +0000 (14:02 +0530)]
event/dpaa2: fix default queue configuration

Test vector expect only one type of scheduling as default.
The old code is provide support scheduling types instead of default.

Fixes: 13370a3877a5 ("eventdev: fix inconsistency in queue config")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
4 years agotest/event_crypto: check adapter config failure
Hemant Agrawal [Fri, 27 Sep 2019 08:16:06 +0000 (13:46 +0530)]
test/event_crypto: check adapter config failure

Return error when the adapter creation fails.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
4 years agotest/event_crypto: change cipher algo
Hemant Agrawal [Fri, 27 Sep 2019 08:16:05 +0000 (13:46 +0530)]
test/event_crypto: change cipher algo

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.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
4 years agotest/event_crypto: fix mempool name
Hemant Agrawal [Fri, 27 Sep 2019 08:16:04 +0000 (13:46 +0530)]
test/event_crypto: fix mempool name

The longer mempool name size is causing error in
rte_mempool_create_empty for dpaa1

ret = snprintf(mz_name, sizeof(mz_name), RTE_MEMPOOL_MZ_FORMAT, name);
This patch reduce the size of mempool name string

Fixes: 24054e3640a2 ("test/crypto: use separate session mempools")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
4 years agoevent/octeontx2: fix Rx adapter capabilities
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.

Fixes: 37720fc1fba8 ("event/octeontx2: add Rx adapter")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
4 years agoevent/sw: fix xstats reset value
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>
4 years agotest/event_crypto: no service core when HW support available
Hemant Agrawal [Fri, 6 Sep 2019 10:44:34 +0000 (16:14 +0530)]
test/event_crypto: no service core when HW support available

If HW support is available, service core shall not come
into play by default. This shall be for both FWD/NEW modes.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
4 years agotest/event_crypto: check session init failure
Hemant Agrawal [Fri, 6 Sep 2019 10:42:28 +0000 (16:12 +0530)]
test/event_crypto: check session init failure

Mismatch in algo or sec capability can cause session to fail.
This patch handle it and return error timely.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
4 years agonet/mlx5: fix direct call to rdma-core library
Viacheslav Ovsiienko [Mon, 7 Oct 2019 13:58:22 +0000 (13:58 +0000)]
net/mlx5: fix direct call to rdma-core library

The routine mlx5dv_query_devx_port() was called directly
instead of using the mlx5 glue thunk.

Fixes: d5c06b1b10ae ("net/mlx5: query vport index match mode and parameters")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
4 years agonet/mlx5: fix device scan within switch domain
Viacheslav Ovsiienko [Mon, 7 Oct 2019 13:56:19 +0000 (13:56 +0000)]
net/mlx5: fix device scan within switch domain

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.

Fixes: f7e95215ac7c ("net/mlx5: extend switch domain searching range")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
4 years agonet/mlx5: adjust inline setting for large Tx queue sizes
Viacheslav Ovsiienko [Tue, 1 Oct 2019 06:53:37 +0000 (06:53 +0000)]
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.

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
4 years agonet/cxgbe: fix races on flow API operations
Rahul Lakkireddy [Fri, 4 Oct 2019 14:24:17 +0000 (19:54 +0530)]
net/cxgbe: fix races on flow API operations

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.

Fixes: ee61f5113b17 ("net/cxgbe: parse and validate flows")
Fixes: 9eb2c9a48072 ("net/cxgbe: implement flow create operation")
Fixes: da23bc9d33f4 ("net/cxgbe: implement flow destroy operation")
Fixes: 8d3c12e19368 ("net/cxgbe: implement flow query operation")
Fixes: 86910379d335 ("net/cxgbe: implement flow flush operation")
Cc: stable@dpdk.org
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
4 years agonet/bnxt: improve CPR handling in vector path
Lance Richardson [Fri, 4 Oct 2019 03:49:03 +0000 (20:49 -0700)]
net/bnxt: improve CPR handling in vector path

Reduce overhead of CPR descriptor validity checking in vector
receive and transmit functions.

Preserve raw cpr consumer index in vector transmit completion
function.

Remove an unneeded prefetch (per benchmarking) from vector
transmit completion function.

Fixes: bc4a000f2f53 ("net/bnxt: implement SSE vector mode")
Cc: stable@dpdk.org
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
4 years agonet/bnxt: advertise scatter Rx offload capability
Lance Richardson [Fri, 4 Oct 2019 03:49:02 +0000 (20:49 -0700)]
net/bnxt: advertise scatter Rx offload capability

Scattered receive is supported but not included in receive offload
capabilities. Fix by adding it and including in scattered receive
calculation.

Fixes: 9c1507d96ab8 ("net/bnxt: switch to the new offload API")
Cc: stable@dpdk.org
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
4 years agonet/bnxt: fix default Rx queue for Thor
Lance Richardson [Fri, 4 Oct 2019 03:49:01 +0000 (20:49 -0700)]
net/bnxt: fix default Rx queue for Thor

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.

Fixes: f8168ca0e690 ("net/bnxt: support thor controller")
Cc: stable@dpdk.org
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
4 years agonet/bnxt: fix stats context calculation
Lance Richardson [Fri, 4 Oct 2019 03:49:00 +0000 (20:49 -0700)]
net/bnxt: fix stats context calculation

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.

Fixes: bd0a14c99f65 ("net/bnxt: use dedicated CPR for async events")
Cc: stable@dpdk.org
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
4 years agonet/bnxt: use common NQ ring
Lance Richardson [Fri, 4 Oct 2019 03:48:59 +0000 (20:48 -0700)]
net/bnxt: use common NQ ring

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.

Fixes: f8168ca0e690 ("net/bnxt: support thor controller")
Cc: stable@dpdk.org
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
4 years agonet/bnxt: support CoS classification
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.

Signed-off-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
Reviewed-by: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
4 years agonet/bnxt: support LRO on Thor adapters
Lance Richardson [Fri, 4 Oct 2019 03:48:57 +0000 (20:48 -0700)]
net/bnxt: support LRO on Thor adapters

Add support for LRO for adapters based on Thor (BCM57508).

Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
4 years agonet/bnxt: fix ring alignment for Thor-based adapters
Lance Richardson [Fri, 4 Oct 2019 03:48:56 +0000 (20:48 -0700)]
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.

Fixes: f8168ca0e690 ("net/bnxt: support thor controller")
Cc: stable@dpdk.org
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
4 years agonet/bnxt: increase TQM entry allocation
Lance Richardson [Fri, 4 Oct 2019 03:48:55 +0000 (20:48 -0700)]
net/bnxt: increase TQM entry allocation

The current TQM backing store size isn't sufficient to allow 512
transmit rings. Fix by correcting TQM SP queue size calculation.

Fixes: f8168ca0e690 ("net/bnxt: support thor controller")
Cc: stable@dpdk.org
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
4 years agonet/atlantic: add FW mailbox guard mutex
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>
4 years agonet/atlantic: fix reported flow control mode
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>
4 years agonet/atlantic: exclude MACsec counters from xstats
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>
4 years agonet/octeontx2: fix CQE ring prefetch on wrap around
Kommula Shiva Shankar [Fri, 13 Sep 2019 15:31:12 +0000 (21:01 +0530)]
net/octeontx2: fix CQE ring prefetch on wrap around

When computing the head of CQE ring of prefetch, use qmask to point to
the correct head index on wrap around case.

Fixes: cc4d7693f2d9 ("net/octeontx2: support Rx")
Cc: stable@dpdk.org
Signed-off-by: Kommula Shiva Shankar <kshankar@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
4 years agonet/bnx2x: update to latest FW 7.13.11
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.

Update the PMD version to 1.1.0.1.

Signed-off-by: Rasesh Mody <rmody@marvell.com>
4 years agonet/bnx2x: update HSI
Rasesh Mody [Wed, 2 Oct 2019 19:14:55 +0000 (12:14 -0700)]
net/bnx2x: update HSI

Update hardware software common base driver code in preparation to
update the firmware to version 7.13.11.

Signed-off-by: Rasesh Mody <rmody@marvell.com>
4 years agonet/bnx2x: update and reorganize HW registers
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.

Signed-off-by: Rasesh Mody <rmody@marvell.com>
4 years agonet/qede: limit Rx ring index read for debug
David Marchand [Fri, 27 Sep 2019 11:28:49 +0000 (13:28 +0200)]
net/qede: limit Rx ring index read for debug

Caught by clang, this idx value is only used for a debug message when
the mbufs allocation fails.
No need to use idx as a temporary storage.

Fixes: 8f2312474529 ("net/qede: fix performance bottleneck in Rx path")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Rasesh Mody <rmody@marvell.com>
4 years agonet/octeontx2: support GRE TSO offload
Nithin Dabilpuram [Wed, 25 Sep 2019 04:54:07 +0000 (10:24 +0530)]
net/octeontx2: support GRE TSO offload

Extends existing TSO support to GRE tunnel on the
same SoC revisions.

Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
4 years agonet/octeontx2: support TSO offload
Nithin Dabilpuram [Wed, 25 Sep 2019 04:36:29 +0000 (10:06 +0530)]
net/octeontx2: support TSO offload

Add support to below TCP segmentation offloads for
96XX A1 onwards and 95xx B0 onwards.
- TCPv4, TCPv6
- VXLAN[v4 | v6][v4 | v6]
- GENEVE[v4 | v6][v4 | v6]

This patch also modifies a fastpath function to be forced
inline due to performance reasons for multi-seg mode.

Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
4 years agonet/octeontx2: extract NVGRE as ltype
Kiran Kumar K [Wed, 11 Sep 2019 09:19:25 +0000 (14:49 +0530)]
net/octeontx2: extract NVGRE as ltype

Adding change to sync RTE Flow with KPU profile to extract
NVGRE as ltype.

Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>