Handle this corner case with TCP connection closing:
1) Client A connects, it is accepted and can send data to us
2) Client B connects, the application needs to call accept()
before we will receive any data from client A to the application.
The app has not yet called accept() at this point (for
whatever reason).
3) Client B then disconnects and we receive FIN. The connection
cleanup is a bit tricky as the client is in half-connected state
meaning that the connection is in established state but the
accept_q in socket queue contains still data which needs to be
cleared.
4) Client A then disconnects, all data is sent etc
The above was not working correctly as the system did not handle the
step 3) properly. The client B was accepted in the application even
if the connection was closing.
After this commit, the commit called "net: tcp: Accept connections
only in LISTENING state" and related other commits are no longer
needed and are reverted.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Do not try to re-use net_context.user_data field as in many places
(like in accept) it is expected to contain pointer to net_context.
Storing the socket flags will corrupt the value. To simplify and
make things less error prone, use socket specific field in net_context
to store the socket flags.
Fixes#19191
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The net_context_send API documentation was missing the fact,
that the callback might be called after the net_context_send()
has returned. Also currently the timeout value is not used properly
anywhere in the code. This is left like this as there is not much
use for the timeout value atm. This might be fixed later if there
is proper use case for it.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The SO_SOCKS5 socket option can be used by the application to
set the SOCKS5 proxy details. These details will be used when
connecting to peer.
Signed-off-by: Ravi kumar Veeramally <ravikumar.veeramally@linux.intel.com>
The SO_TXTIME socket option can be used by the application to
tell the network device driver the exact moment when the
network packet should be sent.
This feature is also implemented in Linux.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
move atomic.h to sys/atomic.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
If the socket is closed, then do CAN detach if that is needed.
This way the CAN interrupts are not received if there are no
CAN sockets listening the data.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
And also to the relevant callbacks.
That parameter is not used anywhere so it is useless.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Now that legacy functions are removew, let's rename the new functions by
removing the _new suffix.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Former net_context_send/sendto and net_context_create_ipv4/ipv6 are now
unused and can be removed.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Though core system is able to manage packet timestamping internaly (gptp
requires it for instance), it might be necessary to enable/disable
packet timestamping from net context directly.
Currently this will be only used by the tx timestamp test. So this
support is disabled by default. (And gptp does not require it anyway).
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
When compiling with C++ support, a build error occur:
net_context.h: In function ‘net_sock_type net_context_get_type(net_context*)’:
net_context.h:402:75: error: invalid conversion from ‘long unsigned int’ to ‘net_sock_type’ [-fpermissive]
enum net_sock_type t = ((context->flags & NET_CONTEXT_TYPE) >> 6);
Let's fix that using a cast.
Signed-off-by: Loic Poulain <loic.poulain@linaro.org>
Newly added net_context_send_new()/net_context_sendto_new() take
void *buf, size_t len params but in docstring, refer to "network
buffer", which is apperently copy-paste artifact.
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
Since net_context_sendto_new() does not take a net_pkt anymore, the only
way to set net_pkt's ttl/hop_limit is to pass it through net_context.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
As the value 0 is a valid network interface index, we cannot use
unsigned value for interface index as that would not allow to
distinguish an invalid value. So make interface index a signed
8-bit value which is ok as we do not expect to have more than 127
network interfaces in the system.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
This commit adds basic packet socket support to net_context and
allows application to receive or send network packets in raw
format.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Ravi kumar Veeramally <ravikumar.veeramally@linux.intel.com>
If status is 0, both ip_hdr and proto_hdr will own a pointer to the
relevant IP and Protocol headers. In order to know which of ipv4/ipv6
and udp/tcp one will need to use respectively net_pkt_family(pkt) and
net_context_get_ip_proto(context).
Having access to those headers directly, many callbacks will not need
to parse the packet again no get the src/dst addresses or the src/dst
ports. This will be change after this commit.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
These will be specifically needed in TCP, as well as being used in
context internally.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
It's not anymore up to user to provide the pkt. Context will build the
packet according to its metadata and provided buffer and length.
It currently supports only IPv4 and UDP.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
If the net_context functions are accessed from preemptive priority,
then we need to protect various internal resources.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
c++ does not allow implicit conversions and setting -fpermissive just
causes a huge load of warnings to appear and hides real errors.
This commit converts those implicit conversions to c-style explicit
conversions.
Signed-off-by: Alexander Polleti <metapsycholo@gmail.com>
Any word started with underscore followed by and uppercase letter or a
second underscore is a reserved word according with C99.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move struct members around in networking code so that we avoid
unnecessary holes inside structs. No functionality changes by
this commit.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The wifi_winc1500 driver's socket id is stored in
net_context->user_data, which may be overwritten later at
the socket layer, which also uses the net_context->user_data
field to store socket flags.
This patch introduces a dedicated offload_context field
for use by offload drivers, and updates the wifi_winc1500 offload
driver to use this field instead of user_data.
Fixes#8820
Signed-off-by: Gil Pitney <gil.pitney@linaro.org>
Add tls_context structure that stored data required by TLS socket
implementation. This structure is allocated from global pool during
socket creation and freed during socket closure.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Add context option support and implement PRIORITY option that
can be used to classify the network traffic to different trafic
classes according to said priority value.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Of these, only struct net_ipv6_nbr_data::send_ns is a descriptive
change:
send_ns is used for timing Neighbor Solicitations in general, not
just for DAD.
The rest are typo/grammar fixes.
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
Describe details and usecase for using this function. This follows
earlier updates for macros used to define buffers used by this
function (in 09b967366).
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
The server is able to listen and serve multiple incoming
connections. This commit does not add support for multiple
incoming TLS connections.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Fix misspellings in .h files missed during code reviews
and affecting generated API documentation
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
This fixes the existing situation that "if application buffers data,
it's the problem of application". It's actually the problem of the
stack, as it doesn't allow application to control receive window,
and without this control, any buffer will overflow, peer packets
will be dropped, peer won't receive acks for them, and will employ
exponential backoff, the connection will crawl to a halt.
This patch adds net_context_tcp_recved() function which an
application must explicitly call when it *processes* data, to
advance receive window.
Jira: ZEP-1999
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
Document clearly how and in what context, the various callbacks
in net_context API are being called.
Jira: ZEP-2352
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
No need to print errors if assinging null values into net_buf
pools as this is a normal condition if those pools are not used.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
By moving user_data member at the beginning of structure. With
refcount at the beginning, reliable passsing of contexts via
FIFO was just impossible. (Queuing contexts to a FIFO is required
for BSD Sockets API).
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
Two changes are required so far:
* There's unavoidable need to have a per-socket queue of packets
(for data sockets) or pending connections (for listening sockets).
These queues share the same space (as a C union).
* There's a need to track "EOF" status of connection, synchronized
with a queue of pending packets (i.e. EOF status should be processed
only when all pending packets are processed). A natural place to
store it per-packet then, and we had a "sent" bit which was used
only for outgoing packets, recast it as "eof" for incoming socket
packets.
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
Convert code to use u{8,16,32,64}_t and s{8,16,32,64}_t instead of C99
integer types.
Jira: ZEP-2051
Change-Id: I4ec03eb2183d59ef86ea2c20d956e5d272656837
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
- net_pkt becomes a stand-alone structure with network packet meta
information.
- network packet data is still managed through net_buf, mostly named
'frag'.
- net_pkt memory management is done through k_mem_slab
- function got introduced or relevantly renamed to target eithe net_pkt
or net_buf fragments.
- net_buf's sent_list ends up in net_pkt now, and thus helps to save
memory when TCP is enabled.
Change-Id: Ibd5c17df4f75891dec79db723a4c9fc704eb843d
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
There have been long lasting confusion between net_buf and net_nbuf.
While the first is actually a buffer, the second one is not. It's a
network buffer descriptor. More precisely it provides meta data about a
network packet, and holds the chain of buffer fragments made of net_buf.
Thus renaming net_nbuf to net_pkt and all names around it as well
(function, Kconfig option, ..).
Though net_pkt if the new name, it still inherit its logic from net_buf.
'
This patch is the first of a serie that will separate completely net_pkt
from net_buf.
Change-Id: Iecb32d2a0d8f4647692e5328e54b5c35454194cd
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
User can configure its own pools for data that needs to
be transmitted out (TX). This helps to avoid deadlocking
the system if user space application uses all the buffers
in the system, and the core IP stack tries to get buffer
that needs to be sent out.
By default the net_buf pool support in net_context is turned
off as application developer needs to create the pools and
tie them to desired contexts.
Change-Id: Ida4a1771d34d6c250974e56fba4f0e0b2592cb29
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Current behaviour has an issue when UDP context is created with local
port number 0, net_conn_input() happens to treat zero port as
a wildcard ("receive packets for all ports"). net_context_bind()
for a UDP context doesn't affect its existing connection in any way.
Proposed solution is, context should be created with a random free
port assigned and bind() updates connection information from context.
Jira: ZEP-1644
Change-Id: Idb3592b58c831d986763312077b0dcdd95850bc9
Signed-off-by: Ravi kumar Veeramally <ravikumar.veeramally@linux.intel.com>
