nuttx/net/icmp/icmp_input.c

422 lines
12 KiB
C
Raw Normal View History

/****************************************************************************
* net/icmp/icmp_input.c
* Handling incoming ICMP input
*
* Copyright (C) 2007-2009, 2012, 2014-2015, 2017, 2019 Gregory Nutt. All
* rights reserved.
* Author: Gregory Nutt <gnutt@nuttx.org>
*
* Adapted for NuttX from logic in uIP which also has a BSD-like license:
*
* Original author Adam Dunkels <adam@dunkels.com>
* Copyright () 2001-2003, Adam Dunkels.
* All rights reserved.
*
* 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.
* 3. The name of the author may not be used to endorse or promote
* products derived from this software without specific prior
* written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 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.
*
****************************************************************************/
/****************************************************************************
* Included Files
****************************************************************************/
#include <nuttx/config.h>
#ifdef CONFIG_NET
#include <stdint.h>
#include <string.h>
#include <debug.h>
nuttx/net:Support to PMTUD RFC 1191 When a router is unable to forward a datagram because it exceeds the MTU of the next-hop network and its Don't Fragment bit is set, the router is required to return an ICMP Destination Unreachable message to the source of the datagram, with the Code indicating "fragmentation needed and DF set". To support the Path MTU Discovery technique specified in this memo, the router MUST include the MTU of that next-hop network in the low-order 16 bits of the ICMP header field that is labelled "unused" in the ICMP specification [7]. The high-order 16 bits remain unused, and MUST be set to zero. Thus, the message has the following format: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 3 | Code = 4 | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | unused = 0 | Next-Hop MTU | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Internet Header + 64 bits of Original Datagram Data | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ RFC 1185 Packet Too Big Message 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Code | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | MTU | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | As much of invoking packet | + as will fit without the ICMPv6 packet + | exceeding 576 octets | IPv6 Fields: Destination Address Copied from the Source Address field of the invoking packet. ICMPv6 Fields: Type 2 Code 0 MTU The Maximum Transmission Unit of the next-hop link. Description A Packet Too Big MUST be sent by a router in response to a packet that it cannot forward because the packet is larger than the MTU of the outgoing link. The information in this message is used as part of the Path MTU Discovery process [RFC-1191]. Signed-off-by: wangchen <wangchen41@xiaomi.com>
2023-05-11 13:32:30 +02:00
#include <sys/time.h>
#include <net/if.h>
#include <arpa/inet.h>
#include <nuttx/net/netconfig.h>
#include <nuttx/net/netdev.h>
#include <nuttx/net/netstats.h>
#include <nuttx/net/ip.h>
#include "devif/devif.h"
2014-06-25 17:12:47 +02:00
#include "icmp/icmp.h"
#include "utils/utils.h"
#ifdef CONFIG_NET_ICMP
/****************************************************************************
* Pre-processor Definitions
****************************************************************************/
#ifdef CONFIG_NET_ICMP_SOCKET
/****************************************************************************
* Private Types
****************************************************************************/
struct icmp_deliver_s
{
FAR struct net_driver_s *dev; /* Current network device */
uint16_t iphdrlen; /* The size of the IPv4 header */
bool delivered; /* Whether the message is delivered */
};
/****************************************************************************
* Private Functions
****************************************************************************/
static bool icmp_filter(uint32_t filter, uint8_t type)
{
if (type < 32)
{
return ((1u << type) & filter) != 0;
}
/* Do not block unknown ICMP types */
return 0;
}
/****************************************************************************
* Name: icmp_datahandler
*
* Description:
* Handle ICMP echo replies that are not accepted by the application.
*
* Input Parameters:
* dev - Device instance only the input packet in d_buf, length = d_len;
* conn - A pointer to the ICMP connection structure
* iphdrlen - The size of the IPv4 header
*
* Returned Value:
* The number of bytes actually buffered is returned. This will be either
* zero or equal to buflen; partial packets are not buffered.
*
****************************************************************************/
static uint16_t icmp_datahandler(FAR struct net_driver_s *dev,
FAR struct icmp_conn_s *conn,
uint16_t iphdrlen)
{
FAR struct ipv4_hdr_s *ipv4;
struct sockaddr_in inaddr;
FAR struct iob_s *iob;
net/l2/l3/l4: add support of iob offload 1. Add new config CONFIG_NET_LL_GUARDSIZE to isolation of l2 stack, which will benefit l3(IP) layer for multi-MAC(l2) implementation, especially in some NICs such as celluler net driver. new configuration options: CONFIG_NET_LL_GUARDSIZE CONFIG_NET_LL_GUARDSIZE will reserved l2 buffer header size of network buffer to isolate the L2/L3 (MAC/IP) data on network layer, which will be beneficial to L3 network layer protocol transparent transmission and forwarding ------------------------------------------------------------ Layout of frist iob entry: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- iob | Reserved | io_len | ------------------------------------------------- ------------------------------------------------------------- Layout of different NICs implementation: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- Ethernet | Reserved | ETH_HDRLEN | io_len | ---------------------------------|--------------- 8021Q | Reserved | ETH_8021Q_HDRLEN | io_len | ---------------------------------|--------------- ipforward | Reserved | io_len | ------------------------------------------------- -------------------------------------------------------------------- 2. Support iob offload to l2 driver to avoid unnecessary memory copy Support send/receive iob vectors directly between the NICs and l3/l4 stack to avoid unnecessary memory copies, especially on hardware that supports Scatter/gather, which can greatly improve performance. new interface to support iob offload: ------------------------------------------ | IOB version | original | |----------------------------------------| | devif_iob_poll() | devif_poll() | | ... | ... | ------------------------------------------ -------------------------------------------------------------------- 1> NIC hardware support Scatter/gather transfer TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | dev->d_iob: | --------------- --------------- io_data iob1 | | | iob3 | | | \ --------------- --------------- --------------- | --------------- | iob0 | | | | iob2 | | | | --------------- | --------------- | \ | / / \ | / / ---------------------------------------------- NICs io vector | | | | | | | | | | ---------------------------------------------- RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to each iobs) -------------------------------------------------------------------- 2> CONFIG_IOB_BUFSIZE is greater than MTU: TX: "(CONFIG_IOB_BUFSIZE) > (MAX_NETDEV_PKTSIZE + CONFIG_NET_GUARDSIZE + CONFIG_NET_LL_GUARDSIZE)" tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | "NIC"_send() (dev->d_iob->io_data[CONFIG_NET_LL_GUARDSIZE - NET_LL_HDRLEN(dev)]) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to io_data) -------------------------------------------------------------------- 3> Compatible with all old flat buffer NICs TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll(devif_poll_callback()) devif_poll_callback() /* new interface, gather iobs to flat buffer */ / \ / \ devif_poll("NIC"_txpoll) "NIC"_send()(dev->d_buf) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | netdev_input() /* new interface, Scatter/gather flat/iob buffer */ | | pkt/ipv[4|6]_input()/... | | NICs io vector receive(Orignal flat buffer) 3. Iperf passthrough on NuttX simulator: ------------------------------------------------- | Protocol | Server | Client | | |-----------------------------------------------| | TCP | 813 | 834 | Mbits/sec | | TCP(Offload) | 1720 | 1100 | Mbits/sec | | UDP | 22 | 757 | Mbits/sec | | UDP(Offload) | 25 | 1250 | Mbits/sec | ------------------------------------------------- Signed-off-by: chao an <anchao@xiaomi.com>
2022-11-23 12:39:40 +01:00
uint16_t buflen;
int ret;
iob = iob_tryalloc(false);
if (iob == NULL)
{
return -ENOMEM;
}
/* Put the IPv4 address at the beginning of the read-ahead buffer */
net/l2/l3/l4: add support of iob offload 1. Add new config CONFIG_NET_LL_GUARDSIZE to isolation of l2 stack, which will benefit l3(IP) layer for multi-MAC(l2) implementation, especially in some NICs such as celluler net driver. new configuration options: CONFIG_NET_LL_GUARDSIZE CONFIG_NET_LL_GUARDSIZE will reserved l2 buffer header size of network buffer to isolate the L2/L3 (MAC/IP) data on network layer, which will be beneficial to L3 network layer protocol transparent transmission and forwarding ------------------------------------------------------------ Layout of frist iob entry: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- iob | Reserved | io_len | ------------------------------------------------- ------------------------------------------------------------- Layout of different NICs implementation: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- Ethernet | Reserved | ETH_HDRLEN | io_len | ---------------------------------|--------------- 8021Q | Reserved | ETH_8021Q_HDRLEN | io_len | ---------------------------------|--------------- ipforward | Reserved | io_len | ------------------------------------------------- -------------------------------------------------------------------- 2. Support iob offload to l2 driver to avoid unnecessary memory copy Support send/receive iob vectors directly between the NICs and l3/l4 stack to avoid unnecessary memory copies, especially on hardware that supports Scatter/gather, which can greatly improve performance. new interface to support iob offload: ------------------------------------------ | IOB version | original | |----------------------------------------| | devif_iob_poll() | devif_poll() | | ... | ... | ------------------------------------------ -------------------------------------------------------------------- 1> NIC hardware support Scatter/gather transfer TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | dev->d_iob: | --------------- --------------- io_data iob1 | | | iob3 | | | \ --------------- --------------- --------------- | --------------- | iob0 | | | | iob2 | | | | --------------- | --------------- | \ | / / \ | / / ---------------------------------------------- NICs io vector | | | | | | | | | | ---------------------------------------------- RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to each iobs) -------------------------------------------------------------------- 2> CONFIG_IOB_BUFSIZE is greater than MTU: TX: "(CONFIG_IOB_BUFSIZE) > (MAX_NETDEV_PKTSIZE + CONFIG_NET_GUARDSIZE + CONFIG_NET_LL_GUARDSIZE)" tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | "NIC"_send() (dev->d_iob->io_data[CONFIG_NET_LL_GUARDSIZE - NET_LL_HDRLEN(dev)]) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to io_data) -------------------------------------------------------------------- 3> Compatible with all old flat buffer NICs TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll(devif_poll_callback()) devif_poll_callback() /* new interface, gather iobs to flat buffer */ / \ / \ devif_poll("NIC"_txpoll) "NIC"_send()(dev->d_buf) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | netdev_input() /* new interface, Scatter/gather flat/iob buffer */ | | pkt/ipv[4|6]_input()/... | | NICs io vector receive(Orignal flat buffer) 3. Iperf passthrough on NuttX simulator: ------------------------------------------------- | Protocol | Server | Client | | |-----------------------------------------------| | TCP | 813 | 834 | Mbits/sec | | TCP(Offload) | 1720 | 1100 | Mbits/sec | | UDP | 22 | 757 | Mbits/sec | | UDP(Offload) | 25 | 1250 | Mbits/sec | ------------------------------------------------- Signed-off-by: chao an <anchao@xiaomi.com>
2022-11-23 12:39:40 +01:00
ipv4 = IPv4BUF;
inaddr.sin_family = AF_INET;
inaddr.sin_port = 0;
net_ipv4addr_copy(inaddr.sin_addr.s_addr,
net_ip4addr_conv32(ipv4->srcipaddr));
memset(inaddr.sin_zero, 0, sizeof(inaddr.sin_zero));
net/l2/l3/l4: add support of iob offload 1. Add new config CONFIG_NET_LL_GUARDSIZE to isolation of l2 stack, which will benefit l3(IP) layer for multi-MAC(l2) implementation, especially in some NICs such as celluler net driver. new configuration options: CONFIG_NET_LL_GUARDSIZE CONFIG_NET_LL_GUARDSIZE will reserved l2 buffer header size of network buffer to isolate the L2/L3 (MAC/IP) data on network layer, which will be beneficial to L3 network layer protocol transparent transmission and forwarding ------------------------------------------------------------ Layout of frist iob entry: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- iob | Reserved | io_len | ------------------------------------------------- ------------------------------------------------------------- Layout of different NICs implementation: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- Ethernet | Reserved | ETH_HDRLEN | io_len | ---------------------------------|--------------- 8021Q | Reserved | ETH_8021Q_HDRLEN | io_len | ---------------------------------|--------------- ipforward | Reserved | io_len | ------------------------------------------------- -------------------------------------------------------------------- 2. Support iob offload to l2 driver to avoid unnecessary memory copy Support send/receive iob vectors directly between the NICs and l3/l4 stack to avoid unnecessary memory copies, especially on hardware that supports Scatter/gather, which can greatly improve performance. new interface to support iob offload: ------------------------------------------ | IOB version | original | |----------------------------------------| | devif_iob_poll() | devif_poll() | | ... | ... | ------------------------------------------ -------------------------------------------------------------------- 1> NIC hardware support Scatter/gather transfer TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | dev->d_iob: | --------------- --------------- io_data iob1 | | | iob3 | | | \ --------------- --------------- --------------- | --------------- | iob0 | | | | iob2 | | | | --------------- | --------------- | \ | / / \ | / / ---------------------------------------------- NICs io vector | | | | | | | | | | ---------------------------------------------- RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to each iobs) -------------------------------------------------------------------- 2> CONFIG_IOB_BUFSIZE is greater than MTU: TX: "(CONFIG_IOB_BUFSIZE) > (MAX_NETDEV_PKTSIZE + CONFIG_NET_GUARDSIZE + CONFIG_NET_LL_GUARDSIZE)" tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | "NIC"_send() (dev->d_iob->io_data[CONFIG_NET_LL_GUARDSIZE - NET_LL_HDRLEN(dev)]) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to io_data) -------------------------------------------------------------------- 3> Compatible with all old flat buffer NICs TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll(devif_poll_callback()) devif_poll_callback() /* new interface, gather iobs to flat buffer */ / \ / \ devif_poll("NIC"_txpoll) "NIC"_send()(dev->d_buf) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | netdev_input() /* new interface, Scatter/gather flat/iob buffer */ | | pkt/ipv[4|6]_input()/... | | NICs io vector receive(Orignal flat buffer) 3. Iperf passthrough on NuttX simulator: ------------------------------------------------- | Protocol | Server | Client | | |-----------------------------------------------| | TCP | 813 | 834 | Mbits/sec | | TCP(Offload) | 1720 | 1100 | Mbits/sec | | UDP | 22 | 757 | Mbits/sec | | UDP(Offload) | 25 | 1250 | Mbits/sec | ------------------------------------------------- Signed-off-by: chao an <anchao@xiaomi.com>
2022-11-23 12:39:40 +01:00
/* Copy the src address info into the front of I/O buffer chain which
* overwrites the contents of the packet header field.
*/
memcpy(iob->io_data, &inaddr, sizeof(struct sockaddr_in));
iob_reserve(iob, sizeof(struct sockaddr_in));
/* Copy the ICMP message into the I/O buffer chain (without waiting) */
ret = iob_clone_partial(dev->d_iob, dev->d_iob->io_pktlen,
0, iob, 0, true, false);
if (ret < 0)
{
iob_free_chain(iob);
return ret;
}
net/l2/l3/l4: add support of iob offload 1. Add new config CONFIG_NET_LL_GUARDSIZE to isolation of l2 stack, which will benefit l3(IP) layer for multi-MAC(l2) implementation, especially in some NICs such as celluler net driver. new configuration options: CONFIG_NET_LL_GUARDSIZE CONFIG_NET_LL_GUARDSIZE will reserved l2 buffer header size of network buffer to isolate the L2/L3 (MAC/IP) data on network layer, which will be beneficial to L3 network layer protocol transparent transmission and forwarding ------------------------------------------------------------ Layout of frist iob entry: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- iob | Reserved | io_len | ------------------------------------------------- ------------------------------------------------------------- Layout of different NICs implementation: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- Ethernet | Reserved | ETH_HDRLEN | io_len | ---------------------------------|--------------- 8021Q | Reserved | ETH_8021Q_HDRLEN | io_len | ---------------------------------|--------------- ipforward | Reserved | io_len | ------------------------------------------------- -------------------------------------------------------------------- 2. Support iob offload to l2 driver to avoid unnecessary memory copy Support send/receive iob vectors directly between the NICs and l3/l4 stack to avoid unnecessary memory copies, especially on hardware that supports Scatter/gather, which can greatly improve performance. new interface to support iob offload: ------------------------------------------ | IOB version | original | |----------------------------------------| | devif_iob_poll() | devif_poll() | | ... | ... | ------------------------------------------ -------------------------------------------------------------------- 1> NIC hardware support Scatter/gather transfer TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | dev->d_iob: | --------------- --------------- io_data iob1 | | | iob3 | | | \ --------------- --------------- --------------- | --------------- | iob0 | | | | iob2 | | | | --------------- | --------------- | \ | / / \ | / / ---------------------------------------------- NICs io vector | | | | | | | | | | ---------------------------------------------- RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to each iobs) -------------------------------------------------------------------- 2> CONFIG_IOB_BUFSIZE is greater than MTU: TX: "(CONFIG_IOB_BUFSIZE) > (MAX_NETDEV_PKTSIZE + CONFIG_NET_GUARDSIZE + CONFIG_NET_LL_GUARDSIZE)" tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | "NIC"_send() (dev->d_iob->io_data[CONFIG_NET_LL_GUARDSIZE - NET_LL_HDRLEN(dev)]) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to io_data) -------------------------------------------------------------------- 3> Compatible with all old flat buffer NICs TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll(devif_poll_callback()) devif_poll_callback() /* new interface, gather iobs to flat buffer */ / \ / \ devif_poll("NIC"_txpoll) "NIC"_send()(dev->d_buf) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | netdev_input() /* new interface, Scatter/gather flat/iob buffer */ | | pkt/ipv[4|6]_input()/... | | NICs io vector receive(Orignal flat buffer) 3. Iperf passthrough on NuttX simulator: ------------------------------------------------- | Protocol | Server | Client | | |-----------------------------------------------| | TCP | 813 | 834 | Mbits/sec | | TCP(Offload) | 1720 | 1100 | Mbits/sec | | UDP | 22 | 757 | Mbits/sec | | UDP(Offload) | 25 | 1250 | Mbits/sec | ------------------------------------------------- Signed-off-by: chao an <anchao@xiaomi.com>
2022-11-23 12:39:40 +01:00
buflen = dev->d_len;
/* Add the new I/O buffer chain to the tail of the read-ahead queue (again
* without waiting).
*/
ret = iob_tryadd_queue(iob, &conn->readahead);
if (ret < 0)
{
nerr("ERROR: Failed to queue the I/O buffer chain: %d\n", ret);
net/l2/l3/l4: add support of iob offload 1. Add new config CONFIG_NET_LL_GUARDSIZE to isolation of l2 stack, which will benefit l3(IP) layer for multi-MAC(l2) implementation, especially in some NICs such as celluler net driver. new configuration options: CONFIG_NET_LL_GUARDSIZE CONFIG_NET_LL_GUARDSIZE will reserved l2 buffer header size of network buffer to isolate the L2/L3 (MAC/IP) data on network layer, which will be beneficial to L3 network layer protocol transparent transmission and forwarding ------------------------------------------------------------ Layout of frist iob entry: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- iob | Reserved | io_len | ------------------------------------------------- ------------------------------------------------------------- Layout of different NICs implementation: iob_data (aligned by CONFIG_IOB_ALIGNMENT) | | io_offset(CONFIG_NET_LL_GUARDSIZE) | | ------------------------------------------------- Ethernet | Reserved | ETH_HDRLEN | io_len | ---------------------------------|--------------- 8021Q | Reserved | ETH_8021Q_HDRLEN | io_len | ---------------------------------|--------------- ipforward | Reserved | io_len | ------------------------------------------------- -------------------------------------------------------------------- 2. Support iob offload to l2 driver to avoid unnecessary memory copy Support send/receive iob vectors directly between the NICs and l3/l4 stack to avoid unnecessary memory copies, especially on hardware that supports Scatter/gather, which can greatly improve performance. new interface to support iob offload: ------------------------------------------ | IOB version | original | |----------------------------------------| | devif_iob_poll() | devif_poll() | | ... | ... | ------------------------------------------ -------------------------------------------------------------------- 1> NIC hardware support Scatter/gather transfer TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | dev->d_iob: | --------------- --------------- io_data iob1 | | | iob3 | | | \ --------------- --------------- --------------- | --------------- | iob0 | | | | iob2 | | | | --------------- | --------------- | \ | / / \ | / / ---------------------------------------------- NICs io vector | | | | | | | | | | ---------------------------------------------- RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to each iobs) -------------------------------------------------------------------- 2> CONFIG_IOB_BUFSIZE is greater than MTU: TX: "(CONFIG_IOB_BUFSIZE) > (MAX_NETDEV_PKTSIZE + CONFIG_NET_GUARDSIZE + CONFIG_NET_LL_GUARDSIZE)" tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll | "NIC"_send() (dev->d_iob->io_data[CONFIG_NET_LL_GUARDSIZE - NET_LL_HDRLEN(dev)]) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | pkt/ipv[4/6]_input()/... | | NICs io vector receive(iov_base to io_data) -------------------------------------------------------------------- 3> Compatible with all old flat buffer NICs TX: tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4) / \ / \ devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...) / \ (copy:tcp) / \ devif_iob_poll(devif_poll_callback()) devif_poll_callback() /* new interface, gather iobs to flat buffer */ / \ / \ devif_poll("NIC"_txpoll) "NIC"_send()(dev->d_buf) RX: [tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead) | | [tcp|udp|icmp|...]_ipv[4|6]_in()/... | | netdev_input() /* new interface, Scatter/gather flat/iob buffer */ | | pkt/ipv[4|6]_input()/... | | NICs io vector receive(Orignal flat buffer) 3. Iperf passthrough on NuttX simulator: ------------------------------------------------- | Protocol | Server | Client | | |-----------------------------------------------| | TCP | 813 | 834 | Mbits/sec | | TCP(Offload) | 1720 | 1100 | Mbits/sec | | UDP | 22 | 757 | Mbits/sec | | UDP(Offload) | 25 | 1250 | Mbits/sec | ------------------------------------------------- Signed-off-by: chao an <anchao@xiaomi.com>
2022-11-23 12:39:40 +01:00
iob_free_chain(iob);
}
else
{
ninfo("Buffered %d bytes\n", buflen);
}
return buflen;
}
/****************************************************************************
* Name: icmp_delivery_callback
*
* Description:
* Copy the icmp package to the application according to the filter
* conditions, but ICMP_ECHO_REPLY is a special message type, if there
* is an application waiting, it will also copy.
*
* Input Parameters:
* conn - A pointer to the ICMP connection structure.
* arg - The context information
*
****************************************************************************/
static int icmp_delivery_callback(FAR struct icmp_conn_s *conn,
FAR void *arg)
{
FAR struct icmp_deliver_s *info = arg;
FAR struct net_driver_s *dev = info->dev;
FAR struct icmp_hdr_s *icmp = IPBUF(info->iphdrlen);
if (icmp_filter(conn->filter, icmp->type) &&
(icmp->type != ICMP_ECHO_REPLY || conn->id != icmp->id ||
conn->dev != dev))
{
return 0;
}
info->delivered = true;
if (devif_conn_event(dev, ICMP_NEWDATA, conn->sconn.list) == ICMP_NEWDATA)
{
icmp_datahandler(dev, conn, info->iphdrlen);
}
return 0;
}
/****************************************************************************
* Name: icmp_deliver
*
* Description:
* Copy the icmp package to the application according to the filter
* conditions, but ICMP_ECHO_REPLY is a special message type, if there
* is an application waiting, it will also copy.
*
* Input Parameters:
* dev - Reference to a device driver structure.
* iphdrlen - The size of the IP header. This may be larger than
* IPv4_HDRLEN if IP option are present.
*
****************************************************************************/
static bool icmp_deliver(FAR struct net_driver_s *dev, uint16_t iphdrlen)
{
struct icmp_deliver_s info;
info.dev = dev;
info.iphdrlen = iphdrlen;
info.delivered = false;
icmp_foreach(icmp_delivery_callback, &info);
return info.delivered;
}
#endif
/****************************************************************************
* Public Functions
****************************************************************************/
/****************************************************************************
2014-06-25 17:12:47 +02:00
* Name: icmp_input
*
* Description:
* Handle incoming ICMP input
*
* Input Parameters:
* dev - The device driver structure containing the received ICMP
* packet
*
* Returned Value:
* None
*
* Assumptions:
* The network is locked.
*
****************************************************************************/
2014-06-28 00:48:12 +02:00
void icmp_input(FAR struct net_driver_s *dev)
{
FAR struct ipv4_hdr_s *ipv4 = IPv4BUF;
FAR struct icmp_hdr_s *icmp;
/* Get the IP header length (accounting for possible options). */
uint16_t iphdrlen = (ipv4->vhl & IPv4_HLMASK) << 2;
#ifdef CONFIG_NET_ICMP_SOCKET
bool delivered = icmp_deliver(dev, iphdrlen);
#endif
#ifdef CONFIG_NET_STATISTICS
g_netstats.icmp.recv++;
#endif
/* The ICMP header immediately follows the IP header */
icmp = IPBUF(iphdrlen);
/* ICMP echo (i.e., ping) processing. This is simple, we only change the
* ICMP type from ECHO to ECHO_REPLY and adjust the ICMP checksum before
* we return the packet.
*/
if (icmp->type == ICMP_ECHO_REQUEST)
{
/* Change the ICMP type */
icmp->type = ICMP_ECHO_REPLY;
/* Swap IP addresses. */
net_ipv4addr_hdrcopy(ipv4->destipaddr, ipv4->srcipaddr);
net_ipv4addr_hdrcopy(ipv4->srcipaddr, &dev->d_ipaddr);
/* Recalculate the ICMP checksum */
#if 0
/* Get the IP header length (accounting for possible options). */
iphdrlen = (ipv4->vhl & IPv4_HLMASK) << 2;
/* The slow way... sum over the ICMP message */
icmp->icmpchksum = 0;
icmp->icmpchksum = ~icmp_chksum(dev,
(((uint16_t)ipv4->len[0] << 8) |
(uint16_t)ipv4->len[1]) - iphdrlen);
if (icmp->icmpchksum == 0)
{
icmp->icmpchksum = 0xffff;
}
#else
/* The quick way -- Since only the type has changed, just adjust the
* checksum for the change of type
*/
if (icmp->icmpchksum >= HTONS(0xffff - (ICMP_ECHO_REQUEST << 8)))
{
icmp->icmpchksum += HTONS(ICMP_ECHO_REQUEST << 8) + 1;
}
else
{
icmp->icmpchksum += HTONS(ICMP_ECHO_REQUEST << 8);
}
#endif
ninfo("Outgoing ICMP packet length: %d (%d)\n",
dev->d_len, (ipv4->len[0] << 8) | ipv4->len[1]);
#ifdef CONFIG_NET_STATISTICS
g_netstats.icmp.sent++;
g_netstats.ipv4.sent++;
#endif
}
nuttx/net:Support to PMTUD RFC 1191 When a router is unable to forward a datagram because it exceeds the MTU of the next-hop network and its Don't Fragment bit is set, the router is required to return an ICMP Destination Unreachable message to the source of the datagram, with the Code indicating "fragmentation needed and DF set". To support the Path MTU Discovery technique specified in this memo, the router MUST include the MTU of that next-hop network in the low-order 16 bits of the ICMP header field that is labelled "unused" in the ICMP specification [7]. The high-order 16 bits remain unused, and MUST be set to zero. Thus, the message has the following format: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 3 | Code = 4 | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | unused = 0 | Next-Hop MTU | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Internet Header + 64 bits of Original Datagram Data | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ RFC 1185 Packet Too Big Message 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Code | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | MTU | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | As much of invoking packet | + as will fit without the ICMPv6 packet + | exceeding 576 octets | IPv6 Fields: Destination Address Copied from the Source Address field of the invoking packet. ICMPv6 Fields: Type 2 Code 0 MTU The Maximum Transmission Unit of the next-hop link. Description A Packet Too Big MUST be sent by a router in response to a packet that it cannot forward because the packet is larger than the MTU of the outgoing link. The information in this message is used as part of the Path MTU Discovery process [RFC-1191]. Signed-off-by: wangchen <wangchen41@xiaomi.com>
2023-05-11 13:32:30 +02:00
#if CONFIG_NET_ICMP_PMTU_ENTRIES > 0
else if (icmp->type == ICMP_DEST_UNREACHABLE)
{
if (icmp->icode == ICMP_FRAG_NEEDED)
{
FAR struct icmp_pmtu_entry *entry;
FAR struct ipv4_hdr_s *inner;
int mtu;
mtu = ntohs(icmp->data[0]) << 16 | ntohs(icmp->data[1]);
if (mtu <= 0)
{
goto typeerr;
}
inner = (FAR struct ipv4_hdr_s *)(icmp + 1);
entry = icmpv4_find_pmtu_entry(
net_ip4addr_conv32(inner->destipaddr));
if (entry == NULL)
{
icmpv4_add_pmtu_entry(
net_ip4addr_conv32(inner->destipaddr), mtu);
}
else
{
entry->pmtu = mtu;
}
goto icmp_send_nothing;
}
}
#endif
/* Otherwise the ICMP input was not processed */
else
{
#ifdef CONFIG_NET_ICMP_SOCKET
if (delivered)
{
goto icmp_send_nothing;
}
else if (icmp->type == ICMP_ECHO_REQUEST)
{
goto drop;
}
#endif
nwarn("WARNING: Unknown ICMP cmd: %d\n", icmp->type);
goto typeerr;
}
return;
typeerr:
#ifdef CONFIG_NET_STATISTICS
g_netstats.icmp.typeerr++;
#endif
2017-10-24 23:46:32 +02:00
#ifdef CONFIG_NET_ICMP_SOCKET
drop:
#ifdef CONFIG_NET_STATISTICS
g_netstats.icmp.drop++;
#endif
#endif
nuttx/net:Support to PMTUD RFC 1191 When a router is unable to forward a datagram because it exceeds the MTU of the next-hop network and its Don't Fragment bit is set, the router is required to return an ICMP Destination Unreachable message to the source of the datagram, with the Code indicating "fragmentation needed and DF set". To support the Path MTU Discovery technique specified in this memo, the router MUST include the MTU of that next-hop network in the low-order 16 bits of the ICMP header field that is labelled "unused" in the ICMP specification [7]. The high-order 16 bits remain unused, and MUST be set to zero. Thus, the message has the following format: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 3 | Code = 4 | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | unused = 0 | Next-Hop MTU | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Internet Header + 64 bits of Original Datagram Data | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ RFC 1185 Packet Too Big Message 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Code | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | MTU | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | As much of invoking packet | + as will fit without the ICMPv6 packet + | exceeding 576 octets | IPv6 Fields: Destination Address Copied from the Source Address field of the invoking packet. ICMPv6 Fields: Type 2 Code 0 MTU The Maximum Transmission Unit of the next-hop link. Description A Packet Too Big MUST be sent by a router in response to a packet that it cannot forward because the packet is larger than the MTU of the outgoing link. The information in this message is used as part of the Path MTU Discovery process [RFC-1191]. Signed-off-by: wangchen <wangchen41@xiaomi.com>
2023-05-11 13:32:30 +02:00
#if defined(CONFIG_NET_ICMP_SOCKET) || CONFIG_NET_ICMP_PMTU_ENTRIES > 0
nuttx/net:Support to PMTUD RFC 1191 When a router is unable to forward a datagram because it exceeds the MTU of the next-hop network and its Don't Fragment bit is set, the router is required to return an ICMP Destination Unreachable message to the source of the datagram, with the Code indicating "fragmentation needed and DF set". To support the Path MTU Discovery technique specified in this memo, the router MUST include the MTU of that next-hop network in the low-order 16 bits of the ICMP header field that is labelled "unused" in the ICMP specification [7]. The high-order 16 bits remain unused, and MUST be set to zero. Thus, the message has the following format: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 3 | Code = 4 | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | unused = 0 | Next-Hop MTU | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Internet Header + 64 bits of Original Datagram Data | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ RFC 1185 Packet Too Big Message 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Code | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | MTU | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | As much of invoking packet | + as will fit without the ICMPv6 packet + | exceeding 576 octets | IPv6 Fields: Destination Address Copied from the Source Address field of the invoking packet. ICMPv6 Fields: Type 2 Code 0 MTU The Maximum Transmission Unit of the next-hop link. Description A Packet Too Big MUST be sent by a router in response to a packet that it cannot forward because the packet is larger than the MTU of the outgoing link. The information in this message is used as part of the Path MTU Discovery process [RFC-1191]. Signed-off-by: wangchen <wangchen41@xiaomi.com>
2023-05-11 13:32:30 +02:00
icmp_send_nothing:
#endif
dev->d_len = 0;
}
#endif /* CONFIG_NET_ICMP */
#endif /* CONFIG_NET */