Qi Zhang [Mon, 22 Oct 2018 06:15:16 +0000 (14:15 +0800)]
eal: fix bus name read for removal in multi-process
A crash may appear when removing some PCI devices because
dev->devargs is not always initialized. So use dev->bus instead of
dev->devargs->bus when building devargs string to remove a device.
Fixes: 244d5130719c ("eal: enable hotplug on multi-process") Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Qi Zhang [Mon, 22 Oct 2018 05:47:11 +0000 (13:47 +0800)]
bus/vdev: fix uninitialized device bus
Device bus should be initialized after bus scan.
While it does not happened when scan vdev from secondary process,
that cause segment fault at rte_dev_probe when call dev->bus->xxx.
Fixes: cdb068f031c6 ("bus/vdev: scan by multi-process channel") Cc: stable@dpdk.org Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Thomas Monjalon <thomas@monjalon.net>
Anatoly Burakov [Fri, 5 Oct 2018 08:29:44 +0000 (09:29 +0100)]
mem: improve segment list preallocation
Current code to preallocate segment lists is trying to do
everything in one go, and thus ends up being convoluted,
hard to understand, and, most importantly, does not scale beyond
initial assumptions about number of NUMA nodes and number of
page sizes, and therefore has issues on some configurations.
Instead of fixing these issues in the existing code, simply
rewrite it to be slightly less clever but much more logical, and
provide ample comments to explain exactly what is going on.
We cannot use the same approach for 32-bit code because the
limitations of the target dictate current socket-centric
approach rather than type-centric approach we use on 64-bit
target, so 32-bit code is left unmodified. FreeBSD doesn't
support NUMA so there's no complexity involved there, and thus
its code is much more readable and not worth changing.
Anatoly Burakov [Thu, 4 Oct 2018 10:20:39 +0000 (11:20 +0100)]
eal: improve musl compatibility of thread log
Musl complains about pthread id being of wrong size, because on
musl, pthread_t is a struct pointer, not an unsigned int. Fix the
printing code by casting pthread id to unsigned pointer type and
adjusting the format specifier to be of appropriate size.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Anatoly Burakov [Thu, 4 Oct 2018 10:20:38 +0000 (11:20 +0100)]
eal: improve musl compatibility of string functions
Musl wraps various string functions such as strlcpy in order to
harden them. However, the fortify wrappers are included without
including the actual string functions being wrapped, which
throws missing definition compile errors. Fix by including
string.h in string functions header.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Anatoly Burakov [Thu, 4 Oct 2018 10:20:33 +0000 (11:20 +0100)]
mk: build with _GNU_SOURCE defined by default
We use _GNU_SOURCE all over the place, but often times we miss
defining it, resulting in broken builds on musl. Rather than
fixing every library's and driver's and application's makefile,
fix it by simply defining _GNU_SOURCE by default for all
builds.
Remove all usages of _GNU_SOURCE in source files and makefiles,
and also fixup a couple of instances of using __USE_GNU instead
of _GNU_SOURCE.
Thomas Monjalon [Wed, 17 Oct 2018 23:42:52 +0000 (01:42 +0200)]
devargs: fix freeing during device removal
After calling unplug function of a bus, the device is expected
to be freed. It is too late for getting devargs to remove.
Anyway, the buses which implement unplug are already freeing
the devargs, except the PCI bus.
So the call to rte_devargs_remove() is removed from EAL and
added in PCI.
Fixes: 2effa126fbd8 ("devargs: simplify parameters of removal function") Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Raslan Darawsheh [Wed, 17 Oct 2018 15:22:11 +0000 (18:22 +0300)]
app/testpmd: set packet dump based on verbosity level
when changing verbosity level it will configure rx/tx callbacks to dump
packets based on the verbosity value as following:
1- dump only received packets:
testpmd> set verbose 1
2- dump only sent packets:
testpmd> set verbose 2
3- dump sent and received packets:
testpmd> set verbose (any number > 2)
4- disable dump
testpmd> set verbose 0
Signed-off-by: Raslan Darawsheh <rasland@mellanox.com> Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Phil Yang [Wed, 17 Oct 2018 01:36:30 +0000 (09:36 +0800)]
app/testpmd: fix physical port socket initialization
Once the lcore list setting excluded the socket which physical device
attached, it will cause failure. Meanwhile, it will disable Testpmd
cross NUMA scenario.
Fixes: dbfb8ec7094c ("app/testpmd: optimize mbuf pool allocation") Cc: stable@dpdk.org Signed-off-by: Phil Yang <phil.yang@arm.com> Reviewed-by: Gavin Hu <gavin.hu@arm.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Shahaf Shuler [Tue, 16 Oct 2018 06:05:17 +0000 (09:05 +0300)]
net/mlx5: fix build on Arm
On some ARM environment, the below compilation error will be seen
dpdk/drivers/net/mlx5/mlx5_flow_dv.c: In function
'flow_dv_translate_item_nvgre':
/tmp/dpdk/drivers/net/mlx5/mlx5_flow_dv.c:785:22: error: pointer targets
in initialization differ in signedness [-Werror=pointer-sign]
const char *tni_v = nvgre_v->tni;
The reason for this error is that nvgre_v->tni is defined as byte array
in size of 3B. However the code in the function iterate till the 4B in
order to copy/set also the subsequent field after it (flow_id)
Fixing by pointing to this struct from a different pointer.
Fixes: fc2c498ccb94 ("net/mlx5: add Direct Verbs translate items") Signed-off-by: Shahaf Shuler <shahafs@mellanox.com> Acked-by: Ori Kam <orika@mellanox.com>
Postcopy live-migration feature requires the application to
not populate the guest memory. As the vhost library cannot
prevent the application to that (e.g. preventing the
application to call mlockall()), the feature is disabled by
default.
The application should only enable the feature if it does not
force the guest memory to be populated.
In case the user passes the RTE_VHOST_USER_POSTCOPY_SUPPORT
flag at registration but the feature was not compiled,
registration fails.
For the same reason, postcopy and dequeue zero copy features
are not compatible, so don't advertize postcopy support if
dequeue zero copy is requested.
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Maxime Coquelin [Fri, 12 Oct 2018 12:40:42 +0000 (14:40 +0200)]
vhost: avoid useless VhostUserMemory copy
The VHOST_USER_SET_MEM_TABLE payload is copied when handled,
whereas it could directly be referenced.
This is not very important, but next, we'll need to update the
payload and send it back to Qemu.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com> Acked-by: Ilya Maximets <i.maximets@samsung.com>
Maxime Coquelin [Fri, 12 Oct 2018 12:40:34 +0000 (14:40 +0200)]
vhost: define postcopy protocol flag
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com> Acked-by: Ilya Maximets <i.maximets@samsung.com> Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
Maxime Coquelin [Fri, 12 Oct 2018 12:40:32 +0000 (14:40 +0200)]
vhost: fix payload size of reply
QEMU doesn't expect any payload for the reply of
VHOST_USER_SET_LOG_BASE request, so don't send any.
Note that the Vhost-user specification isn't clear about
it and would need to be fixed.
Maxime Coquelin [Fri, 12 Oct 2018 12:40:31 +0000 (14:40 +0200)]
vhost: clarify reply-ack in case a reply was already sent
For messages that require a reply, a second ack should not be
sent when reply-ack protocol feature is negotiated, even if
the corresponding flag is set in the message.
The code is compliant with the spec but it isn't clear it is,
so this patch adds a comment to make it explicit.
Maxime Coquelin [Fri, 12 Oct 2018 12:40:30 +0000 (14:40 +0200)]
vhost: fix return code of messages requiring replies
VHOST_USER_GET_PROTOCOL_FEATURES, VHOST_USER_GET_VRING_BASE
and VHOST_USER_SET_LOG_BASE require replies, so their handlers
should return VH_RESULT_REPLY, not VH_RESULT_OK.
Fixes: 0bff510b5ea6 ("vhost: unify message handling function signature") Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com> Acked-by: Ilya Maximets <i.maximets@samsung.com> Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
Maxime Coquelin [Fri, 12 Oct 2018 12:40:29 +0000 (14:40 +0200)]
vhost: fix messages results handling
Return of message handling has now changed to an enum that can
take non-negative value that is not zero in case a reply is
needed. But the code checking the variable afterwards has not
been updated, leading to success messages handling being
treated as errors.
External post and pre callbacks return type needs also to be
changed to the new enum, so that its handling is consistent.
This is done in this patch alongside with the convertion of
its only user, vhost-crypto backend.
Fixes: 0bff510b5ea6 ("vhost: unify message handling function signature") Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Tiwei Bie [Thu, 11 Oct 2018 14:22:34 +0000 (22:22 +0800)]
vhost: fix notification for packed ring
The notification can't be disabled in packed ring when
application tries to disable notification, because the
device event flags field is overwritten by an unexpected
value. This patch fixes this issue.
Fixes: b1cce26af1dc ("vhost: add notification for packed ring") Cc: stable@dpdk.org Signed-off-by: Tiwei Bie <tiwei.bie@intel.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Acked-by: Jason Wang <jasowang@redhat.com>
The corresponding rte_flow_item_eth must be present in
rte_flow pattern
Only support modify outer layer MAC address
The example testpmd command is:
flow create 0 transfer ingress
pattern eth / ipv4 / udp dst is 7000 / end
actions set_mac_dst mac_addr dd:00:aa:11:bb:33 /
set_mac_src mac_addr bb:00:cc:11:aa:22 /
port_id id 1 / end
Signed-off-by: Xiaoyu Min <jackmin@mellanox.com> Acked-by: Yongseok Koh <yskoh@mellanox.com>
Alejandro Lucero [Mon, 15 Oct 2018 14:11:02 +0000 (15:11 +0100)]
ethdev: add field for device data per process
Primary and secondary processes share a per-device private data. With
current design it is not possible to have data per-device per-process.
This is required for handling properly the CPP interface inside the NFP
PMD with multiprocess support.
There is also at least another PMD driver, tap, with similar
requirements for per-process device data.
Xiaoyun Li [Mon, 15 Oct 2018 04:36:34 +0000 (00:36 -0400)]
doc: add ixgbe known issue with legacy interrrupt
When using uio_pci_generic module or using legacy interrupt mode of
igb_uio or vfio, X550 cannot get interrupts. Because the Interrupt
Status bit is not implemented, then the irq cannot be handled correctly
and cannot report the event fd to DPDK apps.
Add this hw limitation and details into ixgbe known issue.
Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com> Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Rahul Lakkireddy [Mon, 15 Oct 2018 14:05:13 +0000 (19:35 +0530)]
net/cxgbe: add action to swap MAC addresses
Implement RTE_FLOW_ACTION_TYPE_MAC_SWAP to offload flows with
action to swap the source and destination MAC addresses in the
outermost Ethernet header.
Phil Yang [Fri, 12 Oct 2018 09:34:55 +0000 (17:34 +0800)]
app/testpmd: fix vdev socket initialization
The cmdline settings of port-numa-config and rxring-numa-config have
been flushed by the following init_config. If we don't configure the
port-numa-config, the virtual device will allocate the device ports to
socket 0. It will cause failure when the socket 0 is unavailable.
Now that logging of initialization is controlled by dynamic debug
level, remove the #ifdef for RTE_LIBRTE_QEDE_DEBUG_INIT since that
config option does not exist in common_base and is not documented.
Yongseok Koh [Fri, 12 Oct 2018 08:42:33 +0000 (08:42 +0000)]
net/mlx5: support multiple groups and jump action
rte_flow has 'group' attribute and 'jump' action in order to support
multiple groups. This feature is known as multi-table support ('chain' in
linux TC flower) in general because a group means a table of flows. Example
commands are:
flow create 0 transfer priority 1 ingress
pattern eth / vlan vid is 100 / end
actions jump group 1 / end
flow create 0 transfer priority 1 ingress
pattern eth / vlan vid is 200 / end
actions jump group 2 / end
flow create 0 transfer group 1 priority 2 ingress
pattern eth / vlan vid is 100 /
ipv4 dst spec 192.168.40.0 dst prefix 24 / end
actions drop / end
flow create 0 transfer group 1 priority 2 ingress
pattern end
actions of_pop_vlan / port_id id 1 / end
flow create 0 transfer group 2 priority 2 ingress
pattern eth / vlan vid is 200 /
ipv4 dst spec 192.168.40.0 dst prefix 24 / end
actions of_pop_vlan / port_id id 2 / end
flow create 0 transfer group 2 priority 2 ingress
pattern end
actions port_id id 2 / end
With theses flows, if a packet having vlan 200 and src_ip as 192.168.40.1,
this packet will firstly hit the 1st flow. Then it will hit the 5th flow
because of the 'jump' action. As a result, the packet will be forwarded to
port 2 (VF representor) with vlan tag being stripped off. If the packet had
vlan 100 instead, it would be dropped by the 3rd flow.
Igor Russkikh [Fri, 12 Oct 2018 11:09:14 +0000 (11:09 +0000)]
net/atlantic: implement firmware operations
AQC NICs comes in fields with two major
FW generations: 1x and 3x.
This is part of linux atlantic driver shared code,
responsible for internal NIC firmware interactions,
including link management ops, FW initialization,
various lifecycle features.
Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: Pavel Belous <pavel.belous@aquantia.com>
John Daley [Wed, 10 Oct 2018 22:11:31 +0000 (15:11 -0700)]
net/enic: fix counter action
- track whether counter DMAs are active or not so they can be stopped if
needed before DMA memory is freed
- fix counter DMA shut-down by changing vnic_dev_counter_dma_cfg() to
take the number of counters to DMA instead of high counter index and
use num counters = 0 to shut off DMAs
- remove unnecessary checks that DMA counter memory is valid and that
counter DMAs are in use
- change the minimum DMA period to match what 1400 series adapter is
capable of
- fix comments and change a couple variable names to make more sense
Fixes: 86df6c4e2fce ("net/enic: support flow counter action") Signed-off-by: John Daley <johndale@cisco.com> Reviewed-by: Hyong Youb Kim <hyonkim@cisco.com>
Fiona Trahe [Thu, 11 Oct 2018 17:14:05 +0000 (18:14 +0100)]
common/qat: fix failure to create PMD
If QAT crypto pmd failed to be created due to reaching MAX
cryptodevs it prevented QAT comp PMD being created. And vice versa.
Change to warning in these cases and allow the other PMD to be created.
Fixes: c0c90bc4cade ("compress/qat: add create and destroy functions") Cc: stable@dpdk.org Signed-off-by: Fiona Trahe <fiona.trahe@intel.com> Acked-by: Tomasz Cel <tomaszx.cel@intel.com>
Fan Zhang [Mon, 15 Oct 2018 11:28:16 +0000 (12:28 +0100)]
cryptodev: fix library version
This patch fixes the cryptodev library version number that was
missed updating in DPDK 18.08.
Fixes: a4493be5bdfa ("cryptodev: replace bus specific struct with generic dev") Cc: stable@dpdk.org Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Hemant Agrawal [Fri, 12 Oct 2018 14:40:54 +0000 (20:10 +0530)]
crypto/caam_jr: add security offload
This patch provides the support for protocol offload
to the hardware. following security operations are
added:
- session_create
- session_destroy
- capabilities_get
Gagandeep Singh [Fri, 12 Oct 2018 14:40:51 +0000 (20:10 +0530)]
crypto/caam_jr: add enqueue/dequeue operations
This patch add support for :
1. creating run time sec hw decriptors for a given request.
2. enqueue operation to the caam jr ring
3. dequeue operation from the caam jr ring in poll mode
4. creating a crypto protocol descriptor for session - first time.
git.droids-corp.org / dpdk.git / log