Gregory Nutt
|
dec7ecbd56
net/tcp/tcp_recvwindow.c: In order to receive data we must not only have IOBs available, but we must also have at least one IOB chain qentry available. Otherwise, we will advertise that we an buffer a lot of data when, in fact, we cannot. This is an expermental fix to a performance problem noted by Masayuki Ishikawa.
|
6 years ago |
Gregory Nutt
|
b3f0aab00a
Squashed commit of the following:
|
6 years ago |
Gregory Nutt
|
76eec53e4f
mm/iob: iob_navail() was returning the number of free IOB chain queue entries, not the number of free IOBs. Completely misnamed.
|
6 years ago |
Gregory Nutt
|
5764d5a529
net/tcp: Back out part of commit d9443888881e5d638a08ab3f6fdc203c9f9f475c. I see a few places that say that the maximum receive window size is 32,767 (INT16_MAX), but most say that it is 65,535 (UINT16_MAX).
|
6 years ago |
Gregory Nutt
|
d944388888
configs/viewtool-stm32f107: Add a TCP performance test using apps/examples/tcpblaster.
|
6 years ago |
Gregory Nutt
|
22cd0d47fa
This commit attempts remove some long standard confusion in naming and some actual problems that result from the naming confusion. The basic problem is the standard MTU does not include the size of the Ethernet header. For clarity, I changed the naming of most things called MTU to PKTSIZE. For example, CONFIG_NET_ETH_MTU is now CONFIG_NET_ETH_PKTSIZE.
|
6 years ago |
Gregory Nutt
|
e59b26370d
Squashed commit of the following:
|
6 years ago |
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.
|
6 years ago |