Ferruh Yigit [Wed, 20 May 2020 10:22:22 +0000 (11:22 +0100)]
bus/fslmc: drop inline from non-static functions
There is no point in having non-static (and non-extern) inline
functions.
Also this breaks the build for the ICC [1] because of the 'internal'
symbol check.
When function is 'inline' ICC is ignoring 'section' attribute and not
putting function to 'internal' section which breaks 'check-symbols.sh'
script with below error.
[1]
qbman_swp_dqrr_next is not flagged as internal
but is listed in version map
Please add __rte_internal to the definition of qbman_swp_dqrr_next
qbman_swp_enqueue_multiple is not flagged as internal
but is listed in version map
Please add __rte_internal to the definition of qbman_swp_enqueue_multiple
qbman_swp_enqueue_multiple_desc is not flagged as internal
but is listed in version map
Please add __rte_internal to the definition of qbman_swp_enqueue_multiple_desc
qbman_swp_enqueue_multiple_fd is not flagged as internal
but is listed in version map
Please add __rte_internal to the definition of qbman_swp_enqueue_multiple_fd
qbman_swp_pull is not flagged as internal
but is listed in version map
Please add __rte_internal to the definition of qbman_swp_pull
qbman_swp_release is not flagged as internal
but is listed in version map
Please add __rte_internal to the definition of qbman_swp_release
Jeff Guo [Tue, 19 May 2020 14:23:50 +0000 (10:23 -0400)]
net/iavf: fix VF reset for RSS
Since there are some default rss configure in kernel PF/VF but not DPDK
IAVF, if these configurations be modified by VF and then VF reset, this
default rss configurations can not be reset to default by IAVF. So need
to add default rss set in IAVF hash initial process.
Fixes: 7be10c3004be ("net/iavf: add RSS configuration for VF") Cc: stable@dpdk.org Signed-off-by: Jeff Guo <jia.guo@intel.com> Acked-by: Qi Zhang <qi.z.zhang@intel.com> Tested-by: Zhiwei He <zhiwei.he@intel.com>
Qiming Yang [Mon, 18 May 2020 05:45:53 +0000 (13:45 +0800)]
net/i40e: fix queue related exception handling
There should have different behavior in queue start fail and stop fail
case. When queue start fail, all the next actions should be terminated
and then started queues should be cleared. But for queue stop stage, one
queue stop fail should not end other queues stop. This patch fixed that
issue in PF and VF.
Fixes: b6583ee40265 ("i40e: full VMDQ pools support") Fixes: 3f6a696f1054 ("i40evf: queue start and stop") Cc: stable@dpdk.org Signed-off-by: Qiming Yang <qiming.yang@intel.com> Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Qi Zhang [Wed, 13 May 2020 12:11:44 +0000 (20:11 +0800)]
net/ice: fix RSS for GTPU
All supported pattern for GTPU include extend header:
pattern_eth_ipv4_gtpu_eh_ipv4
pattern_eth_ipv4_gtpu_eh_ipv4_udp
pattern_eth_ipv4_gtpu_eh_ipv4_tcp
So the RSS rule should only take effect on GTPU packet that contains
extend header. The patch fix above issue and also allow inner l4 port
as input set.
Fixes: c08a72c79c7f ("net/ice: fix pattern name of GTPU with extension header") Cc: stable@dpdk.org Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com> Tested-by: Simei Su <simei.su@intel.com>
Wei Zhao [Mon, 18 May 2020 08:00:51 +0000 (16:00 +0800)]
net/i40e: fix wild pointer
In i40e PMD code of function i40e_res_pool_free(), if valid_entry
is freed by "rte_free(valid_entry);" in the code, then the following
code for pool update may still use the wild pointer "valid_entry"
for pool info update. It seems has the risk of core dump for
using wild pointer operation, we should avoid this risk.
Fixes: 4861cde46116 ("i40e: new poll mode driver") Cc: stable@dpdk.org Signed-off-by: Wei Zhao <wei.zhao1@intel.com> Reviewed-by: Jeff Guo <jia.guo@intel.com> Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
Check returned value after strtok()
CID 355674 (#1 of 1): Dereference null return value (NULL_RETURNS)
4. dereference: Dereferencing a pointer that might be NULL s when
calling inet_pton
Fixes: 103809d032cd ("app/test-fib: add test application for FIB") Cc: stable@dpdk.org Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
Haiyue Wang [Thu, 30 Apr 2020 05:46:05 +0000 (13:46 +0800)]
devtools: handle internal version in symbols check
INTERNAL is a newly introduced version, update the shell script that
checks whether built libraries are versioned with expected ABI
(current ABI, current ABI + 1, EXPERIMENTAL, or INTERNAL).
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com> Acked-by: David Marchand <david.marchand@redhat.com>
Pavan Nikhilesh [Mon, 11 May 2020 10:07:38 +0000 (15:37 +0530)]
mempool/octeontx2: add devargs to lock context in cache
Add device arguments to lock NPA aura and pool contexts in NDC cache.
The device args take hexadecimal bitmask where each bit represent the
corresponding aura/pool id.
Example:
-w 0002:02:00.0,npa_lock_mask=0xf // Lock first 4 aura/pool ctx
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Bruce Richardson [Mon, 27 Apr 2020 14:25:20 +0000 (15:25 +0100)]
mk: add note about make system deprecation
When anyone uses the make build system, they are to be informed
about upcoming plans to deprecate and subsequently remove that
system and to use meson and ninja instead.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Anatoly Burakov [Mon, 11 May 2020 10:25:13 +0000 (11:25 +0100)]
kernel/linux: error out on module build failure
Now that kernel modules aren't built by default, we can be more
strict with their build process, and fail the build if they were
requested to be built, but weren't.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Darek Stojaczyk [Thu, 7 May 2020 10:13:50 +0000 (12:13 +0200)]
build: skip config of explicitly disabled drivers
Even when a PMD was disabled with meson's disable_drivers option
its config file was still being parsed. Some of the PMD configs
attempt to find a library they depend on and parse its header files
with certain assumptions. If the library is found, but it's simply
too old to contain the necessary header files, the meson build
fails and it can only be fixed by either updating that library, or
expanding the meson script for the faulty PMD.
While the latter should be still done for the sake of DPDK quality,
an intermediate solution would be to skip building the faulty PMD
- there's a chance we don't need it. That's what this patch allows.
Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Ferruh Yigit [Tue, 19 May 2020 12:16:29 +0000 (13:16 +0100)]
meter: provide experimental alias for matured API
On v20.02 some meter APIs have been matured and symbols moved from
EXPERIMENTAL to DPDK_20.0.1 block.
This can break the applications that were using these mentioned APIs on
v19.11. Although there is no modification on the APIs and the action is
positive and matures the APIs, the affect can be negative to
applications.
This patch provides aliasing by duplicating the existing and versioned
symbols as experimental.
Since symbols moved from DPDK_20.0.1 to DPDK_21 block in the v20.05, the
aliasing done between EXPERIMENTAL and DPDK_21.
With DPDK_21 ABI (DPDK v20.11) all aliasing will be removed and only
stable version of the APIs will remain.
Fixes: 30512af820fe ("meter: remove experimental flag from RFC4115 trTCM API") Cc: stable@dpdk.org Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com> Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com> Acked-by: Ray Kinsella <mdr@ashroe.eu>
Muhammad Bilal [Fri, 15 May 2020 16:50:54 +0000 (21:50 +0500)]
fix same typo in multiple places
Removed the typing error in doc/guides/eventdevs/index.rst,
drivers/net/mlx5/mlx5.c and in lib/librte_vhost/rte_vhost.h
Bugzilla ID: 477 Fixes: 0857b9421138 ("doc: add event device and software eventdev") Fixes: 039253166a57 ("vhost: add device op when notification to guest is sent") Fixes: ad74bc619504 ("net/mlx5: support multiport IB device during probing") Cc: stable@dpdk.org Signed-off-by: Muhammad Bilal <m.bilal@emumba.com>
Hemant Agrawal [Fri, 15 May 2020 09:47:51 +0000 (15:17 +0530)]
bus/fslmc: add accessor for MCP
Currently rte_mcp_ptr_list is being shared as a variable
across libs. This is only used in control path.
This patch change it to a exported function based access.
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com> Acked-by: Ray Kinsella <mdr@ashroe.eu>
Hemant Agrawal [Fri, 15 May 2020 09:47:50 +0000 (15:17 +0530)]
bus/dpaa: add accessor for netcfg
This patch changes the export of fman port config
as function call instead of direct variable access.
This is in control path, so it will not impact perf.
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com> Acked-by: Ray Kinsella <mdr@ashroe.eu>
Hemant Agrawal [Fri, 15 May 2020 09:47:48 +0000 (15:17 +0530)]
bus/dpaa: add accessor for qman portal ids
qman caam and pool portal ids are only used in control
path. This patch changes their inter library access to
function call instead of direct shared variable.
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com> Acked-by: Ray Kinsella <mdr@ashroe.eu>
Hemant Agrawal [Fri, 15 May 2020 09:47:42 +0000 (15:17 +0530)]
bus/dpaa: mark internal symbols
This patch moves the internal symbols to INTERNAL sections
so that any change in them is not reported as ABI breakage.
This patch also removes two symbols, which are not to be exported.
rte_dpaa_mem_ptov - static inline in the headerfile
fman_ccsr_map_fd - local shared variable.
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com> Acked-by: Ray Kinsella <mdr@ashroe.eu>
Ciara Power [Tue, 12 May 2020 15:29:02 +0000 (16:29 +0100)]
telemetry: fix buffer overrun if max bytes read
If 1024 bytes were received over the socket, this caused
buffer_recvf[bytes] to overrun the array. The size of the buffer - 1 is
now passed to the read function.
Coverity issue: 358442 Fixes: b80fe1805eee ("telemetry: introduce backward compatibility") Signed-off-by: Ciara Power <ciara.power@intel.com> Acked-by: Kevin Laatz <kevin.laatz@intel.com>
Ciara Power [Tue, 12 May 2020 15:28:58 +0000 (16:28 +0100)]
telemetry: keep threads separate from data plane
The threads for listening on the telemetry sockets are control threads
and should be separated from those on the data plane. Since telemetry
cannot use the rte_ctrl_thread_create() API, as it does not depend on
EAL, we pass the ctrl thread cpu_set to telemetry init and use it
directly to ensure that telemetry cannot interfere with the data plane
threads.
Signed-off-by: Ciara Power <ciara.power@intel.com> Acked-by: Kevin Laatz <kevin.laatz@intel.com>
Gaetan Rivet [Wed, 13 May 2020 10:43:01 +0000 (12:43 +0200)]
doc: fix typos in ABI policy
Some errors in the document:
* API instead of ABI once.
Some typos:
* __rte_depreciated instead of __rte_deprecated.
* missing ```` around value.
* inconsistent reference to major ABI version, most
of the time described without the minor appended, except once.
Verbosity and grammar:
* Long sentences that would be better cut short.
* Comma abuse.
* 'May' used where 'can' seems more fitting.
I'm not a native speaker though, so grain of salt applies.
Fixes: fdf7471cccb8 ("doc: introduce major ABI versions") Cc: stable@dpdk.org Signed-off-by: Gaetan Rivet <grive@u256.net> Acked-by: Ray Kinsella <mdr@ashroe.eu>
Gaetan Rivet [Wed, 13 May 2020 10:47:51 +0000 (12:47 +0200)]
pci: explain how empty strings are rejected in DBDF
Empty strings are forbidden as input to rte_pci_addr_parse().
It is explicitly enforced in BDF parsing as parsing the bus
field will immediately fail. The related check is commented.
It is implicitly enforced in DBDF parsing, as the domain would be
parsed to 0 without error, but the check `end[0] != ':'` afterward
will return -EINVAL.
Enforcing consistency between parsers by reading the code is not helped
by this property being implicit. Add a comment to explain.
Darek Stojaczyk [Tue, 12 May 2020 13:30:57 +0000 (15:30 +0200)]
pci: accept 32-bit domain numbers
The parsing code was bailing on domains greater than UINT16_MAX,
but domain numbers like that are still valid and present on some systems.
One example is Intel VMD (Volume Management Device), which acts somewhat
as a software-managed PCI switch and its upstream linux driver assigns
all downstream devices a PCI domain of 0x10000.
Parsing a BDF like 10000:01:00.0 was failing before. To fix it, increase
the upper limit of domain number to UINT32_MAX. This matches the size of
struct rte_pci_addr->domain (uint32).
Ting Xu [Tue, 12 May 2020 10:13:56 +0000 (10:13 +0000)]
app/testpmd: fix DCB set
When set DCB in testpmd, there is a segmentation fault. It is
because the local variable rss_conf in get_eth_dcb_conf()
is not cleared, so that the pointer member variable rss_key has
a random address, which leads to an error in the following
processing. This patch initialized the local variable rss_conf
to avoid this situation.
Bing Zhao [Fri, 15 May 2020 11:44:44 +0000 (19:44 +0800)]
net/mlx5: fix port action resource initialization
After memory optimization, the organization of some resources are
changed from pointer based LIST to the index based ILIST. A lot of
code parts are touched due to such change.
Some static code checking and analysis tool will complain and raise
a false warning on the uninitialized value using. E.g. in the port
action registering function, the stack variable will be used as the
right value with some uninitialized field to initialize variable
allocated from heap. But indeed, it is not an error because all the
fields set with the uninitialized value will be overwritten in the
following code part and the macros. All the fields will be used as
the left value explicitly.
It makes no sense to clear the stack variable to 0 in this case,
and the extra memset will introduce some cycles overhead. It just
needs to ignore the false warning from the tool, if any.
Fixes: f3faf9ea1185 ("net/mlx5: convert port id action to indexed") Signed-off-by: Bing Zhao <bingz@mellanox.com> Reviewed-by: Matan Azrad <matan@mellanox.com> Reviewed-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Bing Zhao [Fri, 15 May 2020 11:44:43 +0000 (19:44 +0800)]
net/mlx5: fix port action assert timing
After memory optimization, some action object handles are changed to
index to save the overhead. Assertion in debug mode will be helpful
for trouble shooting.
In the current implementation, only one port action is supported in
switchdev mode for one device flow. In debug mode, an assertion will
be used to check the if the port action is none, and it should
locate before the port action resource registration but not after
it. The action index in the handle should be 0 before registration.
Or else it will always cause a failure because the port action is
registered and the index is not 0.
Fixes: f3faf9ea1185 ("net/mlx5: convert port id action to indexed") Signed-off-by: Bing Zhao <bingz@mellanox.com> Reviewed-by: Matan Azrad <matan@mellanox.com> Reviewed-by: Suanming Mou <suanmingm@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
common/mlx5: fix netlink buffer allocation from stack
The buffer size to receive netlink reply messages is relatively
large (32K), and it is allocated on the stack and it might
break in application is using smaller per-thread stacks.
This patch allocates temporary buffer from heap.
Fixes: ccdcba53a3f4 ("net/mlx5: use Netlink to add/remove MAC addresses") Cc: stable@dpdk.org Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com> Acked-by: Matan Azrad <matan@mellanox.com>
The MLX5 device supports up to MLX5_MAX_MAC_ADDRESSES (256) MAC
addresses. The code flushes all MAC devices.
If DPDK is compiled with MLX5_DEBUG this would an assert.
PANIC in mlx5_nl_mac_addr_flush():
line 775 assert "(size_t)(i) < sizeof(mac_own) * 8" failed
The root cause is that mac_own is a pointer and is being used as
a bitmap array. The sizeof(mac_own) would therefore be 64 but the
number of entries to be flushed would be 256.
There is a whole set of asserts in MLX5 netlink code with
the same bug; that should just be changed into proper error checks.
Suanming Mou [Fri, 15 May 2020 08:36:03 +0000 (16:36 +0800)]
net/mlx5: fix shared flow counter lookup
Currently, the shared counter search uses the wrong nested index which
is used by the pool index. The incorrect nested index using causes the
search go to incorrect counter pool is not existed.
Add the counter index to fix the incorrect nested use case.
Bing Zhao [Fri, 15 May 2020 08:13:58 +0000 (16:13 +0800)]
net/mlx5: fix doorbell bitmap management offsets
The doorbell record is organized with page and bitmap. When some new
doorbell needs to be associated with a queue, the bit will be set
in the bitmap to indicate the corresponding doorbell occupied. A
counter is used to record the number of doorbell occupied to speed
up the searching.
If the number reaches the maximal value of a pre-defined number of a
page, a new page will be allocated. If not, then the bitmap will be
checked to find a free one.
The LSHIFT and OR (AND NOT) operations are used to update the bitmap
of a page. But 1 will be treated as a signed integer when compiling.
When the shift number is 31, the shifted value will be considered as
negative. Then a wrong extension will be done when setting it to a
64-bits variable. All the upper 32-bits will be set to 1 by such
extension.
Then a wrong offset value will be calculated because of this. The
next 64 bits will be also treated as the bitmap and get corrupted
through the bit set operation.
The immediate value 1 needs to be used as 64 bits width explicitly.
Suanming Mou [Fri, 15 May 2020 00:48:15 +0000 (08:48 +0800)]
net/mlx5: fix Verbs counter pool allocation
When create the Verbs flows with counter, randomly SEGSEV will also
comes. The reason is that the counter pool memory is not allocated
sufficiently and initialized correctly in Verbs case.
As the mlx5_flow_counter array member is moved out of the counter pool
struct, the counter pool memory layout currently contain implicitly
with mlx5_flow_counter, mlx5_age_param(if the pool is an age pool),
mlx5_flow_counter_ext(if the pool is a none batch pool). When allocate
the pool memory, the pool size should be calculated based on the pool
type accordingly.
Currently, for Verbs counter pool, both mlx5_flow_counter and
mlx5_flow_counter_ext need to be taken into account in the pool size.
And the pool type should also be initialized as CNT_POOL_TYPE_EXT.
This patch add the missing size and type for the Verbs counter pool.
Ting Xu [Mon, 11 May 2020 15:27:48 +0000 (15:27 +0000)]
net/iavf: fix RXDID setting for Rx queue
CVL kernel PF configures all reserved queues for VF, including
Rx queue RXDID. The number of reserved queues is the maximum
between Tx and Rx queues. If the number of the enabled Rx queues
is less than that of reserved queues, required RXDID will only
be set for those enabled, but default value (0) is set for others.
However, RXDID 0 (legacy 16byte descriptor) is not supported now,
PF will return error when configuring those disabled VF queues.
In this patch, required RXDID is set for all reserved Rx queues,
no matter enabled or not. In this way, PF will configure Rx
queues correctly without reporting error.
Fixes: b8b4c54ef9b0 ("net/iavf: support flexible Rx descriptor in normal path") Signed-off-by: Ting Xu <ting.xu@intel.com> Tested-by: Xiaoxiao Zeng <xiaoxiaox.zeng@intel.com> Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Ajit Khaparde [Fri, 15 May 2020 08:31:59 +0000 (14:01 +0530)]
net/bnxt: fix allocation of flow stat related structs
Consolidate flow stat related structs for performance improvement.
The intention of this patch is to reduce the size struct bnxt which
had grown because of recent changes and was impacting performance.
Rasesh Mody [Fri, 15 May 2020 06:34:19 +0000 (23:34 -0700)]
net/qede: fix port reconfiguration
This patch fixes deallocation of all fastpath resources unconditionally,
when re-configuring the device. When re-allocating resources PMD depends
on application to explicitly setup the Rx/Tx queue.
Deallocation of all the resources is only required if the Rx/Tx queue
configuration changes. For other scenarios like KNI MTU change we'd keep
existing configuration.
Fixes: 8de0c4201926 ("net/qede: fix odd number of queues usage in 100G mode") Fixes: dd28bc8c6ef4 ("net/qede: fix VF port creation sequence") Cc: stable@dpdk.org Signed-off-by: Rasesh Mody <rmody@marvell.com> Signed-off-by: Igor Russkikh <irusskikh@marvell.com>
Dharmik Thakkar [Fri, 8 May 2020 22:38:25 +0000 (22:38 +0000)]
app/testpmd: print clock with CPU cycles per packet
On aarch64 platforms, the cycles are counted using either a
low-resolution generic counter or a high-resolution PMU cycle counter.
Print the clock frequency along with CPU cycles/packet to identify which
cycle counter is being used.
Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Reviewed-by: Phil Yang <phil.yang@arm.com> Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Jin Yu [Fri, 15 May 2020 14:45:02 +0000 (22:45 +0800)]
examples/vhost_blk: fix not null-terminated buffer
In vhost_blk_bdev_construct: The string buffer may not have
a null terminator if the source string's length is equal to
the buffer size.
Fixes: 91d3e2d42997 ("examples/vhost_blk: refactor to increase readability") Signed-off-by: Jin Yu <jin.yu@intel.com> Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
Marvin Liu [Fri, 15 May 2020 02:35:07 +0000 (10:35 +0800)]
net/virtio: fix unexpected event after reconnect
Event notification should be disabled after virtqueue initialization and
enabled by calling rte_eth_dev_rx_intr_enable later. When virtio user
device reconnecting to vhost, virtqueue_disable_intr should be called to
disable event notification.
Fixes: 6ebbf4109f35 ("net/virtio-user: fix packed ring server mode") Cc: stable@dpdk.org Signed-off-by: Marvin Liu <yong.liu@intel.com> Acked-by: Xiao Wang <xiao.w.wang@intel.com> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
vhost buffer allocation is successful for packets that fit
into a linear buffer. If it fails, vhost library is expected
to drop the current packet and skip to the next.
The patch fixes the error scenario by skipping to next packet.
Note: Drop counters are not currently supported.
Fixes: c3ff0ac70acb ("vhost: improve performance by supporting large buffer") Cc: stable@dpdk.org Signed-off-by: Sivaprasad Tummala <sivaprasad.tummala@intel.com> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Xiaoyun Wang [Thu, 14 May 2020 09:29:19 +0000 (17:29 +0800)]
net/hinic: optimize RSS RETA table update
Before updating RSS indirection table, firstly determine whether
rq num in RETA table is legal, if it is invalid(such as exceeding
the maximum rxq num), driver will not update hw indirection
table and return fail.
Signed-off-by: Xiaoyun Wang <cloud.wangxiaoyun@huawei.com>
Xiaoyun Wang [Thu, 14 May 2020 09:29:18 +0000 (17:29 +0800)]
net/hinic: update code style
The patch modifies the comments of structures or functions, and adds
space for comments, removes extra empty lines and adjusts the
print level for VF branches in one sdk interface.
Signed-off-by: Xiaoyun Wang <cloud.wangxiaoyun@huawei.com>
Xiaoyun Wang [Thu, 14 May 2020 09:29:16 +0000 (17:29 +0800)]
net/hinic: fix queues resource free
Adds tx_queues and rx_queues non-null judgment before free tx or
rx resources, because some app may set tx_queues or rx_queues to
be null before call free resource interfaces, which may cause
a segfault.
The VMXNet3 protocol has a start-of-packet (SOP) and end-of-packet (EOP)
marker. If there was a bug where mbuf arrived without SOP the code that
chains the mbuf would dereference a null pointer.
Also, record any mbuf's dropped in statistics.
Although did the initial code no longer have access to VMware.
Compile tested only!
Coverity issue: 124563 Fixes: 8ee787ce80a8 ("vmxnet3: remove asserts that confuse coverity") Cc: stable@dpdk.org Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Yong Wang <yongwang@vmware.com>
Dekel Peled [Wed, 13 May 2020 19:49:19 +0000 (22:49 +0300)]
net/mlx5: fix VLAN flow action with wildcard VLAN item
Previous patch added support of VLAN item without VLAN ID value,
i.e. using wildcard VLAN item, to match VLAN with any VLAN ID.
The implication on VLAN actions was not taken into consideration.
VLAN actions (e.g. push vlan) use the VLAN ID value in the VLAN item,
and expect it to be valid.
This patch updates function flow_dev_get_vlan_info_from_items() to
check the VLAN item contents before trying to use it.
Fixes: 92818d839e8e ("net/mlx5: fix match on empty VLAN item in DV mode") Cc: stable@dpdk.org Signed-off-by: Dekel Peled <dekelp@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Matan Azrad [Tue, 12 May 2020 12:52:13 +0000 (12:52 +0000)]
net/mlx5: fix flow counter container resize
The design of counter container resize used double buffer algorithm in
order to synchronize between the query thread to the control thread.
When the control thread detected resize need, it created new bigger
buffer for the counter pools in a new container and change the container
index atomically.
In case the query thread had not detect the previous resize before a new
one need was detected by the control thread, the control thread returned
EAGAIN to the flow creation API used a COUNT action.
The rte_flow API doesn't allow unblocked commands and doesn't expect to
get EAGAIN error type.
So, when a lot of flows were created between 2 different periodic
queries, 2 different resizes might try to be created and caused EAGAIN
error.
This behavior may blame flow creations.
Change the synchronization way to use lock instead of double buffer
algorithm.
The critical section of this lock is very small, so flow insertion
rate should not be decreased.
Fixes: ebbac312e448 ("net/mlx5: resize a full counter container") Cc: stable@dpdk.org Signed-off-by: Matan Azrad <matan@mellanox.com> Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Thomas Monjalon [Mon, 2 Mar 2020 23:08:53 +0000 (00:08 +0100)]
doc: add required Linux capabilities in mlx5 guide
If running DPDK as non-root, some extra capabilities may be required.
The Mellanox devices, using a bifurcated model with Linux drivers,
have some specific requirements summarized in mlx5 PMD guide.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net> Reviewed-by: Raslan Darawsheh <rasland@mellanox.com>
Rasesh Mody [Thu, 14 May 2020 04:09:02 +0000 (21:09 -0700)]
net/qede: fix link state configuration
Move link state enable/disable to dev_start() and dev_stop()
respectively. This will ensure when devices are stopped,
link status will be appropriately shown as down.
Jeff Guo [Wed, 13 May 2020 20:21:27 +0000 (16:21 -0400)]
common/iavf: fix RSS algorithm type
A typo need to be fixed for Simple XOR algorithm configuration.
Fixes: 57094d594324 ("common/iavf: support advanced RSS input set change") Signed-off-by: Jeff Guo <jia.guo@intel.com> Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Shiri Kuzin [Tue, 12 May 2020 12:21:46 +0000 (15:21 +0300)]
common/mlx5: disable relaxed ordering in unsuitable CPUs
Relaxed ordering is a PCI optimization that enables reordering
reads/writes in order to improve performance.
Relaxed ordering was enabled for all processors causing
a degradation in performance in Haswell and Broadwell processors
that don't support this optimization.
In order to avoid that we check if the processor is Haswell
or Broadwell and if so we disable relaxed ordering.
Shougang Wang [Wed, 13 May 2020 02:54:50 +0000 (02:54 +0000)]
net/i40e: fix flow director MAC rule creation
This patch fixes the issue that the flow which both source MAC mask
and destination MAC mask are all zeros can not be created.
Fixes: ea0c22fd8227 ("net/i40e: enable MAC address as flow director input set") Signed-off-by: Shougang Wang <shougangx.wang@intel.com> Tested-by: Hailin Xu <hailinx.xu@intel.com> Reviewed-by: Jeff Guo <jia.guo@intel.com>
Guinan Sun [Tue, 12 May 2020 08:13:06 +0000 (08:13 +0000)]
net/ixgbe: fix statistics in flow control mode
The register autoneg can't be updated synchronously with flow control
mode setting in the state of port start, so NIC statistics error occurs.
AUTO_NEG Advt register should be updated by ixgbe_setup_fc() when
enabling flow control.
The patch fixes the issue.
Fixes: a524f550da6e ("net/ixgbe: fix flow control mode setting") Cc: stable@dpdk.org Signed-off-by: Guinan Sun <guinanx.sun@intel.com> Reviewed-by: Wei Zhao <wei.zhao1@intel.com>
git.droids-corp.org / dpdk.git / log