This commit adds a check, in the `dns_read` function, before
dereferencing the query index returned by the `dns_validate_msg`
function.
This fixes the warnings generated by the GCC 12 such as:
error: array subscript -1 is below array bounds of
'struct dns_pending_query[5]' [-Werror=array-bounds]
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
Logging v1 has been removed and log_strdup wrapper function is no
longer needed. Removing the function and its use in the tree.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Return ret error code instead of always returning zero causing
warnings:
...
subsys/net/lib/dns/resolve.c:975:6: warning: variable 'ret' set but
not used [-Wunused-but-set-variable]
int ret = 0;
^
...
Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
In order to bring consistency in-tree, migrate all subsystems code to
the new prefix <zephyr/...>. Note that the conversion has been scripted,
refer to zephyrproject-rtos#45388 for more details.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
A common pattern here was to take the work item as the subfield of a
containing object. But the contained field is not a k_work, it's a
k_work_delayable.
Things were working only because the work field was first, so the
pointers had the same value. Do things right and fix things to
produce correct code if/when that field ever moves within delayable.
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Fixed mutli-IP DNS resolution as previously the same IP address was
used to populate all AI entries and added DNS_RESOLVER_AI_MAX_ENTRIES
config entry to define max number of IP addresses per DNS name to be
handled.
Signed-off-by: Ievgen Glinchuk <john.iceblink@gmail.com>
So far there was no dedicated mechanism for replacing DNS servers with
new list. Add dns_resolve_reconfigure() API that allows to achieve that
in a thread-safe manner.
Introduce 3rd state in DNS context lifetime by converting from 'bool
is_used' to 'enum dns_resolve_context_state state'. This new
DEACTIVATING state allows to mark a DNS context as busy and safely close
context without holding lock. Closing DNS context with released lock
prevents deadlock in case net_context_close() has to synchronize with a
separate thread executing handler passed to net_context_recv() (which is
the case for example with ESP-AT WiFi driver).
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
ctx->is_used member seemed to be used sometimes within a mutex acquired
block, sometimes not. Make it consistent by always using it with
acquired mutex.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Query timeout handler is rescheduled if DNS context mutex is locked. So
far there was no timeout used, which means that work is simply put at
the end of system workqueue. This solves cases when mutex is locked by
any higher priority cooperative threads.
If however mutex was locked in application code within lower priority
thread (which is very likely) and query timeout has expired in the
meantime, then system workqueue is busy looping by calling query timeout
handler and trying to acquire DNS context lock.
Reschedule query timeout handler with 10ms delay, so that all
threads, including those with lower priorities, have a chance to move
forward and release DNS context lock.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Switch to the new API for delayed work related to DNS queries.
In the previous solution it was assumed that the work item could be
immediately cancelled at the point the query slot was released. This
is not true. We need a secondary condition to record the fact that
the query was completed while the work item was still pending, and an
additional check to detect when the work item completed and the slot
reclaimed.
Also annotate functions to indicate when they require the lock on
query content to be held, add some helpers that abstract core
operations like invoking a callback or releasing a query slot, and fix
some more cases where query slot content was accessed outside of the
new lock infrastructure.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
It is documented that using transient information like whether a work
item is pending or a delayed work item has time left to determine the
state of the work item before subsequent reconfiguration is prone to
race conditions, and known to produce unexpected behavior in the
presence of preemptive threads, SMP, or use of the work item from
interrupts. As a best practice such pre-validation steps should be
avoided unless algorithmically necessary.
All comparisons of remaining delayed time before canceling a delayed
work item in this module appear to be optimizations subject to the
above race conditions. Remove the checks so that only the inherent
race conditions in the implementation of canceling a work item remain.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Add more tests to verify that we discard malformed packets.
In order to simplify the testing, separate message validation to
dns_validate_msg() function in resolve.c. Allow that function to
be called from unit test. This way we can construct invalid DNS
messages in unit test and verify that they are discarded when
needed.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
-Wimplicit-fallthrough=2 requires a fallthrough comment or a compiler
to tells gcc that this happens intentionally.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Unit tests were failing to build because random header was included by
kernel_includes.h. The problem is that rand32.h includes a generated
file that is either not generated or not included when building unit
tests. Also, it is better to limit the scope of this file to where it is
used.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
It might happen that while we are waiting for the response
to one of our query, we receive a query to resolve another name.
Previously this would make the current name resolution to fail because
only responses were expected to be received.
Signed-off-by: Léonard Bise <leonard.bise@gmail.com>
As mDNS requests set DNS id to 0, we cannot use it to match
the DNS response packet. In order to allow this functionality,
create a hash from query name and type, and use that together
with DNS id to match request and response.
Fixes#21914
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
If we are returned IPv4 address but the hints says IPv6, then
return error as currently we do not support AI_V4MAPPED addresses.
Same check for IPv6 if we want only IPv4 address.
Fixes#18870
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The mDNS packet receive had issues:
* The DNS id needs to be 0 for both sending and receiving, we did
not accepted 0 incoming id.
* The mDNS response does not have any questions in it so we just
need to skip the question count checks in response.
* Skip the Cache-Flush bit in Class field so that we can properly
parse CLASS_IN value.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
In DNS server init, print information whether the DNS server
is mDNS or LLMNR one. This way we do not need to remember what
IP addresses are used either of them.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The source UDP port in all Multicast DNS responses MUST be 5353
as described in RFC 6762 chapter 6.
Fixes#18732
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
If DNS server(s) are added or removed e.g., as part of DHCP
processing, send newly defined net-mgmt events so that
a user application may get this information.
Fixes#16924
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>
So instead of building the net_pkt by itself, dns will pass its message
buffer to net_context_sendto_new(), which in turn will build the net_pkt
relevantly.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
dns_write was in the middle of this, let's just move it to a more
relevant place so we logically have macros, then local variables then
functions.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
There are issues using lowercase min and max macros when compiling a C++
application with a third-party toolchain such as GNU ARM Embedded when
using some STL headers i.e. <chrono>.
This is because there are actual C++ functions called min and max
defined in some of the STL headers and these macros interfere with them.
By changing the macros to UPPERCASE, which is consistent with almost all
other pre-processor macros this naming conflict is avoided.
All files that use these macros have been updated.
Signed-off-by: Carlos Stuart <carlosstuart1970@gmail.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>
Remove network specific default and max log level setting
and start to use the zephyr logging values for those.
Remove LOG_MODULE_REGISTER() from net_core.h and place the
calls into .c files. This is done in order to avoid weird
compiler errors in some cases and to make the code look similar
as other subsystems.
Fixes#11343Fixes#11659
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Unify the function naming for various network checking functions.
For example:
net_is_ipv6_addr_loopback() -> net_ipv6_is_addr_loopback()
net_is_my_ipv6_maddr() -> net_ipv6_is_my_maddr()
etc.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
As the debugging print calls are async, all the strings that might
be overwritten must use log_strdup() which will create a copy
of the printable string.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The return of memset is never checked. This patch explicitly ignore
the return to avoid MISRA-C violations.
The only directory excluded directory was ext/* since it contains
only imported code.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Instead of waiting forever for a network buffer, have a timeout
when allocating net_buf. This way we cannot left hanging for a
long time waiting for a buffer and possibly deadlock the system.
This commit adds checks to L2 and network support libraries.
Fixes#7571
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
This implements LLMNR client from RFC 4795. This means that caller
is able to resolve DNS resource records using multicast DNS.
The LLMNR is used in Windows networks.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Callers of dns_get_addr_info() depends on receiving an end-of-results
callback (where 'info' is NULL) when finished. Do this on failure in
cb_recv() instead of pointing to an empty object.
This fix will avoid a deadlock situation in getaddrinfo().
A reply code other than "no error" in the DNS response is a typical
case when the lookup failed and this occurs.
Moved "struct dns_addrinfo info" variable from cb_recv() to dns_read()
because it was now unused in cb_recv().
Adding tests for this requires network access and a DNS server,
or a mechanism for generating a DNS response with a reply code.
Signed-off-by: Stig Bjørlykke <stig.bjorlykke@nordicsemi.no>
In some cases, we need to initialize DNS servers from a binary
addresses, e.g. in case of DHCP processing. With existing API,
such addresses would need to be converted to strings, just to
be converted back to struct sockaddr in dns_resolve_init().
This is not efficient, and with a number of addresses quite
cumbersome. So instead, allow to pass DNS server either as
strings, or as struct sockaddr's (or both).
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
If IPv6 is disabled, then it is useless to try to resolve
IPv6 address because "struct sockaddr" does not have enough
space to store IPv6 address.
Fixes#1487
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
This implements mDNS client from RFC 6762. What this means that
caller is able to resolve "hostname.local" names using multicast DNS.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
If the query name is already in numeric format, there is no
need to send the query string to DNS server as we can just
convert it ourselves.
Jira: ZEP-2562
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
