hardenedbsd/HardenedBSD
1
0
Fork 0
mirror of https://git.hardenedbsd.org/hardenedbsd/HardenedBSD.git synced 2024-11-24 09:13:37 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity
a8902dbb87
HardenedBSD/sys/netlink/netlink_message_writer.h

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

307 lines
9.0 KiB
C
Raw Normal View History

netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
/*-
spdx: The BSD-2-Clause-FreeBSD identifier is obsolete, drop -FreeBSD The SPDX folks have obsoleted the BSD-2-Clause-FreeBSD identifier. Catch up to that fact and revert to their recommended match of BSD-2-Clause. Discussed with: pfg MFC After: 3 days Sponsored by: Netflix
2023-05-10 17:40:58 +02:00
* SPDX-License-Identifier: BSD-2-Clause
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
*
* Copyright (c) 2021 Ng Peng Nam Sean
* Copyright (c) 2022 Alexander V. Chernikov <melifaro@FreeBSD.org>
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#ifndef _NETLINK_NETLINK_MESSAGE_WRITER_H_
#define _NETLINK_NETLINK_MESSAGE_WRITER_H_
netlink: make test-includes happy by hiding most of the header contents under _KERNEL.
2022-10-01 19:01:53 +02:00
#ifdef _KERNEL
carp: support unicast Allow users to configure the address to send carp messages to. This allows carp to be used in unicast mode, which is useful in certain virtual configurations (e.g. AWS, VMWare ESXi, ...) Reviewed by: melifaro Sponsored by: Rubicon Communications, LLC ("Netgate") Differential Revision: https://reviews.freebsd.org/D38940
2023-03-15 13:31:45 +01:00
netlink: make nlattr_add_in[6]_addr inline MFC after: 2 weeks
2023-03-27 13:53:34 +02:00
#include <netinet/in.h>
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
/*
* It is not meant to be included directly
*/
netlink: use protocol specific receive buffer Implement Netlink socket receive buffer as a simple TAILQ of nl_buf's, same part of struct sockbuf that is used for send buffer already. This shaves a lot of code and a lot of extra processing. The pcb rids of the I/O queues as the socket buffer is exactly the queue. The message writer is simplified a lot, as we now always deal with linear buf. Notion of different buffer types goes away as way as different kinds of writers. The only things remaining are: a socket writer and a group writer. The impact on the network stack is that we no longer use mbufs, so a workaround from d18715475071 disappears. Note on message throttling. Now the taskqueue throttling mechanism needs to look at both socket buffers protected by their respective locks and on flags in the pcb that are protected by the pcb lock. There is definitely some room for optimization, but this changes tries to preserve as much as possible. Note on new nl_soreceive(). It emulates soreceive_generic(). It must undergo further optimization, see large comment put in there. Note on tests/sys/netlink/test_netlink_message_writer.py. This test boiled down almost to nothing with mbufs removed. However, I left it with minimal functionality (it basically checks that allocating N bytes we get N bytes) as it is one of not so many examples of ktest framework that allows to test KPIs with python. Note on Linux support. It got much simplier: Netlink message writer loses notion of Linux support lifetime, it is same regardless of process ABI. On socket write from Linux process we perform conversion immediately in nl_receive_message() and on an output conversion to Linux happens in in nl_send_one(). XXX: both conversions use M_NOWAIT allocation, which used to be the case before this change, too. Reviewed by: melifaro Differential Revision: https://reviews.freebsd.org/D42524
2024-01-02 22:04:01 +01:00
struct nl_buf;
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
struct nl_writer;
netlink: use protocol specific receive buffer Implement Netlink socket receive buffer as a simple TAILQ of nl_buf's, same part of struct sockbuf that is used for send buffer already. This shaves a lot of code and a lot of extra processing. The pcb rids of the I/O queues as the socket buffer is exactly the queue. The message writer is simplified a lot, as we now always deal with linear buf. Notion of different buffer types goes away as way as different kinds of writers. The only things remaining are: a socket writer and a group writer. The impact on the network stack is that we no longer use mbufs, so a workaround from d18715475071 disappears. Note on message throttling. Now the taskqueue throttling mechanism needs to look at both socket buffers protected by their respective locks and on flags in the pcb that are protected by the pcb lock. There is definitely some room for optimization, but this changes tries to preserve as much as possible. Note on new nl_soreceive(). It emulates soreceive_generic(). It must undergo further optimization, see large comment put in there. Note on tests/sys/netlink/test_netlink_message_writer.py. This test boiled down almost to nothing with mbufs removed. However, I left it with minimal functionality (it basically checks that allocating N bytes we get N bytes) as it is one of not so many examples of ktest framework that allows to test KPIs with python. Note on Linux support. It got much simplier: Netlink message writer loses notion of Linux support lifetime, it is same regardless of process ABI. On socket write from Linux process we perform conversion immediately in nl_receive_message() and on an output conversion to Linux happens in in nl_send_one(). XXX: both conversions use M_NOWAIT allocation, which used to be the case before this change, too. Reviewed by: melifaro Differential Revision: https://reviews.freebsd.org/D42524
2024-01-02 22:04:01 +01:00
typedef bool nl_writer_cb(struct nl_writer *nw);
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
struct nl_writer {
netlink: use protocol specific receive buffer Implement Netlink socket receive buffer as a simple TAILQ of nl_buf's, same part of struct sockbuf that is used for send buffer already. This shaves a lot of code and a lot of extra processing. The pcb rids of the I/O queues as the socket buffer is exactly the queue. The message writer is simplified a lot, as we now always deal with linear buf. Notion of different buffer types goes away as way as different kinds of writers. The only things remaining are: a socket writer and a group writer. The impact on the network stack is that we no longer use mbufs, so a workaround from d18715475071 disappears. Note on message throttling. Now the taskqueue throttling mechanism needs to look at both socket buffers protected by their respective locks and on flags in the pcb that are protected by the pcb lock. There is definitely some room for optimization, but this changes tries to preserve as much as possible. Note on new nl_soreceive(). It emulates soreceive_generic(). It must undergo further optimization, see large comment put in there. Note on tests/sys/netlink/test_netlink_message_writer.py. This test boiled down almost to nothing with mbufs removed. However, I left it with minimal functionality (it basically checks that allocating N bytes we get N bytes) as it is one of not so many examples of ktest framework that allows to test KPIs with python. Note on Linux support. It got much simplier: Netlink message writer loses notion of Linux support lifetime, it is same regardless of process ABI. On socket write from Linux process we perform conversion immediately in nl_receive_message() and on an output conversion to Linux happens in in nl_send_one(). XXX: both conversions use M_NOWAIT allocation, which used to be the case before this change, too. Reviewed by: melifaro Differential Revision: https://reviews.freebsd.org/D42524
2024-01-02 22:04:01 +01:00
struct nl_buf *buf; /* Underlying storage pointer */
struct nlmsghdr *hdr; /* Pointer to the currently-filled msg */
nl_writer_cb *cb; /* Callback to flush data */
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
union {
netlink: use protocol specific receive buffer Implement Netlink socket receive buffer as a simple TAILQ of nl_buf's, same part of struct sockbuf that is used for send buffer already. This shaves a lot of code and a lot of extra processing. The pcb rids of the I/O queues as the socket buffer is exactly the queue. The message writer is simplified a lot, as we now always deal with linear buf. Notion of different buffer types goes away as way as different kinds of writers. The only things remaining are: a socket writer and a group writer. The impact on the network stack is that we no longer use mbufs, so a workaround from d18715475071 disappears. Note on message throttling. Now the taskqueue throttling mechanism needs to look at both socket buffers protected by their respective locks and on flags in the pcb that are protected by the pcb lock. There is definitely some room for optimization, but this changes tries to preserve as much as possible. Note on new nl_soreceive(). It emulates soreceive_generic(). It must undergo further optimization, see large comment put in there. Note on tests/sys/netlink/test_netlink_message_writer.py. This test boiled down almost to nothing with mbufs removed. However, I left it with minimal functionality (it basically checks that allocating N bytes we get N bytes) as it is one of not so many examples of ktest framework that allows to test KPIs with python. Note on Linux support. It got much simplier: Netlink message writer loses notion of Linux support lifetime, it is same regardless of process ABI. On socket write from Linux process we perform conversion immediately in nl_receive_message() and on an output conversion to Linux happens in in nl_send_one(). XXX: both conversions use M_NOWAIT allocation, which used to be the case before this change, too. Reviewed by: melifaro Differential Revision: https://reviews.freebsd.org/D42524
2024-01-02 22:04:01 +01:00
struct nlpcb *nlp;
netlink: make netlink work correctly on CHERI. Current Netlink message writer code relies on executing callbacks with arbitrary data (pointer or integer) to flush the completed messages. This arbitrary data is stored as a union of { void *, uint64_t }. At some stage, the message flushing code copied this data, using direct uint64_t assignment instead of copying the union. It lead to failure on CHERI, as sizeof(pointer) == 16 there. Fix the code by making union non-anonymous and copying it entirely. Reviewed by: br, jhb, jrtc27 Differential Revision: https://reviews.freebsd.org/D39557 MFC after: 2 weeks
2023-04-13 13:53:49 +02:00
struct {
uint16_t proto;
uint16_t id;
} group;
netlink: use protocol specific receive buffer Implement Netlink socket receive buffer as a simple TAILQ of nl_buf's, same part of struct sockbuf that is used for send buffer already. This shaves a lot of code and a lot of extra processing. The pcb rids of the I/O queues as the socket buffer is exactly the queue. The message writer is simplified a lot, as we now always deal with linear buf. Notion of different buffer types goes away as way as different kinds of writers. The only things remaining are: a socket writer and a group writer. The impact on the network stack is that we no longer use mbufs, so a workaround from d18715475071 disappears. Note on message throttling. Now the taskqueue throttling mechanism needs to look at both socket buffers protected by their respective locks and on flags in the pcb that are protected by the pcb lock. There is definitely some room for optimization, but this changes tries to preserve as much as possible. Note on new nl_soreceive(). It emulates soreceive_generic(). It must undergo further optimization, see large comment put in there. Note on tests/sys/netlink/test_netlink_message_writer.py. This test boiled down almost to nothing with mbufs removed. However, I left it with minimal functionality (it basically checks that allocating N bytes we get N bytes) as it is one of not so many examples of ktest framework that allows to test KPIs with python. Note on Linux support. It got much simplier: Netlink message writer loses notion of Linux support lifetime, it is same regardless of process ABI. On socket write from Linux process we perform conversion immediately in nl_receive_message() and on an output conversion to Linux happens in in nl_send_one(). XXX: both conversions use M_NOWAIT allocation, which used to be the case before this change, too. Reviewed by: melifaro Differential Revision: https://reviews.freebsd.org/D42524
2024-01-02 22:04:01 +01:00
};
u_int num_messages; /* Number of messages in the buffer */
int malloc_flag; /* M_WAITOK or M_NOWAIT */
bool ignore_limit; /* If true, ignores RCVBUF limit */
bool enomem; /* True if ENOMEM occured */
bool suppress_ack; /* If true, don't send NLMSG_ERR */
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
};
#define NLMSG_SMALL 128
#define NLMSG_LARGE 2048
/* Message and attribute writing */
netlink: add netlink KPI to the kernel by default This change does the following: Base Netlink KPIs (ability to register the family, parse and/or write a Netlink message) are always present in the kernel. Specifically, * Implementation of genetlink family/group registration/removal, some base accessors (netlink_generic_kpi.c, 260 LoC) are compiled in unconditionally. * Basic TLV parser functions (netlink_message_parser.c, 507 LoC) are compiled in unconditionally. * Glue functions (netlink<>rtsock), malloc/core sysctl definitions (netlink_glue.c, 259 LoC) are compiled in unconditionally. * The rest of the KPI _functions_ are defined in the netlink_glue.c, but their implementation calls a pointer to either the stub function or the actual function, depending on whether the module is loaded or not. This approach allows to have only 1k LoC out of ~3.7k LoC (current sys/netlink implementation) in the kernel, which will not grow further. It also allows for the generic netlink kernel customers to load successfully without requiring Netlink module and operate correctly once Netlink module is loaded. Reviewed by: imp MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D39269
2023-03-27 13:59:30 +02:00
#if defined(NETLINK) || defined(NETLINK_MODULE)
/* Provide optimized calls to the functions inside the same linking unit */
bool _nlmsg_get_unicast_writer(struct nl_writer *nw, int expected_size, struct nlpcb *nlp);
bool _nlmsg_get_group_writer(struct nl_writer *nw, int expected_size, int proto, int group_id);
bool _nlmsg_flush(struct nl_writer *nw);
void _nlmsg_ignore_limit(struct nl_writer *nw);
netlink: use protocol specific receive buffer Implement Netlink socket receive buffer as a simple TAILQ of nl_buf's, same part of struct sockbuf that is used for send buffer already. This shaves a lot of code and a lot of extra processing. The pcb rids of the I/O queues as the socket buffer is exactly the queue. The message writer is simplified a lot, as we now always deal with linear buf. Notion of different buffer types goes away as way as different kinds of writers. The only things remaining are: a socket writer and a group writer. The impact on the network stack is that we no longer use mbufs, so a workaround from d18715475071 disappears. Note on message throttling. Now the taskqueue throttling mechanism needs to look at both socket buffers protected by their respective locks and on flags in the pcb that are protected by the pcb lock. There is definitely some room for optimization, but this changes tries to preserve as much as possible. Note on new nl_soreceive(). It emulates soreceive_generic(). It must undergo further optimization, see large comment put in there. Note on tests/sys/netlink/test_netlink_message_writer.py. This test boiled down almost to nothing with mbufs removed. However, I left it with minimal functionality (it basically checks that allocating N bytes we get N bytes) as it is one of not so many examples of ktest framework that allows to test KPIs with python. Note on Linux support. It got much simplier: Netlink message writer loses notion of Linux support lifetime, it is same regardless of process ABI. On socket write from Linux process we perform conversion immediately in nl_receive_message() and on an output conversion to Linux happens in in nl_send_one(). XXX: both conversions use M_NOWAIT allocation, which used to be the case before this change, too. Reviewed by: melifaro Differential Revision: https://reviews.freebsd.org/D42524
2024-01-02 22:04:01 +01:00
bool _nlmsg_refill_buffer(struct nl_writer *nw, u_int required_len);
netlink: add netlink KPI to the kernel by default This change does the following: Base Netlink KPIs (ability to register the family, parse and/or write a Netlink message) are always present in the kernel. Specifically, * Implementation of genetlink family/group registration/removal, some base accessors (netlink_generic_kpi.c, 260 LoC) are compiled in unconditionally. * Basic TLV parser functions (netlink_message_parser.c, 507 LoC) are compiled in unconditionally. * Glue functions (netlink<>rtsock), malloc/core sysctl definitions (netlink_glue.c, 259 LoC) are compiled in unconditionally. * The rest of the KPI _functions_ are defined in the netlink_glue.c, but their implementation calls a pointer to either the stub function or the actual function, depending on whether the module is loaded or not. This approach allows to have only 1k LoC out of ~3.7k LoC (current sys/netlink implementation) in the kernel, which will not grow further. It also allows for the generic netlink kernel customers to load successfully without requiring Netlink module and operate correctly once Netlink module is loaded. Reviewed by: imp MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D39269
2023-03-27 13:59:30 +02:00
bool _nlmsg_add(struct nl_writer *nw, uint32_t portid, uint32_t seq, uint16_t type,
uint16_t flags, uint32_t len);
bool _nlmsg_end(struct nl_writer *nw);
void _nlmsg_abort(struct nl_writer *nw);
bool _nlmsg_end_dump(struct nl_writer *nw, int error, struct nlmsghdr *hdr);
static inline bool
nlmsg_get_unicast_writer(struct nl_writer *nw, int expected_size, struct nlpcb *nlp)
{
return (_nlmsg_get_unicast_writer(nw, expected_size, nlp));
}
static inline bool
nlmsg_get_group_writer(struct nl_writer *nw, int expected_size, int proto, int group_id)
{
return (_nlmsg_get_group_writer(nw, expected_size, proto, group_id));
}
static inline bool
nlmsg_flush(struct nl_writer *nw)
{
return (_nlmsg_flush(nw));
}
static inline void
nlmsg_ignore_limit(struct nl_writer *nw)
{
_nlmsg_ignore_limit(nw);
}
static inline bool
HBSD: Use an unsigned integer when dealing with lengths in netlink There's certainly more to change here, but this is a start. Signed-off-by: Shawn Webb <shawn.webb@hardenedbsd.org> MFC-to: 14-STABLE
2024-01-04 23:07:17 +01:00
nlmsg_refill_buffer(struct nl_writer *nw, u_int required_size)
netlink: add netlink KPI to the kernel by default This change does the following: Base Netlink KPIs (ability to register the family, parse and/or write a Netlink message) are always present in the kernel. Specifically, * Implementation of genetlink family/group registration/removal, some base accessors (netlink_generic_kpi.c, 260 LoC) are compiled in unconditionally. * Basic TLV parser functions (netlink_message_parser.c, 507 LoC) are compiled in unconditionally. * Glue functions (netlink<>rtsock), malloc/core sysctl definitions (netlink_glue.c, 259 LoC) are compiled in unconditionally. * The rest of the KPI _functions_ are defined in the netlink_glue.c, but their implementation calls a pointer to either the stub function or the actual function, depending on whether the module is loaded or not. This approach allows to have only 1k LoC out of ~3.7k LoC (current sys/netlink implementation) in the kernel, which will not grow further. It also allows for the generic netlink kernel customers to load successfully without requiring Netlink module and operate correctly once Netlink module is loaded. Reviewed by: imp MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D39269
2023-03-27 13:59:30 +02:00
{
return (_nlmsg_refill_buffer(nw, required_size));
}
static inline bool
nlmsg_add(struct nl_writer *nw, uint32_t portid, uint32_t seq, uint16_t type,
uint16_t flags, uint32_t len)
{
return (_nlmsg_add(nw, portid, seq, type, flags, len));
}
static inline bool
nlmsg_end(struct nl_writer *nw)
{
return (_nlmsg_end(nw));
}
static inline void
nlmsg_abort(struct nl_writer *nw)
{
return (_nlmsg_abort(nw));
}
static inline bool
nlmsg_end_dump(struct nl_writer *nw, int error, struct nlmsghdr *hdr)
{
return (_nlmsg_end_dump(nw, error, hdr));
}
#else
/* Provide access to the functions via netlink_glue.c */
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
bool nlmsg_get_unicast_writer(struct nl_writer *nw, int expected_size, struct nlpcb *nlp);
bool nlmsg_get_group_writer(struct nl_writer *nw, int expected_size, int proto, int group_id);
bool nlmsg_get_chain_writer(struct nl_writer *nw, int expected_size, struct mbuf **pm);
bool nlmsg_flush(struct nl_writer *nw);
void nlmsg_ignore_limit(struct nl_writer *nw);
bool nlmsg_refill_buffer(struct nl_writer *nw, int required_size);
bool nlmsg_add(struct nl_writer *nw, uint32_t portid, uint32_t seq, uint16_t type,
uint16_t flags, uint32_t len);
bool nlmsg_end(struct nl_writer *nw);
void nlmsg_abort(struct nl_writer *nw);
bool nlmsg_end_dump(struct nl_writer *nw, int error, struct nlmsghdr *hdr);
netlink: add netlink KPI to the kernel by default This change does the following: Base Netlink KPIs (ability to register the family, parse and/or write a Netlink message) are always present in the kernel. Specifically, * Implementation of genetlink family/group registration/removal, some base accessors (netlink_generic_kpi.c, 260 LoC) are compiled in unconditionally. * Basic TLV parser functions (netlink_message_parser.c, 507 LoC) are compiled in unconditionally. * Glue functions (netlink<>rtsock), malloc/core sysctl definitions (netlink_glue.c, 259 LoC) are compiled in unconditionally. * The rest of the KPI _functions_ are defined in the netlink_glue.c, but their implementation calls a pointer to either the stub function or the actual function, depending on whether the module is loaded or not. This approach allows to have only 1k LoC out of ~3.7k LoC (current sys/netlink implementation) in the kernel, which will not grow further. It also allows for the generic netlink kernel customers to load successfully without requiring Netlink module and operate correctly once Netlink module is loaded. Reviewed by: imp MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D39269
2023-03-27 13:59:30 +02:00
#endif /* defined(NETLINK) || defined(NETLINK_MODULE) */
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
static inline bool
nlmsg_reply(struct nl_writer *nw, const struct nlmsghdr *hdr, int payload_len)
{
return (nlmsg_add(nw, hdr->nlmsg_pid, hdr->nlmsg_seq, hdr->nlmsg_type,
hdr->nlmsg_flags, payload_len));
}
/*
* KPI similar to mtodo():
* current (uncompleted) header is guaranteed to be contiguous,
* but can be reallocated, thus pointers may need to be readjusted.
*/
netlink: uninline some KPI functions that work with struct nl_writer These functions work with a buffer embedded into nl_writer, which is going to go opaque with upcoming changes. Make them private to the netlink module. No functional change intended. Reviewed by: melifaro Differential Revision: https://reviews.freebsd.org/D42523
2024-01-02 22:03:40 +01:00
u_int nlattr_save_offset(const struct nl_writer *nw);
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
static inline void *
_nlattr_restore_offset(const struct nl_writer *nw, int off)
{
return ((void *)((char *)nw->hdr + off));
}
#define nlattr_restore_offset(_ns, _off, _t) ((_t *)_nlattr_restore_offset(_ns, _off))
static inline void
nlattr_set_len(const struct nl_writer *nw, int off)
{
struct nlattr *nla = nlattr_restore_offset(nw, off, struct nlattr);
nla->nla_len = nlattr_save_offset(nw) - off;
}
netlink: uninline some KPI functions that work with struct nl_writer These functions work with a buffer embedded into nl_writer, which is going to go opaque with upcoming changes. Make them private to the netlink module. No functional change intended. Reviewed by: melifaro Differential Revision: https://reviews.freebsd.org/D42523
2024-01-02 22:03:40 +01:00
void *nlmsg_reserve_data_raw(struct nl_writer *nw, size_t sz);
netlink: cleanup netlink_writer code * Remove unused nlattr_add_nla() - that's a duplicate of nlattr_add_raw(). * Calculate alignment only once in nlmsg_reserve_data_raw() MFC after: 2 weeks
2023-03-07 18:42:27 +01:00
#define nlmsg_reserve_object(_ns, _t) ((_t *)nlmsg_reserve_data_raw(_ns, sizeof(_t)))
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
#define nlmsg_reserve_data(_ns, _sz, _t) ((_t *)nlmsg_reserve_data_raw(_ns, _sz))
static inline int
nlattr_add_nested(struct nl_writer *nw, uint16_t nla_type)
{
int off = nlattr_save_offset(nw);
struct nlattr *nla = nlmsg_reserve_data(nw, sizeof(struct nlattr), struct nlattr);
if (__predict_false(nla == NULL))
return (0);
nla->nla_type = nla_type;
return (off);
}
static inline void *
_nlmsg_reserve_attr(struct nl_writer *nw, uint16_t nla_type, uint16_t sz)
{
sz += sizeof(struct nlattr);
struct nlattr *nla = nlmsg_reserve_data(nw, sz, struct nlattr);
if (__predict_false(nla == NULL))
return (NULL);
nla->nla_type = nla_type;
nla->nla_len = sz;
return ((void *)(nla + 1));
}
#define nlmsg_reserve_attr(_ns, _at, _t) ((_t *)_nlmsg_reserve_attr(_ns, _at, NLA_ALIGN(sizeof(_t))))
netlink: uninline some KPI functions that work with struct nl_writer These functions work with a buffer embedded into nl_writer, which is going to go opaque with upcoming changes. Make them private to the netlink module. No functional change intended. Reviewed by: melifaro Differential Revision: https://reviews.freebsd.org/D42523
2024-01-02 22:03:40 +01:00
bool nlattr_add(struct nl_writer *nw, int attr_type, int attr_len,
const void *data);
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
netlink: return optional metadata with the operation result. Some operations like interface creation may need to return metadata - in this case, interface name - back to the caller if the operation is successful. This change implements attaching an `NLMSGERR_ATTR_COOKIE` nla to the operation reply message via `nlmsg_report_cookie()`. Additionally, on successful interface creation, interface index and interface name are returned in the `IFLA_NEW_IFINDEX` and `IFLA_IFNAME TLVs, encapsulated in the `NLMSGERR_ATTR_COOKIE`. Reviewed By: pauamma Differential Revision: https://reviews.freebsd.org/D38283 MFC after: 1 week
2023-02-09 15:53:44 +01:00
static inline bool
nlattr_add_raw(struct nl_writer *nw, const struct nlattr *nla_src)
{
int attr_len = nla_src->nla_len - sizeof(struct nlattr);
MPASS(attr_len >= 0);
return (nlattr_add(nw, nla_src->nla_type, attr_len, (const void *)(nla_src + 1)));
}
pf: convert DIOCGETSTATUS to netlink Introduce pfctl_get_status_h() because we need the pfctl_handle. In this variant use netlink to obtain the information. Sponsored by: Rubicon Communications, LLC ("Netgate")
2024-03-23 07:31:51 +01:00
static inline bool
nlattr_add_bool(struct nl_writer *nw, int attrtype, bool value)
{
return (nlattr_add(nw, attrtype, sizeof(bool), &value));
}
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
static inline bool
nlattr_add_u8(struct nl_writer *nw, int attrtype, uint8_t value)
{
return (nlattr_add(nw, attrtype, sizeof(uint8_t), &value));
}
static inline bool
nlattr_add_u16(struct nl_writer *nw, int attrtype, uint16_t value)
{
return (nlattr_add(nw, attrtype, sizeof(uint16_t), &value));
}
static inline bool
nlattr_add_u32(struct nl_writer *nw, int attrtype, uint32_t value)
{
return (nlattr_add(nw, attrtype, sizeof(uint32_t), &value));
}
static inline bool
nlattr_add_u64(struct nl_writer *nw, int attrtype, uint64_t value)
{
return (nlattr_add(nw, attrtype, sizeof(uint64_t), &value));
}
static inline bool
nlattr_add_s8(struct nl_writer *nw, int attrtype, int8_t value)
{
return (nlattr_add(nw, attrtype, sizeof(int8_t), &value));
}
static inline bool
nlattr_add_s16(struct nl_writer *nw, int attrtype, int16_t value)
{
return (nlattr_add(nw, attrtype, sizeof(int16_t), &value));
}
static inline bool
nlattr_add_s32(struct nl_writer *nw, int attrtype, int32_t value)
{
return (nlattr_add(nw, attrtype, sizeof(int32_t), &value));
}
static inline bool
nlattr_add_s64(struct nl_writer *nw, int attrtype, int64_t value)
{
return (nlattr_add(nw, attrtype, sizeof(int64_t), &value));
}
static inline bool
nlattr_add_flag(struct nl_writer *nw, int attrtype)
{
return (nlattr_add(nw, attrtype, 0, NULL));
}
static inline bool
nlattr_add_string(struct nl_writer *nw, int attrtype, const char *str)
{
return (nlattr_add(nw, attrtype, strlen(str) + 1, str));
}
netlink: make nlattr_add_in[6]_addr inline MFC after: 2 weeks
2023-03-27 13:53:34 +02:00
static inline bool
nlattr_add_in_addr(struct nl_writer *nw, int attrtype, const struct in_addr *in)
{
return (nlattr_add(nw, attrtype, sizeof(*in), in));
}
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
netlink: make nlattr_add_in[6]_addr inline MFC after: 2 weeks
2023-03-27 13:53:34 +02:00
static inline bool
nlattr_add_in6_addr(struct nl_writer *nw, int attrtype, const struct in6_addr *in6)
{
return (nlattr_add(nw, attrtype, sizeof(*in6), in6));
}
netlink: add netlink support Netlinks is a communication protocol currently used in Linux kernel to modify, read and subscribe for nearly all networking state. Interfaces, addresses, routes, firewall, fibs, vnets, etc are controlled via netlink. It is async, TLV-based protocol, providing 1-1 and 1-many communications. The current implementation supports the subset of NETLINK_ROUTE family. To be more specific, the following is supported: * Dumps: - routes - nexthops / nexthop groups - interfaces - interface addresses - neighbors (arp/ndp) * Notifications: - interface arrival/departure - interface address arrival/departure - route addition/deletion * Modifications: - adding/deleting routes - adding/deleting nexthops/nexthops groups - adding/deleting neghbors - adding/deleting interfaces (basic support only) * Rtsock interaction - route events are bridged both ways The implementation also supports the NETLINK_GENERIC family framework. Implementation notes: Netlink is implemented via loadable/unloadable kernel module, not touching many kernel parts. Each netlink socket uses dedicated taskqueue to support async operations that can sleep, such as interface creation. All message processing is performed within these taskqueues. Compatibility: Most of the Netlink data models specified above maps to FreeBSD concepts nicely. Unmodified ip(8) binary correctly works with interfaces, addresses, routes, nexthops and nexthop groups. Some software such as net/bird require header-only modifications to compile and work with FreeBSD netlink. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D36002 MFC after: 2 months
2022-01-20 22:39:21 +01:00
#endif
netlink: make test-includes happy by hiding most of the header contents under _KERNEL.
2022-10-01 19:01:53 +02:00
#endif
Reference in New Issue Copy Permalink