Hi,
For a prof of concept I've added an initial repository file [1] and some
module files [2] for the xpcc drivers. Now you can do all the things you
always wanted [3] by using lbuild:
Display all the available targets:
$ ../library-builder/scripts/lbuild-discover -r"source/repo.lb"
--discover=repository:options
modm-platform:target = [at90can128, ... long list of all targets ...,
stm32f779ni]
Display all available modules for a specific target:
$ ../library-builder/scripts/lbuild-discover -r"source/repo.lb"
--discover=modules -D":target=stm32f303k6"
modm-platform:adc
modm-platform:can
modm-platform:clock
modm-platform:core
modm-platform:dma
modm-platform:gpio
modm-platform:i2c
modm-platform:id
modm-platform:spi
modm-platform:spi_uart
modm-platform:timer
modm-platform:uart
Get the available options for those modules together with their current
value and possible values:
$ ../library-builder/scripts/lbuild-discover -r"source/repo.lb"
--discover=module:options -D":target=stm32f303k6"
modm-platform:core:allocator = newlib [block_allocator, newlib, tlsf]
modm-platform:core:enable_gpio = True [True, False]
modm-platform:core:main_stack_size = 3040 [512 ... 8192]
modm-platform:core:vector_table_in_ram = False [True, False]
To check it out yourself you need the following repositories cloned in the
same folder:
https://github.com/dergraaf/modm-platformhttps://github.com/dergraaf/library-builderhttps://github.com/dergraaf/xpcc in branch feature/lbuild
Then you can run:
$ cd modm-platform
$ make discover
Building doesn't work yet because the mapping between the device file
content and substitutions for the jinja2 templates is not exactly the same
as before. Therefore the next step is to clean up that mapping and remove
the special cases which are at the moment hard-coded in the xpcc python
build tools.
E.g. '<core>cortex-m4f</core>' gets mapped to
'properties["core"][0]["value"]' which is quite cumbersome [4]. The correct
thing would be something like 'properties["core"]'. But that leads to
problems if there would be a second 'core' tag. Like there is for the
'define' tags.
Any idea on how to proceed with that?
Cheers,
Fabian
Btw. should I create a pull request for the discussions about the code?
Unfortunately the interesting parts are spread over the three repository,
therefore it would only work for the driver files.
[1] https://github.com/dergraaf/modm-platform/blob/master/source/repo.lb
[2]
https://github.com/dergraaf/xpcc/commit/309655a5e1e960f6af9ac43dfc50bbd79ab…
[3] DISCLAIMER: May not be all the things you always wanted.
[4]
https://github.com/dergraaf/xpcc/commit/309655a5e1e960f6af9ac43dfc50bbd79ab…
Hi,
in an effort to make xpcc build system independent I had a look at the
platform generation and therefore the XML device files. To understand
how they are set up I wrote a schema against all of them can be validated:
https://github.com/dergraaf/modm-platform/blob/master/tools/device/device/r…
The schema could be refined a bit more (not only string type for
attributes) but should be good starting point. Is there a documentation
available what the different attributes mean? Especially for the
attributes marked as optional the use differs quite a lot between the
different architectures.
Cheers,
Fabian
Hi all,
I'm a new subscriber--my name is Shane Leonard, and I'm an electrical
engineering
master's student specializing in embedded systems at Stanford. I came
across XPCC in a personal quest to come up with a creative hardware
register access scheme, and your blog post on the subject was by far the
best I could find :) I'm currently rolling a basic platform for a
development board I made for the ATSAM4L, and as it turns out, I find
myself looking at your codebase quite a lot for inspiration. I know the AVR
and STM platforms are fairly removed from the Atmel SAM family, but I
wanted to gauge your opinions on whether it could be a good idea to fork
XPCC and start scratching together a framework for SAM or more specifically
SAM4L. I'll likely end up doing it for my own purposes either way, but I
would probably write more production-quality code from the start if I
suspected a chance for this to make it into the main codebase.
Thanks, and keep up the awesome work!
Shane
