Commit Graph

649 Commits

Author SHA1 Message Date
Alexander Lunev
64dd669749 net/tcp/sendfile: retransmit only one the earliest not acknowledged segment
(according to RFC 6298 (5.4)). The issue is the same as it was in tcp_send_unbuffered.c.
2022-01-20 18:37:39 +08:00
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
6bb7a92a9a net/tcp/tcp_send*: added debug asserts for TCP_ACKDATA, TCP_REXMIT and TCP_DISCONN_EVENTS flags 2022-01-19 10:45:38 +08:00
Alexander Lunev
79e609a8c8 net/tcp/sendfile: swapped the location of TCP_DISCONN_EVENTS and TCP_ACKDATA conditions towards tcp_send_unbuffered.c unification 2022-01-19 00:13:38 +08:00
Petro Karashchenko
9551de7115 net: use HTONS, NTOHS, HTONL, NTOHL macro in kernel code
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2022-01-18 10:59:47 +01:00
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
Alexander Lunev
f61f276120 net/tcp/sendfile: TCP retransmission could not start because of incorrect snd_ackcb callback handling:
Both the snd_ackcb and snd_datacb callbacks were created and destroyed right after sending every packet.
Whenever TCP_REXMIT event occurred due to TCP send timeout, TCP_REXMIT was ignored because
snd_ackcb callback had been destroyed by the time.
The issue is fixed as follows:
- both the snd_ackcb and snd_datacb callbacks are combined into one snd_cb callback
  (the same way as in tcp_send_unbuffered.c).
- the snd_cb callback lives until all requested data (via sendfile) is sent,
  including all ACKs and possible retransmissions.

As a positive side effect of the code optimization / fix, sendfile TCP payload throughput is increased.
2022-01-18 02:03:40 +08:00
Alexander Lunev
0f080cdeaf net/tcp/sendfile: NET_TCP_WRITE_BUFFERS and NET_SENDFILE were inconsistent with each other:
tcp_sendfile() reads data directly from a file and does not use NET_TCP_WRITE_BUFFERS data flow
even if CONFIG_NET_TCP_WRITE_BUFFERS option is enabled.
Despite this, tcp_sendfile relied on NET_TCP_WRITE_BUFFERS specific flow control variables that
were idle during sendfile operation. Thus it was a total inconsistency.

E.g. because of the issue, TCP socket used by sendfile() operation never issued
FIN packet on close() command, and the TCP connection hung up.

As a result of the fix, simultaneously enabled CONFIG_NET_TCP_WRITE_BUFFERS and
CONFIG_NET_SENDFILE options can coexist.
2022-01-17 01:42:41 +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
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
Alexander Lunev
0afb1d8dbb net/tcp(unbuffered): fast retransmit on duplicate acknowledgments 2022-01-02 23:25:09 +08:00
chao.an
38b7b3d26a net/tcp: add support for CONFIG_NET_ALLOC_CONNS 2022-01-01 20:40:02 +08:00
Alexander Lunev
2b60468845 net/tcp(unbuffered): removed excessive overwrites of conn->sndseq
(conn->sndseq was updated in multiple places that was unreasonable and complicated).
2021-12-29 05:35:23 -06:00
Alexander Lunev
e68ffb9f99 net/tcp(unbuffered): fixed an issue with unackseq calculation.
Wrong unackseq calculation locked conn->tx_unacked at non-zero values
even if all ACKs were received. Thus unbuffered psock_tcp_send() never completed.
2021-12-27 20:59:48 -06:00
Alexander Lunev
19dc121a4f net/tcp(unbuffered): fixed an issue with tx_unacked overflow that occurred if NET_TCP_WINDOW_SCALE option was enabled.
If the remote TCP receiver advertised TCP window size greater than 64 KB
and TCP ACK packets returned to the NuttX TCP sender with a significant delay,
tx_unacked variable overflowed and further TCP send stalled forever
(until TCP re-connection).
2021-12-27 11:05:01 -06:00
chao.an
0ee7400fdf net/tcp: fix send deadlock if disconnect
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-12-16 01:29:10 -06:00
Alexander Lunev
1e07b6d528 net/tcp(unbuffered): retransmit only one the earliest not acknowledged segment
(according to RFC 6298 (5.4)).
2021-11-10 12:21:07 -06:00
YAMAMOTO Takashi
ecd6a3572b net/tcp/Kconfig: Remove NET_TCP_SPLIT
While it's a neat idea, it doesn't work well in reality.

* Many of modern tcp stacks don't obey the "ack every other packet"
  rule these days. (Linux, macOS, ...)

* Even if a traditional TCP implementation is assumed, we can't
  predict/control which packets are acked reliably. For example,
  window updates can easily mess up our strategy.
2021-11-04 13:32:57 -05:00
YAMAMOTO Takashi
28d168e1b8 tcp_send_unbuffered.c: Fix nxstyle errors 2021-11-04 13:32:57 -05:00
YAMAMOTO Takashi
1550a525e9 tcp_send_unbuffered.c: unifdef -UCONFIG_NET_TCP_SPLIT 2021-11-04 13:32:57 -05:00
Alexander Lunev
1e25602678 net/can,icmp,icmpv6,tcp,tcp_timer,udp: device should poll only those connections that are bound to the device.
tcp_timer: eliminated false decrements of conn->timer in case of multiple network adapters.
The false timer decrements sometimes provoked TCP spurious retransmissions due to premature timeouts.
2021-10-11 23:09:00 -07:00
chao.an
c132e5bed4 net/tcp: sanity check for the listen address
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-09-23 23:07:57 -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
Alin Jerpelea
91a5f90a7f author: UVC Ingenieure : update licenses to Apache
Gregory Nutt has submitted the SGA
UVC Ingenieure has submitted the SGA
Max Holtzberg has submitted the ICLA

as a result we can migrate the licenses to Apache.

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
2021-09-15 15:57:55 +08:00
YAMAMOTO Takashi
42f1851ca6 tcp_send_buffered.c: Fix snd_wnd
snd_wnd is an offset from the acked sequence number.
2021-08-25 20:56:05 +08:00
YAMAMOTO Takashi
1b82f1c749 tcp_input: snd_wnd processing
* Do not accept the window in old segments.
  Implement SND.WL1/WL2 things in the RFC.

* Do not accept the window in the segment w/o ACK bit set.
  The window is an offset from the ack seq.
  (maybe it's simpler to just drop segments w/o ACK though)

* Subtract snd_wnd by the amount of the ack advancement.
2021-08-25 20:56:05 +08:00
chao.an
9701a678bd net/tcp: add nonblock connect(2) support
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-08-19 19:19:05 -07:00
YAMAMOTO Takashi
b815a2c3a8 tcp_input: Don't put back sndseq on an old ack 2021-08-06 21:17:25 -07:00
YAMAMOTO Takashi
e53f989997 tcp_rexmit: advance conn->sndseq
Otherwise, we use an old sequence number when sending
non-data segment. (eg. window update)
The peer might consider such a segment stale and ignore.
2021-08-06 21:17:25 -07:00
YAMAMOTO Takashi
b5bb0d56ad tcp: fix an assertion in "fix iob allocation deadlock" commit
Fix a wrong assertion in:

```
commit 98ec46d726
Author: YAMAMOTO Takashi <yamamoto@midokura.com>
Date:   Tue Jul 20 09:10:43 2021 +0900

    tcp_send_buffered.c: fix iob allocation deadlock

    Ensure to put the wrb back onto the write_q when blocking
    on iob allocation. Otherwise, it can deadlock with other
    threads doing the same thing.
```

I forget to submit this with https://github.com/apache/incubator-nuttx/pull/4257
2021-08-05 23:50:06 -07:00
YAMAMOTO Takashi
1d3594ba07 tcp_send_buffered.c: Fix broken retransmit
With an applictation using mbedtls, I observed retransmitted segments
with corrupted user data, detected by the peer tls during mac processing.

Looking at the packet dump, I suspect that a wrb which has been put back
onto the write_q for retransmission was partially sent but fully acked.
Note: it's normal for a retransmission to be acked before sent.

In that case, the bug fixed in this commit would cause the wrb have
a wrong sequence number, possibly the same as the next wrb. It matches
what I saw in the packet dump. That is, the broken segments contain the
payload identical to one of the previous segment.
2021-08-02 11:30:01 -07:00
YAMAMOTO Takashi
98ec46d726 tcp_send_buffered.c: fix iob allocation deadlock
Ensure to put the wrb back onto the write_q when blocking
on iob allocation. Otherwise, it can deadlock with other
threads doing the same thing.
2021-08-02 11:25:55 -07:00
chao.an
7d4502aca6 net/socket: add SO_SNDBUF support
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-07-20 20:24:58 -07:00
chao.an
f52757f90a net/wrb: add tcp/udp_inqueue_wrb_size() interface
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-07-20 20:24:58 -07:00
YAMAMOTO Takashi
09f3a1ec8e tcp_send_buffered: throttle IOB allocations for send
Consider a bi-directional TCP connection:

1. we use all IOBs for tx queue
2. we advertize zero recv window because we have no free IOBs
3. if the peer tcp does the same thing,
   both sides advertize zero window and can not drain the tx queue.

For a similar stall to happen, the peer doesn't need to be
a naive tcp implementation like nuttx. A naive application blocking
on send() without draining its read buffer is enough.
(Probably such an application should be fixed to drain rx even
when tx is full. However, it's another story.)

This commit avoids the situation by prevent tx from grabbing
the all IOBs in the first place. (assuming CONFIG_IOB_THROTTLE > 0)
2021-07-14 15:08:18 +08:00
chao.an
83f7c08f65 net/tcp: only print the error when disable the TCP_NODELAY
Since we do not have the Nagle's algorithm,
the TCP_NODELAY socket option is enabled by default.

Change-Id: I0c8619bb06cf418f7eded5bd72ac512b349cacc5
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-07-13 09:44:19 -03:00
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
Xiang Xiao
906cb8b0f4 net/tcp: tcp_sendfile need restore the file location at the end
quote from https://man7.org/linux/man-pages/man2/sendfile.2.html:
If offset is not NULL, then it points to a variable holding the
file offset from which sendfile() will start reading data from
in_fd.  When sendfile() returns, this variable will be set to the
offset of the byte following the last byte that was read.  If
offset is not NULL, then sendfile() does not modify the file
offset of in_fd; otherwise the file offset is adjusted to reflect
the number of bytes read from in_fd.
If offset is NULL, then data will be read from in_fd starting at
the file offset, and the file offset will be updated by the call.

The change also align with the implementation at:
libs/libc/misc/lib_sendfile.c

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I607944f40b04f76731af7b205dcd319b0637fa04
2021-07-12 05:20:45 -07:00
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
chao.an
87bffc190c net/tcp: remove the invalid break during tcp option loop
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-07-07 03:55:41 -05:00
chao.an
b901f22c27 net/socket: add SO_RCVBUF support
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-07-06 01:44:55 -05:00
chao.an
eabe535de7 net/inet: add support of FIONREAD
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-07-05 06:20:52 -05:00
Anthony Merlino
a885b24cc1 Attempt to fix race condition reported in issue #3647 2021-07-04 08:54:15 -05:00
YAMAMOTO Takashi
669619a06a tcp_close: Fix a race with passive close
tcp_close disposes the connection immediately if it's called in
TCP_LAST_ACK. If it happens, we will end up with responding the
last ACK with a RST.

This commit fixes it by making tcp_close wait for the completion
of the passive close.
2021-07-02 13:54:15 +09:00
YAMAMOTO Takashi
c7ba75697c tcp_recvwindow.c: Use iob_tailroom to replace the home grown one 2021-06-30 06:40:13 -05:00
YAMAMOTO Takashi
52c237cb5f net/tcp/tcp.h: Update a comment about readahead 2021-06-30 06:40:13 -05:00
YAMAMOTO Takashi
08e9dff0e9 tcp_close: disable send callback before sending FIN
This fixes connection closing issues with CONFIG_NET_TCP_WRITE_BUFFERS.

Because TCP_CLOSE is used for both of input and output for tcp_callback,
the close callback and the send callback confuses each other as
the following. As it effectively disposes the connection immediately,
we end up with responding to the consequent ACK and FIN/ACK from the peer
with RSTs.

tcp_timer
    -> tcp_close_eventhandler
        returns TCP_CLOSE (meaning an active close)
    -> psock_send_eventhandler
        called with TCP_CLOSE from tcp_close_eventhandler, misinterpet as
        a passive close.
        -> tcp_lost_connection
            -> tcp_shutdown_monitor
                -> tcp_callback
                    -> tcp_close_eventhandler
                        misinterpret TCP_CLOSE from itself as
                        a passive close
2021-06-30 06:39:13 -05:00
YAMAMOTO Takashi
326a8ef0a2 tcp_close_disconnect: don't nullify sndcb
It isn't necessary and I plan to use the value later in
the close processing.
2021-06-30 06:39:13 -05:00
YAMAMOTO Takashi
8472430f22 tcp_close: replace scaring comments 2021-06-30 06:39:13 -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
98e7c6924d tcp: always responds to keep-alive segments
* It doesn't make sense to have this conditional on our own
  SO_KEEPALIVE support. (CONFIG_NET_TCP_KEEPALIVE)
  Actually we don't have a control on the peer tcp stack,
  who decides to send us keep-alive probes.

* We should respond them for non ESTABLISHED states. eg. FIN_WAIT_2
  See also:
  https://github.com/apache/incubator-nuttx/pull/3919#issuecomment-868248576
2021-06-30 11:52:08 +09:00
YAMAMOTO Takashi
4878b7729c tcp: simplify readahead
Do not bother to preserve segment boundaries in the tcp
readahead queues.

* Avoid wasting the tail IOB space for each segments.
  Instead, pack the newly received data into the tail space
  of the last IOB. Also, advertise the tail space as
  a part of the window.

* Use IOB chain directly. Eliminate IOB queue overhead.

* Allow to accept only a part of a segment.

* This change improves the memory efficiency.
  And probably more importantly, allows less-confusing
  recv window advertisement behavior.
  Previously, even when we advertise N bytes window,
  we often couldn't actually accept N bytes. Depending on
  the segment sizes and IOB configurations, it was causing
  segment drops.
  Also, the previous code was moving the right edge of the
  window back and forth too often, even when nothing in
  the system was competing on the IOBs. Shrinking the
  window that way is a kinda well known recipe to confuse
  the peer stack.
2021-06-30 06:22:14 +09:00
YAMAMOTO Takashi
0886257eb4 tcp_input: Accept segments spanning over rcvseq 2021-06-30 06:22:14 +09:00
YAMAMOTO Takashi
eeafe070ec tcp.h: Add TCP_SEQ_ADD 2021-06-30 06:22:14 +09:00
YAMAMOTO Takashi
022a2490d1 tcp: Change the way to advance rcvseq
* Move the code to advance rcvseq for user data from tcp_input
  to receive handlers.
  Motivation: allow partial ack.

* If we drop a segment, ignore FIN as well. Note than tcp FIN bit is
  logically after the user data in the same segment.
2021-06-30 06:22:14 +09:00
YAMAMOTO Takashi
1448de4451 tcp_should_send_recvwindow: Remove function name from ninfo()
ninfo() itself usually prefixes the function name
automatically for us.
2021-06-18 00:47:47 -05:00
YAMAMOTO Takashi
ca0f2bdb95 tcp_get_recvwindow: Make this match the reality (tcp_datahandler)
It doesn't make much sense to advertize more than we can actually
accept, especially when we don't queue partial segments.
2021-06-15 01:17:38 -05:00
YAMAMOTO Takashi
af64912833 tcp_get_recvwindow: use tcp_rx_mss
Just to reduce code duplication.
No functional changes are intended.
2021-06-15 01:17:38 -05:00
YAMAMOTO Takashi
64676641cb tcp_datahandler: try throttled=false on iob_trycopyin failure as well
I assume this was just an oversight because I couldn't
find any obvious reason to special-case only the first IOB.

The commit message of the original commit is cited below.

```
commit bf21056001
Author: chao.an <anchao@xiaomi.com>
Date:   Fri Nov 27 09:50:38 2020 +0800

    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>
```
2021-06-15 01:17:38 -05:00
YAMAMOTO Takashi
0347cd3f09 tcp_should_send_recvwindow: Add a few ninfo() 2021-06-13 21:20:24 -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
7d82e7a7c4 tcp_input: fix a confusing variable name and a comment
It looks like a copy-and-paste mistake.
2021-06-10 22:47:04 -05:00
YAMAMOTO Takashi
eb00e00e48 tcp: Use the tcp seq macros in some obvious places 2021-06-10 22:47:04 -05:00
YAMAMOTO Takashi
433a2b27d9 tcp: add macros to deal with sequence number wraparound 2021-06-10 22:47:04 -05:00
Xiang Xiao
5b2a17b892 Include assert.h in necessary place
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-06-08 13:06:08 -07:00
YAMAMOTO Takashi
69b3f034a4 tcp: Move buffered/unbuffered common code to tcp_send.c 2021-06-03 21:33:10 -05:00
YAMAMOTO Takashi
7d33c01a0a tcp_get_recvwindow: Add a revisit comment 2021-06-03 21:32:34 -05:00
YAMAMOTO Takashi
7ac6c0a8de tcp_data_event: Add a comment 2021-05-31 01:37:51 -05:00
YAMAMOTO Takashi
92328792fd tcp_data_event: Fix an indent 2021-05-31 01:37:51 -05:00
YAMAMOTO Takashi
2ce0457edb tcp_get_recvwindow: Add a comment 2021-05-31 01:37:51 -05:00
YAMAMOTO Takashi
0c606ecb8e psock_tcp_recvfrom: Add a comment about window updates 2021-05-31 01:37:51 -05:00
Alin Jerpelea
b3ad98c89a net: update licenses to Apache
Gregory Nutt is the copyright holder for those files and he has submitted the
SGA as a result we can migrate the licenses to Apache.

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
2021-05-27 08:07:25 +09:00
Xiang Xiao
001e7c3e76 sched: Don't include nuttx/sched.h inside sched.h
But let nuttx/sched.h include sched.h instead to
avoid expose nuttx kernel API to userspace.

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-05-24 12:11:53 +09:00
chao.an
48b0e48cd4 net/tcp: set/get TCP_KEEPINTVL/IDLE value as BSD style
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-05-22 09:01:18 -05:00
chao.an
a876f0253a net/tcp: recounter the ack counter during obtain newdata
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-05-21 18:02:53 -03:00
YAMAMOTO Takashi
acc3596adc tcp_netpoll.c: Fix a performance issue with CONFIG_NET_TCP_WRITE_BUFFERS
Tested with a modified version of webclient, which uses non-blocking i/o.
The packet dumps look more reasonable with this change.
2021-04-05 06:16:46 -05:00
chao.an
621242e890 net/tcp: support bind the same port with different domain
Reference here:
https://man7.org/linux/man-pages/man7/ipv6.7.html

IPV6_V6ONLY slice

Signed-off-by: chao.an <anchao@xiaomi.com>
2021-04-01 20:05:14 -04:00
YAMAMOTO Takashi
09869e5d41 net/tcp/tcp.h: Remove unused extern g_netdevices 2021-03-30 12:27:50 -05:00
YAMAMOTO Takashi
1c29a2e8e8 net/tcp/tcp_send_buffered.c: Fix non-blocking I/O
My recent changes to buffered tcp send broke this. [1]

One of my local apps using non-blocking tcp is working
again with this fix.

[1]
```
commit 837e1a72a4
Author: YAMAMOTO Takashi <yamamoto@midokura.com>
Date:   Mon Mar 15 16:19:42 2021 +0900

    tcp_send_buffered.c: improve tcp write buffering
```
2021-03-30 01:12:55 -05:00
YAMAMOTO Takashi
271e748ba5 tcp_send_buffered.c: Add a bit more info to an ninfo() 2021-03-30 01:12:55 -05:00
YAMAMOTO Takashi
a2840b6354 tcp_send_buffered.c: Add an assertion 2021-03-30 01:12:55 -05:00
YAMAMOTO Takashi
ef9adcf399 tcp_send_buffered.c: Remove dead code 2021-03-30 01:12:55 -05:00
YAMAMOTO Takashi
837e1a72a4 tcp_send_buffered.c: improve tcp write buffering
* Send data chunk-by-chunk
  Note: A stream socket doesn't have atomicity requirement.

* Increase the chance to use full-sized segments

Benchmark numbers in my environment:

* Over ESP32 wifi
* The peer is NetBSD, which has traditional delayed ack TCP
* iperf uses 16384 bytes buffer

---

without this patch,
CONFIG_IOB_NBUFFERS=36
CONFIG_IOB_BUFSIZE=196

does not work.
see https://github.com/apache/incubator-nuttx/pull/2772#discussion_r592820639

---

without this patch,
CONFIG_IOB_NBUFFERS=128
CONFIG_IOB_BUFSIZE=196
```
nsh> iperf -c 192.168.8.1
       IP: 192.168.8.103

 mode=tcp-client sip=192.168.8.103:5001,dip=192.168.8.1:5001, interval=3, time=30

        Interval Bandwidth

   0-   3 sec,  4.11 Mbits/sec
   3-   6 sec,  4.63 Mbits/sec
   6-   9 sec,  4.89 Mbits/sec
   9-  12 sec,  4.63 Mbits/sec
  12-  15 sec,  4.85 Mbits/sec
  15-  18 sec,  4.85 Mbits/sec
  18-  21 sec,  5.02 Mbits/sec
  21-  24 sec,  3.67 Mbits/sec
  24-  27 sec,  4.94 Mbits/sec
  27-  30 sec,  4.81 Mbits/sec
   0-  30 sec,  4.64 Mbits/sec
nsh>
```

---

with this patch,
CONFIG_IOB_NBUFFERS=36
CONFIG_IOB_BUFSIZE=196
```
nsh> iperf -c 192.168.8.1
       IP: 192.168.8.103

 mode=tcp-client sip=192.168.8.103:5001,dip=192.168.8.1:5001, interval=3, time=30

        Interval Bandwidth

   0-   3 sec,  5.33 Mbits/sec
   3-   6 sec,  5.59 Mbits/sec
   6-   9 sec,  5.55 Mbits/sec
   9-  12 sec,  5.59 Mbits/sec
  12-  15 sec,  5.59 Mbits/sec
  15-  18 sec,  5.72 Mbits/sec
  18-  21 sec,  5.68 Mbits/sec
  21-  24 sec,  5.29 Mbits/sec
  24-  27 sec,  4.67 Mbits/sec
  27-  30 sec,  4.50 Mbits/sec
   0-  30 sec,  5.35 Mbits/sec
nsh>
```

---

with this patch,
CONFIG_IOB_NBUFFERS=128
CONFIG_IOB_BUFSIZE=196
```
nsh> iperf -c 192.168.8.1
       IP: 192.168.8.103

 mode=tcp-client sip=192.168.8.103:5001,dip=192.168.8.1:5001, interval=3, time=30

        Interval Bandwidth

   0-   3 sec,  5.51 Mbits/sec
   3-   6 sec,  4.67 Mbits/sec
   6-   9 sec,  4.54 Mbits/sec
   9-  12 sec,  5.42 Mbits/sec
  12-  15 sec,  5.37 Mbits/sec
  15-  18 sec,  5.11 Mbits/sec
  18-  21 sec,  5.07 Mbits/sec
  21-  24 sec,  5.29 Mbits/sec
  24-  27 sec,  5.77 Mbits/sec
  27-  30 sec,  4.63 Mbits/sec
   0-  30 sec,  5.14 Mbits/sec
nsh>
```
2021-03-22 01:12:59 -07:00
chao.an
e03218ab71 net/tcp: reset the connection ref count before tcp_free()
reset the connection refcount if SYN retry count has elapsed

Assertion:

up_assert: Assertion failed at file:tcp/tcp_conn.c line: 764 task: netdev_wq

N/A

Signed-off-by: chao.an <anchao@xiaomi.com>
2021-03-22 10:55:30 +09:00
chao.an
a5613e6008 net/tcp: correct the port byte order
1. unify the byte order to network
2. Do not re-select the port if the local port has been bind()

Signed-off-by: chao.an <anchao@xiaomi.com>
2021-03-20 09:13:18 -07:00
YAMAMOTO Takashi
45098769e7 tcp_sendfile.c: Remove an unused copy of CONFIG_NET_TCP_SPLIT_SIZE 2021-03-15 04:52:58 -07:00
Masayuki Ishikawa
e65d5d8c73 Revert "net: tcp: Add DEBUGASSERT() in psock_tcp_send()"
Summary:
- Based on the discussion (PR#2772), let me revert the commit

Impact:
- None

Testing:
- N/A

This reverts commit ec8bf5c8c1.

Suggested-by: YAMAMOTO Takashi <yamamoto@midokura.com>
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
2021-03-12 16:23:01 +08:00
Gustavo Henrique Nihei
330eff36d7 sourcefiles: Fix relative path in file header 2021-03-09 23:18:28 +08:00
Jiuzhu Dong
4d5a964f29 net: unify socket into file descriptor
Change-Id: I9bcd21564e6c97d3edbb38aed1748c114160ea36
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2021-03-03 19:01:41 -08:00
Alin Jerpelea
37d5c1b0d9 net: Author Gregory Nutt: update licenses to Apache
Gregory Nutt has submitted the SGA and we can migrate the licenses
 to Apache.

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
2021-02-20 00:38:18 -08:00
Alin Jerpelea
7dc7ef45f1 net: nxstyle fixes
Nxstyle fixes to pass CI.

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
2021-02-20 00:38:18 -08:00
Masayuki Ishikawa
ec8bf5c8c1 net: tcp: Add DEBUGASSERT() in psock_tcp_send()
Summary:
- This commit adds DEBUGASSERT() to check the IOB size

Impact:
- None

Testing:
- Tested with sabre-6quad:netnsh with QEMU

Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
2021-01-29 00:05:01 -08:00
Abdelatif Guettouche
7e3d4a5f29 net: Remove duplicate forward references.
Signed-off-by: Abdelatif Guettouche <abdelatif.guettouche@espressif.com>
2021-01-16 07:39:04 -08:00
Abdelatif Guettouche
8e4397968c net/ & esp32/wlan: Fix some typos and nxstyle issues.
Signed-off-by: Abdelatif Guettouche <abdelatif.guettouche@espressif.com>
2021-01-14 07:57:27 -06:00
chao.an
ae5b727580 net/tcp/handshake: send the SYN immediately.
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-12-18 14:16:11 +09:00
chao.an
5e9e50991c net/tcp: send the ack on nonblock mode
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-12-18 14:16:11 +09:00
Juha Niskanen
de1ad1fdb3 net: fix typos, incorrect comments, nxstyle
Signed-off-by: Juha Niskanen <juha.niskanen@haltian.com>
2020-12-13 09:06:28 -06:00
chao.an
8d0118569c [Performance]net/tcp: send the ACK in time after obtain ahead buffer from iobs
Request the TCP ACK to estimate the receive window after handle
any data already buffered in a read-ahead buffer.

Change-Id: Id998a1125dd2991d73ba4bef081ddcb7adea4f0d
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-12-10 12:23:47 +09:00
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
794a6ec23d net/tcp: rename the winszie to snd_wnd to make the semantics more accurate
Change-Id: I8fdc7cf78a7f2cd53a30ef1de702b1a697c43238
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-12-10 12:23:47 +09:00
Gregory Nutt
45699e2701 net/tcp/tcp_input.c: Correct bad check of urgent data length
Urgent data preceded "normal" data in the TCP payload.  If the urgent data is larger than the size of the TCP payload, this indicates that the entire payload is urgent data and that urgent data continues in the next packet.

This case was handled correctly for the case where urgent data was present but was not being handled correctly in the case where the urgent data was NOT present.
2020-12-09 07:25:14 -08:00
YAMAMOTO Takashi
cee4261c0d net/tcp/tcp_sendfile.c: Fix syslog formats 2020-12-05 08:13:32 -06:00
chao.an
c2b0006dcd net/tcp: implement the fast retransmit
RFC2001: TCP Slow Start, Congestion Avoidance, Fast Retransmit,
         and Fast Recovery Algorithms

...

3.  Fast Retransmit
  Modifications to the congestion avoidance algorithm were proposed in
  1990 [3].  Before describing the change, realize that TCP may
  generate an immediate acknowledgment (a duplicate ACK) when an out-
  of-order segment is received (Section 4.2.2.21 of [1], with a note
  that one reason for doing so was for the experimental fast-
  retransmit algorithm).  This duplicate ACK should not be delayed.
  The purpose of this duplicate ACK is to let the other end know that a
  segment was received out of order, and to tell it what sequence
  number is expected.

  Since TCP does not know whether a duplicate ACK is caused by a lost
  segment or just a reordering of segments, it waits for a small number
  of duplicate ACKs to be received.  It is assumed that if there is
  just a reordering of the segments, there will be only one or two
  duplicate ACKs before the reordered segment is processed, which will
  then generate a new ACK.  If three or more duplicate ACKs are
  received in a row, it is a strong indication that a segment has been
  lost.  TCP then performs a retransmission of what appears to be the
  missing segment, without waiting for a retransmission timer to
  expire.

Change-Id: Ie2cbcecab507c3d831f74390a6a85e0c5c8e0652
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-12-01 11:36:10 -06:00
chao.an
234d3df19f net/tcp: correct the available iobs count if throttle is enabled
The number of available iobs is already sub in iob_navail(true) on line 114

net/tcp/tcp_recvwindow.c:
...
 73 uint16_t tcp_get_recvwindow(FAR struct net_driver_s *dev)
...
114   niob_avail    = iob_navail(true);

Change-Id: I230927904d8db08ed8d95d7fa18c5c5fce08aa5e
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-11-28 23:24:44 -06:00
chao.an
4305718691 net/tcp: fix nxstyle warnings
Change-Id: Id885fefc80caa12806c73778e632c8b0c5888be0
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-11-28 00:03:47 -06:00
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
YAMAMOTO Takashi
cc9b3a3ec5 net/tcp/tcp_appsend.c: Fix syslog formats
Note: tx_unacked is either uint16_t or uint32_t depending on the config.
2020-11-27 05:18:57 -06:00
YAMAMOTO Takashi
cbdbb369c0 net/tcp/tcp_input.c: Fix syslog formats
Note: tx_unacked is either uint16_t or uint32_t depending on the config.
2020-11-27 05:18:57 -06:00
YAMAMOTO Takashi
af6719965f net/tcp/tcp_appsend.c: Fix syslog formats 2020-11-25 05:11:26 -08:00
YAMAMOTO Takashi
390f766b36 net/tcp/tcp_appsend.c: Appease nxstyle 2020-11-25 05:11:26 -08:00
YAMAMOTO Takashi
40d4c03a8c net/tcp/tcp_input.c: Fix syslog formats 2020-11-25 05:11:26 -08:00
YAMAMOTO Takashi
17654f592e net/tcp/tcp_send_buffered.c: Fix syslog formats 2020-11-25 05:11:26 -08:00
YAMAMOTO Takashi
634713cec4 net/tcp/tcp_sendfile.c: Fix syslog formats 2020-11-24 22:31:33 -08:00
YAMAMOTO Takashi
ee1c914277 net/tcp/tcp_input.c: Fix syslog formats 2020-11-23 05:00:10 -08:00
YAMAMOTO Takashi
28fda4e5bb net/tcp/tcp_send_unbuffered.c: Fix syslog formats 2020-11-23 05:00:10 -08:00
YAMAMOTO Takashi
8cb6790cef net/tcp/tcp_send_unbuffered.c: Fix a syslog format 2020-11-20 22:22:53 -08:00
YAMAMOTO Takashi
8a23f98233 net/tcp/tcp_wrbuffer_dump.c: Fix a syslog format 2020-11-20 22:22:53 -08:00
YAMAMOTO Takashi
ba1b6876b4 net/tcp/tcp_send_buffered.c: Fix a syslog format error 2020-11-20 22:22:53 -08:00
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
chao.an
95be9eaf4b net/callback: remove the assert check of conn instance
remove the connection assertion since the instance will be invalid
if the network device has been taken down.

net/netdev/netdev_ioctl.c:

1847 void netdev_ifdown(FAR struct net_driver_s *dev)
1848 {
...
1871       /* Notify clients that the network has been taken down */
1872
1873       devif_dev_event(dev, NULL, NETDEV_DOWN);
...
1883 }

Change-Id: I492b97b5ebe035ea67bbdd7ed635cb13d085e89c
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-11-10 19:40:41 -08:00
chao.an
a22f5bb8b0 net/tcp: fix nxstyle warning
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-10-28 11:41:16 -07:00
chao.an
d78bf36740 net/tcp: fix tcp socket close timeout if loss wireless connection
In the current net stack implementation, there is no mechanism
for notifying the loss of the wireless connection, if the network
is disconnected then application sends data packets through tcp,
the tcp_timer will keep retrying fetch the ack for awhile, the
connection status will not be able to be switched timely.

Change-Id: I84d1121527edafc6ee6ad56ba164838694e7e11c
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-10-28 11:41:16 -07:00
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
Nathan Hartman
80ce7800a9 Sources and Docs: Fix typos and nxstyle issues
Documentation/contributing/coding_style.rst:

    * Fix repeated words: ("this this").
    * Remove trailing spaces.

boards/z80/z80/z80sim/README.txt:

    * Fix repeated words: ("this this") and rewrap lines.

graphics/Kconfig,
libs/libc/math/Kconfig:

    * Fix repeated words: ("this this").

arch/arm/src/armv7-a/arm_assert.c,
arch/arm/src/armv7-r/arm_assert.c,
arch/arm/src/imxrt/imxrt_enet.c,
arch/arm/src/kinetis/kinetis_enet.c,
arch/arm/src/kinetis/kinetis_flexcan.c,
arch/arm/src/s32k1xx/s32k1xx_enet.c,
arch/arm/src/s32k1xx/s32k1xx_flexcan.c,
arch/arm/src/stm32/stm32_pwm.c,
arch/arm/src/stm32h7/stm32_pwm.c,
arch/arm/src/stm32l4/stm32l4_pwm.c,
arch/renesas/src/rx65n/rx65n_usbdev.c,
binfmt/libnxflat/libnxflat_bind.c,
drivers/pipes/pipe_common.c,
net/igmp/igmp_input.c,
net/tcp/tcp_conn.c,
sched/sched/sched_roundrobin.c:

    * Fix typo in comment ("this this").

arch/arm/src/cxd56xx/cxd56_usbdev.c,
arch/arm/src/lc823450/lc823450_usbdev.c:

    * Fix typo in comment and rewrap lines.

arch/arm/src/imxrt/imxrt_usbdev.c,
arch/arm/src/stm32/stm32_dac.c,
arch/arm/src/stm32f0l0g0/stm32_pwm.c,
arch/arm/src/stm32f7/stm32_pwm.c,
arch/arm/src/tiva/lm/lm4f_gpio.h,
fs/nxffs/nxffs_write.c,
include/nuttx/analog/pga11x.h,
include/nuttx/usb/usbdev.h,
net/mld/mld_join.c:

    * Fix typo in comment ("this this").
    * Fix nxstyle issues.
2020-10-02 04:54:52 +02:00
YAMAMOTO Takashi
4f7745ae1c net/tcp/tcp_timer.c: Remove redundant assignments
Found by clang-check:

tcp/tcp_timer.c:185:15: warning: Value stored to 'result' is never read
              result = tcp_callback(dev, conn, TCP_TIMEDOUT);
              ^        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
tcp/tcp_timer.c:264:23: warning: Value stored to 'result' is never read
                      result = tcp_callback(dev, listener, TCP_TIMEDOUT);
                      ^        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
tcp/tcp_timer.c:300:19: warning: Value stored to 'result' is never read
                  result = tcp_callback(dev, conn, TCP_TIMEDOUT);
                  ^        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3 warnings generated.
2020-08-03 22:52:31 -05:00
Masayuki Ishikawa
c7500c5c9d net: tcp: Fix port info in warning message in tcp_input.c
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
2020-05-19 13:36:45 +08:00
Gregory Nutt
57bc329aac Run nxstyle all .c and .h files modified by PR. 2020-05-17 14:01:00 -03:00
Gregory Nutt
a569006fd8 sched/: Make more naming consistent
Rename various functions per the quidelines of https://cwiki.apache.org/confluence/display/NUTTX/Naming+of+OS+Internal+Functions

    nxsem_setprotocol -> nxsem_set_protocol
    nxsem_getprotocol -> nxsem_get_protocol
    nxsem_getvalue -> nxsem_get_value
2020-05-17 14:01:00 -03:00
chao.an
7a62a59dec net/conn: generate port base dynamically
In some extreme scenarios(eg. crash, reboot, reset, etc...),
an established connection cannot guarantee that the port can be
closed properly, if we try to reconnect quickly after reset, the
connection will fail since the current port is same as the
previous one, the previous port connection has been hold on server side.

dynamically apply for the port base to avoid duplication.

Change-Id: I0089244b2707ea61f553a4dae09c7af3649c70bd
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-05-15 07:27:01 -06:00
chao.an
a5868864bc net: remove unnecessary spaces
minor style fix

Change-Id: I06b6fe48628440490090b89a39baf01481f79b79
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-05-14 07:54:57 -06:00
Xiang Xiao
63e3054ced Don't need monitor IOB buffer empty event for POLLOUT implementation
It's enough to check the buffer available in the net event handler

Change-Id: I2d7c7a03675cf6eff6ffb42a81b7c7245253e92c
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-13 06:50:07 -06:00
Xiang Xiao
517974787f Rename clock_systime[r|spec] to clock_systime_[ticks|timespec]
follow up the new naming convention:
https://cwiki.apache.org/confluence/display/NUTTX/Naming+of+OS+Internal+Functions
2020-05-10 14:35:50 -06:00
chao.an
0887203c57 tcp/conn: request arp before the 3-way handshake
Since the request address was not properly resolved before the handshake,
every time of connection, the handshake data will be overwitten into
arp packet and retransmitted until the next tcp timer.
Request the arp address before the handshake to avoid the retransmission.

Change-Id: I80118b9a8096c126c8e16cdf2f7b3d98fca92437
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-05-09 07:25:03 -06:00
Xiang Xiao
b7d922960f Fix nxstyle issue
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-08 07:20:49 -06:00
GAEHWILER Reto
2ba2846ced net/tcp/tcp_input.c: Fix of faulty TCP sequence increments
Sequence numbers must not be incremented in the case of retransmitted packets
2020-03-03 09:15:18 -06:00
patacongo
861efdf8a3
net/: Whenever the network initializes an IPv4 address, it must clear sin_zero. 2020-02-25 15:53:39 +01:00
Alan Carvalho de Assis
d74ca3ab82 Make the tcp_backlog error message clear
The tcp_backlog is used when there is not process running to accept a new connection, but it is limited by the number of allocated backlog containers. The current error message induces the user to believe there is not free memory to allocate a new container, but actually the containers were allocated during the initialization and were available until the last element of the list has been removed to use.
2020-02-24 11:35:01 -06:00
Xiang Xiao
cde88cabcc Run codespell -w with the latest dictonary again
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-02-23 22:27:46 +01:00
Xiang Xiao
bd4e8e19d3 Run codespell -w against all files
and fix the wrong correction
2020-02-22 14:45:07 -06:00
chao.an
c65d8e6a23 net/socket: add MSG_DONTWAIT support
MSG_DONTWAIT (since Linux 2.2)
  Enables nonblocking operation; if the operation would block, the
  call fails with the error EAGAIN or EWOULDBLOCK. This provides
  similar behavior to setting the O_NONBLOCK flag (via the fcntl(2)
  F_SETFL operation), but differs in that MSG_DONTWAIT is a per-call
  option, whereas O_NONBLOCK is a setting on the open file description
  (see open(2)), which will affect all threads in the calling process
  and as well as other processes that hold file descriptors referring
  to the same open file description.
2020-02-19 12:21:28 -06:00
Juha Niskanen
15b78abccf Fix typos in comments 2020-02-14 08:50:45 -06:00
Xiang Xiao
5c80b94820 Replace #include <semaphore.h> to #include <nuttx/semaphore.h>
Since the kernel side should call nxsem_xxx instead and remove the unused inclusion
2020-02-01 08:27:30 -06:00
Greg Nutt
6b413ec241 net/tcp: Fix errors found in build testing.
Author: Gregory Nutt <gnutt@nuttx.org>

    net/tcp: Fix errors found in build testing.

    Recent re-organization moved some functions from net/inet to net/tcp and net/udp.  This include references to nxsem_wait(), SEM_PRIO_NONE, and other internal NuttX semaphore functions.  These all failed to compile because nuttx/semaphore.h was not included in any of the files.
2020-01-22 12:29:26 -03:00
Gregory Nutt
6d4b86ff06 net/tcp/tcp.h: Correct spacing error introduced with the last PR. 2020-01-21 08:30:39 -06:00
Xiang Xiao
e75b5e9d86 net/tcp and udp: Move tcp/udp recvfrom into tcp/udp folder
Move tcp/udp recvfrom into tcp/udp folder and remove inet_recvfrom.c
2020-01-21 08:30:39 -06:00
Xiang Xiao
e869a10c18 net/tcp, udp: Move tcp/udp close operation into tcp/udp folder
Move tcp/udp close operation into tcp/udp folder and remove inet_close.c
2020-01-21 08:30:39 -06:00
chao.an
a4aa8ae491 wqueue/notifier: update the work notifier usage
usage changed after commit 90c52e6f8f

Change-Id: Ifb0d739b046a6794b5b3ac177f489fb9a1c5c799
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-01-21 14:16:19 +08:00
Gregory Nutt
bb5ad4df23 net/tcp/tcp_getsockopt.c: Fix compilation failure found in build testing. Error: invalia application of 'sizeof' to incomplete type 'struct timeval'. Fixed by including sys/time.h. Same problem and same fix as for setsockopt earlier." 2020-01-11 17:20:02 -03:00
Gregory Nutt
3374772580 net/tcp/tcp_setsockopt.c: Fix compilation failure found in build testing. Error: invalia application of 'sizeof' to incomplete type 'struct timeval'. Fixed by including sys/time.h. 2020-01-11 16:29:25 -03:00
Gregory Nutt
45a83e9481 net/tcp/tcp_send_unbuffered.c: Correct a bad signal number
There is no errno ETIMEOUT.  Probably should be ETIMEDOUT.
2020-01-11 15:36:51 -03:00
Gregory Nutt
98f431d8d8 net/: Run all .c files under net/ through tools/nxstyle and fix all resulting complaints. 2020-01-11 13:41:33 -03:00
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
346336bb9e Make the read ahead buffer unselectable
Here is the email loop talk about why it is better to remove the option:
https://groups.google.com/forum/#!topic/nuttx/AaNkS7oU6R0

Change-Id: Ib66c037752149ad4b2787ef447f966c77aa12aad
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-01-11 08:24:49 -06:00
Xiang Xiao
6a3c2aded6 Fix wait loop and void cast (#24)
* Simplify EINTR/ECANCEL error handling

1. Add semaphore uninterruptible wait function
2 .Replace semaphore wait loop with a single uninterruptible wait
3. Replace all sem_xxx to nxsem_xxx

* Unify the void cast usage

1. Remove void cast for function because many place ignore the returned value witout cast
2. Replace void cast for variable with UNUSED macro
2020-01-02 10:54:43 -06:00
Masayuki Ishikawa
6e7c761fc8 net: tcp: Fix compile error in tcp.h 2020-01-02 09:36:25 -06:00
Xiang Xiao
02b652ce1c Minor fix (#23)
* fix ieee802154/ieee802154_input.c:179:7: error: too few arguments to function 'iob_free'
       iob_free(container->ic_iob);
       ^~~~~~~~

ieee802154/ieee802154_input.c:180:7: error: too many arguments to function 'ieee802154_container_free'
       ieee802154_container_free(container, IOBUSER_NET_SOCK_IEEE802154);
       ^~~~~~~~~~~~~~~~~~~~~~~~~

* fix udp/udp_netpoll.c:327:10: warning: 'ret' may be used uninitialized in this function [-Wmaybe-uninitialized]

* fix local/local_netpoll.c:154:15: warning: implicit declaration of function 'nxsem_post'; did you mean 'sem_post'? [-Wimplicit-function-declaration]
               nxsem_post(fds->sem);
               ^~~~~~~~~~
               sem_post
2020-01-02 09:35:50 -06:00
Xiang Xiao
90c52e6f8f Squashed commit of the following:
Author: Gregory Nutt <gnutt@nuttx.org>

    Run all .h and .c files modified in last PR through nxstyle.

Author: Xiang Xiao <xiaoxiang@xiaomi.com>

    Net cleanup (#17)

    * Fix the semaphore usage issue found in tcp/udp

    1. The count semaphore need disable priority inheritance
    2. Loop again if net_lockedwait return -EINTR
    3. Call nxsem_trywait to avoid the race condition
    4. Call nxsem_post instead of sem_post

    * Put the work notifier into free list to avoid the heap fragment in the long run.  Since the allocation strategy is encapsulated internally, we can even refine the implementation later.

    * Network stack shouldn't allocate memory in the poll implementation to avoid the heap fragment in the long run, other modification include:

    1. Select MM_IOB automatically since ICMP[v6] socket can't work without the read ahead buffer
    2. Remove the net lock since xxx_callback_free already do the same thing
    3. TCP/UDP poll should work even the read ahead buffer isn't enabled at all

    * Add NET_ prefix for UDP_NOTIFIER and TCP_NOTIFIER option to align with other UDP/TCP option convention

    * Remove the unused _SF_[IDLE|ACCEPT|SEND|RECV|MASK] flags since there are code to set/clear these flags, but nobody check them.
2019-12-31 09:26:14 -06:00
chao.an
8dd4b27973 tcp_conn: Check if the remote address is unspecified
Change-Id: I6f547bb4bfb3bb621573db9097a531ce2260e794
Signed-off-by: chao.an <anchao@xiaomi.com>
2019-12-24 10:48:00 -06:00
Gregory Nutt
e5e52446ca net/tcp/tcp_send.c: Trival update to some comments. 2019-12-11 11:19:32 -06:00
Gregory Nutt
c2642eef3c net/tcp/tcp_appsend.c: Enhance delayed ACK algorithm with behavior of Windows delayed ACK described at https://support.microsoft.com/en-nz/help/214397/design-issues-sending-small-data-segments-over-tcp-with-winsock: 'If there are data to be sent in the same direction as the ACK before the second data packet is received and the delay timer expires, the ACK is piggybacked with the data segment and sent immediately.' 2019-12-09 08:45:52 -06:00
Gregory Nutt
6a0ab27bd5 net/tcp/tcp_appsend.c: Update some comments. 2019-12-09 06:06:17 -06:00
Gregory Nutt
8cd5e04d53 net/tcp: Minor design improvement to TCP delayed ACKs. 2019-12-08 14:17:55 -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
Xiang Xiao
1905e01fda net/: ICMP/ICMPv6/TCP/UDP poll shouldn't set POLLHUP and POLLOUT at the same time the standard require that only report POLLHUP:
https://pubs.opengroup.org/onlinepubs/009695399/functions/poll.html.
2019-11-25 09:59:50 -06:00
Xiang Xiao
af9c67ab58 net/tcp/tcp_netpoll.c and net/udp/udp_netpoll.c: In [tcp|udp]_iob_work fix the wrong condition logic (& vs &&). 2019-11-25 09:32:24 -06:00
Xiang Xiao
98fc60eb75 [tcp|udp]_poll_eventhandler check psock_[tcp|udp]_cansend before report POLLOUT. Change the unbuffered psock_[tcp|udp]_cansend return OK to unify the code logic and remove the unnecessary [tcp|udp]_poll_txnotify call. 2019-11-24 10:30:48 -06:00
Xiang Xiao
673f812c4e net/tcp/tcp_netpoll.c: Monitor TCP_POLL/TCP_NEWDATA/TCP_BACKLOG per the request like UDP counterpart. 2019-11-24 10:20:01 -06:00
Gregory Nutt
9efadaefc1 net/tcp: Be consistent with units of TIME_WAIT. Units were unspecified in tcp/Kconfig, but assumed to be in units of half seconds in tcp/timer.h. include/nuttx/netconfig does not indicate the units but is apparently assuming seconds. This commit unifies all delays to clearly specified units of seconds. 2019-11-24 09:19:54 -06:00
Gregory Nutt
3b275fcf4e net/: Run nxstyle against all C files. 2019-10-25 11:31:42 -06:00
Xiang Xiao
5fd8f78bf9 net/ipforward, tcp, and udp: Fix a chicken and egg problem by eliminating the check of the arp/neighbor tables before packet transmission
1. For buffered tcp/udp case, if CONFIG_NET_ARP_SEND/CONFIG_NET_ARP_IPIN/CONFIG_NET_ICMPv6_NEIGHBOR isn't enabled and the table doesn't contain ip<->ethaddr mapping yet, the logic will skip the realtransmission and then arp/neighbor can't steal the final buffer to generate arp/icmpv6 packet.
2.for all other case, the tcp layer or user program should already contain the retransmit logic, the check is redundancy and may generate many duplicated packets if arp/icmpv6 response is too slow because the cursor stop forward. If user still concern about the very first packet lost, he could fix the issue by enabling CONFIG_NET_ARP_SEND/CONFIG_NET_ICMPv6_NEIGHBOR at begin.
2019-09-18 12:33:41 -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
6266e067e9 net/: Re-order the content of all address-family socket 'connection' structures so that they begin with a comomon prologue. This permits better use of logic for different address family types. 2019-09-01 08:47:01 -06:00
Gregory Nutt
a52ceac13e net/: Now handles reception of IPv4 packets with larger IPv4 headers containing options. 2019-08-31 12:25:30 -06:00
zhangyuan7
b33fc302f0 net/inet and net/tcp: Fix tcp close flow; free the connection after all tcp close process finished. 2019-08-30 06:44:43 -06:00
Gregory Nutt
ac151e05e7 Fix warnings found in build testing. 2019-08-27 16:20:39 -06:00
zhangyuan7
ba598841e5 net/tcp/Kconfig, include/nuttx/net/netconfig.h: Make RTO and WAIT_TIME configurable. This fix is not really correct: TCP RTO really should be calculated dynamically for each TCP connection: http://sgros.blogspot.com/2012/02/calculating-tcp-rto.html. 2019-08-26 11:03:27 -06:00
biantao
f1c46b4498 net/tcp/tcp_send.c: Fix RST packet with wrong ack number. 2019-08-26 10:16:43 -06:00
Anthony Merlino
70404ed0dc Merged in antmerlino/nuttx/iobinstrumentation (pull request #1001)
Iobinstrumentation

* mm/iob: Introduces producer/consumer id to every iob call. This is so that the calls can be instrumented to monitor the IOB resources.

* iob instrumentation - Merges producer/consumer enumeration for simpler IOB user.

* fs/procfs: Starts adding support for /proc/iobinfo

* fs/procfs: Finishes first pass of simple IOB user stastics and /proc/iobinfo entry

Approved-by: Gregory Nutt <gnutt@nuttx.org>
2019-08-16 22:42:25 +00:00
Gregory Nutt
124212624d net: Minor fix in error checking in TCP/UDP Tx drain logic 2019-07-28 07:39:55 -06:00
Gregory Nutt
361d85ae35 net/tcp and udp: Fix errors in the new implementation of SO_LINGER. The tcp_drain() and udp_drain() functions were casting the working argument to the wrong type, resulting in hangs and abnormal behavior. There is a complexity in the tcp drain logic when the remote peer closes the socket before all Tx data has been flushed. Sometimes we are not notified of this case and wait the entire timeout unnecessarily. There is a workaround in place in tcp_txdrain(), but this really should be revisited. 2019-07-27 10:26:52 -06:00
Gregory Nutt
1346f29151 net/: Fix alignment and spacing problems found by tools/nxstyle. 2019-07-02 18:02:23 -06:00
Gregory Nutt
574595dc32 Still fixing new warnings found in build testing. 2019-07-01 15:56:34 -06:00
Gregory Nutt
8774977f4d Fix warnings found in build testing. 2019-07-01 15:22:42 -06:00
Gregory Nutt
de5a6163d5 This commit implements a proper version of SO_LINGER. Not sufficiently tested on initial commit.
Squashed commit of the following:

    net/: Fix some naming inconsistencies, Fix final compilation issies.

    net/inet/inet_close():  Now that we have logic to drain the buffered TX data, we can implement a proper lingering close.

    net/inet,tcp,udp:  Add functions to wait for write buffers to drain.

    net/udp:  Add support for notification when the UDP write buffer becomes empty.

    net/tcp:  Add support for notification when the TCP write buffer becomes empty.
2019-07-01 12:25:32 -06:00
Gregory Nutt
d0c569d7b5 net/tcp/tcp_send_buffered.c: Fix a warning reported in build testing. 2019-06-21 13:48:27 -06:00
Gregory Nutt
2cf195e6d1 configs: Fix included path used in all arch/z80 XDS-II builds (ez80 and z8). 2019-06-02 16:16:44 -06:00
Valmantas Paliksa
d3cedfb823 net/tcp/tcp_send_buffered.c: Fix deadlock in iob_copyin when iob buffers are exhausted and network lock is taken. 2019-05-31 07:29:17 -06:00
Gregory Nutt
b49be4bb20 Squashed commit of the following:
arch/:  Removed all references to CONFIG_DISABLE_POLL.  The standard POSIX poll() can not longer be disabled.
    sched/ audio/ crypto/:  Removed all references to CONFIG_DISABLE_POLL.  The standard POSIX poll() can not longer be disabled.
    Documentation/:  Removed all references to CONFIG_DISABLE_POLL.  The standard POSIX poll() can not longer be disabled.
    fs/:  Removed all references to CONFIG_DISABLE_POLL.  The standard POSIX poll() can not longer be disabled.
    graphics/:  Removed all references to CONFIG_DISABLE_POLL.  The standard POSIX poll() can not longer be disabled.
    net/:  Removed all references to CONFIG_DISABLE_POLL.  The standard POSIX poll() can not longer be disabled.
    drivers/:  Removed all references to CONFIG_DISABLE_POLL.  The standard POSIX poll() can not longer be disabled.
    include/, syscall/, wireless/:  Removed all references to CONFIG_DISABLE_POLL.  The standard POSIX poll() can not longer be disabled.
    configs/:  Remove all references to CONFIG_DISABLE_POLL.  Standard POSIX poll can no longer be disabled.
2019-05-21 18:57:54 -06:00
Gregory Nutt
f6b00e1966 tools/nxstyle.c: Fix logic error that prevent detecion of '/' and '/=' as operators. net/: Minor updates resulting from testing tools/nxstyle. 2019-03-11 12:48:39 -06:00
Gregory Nutt
69056d4053 net/: The value ERROR should never be returned from internal OS functions. That is reserved for returning values to appliations with the errno value set. Within the OS, errors are returned with a negated errno value ALWAYS. 2019-02-14 15:38:36 -06:00
Gregory Nutt
359753adee net/tcp/tcp_send_buffered.c: Fix assertion crash when window size is zero. 2019-02-14 08:39:16 -06:00
Gregory Nutt
0bc800d71f net/tcp/tcp.h: Fix a muffed edit to conditional found in build testing. 2019-02-11 15:54:31 -06:00
Gregory Nutt
efe65749ce Fix condition logic: The setup seems to support a network without sockets. That is not the case.
Squashed commit of the following:

    sched/sched/sched_getsockets.c:  Fix an error in conditional compilation.
    fs/:  Remove all conditional logic based on CONFIG_NSOCKET_DESCRIPTORS == 0
    Documentation/:  Remove all references to CONFIG_NSOCKET_DESCRIPTORS == 0
    include/:  Remove all conditional logic based on CONFIG_NSOCKET_DESCRIPTORS == 0
    libs/:  Remove all conditional logic based on CONFIG_NSOCKET_DESCRIPTORS == 0
    net/:  Remove all conditional logic based on CONFIG_NSOCKET_DESCRIPTORS == 0
    sched/:  Remove all conditional logic based on CONFIG_NSOCKET_DESCRIPTORS == 0
    syscall/:  Remove all conditional logic based on CONFIG_NSOCKET_DESCRIPTORS == 0
    tools/:  Fixups for CONFIG_NSOCKET_DESCRIPTORS no longer used to disable sockets.
2019-02-11 15:47:25 -06:00
Xiang Xiao
fe421022e2 sched/wqueue/kwork_notifier.c and several Kconfig files: Notifier should work with either lpwork or hpwork and other minor typo fix. 2019-01-27 11:02:56 -06:00
Xiang Xiao
0a6e234962 net/neighbor: neighbor_lookup() checks if the target IP belongs to one of the local network devices. 2018-11-09 14:14:15 -06:00
zhangyuan7
f036e2a32a net/tcp/tcp_backlog.c: Add SOMAXCONN definition and implementaton per OpenGroup.org. 2018-11-09 11:20:33 -06:00
Xiang Xiao
d2cfd398ba Fix compiler error and warning when CONFIG_NET_SENDFILE=y 2018-11-09 11:17:43 -06:00