2020-07-20 16:41:49 +02:00
|
|
|
#define UNIX_SOCK_PATH "/tmp/passt.socket"
|
2020-07-20 16:27:43 +02:00
|
|
|
|
2021-04-22 13:39:36 +02:00
|
|
|
/**
|
|
|
|
|
* struct tap_msg - Generic message descriptor for arrays of messages
|
|
|
|
|
* @start: Pointer to message start
|
|
|
|
|
* @l4_start: Pointer to L4 header
|
|
|
|
|
* @len: Message length, with L2 headers
|
|
|
|
|
* @l4_len: Message length, with L4 headers
|
|
|
|
|
*/
|
|
|
|
|
struct tap_msg {
|
|
|
|
|
char *start;
|
|
|
|
|
char *l4h;
|
|
|
|
|
size_t len;
|
|
|
|
|
size_t l4_len;
|
|
|
|
|
};
|
|
|
|
|
|
2021-03-17 10:57:44 +01:00
|
|
|
#include "icmp.h"
|
2021-03-17 10:57:41 +01:00
|
|
|
#include "tcp.h"
|
passt: Spare some syscalls, add some optimisations from profiling
Avoid a bunch of syscalls on forwarding paths by:
- storing minimum and maximum file descriptor numbers for each
protocol, fall back to SO_PROTOCOL query only on overlaps
- allocating a larger receive buffer -- this can result in more
coalesced packets than sendmmsg() can take (UIO_MAXIOV, i.e. 1024),
so make sure we don't exceed that within a single call to protocol
tap handlers
- nesting the handling loop in tap_handler() in the receive loop,
so that we have better chances of filling our receive buffer in
fewer calls
- skipping the recvfrom() in the UDP handler on EPOLLERR -- there's
nothing to be done in that case
and while at it:
- restore the 20ms timer interval for periodic (TCP) events, I
accidentally changed that to 100ms in an earlier commit
- attempt using SO_ZEROCOPY for UDP -- if it's not available,
sendmmsg() will succeed anyway
- fix the handling of the status code from sendmmsg(), if it fails,
we'll try to discard the first message, hence return 1 from the
UDP handler
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
2021-04-23 22:22:37 +02:00
|
|
|
#include "udp.h"
|
2021-03-17 10:57:41 +01:00
|
|
|
|
2020-07-20 16:27:43 +02:00
|
|
|
/**
|
|
|
|
|
* struct ctx - Execution context
|
|
|
|
|
* @epollfd: file descriptor for epoll instance
|
|
|
|
|
* @fd_unix: AF_UNIX socket for tap file descriptor
|
2020-07-21 10:48:24 +02:00
|
|
|
* @v4: Enable IPv4 transport
|
passt: New design and implementation with native Layer 4 sockets
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>
2021-02-16 07:25:09 +01:00
|
|
|
* @mac: Host MAC address
|
|
|
|
|
* @mac_guest: Guest MAC address
|
2020-07-20 16:27:43 +02:00
|
|
|
* @addr4: IPv4 address for external, routable interface
|
|
|
|
|
* @mask4: IPv4 netmask, network order
|
|
|
|
|
* @gw4: Default IPv4 gateway, network order
|
|
|
|
|
* @dns4: IPv4 DNS address, network order
|
2020-07-21 10:48:24 +02:00
|
|
|
* @v6: Enable IPv6 transport
|
|
|
|
|
* @addr6: IPv6 address for external, routable interface
|
|
|
|
|
* @gw6: Default IPv6 gateway
|
|
|
|
|
* @dns4: IPv6 DNS address
|
2020-07-20 16:27:43 +02:00
|
|
|
* @ifn: Name of routable interface
|
|
|
|
|
*/
|
|
|
|
|
struct ctx {
|
|
|
|
|
int epollfd;
|
|
|
|
|
int fd_unix;
|
|
|
|
|
unsigned char mac[ETH_ALEN];
|
passt: New design and implementation with native Layer 4 sockets
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>
2021-02-16 07:25:09 +01:00
|
|
|
unsigned char mac_guest[ETH_ALEN];
|
2020-07-21 10:48:24 +02:00
|
|
|
|
|
|
|
|
int v4;
|
2020-07-20 16:27:43 +02:00
|
|
|
unsigned long addr4;
|
|
|
|
|
unsigned long mask4;
|
|
|
|
|
unsigned long gw4;
|
|
|
|
|
unsigned long dns4;
|
2020-07-21 10:48:24 +02:00
|
|
|
|
|
|
|
|
int v6;
|
|
|
|
|
struct in6_addr addr6;
|
passt: New design and implementation with native Layer 4 sockets
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>
2021-02-16 07:25:09 +01:00
|
|
|
struct in6_addr addr6_guest;
|
2020-07-21 10:48:24 +02:00
|
|
|
struct in6_addr gw6;
|
|
|
|
|
struct in6_addr dns6;
|
|
|
|
|
|
2020-07-20 16:27:43 +02:00
|
|
|
char ifn[IF_NAMESIZE];
|
2021-03-17 10:57:41 +01:00
|
|
|
|
2021-03-17 10:57:44 +01:00
|
|
|
struct icmp_ctx icmp;
|
2021-03-17 10:57:41 +01:00
|
|
|
struct tcp_ctx tcp;
|
passt: Spare some syscalls, add some optimisations from profiling
Avoid a bunch of syscalls on forwarding paths by:
- storing minimum and maximum file descriptor numbers for each
protocol, fall back to SO_PROTOCOL query only on overlaps
- allocating a larger receive buffer -- this can result in more
coalesced packets than sendmmsg() can take (UIO_MAXIOV, i.e. 1024),
so make sure we don't exceed that within a single call to protocol
tap handlers
- nesting the handling loop in tap_handler() in the receive loop,
so that we have better chances of filling our receive buffer in
fewer calls
- skipping the recvfrom() in the UDP handler on EPOLLERR -- there's
nothing to be done in that case
and while at it:
- restore the 20ms timer interval for periodic (TCP) events, I
accidentally changed that to 100ms in an earlier commit
- attempt using SO_ZEROCOPY for UDP -- if it's not available,
sendmmsg() will succeed anyway
- fix the handling of the status code from sendmmsg(), if it fails,
we'll try to discard the first message, hence return 1 from the
UDP handler
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
2021-04-23 22:22:37 +02:00
|
|
|
struct tcp_ctx udp;
|
2020-07-20 16:27:43 +02:00
|
|
|
};
|