Bruce Richardson [Thu, 10 Jan 2019 10:37:26 +0000 (10:37 +0000)]
devtools: fix build check for whether meson has run
The current check to see whether we need to call meson or just ninja
simply checked if the build directory existed. However, if meson was run
but failed, the build directory would still exist. We can fix this by
instead checking for the build.ninja file inside the directory. Once that
is present, we can use ninja safely and let it worry about rerunning
meson if necessary.
Fixes: a55277a788df ("devtools: add test script for meson builds") Cc: stable@dpdk.org Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Luca Boccassi <bluca@debian.org>
Bruce Richardson [Thu, 10 Jan 2019 10:29:19 +0000 (10:29 +0000)]
build: fix variable name in dependency error message
The variable name in the error message had an extra '_' which caused
an actual meson error when the message would otherwise be printed to
give meaningful information about what was going wrong.
Fixes: 203b61dc5e18 ("build: improve error message for missing dependency") Cc: stable@dpdk.org Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Luca Boccassi <bluca@debian.org>
Luca Boccassi [Fri, 11 Jan 2019 16:12:44 +0000 (16:12 +0000)]
build: use static deps for pkg-config libs.private
Dependencies of the RTE libraries were not being added to the
Requires.private field of the pc file since the variable used for
dynamic linking was passed to the related field of pkg.generate.
Use the static one so that dependencies are included.
Fixes: 57ae0ec62620 ("build: add dependency on telemetry to apps with meson") Cc: stable@dpdk.org Signed-off-by: Luca Boccassi <bluca@debian.org>
Akhil Goyal [Wed, 9 Jan 2019 15:14:21 +0000 (15:14 +0000)]
crypto/dpaa2_sec: fix FLC address for physical mode
Fixes: 547a4d40e7bf ("crypto/dpaa2_sec: support out of place protocol offload") Fixes: 0a23d4b6f4c2 ("crypto/dpaa2_sec: support protocol offload IPsec") Cc: stable@dpdk.org Signed-off-by: Akhil Goyal <akhil.goyal@nxp.com>
Akhil Goyal [Wed, 9 Jan 2019 15:14:17 +0000 (15:14 +0000)]
crypto/dpaa_sec: support same session flows on multi-cores
In dpaa_sec, each session should have a separate in-queue,
and each in-queue should be mapped to an outq.
So if multiple flows of same SA comes to different cores
due to rss distribution of NIC, enqueue from any core will
try to do re-assignment of inq - outq mapping which will fail.
In this patch, for each core a separate inq is allocated and
used from each of the core. But the number of outq will remain
the same and we save the pointer to outq in session for each
of the cores.
examples/ipsec-secgw: add scripts for functional test
The purpose of these scripts is to automate ipsec-secgw functional testing.
The scripts require two machines (SUT and DUT) connected through
at least 2 NICs and running linux (so far tested only on Ubuntu 18.04).
Introduced test-cases for the following scenarios:
- Transport/Tunnel modes
- AES-CBC SHA1
- AES-GCM
- ESN on/off
- legacy/librte_ipsec code path
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Radu Nicolau <radu.nicolau@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
examples/ipsec-secgw: make data-path to use IPsec library
Changes to make ipsec-secgw data-path code to utilize librte_ipsec library.
Note that right now by default current (non-librte_ipsec) code-path will
be used. User has to run application with new command-line option ('-l')
to enable new codepath.
Signed-off-by: Mohammad Abdul Awal <mohammad.abdul.awal@intel.com> Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Radu Nicolau <radu.nicolau@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
examples/ipsec-secgw: make app to use IPsec library
Changes to make ipsec-secgw to utilize librte_ipsec library.
That patch provides:
- changes in the related data structures.
- changes in the initialization code.
- new command-line parameters to enable librte_ipsec codepath
and related features.
Note that right now by default current (non-librte_ipsec) code-path will
be used. User has to run application with new command-line option ('-l')
to enable new codepath.
The main reason for that:
- current librte_ipsec doesn't support all ipsec algorithms
and features that the app does.
- allow users to run both versions in parallel for some time
to figure out any functional or performance degradation with the
new code.
It is planned to deprecate and remove non-librte_ipsec code path
in future releases.
Signed-off-by: Mohammad Abdul Awal <mohammad.abdul.awal@intel.com> Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Radu Nicolau <radu.nicolau@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
examples/ipsec-secgw: fix outbound codepath for single SA
Looking at process_pkts_outbound_nosp() there seems few issues:
- accessing mbuf after it was freed
- invoking ipsec_outbound() for ipv4 packets only
- copying number of packets, but not the mbuf pointers itself
that patch provides fixes for that issues.
Fixes: 906257e965b7 ("examples/ipsec-secgw: support IPv6") Cc: stable@dpdk.org Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Radu Nicolau <radu.nicolau@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
examples/ipsec-secgw: fix crypto-op might never get dequeued
In some cases crypto-ops could never be dequeued from the crypto-device.
The easiest way to reproduce:
start ipsec-secgw with crypto-dev and send to it less then 32 packets.
none packets will be forwarded.
Reason for that is that the application does dequeue() from crypto-queues
only when new packets arrive.
This patch makes the app to call dequeue() on a regular basis.
Also to make code cleaner and easier to understand,
it separates crypto-dev enqueue() and dequeue() code paths.
pkt_process() now only enqueues packets into crypto device,
dequeuing and final processing is done by drain_crypto_queues().
Fixes: c64278c0c18b ("examples/ipsec-secgw: rework processing loop") Cc: stable@dpdk.org Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Radu Nicolau <radu.nicolau@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
examples/ipsec-secgw: allow to specify neighbour MAC address
In some cases it is useful to allow user to specify destination
ether address for outgoing packets.
This patch adds such ability by introducing new 'neigh' config
file option.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Radu Nicolau <radu.nicolau@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
examples/ipsec-secgw: allow disabling some Rx/Tx offloads
Right now ipsec-secgw always enables TX offloads
(DEV_TX_OFFLOAD_MULTI_SEGS, DEV_TX_OFFLOAD_SECURITY),
even when they are not requested by the config.
That causes many PMD to choose full-featured TX function,
which in many cases is much slower then one without offloads.
That patch adds ability for the user to disable unneeded HW offloads.
If DEV_TX_OFFLOAD_IPV4_CKSUM is disabled by user, then
SW version of ip cksum calculation is used.
That allows to use vector TX function, when inline-ipsec is not
requested.
Signed-off-by: Remy Horton <remy.horton@intel.com> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Radu Nicolau <radu.nicolau@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Create functional test for librte_ipsec.
Note that the test requires null crypto pmd to pass successfully.
Signed-off-by: Mohammad Abdul Awal <mohammad.abdul.awal@intel.com> Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Declan Doherty <declan.doherty@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
ipsec: rework SA replay window/SQN for MT environment
With these changes functions:
- rte_ipsec_pkt_crypto_prepare
- rte_ipsec_pkt_process
can be safely used in MT environment, as long as the user can guarantee
that they obey multiple readers/single writer model for SQN+replay_window
operations.
To be more specific:
for outbound SA there are no restrictions.
for inbound SA the caller has to guarantee that at any given moment
only one thread is executing rte_ipsec_pkt_process() for given SA.
Note that it is caller responsibility to maintain correct order
of packets to be processed.
Provide implementation for rte_ipsec_pkt_crypto_prepare() and
rte_ipsec_pkt_process().
Current implementation:
- supports ESP protocol tunnel mode.
- supports ESP protocol transport mode.
- supports ESN and replay window.
- supports algorithms: AES-CBC, AES-GCM, HMAC-SHA1, NULL.
- covers all currently defined security session types:
- RTE_SECURITY_ACTION_TYPE_NONE
- RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO
- RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL
- RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL
For first two types SQN check/update is done by SW (inside the library).
For last two type it is HW/PMD responsibility.
Signed-off-by: Mohammad Abdul Awal <mohammad.abdul.awal@intel.com> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Declan Doherty <declan.doherty@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Introduce Security Association (SA-level) data-path API
Operates at SA level, provides functions to:
- initialize/teardown SA object
- process inbound/outbound ESP/AH packets associated with the given SA
(decrypt/encrypt, authenticate, check integrity,
add/remove ESP/AH related headers and data, etc.).
Signed-off-by: Mohammad Abdul Awal <mohammad.abdul.awal@intel.com> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Declan Doherty <declan.doherty@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Introduce librte_ipsec library.
The library is supposed to utilize existing DPDK crypto-dev and
security API to provide application with transparent IPsec processing API.
That initial commit provides some base API to manage
IPsec Security Association (SA) object.
Signed-off-by: Mohammad Abdul Awal <mohammad.abdul.awal@intel.com> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Declan Doherty <declan.doherty@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Mohammad Abdul Awal <mohammad.abdul.awal@intel.com> Acked-by: Declan Doherty <declan.doherty@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
security: add opaque userdata pointer into security session
Add 'uint64_t opaque_data' inside struct rte_security_session.
That allows upper layer to easily associate some user defined
data with the session.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Acked-by: Mohammad Abdul Awal <mohammad.abdul.awal@intel.com> Acked-by: Declan Doherty <declan.doherty@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Fan Zhang [Thu, 10 Jan 2019 14:50:21 +0000 (14:50 +0000)]
cryptodev: add reference count to session private data
This patch adds a refcnt field to every session private data in the
cryptodev symmetric session. The counter is used to prevent freeing
symmetric session blindly before it is not cleared by every type of
crypto device in use.
Fan Zhang [Thu, 10 Jan 2019 14:50:20 +0000 (14:50 +0000)]
cryptodev: add user data size to symmetric session
This patch adds a user_data_sz field to cryptodev symmetric session.
The field is used to check if reading or writing the session's user
data field is eligible.
Fan Zhang [Thu, 10 Jan 2019 14:50:19 +0000 (14:50 +0000)]
cryptodev: update symmetric session structure
This patch updates the rte_cryptodev_sym_session structure for
cryptodev library. The updates include a changed session private
data array and an added nb_drivers field. They are used to
calculate the correct session header size and ensure safe access
of the session private data.
Fan Zhang [Thu, 10 Jan 2019 14:50:17 +0000 (14:50 +0000)]
test/crypto: use separate session mempools
This patch uses the two session mempool approach to crypto unit
test. One mempool is for session header objects, and the other is
for session private data.
Fan Zhang [Thu, 10 Jan 2019 14:50:16 +0000 (14:50 +0000)]
vhost/crypto: use separate session mempools
This patch uses the two session mempool approach to vhost crypto.
One mempool is for session header objects, and the other is for
session private data.
Fan Zhang [Thu, 10 Jan 2019 14:50:15 +0000 (14:50 +0000)]
examples: use separate crypto session mempools
This patch uses the two session mempool approach to all cryptodev
sample applications. One mempool is for session header objects, and
the other is for session private data.
Fan Zhang [Thu, 10 Jan 2019 14:50:13 +0000 (14:50 +0000)]
app/crypto-perf: use separate session mempools
This patch uses the two session mempool approach to crypto perf
application. One mempool is for session header objects, and the other
is for session private data.
Fan Zhang [Thu, 10 Jan 2019 14:50:12 +0000 (14:50 +0000)]
cryptodev: add sym session mempool create
This patch adds a new API "rte_cryptodev_sym_session_pool_create()" to
cryptodev library. All applications are required to use this API to
create sym session mempool as it adds private data and nb_drivers
information to the mempool private data.
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com> Acked-by: Fiona Trahe <fiona.trahe@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Fan Zhang [Thu, 20 Dec 2018 12:22:15 +0000 (12:22 +0000)]
crypto/aesni_mb: support plain SHA
This patch adds the plain SHA1, SHA224, SHA256, SHA384, and SHA512
algorithms support to AESNI-MB PMD. The cryptodev unit test and
documentation are updated accordingly.
Tomasz Jozwiak [Thu, 27 Dec 2018 10:42:19 +0000 (11:42 +0100)]
compress/qat: fix returned status on overflow
This patch fixes correct status in case of overflow on
QAT is detected.
In that case RTE_COMP_OP_STATUS_OUT_OF_SPACE_TERMINATED value is set in
rte_comp_op.status field instead of RTE_COMP_OP_STATUS_ERROR
Fixes: 32842f2a6d7d ("compress/qat: create FW request and process response") Cc: stable@dpdk.org Signed-off-by: Tomasz Jozwiak <tomaszx.jozwiak@intel.com> Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Fan Zhang [Thu, 20 Dec 2018 11:56:45 +0000 (11:56 +0000)]
doc: update AESNI_MB guide
This patch updates the AESNI-MB PMD document with the new intel-ipsec-mb
version number, the release note with new library version support, and
the deprecation notice for removing some library version support in
19.05 release.
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Fan Zhang [Thu, 20 Dec 2018 11:56:44 +0000 (11:56 +0000)]
crypto/aesni_mb: use architecture independent macros
This patch duplicates the original rte_aesni_mb_pmd*.c files and replaces
the function calls provided by intel-ipsec-mb library into
architecture-independent macros. The build systems are updated to choose
compiling either rte_aesni_mb_pmd*.c or rte_aesni_mb_pmd*_compat.c based
on the installed intel-ipsec-mb version. For the intel-ipsec-mb older
than 0.52.0 rte_aesni_mb_pmd*_compat.c will be compiled, otherwise
rte_aesni_mb_pmd*.c will be compiled.
Fan Zhang [Thu, 20 Dec 2018 11:56:43 +0000 (11:56 +0000)]
crypto/aesni_mb: rename files
This patch renames the rte_aesni_mb_pmd*.c to
rte_aesni_mb_pmd*_compat.c to indicate those files are for
compatible with older versoin of intel-ipsec-mb library
only.
Fan Zhang [Wed, 9 Jan 2019 12:11:12 +0000 (12:11 +0000)]
doc: fix AESNI_MB guide
This patch fixes the incorrect intel multi buffer library support
version and missed limitation of GCM algorithm support.
Fixes: 0e9f8507afcb ("crypto/aesni_mb: support AES-GCM algorithm") Cc: stable@dpdk.org Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Fan Zhang [Fri, 21 Dec 2018 14:11:05 +0000 (14:11 +0000)]
drivers/crypto: fix PMDs memory leak
This patch fixes the memory leak during queue pair release.
Originally the operation ring is not freed when releasing
queue pair, causing the next queue_pair configure call fail
and memory leak.
Fixes: eec136f3c54f ("aesni_gcm: add driver for AES-GCM crypto operations") Fixes: cf7685d68f00 ("crypto/zuc: add driver for ZUC library") Fixes: d61f70b4c918 ("crypto/libcrypto: add driver for OpenSSL library") Fixes: 3aafc423cf4d ("snow3g: add driver for SNOW 3G library") Fixes: 94b0ad8e0aa5 ("null_crypto: add driver for null crypto operations") Cc: stable@dpdk.org Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Lee Daly [Tue, 8 Jan 2019 16:20:43 +0000 (16:20 +0000)]
test/compress: add checksum tests
This patch adds a test which examines what type of checksum the PMD
supports, Adler, CRC32 or Adler32_CRC32
and tests that feature if the PMD supports it.
Signed-off-by: Lee Daly <lee.daly@intel.com> Acked-by: Tomasz Jozwiak <tomaszx.jozwiak@intel.com> Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Vipin Varghese [Thu, 10 Jan 2019 21:36:45 +0000 (03:06 +0530)]
app/procinfo: add --iter-mempool
Function iter_mempool is used for iterating slected mempool elements
and display the contents of elements for a max of 256 bytes. In case
of invalid or no name for mempool, no information is displayed.
Vipin Varghese [Thu, 10 Jan 2019 21:36:40 +0000 (03:06 +0530)]
app/procinfo: add --show-port
Function show_port is used for displaying the port PMD information under
primary process. The information shows basic, per queue and security.
Also helper MACRO 'MAX_STRING_LEN, STATS_BDR_FMT and STATS_BDR_STR' and
meson build file is updated for new functionality.
Add support to run the Soft NIC device on service cores, based on the
value of the sc parameter:
0: The current Soft NIC device is run explicitly by the application. The
firmware creates one or several pipelines for this dev and maps them
to CPU cores that should not be service cores. The app is required to
call rte_pmd_softnic_run() to make this dev work.
1 (default): The current device is run on the service cores transparently
to the app. The firmware creates one or several pipelines for this dev
current device and maps them to CPU cores that should be service cores.
Each of these service cores is calling rte_pmd_softnic_run()
in order to make this dev work with no app intervention.
Gavin Hu [Fri, 21 Dec 2018 06:02:56 +0000 (14:02 +0800)]
devtools: fix wrong headline lowercase for arm
Change to the new wording according to Arm corp:
https://mails.dpdk.org/archives/dev/2018-October/115174.html
otherwise, "Wrong headline lowercase" may be falsely reported.
Fixes: f146ada21d1e ("scripts: add more git log checks") Cc: stable@dpdk.org Signed-off-by: Gavin Hu <gavin.hu@arm.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Jerin Jacob [Thu, 27 Dec 2018 06:06:48 +0000 (06:06 +0000)]
net/ice: fix build with debug enabled
When RTE_LIBRTE_MBUF_DEBUG enabled, rte_mbuf_sanity_check()
function defined in rte_mbuf.so library,
add it while linking the librte_pmd_ice.so library to fix
the build issue.
error log:
/usr/bin/ld: ice_rxtx.o: in function `ice_recv_pkts':
ice_rxtx.c:(.text+0x1d97): undefined reference to
`rte_mbuf_sanity_check'
/usr/bin/ld: ice_rxtx.o: in function
`ice_tx_queue_release_mbufs':
ice_rxtx.c:(.text+0x21e0): undefined reference to
`rte_mbuf_sanity_check'
/usr/bin/ld: ice_rxtx.c:(.text+0x22b3): undefined reference
to `rte_mbuf_sanity_check'
/usr/bin/ld: ice_rxtx.c:(.text+0x24a0): undefined reference
to `rte_mbuf_sanity_check'
/usr/bin/ld: ice_rxtx.c:(.text+0x2640): undefined reference
to `rte_mbuf_sanity_check'
/usr/bin/ld: ice_rxtx.o:ice_rxtx.c:(.text+0x299f):
more undefined references to `rte_mbuf_sanity_check' follow
Fixes: 50370662b727 ("net/ice: support device and queue ops") Signed-off-by: Jerin Jacob <jerinj@marvell.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Rosen Xu [Mon, 26 Nov 2018 08:35:57 +0000 (16:35 +0800)]
app/testpmd: expand RED queue thresholds to 64 bits
There's very commonly that more than 4G DDR memory in NIC for HQoS,
so right now the queue threshold size of RED needs to expand to
uint64_t. For struct rte_tm_red_params, it has been fixed, but for
test-pmd TM configuration, it hasn't been fixed. This patch fixes it.
Fixes: bddc2f40b594 ("app/testpmd: add commands for shaper and wred profiles") Cc: stable@dpdk.org Signed-off-by: Rosen Xu <rosen.xu@intel.com> Reviewed-by: Jasvinder Singh <jasvinder.singh@intel.com>
Reshma Pattan [Thu, 20 Dec 2018 12:16:09 +0000 (12:16 +0000)]
mbuf: implement generic format for sched field
This patch implements the changes proposed in the deprecation
notes [1][2].
librte_mbuf changes:
The mbuf->hash.sched field is updated to support generic
definition in line with the ethdev traffic manager and meter APIs.
The new generic format contains: queue ID, traffic class, color.
Added public APIs to set and get these new fields to and from mbuf.
librte_sched changes:
In addtion, following API functions of the sched library have
been modified with an additional parameter of type struct
rte_sched_port to accommodate the changes made to mbuf sched field.
(i)rte_sched_port_pkt_write()
(ii) rte_sched_port_pkt_read_tree_path()
librte_pipeline, qos_sched UT, qos_sched app are updated
to make use of new changes.
Also mbuf->hash.txadapter has been added for eventdev txq,
rte_event_eth_tx_adapter_txq_set and rte_event_eth_tx_adapter_txq_get()
are updated to use mbuf->hash.txadapter.txq.
doc:
Release notes updated.
Removed deprecation notice for mbuf->hash.sched and sched API.
Reshma Pattan [Thu, 20 Dec 2018 12:16:08 +0000 (12:16 +0000)]
meter: unify packet color definition
Added new rte_color definition in librte_meter to
consolidate color definition which is currently replicated
in various places such as rte_meter.h, rte_tm.h and rte_mtr.h
Created aliases for rte_tm_color, rte_mtr_color and rte_meter_color
to use new rte_color values.
The definitions of rte_tm_color, rte_mtr_color and rte_meter_color
will be deprecated in future.
Bruce Richardson [Wed, 19 Dec 2018 11:59:50 +0000 (11:59 +0000)]
telemetry: fix using ports of different types
Different NIC ports can have different numbers of xstats on them, which
means that we can't just use the xstats list from the first port registered
in the telemetry library. Instead, we need to check the type of each port -
by checking its ops structure pointer - and register each port type once
with the metrics lib.
Fixes: fdbdb3f9ce46 ("telemetry: add initial connection socket") Cc: stable@dpdk.org Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Kevin Laatz <kevin.laatz@intel.com>
Tiago Lam [Tue, 20 Nov 2018 10:26:29 +0000 (10:26 +0000)]
net/af_packet: fix setting MTU decrements sockaddr twice
When setting the MTU, eth_dev_mtu_set() is called to validate the
provided MTU. As part of that, it calculates the useful area to store
data and compares it against the MTU, to guarantee that there's enough
space to store the data. It calculates that as:
"tp_frame_size - TPACKET2_HDRLEN - sizeof(struct sockaddr_ll)"
However, the TPACKET2_HDRLEN macro already increments sizeof(struct
sockaddr_ll) internally, meaning the useuful area of data above will
have sizeof(struct sockaddr_ll) decremented twice.
Instead, the useful area of data should be calculated as:
"tp_frame_size - TPACKET2_HDRLEN"
This makes sure that there's enough useful area to fit the provided MTU
after excluding tpacket2_hdr and sockaddr_ll.
Fixes: cc68ac4847bc ("net/af_packet: support MTU change") Cc: stable@dpdk.org Signed-off-by: Tiago Lam <tiago.lam@intel.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Maxime Coquelin [Thu, 20 Dec 2018 16:47:55 +0000 (17:47 +0100)]
vhost: batch used descs chains write-back with packed ring
Instead of writing back descriptors chains in order, let's
write the first chain flags last in order to improve batching.
Also, move the write barrier in logging cache sync, so that it
is done only when logging is enabled. It means there is now
one more barrier for split ring when logging is enabled.
With Kernel's pktgen benchmark, ~3% performance gain is measured.
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
There is a control dependency between steps 1 and step 2.
2 could be speculatively executed before 1, which could result
in 'id' to not be updated yet.
Fixes: 2f3225a7d69b ("vhost: add vector filling support for packed ring") Cc: stable@dpdk.org Reported-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com> Acked-by: Ilya Maximets <i.maximets@samsung.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Tiwei Bie <tiwei.bie@intel.com>
There is a control dependency between step 1 and steps 3 & 4,
3 could be speculatively executed before 1, which could result
in 'idx' to not being updated yet.
Fixes: 4796ad63ba1f ("examples/vhost: import userspace vhost application") Cc: stable@dpdk.org Reported-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com> Acked-by: Ilya Maximets <i.maximets@samsung.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Tiwei Bie <tiwei.bie@intel.com>
The common data freeing has been moved to rte_eth_dev_release_port(),
so freeing mac_addrs like this in eth_dev_close() is unnecessary and
will cause double free.
Fixes: e16adf08e54d ("ethdev: free all common data when releasing port") Cc: stable@dpdk.org Signed-off-by: Hideyuki Yamashita <yamashita.hideyuki@po.ntt-tx.co.jp> Signed-off-by: Yasufumi Ogawa <ogawa.yasufumi@lab.ntt.co.jp> Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
The failsafe driver spews messages every time it is polling
making debugging other things more difficult. Just remove it.
Example:
net_failsafe: Hot-plug mutex was locked by thread 140182066431744 by the hot-plug alarm
net_failsafe: Hot-plug mutex was locked by thread 140182066431744 by the hot-plug alarm
...
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Bruce Richardson [Mon, 17 Dec 2018 15:50:05 +0000 (15:50 +0000)]
net/tap: add buffer overflow checks before checksum
The checksum calculation APIs take only the packet headers pointers as
parameters, so they assume that the lengths reported in those headers
are correct. However, a malicious packet could claim to be far larger
than it is, so we need to check the header lengths in the driver before
calling the checksum API.
A better fix would be to allow the lengths to be passed into the API
function, but that would be an API break, so fixing in TAP driver for
now.
Bruce Richardson [Mon, 17 Dec 2018 15:50:04 +0000 (15:50 +0000)]
net: fix underflow for checksum of invalid IPv4 packets
If we receive a packet with an invalid IP header, where the total packet
length is reported as less than the IP header length, we would end up
getting an underflow in the length subtraction.
This could cause us to checksum e.g. 4GB of data in the case where the
result of the subtraction was -1.
We fix this by having the function return 0 - an invalid sum - when
the length is less than the header length.
Fixes: af75078fece3 ("first public release") Fixes: 6006818cfb26 ("net: new checksum functions") Cc: stable@dpdk.org Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Yongseok Koh [Wed, 12 Dec 2018 11:11:32 +0000 (03:11 -0800)]
net/mlx5: fix Multi-Packet RQ mempool free
When MPRQ mempool is freed, the pointer stored in priv structure must be
reset to null. Otherwise, the mempool can be freed again if the port is
restarted.
Zhirun Yan [Thu, 20 Dec 2018 11:01:29 +0000 (11:01 +0000)]
net/i40e: support PF respond VF request more queues
This patch respond the VIRTCHNL_OP_REQUEST_QUEUES msg from VF, and
process to allocated more queues for the requested VF. If successful,
PF will notify VF to reset. If unsuccessful, PF will send message to
inform VF.
Signed-off-by: Zhirun Yan <zhirun.yan@intel.com> Signed-off-by: Haiyue Wang <haiyue.wang@intel.com> Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Zhirun Yan [Thu, 20 Dec 2018 11:01:28 +0000 (11:01 +0000)]
net/i40e: support VF request more queues
Before this patch, VF gets a default number of queues from the PF.
This patch enables VF to request a different number. When VF configures
more queues, it will send VIRTCHNL_OP_REQUEST_QUEUES to PF to request
more queues, if success, PF will reset the VF.
User can run "port stop all", "port config port_id rxq/txq queue_num"
and "port start all" to reconfigure queue number.
Signed-off-by: Zhirun Yan <zhirun.yan@intel.com> Signed-off-by: Haiyue Wang <haiyue.wang@intel.com> Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Rahul Lakkireddy [Wed, 19 Dec 2018 16:28:26 +0000 (21:58 +0530)]
net/cxgbe: fix other misc build issues for Windows
Fix following build errors reported by Intel C++ compiler in Windows
build.
C:\> t4_hw.c(5105): warning #147: declaration is incompatible with
"int t4_bar2_sge_qregs(struct adapter *, unsigned int, unsigned int,
u64={uint64_t={unsigned __int64}} *, unsigned int *)"
(declared at line 524 of "..\..\..\..\drivers\net\cxgbe\base\common.h")
int t4_bar2_sge_qregs(struct adapter *adapter, unsigned int qid,
^
C:\> sge.c(400): error : expression must be a pointer to a complete
object type
(uint16_t)(RTE_PTR_ALIGN((char *)mbuf->buf_addr +
^
Build Environment:
1. Target OS: Microsoft Windows Server 2016
2. Compiler: Intel C++ Compiler from Intel Parallel Studio XE 2019 [1]
3. Development Tools:
3.1 Microsoft Visual Studio 2017 Professional
3.2 Windows Software Development Kit (SDK) v10.0.17763
3.3 Windows Driver Kit (WDK) v10.0.17763
git.droids-corp.org / dpdk.git / log