If an app request call-back modifies any data associated with a resource,
then multiple threads can attempt to unsafely modify the data at the same time.
Lock the resource when calling the app request handler (as global_lock is
unlocked) to prevent concurrent changes to the resource information.
If the app request handler is definately known to not modify the resource
information, then COAP_RESOURCE_SAFE_REQUEST_HANDLER can be set in the flags
for coap_resource_init(). Usually if there is a PUT, POST, PATCH or IPATCH
handler it is unsafe to set this. Setting this will reduce lock contention.
Include common system header files in coap_internal.h based on HAVE_*H
definitions.
Remove these system header files from include/*/*.h.
Add in system header files where appropriate in various apps.
Add some improvements to make secure coap work on Zephyr 4.3.0
with mbedtls (not using Zephyr TLS protocol options).
We now set the build options to zephyr, so we make use of the
internal mbedtl in zephyr, so we do not require a specific header
file anymore.
In older versions of Zephyr, where mbedtls does not expose the
configuration, we can force-set the macro in the C flags, so that
mbedtls will have our required function.
This way, we do not need to change the mbedtls included in Zephyr,
while still supporting a few older versions.
Function coap_context_set_max_body_size() added.
For servers, if body size is too large, a 4.13 response is generated.
For clients, if the body size is too large, the server gets a RST and
no date is passed to the application. A new event COAP_EVENT_BLOCK_ISSUE
is generated.
Add in new functions
coap_new_client_session3()
coap_new_client_session_pki3()
coap_new_client_session_psk3()
coap_new_client_session_oscore3()
coap_new_client_session_oscore_pki3()
coap_new_client_session_oscore_psk3()
which effectively does coap_session_set_app_data2() and
coap_ws_set_host_request() early on in the session set up cycle.
The first coap_send(pdu) is held back until after the optional tests for
Extended Token or Q-Block support are complete with the return from coap_send()
containing the held back pdu's mid.
If there are any errors with this PDU when subsequentially sent, the
event COAP_EVENT_FIRST_PDU_FAIL gets raised.
Use correct input function for netif_add() in examples.
Invoke coap_io_process() as soon as possibile following TCP input /
error.
Update rcvd window size after receipt of data.
Add in new function coap_add_data_large_request_app() that provides a call-back
to get individual blocks rather than having a large buffer held in
(contrained) memory.
If we correctly determine the presence of the sys/time.h and time.h
files using the HAVE_TIME_H and HAVE_SYS_TIME_H macros, then the LWIP
compilation will result in a duplicate function error. Perhaps previous
versions of LWIP did not have these files, in which case everything
would have worked correctly.
These changes add the necessary ifdefs and keep the optimal timer
function for LWIP.
Adds missed #include <time.h> lines to examples and documentation.
Adds the detection of the sys/select.h header to the CMake and autoconf.
Host names larger than 255 bytes will cause an internal buffer overflow.
Hostnames provided to coap_resolve_address_info() now have their length validated.
Discovered by SecMate (https://secmate.dev).
Sanity check host lengths when parsing a CoAP URI when using the coap_split_uri()
function.
Some of the Zephyr include files need to be C++, not C based. The 'extern "C"'
wrapper in include/coap3/coap.h means that all #include are treated as being C.
Remove 'extern "C"' wrapper from include/coap3/coap.h and add wrapper to
include files after #include and before any function / structure definitions.
Add in a C++ build test for CI.
Let libcoap do the necessary multi-thread locking during resource
generation, lookup and deletion.
Supported using a new handler set up by coap_register_dynamic_resource_handler().
Server to Client is already supported, this provides the reverse with
coap_session_set_type_server().
Add in call-home support to coap-client and coap-server for testing.
Addition of a new event COAP_EVENT_SERVER_SESSION_CONNECTED.
Changes:
- Remove duplicate OBSERVE_PERSIST configuration block
Improve example build:
- Make virtual environment activation optional in Makefile
- Remove --fresh flag unsupported in newer west versions
- Copy mbedTLS config header to workspace include directory
- Add TARGET check for mbedTLS before configuring include directories
- Add mbedTLS linking to coap-3 target
Add POSIX API support:
- Enable CONFIG_POSIX_API=y in examples/zephyr
- Replace direct Linux/Windows system calls with Zephyr POSIX API
equivalents
- Add Zephyr's sys_rand_get() to coap_prng_default()
- Define missing network macros (IN_MULTICAST, IN6_IS_ADDR_MULTICAST,
etc.)
- Add FIONBIO and OPTVAL_T/OPTVAL_GT macro definitions for Zephyr
- Add native Zephyr timing functions to replace missing mbedTLS timing
support
- Fix fd_set mapping using zephyr/net/socket_select.h
- Exclude strnlen redefinition conflicts for Zephyr
- Add proper IPv6 packet info and address mapped definitions
(Patch) Add DNS resolution and net setup to example client:
- Enable CONFIG_DNS_RESOLVER for hostname resolution
- Add Linux networking setup instructions for native_sim board
- Configure static IP gateway and DNS server (8.8.8.8)
- Add DHCP vs static IP configuration documentation
Co-authored-by: mrdeep1
Move the MinGW #if tests to after where mswsock.h is included.
Tidy up documentation.
Tested against MSYS2 UCRT64, CLANG64, MINGW32 and MINGW64 on Windows 11.
Fixed a couple of minor build issues found.
In particular for TCP support.
Also added in retrying to get the lwip git 3 times for the CI tests as the
lwip host server is currently not that reliable, returning 502 gateway errors.
Add in limited caching for multiple clients requesting an Observe subscription
for the same upstream session resource so that only a single observation request
is sent upstream and any unsolicited response is sent back to all the
requesting clients.
This requires a change in the recently added handler registered to
coap_register_proxy_response_handler(), which no longer should no longer call
coap_proxy_forward_response(), but just return a PDU for forwarding on.
When initating a multicast request, multiple unicast responses can occur.
If the unicast response subsequently generates further traffic (in other words
a BLOCK1 or BLOCK2 transfer) to complete the request/response sequence, each
unicast transfer can behave differently.
Set up different coap_lg_crcv_t for each unicast response for the session.
Set up different coap_lg_xmit_t for each unicast request if doing large transmits.
Reference initial transmitted request PDU rather than maintaining copies of the
PDU in coap_lg_crcv_t or coap_lg_xmit_t.
Jon Shallow