The gpio pins numbers aren't just matching the physical pin numbers...
on some distros like raspbian, with a kernel explicitly setting the
base to 0, it works, but for kernels using the first free slot it
doesn't (see gpiochip_find_base() in kernel's gpiolib.c).
To get the proper gpio pin value, we thus need to find the base for
the corresponding gpiochip. For the raspberry pi, we search for the
chip with a label containing 'bcm2835'.
Signed-off-by: Cédric Bosdonnat <cbosdonnat@suse.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
While some distributions have a modified /proc/cpuinfo including
the board revision for raspberry pi, most distributions don't
have this.
In order to guess the raspberry pi version in such cases, use
the /proc/device-tree/compatible content as a fallback solution.
The values expected from this file have been taken from
upstream kernel documentation:
Documentation/devicetree/bindings/arm/bcm/brcm,bcm2835.txt
Signed-off-by: Cédric Bosdonnat <cbosdonnat@suse.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
As per Documentation/ABI/testing/sysfs-firmware-ofw, perfer
/proc/device-tree over /sys/firmware/devicetree/base as this
is the stable one.
Signed-off-by: Cédric Bosdonnat <cbosdonnat@suse.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Change PLATFORM_RASPBERRY_PI_ZERO into an integer to avoid
build error when comparing it to integer platform_detected.
Signed-off-by: Cédric Bosdonnat <cbosdonnat@suse.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Like the beaglebone, the phyBOARD-Wega also got an am335x.
So I merged the four mmap-functions for gpio_context and some
identical defines in a separate header and c-file.
The new platform support-files are based on beaglebone-files.
The documentation of phyBOARD-Wega is still in process,
but for now there are enough informations.
At this time it is possible to use GPIO-Pins and Uart0
(tested via python with mraa). The code for using SPI, I2C and
PWM is also still in process and not tested yet.
Signed-off-by: Norbert Wesp <nwesp@phytec.de>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Every 96Boards baseboard uses the same expansion connector, but the
current 96Boards initialization code assigns the SoC specific GPIO
numbers instead of being consistent across all boards. Change the code
to assign 96Boards GPIO-A through GPIO-L to MRAA GPIOs 23-34.
v2: Create info for all the LS connector pins, not just the GPIO pins.
This gives us room to describe other pins in the future.
Signed-off-by: Grant Likely <grant.likely@linaro.org>
Jorge Ramirez-Ortiz <jorge.ramirez-ortiz@linaro.org>
Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Somewhere between kernel v3.18 and v4.4 the output of /proc/cpuinfo
stopped showing the board model. This confused MRAA when running on
HiKey. Eventually we want to do a better job of getting the kernel to
express hardware information to userspace, but in the mean time fix the
test so that MRAA will work on HiKey.
Signed-off-by: Grant Likely <grant.likely@linaro.org>
Cc: Jorge Ramirez-Ortiz <jorge.ramirez-ortiz@linaro.org>
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Brendan Le Foll <brendan.le.foll@intel.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Switch to using calloc on all calls to malloc where the memory isn't
initialized. For things like the mraa_board_t, not allocating all to zero
causes issues such as with the sub_platform member, where if that's not zero
mraa_get_platform_type will try to dereference a random memory location for the
sub_platform->platform_name, which can result in segmentation faults and other
issues.
Note that in some places where immediately after the malloc call is a copy
operation, there is no need for calloc, as all the memory gets overwritten
anyways, but in cases where there may or may not be memory written to (such as
in mraa_file_contains, with reading from a file), even though in most cases the
memory is overwritten, it could be the case that the read operation does
nothing, but the memory still has non-zero values, by virtue of the fact it
wasn't overwritten.
Signed-off-by: Ian Johnson <ijohnson@wolfram.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
This patch adds support to 96boards. 96boards is an open platform
specification. This spec strandardizes the pins on the external
connectors like Low speed and High speed. Given these pins are standard
across multiple boards, I think adding a generic 96boards made sense to
me.
The idea behind adding this generic board file is to make mraa work on
most of the 96boards with minimal changes to board support in libmraa.
This patch adds support to the LS expansion connector which has got 12
gpio pins + 2 i2c + 1 spi and 2 uarts.
For now I have added Dragaon board DB401c support as part of this patch
and is tested.
More info about board @ https://www.96boards.org/products/ce/dragonboard410c/
Long term plan is to get all this configuration from the /sys and
populate the board specifics dynamically, which is bit easy with
96boards specs in-place.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
NPM builds don't support cross compiling and will always compile for the host.
Binding.gyp now uses mraa_NPM_SRCS instead of mraa_LIB_SRCS_NOAUTO which
contains all platforms. Cmake now uses mraa_LIB_PLAT_SRCS_NOAUTO to provide all
the sources. NPM builds will build objects for all platforms reguardless of
space which does meann slightly bigger binaries.
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
The latest Raspberry Pi 2 revision string is missing from the discovery
function mraa_raspberry_pi() and results in incorrect pin mapping defaulting to
PI_B Rev 1.
Signed-off-by: Paul Barrette <paul.barrette@windriver.com>
Signed-off-by: Paul Barrette <paulbarrette@gmail.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
src/arm/CMakeLists.txt added beaglebone.c as a dependency src/arm/arm.c added
initialization of beaglebone black platform src/arm/beaglebone.c mraa support
for the Beaglebone Black Rev B+C includes support for mmap access to gpio
Signed-off-by: Michael Ring <mail@michael-ring.org>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Includes SPI,I2C,GPIO and mmap access to GPIO. This commit removes the _b
suffix to the rpi board file
Signed-off-by: Michael Ring <mail@michael-ring.org>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Used for getting the path to the character device under linux for uart
device. i.e. "/dev/ttyS0"
Adds paths to existing platforms.
Closes#84
Signed-off-by: Thomas Ingleby <thomas.c.ingleby@intel.com>
They needed offsetting by +1, due to starting the config at 1 instead of
0 as it was during development
Signed-off-by: Thomas Ingleby <thomas.c.ingleby@intel.com>
