Ferruh Yigit [Thu, 17 Nov 2016 13:27:21 +0000 (13:27 +0000)]
doc: announce KNI vhost removal
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Yuanhan Liu [Mon, 23 Jan 2017 13:04:42 +0000 (21:04 +0800)]
doc: announce API changes for vhost
I made a vhost ABI/API refactoring at v16.04, meant to avoid such issue
forever. Well, apparently, I lied.
People are looking for more vhost-user options now days, other than
vhost-user net only. For example, SPDK (Storage Performance Development
Kit) are looking for chance of vhost-user SCSI and vhost-user block.
Apparently, they also need a vhost-user backend, while DPDK already
has a (mature enough) backend, they don't want to implement it again
from scratch. They want to leverage the one DPDK provides.
However, the last refactoring hasn't done that right, at least it's
not friendly for extending vhost-user to add more devices support.
For example, different virtio devices has its own feature set, while
APIs like rte_vhost_feature_disable(feature_mask) have no option to
tell the device type. Thus, a more proper API should look like:
rte_vhost_feature_disable(device_type, feature_mask);
Besides that, few public files and structures should be renamed, to
not let it bind to virtio-net. Specifically, they are:
- virtio_net_device_ops --> vhost_device_ops
- rte_virtio_net.h --> rte_vhost.h
Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Bernard Iremonger [Thu, 5 Jan 2017 15:25:36 +0000 (15:25 +0000)]
doc: announce move of ethdev bypass function to ixgbe API
In 17.05, nine rte_eth_dev_* functions for bypass control,
and implemented only in ixgbe, will be removed from ethdev,
renamed and moved to the ixgbe PMD-specific API.
Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Thomas Monjalon [Mon, 13 Feb 2017 14:26:44 +0000 (15:26 +0100)]
doc: postpone API change in ethdev
The change of _rte_eth_dev_callback_process has not been done in 17.02.
Let's postpone to 17.05.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Shreyansh Jain [Mon, 13 Feb 2017 12:01:48 +0000 (17:31 +0530)]
doc: announce VFIO symbols renaming for export
Some vfio symbols need to be exported outside librte_eal.
For that, they need to be renamed to rte_* naming convention.
Signed-off-by: Shreyansh Jain <shreyansh.jain@nxp.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Thomas Monjalon [Mon, 13 Feb 2017 16:20:34 +0000 (17:20 +0100)]
doc: announce API changes to implement the bus model
The new bus model has been proposed in 17.02 without being used.
The big rework should happen in 17.05.
Suggested-by: Shreyansh Jain <shreyansh.jain@nxp.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Bruce Richardson [Mon, 13 Feb 2017 17:38:30 +0000 (17:38 +0000)]
doc: announce ABI change for ring library
Document proposed changes for the rings code in the next release.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Thomas Monjalon [Mon, 13 Feb 2017 10:52:32 +0000 (11:52 +0100)]
doc: announce TILE-Gx removal
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Dmitriy Yakovlev [Tue, 7 Feb 2017 02:51:06 +0000 (05:51 +0300)]
cfgfile: fix uninitialized variable on load error
Uninitialized scalar variable.
Using uninitialized value cfg->sections[curr_section]->num_entries
when calling rte_cfgfile_close.
And memory in variables cfg->sections[curr_section],
sect->entries[curr_entry] maybe not equal NULL.
We must decrement counters curr_section, curr_entry when failed to realloc.
Fixes:
eaafbad419bf ("cfgfile: library to interpret config files")
Signed-off-by: Dmitriy Yakovlev <bombermag@gmail.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Jianfeng Tan [Tue, 14 Feb 2017 11:55:34 +0000 (11:55 +0000)]
doc: fix images of virtio-user tutorials
The images by below two commits are very unclear. Fix it.
Fixes:
50665deebda ("doc: add guide to use virtio-user for container networking")
Fixes:
0ba3870e755 ("doc: add guide to use virtio-user as exceptional path")
Suggested-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Thomas Monjalon [Tue, 14 Feb 2017 13:57:37 +0000 (14:57 +0100)]
doc: introduce tested platforms as combinations
Add some text and rearrange lists to make sure it is clear that the
tested platforms listed in the release notes are some combinations
of the items in each group.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Shreyansh Jain [Mon, 13 Feb 2017 11:29:48 +0000 (16:59 +0530)]
doc: add EAL bus support in release notes
Signed-off-by: Shreyansh Jain <shreyansh.jain@nxp.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Yulong Pei [Tue, 14 Feb 2017 04:14:14 +0000 (12:14 +0800)]
doc: add tested Intel platforms in the release notes
Add tested Intel platforms with Intel NICs to the release note.
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Nelio Laranjeiro [Thu, 9 Feb 2017 08:32:01 +0000 (09:32 +0100)]
doc: update release notes for mlx5
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Nelio Laranjeiro [Thu, 9 Feb 2017 08:32:00 +0000 (09:32 +0100)]
doc: update release notes for mlx4
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Nelio Laranjeiro [Thu, 9 Feb 2017 08:31:59 +0000 (09:31 +0100)]
doc: merge test sections in release notes
These sections do not provide the exact tests that were done nor whether
specific NICs are supported by all platforms.
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Qi Zhang [Thu, 9 Feb 2017 19:59:43 +0000 (14:59 -0500)]
vfio: fix maximum number of interrupt for MSI-X
The max number of interrupt request is possible
be changed after rte_intr_callback_register, so
in get_max_intr, we need to check if necessary to
update the max_intr.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Andrew Rybchenko [Mon, 13 Feb 2017 10:32:23 +0000 (10:32 +0000)]
ethdev: fix typo in UDP tunnel API description
Fixes:
1cbe755fef47 ("ethdev: rename UDP tunnel port functions")
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
David Hunt [Mon, 6 Feb 2017 08:08:29 +0000 (08:08 +0000)]
doc: add distributor library API change notice
Given that the packet distributor library improvements (1) will
not be in 17.02, I plan on doing some consolidation of the
API for burst operation for 17.05, merging the two api's into
one, with options for single or burst operation.
(1) http://dpdk.org/dev/patchwork/patch/19911/
Signed-off-by: David Hunt <david.hunt@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Shreyansh Jain [Mon, 13 Feb 2017 11:55:44 +0000 (17:25 +0530)]
doc: remove deprecation notice for bus abstraction
Signed-off-by: Shreyansh Jain <shreyansh.jain@nxp.com>
Olivier Matz [Mon, 13 Feb 2017 11:05:26 +0000 (12:05 +0100)]
doc: postpone ABI changes to 17.05
Postpone the ABI changes for mempool and mbuf that were planned
for 17.02 to 17.05.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Thomas Monjalon [Mon, 13 Feb 2017 10:54:34 +0000 (11:54 +0100)]
doc: remove announce of Tx preparation
The feature is part of 17.02, so the ABI changes notice can be removed.
Fixes:
4fb7e803eb1a ("ethdev: add Tx preparation")
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Thomas Monjalon [Fri, 10 Feb 2017 16:15:32 +0000 (17:15 +0100)]
version: 17.02-rc3
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Pablo de Lara [Fri, 10 Feb 2017 12:33:42 +0000 (12:33 +0000)]
doc: clarify multi-buffer crypto library version support
AES-NI MB PMD uses external Multi-Buffer library,
which is hosted in github, but the version was not specified
in the documentation.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
Fan Zhang [Thu, 9 Feb 2017 18:46:21 +0000 (18:46 +0000)]
crypto/scheduler: fix session backup
Fixes the missed session backup during enqueue.
Fixes:
100e4f7e44ab ("crypto/scheduler: add round-robin mode")
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Fan Zhang [Thu, 9 Feb 2017 18:49:59 +0000 (18:49 +0000)]
crypto/scheduler: fix initialization
Fixes the wrong slave initialization issue on start-up
Fixes:
100e4f7e44ab ("crypto/scheduler: add round-robin mode")
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Slawomir Mrozowicz [Fri, 10 Feb 2017 14:22:30 +0000 (15:22 +0100)]
app/crypto-perf: fix null dereference
Dereferencing a pointer that might be null key_token when calling strstr.
Check if the pointer is null before.
Coverity issue: 141071
Fixes:
f8be1786b1b8 ("app/crypto-perf: introduce performance test application")
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Jacek Piasecki [Fri, 10 Feb 2017 13:26:15 +0000 (14:26 +0100)]
app/crypto-perf: fix size of expression
This commit fixes problem of passing a pointer to sizeof() function.
Now the size of enabled_cdevs structure is passed by RTE_CRYPTO_MAX_DEVS.
Coverity issue: 141068
Fixes:
f8be1786b1b8 ("app/crypto-perf: introduce performance test application")
Signed-off-by: Jacek Piasecki <jacekx.piasecki@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Slawomir Mrozowicz [Thu, 9 Feb 2017 13:57:29 +0000 (14:57 +0100)]
app/crypto-perf: fix crash for wireless algorithms
This commit fixes segmentation fault that happens when
using KASUMI, SNOW3G or ZUC authentication algorithms,
as they need authentication keys and was not being set.
It also set value of authentication key to cipher key
when using GMAC algorithm.
Fixes:
f8be1786b1b8 ("app/crypto-perf: introduce performance test application")
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Aleksander Gajewski [Fri, 10 Feb 2017 09:22:32 +0000 (10:22 +0100)]
app/crypto-perf: fix uninitialized variable
This commit ixes problem with uninitialized nb_cryptodevs variable by
initialize it with 0 value. Program could jump to err label
without running cperf_initialize_cryptodev() function. Also assign 0
value to nb_cryptodevs after cperf_initialize_cryptodev() when value is
negative.
Coverity issue: 141073
Fixes:
f8be1786b1b8 ("app/crypto-perf: introduce performance test application")
Signed-off-by: Aleksander Gajewski <aleksanderx.gajewski@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Pablo de Lara [Thu, 9 Feb 2017 12:27:45 +0000 (12:27 +0000)]
examples/l2fwd-crypto: fix padding
L2fwd-crypto app was padding an incoming buffer,
to be aligned with the algorithm block size, in all cases.
This was not the right approach, as padding is only necessary
when using block cipher algorithms, such as AES-CBC.
In case of using a stream cipher algorithm, such as SNOW3G UEA2,
there is no need to include padding and increase the buffer size.
Fixes:
387259bd6c67 ("examples/l2fwd-crypto: add sample application")
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Fan Zhang <roy.fan.zhang@intel.com>
Pablo de Lara [Tue, 7 Feb 2017 14:17:25 +0000 (14:17 +0000)]
examples/l2fwd-crypto: fix overflow
This commit fixes an array overflow when number of crypto devices
is higher than 32.
Fixes:
387259bd6c67 ("examples/l2fwd-crypto: add sample application")
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Fan Zhang <roy.fan.zhang@intel.com>
Jacek Piasecki [Wed, 8 Feb 2017 16:04:40 +0000 (17:04 +0100)]
app/crypto-perf: fix big parameter passed by value
Structure opts and structure test_vec are now passed by pointer to
the cperf_check_test_vector.
Coverity issue: 141072
Fixes:
f8be1786b1b8 ("app/crypto-perf: introduce performance test application")
Signed-off-by: Jacek Piasecki <jacekx.piasecki@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Pablo de Lara [Tue, 7 Feb 2017 22:49:58 +0000 (22:49 +0000)]
drivers/crypto: fix different auth/cipher keys
When ciphering and authenticating in the same operation
(cipher-then-auth or auth-then-cipher),
the cipher key and authentication key were set with the same
key, in SNOW3G, KASUMI and ZUC PMDs.
They were using the key of the first transform structure,
instead of using the keys of the two different transform
structures.
This is not a big issue, since usually, the same key is
used for ciphering and authentication, but keys may be different.
Fixes:
3aafc423cf4d ("snow3g: add driver for SNOW 3G library")
Fixes:
2773c86d061a ("crypto/kasumi: add driver for KASUMI library")
Fixes:
cf7685d68f00 ("crypto/zuc: add driver for ZUC library")
Cc: stable@dpdk.org
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Deepak Kumar Jain <deepak.k.jain@intel.com>
Fiona Trahe [Tue, 7 Feb 2017 16:32:55 +0000 (16:32 +0000)]
doc: add limitations section to cryptoperf guide
Add limitations to use of the dpdk-test-crypto-perf
tool for hardware accelerator measurements
Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Kuba Kozak [Tue, 7 Feb 2017 12:20:16 +0000 (13:20 +0100)]
app/crypto-perf: add options parsing check
Added total_ops value validation in parse_total_ops() function.
Coverity issue: 141070
Fixes:
f8be1786b1b8 ("app/crypto-perf: introduce performance test application")
Signed-off-by: Kuba Kozak <kubax.kozak@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Aleksander Gajewski [Tue, 7 Feb 2017 09:20:18 +0000 (10:20 +0100)]
app/crypto-perf: fix string not null terminated
This commit fixes the case where the string buffer may not have
a null terminator if the source string's length is equal to the
buffer size.
Coverity issue: 141069
Fixes:
f8be1786b1b8 ("app/crypto-perf: introduce performance test application")
Signed-off-by: Aleksander Gajewski <aleksanderx.gajewski@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Slawomir Mrozowicz [Fri, 3 Feb 2017 15:55:02 +0000 (16:55 +0100)]
cryptodev: fix crash when querying device by name
This patch fixes a segmentation fault in function
rte_cryptodev_devices_get(), due to incorrect driver name path.
It reworks the function to use correct types and clean up
for visibility.
Coverity issue: 141067
Fixes:
38227c0e3ad2 ("cryptodev: retrieve device info")
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Pablo de Lara [Fri, 27 Jan 2017 13:16:08 +0000 (13:16 +0000)]
crypto/aesni_mb: add missing digest length
AESNI MB now supports authentication only operations,
but the array containing all the possible digest lengths
was not updated to reflect the new algorithm NULL_HASH,
causing an "Out-of-bounds access".
Coverity issue: 140977
Fixes:
8772c3f713e2 ("crypto/aesni_mb: add single operation functionality")
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
Wenzhuo Lu [Fri, 10 Feb 2017 05:25:53 +0000 (13:25 +0800)]
net/i40e: fix TC bandwidth definition
The range of TC bandwidth is 0 ~ 800, it's 16bits not 8bits.
Fixes:
c8b9a3e3fe1b ("i40e: support DCB mode")
Cc: stable@dpdk.org
Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Wei Zhao [Fri, 10 Feb 2017 01:37:58 +0000 (09:37 +0800)]
net/ixgbe: fix VLAN mask TCI in flow rule parser
Use vlan_mask->tci as big endian since this is how rte flow defines it.
Fixes:
11777435c727 ("net/ixgbe: parse flow director filter")
Fixes:
37ed39b4e67d ("net/ixgbe: add TCI mask check for flow director")
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Jingjing Wu [Thu, 9 Feb 2017 06:41:54 +0000 (14:41 +0800)]
doc: add known issue for i40e statistics
Rx statistic is incorrect when packet is oversize.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Wenzhuo Lu [Thu, 9 Feb 2017 01:28:00 +0000 (09:28 +0800)]
doc: announce ixgbe MTU setting limitation
Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Wenzhuo Lu [Thu, 9 Feb 2017 01:27:59 +0000 (09:27 +0800)]
doc: fix titles of ixgbe guide
The framework of the ixgbe doc is not appropriate,
adjust it.
Fixes:
972e365bfe7a ("doc: nics guide")
CC: stable@dpdk.org
Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Shahaf Shuler [Thu, 9 Feb 2017 12:29:54 +0000 (14:29 +0200)]
net/mlx5: fix link status query
Trying to query the link status through the new ETHTOOL_GLINKSETTINGS
ioctl available since Linux 4.5 was always failing due to a kernel bug
fixed since version 4.9.
This commit also addresses a common issue where the headers version used
at compile time differs from that of the kernel on the target system, by
always defining missing symbols and moving the kernel version check at run
time.
Fixes:
188408719888 ("net/mlx5: fix support for newer link speeds")
CC: stable@dpdk.org
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Rasesh Mody [Thu, 9 Feb 2017 07:06:44 +0000 (23:06 -0800)]
net/bnx2x: increase release version
This patch bumps bnx2x PMD version to 1.0.5.1 to track the recent
fixes to the driver.
Signed-off-by: Rasesh Mody <rasesh.mody@cavium.com>
Ido Barnea [Thu, 2 Feb 2017 12:08:22 +0000 (12:08 +0000)]
net/vmxnet3: add speed capability
Signed-off-by: Ido Barnea <ibarnea@cisco.com>
Acked-by: Yong Wang <yongwang@vmware.com>
Jianbo Liu [Thu, 9 Feb 2017 04:05:26 +0000 (12:05 +0800)]
net/ixgbe: fix received packets number for ARM NEON
Vector PMD will check 4 descs in one time, but the statuses are not
consistent because the memory allocated for RX descriptors is cacheable
huagepage.
This patch is to calculate the number of received packets by scann DD bit
sequentially, and stops when meeting the first packet with DD bit unset.
Fixes:
b20971b6cca0 ("net/ixgbe: implement vector driver for ARM")
Cc: stable@dpdk.org
Signed-off-by: Jianbo Liu <jianbo.liu@linaro.org>
Jianbo Liu [Thu, 9 Feb 2017 04:05:25 +0000 (12:05 +0800)]
net/ixgbe: fix received packets number for ARM
To get better performance, Rx bulk alloc recv function will scan 8 descs
in one time, but the statuses are not consistent on ARM platform because
the memory allocated for Rx descriptors is cacheable hugepages.
This patch is to calculate the number of received packets by scan DD bit
sequentially, and stops when meeting the first packet with DD bit unset.
Fixes:
7431041062b9 ("ixgbe: allow rx bulk alloc")
Cc: stable@dpdk.org
Signed-off-by: Jianbo Liu <jianbo.liu@linaro.org>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
John Daley [Thu, 9 Feb 2017 00:40:09 +0000 (16:40 -0800)]
net/enic: fix hardcoding of some flow director masks
Hard coded mask values were being used for several of the IPv4 and IPv6
fields. Use the values in the rte_eth_fdir_masks structure provided by the
caller.
Fixes:
dfbd6a9cb504 ("net/enic: extend flow director support for 1300 series")
Cc: stable@dpdk.org
Signed-off-by: John Daley <johndale@cisco.com>
Jingjing Wu [Wed, 8 Feb 2017 05:05:53 +0000 (13:05 +0800)]
mbuf: fix bitmask of Tx offload flags
Add missed PKT_TX_MACSEC and PKT_TX_IEEE1588_TMST flags to bitmask of
all supported packet Tx offload features flags.
Fixes:
4fb7e803eb1a ("ethdev: add Tx preparation")
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Jingjing Wu [Wed, 8 Feb 2017 05:05:52 +0000 (13:05 +0800)]
net/e1000: fix bitmask of supported Tx flags
Add missed PKT_TX_IEEE1588_TMST to bitmask of all supported
packet Tx flags.
Fixes:
2b76648872c9 ("net/e1000: add Tx preparation")
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Jingjing Wu [Wed, 8 Feb 2017 05:05:51 +0000 (13:05 +0800)]
net/ixgbe: fix bitmask of supported Tx flags
Add missed PKT_TX_IEEE1588_TMST to bitmask of all supported
packet Tx flags.
Fixes:
7829b8d52be0 ("net/ixgbe: add Tx preparation")
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Jingjing Wu [Wed, 8 Feb 2017 05:05:50 +0000 (13:05 +0800)]
net/i40e: fix bitmask of supported Tx flags
PKT_TX_TUNNEL_MASK and PKT_TX_IEEE1588_TMST are missed in bitmask
of all supported packet Tx flags by i40e. It will cause packet preparing
fail when sending tunnel packets with Tx offload.
This patch fixes it.
Fixes:
3f33e643e5c6 ("net/i40e: add Tx preparation")
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Qiming Yang [Wed, 8 Feb 2017 09:14:24 +0000 (17:14 +0800)]
net/i40e: fix link update delay
Fix the redundant delay in function link update. There is no need to
call rte_delay_ms and hold CPU for 100ms when link status is up.
Fixes:
263333bbb7a9 ("i40e: fix link status timeout")
Cc: stable@dpdk.org
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Yongseok Koh [Thu, 2 Feb 2017 18:43:32 +0000 (10:43 -0800)]
net/mlx5: fix updating total length of multi-packet send
The total length field in descriptor of inlined multi-packet send must be
updated before closing a session. There's possibility of updating it
afterward. This bug might cause one packet out of MLX5_MPW_DSEG_MAX gets
silently dropped by HW and impact performance, especially lossless test.
Fixes:
230189d9ff22 ("net/mlx5: support multi-packet send")
Cc: stable@dpdk.org
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Wenzhuo Lu [Wed, 8 Feb 2017 14:28:12 +0000 (14:28 +0000)]
net/i40e: rework port check for driver-specific API
Change is_i40e_pmd to is_device_supported to make it more generic.
Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Tom Crugnale [Tue, 6 Dec 2016 20:16:21 +0000 (15:16 -0500)]
net/i40evf: fix reporting of imissed packets
Missed packets on RX were erroneously being assigned to the ierrors
struct member. Change it to be assigned to imissed.
Fixes:
4861cde46116 ("i40e: new poll mode driver")
Cc: stable@dpdk.org
Signed-off-by: Tom Crugnale <tcrugnale@sandvine.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Jeff Guo [Tue, 7 Feb 2017 03:52:19 +0000 (11:52 +0800)]
net/i40e: fix ethertype filter on X722
The GL_SWR_PRI_JOIN_MAP registers are effective on filters, changing
the register's default value will fail the ethertype filter.
The GL_SWR_PRI_JOIN_MAP values are different for each NIC, and current
X722 register values are wrong.
Fix X722 ethertype filter by setting registers to X722 default NVM
values.
Fixes:
92fbf2cbdff4 ("i40e: support X722 and its A0 hardware")
Cc: stable@dpdk.org
Signed-off-by: Jeff Guo <jia.guo@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Jeff Guo [Wed, 8 Feb 2017 09:50:16 +0000 (17:50 +0800)]
drivers/net: fix device configuration
dev_flags is wrongly overwritten with RTE_ETH_DEV_DETACHABLE value
in drivers after rte_eth_copy_pci_info().
Previous values of the dev_flags set in rte_eth_copy_pci_info(),
like RTE_ETH_DEV_INTR_LSC, are get lost. That will fail the device
configuration.
Fix by preventing dev_flags overwritten.
Fixes:
22dda618c00c ("pci: separate detaching ethernet ports from PCI devices")
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Signed-off-by: Jeff Guo <jia.guo@intel.com>
Beilei Xing [Mon, 6 Feb 2017 08:52:47 +0000 (16:52 +0800)]
net/i40e: fix tunnel filter
Creating IPv4 flow and IPv6 flow will cause conflict error.
Root cause is there's no IP info included in tunnel filter
input.
Fixes:
425c3325f0b0 ("net/i40e: store tunnel filter")
Fixes:
d416530e6358 ("net/i40e: parse tunnel filter")
Signed-off-by: Beilei Xing <beilei.xing@intel.com>
Beilei Xing [Mon, 6 Feb 2017 05:29:56 +0000 (13:29 +0800)]
net/i40e: fix parsing tunnel filter
VNI of VXLAN is parsed wrongly. The root cause is that
array VNI in item VXLAN uses network byte ordering.
Fixes:
d416530e6358 ("net/i40e: parse tunnel filter")
Signed-off-by: Beilei Xing <beilei.xing@intel.com>
Jakub Palider [Mon, 6 Feb 2017 11:56:56 +0000 (12:56 +0100)]
net/ena: fix setting host attributes
The hardware may reject adding host_info in case support for
host_info is missing in the list of supported features. On the
other hand the list of supported features may contain support
for the host_info - typical bootstrap problem.
This patch solves it by removing check against support for
host_info attribute and improves error handling by reacting
only to host attribute write failure to the hardware.
Fixes:
99ecfbf845b3 ("ena: import communication layer")
Cc: stable@dpdk.org
Signed-off-by: Jakub Palider <jpa@semihalf.com>
Wenzhuo Lu [Tue, 7 Feb 2017 06:33:26 +0000 (14:33 +0800)]
net/ixgbe: rework port check for driver-specific API
For PMD specific API it is required to check if provided port id is for
a supported device.
It's not appropriate to call rte_eth_dev_info_get in PMD, as
rte_eth_dev_info_get need to get info from PMD.
Remove rte_eth_dev_info_get from PMD code and get the info directly.
Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Keith Wiles [Mon, 6 Feb 2017 19:40:38 +0000 (13:40 -0600)]
net/tap: move closing file descriptors to close function
Remove closing fds code from pmd stop routine.
Signed-off-by: Keith Wiles <keith.wiles@intel.com>
Acked-by: Pascal Mazon <pascal.mazon@6wind.com>
Keith Wiles [Mon, 6 Feb 2017 19:40:37 +0000 (13:40 -0600)]
net/tap: move link down before close
Fixes:
f457b472b1f2 ("net/tap: add link up and down operations")
Signed-off-by: Keith Wiles <keith.wiles@intel.com>
Acked-by: Pascal Mazon <pascal.mazon@6wind.com>
Keith Wiles [Mon, 6 Feb 2017 19:40:36 +0000 (13:40 -0600)]
net/tap: cleanup log messages
Signed-off-by: Keith Wiles <keith.wiles@intel.com>
Acked-by: Pascal Mazon <pascal.mazon@6wind.com>
Keith Wiles [Mon, 6 Feb 2017 19:40:35 +0000 (13:40 -0600)]
net/tap: fix multi-queue support
At the same time remove the code which created the first device queue
at probe time. Now all queues are created during queue setup calls.
Fixes:
02f96a0a82d1 ("net/tap: add TUN/TAP device PMD")
Signed-off-by: Keith Wiles <keith.wiles@intel.com>
Acked-by: Pascal Mazon <pascal.mazon@6wind.com>
Keith Wiles [Mon, 6 Feb 2017 19:40:34 +0000 (13:40 -0600)]
net/tap: remove unused variable
Signed-off-by: Keith Wiles <keith.wiles@intel.com>
Acked-by: Pascal Mazon <pascal.mazon@6wind.com>
Keith Wiles [Mon, 6 Feb 2017 19:40:33 +0000 (13:40 -0600)]
net/tap: remove redundant file descriptor array
Signed-off-by: Keith Wiles <keith.wiles@intel.com>
Acked-by: Pascal Mazon <pascal.mazon@6wind.com>
Keith Wiles [Mon, 6 Feb 2017 19:40:32 +0000 (13:40 -0600)]
net/tap: fix guide for device name
Fixes:
02f96a0a82d1 ("net/tap: add TUN/TAP device PMD")
Signed-off-by: Keith Wiles <keith.wiles@intel.com>
Acked-by: Pascal Mazon <pascal.mazon@6wind.com>
Wei Zhao [Fri, 3 Feb 2017 08:44:49 +0000 (16:44 +0800)]
net/ixgbe: fix VXLAN parsing for flow director
VNI of VXLAN is parsed wrongly. The root cause is that
VNI array in VXLAN item also uses network byte ordering.
Fixes:
11777435c727 ("net/ixgbe: parse flow director filter")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Wei Zhao [Fri, 3 Feb 2017 08:43:57 +0000 (16:43 +0800)]
net/ixgbe: add TCI mask check for flow director
Add more check on the tci mask of VLAN and VXLAN parser
in fdir filter rule pattern parser. If such check not added,
it maybe cause error in fdir configuration set check.
Fixes:
11777435c727 ("net/ixgbe: parse flow director filter")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
John Daley [Fri, 3 Feb 2017 01:18:40 +0000 (17:18 -0800)]
net/enic: fix memory leak with oversized Tx packets
If a packet send is attempted with a packet larger than the NIC
is capable of processing (9208) it will be dropped with no
completion descriptor returned or completion index update, which
will lead to an mbuf leak and eventual hang.
Drop and count oversized Tx packets in the Tx burst function and
dereference/free the mbuf without sending it to the NIC.
Since the maximum Rx and Tx packet sizes are different on enic
and are now both being used, make the define ENIC_DEFAULT_MAX_PKT_SIZE
be 2 defines, one for Rx and one for Tx.
Fixes:
fefed3d1e62c ("enic: new driver")
Cc: stable@dpdk.org
Signed-off-by: John Daley <johndale@cisco.com>
Pascal Mazon [Thu, 2 Feb 2017 16:18:04 +0000 (17:18 +0100)]
net/tap: support promiscuous and allmulticast
Signed-off-by: Pascal Mazon <pascal.mazon@6wind.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
Pascal Mazon [Thu, 2 Feb 2017 16:18:03 +0000 (17:18 +0100)]
net/tap: add link up and down operations
Signed-off-by: Pascal Mazon <pascal.mazon@6wind.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
Pascal Mazon [Thu, 2 Feb 2017 16:18:02 +0000 (17:18 +0100)]
net/tap: display name after parsing
The probe parses for user-defined iface name. Let's use that value.
Signed-off-by: Pascal Mazon <pascal.mazon@6wind.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
Pascal Mazon [Thu, 2 Feb 2017 16:18:01 +0000 (17:18 +0100)]
net/tap: keep kernel-assigned MAC address
There's no point in having a different internal MAC address than the one
provided by the kernel when creating the netdevice.
Signed-off-by: Pascal Mazon <pascal.mazon@6wind.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
Pascal Mazon [Thu, 2 Feb 2017 16:18:00 +0000 (17:18 +0100)]
net/tap: skip overwritten file descriptor assignment
pmd->fds[0], pmd->rxq[0] and pmd->txq[0] are set a couple of lines after
the for loop that initializes them to -1.
Signed-off-by: Pascal Mazon <pascal.mazon@6wind.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
Pascal Mazon [Thu, 2 Feb 2017 16:17:59 +0000 (17:17 +0100)]
net/tap: fix log level of errors
Signed-off-by: Pascal Mazon <pascal.mazon@6wind.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
Pascal Mazon [Thu, 2 Feb 2017 16:17:58 +0000 (17:17 +0100)]
net/tap: fix name
dev->data->name contains the device name, e.g. "net_tap0".
dev->data->dev_private->name contains the actual iface name,
e.g. "dtap0".
In any case, the name must to be consistent with the tun_alloc() call in
eth_dev_tap_create().
Signed-off-by: Pascal Mazon <pascal.mazon@6wind.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
Nélio Laranjeiro [Thu, 2 Feb 2017 10:34:13 +0000 (11:34 +0100)]
net/mlx5: fix inline WQE consumption
For some sizes of packets, the number of bytes copied in the work queue
element could be greater than the available size of the inline. In such
situation it could consume one more work queue element where it should
not.
Fixes:
0e8679fcddc4 ("net/mlx5: fix inline logic")
Cc: stable@dpdk.org
Reported-by: Elad Persiko <eladpe@mellanox.com>
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Nélio Laranjeiro [Thu, 2 Feb 2017 10:34:12 +0000 (11:34 +0100)]
net/mlx5: fix Tx WQE corruption caused by starvation
Fixes an issue which may occurs with the inline feature activated and a
packet greater than the max_inline requested.
In such situation, more work request elements can be consumed and in the
worst case override some still handled by the NIC, this can result in
sending garbage on the network or putting the work queue in error.
Fixes:
2a66cf378954 ("net/mlx5: support inline send")
Cc: stable@dpdk.org
Reported-by: Elad Persiko <eladpe@mellanox.com>
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
Nélio Laranjeiro [Thu, 2 Feb 2017 10:34:11 +0000 (11:34 +0100)]
net/mlx5: fix ethernet header re-writing
First two bytes of the Ethernet header was written twice at the same place.
Fixes:
b8fe952ec5b6 ("net/mlx5: prepare Tx vectorization")
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Ivan Malov [Wed, 1 Feb 2017 08:17:11 +0000 (08:17 +0000)]
net/sfc: fix setting empty multicast list
The patch is to fix sfc_set_mc_addr_list() behaviour in order
to make it accept an empty multicast address list thus making
it possible to remove multicast addresses inserted previously
Fixes:
0fa0070e4391 ("net/sfc: support multicast addresses list controls")
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Nélio Laranjeiro [Tue, 31 Jan 2017 16:02:48 +0000 (17:02 +0100)]
net/mlx5: fix flow rule creation if ports are stopped
Adding a flow when the port is stopped ends in an inconsistent situation
where the queue can receive traffic when it should not.
Record new rules and apply them as soon as the port is started.
Fixes:
2097d0d1e2cc ("net/mlx5: support basic flow items and actions")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Nélio Laranjeiro [Tue, 31 Jan 2017 16:02:47 +0000 (17:02 +0100)]
net/mlx5: fix verification of mark action
A configuration structure for the MARK action must always be specified.
Fixes:
ea3bc3b1df94 ("net/mlx5: support mark flow action")
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Nélio Laranjeiro [Tue, 31 Jan 2017 16:02:46 +0000 (17:02 +0100)]
net/mlx5: use flow API default masks
Default masks were introduced in the API after its implementation in this
PMD.
Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Shahaf Shuler [Tue, 31 Jan 2017 13:13:38 +0000 (15:13 +0200)]
net/mlx5: fix inconsistent link status
Querying the link status can end up being in an inconsistent state,
like the port is reporting speed although it is down.
For this case another query is scheduled.
A race condition can occur between the scheduled query and link
status interrupt handlers.
When the scheduled query by-pass interrupt handlers, the link status
will be stuck in an inconsistent state.
This patch addresses the race condition by not blocking link status
queries in case delayed query is used.
Fixes:
198a3c339a8f ("mlx5: handle link status interrupts")
Cc: stable@dpdk.org
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Yong Wang [Fri, 10 Feb 2017 13:53:17 +0000 (08:53 -0500)]
pci: fix UIO interrupt file descriptor check before close
The "dev->intr_handle.fd" is possibly a negative value while it is
passed as an argument to function "close". Fix the check to the fd.
Fixes:
5a60a7ffc801 ("pci: introduce functions to alloc and free uio resource")
Signed-off-by: Yong Wang <wang.yong19@zte.com.cn>
Ido Barnea [Thu, 2 Feb 2017 12:05:07 +0000 (12:05 +0000)]
net/virtio: add speed capability
The chosen fake capability (10G) is consistent with the reported
link speed in virtio_dev_link_update():
link.link_speed = SPEED_10G;
The feature is not marked in doc/guides/nics/features/virtio.ini
because it is only a fake value.
Signed-off-by: Ido Barnea <ibarnea@cisco.com>
[Thomas: comments added]
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Jianfeng Tan [Thu, 9 Feb 2017 16:06:46 +0000 (16:06 +0000)]
doc: postpone ABI changes in igb_uio
This ABI changes to remove iomem and ioport mapping in igb_uio. The
purpose of this changes was to fix a bug: when DPDK app crashes,
those devices by igb_uio are not stopped either DPDK PMD driver or
igb_uio driver.
Then it has been pointed out by Stephen Hemminger that it has
backward compatibility issue: cannot run old version DPDK on
modified igb_uio.
However, we still have not figure out a new way to fix this bug
without this change. Let's postpone this deprecation announcement
in case this change cannot be avoided.
Fixes:
3bac1dbc1ed ("doc: announce iomem and ioport removal from igb_uio")
Suggested-by: Stephen Hemminger <stephen@networkplumber.org>
Suggested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Suggested-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Jianfeng Tan [Thu, 9 Feb 2017 16:06:45 +0000 (16:06 +0000)]
doc: add guide to use virtio-user as exceptional path
Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
Jianfeng Tan [Thu, 9 Feb 2017 16:06:44 +0000 (16:06 +0000)]
doc: add guide to use virtio-user for container networking
Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
Ferruh Yigit [Tue, 31 Jan 2017 15:04:48 +0000 (15:04 +0000)]
mk: move PMD libraries linking to applications
Some PMDs provide device specific APIs. Bond and xenvirt are existing
samples for this.
And since these are PMD libraries, there are two options on how to link
them for shared library build:
1- They can be linked to all applications by default, using common
rte.app.mk file.
2- They can be explicitly linked to applications that use device
specific API.
Currently option one is in use, this patch switches to the option two.
Moves library linking to the Makefile of application Makefile that uses
device specific API.
This prevent these PMD libraries to be a dependency to applications
that don't use these device specific APIs.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Ferruh Yigit [Tue, 31 Jan 2017 15:04:47 +0000 (15:04 +0000)]
app/testpmd: use LDLIBS in makefile
_LDLIBS is for internal usage, convert to LDLIBS usage.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Ferruh Yigit [Tue, 31 Jan 2017 15:01:32 +0000 (15:01 +0000)]
mk: link more libraries without --whole-archive
During app build with static library, some libraries wrapped with
--whole-archive compiler flag.
Wrapped libraries are mainly PMD libraries, this is required because PMD
APIs not called directly but run through callbacks registered via
constructor functions.
Also some set of libraries, depends to the PMD libraries needs this,
because of same reason.
All the libraries used by a plugin (any driver) must be in
--whole-archive to ensure that every symbols will be available for the
plugin.
But other libraries can be out of this flag, and this saves some bytes
in final binary.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Qiming Yang [Wed, 4 Jan 2017 12:18:15 +0000 (20:18 +0800)]
examples/ethtool: display bus information
This patch enhances the ethtool example to support to show
bus information, in the same way that the Linux kernel
ethtool does.
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Acked-by: Remy Horton <remy.horton@intel.com>
Alan Dewar [Mon, 6 Feb 2017 17:32:42 +0000 (18:32 +0100)]
sched: fix crash when freeing port
Prevent a segmentation fault in rte_sched_port_free by only accessing
the port structure after the NULL pointer check has been made.
Fixes:
7b3c4f35 ("sched: fix releasing enqueued packets")
Cc: stable@dpdk.org
Signed-off-by: Alan Dewar <adewar@brocade.com>
Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Patrick MacArthur [Thu, 26 Jan 2017 23:05:21 +0000 (18:05 -0500)]
vfio: fix file descriptor leak in multi-process
When a secondary process wants access to the VFIO container file
descriptor, the primary process calls vfio_get_container_fd() which
always opens an entirely new file descriptor on /dev/vfio/vfio.
However, once the file descriptor has been passed to the subprocess, it
is effectively duplicated, meaning that the copy of the file descriptor
in the primary process is no longer needed. However, the primary
process does not close the duplicate fd, which results in a resource
leak.
This can be reproduced by starting a primary process with a small
RLIMIT_NOFILE limit configured to use VFIO for at least one device, and
repeatedly launching secondary processes until the file descriptor limit
is exceeded.
Fix the resource leak by closing the local vfio container file
descriptor after passing it to the secondary process.
Fixes:
2f4adfad0a69 ("vfio: add multiprocess support")
Cc: stable@dpdk.org
Signed-off-by: Patrick MacArthur <patrick@patrickmacarthur.net>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>