Ivan Malov [Wed, 6 Jan 2021 10:06:01 +0000 (13:06 +0300)]
common/sfc_efx/base: enhance field ID check in field set API
A field ID passed to the API may point to a gap in the array
of field descriptors. Turn down such invocations as improper.
Fixes: 370ed675a952 ("common/sfc_efx/base: support setting PPORT in match spec") Cc: stable@dpdk.org Reviewed-by: Andy Moreton <amoreton@xilinx.com> Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru> Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Ivan Malov [Wed, 6 Jan 2021 10:06:00 +0000 (13:06 +0300)]
common/sfc_efx/base: fix MAE match spec class comparison API
The helper exits once it encounters a field which hasn't its
capability status reported by the FW. Handle the corner case
when the two mask-value pairs match for the field, which, in
the absence of capability information, is sufficient to deem
the class unaffected by the field. Explain this in a comment.
Fixes: bb71f7e0a35a ("common/sfc_efx/base: add match specs class comparison API") Cc: stable@dpdk.org Reviewed-by: Andy Moreton <amoreton@xilinx.com> Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru> Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Ivan Malov [Wed, 6 Jan 2021 10:05:59 +0000 (13:05 +0300)]
common/sfc_efx/base: fix MAE match spec validation helper
A particular FW version is aware of some set of match fields.
Depending on FW configuration and match specification type, a
known field may not necessarily be allowed to have a non-zero
mask. FW communicates such restrictions via field capabilities
MCDI. Newer FW may be aware of more fields. For such fields,
older FW simply does not report any capabilities.
A situation may occur when libefx is aware of a match field
which the FW is unaware of (eg., older FW), that is, FW does
not report capability status for this field. In this case,
libefx must consider such field as unsupported and demand
all-zeros mask for it when validating match specifications.
Currently, the helper in question simply exits and reports
that the specification is valid when it encounters a field
with no capability status available. This is clearly wrong.
Introduce the missing check to fix the problem.
Fixes: 34285fd0891d ("common/sfc_efx/base: add match spec validate API") Cc: stable@dpdk.org Reviewed-by: Andy Moreton <amoreton@xilinx.com> Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru> Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Ivan Malov [Mon, 28 Dec 2020 03:00:23 +0000 (06:00 +0300)]
common/sfc_efx/base: fix MPORT related byte order handling
MPORT values derived by helper functions are little-endian.
At the same time, MCDIs which consume these values perform
one more host-order to little-endian conversion internally.
Fix the helper functions to return host-order MPORT values.
Fixes: 370ed675a952 ("common/sfc_efx/base: support setting PPORT in match spec") Fixes: bb024542fffd ("common/sfc_efx/base: add API for adding action drop") Fixes: 097058033f03 ("common/sfc_efx/base: add API to get mport of PF/VF") Cc: stable@dpdk.org Reported-by: Andy Moreton <amoreton@xilinx.com> Reviewed-by: Andy Moreton <amoreton@xilinx.com> Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru> Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Karra Satwik [Sun, 20 Dec 2020 22:47:41 +0000 (04:17 +0530)]
net/cxgbe: rework and simplify link handling
Rework and simplify link handling code. Remove redundant
variables in link configuration structure and directly
extract information from the 32-bit link capabilities.
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com> Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Balazs Nemeth [Wed, 23 Dec 2020 14:15:49 +0000 (15:15 +0100)]
net/qede: fix promiscuous enable
When calling rte_eth_promiscuous_enable(port_id) followed by
rte_eth_allmulticast_enable(port_id), the port is not in promisc mode
anymore. This patch ensures that promisc mode takes precedence over
allmulticast mode fixing the regression introduced by b10231aed1ed.
Fixes: b10231aed1ed ("net/qede: fix multicast drop in promiscuous mode") Cc: stable@dpdk.org Signed-off-by: Balazs Nemeth <bnemeth@redhat.com> Acked-by: Rasesh Mody <rmody@marvell.com>
This patch add support for generating GTP PDU container
session option for the raw encap and raw decap sets.
The generated options is single 32-bit word with
minimal length specified as 4, no extra fields in the
option are supported. The option item must be preceded
with the GTP item itself, and GTP item flags must be
set accordingly:
- E flag must be set, we are going to provide extension
- S flag must be reset, because GTP item does not
provide the value for SQN field, we can't fill this one
- PN flag must be reset, no N-PDU value provided by
GTP item either
The raw set example:
set raw_encap 2 eth / ipv4 / udp /
gtp v_pt_rsv_flags is 0x34 / gtp_psc / end_set
Please, note - value 0x34 provides the required flag combination.
Signed-off-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com> Acked-by: Ori Kam <orika@nvidia.com>
This patch introduces the GTP header individual flag bit fields
and the header optional word with N-PDU number, Sequence Number
and Next Extension Header type.
Signed-off-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com> Acked-by: Ori Kam <orika@nvidia.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Min Hu (Connor) [Wed, 6 Jan 2021 03:46:31 +0000 (11:46 +0800)]
net/hns3: fix crash with multi-process
In current version, procedure of saving eth_dev in
hns3 PMD init will be called more than twice, one
for primary, the other for secondary. That will cause
segmentation fault in Multi-process as eth_dev will
be changed in secondary process, which is different
from one in primary process.
The initial problem was access to 'rte_eth_devices'
global variable, which is wrong. But current approach
can cause problem for the secondaries, moving 'eth_dev'
to process private can work but before making things
more complex.
This patch deserted the procedure of saving eth_dev in
hns3 PMD init. Instead, it creates an internal function
that gets "struct hns3_hw" as parameter and it can be
called internally without knowing 'eth_dev'and the
.dev_ops can be wrapper to this.
Fixes: 2390bf217f4d ("net/hns3: fix FEC state query") Cc: stable@dpdk.org Signed-off-by: Min Hu (Connor) <humin29@huawei.com> Signed-off-by: Lijun Ou <oulijun@huawei.com>
Huisong Li [Wed, 6 Jan 2021 03:46:30 +0000 (11:46 +0800)]
net/hns3: fix Rx/Tx errors stats
Abnormal errors stats in Rx/Tx datapath are statistics
items in driver, and displayed in xstats. They should
be cleared by the rte_eth_xstats_reset api, instead of
the rte_eth_stats_reset.
Fixes: c4b7d6761d01 ("net/hns3: get Tx abnormal errors in xstats") Fixes: 521ab3e93361 ("net/hns3: add simple Rx path") Fixes: bba636698316 ("net/hns3: support Rx/Tx and related operations") Cc: stable@dpdk.org Signed-off-by: Huisong Li <lihuisong@huawei.com> Signed-off-by: Lijun Ou <oulijun@huawei.com>
Huisong Li [Wed, 6 Jan 2021 03:46:29 +0000 (11:46 +0800)]
net/hns3: fix error code in xstats
The ethdev API has processed the failure to obtain
xstats statistics. Therefore, driver should return
an error code instead of 0 in 'hns3_dev_xstats_get'
API.
Fixes: 8839c5e202f3 ("net/hns3: support device stats") Cc: stable@dpdk.org Signed-off-by: Huisong Li <lihuisong@huawei.com> Signed-off-by: Lijun Ou <oulijun@huawei.com>
Huisong Li [Wed, 6 Jan 2021 03:46:28 +0000 (11:46 +0800)]
net/hns3: fix xstats with id and names
Currently, validity check for ids and values in the
hns3_dev_xstats_get_by_id API is incorrect, which will
cause a problem. Namely, if the ID range of the xstats
stats item does not include the basic stats item, the
app can not obtain the corresponding xstats statistics
in hns3_dev_xstats_get_by_id.
Similarly, the hns3_dev_xstats_get_names_by_id interface
also has a problem.
Although the input parameter verification code cannot be
executed due to the implementation of the ethdev framework
interface, the driver needs to ensure the correctness of
the input parameters.
Fixes: 8839c5e202f3 ("net/hns3: support device stats") Cc: stable@dpdk.org Signed-off-by: Huisong Li <lihuisong@huawei.com> Signed-off-by: Lijun Ou <oulijun@huawei.com>
Lijun Ou [Wed, 6 Jan 2021 03:46:27 +0000 (11:46 +0800)]
net/hns3: fix interception with flow director
The rte_fdir_conf structure has deprecated and users need
to use the specified rule parameters of rte_flow structure
when configure a flow rule. As a result, it is incorrectly
used in the rte_flow API.
Fixes: fcba820d9b9e ("net/hns3: support flow director") Cc: stable@dpdk.org Signed-off-by: Lijun Ou <oulijun@huawei.com>
Leyi Rong [Thu, 14 Jan 2021 06:39:51 +0000 (14:39 +0800)]
net/i40e: optimize Tx by using AVX512
Optimize Tx path by using AVX512 instructions and vectorize the
tx free bufs process.
Signed-off-by: Leyi Rong <leyi.rong@intel.com> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Leyi Rong [Thu, 14 Jan 2021 06:39:50 +0000 (14:39 +0800)]
net/i40e: add AVX512 vector path
Add AVX512 support for i40e PMD. This patch adds i40e_rxtx_vec_avx512.c
to support i40e AVX512 vPMD.
This patch aims to enable AVX512 on i40e vPMD. Main changes are focus
on Rx path compared with AVX2 vPMD.
Signed-off-by: Leyi Rong <leyi.rong@intel.com> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Leyi Rong [Thu, 14 Jan 2021 06:39:49 +0000 (14:39 +0800)]
net/i40e: remove devarg use-latest-supported-vec
As eal parameter --force-max-simd-bitwidth is already introduced,
to make it more clear when setting rx/tx function, remove
devarg use-latest-supported-vec support.
Signed-off-by: Leyi Rong <leyi.rong@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Weifeng Li [Sun, 10 Jan 2021 08:34:17 +0000 (16:34 +0800)]
net/i40e: fix X722 for 802.1ad frames ability
I40E_DEV_ID_SFP_I_X722 does not support 802.1ad frames ability,
so I40E_HW_FLAG_802_1AD_CAPABLE should not be set.
The patch also correct the comment for I40E_HW_FLAG_802_1AD_CAPABLE
configure.
Fixes: 9efa8d28b4da ("net/i40e: fix SFP X722 with FW4.16") Cc: stable@dpdk.org Signed-off-by: Weifeng Li <liweifeng96@126.com> Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Hemant Agrawal [Thu, 14 Jan 2021 07:05:02 +0000 (12:35 +0530)]
crypto/dpaa2_sec: support AES-CMAC integrity check
This patch adds support for AES_CMAC integrity in non-security mode.
This patch modifies the camm flib to handles the AES CMAC
without conflicting the proto ALG operations. i.e. by creating
another ALG operation routine.
Hemant Agrawal [Thu, 14 Jan 2021 07:05:01 +0000 (12:35 +0530)]
crypto/dpaa_sec: reduce log on queue closure
If for some reason the queue is not close properly,
specially in test cases.
The QUEUE retire prints are flooding the screen.
They are not really required as WARNING.
The descriptor sharing needed to be changed for ZUC+ZUC as we
were getting invalid CHA combination error due to sharing
being done on DECOs simultaneously.
This patch adds APIs to add/remove callback functions on crypto
enqueue/dequeue burst. The callback function will be called for
each burst of crypto ops received/sent on a given crypto device
queue pair.
Tejasree Kondoj [Thu, 31 Dec 2020 17:52:57 +0000 (23:22 +0530)]
crypto/octeontx2: support CN98xx
CN98xx SoC comes up with two CPT blocks wrt
CN96xx, CN93xx, to achieve higher performance.
Adding support to allocate all LFs of VF with even BDF from CPT0
and all LFs of VF with odd BDF from CPT1.
If LFs are not available in one block then they will be allocated
from alternate block.
Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com> Acked-by: Anoob Joseph <anoobj@marvell.com>
Ori Kam [Thu, 7 Jan 2021 16:57:08 +0000 (18:57 +0200)]
regex/mlx5: support combined rule file
The rof file holds programming instructions for
a given HW version.
In order to support future generation of HW it
was decided that the rof file will hold number
of rule configurations, and the driver will use
the one that matches the HW version.
In current code we force sync after each write block.
This has impact on performance.
The solution is to move the sync to the end of the
entire programming sequence.
Checkpatch prefers 'unsigned int' over bare 'unsigned'.
Reword the error messages for brevity and clarity
so they don't have to be split across multiple lines.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Harman Kalra [Tue, 12 Jan 2021 18:24:46 +0000 (23:54 +0530)]
examples/l3fwd: remove limitation on Tx queue count
In l3fwd no of transmit queues is calculated based on no of
lcores with which it is launched. Hence maximum no of tx
queues possible per port should depend on RTE_MAX_LCORE value.
Fixes: af75078fece3 ("first public release") Cc: stable@dpdk.org Signed-off-by: Harman Kalra <hkalra@marvell.com> Reviewed-by: David Marchand <david.marchand@redhat.com>
Anatoly Burakov [Thu, 14 Jan 2021 15:02:45 +0000 (15:02 +0000)]
fbarray: fix overlap check
When we're attaching fbarrays in secondary processes, we check for
whether the intended memory address for the fbarray is already in use by
some other, local fbarray. However, the check for end-overlap (i.e. to
see if our memory area's end overlaps with some other fbarray) is
incorrectly counting end offset as part of the overlap. Fix the check.
Anatoly Burakov [Tue, 10 Nov 2020 15:41:36 +0000 (15:41 +0000)]
eal/linux: improve no hugepages logging
When no hugepages are found, we log a message about it, but we never
specify on which node. We also implicitly declare the page size based
on the directory name, but that's not very user friendly.
Fix both by changing the text of the message to note the NUMA node (if
applicable) and explicitly mention page size in kilobytes.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Reviewed-by: David Marchand <david.marchand@redhat.com>
Liang Ma [Thu, 14 Jan 2021 14:46:08 +0000 (14:46 +0000)]
ethdev: add simple power management API
Add a simple API to allow getting the monitor conditions for
power-optimized monitoring of the Rx queues from the PMD, as well as
release notes information.
Signed-off-by: Liang Ma <liang.j.ma@intel.com> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Anatoly Burakov [Thu, 14 Jan 2021 14:46:07 +0000 (14:46 +0000)]
eal: add monitor wakeup function
Now that we have everything in a C file, we can store the information
about our sleep, and have a native mechanism to wake up the sleeping
core. This mechanism would however only wake up a core that's sleeping
while monitoring - waking up from `rte_power_pause` won't work.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Anatoly Burakov [Thu, 14 Jan 2021 14:46:06 +0000 (14:46 +0000)]
eal: remove sync version of power monitor
Currently, the "sync" version of power monitor intrinsic is supposed to
be used for purposes of waking up a sleeping core. However, there are
better ways to achieve the same result, so remove the unneeded function.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Anatoly Burakov [Thu, 14 Jan 2021 14:46:04 +0000 (14:46 +0000)]
eal: avoid invalid power intrinsics API usage
Currently, the API documentation mandates that if the user wants to use
the power management intrinsics, they need to call the
`rte_cpu_get_intrinsics_support` API and check support for specific
intrinsics.
However, if the user does not do that, it is possible to get illegal
instruction error because we're using raw instruction opcodes, which may
or may not be supported at runtime.
Now that we have everything in a C file, we can check for support at
startup and prevent the user from possibly encountering illegal
instruction errors.
We also add return values to the API's as well, because why not.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Anatoly Burakov [Thu, 14 Jan 2021 14:46:03 +0000 (14:46 +0000)]
eal: uninline power intrinsics
Currently, power intrinsics are inline functions. Make them part of the
ABI so that we can have various internal data associated with them
without exposing said data to the outside world.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Juraj Linkeš [Fri, 15 Jan 2021 13:26:06 +0000 (14:26 +0100)]
config/arm: add core and NUMA counts to cross files
Add support for setting core count and numa nodes in cross files. The
values specified in cross files will override the default values.
Also add missing default values to Arm config.
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Tested-by: Vimal Chungath <vcchunga@amazon.com> Tested-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Juraj Linkeš [Fri, 15 Jan 2021 13:26:04 +0000 (14:26 +0100)]
config/arm: isolate generic build
Use generic configuration for the only build where it makes sense - the
generic build. For other builds, if we don't know either of implementer
ID or part number, the build is not supported.
Add part numbers to cross files where fallback to generic configuration
is assumed.
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Tested-by: Vimal Chungath <vcchunga@amazon.com> Tested-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Juraj Linkeš [Fri, 15 Jan 2021 13:26:03 +0000 (14:26 +0100)]
config/arm: organize config into dict
Use dictionary lookup instead of checking for existing variables,
iterating over all elements in the list or checking lists for optional
configuration. Move variable contents into the dictionary for variables
that would be referenced only once.
Fallback to generic part number if the discovered part number is
unknown.
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Tested-by: Vimal Chungath <vcchunga@amazon.com> Tested-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Juraj Linkeš [Fri, 15 Jan 2021 13:26:01 +0000 (14:26 +0100)]
config/arm: reformat and move config and comments
Change formatting so that it's more consistent and readable, add/modify
comments/stdout messages, move configuration options to more appropriate
places and make the order consistent according to these rules:
1. First list generic configuration options, then list options that may
be overwritten. List SoC-specific options last.
2. For SoC-specific options, list number of cores before the number of
NUMA nodes, to make it consistent with config/meson.build.
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Tested-by: Vimal Chungath <vcchunga@amazon.com> Tested-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Juraj Linkeš [Fri, 15 Jan 2021 13:25:59 +0000 (14:25 +0100)]
config/arm: rename variables
Rename Arm build variables and values so that they better conform to Arm
specifications. Also rename generically sounding variable to names that
better capture what the variables hold.
Rename machine_args_generic to part_number_config_arm since the
variable contains more than just the generic machine args and is used
mainly as the fallback arm configuration.
Rename the default machine args to generic machine args to reflect that.
The rest of the variables are self-explanatory.
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech> Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Jerin Jacob <jerinj@marvell.com> Tested-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Tested-by: Vimal Chungath <vcchunga@amazon.com> Tested-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Currently bitmap line not empty check API assumes cache line
of 64B and only checks 8 slabs. Since in 128B cacheline, we
have 16 slabs per cacheline, rte_bitmap_clear() will mark
complete line as empty as soon as 8 slabs are empty thereby
breaking bitmap scan functionality. Fix it by defining new
__rte_bitmap_line_not_empty() for 128B cacheline platform.
Long Li [Mon, 21 Dec 2020 21:33:22 +0000 (13:33 -0800)]
net/netvsc: support VF device hot add/remove
When a VF device is present, netvsc can send or receive packets over the
VF device. The VF device driver communicates directly with the PCI device
via the PF from the host hypervisor. This is faster than exchanging data
with netvsp via vmbus, i.e. syntheic path.
In Azure and Hyper-v environments, VF device can be hot added or hot
removed at anytime while guest VM is running. This patch improves netvsc
to support VF device hot add/remove.
1. netvsc monitors all system hot add activities over the PCI bus. When it
detects a VF device is added to the system and is managed under this
netvsc device, it asks EAL to probe and start this VF device, then it
attaches and switches data path to the VF device.
2. After a VF device is attached to netvsc, netvsc monitors this device on
hot remove. When this VF device is hot removed, netvsc switches data path
to synthetic, stops this VF device and removes it from EAL.
3. If any failure happens during a VF device hot remove or add, the netvsc
falls back to synthetic path for all data traffic.
Long Li [Mon, 21 Dec 2020 21:32:36 +0000 (13:32 -0800)]
eal/linux: allow multiple starts of event monitor
In some cases, a device or infrastructure may want to enable hotplug
but application may also try and start hotplug as well. Therefore
change the monitor_started from a boolean into a reference count.
When running one test (via DPDK_TEST) the test program
would leave the terminal in raw mode. This was because
it was setting up cmdline to do interactive input.
The fix is to use cmdline_new() for the interactive case.
This also fixes a memory leak because the test
runner was never calling cmdline_free().
Fixes: 9b848774a5dc ("test: use env variable to run tests") Cc: stable@dpdk.org Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Tested-by: Harry van Haaren <harry.van.haaren@intel.com>
Thomas Monjalon [Wed, 2 Dec 2020 16:47:03 +0000 (17:47 +0100)]
devtools: reduce ABI checks and static binaries
When testing compilation and checking ABI compatibility,
there is no real need of static binaries eating disks.
The static linkage of applications was already well tested,
though the static examples tested with meson were limited to "l3fwd" only.
The static build test with make is limited to "helloworld" example.
The ABI compatibility is checked on shared libraries,
and there is no need to test again on similar builds.
A new parameter is added to the function "build",
so the ABI check is enabled only for native gcc and clang shared builds,
32-bit, generic armv8 and ppc cross compilations.
In other words, it is disabled for some static builds and some Arm ones.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net> Acked-by: David Marchand <david.marchand@redhat.com>
Olivier Matz [Wed, 13 Jan 2021 08:28:06 +0000 (09:28 +0100)]
service: propagate init error in EAL
Currently, when rte_service_init() fails at initialization, the
application always gets a ENOEXEC error code. For example, with testpmd,
this is displayed as:
Cannot init EAL: Exec format error
This error code does not describe the real issue. Instead, use the error
code returned by the function.
Fixes: e39824500825 ("service: initialize with EAL") Cc: stable@dpdk.org Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
Tyler Retzlaff [Thu, 14 Jan 2021 21:22:35 +0000 (13:22 -0800)]
bus/pci: fix build with Windows SDK >= 10.0.20253
NetUIO device class and interface GUIDs are defined in system
headers starting from platform SDK v10.0.20253. Inspect SDK
version to avoid redefinition.
Pre-release SDKs do not promise compatibility and a narrow
subset of SDKs may still be subject to redefinition.
Gaetan Rivet [Mon, 9 Nov 2020 13:37:55 +0000 (14:37 +0100)]
bus/dpaa: optimize device name parsing
Device name parsing is done on all buses during device iterations at
either EAL or ethdev levels.
When a bus implements device name parsing slowly, all iterations are
impacted. Efficient implementation is important.
The DPAA bus device name parsing has two issues: it allocates dynamic
memory and uses snprintf without a real need for it. Both can be
avoided, which improves the parsing performance.
Yicai Lu [Wed, 16 Dec 2020 13:36:30 +0000 (21:36 +0800)]
ip_frag: remove padding length of fragment
In some situations, we would get several ip fragments, which total
data length is less than min_ip_len(64) and padding with zeros.
We simulated intermediate fragments by modifying the MTU.
To illustrate the problem, we simplify the packet format and
ignore the impact of the packet header.In namespace2,
a packet whose data length is 1520 is sent.
When the packet passes tap2, the packet is divided into two
fragments: fragment A and B, similar to (1520 = 1510 + 10).
When the packet passes tap3, the larger fragment packet A is
divided into two fragments A1 and A2, similar to (1510 = 1500 + 10).
Finally, the bond interface receives three fragments:
A1, A2, and B (1520 = 1500 + 10 + 10).
One fragmented packet A2 is smaller than the minimum Ethernet
frame length, so it needs to be padded.
When processing the preceding packets above,
DPDK would aggregate fragmented packets A2 and B.
And error packets are generated, which padding(zero)
is displayed in the middle of the packet.
A2 + B:
0000 fa 16 3e 9f fb 82 fa 47 b2 57 dc 20 08 00 45 00
0010 00 33 b4 66 00 ba 3f 01 c1 a5 01 01 01 01 02 01
0020 01 02 c0 c1 c2 c3 c4 c5 c6 c7 00 00 00 00 00 00
0030 00 00 00 00 00 00 00 00 00 00 00 00 c8 c9 ca cb
0040 cc cd ce cf d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 da db
0050 dc dd de df e0 e1 e2 e3 e4 e5 e6
So, we would calculate the length of padding, and remove
the padding in pkt_len and data_len before aggregation.
And also we have the fix for both ipv4 and ipv6.
Fixes: 7f0983ee331c ("ip_frag: check fragment length of incoming packet") Cc: stable@dpdk.org Signed-off-by: Yicai Lu <luyicai@huawei.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
git.droids-corp.org / dpdk.git / log