Remove the workaround for clang 3.6.2 that did not support
'cortex-m0plus'.
clang 3.8 was already supporting it according to the PR introducing the check.
clang >=3.8 is avaible since ubuntu-xenial and debian-stretch.
The current ubuntu-bionic has clang 6 and debian-buster clang 7.
This removes overwriting 'CPU_ARCH'.
Add a 'dependency-debug' and a 'DEPENDENCY_DEBUG=1' option for
'info-boards-supported' to save some variables used when resolving
dependencies.
Print some some 'sorted' variables to simplify comparing the actual value
when the parsing order changed.
This should help tracking changes introduced when refactoring the
dependency parsing.
Microchip offers ready-to-use modules with the mrf24j40 chip.
All but the MRF24J40MA integrate an external PA/LNA, they also come
with an RF shield.
If the PA/LNA is not enabled, the signal off these modules is really
poor.
This adds pseudomodules so that the PA/LNA is automatically enabled
when the appropriate module is used.
Do not remove the '-D' and '-U' values from CFLAGS.
This prevents issues where a '-D' could contain a space.
Some values way be duplicated from the 'riotbuild.h' header and the
command line but with the same value so without conflict.
To not put too many things in the command line, the -DMODULE_NAME are
only put in CFLAGS_WITH_MACROS.
Also, as now, the deferred value of CFLAGS is used for 'riotbuild.h',
macros set after the inclusion of `Makefile.include` will be taken into
account.
This commit overrides the default flash recipe with an edbg specific
one.
The new recipe first verifies the flashfile agains the device flash. If
it is already on there, it won't get overridden.
edbg takes care of resetting the device in any case.
pic32prog is a program for flashing pic32 boards from command line on Linux.
It works with:
* Microchip PICkit2
* Microchip PICkit3 with script firmware.
* Other ones: https://github.com/sergev/pic32prog/wiki
Depending on the PROGRAMMER used, the file used for flashing may be
different, like openocd/jlink, so the file being generated when
compiling should be different.
Because of this, when building in docker, the build system needs to know
for which programmer you are building for.
This manifested for 'nordic_softdevice' applications when using 'openocd'
but is not limited to this case.
This adds a new target "archive-check".
Thin archives should be created with relative paths so that archives created
in a build container are useful in the host. This check ensures there are no
absolute paths inside thin archives.
Normal, or thick archives contain a copy of the object code. Thin
archives, on the contrary, are just an index to the .o files.
This patch does two things:
1. Change ARFLAGS to enable the "T" options.
2. Call AR with all relative paths.
The second step is necessary because the build system handles all
absolute paths. If the index in the thin archive contains absolute
paths, archives created in docker are no usable outside, and moving
the objects breaks the archive.
If all arguments to AR are relative, the resulting archive contains
filenames *relative to the .a file* and nothing should break as long
as the relative location of the .a and .o remains unchanged.
Compilation time is unchanged, but disc usage is reduced by approximately
50%. These are the result of a full RIOT build:
| Thin Archive | no | yes | Savings (%) |
| -------------- | ------: | ----: | ----------- |
| pkg (10e6 KiB) | 1 790 | 905 | 49% |
| Non pkg | 71 | 71 | 1% |
| Total | 1 812 | 976 | 46 % |
The default channel configuration is managed within the Makefiles of many
examples and tests. This commit moves the default channel logic to
`makefiles/default-channel.ink.mk` to ease maintenance. All Makefiles that
previously managed the default channel configuration themself have been
updated to include that file.
The default channel configuration of the cc110x has been added
Use the -gz option to compress ELF sections containing DWARF information.
This saves around 50% of disk space, without any side effects.
See https://gcc.gnu.org/onlinedocs/gcc-9.2.0/gcc/Debugging-Options.html#Debugging-Options
for more infomation on this option.
Some platforms have an outdated toolchain that does not support -gz so
the flag is blacklisted there. Even then, the results are quite impressive.
I used @cladmi's `buildtest` branch (https://github.com/cladmi/RIOT/tree/wip/du/buildtest)
with this change and compiled the `examples/default` application:
```
$ BUILD_IN_DOCKER=1 DOCKER="sudo docker" make -C examples/default buildtest-indocker
```
The size was obtained with:
```
$ find output -name "*.bin.bindirsize" -type f -exec tail -n1 '{}' \; | cut -f 1 | awk '{s+=$1} END {printf "%.0f", s}'
```
Results:
- Vanilla: 10328112 KB (~10GB).
- with -gz: 4982788 KB (~5GB).
This was inspired by #8496.
The current code for detecting whether colors should be used by default
was not working. Normally what programs such as git or grep do is check
if they are writing to a pipe.
To see the issue by yourself:
$ BOARD=samr21-xpro make -C tests/periph_qdec > I_SHOULD_NOT_HAVE_COLORS 2>&1
This commit uses MAKE_TERMOUT, MAKE_TERMERR and if ANY of them is NOT
a terminal, colors are disabled.
Update the optional flags to use 'OPTIONAL_CFLAGS' instead of evaluating
everytime.
They are now blacklisted by architecture/toolchain according to the
current docker reference image.
Handle declaring OPTIONAL_CFLAGS and blacklisting them with
OPTIONAL_CFLAGS_BLACKLIST.
This should replace checking everytime if options are supported.
Add make only function to convert strings to lowercase and uppercase.
This can replace the `$(shell echo $(var) | tr 'a-z-' 'A-Z_')` pattern.
Using the 'make' implementation results in being around 100 times faster.
CFLAGS_CPU is required for 'pkg/openthread'. So declare it as a global
compilation variable.
The goal is to move the export to `makefiles/vars.inc.mk` and remove the
local exports that could be in boards/cpu/arch.
Add a variable for `pyterm` specific flags that are not handled by other
terminals.
This will prevent issues with boards that have options only supported by
`pyterm` and setting `pyterm` options from the environment.
Rename TMP006 to TMP00x
Add TMP007 sensor support to TMP00X
Change uint8_t reg to uint16_t
Add to doxygen documentation group
Expose compile time configurations
Move defines from .c to .h
Change double to float, because double is not needed
Add TMP007 register information
The value must be saved from the original value to allow restoring it.
This has currently no impact on 'info-boards-supported' as it is
currently ignored.
By default if a test is available and not blacklisted it will be run by
murdock.
This will move from a whitelisting everything that works, to
blacklisting what fails. This way, failing tests will be easier to
track instead of being silently just not run.
It could even allow introducing failing tests before a fix is available.
Only set `TEST_ON_CI_ENABLED` if the application has tests.
Currently `TEST_ON_CI_WHITELIST` was only set if there were tests.
This is a pre-step to have `TEST_ON_CI_WHITELIST` being `all` by
default.
The cc110x driver has been re-written from scratch to overcome the limitations
of the old driver. The main motivation of the rewrite was to achieve better
maintainability by a detailed documentation, reduce the complexity and the
overhead of the SPI communication with the device, and to allow to
simultaneously use transceivers with different configuration regarding the used
base band, the channel bandwidth, the modulation rate, and the channel map.
Features of this driver include:
- Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers.
- Detailed documentation of every aspect of this driver.
- An easy to use configuration API that allows setting the transceiver
configuration (modulation rate, channel bandwidth, base frequency) and the
channel map.
- Fast channel hopping by pre-calibration of the channels during device
configuration (so that no calibration is needed during hopping).
- Simplified SPI communication: Only during start-up the MCU has to wait
for the transceiver to be ready (for the power regulators and the crystal
to stabilize). The old driver did this for every SPI transfer, which
resulted in complex communication code. This driver will wait on start up
for the transceiver to power up and then use RIOT's SPI API like every other
driver. (Not only the data sheet states that this is fine, it also proved to
be reliable in practise.)
- Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was
about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms
(depending on SPI clock and on CPU performance) (measured with ping6).
- Increased reliability: The preamble size and the sync word size have been
doubled compared to the old driver (preamble: 8 bytes instead of 4,
sync word: 4 byte instead of 2). The new values are the once recommended by
the data sheet for reliable communication.
- Basic diagnostic during driver initialization to detect common issues as
SPI communication issues and GDO pin configuration/wiring issues.
- TX power configuration with netdev_driver_t::set() API-integration
- Calls to netdev_driver_t::send() block until the transmission has completed
to ease the use of the API (implemented without busy waiting, so that the
MCU can enter lower power states or other threads can be executed).
This allows passing other arguments as environment variables in
DOCKER_ENVIRONMENT_CMDLINE and as command line override in
DOCKER_OVERRIDE_CMDLINE without using the automatic detection.
This will allow passing USEMODULE to docker as its value is usually set
in an application Makefile so has its origin changed.
This allows setting other variables that should be exported to the
docker build.
As for other variables, they must still be unmodified before parsing
`docker.inc.mk` to be exported.
The value must be saved from the original value to allow restoring it.
This has currently no impact as 'makefiles/defaultmodules.inc.mk'
and 'DISABLE_MODULE' are ignored in this parsing.
This remove executing buildtest `for` loop in docker.
When building completely in docker, 'buildtest' would hide issues when
the host toolchain would be used when doing `make all` directly.
It has the consequence that it now starts a container for each
compilation which is slower.
The previous behavior can be reproduced by using
BUILD_IN_DOCKER=1 make buildtest-indocker
A side effect is also that now `BUILDTEST_MAKE_REDIRECT` would work when
doing `buildtest` with docker.
Add a 'buildtest-indocker' that forces executing 'buildtest' for loop
completely inside the container.
It prevents starting one container per compilation wich is slower but
it could hide errors where the host toolchain would be used
It is currently equivalent to `buildtest` but will change when the
`buidtest` handling will be move outside of `BUILD_IN_DOCKER`.
Display an error when executed without BUILD_IN_DOCKER=1.
