This code is marked as dead when BUF_SZ is a multiple of 4 (current
situation with unit tests)
This hopefully fixes a -Wunreachable-code failure found by Travis CI
This migrates the sendmsg TCP test from socket examples (task #14408)
to socket unit tests
Additionally, this adds support for testing recvmsg, creating a TCP
test for both sendmsg/recvmsg (referred to as msgapi test)
This also makes a small change to msgapi UDP to clear the receive
buffer after verifying the previous datagram
Using {0} broke Travis CI even though this should be correct for
initializing struct msghdr (see example in Linxu man pages:
http://man7.org/linux/man-pages/man3/cmsg.3.html)
Just use memset for now which is the common approach in LwIP codebase
This converts the sendmsg test to use recvmsg for receiving, thus
exercising both sendmsg and recvmsg in a single test
This also adjusts the test naming to communicate all message APIs
(sendmsg/recvmsg) are being tested
This commit moves the sendmsg UDP test from socket examples to socket
unit tests
The test has been converted to send/receive on the loopback interface
and also test a connected sendmsg with NULL msg_name
test_tcp_fast_rexmit_wraparound correctness relied on the congestion
window being in congestion avoidance so that only a single TCP_MSS
segment is sent upon ACKing the first segment.
It's not known whether the test was relying tcp_alloc() to set ssthresh
to 0 and thus start in congestion avoidance or if the test was working by
accident until changes in b90a54f989
This fixes the test by enforcing the requirement of starting in
congestion avoidance
Signed-off-by: goldsimon <goldsimon@gmx.de>
If a locally generated TCP SYN packet is replied to with an ACK
packet, lwIP immediately sends a RST packet followed by resending the
SYN packet. This is expected, but on loopback interfaces the resent
SYN packet may immediately get another ACK reply, typically when the
other endpoint is in TIME_WAIT state (which ignores the RSTs). The
result is an endless loop of SYN, ACK, RST packets.
This patch applies the normal SYN retransmission limit in this
scenario, such that the endless loop is limited to a brief storm.
This commit changes ssthresh to be the largest effective congestion
window (amount of in-flight data). This follows the guidance of RFC
5681 which recommends setting ssthresh arbitrarily high.
LwIP was previously using the receive window value at the end of the
3-way handshake and in the case of an active open where the receiver
used window scaling and/or window auto-tuning, this resulted in a very
small ssthresh value even though the window ramped up once the connection
was established
User should not use ppp_connect or ppp_listen return value to retry
later, it must wait for the callback to be called. This is primarily
done this way to have a consistent behavior with and without the
holdoff feature.
Remove returned error value from PPP link level API connect and listen
callbacks because we are not using them anymore, then make ppp_connect
or ppp_listen to always return ERR_OK, thus we are not breaking the PPP
user API.
We don't need the return code here, all PPP link level drivers can't
fail at all (e.g. PPPoS) or retry if necessary (PPPoE and PPPoL2TP).
