This change makes it possible to delete rules after inserting them,
without needing to query the rules first. Rules can be deleted both
before and after they are flushed. Additionally, this allows positioning
a new rule next to an existing rule, both before and after the existing
rule is flushed.
There are two ways to refer to a rule: Either by ID or by handle. The ID
is assigned by userspace, and is only valid within a transaction, so it
can only be used before the flush. The handle is assigned by the kernel
when the transaction is committed, and can thus only be used after the
flush. We thus need to set an ID on each newly created rule, and
retrieve the handle of the rule during the flush.
There was an existing mechanism to allocate IDs for sets, but this was
using a global counter without any synchronization to prevent data
races. I replaced this by a new mechanism which uses a connection-scoped
counter.
I implemented a new mechanism for retrieving replies in Flush, and
handling these replies by adding a callback to netlink messages. There
was some existing code to handle "overrun", which I deleted, because it
was nonsensical and just worked by accident. NLMSG_OVERRUN is in fact
not a flag, but a complete message type, so the (re&netlink.Overrun)
masking makes no sense. Even better, NLMSG_OVERRUN is never actually
used by Linux. What this code was actually doing was skipping over the
NFT_MSG_NEWRULE replies, and possibly a NFT_MSG_NEWGEN reply.
I had to update all existing tests which compared generated netlink
messages against a reference, by inserting the newly added ID attribute.
We also need to generate replies for the NFT_MSG_NEWRULE messages with a
handle added.
Jan Schär
Steffen Vogel
Michael Stapelberg