Change-Id: Ie223f458776c81439db8cd4c3a25b4753599588b
Signed-off-by: Anas Nashif <anas.nashif@intel.com>

Change-Id: I66c71f7322e1529d7b5bfda5b29eeb078b13daf1
Signed-off-by: Anas Nashif <anas.nashif@intel.com>

Or else, PINMUX_NUM_PINS is not found.

Change-Id: I48c173395f2cda72e8915c46aad964e83509a605
Signed-off-by: Daniel Leung <daniel.leung@intel.com>

Set the target to ARC to speed up flashing. the difference:

before
 downloaded 27320 bytes in 50.685825s (0.526 KiB/s)

after

 downloaded 27320 bytes in 3.396626s (7.855 KiB/s)

Change-Id: I53ca26f97eefd40e869662b137f5a659082aa478
Signed-off-by: Anas Nashif <anas.nashif@intel.com>

Change-Id: I72a4947119f1f2b6fcd59fadbe38a54b6bc1ea61
Signed-off-by: Maciej Borzecki <maciek.borzecki@gmail.com>

Exclude STM32F103RB based platforms from the test. The SRAM is overflown
by 14kB.

Change-Id: I37c2905ed2706a89d9a199bc49518937659ac997
Signed-off-by: Maciej Borzecki <maciek.borzecki@gmail.com>

For STM32F103RB targets, the SRAM is overflown by 2-3kB in microkernel
test. Platforms with this SoC can still be built for in the nanokernel test.

Change-Id: I012b93cf8dfec74292f7ab228f4b2fca1a4f3444
Signed-off-by: Maciej Borzecki <maciek.borzecki@gmail.com>

Exclude STM32F103RB platform from the test. The SRAM is overflown by
>20kB, while the CPU itself has 20kB of SRAM.

Change-Id: Ic9aad6b88d517b62f4a18901cd698ba9a9defb40
Signed-off-by: Maciej Borzecki <maciek.borzecki@gmail.com>

Disable the test for STM32 based platforms until an implementation of
timestamp counter becomes available.

Change-Id: I2e50dac36dbfdc61081610c0e0cf1ace8892f602
Signed-off-by: Maciej Borzecki <maciek.borzecki@gmail.com>

Take advantage of shell_init() accepting const commands array. This
moves ~280 bytes from SRAM to the text section.

Change-Id: Id64ee766e3c6cf7ce4cc623a1e21d3dacf33f050
Signed-off-by: Maciek Borzecki <maciek.borzecki@gmail.com>

Take advantage of shell_init() accepting const commands array.

Change-Id: Ie8c023de62cda75f7057184555806401f0381a1a
Signed-off-by: Maciej Borzecki <maciek.borzecki@gmail.com>

The commands array is not expected to change and can be declared
const. This allows callers to cleanly declare their commands lists as
const, effectively moving the structures from SRAM to code section.

Change-Id: Ie1710622b8cfa609e129eb79712f910f1d1aace3
Signed-off-by: Maciej Borzecki <maciek.borzecki@gmail.com>

Enable build tests for Nucleo-F103RB and STM32 MINI A15 platforms.

Change-Id: I26ccdb6edd537f5dfc9c0120e5a6d4ee1ed63dbb
Signed-off-by: Maciek Borzecki <maciek.borzecki@gmail.com>

The comments in the sensor header specify that pressure should be
expressed in kPa, but the bmp280 driver returns a value in Pa.

Change-Id: I6d5346db250d1a01a1e5e31fb1d8685ab5dc405b
Signed-off-by: Vlad Dogaru <vlad.dogaru@intel.com>

SPI flash depends on SPI. Enable the SPI flash driver only when
SPI is enabled.

Change-Id: I902588b806a4a5773fddb58a7567a8e0d4ba9fe0
Signed-off-by: Baohong Liu <baohong.liu@intel.com>

Fixes this built error:
zephyr/drivers/pinmux/dev/pinmux_dev_quark_mcu.c:138:18:
error: 'PINMUX_BASE_ADDR' undeclared here (not in a function)
.base_address = PINMUX_BASE_ADDR,

Change-Id: I54f15bc262cb887117f7b56660bcf85983daa877
Signed-off-by: liu.lei <lei.a.liu@intel.com>

From now on user/app has possibility to dismiss Passkey Notify authentication
on acceptor side even with DisplayOnly interface. The action on 'Cancel' API
in this case is disconnection.

Change-Id: I4be198482c23c0ccaeb0112b72ff269037e03583
Signed-off-by: Arkadiusz Lichwa <arkadiusz.lichwa@tieto.com>

Some devices may actually lose track of CCC setting so this is required
to work around the problem, in case the device does work track CCC
properly the extra write shall not cause anything more than one extra
round trip thus it is probably work doing it anyway.

Change-Id: I9e5ed3fa459e4617c6fae094d0ce0f80cb2682e4
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Implement bt_gatt_attr_read_cud() making working peripheral_esp app.

Change-Id: Ic9634bf31a39ae9cd55279de34ee4c0c0995a4dd
Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>

Deal with controllers that don't support controller to host control.
This may potentially lead to dropped packets so make a clear warning
log of the issue.

Change-Id: Ie8fdaed826a072fd157343721222f618328d59a3
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>

Reading supported commands isn't an LE-specific feature, so the
command should be in common_init() rather than le_init().

Change-Id: I613bbe8d39b4c2b6dadc45a710bc59568ec9b488
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>

Change-Id: I36fd9b96d247184b4a986467f1609a9c3b0cbd9f
Signed-off-by: Anas Nashif <anas.nashif@intel.com>

Splits the test into two clear sub-categories: microkernel and nanokernel.
This is done to maintain consistency with other tests that do the same
(e.g. test_obj_tracing, test_sema, test_stackprot, test_timer).

Change-Id: Iddb1ec8d569a9d953fb8af5ce08e87b51995f821
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>

It will extensively test all provided functions within a normal usage of
the API.

Change-Id: I723203a29c3f3416b464030a7fe34eac5fff6095
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>

As for double linked list provided through misc/dlist.h, this provides a
generic API for single linked list in misc/slist.h. It follows the same
naming rule as in dlist.h

Change-Id: I955bd16a201bc9987c29f5a9e3e3d8447682a71e
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>

Fixes line length and whitespace issues flagged by the checkpatch tool.

Change-Id: Id49cd5341571ac7893929a2836fe5e06166abe06
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>

When an ethernet frame is larger than UIP_BUFSIZE,
a net_buf is allocated but never released. Therefore
after few bad frames, no more RX network packet
can be received.

Fixed by allocating the net_buf after checking
the frame length.

Change-Id: I436487e3c26d739de347b4db6facc3a3dbebbe75
Signed-off-by: Sebastien Griffoul <sebastien.griffoul@intel.com>

Reduces the amount of memory allocated for fiber's stack.
This permits the test_context to compile on boards like
nucleo-f103rb. Sanitycheck script passes all test for
test-context.

This patch was verified on the emulated platforms and not
against real hardware boards.

Change-Id: I219f63063ee2dca5b0326e25141d8b37f4cd1d74
Signed-off-by: Yannis Damigos <giannis.damigos@gmail.com>

Reduces the amount of memory allocated for task's stack.
This permits the test_pool to compile on boards like
nucleo-f103rb. Sanitycheck script passes all test for test-nano.

This patch was verified on the emulated platforms and not
against real hardware boards.

Change-Id: Ib8041f6d91e8ffc5fcb16dc73de1f7662d9596da
Signed-off-by: Yannis Damigos <giannis.damigos@gmail.com>

Reduces the amount of memory allocated for stack (both fiber
and task). This permits test_sema and test_mutex to compile
on boards like nucleo-f103rb. Sanitycheck script passes all
tests for test_sema and test_mutex.

This patch was verified on the emulated platforms and not
against real hardware boards.

Change-Id: Ie25288bbbbfa64bfc5f7463639bfb09639cc184f
Signed-off-by: Yannis Damigos <giannis.damigos@gmail.com>

Remove those kconfig options that are SoC specific, and
should not be configurable via kconfig.

This also separates IRQ_PRI into one for DW and one
for QMSI, to follow the convention of every other
drivers.

Change-Id: I338f819f71c18fa9e17015e8a588a3d0207350c6
Signed-off-by: Daniel Leung <daniel.leung@intel.com>

Remove those kconfig options that are SoC specific, and
should not be configurable via kconfig.

Change-Id: Ic73783189db57059d2b7f3727e4802e1b2e27931
Signed-off-by: Daniel Leung <daniel.leung@intel.com>

Remove those kconfig options that are SoC specific, and
should not be configurable via kconfig.

Change-Id: Ifd65097a65f80539cac073f95aadc2d8e42efb9f
Signed-off-by: Daniel Leung <daniel.leung@intel.com>

This extends the pwm_dw sample app to build for
FRDM_K64F board also. Because of this, pwm_dw
has been renamed to pwm.

This allows us to put it through sanity test
to catch build breakage.

Change-Id: Ie4d7442af43ede58f4cec0ec28643384a685321a
Signed-off-by: Daniel Leung <daniel.leung@intel.com>

Remove those kconfig options that are SoC specific, and
should not be configurable via kconfig.

Change-Id: Ifdbb5e3a997795ef577350d88f8cb06877eb6463
Signed-off-by: Daniel Leung <daniel.leung@intel.com>

Remove those kconfig options that are SoC specific, and
should not be configurable via kconfig.

Change-Id: Ib8158f00a6c6616360ddbcf63981f1a85911c1b9
Signed-off-by: Daniel Leung <daniel.leung@intel.com>

On Galileo Gen2 board SPI uses GPIO pin for CS.
Thus spi_intel driver must be initialized after gpio_dw,
so it's init level must be more that gpio_dw's one.

Change-Id: If2e5fb1106afe01e5567cf3fe72063bdc94ad3d2
Signed-off-by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>

Change-Id: Ifd977289c92d3c88ee2ca93dd964307372d49ca0
Signed-off-by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>

GCC enables unaligned memory access for ARMv6 and above.
Once unaligned memory access is allowed, there's no need
for the unaligned memory access trap.

The change prevents the situation when the compiler
generates code that uses unaligned memory access, but
the CPU generates an exception when this code runs.

Change-Id: Id33f2264c631772e5c561e76fb579d8b7bc26e1e
Signed-off-by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>

This patch extends the i2c shim driver so it supports the 'default
configuration' infrastructure which enables the user to specify a
default configuration for the I2C device. The default configuration
is set during driver initialization.

This patch also changes Quark SE and D2000 Kconfig.defconfig files so
the i2c default configuration is set to '0x12' which means standard
bus speed, 7-bit addressing and master mode. This is the same value
used when DW driver is selected, by the way.

Change-Id: I06e0dc3c29e8da2f3317db5bef285177f2e92c9a
Signed-off-by: Andre Guedes <andre.guedes@intel.com>