nuttx/net/utils/utils.h

490 lines
16 KiB
C
Raw Permalink Normal View History

/****************************************************************************
* net/utils/utils.h
*
* SPDX-License-Identifier: Apache-2.0
*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership. The
* ASF licenses this file to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance with the
* License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
* License for the specific language governing permissions and limitations
* under the License.
*
****************************************************************************/
#ifndef __NET_UTILS_UTILS_H
#define __NET_UTILS_UTILS_H
/****************************************************************************
* Included Files
****************************************************************************/
#include <nuttx/config.h>
#include <nuttx/net/net.h>
#include <nuttx/net/ip.h>
#include <nuttx/net/netdev.h>
/****************************************************************************
* Pre-processor Definitions
****************************************************************************/
/* Some utils for port selection */
#define NET_PORT_RANDOM_INIT(port) \
do \
{ \
net_getrandom(&(port), sizeof(port)); \
(port) = (port) % (CONFIG_NET_DEFAULT_MAX_PORT - \
CONFIG_NET_DEFAULT_MIN_PORT + 1); \
(port) += CONFIG_NET_DEFAULT_MIN_PORT; \
} while (0)
/* Get next net port number, and make sure that the port number is within
* range. In host byte order.
*/
#define NET_PORT_NEXT_H(hport) \
do \
{ \
++(hport); \
if ((hport) > CONFIG_NET_DEFAULT_MAX_PORT || \
(hport) < CONFIG_NET_DEFAULT_MIN_PORT) \
{ \
(hport) = CONFIG_NET_DEFAULT_MIN_PORT; \
} \
} while (0)
/* Get next net port number, and make sure that the port number is within
* range. In both network & host byte order.
*/
#define NET_PORT_NEXT_NH(nport, hport) \
do \
{ \
NET_PORT_NEXT_H(hport); \
(nport) = HTONS(hport); \
} while (0)
/****************************************************************************
* Public Types
****************************************************************************/
/* These values control the behavior of net_timeval2desc */
enum tv2ds_remainder_e
{
TV2DS_TRUNC = 0, /* Truncate microsecond remainder */
TV2DS_ROUND, /* Round to the nearest full decisecond */
TV2DS_CEIL /* Force to next larger full decisecond */
};
/****************************************************************************
* Public Data
****************************************************************************/
#ifdef __cplusplus
#define EXTERN extern "C"
extern "C"
{
#else
#define EXTERN extern
#endif
/****************************************************************************
* Public Function Prototypes
****************************************************************************/
struct net_driver_s; /* Forward reference */
struct timeval; /* Forward reference */
/****************************************************************************
* Name: net_breaklock
*
* Description:
* Break the lock, return information needed to restore re-entrant lock
* state.
*
****************************************************************************/
int net_breaklock(FAR unsigned int *count);
/****************************************************************************
* Name: net_restorelock
*
* Description:
* Restore the locked state
*
* Returned Value:
* Zero (OK) is returned on success; a negated errno value is returned on
* failured (probably -ECANCELED).
*
****************************************************************************/
int net_restorelock(unsigned int count);
/****************************************************************************
* Name: net_dsec2timeval
*
* Description:
* Convert a decisecond value to a struct timeval. Needed by getsockopt()
* to report timeout values.
*
* Input Parameters:
* dsec The decisecond value to convert
* tv The struct timeval to receive the converted value
*
* Returned Value:
* None
*
* Assumptions:
*
****************************************************************************/
void net_dsec2timeval(uint16_t dsec, FAR struct timeval *tv);
/****************************************************************************
* Name: net_dsec2tick
*
* Description:
* Convert a decisecond value to a system clock ticks. Used by IGMP logic.
*
* Input Parameters:
* dsec The decisecond value to convert
*
* Returned Value:
* The decisecond value expressed as system clock ticks
*
****************************************************************************/
unsigned int net_dsec2tick(int dsec);
/****************************************************************************
* Name: net_timeval2dsec
*
* Description:
* Convert a struct timeval to deciseconds. Needed by setsockopt() to
* save new timeout values.
*
* Input Parameters:
* tv - The struct timeval to convert
* remainder - Determines how to handler the microsecond remainder
*
* Returned Value:
* The converted value
*
* Assumptions:
*
****************************************************************************/
unsigned int net_timeval2dsec(FAR struct timeval *tv,
enum tv2ds_remainder_e remainder);
/****************************************************************************
* Name: net_getrandom
*
* Description:
* Fill a buffer of arbitrary length with randomness. This function is
* guaranteed to be success.
*
* Input Parameters:
* bytes - Buffer for returned random bytes
* nbytes - Number of bytes requested.
*
****************************************************************************/
void net_getrandom(FAR void *bytes, size_t nbytes);
/****************************************************************************
* Name: net_ipv4_mask2pref
*
* Description:
* Convert a 32-bit netmask to a prefix length. The NuttX IPv4
* networking uses 32-bit network masks internally. This function
* converts the IPv4 netmask to a prefix length.
*
* The prefix length is the number of MS '1' bits on in the netmask.
* This, of course, assumes that all MS bits are '1' and all LS bits are
* '0' with no intermixed 1's and 0's. This function searches from the MS
* bit until the first '0' is found (this does not necessary mean that
* there might not be additional '1' bits following the firs '0', but that
* will be a malformed netmask.
*
* Input Parameters:
* mask An IPv4 netmask in the form of in_addr_t
*
* Returned Value:
* The prefix length, range 0-32 on success; This function will not
* fail.
*
****************************************************************************/
#ifdef CONFIG_NET_IPv4
uint8_t net_ipv4_mask2pref(in_addr_t mask);
#endif
/****************************************************************************
* Name: net_ipv6_common_pref
*
* Description:
* Calculate the common prefix length of two IPv6 addresses.
*
* Input Parameters:
* a1,a2 Points to IPv6 addresses in the form of uint16_t[8]
*
* Returned Value:
* The common prefix length, range 0-128 on success; This function will
* not fail.
*
****************************************************************************/
#ifdef CONFIG_NET_IPv6
uint8_t net_ipv6_common_pref(FAR const uint16_t *a1, FAR const uint16_t *a2);
#endif
/****************************************************************************
* Name: net_ipv6_mask2pref
*
* Description:
* Convert a 128-bit netmask to a prefix length. The NuttX IPv6
* networking uses 128-bit network masks internally. This function
* converts the IPv6 netmask to a prefix length.
*
* The prefix length is the number of MS '1' bits on in the netmask.
* This, of course, assumes that all MS bits are '1' and all LS bits are
* '0' with no intermixed 1's and 0's. This function searches from the MS
* bit until the first '0' is found (this does not necessary mean that
* there might not be additional '1' bits following the firs '0', but that
* will be a malformed netmask.
*
* Input Parameters:
* mask Points to an IPv6 netmask in the form of uint16_t[8]
*
* Returned Value:
* The prefix length, range 0-128 on success; This function will not
* fail.
*
****************************************************************************/
#ifdef CONFIG_NET_IPv6
uint8_t net_ipv6_mask2pref(FAR const uint16_t *mask);
#endif
/****************************************************************************
* Name: net_ipv6_pref2mask
*
* Description:
* Convert a IPv6 prefix length to a network mask. The prefix length
* specifies the number of MS bits under mask (0-128)
*
* Input Parameters:
* mask - The location to return the netmask.
* preflen - Determines the width of the netmask (in bits). Range 0-128
*
* Returned Value:
* None
*
****************************************************************************/
#ifdef CONFIG_NET_IPv6
void net_ipv6_pref2mask(net_ipv6addr_t mask, uint8_t preflen);
#endif
/****************************************************************************
* Name: net_ipv6_payload
*
* Description:
* Given a pointer to the IPv6 header, this function will return a pointer
* to the beginning of the L4 payload.
*
* Input Parameters:
* ipv6 - A pointer to the IPv6 header.
* proto - The location to return the protocol number in the IPv6 header.
*
* Returned Value:
* A pointer to the beginning of the payload.
*
****************************************************************************/
#ifdef CONFIG_NET_IPv6
FAR void *net_ipv6_payload(FAR struct ipv6_hdr_s *ipv6, FAR uint8_t *proto);
#endif
/****************************************************************************
* Name: net_iob_concat
*
* Description:
* Concatenate iob_s chain iob2 to iob1, if CONFIG_NET_RECV_PACK is
* endabled, pack all data in the I/O buffer chain.
*
* Returned Value:
* The number of bytes actually buffered is returned. This will be either
* zero or equal to iob1->io_pktlen.
*
****************************************************************************/
#ifdef CONFIG_MM_IOB
uint16_t net_iob_concat(FAR struct iob_s **iob1, FAR struct iob_s **iob2);
#endif
/****************************************************************************
* Name: net_chksum_adjust
*
* Description:
* Adjusts the checksum of a packet without having to completely
* recalculate it, as described in RFC 3022, Section 4.2, Page 9.
*
* Input Parameters:
* chksum - points to the chksum in the packet
* optr - points to the old data in the packet
* olen - length of old data
* nptr - points to the new data in the packet
* nlen - length of new data
*
* Limitations:
* The algorithm is applicable only for even offsets and even lengths.
*
****************************************************************************/
void net_chksum_adjust(FAR uint16_t *chksum,
FAR const uint16_t *optr, ssize_t olen,
FAR const uint16_t *nptr, ssize_t nlen);
/****************************************************************************
* Name: tcp_chksum, tcp_ipv4_chksum, and tcp_ipv6_chksum
*
* Description:
* Calculate the TCP checksum of the packet in d_buf and d_appdata.
*
* The TCP checksum is the Internet checksum of data contents of the
* TCP segment, and a pseudo-header as defined in RFC793.
*
* Note: The d_appdata pointer that points to the packet data may
* point anywhere in memory, so it is not possible to simply calculate
* the Internet checksum of the contents of the d_buf buffer.
*
* Returned Value:
* The TCP checksum of the TCP segment in d_buf and pointed to by
* d_appdata.
*
****************************************************************************/
#ifdef CONFIG_NET_IPv4
uint16_t tcp_ipv4_chksum(FAR struct net_driver_s *dev);
#endif
#ifdef CONFIG_NET_IPv6
uint16_t tcp_ipv6_chksum(FAR struct net_driver_s *dev);
#endif
#if defined(CONFIG_NET_IPv4) && defined(CONFIG_NET_IPv6)
uint16_t tcp_chksum(FAR struct net_driver_s *dev);
#elif defined(CONFIG_NET_IPv4)
# define tcp_chksum(d) tcp_ipv4_chksum(d)
#else /* if defined(CONFIG_NET_IPv6) */
# define tcp_chksum(d) tcp_ipv6_chksum(d)
#endif
/****************************************************************************
* Name: udp_ipv4_chksum
*
* Description:
* Calculate the UDP/IPv4 checksum of the packet in d_buf and d_appdata.
*
****************************************************************************/
#if defined(CONFIG_NET_UDP_CHECKSUMS) && defined(CONFIG_NET_IPv4)
uint16_t udp_ipv4_chksum(FAR struct net_driver_s *dev);
#endif
/****************************************************************************
* Name: udp_ipv6_chksum
*
* Description:
* Calculate the UDP/IPv6 checksum of the packet in d_buf and d_appdata.
*
****************************************************************************/
#if defined(CONFIG_NET_UDP_CHECKSUMS) && defined(CONFIG_NET_IPv6)
uint16_t udp_ipv6_chksum(FAR struct net_driver_s *dev);
#endif
/****************************************************************************
* Name: icmp_chksum
*
* Description:
* Calculate the checksum of the IPv4 ICMP message
*
****************************************************************************/
#ifdef CONFIG_NET_ICMP
uint16_t icmp_chksum(FAR struct net_driver_s *dev, int len);
#endif
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
/****************************************************************************
* Name: icmp_chksum_iob
*
* Description:
* Calculate the checksum of the ICMP message, the input can be an I/O
* buffer chain
*
****************************************************************************/
#if defined(CONFIG_NET_ICMP) && defined(CONFIG_MM_IOB)
uint16_t icmp_chksum_iob(FAR struct iob_s *iob);
#endif /* CONFIG_NET_ICMP && CONFIG_MM_IOB */
/****************************************************************************
* Name: icmpv6_chksum
*
* Description:
* Calculate the checksum of the ICMPv6 message
*
****************************************************************************/
#ifdef CONFIG_NET_ICMPv6
uint16_t icmpv6_chksum(FAR struct net_driver_s *dev, unsigned int iplen);
#endif
/****************************************************************************
* Name: cmsg_append
*
* Description:
* Append specified data into the control message, msg_control and
* msg_controllen will be changed to the appropriate value when success
*
* Input Parameters:
* msg - Buffer to receive the message.
* level - The level of control message.
* type - The type of control message.
* value - If the value is not NULL, this interface copies the data
* to the appropriate location in msg_control, and modify
* msg_control and msg_controllen.
* If the value is NULL, just modify the corresponding value
* of msg.
* value_len - The value length of control message.
*
* Returned Value:
* On success, a pointer to the start address of control message data,
* the caller can copy the data in.
* On failure, return NULL, because of msg_controllen is not enough
*
****************************************************************************/
FAR void *cmsg_append(FAR struct msghdr *msg, int level, int type,
FAR void *value, int value_len);
#undef EXTERN
#ifdef __cplusplus
}
#endif
#endif /* __NET_UTILS_UTILS_H */