18918 Commits

Author SHA1 Message Date
David Marchand
8ff0003c19 net/bonding: fix out of bound access in LACP mode
We'd better consolidate the fast queue and the normal tx burst functions
under a common inline wrapper for maintenance.

But looking closer at the bufs_slave_port_idxs[] mapping array in those
tx burst functions, its size is invalid since up to nb_bufs are handled
here.
A previous patch [1] fixed this issue for balance tx burst function
without mentioning it.

802.3ad and balance modes are functionally equivalent on transmit.
The only difference is on the slave id distribution.
Add an additional inline wrapper to consolidate even more and fix this
issue.

[1]: https://git.dpdk.org/dpdk/commit/?id=c5224f623431

Fixes: 09150784a776 ("net/bonding: burst mode hash calculation")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Chas Williams <chas3@att.com>
2019-08-22 18:45:49 +02:00
David Marchand
e7fd5a3bbe version: 19.11-rc0
Start a new release cycle with empty release notes.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
2019-08-13 14:59:02 +02:00
Thomas Monjalon
b81660d09e version: 19.08.0
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
v19.08
2019-08-11 11:37:19 +02:00
Rami Rosen
78b04aaf77 doc: add guidelines for initial PMD submission
This patch for DPDK Contributor's Guidelines indicates the repos
against which a new PMD should be prepared; for example, for new
network ethernet PMDs it should be dpdk-next-net, and for new crypto
PMDs it should be dpdk-next-crypto. For other new PMDs, the
contributor should refer to the MAINTAINERS file. Though this may seem
obvious, it is not mentioned in DPDK documentation.

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>

Last contribution, last commit closing the release.
Thank you Rami, you will be missed.
2019-08-11 11:34:24 +02:00
Jasvinder Singh
f6feb39c20 examples/qos_meter: fix color type conversion
In APP_MODE_SRTCM_COLOR_AWARE mode, sample app compilation fails
due to wrong meter color type conversion.

error log-
/qos_meter/main.c:error: conversion to incomplete type
(enum rte_meter_color) input_color);

Fixes: c1656328dbc2 ("meter: replace color definitions")
Cc: stable@dpdk.org

Reported-by: Yuan Peng <yuan.peng@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
2019-08-11 00:07:31 +02:00
Lance Richardson
3227d09853 net/bnxt: revert fix traffic stall on stop/start
This reverts commit aa2c00702bad7b2c742e11a86cb9dbbb8364fd88, which
introduced the possibility of an invalid address exception when running
an application with a stopped receive queue. The issues with rxq stop/start
 will be revisited in the 19.11 release timeframe.

Fixes: aa2c00702bad ("net/bnxt: fix traffic stall on Rx queue stop/start")
Cc: stable@dpdk.org

Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
2019-08-10 23:55:07 +02:00
Matan Azrad
200e96c7df doc: announce ethdev ABI change for LRO fields
It may be needed by the user to limit the LRO session packet size.
In order to allow the above limitation, a new Rx configuration may be
added for the maximum LRO session size.

A new capability may be added too to expose the maximum LRO session size
supported by the port.

Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
2019-08-10 23:36:24 +02:00
Pavan Nikhilesh
c5b2e78d11 doc: announce ethdev API changes in offload flags
Add new offload flags ``DEV_RX_OFFLOAD_RSS_HASH`` and
``DEV_RX_OFFLOAD_FLOW_MARK``.
Add new function ``rte_eth_dev_set_supported_ptypes`` to allow application
to set specific ptypes to be updated in ``rte_mbuf::packet_type``

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2019-08-10 23:07:12 +02:00
Thomas Monjalon
d00c149791 doc: announce removal of old port count function
The function rte_eth_dev_count() was marked as deprecated in DPDK 18.05
in commit d9a42a69febf ("ethdev: deprecate port count function").
It is planned to be removed after the next LTS release.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: David Marchand <david.marchand@redhat.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
2019-08-10 22:34:39 +02:00
Andrew Rybchenko
fbe7ac9f3a doc: announce ethdev functions will not return void
void return value is bad for get API (like rte_eth_dev_info-get())
since caller does not know if the function does its job or not and
output value is filled in.

void return value is bad for state changing API (like
rte_eth_promiscuous_enable()) since caller should use get API
to understand if state is really changed.

Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Jerin Jacob <jerinj@marvell.com>
2019-08-10 22:33:38 +02:00
Stephen Hemminger
9c8a4aba02 doc: announce ethernet structures alignment changes
Tell users about upcoming changes to rte_ether_addr and
rte_ether_header.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
2019-08-10 22:21:31 +02:00
David Marchand
07f08a96f5 doc: announce lcore config struct removal from ABI
New accessors have been introduced to provide the hidden information.
This symbol can now be kept internal.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
2019-08-08 17:55:22 +02:00
David Marchand
23be1cbfa3 doc: announce malloc virt2phys function removal
This symbol has been deprecated for quite some time.
Let's drop it as soon as possible.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2019-08-08 17:54:53 +02:00
David Marchand
411b586f0e doc: announce CPU flags check function removal
This symbol has been deprecated for quite some time.
Let's drop it as soon as possible.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Aaron Conole <aconole@redhat.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
2019-08-08 17:54:08 +02:00
Andrius Sirvys
a667070b09 usertools: fix input handling in telemetry script
This commit removes the unnecesarry ast.literal_eval() function call
from the input handling, which now relies just on raw_input() to get
its input.

Fixes: 53f293c9a783 ("usertools: replace unsafe input function")
Cc: stable@dpdk.org

Signed-off-by: Andrius Sirvys <andrius.sirvys@intel.com>
Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
2019-08-08 17:42:24 +02:00
Gagandeep Singh
47caefc163 eal: increase maximum different hugepage sizes on Arm
ARM is supporting maximum 4 hugepage sizes (64K, 2M, 32M
and 1G) when granule is 4KB since very long and DPDK
support maximum 3 hugepage sizes.

With all 4 hugepage sizes enabled, applications and some
stacks like VPP which are working over DPDK and using
"in-memory" eal option, or using separate mount points
on ARM based platform, fails at huge page initialization,
reporting error messages from eal:

EAL: FATAL: Cannot get hugepage information.
EAL: Cannot get hugepage information.
EAL: Error - exiting with code: 1

This issue is originated from Linux 5.0
(a21b0b78eaf7 "arm64: hugetlb: Register hugepages during arch init")
where kernel is by default creating directories for each supported
hugepage size in /sys/kernel/mm/hugepages/

On earlier Stable Kernel LTR's, the directories visible in
/sys/kernel/mm/hugepages/ were dependent upon what hugepage
sizes are configured at boot time.

This change increases the maximum supported hugepage sizes
to 4 for ARM based platforms.

Cc: stable@dpdk.org

Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
2019-08-08 17:25:14 +02:00
Congwen Zhang
1861e3a66f net/ixgbe/base: fix product version check
The "and" condition offset == 0 && offset == NVM_INVALID_PTR
 can never be true.

Fixes: cf3af5aa56c9 ("net/ixgbe/base: add functions to get version info")
Cc: stable@dpdk.org

Signed-off-by: Congwen Zhang <zhang.congwen@zte.com.cn>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
2019-08-08 16:55:19 +02:00
Viacheslav Ovsiienko
4dec9c79c2 net/mlx5: fix completion request for multi-segment
The copying of sent mbufs pointers might be deferred to the end of
tx_burst() routine to be copied in one call of rte_memcpy.
For the multi segment packets this optimization is not applicable,
because number of packets does not match with number of mbufs and
we do not have linear array of pointers in pkts parameter.

The completion request generating routine wrongly took into account
the inconsistent (for multi-segment packets) deferred pointer copying.

Fixes: 5a93e173b874 ("net/mlx5: fix Tx completion request generation")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-07 15:28:09 +02:00
Viacheslav Ovsiienko
3a418d1d3d doc: add limitation with mlx5 Tx inline settings
Introduces the possible limitations on maximal Tx queue
size in descriptors if Tx inline data are enabled.

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-08 16:53:49 +02:00
Viacheslav Ovsiienko
e98e44bace doc: update Tx inline settings in mlx5 guide
This patch updates mlx5 documentation in parts:

  - txq_inline_min parameter is described in more details,
    values are fixed

  - maximal amount of segments in multi-segment packets.

Fixes: 38b4b397a57d ("net/mlx5: add Tx configuration and setup")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Raslan Darawsheh <rasland@mellanox.com>
2019-08-07 15:27:45 +02:00
Aaron Conole
7fb1e84570 test/mcslock: wait for lcore completion
It's possible that the mcsunlock occurs before the test_mcslock_try has
a chance to execute, which will result in the trylock being successful,
making the test case fail.  Fix this by waiting until all lcores have
completed their test before unlocking the master lock.

Fixes: 32dcb9fd2a22 ("test/mcslock: add MCS queued lock unit test")

Signed-off-by: Aaron Conole <aconole@redhat.com>
Acked-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
2019-08-08 12:19:59 +02:00
Amit Gupta
8b36297d15 app/testpmd: fix latency stats deinit on signal
On receiving signal, testpmd showing warning as
"LATENCY_STATS: failed to remove Rx/Tx callback"
because rte_latencystats_uninit is called without
checking if latencystats is enabled or not.

After this fix, rte_latencystats_uninit will be
called only if latencystats is enabled.

Fixes: 62d3216d6194 ("app/testpmd: add latency statistics calculation")
Cc: stable@dpdk.org

Signed-off-by: Amit Gupta <agupta3@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Tested-by: Jerin Jacob <jerinj@marvell.com>
2019-08-08 12:13:05 +02:00
Jerin Jacob
de4473d911 net/memif: fix build with gcc 9.1
gcc-9 is stricter on NULL arguments for printf.
Fix the following build error by avoiding NULL argument to printf.

In file included from drivers/net/memif/memif_socket.c:26:
In function 'memif_socket_create',
inlined from 'memif_socket_init' at net/memif/memif_socket.c:965:12:
net/memif/rte_eth_memif.h:35:2: error:
'%s' directive argument is null [-Werror=format-overflow=]
   35 |  rte_log(RTE_LOG_ ## level, memif_logtype, \
      |  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   36 |   "%s(): " fmt "\n", __func__, ##args)
      |   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Fixes: 09c7e63a71f9 ("net/memif: introduce memory interface PMD")

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
2019-08-07 10:21:33 +02:00
Thomas Monjalon
0710d87b7f version: 19.08-rc4
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
v19.08-rc4
2019-08-06 23:50:31 +02:00
John McNamara
e090fdb41a doc: advise patching third party dependencies
Added a note to the getting started guides about patching third
party libraries/dependencies to avoid any known vulnerabilities.

Signed-off-by: John McNamara <john.mcnamara@intel.com>
2019-08-06 23:48:54 +02:00
Thomas Monjalon
030febb664 doc: remove deprecated ethdev features
As legacy filter API "filter_ctrl" is superseded since 2017
by the rte_flow API, and got the deprecated attribute in DPDK 19.05,
it is time to remove the associated features from the matrix.
Not documenting deprecated features as supported will avoid confusion.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
2019-08-06 23:41:55 +02:00
Nicolas Chautru
be840d083f app/bbdev: fix bulk allocation checks
Returned value from rte_bbdev_enc_op_alloc_bulk not checked consistently.

Coverity issue: 344986
Fixes: d819c08327f3 ("app/bbdev: update for 5GNR")

Signed-off-by: Nicolas Chautru <nicolas.chautru@intel.com>
2019-08-06 23:36:34 +02:00
Nicolas Chautru
4e8cd9041e baseband/turbo_sw: remove dead code
Error handling code cannot logically be reached.

Coverity issue: 344980, 344992
Fixes: c769c711757a ("baseband/turbo_sw: extend for 5G")

Signed-off-by: Nicolas Chautru <nicolas.chautru@intel.com>
2019-08-06 23:33:14 +02:00
Viacheslav Ovsiienko
0e3d0525b2 net/mlx5: fix memory event callback list
The shared Infiniband device context should be included
into memory event callback list only once on context creation,
and removed from the list only once on context destroying.
Multiple insertions of the same object caused the infinite
loop on the list processing.

Fixes: ccb3815346e6 ("net/mlx5: update memory event callback for shared context")
Cc: stable@dpdk.org

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-06 17:42:12 +02:00
Xiaoyu Min
c3318f529c net/mlx5: fix RSS expand for IP-in-IP
The RSS expand function for IP-in-IP tunnel type is missed,
which leads to create following flow failed:

   flow create 0 ingress pattern eth / ipv4 proto is 4 /
        ipv4 / udp / end actions rss queues 0 1 end level 2
	types ip ipv4-other udp ipv4 ipv4-frag end /
	mark id 221 / count / end

In order to make RSS expand function working correctly,
now the way to check whether a IP tunnel existing is to
check whether there is the second IPv4/IPv6 item and whether the
first IPv4/IPv6 item's next protocl is IPPROTO_IPIP/IPPROTO_IPV6.
For example:
  ... pattern eth / ipv4 proto is 4 / ipv4 / ....

Fixes: 5e33bebdd8d3 ("net/mlx5: support IP-in-IP tunnel")

Signed-off-by: Xiaoyu Min <jackmin@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-06 17:42:12 +02:00
Thomas Monjalon
4710b1133e doc: add more details about mlx5 offloads
Add firmware config for MPLS and DevX (required by LRO and DR).

Add a table for queue offloads requirements.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Raslan Darawsheh <rasland@mellanox.com>
2019-08-06 17:42:12 +02:00
Thomas Monjalon
bdc3a54225 doc: add mlx5 design details
Some details about libibverbs were missing:
- automatic detection by meson
- main ways to access the device

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Raslan Darawsheh <rasland@mellanox.com>
2019-08-06 17:42:12 +02:00
Thomas Monjalon
499b461f90 doc: remove useless console syntax in mlx guides
It is not needed to use "console" syntax highlighting
for literal blocks.
The file is easier to read by removing the code-block lines
and simply having double colons in previous line.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Raslan Darawsheh <rasland@mellanox.com>
2019-08-06 17:42:12 +02:00
Thomas Monjalon
23bdbfec2b doc: fix wording and formatting of mlx5 guide
These are simple fixes of punctuation, anchor placement
or wording.
The table format is fixed to avoid having a long line
in the first column.

Fixes: 909be50a34d0 ("doc: update Mellanox guides and release notes")
Cc: stable@dpdk.org

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Raslan Darawsheh <rasland@mellanox.com>
2019-08-06 17:42:12 +02:00
Thomas Monjalon
ae28c04547 doc: improve firmware configuration in mlx5 guide
The command mlxconfig was not enough explained and too much verbose
at the same time.
The syntax is now explained in introduction before listing the options,
without repeating the commands.
Some options, which are explained elsewhere in the doc,
are added to this list.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Raslan Darawsheh <rasland@mellanox.com>
2019-08-06 17:42:12 +02:00
Dekel Peled
1f1eb0949c net/mlx5: fix UDP checksum zeroing
Function flow_dv_zero_encap_udp_csum() uses a while loop to iterate
over vlan items in flow rule.
Pointer next_hdr is incremented to the next item before it is used,
so the first item is skipped.

This patch moves the incrementing of next_hdr to the correct place.

Fixes: bf1d7d9a033a ("net/mlx5: zero out UDP checksum in encapsulation")

Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-06 17:42:12 +02:00
Xiaoyu Min
6fd05da9ef net/mlx5: fix link speed info when link is down
When the link is down, the link speed returned by ethtool is
UINT32_MAX and the link status is 0.

In this case, the DPDK ethdev link speed should be set to
ETH_SPEED_NUM_NONE.
Otherwise since link speed is non-zero but link status is zero, this
is an inconsistent situation and -EAGAIN is returned, which is not right.

Fixes: 188408719888 ("net/mlx5: fix support for newer link speeds")
Cc: stable@dpdk.org

Signed-off-by: Xiaoyu Min <jackmin@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-06 17:42:12 +02:00
Viacheslav Ovsiienko
f32a3f5216 net/mlx5: fix completion queue overflow for large burst
There is the limit on completion descriptor fetch to improve
latency. If burst size is large there might be not enough
resources freed in completion processing. This fix reiterates
sending loop and allows multiple completion descriptor fetch
and processing.

Fixes: 18a1c20044c0 ("net/mlx5: implement Tx burst template")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
2019-08-06 17:42:12 +02:00
Viacheslav Ovsiienko
3502b05925 net/mlx5: fix packet size inline settings
This patch fixes the default settings for packet size to inline
with Enhanced Multi-Packet Write feature, allowing 256B packets
to be inlined with Out-Of-the-Box settings.

Fixes: 50724e1bba76 ("net/mlx5: update Tx definitions")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
2019-08-06 17:42:12 +02:00
Viacheslav Ovsiienko
c6f0485646 net/mlx5: fix inline data settings
If the minimal inline data are required the data inline feature
must be engaged. There were the incorrect settings enabling the
entire small packet inline (in size up to 82B) which may result
in sending rate declining if there is no enough cores. The same
problem was raised if inline was enabled to support VLAN tag
insertion by software.

Fixes: 38b4b397a57d ("net/mlx5: add Tx configuration and setup")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
2019-08-06 17:42:12 +02:00
Viacheslav Ovsiienko
da1df1ccab net/mlx5: fix completion queue drain loop
The completion loop speed optimizations for error-free
operations are done - no CQE field fetch on each loop
iteration. Also, code size is oprimized - the flush
buffers routine is invoked once.

Fixes: 318ea4cfa1b1 ("net/mlx5: fix Tx completion descriptors fetching loop")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
2019-08-06 17:42:12 +02:00
Viacheslav Ovsiienko
7014ef5bba net/mlx5: fix inline data length assert
The debug assert wrongly triggers on the inline data 18B,
this should be passed successfully.

Fixes: 18a1c20044c0 ("net/mlx5: implement Tx burst template")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
2019-08-06 17:42:12 +02:00
Viacheslav Ovsiienko
614de6c898 net/mlx5: fix default minimal data inline
The patch [Fixes] sets the default value of required minimal
inline data to 0 bytes. On some configurations (depends
on switchdev/legacy settings and FW version/settings)
the ConnectX-4LX NIC requires minimal 18 bytes of
Tx descriptor inline data to operate correctly.

Wrongly set to 0 default value may prevent NIC from operating
with out-of-the-box settings, this patch reverts default
value for ConnectX-4LX back to 18 bytes (inline L2).

Fixes: 9f350504bb32 ("net/mlx5: fix ConnectX-4LX minimal inline data limit")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
2019-08-06 17:42:12 +02:00
Xiaoyu Min
b12c7b2368 net/mlx5: fix VLAN inner type matching on DR/DV
The rte_flow_item_vlan has the inner_type, which is missing on
DR/DV flow engine.

By adding this support, the example testpmd commands could be:

 - matching all vlan traffic with id 2:

 testpmd> flow create 0 ingress pattern eth / vlan vid is 2 / end
          actions queue index 2 / end

 - matching all ipv4 traffic in vlan with id 2:

 testpmd> flow create 0 ingress pattern eth / vlan vid is 2
          inner_type is 0x0800 / end actions queue index 2 / end

Fixes: fc2c498ccb94 ("net/mlx5: add Direct Verbs translate items")
Cc: stable@dpdk.org

Signed-off-by: Xiaoyu Min <jackmin@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-06 17:42:12 +02:00
Dekel Peled
9b08de133e net/mlx5: fix flow rule configuration
Some flow rules were not configured.

Part of the code in function flow_dv_matcher_enable() is enclosed in
'#ifdef HAVE_MLX5DV_DR' preprocessor directive.
Using this directive is not needed here, and prevents compilation of
relevant code.

This patch removes the unnecessary preprocessor directive.

Fixes: 4f84a19779ca ("net/mlx5: add Direct Rules API")
Cc: stable@dpdk.org

Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-06 17:42:12 +02:00
Dekel Peled
a3970d31e6 net/mlx5: fix validation of VLAN PCP item
Function mlx5_flow_validate_item_vlan() validates the user setting
is supported by NIC, using a mask with TCI mask 0x0fff.
This check will reject a flow rule specifying a vlan pcp item.

This patch updates mlx5_flow_validate_item_vlan() to use mask 0xffff,
so flow rules with vlan pcp item are accepted.

Fixes: 23c1d42c7138 ("net/mlx5: split flow validation to dedicated function")
Cc: stable@dpdk.org

Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-06 17:42:12 +02:00
Stephen Hemminger
164cad78b0 net/mlx4: fix crash on info query in secondary process
mlx4_dev_info_get calls mlx4_get_ifname, but mlx4_get_ifname
uses priv->ctx which is not a valid pointer in a secondary
process. The fix is to cache the value in primary.

In the primary process, get and store the interface index of
the device so that secondary process can see it.

Bugzilla ID: 320
Fixes: 61cbdd419478 ("net/mlx4: separate device control functions")
Cc: stable@dpdk.org

Reported-by: Suyang Ju <sju@paloaltonetworks.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Matan Azrad <matan@mellanox.com>
2019-08-06 17:42:12 +02:00
David Christensen
20215627cd net/mlx5: fix Tx inline minimum for ConnectX-5
The function mlx5_set_min_inline() includes a switch() that checks
various PCI device IDs in order to set the txq_inline_min value.  No
value is set when the PCI device ID matches the ConnectX-5 adapters,
resulting in an assert() failure later in the function
mlx5_set_txlimit_params().

This error was encountered on an IBM Power 9 system running RHEL 7.6
w/o Mellanox OFED installed.

Fixes: 38b4b397a57d ("net/mlx5: add Tx configuration and setup")

Signed-off-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-06 17:42:12 +02:00
Dekel Peled
b857ed01ca net/mlx5: fix limit of direct rules tables number
MLX5 PMD limits the number of SW steering tables to 32.
This patch updates the limit to 65535, to allow wide range of values.

Fixes: e2b4925ef7c1 ("net/mlx5: support Direct Rules E-Switch")
Cc: stable@dpdk.org

Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
2019-08-06 17:42:12 +02:00
Viacheslav Ovsiienko
dfedf3e3f9 net/mlx5: add workaround for VLAN in virtual machine
On some virtual setups (particularly on ESXi) when we have SR-IOV and
E-Switch enabled there is the problem to receive VLAN traffic on VF
interfaces. The NIC driver in ESXi hypervisor does not setup E-Switch
vport setting correctly and VLAN traffic targeted to VF is dropped.

The patch provides the temporary workaround - if the rule
containing the VLAN pattern is being installed for VF the VLAN
network interface over VF is created, like the command does:

  ip link add link vf.if name mlx5.wa.1.100 type vlan id 100

The PMD in DPDK maintains the database of created VLAN interfaces
for each existing VF and requested VLAN tags. When all of the RTE
Flows using the given VLAN tag are removed the created VLAN interface
with this VLAN tag is deleted.

The name of created VLAN interface follows the format:

  evmlx.d1.d2, where d1 is VF interface ifindex, d2 - VLAN ifindex

Implementation limitations:

- mask in rules is ignored, rule must specify VLAN tags exactly,
  no wildcards (which are implemented by the masks) are allowed

- virtual environment is detected via rte_hypervisor() call,
  and the type of hypervisor is checked. Currently we engage
  the workaround for ESXi and unrecognized hypervisors (which
  always happen on platforms other than x86 - it means workaround
  applied for the Flow over PCI VF). There are no confirmed data
  the other hypervisors (HyperV, Qemu) need this workaround,
  we are trying to reduce the list of configurations on those
  workaround should be applied.

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
2019-08-06 17:42:12 +02:00