GCC complains about struct sockaddr accesses due to the various
address-family-specific variants being of differing sizes. Let's not
mess with code (which looks correct), just silence the compiler.
Signed-off-by: Keith Packard <keithp@keithp.com>
If the dispatcher is muxing the connection i.e., so there are two
services for the same port, then mark service socket descriptor
of the service with the socket number so that "net sockets" shell
command can show a proper value for it.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
The socket services API has a limitation where the user data is shared
between file descriptors described in the same service.
This can cause problem in DNS dispatcher where each listened socket
needs to have their own dispatcher struct set as user data so that we
can map between dispatcher context and socket. Solve this by always
have a dispatcher table as user data, and then have the actual mapping
done via the dispatcher table when receiving data to the dispatcher socket.
Fixes#78146
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Move the network buffer header file from zephyr/net/buf.h to
zephyr/net_buf.h as the implementation now lives outside of the networking
subsystem.
Add (deprecated) zephyr/net/buf.h header to maintain compatibility with old
file path.
Signed-off-by: Henrik Brix Andersen <henrik@brixandersen.dk>
Utilize a code spell-checking tool to scan for and correct spelling errors
in all files within the `subsys/net/lib` directory.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
If we have configured the DNS dispatcher to be only as a
responder but receive a query response, or if we are only
as a resolver but receive a query, then the dispatcher just
ignores the packet and returns -ENOENT.
Unfortunately we print an error message in this case
[00:10:18.818,000] <err> net_dns_dispatcher: DNS recv error (-2)
which is totally unnecessary and causes confusion so do not
print an error message in this case.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Mask the DNS_CLASS_FLUSH value when checking if the DNS_CLASS_IN
is set when unpacking a query.
Fixes#74829
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
The network interface name that is copied to if_req struct might
be missing terminating null.
Fixes#74795
Coverity-CID: 368797
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
A few issues were identified with DNS_SD segment buffers, making it work
"by chance" (depending on memory layout):
- size check in dns_sd_query_extract() did not take NULL termination
into account, and in result could overflow provided buffer
- the proto_buf in send_sd_response() can either be used to parse
protocol or domain, depending on number of segments in the query.
It should therefore be large enough to hold either.
- Similarily, instance_buf should be able to hold
DNS_SD_INSTANCE_MAX_SIZE, not DNS_SD_SERVICE_MAX_SIZE.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The family variable was not initialized and could potentially
be left like that. This could only affect error printout.
Fixes#74796
Coverity-CID: 368799
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
We should return the errno value in such case, as ret will always be -1
on failure, which might misleading, as it does not give any hint on what
failed.
Also bump the log level of the error message, so the failure is visible
w/o needing to enable debug logs. Such a failure is fatal for the DNS
subsystem, so error level is justified here.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The `dns_data` buffer, allocated by the DNS dispatcher was dereferenced
twice - once in registered DNS handler, second time in the dispatcher
itself.
Since the buffer was allocated by the dispatcher, and it's not really
guaranteed that the buffer will be freed in the registered handler (this
depends on the processing outcome, the function may return early w/o
freeing the net buf in case of errors), it makes most sense for the
dispatcher to keep ownership of the buffer. Hence, the registered
handlers will no longer release the buffer provided in any case, and the
dispatcher will free it on exit.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Instead of printing either A or AAAA resource query type,
print the correct query type value.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Allow mDNS resolver and responder to to be used at the same
time so that both can use the port 5353. This requires
a DNS traffic dispatcher which affects also the normal DNS
resolver.
Fixes#72553
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
The socket allocation count was incorrectly calculated as
we have only one socket per listened port.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
If DNS_EAI_ALLDONE is returned, it indicates that the request
was done and should not be considered an error.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Instead of using net_context API directly, the LLMNR responder is
changed to use the socket service API. This allows DNS access
for offloaded sockets and can lower overall memory consumption.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Instead of using net_context API directly, the mDNS responder is
changed to use the socket service API. This allows DNS access
for offloaded sockets and can lower overall memory consumption.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Instead of using net_context API directly, the DNS resolver is
changed to use the socket service API. This allows DNS access
for offloaded sockets and can lower overall memory consumption.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
The size field in net_buf should not be used directly as then
the optional headroom will not be taken into account.
There is the net_buf_max_len() API that should be used instead.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
The "struct sockaddr" should only be used in casts and never
as a standalone variable because it might not have enough
space allocated for all the protocol specific fields.
So refactor the port_in_use() function to reflect that.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Check if the interface is down before joining IPv4 mDNS multicast
group. This avoids warning later when the network packet cannot be sent.
Fixes#71121
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Current implementation of mDNS responder does makes it mandatory to have
all the records set at compile time. It is not suited well for applications
that have to publish/unpublish or change records in runtime, e.g. data
received from the network.
Signed-off-by: Konrad Derda <konrad.derda@nordicsemi.no>
In order to reduce dns resolve requests when using the dns
resolver an optional cache was introduced. This cache
retains query requests for the TTL duration and therefore
prevents premature refetching of DNS RRs.
Signed-off-by: Carlo Kirchmeier <carlo.kirchmeier@zuehlke.com>
Adds an option to disable the automatic initialization of the default
dns_context. This lets applications use the default context, while also
managing the `init` and `close` functions.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
The original idea was to check that we have enough network
interfaces in the system. The check needs to verify max IPv4
and IPv6 supported interfaces instead of always checking
IPv6 one.
Fixes: #66843
Coverity-CID: 334899
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
We are creating a multicast address in mDNS or LLMNR
responder so set the TTL or hoplimit using the multicast
variant API.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
We specifically set TTL/hoplimit to 1 for LLMNR,
but only want to set it if in that specific case.
We must not pass TTL/hoplimit value 0 as that would
cause the packet to be dropped.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Added igmpv3 support based on the already existing structure for igmpv2.
The already existing api is not modified to prevent breaking exisiting
applications.
Signed-off-by: Ibe Van de Veire <ibe.vandeveire@basalte.be>
No issues with gcc but clang gives this error for
the *v4 variable few lines below.
.../lib/dns/mdns_responder.c:712:2: error: expected expression
struct net_context *v4;
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Instead of just listening first network interface in the system,
install a multicast listener to all available network interfaces.
Fixes#18748
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
rand32.h does not make much sense, since the random subsystem
provides more APIs than just getting a random 32 bits value.
Rename it to random.h and get consistently with other
subsystems.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
If an IPv4 address is requested on an IPv6 interface or vice versa
use a sane default fallback address to send the response.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
The number of buffer provided was verified in the final else block of a
long validation sequence. It would never be executed though, as one of
the conditions before would always evaluate to true.
As the number of buffers provided verification appears to be significant
in this case, as the buffers are referenced during other validations,
move this check at the beginning of the sequence instead. This also
eliminates the dead-code problem.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Until now iterable sections APIs have been part of the toolchain
(common) headers. They are not strictly related to a toolchain, they
just rely on linker providing support for sections. Most files relied on
indirect includes to access the API, now, it is included as needed.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
This commit simplifies the arguments for sending dns responses
and fixes an issue where both the ipv4 and ipv6 member of a header
union were accessed.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
Duplicate logging prints make it hard to find the origin.
This commit does a cleanup of the debug messages printed.
Also change the level of an invalid record to debug as this
is out of our control.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
The init infrastructure, found in `init.h`, is currently used by:
- `SYS_INIT`: to call functions before `main`
- `DEVICE_*`: to initialize devices
They are all sorted according to an initialization level + a priority.
`SYS_INIT` calls are really orthogonal to devices, however, the required
function signature requires a `const struct device *dev` as a first
argument. The only reason for that is because the same init machinery is
used by devices, so we have something like:
```c
struct init_entry {
int (*init)(const struct device *dev);
/* only set by DEVICE_*, otherwise NULL */
const struct device *dev;
}
```
As a result, we end up with such weird/ugly pattern:
```c
static int my_init(const struct device *dev)
{
/* always NULL! add ARG_UNUSED to avoid compiler warning */
ARG_UNUSED(dev);
...
}
```
This is really a result of poor internals isolation. This patch proposes
a to make init entries more flexible so that they can accept sytem
initialization calls like this:
```c
static int my_init(void)
{
...
}
```
This is achieved using a union:
```c
union init_function {
/* for SYS_INIT, used when init_entry.dev == NULL */
int (*sys)(void);
/* for DEVICE*, used when init_entry.dev != NULL */
int (*dev)(const struct device *dev);
};
struct init_entry {
/* stores init function (either for SYS_INIT or DEVICE*)
union init_function init_fn;
/* stores device pointer for DEVICE*, NULL for SYS_INIT. Allows
* to know which union entry to call.
*/
const struct device *dev;
}
```
This solution **does not increase ROM usage**, and allows to offer clean
public APIs for both SYS_INIT and DEVICE*. Note that however, init
machinery keeps a coupling with devices.
**NOTE**: This is a breaking change! All `SYS_INIT` functions will need
to be converted to the new signature. See the script offered in the
following commit.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
init: convert SYS_INIT functions to the new signature
Conversion scripted using scripts/utils/migrate_sys_init.py.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
manifest: update projects for SYS_INIT changes
Update modules with updated SYS_INIT calls:
- hal_ti
- lvgl
- sof
- TraceRecorderSource
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
tests: devicetree: devices: adjust test
Adjust test according to the recently introduced SYS_INIT
infrastructure.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
tests: kernel: threads: adjust SYS_INIT call
Adjust to the new signature: int (*init_fn)(void);
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
We get compile warnings of the form:
error: converting the result of
'<<' to a boolean; did you mean
'((__aeabi_ctype_table_ + 1)[(byte)] << 28) != 0'?
[-Werror,-Wint-in-bool-context]
if (!isprint(byte)) {
^
Since isprint (and the other is* functions) return an int, change check
to an explicit test against the return value.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
In dns_resolve_reconfigure() check if the DNS servers already exist
before cancel all ongoing queries. This will solve an issue with
getaddrinfo() returning DNS_EAI_CANCELED when receiving a retransmitted
DHCP offer and when receiving a IPv6 Router Advertisement.
Signed-off-by: Stig Bjørlykke <stig.bjorlykke@nordicsemi.no>
