| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
Flooding a virtio-net interface connected to a socket back-end results
in a TX stall I'm still debugging. The stall goes away by setting a
higher value for x-txburst (256 by default).
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
| |
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
| |
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
| |
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
On pc-q35, pci.2 is usually configured by libvirt as a hotplug bus,
so we can't use address 0x0 there. Look for free busses starting from
pci.3 instead.
While at it, add proper error reporting for passt probing, and add
some comments to structs that were previously missing.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
| |
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
| |
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
|
|
|
|
| |
As libvirt can pass e1000e (not just e1000) devices as well,
make sure we also drop those network devices from the command
line before adding the parameters we need.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
|
|
| |
...3 was a left-over from a test.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
The previous approach wasn't really robust: adding a -netdev option
without libvirt knowing could result in clashes with other devices.
Drop network devices from command line, check the available busses
and addresses from all -device options according to the -machine
parameter, and add a virtio-net device using an available address
or bus. Then, add a corresponding -netdev socket option.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
|
|
|
|
| |
If a socket netdev parameter is already passed, don't touch the command
line. If it's not, add it, taking the id= reference from a netdev=
parameter, if any.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
|
|
|
|
|
|
| |
The name-guessing loop should iterate over names, not single
characters. Also add /usr/libexec/qemu-kvm as full path for
execvp(): execvp() won't find it if it's not in $PATH, which
is the reason why it shouldn't be under /usr/libexec/, but this
seems to be the case for some current version of Fedora.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It might be impractical to pass options to qrap when using libvirt,
because the <emulator/> tag expects a path to an executable, without
further arguments.
If the first argument is not a plausible socket number, and the
second argument is not a valid executable, look up a qemu command
from a list of possible names, then start it patching the command line
to include the -netdev fd= parameter corresponding to the AF_UNIX
domain socket we just opened.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is a reimplementation, partially building on the earlier draft,
that uses L4 sockets (SOCK_DGRAM, SOCK_STREAM) instead of SOCK_RAW,
providing L4-L2 translation functionality without requiring any
security capability.
Conceptually, this follows the design presented at:
https://gitlab.com/abologna/kubevirt-and-kvm/-/blob/master/Networking.md
The most significant novelty here comes from TCP and UDP translation
layers. In particular, the TCP state and translation logic follows
the intent of being minimalistic, without reimplementing a full TCP
stack in either direction, and synchronising as much as possible the
TCP dynamic and flows between guest and host kernel.
Another important introduction concerns addressing, port translation
and forwarding. The Layer 4 implementations now attempt to bind on
all unbound ports, in order to forward connections in a transparent
way.
While at it:
- the qemu 'tap' back-end can't be used as-is by qrap anymore,
because of explicit checks now introduced in qemu to ensure that
the corresponding file descriptor is actually a tap device. For
this reason, qrap now operates on a 'socket' back-end type,
accounting for and building the additional header reporting
frame length
- provide a demo script that sets up namespaces, addresses and
routes, and starts the daemon. A virtual machine started in the
network namespace, wrapped by qrap, will now directly interface
with passt and communicate using Layer 4 sockets provided by the
host kernel.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
| |
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
|
|
| |
Plug A Simple Socket Transport.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|
|
We can bypass a full-fledged network interface between qemu and merd by
connecting the qemu tap file descriptor to a provided UNIX domain
socket: this could be implemented in qemu eventually, qrap covers this
meanwhile.
This also avoids the need for the AF_PACKET socket towards the guest.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
|