In order to restore i2c context after entering
SYS_PM_DEEP_SLEEP, suspend and resume functions
are called.
The following parameters are restored:
- All registers for the I2C device.
- I2C clocks.
I2C depends on the DMA which is enabled separately.
The suspend/resume functionality is implemented in
the QMSI shim layer as a fast and temporary solution,
it will be removed and migrated to QMSI later.
Change-Id: I22ca2fb9884109f7b5ca75af572811187a531403
Signed-off-by: Julien Delayen <julien.delayen@intel.com>
In order to restore watchdog context after entering
SYS_PM_DEEP_SLEEP, suspend and resume functions
are called.
The following parameters are restored:
- All registers for the WDT device.
- Watchdog interrupt mask.
The suspend/resume functionality is implemented in
the QMSI shim layer as a fast and temporary solution,
it will be removed and migrated to QMSI later.
Change-Id: I6e4e30a797b0ca38ce727d50fa51d1b2f8330405
Signed-off-by: Julien Delayen <julien.delayen@intel.com>
DEFINE_DEVICE_PM macro was not defining device_pm_ops
as 'static'. Fixes the issue and impacted areas.
Jira: ZEP-639
Change-Id: I5e1de6af97bf7b2b690af0c81034ce167e655e43
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
Change tags to reflect the type of samples, in this case it is all
about sensors.
Change-Id: I9f87e8683dd52fd757cd07ba5aa9fad99d17ab38
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Fix a compilation warning. Compilation of the i2c driver
requires QMSI dma object file. As a solution, the current
code has "selects DMA_QMSI" in i2c Kconfig. This results
in a compilation warning since "select DMA" is also
required. Adding "select DMA" and "select DMA_QMSI"
together in i2c Kconfig can solve the compilation issue.
But, this will cause the dma driver be unnecessarily
enabled. So, the solution is to fix the issue in the QMSI
part instead of in the shim driver.
Jira: ZEP-777 ZEP-778
Change-Id: Id4c1f749bf71eece1e70583bd586a69b596768b2
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
Remove this app, since the sensor is not on the
board any more.
Jira: ZEP-763
Change-Id: I931acfe9f01b817bc0b24a4e0745bf759e53da90
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
Remove the function wrappers around the 64 bit math and just use C
syntax natively, combining ops where appropriate (e.g. there was a
sequence implementing "(x<<2+x)<<2" to do "multiply by 10"). The
_ldiv5 and _rlrshift routines are non-standard ops that provide useful
abstraction, so they remain as separate functions.
Change-Id: I4d83847348fdd7be09887b833c8ccbd2aa1e4182
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The _to_float() implementation had a somewhat kludgey hand-written 64
bit math implementation, which is unhelpful on Zephyr as all our
toolchains provide a working uint64_t runtime. This is at best just
dupicated code from libgcc, and at worst less efficient.
This patch replaces the existing 64 bit minilibrary but keeps the
uint32_t[2] API as is for ease of validation and review.
One exception is _ldiv5, a specialized divide-by-five implementation.
The 64 bit division routines are large on some architectures (ARM and
ARC in particular), not pulled in by a default Zephyr build, and will
swamp the benefit from this patch. So this includes a
refactored/improved _ldiv5 which leverages libgcc for multiword shifts
instead of just using raw division.
Note also the "noinline" attribute on _ladd(). This is a workaround
for an apparent compiler bug when built with -Og or -Os (hand-hacking
the Makefiles to build with -O0 works), perhaps due to my aliasing the
int array with a long long. This will go away in phase 2.
Change-Id: I63e8c82dabe2bfaa75b63ddb59e5f11d51be538e
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The _to_float routine was apparently written to be able to take a 32
bit float bit representationa as well as a 64 bit double. But in a
printf routine, that can never happen per the C standard (where floats
are always promoted to doubles in varargs context).
This was just hard-configured to 1 at the top of the file, and nothing
else in the project sets "DOUBLE" to try to change it. Just remove
it. If we ever want code to convert a float to a double in memory so
we can use this routine, we have it in libgcc. Or even in hardware on
the FPU where available.
Change-Id: I796814c0fce3ce96faa34fde8da411a28c826699
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Hex, octal and decimal all had separately implemented reduction loops
to generate strings. With only a little work these can all be unified
to a single implementation that works with an arbitrary base.
Performance is probably a little lower owing to the fact that
hex/octal now requires a division per character, and the extra
"reverse the string" trick at the end of the conversion. But code
size savings are substantial.
Change-Id: I11ff376aeca1483f974d328271e19918221b2a41
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The Sensor Subsystem has different IRQ numbers for the same device
across CPUs. QMSI declares these, but doesn't do anything to try to
map it automatically, so we have to do it in the driver.
Similarly the SCSS interrupt routing (the SoC-wide "APIC", which is
distinct from the per-CPU interrupt masking hardware) needs to know
which CPU we're trying to unmask an interrupt for.
Change-Id: I27e5ec44f324f869e16a92e7ef40a22a305d98a2
Jira: ZEP-546
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Don't try to detect the UART device manually, the list is missing a
bunch of boards (the Quark SE sensor subsystem variant, and both
Arduino 101 halves). There's a config variable for that.
Change-Id: I903e52dcb6025e5da824faadb5e4bc59221fa210
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The ARC side should use the same console UART as x86 by default if we
want identical behavior.
Change-Id: I067860581cfd93d97ffad3d8f0bc5591f555e3ce
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
This output is going to a terminal device, not a Unix tty descriptor
which will cook the output for us. Newline moves the cursor down.
Carriage return moves it to the start of a line. We need both.
Change-Id: If93d1a85d16cea93b4788fa55e694a7b77055bfe
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Overwhelming industry practice when using two-byte line termination is
to end the line with "\r\n" and not "\n\r".
Also remove needless casts. GCC does not warn on int-to-char
conversions.
Change-Id: If01a2e5ca327aacebcbcaf384af9c7131119689b
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
These impede debugging and we have CONFIG_OMIT_FRAME_POINTER
now which does this globally for the entire kernel.
Change-Id: I46939223e27dd298ca3ed162ff5790cb2e9ed2a2
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
There is a total of 39 IRQs defined at nrf52/soc_irq.h.
Change-Id: Id478fb15a07cfdecaa6cc136730d20017b8752c5
Signed-off-by: Ricardo Salveti <ricardo.salveti@linaro.org>
Quark D2000 and SE based boards (but Arduino 101) use QMSI bootloader
by default. QMSI bootloader sets up GDT in the so-called 'basic flat
model' just like Zephyr does by default.
This patch changes Quark D2000 and SE boards default configuration
so they rely on QMSI bootloader and we don't sets up GDT twice.
Change-Id: Ic6e520148b732bd48c00657c6c8138a8d865faef
Signed-off-by: Andre Guedes <andre.guedes@intel.com>
sys_put_le64 stores a 64-bit unsigned integer into an
arbitrary memory location and is alignment-safe for any
architecture.
Change-Id: Ifd9871a509b9cab05a59d81f4917c68dda3cc824
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Zephyr uses the last priority level for the PendSV exception, but
sharing prio can still be useful on systems with a reduced set of
priorities, like Cortex-M0/M0+.
Change-Id: I767527419dcd8f67c2b406756b9208afd3b96de0
Signed-off-by: Ricardo Salveti <ricardo.salveti@linaro.org>
Add documentation for OS X users that want to compile
using an available 3rd-party cross compiler and also
complete the section to describe the existing support
for the GCC ARM Embedded toolchain.
Change-Id: I97d27d46c0048a49029dbe4d66fe141755c226e9
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
_irq_controller_isr_vector_get() now returns -1 if it couldn't
determine which vector was activated.
Issue: ZEP-602
Change-Id: Ib0f5dbc3b68cc5e2c3a23121530e178aede20d06
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
You can't query the LOAPIC for every kind of interrupt that fires,
it has no idea about IRQs that were generated by an 'int' instruction
for example. Extend the semantics of _irq_controller_isr_vector_get()
to return -1 if the vector can't be identified.
Issue: ZEP-602
Change-Id: I1174aa62fbedffdcd329d60da8ef14fabb042dc3
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Change the signal ramp up/down config parameters in i2c driver
module to SoC specific.
Jira: ZEP-753
Change-Id: Ie01f1d890a7133d30ea53eee07f60354734a8571
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
Fix "TCF: especify ARCH when building" missed feeding ARCH to the
initconfig phase, that some projects require.
Change-Id: Ie44c9056d2d44198abbac6aeb6e0df05aff4dc36
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
By default TCF was running test single core test cases in both Quark
SE cores at the same time.
This worked because the ARC output came through the IPM console to
x86, which enabled them to coexist.
However, after default configuration changes, this is no longer
possible without adding extra hardware to the current test setup,
which is cost prohibitive, and thus we are driving ARC output to the
same port as the x86 output.
This would make the output single-core test cases that are ran at the
same time in the x86 and ARC cores mix their outputs.
So we have that scenario disabled (and makes us run less test
cases). However, multicore test cases still can be run by selecting:
@targets bsp_model:x86+arc
Change-Id: I5f6b43246a24d0b263fd4a47866c7f0f99d20505
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
The current code redirects the ARC output to UART0.
Our test setup is laid out with a single serial port (which is mostly
UART1). For simplicity (as the extra cabling would be prohibitive), we
will change the default test building to redirect output back to ARC1.
This implies that the output from test cases will be intermixed when
test cases are not prepared to share the serial port, so running of
testcases combined in x86+arc will have to be disabled by default and
only done in testcases which declare they need to do so.
Change-Id: I5a6b79ecb48ecfb2d711be7f5fdbe08b4c3d34e9
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Currently it plucks configuration strings in/out of the $O/.config and
that is quite dirty.
There is a cleaner way to do that with configuration fragments which
is now enabled by being able to include them from the object
directory.
Change-Id: Id51881391293ace89b58153c5de7db1afe67118a
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Current setup includes configuration fragments (*.conf) from the
source directory.
This patch enables said inclusion to also happen from the output
directory; this enables automatic build systems and validation to set
configuration fragments without modifying the source tree.
This is important so that the source tree remains pristine from the
standpoint of source control systems.
Change-Id: Ic4a896afad2b8525c13dd6c6d9081f514b18e7aa
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Without this line break, when CONFIG_NET_BUF_DEBUG is enabled, the
output may get confusing.
Change-Id: Ic467511f1dbbd283710f51ef14f94e0269c7c0ea
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
- Set default board to Quark SE Debvboard
Jira: ZEP-690, ZEP-760, ZEP-782
Change-Id: I7f7cc45b527a116f88cf45796e46b429ae6f0060
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
The 'z' length specifier is the appropriate one to be used with size_t
(%zu) and ssize_t (%zd) types. Having support for this in our libc
means that we can utilize the compiler format string checks
(__printf_like) without getting warnings of incorrect format
specifiers for size_t and ssize_t variables.
Change-Id: I73fec0145692e0a59934cab548caf24c1c16a3df
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
The i2c_burst_read function transfers two messages to the I2C device.
The first message writes the internal address of the device, the second
message then reads bytes. A repeated start (restart) is needed to change
the direction from write to read, but the restart flag should be set in
the second message rather than the first.
Change-Id: I7d48de2f4866e3b514f689f41ce5d28eba90c39f
Signed-off-by: Maureen Helm <maureen.helm@nxp.com>
Add a shim layer around the ksdk I2C driver to adapt it to the Zephyr
I2C interface. Currently only supports master mode.
Jira: ZEP-717
Change-Id: I704b8c38e22e456bb9fa4325682b2a354a27a7ba
Signed-off-by: Maureen Helm <maureen.helm@nxp.com>
Paranoid with older compare set logic, the changes in this
commit is better in avoiding compares being set to stale
count values (values in past). Compares shall always be set
in future w.r.t to current count value in the counter h/w.
No longer using fp_worker_sched() to explicitly run the
worker because there may be no ticks elapsed when worker
runs, which leads to ticker job running thereafter not to
update the compare to a new value, probably leading to a
stalled controller roles.
Change-id: If07f648357585ebb80c35d1320210eac3137781c
Signed-off-by: Vinayak Chettimada <vinayak.kariappa.chettimada@nordicsemi.no>
The old __CYGWIN32__ does not work with the 64-bit version
of Cygiwn. Using __CYGWIN__ instead allows us to build
gen_idt.c under 64-bit Cygwin.
Change-Id: I6590767153e929764f0b448f74bb8e3f2e593a6a
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
On OS X (macOS), "make menuconfig" fails with missing
linker symbols. By specifying the "-l" linker options
with the menu, panel and ncurses libraries directly if
the autodetection fails, it builds and links properly.
Note that this might denote a problem with
"check-lxdialog.sh".
Change-Id: Ib2721646cc01c3e977911d8e6d0c8303dcedbc58
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
We no longer produce nanokernel.elf or microkernel.elf.
Change-Id: I37cb985ddc90f4129f122b9209d306898a71a596
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
This adds method which initialize data pool for L2CAP data.
Change-Id: I2b170489aa4b2afc3c1f962591f377e23d147161
Signed-off-by: Grzegorz Kolodziejczyk <grzegorz.kolodziejczyk@tieto.com>
Signed-off-by: Mariusz Skamra <mariusz.skamra@tieto.com>
Currently bt_rfcomm_dlc() is returning error code and dlc is returned
as output param. Since caller does not need to check error, it can
return NULL if some error happens
Change-Id: I86e85b67545edb960b4b745267e6645017373bf6
Signed-off-by: Jaganath Kanakkassery <jaganathx.kanakkassery@intel.com>
Data sent by user will be queued in a FIFO. A TX fiber will be
started for each DLC which process this FIFO and write to L2CAP.
Fiber will sleep in two scenarios - no buffer in FIFO, no TX
credit. Credit is handled using semaphore. So if credit is 0
then fiber will wait on semaphore and will be scheduled when
it receives credit.
Change-Id: Id6e796eed594b28d6fb6e4259d3ed52634db9335
Signed-off-by: Jaganath Kanakkassery <jaganathx.kanakkassery@intel.com>
ccc cfg is allocated when a peer updates a ccc value. This
fix will retain a cfg only if the new value is not default.
Change-id: I586082818145e43c771a6fccdb0bf2b3cecdd30c
Signed-off-by: Vinayak Chettimada <vinayak.kariappa.chettimada@nordicsemi.no>
Start using Zephyr's include/bluetooth/hci.h for HCI command
handling.
This is the first commit in a series that transitions from
the structures in hci.c to the ones in hci.h.
Jira: ZEP-726
Change-Id: I1d79586b85d25c8ef707b8a2b19a27a77b08b819
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
