AgeCommit message (Collapse)Author
2019-05-13doc: update release notes for 19.05HEADwindpdk-next-devmasterJohn McNamara
Fix grammar, spelling and formatting of DPDK 19.05 release notes. Signed-off-by: John McNamara <>
2019-05-13doc: update Mellanox guides and release notesOri Kam
This patch adds some missing features to Mellanox drivers release notes. It also updates the mlx5/mlx4 documentations. Fixes: d85b204b5dba ("doc: update release notes for Mellanox drivers") Signed-off-by: Ori Kam <> Acked-by: Shahaf Shuler <>
2019-05-13doc: fix JSON interface for power sampleLukasz Krakowiak
Updated doc for JSON sample code related to vm_power_manager fifo interface: "command": "destroy", "command": "power". There is no code change to go with this doc update, it is to fix the docs to match the implementation in the code. Fixes: a63504a90f ("examples/power: add JSON string handling") Cc: Signed-off-by: Lukasz Krakowiak <> Acked-by: David Hunt <>
2019-05-13doc: fix doxygen for timer APIErik Gabriel Carrillo
Now that some of the symbols in the timer lib are versioned, the Doxygen documentation that is generated is incorrect. Group all versioned symbols, listing the generic name first, and remove comments for older versions of symbols. Fixes: c0749f7096c7 ("timer: allow management in shared memory") Signed-off-by: Erik Gabriel Carrillo <>
2019-05-13app/testpmd: fix offload flags after port configWei Zhao
There is an error in function rxtx_port_config(), which may overwrite offloads configuration get from function launch_args_parse() when run testpmd app. So rxtx_port_config() should do "or" for port offloads. Fixes: d44f8a485f5d ("app/testpmd: enable per queue configure") Cc: Signed-off-by: Wei Zhao <> Tested-by: Peng Yuan <>
2019-05-12net/mlx5: fix device removal handler for multiportViacheslav Ovsiienko
IBV_EVENT_DEVICE_FATAL event is generated by the driver once for the entire multiport Infiniband device, not for each existing ports. The port index is zero and it causes dropping the device removal event. We should invoke the removal event processing routine for each port we have installed handler for. Fixes: 028b2a28c3cb ("net/mlx5: update event handler for multiport IB devices") Signed-off-by: Viacheslav Ovsiienko <> Acked-by: Shahaf Shuler <>
2019-05-13event/dsw: ignore scheduling type for single-link queuesMattias Rönnblom
The scheduling type parameter is not applicable for single link queues. DSW would, at the time of rte_event_queue_setup(), erroneously verify that scheduling type was one of the supported types, and returned -ENOTSUP in case of RTE_SCHED_TYPE_ORDERED. Fixes: 4540ee9c68 ("event/dsw: add device and queue configuration") Cc: Signed-off-by: Mattias Rönnblom <> Acked-by: Gage Eads <>
2019-05-13eventdev: fix Rx adapter event flushNikhil Rao
The Rx adapter flushes events only if it has BATCH_SIZE events buffered where BATCH_SIZE is set to 32, e.g., if a single packet is sent, it is never passed to eventdev. Fix this issue by adding an event buffer flush either when a Rx queue is found to be empty or the adapter service function has processed the max number of packets for an invocation. Bugzilla ID: 277 Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter") Cc: Reported-by: Matias Elo <> Signed-off-by: Nikhil Rao <> Reviewed-by: Mattias Rönnblom <> Tested-by: Matias Elo <>
2019-05-10devtools: select patches to check with git rangeDavid Marchand
Rather than default to origin/master.., it can be handy to choose the range you want to check. Example on a branch rebased on next-net: Before: $ ./devtools/ ... ... 67/69 valid patches After: $ ./devtools/ -r next-net/master.. 3/3 valid patches Signed-off-by: David Marchand <> Acked-by: Thomas Monjalon <>
2019-05-10version: 19.05-rc4Thomas Monjalon
Signed-off-by: Thomas Monjalon <>
2019-05-10doc: update recommended kernel version for i40eLijuan Tu
add recommended DPDK/kernel driver/firmware version matching list for i40e for 19.05 Signed-off-by: Lijuan Tu <> Acked-by: John McNamara <>
2019-05-10doc: add tested Intel platforms with Intel NICsLijuan Tu
Add tested Intel platforms with Intel NICs to v19.05 release note. Signed-off-by: Lijuan Tu <> Acked-by: John McNamara <>
2019-05-10doc: add tested platforms with Mellanox NICsRaslan Darawsheh
Signed-off-by: Raslan Darawsheh <>
2019-05-10doc: add Mellanox BlueField platform guideYongseok Koh
Platform specific guide for Mellanox BlueField SoC is added. Signed-off-by: Yongseok Koh <> Acked-by: John McNamara <>
2019-05-10doc: update release notes for Mellanox driversYongseok Koh
mlx4/mlx5 release note for v19.05 Signed-off-by: Yongseok Koh <>
2019-05-10doc: update release notes for QAT PMDsFiona Trahe
Added release note entry for QAT compression PMD Clarified that previous entry was for QAT sym crypto PMD Signed-off-by: Fiona Trahe <> Signed-off-by: Tomasz Jozwiak <> Acked-by: John McNamara <>
2019-05-10doc: remove redundant information from debug guideVipin Varghese
Remove redundant information from section Performance issue isolation. Re-word for section header for packet capture. Fixes: 08db7bde1617 ("doc: add guide for debug and troubleshoot") Signed-off-by: Vipin Varghese <> Acked-by: John McNamara <>
2019-05-10doc: fix formatting in testpmd guideKevin Traynor
Minor formatting error related to code block noticed when reading the doc. Fix it and some other errors. Fixes: c7217b9dd8ef ("app/testpmd: change log level at run time") Fixes: 3c272b280a50 ("app/testpmd: add commands for RSS queue region") Fixes: e38ea44f1714 ("app/testpmd: add configuration for input set") Fixes: 08e0b3440baf ("app/testpmd: add option to configure UDP tunnel port") Fixes: e977e4199a8d ("app/testpmd: add commands to load/unload BPF filters") Cc: Signed-off-by: Kevin Traynor <> Acked-by: Bernard Iremonger <>
2019-05-10devtools: fix check of symbol added as stable APIDavid Marchand
The incriminated commit broke the detection of new symbols skipping the EXPERIMENTAL step before entering a stable ABI section. sed won't return an error, check a null output instead. Fixes: 3630757803ab ("devtools: accept experimental symbol promotion") Cc: Signed-off-by: David Marchand <>
2019-05-10devtools: fix symbol name in check logDavid Marchand
We have an incorrect variable name in this log. Fixes: 4bec48184e33 ("devtools: add checks for ABI symbol addition") Cc: Signed-off-by: David Marchand <>
2019-05-10devtools: skip warning for reordered symbolsDavid Marchand
No need to shout when we are just reordering symbols in a section. Signed-off-by: David Marchand <>
2019-05-10examples/ipsec-secgw: fix build error logMarcin Smoczynski
Fix invalid indentation - extra whitespace before error directive which is causing syntax error when no pkgconfig file for the DPDK is found and RTE_SDK is not specified. Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample application") Cc: Signed-off-by: Marcin Smoczynski <>
2019-05-10doc: add timer library changes to release notesErik Gabriel Carrillo
Signed-off-by: Erik Gabriel Carrillo <> Acked-by: John McNamara <>
2019-05-10timer: allow first subsystem init from secondaryErik Gabriel Carrillo
Since memzones can be reserved from secondary processes as well as primary processes, if the first call to the timer subsystem init function occurs in a secondary process, we should allow it to succeed. Fixes: c0749f7096c7 ("timer: allow management in shared memory") Signed-off-by: Erik Gabriel Carrillo <>
2019-05-09test/hash: check freeing key with positionDharmik Thakkar
This patch adds a unit test for rte_hash_free_key_with_position(). Suggested-by: Linfan <> Signed-off-by: Dharmik Thakkar <> Acked-by: Yipeng Wang <>
2019-05-09hash: fix total entries countDharmik Thakkar
In rte_hash, with current implementation, it is possible that keys are stored at indexes greater than the number of total entries. Currently, in rte_hash_free_key_with_position(), due to incorrect computation of total_entries, application cannot free keys with indexes greater than the number of total entries. This patch fixes this incorrect computation of total_entries. Bugzilla ID: 261 Fixes: 9d033dac7d7c ("hash: support no free on delete") Cc: Reported-by: Linfan <> Suggested-by: Linfan <> Signed-off-by: Dharmik Thakkar <> Acked-by: Yipeng Wang <>
2019-05-09hash: fix position returned in free slotsDharmik Thakkar
Currently, in rte_hash_free_key_with_position(), the position returned to the ring of free_slots leads to an unexpected conflict with a key already in use. This patch fixes incorrect position returned to the ring of free_slots. Bugzilla ID: 261 Fixes: 9d033dac7d7c ("hash: support no free on delete") Cc: Reported-by: Linfan <> Suggested-by: Linfan <> Signed-off-by: Dharmik Thakkar <> Acked-by: Yipeng Wang <>
2019-05-09test/barrier: fix for Power CPUsDavid Christensen
The memory barrier test fails on IBM Power 9 systems. Add additional barriers to accommodate the weakly ordered model used on Power CPUs. Fixes: 93da5b59afc9 ("test: introduce memory barrier test case") Cc: Signed-off-by: David Christensen <> Acked-by: Konstantin Ananyev <>
2019-05-09test/barrier: enlarge variables to 64 bitsDavid Christensen
Memory barrier failures can be intermittent. Increase the size of the sum/val/iteration variables to allow tests that can run for days so that sporadic errors can be identified. Signed-off-by: David Christensen <> Acked-by: Konstantin Ananyev <>
2019-05-09test/barrier: fix allocation checkDavid Christensen
Code tested calloc failures for pt & lpt variables but not for the sum variable. Add a test for calloc failure of sum. Fixes: 93da5b59afc9 ("test: introduce memory barrier test case") Cc: Signed-off-by: David Christensen <> Acked-by: Konstantin Ananyev <>
2019-05-09test/barrier: fix typo in logDavid Christensen
Change "much" to "match" in a printf. Fixes: 93da5b59afc9 ("test: introduce memory barrier test case") Cc: Signed-off-by: David Christensen <> Acked-by: Konstantin Ananyev <>
2019-05-09test: call timer subsystem finalize at exitErik Gabriel Carrillo
The eal_flags_autotest checks that no hugepage map files are left behind after a process exits, which can only be the case if all allocations made from DPDK heaps were freed back to the freelist, resulting in the hugepage map files being unlinked automatically. Add a call to rte_timer_subsystem_finalize() at application exit time to release allocations now made by the timer library. Fixes: c0749f7096c7 ("timer: allow management in shared memory") Signed-off-by: Erik Gabriel Carrillo <>
2019-05-09power: fix cache line alignmentMattias Rönnblom
The ACPI and PState CPU frequency scaling drivers used the __rte_cache_aligned attribute without including rte_memory.h, which turns what looks as the declaration of a cache line-aligned struct into a non-aligned struct declaration and the definition of an instance of the struct. Fixes: e6c6dc0f96 ("power: add p-state driver compatibility") Fixes: 445c6528b5 ("power: common interface for guest and host") Cc: Signed-off-by: Mattias Rönnblom <>
2019-05-09power: fix resource leakLiang Ma
Fix the resource leaking issue Coverity issue: 337668 Fixes: b60fd5f8b1ce8f0a2c ("power: add bit for high frequency cores") Signed-off-by: Liang Ma <> Tested-by: David Hunt <>
2019-05-09ipc: add warnings about correct API usageAnatoly Burakov
When handling synchronous or asynchronous requests, the reply must be sent explicitly even if the result of the operation is an error, to avoid the other side timing out. Make note of this in documentation explicitly. Cc: Signed-off-by: Anatoly Burakov <>
2019-05-09ipc: add warnings about not using IPC with memory APIAnatoly Burakov
IPC and memory-related API's should not be mixed because memory relies on IPC internally. Add explicit warnings to IPC API and to the documentation about this. Cc: Signed-off-by: Anatoly Burakov <>
2019-05-09doc: fix typo in IPC guideAnatoly Burakov
The word "synchronous" appears twice. Fix it. Fixes: e22266669e86 ("doc: add IPC guide") Cc: Signed-off-by: Anatoly Burakov <> Acked-by: John McNamara <>
2019-05-09ipc: unlock on failureAaron Conole
Coverity issue: 340076 Fixes: a2a06860b8c4 ("ipc: fix memory leak on request failure") Cc: Signed-off-by: Aaron Conole <> Reviewed-by: David Marchand <> Tested-by: Darek Stojaczyk <> Acked-by: Anatoly Burakov <>
2019-05-09test/crypto: return not supported instead of skippedAyuj Verma
Currently some tests return TEST_SKIPPED/-1 when tests or params are not supported for particular PMD because of which tests adds to FAILED test counter in place of Skipped/Unsupported counter. Since unsupported test is not a failure case, replace return value TEST_SKIPPED/-1 with -ENOTSUP - Return -ENOTSUP for unsupported tests - add NULL check for rte_cryptodev_asym_capability_get() - Typo correction Signed-off-by: Ayuj Verma <> Signed-off-by: Shally Verma <>
2019-05-09ipsec: fix headers installMarcin Smoczynski
Invalid statement is used to indicate header files to install. Fixed the statement and reformatted recipe file. Signed-off-by: Marcin Smoczynski <> Acked-by: Bruce Richardson <>
2019-05-08doc: fix typo in mlx5 guideYongseok Koh
Fixes: 43e9d9794cde ("net/mlx5: support upstream rdma-core") Cc: Signed-off-by: Yongseok Koh <> Reviewed-by: Ferruh Yigit <>
2019-05-08app/testpmd: fix mbuf leak with multi-segment TxAndrew Rybchenko
The last mbuf allocated in bulk is never used and never freed. Fixes: 01b645dcff7f ("app/testpmd: move txonly prepare in separate function") Fixes: 561ddcf8d099 ("app/testpmd: allocate txonly segments per bulk") Signed-off-by: Andrew Rybchenko <> Reviewed-by: Ferruh Yigit <>
2019-05-08net/mlx5: fix errno for unsupported Direct RulesOri Kam
When Direct Rules API is not supported we don't set the errno. This results in failing the function but with errno equals to zero. The result of this is that a function that failed, is considered as a function that worked correctly. This commit fixes this issue by setting the errno to ENOTSUP and returning this error when error value should be returned. Since RDMA-CORE are returning positive errno we are also returning positive error values. Fixes: 4f84a19779ca ("net/mlx5: add Direct Rules API") Signed-off-by: Ori Kam <> Acked-by: Shahaf Shuler <>
2019-05-08net/mlx5: fix release of Rx queue objectDekel Peled
Function mlx5_rx_intr_disable() calls mlx5_rxq_ibv_get() and performs some actions on the returned rxq_ibv. It doesn't release the rxq_ibv when all is completed with success. This patch adds call to mlx5_rxq_ibv_release() where it's missing. Fixes: 09cb5b581762 ("net/mlx5: separate DPDK from verbs Rx queue objects") Cc: Signed-off-by: Dekel Peled <> Acked-by: Shahaf Shuler <>
2019-05-08net/mlx5: fix Multi-Packet RQ mempool nameYongseok Koh
Currently, the name of MPRQ mempool is set by snprintf(name, sizeof(name), "%s-mprq", dev->device->name); For port representor, the name is duplicate of its master and failed to create such a mempool having the same name. Port ID is used in the name instead. Fixes: 7d6bf6b866b8 ("net/mlx5: add Multi-Packet Rx support") Cc: Signed-off-by: Yongseok Koh <> Acked-by: Shahaf Shuler <>
2019-05-08net/mlx5: fix init with zero Rx queueDekel Peled
Recent patch [1] added, at the end of mlx5_dev_configure(), a call to mlx5_proc_priv_init(), initializing process_private data of eth_dev. This call is not reached if PMD is started with zero Rx queues. In this case mlx5_dev_configure() returns earlier due to the check: if (rxqs_n == priv->rxqs_n) return 0; In such a scenario, later references to uninitialized process_private data will result in segmentation fault. For example see in function txq_uar_init(). This patch changes the check logic. The following code is executed if (rxqs_n != priv->rxqs_n), and skipped otherwise. Function mlx5_proc_priv_init() is always invoked, to ensure process_private data is initialized. [1] Fixes: 120dc4a7dcd3 ("net/mlx5: remove device register remap") Cc: Signed-off-by: Dekel Peled <> Acked-by: Yongseok Koh <>
2019-05-07net/iavf: enable more link speedQi Zhang
Enable advanced link speed mode (VIRTCHNL_VF_CAP_ADV_LINK_SPEED) so iavf PMD can identify more link speed that reported by pf. Cc: Signed-off-by: Qi Zhang <> Acked-by: Wei Zhao <>
2019-05-07net/i40e: add device id for X710/XXV710 of ipn3keRosen Xu
New PCI device ids are created to support X710/XXV710 of Intel FPGA Programmable Acceleration card N3000, also called ipn3ke. Signed-off-by: Rosen Xu <> Acked-by: Qi Zhang <>
2019-05-07net/ice: fix EEPROM range checkXiao Wang
The last word should not cross shadow RAM boundary. Fixes: 68a1ab82ad74 ("net/ice: speed up to retrieve EEPROM") Cc: Signed-off-by: Xiao Wang <> Acked-by: Qi Zhang <>
2019-05-07net/ice: clean up minor issuesXiao Wang
This patch is a cleanup on comment, variable modifier, coding style. Signed-off-by: Xiao Wang <> Acked-by: Qi Zhang <>