9d3148406c
Squashed commit of the following:
Fix up some final compile isses.
net/netdev: Convert the network down notification logic to use the new wqueue-based notification factility.
net/udp: Convert the UDP readahead notification logic to use the new wqueue-based notification factility.
net/tcp: Convert the TCP readahead notification logic to use the new wqueue-based notification factility.
mm/iob: Convert the IOB notification logic to use the new wqueue-based notification factility.
sched/wqueue: Signals are not good IPCs to support the target poll functionality for several reasons including the amount of data that can be passed with a signal and in the fact that in protected and kernel modes, user threads executing signal handlers in protected, kernel memory is problematic. Instead, convert the same logic to perform the notifications via function callback on the high priority work queue.
186 lines
5.8 KiB
Plaintext
186 lines
5.8 KiB
Plaintext
#
|
|
# For a description of the syntax of this configuration file,
|
|
# see the file kconfig-language.txt in the NuttX tools repository.
|
|
#
|
|
|
|
menu "TCP/IP Networking"
|
|
|
|
config NET_TCP
|
|
bool "TCP/IP Networking"
|
|
default n
|
|
---help---
|
|
Enable or disable TCP networking support.
|
|
|
|
config NET_TCP_NO_STACK
|
|
bool "Disable TCP/IP Stack"
|
|
default n
|
|
select NET_TCP
|
|
---help---
|
|
Build without TCP/IP stack even if TCP networking support enabled.
|
|
|
|
if NET_TCP && !NET_TCP_NO_STACK
|
|
|
|
config NET_TCP_KEEPALIVE
|
|
bool "TCP/IP Keep-alive support"
|
|
default n
|
|
select NET_TCPPROTO_OPTIONS
|
|
---help---
|
|
Enable support for the SO_KEEPALIVE socket option
|
|
|
|
config NET_TCPURGDATA
|
|
bool "Urgent data"
|
|
default n
|
|
---help---
|
|
Determines if support for TCP urgent data notification should be
|
|
compiled in. Urgent data (out-of-band data) is a rarely used TCP feature
|
|
that is very seldom would be required.
|
|
|
|
config NET_TCP_CONNS
|
|
int "Number of TCP/IP connections"
|
|
default 8
|
|
---help---
|
|
Maximum number of TCP/IP connections (all tasks)
|
|
|
|
config NET_MAX_LISTENPORTS
|
|
int "Number of listening ports"
|
|
default 20
|
|
---help---
|
|
Maximum number of listening TCP/IP ports (all tasks). Default: 20
|
|
|
|
config NET_TCP_READAHEAD
|
|
bool "Enable TCP/IP read-ahead buffering"
|
|
default y
|
|
select NET_READAHEAD
|
|
select MM_IOB
|
|
---help---
|
|
Read-ahead buffers allows buffering of TCP/IP packets when there is no
|
|
receive in place to catch the TCP packet. In that case, the packet
|
|
will be retained in the NuttX read-ahead buffers.
|
|
|
|
You might want to disable TCP/IP read-ahead buffering on a highly
|
|
memory constrained system that does not have any TCP/IP packet rate
|
|
issues. But, if disabled, there will probably be more packet
|
|
retransmissions or even packet loss.
|
|
|
|
Make sure that you check the setting in the I/O Buffering menu.
|
|
These settings are critical to the reasonable operation of read-
|
|
ahead buffering.
|
|
|
|
config TCP_READAHEAD_NOTIFIER
|
|
bool "Support TCP read-ahead notifications"
|
|
default n
|
|
depends on NET_TCP_READAHEAD && SCHED_HPWORK
|
|
---help---
|
|
Enable building of TCP read-ahead notifier logic that will execute a
|
|
worker function on the high priority work queue when read-ahead data
|
|
is available. This is is a general purpose notifier, but was
|
|
developed specifically to support poll() logic where the poll must
|
|
wait for read-ahead data to become available.
|
|
|
|
config NET_TCP_WRITE_BUFFERS
|
|
bool "Enable TCP/IP write buffering"
|
|
default n
|
|
select NET_WRITE_BUFFERS
|
|
select MM_IOB
|
|
---help---
|
|
Write buffers allows buffering of ongoing TCP/IP packets, providing
|
|
for higher performance, streamed output.
|
|
|
|
You might want to disable TCP/IP write buffering on a highly memory
|
|
memory constrained system where there are no performance issues.
|
|
|
|
if NET_TCP_WRITE_BUFFERS
|
|
|
|
config NET_TCP_NWRBCHAINS
|
|
int "Number of pre-allocated I/O buffer chain heads"
|
|
default 8
|
|
---help---
|
|
These tiny nodes are used as "containers" to support queuing of
|
|
TCP write buffers. This setting will limit the number of TCP write
|
|
operations that can be "in-flight" at any give time. So a good
|
|
choice for this value would be the same as the maximum number of
|
|
TCP connections.
|
|
|
|
config NET_TCP_WRBUFFER_DEBUG
|
|
bool "Force write buffer debug"
|
|
default n
|
|
depends on DEBUG_FEATURES
|
|
select IOB_DEBUG
|
|
---help---
|
|
This option will force debug output from TCP write buffer logic,
|
|
even without network debug output. This is not normally something
|
|
that would want to do but is convenient if you are debugging the
|
|
write buffer logic and do not want to get overloaded with other
|
|
network-related debug output.
|
|
|
|
config NET_TCP_WRBUFFER_DUMP
|
|
bool "Force write buffer dump"
|
|
default n
|
|
depends on DEBUG_NET || NET_TCP_WRBUFFER_DEBUG
|
|
select IOB_DEBUG
|
|
---help---
|
|
Dump the contents of the write buffers. You do not want to do this
|
|
unless you really want to analyze the write buffer transfers in
|
|
detail.
|
|
|
|
endif # NET_TCP_WRITE_BUFFERS
|
|
|
|
config NET_TCP_RECVDELAY
|
|
int "TCP Rx delay"
|
|
default 0
|
|
---help---
|
|
If NET_TCP_READAHEAD_BUFFERS is undefined, then there will be no buffering
|
|
of TCP/IP packets: Any TCP/IP packet received will be ACKed, but its contents
|
|
will be dropped in the bit-bucket.
|
|
|
|
One low-performance option is delay for a short period of time after a
|
|
TCP/IP packet is received to see if another comes right behind it. Then
|
|
the packet data from both can be combined. This option only makes since
|
|
if performance is not an issue and you need to handle short bursts of
|
|
small, back-to-back packets. The delay is in units of deciseconds.
|
|
|
|
config NET_TCPBACKLOG
|
|
bool "TCP/IP backlog support"
|
|
default n
|
|
---help---
|
|
Incoming connections pend in a backlog until accept() is called.
|
|
The size of the backlog is selected when listen() is called.
|
|
|
|
config NET_TCP_SPLIT
|
|
bool "Enable packet splitting"
|
|
default n
|
|
depends on !NET_TCP_WRITE_BUFFERS
|
|
---help---
|
|
send() will not return until the transfer has been ACKed by the
|
|
recipient. But under RFC 1122, the host need not ACK each packet
|
|
immediately; the host may wait for 500 MS before ACKing. This
|
|
combination can cause very slow performance with small transfers are
|
|
made to an RFC 1122 client. However, the RFC 1122 must ACK at least
|
|
every second (odd) packet.
|
|
|
|
This option enables logic to trick the RFC 1122 host be exploiting
|
|
this last RFC 1122 requirement: If an odd number of packets were to
|
|
be sent, then send() will split the last even packet to guarantee
|
|
that an even number of packets will be sent and the RFC 1122 host
|
|
will ACK the final packet immediately.
|
|
|
|
if NET_TCP_SPLIT
|
|
|
|
config NET_TCP_SPLIT_SIZE
|
|
int "Split size threshold"
|
|
default 40
|
|
---help---
|
|
Packets of this size or smaller than this will not be split.
|
|
|
|
endif # NET_TCP_SPLIT
|
|
|
|
config NET_SENDFILE
|
|
bool "Optimized network sendfile()"
|
|
default n
|
|
---help---
|
|
Support larger, higher performance sendfile() for transferring
|
|
files out a TCP connection.
|
|
|
|
endif # NET_TCP && !NET_TCP_NO_STACK
|
|
endmenu # TCP/IP Networking
|