Closes: #28331
Configurable initialization priority is useful when using CDC ACM
serial device, for example.
Signed-off-by: Ievgenii Meshcheriakov <ievgenii.meshcheriakov@nordicsemi.no>
Shell was accepting CONFIG_SHELL_ARGC_MAX+1 arguments which lead to
memory corruption and fault. Added error message reported when number
of provided arguments exceeds CONFIG_SHELL_ARGC_MAX.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
This fix prevents a scenario when command callback calls shell_fprintf
after the shell cleared the command context flag and before setting
the prompt correctly.
Fixes#27522
Signed-off-by: Jakub Rzeszutko <jakub.rzeszutko@nordisemi.no>
shell_fprintf requires that formatted output be emitted with a
putchar()-like output function. Newlib does not provide such a
capability. Zephyr provides two solutions: z_prf() which is part of
minimal libc and handles floating point formatting, and z_vprintk()
which is core and does not support floating point.
Move z_prf() out of minimal libc into the core lib area, and use it
unconditionally in the shell.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
The smp_shell_rx_byte has been renamed to smp_shell_rx_bytes and now
accepts data buffer pointer and its size as parameters. Return value
has been changed to size_t and represents number of bytes processed from
the given buffer.
The change has been done to more efficiently serve most common scenario
when the function is called in loop to process buffer, byte by byte.
Previously such operation required passing each byte separately,
with the change the function will work directly on source buffer
reducing number of calls and byte copy operations.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
Now that device_api attribute is unmodified at runtime, as well as all
the other attributes, it is possible to switch all device driver
instance to be constant.
A coccinelle rule is used for this:
@r_const_dev_1
disable optional_qualifier
@
@@
-struct device *
+const struct device *
@r_const_dev_2
disable optional_qualifier
@
@@
-struct device * const
+const struct device *
Fixes#27399
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@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>
In case when SMP has been enabled, the shell_thread would be signaled
even if no data has been read from fifo.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
In case when shell ring buffer gets full the data may still be taken
from the UART fifo, byte by byte, and accepted by the SMP back-end,
if the back-end has been enabled.
Unfortunately the uart_rx_handle failed to check if it has been
successfull in reading a byte from the fifo, before passing it to
the SMP for processing, which could lead to processing random stack data
as a part of an SMP frame.
Additinally, when the SMP would have accepted the byte,
the uart_rx_handle would fail to signal shell_thread, that the SMP has
internally buffered data that could be processed.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
So far SMP shell transport was initialized in APPLICATION run level, but
shell over UART was initialized in POST_KERNEL. This could end up in
situation when received frames were scheduled for further processing in
SMP layer, when it was not initialized yet.
Export smp_shell_init() function declaration and call it before shell is
initialized with all its receive data handlers. This prevents situation
when data is scheduled for processing in SMP layer, when that one is not
ready yet.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Shell will display a warning message if there was a request to print
a message on the not initialized shell backend.
Signed-off-by: Jakub Rzeszutko <jakub.rzeszutko@nordisemi.no>
Disable shell print functions if the shell is not initialized.
Update tests accordingly.
Fixes#27161
Signed-off-by: Jakub Rzeszutko <jakub.rzeszutko@nordisemi.no>
Signed-off-by: Steven Slupsky <sslupsky@gmail.com>
This commit adds comment clarifying how does DTS chosen node,
'zephyr,shell-uart', impacts default value of option assigning UART
port to the shell.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The definition of the help command has been changed
so that it does not accept the arguments.
Thanks to this it cannot be marked by the "select" command.
Signed-off-by: Jakub Rzeszutko <jakub.rzeszutko@nordisemi.no>
If the command buffer is empty the shell was trying to read a string
from a random memory location. Added a condition checking "argc" value
when the select command is executed.
Fixes: #27227
Signed-off-by: Jakub Rzeszutko <jakub.rzeszutko@nordisemi.no>
By adding a prompt, users can alter the default value
of SHELL_LOG_BACKEND.
While it makes sense that when you have a shell, the log
messages will, by default, be shown there,
sometimes you want to disable this selection.
For example, when you have a telnet shell (SHELL_BACKEND_TELNET)
and want to debug the network subsys.
Signed-off-by: David D <a8961713@gmail.com>
It is not needed by the command to have subcommand in order to be used
for prompting. Removed this condition from the shell engine.
Fixes#21819
Signed-off-by: Jakub Rzeszutko <jakub.rzeszutko@nordisemi.no>
If shell is enabled then enable all sub-shells if their dependencies are
satisfied. This was done for some modules and subsystems but was not
consistent.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Currently this is useful only for some internal applications that
iterate over the device table, since applications can't get access to
a device that isn't ready, and devices can't be made unready. So it's
introduced as internal API that may be exposed as device_ready() when
those conditions change.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Several shell modules use cloned code to iterate over all devices and
identify the nth instance that meets some criteria. The code was
repetitive and included various errors. Abstract to a helper function
that performs the check consistently.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
The extern array declaration of size 0 gives a warning when
compiling with GCC. Updated to use [] rather than [0].
Signed-off-by: Emil Gydesen <emil_gydesen@bose.com>
After recent changes to shell, there is no more "no_such_command:
command not found" message when executing non-existing command. Restore
that message, so users are warned once again about wrong command,
instead of silently ignoring their request.
Fixes: 512de5ecac ("shell: Refactor command execution to enable raw
arguments")
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Some Kconfig defined devices may be defined using dt_chosen_label
function. Since there is no way to ensure a device enabled in dts
is also defined in Kconfig, it may happen that instance is not
actually defined.
In this case device_get_binding might return 0, leading to undefined
behavior in the function that calls it.
When not already done, systematically check return of function
device_get_binding on devices defined through dt_chosen_label macro.
Trigger ASSERT when required and return error when possible.
Fixes#20068
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
Default values were fixed in the code. Moved to Kconfig to allow
customized configuration. Custom configuration may be used to prevent
line breaking injected on terminal width.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
When the device driver model got introduced, there were no concept of
SYS_INIT() which can be seen as software service. These were introduced
afterwards and reusing the device infrastructure for simplicity.
However, it meant to allocate a bit too much for something that only
required an initialization function to be called at right time.
Thus refactoring the devices structures relevantly:
- introducing struct init_entry which is a generic init end-point
- struct deviceconfig is removed and struct device owns everything now.
- SYS_INIT() generates only a struct init_entry via calling
INIT_ENTRY_DEFINE()
- DEVICE_AND_API_INIT() generates a struct device and calls
INIT_ENTRY_DEFINE()
- init objects sections are in ROM
- device objects sections are in RAM (but will end up in ROM once they
will be 'constified')
It also generate a tiny memory gain on both ROM and RAM, which is nice.
Perhaps kernel/device.c could be renamed to something more relevant.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Added special flag that can be used to indicate that optional
arguments are passed without any parsing (e.g. quotation marks
removal). Modified execute command to parse command line buffer
argument by argument.
After this change it is possible to forward whole command to
command handler (using select).
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
At present it is not possible to write a printf()-like function in
board code which outputs to the shell. Add shell_vfprintf() to permit
this.
Signed-off-by: Simon Glass <sjg@chromium.org>
It is useful to run tests which generate shell output and check that it
is correct. Update the existing 'dummy' backend to support this.
It works by retaining the output in a small buffer so that it can be
read and checked by the test.
Signed-off-by: Simon Glass <sjg@chromium.org>
Improve hexdump output of shell_hexdump() to match what is currently
printed by LOG_HEXDUMP*() family of functions. That way string buffers
that are sent/received using shell commands can be easily presented
using shell_hexdump() without loosing user experience (with previous
shell_hexdump() that printed only hex bytes) and code validity (printed
buffers are not always NULL terminated with only printable characters).
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
New function allows to set from the code the root command. It is
an equivalent of calling 'select <rootcmd>' except it sets command
for all shell instances.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Kernel timeouts have always been a 32 bit integer despite the
existence of generation macros, and existing code has been
inconsistent about using them. Upcoming commits are going to make the
timeout arguments opaque, so fix things up to be rigorously correct.
Changes include:
+ Adding a K_TIMEOUT_EQ() macro for code that needs to compare timeout
values for equality (e.g. with K_FOREVER or K_NO_WAIT).
+ Adding a k_msleep() synonym for k_sleep() which can continue to take
integral arguments as k_sleep() moves away to timeout arguments.
+ Pervasively using the K_MSEC(), K_SECONDS(), et. al. macros to
generate timeout arguments.
+ Removing the usage of K_NO_WAIT as the final argument to
K_THREAD_DEFINE(). This is just a count of milliseconds and we need
to use a zero.
This patch include no logic changes and should not affect generated
code at all.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Replace all occurences of BUILD_ASSERT_MSG() with BUILD_ASSERT()
as a result of merging BUILD_ASSERT() and BUILD_ASSERT_MSG().
Signed-off-by: Oleg Zhurakivskyy <oleg.zhurakivskyy@intel.com>
There is a new init level SMP that was just added, and this module needs
to take it into account.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
This reverts commit 8739517107.
Pull Request #23437 was merged by mistake with an invalid manifest.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
