Age | Commit message (Collapse) | Author | Files | Lines |
|
Type: improvement
DBGvpp# set loggin class virtio level debug
DBGvpp# create int virtio 0000:00:03.0 tx-queue-size 1024
show virtio pci
```
Virtqueue (TX) 1
qsz 1024, last_used_idx 0, desc_next 0, desc_in_use 0
avail.flags 0x1 avail.idx 0 used.flags 0x0 used.idx 0
```
show logging
```
2022/07/22 23:20:22:557 debug virtio 0000:00:03.0: tx-queue: number 1, default-size 256
2022/07/22 23:20:22:557 debug virtio 0000:00:03.0: tx-queue: number 1, new size 1024
```
Change-Id: Ib1a3ebe742b3a6c9fe72bd1c5accfe07682cbdd1
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
If a frame should be sent to 'ethernet-input' set the
ETH_INPUT_FRAME_F_SINGLE_SW_IF_IDX flag. It will force 'ethernet-input'
to use a fast-path for such frames.
This patch also aligns the behaviour with other input nodes.
Type: improvement
Change-Id: Icff0fa31204d5304a2ea0a4f4e7bc418dedbfe32
Signed-off-by: Stanislav Zaikin <zstaseg@gmail.com>
|
|
Type: fix
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: Idada695432d2bfac8808f35f1e8cd16f84d963c6
|
|
Type: style
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: Ie02d068122ab8f2c6049754f28722d851ae9b3f1
|
|
Type: improvement
Change-Id: I30e66370c927afeb62ba3a2b3334bdc2a31d4561
Signed-off-by: Benoît Ganne <bganne@cisco.com>
|
|
convert vhost device driver to a plugin as described in
https://jira.fd.io/browse/VPP-2065
Type: improvement
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Ibfe2f351bcaed36a04b136d082ae414145dd37b5
|
|
Type: fix
virtio transmit node uses header offsets to set the
appropriate metadata for packet with offloads. But
l2_hdr_offset is not correctly set by previous node(s).
This patch makes use of curren_data field as l2_hdr_offset.
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: Id2f41a7650be45c9cc1ebacc7bec298be79bf218
|
|
Type: fix
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: I7003162ee72c39c44be1e7ff0d27f4416b3eca5a
|
|
Type: fix
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: Ibb4b17b954c55bcb37ede57d398966f244735c3d
|
|
Type: improvement
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: I21a701a556b88a9d81f0e074a59fa34b3746b1d9
|
|
It may contain garbage in debug builds resulting in wrong
gho detected flags and offsets.
Type: fix
Signed-off-by: Vladislav Grishenko <themiron@yandex-team.ru>
Change-Id: Ia79633262185016f527e7dc6c67334cda6f055f2
|
|
Type: improvement
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: Ie63c360f2e42e9e5799f5c536453823ea95ed3b8
|
|
single struct to hold all api handler, flags, etc.
Provide functions to toggle flags instead of writing directly to
internal data.
Type: refactor
Change-Id: I4730d7290e57489de8eda34a72211527e015b721
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
We allocate vring_spinlock for all vrings in the vui, not just the ones
being used. So when we free the vui, we have to free vring_spinlock for
all vrings, not just the one being used.
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I3951fda71ce6e11474b04302116ea9e08b404758
|
|
Type: fix
set use_custom_mac for args in create_vhost_user_if_v2 API
Add testcase for custom mac-address
Signed-off-by: Fahad Naeem <fahadnaeemkhan@gmail.com>
Change-Id: Iac64d818e0f1e6d36187fe769ee33d202aaafd05
Signed-off-by: Fahad Naeem <fahadnaeemkhan@gmail.com>
|
|
Type: refactor
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: I3788cc857023fafcc8eb6d6ff4524425026a75d8
|
|
Type: refactor
IP4 does not depend on TCP (it's the other way around).
This upside down dependency leads to some nasty circular includes when trying to use ip46_address.h in interface.h
Signed-off-by: Neale Ranns <neale@graphiant.com>
Change-Id: I4a1bd21543b08b9c1cf1e5563da738414734a878
|
|
Coverity complains the expression, j % vif->num_txq, may encounter
divide by zero. While there is little chance that vif->num_txq is
zero, it is easy to prevent divide by zero if vif->num_txq is ever
zero.
Type: fix
Fixes: I337ec63d0868f665329d68eadf1744e080b73a0d
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I2e91f296737ce266ab70fffc1f442cc600724fa2
|
|
Type: fix
During the interface creation time, (by default) admin-up
flag is locally set for tap and virtio interfaces.
While, in VPP the state of these interfaces are still
admin-down. User needs to explicitly call
'set interface state <interface-name> up' to admin-up the
newly created tap or virtio interface(s) in VPP. So, this
behavior is inconsistent.
This patch fixes the issue to have consistent behavior
for given interface between local and global administration
state.
Change-Id: Ifd8904a09fbdbe7b386874ac3231dc0527064518
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: improvement
Signed-off-by: Damjan Marion <damarion@cisco.com>
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: Ie595e69af8657b0ee18a84ac71c5d433108d9ef8
|
|
Prep for supporting multiple callbacks, optional args, etc.
Type: improvement
Change-Id: I96244c098712e8213374678623f12527b0e7f387
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Make it shorter to type, easier to debug, make adding callbacks in
future simpler.
Type: improvement
Change-Id: I6cdd6375e36da23bd452a7c7273ff42789e94433
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Type: improvement
Change-Id: I337ec63d0868f665329d68eadf1744e080b73a0d
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: refactor
Change-Id: Ia13a9cf6480aac280f3d287c59908e84c29c3443
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: fix
Fixes: d96b28ac0917 ("virtio: improve input node performance")
vlib_buffer_t is defined on stack to be used for input feature arc
lookup once per frame call for performance reasons. The definition
is missing the initialization to avoid the assignment of garbage value.
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: Ie7930186bbd8240c44d257808ebd31e06e854b29
|
|
Type: improvement
Change-Id: Iac01d7830b53819ace8f199554be10ab89ecdb97
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
|
|
Type: fix
Tap/Tun interfaces do not have control queue.
This patch removes the support of control queue
from virtio_show() which is used by show tap/tun cli.
Change-Id: Ib89144ad488ed548fb1ce50ee232a1b8659ccf29
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: improvement
Linux uses pseudo header checksum when checksum of l4 is offloaded.
This patch adds similar support in virtual interfaces.
Change-Id: I6a94d1104e59356f95057e7c122e3be9cd8659a3
Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: fix
For some virtio backends, packets don't seem to flow
if all the queues that have been requested are not
initialized
Change-Id: I41765d668497ff954e6d69d36836590947fd93b1
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
|
|
Type: fix
virtio have three different interfaces (tap, tun and native virtio).
virtio_show() is used by CLI commands to show information about these
interfaces. It uses interface type to print interface specific information.
virtio_show() should use proper type for virtio interfaces (TAP, TUN, VIRTIO).
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: I2043dc2cfe2e2f6c7c9348b8494aa4a27cab31f3
|
|
Take advantage of the new TX infra and support manual thread placement
Type: improvement
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Id8023846a2eb59125fcf2f80f4f11eb681cf14dc
|
|
- add format_vnet_buffer and format_vnet_buffer_no_chain to mirror
format_vlib_buffer and format_vlib_buffer_no_chain
- format_vnet_buffer used to be the "no chain" version, replace all of
its current use with the corresponding format_vnet_buffer_no_chain
- add a function to dump vnet buffer details from gdb
Type: improvement
Change-Id: I143ce845f80e7ef937ea33a557b6e3b5988c5b8f
Signed-off-by: Benoît Ganne <bganne@cisco.com>
|
|
vhost interrupt mode support with mq does not work when coalesce frame
is not configured to 0.
When packed-ring is configured, we were also setting the wrong flag
for want-interrupt.
No need to trigger an interrupt to RX infra if there is at least one
queue in the same thread that is doing polling.
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I8e64250f2aa6bf611a777a53e4d2c1076d83305f
|
|
Type: refactor
Change-Id: Id10cbf52e8f2dd809080a228d8fa282308be84ac
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Use autogenerated code.
Does not change API definitions.
Type: improvement
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
Change-Id: I4f9316b16f16a48e2042aa17db596bfd181bd314
|
|
Use autogenerated code.
Does not change API definitions.
Type: improvement
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
Change-Id: I047310b8c9fcc51dcfb187710ff59b7895abe217
|
|
Type: style
reduce the number of files recompiled after changing pg.h from 1110 to
102.
Signed-off-by: Neale Ranns <neale@graphiant.com>
Change-Id: I50611eba818eeb3a2dffd437a3c72c77766bed80
|
|
Type: fix
virtio uses indirect descriptors for chain buffers.
indirect descriptor chain is mapped on a vlib_buffer_t.
Single descriptor is 16 bytes and vlib_buffer_t has
2048 bytes space. So maximum long chain can have
128 (=2048/16) indirect descriptors.
This patch adds check to make sure descriptors chain
len should not exceed 128.
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: I060cfb7709568f42c9b5634527172690ce66a1a3
|
|
When vhost and another interface, say tap, are configured for interrupt
mode, performance number may be very low from vhost.
Further analysis discovers that when vhost posts an interrupt to the
RX infra, there is a 10 msec delay in waking up its input routine.
The delay is due to vhost posts the interrupt from the main thread
which tries to wake up the worker thread.
The fix is for vhost to move the posting interrupt call to the
corresponding input worker thread by calling
vnet_hw_if_set_rx_queue_file_index() to set it up.
While at it, streamline the function vhost_user_kickfd_read_ready()
since it will be called from the worker thread.
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I9beedcd33e1558c8335da4ee7fadc51c29ee4589
|
|
Type: fix
Change-Id: I433fe3799975fe3ba00fa30226f6e8dae34e88fc
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: improvement
Some tests i.e. ipsec see performance regression when offload flags
are moved to 2nd cacheline. This patch moves them back to 1st cacheline.
Change-Id: I6ead45ff6d2c467b0d248f409e27c2ba31758741
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: improvement
Change-Id: I4ac6b1a6553e26a5c26b6761fa3c4cc5d4925486
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: fix
virtio/tap interfaces set the empty buffers in the input node
for receiving data. Backend uses those buffers, fills them with
data and notifies the virtio/tap driver. But virtio/tap driver
gets into stall state if interface is created and configured
through exec script on VPP startup.conf and put the interface in
interrupt mode while VPP is only configured with main thread.
This patch fixes the problem by prefilling buffers during the
interface creation.
Change-Id: Ibc4d0e70e127ccc4b7cf8b2b18406ae4b02c73b4
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
trajectory trace has been broken for a while because we used to save the
buffer trajectory in a vector pointed to in opaque2. This does not work
well when opaque2 is copied (eg. because of a clone) as 2 buffers end up
sharing the same vector.
This dedicates a full cacheline in the buffer metadata instead when
trajectory is compiled in. No dynamic allocation, no sharing, no tears.
Type: refactor
Change-Id: I6a028ca1b48d38f393a36979e5e452c2dd48ad3f
Signed-off-by: Benoît Ganne <bganne@cisco.com>
|
|
Patch 24601 introduced description into template. In vhost_user_process,
template is reused inside the while loop to call clib_file_add. But
description is allocated outside of the loop only once. As a result, the
same storage for description is being referenced by all instances of call
to clib_file_add. As long as we don't call clib_file_del, we may be
good with multiple fds sharing the same storage for the description. When
one of the fds disconnects and frees the description, the other fds is
holding onto the free memory pointer. Bad news eventually happens when
another fd disconnects and frees the description that was already free
previously.
The fix is to move the allocation of description inside the loop to avoid
sharing.
Type: fix
Fixes: gerrit.fd.io/r/c/vpp/+/24601
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Ie670931acdc2c7b851982d98fd0d837284a19036
|
|
Type: improvement
Change-Id: I56dadb852f63da0597ddf4db395a22afdf2d9cc0
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Type: improvement
This patch adds flags to represent the modern NICs capabilities.
Change-Id: I96d38d9ab7eac55974d72795cd100d8337168e1e
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: improvement
Change-Id: I8322bca1a9aa75c97c0fe2ff24b2f65fc43242ce
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: improvement
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: Ic25ffe9c8e37826733cfb9e62cefb491bb3322bc
|
|
Type: refactor
This patch refactors the offload flags in vlib_buffer_t.
There are two main reasons behind this refactoring.
First, offload flags are insufficient to represent outer
and inner headers offloads. Second, room for these flags
in first cacheline of vlib_buffer_t is also limited.
This patch introduces a generic offload flag in first
cacheline. And detailed offload flags in 2nd cacheline
of the structure for performance optimization.
Change-Id: Icc363a142fb9208ec7113ab5bbfc8230181f6004
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|