For legibility, wrap text paragraphs so every line is at most 70
characters long.
To check the generated documentation does not change, I compared the
generated HTML before and after this commit with "wdiff -3". Finds no
differences. Comparing with diff is not useful, as the refilled
paragraphs are visible there.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240322140910.328840-11-armbru@redhat.com>
By convention, we indent the second and subsequent lines of
descriptions and tagged sections, except for examples.
Turn this into a hard rule, and apply it to examples, too.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240216145841.2099240-11-armbru@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
[Straightforward conflicts in qapi/migration.json resolved]
Use of String is problematic, because it results in awkward interface
documentation. The previous commit cleaned up one instance.
Move String out of common.json next to its remaining users in net.json
to discourage reuse elsewhere.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240205074709.3613229-15-armbru@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Commit e050e42678 (qapi: Use explicit bulleted lists) added list
markup to correct bad rendering:
A JSON block comment like this:
Returns: nothing on success
If @node is not a valid block device, DeviceNotFound
If @name is not found, GenericError with an explanation
renders like this:
Returns: nothing on success If node is not a valid block device,
DeviceNotFound If name is not found, GenericError with an explanation
because whitespace is not significant.
Use an actual bulleted list, so that the formatting is correct.
It missed a few instances. Commit a937b6aa73 (qapi: Reformat doc
comments to conform to current conventions) then reflowed them.
Revert the reflowing, and add list markup.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240120095327.666239-6-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
AF_XDP is a network socket family that allows communication directly
with the network device driver in the kernel, bypassing most or all
of the kernel networking stack. In the essence, the technology is
pretty similar to netmap. But, unlike netmap, AF_XDP is Linux-native
and works with any network interfaces without driver modifications.
Unlike vhost-based backends (kernel, user, vdpa), AF_XDP doesn't
require access to character devices or unix sockets. Only access to
the network interface itself is necessary.
This patch implements a network backend that communicates with the
kernel by creating an AF_XDP socket. A chunk of userspace memory
is shared between QEMU and the host kernel. 4 ring buffers (Tx, Rx,
Fill and Completion) are placed in that memory along with a pool of
memory buffers for the packet data. Data transmission is done by
allocating one of the buffers, copying packet data into it and
placing the pointer into Tx ring. After transmission, device will
return the buffer via Completion ring. On Rx, device will take
a buffer form a pre-populated Fill ring, write the packet data into
it and place the buffer into Rx ring.
AF_XDP network backend takes on the communication with the host
kernel and the network interface and forwards packets to/from the
peer device in QEMU.
Usage example:
-device virtio-net-pci,netdev=guest1,mac=00:16:35:AF:AA:5C
-netdev af-xdp,ifname=ens6f1np1,id=guest1,mode=native,queues=1
XDP program bridges the socket with a network interface. It can be
attached to the interface in 2 different modes:
1. skb - this mode should work for any interface and doesn't require
driver support. With a caveat of lower performance.
2. native - this does require support from the driver and allows to
bypass skb allocation in the kernel and potentially use
zero-copy while getting packets in/out userspace.
By default, QEMU will try to use native mode and fall back to skb.
Mode can be forced via 'mode' option. To force 'copy' even in native
mode, use 'force-copy=on' option. This might be useful if there is
some issue with the driver.
Option 'queues=N' allows to specify how many device queues should
be open. Note that all the queues that are not open are still
functional and can receive traffic, but it will not be delivered to
QEMU. So, the number of device queues should generally match the
QEMU configuration, unless the device is shared with something
else and the traffic re-direction to appropriate queues is correctly
configured on a device level (e.g. with ethtool -N).
'start-queue=M' option can be used to specify from which queue id
QEMU should start configuring 'N' queues. It might also be necessary
to use this option with certain NICs, e.g. MLX5 NICs. See the docs
for examples.
In a general case QEMU will need CAP_NET_ADMIN and CAP_SYS_ADMIN
or CAP_BPF capabilities in order to load default XSK/XDP programs to
the network interface and configure BPF maps. It is possible, however,
to run with no capabilities. For that to work, an external process
with enough capabilities will need to pre-load default XSK program,
create AF_XDP sockets and pass their file descriptors to QEMU process
on startup via 'sock-fds' option. Network backend will need to be
configured with 'inhibit=on' to avoid loading of the program.
QEMU will need 32 MB of locked memory (RLIMIT_MEMLOCK) per queue
or CAP_IPC_LOCK.
There are few performance challenges with the current network backends.
First is that they do not support IO threads. This means that data
path is handled by the main thread in QEMU and may slow down other
work or may be slowed down by some other work. This also means that
taking advantage of multi-queue is generally not possible today.
Another thing is that data path is going through the device emulation
code, which is not really optimized for performance. The fastest
"frontend" device is virtio-net. But it's not optimized for heavy
traffic either, because it expects such use-cases to be handled via
some implementation of vhost (user, kernel, vdpa). In practice, we
have virtio notifications and rcu lock/unlock on a per-packet basis
and not very efficient accesses to the guest memory. Communication
channels between backend and frontend devices do not allow passing
more than one packet at a time as well.
Some of these challenges can be avoided in the future by adding better
batching into device emulation or by implementing vhost-af-xdp variant.
There are also a few kernel limitations. AF_XDP sockets do not
support any kinds of checksum or segmentation offloading. Buffers
are limited to a page size (4K), i.e. MTU is limited. Multi-buffer
support implementation for AF_XDP is in progress, but not ready yet.
Also, transmission in all non-zero-copy modes is synchronous, i.e.
done in a syscall. That doesn't allow high packet rates on virtual
interfaces.
However, keeping in mind all of these challenges, current implementation
of the AF_XDP backend shows a decent performance while running on top
of a physical NIC with zero-copy support.
Test setup:
2 VMs running on 2 physical hosts connected via ConnectX6-Dx card.
Network backend is configured to open the NIC directly in native mode.
The driver supports zero-copy. NIC is configured to use 1 queue.
Inside a VM - iperf3 for basic TCP performance testing and dpdk-testpmd
for PPS testing.
iperf3 result:
TCP stream : 19.1 Gbps
dpdk-testpmd (single queue, single CPU core, 64 B packets) results:
Tx only : 3.4 Mpps
Rx only : 2.0 Mpps
L2 FWD Loopback : 1.5 Mpps
In skb mode the same setup shows much lower performance, similar to
the setup where pair of physical NICs is replaced with veth pair:
iperf3 result:
TCP stream : 9 Gbps
dpdk-testpmd (single queue, single CPU core, 64 B packets) results:
Tx only : 1.2 Mpps
Rx only : 1.0 Mpps
L2 FWD Loopback : 0.7 Mpps
Results in skb mode or over the veth are close to results of a tap
backend with vhost=on and disabled segmentation offloading bridged
with a NIC.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> (docker/lcitool)
Signed-off-by: Jason Wang <jasowang@redhat.com>
Since commit a937b6aa73 (qapi: Reformat doc comments to conform to
current conventions), a number of comments not conforming to the
current formatting conventions were added. No problem, just sweep
the entire documentation once more.
To check the generated documentation does not change, I compared the
generated HTML before and after this commit with "wdiff -3". Finds no
differences. Comparing with diff is not useful, as the reflown
paragraphs are visible there.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20230720071610.1096458-7-armbru@redhat.com>
Change
# @name: Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed
# do eiusmod tempor incididunt ut labore et dolore magna aliqua.
to
# @name: Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed
# do eiusmod tempor incididunt ut labore et dolore magna aliqua.
See recent commit "qapi: Relax doc string @name: description
indentation rules" for rationale.
Reflow paragraphs to 70 columns width, and consistently use two spaces
to separate sentences.
To check the generated documentation does not change, I compared the
generated HTML before and after this commit with "wdiff -3". Finds no
differences. Comparing with diff is not useful, as the reflown
paragraphs are visible there.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20230428105429.1687850-18-armbru@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Acked-by: Lukas Straub <lukasstraub2@web.de>
[Straightforward conflicts in qapi/audio.json qapi/misc-target.json
qapi/run-state.json resolved]
MemoryDeviceInfoKind, NetClientDriver, and GuestPanicAction mention
some members only in ad hoc since documentation. The generated
documentation shows these members as "Not documented".
Replace by formal member documentation.
Add actual documentation text for the GuestPanicAction members, to
match existing member documentation there. For the others, merely
move existing "since" information.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20230425064223.820979-14-armbru@redhat.com>
rST parses something like
first line
second line
as a definition list item, where "first line" is the term being
defined by "second line".
This bites us in a couple of places. Here's one:
# @bps_max: total throughput limit during bursts,
# in bytes (Since 1.7)
scripts/qapi/parser.py parses this into an "argument section" with
name "bps_max" and text
total throughput limit during bursts,
in bytes (Since 1.7)
docs/sphinx/qapidoc.py duly passes the text to the rST parser, which
parses it as another definition list. Comes out as nested
definitions: term "bps_max: int (optional)" defined as term "total
throughput limit during bursts," defined as "in bytes (Since 1.7)".
rST truly is the Perl of ASCII-based markups.
Fix by deleting the extra indentation.
Fixes: 26ec4e53f2 (qapi: Fix indent level on doc comments in json files)
Fixes: c0ac533b6f (qapi: Stop using whitespace for alignment in comments)
Fixes: 81ad2964e9 (net/vmnet: add vmnet backends to qapi/net)
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20230425064223.820979-11-armbru@redhat.com>
A few examples neglect to prefix QMP input with '->'. Fix that.
Two examples have extra space after '<-'. Delete it.
A few examples neglect to show output. Provide some. The example
output for query-vcpu-dirty-limit could use further improvement. Add
a TODO comment.
Use "Examples:" instead of "Example:" where multiple examples are
given.
One example section numbers its two examples. Not done elsewhere;
drop.
Another example section separates them with "or". Likewise.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20230425064223.820979-8-armbru@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
In stream mode, if the server shuts down there is currently
no way to reconnect the client to a new server without removing
the NIC device and the netdev backend (or to reboot).
This patch introduces a reconnect option that specifies a delay
to try to reconnect with the same parameters.
Add a new test in qtest to test the reconnect option and the
connect/disconnect events.
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
The netdev reports NETDEV_STREAM_CONNECTED event when the backend
is connected, and NETDEV_STREAM_DISCONNECTED when it is disconnected.
The NETDEV_STREAM_CONNECTED event includes the destination address.
This allows a system manager like libvirt to detect when the server
fails.
For instance with passt:
{ 'execute': 'qmp_capabilities' }
{ "return": { } }
{ "timestamp": { "seconds": 1666341395, "microseconds": 505347 },
"event": "NETDEV_STREAM_CONNECTED",
"data": { "netdev-id": "netdev0",
"addr": { "path": "/tmp/passt_1.socket", "type": "unix" } } }
[killing passt here]
{ "timestamp": { "seconds": 1666341430, "microseconds": 968694 },
"event": "NETDEV_STREAM_DISCONNECTED",
"data": { "netdev-id": "netdev0" } }
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Copied from socket netdev file and modified to use SocketAddress
to be able to introduce new features like unix socket.
"udp" and "mcast" are squashed into dgram netdev, multicast is detected
according to the IP address type.
"listen" and "connect" modes are managed by stream netdev. An optional
parameter "server" defines the mode (off by default)
The two new types need to be parsed the modern way with -netdev, because
with the traditional way, the "type" field of netdev structure collides with
the "type" field of SocketAddress and prevents the correct evaluation of the
command line option. Moreover the traditional way doesn't allow to use
the same type (SocketAddress) several times with the -netdev option
(needed to specify "local" and "remote" addresses).
The previous commit paved the way for parsing the modern way, but
omitted one detail: how to pick modern vs. traditional, in
netdev_is_modern().
We want to pick based on the value of parameter "type". But how to
extract it from the option argument?
Parsing the option argument, either the modern or the traditional way,
extracts it for us, but only if parsing succeeds.
If parsing fails, there is no good option. No matter which parser we
pick, it'll be the wrong one for some arguments, and the error
reporting will be confusing.
Fortunately, the traditional parser accepts *anything* when called in
a certain way. This maximizes our chance to extract the value of
"type", and in turn minimizes the risk of confusing error reporting.
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Similar to other vhost backends, vhostfd can be passed to vhost-vdpa
backend as another parameter to instantiate vhost-vdpa net client.
This would benefit the use case where only open file descriptors, as
opposed to raw vhost-vdpa device paths, are accessible from the QEMU
process.
(qemu) netdev_add type=vhost-vdpa,vhostfd=61,id=vhost-vdpa1
Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Example output has the optional member @dnssearch as string type. It
should be an array of String objects instead. Fix it.
For reference, see NetdevUserOptions.
Signed-off-by: Victor Toso <victortoso@redhat.com>
Message-Id: <20220401110712.26911-1-victortoso@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
[Commit message tweaked for precision]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Creating and destroying network backend does not require a fully
constructed machine. Allow the related monitor commands to run before
machine initialization has concluded.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
