Commit Graph

65 Commits

Author SHA1 Message Date
chao an
62004a28a6 net/d_buf: remove d_buf reference from l3/l4
l3/l4 stack will decouple the reference of d_buf gradually, Only legacy
devices still retain d_buf support, new net devices will use d_iob

Signed-off-by: chao an <anchao@xiaomi.com>
2022-12-04 20:37:14 +08:00
chao an
34d2cde8a8 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>
2022-12-03 11:47:04 +08:00
chao an
615845d9a6 net/devif: add common interface to build L2 headers
So we can hide the implementation details of L2 NICs.

Signed-off-by: chao an <anchao@xiaomi.com>
2022-11-28 14:44:15 +01:00
chao an
8850dee746 net/devif: move preprocess of txpoll into common code
Signed-off-by: chao an <anchao@xiaomi.com>
2022-11-27 12:11:12 +08:00
chao an
db88554d7d net/devif: add common funtion to build L2 headers before sending
Generic callback before device output to build L2 headers before sending

Signed-off-by: chao an <anchao@xiaomi.com>
2022-11-27 12:11:12 +08:00
Fotis Panagiotopoulos
143d1322ea Added handling of MSG_WAITALL flag in TCP recv. 2022-10-13 18:22:05 +08:00
chao.an
162fcd10ca net: cleanup pvconn reference to avoid confuse
More reference:
https://github.com/apache/incubator-nuttx/pull/5252
https://github.com/apache/incubator-nuttx/pull/5434

Signed-off-by: chao.an <anchao@xiaomi.com>
2022-08-26 20:58:11 +08:00
Alexander Lunev
5b13797cce net/tcp/tcp_send*: reliably obtain the TCP connection pointer in TCP event handlers
Do not use pvconn argument to get the TCP connection pointer because pvconn is
normally NULL for some events like NETDEV_DOWN. Instead, the TCP connection pointer
can be reliably obtained from the corresponding TCP socket.
2022-01-18 16:14:38 +08:00
Petro Karashchenko
8d3bf05fd2 include: fix double include pre-processor guards
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2022-01-16 11:11:14 -03:00
Alexander Lunev
4ac7945676 net/devif/devif_callback.c: made the connection event list doubly linked.
The resulting time complexities are as follows:
* devif_callback_alloc() time complexity is O(1) (i.e. O(n) to fill the whole list).
* devif_callback_free() time complexity is O(1) (i.e. O(n) to empty the whole list).
* devif_conn_event() time complexity is O(n).
2021-09-21 03:51:06 -07:00
Alexander Lunev
36fbedcbfc net/devif/devif_callback.c: corrected the connection event list to work as FIFO instead of LIFO.
In case of enabled packet forwarding mode, packets were forwarded in a reverse order
because of LIFO behavior of the connection event list.
The issue exposed only during high network traffic. Thus the event list started to grow
that resulted in changing the order of packets inside of groups of several packets
like the following: 3, 2, 1, 6, 5, 4, 8, 7 etc.

Remarks concerning the connection event list implementation:
* Now the queue (list) is FIFO as it should be.
* The list is singly linked.
* The list has a head pointer (inside of outer net_driver_s structure),
  and a tail pointer is added into outer net_driver_s structure.
* The list item is devif_callback_s structure.
  It still has two pointers to two different list chains (*nxtconn and *nxtdev).
* As before the first argument (*dev) of the list functions can be NULL,
  while the other argument (*list) is effective (not NULL).
* An extra (*tail) argument is added to devif_callback_alloc()
  and devif_conn_callback_free() functions.
* devif_callback_alloc() time complexity is O(1) (i.e. O(n) to fill the whole list).
* devif_callback_free() time complexity is O(n) (i.e. O(n^2) to empty the whole list).
* devif_conn_event() time complexity is O(n).
2021-09-18 21:01:39 -05:00
Brennan Ashton
68b526b335 tcp: Remove incomplete support for TCP reassembly 2020-12-09 07:25:26 -08:00
Peter van der Perk
55d9e5f7af net: Add SocketCAN support 2020-06-15 08:07:19 -06:00
Xiang Xiao
5c5c08efcd network: simplify the timeout process logic
1.Consolidate absolute to relative timeout conversion into one place(_net_timedwait)
2.Drive the wait timeout logic by net_timedwait instead of devif_timer
This patch help us remove devif_timer(period tick) to save the power in the future.

Change-Id: I534748a5d767ca6da8a7843c3c2f993ed9ea77d4
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-01-11 08:24:49 -06:00
Xiang Xiao
87cf5c58ae Correct some problems with network timed events when there are multiple network devices in the configuration.
Squashed commit of the following:

Author: Gregory Nutt <gnutt@nuttx.org>

    Ran nxstyle against many of the affected files.  But this job was too big for today.  Many of the network drivers under arch are highly non-compiant and generate many, many faults from nxstyle.  Those will have to be visited again another day.

Author: Xiang Xiao <xiaoxiang@xiaomi.com>

    This effects all network drivers as well as timing related portions of net/: devif_poll_tcp_timer shouldn't be skipped in the multiple card case.  devif_timer will be called multiple time in one period if the multiple card exist, the elapsed time calculated for the first callback is right, but the flowing callback in the same period is wrong(very short) because the global variable g_polltimer is used in the calculation.  So let's pass the delay time to devif_timer and remove g_polltimer.
2019-12-24 10:37:30 -06:00
Xiang Xiao
e1e192fc33 net/: icmp[v6] fix ping[6] complain 'WARNING: Received after timeout'. The root cause is that icmp[v6]_pollsetup monitors ICMP[v6]_NEWDATA, but icmp[v6]_input reports ICMP[v6]_ECHOREPLY. This change lets icmp[v6]_input report report ICMP[v6]_NEWDATA to fix this issue and remove ICMP[v6]_ECHOREPLY to avoid the wrong usage in the future. 2019-11-27 08:09:51 -06:00
Gregory Nutt
433f29eea2 net/: Rename CONFIG_TCP_REASSEMBLY to CONFIG_IPv4_REASSEMBLY. Add some fixes to get a clen compile with CONFIG_IPv4_REASSEMBLY enabled. There are several problems with the current implementation: It is untested (and depends on CONFIG_EXPERMIMENTAL). It uses some Ethernet specific definitions (and depends on CONFIG_NET_ETHERNET). There is only a single reassembly buffer. The last two issues prevent use of this feature in any context where IPv4 packets may be reassembled for multiple network devices concurrently). 2018-06-24 12:06:12 -06:00
Gregory Nutt
8fdbb1e0a4 Elimate use of the non-standard type systime_t and replace it the equivalent, standard type clock_t
Squashed commit of the following:

    sched:  Rename all use of system_t to clock_t.
    syscall:  Rename all use of system_t to clock_t.
    net:  Rename all use of system_t to clock_t.
    libs:  Rename all use of system_t to clock_t.
    fs:  Rename all use of system_t to clock_t.
    drivers:  Rename all use of system_t to clock_t.
    arch:  Rename all use of system_t to clock_t.
    include:  Remove definition of systime_t; rename all use of system_t to clock_t.
2018-06-16 12:16:13 -06:00
Gregory Nutt
0c007be4bd This comment adds (1) basic support for AF_BLUETOOTH sockets. The logic compiles but is still incomplete. Support for Bluetooth is general is still dependent on CONFIG_EXPERMIMENTAL because it is not yet ready for used.
Squashed commit of the following:

    wireless/bluetooth:  Some small changes that gets to a clean compile by just eliminating some incorrect implementations (still with a lot of warnings.  The logic is still incomplete but now not so lethal.
    wireless/bluetooth:  Restructuring:  Connection interfaces should internal to wireless/bluetooth.  include/nuttx/wireless/bt_conn.h removed and merged with wireless/bluetooth/bt_conn.h.  Several fix to get closer to bt_netdev.c compiling.  Need to design some not interfaces and use some existing interfaces to send and receiv packets.
    wireless/bluetooth: Some organization with some network device compile errors fixed.  Still not even close to compiling.
    net/bluetooth:  Fix numerous compile issues; Still open design issues with regard to the interface with the Bluetooth stack.
    wireless/bluetooth:  Create bt_netdev.c with a crude copy of mac802154_netdev.c.  Does not not even compile yet.
    include/nuttx/net:  Add bluetooth.h.  Content is not yet correct.
    net/netpackets:  Add bluetooth.h.  Update net/bluetooth to use new socket address definition.
    net/bluetooth:  Some fixes for initial build.
    net/bluetooth:  Add initial support for Bluetooth sockets.  The initial cut is just the a clone of the IEEE 802.15.4 socket support with name changes.
    net/ieee802154:  Fix some typos noted when cloning to create net/bluetooth.
2018-03-31 14:55:03 -06:00
Gregory Nutt
b54ffe858a Standardization of some function headers. 2018-03-13 09:52:27 -06:00
Gregory Nutt
7cf88d7dbd Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
Gregory Nutt
cff4ac6845 Squashed commit of the following:
commit 2a3ab1652a2c95bcfc8be8380fc7cbdcb6472938
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Sat Aug 19 08:44:31 2017 -0600

    PF_IEEE802154: Finish some missing bind() logic.  Add configs/sim configuration for testing.

commit 59be4b846a6e3bfe82087a888e3fdac9c3c414e5
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 19:30:04 2017 -0600

    PF_IEEE802154:  More renaming to decouple 6LoPAN from radios in general.

commit 69fabb1aea76e54381bdc13de28a3f1441fb42f4
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 19:21:11 2017 -0600

    PF_IEEE802154: Missed a few renamings.

commit ff0af1bb25567720934cc1c2a721ccd92cc35f89
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 17:46:58 2017 -0600

    PF_IEEE802154:  A few bugfixes

commit 01c7c84afd00cf907d280d30cfaf0fb2cf90e02e
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 17:01:31 2017 -0600

    PF_IEEE802154:  A few bugfixes

commit dcef4056d1c1488c93151135f3b7106977faa903
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 16:31:05 2017 -0600

    PF_IEEE802154: Bring in framework for sendto/recvfrom.  Currently just a crude port of functions from net/pkt and do not provide the implemenation needed.

commit 68c5b7b6dd3ab7eb2d0c19890abb38e6561b140e
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 15:18:31 2017 -0600

    Trivial fix to typo in comment

commit fd0af534c089569ccdbd59f13b85453de0a653ad
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 15:07:20 2017 -0600

    PF_IEEE802154: Add device lookup logic; Rename some things that used to be used only by 6LoWPAN but now must be shared with PF_IEEE802154 and need more generic naming.

commit 4fc80a1659f1c699b050775cefc7f6b631b46114
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 13:49:54 2017 -0600

    PF_IEEE802154: Add driver poll logic.

commit d83f71992df8764faa93f9425f1a7602a758f730
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 13:28:59 2017 -0600

    PF_IEEE802154:  Add frame input function.

commit 77561b8c4d5d7be1f8d8eb62cf1a07229afe2048
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 12:46:29 2017 -0600

    PF_IEEE802154: Socket type should be SOCK_DGRAM.  Hook in socket interface.

commit c0f90350282e9905d7f26a1b30f04cc6d6794648
Merge: 8b518abfd0 169c55e546
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 09:36:32 2017 -0600

    Merge remote-tracking branch 'origin/master' into pf_ieee802154

commit 8b518abfd07d492f5148f2c5fdf65604de9822da
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 09:35:39 2017 -0600

    PF_IEEE802154: Add initialization and connection management logic.

commit 98b62620b3cb420041d8ad14204f9410a8aace8c
Author: Gregory Nutt <gnutt@nuttx.org>
Date:   Fri Aug 18 07:52:51 2017 -0600

    PF_IEEE802154: Add basic build support and socket interface framework.
2017-08-19 08:48:52 -06:00
Gregory Nutt
835ae2999d networking: Move a little more to the inet/ subdirectory. 2017-08-07 13:03:23 -06:00
Gregory Nutt
bea75baaab IP forwarding: Two bugfixes (1) IPFWD poll event must be unique and different from other device poll events otherwise, some other waiting task might get the poll, (2) Add logic necessary to forward 6LoWPAN packets. 2017-07-08 09:56:08 -06:00
Gregory Nutt
aa2e9c15a5 IP forwarding: Major rearchitecting of the outgoing portion of the IP forwarding logic necessary into to properly received device-related forwarding events. 2017-07-07 20:19:26 -06:00
Gregory Nutt
04716a65a5 IP forwading: Add optional support to forward broadcast and multicast packets. 2017-07-06 07:26:01 -06:00
Gregory Nutt
b297066eb9 IP Forwarding: Add IPv4 packet forwarding logic. Initial commit is an untested clone of the IPv6 forwarding logic with a few minor logic changes for IPv4. 2017-07-05 15:12:29 -06:00
Gregory Nutt
a93f130002 6LoWPAN: Add configurable support for 6LoWPAN star topology. With this change, the endpoints which are the 'points' of the star will forward all traffic to the coordinator. The coordinator is assumed to be the 'hub' of the star. This function also implements IPv6 forwarding. That forwarding implementation is minimal, just enough to support the forwarding requirements of the star hub node. 2017-06-28 15:39:21 -06:00
Gregory Nutt
0de294a586 Fix lots of occurrences of 'the the', 'the there', 'the these', 'the then', 'the they. 2017-05-11 13:35:56 -06:00
Gregory Nutt
2043e1a114 IOBs: Move from driver/iob to a better location in mm/iob 2017-05-09 07:35:30 -06:00
Gregory Nutt
d5207efb5a Be consistent... Use Name: consistent in function headers vs Function: 2017-04-21 16:33:14 -06:00
Gregory Nutt
bcc6b61fc1 Move include/nuttx/net/iob.h to include/drivers/iob.h; rename CONFIG_NET_IOB to CONFIG_DRIVERS_IOB 2017-04-20 14:53:30 -06:00
Gregory Nutt
732f0855c6 6loWPAN: Fleshes out framwork for IEEE802.15.4 send. But still has some gaping holes. 2017-03-31 15:09:07 -06:00
Jussi Kivilinna
cd3c9634c8 Add user-space networking stack API (usrsock)
User-space networking stack API allows user-space daemon to
provide TCP/IP stack implementation for NuttX network.

Main use for this is to allow use and seamless integration of
HW-provided TCP/IP stacks to NuttX.

For example, user-space daemon can translate /dev/usrsock
API requests to HW TCP/IP API requests while rest of the
user-space can access standard socket API, with socket
descriptors that can be used with NuttX system calls.
2017-03-31 08:58:14 -06:00
Gregory Nutt
4f208600aa Replace confusing references to uIP with just 'the network' 2016-05-30 09:31:44 -06:00
Gregory Nutt
e4e6181098 net/: IEEE 802.15-4 basic build build structure. 2016-02-25 10:43:51 -06:00
Gregory Nutt
dea24c503a Networking: Remove the HSEC argument from devif_timer. 2016-02-03 12:12:41 -06:00
Gregory Nutt
3bcdb218ff Networking: Get rid of the tcp_mss macro. It is confusing and only obfuscates what is really going on 2015-08-27 08:39:17 -06:00
Gregory Nutt
8f7752d956 Networking: Modify how callback structures are stored to avoid another potential use of a stal pointer. 2015-06-03 08:11:57 -06:00
Gregory Nutt
04a661a97c TCP networking: Add support for network driver events 2015-05-30 09:12:27 -06:00
Gregory Nutt
e81f279315 Networking: Modify event list handling: Now there are two event lists each device structure: (1) One is for ARP and ICMP data related evetns, the other is for device related events. Callback allocation/free routines no accept a device paramter as well as a list: If the device paramter is added, then the callback goes into both the connection-related liast AND the device event list. Thus each socket type can received both custom data-related events as well as common device related events. 2015-05-28 12:01:38 -06:00
Gregory Nutt
3260db7dbe Reorder some network events bitset to make the event class clearer 2015-05-28 08:55:36 -06:00
Gregory Nutt
4fd0192a3e Networking: No that ARP, ICMP, and ICMPv6 reside in the same list we need to make the definitions of ARP_POLL, ICMP_POLL, and ICMPv6_POLL different because there are small differences in the context for each. 2015-05-27 15:04:51 -06:00
Gregory Nutt
b94321cfb4 Extend device specific callbacks to ICMPv6 and ARP. Fix some IPv6 compilation errors that have crept in 2015-05-27 11:39:44 -06:00
Gregory Nutt
8ecb84ed89 Improve icmp handling if MULTINIC enabled. From Max Neklyudov 2015-05-27 09:26:00 -06:00
Gregory Nutt
24d800398e Networking: Break out Ethernet definitions into a separate file; add IPv6 multicast addresses as common globals, Ethernet drivers need to filter link-local, all nodes Ethernet address 2015-02-04 14:51:20 -06:00
Gregory Nutt
66f0198e3f Networking: Add support for sending ICMPv6 Router Solicitation and receiving Router advertisement. This is a part of the ICMPv6 auto-configuration logic is still still not complete 2015-02-03 13:25:28 -06:00
Gregory Nutt
a07a764d42 ICMPv6: Adds basic logic to support verification that we have the IPv6 address in the Neighbor Table. This feature is important because otherwise the first packet sent to a remote peer will fail (it will be replaced with a Neighbor Solicitation). 2015-02-02 13:44:31 -06:00
Gregory Nutt
b1b8c0e47d IPv6: More framework for automatic neighbor solicition. I think this also corrects and error in a broadcast/multicast address chedk 2015-02-02 11:34:51 -06:00
Gregory Nutt
3385082b99 IPv6: Remove an unusd file (kruft from original IPv6 develoment); Add inital hooks to support automatic Neighbor Solitication 2015-02-02 10:49:50 -06:00