The commit cb6696d220 ("drivers: update registration macro usage")
changes the name from virtio-user to virtio_user, because hyphen
cannot be used in a C symbol name. However, this commit does not
update the strings in docs and source code, which could lead to
failure to start this device as per the docs.
This patch updates related strings in the docs and source code.
Fixes: cb6696d220 ("drivers: update registration macro usage")
Reported-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Sometimes app just wants to update the RSS hash function and no RSS key
update is needed, but fm10k pmd will return EINVAL for this case.
If the rss_key is NULL, we don't need to check the rss_key_len.
Fixes: 57033cdf8f ("fm10k: add PF RSS")
Reported-by: Xueqin Lin <xueqin.lin@intel.com>
Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Rx loop inside _recv_raw_pkts_vec() ignores nb_pkts argument and always
tries to receive RTE_I40E_VPMD_RX_BURST (32) packets. This is a violation
of rte_eth_rx_burst() API and can lead to memory corruption (out-of-bounds
writes to struct rte_mbuf **rx_pkts) if nb_pkts is less than 32.
Fix this by actually using nb_pkts inside the loop.
Fixes: 9ed94e5bb0 ("i40e: add vector Rx")
Signed-off-by: Sergey Dyasly <s.dyasly@samsung.com>
Acked-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Adam Bynes <adambynes@outlook.com>
The configure function enicpmd_dev_configure() was not paying attention
to the rxmode VLAN strip bit. Set the VLAN strip mode according to the bit.
Fixes: fefed3d1e6 ("enic: new driver")
Signed-off-by: John Daley <johndale@cisco.com>
Reviewed-by: David Harton <dharton@cisco.com>
Tested-by: David Harton <dharton@cisco.com>
Initialize the mbuf data offset to RTE_PKTMBUF_HEADROOM as the
enic takes ownership of them. If allocated mbufs had some offset
other than RTE_PKTMBUF_HEADROOM, the application would read mbuf
data starting at the wrong place and misinterpret the packet.
Fixes: 856d7ba7ed ("net/enic: support scattered Rx")
Reviewed-by: Nelson Escobar <neescoba@cisco.com>
Signed-off-by: John Daley <johndale@cisco.com>
drivers/net/ena/base/ena_com.c(346):
error #3656: variable "dev_node" may be used before its value is set
ENA_MEM_ALLOC_COHERENT_NODE(ena_dev->dmadev,
^
drivers/net/ena/base/ena_com.c(399):
error #3656: variable "prev_node" may be used before its value is set
ENA_MEM_ALLOC_COHERENT_NODE(ena_dev->dmadev,
^
Fixes: 3d3edc265f ("net/ena: make coherent memory allocation NUMA-aware")
Reported-by: Eoin Breen <eoin.breen@intel.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Jan Medala <jan@semihalf.com>
The rxq/txq for the queue_release callback could be NULL, say when
rte_eth_dev_configure() fails that the queue is not setup at all.
Do a simple NULL check would fix the crash issue.
Fixes: 01ad44fd37 ("net/virtio: split Rx/Tx queue")
Reported-by: Olivier Matz <olivier.matz@6wind.com>
Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
The support of virtio-user changed the way the mbuf dma address is
retrieved, using a physical address in case of virtio-pci and a virtual
address in case of virtio-user.
This change introduced some possible memory corruption in packets,
replacing:
m->buf_physaddr + RTE_PKTMBUF_HEADROOM
by:
m->buf_physaddr + m->data_off (through a macro)
This patch fixes this issue, restoring the original behavior.
By the way, it also rework the macros, adding a "VIRTIO_" prefix and
API comments.
Fixes: f24f8f9fee ("net/virtio: allow virtual address to fill vring descriptors")
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Jianfeng Tan <jianfeng.tan@intel.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Currently while calling of 'rte_vhost_driver_unregister()' connection
to QEMU will not be closed. This leads to inability to register driver
again and reconnect to same virtual machine.
This scenario is reproducible with OVS. While executing of the following
command vhost port will be re-created (will be executed
'rte_vhost_driver_register()' followed by 'rte_vhost_driver_unregister()')
network will be broken and QEMU possibly will crash:
ovs-vsctl set Interface vhost1 ofport_request=15
Fix this by closing all established connections on driver unregister and
removing of pending connections from reconnection list.
Fixes: 64ab701c3d ("vhost: add vhost-user client mode")
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
If something abnormal happened to QEMU, 'connect()' can block calling
thread (e.g. main thread of OVS) forever or for a really long time.
This can break whole application or block the reconnection thread.
Example with OVS:
ovs_rcu(urcu2)|WARN|blocked 512000 ms waiting for main to quiesce
(gdb) bt
#0 connect () from /lib64/libpthread.so.0
#1 vhost_user_create_client (vsocket=0xa816e0)
#2 rte_vhost_driver_register
#3 netdev_dpdk_vhost_user_construct
#4 netdev_open (name=0xa664b0 "vhost1")
[...]
#11 main
Fix that by setting non-blocking mode for client sockets for connection.
Fixes: 64ab701c3d ("vhost: add vhost-user client mode")
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
After doing a driver callout to fill in the driver specific
parts of struct rte_eth_stats, rte_eth_stats_get() overwrites
the rx_nombuf member regardless of whether the driver itself
has assigned a value. Any driver-assigned value should take
priority.
Fixes: af75078fec ("first public release")
Signed-off-by: Remy Horton <remy.horton@intel.com>
The autotest lists, requirements and distribution needs a big rework
to reduce the amount of cores and memory required.
The root cause is not addressed yet.
This patch just increase some memory allocation for some greedy tests
which often fail because of memory fragmentation:
LPM6 and reentrancy tests in groups 3 and 6 respectively.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Tested-by: Olivier Matz <olivier.matz@6wind.com>
The unavailable tests are filtered out by autotest by looking for
the symbols in the binary:
PCI autotest: Skipped [Not Available] [00m 00s]
Malloc autotest: Success [00m 00s]
It results to skip everything if the binary has no symbol (stripped):
PCI autotest: Skipped [Not Available] [00m 00s]
Malloc autotest: Skipped [Not Available] [00m 00s]
This case is handled by getting back to the old behaviour if the binary
has no symbol information:
PCI autotest: Fail [Not found] [00m 00s]
Malloc autotest: Success [00m 00s]
Fixes: d553c8f2b1 ("app/test: filter out unavailable tests")
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Tested-by: Olivier Matz <olivier.matz@6wind.com>
test_mempool func not using pointer variable 'mp_ext' and incorrectly
freed. So removing ptr var. Now freeing mp_stack var.
Fixes: ea151eb48a ("app/test: migrate custom mempool handler to stack handler")
Signed-off-by: Santosh Shukla <santosh.shukla@caviumnetworks.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
rte_ring_create expects the size of the ring to
be a power of 2. REFCNT_RING_SIZE value is not
power of 2 in-case if RTE_MAX_LCORE == 96.
Fix it by aligning the size to next power of 2 value.
Fixes: af75078f ("first public release")
Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Consumer queue dequeuing must be guaranteed to be done fully before
the tail is updated. This is not guaranteed with a read barrier,
changed to a write barrier just before tail update which in practice
guarantees correct order of reads and writes.
Signed-off-by: Juhamatti Kuusisaari <juhamatti.kuusisaari@coriant.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The new mempool handler interface forgets to register the free() function
of the ops. Introduced in this patch:
Fixes: 449c49b93a ("mempool: support handler operations")
Signed-off-by: Zoltan Kiss <zoltan.kiss@schaman.hu>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
A recent patch brought up an issue about the size of the 'name' fields:
85cf0079 mem: avoid memzone/mempool/ring name truncation
These relations should be observed:
1. Each ring creates a memzone with a prefixed name:
RTE_RING_NAMESIZE <= RTE_MEMZONE_NAMESIZE - strlen(RTE_RING_MZ_PREFIX)
2. There are some mempool handlers which create a ring with a prefixed
name:
RTE_MEMPOOL_NAMESIZE <= RTE_RING_NAMESIZE - strlen(RTE_MEMPOOL_MZ_PREFIX)
3. A mempool can create up to RTE_MAX_MEMZONE pre and postfixed memzones:
sprintf(postfix, "_%d", RTE_MAX_MEMZONE)
RTE_MEMPOOL_NAMESIZE <= RTE_MEMZONE_NAMESIZE -
strlen(RTE_MEMPOOL_MZ_PREFIX) - strlen(postfix)
Setting all of them to 32 hides this restriction from the application.
This patch decreases the mempool and ring string size to accommodate for
these prefixes, but it doesn't apply the 3rd constraint. Applications
relying on these constants need to be recompiled, otherwise they'll run
into ENAMETOOLONG issues.
The size of the arrays are kept 32 for ABI compatibility, it can be
decreased next time the ABI changes.
Signed-off-by: Zoltan Kiss <zoltan.kiss@schaman.hu>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
(strlen(name) == sizeof(mz->name) - 1) is a valid case, change the
condition to reflect that.
Move it earlier to avoid lookup with invalid name.
Change errno to ENAMETOOLONG.
Fixes: 85cf0079 ("mem: avoid memzone/mempool/ring name truncation")
Signed-off-by: Zoltan Kiss <zoltan.kiss@schaman.hu>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
On weak memory order architecture like POWER, rte_smp_wmb/rte_smp_rmb
need to use CPU instructions, not compiler barrier. This patch fixes
this. Also, to improve performance on PPC64, use light weight sync
instruction instead of sync instruction.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
The sed syntax of '0,/regexp/' is GNU specific and fails with
non GNU sed in FreeBSD.
To solve the issue we can use awk instead to remove duplicates.
The awk script basically keeps the last config value, while
maintaining order and comments from original config file.
Fixes: b2063f104d ("mk: filter duplicate configuration entries")
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Acked-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>
There is an error when linking static EAL library with an application:
eal_alarm.c:(.text+0xd7): undefined reference to `clock_gettime'
eal_alarm.c:(.text+0x20f): undefined reference to `clock_gettime'
eal_timer.c:(.text+0x108): undefined reference to `clock_gettime'
eal_timer.c:(.text+0x146): undefined reference to `clock_gettime'
The function clock_gettime() is in librt for old glibc.
Fixes: 281948b475 ("mk: fix missing librt dependencies")
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Tested-by: Yongjie Gu <yongjiex.gu@intel.com>
The correct mailing list dev@dpdk.org, not dev@dpkg.org.
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Reviewed-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Added a missing note about dependencies on libpcap and
CONFIG_RTE_LIBRTE_PMD_PCAP flag that pdump tool has.
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Fixed default socket path name "/var/run" to "/var/run/.dpdk" and
"$HOME" to "~/.dpdk".
Fixes: 278f945402 ("pdump: add new library for packet capture")
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Since users of the pdump library and tool can chose to have their own
server and client paths, it is must for the pdump tool to use the same
server socket path that was used by primary application while
initializing packet capture framework by rte_pdump_init() or
rte_pdump_set_socket_dir() APIs.
To pass the socket path info to pdump tool a new optional command
line options "server-socket-path" and "client-socket-path" are added.
"client-socket-path" is also added, if the users want to have client
sockets in their own defined paths.
Updated pdump tool guide with the new changes.
Fixes: caa7028276 ("app/pdump: add tool for packet capturing")
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The changes include
1)If mkdir fails for user passed socket paths log error and return.
2)At some places return value was set to errno and that non-negative number
was returned, but the intention was to return negative value.
So now rte_errno was set to errno and returning the actual negative value
that the APIs has returned.
Fixes: 278f945402 ("pdump: add new library for packet capture")
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
When sending Inbound non IPSec traffic that matches an Inbound Security
Policy set to Protect, the code will check that the SPI of the packet
and the associated Security Association match.
That check should only be done for IPSec packets and results in SEGFAULT
when done on non IPSec packets.
Fixes: 906257e965 ("examples/ipsec-secgw: support IPv6")
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
This patch improve generate_random_key() function by replacing rand()
function with reading from /dev/urandom.
CID 120136 : Calling risky function (DC.WEAK_CRYPTO)
dont_call: rand should not be used for security related applications, as
linear congruential algorithms are too easy to break
Coverity issue: 120136
Signed-off-by: Piotr Azarewicz <piotrx.t.azarewicz@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
Zero packets can be returned by rte_eth_rx_burst() and
rte_distributor_returned_pkts() inside lcore_rx(), so
for zero packet scenario instead of proceeding to
next operations we should continue to the next iteration of the
loop to avoid unnecessary processing overhead which is causing
rx packets to be dropped and hence distributor failing to forward the
packets.
Fixes: 07db4a97 ("examples/distributor: new sample app")
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Inside flush_fwd_rx_queues removed redundant prev_tsc update with cur_tsc,
as prev_tsc value is always updated with rte_rdtsc() in for loop.
Coverity issue: 127797
Fixes: f487715f36 ("app/testpmd: add timeout in Rx queue flushing")
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
When enic's interrupt handler is called indicating an error, it scans
through the receive queues (RQs) on the adapter looking for errors.
But since the inclusion of rx scatter, some of the RQs may not be in
use, and you shouldn't check them for errors.
Fixes: 856d7ba7ed ("net/enic: support scattered Rx")
Signed-off-by: Nelson Escobar <neescoba@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
The calculation of truncated packets didn't take into account packet
errors due to the adapter not having buffers, causing both the
ipackets, and imissed counts to be wrong if such errors occurred. In
order to properly calculate the number of packets truncated, we need
to subtract the count of errors due to no buffers.
Fixes: c44d9f01ad ("net/enic: count truncated packets")
Signed-off-by: Nelson Escobar <neescoba@cisco.com>
The exported device start and stop functions where not setting the queue
states to RTE_ETH_QUEUE_STATE_STARTED and RTE_ETH_QUEUE_STATE_STOPPED.
After starting the device, the RTE queue stop function would not call
the enic queue stop function since queue was already marked as stopped.
Put queue state updates in the lower level queue start/stop functions
which are called by both device and queue start/stop functions.
Fixes: fefed3d1e6 ("enic: new driver")
Reviewed-by: Nelson Escobar <neescoba@cisco.com>
Signed-off-by: John Daley <johndale@cisco.com>
The check that all Tx and Rx queues were set up was not
adequate when reconfiguring with a different number of
queues. Only the number of completion queues (CQs) was
being used to make the determination, but the CQ array
is shared between the underlying Rx and Tx queues.
Check that the internal Rx, Tx and CQs are all set up
before completing port configuration.
Fixes: fefed3d1e6 ("enic: new driver")
Signed-off-by: John Daley <johndale@cisco.com>
Reviewed-by: Nelson Escobar <neescoba@cisco.com>
One instance of a filter add failure was not incrementing the
the fail counter.
Fixes: 4c2c7bf41f ("net/enic: fix negative array index write")
Signed-off-by: John Daley <johndale@cisco.com>
In the burst Tx cleanup function, the reference count in mbufs
returned to the pool should to be decremented before they are
returned. Decrementing is not done by rte_mempool_put_bulk()
so it must be done separately using __rte_pktmbuf_prefree_seg().
Also when returning unsent buffers when the device is stopped
use rte_mbuf_free_seg() instead of rte_mempool_put() so that
reference counts are properly decremented.
Fixes: 36935afbc5 ("net/enic: refactor Tx mbuf recycling")
Reviewed-by: Nelson Escobar <neescoba@cisco.com>
Signed-off-by: John Daley <johndale@cisco.com>
The enic PMD code has diverged from code that was once
shared with the enic kernel mode driver for performance
reasons. It is confusing and misleading to print the
internal version number. Remove it.
Fixes: fefed3d1e6 ("enic: new driver")
Signed-off-by: John Daley <johndale@cisco.com>
RHEL 7.1's GCC for POWER8 reports the following error in one rte_memcpy()
macro call by mlx5:
error: array subscript is above array bounds [-Werror=array-bounds]
It appears to be a GCC bug which can be worked around by making parentheses
more explicit.
Signed-off-by: Olga Shern <olgas@mellanox.com>
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
In multi-queue + FTAG use case, we need to turn tx_ftag_en on for all
the Tx queues.
Fixes: 7958b1310d ("fm10k: enable FTAG based forwarding")
Reported-by: Ricky Li <ricky.li@intel.com>
Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
