Bruce Richardson [Mon, 25 Nov 2019 15:36:20 +0000 (15:36 +0000)]
examples/ioat: handle error when querying number of stats
To get the amount of memory needed for stats, we call the xstats_get_names
function with a NULL parameter, which can return -1 on error. This negative
value was not previously handled correctly, so we adjust things to quit the
stats printing routine if this basic call fails.
Bruce Richardson [Mon, 25 Nov 2019 15:23:03 +0000 (15:23 +0000)]
examples/ioat: handle failure case for ioat dequeue
On a failure with the ioat dequeue, -1 is returned, which was not properly
handled. This could lead to out-of-bounds reads on a later loop which
assumed a positive return value. Fix this by treating an error as though a
dequeue of 0 had been encountered.
CCP can support varied digest sizes ranging from 1 to some max value
But the current code support only fixed max values.
This patch updates the minimum digest sizes to 1
Fixes: e0d88a394e ("crypto/ccp: support run-time CPU based auth") Cc: stable@dpdk.org Signed-off-by: Amaranath Somalapuram <asomalap@amd.com>
CCP driver was scheduling only one CCP in a single burst(enqueue).
Effective throughput was limited to 1 CCP performance.
Scheduling multiple ccp within one burst will increase the ccp performance.
this changes will divide the enqueue packets equally among the multiple CCP
Fixes: e0d88a394e ("crypto/ccp: support run-time CPU based auth") Cc: stable@dpdk.org Signed-off-by: Amaranath Somalapuram <asomalap@amd.com>
when ccp_auth_opt=1 is set and if authentication error occurred,
ccp driver crash. The enqueue referance count nb_ops miss match
with dequeue nb_ops on authentication error.
Fixes: e0d88a394e ("crypto/ccp: support run-time CPU based auth") Cc: stable@dpdk.org Signed-off-by: Amaranath Somalapuram <asomalap@amd.com>
Hemant Agrawal [Thu, 7 Nov 2019 09:01:21 +0000 (14:31 +0530)]
test/event_crypto: fix check for HW support
When HW is available but the mode mismatches, it is better
to skip the test case.
Fixes: aa026482baf9 ("test/event_crypto: no service core when HW support available") Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com> Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
Pavan Nikhilesh [Fri, 22 Nov 2019 15:44:27 +0000 (21:14 +0530)]
event/octeontx2: use opposite bucket for current chunk
Since TIM buckets are always aligned to 32B and our cache line size being
128B, we will always have a cache miss when reading current_chunk pointer.
Avoid the cache miss by storing the current_chunk pointer in the bucket
opposite to the current bucket.
Pavan Nikhilesh [Fri, 22 Nov 2019 15:44:26 +0000 (21:14 +0530)]
event/octeontx2: fix HW timer race condition
Fix HW race condition observed when timeout resolution is low (<5us).
When HW traverses a given TIM bucket it will clear chunk_remainder,
but since SW always decreases the chunk_remainder at the start of the
arm routine it might cause a race where SW updates chunk_remainder
after HW has cleared it that lead to nasty side effects.
Pavan Nikhilesh [Wed, 20 Nov 2019 03:48:02 +0000 (09:18 +0530)]
common/octeontx: update mbox to version 1.1.3
Sync mail box data structures to version 1.1.3.
Add mail box version verification and defer initializing octeontx
devices if mail box version mismatches.
Update OCTEON TX limitaion with max mempool size used.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Reviewed-by: Jerin Jacob <jerinj@marvell.com>
David Marchand [Mon, 25 Nov 2019 08:10:07 +0000 (09:10 +0100)]
buildtools: fix build with coverage
A compiler can reuse a variable name and prefix it when instrumenting
with coverage.
Example:
$ make defconfig T=x86_64-native-linux-gcc O=master
$ make EXTRA_CFLAGS='--coverage' O=master
[...]
CC rte_flow.o
rte_flow_dynf_metadata_offs is not flagged as experimental but is listed
in version map
Please add __rte_experimental to the definition of
rte_flow_dynf_metadata_offs
Protect against this by adding a space character in the pattern.
Fixes: a4bcd61de82d ("buildtools: add script to check experimental API exports") Cc: stable@dpdk.org Reported-by: Andrew Rybchenko <arybchenko@solarflare.com> Signed-off-by: David Marchand <david.marchand@redhat.com> Tested-by: Andrew Rybchenko <arybchenko@solarflare.com> Acked-by: Neil Horman <nhorman@tuxdriver.com>
Anatoly Burakov [Thu, 14 Nov 2019 14:13:06 +0000 (14:13 +0000)]
mem: clarify documentation of virt2iova behaviour
It may not be immediately clear that rte_mem_virt2iova does not actually
check the internal memseg table, and will instead either return VA (in
IOVA as VA mode), or will fall back to kernel page table walk (in IOVA
as PA mode).
Add a note to API documentation indicating the above.
Remove trailing blank lines. They serve no purpose and are just
editor leftovers.
These can cause git to complain about whitespace errors during merges.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
examples/multi_process: fix client crash with sparse ports
The mp_client crashes if run on Azure or any system where ethdev
ports are owned. In that case, the tx_buffer and tx_stats for the
real port were initialized correctly, but the wrong port was used.
For example if the server has Ports 3 and 5. Then calling
rte_eth_tx_buffer_flush on any other buffer will dereference null
because the tx buffer for that port was not allocated.
Also:
- the flush code is common enough that it should not be marked
unlikely
- combine conditions to reduce indentation
- avoid unnecessary if() if sent is zero.
Fixes: e2366e74e029 ("examples: use buffered Tx") Cc: stable@dpdk.org Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com> Acked-by: Matan Azrad <matan@mellanox.com>
Gavin Hu [Mon, 25 Nov 2019 03:36:33 +0000 (11:36 +0800)]
test/rcu: fix build for small number of cores
If the RTE_MAX_LCORE is less than 10, a compilation error is generated:
app/test/test_rcu_qsbr.c:234:10: error: comparison of integer
expressions of different signedness: ‘unsigned int’ and ‘int’
[-Werror=sign-compare]
The cause is (RTE_MAX_LCORE - 10) results in a negative value.
To fix, use rte_rand() to find a number between 0 and RTE_MAX_LCORE.
Fixes: b87089b0bb19 ("test/rcu: add API and functional tests") Cc: stable@dpdk.org Signed-off-by: Gavin Hu <gavin.hu@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Reviewed-by: Steve Capper <steve.capper@arm.com> Reviewed-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Ferruh Yigit [Fri, 22 Nov 2019 11:30:23 +0000 (11:30 +0000)]
mk: remove library search path from binary
This patch functionally reverts the patch in fixes line to not have any
hardcoded library path in the final binary for the security reasons, in
case this binary distributed to production environment.
RPATH only added in RTE_DEVEL_BUILD case and this binary shouldn't
distributed, but still removing it to be cautious.
Fixes: 8919f73bcbaa ("mk: add build directory to library search path") Cc: stable@dpdk.org Suggested-by: Bruce Richardson <bruce.richardson@intel.com> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Bruce Richardson [Sat, 23 Nov 2019 02:59:59 +0000 (18:59 -0800)]
build: fix Windows build
While most windows apps can handle both "\" and "/" as path separators,
"more" is treating the "/" as the start of a command-line flag in this
case, causing errors.
Fixes: cba806e07d6f ("build: change ABI versioning to global") Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Reviewed-by: Ranjit Menon <ranjit.menon@intel.com> Tested-by: Pallavi Kadam <pallavi.kadam@intel.com> Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Thomas Monjalon [Fri, 22 Nov 2019 13:38:28 +0000 (14:38 +0100)]
devtools: disable automatic probing in null testing
The script test-null.sh is supposed to do a quick and simple
run of testpmd with null PMD only, for sanity check.
As it is not supposed to test probing of any other PMD,
physical device probing is switched to whitelist mode
by using a fake PCI address (0:0.0).
It will also help to keep memory usage stable across platforms.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net> Tested-by: Ferruh Yigit <ferruh.yigit@intel.com> Acked-by: David Marchand <david.marchand@redhat.com>
Mattias Rönnblom [Thu, 14 Nov 2019 14:10:36 +0000 (15:10 +0100)]
power: handle frequency increase with turbo disabled
Calling pstate's or acpi's rte_power_freq_up() when on the highest
non-turbo frequency results in an error, if turbo is enabled in the BIOS,
but disabled via the power library.
The error is in the form of a return code and a RTE_LOG() entry
on the ERR level.
According to the API documentation, the frequency is scaled up
"according to the available frequencies". In case turbo is disabled,
that frequency is not available. This patch's rte_power_freq_up()
behaviour is also consistent with how rte_power_freq_max() is
implemented (i.e. the highest non-turbo frequency is set, in case
turbo is disabled).
Fixes: 445c6528b55f ("power: common interface for guest and host") Fixes: e6c6dc0f96c8 ("power: add p-state driver compatibility") Cc: stable@dpdk.org Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com> Tested-by: David Hunt <david.hunt@intel.com> Acked-by: David Hunt <david.hunt@intel.com> Reviewed-by: Liang Ma <liang.j.ma@intel.com>
Matan Azrad [Tue, 12 Nov 2019 08:47:38 +0000 (08:47 +0000)]
bus/pci: clear driver on detach
When probing, a bus is responsible for filling the driver field to
indicate the device is bound.
Let's clear this field when detaching to keep a consistent behavior.
This is not a fix per se, since the device is freed when detaching.
But at least clearing the field has been added to remind that the
driver field has a meaning for the EAL.
Signed-off-by: Matan Azrad <matan@mellanox.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Ferruh Yigit [Wed, 20 Nov 2019 17:22:26 +0000 (17:22 +0000)]
kni: increase kernel version requirement for VA
A build error reported related to the selected 'get_user_pages_remote()'
kernel API:
.../kernel/linux/kni/kni_dev.h:113:8:
error: too few arguments to function ‘get_user_pages_remote’
ret = get_user_pages_remote(tsk, tsk->mm, iova, 1
^~~~~~~~~~~~~~~~~~~~~
Currently there are three versions of the 'get_user_pages_remote()'
supported, based on kernel version < 4.9, = 4.9, > 4.9.
These version based checks are not working fine with the distro kernels
which is the cause of reported build error. The error reported by the
kernel version 4.8, but it is using API defined in > 4.9.
To be able to take control of this, and possible more, related build
error, increasing the minimum supported kernel version for iova=va with
KNI to kernel version 4.9.
This leaves us with single version of the kernel API and more manageable.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com> Reviewed-by: David Marchand <david.marchand@redhat.com>
Ruifeng Wang [Fri, 15 Nov 2019 06:47:54 +0000 (14:47 +0800)]
bpf/arm: fix clang build
Clang has different prototype for __builtin___clear_cache().
It requires 'char *' parameters while gcc requires 'void *'.
Clang version 8.0 was used.
Warning messages during build:
../lib/librte_bpf/bpf_jit_arm64.c:1438:26: warning: incompatible pointer
types passing 'uint32_t *' (aka 'unsigned int *') to parameter of type
'char *' [-Wincompatible-pointer-types]
__builtin___clear_cache(ctx.ins, ctx.ins + ctx.idx);
^~~~~~~
../lib/librte_bpf/bpf_jit_arm64.c:1438:35: warning: incompatible pointer
types passing 'uint32_t *' (aka 'unsigned int *') to parameter of type
'char *' [-Wincompatible-pointer-types]
__builtin___clear_cache(ctx.ins, ctx.ins + ctx.idx);
^~~~~~~~~~~~~~~~~
Fixes: f3e516772464 ("bpf/arm: add prologue and epilogue") Cc: jerinj@marvell.com Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com> Reviewed-by: Phil Yang <phil.yang@arm.com> Reviewed-by: Gavin Hu <gavin.hu@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Jerry Hao OS [Mon, 14 Oct 2019 09:34:18 +0000 (17:34 +0800)]
config: add eMAG
Add the make and meson based build infrastructure for the eMAG platform
from Ampere Computing corp., which is a 64-bit ARM processor with 32
Armv8 64-bit CPU cores. For more information, refer to:
https://amperecomputing.com/product/
Signed-off-by: Jerry Hao OS <jerryhao@os.amperecomputing.com> Signed-off-by: Gavin Hu <gavin.hu@arm.com> Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com> Reviewed-by: Phil Yang <phil.yang@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Gavin Hu [Mon, 11 Nov 2019 05:41:50 +0000 (13:41 +0800)]
config: add Arm Neoverse N1 SDP
Arm N1 SDP is an infrastructure segment development platform
based on armv8.2-a Neoverse N1 CPU. For more information, refer to:
https://community.arm.com/developer/tools-software/oss-platforms/w/
docs/440/neoverse-n1-sdp
Signed-off-by: Gavin Hu <gavin.hu@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Reviewed-by: Steve Capper <steve.capper@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Thomas Monjalon [Wed, 20 Nov 2019 23:05:55 +0000 (00:05 +0100)]
devtools: reduce list of Arm builds tested with meson
The list of Arm configs is growing:
config/arm/arm64_armada_linux_gcc
config/arm/arm64_armv8_linux_gcc
config/arm/arm64_bluefield_linux_gcc
config/arm/arm64_dpaa_linux_gcc
config/arm/arm64_emag_linux_gcc
config/arm/arm64_n1sdp_linux_gcc
config/arm/arm64_octeontx2_linux_gcc
config/arm/arm64_thunderx2_linux_gcc
config/arm/arm64_thunderx_linux_gcc
In order to keep testing time reasonable,
and also because n1sdp is merged without a related fix in tests,
the list of configs is reduced in the script test-meson-builds.sh.
The list of tested Arm builds becomes:
build-arm64-host-clang (armv8a)
build-arm64-bluefield
build-arm64-dpaa
build-arm64-octeontx2
Signed-off-by: Thomas Monjalon <thomas@monjalon.net> Acked-by: David Marchand <david.marchand@redhat.com>
Signed-off-by: Marcin Baran <marcinx.baran@intel.com> Signed-off-by: Pawel Modrak <pawelx.modrak@intel.com> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Marcin Baran [Wed, 20 Nov 2019 17:23:36 +0000 (17:23 +0000)]
distributor: rename v2.0 ABI to _single suffix
The original ABI versioning was slightly misleading in that the
DPDK 2.0 ABI was really a single mode for the distributor, and is
used as such throughout the distributor code.
Fix this by renaming all _v20 API's to _single API's, and remove
symbol versioning.
Signed-off-by: Marcin Baran <marcinx.baran@intel.com> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: David Hunt <david.hunt@intel.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Marcin Baran [Wed, 20 Nov 2019 17:23:35 +0000 (17:23 +0000)]
distributor: remove deprecated code
Remove code for old ABI versions ahead of ABI version bump.
Signed-off-by: Marcin Baran <marcinx.baran@intel.com> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: David Hunt <david.hunt@intel.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Marcin Baran [Wed, 20 Nov 2019 17:23:34 +0000 (17:23 +0000)]
lpm: remove deprecated code
Remove code for old ABI versions ahead of ABI version bump.
Signed-off-by: Marcin Baran <marcinx.baran@intel.com> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Marcin Baran [Wed, 20 Nov 2019 17:23:33 +0000 (17:23 +0000)]
timer: remove deprecated code
Remove code for old ABI versions ahead of ABI version bump.
Signed-off-by: Marcin Baran <marcinx.baran@intel.com> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Erik Gabriel Carrillo <erik.g.carrillo@intel.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Anatoly Burakov [Wed, 20 Nov 2019 17:23:32 +0000 (17:23 +0000)]
buildtools: add ABI version update script
In order to facilitate mass updating of version files, add a shell
script that recurses into lib/ and drivers/ directories and calls
the ABI version update script.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Pawel Modrak [Wed, 20 Nov 2019 17:23:31 +0000 (17:23 +0000)]
buildtools: add symbols map update script
Add a script that automatically merges all stable ABI's under one
ABI section with the new version, while leaving experimental
section exactly as it is.
Signed-off-by: Pawel Modrak <pawelx.modrak@intel.com> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Anatoly Burakov [Wed, 20 Nov 2019 17:23:30 +0000 (17:23 +0000)]
build: remove individual library versions
Since the library versioning for both stable and experimental ABI's is
now managed globally, the LIBABIVER and version variables no longer
serve any useful purpose, and can be removed.
The replacement in Makefiles was done using the following regex:
^(#.*\n)?LIBABIVER\s*:=\s*\d+\n(\s*\n)?
(LIBABIVER := numbers, optionally preceded by a comment and optionally
succeeded by an empty line)
The replacement for meson files was done using the following regex:
^(#.*\n)?version\s*=\s*\d+\n(\s*\n)?
(version = numbers, optionally preceded by a comment and optionally
succeeded by an empty line)
[David]: those variables are manually removed for the files:
- drivers/common/qat/Makefile
- lib/librte_eal/meson.build
[David]: the LIBABIVER is restored for the external ethtool example
library.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Signed-off-by: David Marchand <david.marchand@redhat.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Marcin Baran [Wed, 20 Nov 2019 17:23:28 +0000 (17:23 +0000)]
build: change ABI versioning to global
As per new ABI policy [1], all of the libraries are now versioned using
one global ABI version. Stable libraries use the MAJOR.MINOR ABI
version for their shared objects, while experimental libraries
use the 0.MAJORMINOR convention for their versioning.
Experimental library versioning is managed globally. Changes in this
patch implement the necessary steps to enable that.
The CONFIG_RTE_MAJOR_ABI option was introduced to permit multiple
DPDK versions installed side by side. The problem is now addressed
through the new ABI policy, and thus can be removed.
[David] For external libraries relying on Makefile, LIBABIVER is
preserved to avoid using DPDK global ABI version.
Signed-off-by: Marcin Baran <marcinx.baran@intel.com> Signed-off-by: Pawel Modrak <pawelx.modrak@intel.com> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com> Signed-off-by: David Marchand <david.marchand@redhat.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Bruce Richardson [Fri, 15 Nov 2019 15:17:00 +0000 (15:17 +0000)]
examples: hide error for missing pkg-config path flag
Some versions of pkg-config don't support the --path flag, which is not a
fatal error when building the apps. Without the flag, the makefile just
cannot track the .pc file of DPDK as a dependency of the build. Therefore,
we can ignore the error and suppress it by redirecting to /dev/null the
stderr from that call to pkg-config.
Fixes: 22119c4591a0 ("examples: use pkg-config in makefiles") Cc: stable@dpdk.org Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Bruce Richardson [Fri, 15 Nov 2019 15:16:59 +0000 (15:16 +0000)]
examples: fix build with old pkg-config
Not all versions of pkg-config in distros have support for the
--define-prefix flag [1], causing errors when building examples manually or
with test-meson-builds.sh script [2].
For the former case, we need to remove the hard-coded use of the flag in
the Makefiles.
For the latter case, the flag is necessary for builds to succeed, so we
skip the tests when it's not present, passing it as part of the pkg-config
command if it is supported.
[1]
CentOS Linux release 7.7.1908 (Core)
pkg-config version 0.27.1
[2]
## Building cmdline
Unknown option --define-prefix
gmake: Entering directory
`...ild-x86-default/install-root/usr/local/share/dpdk/examples/cmdline'
rm -f build/cmdline build/cmdline-static build/cmdline-shared
test -d build && rmdir -p build || true
Unknown option --define-prefix
Unknown option --define-prefix
gcc -O3 main.c commands.c parse_obj_list.c -o build/cmdline-shared
main.c:14:28: fatal error: cmdline_rdline.h: No such file or directory
Fixes: ca9268529d2b ("examples: support relocated DPDK install") Fixes: 7f80a2102bbb ("devtools: test pkg-config file") Cc: stable@dpdk.org Reported-by: Ferruh Yigit <ferruh.yigit@intel.com> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Matan Azrad [Tue, 19 Nov 2019 15:32:28 +0000 (15:32 +0000)]
net/mlx5: fix L3 encapsulation flow validation
In order to configure L3 encapsulation\decapsulation flow to mlx5
devices, 2 actions should be added to the flow actions list:
RTE_FLOW_ACTION_TYPE_RAW_DECAP and RTE_FLOW_ACTION_TYPE_RAW_ENCAP.
One of the validations for this scenario is to check that modify actions
is not done before the L3 decapsulation, because it doesn't make sense
to decapsulate a modified data.
Wrongly, this check was done for the case of L3 encapsulation what
causes a validation failure in modify + L3 encapsulation flow.
Ignore this check in case of L3 encapsulation.
Fixes: 4bb14c83df95 ("net/mlx5: support modify header using Direct Verbs") Cc: stable@dpdk.org Signed-off-by: Matan Azrad <matan@mellanox.com>
Ajit Khaparde [Wed, 20 Nov 2019 03:00:23 +0000 (19:00 -0800)]
net/bnxt: fix flow check for null spec and mask
bnxt_validate_and_parse_flow_type already has protocol specific NULL
checks for rte_flow_item spec and mask and take actions accordingly.
A check at the top of the loop is redundant and is preventing the
protocol specific checks from being executed.
Rahul Gupta [Fri, 15 Nov 2019 09:40:09 +0000 (15:10 +0530)]
net/bnxt: fix jumbo frame configuration in firmware
In order to prevent reconfiguration of firmware resources for every
MTU change, configure FW with max MTU value using hwrm_func_cfg
to support all frame sizes. There is no need to overwrite the driver
level MTU variable data->mtu with the FW MTU.
Fixes: 905cd45ce30e ("net/bnxt: use configured MTU during load") Cc: stable@dpdk.org Signed-off-by: Rahul Gupta <rahul.gupta@broadcom.com> Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com> Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Ajit Khaparde [Mon, 18 Nov 2019 22:23:45 +0000 (14:23 -0800)]
net/bnxt: fix resource qcaps with older FW
On some old versions of FW, bnxt_hwrm_func_resc_qcaps can return an
error. This is because the command was not implemented completely
in FW till the subsequent version. Ignore the error and continue with
the driver initialization.
Qi Zhang [Tue, 19 Nov 2019 06:14:42 +0000 (14:14 +0800)]
net/ice: add flow mark hint
Since not all data paths support flow mark, the driver needs
a hint from application to select the correct data path if
flow mark is required. The patch introduces a devarg
"flow-mark-support" as a workaround solution, since a standard
way is still ongoing.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com>
Yahui Cao [Mon, 18 Nov 2019 22:23:43 +0000 (06:23 +0800)]
net/ice: fix flow director rule after device stop
By moving irq enable/disable from device start/stop to FDIR
setup/teardown, FDIR queue irq config is independent of LAN queue irq
config. So device stop will not cause FDIR rule failure.
Qi Zhang [Tue, 19 Nov 2019 01:07:02 +0000 (09:07 +0800)]
net/ice: fix flow type selection for flow director
The FDIR parser will select ICE_FLTR_PTYPE_NONF_IPV4_OTHER as flow type
for an IPv4 UDP flow with empty l4 matching field which is not correct.
Same issues happens on all the combination between IPv4/IPv6 and
UDP/TCP/SCTP cases.
The patch fixes all the wrong flow ptype selections.
Fixes: f5cafa961fae ("net/ice: add flow director create and destroy") Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Beilei Xing <beilei.xing@intel.com>
As the result of testing it was found that some hosts have
the performance penalty imposed by required write memory barrier
after doorbell writing. Before 19.08 release there was some
heuristics to decide whether write memory barrier should be
performed. For the bursts of recommended size (or multiple)
it was supposed there were some extra ongoing packets in the
next burst and write memory barrier may be skipped (supposed
to be performed in the next burst, at least after descriptor
writing).
This patch restores that behaviour, the devargs tx_db_nc=2
must be specified to engage this performance tuning feature.
Dekel Peled [Thu, 24 Oct 2019 12:52:53 +0000 (15:52 +0300)]
net/mlx5: change default flow engine to DV
The default flow engine is Verbs flow engine, for legacy reasons.
This patch changes the default to DV flow engine (dv_flow_en = 1).
Documentation is updated accordingly.
The rdma_core routine mlx5dv_dr_create_flow_action_dest_vport()
requires the vport id parameter to create port action.
The register c[0] value was used to deduce the port id value
and it fails in bonding configuration. The correct way is
to apply vport_num value queried from the rdma_core library.
Matan Azrad [Sun, 17 Nov 2019 12:14:54 +0000 (12:14 +0000)]
net/mlx5: fix flow table hash list conversion
For the case when DR is not supported and DV is supported:
multi-tables feature is off.
In this case, only table 0 is supported.
Table 0 structure wrongly was not created what prevented any
matcher object to be created and even caused crashes.
Create the table hash list in DV case too.
Create table zero empty structure for each domain when DR is not
supported.
Allow NULL DR internal table object to be used.
Jakub Grajciar [Mon, 4 Nov 2019 11:03:00 +0000 (12:03 +0100)]
net/memif: support zero-copy slave
Zero-copy slave support for memif PMD.
Slave interface exposes DPDK memory to
master interface. Only single file segments
are supported (EAL option --single-file-segments).
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Dekel Peled [Thu, 14 Nov 2019 13:59:13 +0000 (15:59 +0200)]
app/testpmd: fix cleanup of Tx metadata offload
Commit in fixes tag removed the match_metadata Tx offload.
This patch removes the option to select this offload from testpmd
menu, help text and documentation.
It also modifies the cmd_show_tx_metadata_parsed() function, to
display the value correctly, and the dump_pkt_burst() function to
display the relevant (Tx/Rx) metadata only.
Andrew Rybchenko [Thu, 14 Nov 2019 16:40:52 +0000 (16:40 +0000)]
ethdev: improve message about not disabled offload
Avoid usaged of "failed" in the message about not requested but
enabled offload, since it is not a failure.
Fixes: 1daa33805824 ("ethdev: validate offloads set by PMD") Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com> Acked-by: Ferruh Yigit <ferruh.yigit@intel.com> Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Andrew Rybchenko [Thu, 14 Nov 2019 16:40:51 +0000 (16:40 +0000)]
ethdev: decrease verbosity of not disabled offload logs
Right now a PMD decides if it is critical that an offload cannot
be disabled (i.e. not requested, but still enabled). If PMD treaks
it as OK, we should not spam logs with corresponding messages
by default. Default log level in ethdev is INFO, so change the
message level to DEBUG.
Fixes: 1daa33805824 ("ethdev: validate offloads set by PMD") Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com> Acked-by: Ferruh Yigit <ferruh.yigit@intel.com> Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Pavan Nikhilesh [Thu, 7 Nov 2019 02:52:54 +0000 (08:22 +0530)]
net/octeontx2: support reduced set of packet types
Add support to set supported ptypes for octeontx2 and
remove devarg scheme to disable ptype parsing support as
application can use rte_eth_dev_set_ptypes() normative API
to enable the same use case.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Harman Kalra [Mon, 11 Nov 2019 05:48:54 +0000 (05:48 +0000)]
net/octeontx2: fix PTP configurations for VF
Issue has been observed if PTP is already enabled on PF and
later VFs are configured. Since PTP requires mbuf data off
to be shifted by 8 bytes, due to this l3fwd/l2fwd was not
working with VFs.
Also some extra garbage bytes were observed in packet data
when ptp was enabled.
Fixes: b5dc3140448e ("net/octeontx2: support base PTP") Cc: stable@dpdk.org Signed-off-by: Harman Kalra <hkalra@marvell.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Anoob Joseph [Mon, 11 Nov 2019 13:21:14 +0000 (18:51 +0530)]
net/octeontx2: fix error handling after CQ init
After otx2_nix_register_cq_irqs() is called and the IRQs are setup,
otx2_nix_unregister_cq_irqs() need to be called in the subsequent error
exit paths.
Fixes: d34db5ccbf30 ("net/octeontx2: fix driver reconfiguration") Cc: stable@dpdk.org Signed-off-by: Anoob Joseph <anoobj@marvell.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
git.droids-corp.org / dpdk.git / log