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>
Since on most 3rd party breakout boards the i2c bus exposed is 1 and not 6, it
makes more sense to have that as the default i2c bus
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
The MRAA library needs the debugfs interface in order to access GPIO
pin modes attributes. Mounting the debugfs is not always desirable so this
patch instructs libmraa to first check for the required attributes in
the normal SYSFS CLASS GPIO interface and to use debugfs as a fallback.
Signed-off-by: Mihai Serban <mihai.serban@intel.com>
Signed-off-by: Constantin Musca <constantin.musca@intel.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Calling mraa_intel_edison_misc_spi() is not really needed, because the
GPIO 10..13 pins will be configured correctly when the user actually
inits them as GPIO pins. When using these pins for GPIO,
mraa_setup_mux_mapped() and mraa_intel_edison_gpio_init_post() will do
all this work based on the pin map and pinmodes for the Edison Arduino
board.
On the contrary, this function would break any user of the SPI bus
that is already running.
Signed-off-by: Jonas Norling <jonas.norling@connode.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
There are technically two i2c buses that are user accesible on minnowboard max
but the other one is on the high speed expansion header that I've never tested
and it's not mapped in mraa so leave this to 1 until someone enables it.
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Fixes#251. default_i2c_bus was set to 7 before setting actual parameters for
it leading to obvious failure as 0 != 7
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>
Edison SPI driver doesn't support LSB_FIRST mode, we will now return
_FEATURE_NOT_SUPPORTED instead of _INVALID_RESOURCE
for attempts to enable it.
Closes#22.
Signed-off-by: Alex Tereschenko <alext.mkrs@gmail.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Galileo Gen1 doesn't support LSB_FIRST mode, so let's
return MRAA_ERROR_FEATURE_NOT_SUPPORTED instead of <...>_INVALID_RESOURCE.
Closes#178.
Signed-off-by: Alex Tereschenko <alext.mkrs@gmail.com>
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Gpio 111 controls the use of hardware CS by the edison kernel's SPI driver.
This is an issue as that CS will go high between every byte transmitted. The
solution is to let gpio 111 alone and let the driver decide what is best. This
fixes#137
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
This reworks the mraa_intel_edison_i2c_freq() function. i2c_dw_sysnode is a
folder not a file so we need to grab the mode file from that folder. We now
also support i2c-1 frequency changing
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
This stops the legacy beahviour that was to initialise as a gen1 galileo if we
failed to find a valid dmi name. Closes#142
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Not defining this causes undefined behaviour leading to the check mux_total
sometimes being true and asking the kernel to export all sorts of garbage gpio
Signed-off-by: Brendan Le Foll <brendan.le.foll@intel.com>
Instead of plain char* to avoid need to malloc and do all the needed
error checking needed that goes along with that.
Signed-off-by: Thomas Ingleby <thomas.c.ingleby@intel.com>
