This patch removes includes that are not used. The removals were found with
include-what-you-use which checks if any of the symbols from a header is
in use.
Like Infiniband. See RFC 4390 section 2.1 for details on DHCP
and Infiniband; chaddr is zeroed, hlen is set to 0, and htype
is set to ARPHRD_INFINIBAND because IB hardware addresses
are 20 bytes in length.
This allows the sockets to be bound to a specific address before it is configured,
also enable SO_REUSEADDR to allow multiple DHCP clients to run at the same time.
We already ignore IP fragments, because we expect that Fragment
offset (FO) field is not set. However first fragment in a fragmented IP
flow will have all zeroes in FO field. We should ignore such packet as
well, thus we need to look at MF flag in the IP header. Checking MF flag
will filter out all except last packet in fragmented flows. Last one
will be ruled out by next check for value of FO.
Check that received DHCP packets actually include our MAC address in
chaddr field. BPF interpreter has 32 bit wide registers but MAC address
is 48 bits long so we have to do check in two steps.
There is no need to explicitly check version of L3 protocol in the
ethernet header because we bind socket with .sll_protocol set to
ETH_P_IP, thus we only receive IPv4 packets on the socket.
For this to work nicely we need to use REUSEADDR so that more than one socket
can be open at the same time. Also, we request the ifindex to be appended
to incoming messages, so we know whence it came.
Try a bit harder to make the kernel drop packets not for us. This should reduce
the number of wakeups from n^2 to n in the number of dhcp clients, which admittedly
only makes a differenc in very extreme cases.
If they are too small to fit the IP+UDP+DHCP headers they can be of no use, so
don't waste resources parsing them. This is at the cost of losing some verbosity
in the logging.
Filter out everything except UDP packets destined for the DHCP client port,
this should avoid the vast majority of spurious wakeups.
Filter based on [0], with permission.
Possible improvemnts: also check for the DHCP magic cookie to drop invalid
packets. Check for our xid to filter out packets destined for other clients.
[0]: <https://github.com/ambrop72/badvpn/blob/master/dhcpclient/BDHCPClient.c#L57>
Passing the protocol to socket() is redundant as it will be specified again in
bind(). Dropping the redundancy reduces the cost of bind() from ~30ms to ~0ms.
For details see [0].
networkd in a container (i.e., with next to no network latency) can now
negotiate a DHCP lease in 0.7 - 5 ms.
Thanks to Kay for help with debugging and to Daniel Borkmann for the pointer
to fix the problem.
[0]: <https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=902fefb82ef72a50c78cb4a20cc954b037a98d1c>
safe_close() automatically becomes a NOP when a negative fd is passed,
and returns -1 unconditionally. This makes it easy to write lines like
this:
fd = safe_close(fd);
Which will close an fd if it is open, and reset the fd variable
correctly.
By making use of this new scheme we can drop a > 200 lines of code that
was required to test for non-negative fds or to reset the closed fd
variable afterwards.
Lennart Poettering
Thomas Hindoe Paaboel Andersen
Torstein Husebø
Dan Williams
Tom Gundersen
Michal Sekletar