sergiotarxz/nuttx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
47,391 Commits 1 Branch 0 Tags 291 MiB
11de08de27
Commit Graph

74 Commits

Author SHA1 Message Date
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
liyi
391b501639 net: extract l3 header build code into new functions 
Signed-off-by: liyi <liyi25@xiaomi.com>
 2022-11-29 18:36:15 +08:00
chao an
a8d3286258 net: move device buffer define to common header 
Signed-off-by: chao an <anchao@xiaomi.com>
 2022-10-28 00:32:16 -04:00
Petro Karashchenko
08043fb5bc net: unify FAR keyword usage for all net buffer memory mapped buffers 
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
 2022-01-20 01:42:56 +08:00
Alexander Lunev
e9ab3adf23 net/tcp(unbuffered): advance sndseq by +1 because SYN and FIN occupy one sequence number (RFC 793) 2022-01-03 12:18:44 +09:00
Huang Qi
e5c278981a net: Rename IP_TTL to IP_TTL_DEFAULT 
Since a SOL option IP_TTL exist, we should rename this IP_TTL
in netconfig.h to avoid confusion.

Signed-off-by: Huang Qi <huangqi3@xiaomi.com>
Change-Id: Ib04c36553f23bce8d362e97294a8b83eaa050cf3
 2021-07-12 16:30:37 -03:00
chao.an
d4ce70979e net/tcp: change all window relative value type to uint32_t 
1. change all window relative value type to uint32_t
2. move window range validity check(UINT16_MAX) before assembling TCP header

Signed-off-by: chao.an <anchao@xiaomi.com>
 2021-07-07 03:55:41 -05:00
chao.an
aab03ef86d net/tcp: add window scale support 
Reference here:
https://tools.ietf.org/html/rfc1323

Signed-off-by: chao.an <anchao@xiaomi.com>
 2021-07-07 03:55:41 -05:00
chao.an
a5cdc4e69b net/tcp: change the tcp optdata to dynamic arrays 
Signed-off-by: chao.an <anchao@xiaomi.com>
 2021-07-07 03:55:41 -05:00
YAMAMOTO Takashi
1ce13ee731 tcp_reset: Don't copy the peer window 
The current code just leave the window value from the segment
from the peer. It doesn't make sense.

Instead, always use 0.
This matches what NetBSD and Linux do.
(As far as I read their code correctly.)
 2021-06-29 22:23:48 -05:00
YAMAMOTO Takashi
14ec75e7fc tcp: window update improvements 
* Fixes the case where the window was small but not zero.

* tcp_recvfrom: Remove tcp_ackhandler. Instead, simply schedule TX for
  a possible window update and make tcp_appsend decide.

* Replace rcv_wnd (the last advertized window size value) with
  rcv_adv. (the window edge sequence number advertized to the peer)
  rcv_wnd was complicated to deal with because its base (rcvseq) is
  also moving.

* tcp_appsend: Send a window update even if there are no other reasons
  to send an ack.
  Namely, send an update if it increases the window by
    * 2 * mss
    * or the half of the max possible window size
 2021-06-13 21:20:24 -05:00
YAMAMOTO Takashi
1f6fdf04b7 tcp: Extract MSS calculation from tcp_synack 
I plan to use it for recv window update decision logic.
 2021-06-13 21:20:24 -05:00
YAMAMOTO Takashi
69b3f034a4 tcp: Move buffered/unbuffered common code to tcp_send.c 2021-06-03 21:33:10 -05:00
chao.an
881dd9d62d net/tcp: add a member to record the current receiving window 
Change-Id: Ic4c46d643a905fdd3a828e563eab4814da70dbe5
Signed-off-by: chao.an <anchao@xiaomi.com>
 2020-12-10 12:23:47 +09:00
chao.an
bf21056001 net/tcp: fallback to unthrottle pool to avoid deadlock 
Add a fallback mechanism to ensure that there are still available
iobs for an free connection, Guarantees all connections will have
a minimum threshold iob to keep the connection not be hanged.

Change-Id: I59bed98d135ccd1f16264b9ccacdd1b0d91261de
Signed-off-by: chao.an <anchao@xiaomi.com>
 2020-11-28 00:03:47 -06:00
GAEHWILER Reto
cbd4e90781 net/tcp: Rectified keepalive fix 
Simplification of previous fix. Also fixes the case when new data should
be sent while a keepalive is not yet acked.
 2020-11-18 17:40:42 -03:00
GAEHWILER Reto
83745652c4 TCP-stack fix for stalled tcp sockets due to broken keepalive 
Fixes an issue where tcp sockets with activated keepalives stalled and
were not properly closed. Poll would not indicate a POLLHUP and therefore
locks down the application.

* tcp_conn_s.tcp_conn_s & tcp_conn_s.keepintvl changed to uint32_t
  According RFC1122 keepidle MUST have a default of 2 hours.
 2020-10-27 11:21:56 -07:00
Gregory Nutt
e5e52446ca net/tcp/tcp_send.c: Trival update to some comments. 2019-12-11 11:19:32 -06:00
Gregory Nutt
66ef6d143a This commit adds an initial implemented of TCP delayed ACKs as specified in RFC 1122. 
Squashed commit of the following:

    net/tmp:  Rename the unacked field of the tcp connection structure to tx_unacked.  Too confusing with the implementation of delayed RX ACKs.

    net/tcp:  Initial implementation of TCP delayed ACKs.

    net/tcp:  Add delayed ACK configuration selection.  Rename tcp_ack() to tcp_synack().  It may or may not send a ACK.  It will always send SYN or SYN/ACK.
 2019-12-08 13:13:51 -06:00
Gregory Nutt
7871e983b8 net/tcp/tcp_send.c: Commit a52ceac13e broke IPv4 sending. In cp_send.c:tcp_ipv4_sendcomplete(), ‘ipv4->vhl’ now needs to be configured before call to tcp_ipv4_chksum(). Noted by Jussi Kivilinna in comit comments. 2019-09-06 09:48:42 -06:00
Gregory Nutt
ac151e05e7 Fix warnings found in build testing. 2019-08-27 16:20:39 -06:00
biantao
f1c46b4498 net/tcp/tcp_send.c: Fix RST packet with wrong ack number. 2019-08-26 10:16:43 -06:00
Gregory Nutt
e59b26370d Squashed commit of the following: 
Fix a few typo/compilation problems.
    net/:  Remove all CONFIG_NET_xxx_TCP_RECVWNDO configuration variables.  They were used only to initialize the d_recwndo of the network device structure which no longer exists.
    net/:  Remove the device TCP receive window field (d_recvwndo) from the device structure.  That value is no longer retained, but is calculated dynamically.
    Remove some dangling references to CONFIG_NET_TCP_RWND_CONTROL.
    net/tcp:  Take read-ahead throttling into account when calculating the TCP receive window size.
    net/tcp: tcp_get_recvwindow() now returns the receive window size directly (vs. indirectly via the device structure).
    net/tcp:  Remove CONFIG_NET_TCP_RWND_CONTROL.  TCP window algorithm is now trigged only by CONFIG_NET_TCP_READAHEAD.
 2018-07-01 07:59:33 -06:00
Gregory Nutt
b32d8b1714 net/tcp and sixlowpan: Separate the the TCP receive window calculations to a separate header file. It must also be used by the special 6LoWPAN TCP logic. 2018-06-30 20:50:07 -06:00
Gregory Nutt
25686a9dac Remove if #if 0 logic. Update some comments. 2018-06-24 15:50:45 -06:00
Gregory Nutt
715d561c41 Trivial spelling corrections in comments; update some comments in a Kconfig file. 2018-06-24 15:01:49 -06:00
Gregory Nutt
0786b5d053 net/tcp: Re-think CONFIG_NET_TCP_RWND_CONTROL TCP windowing controls. 2018-06-24 14:46:12 -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
Masayuki Ishikawa
30070b06df Merged in masayuki2009/nuttx.nuttx/tcp_rcvwnd_control (pull request #555) 
net/tcp: Introduce tcp receive window control based on I/O buffer

NOTE: The algorithm is still experimental but useful for http streaming.

Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>

Approved-by: Gregory Nutt <gnutt@nuttx.org>
 2017-12-18 12:11:52 +00:00
Gregory Nutt
835ae2999d networking: Move a little more to the inet/ subdirectory. 2017-08-07 13:03:23 -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
64933246c3 6loWPAN: Tie 6loWPAN send into common socket send logic. 2017-03-28 14:08:54 -06:00
Gregory Nutt
43eb04bb8f Without lowsyslog() *llinfo() is not useful. Eliminate and replace with *info(). 2016-06-20 11:59:15 -06:00
Gregory Nutt
fc3540cffe Replace all occurrences of vdbg with vinfo 2016-06-11 11:59:51 -06:00
Gregory Nutt
2c95fef501 Remove some empty code section comments 2016-02-26 07:35:55 -06:00
Gregory Nutt
0fb035f76b Standardize some naming in code section comments 2016-02-21 18:09:04 -06:00
Gregory Nutt
cb9e27c3b0 Standardize naming used for public data and function groupings 2015-10-02 16:30:35 -06:00
Gregory Nutt
f4bb7f14e1 Networking: Clean up network status collection and presentation for IPv6 2015-01-24 08:26:12 -06:00
Gregory Nutt
eac4ea7ae5 Networking: Back out a previous change. It is incorrect 2015-01-22 12:10:16 -06:00
Gregory Nutt
46fc574f0f Forgot to remove some debug instrumentation in the last commit 2015-01-22 10:35:11 -06:00
Gregory Nutt
7106469191 Armv7-M: Remove Px4-only setting of stack to 0xff. This is incompatible with standard NuttX stack montitoring logic 2015-01-22 10:09:10 -06:00
Gregory Nutt
5e23a855ec More typos in debug statements 2015-01-21 10:32:59 -06:00
Gregory Nutt
354bc626ab Fix a typo introduced in the last commit 2015-01-21 10:31:18 -06:00
Gregory Nutt
3202819328 IPv6 -- various fixed from problems found in testing 2015-01-21 10:29:18 -06:00
Gregory Nutt
2842b5c5d6 Networking: Fix some problems with debug is enabled; fix a warning 2015-01-19 10:16:37 -06:00
Gregory Nutt
f7663ef0ab Networking: Final detangle off IPv4 and IPv6 TCP/UDP send logic. The Networking subsystem now compiles with IPv6 enabled 2015-01-18 08:56:05 -06:00
Gregory Nutt
8f8259a0d6 Networking: UDP and TCP MSS depends on the IP header size (as well as the link layer header size) and cannot be represented with a single value. 2015-01-16 15:03:10 -06:00
Gregory Nutt
2663538b0a Networking: Replace all references to the macros net_ipaddr_copy, net_ipaddr_hdrcopy, net_ipaddr_cmp, net_ipaddr_hdrcmp, and net_ipaddr_maskcmp with the appropriate IPv4 or IPv6 version of the macro (such as net_ipv4addr_copy). The goal is to support both IPv4 and IPv6 simultaneously. This requires that the macros be distinct and not conditionally defined to one on or the other. 2015-01-16 13:01:08 -06:00
Gregory Nutt
bee89be4f4 Networking: Drivers can have both IPv4 and IPv6 addesses, but a socket can only only one or the other; The socket connnection structures need to include a union of IPv4 and IPv6 addresses for the local address binding and for the remote address connections 2015-01-16 10:01:54 -06:00