I have a patch against commit d378220c96
extending libnl with a facility to receive generic netlink messages sent
to multicast groups.
Essentially it add one new function genl_ctrl_resolve_grp which
prototype looks like this
int genl_ctrl_resolve_grp(struct nl_sock *sk, const char *family_name,
const char *grp_name)
It resolves the family name and the group name to group id. Then
the returned id can be used in nl_socket_add_membership to subscribe
to multicast messages.
Besides that it adds two more functions
uint32_t nl_socket_get_peer_groups(struct nl_sock *sk)
void nl_socket_set_peer_groups(struct nl_sock *sk, uint32_t groups)
allowing to modify the socket peer groups field. So it's possible to
multicast messages from the user space using the legacy interface.
Looks like there is no way (or I was not able to find one?) to modify
the netlink socket destination group from the user space, when the
group id is greater then 32.
Create new function nl_send_iovec() to be used to send multiple 'struct iovec'
through the netlink socket. This will be used for NF_QUEUE, to send
packet payload of a modified packet.
Refactor nl_send() to use nl_send_iovec() sending a single struct iovec.
Create new function nl_auto_complete() by refactoring nl_send_auto_complete(),
so other functions that call nl_send may also use nl_auto_complete()
Signed-off-by: Karl Hiramoto <karl@hiramoto.org>
The idea of a common handle is long revised and only misleading,
nl_handle really represents a socket with some additional
action handlers assigned to it.
Alias for nl_handle is kept for backwards compatibility.
In order for the interface to become more thread safe, the error
handling was revised to no longer depend on a static errno and
error string buffer.
This patch converts all error paths to return a libnl specific
error code which can be translated to a error message using
nl_geterror(int error). The functions nl_error() and
nl_get_errno() are therefore obsolete.
This change required various sets of function prototypes to be
changed in order to return an error code, the most prominent
are:
struct nl_cache *foo_alloc_cache(...);
changed to:
int foo_alloc_cache(..., struct nl_cache **);
struct nl_msg *foo_build_request(...);
changed to:
int foo_build_request(..., struct nl_msg **);
struct foo *foo_parse(...);
changed to:
int foo_parse(..., struct foo **);
This pretty much only leaves trivial allocation functions to
still return a pointer object which can still return NULL to
signal out of memory.
This change is a serious API and ABI breaker, sorry!
As pointed out by Regis Hanna, a considerable performance gain can be
achieved by using malloc() over calloc() when allocating netlink message
buffers. This is likely due to the fact that we use a complete page for
each message.
I stepped over libnl always freeing the messages and it
kind of made it awkward to reuse the message data without
reallocating.
The basic idea is: if a callback return value has a bit set,
don't free that message. The calling application owns it.
By default, things stay as before (messages are freed).
Signed-off-by: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Patrick McHardy reported a problem where pointers to the
payload of a netlink message as returned by f.e. the
nesting helpers become stale when the payload data
chunk is reallocated.
In order to avoid further problems, the payload chunk is
no longer extended on the fly. Instead the allocation is
made during netlink message object allocation time with
a default size of a page which should be fine for the
majority of all users. Additionally the functions
nlmsg_alloc_size() and nlmsg_set_default_size() have been
added to allocate messages of a particular length and to
modify the default message size.
Fixes an off-by-one when releasing local ports. Fixes nl_connect()
to properly close the socket upon failure. Return EBADFD if
operations are performed on unconnected sockets where appropriate.
Makes nl_handle_alloc() return an error if all local ports are
used up.
Thomas Graf